the michaelis-menten kinetics and errors in estimated reaction constants … · 2018-09-04 ·...

1267

J. Indian Chem. Soc.,Vol. 94, December 2017, pp. 1267-1278

The Michaelis-Menten kinetics and errors in estimated reaction constants :A reappraisal†

Sharmistha Dhatt and Kamal Bhattacharyya*

Department of Chemistry, University of Calcutta, Kolkata-700 009, India

E-mail : [email protected], [email protected]

Manuscript received 04 November 2017, accepted 17 November 2017

Abstract : Reaction constants in traditional Michaelis-Menten type enzyme kinetics are most often determinedthrough a few linear plots. Such graphical plots sometimes provide reasonably good data. But, for a com-parative survey, visual examinations are not enough. Instead, it is always advisable to go simultaneously fora few associated numerical tests. They can assess how far the measured values of the quantities sought arereliable, furnishing along with appropriate error estimates that can be checked against the corresponding non-linear methods. Here, we explicitly deal with a few such situations to stress the importance of these tests withspecific examples. Also, we advocate a new scheme to mainly highlight how cases may appear with negativereaction constants and explore the origin of such bizarre findings. This exercise, in turn, might even judgethe quality of the input data set and modify it to suit the needs. Pilot calculations bring the worth to light.

Keywords : Enzyme kinetics, Lineweaver-Burk type plots, negative reaction constants, reliability of kineticdata, data-set errors.

1. Introduction

An enzyme-catalyzed reaction is usually assumedto follow the Michaelis-Menten (MM)1 mechanism,given by

k k

k

1 2

1

E S C P E

.

Here, the substrate S combines with the enzyme E toform the complex C that subsequently yields the prod-uct P, regenerating the catalyst E. Denoting the con-centrations of the species of interest by the corre-sponding small letters like s, e, c, etc., the relevantrate equations are obtained as

dc/dt = k1e.s–(k–1+k2)c;

V=dp/dt=k2c. (1)

Note that the rate of product formation V in (1) andthe concentrations of all the species involved are ac-tually time-dependent. In other words, V = V(t), c =c(t), etc. The initial conditions, viz., s = s0 and e =

†Acharya P. C. Ray Memorial Lecture (2016).

e0 at t = 0, allow us to write the following two con-servation equations, in addition to (1), for complete-ness of the problem :

e0=e+c,

s0=s+c+p. (2)

All other rate equations, i.e., expressions for de/dtand ds/dt, follow from (1) and (2). From experiments,however, it is not easy to estimate the individual rateconstants like k1, k–1 or k2 in (1). Instead, one findsthat two characteristic reaction constants, defined by

Vm= k2e0, Km= (k–1+k2)/k1, (3)

are obtainable. They turn out to be really importanttoo in all discussions of enzyme kinetics1–3. In (3),Vm signifies the maximum possible rate assuming, ofcourse, that e0 s0, while Km stands for the Michae-lis constant. If e0 s0 is obeyed, Vm still remains aconstant, but the maximum possible rate will then begiven by k2s0. The governing equation connecting

J. Indian Chem. Soc., Vol. 94, December 2017

1268

rate and these reaction constants is usually written as(see later for justifications)

VmsV = ————. (4)

Km+s

Considerable attention has therefore been paid on (4)over the years. Several linearized versions of (4) havealso appeared to simplify the task of obtaining Vmand Km. However, these are often beset with prob-lems like the accuracy of such linear plots2–6 in esti-mating the reaction constants, the reasons behind theirdeviations from linearity7, their comparative perfor-mances8,9, etc. In this respect, a few instructive edu-cational notes10 are also available.

It is easy to notice that (4) may be rearranged forlinearity in a number of ways. Of these, a few rel-evant ones are as follows : First, we have the Hanes-Woolf (HW)11 method, where sV–1 is plotted againsts. The Lineweaver-Burk (LB) plot12 comes second,which recommends a plot of V–1 vs. s–1. The thirdscheme refers to a plot of V vs. Vs–1 and it is theEadie-Hofstee (EH)13 proposal. We summarize theserearrangements below :

s Km sHW : —— = —— + ——

V Vm Vm

1 Km 1LB : —— = ——— + —— (5)

V Vms Vm

VKmEH : V = Vm – ———

s

With time, however, it has been noted that the LBmethod received the most attention1. Indeed, the LBscheme still remains a premier tool in various studieson enzymatic reactions14 though some authors stronglydisagree4,8,9. Hence, the endeavor to somehow ex-tract good reaction constants is continuing15,16. Theproblem is thus an open one.

One important reason behind the observed dispa-rity among the values of reaction constants obtained

by various means is, of course, the rather uncriticaluse of (4). Therefore, before proceeding further, weshould pay some attention to the arrival of (4) fromthe basic eqs. (1) and (2). First, let us concentrate onthe time t = at which the rate of reaction is maxi-mum. Suppose, this occurs at a definite c=c– , corre-sponding to the maximum attainable value of c [c– min (s0,e0)] for specific values of the rate constantsand the initial conditions, that ensures

dc/dt = 0, t=. (6)

For convenience, the concentrations of other relevantspecies at this time (t=) may be denoted by e—, s—

etc. Condition (6) is the most general one. Puttingthis condition in (1), we obtain

k1e—.s— – (k–1+k2)c

— = 0;

V—

= k2c—. (7)

Using the conservation for e0 in (2), one finds from(7) the exact equation

Vms—

V—

= ————— ; t = . (8)Km + s—

This equation looks precisely like (4). The lesson,therefore, is that (4) should strictly be used only att=. Notably, (8) does not demand one to invoke thesteady-state approximation (SSA). Secondly, if theSSA holds over a time span, this implies that, inplace of (6), we have

dc/dt 0, 1 t 2. (9)

Thus, over a range of time, the complex concentra-tion, and hence the rate, remains almost constant atV(t) V— though s may considerably deviate from s0.Here, we have the better (see below) option of em-ploying V(t) and the corresponding s(t) in (4). Thirdly,the most popular version of (4) reads as

Vms0V = ————. (10)

Km+s0

It involves, apart from the SSA, further approxima-tions like e0<<s0 and c

—2 0. The significance of V

Dhatt et al. : The Michaelis-Menten kinetics and errors in estimated reaction constants : A reappraisal

1269

in (10) is just its virtual constancy, and no referenceto time is explicit here. In essence, therefore, the keyeq. (10) rests on the following assumptions : (i) TheSSA is valid. (ii) The time 2 is small enough toensure that p— 0. (iii) The condition e0<<s0 ismaintained. (iv) The stipulation c—2 0 holds. (v)Finally, we need to also guarantee that the measuredrate is almost constant within the concerned (1 t 2) time interval. We shall see, if one disregards theabove constraints in the course of employing (4), orany of its linearized versions appearing in (5), conse-quences may be dangerous.

A critical survey of the relevant literature2–10 re-veals that discussions on the following points are stilllacking : (i) How do the HW, LB and EH methodstheoretically compare? How far does the nonlinearfitting procedure involving (4) fare over the conven-tional linear exercises based on (5)? Is there any quan-titative measure? (ii) Is there any other alternative tothe nonlinear fitting? (iii) Why does one sometimesfind negative reaction constants and when? (iv) Howshould one ensure the quality of data set being handled?Is there a way to improve it?

In what follows, we like to take up the above is-sues for clarifications. Obviously, our attention willbe paid to numerical procedures, not to the drawingof the plots or their visual deviations from linearity.

2. Experimental

2.1. The numerical experiments

The set of coupled nonlinear differential equationsthat follows from (1) and (2) is solved stepwise viathe fourth-order Runge-Kutta method17. The stabilityof the numerical experiment is checked against thesatisfaction of conservations in (2). The time step hasbeen fixed at 10–6 min. We start with s0 = 10 moland follow the dynamics up to 20 min. This is onerun. Then, gradually, s0 is increased in steps of 10units. Six such runs are performed to prepare a set ateach of two fixed e0 (one at 1 mol and the other at0.1 mol) values and the results are recorded.

2.1.1. Choice of rate constants

To explore the relative performances, we need tochoose the sets rather judiciously. In the present case,the rate constants are taken as k1 = 10.8 mol–1

min–1, k–1 = 10.0 min–1 and k2 = 0.8 min–1 18.Thus, when we fix e0 at 1 mol, Vm = 0.8 molmin–1 and Km = 1.0 mol follow. These exact val-ues will later be compared against the findings fromdifferent schemes. Henceforth, we shall continue withthe corresponding units for other quantities, but maynot explicitly refer to them, for brevity. Notably, theabove choice of the rate constants has been known18

to lead often to marked errors in estimated Km, inparticular. So, we felt obliged to consider such val-ues for exploration.

2.1.2. The sets

It has already been remarked that one should actu-ally use V

— and s— in places of bare V and s in (4).

This forms the ideal set. This is the set at t = . Thenext worse alternative is to employ V(t) and the cor-responding s(t), without checking the restriction ofconstancy of the former (at SSA). This class is repre-sented also by sets A. However, provided that theconditions listed below (10) are obeyed, we mayemploy s0 in (4) for s. Thus, we have another set (setB) at hand for every set A.

2.2. The methods

2.2.1. Linear methods

Here, the left side of (5) is fitted via a least squaresmethod (LSM) as a function of the variable at theright side. The calculated slope and intercept thenyield the reaction constants in (3). This is most popu-lar numerical scheme for any of the (HW, LB or EH)methods. However, the plain graphical approach doesnot offer any quantitative estimate of error. Theoreti-cally, an overall measure of error for the calculatedVm and Km may be found from the expression

Mm j

jm jj

V sV

M K s1

1

(11)


1270

where M observations form a set. Thus, (11) offers away of judging the comparative performances of theHW, LB and EH procedures via numerical means.

In view of the widespread belief that SSA is betterobeyed with larger s0, reducing, as a consequence,the corresponding V–1 to be employed in LB calcula-tions, it has been suggested that a better strategy spe-cifically for LB would be to employ a weighted LB(WLB) method where V4 acts as the weight factor6.Here, the graphical approach does not help. But, onemay carry out the procedure in the same manner asthe plain LSM. Only, the averages are computed withthe weight factors. It is also recommended as a closealternative6 to the nonlinear fit (NLF) method. Hav-ing obtained the values of Vm and Km via this scheme,one may again stick to (11) for finding the overallerror involved.

2.2.2. The nonlinear method

As the next option, one may go for a NLF of data.In this scheme, one tries to minimize the error in(11) by choosing suitable Vm and Km. It may be ac-complished as follows. (i) Start from some arbitraryKm. Calculate the two average values N and D, de-fined by

N = V , D=s/(Km+s), (12)

to obtain Vm as Vm = N/D. (ii) Allow random varia-tions of Vm around this value to find a minimum valuefor . (iii) Save the values of , Vm and Km. (iv)Change Km systematically in both directions to repeatsteps (i) to (ii) and, only if the new is less than theprevious minimum, proceed to step (iii). The recipewill finally yield a minimum along with the optimalVm and Km.

2.2.3. A new alternative

Keeping aside schemes like the LSM or NLF, thereis another possibility to get the reaction constants withinthe MM framework. It does not seem to have beenexplored at all. In a set, we have M points (Vj, sj).Consider, for example the case with i=1, j=2. Us-

ing (4), one can solve for the two unknowns fromthis pair to obtain the results

s2–s1Vm(1,2)=——————— ;

s2/V2–s1/V1

m mV KV V s s1 2 1 2

1 1 1 1(1,2) (1,2) .

(13)

If a set is arranged in order of increasing s and V, wenotice from the second expression in (13) that Vm(i, j)/Km(i, j) is always positive. However, bare Vm(i, j) ispositive only when the restriction

sj/si > Vj/Vi, j > i, (14)

is maintained. Otherwise, both Vm and Km becomenegative. We shall see later the emergence of such asituation. For the points (Vj, sj), j = 1, 2, .. M, onewill get M(M–1)/2 values of Vm and Km on applying(13) in a pair wise manner. We call it the pair wisesolution (PS) scheme. The results for Vm and Km maybe separately averaged, obtaining along with theirstandard deviations (Vm and Km). This PS schemethus offers, very unlike the other schemes sketchedearlier, independent errors of the reaction constants.

3. Results and discussion

3.1. Validity of the steady-state approximation

Given the rate constants and initial conditions, it isnot always transparent how one could ensure the va-lidity of the SSA within the MM framework. Variouscriteria have been put forward from time to time. Theadequacy of such riders, along with a number of freshones, has been tested elsewhere19 quite elaboratelyfor closed systems. The case of open systems, how-ever, is entirely different20. Concerning the presentstudy, we display the dynamics for a few runs in Fig.1. It is pretty clear that the SSA is not always obeyed.The choice is deliberate, but is required to show theintricacies of the schemes in vogue. Clearly, the SSAis worst for the case of s0 equal to 10 units and is bestwhen s0 = 60. Fig. 2 shows the temporal profiles of


1271

these runs for the relative substrate concentrations/s0. Notably, the ‘reactant stationary approximation’(RSA)18 is not valid for any run, though SSA is re-spected at least for the run with s0 = 60.

3.2. Use of exact data

In Table 1, we display the first set of relevantresults for our study. Here, the time t = refers tothe maximum point of the rate-vs.-time plot, i.e., where(dv/dt)=0. Hence, (6) and (8) are exactly obeyed.For this set, we take e0 = 1.0. We find everywhere << 1 min. No approximation is made anywhere onthe computed data too (correct up to 14–15 decimal

places). This, therefore, forms the best set and itshould be considered for an ideal study.

Results of the above set are presented in Table 2.It is particularly comforting to note that all the threelinear methods based on LSM perform almost com-parably with the NLF scheme. We also notice thatestimated Vm is much less susceptible to errors thanKm. This particular feature is retained in all the stud-ies to follow. The moral is clear. If we use near-exact

Table 2. Results of employing the LSM vs. the NLFmethod to obtain Vm and Km using the best set of data

displayed in Table 1

Method VM KM EH 0.799 993 0.999 723 3.92 E-06

HW 0.799 999 0.999 924 4.04 E-06

LB 0.799 993 0.999 713 3.93 E-06

NLF 0.799 995 0.999 700 3.22 E-06

Exact 0.8 1.0 –

Table 1. Exact results of runs with rates V at times t = and the corresponding substrate concentrations starting

from varying initial substrate concentrations (in mol) s0at a fixed e0 equal to 1 mol. The time t = refers to themaximum point of the rate-vs.-time plot, i.e., where V(t) ismaximum; corresponding s values at are also tabulated

s0 t = s V

10.0 9.040986588168579 0.720344622657550

20.0 19.01538836571947 0.760023566573087

30.0 29.00473769964218 0.773337937185753

40.0 38.99558317935295 0.779998670263430

50.0 48.99791693764029 0.783999727616150

60.0 58.99427138294789 0.786665660507531

Fig. 2. Temporal profile of the relative substrate concentra-tion at e0 = 1.0 that shows the failure of the reactantstationary approximation.

data for (s—,V—

) in (4), good estimates are really ob-tained via any one of the linear schemes. The aboveassertion is true irrespective of whether the SSA orthe RSA is obeyed or not. Moreover, the use of NLFmay not offer any extra advantage in such situations.

3.3. Use of approximate data

Trouble arises when we approximate the data setby truncation. In Table 3, we display sample resultsfor our computed data. From the data for different

Fig. 1. A plot of the temporal behavior of the calculated rate,showing that the SSA is not maintained at lower s0.Runs are conducted at e0 = 1.0 and the rate constantsare given in Sec. 2.1.1.


1272

runs summarized in Table 3, the following sets maybe considered for study : (i) set of values (truncated)for (s—,V

—) at t = [to be called set 1A]; (ii) set of

values for (s0,V—

) that is more traditional [to be calledset 1B]; (iii) set of values for (s, V) where V(t) ands(t) are measured at t >> [sets 2A – 4A]; (iv) set ofvalues for (s0, V) where V(t) is assumed constant andmeasured at t >> [to be called sets 2B – 4B]. Allthese 8 sets are then employed to extract values of Vmand Km by employing the linear schemes mentionedabove. For consistency, we take s and V correct up to2 and 3 decimal places, respectively, and report thereaction constants up to 3 decimal places or 4 digits,as appropriate.

3.3.1. Performance of linear methods

Results from the sets in Table 3 are displayed inTable 4. All the three LSM-based strategies, viz., the

EH, HW and LB, are employed everywhere and theoverall errors are noted. For comparative purposes,such calculated errors are far superior and more quan-titative than those found from visual plots. Note thatthe performance is best for set 1A, as expected. Cal-culated values reflect that set 4B is not just theworst one; here, we first encounter negative reactionconstants as well. Besides, there is a gradual rise in as t increases. The error also rises in going from setA to set B. As experienced earlier, Vm is again foundto be more robust than Km under any circumstances.An added point is, the HW scheme usually overesti-mates Km. We also notice that the use of s(t) and thecorresponding V(t), even if t >> (e.g., at t = 20min), do not seem to cause very serious errors in (4)or (5), provided we measure s(t) and V(t) for all runsat the same t. This is reflected from the observedmuch better performance of set 4A than set 4B.

Table 3. Results of runs at different times with varying initial substrate concentrations (in mol) s0 at a fixed e0 = 1mol. The rate V at a certain time t and the substrate concentration at that instant s are displayed, along with s0 for the

specific run. The sets are numbered and the values are truncated

s0 t = (1A) t = 5 m (2A) t = 10 m (3A) t = 20 m (4A)

s V s V s V s V

10.0 9.04 0.720 5.64 0.680 2.58 0.578 0.09 0.066

20.0 19.02 0.760 15.29 0.751 11.58 0.736 4.61 0.658

30.0 29.00 0.773 25.18 0.769 21.35 0.764 13.81 0.746

40.0 39.00 0.780 35.13 0.778 31.26 0.775 23.55 0.767

50.0 49.00 0.784 45.11 0.783 41.20 0.781 33.41 0.777

60.0 58.99 0.787 55.09 0.786 51.16 0.785 43.33 0.782

Table 4. Employment of the LSM to obtain Vm and Km from the data sets given in Table 3. Sets A employ the actualvalues of s at the times of rate (V) measurements, while sets B rely on the more traditional choice of s0 in place of s. The

sets are numbered in order of increasing time. Absolute error () in each case reveals the adequacySet Method

EH HW LB

Vm Km Vm Km Vm Km 1A 0.800 1.007 1.25 E-04 0.800 1.015 2.16 E-04 0.800 1.007 1.24 E-04

2A 0.800 0.996 2.10 E-04 0.800 1.007 2.97 E-04 0.800 0.996 2.10 E-04

3A 0.800 0.990 2.87 E-04 0.800 1.005 5.46 E-04 0.800 0.990 2.80 E-04

4A 0.800 1.001 2.36 E-04 0.800 1.001 2.19 E-04 0.800 1.001 2.46 E-04

1B 0.802 1.135 3.39 E-04 0.802 1.116 3.77 E-04 0.802 1.136 3.40 E-04

2B 0.816 1.935 3.33 E-03 0.810 1.703 3.71 E-03 0.816 1.955 3.36 E-03

3B 0.863 4.503 1.51 E-02 0.836 3.440 1.55 E-02 0.871 4.809 1.58 E-02

4B 0.341 –15.59 4.02 E-01 –1.345 –128.1 2.39 E-01 –0.314 –52.49 1.64 E-00


1273

3.3.2. Primary sources of errors

The first source of error lies in the lack of respectfor set 1A. One may next opt for set 1B if measure-ments of s(t) become troublesome. At large times,

wildly once we replace s by s0. A good reason thatexplains the discrepancy is the following. If e0 << s0is not obeyed, SSA fails to hold. So, after a prettylong time ( << t = 20), s(t) will differ significantlyfrom s0. Had e0 been further reduced from 1.0 to0.1, keeping s0-values for the sets unchanged, theSSA would have been obeyed for all the runs (seeTable 11 later). As a consequence, the errors incurredby the substitution of s0 for s(t) is much more seriousat higher e0, our current case of concern. Fig. 3 re-veals it quite transparently; here, we show the per-cent errors for the said substitution in two cases.

3.3.3. Performance of the nonlinear method

The real advantage of going over to NLF is clearonce we compare the absolute errors in Table 4 withthose in Table 6. Particularly, in respect of set 4B,considerable betterment has been achieved by adopt-ing the NLF procedure. The negative results have alldisappeared now. The absolute error has decreasedsignificantly. A sane value for Vm has emerged, thoughKm is not quite good. Therefore, when we deal spe-cifically with a poor set of data, we need to definitelyimplement the NLF and check the outcomes againstthe LSM.

3.3.4. The weighted LB method

By now, it is clear that the set 1A is the best and4B is the worst among all those given in Table 3. So,for brevity, we may discuss about these sets only. Inorder to additionally note the importance of the re-placement of s0 for s(t), one may include sets 1B and4A. Results of employing the WLB scheme on thesesets are shown in Table 7. It is comforting to observe

Table 5. Magnifications in percentage errors at large times(set 4) for the various linear schemes owing to the replace-

ment of the actual values of s (set 4A) by s0 (set 4B)

Property Method Set

4A 4B

Vm EH 0.02 57.32

HW 0.01 268.18

LB 0.03 139.19

Km EH 0.10 1658.54

HW 0.06 12913.70

LB 0.12 5349.23

Fig. 3. Plots showing a significant reduction of percentageerror in substituting s0 for s at t = 20 min when thecondition e0 << s0 is obeyed. The ordinate is in thelogarithmic scale.

Table 6. Results of applying the NLF method on all the sets displayed in Table 4 for the various linear schemes. Abso-lute error () in each case reveals the worth

Set NLF Set NLF

Vm Km Vm Km 1A 0.800 1.006 1.15 E-04 1B 0.802 1.143 3.14 E-04

2A 0.800 0.996 1.89 E-04 2B 0.817 2.019 3.19 E-03

3A 0.800 0.989 2.69 E-04 3B 0.872 4.992 1.50 E-02

4A 0.800 0.995 2.18 E-04 4B 0.884 6.855 8.48 E-02

situations change. Concentrating only on the sets 4Aand 4B, we display in Table 5 how the calculatedpercent errors for the two reaction constants behave

JICS-3


1274

that the negative results have again disappeared fromset 4B. Moreover, the error is close to the NLF methodand the values found are quite good; indeed, a muchbetter Km is found. Thus, the WLB procedure showsthat it can also compete with the NLF method whenwe deal with a very poor data set. One may findadditionally that the NLF fares only marginally thanthe WLB in terms of , as had been rightly pointedout elsewhere6.

3.3.5. Performance of our scheme

While we have noted the suitability of a few well-known schemes already in vogue, it is still not clearwhy the plain linear schemes have mostly (except forVm by the EH method; see Table 4) yielded negativereaction constants for the set 4B. Another problem isthat the goodness of computed Vm or Km could not beseparately gauged; one could at most obtain an over-all error . We need to explore as well whether thereexists any simple alternative to the linear schemesthat can assess a priori the goodness of a chosen

Table 8. Results of adopting the PS scheme on set 1A.The upper and lower entries display values of Vm and Km,

respectively

2 3 4 5 6

1 0.800 0.800 0.800 0.800 0.800

1.008 1.000 1.006 1.006 1.010

2 0.799 0.800 0.800 0.801

0.977 1.002 1.002 1.014

3 0.801 0.801 0.801

1.052 1.032 1.052

4 0.8 0.801

1.0 1.051

5 0.802

1.128

input data set and try to modify it. These issues willconcern us right now.

3.3.5.1. The emergence of negative reaction con-stants

Results of the PS scheme, outlined in Sec. 2.2.3,are first shown for the set 1A in Table 8. As ex-pected, scatter in the estimated values of either Vm(i,j) or Km(i, j) is little in case 1A. It is also important towitness that Vm(i, j) values differ from the exact Vm(equal to 0.8) only in the third decimal place, if at all,while the Km(i, j) data are mostly affected from truevalue at the second decimal place. Thus, we againnotice the relative robustness of Vm in comparisonwith Km. However, when this very scheme is appliedto set 4B, the values behave wildly, as Table 9 re-veals. Particularly, negative values show up in theentire first row. We now know that the origin lies inthe violation of condition (14). Note also that thescatter is still much less for Vm in comparison withKm. Therefore, this becomes a general feature; mea-sured Km is always more susceptible to error, specifi-cally when we try to simplify (4) by (10).

3.3.5.2. Goodness of individual reaction constant

As indicated earlier, the PS scheme yields 15 (6×5/2) different values for either Vm or Km for a set of 6members. These values can be averaged. They should

Table 9. Results of adopting the PS scheme on set 4B. Theupper and lower entries display values of Vm and Km,

respectively

2 3 4 5 6

1 –0.083 –0.180 –0.302 –0.459 –0.669

–22.510 –37.226 –55.746 –79.530 –111.295

2 1.018 0.919 0.884 0.863

10.954 7.942 6.855 6.242

3 0.838 0.829 0.822

3.690 3.324 3.042

4 0.820 0.814

2.751 2.442

5 0.808

1.995

Table 7. Results of employing the WLB prescription onthe extreme sets 1 and 4

Set Vm Km 1A 0.800 1.009 1.47 E-04

4A 0.800 0.999 2.26 E-04

1B 0.802 1.124 3.58 E-04

4B 0.837 3.885 9.89 E-02


1275

be ideally called V–––

m and K–––

m, but we shall call themplainly as Vm or Km for the sake of brevity, unlessspecific need arises. The standard deviation (SD) canbe calculated as well. In Table 10, we summarize afew findings. Since the averaging procedure is quitedifferent in the LSM from the PS scheme, the effectsdiffer. In fact, the averaging via LSM in cases ofEH, HW and LB schemes are all very dissimilar. Asa result, data found for PS in Table 10 upon averag-ing do not have any parallel in the EH, HW or LBcalculations (see Table 4). However, this table tellsus first about the sanctity of the individual estimatedvalues. If the SD values are termed as Vm and Km,we should theoretically have Vm<<|V

–––m| and

Km<<|K–––

m|. Even modest estimates should satisfy Vm/|V

–––m|<0.1 and Km/|K

–––m|<0.1. Note that,

in this respect, set 4B is miserable. The scatters aresuch severe that Vm can lie anywhere within 0.462 ±0.580 and Km within -17.138 ± 36.172. Table 10thus also furnishes separate error estimate for eachof the two reaction constants. As an alternative, onemay look for the overall error here too by puttingthese average estimates in (11). But, we do not intendto assess it here.

3.3.5.3. Reliability of input data

We have learnt from the discussion above, alongwith Tables 4, 6 and 10, that sets 1A to 4A and 1B to4B are progressively worse in respect of calculationsof the reaction constants. Indeed, the PS scheme alonetells us nicely about the quality of a set. A reliable setshould obey the conditions Vm<<|V

–––m| and

Km<<|K–––

m|. In other words, goodness of a setmay be judged on the basis of the PS tables. Forexample, one can conclude safely from Tables 9 and10 that the set 4B should not be employed in calculat-ing Km and Vm by any method. This is a definiteadvantage to be exploited in the next subsection.

At lower e0, the SSA is satisfied in a much betterway. Hence, the replacement of s by s0 is not thatserious (cf. Fig. 3). Table 11 shows, how the disas-trous situation for set 4B can be avoided by using e0= 0.1 in the runs.

3.3.5.4. A route to modify the data set

A glance at Table 9 would convince one that themost undesirable (negative) results follow only whenthe PS involves i = 1 (row 1). This is not unexpectedin view of the largest deviation from SSA shown inFig. 1 for the case of s0 = 10, the first member ofany set. Indeed, this causes a big difference betweens(t) and s0, which is responsible for the violation of(14). Once we are certain about this fact, we canremove this run to form a modified set for consider-able consistency. In Table 12, relevant results are

Table 11. Results of runs at t = 20 min with varying initial substrate concentrations (in mol) s0 at a fixed e0 = 0.1 mol. The rate V(t) and s(t) are displayed. Exact values are : Vm = 0.08, Km = 1.0. Comparing the outcomes with set

4B, the need for satisfaction of the SSA is obvious. Errors for others and SD values for PS scheme are shown

s0 s V Method Vm Km Error

10.0 8.47 0.072 EH 0.080 1.126 1.95 E-04

20.0 18.38 0.076 HW 0.080 1.192 2.43 E-04

30.0 28.36 0.077 LB 0.080 1.126 1.95 E-04

40.0 38.34 0.078 WLB 0.080 1.140 1.97 E-04

50.0 48.33 0.078 NLF 0.080 1.143 1.83 E-04

60.0 58.33 0.079 PS 0.080 1.303 0.001, 0.839

Table 10. Average values of reaction constants along withtheir SD values obtained from the PS scheme are displayed

for selected sets

Set Vm Vm Km Km

1A 0.800 6.81 E-04 1.023 3.53 E-02

4A 0.800 7.44 E-04 1.002 2.42 E-02

1B 0.802 1.02 E-03 1.111 3.77 E-02

4B 0.462 5.80 E-01 –17.138 36.172


1276

noted. One may appreciate that, although errors arestill sizeable, the values do not differ much in movingfrom one method to the other. Results in Tables 4, 6,7 and 10 for the parent set 4B lend credence to thisobservation.

3.3.5.5. Analysis of some other data

A thorough statistical comparison among severallinear transformations of (4) had earlier been made8

after introducing different kinds of errors in datathrough computer programs. These authors have alsonoted the appearance of negative reaction constantsfrom the LB scheme when specific types of errors areincorporated. The parent set is given in Table 13. Wehave found that the results obtained from variousmethods are very similar in character. This againconfirms our previous assertion (see Tables 1 and 2)that, if the right set of data is taken, results are virtu-ally independent of the chosen scheme.

Another exhaustive statistical analysis4 used data

Table 14. Enzyme kinetics data for the reaction ofnicotinamide mononucleotide21 at pH 4.95, and perfor-mances of various methods under scrutiny. Provisional

values are Vm = 0.679, Km = 0.569. Errors for othersand SD values for the PS scheme are shown

s V Method Vm Km Error

0.138 0.148 EH 0.626 0.490 1.27 E-02

0.220 0.171 HW 0.685 0.582 9.24 E-03

0.291 0.234 LB 0.585 0.441 1.65 E-02

0.560 0.324 WLB 0.680 0.571 9.27 E-03

0.766 0.390 NLF 0.696 0.601 8.70 E-03

1.460 0.493 PS 0.491 0.358 0.592, 0.754

of the substrate nicotinamide mononucleotide21 at pH4.95. Here, however, the situation is quite weak. Weshow in Table 14 the parent data set and results de-rived via the schemes discussed so far. One notes thatonly the HW and WLB methods come close to theprovisional values. Even the NLF scheme performs

Table 15. Results of adopting the PS scheme on the setdisplayed in Table 14. The upper and lower entries exhibit

values of Vm and Km, respectively

2 3 4 5 6

1 0.232 0.492 0.530 0.609 0.652

0.078 0.320 0.356 0.430 0.470

2 –1.653 0.770 0.806 0.740

–2.346 0.770 0.817 0.732

3 0.555 0.659 0.680

0.399 0.529 0.555

4 0.874 0.730

0.951 0.702

5 0.696

0.601

Table 12. Results of the reaction constants obtained byusing various methods, with the corresponding errors, forthe modified set 4B with the rate for s0 = 10 dropped. In

the PS scheme, individual SD is recorded

Method Vm Km Error

EH 0.875 6.090 1.08 E-02

HW 0.856 5.178 1.08 E-02

LB 0.881 6.405 1.10 E-02

WLB 0.837 3.885 1.13 E-02

NLF 0.884 6.855 1.01 E-02

PS 0.861 4.924 0.062, 2.793

Table 13. A set has been taken from an earlier work8 toshow the performances of various methods under scrutiny.

Exact values are : Vm = 30, Km = 15. Errors forothers and SD values for the PS scheme are shown. TheWLB scheme performs almost comparably with the NLF

one, and hence not displayed

s V Method Vm Km Error

2.5 4.29 EH 29.98 14.98 3.27 E-03

5.0 7.5 HW 30.00 14.99 2.58 E-03

10.0 12.0 LB 29.95 14.96 5.57 E-03

20.0 17.14 NLF 30.00 15.00 1.69 E-03

40.0 21.82 PS 29.97 14.97 0.063, 0.044

poorly. This calls for a thorough scrutiny, and welike to highlight the PS scheme at this juncture. Re-sults show considerable SD for either property, largerthan the individual averages. So, we proceed furtherto examine the data in PS Table 15. It immediatelyshows some problems with the second row due to theappearance of negative values. Thus, the chosen setitself has problems with i = 2. Therefore, we againmodify this set by deleting the second (s, V) entryfrom Table 14. Analysis of this fresh set brings to


1277

light sufficiently better estimates. These data are com-piled in Table 16. Note in particular the effects on Vm/|V

–––m| and Km/|K

–––m| in the PS scheme by

dropping the second entry (compare Table 14). Thescatter is now much less as well. The advantage ofintroducing the PS is now quite apparent.

vorable situations, modify the set towards betterment.Two cases in point are our set 4B and the set in Table14; both have thus been modified to gain advantage.

Acknowledgement

SD wishes to thank CSIR, India, for a researchassociateship.

Note added in proof :

The MM form given by eq. (4) with just two ‘ef-fective’ reaction constants appears in a variety of con-texts that are much more complex in character (see,e.g., I. Barel and F. L. H. Brown, J. Chem. Phys.,2017, 146, 014101, and references quoted therein).In all such cases, goodness of the estimated values ofeach such constant would pose similar problems ashighlighted here.

References

1. K. A. Johnson and R. S. Goody, Biochem., 2011, 50,8264.

2. S. Schnell and P. K. Maini, Comments Theor. Biol.,2003, 8, 169.

3. A. Cornish-Bowden, Persp. Sci., 2015, 4, 3.

4. G. N. Wilkinson, Biochem. J., 1961, 80, 324.

5. R. Eisenthal and A. Cornish-Bowden Biochem. J., 1974,139, 715.

6. A. Cornish-Bowden, J. Theor. Biol., 1991, 153, 437.

7. P. C. Engel and W. Ferdinand, Biochem. J., 1973,131, 97; L-H. Wang, M-S. Wang, X-A. Zeng, D-M.Gong and Y-B. Huang, Biochim. Biophys. Acta, 2017,1861, 3189.

8. J. E. Dowd and D. S. Riggs, J. Biol. Chem., 1965,240, 863.

9. G. L. Atkins and I. A. Nimmo, Biochem. J., 1975,149, 775.

10. N. C. Price, Biochem. Edu., 1985, 13, 81; R. J.Ritchie and T. Pravan, Biochem. Edu., 1996, 24,196; J. K. Harper and E. C. Heider, J. Chem.Educ., 2017, 94, 610.

11. C. S. Hanes, Biochem. J., 1932, 26, 1406.

12. H. Lineweaver, D. Burk and W. E. Deming, J.Am. Chem. Soc., 1934, 56, 225; H. Lineweaverand D. Burk, J. Am. Chem. Soc., 1934, 56, 658.

13. G. S. Eadie, J. Biol. Chem., 1942, 146, 85; B.H. J. Hofstee, Science, 1952, 116, 329; B. H. J.Hofstee, Nature, 1959, 184, 1296.

Table 16. Results obtained by removing the secondmember in the set displayed in Table 14. Errors forothers and SD values for the PS scheme are shown

Method Vm Km Error

EH 0.623 0.463 9.97 E-03

HW 0.664 0.530 7.56 E-03

LB 0.592 0.421 1.30 E-02

WLB 0.666 0.538 7.08 E-03

NLF 0.680 0.555 6.95 E-03

PS 0.648 0.531 0.105, 0.178

4. Concluding remarks

In summary, we wished to first highlight the quan-titative measure in (11) that showed the relativemerits of both the linear and nonlinear schemes. Agree-ment of the values of estimated reaction constantsamong the various linear methods reflects the qualityof a set. While the NLF procedure always fares overthe conventional LSM-based numerical exercises, forvery good-quality data sets (see, e.g., Tables 1, 2and 13), one need not invoke the NLF scheme. There-fore, the failure to obey the RSA is not the real reasonfor marked errors in estimated Km, as is sometimesbelieved18. We found also that the WLB is almost avirtual alternative to the NLF6. Secondly, an alto-gether different kind of approach, the PS scheme, hasbeen advocated here. It shows very clearly the good-ness of a set and also why one sometimes finds nega-tive reaction constants (see, e.g., Tables 9 and 15),not just in the LB scheme22, but elsewhere too. Anadded advantage of this scheme is that, here indi-vidual errors in terms of the SD values for both Kmand Vm are found (see, e.g., Tables 10–14 and 16).Finally, on the basis of the PS table introduced here,we demonstrated that it is not only possible to ensurethe quality of data set being handled but also, in fa-


1278

14. W. Liang, A. P. Fernandes, A. Holmgren, X. Liand L. Zhong, FEBS, 2016, 283, 446.

15. W. Stroberg and S. Schnell, Biophys. Chem.,2016, 219, 17.

16. M. L. R. González, S. Cornell-Kennon, E. Schaeferand P. Kuzmi

c

, Anal. Biochem., 2017, 518, 16.

17. W. H. Press, S. A. Teukolsky, W. T. Vetterlingand B. P. Flannery, "Numerical Recipes : The Artof Scientific Computing", CUP, 3rd ed., 2007.

18. S. M. Hanson and S. Schnell, J. Phys. Chem.(A), 2008, 112, 8654; S. Schnell, FEBS, 2014,

281, 464.

19. S. Dhatt and K. Bhattacharyya, J. Math. Chem.,2013, 51, 1467; K. Bhattacharyya and S. Dhatt,MATCH Commun. Math. Comput. Chem., 2013,70, 759.

20. K. Banerjee and K. Bhattacharyya, Chem. Phys.,2014, 438, 1; J. Chem. Phys., 2015, 43, 235102.

21. M. R. Atkinson, J. F. Jackson and R. K. Morton,Biochem. J., 1961, 80, 318.

22. S. Dhatt and K. Bhattacharyya, ARXIV, 1710.08131.

1279


Theory of chronocoulometry of a pseudo first-order catalytic process at arough electrode†

Md. Merajul Islam and Rama Kant*

Department of Chemistry, University of Delhi, Delhi-110 007, India

E-mail : [email protected]

Manuscript received 25 October 2017, accepted 01 November 2017

Abstract : A theoretical model for single potential step charge-transients has been developed for the pseudofirst-order catalytic coupled (EC’) reaction mechanism at rough electrode-electrolyte interface. A second-or-der perturbation solution in surface prole for an arbitrary rough surface is obtained. The expression for thecharge-transient at the rough electrode-electrolyte interface for band limited self-affine fractal roughness isobtained. An extension of the generalized Danckwerts’ expression in terms of the charge is presented. Ourresults show an enhancement in the charge-transients due to the roughness present at the electrode surface.The morphological characteristics show strong dependency in the anomalous intermediate regime obtained fromthe power-spectrum of roughness while the long-time regime is the time dependent kinetic controlled. Thetraditional plot of charge-transient does not provide the information related to the morphology and its pos-sible kinetic influences.

Keywords : Power spectrum of roughness, fractal electrode, Chronocoulometry, catalytic process, chargetransfer.

Introduction

The surface morphology or roughness of an elec-trode affects the spectroelectrochemical responses. Theroughness effect is inevitable especially, if the elec-trodes are made of solid materials. The rough elec-trodes have potential to be used in fundamental andapplied fields1–6. This is possible because rough sur-faces show enhanced surface reactions due to the in-crease in the effective area. Rough surface is alsocapable to separate electrons and holes from combi-nation because of generation of channel-like structurein the vicinity of the electrode. Hence, rough elec-trode surface structure is favorable to enhance theperformance of light emitting diodes and solar celldevices1.

In our previous paper, we have generalized theAnson equation that describes the potentiostatic re-sponse of a reversible charge transfer reaction for

†Acharya J. C. Ghosh Memorial Lecture (2016).

rough electrode. It explains the observed deviation inthe Anson plot from linearity due to the roughnesspresent at the electrode surface7. An important reac-tion mechanism is where the electrochemical step iscoupled with bulk kinetics which is the catalyticcoupled reaction mechanism used in many fundamen-tal as well as in applied fields8–10,12–21. Saveant etal.9 discussed the potentiostatic and cyclicvoltammetric responses of this mechanism throughvarious zones attributing to the extent or reversibilityof the E step as well as the dependence of the re-sponse on the concentration of the electrochemicallyinactive species. Here, we focus on the influence ofubiquitous electrode roughness on the response of suchpseudo first-order catalytic coupled reaction mecha-nism under the potentiostatic perturbation.

The pseudo first-order catalytic-coupled (EC’) re-action mechanism scheme is one of the most impor-


1280

tant class of catalytic coupled reaction which is repre-sented as8–10 :

kO + ne– R + z — O + z (1)

where k is homogeneous rate constant, z and z areelectrochemically inactive at the electrode potential atwhich reaction proceeds. The first step is the rever-sible redox reaction at the electrode surface and thesecond step is the chemically coupled reaction withthe first step.

Chronocoulometry is an effective technique forstudying homogeneous chemical reactions that arecoupled to the heterogeneous electron transfer reac-tion. It is well known that, the chronocoulometry tech-nique is advantageous over chronoamperometry22. Thestudies such as22 : mechanism, rate constant, fara-daic and non-faradaic components separation, adsorp-tion studies23–28 can be easily done by usingchronocoulometric technique. The chronocoulometricstudy for EC’ reaction mechanism was first done byJ. H. Christie in 1967 for the condition of doublepotential-step29. He proposed an expression for thecharge-transient for the pseudo first-order EC’ reac-

tion mechanism for single-step also at smooth elec-trode, which can be written as29 :

QC(t) = QP(t) ktkt e kt k

kt

2 (1 2 )erf t

4

,

(2)

where QC(t) is the total catalytic charge-transient, t isthe time, and “erf” is the error function30. QP(t) isthe potentiostatic charge-transient for planar surfacewhich is given by the Anson and is known as Ansonequation22 :

QP(t) =

snF A C Dt02

, (3)

where D is the diffusion co-efficient of the species, nis the number of electron(s) involved in the redoxreaction, F is Faraday constant, A0 is the electrodegeometric area and Cs is the bulk concentration of thespecies. This pioneer work of Christie29 has given abasis to study many aspects of catalytic coupled reac-tion mechanism by using chronocoulometry technique.

The recent contribution in theoretical and experi-mental studies in EC’ mechanism such as to deter-mine the kinetic parameters of EC’ mechanism insquare wave voltammetry (SWV)31,32 and its surfacecatalytic mechanism33, analysis of two-step redoxmechanisms in protein-film by cyclic staircasevoltammetry (SCV)34 and comparative study of mecha-nism by SWV and SCV35 are significant and furtherexplored the understanding of EC’ mechanism.

The recent interest in realistic fractal geometrywhich use the power spectrum of roughness approach,has opened a new horizon to understand the influenceof surface disorder on spectroelectrochemical re-sponses. This includes the localization phenomenonof impedance on a Weierstrass fractal surface36,anomalous electric double-layer dynamics in ionic liq-uids37, anomalous response in cyclic staircasevoltammetry38, generalization of Randles-Ershler ad-mittance for an arbitrary topography electrode39,double potential step chronoamperometry response for

Fig. 1. Schematic diagram of EC’ reaction mechanism tak-ing place at a rough electrode-electrolyte interface.The various morphological and phenomenologicalcharacteristics are shown. The various characteristicsare as : DH is fractal dimension, l is width of theinterface, l and L are lower and upper cut-off length

scales, respectively. Dt is diffusion-width,

Dk

is the reaction layer thickness and k is the homoge-neous rate constant. Here z is the catalyst which isinactive at the potential at which reaction is occur-ring.

Islam et al. : Theory of chronocoulometry of a pseudo first-order catalytic process at a rough electrode

1281

finite fractal and nonfractal electrode roughness withand without ohmic effect40, dynamics of the electricdouble layer at a heterogeneous and rough electrode41,absorbance-transient for pseudo first-order catalyticcoupled mechanism42, generalization of Gouy-Chapman-Stern model for a morphologically com-plex electrode43, Debye-Falkenhagen Dynamics44 andothers45–50.

Chronocoulometry and chronoabsorptometry showlinear relationship to each other and have mathemati-cal isomorphism22,51. Absorbance is the characteris-tic of the integrated concentration of a given specieswhereas, charge represents the integrated flux of thespecies at the electrode surface. Thus, the influenceof the kinetic process occurring in the diffusion layeron absorbance-time behavior is quite different fromthat of the charge-time52. Hence, it is of great impor-tance to analyze the problem from chronocoulometricpoint of view also, along with the chronoabsorptometrymethod42. Due to the mathematical isomorphism be-tween these two techniques, the detailed calculationsare not provided here and the formalism of the prob-lem can be found in Ref. 35.

The present paper is organized as follows : in thefirst section, the charge transient is obtained for anEC’ reaction mechanism at an arbitrary and randomroughness profile, in the second section, thechronocoulometric expression for an EC’ reactionmechanism at finite self-affine fractal rough electrodeis obtained followed by the results and discussion.Finally, on the basis of results and discussion, con-clusions are drawn.

EC’ reaction mechanism at random rough elec-trodes

In this section, the chronocoulometric model foran EC’ reaction mechanism at random rough elec-trode is obtained. The random morphology of therough electrode surface or interface can be satisfacto-rily described as centered Gaussian fields which isstatistically homogeneous surface (r

||), where r

|| is

a two dimensional space vector. The surface struc-

ture factor is used to understand the statistics for suchGaussian fields or surfaces. The surface structure factoris defined as the Fourier transform of the surface

correlation function53,54 and is given as : |(K||)|2,

where K|| is the wave-vector in two dimension and .

represents the ensemble average. It is also known asthe power-spectrum of roughness and is quantified asthe ensemble average over all the possible configura-tions. The equation which governs the ensemble av-eraged Fourier transformed surface profile and itscorrelation is given as :

(K||)= 0

(K||)(K

||)

= (2)2(K||+K

||)|(K

||)|2, (4)

where (K||)is the Dirac delta function in K

||.

The realistic finite fractal concept is used to char-acterize the roughness of the electrode surface. Thefinite fractals have two cut-off length scales, i.e. lowerand upper cut-off length scales and the range of thesetwo cut-off lengths show self-affine fractal behavior.The upper cut-off length scale is nothing but the cor-relation length. The fractals without these band limi-ted cut-off length scales are known as idealized fractalsand suffer from mathematical difficulties such as non-differentiability and non-homogeneity. But band lim-ited finite fractal does not suffer from these math-ematical difficulties and uses power-spectrum of rough-ness with band limited power law.

To get the physical insights, the moments of power-spectrum are used which is related to the mean squarederivatives of surface profile. The 2j-th moments ofpower-spectrum is given as53 :

m2j =

21

(2 )

d2K||K||2j|(K

||)|2=

(j||(r||))2, (5)

JICS-4


1282

where j||(/x, /y) is the two dimensional gra-

dient operator. The moments of power-spectrum arerelated to the various morphological characteristicsof a rough surface such as : the mean square (MS)width (m0), the MS gradient (m2) and the MS curva-ture (m4) etc. SEM/AFM studies are used to get thesurface roughness correlation function, power-spec-trum and its various moments48.

Theory of chronocoulometry for an EC’ reactionmechanism at rough electrode-electrolyte interface

The concentration profile must be known aroundthe arbitrary electrode profile to calculate the charge-transient for rough electrode-electrolyte interface. Theconcentration profile for an EC’ reaction satisfies thediffusion equation as :

— t

Ci (r, t) = Di

2 Ci (r, t) – kCi (r

, t), (6)

where k is the pseudo first-order homogeneous rateconstant, i O,R and represents the oxidized or re-duced species, respectively, Ci() is the differencebetween surface and bulk concentration, Di is diffu-sion-coefficient (for simplicity we assume DO = DR= D) and r

is the three dimensional vector, (r

(x,

y, z). In our calculation, we assume C (r, t) = CO

(r, t) = –CR(r

, t). To complete the transformation

of the boundary value problem to a simpler problem,we need to transform the initial and the bulk bound-ary problems. The transformed initial and bulk bound-ary problems are given as : C*(r

, t = 0) = 0 and

C*(r||, z , t) = 0. The surface boundary condi-

tion for both the processes (cathodic or anodic) aresame. So, the transformed boundary condition is writ-ten as :

ektC0O

C*(r, t = –ektCs = – —————, (7)

1+

where = exp(–nf (Ei – E0)), Ei and E0 are the

initial and the formal potentials, respectively, f = F= RT, F is Faraday constant, R is the gas constant, Tis the absolute temperature, C0

O is the bulk concen-tration of the oxidized species and n is the number ofelectron(s) involved in the first step redox reaction of

eq. (1).

The total charge at rough electrode-electrolyte in-terface can be related to the concentration profile bythe following expression :

Q(t) = nFD

S

dt

0

t

0

dS0n CO(z = (r||), t). (8)

The concentration in the above equation (eq. (8)) isobtained as the perturbation solution upto second or-der in the surface profile. The concentration can bewritten in Fourier and Laplace transform domains fora fluctuating two-dimensional (2-D) rough surface as:

CO(K||, z, p) =

sCp

exp(–q||z)

{(2)2(K||)+qk

(K||)+

kq 2

22(2 )

dK||(K

|| – K

||)(K

||)

[2||,||– 1]}+O(

3), (9)

where (K

||) is the two dimensional Fourier trans-

form of the rough surface profile given as :

(K

||) = dr

||e–jK

||r

|| (r||); j =

1

q = (p/D)1/2

qk = [q2 + k/D]1/2

q|| = qk

|| = [q2

k+K

||2]1/2

q||,|| = qk

||,||= [q2

k+(K

||–K

||)]1/2, (10)

where is the homogeneous rate constant, K|| =|K

||| and (K

||) is the two-dimensional Dirac delta

function in vector K

||.

The expression for the total charge at rough elec-trode-electrolyte interface in the Laplace transformdomain upto second order perturbation is given as :

Q(p) =

S 0

d2r||{[Q0 (2)

2(K||)+Q

1(K||)

Q2(K||)(K||

– K||)]; K

|| r

||}, (11)


1283

where p is the Laplace transform variable and S0

d2r

||(.)

is the integral over the xy plane, S0 is the projection

of the surface on the reference surface z = 0. Q0,

Q1 and Q

2 are the operators for the smooth surface,

the first-order and the second-order terms in surfaceprofile in Laplace transformed domain, respectivelyand are defined as :

Q0

s knFDC q

p 2;

Q1 Q

0 (q

||–qk)

Q2

snFDC

p2 2(2 )

d2K||[q

||q||,||

– 1—2 (q2

k+qkq||) – (K

||

– K

||)2

–K

||.(K||

– K

||)]. (12)

Eq. (11) is used to predict the chronocoulometric re-sponse after taking the inverse Laplace transform andthe inverse Fourier transform for a known surfaceprofile. This expression provides insight into the re-sponse of a nanometer to micrometer scale structuredelectrode that has an arbitrary geometry profile orrough electrode with known profile.

The total charge-transient for a random surfacemorphology is obtained by taking the ensemble aver-aged charge-transient over all possible surface con-figurations. The ensemble averaged charge-transienthas the information of surface morphology throughthe surface structure factor or power-spectrum ofroughness. The Laplace transformed chronocoulomet-ric response for the two dimensional statistically ho-mogeneous random rough electrode is obtained byensemble-averaging over all possible surface configu-rations on eq. (11) as :

s k knFA DC q qQ p

p0

2 2( ) 1

(2 )

d2K||[q||– qk]|

(K||)|2}. (13)

Above equation (eq. (13)) can be written in terms of

the Laplace transformed generalized Cottrell cur-rent53,55 (Igc(p)) for the rough electrode as :

gck

Q p I p kp p1

( ) 1 ( )

(14)

sgc

nFA DC qI p

q02 2

( ) 1(2 )

d2K||[q

||– q]|

(K||)|2}. (15)

The inverse Laplace transform of eq. (14) for pseudofirst-order catalytic coupled reaction mechanism canbe rearranged as :

t tkt

gcQ t dt k dt e I t0 0

( ) 1 ( )

t tkt

gad

dt k dt e Q tdt

0 0

1 ( )

tkt

gak dt e Q t

2

0

1 ( )

, (16)

where Igc (t) = d—dt Qga (t). Here Qga (t) is thetotal ensemble averaged charge transient when nochemical reaction takes place in the second step ofthe pseudo-first-order catalytic coupled reaction (seeeq. (1)), known as generalized Anson equation7.

Equation (16) is the generalization of theDanckwerts’ expression56 in terms of the charge atrough electrode which is the first of its kind. It re-lates the charge-transients with coupled homogeneousreaction to the generalized Anson equation7 of roughelectrode. The eq. (16) can be rearranged as :

tkt kt

ga t gaQ t e Q k dt e Q t( )0

( ) 2 ( )

t tkt

gak dt dt e Q t2

0 0

( ) (17)


1284

Here Q(t) is the total ensemble averaged charge tran-sient for the pseudo first-order catalytic coupled (EC’)reaction mechanism. The second term in the curlybracket in eq. (16) introduces the kinetic effects ofthe EC’ reaction. The eq. (17) is the extension of theDanckwerts’ expression56 in terms of the charge atrough electrode-electrolyte interface. The simplestDanckwerts’ expression requires a linear time-inde-pendent operator while eq. (16) is valid for square ofthe operator. In the above eq. (16), {.} represents anoperator. From eq. (17), it is clear that, it accountsthe kinetics of the simple reversible redox reaction(pure Nernstian case) as well (first term). This exten-sion of Danckwert’s expression could prove useful inthe study of an EC’ reaction mechanism at roughelectrode-electrolyte interface.

The quantity Qga (t) is the total ensemble aver-aged charge transient for purely diffusion-controlledreversible redox reaction and is given as7 :

sga

nFA C DtQ t 02

( )

(1+R|(K||)|2), (18)

where operator R brings in dynamic effect of surface

roughness on the charge through its action on powerspectrum of roughness of electrode, given by

RDt2

1ˆ2(2 )

d2K

||(1 – exp(–K2||Dt) –

K2||Dt erf ( K2

||Dt)). (19)

Chronocoulogram at finite self-affine fractal elec-trodes

The development of fractal geometry was asignicant breakthrough in the description of irregu-larity present in the nature57. The topography of arough physical surface is often random but exhibitsstatistical self-resemblance over a range of scale be-cause the random processes that formed the surfaceshow the scale invariance57,58. Such kind of surfaceor topography is described by using fractal geometry.

The fractal surfaces can be self-similar or self-affinein nature57. The self-affine fractal surfaces are thosesurfaces in which the surface functions remain statis-tically invariant under scale transformation : (x, y, z) (x, y, Hz) where H is the roughness Hurst’sexponent. Hurst’s exponent describes the persistenceand anti-persistence behavior in the height fluctuationand is related to fractal dimension (DH) as : DH =3–H for isotropic self-affine surface.

Surfaces with limited scales of invariance prop-erty are usually described by the stationary Gausianrandom processes with a power-law spectrum over alimited range of wave-numbers59. The roughness spec-trum of an approximately self-affine isotropic (ran-dom) fractal surface is described in terms of a limitedband of wave-numbers (K

||) which follows a power-

law function as60,61 :

|(K||)|2 = |K

|||2DH–7,

1/L |K||| 1/l, (20)

The above expression (eq. (20)) of power-spectrumwhich is used to describe the roughness and consistsof four physical characteristics. The four physicalcharacteristics are fractal dimension (DH) which isthe global property of roughness and describes scaleinvariance property of the roughness. The lower cut-off length scale (l) which is the length above whichsurface shows fractal behavior. Similarly, the uppercut-off length scale (L) is the length below whichsurface shows fractal behavior. The ratio, L/l (=),is known as range of fractality and it must be largeenough to be a fractal surface. The fourth physicalcharacteristics is the strength of fractality () whichis related to the topothesy of fractal60–62 and 0means no roughness. It is related to the width of theinterface through the topothesy length (l ) as : =l

2DH–3. These realistic fractal characteristics can beobtained from AFM and SEM measurements on thesurface and its power-spectrum48. The idealized fractaldescription of anomalous power-law behavior in thecharge-transient does not provide proper characte-rization since their time exponent is simply assumed


1285

to be function of DH only.

To get the physical insights into the rough surfacemorphology, the moments of power-spectrum are used.The moments of power-spectrum provide the physi-cal quantities such as : mean square (MS) width ofinterface (m0), MS gradient (m2), MS curvature (m4)and so on. The general expression for the 2j-th mo-ment of the power-spectrum for a band-limited fractalcan be written as63 :

j j

jl

mj j

2( ) 2( )

2L

4

, (21)

where = DH – 5/2.

The quantitative measure of the roughness of thesurface is obtained by knowing the roughness factor,R*. This gives the extent of deviation of electrochemi-cal responses (current, charge, absorbance, etc.) fromthe smooth surface. The roughness factor is relatedto MS gradient (m2) obtained from moment of power-spectrum as discused above. The rough surfaces withsmall m2 has small also, which are referred as lowroughness surfaces. The rough surfaces with largem2 means large and has large roughness surfaces.The expression for the roughness factor for low rough-ness surfaces is given by R* 1 + m2 similarly, theroughness factor for the large roughness surfaces is

R* =

m 2 /2

.

The total interfacial ensemble averaged charge-transient for an EC’ process is obtained by solvingthe eq. (17) which is the general form of the charge-transient for rough electrode surfaces. Thus, the firstterm of the eq. (17) is the ensemble averaged charge-transient for the purely diffusion-limited reaction atan approximately self-affine fractal electrode surface,takes the form as7 :

Qga(t) = QP(t) (1 – RF1(t)+RF2(t, l) – RF2(t, L))

HDm

R tDt Dt0

F1 3/2( )

2 8 ( )

HD Dt Dt L2 2( 5/2, / , / ), H

F D tH

R tD

2 2 3 *( , )

8 ( 2)

HDHt t D t*2erf( *) ( 3/2, *) , (22)

where is the cut-off length (either l or L) and di-mensionless time t* = D— t—

2 , (, x0, x1) = (,

x0)– (, x1) = (, x1) – (, x0), (, xi) and (, xi) are the incomplete Gamma functions30. Theeq. (22) represents the exact form of the charge7 be-cause the first term of eq. (17) can be integrated withrespect to time. The second and the third terms of eq.(17) has difficulty to integrate with respect to thetime with the exact form. This is due to the fact thatthese terms bring the kinetic contributions. Thus, tosolve these two terms, an approximated form of thecharge is used to overcome this difficulty for longertime. The longer time is the kinetic control and thisapproximation works good for the longer time re-gime. So we use this approximation to solve the sec-ond and the third terms of eq. (17). The expressionfor the approximate form of the charge used to solvethese two terms is given as7 :

sga

nFA C DtQ t 02

( )

Dt Dt

2

1( )

18 8 (2 1)( )

. (23)

The above eq. (23) is valid for t >> ti, where ti is theinner fractal crossover time. It is the time after whichthe anomalous intermediate time regime is observedin the charge-transient.

The short-time contribution to the total charge-transient, i.e. Q(ti) is very small for all t >> ti.Hence, it is ignored in the above calculation.

After using eq. (23), the second term (say, Q1(t))in the eq. (17) takes the form as :

tkQ t k dt e Q tt

1 ga0

( ) 2 ( )


1286

ktsk nFA C Dt kt t ekk

0

3

2 2 erf

2

kt

D k

2 erf

8

k kt

D

1/2

1( ) (1/2 , )

8 (2 1)

(24)

and the third term (say, Q2(t)) in the eq. (17) for self-affine fractal electrode becomes as :

t tkt

gaQ t k dt dt e Q t22

0 0

( ) ( )

sk nFA C D ktt

kk

20

3

2 erf 322

kt ktt e te

D k kk

2

23

82

kt kt

k D

( )1erf

2 (2 1)

t kt ktk1

(1/2 , ) (3/2 , ) . (25)

For the limiting case that is, at k = 0 the second andthe third terms in the eq. (17) becomes zero and eq.(17) is reduced to purely generalized Anson equa-tion7. Under this condition, the eq. (17) becomes sameas eq. (18) in Ref. 7. Thus we can say that, eq. (17)is also valid for the simple reversible redox systems(purely Nernstian case) under the limiting case. If theroughness is zero this equation becomes Anson equa-tion for planar case (eq. (3)). The differentiation ofeq. (17) with respect to the time will give current.Hence, the differential form of eq. (17) is thechronoamperometric (current-time relationship) re-sponse for an EC’ reaction mechanism at rough elec-

trode-electrolyte interface. Thus, eq. (17) providesmore physical insights into the EC’ reaction mecha-nism with proper characterization of rough electrode-electrolyte interface in terms of fractal morphologicalcharacteristics.

Results and discussion

Fig. 1 depicts the various morphological and phe-nomenological characteristics which control thechronocoulometric responses of an EC’ reactionmechanism at fractally rough electrode-electrolyteinterface. As discussed earlier, the four morphologi-cal characteristics are fractal dimension (DH), strengthof fractality (), lower (l) and upper (L) cut-off lengthscales. Out of these, DH, and l contribute stronglyto the chronocoulometric responses while L showsweak dependency for large fractal range, (= L/lratio).

In Fig. 2, the effect of fractal dimension on charge-transient for an EC’ reaction mechanism taking placeat a rough electrode-electrolyte interface is analyzed.Since, fractal dimension is the global property of therough surface thus, the increase in the fractal dimen-sion means increase in the roughness or fractal natureof the surface. The double logarithmic plot of chargeand time clearly demonstrates that, as the fractal di-mension increases, the charge-transients correspond-ingly enhance. The fractal rough surface has moreeffective area and due to the fractal nature of thesurface the reaction taking place at the surface be-comes more localized. As the roughness increases,the localized reaction becomes more and more promi-nent which increases the generation of the redox spe-cies even more at the electrode surface. Thus, charge-transient enhances signicantly, if the fractal nature ofthe rough surface increases. The first plot from bot-tom is generated by using eq. (2) which is the planarcatalytic response.

The two time regimes can be seen easily in thedouble logarithmic plot of charge and time, i.e. anoma-lous intermediate and long-time regimes. The inter-


1287

mediate-time regime shows anomalous power law be-havior and is dependent on fractal morphologicalcharacteristics. There is very weak dependency ofthe kinetics in this region and gives the informationabout the surface morphology. Thus, anomalous in-termediate-time regime provides the information aboutthe fractal nature of the rough electrode surface. Thelong-time regime is the time dependent kinetic con-trolled.

Fig. 3 depicts the effect of strength of fractality oncharge-transient. The strength of fractality is relatedto the width of the interface through topothesy length,l. The larger width of the interface leads to the morerough surface. The increase in the strength of fractality,enhances the charge-transients due to increase in thewidth of the interface hence, roughness of the sur-face. The double logarithmic plot of charge and timeclearly demonstrates the two time regimes in it. Thatis, the anomalous intermediate and long time regimes.

The intermediate-time regime shows weak dependencyof strength of fractality with time. Finally, it mergeswith planar response in the long-time regime. Thefirst plot from bottom is of planar catalytic response(eq. (2)).

In Fig. 4, the effect of lower cut-off length scaleon charge-transient is analyzed. The decrease in thelower cut-off length scale increases the self-affinefractal nature of the rough-surface. Due to decreasein the lower cut-off length scale, roughness of thesurface increases which enhances the charge-transient.Here it is important to understand the relationshipbetween the lower cut-off length scale and the inner-transition time (ti). The inner-transition time is thetime after which the anomalous intermediate-time re-gime is started. The short-time domain is stronglydependent on lower cut-off length scale while long-time domain is controlled by width of the interface(h). The width of the interface is a strong function ofupper cut-off length scale (L) and has no impact of

Fig. 2. Effect of fractal dimension (DH) on charge-transientsfor an EC’ reaction mechanism taking place at a roughelectrode-electrolyte interface. The values of DH aretaken as : 2.10, 2.15, 2.20, 2.25. The other param-eters are taken as : l = 30 nm, L = 5 m, = 10–5

(a.u.), k = 1 s–1, Cs = 5×10–6, mol cm–3, D =5×10–6, cm2 s–1, A0 = 0.15 cm2, n = 1.

Fig. 3. Effect of strength of fractality () on charge-tran-sients. The values of (a.u.) are taken as : 0, 2×0,4×0, 8×0 where 0 = 10–6. The other param-eters are taken as : DH = 2.2, l = 30 nm; L = 5 m;k = 1 s–1, Cs = 5×10–6 mol cm–3, D = 5×10–6

cm2 s–1, A0 = 0.15 cm2, n = 1.


1288

lower cut-off length scale in the long-time domain.Therefore, the inner transition-time decreases withdecrease in the lower cut-off length scale while itincreases as the lower cut-off length increases. Toobserve the roughness effect, the diffusion layer widthmust be comparable to the surface irregularities. Weare interested to observe the roughness in the timeregion of 10–3 s to 10 s. The value of the inner-transition time is often much less than 10–3 s, below10–3 s the double-layer region becomes prominentand causes an additional effect on the charge-tran-sient. The magnitude of double-layer charging withohmic drop can be corrected in the experimental dataas reported in the recent publication64.

Fig. 5 shows the effect of kinetics (homogeneousrate constant) of the EC’ reaction mechanism oncharge-transient at rough electrode-electrolyte inter-face. From Fig. 5 it is clear that, as the value ofhomogeneous rate constant increases, first it showsno (especially at the lower values) or weak depen-

dency on charge-transient. It becomes plateau andafter that enhances rapidly in the long-time regime.This plateau becomes more and more prominent asthe value of the rate constant increases. At signifi-cantly larger value of rate constant, initially the charge-transient decreases slightly and then increases rapidlyin the long time regime. The time at which this rapidincrease in charge-transient is observed, keep on de-creasing as the rate constant increases and is roughlycorresponds to 1/k value. In the early time, there isno effect of the rate constant. Anomalous intermedi-ate-time regime shows very weak dependency on rateconstant. But the long-time domain shows strong de-pendency of rate constant on charge-transients. Thisclearly demonstrates that the long time regime is timedependent kinetic controlled regime. The rapid in-crease of charge-transient in the long time regime isdue to the catalytically coupled reaction. The inset isthe traditional plot (Q(t) vs t1/2) of the corresponding

Fig. 4. Effect of lower cut-off length scale (l) on charge-transients while other parameters are kept constant.The values of l (nm) are taken as : 15, 30, 60, 200.The other parameters are as : DH = 2.2, =10–5

(a.u.), L = 5 m, k = 1 s–1, Cs = 5×10–6 mol cm–3,D = 5×10–6 cm2 s–1, A0 = 0.15 cm2, n = 1.

Fig. 5. Effect of homogeneous rate constant (k) on charge-transients while keeping other morphological charac-teristics are constant. The values of k (s–1) are takenas : 5, 30, 60, 100, 250. The other parameters aretaken as : DH = 2.2, = 10–5 (a.u.), l = 30 nm, L= 5 m, Cs = 5×10–6 mol cm–3, D = 5×10–6 cm2

s–1, A0 = 0.15 cm2, n = 1. The inset plot representsthe traditional Q(t) vs t1/2 plot.


1289

curves. There are no characteristic features observedin this presentation as the double logarithmic curvesprovide. Thus, we can observe the roughness effectclearly in the double logarithmic scale while it is notseen in the traditional plots.

Conclusions

In summary, a theoretical model for single poten-tial step charge-transients for pseudo first-order cata-lytic coupled reaction (EC’) mechanism at the roughelectrode-electrolyte interface is developed. The theorypresented here uses approach based on the powerspectrum of roughness and addresses the detail analy-sis of chronocoulometric response for the finite fractalelectrode roughness. The important conclusions areas follows : (i) the charge-transient response enhanceswith the increase in the surface roughness, (ii) thefractal nature of the rough surfaces shows enhancedcharge-transient with increase in fractal dimension,topothesy length and decrease in the minute scale offractality, (iii) anomalous intermediate-time regimeshows strong dependency on these fractal characte-ristics, (iv) the long-time regime is bulk kinetic con-trolled, (v) the time window available for Nernstianor diffusion control regime in EC’ reaction is D/k2 <t < k–1. Hence, large value of D/k2 will curtail theNernstian regime in EC’ reaction, (vi) the influenceof roughness will be maximum for the smaller grainsize (l) while the width of roughness (l) is greater

than

D/k

. Finally, the roughness enhances sensi-

tivity to EC’ reaction.

References

1. S. V. Novikov, Macromol. Symp., 2004, 212, 191.

2. F. A. Castro, H. Benmansour, C. F. O. Graeff, F.Nuesch, E. Tutis and R. Hany, Chem. Mater., 2006,18, 5504.

3. L. Fojt and S. Hason, J. Electroanal. Chem., 2006,586, 136.

4. K. A. Ritter and J. W. Lyding, Nat. Mater., 2009, 8,235.

5. S. M. Mitrovski, L. C. C. Elliot and R. G. Nuzzo,Langmuir, 2004, 20, 6974.

6. R. Kant and M. M. Islam, J. Phys. Chem. (C), 2010,

114, 19357.

7. M. M. Islam and R. Kant, Electrochim. Acta, 2010,56, 4467.

8. P. Delahay, "New Instrumental Methods in Electroche-mistry", Interscience, New York, 1954.

9. J-M. Savéant, "Elements of Molecular and BiomolecularElectrochemistry", Wiley-Interscience, New Jersey,2006.

10. J. D. E. McIntyre, J. Phys. Chem., 1967, 71,1196.

11. M. M. Islam, "Theoretical Models for Spectro-electrochemical Absorbance and Chronocoulo-meteric Transients under Potentiostatic Conditionon Rough Electrodes", University of Delhi, Delhi,India, 2012.

12. D. M. H. J. Kern, J. Am. Chem. Soc., 1953, 76,1011.

13. A. C. Testa and W. H. Reinmuth, Anal. Chem.,1961, 33, 1320.

14. K. B. Wilberg and T. P. Lewis, J. Am. Chem.Soc., 1970, 392, 7154.

15. R. S. Nicolson and I. Shain, Anal. Chem., 1964,36, 706.

16. H. Y. Cheng and R. L. McCreery, Anal. Chem.,1978, 50, 645.

17. M. Mohammad, Anal. Chem., 1977, 49, 60.

18. E. Streckhan, Top. Curr. Chem., 1987, 142, 1.

19. S. Olivero, J. C. Ciinet and E. Duach, Tetrahe-dron Lett., 1995, 36, 4429.

20. A. P. Doherty and K. Scott, J. Electroanal.Chem., 1998, 442, 35.

21. C. F. Hogan, A. M. Bond, J. C. Myland and K.B. Oldham, Anal. Chem., 2004, 76, 2256.

22. A. J. Bard and L. R. Faulkner, "ElectrochemicalMethods : Fundamentals and Application", Willey,New York, 1980.

23. D. H. Evans and M. J. Kelly, Anal. Chem.,1982, 54, 1727.

24. R. A. Osteryoung and F. C. Anson, Anal. Chem.,1964, 36, 975.

25. F. C. Anson, Anal. Chem., 1964, 36, 932.

26. B. Case and F. C. Anson, J. Phys. Chem., 1967,71, 402.

27. J. H. Christie, G. Lauer and R. A. Osteryoung, J.Electroanal. Chem., 1964, 7, 60.

28. F. C. Anson, Anal. Chem. 1966, 38, 54.

29. J. H. Christie, J. Electroanal. Chem., 1967, 13,79.

30. M.Abramowitz and A. Stegan, (Eds.), "Handbookof Mathematical Functions", Dover PublicationsInc., New York, 1972.

JICS-5


1290

31. R. Gulaboski and V. Mireceski, Electrochim.Acta, 2015, 167, 219.

32. V. Mireceski, R. Gulaboski, M. Lovric, I. Bogeski,R. Kappl and M. Hoth, Electroanalysis, 2013, 25,2411.

33. V. Mireceski and R. Gulaboski, Electroanalysis,2001, 13, 1326.

34. R. Gulaboski, P. Kokoskarova and S. Mitrev,Electrochim. Acta, 2012, 69, 86.

35. V. Mireceski, R. Gulaboski and J. Solid, StateElectrochem., 2003, 7, 157.

36. R. Kant, S. Dhillon and R. Kumar, J. Phys.Chem. (B), 2015, DOI: 10.1021/jp512297f.

37. M. B. Singh and R. Kant, J. Phys. Chem. (C),2014, 118, 8766.

38. Parveen and R. Kant, J. Phys. Chem. (C), 2014,118, 26599.

39. R. Kant and M. B. Singh, Electrochim. Acta,2015, 163, 310.

40. S. Dhillon and R. Kant, Electrochim. Acta, 2014,129, 245.

41. M. B. Singh and R. Kant, J. Phys. Chem. (C),2014, 118, 5122.

42. R. Kant and M. M. Islam, J. Electroanal. Chem.,2014, 713, 82.

43. R. Kant and M. B. Singh, Phys. Rev. (E), 2013,88, 052303.

44. M. B. Singh and R. Kant, J. Electroanal. Chem.,2013, 704, 197.

45. S. Dhillon and R. Kant, Electroanalysis, 2014,26, 2350.

46. Parveen and R. Kant, Electrochim. Acta, 2013,111, 223.

47. M. B. Singh and R. Kant, Proc. R. Soc. (A),2013, 469, 20130163, doi:10.1098/rspa.2013.0163.

48. S. Dhillon and R. Kant, Appl. Surf. Sci., 2013,282, 105.

49. S. Srivastav, S. Dhillon, R. Kumar and R. Kant,J. Phys. Chem. (C), 2013, 117, 8694.

50. R. Kant, M. Sarathbabu and S. Srivastav,Electrochim. Acta, 2013, 95, 237.

51. R. Kant and M. M. Islam, J. Phys. Chem. (C),2010, 114, 19357.

52. J. Niu and S. Dong, Rev. Anal. Chem., 1996, 15,1.

53. R. Kant and S. K. Rangarajan, J. Electroanal.Chem., 1994, 368, 1.

54. S. K. Jha, A. Sangal and R. Kant, J. Electroanal.Chem., 2008, 615, 180.

55. R. Kant, Phys. Rev. Lett., 1993, 70, 4094.

56. P. V. Danckwerts, Trans. Faraday Soc., 1951,47, 1014.

57. J. Feder, "Fractals", Plenum, New York, 1988.

58. J. M. Williams and T. P. Beebe (Jr.), J. Phys.Chem., 1993, 97, 6249.

59. R. Kant, Phys. Rev. (E), 1996, 53, 5749.

60. R. S. Sayles and T. R. Thomas, Nature, 1978,271, 431.

61. O. I. Yordonov and I. S. Atansov, Eur. Phys. J.(B), 2002, 29, 211.

62. D. Vandembroucq, A. C. Boccara and S. Roux,Europhys. Lett., 1995, 30, 209.

63. R. Kant and S. K. Jha, J. Phys. Chem. (C),2007, 111, 14040.

64. S. Srivastav and R. Kant, Electrochim. Acta,2015, 180, 208.

1291


Metal-based ionic liquids : Their properties, applications andfuture prospects†

Arvind Kumar*a,b, Praveen Singh Gehlota, Krishnaiah Damarlaa, Mohit J. Mehtab andAkshay Kulshresthab

aAcademy of Scientific and Innovative Research (AcSIR), Salt and Marine Chemicals Discipline,Central Salt and Marine Chemicals Research Institute,Council of Scientific and Industrial Research (CSIR), G. B. Marg, Bhavnagar-364 002, Gujarat, IndiabSalt and Marine Chemicals Discipline, Central Salt and Marine Chemicals Research Institute,Council of Scientific and Industrial Research (CSIR), G. B. Marg, Bhavnagar-364 002, Gujarat, India

E-mail : [email protected], [email protected] Fax : 91-278-2567562


Abstract : Metal-based ionic liquids (M-ILs) is new emerging class of ionic liquids. M-ILs comprises of vari-ous transition metals or lanthanides either in cations or anions. The involvement of such metals makes thema new type of ILs materials which have properties similar to conventional ILs such as low vapour pressure,high thermal stability, and high ionic conductivity with additional features such as magnetic responsivenessand luminescence activity. Due to such amazing properties, M-ILs have been gained considerable attention inscientific community for applied research. In this review, we have drafted a general introduction, idea aboutinvolved cations and metals, brief discussion on characterization of metal containing anions and have out-lined the important scopes in catalysis, nanomaterial synthesis, colloidal formulation, biomass processing, mi-cro-extraction process, magnetic separation of biomolecules, and other potential applications anticipated infuture from M-ILs.

Keywords : Metal-based ionic liquids, paramagnetic ionic liquids, micro-emulsion, magnetic surfactants,magnetic susceptibility, PCR, micro-extraction.

1. Introduction

Currently, a major research drive is underway inindustry and academia to substitute more environ-mentally friendly technologies for traditional ones inwhich damaging and volatile organic solvents areheavily used. The exponential growth of environmentalissues year by year is a key point of research anddiscussion1. The concept of Green Chemistry (alsotermed as a sustainable chemistry) now has been sub-jected to eliminate the toxic organic substance andpromote the innovation of new zero west technologyfor an industrial process2. Thus, ILs (boon of green

†Professor Dhananjay Nasipuri Memorial Lecture (2016).

chemistry) are considered as environmentally friendlyefficient substitutes for such volatile toxic organicsubstance and solvents, not only because of their lowvapour pressures but more importantly because oftheir ability to act as catalysts and good solvationnature. Apart from these the ILs hold several otherattractive properties, including chemical and thermalstability, non-flammability, high ionic conductivity,and a wide electrochemical potential window.

ILs usually entail of inorganic anions and usuallynitrogen-containing organic cations, and their chemi-cal and physical properties can be finely tuned for a


1292

range of applications by varying the cations or an-ions. Varying the anion or cation their melting point,liquidus range and other property changes dramati-cally. For example if we replace Cl– from Br– to the[C4mim]+ cation then it will turn in to the liquidbelow room temperature or the hydrophilic naturewith BF4 can be a changed into hydrophobic naturevia replacement witha NTf2 anion. This is how ILswith desired properties are accessible and have beenexplored in many applications.

The first generation of ILs was synthesized withchloroaluminate and another metal halide where 1,3-dialkyl imidazolium and N-alkyl pyridinium cationswere at crowning3. The ILs of first-generation sufferseveral drawback including instability in water, notsuitability for biotransformations, oxygen-sensitivity4

and hygroscopic nature5.

The shortcomings of first-generation ILs now havebeen overcome by the smart choice of cations,anions, or ligands. The use of transition and lanthanidemetals in anionic part of ILs have come into picturedue to versatile nature of these elements. It is well-known that transition and lanthanide metals are usedfor catalytic activity due to variable oxidation states.Inherent magnetic and optical properties are also akey factor for interest. Halometalate and coordina-tion compounds of transition metals are also now havebeen used for task-specific ILs synthesis. Desired prop-erties say magnetic, optical, luminescent, electrochemi-cal and catalytic as an inherent in ILs anion ofhalometallate and organic ligand-metal complex havebeen used. These ILs, therefore, now has emerged asthe new functional active class of ILs. Some reviewson magnetic, catalytic, pharmaceutical and energeticproperties have been published6. Such anion or or-ganic complex component ions allow the fine anddual tuning of liquid properties of the resulting ILs,by the chemical modification of both cation and an-ion.

In this review, we focus on various metal-basedionic liquids (M-ILs) which have been developed tillnow. The review comprises use of cations and anionscontaining metal to the synthesis of M-ILs, their char-acterization and their versatile applications.

2. Choice of cations and metals

Since there is numerous choice for cations and somany unique cations have been used to meet criteriaof ILs for the desired applications. Four major classesof the cation are highlighted till date ammonium,imidazolium, pyridinium, sulfonium, and phospho-nium.

Subsequently, more biodegradable ILs have beenreported, including (cation as) amino acid-based chiral,cholinium cations, and dications. Moreover, ILs con-sisting of ammonium ions have also been reported,namely, Aliquat 336, long alkyl chain containingammonium and imidazolium cations. Apart from these,other cations such as phenthrolinium, guanidiniumcation (Fig. 1) have also been reported.

The role of cations is very important if IL for atask specific application. For example, if the anionicpart of IL is showing catalytic activity then the choiceof cation should be in such a way that it dissolves allthe reactants molecules. Combination of cation withany anions will be effective if they will show syner-gic effect, no adverse effect on activity of counterion, able to reduce the melting point, able to reducethe negative effect of counter ion, low viscosity, in-ertness to moisture etc. A great choice of cation brings

Fig. 1. Various cations used to prepare metal-based ionic liq-uids.

Arvind Kumar et al. : Metal-based ionic liquids : Their properties, applications and future prospects

1293

the ability to solubilize a large array of diverse mol-ecules along with the immiscibility both in water andnonpolar organic solvents which further adds to theirusefulness in various streams of science.

Similarly, the choice of anions is also important ifdesired properties in the ILs are to be achieved. Foranions, there is no tactic question for selection, basedon property; metal can be chosen. Some transitionmetals like V, Mn, Cr, Co, Fe, Ni, and Zn are usedin catalytic as well as magnetic applications. Somelanthanides including Eu, Yr, Tt are used as a lumi-nescent reflector. Lanthanide Gd, Ho, and Dy arealso used for magnetic properties.

In a class of halometalate, transition metal hasbeen used as general formula [MXy]

n– where X isCl, or Br. Whereas, in the class of organic metalcomplex, metal is chelated with appropriate ligand.Due to variable oxidation, transition and lanthanidemetals can be used in cation or anion complex. Fer-rocene-based ILs are example where Fe metal madecationic part of the IL. Apart from these, CN, NTf2,and mixed halo based ILs are also reported. Anion,where mixed halides are used can be represented byformula [MXaYb]

–n where X and Y are Cl or Br butXY.

3. Characterization

It is customary to elucidate the structure verifica-tion of ILs with various techniques. Mostly cationicpart is organic hence it can be verified by NMR, IRand MS techniques. But where the anionic part whichis mostly metal containing, 1H NMR and MS tech-niques may not be suitable. UV-Visible and Ramanspectroscopy has been used for complex elucidation.

Transition elements have transition electrons in theirouter most shell which undergoes in transition whenthey absorb the UV/Visible light. For electronic tran-sition in metals, Orgel and Tanabe-Sugano diagramsare useful. Orgel diagrams are used only for highspin complexes qualitatively whereas Tanabe-Suganodiagrams are useful both for high spin and low spincomplexes. So electronic transition in the anionic metalcomplex can be explained via this diagram. In (Fig.2) UV absorption spectra of 0.5 mM [C8mim]FeCl4in acetonitrile is shown as representative for all

[FeCl4]–.

FeIII, MnII have ad5 system where five unpairedelectrons collectively give equal contribution towardmagnetic and spectral properties. Due to 6S, variousspectral transitions are expected. Fig. 2 is showingthree peaks for FeIII when it was high spin tetrahedralstructure. For the UV-Vis spectra of [FeCl4] anion,the characteristic band has appeared between 450 to700 nm. At 529 nm peak appeared due to 6A1

4T2,at 602 nm peak appeared due to 6A1

4A2 and at 684nm due to 6A14T2. These are characteristic absorp-tion bands for [FeCl4]

– anion. Regarding character-ization by UV-Vis spectra of [MnCl4]

2–, two sets ofabsorption bands were reported by Jared L. Andersonet al.7. The first group of absorption bands may beattributed to transitions to D-term states with strongligand field splitting, whereas the second group ofbands originates from small ligand-field-split G terms.In general, the bands of the G-terms appear with muchlower intensities than those of the D-terms. Thesetwo bands appeared in the 335–500 nm regioncorresponding to the tetrahedral-coordinated [MnCl4]

2–

anions8.

UV-Visible absorption spectroscopy is also usedfor metal halide like [CuCl4]

2–, [NiCl4]2–, [CoCl4]

2–.Here also two absorption bands have appeared one isrelated to the aromatic ring in UV region and anotherone due to electronic transitions. 432 nm for

Fig. 2. UV absorption spectrum of [C8mim]FeCl4 (Synthe-sized and spectrum taken at CSIR-CSMCRI,Bhavnagar).


1294

[NiCl4]2–, 528 nm for [CuCl4]

2– and 682 nm for[CoCl4]

2– as reported by Seham A. Shaban et al.9.

Raman spectroscopy is an excellent tool to analysethe ionic species in their complex nature. Raman spec-tra can also sensitively reveal changes in molecularsymmetry and are thus suited to the identification ofspecies. Furthermore, the intensities of the Ramanlines can be used to empirically relate the concentra-tions of species and one can calculate the solvationnumber of metal ions. The structural confirmation ofcomplex anion such [FeCl4]

– anion can further doneby Raman analysis as shown in Fig. 3.

ing on bond (Mn-Cl)8. Similarly, other tetrahalide ILdiethylammonium terachlorocuorate (DEA)2[CuCl4]was studied by Christian Reber et al. They reportedthe A1 symmetric stretching frequency and the T2 fre-quency are 277 cm–1 and 188 cm–1, respectively forbond (Cl-Cu) at room-temperature11. Further, a zincbased anionic chloro complex was studied by FrankEndres et al. They observed the vibrational stretch-ing frequency on Zn-Cl bond in [ZnCl4]

2– at ~275cm–1. The Raman vibration frequencies for [ZnCl3]

–

can be found at ~285 and 348 cm–1.

Seham A. Shaban et al. have been prepared vari-ous M-ILs and characterized by Raman spectroscopy9.The prepared [CoCl4]

2– has been distinguished bytwo broad absorption bands at 256.61 and 325.23cm–1 assigned to the symmetric Co-Cl stretching vi-bration of [CoCl4]

2– anion12. The Raman spectrumof [NiCl4]

2– exhibiting weak bands at 250 cm–1 isassigned to the symmetric Ni-Cl stretching vibrationof [NiCl4]

2– anion (Fig. 5). Similar technique hasbeen employed for other anion complex of metals toevaluate structure and bonding.

Although mass spectrometry has been commonlyused as a technique for characterisation of ILs wherepositive and negative modes are used to characterizecation and anion of ILs, the anionic part of ILs beingcomplex leads to erroneous conclusions, for example

Fig. 3. Raman scattering spectrum of FeCl4 anion (Spectrumtaken at CSIR-CSMCRI, Bhavnagar).

The characteristic band has appeared at 334 cm–1

which corresponds to the Fe-Cl vibration mode in[FeCl4]

–. This value of 334 cm–1 belongs to the totalsymmetric vibration A1 mode. In Fig. 3, except at334 cm–1 absence of other peaks rules out the possi-bility of any dimerization originating from [FeCl4]

–

anion. When a symmetric [FeCl4]– anion is distorted

and converted into asymmetric [FeCl3Br]– anion bychemical means then other new Raman shift has beenobserved. The appearance of new bands at 224, 245,273, 334, 349 due to Fe-Br bond and a shoulder at400 (Fe-Cl) cm–1 is appeared10.

Raman spectra of the [MnCl4]2– anion (Fig. 4)

showed a characteristic absorption band at 251 cm–1,which corresponds to symmetric vibrational stretch-

Fig. 4. Raman scattering spectrum of [MnCl4]2– anion (Spec-

trum taken at CSIR-CSMCRI, Bhavnagar).


1295

about anionic speciation in chlorometallate ILs or anymetal complex anion. Despite these drawbacks, someresearchers have used this technique for characteriza-tion of metallohalide anion.

Fig. 6 shows the ESI-MS spectra of the synthe-sized metal-containing IL by Feng Yan et al. Theintensive peaks corresponding to [CuCl2Br]–,[FeCl3Br]–, and [ZnCl2Br]– can be observed at m/z211, 239 and 212, respectively13. However presenceof other peaks creates confusion to formation othercomplex.

4. PropertiesThe presence of transition metals as part of mol-

ecule have resulted in expected property of transitionmetals in ILs or have introduced the properties evenwith better performance. All the properties of transi-

tion metals including optical, spectral, magnetism,catalytic, electrochemical etc. have been introducedin ILs. Dependence on metal ion, property of IL canbe tuned accordance to desired application. Like otherILs all M-ILs exhibit same properties including wideliquidus range, high thermal stability, non-flamma-bility, less volatile, non-toxic nature, etc. Some M-ILs which are solid at room temperature also followsthe criteria of ILs. But due to involvement of metalsthese show additional properties and some of themare briefly described here in this review.

4.1. Magnetic properties :

Transition metals and inner transition metals bothexhibit magnetic property due to presence at leastsingle unpaired electron. But other factors such asspin orbital coupling, deviation from regular complexstructure, nature of ligand and surrounded solventsystem vanish the magnetism. The good choice ofmetal containing cation or anion with their counterionwill help to protect the magnetism. However, thismagnetism will differ from other magnetic materials.M-ILs which shows response to the external mag-netic field are known as magnetic ILs or paramag-netic ILs (Fig. 7).

These possess magnetic properties by itself with-out any need for the addition of magnetic nanoparticlesto ILs. So they act as a molecular magnet. Theirparamagnetic property originate either from the an-ion or cation or from both of them depending on themetal ion location. This new class of magnetic fluidshas been synthesized as an attempt to develop inher-ent magnetic property in ILs and has progressivelycontributed to the research field of ‘magnetism in

Fig. 5. Raman scattering spectrum of [NiCl4]2– anion (Spec-

trum taken at CSIR-CSMCRI, Bhavnagar).

Fig. 6. Electrospray ionization mass spectrometry (ESI-MS) of the metal-containing anions, [CuCl2Br]–, [FeCl3Br]–, and[ZnCl2Br]–, coordinated in ionic liquids (Fig. reproduced from Ref. 13, Copyright to ACS, 2017).


1296

liquids’. These single-component ILs are magneticnanoparticle-free unlike ferrofluids, but usually con-sist of complexes of magnetic ions with high mag-netic moment.

Paramagnetic ILs have some additional major ad-vantages of over other conventional magnetic fluidssuch as ferrofluids and magnetorheological fluids asa consequence of their optical transparency, small-line width and high color purity, easy to handle whichmakes them more suitable for many potential opticaland magnetic applications14. Usually, due to mag-netic property and high boiling point, possibility ofmagnetic separation and recycling can be consideredas comparatively greener process. Incorporating dif-ferent types of magnetic metal anions in ILs cause itsstrong sensitivity to magnetic field which creates inthem an additional advantage of its recoverability andreuse15.

All transition metal based ILs with unpaired num-ber have exhibited paramagnetic behavior at roomtemperature. The magnetic susceptibilities of thesehas been measured using a Quantum Design super-conducting quantum interference device (SQUID)16.The transition metal based ILs display simple para-magnetic behavior over the temperature range of 50–350 K, and room temperature magnetic susceptibilityvalues (T) has been found well corresponding withtheir respective spin states17. The magnetic suscepti-bility measurements have been considered to measurethe degree of magnetization of MILs containing dif-ferent transition metals in their structure in responseto an applied external magnetic field quantitatively.The magnetic susceptibility of most of these ILs hasbeen measured at room temperature with sweep ofmagnetic field range (kOe).

The straight line obtained between the magnetic

susceptibility and the applied magnetic field indicatesthe paramagnetic behaviour (Fig. 8). The linear fit-ting of that straight line gives quantitative values ofmagnetic susceptibility18. The values from variousmagnetic ILs has been listed in Table 1.

Fig. 7. Response of the orange-colored ionic liquid 1 to aneodymium magnet (Fig. reproduced from Ref. 23,Copyright to Wiley, 2008).

Table 1. Magnetic susceptibility of the PMILs (Copyright© 2012 Royal Society of Chemistry (From Ref. 18))

Anion Cation Magnetic susceptibility/

mT (emu K.mol–1)

[CoCl4] [P6,6,6,14] 2.10/2.48

[Co(NCS)4] [P6,6,6,14] 2.06

[MnCl4] [P6,6,6,14] 4.23/4.22

[FeCl4] [P6,6,6,14] 4.29/4.34/4.05

[FeCl4] [C2mim] 4.03

[FeCl4] [C4mim] 4.11

[FeCl4] [C4py] 4.34

[GdCl6] [P6,6,6,14] 6.51/7.72

[Dy(SCN)8–x [C6mim] 13.41–14.00

(H2O)x],

(X=0–2)

Fig. 8. Magnetization of Co, Mn, Fe and Gd based PMILswith the function of magnetic field at 300 K (Fig.reproduced from Ref. 19, Copyright to Wiley, 2017).

4.2. Luminescent properties :

Some lanthanide metals like Eu, Sm, Tb and Ybshows luminescent property. Lanthanide mostly usedin form of their complex. Lanthanide complexes haveexceptional optical properties compared to traditionalorganic dyes such as large Stokes shifts, long lumi-nescence lifetimes (s-ms) and sharp emission peaks19.In 2006, Peter Nockemann for the first time reportedthe series of rare earth metal lanthanide from Ln to


1297

Yb based ILs [C4mim]x–3[Ln(NCS)x(H2O)y], (x =6–8, y = 0–2, Ln = Y, La, Pr, Nd, Sm, Eu, Gd,Tb, Ho, Er and Yb)20.

The first europium based IL was prepared by KoenBinnemans et al. in 2006, and that is 1-hexyl-3-methylimidazolium tetrakis(naphthoyltrifluoro-acetonato)europate(III) (Fig. 9). The luminescentionogel was formed from this IL by using silica gelnetwork21.

europiumcholate IL in 2010 and studied the lumines-cent nature of ionogel. They have demonstrated therole of pyrene in luminescent enhancement wherepyrene transferred the energy to the singlet excitedstate (S1). Fig. 11 shows the time-delayed emissionspectra of Eu-cholate gel (5 mM/15 mM) in the pres-

Fig. 9. Luminescence of europium(III)-doped ionogel underultraviolet irradiation (Fig. reproduced from Ref. 22with permission, Copyright 2006, American Chemi-cal Society).

In 2008, Anja-Verena Mudring et al. further dis-closed the NTf2 based ILs with low melting point,[R]x[Eu(Tf2N)3+x] (where, Tf2N=bis-(trifluoro-methanesulfonyl) amide; x = 1 for R = 1-propyl-3-methylimidazolium (C3mim) and 1-butyl-3-methylimidazolium (C4mim); x = 2 for R = 1-butyl-1-methylpyrrolidinium (C4mpyr))22. After Eu basedIL, in 2008, Anja-Verena Mudring furhter reportedthe Dy based IL which have magnetic as well as lu-minescent properties. They synthesized the new ILs[C6mim]5–x[Dy(SCN)8–x(H2O)x] (x = 0–2, C6mim= 1-hexyl-3-methylimidazolium) from [C6mim]SCN,KSCN, and Dy(ClO4)3·6H2O

23. Fig. 10 shows thespectral transition during emission.

In 2009, Koen Binnemans reviewed the variouslanthanide complex based hybrid materials includingOLED and LASER applications24. Uday Maitra etal. have prepared luminescent ionogel using

Fig. 10. Emission spectra of DyIII based ionic liquid with tran-sition assignment of the compounds [C6mim]3[Dy(SCN)6(H2O)2] (green), [C6mim]4[Dy(SCN)7(H2O)] (red), and [C6mim]5[Dy(SCN)8] (black) atroom temperature under excitation at ex = 453 nm(Fig. reproduced from Ref. 23, Copyright to 2008,Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim).

Fig. 11. Time-delayed emission spectra of Eu-cholate gel (5mM/15 mM) in the presence of pyrene when excitedat 335 nm; delay time 0.2 ms; gate time 3 ms (Repro-duced from Ref. 26, Copyright to the Royal Societyof Chemistry, 2010).

JICS-6


1298

ence of pyrene25. Auhors also prepared other lan-thanide cholate based ionogels and characterized themin this study26.

In 2010, Slawomir Pitula and Anja-Verena Mudringdemonstrated the luminescent nature of d block tran-sition metal Mn with Cl and NTf2 anion. Figs. 12and 13 shows the luminescent nature of these MnII

based ILs8.

4.3. Catalytic properties :

Now various ionic liquids have been prepared fororganic transformation by taking advantage of versa-tile properties such as low vapour pressure, high boil-ing point, easy to recovery, and reusability. Most of

the transition and inner transition metals has beenused for various industrially important organic trans-formations. The synergic effect has been uncoveredwhen incorporation such metals is considered in ILproperties. The additional feature like magnetic char-acter has been used, for example, in easy separationof products from the reaction mixture and removal ofcatalyst at the end of reaction. Since iron is one of themost inexpensive metal among transition metals, thescope for iron-catalyzed organic reactions is of greatinterest and constantly growing. All of these featuresopen a new frontier of reinventing organic reactionspracticed in chemical industry leading to improvedprocess performance. M-ILs have been used in chemi-cal reactions in form of catalyst or reaction medium.

In 2001, Valkenberg et al. investigated the cata-

Fig. 12. Emission spectra of [C2mim]2[MnCl4] (top) and[C2mim][Mn(Tf2N)3] (bottom) (Reproduced from Ref.8, Copyright to 2010, Wiley-VCH Verlag GmbH andCo. KGaA, Weinheim).

Fig. 13. From left to right in the top row [C2mim]2 [MnCl4],[C3mim]2 [MnCl4], [C4mim]2 [MnCl4], and [C6mim]2[MnCl4]; in the bottom row [C2mim][Mn(Tf2N)3],[C3mim][Mn(Tf2N)3], [C4mim][Mn(Tf2N)3], and[C6mim][Mn(Tf2N)3] excited at room temperaturewith UV light (ex = 366 nm) (Reproduced fromRef. 8, Copyright to 2010, Wiley-VCH Verlag GmbHand Co. KGaA, Weinheim).


1299

lytic use of [FeCl4]– anion based IL in the Friedel-

Crafts acylation of aromatic compounds as an alter-native to existing homogeneous catalysts27.

magnetic properties and even in both. They preparedthe novel electrochromic and magnetic ILs (Fig. 15)by a simple combination of CoIII, CrIII and FeIII withethylenediaminetetraacetic (EDTA) complexes as an-ions and the cations was used 1-ethyl-3-methylimidazolium [C2mim], 1-butyl-3-methylimidazolium [C4mim], 1-octyl-3-methyl-imidazolium [C4mim], trioctylmethylammonium[N1888] and trihexyltetradecylphosphonium[P6,6,6,14]

30.

Fig. 14. Chitosan supported magnetic ionic liquid as a cata-lyst28 (Fig. reproduced from Ref. 28, Copyright toRSC Advances, 2014).

In 2013, Muraoka et al. first time reported aneffective utilization of M-ILs as a medium in dissolu-tion of crystalline cellulose and recovery of cellulosemagnetically29. More details are described in this re-view under application section.

4.4. Electrochemical properties :

The accessibility of variable oxidation state in tran-sition metals allow to use them in electrochemicalapplication. M-ILs have been disclosed as newelectrochromic materials for the manufacturing ofdifferent electronic devices such as information dis-plays, anti-glare rear view of automobiles, smart win-dows, etc. by taking advantage of photophysical andoptical properties that originate from the various met-als incorporated in the complex anion of correspond-ing IL. In 2011, Luis C. Branco and Fernando Pinaetet al. proposed Co, Cr and Fe based ILs are not onlycapable of behaving as electrolytes but also to makean electrochromic component. The appropriate com-bination of electrochromic and magnetic anions makethem suitable candidate for electrochemical applica-tions. They have also proved that anions or cationswhich have metal with variable oxidation nature arecapable of reversible oxidation/reduction processesinvolving the strong changes in either colour or their

Fig. 15. Visible variations of the compound [C2mim][Co(EDTA)] upon reduction at –0.38 V and 300 min.A liquid U shaped electrochromic cell containing twoimmiscible ionic liquids showing simultaneous oxida-tion reduction (Reproduced from Ref. 30, Copyrightto The Royal Society of Chemistry, 2011).

Akitsu and Einaga et al. in 2006, reported a newphoto-controllable supramolecular system containing[C4mim][FeCl4] and azobenzene as a new photo-functional magnetic material with application formemory devices31.


1300

4.5. Thermo and solvatochromism :

Some of the M-ILs investigated showedthermochromic as well as solvatochromic properties.In 2008, George Z. Chen et al. prepared chloro-NiII

complexes based ILs. Thermochromic and solvato-chromic behaviour was investigated by visual obser-vation and visible-spectroscopy in 1-hydroxyalkyl-3-methylimidazolium [Cn(OH)mim][ClO4], where n =2 or 3 based ionic liquid in the range of 25 to 85 ºC.The hypothesis was given by them that the tetrahe-dral [NiCl4]

– (blue, hot) has been converted intooctahedral structure [NiClx(Cn(OH)mim)y]

z+ (yellowor green, cold) upon where [Cn(OH)mim]+ cation ILhas been used as ligand.

Fig. 16 shows that colour of [C4mim][NiCl4] ischanged upon heating due to evaporation of ethanolwhere [C4mim]Cl stabilized the colour intensity. Fig.16 also confirmed the importance of hydroxyl groupwhich utilized in coordination bonding with Ni metalion32.

tetrachloro-nickelate [C4mim]2[NiCl4] and excess of1-butyl-3-methylimidazolium chloride [C4 mim]Cl hasbeen demonstrated by George Z. Chen in an IL sol-vent, 1-(2-hydroxyethyl)-3-methylimidazoliumtetrafluoroborate [C2(OH)mim]BF4 (Fig. 17)33.

Fig. 16. Photos of [C4mim]2[NiCl4] in ethanol and [C3(OH)-mim]BF4 at the indicated temperatures (Reproducedfrom Ref. 32, Copyright to The Royal Society ofChemistry, 2008).

In 2014, again George Z. Chen et al. disclosedthe Cryo-solvatochromism. The Cryo-solvatochromismis more precise term of thermochromism along withsolvatochromism. Blue to green colure changes inresponse to cooling from room temperature to wellbelow 0 ºC can have many applications. The Cryo-solvatochromism of di-(1-butyl-3-methylimidazolium)

Fig. 17. Photographs of 0.14 mole.L–1 [C4mim]2[NiCl4] and1.4 mole.L–1 [C4mim]Cl in [C2(OH)mim]BF4 at (A)22 ºC and (B) –13 ºC (Reproduced from Ref. 33,Copyright to The Royal Society of Chemistry, 2014).

In 2016, Jun Ding and Xianmao Lu et al. havereported for the first time the thermoresponsive Febased magnetic ILs octyltrimethylammoniumbromotrichloroferrate(III)([N1,1,1,8][FeBrCl3]),dodecyltrimethylammonium tetrachloroferrate(III)([N1,1,1,12][FeCl4]), tetradedecyltrimethylammoniumbromotrichloroferrate(III) ([N1,1,1,14][FeBrCl3]), and8-butyl-1,8-diazabicyclo[5.4.0]undec-7-enebromotrichloroferrate(III) ([DBU-Bu][FeBrCl3]). Allfour magnetic ILs respond to temperature change.They have investigated the thermoresponsive behaviourof magnetic ILs (25 wt%) using UV-Vis transmit-tance at 600 nm with temperature. Fig. 18 shows thelower critical solution temperature (LCST) of[N1,1,1,8][FeCl4] and [DBU-Bu][FeBrCl3] are 60 ºCand 50 ºC respectively, at which the % transmittancesdropped abruptly (LCST is the critical temperaturebelow which the components of a mixture are mis-cible for all compositions).

The concentration of magnetic ILs also interferein the lowering of LCST. The LCST of [N1,1,1,8]-[FeCl4] was reduced from 60 to 38 ºC when concen-tration changed from 25 to 33 wt%. Similarly for[DBU-Bu][FeBrCl3] the LCSTs were 63 and 50 ºCat concentrations of 20 wt% and 25 wt% respec-tively34.


1301

5. Applications5.1. Catalysis :

As described in the Section 4.3 M-ILs have beenused in various organic transformation. M-ILs havebeen used either in form of catalyst or medium. In2005, Andrew P. Abbott et al. investigated the use ofzinc-based IL which was prepared by mixing of 1 eq.of choline chloride and 2 eq. of ZnCl2; [ChCl][ZnCl2]2for the O-acetylation of sugars and cellulose35. Here[ChCl][ZnCl2]2 played the role of solvent as well ascatalyst. In 2005, Donghong Yin et al. prepareddiphenylmethane and its derivatives via Friedel-Craftsbenzylation reaction where 1-butyl-3-methylimidazolium chloride-ZnCl2 ([C4mim]Cl-ZnCl2), 1-butyl-3-methylimidazolium chloride-FeCl3([C4mim]Cl-FeCl3) and 1-butyl-3-methylimidazoliumchloride-FeCl2 ([C4mim]Cl-FeCl2) were used bothreaction media and Lewis acid catalysts. These ILswere conveniently recovered and for recycled.[C4mim]Cl-ZnCl2 was reused at least eight times with-out loss of catalytic activity due to higher moisture36.In 2008, Bica et al. developed the very efficient, ea-sily prepared and reusable iron-containing IL;butylmethylimidazolium tetrachloroferrate(III)[C4mim][FeCl4] as a coherent and magnetic catalyst

for the hydromethylation of the -keto ester in highyield. They also investigated the other transition met-als NiII, CoII, CuII and TiIV for desired reactions37.In 2010, Yu Lin Hu et al. have reported the conver-sion of benzyl chloride and its derivative to benzalde-hyde using surface active magnetic IL [C12mim][FeCl4] with oxidant H5IO6.[C12mim][FeCl4] dem-onstrated the best performance. Authors also investi-gated the catalytic effects of other different alkyl chaincontaining magnetic ILs which may be attributed totheir different abilities of stabilizing and dissolvingH5IO6 and the substrate. Under similar reaction con-ditions, H5IO6 was more soluble in [C12mim][FeCl4],leading to higher effective concentration of the oxi-dant38. In same year 2010, Hui Wang et al. haveshowed the high synergic effect and eco-friendly ap-proach for the glycolysis of the PET plastic by usingthe [C4mim][FeCl4] as an efficient catalyst in ethyl-ene glycol medium39. In continuation to degrade PETplastic, Hui Wang et al. 2015, further investigatedthe high selectivity and recyclability of first transitionmetal based IL12. Amutha Chinnappan et al. 2013,proposed the nanocomposite supported catalytic re-duction of nitroarenes. The nanocomposite was pre-pared from PVDF with nickel-based dicationic IL,

Fig. 18. (a) Change in UV-Vis transmittance with temperature at 25% wt. (b) Change in UV-Vis transmittance with temperatureat different concentrations. Heating of [DBU-Bu][FeBrCl3] solution from 24 to 50 ºC and magnetic separation fromwater at 50 ºC (Reproduced from Ref. 34, Copyright to The Royal Society of Chemistry, 2016).


1302

[C6(mpy)2][NiCl4]2– at room temperature. The

polyvinylidene fluoride (PVDF)-Ni based nanofibercomposite catalyst was proposed more active than otherNi-based catalysts. The use of such nanofibre elimi-nated the need of inert atmosphere, and additionalbase or other additives. The catalytic system showedthe good yields and selectivity (Fig. 19)40.

2015, Lei Yang et al. also carried out catalytic reac-tion from dual magnetic ionic liquid-mesoporus silica.The mesoporous silica was synthesised from M-IL,[C16mim]3PMo12O40 and the triblock copolymer P123as co-templates. In this strategy, the IL, [C16mim]3PMo12O40 was not only employed as the template,but also provided the metal source. Dual magneticIL-mesoporus silica has been used as a catalyst forthe oxidative desulfurization of fuel44. MarenMuntzeck et al. 2016, proposed a three-componentoxidative dehydrogenation tandem reaction via thecoupling and hydroarylation of benzaldehyde, anilineand phenylacetylene to a quinoline derivate which wascatalysed by an iron-containing IL with anions [FeCl4]

–

and [Fe2Cl7]– 45. In 2017, Huiqing Liu et al. showed

the excellent catalytic activity of the magnetic ionicliquid [C4mim][FeCl4] for methanolysis of poly(lacticacid)46. Recently Sachin R. Thawarkar et al. 2017,have synthesized imidazolium based nickel and palla-dium containing catalyst [C4mim]2[PdCl4] and[C4mim]2[NiCl4] and used as effective catalysts forthe reduction of nitroarenes to aminoarenes in thepresence of NaBH4. The catalytic enhancement wasattributed to in situ formation of Pd and Ni metalnanoparticles (NPs) were fromed by the reduction ofrespective M-ILs47.

5.2. Biomass processing :

The depletion of conventional fuels is forcing man-kind to focus on renewable sources for the produc-tion of energy, fuels and chemicals48. Biomass is theone of the renewable sources of conventional fuels.Hence, it is very important to utilize this biomass forits valorization into a basic building block compound.On the other hand cost of such chemicals has in-creased and availability is decreased49. Use of ILs assolvents in biomass dissolution and conversion of dis-

Fig. 19. Reduction of aromatic nitro compounds to the corre-sponding aniline with [C6(mpy)2][NiCl4], NaBH4 andH2O (Reproduced from Ref. 40, Copyright to TheRoyal Society of Chemistry, 2013).

Fig. 20. Reaction scheme (Reproduced from Ref. 42, Copy-right to Tetrahedron Lett., 2016).

In 2014, Soheil Sayyahi et al. prepared paramag-netic 3-(2-hydroxyethyl)-1-methyl imidazoliumbromotrichloro ferrate(III) [C2(OH)mim][FeCl3Br]which was used in the catalytic synthesis of 1-amidoalkyl-2-naphthols via one-pot three-componentcondensation reaction of aromatic aldehydes, 2-naph-thol and acetamide at 85 ºC under solvent-free condi-tions41. In 2014, Ali Khalafi-Nezhad et al. proposeda green and highly efficient chitosan supported mag-netic IL (CSMIL, Fig. 14) which was synthesizedfrom chitosan, methyl imidazole and anhydrous/hy-drous FeCl3

28. The heterogeneous catalyst thus ob-tained was used for the direct conversion of alde-hydes to the corresponding nitriles in the presence ofNH2OH.HCl/dry-CSMIL/MeSO2Cl and amides withNH2OH.HCl/wet-CSMIL/MeSO2Cl28. In 2015,AliAkbari developed an efficient green catalyst forthe one-pot synthesis of 1,2,4,5-tetrasubstituted imi-dazoles via the condensation of benzil, an aromaticaldehyde, aniline and ammonium acetate. The greensalt and magnetic gadolinium-based IL, [C4mim]3[GdCl6] was used for this reaction (Fig. 20)42.

Further in series of condensation reactions, in 2015,Soheil Sayyahi et al. used [Bmim][FeCl4] as a cata-lyst for a synthesis of 2-aryl benzimidazoles and 2-aryl benzothiazoles derivatives with high yield43. In


1303

solved biomass fractions like lignin into chemicals ispromising50.

Specially IL as solvent and metal chloride as acatalyst is being used widely for this type of biom-ass51. The use metal containing ILs for this applica-tion plays dual functions viz. as solvent and as a cata-lyst. There are very few reports available where ILshave been used as both the solvent and catalyst forbiomass processing. The first report of conversion ofbiomass into other chemicals was reported by Lianget al., where they used IL [Et3NH]Cl-AlCl3 alongwith other ILs and converted soybean oil into biodiesel.They have studied different variation of ILs alongwith [Et3NH]Cl in conjugation with metal chlorideslike Fe, Zn, Mg and Sn. They also studied effect ofdifferent anions on conversion of soyabean oil intobiodiesel52. In 2013, Muraoka et al. successfully dis-solved crystalline cellulose using magnetic IL,[Cmmim]FeCl4. They concluded that cationic moietymay play major role behind dissolution of cellulose29.The another report by Saidana et al. states the con-version of oil palm fronds (OPF) into lavulinic acid(Fig. 21). They used acidic ionic liquid [SMIM]FeCl4to convert OPF and glucose into levulinic acid. Theyalso achieved up gradation of levulinic acid to ethyllevulinate through esterification with ethanol over[SMIM]FeCl4

53.

5.3. Magnetic surfactants and their colloidal formu-lations :

There is a vast possibility of tuning the IL struc-tures and it is well known that amphiphilic character

can be introduced in ILs by incorporating long alkylchains to IL moieties. These compounds are struc-tural relatives of common surfactants and have beencoined surfactant ionic liquids (SAILs). Recently, sur-factants responsive to a magnetic field have been de-veloped that offer unique physicochemical properties.The surfactants those shows response toward exter-nal magnetic field are identified as Magnetic IonicLiquid Surfactants (MILSs, Fig. 21). In 2012, JulianEastoe et al. showed that MILSs can be readily pro-duced by mixing an iron trihalide with the appropri-ate cationic surfactant. Surprisingly, even micellarsolutions of these MILSs demonstrate a field response,leading to the intriguing possibility of micellar struc-turing, which would not happen with a normal non-micellizing magnetic IL54. In continuation 2013, thegroup of Paul Brown and Julian Eastoe further pre-pared another kind of MILs which included lanthanidesmetal counterions (Ho, Gd and Ce) also and charac-terized well55. Heteroanions containing MILs havebeen developed from same group where [FeCl3Br]and [AOT] anions has been used simultaneously.Sebastian Polarz et al. reported that aqueous phasebehaviour of MILSs is similar to that of conventionalsurfactants and self-assembly provides a simple meansof preparing well-defined aggregates of metal com-plexes related magnetic surfactant56.

This discovery adds to the armoury of responsivestimuli, allowing for surfactant properties to be con-trolled simply by the switching “on” and “off” of amagnetic field. SQUID magnetometry has been used

Fig. 21. Conversion of OPF to HMF and LA (Work is reproduced from Ref. 53, Copy right to Springer Science+BusinessMedia, New York, 2016).


1304

to elucidate the magnetic phase behavior, and small-angle neutron scattering (SANS) provides evidenceof micellar aggregation in aqueous media. The studyalso reveals that for cationic surfactants in aqueoussystems there appears to be no significant increase inmagnetic susceptibility after micellization55. In 2016,Paul Brown et al. used muon spin relaxation spec-troscopy to study how the aggregation behavior ofmagnetic surfactants containing lanthanide counterionsmay be exploited to create spin glass-like materials.The magnetic behavior of MILs may also be manipu-lated via formation of micelles rather than simple di-lution as well as via design of surfactant moleculararchitecture57. Metal-based ILs have also been usedto contruct microemulsions (MEs) and other colloidalformulations. It is well known, MEs are the thermo-dynamically stable clear transparent solutions con-sists of polar, nonpolar and an emulsifier. The part ofreview has focused on MEs which have utilized metalbased ILs or DES as required components58–60.

Introduction of metal based ILs to MEs leads towider applications, especially for the synthesis of nano-materials based on their magnetic nature.

MEs formed from magnetic ILs are rarely found,and has limited reports till date. MEs with magneticproperties has been formed by employing a magnetic

Fig. 22. Response of liquid droplets to the field froma 0.4 TNdFeB magnet. [C10mimCl] and [C10mimF] = 20wt% (Reproduced from Ref. 54)54.

room temperature ionic liquid i.e. [C4mim][FeCl4]as polar phase, cyclohexane as non-polar phase andan appropriate mixture of ionic surfactant and decanolas a cosurfactant (Fig. 23)61.

When the alkyl chain length of the imidazoliummoiety is increased to C10, the IL, 1-methyl-3-

Fig. 23. (a) Response of the MRTIL containing MEs to thefield gradient of an electromagnet. The sample shownconsists of 31.2 wt% D12-cyclohexane, 46.1 wt%bmim[FeCl4], 11.8 wt% C16mimCl and 10.9 wt%decanol. The magnetic field is oriented parallel to theliquid surface. (b) Pseudo ternary phase diagram (byweight) for mixtures of MRTIL (bmim[FeCl4]) andcyclohexane formulated with C16mimCl and decanol(molar ratio 1 : 2) as surfactant and co-surfactant,respectively. The filled squares mark the detectedphase transitions and the filled circles indicate theSANS samples. The dashed line gives the experimen-tal path for conductivity titration and magnetic mea-surements (Reproduced from Ref. 61, Copyright toThe Royal Society of Chemistry, 2012).


1305

decylimidazolium tetrachloroferrate ([C10mim][FeCl4])proved to be surface active and generated a magneticionic liquid surfactant (MILS)54.

Ammonium and imidazolium based magnetic sur-factants have been studied for their aggregationbehaviour using SANS technique by Eastoe and co-workers55.

The opinion on aggregation of magnetic surfac-tants has been provided by Brown et al.62.

5.4. Material synthesis :

From past to present, studies on nanomaterials haveplayed key role in applied science. Requirement ofthese materials is high due to their high surface area,low size and desired physicochemical properties. Syn-thesis of these advanced materials (functional materi-als, quantum dots, porous materials etc.) using greenermethods is desired. Among such methods, utilizationof IL based MEs as “Nano reactors” is very promis-ing63,64.

Scheme below provides an overview of how theNPs are synthesized in a cyclic manner using IL basedMEs (Scheme 1).

containing mesoporous silica materials were synthe-sized by using different M-ILs, ([C16mim]Cl/CuCl2,[C16mim]Cl/FeCl3, [C16mim]Cl/MnCl2 and[C16mim]Cl/NiCl2 as templates, tetraethyl orthosilicate(TEOS) as silica source at room temperature in acidicmedia (Fig. 24)65.

Scheme 1. General synthetic route for the preparation of nano-particles (NPs) using magnetic based MEs.

Inclusion of metal based ILs for the synthesis ofnanoparticles in an eco-friendly manner is very im-portant. Xiong et al. have reported highly mesoporousmaterials for the first time wherein various metal-

Fig. 24. The formation process of the metal-containingmesoporous silica65 (Copyright Ref. 65 to Royal So-ciety of Chemistry, 2014).

In this review, magnetic ILs (as polar, nonpolarand as an emulsifier or template) based MEs havebeen focussed and explored huge advantages of suchMEs to produce single distributed ultra-fine preferredsize and morphologies, and properties in eco-friendlymanner.

Following are the advantages of M-ILs basedMEs :

(1) It is an eco-friendly method such as recy-clable “Nanoreactors”.

(2) Single distribution in size.

(3) Desired morphological materials (spherical,rod etc.).

(4) Inducing magnetic nature (for desired appli-cations).

JICS-7


1306

(5) Regeneration of M-ILs based MEs.

Limitations of M-ILs based MEs :

(1) M-ILs are somewhat costly materials com-pared to conventional solvents.

(2) Some of M-ILs are toxic in nature.

5.5. Extraction and stability of biomolecules :

ILs have been used to resolve the major challengesin biotechnology including effective control over trans-port and delivery of biomolecules for the regulationof gene suppression, targeted drug delivery66, andprotein separation67. In 2012, Paul Brown et al. dem-onstrated the unprecedented low strength magneticfield-induced migration of DNA and proteins(biomolecules separation) via magnetic surfactant con-jugation. Moreover, near-native conformations of thebiomolecules could be maintained by careful controlover the biomolecule : surfactant stoichiometry. Theyhave used magnetic cationic surfactants DTAF([C15H34N]+[FeCl3Br]–), DTAG ([C15H34N]+

[GdCl3Br]–) and DTAH ([C15H34N]+ [HoCl3Br]–)for this study68. In 2015, Jingcheng Hao et al. re-ported a low strength magnetic field-induced migra-tion of DNA and proteins via the conjunction withonekind of novel surfactant-coated magnetic AuNP(Fig. 25). The advantage of using these nanoparticlesis their easy preparation (one-step modification), goodstability and dispersibility, fast and effective binding,and high migration efficiency. The native conforma-tion of DNA and proteins can be protected during themigration process by carefully controlling the stoi-chiometric ratio of the AuNPs and biomolecules69.

Further, in biotechnology magnetic extractionphases have been employed in nucleic acid analysisas mobile substrates for the rapid extraction of DNA.In magnet-based approaches, the DNA-enriched ex-traction medium is readily isolated and controlled byapplication of an external magnetic field. Function-alized magnetic beads are commonly used in foren-sics and drug discovery applications to increase samplethroughput by eliminating the need for tedious cen-trifugation steps70. In 2015, Jared L. Anderson et al.have studied extraction of DNA using hydrophobicM-ILs. In total, three hydrophobic M-ILs, namely1,12-bis[N-(N-hexadecylbenzimidazolium)-dodecane

bis[(trifluoromethyl)sulfonyl] imide bromotrichlo-ferrate(III) ([(C16BnIM)2C12

2+][NTf2–, FeCl3Br–]),

benzyltrioctylammonium bromotrichloroferrate(III)([(C8)3BnN+][FeCl3Br–]), and trihexyl(tetradecyl)-phosphonium tetra-chloroferrate(III) ([P6,6,6,14

+]-[FeCl4

–]), have been employed for the direct extrac-tion of DNA from an aqueous solution (Fig. 26).They isolated the extraction phase by applying anexternal magnetic field, thereby circumventing time-consuming centrifugation steps.

The optimized M-ILs based extraction proceduresare capable of performing rapid and highly efficientextraction of double-stranded and single-stranded DNAfrom a matrix containing metal ions and protein. Plas-mid DNA (pDNA) has also been extracted from abacterial cell lysate using the M-ILs based methodand shown to be a high quality template for PCR71.In same year 2015 and same group Jared L. Ander-son et al., have been reported a method for M-ILs

Fig. 25. Synthesis of the magnetic gold particles and demon-stration of migration (Reproduced from Ref. 69, Copy-right to The Royal Society of Chemistry, 2015).


1307

based extraction of bacterial plasmid DNA (pDNA)followed by immediate PCR amplification and detec-tion of a target gene (Fig. 27). They designed thePCR buffer to enable the amplification of a targetgene from pDNA-enriched magnetic ionic liquids.According to them by carefully engineering the com-ponents within a PCR mixture, the pDNA-enrichedM-IL could be added directly to a PCR tube for geneamplification without additional sample purification.The results have been demonstrated the feasibility ofinterfacing M-IL extraction solvents with biochemi-cal assays to achieve rapid enrichment and analysisof DNA.

The results have also showed that PCR inhibitioncaused by the cationic and anionic components of twoM-ILs; trihexyl(tetradecyl)phosphonium tetra-chloroferrate(III) ([P6,6,6,14]

+[FeCl4]–) MIL or the

trioctylbenzylammonium bromotrichloroferrate(III)([(C8)3BnN]+[FeCl3Br]–) could be mitigated usingalbumin, iron(III) chelators and by increased buffer

capacity of the PCR mixture. These M-ILs were alsoused to extract PCR amplifiable pDNA from crudebacterial cell lysate without the need for time con-suming sample purification or DNA recovery proce-dures. This study done by them opens up the compat-ibility of M-ILs solvents with bioanalytical techniquesto dramatically reduce the time required for DNAanalysis, making these materials particularly attrac-tive for food safety or other high throughput applica-tions60. In 2015 Jingcheng Hao et al. have constructedfor the first time well-ordered surfactant- DNA hy-brid nanospheres made from double-strand DNA andcationic surfactants with magnetic counterion,[FeCl3Br]–. The magnetic cationic surfactants com-pacted the DNA at high concentrations and built well-ordered nanospheres through aggregation, fusion, andcoagulation processes. They also used light-respon-sive magnetic cationic surfactant to producenanospheres which was further used as a dual con-trollable drug-delivery system. These systems can bestimulated with external magnetic force and alterna-tive UV and visible light (Fig. 28). The formednanospheres had high drug absorption efficiency, slowrelease property, and good biocompatibi-lity. Authorsexplored the potential candidates for effective mag-netic-field-based targeted drug delivery, followed byphotocontrollable drug release72.

Conclusion and Future Visions

In this review, the main focus have been centred tothe synthesis, characterization and applications of metalbased ionic liquids (M-ILs). M-ILs containing tran-

Fig. 26. Reproduced the graphical abstract from Ref. 71.

Fig. 27. Amplification of the MTAP gene within PCR buffers spiked with (a) 20 mMFeCl3, (b) 0.5 mL of [(C8)3BnN]+[Br]–

(lane 3) or 0.5 mL of [P6,6,6,14]+[Cl]– (lanes 4–6).


1308

Chemical Society for providing the opportunity tosubmit the review on metal based ionic liquids.

Abbreviation :ILs Ionic liquids

[PF6]– Hexa fluorophosphate

[NTf2]– Bis trifluorosulfoamide

Cl– Chloride

Br– Bromide

Aliquat

[CN]– Cyanide

NMR Nuclear Magnetic Resonance

IR- Infrared red

MS Mass spectra

[FeCl4]– Tetrachloroferrate(III)

[MnCl4]2– Tetrachloromaganate(II)

[CuCl4]2– Tetrachlorocuparte(II)

[NiCl4]2– Tetrachloronickelate(II)

[CoCl4]2– Tetrachlorocobaltate(II)

[FeCl3Br]– Bromotrichloroferrate(III)

[CuCl2Br]– Bromodichlorocuprate(II)

[ZnCl2Br]– Bromodichlorozincate(II)

DEA Diethyl ammonium

ESI Electron spray ionization

m/z Mass to charge ratio

MILs Magnetic ionic liquids

PMILs Paramagnetic ionic liquid

SQUID Superconducting quantum in-terface device

KOe Kilo Oersteds

EDTA Ethyldiamine tetraacetic acid

LCST Lower critical solution tem-perature

ChCl Choline chloride

PET Polyethylene tetraphthalate

PVDF Polyvinylidenefluoride

MILSs Magnetic ionic liquid surfac-tants

SANS Small-angle neutron scatter-ing

Fig. 28. UV-Visible and magnetic field triggered drug deli-very by using magnetic surfactant (Reprinted fromRef. 72, Copyright to 2015, American Chemical So-ciety).

sition metals like Co, Cr, V, Mn, Fe, Cu, Ni and thelanthanide complexes Gd, Ho, Yb, Tb and Dy havebeen studied and explored for their potential applica-tions. Special properties of M-ILs such optical or lu-minescence, magnetic and catalytic are leading to newinteresting potential applications in both process andproduct engineering. Regarding process engineering,M-ILs may be applied in extraction, separation andas hydrophobic media. In chemical reactions they canact as catalyst, solvent or reaction medium. With ref-erence to product engineering, M-ILs can be used inapplications related to the polymer chemistry, devel-opment of electrochemical and medical devices. Bio-technological and biomedical applications are alsolinked with M-ILs. Many other important applica-tions in fields such as microextraction, environmentalremediation and gas adsorption possess great syntheticimportance. Utility of certain paramagnetic ILs aspromising candidates for CO2 gas adsorption needsto be given special attention because of the growingconcern for global warming and other related causes.

Acknowledgement

Authors are thankfulto the Department of Scienceand Technology (DST), Government of India, for thefinancial support for this work (No. EMR/2016/004747). P. S. Gehlot and Mohit Mehta are thankfulto UGC, Government of India, for providing ResearchFellowships. Authors are also thankful to the Indian


1309

MEs Micro-emulsions

DES Deep eutectic solvent

MRTILs Magnetic room temperatureionic liquid

DTAB Dodecyl trimethylammonium-bromide

NPs Nanoparticles

TEOS Tetraethylorthosilicates

pDNA plasmid DNA

PCR Polymerase chain reaction

References

1. L. Bai, J.-X. Wang and Y. Zhang, Green Chem., 2003,5, 615.

2. V. S. Padalkar, V. D. Gupta, K. R. Phatangare, V. S.Patil, P. G. Umape and N. Sekar, Green Chem. Lett.Rev., 2012, 5, 139.

3. J. S. Wilkes, J. A. Levisky, R. A. Wilson and C. L.Hussey, Inorg. Chem., 1982, 21, 1263.

4. F. Endres and S. Z. El-Abedin, Phys. Chem. Chem.Phys., 2006, 8, 2101.

5. E. Moustafa, S. Zein El-Abedin, A. Shkurankov, E.Zschippang, A. Saad, A. Bund and F. Endres, J. Phys.Chem. (B), 2007, 111, 4693.

6. Z. Fei, T. J. Geldbach, D. Zhao and P. J. Dyson, Chem-istry-A European Journal, 2006, 12, 2122; R. P. Singh,R. D. Verma, D. T. Meshri and J. n. M. Shreeve,Angew. Chem. Int. Ed., 2006, 45, 3584; M. Smiglak,A. Metlen and R. D. Rogers, Acc. Chem. Res., 2007,40, 1182; W. L. Hough and R. D. Rogers, Bull. Chem.Soc. Jpn., 2007, 80, 2262.

7. H. Yu, J. Merib and J. L. Anderson, J. Chromat. (A),2016, 11, 1463.

8. S. Pitula and A. V. Mudring, Chemistry-A EuropeanJournal, 2010, 16, 3355.

9. F. A. Yassin, F. Y. El-Kady, H. S. Ahmed, L. K.Mohamed, S. A. Shaban and A. K. Elfadaly, Egypt. J.Petro., 2015, 24, 103.

10. M. Döbbelin, V. Jovanovski, I. Llarena, L. J. C.Marfil, G. Cabañero, J. Rodriguez and D.Mecerreyes, Polym. Chem., 2011, 2, 1275.

11. Y. Suffren, F.-G. Rollet and C. Reber, Comm.Inorg. Chem., 2011, 32, 246.

12. Q. Wang, Y. Geng, X. Lu and S. Zhang, ACSSust. Chem. Eng., 2015, 3, 340.

13. Z. Zheng, J. Guo, H. Mao, Q. Xu, J. Qin and F.Yan, ACS Biomat. Sci. Eng., 2017.

14. J. Chen, "Application of Ionic Liquids on Rare

Earth Green Separation and Utilization", Springer,2015.

15. S. H. Lee, S. H. Ha, C.-Y. You and Y.-M. Koo,Korean J. Chem. Eng., 2007, 24, 436.

16. M. Li, S. L. De Rooy, D. K. Bwambok, B. El-Zahab, J. F. DiTusa and I. M. Warner, Chem.Commun., 2009, 6922.

17. R. E. Del Sesto, T. M. McCleskey, A. K.Burrell, G. A. Baker, J. D. Thompson, B. L.Scott, J. S. Wilkes and P. Williams, Chem.Commun., 2008, 447.

18. E. Santos, J. Albo, A. Rosatella, C. A. Afonsoand Á. Irabien, J. Chem. Technol. Biotech.,2014, 89, 866.

19. J.-C. G. Bünzli, Chem. Rev., 2010, 110, 2729.

20. J. Alvarez-Vicente, S. Dandil, D. Banerjee, H. N.Gunaratne, S. Gray, S. Felton, G. Srinivasan, A.M. Kaczmarek, R. Van Deun and P. Nockemann,J. Phys. Chem. (B), 2016, 120, 5301.

21. K. Lunstroot, K. Driesen, P. Nockemann, C.Görller-Walrand, K. Binnemans, S. Bellayer, J.Le Bideau and A. Vioux, Chem. Mater., 2006,18, 5711.

22. S. Tang, A. Babai and A. V. Mudring, Angew.Chem. Int. Ed., 2008, 47, 7631.

23. B. Mallick, B. Balke, C. Felser and A. V. Mudring,Angew. Chem. Int. Ed., 2008, 47, 7635.

24. K. Binnemans, Chem. Rev., 2009, 109, 4283.

25. S. Bhowmik, S. Banerjee and U. Maitra, Chem.Commun., 2010, 46, 8642.

26. V. Sajisha and U. Maitra, CHIMIA Int. J. Chem.,2013, 67, 44.

27. M. Valkenberg and W. Hölderich, Green Chem.,2002, 4, 88.

28. A. Khalafi-Nezhad and S. Mohammadi, RSCAdv., 2014, 4, 13782.

29. J. Muraoka, N. Kamiya and Y. Ito, J. Mole.Liq., 2013, 182, 76.

30. A. Branco, L. C. Branco and F. Pina, Chem.Commun., 2011, 47, 2300.

31. T. Akitsu and Y. Einaga, Inorg. Chem. Commun.,2006, 9, 1108.

32. X. Wei, L. Yu, D. Wang, X. Jin and G. Z. Chen,Green Chem., 2008, 10, 296.

33. L. Yu and G. Z. Chen, RSC Adv., 2014, 4,40281.

34. Q. Zhao, T. S. Herng, C. X. Guo, D. Zhao, J.Ding and X. Lu, RSC Adv., 2016, 6, 15731.

35. A. P. Abbott, T. J. Bell, S. Handa and B. Stoddart,Green Chem., 2005, 7, 705.

36. D. Yin, C. Li, L. Tao, N. Yu, S. Hu and D. Yin,


1310

J. Mole. Catal. A : Chem., 2006, 245, 260.

37. K. Bica and P. Gaertner, Eur. J. Org. Chem.,2008, 3453.

38. Y. L. Hu, Q. F. Liu, T. T. Lu and M. Lu, Catal.Commun., 2010, 11, 923.

39. H. Wang, R. Yan, Z. Li, X. Zhang and S. Zhang,Catal. Commun., 2010, 11, 763.

40. A. Chinnappan and H. Kim, RSC Adv., 2013, 3,3399.

41. S. Sayyahi, A. Azin and S. J. Saghanezhad, J.Mole. Liq., 2014, 198, 30.

42. A. Akbari, Tetrahedron Lett., 2016, 57, 431.

43. S. Sayyahi, S. Shabani, S. Ghasemi, A. Azin andS. M. Hasani, Oriental J. Chem., 2015, 31,1773.

44. L. Yang, J. Xiong, H. Li, M. Zhang, W. Jiang,H. Liu, W. Zhu and H. Li, RSC Adv., 2015, 5,104322.

45. M. Muntzeck and R. Wilhelm, Int. J. Mole. Sci.,2016, 17, 860.

46. H. Liu, R. Zhao, X. Song, F Liu., S. Yu, S. Liuand X. Ge, Catal. Lett., 2017, 147, 2298.

47. S. R. Thawarkar, N. D. Khupse and A. Kumar,Chemistry Select, 2017, 2, 6833.

48. M. Dashtban, A. Gilbert and P. Fatehi, RSCAdv., 2014, 4, 2037.

49. P. C. Bruijnincx and B. M. Weckhuysen, Angew.Chem. Int. Ed., 2013, 52, 11980.

50. S. Jia, B. J. Cox, X. Guo, Z. C. Zhang and J. G.Ekerdt, "Catalysis of Lignin Depolymerization inIonic Liquids". In Abstracts of Papers of theAmerican Chemical Society, 240 : Amer. Chem.Soc., 1155 16th st, NW, Washington, DC 20036,USA, 2010.

51. H. Zhao, J. E. Holladay, H. Brown and Z. C.Zhang, Science, 2007, 316, 1597.

52. X. Liang, G. Gong, H. Wu and J. Yang, Fuel,2009, 88, 613.

53. N. A. S. Ramli and N. A. S. Amin, BioEnergyResearch, 2017, 10, 50.

54. P. Brown, A. Bushmelev, C. P. Butts, J. Cheng,J. Eastoe, I. Grillo, R. K. Heenan and A. M.Schmidt, Angew. Chem., 2012, 124, 2464.

55. P. Brown, A. Bushmelev, C. P. Butts, J.-C. Eloi,I. Grillo, P. J. Baker, A. M. Schmidt and J.

Eastoe, Langmuir, 2013, 29, 3246.

56. S. Polarz, C. Bährle, S. Landsmann and A. Klaiber,Angew. Chem. Int. Ed., 2013, 52, 13665.

57. P. Brown, G. N. Smith, E. P. Hernández, C.James, J. Eastoe, W. C. Nunes, C. M. Settens, T.A. Hatton and P. J. Baker, J. Phys. : CondensedMatter, 2016, 28, 176002.

58. T. Khezeli and A. Daneshfar, UltrasonicsSonochemistry, 2017, 38, 590.

59. A. García Saiz, P. Migowski, O. Vallcorba, J.Junquera, J. A. Blanco, J. A. González, M. T.Fernández Díaz, J. Rius, J. Dupont and J.Rodríguez Fernández, Chemistry-A EuropeanJournal, 2014, 20, 72.

60. K. D. Clark, M. M. Yamsek, O. Nacham and J.L. Anderson, Chem. Commun., 2015, 51, 16771.

61. A. Klee, S. Prevost, W. Kunz, R. Schweins, K.Kiefer and M. Gradzielski, Phys. Chem. Chem.Phys., 2012, 14, 15355.

62. P. Brown, T. A. Hatton and J. Eastoe, CurrentOpinion in Colloid & Interface Science, 2015,20, 140.

63. J. Jiang, Y. He, L. Wan, Z. Cui, Z. Cui and P.G. Jessop, Chem. Commun., 2013, 49, 1912.

64. S. Sharma, N. Pal, P. K. Chowdhury, S. Sen andA. K. Ganguli, J. Am. Chem. Soc., 2012, 134,19677.

65. J. Xiong, W. Zhu, W. Ding, P. Wang, Y. Chao,M. Zhang, F. Zhu and H. Li, RSC Adv., 2014, 4,40588.

66. J. Dobson, Gene Therapy, 2006, 13, 283.

67. I. S. Lee, N. Lee, J. Park, B. H. Kim, Y.-W. Yi,T. Kim, T. K. Kim, I. H. Lee, S. R. Paik and T.Hyeon, J. Am. Chem. Soc., 2006, 128, 10658.

68. P. Brown, A. M. Khan, J. P. Armstrong, A. W.Perriman, C. P. Butts and J. Eastoe, AdvancedMaterials, 2012, 24, 6244.

69. L. Xu, L. Feng, S. Dong and J. Hao, Chem.Commun., 2015, 51, 9257.

70. C. J. Frégeau and A. De Moors, Forensic ScienceInternational : Genetics, 2012, 6, 511.

71. K. D. Clark, O. Nacham, H. Yu, T. Li, M. M.Yamsek, D. R. Ronning and J. L. Anderson,Anal. Chem., 2015, 87, 1552.

72. L. Xu, Y. Wang, G. Wei, L. Feng, S. Dong andJ. Hao, Biomacromolecules, 2015, 16, 4004.

1311


An overview on photoinduced bimolecular electron transfer (ET) inconstrained reaction media†

Haridas Pal

Radiation & Photochemistry Division, Bhabha Atomic Research Centre, Mumbai-400 085, India



Abstract : Understanding the kinetic and mechanistic details of the electron transfer (ET) reactions in con-strained media is an area of intense research in chemical sciences. The well known Marcus outer-sphere ETtheory predicted an interesting inversion behavior for the ET rates with reaction exergonicity (–G0), whichremains debatable for last more than fifty years for bimolecular ET reactions, though clear Marcus Inversion(MI) behaviors have been well documented for many intramolecular ET systems. For bimolecular ET reac-tions in conventional solvents, the observed reaction rate at the higher exergonicities usually reach a satura-tion limit equal to the bimolecular diffusional rate (kd), displaying the typical Rehm-Weller behavior thanthe theoretically predict MI behavior. Though many explanations have been put forward to justify the ob-served Rehm-Weller behavior for the bimolecular ET rates with reaction exergonicity, research had been con-tinuing to find out the means that can overcome the limitations associated with the bimolecular ET reactionsand can lead to appearance of the theoretically predicted MI behavior even for the bimolecular ET reactions.With this perspective and considering that the ET reactions are the most fundamental reactions occurringubiquitously in chemistry and biology, extensive studies have been pursed on bimolecular ET reactions forlast many decades to understand all the intricate details that can modulate the kinetics of such fundamentalreactions. For last about one and half decades, our extensive efforts to understand bimolecular photoinducedET (PET) reactions under various experimental conditions have made us to explore the requisite situationsfor the reaction environments that can favorably assist in realizing the MI behavior for bimolecular PET re-actions. It has been comprehended from our studies that using constrained reaction media like microherogeneoussurfactant assemblies like micelles, reverse micelles, etc., or high viscosity and slow relaxing solvents like ionicliquids, the MI behavior is observed very easily for bimolecular PET reactions, assisted jointly by ultraslowreactant diffusion and exceedingly slow solvent relaxation dynamics compared to the ET rates. In this articlewe provide an overview of various aspects of bimolecular ET reactions bringing out how the constrained re-action media help in realizing MI behavior even for such bimolecular ET reactions.

Keywords : Electron transfer reaction, Marcus Inversion behavior, constrained reaction media, one-dimensionaland two-dimensional ET models.

Introduction

Electron transfer (ET) is the most fundamentalreaction ubiquitously occurring in chemistry and bi-ology1–25. All oxidation-reduction processes occur-ring in chemistry and biology invariably involve theET reactions in some forms. In biology, ET plays avery vital role, often acting as a trigger to initiatesequences of the follow-up reactions leading to the de-

†Professor D. P. Chakraborty 60th Birth Anniversary Lecture (2016).

sired outcome of the complex biological processes17–25.Photosynthesis in green plants is the most importantexample of the ET mediated process in the MotherNature that produces food for the whole living world,through the direct utilization of the solar energy. Inchemistry as well as in chemical technology, the ETreactions play very important roles. Artificial solarenergy conversion mechanisms, photography, xerogra-


1312

phy, molecular electronics, photovoltaics, informa-tion storage, and in many such important technologicalareas the ET reactions are directly associated with theoverall chemical processes involved therein1–16,26–32.Due to such immense importance, the ET reactionshave attracted enormous research interests of the sci-entists for many decades to explore various funda-mental aspects of the ET reactions, both experimen-tally and theoretically1–67. Though ET reaction in itsliterary term implies the transfer of an electron froma donor (D) moiety to an acceptor (A) moiety, inregard to the energetics and dynamics involved in theET reactions in various D/A systems are unexpect-edly complex as the actual reaction is controlled by alarge number of energetic/dynamic factors that areyet to be understood very explicitly. Accordingly,studies on the ET reactions in various D/A systemsand different reaction environments still remain to bevery active research area in the contemporary chemi-cal and biological sciences.

For most ET systems involving organic donor (D)and acceptor (A) molecules, the transfer of an elec-tron from the ground state D to the ground state A isnot favorable from energetic considerations. Interest-ingly, however, when either of these reactants (D orA) is promoted to the excited electronic state, e.g.following photoexcitation of the molecule, the ETreaction in the concerned D*/A or A*/D pair oftenbecomes energetically very favorable. This happensbecause the excess energy provided to the D* or A*molecule through photoexcitation acts favorably toincrease the effective reaction exergonicity (-G0) forthe ET process in the D*/A or A*/D pair as comparedto that of the ET reaction involving the D/A pairwhere both the reactants are in the ground state1–44.Accordingly, studies on the photoinduced ET (PET)reactions have been used very extensively to under-stand the mechanistic and kinetic details of the ETreactions involving various D/A systems under vari-ous experimental conditions1–67.

In many organic D/A systems, the PET reactionsusually occur in the fast to ultrafast time scales, lead-

ing the ET times to appear in the nanosecond to sub-picosecond time domains1-54. Accordingly, to explorethe kinetic and dynamic details of such PET reactionsit is essential to use the short or ultra-short laser pulsesto initiate the excitation of D/A system and a simi-larly fast or ultrafast detection technique is requiredto monitor the progress of the PET reaction. Thus,fast and ultrafast fluorescence or transient absorptionbased photochemical techniques are realized to be thebest methods to investigate the kinetics of most PETreactions in organic D/A systems. In the last coupleof decades such fast and ultrafast PET studies in bothhomogeneous and heterogeneous media have beeninvestigated quite extensively, exploring many intri-cate details of the ET reactions1–67. In spite of suchextensive studies, many details of the PET reactions,especially those of the bimolecular PET reactions underconstrained reaction media, are still remained to beunderstood very clearly10–12,68–90. Accordingly studieson bimolecular PET reactions in constrained solventmedia are still the subject of intense research in con-temporary chemical sciences, aiming to arrive at acomprehensive understanding of the various factorsthat control the bimolecular ET reactions under suchtypical reaction environments that have many appliedinterests.

In general, for a PET reaction, the concerned Dand A moieties can either be covalently linked to eachother or they can be isolated molecules randomly dis-tributed in the reaction media. Accordingly, PET re-actions can either be of intramolecular or intermo-lecular (also called bimolecular) in nature. In boththe cases, if for simplicity we assume that the accep-tor moiety is photoexcited to initiate the PET reac-tion, the intramolecular and bimolecular PET reac-tions can be represented as in their simplest forms asshown in Scheme 1. Following this scheme, it is evi-dent that for intramolecular PET the experimentallyestimated rate constant (kobs) following the decay ki-netics of A* in the bound A*-D system would beexactly equal to the rate constant (ket) for the intrinsicET step. For bimolecular PET, however, the experi-

Pal : An overview on photoinduced bimolecular electron transfer (ET) in constrained etc.

1313

mentally estimated kobs following the decay kineticsof free A* cannot often be just equal to the ket for theintrinsic ET step, because, in this case the randomlydistributed free A* and D molecules are first requiredto be brought within the zone of the reaction spherethrough diffusion process to form the precursor (orencounter) complex [A*/D], and the ET reactionscan eventually take place only in these precursor/encounter complexes. Accordingly, in this case, thekobs will be a function of both diffusional rate con-stant (kd) and the intrinsic ET rate constant (ket).Moreover, for bimolecular PET reactions, the con-tact ion-pair state [A–/D+] which is formed by ETreaction in the precursor/encounter complex can sub-sequently undergo number of follow-up reactions,namely, reverse ET reaction (k–et), ion dissociationprocess (kID), charge recombination process (kCR),chemical reactions (kr), etc., and all these processescan influence the experimental kobs value, as all theprocesses are coupled to each other. It is thus evidentthat the observed kinetics of the bimolecular PETreactions would inherently be very complicated ascompared to that of the intramolecular PET reactions.

As it is understandable from the above discus-sions, the overall bimolecular PET reaction shouldbe represented better by the following explicit reac-tion Scheme 2, where intrinsic ET step in the encoun-ter complex is just an elementary step among the otherprocesses associated to the ET reaction in a coupledmanner. Obviously, exploration of the kinetic detailsfor the bimolecular ET reaction in a solution is inher-ently a very complex proposition. Further, studies onthe bimolecular PET reactions become far more com-

plex when such ET reactions are intended to be car-ried out in a constrained reaction medium, i.e. in asolvent medium where the diffusion of the reactantmolecules are quite restricted and also the distribu-tions of the reactant molecules in the medium arequite inhomogeneous, either due to the topology ofthe reaction medium or due to the slow diffusion ofthe reactants and slow relaxation of the solvent sys-tem, as compared to the time-scales of the ET reac-tions. In the following discussion we will systemati-cally present the general aspects of the ET reactions,bringing out sequentially the important features of thebimolecular PET reaction in constrained media, asthis is the topic of our concern for the present reviewarticle10–12,68–90.

Scheme 1. Representations of the intramolecular and bimolecu-lar PET reactions, in their simplest forms, consider-ing that the acceptor molecule is made in the excitedstate following photoexcitation to undergo ET reac-tion with the ground state donor molecule.

Scheme 2. Elementary steps involved in bimolecular PET reac-tions; kd is the bimolecular diffusion controlled rateconstant for encounter complex ([A*/D]) formation,k–d is the diffusion mediated dissociation constant fordisintegration of the [A*/D] complex, ket is the intrin-sic rate constant for ET in the [A*/D] complex, k–et isthe rate constant for reverse ET from the contact ion-pair state [A–/D+], and kp is the sum of all the rateconstants associated with the decay processes of [A–/D+] excluding the reverse ET step.

Theoretical background of Marcus ET theory : In-version behavior for ET rates with reactionexergonicity (–G0)

The phenomenal theoretical treatment to correlatethe rates of the outer-sphere ET reactions (occur with-out making or breaking any bond) with the concernedfree energy changes (G0) of the reactions was pro-vided by Professor Rudolph A. Marcus back in 1956which is now universally known as the Marcus outer-sphere ET theory. Considering the versatility of thistheory and finding its extreme utilizations in under-standing various ET reactions, Professor Marcus wasawarded with the Nobel Prize in chemistry in theyear 1992. Though the original Marcus ET theory

JICS-8


1314

was proposed based on the classical concept of thesimple outer-sphere ET reaction32,45,46, the concept,however, received many refinements during the courseof its long journey, incorporating both semi-classicaland quantum mechanical aspects into the backbone ofthe original outer-sphere ET theory1–10,32,45–67.

Based on the Marcus outer-sphere ET theory, ei-ther in its classical form or on incorporating the semi-classical concept into the original theory, the rateconstant ket for the intrinsic ET step in Scheme 1 shouldbe expressed in its simplified form as1–10,32,45–67,

etG*

k expRT

(1)

where the pre-exponential factor is the effectivefrequency factor for the intrinsic ET reaction, G* isthe free energy of activation, R is the universal gasconstant and T is the absolute temperature. In its ex-plicit form, the frequency factor is in fact the prod-uct of two terms, namely, the nuclear frequency fac-tor, n, and the electronic transmission coefficient,el. The term n represents the frequency at whichthe reactant state passes through the transition state(TS) region of the ET reaction and the term el rep-resents the probability by which the reactant state (R)can transform to the product state (P) as the reactingsystem passes through the TS. Though in the classi-cal form of the Marcus ET theory the parameter elwas considered to be unity, however, following thesemi-classical model of the Marcus ET theory, thevalue of el should be determined by the extent ofelectronic coupling matrix element (Vel) between theR and P states and should be given as1–10,32,45–67,

et 2el s

1

1 4 V /

(2)

where s is the time constant for the solvent relax-ation process around the reacting system and is thetotal reorganization energy associated with the con-cerned ET reaction. It is evident from eq. (2) that if

Vel between the R and P states is very strong (Vel>> 100 cm–1), el will be very close to unity, mak-ing the ET reaction to occur with an adiabatic mecha-nism. On the contrary, if Vel is significantly low (Vel<< 100 cm–1), el becomes much less than unity,causing the ET reaction to proceed following the non-adiabatic mechanism1–10,32,45–67.

In Marcus outer-sphere ET theory, it is consid-ered that along the unified reaction coordinate thefree energy of both reactant (R) and product (P) statesvaries in a quadratic manner. It is further assumedthat the force constant for the quadratic function forthe free energy of both R and P states is effectivelythe same for an outer-sphere ET reaction. Thus, con-sidering the unified reaction coordinate X as the nor-malized coordinate such that the free energy mini-mum for the R state would appear at X = 0 and thatof the P state would appear at X = 1, the free ener-gies for the R and P states (GR and GP, respectively)as a function of X can be expressed as1–10,32,45–67.

GR = X2 (3)

GP = (1 – X)2 + G0 (4)

where is the total reorganization energy of the ETreaction and it is the function of the associated forceconstant (k) for the R and P states (i.e. = k/2) andG0 is the standard free energy change between theminima of the P and R states. The free energy rela-tions for R and P states, represented by eqs. (3) and(4), can also be graphically presented as in Fig. 1 fortheir easier visualization.

Based on eqs. (3) and (4) and considering Panel Ain Fig. 1, if Xc is the crossing point of the GR and GPcurves along the X coordinate, which presents the TSfor the ET reaction, one can write1–10,32,45–67,

Xc2 = (1 – Xc)

2 + G0 (5)

By solving eq. (5) we can therefore express Xc as,

+G0

Xc = ————— (6)2

Accordingly, the expression for the free energy of


1315

activation for the ET reaction can be obtained as,2 20 0

2c

G GG* X

2 4

(7)

Substituting this expression for G* in the simplifiedform of the ket expression in eq. (1), a somewhatmore explicit expression for ket can be given as,

–(+G0)2ket = exp ——————— (8)

4RT

Since the reorganization energy is always a positivequantity, eq. (8) suggests that the ET process wouldturn out to be barrierless (i.e. G* = 0) when theexergonicity (–G0) of the reaction becomes equal to. This would naturally represent the situation for themaximum ket value possible for an ET system. Onecan now easily visualize that for the exergonicity re-gion lower than the barrierless condition, there willbe a gradual decrease in G* with the increasing

exergonicity for the ET reaction (cf. eq. (7)), leadingthe ket value to increase asymptotically with the in-creasing –G0 till one reaches the maximum ket valueat the barrierless condition. This is the region werefer as the normal Marcus region. Similarly, at theexergonicity region higher than the barrierless condi-tion, there would be a gradual increase in G* withthe increasing exergonicity (cf. eq. (7)), leading theket value to decrease gradually on increasing the–G0 beyond the barrierless condition. This is theregion we refer as the Marcus inversion region. It isthus evident from the consideration of the Marcusouter-sphere ET theory that the ket values should fol-low an inverted bell-shaped correlation curve as theexergonicity of the ET reaction is changed from thevery low to the very high values covering thebarrierless condition of the ET reaction, which canpictorially be presented as in Fig. 2 for a qualitativerepresentation.

Experimental evidences of Marcus Inversion be-havior

Though inversion behavior in the ET rates withreaction exergonicity was predicted by ProfessorMarcus in his famous outer-sphere ET theory45,46

well back in the year 1956, in spite of enormousefforts by many researchers it took almost 30 years to

Fig. 1. Quadratic free energy curves for R and P states asconsidered in Marcus outer-sphere ET theory. PanelA is shown to represent the situation involving theclassical ET model and Panel B is shown to representthe situation involving the semi-classical ET model.Different energy terms as appear in eqs. (1), (3) and(4) are indicated in the panels for their easy realiza-tion.

Fig. 2. Graphical presentation of the appearance of Marcusinversion for ET rates as the exergonicity (–G0) ofthe reaction increases. Qualitative presentation of thetype of free energy crossings of the R and P states forthe three typical situations of ET under normal re-gion, barrierless condition and inversion region arealso shown in the figure.


1316

realize this unique Marcus Inversion (MI) behaviorexperimentally for the first time by Miller and co-workers102 in the year 1984. These authors convinc-ingly demonstrated the MI behavior for the first timefollowing the intramolecular charge shift reactions inthe pulse radiolytically produced radical species inthe thoughtfully designed donor-spacer-acceptor (D-S-A) kind of bi-functional chemical systems suitablefor intramolecular ET reaction (cf. Scheme 1). Theexemplary D-S-A kind of bi-functional systems usedby Miller and coworkers102 in their seminal work onMI behavior are shown in Scheme 3. Subsequently,the same research group extended their studies onintramolecular ET reactions involving other moreextended series of D-S-A systems, reconfirming ex-perimentally the presence of MI behavior for intramo-lecular ET reactions103.

Scheme 4. Following these initial reports on the ex-perimental demonstrations of the MI behavior for theET reactions, subsequently there have been manyextensive follow up studies involving different donor-acceptor systems and following various experimentaldesigns, with the aim to realize the MI behavior andalso to understand diverse factors that play the role infacilitating or suppressing the MI behavior for the ETreactions under various experimental conditions105–126.

Scheme 3. The D-S-A kind of chemical systems used by Millersand coworkers102 in their study demonstrating MIbehavior for the first time experimentally using in-tramolecular charge shift reactions in the pulseradiolytically produced radicals in these chemical sys-tems.

Scheme 4. The porphyrin-quinone based intramolecular ET sys-tems used by Wasielewski and coworkers104 in theirstudy demonstrating the MI behavior experimentallyinvolving intramolecular charge recombination reac-tions in the photoinduced radical ion-pairs in thesechemical systems.

Convincing experimental demonstration of MIbehavior for intramolecular ET reaction was also re-ported in the subsequent year by Wasielewski andcoworkers104, following the photoinduced radical ion-pair generations and subsequently monitoring theirrecombination reactions in a series of porphyrin-quinone based intramolecular ET systems with fixed-distance donor-acceptor centers, as are shown in

An interesting observation in regard to the MIbehavior observed experimentally from many exten-sive studies is the asymmetric shape of the constructedbell-shaped curves1–11,32,102–126, though from eq. (7)and (8), as given by Marcus ET theory45,46, the in-version behavior should lead to a very symmetric in-verted bell-shaped correlation. This aspect is qualita-tively demonstrated in Fig. 3 for just a schematicpresentation of the facts. To be mentioned that theasymmetry in the experimental MI curves alwaysappeared due to the less stiff and somewhat non-qua-dratic bending of the curves in the “Marcus InvertedRegion” compared to the quite quadratic bending ofthe curves in the “Normal Marcus Region”, as alsoqualitatively indicated in Fig. 3.


1317

A number of reasons have been put forward torationalize the asymmetric nature of the experimentalMI plots. One of these reasons is the involvement ofthe high frequency (h >> kT) vibrational modes inmodulating the ET reactions, which occur especiallyat the higher exergonicity region, because in this re-gion the energy released from the ET reactions issufficient enough to excite the high frequency vibra-tional modes of the product (P) state formed duringthe progress of the ET reaction1–10,32,45–67. Otherreasons suggested for the observed asymmetry in theexperimental MI plots are the donor-acceptor separa-tion dependent electronic coupling element (Vel) andsolvent reorganization energy (s) for the concernedET systems, especially at the higher exergonicity re-gion1–16,32,26–42,52–67,102–120. The rationale behind theconsideration of influence of the parameters Vel ands towards the asymmetry in the MI plots is that bothof these parameters are strongly dependent of the sepa-ration between the interacting donor-acceptor pairs inthe reaction sphere, as given by the eqs. (9) and (10),respectively, and the fact that the average size of thereaction sphere gradually decreases on increasing thereaction exergonicity1–16,32,26–42,52–67,102–120.

2 2el,r el,0 ADV V exp{ (r )}

(9)

2

sop s A D AD

e 1 1 1 1 12 r r r

(10)

where rAD is the separation between the D and Acenters in the reaction sphere, is the contact dis-tance between D and A, rD and rA are the radii of theD and A molecules, respectively (considered to be aneffective sphere), op and s are the high frequencyand static dielectric constants of the solvent, respec-tively, and e is the charge of an electron. It is real-ized that at very high exergonicity region the donor-acceptor pairs in the reaction sphere having smallerseparations contribute more to the ET reaction thanthose with higher separations. Accordingly, incorpo-rating the effects of Vel and s on the ET kineticseffectively causes an enhancement in the ET rates asotherwise expected without considering these effectsof the D-A separations on the observed ETrates1–16,102–120.

Various type of ET reactions for which MarcusInversion behavior have been observed experimen-tally without any ambiguity are : (i) intramolecularET reactions between covalently linked donor andacceptor moieties (often with a suitable spacer in be-tween the donor and acceptor enters)1–

16,32,35,36,53,113–134 and (ii) charge recombination(CR) reactions taking place in the contact radical ion-pairs subsequent to their formation by a precedingPET reaction102–126. For bimolecular or intermolecularET reactions, however, observed reaction rates (kobs)in conventional homogeneous media hardly show theexpected Marcus inversion behavior. In these casesthe most common observation is that the kobs valueincreases initially with –G0 at the lower exergonicityregion (normal region) but it saturates finally to alimiting value, typically to the bimolecular diffusioncontrolled rate constant (kd; cf. Scheme 2), at thehigher exergonicity region, without showing any in-version behavior1–16,32,35,36,53,113–138. Typical ofsuch behavior is commonly known as the Rehm-Wellerbehavior, after the scientists D. Rehm and A. Weller,who observed such a behavior for the first time in theyear 1970 for bimolecular ET reactions in a homoge-neous solution127. Typical feature of the Rehm-Wellerbehavior in the kobs vs G0 plot is schematically shownin Fig. 4 along with the otherwise expected Marcus

Fig. 3. Qualitative presentation of the asymmetric nature ofthe experimentally observed Marcus inversion curvesas compared to the theoretically predicted symmetricbell-shaped curves.


1318

inversion behavior that remains obscured for bimo-lecular ET reactions in conventional homogeneoussolvents.

Following the reaction steps involved in bimolecularET reactions, as shown in Scheme 2, the experimen-tally observed rate constant, kobs, as would be esti-mated following the decay kinetics of isolated A*molecules, should be expressed in its simplified formas,

d dobs

d d

et eq et

k kk

k k1 1

k K k

d

d

eq

k

k 11

K exp ( G*/RT)

(11)

where Keq = (kd/k–d), the equilibrium constant forthe precursor/encounter complex formation and isthe pre-exponential factor, having the same meaningas discussed with respect to eq. (1), given earlier.Considering eq. (11) and observing that kobs valuesfinally saturates to the kd value at higher exergonicityregion, Rehm and Weller proposed an empirical ex-pression for G* in terms of the G0 and values,as given by eq. (12), which is distinctly different thanthe quadratic relation proposed by the Marcus ETtheory (cf. eq. (7)).

1/22 20 0G GG*

2 2 4

(12)

As suggested from eq. (12), on increasing the reac-tion exergonicity (–G0), the G* value will gradu-ally decrease asymptotically tending toward zero,which can be pictorially represented as in Fig. 5A.Thus, following eqs. (11) and (12), the kobs valuewill gradually increase in an asymptotic manner withthe increasing exergonicity of the reaction, tendingtoward the saturation limit of the kd value at the higherexergonicity region, as pictorially represented in Fig.5B. That the bimolecular ET rates with the reactionexergonicity are usually seen to follow the typical

Fig. 4. Qualitative presentation of the Rehm-Weller behav-ior for the observed bimolecular ET rates in homoge-neous solutions. Expected inversion at very highexergonicity region is also qualitatively indicated.

Fig. 5. Conceptual presentations of (A) changes in the G*with reaction exergonicity (–G0) and (B) the varia-tions of kobs values with reaction exergonicity as gov-erned by Rehm-Weller relations (eqs. (11) and (12))for bimolecular ET reactions in polar solvent media.


1319

Rehm-Weller behavior than the theoretically predictedMarcus Inversion behavior, this aspect had been thesubject of great concern to the researchers for manydecades. This had been especially so because quiteconvincing Marcus Inversion behaviors could be ob-tained experimentally not only for a large number ofintramolecular ET systems but also for the chargerecombination reactions in various contact ion-pairs1–16,32,35,36,53,102–134, though the same inversion be-havior remained always obscured for bimolecular ETreactions in conventional solvents. Accordingly, thelong lasting quest had been, what could be the factorsthat actually resist the observation of the Marcus in-version behavior for bimolecular ET reactions in thesolution phase and is there any possibility to over-come these influences by some suitable experimentaldesigns so that the Marcus Inversion behavior canalso be observed easily for the bimolecular ET reac-tions as well?

Though observation of Marcus inversion behavioris still a rare finding for bimolecular ET reactions, itis however, quite convincingly understood from ex-tensive theoretical and experimental studies that irre-spective of the trends observed for the kobs vs G0

plots, the observed results can be justifiably rational-ized within the framework of the Marcus outer-sphereET theory. Accordingly, Marcus ET theory has beenrealized as the most versatile theoretical basis to ac-commodate all the different findings involving theET reactions, albeit with the incorporations of suit-able refinements, as may be required based on theexperimental conditions. Therefore, in the expressionof kobs for the bimolecular ET reactions, as given byeq. (11), the term G* would better be considered asgiven by eq. (7) in the Marcus ET theory than eq.(12) given empirically by Rehm-Weller. Accordingly,one can expect that for the lower reactionexergonicities, where –G0 << , the G* will bequite high to make the intrinsic ket much lower thankd. In this situation, precursor/encounter complexesare formed faster than the intrinsic ET process, mak-ing the kobs effectively similar to ket (cf. eqs. (8) and(11)). Similarly for the exceedingly higher exergonicity

region, where –G0 >> , the G* will again be-come quite high, making kobs once gain effectivelysimilar to ket (cf. eqs. (8) and (11)). For the interme-diate reaction exergonicities, however, G* being rea-sonably small, the ket value will be high enough suchthat ket >> kd. In these situations, therefore, kobswill be effectively determined by the formation rate(kd) of the encounter complex than the intrinsic ETrate. The typical of these situations, as would be re-flected in the kobs vs G0 correlation curves, areconceptually shown in Fig. 4 along with the concep-tual presentation of the predicted Marcus inversionbehavior and the often encountered Rehm-Weller be-havior observed for most of the bimolecular ET reac-tions conducted in conventional homogeneous solvents.To be mentioned, however, that the involvement of thehigh frequency vibrational modes1–11,32,47–53,63–67 andthe reaction exergonicity dependent changes in theVel and value1–11,32,115–120 (cf. eqs. (9) and (10)),as we have discussed earlier, can often cause the ETrates at higher exergonicity region to be much higherthan expected from the simple quadratic consider-ation of G* expression and thus can move the ex-pected inversion region to much higher exergonicities,causing the Rehm-Weller behavior to be the onlyfeature observed for such bimolecular ET reactions.

Bimolecular electron transfer kinetics studiedthrough fluorescence quenching measurementsunder diffusive and non-diffusive conditions

In most of the bimolecular PET studies in homo-geneous solutions, conventional fluorescence quench-ing measurements are carried out using very lowfluorophore (F) concentration (typically M range),where reasonably lower quencher (Q) concentrations(typically tens of mM range) can be sufficient to giveenough fluorescence quenching to extract the requiredkobs values as the bimolecular quenching constants(kq), defined by the Stern-Volmer quenching kinet-ics127–139, without involving any transient effect (in-stantaneous or ultrafast quenching in the preexistingclose proximity F-Q pairs, without involving diffu-sion) in the quenching process. In these experiments,


1320

the quencher concentrations are still being much higherthan the excited fluorophore concentration, the PETreaction can satisfactorily be considered as a pseudo-first process127–139. Therefore, the observed decaykinetics for the excited F* in the absence and in pres-ence of Q as one would measure from time-resolved(TR) fluorescence studies can be expressed, respec-tively, as,

F* = F*0 exp{–(0–1)t} (13)

FQ* = F*0 exp{–(0–1 + kq[Q])t}

= F*0 exp{–(Q–1)t} (14)

where 0 and Q = {–(0–1 + kq[Q])–1 are the fluo-

rescence lifetimes of the excited F* in the absenceand in the presence of Q. Therefore, in these cases,the standard Stern-Volmer relation from TR fluores-cence quenching results can be expressed as127–139,

0—— = 1 + kq0[Q] (15)Q

To be mentioned here, in regard to the PET reac-tions, the F* here can represent the excited acceptor(A*) while the Q will represent the ground state do-nor (D), as we have discussed earlier. Similarly, onecan also consider F* to represent a D* and Q torepresent a ground state A, if the PET reaction in-volves the excited state of the donor and the groundstate of the acceptor molecules.

As eq. (15) suggests, the 0/Q vs [Q] plot shouldfollow a linear correlation. This linearity is invari-ably observed when quenching process occurs strictlyunder diffusive condition (cf. Scheme 2) and such asituation can be maintained beyond doubt when thequencher concentrations used are reasonably low, typi-cally in tens of mM range127–139. As Scheme 2 andeq. (11) suggest, under diffusive condition, the kq (=kobs) value will reach its maximum limit to kd whenET reaction becomes diffusion controlled (because kd<< intrinsic ket), and this is the situation that leadsto the Rehm-Weller behavior127 discussed earlier. Inthe conventional low viscosity homogeneous solvents,the typical kd value is in the range of about 2×1010

M–1 s–1 127–139. Therefore, for a representativefluorophore for which fluorescence lifetime (0) is

typically about 5 ns, a 50 mM quencher concentra-tion can effectively reduce its lifetime (Q) to as lowas about 0.83 ns, a reduction large enough for aquenching study exclusively under diffusive condi-tion (cf. Scheme 2), completely avoiding any tran-sient effect in the observed quenching kinetics. How-ever, if the quencher concentration in the experimen-tal solution is made substantially higher, say in themolar range, the observed quenching kinetics in thesecases will invariably possess a substantial extent oftransient effect and the quenching process can nolonger be considered solely under diffusive conditionbecause in these situations the non-diffusive effectswill also be associated with the observed quenchingprocess. To be noted here that under completely dif-fusive condition, the quenching of the steady-state(SS) fluorescence intensity of F* will also follow alinear Stern-Volmer relation, which can be expressedin this case as127–139,

I0—— = 1 + kq0[Q] (16)IQ

where I0 and IQ are the fluorescence intensities in theabsence and in presence of Q, respectively. Mostimportantly, under completely diffusive situation theeq. (16) becomes just equal to eq. (15), i.e.127–139,

I0 0—— = —— = 1 + kq0[Q] (17)IQ Q

Thus, validation of the equality of the time-resolve(0/Q) and steady-state (I0/IQ) quenching kinetics asgiven by eq. (17) is an important criterion to ascer-tain that a bimolecular quenching process actuallyoccurs under a completely diffusive condition with-out involving any transient effect.

In conventional low viscosity homogeneous sol-vents, maintaining the situation represented by eq.(17) for bimolecular ET reaction is quite easy be-cause in these solvents kd value is quite large(~2×1010 M–1) such that fairly substantial extent offluorescence quenching can be achieved just by usingfew tens of mM quencher concentration. However,when such a bimolecular ET reaction is carried out ina very high viscosity solvent or in a constrained mi-


1321

cro-heterogeneous medium, where kd value in inher-ently very low, use of very high concentrations of thequencher becomes unavoidable to achieve a reason-able extent of fluorescence quenching for a meaning-ful kinetic analysis to extract the required quenchingrate constant. In these cases, obviously, a significantextent of preexisting close proximity F-Q pairs willbe present in the system that can undergo instanta-neous or ultrafast quenching process followingphotoexcitation, without involving any diffusion ofthe reactants and accordingly a substantial extent oftransient effect will be present in the observed quench-ing kinetics for these systems. Understandably thus,for these ET systems the linear Stern-Volmer rela-tionships, as given by eqs. (15), (16) and (17), maynot be truly applicable. Accordingly, in these cases,the ET kinetics to be dealt with a quite different ap-proach than using the diffusion reaction kinetics asdepicted in Scheme 2 and expressed by eq. (11).

As we understand from the discussion above, theET reactions in very high viscosity solvents or inconstrained micro-heterogeneous media would invari-ably occur either under non-diffusive condition orunder a restricted diffusion condition, such that thetransient effect in the observed quenching kineticsbecomes inevitable. Interestingly, in these cases, theobserved bimolecular ET kinetics are expected to dis-play the Marcus Inversion type of behavior quite eas-ily, because the quenching components arising fromtransient effects in the preexisting close-proximity F-Q pairs would effectively be equivalent to the in-tramolecular kind of ET reactions as occur in thechemically bound D-A systems, for which Marcusinversion behavior is a quite common observation, asin these cases the kobs values become the direct mea-sures of the intrinsic ket rates (cf. Scheme 1 and re-lated discussion). To be mentioned here that the ETreactions under the so-called restricted diffusion con-ditions are directly relevant to many ET processesinvolved in the biological systems, e.g. in photosyn-thesis, cellular respiration, redox-mediated enzymecatalysis, and so on1–32. Moreover, ET reactions un-der constrained conditions are also having relevanceto various applied areas like information storage, solid

state electronics, molecular electronics, sensing, ca-talysis, solar energy conversion, photovoltaic, bio-technology, etc.1–32,57–67. Considering all these, theET reactions under non-diffusive or restricted diffu-sion conditions are the interesting and important re-search topics and quite extensive studies have beencarried out on such ET reactions to understand theinsights of these reactions, aiming their desired con-trols and efficient utilizations. In the forthcoming sec-tions we will discuss different situations of the bimo-lecular ET reactions in constrained reaction media,imposing the situations of non-diffusive or restricteddiffusion conditions for the observed ET reactions,bringing out the intriguing features of these reactionsthat are never observed for bimolecular ET reactionsin conventional low viscosity homogeneous solvents,investigating the reactions exclusively under the dif-fusive conditions (cf. Scheme 2).

Non-diffusive conditions for bimolecular electrontransfer reactions in micro-heterogeneous media

PET reaction in constrained micro-heterogeneousmedia, like micelles, reverse micelles, vesicles, etc.have attracted considerable research interests for lastcouple of decades, because of the structural resemblesof these assemblies with various biological mem-branes77–90,140–163. It is generally perceived that suchconstrained reaction media can help in lingering thelifetime of the radical ion pairs produced in the for-ward PET reactions quite significantly, by preventingor retarding the back ET or charge recombinationreactions, and thereby proving the primary species ofthe PET reaction to have longer survival times tomake them available for the desired follow-up reac-tions related to their applications in solar energy con-versions, photovoltaics, biotechnology, informationstorage, solid state electronics, molecular electronics,sensing, catalysis, and many others1–32,40–44,140–163.Among different micro-heterogeneous media, themicelles and reverse micelles have attracted the mostattentions of the researchers because these systemscan be easily formed by using appropriate surfactantsolutions and most of the organic donor (D) and ac-ceptor (A) molecules show their preferential solubil-ity inside these self-assembled systems than in the

JICS-9


1322

bulk aqueous phase, making the desired PET reac-tions to be carried out quite efficiently within theseassemblies. The investigations of the PET reactionsin micro-heterogeneous media are mainly aimed tounderstand the effect of the topology of these mediaon the mechanism and dynamics of the ET reactionsand also to explore the possible controls of the ETkinetics through the use of simple tuning methodolo-gies of these self-assembled systems. In the forth-coming discussions, we will bring out several intrigu-ing results that have emerged for the PET studiesin such constrained micro-heterogeneous media,especially in regard to the theoretically predictedMarcus inversion behavior for the bimolecular ETreactions77–90,140–163.

For the last about one and half decades, numberof research groups, including ours, have carried outextensive studies on bimolecular PET reactions invarious self-assembled micro-heterogeneous media tounderstand the effect of such self-assembled systemson the energetic and kinetics of the ET reac-tions10,69,77–90,140–163. In these micro-heterogeneousmedia, the reactant molecules, i.e. D and A, are mainlysolubilized within a small region of the self-assembledsystems, e.g. within the Stern layer or palisade layerof the micelles, where the surfactant molecules en-tangle the D and A molecules strongly retarding theirdiffusion process. Accordingly, the mobility/diffusionof these reactant molecules inside these micro-hetero-geneous media becomes highly restricted. In fact, fromour elaborate studies, it has been realized that thediffusion of the reactant molecules within these self-assembled micro-heterogeneous media is so drasti-cally retarded that the bimolecular PET reactions inthese cases effectively occur under a non-diffusivecondition10,69,77–80,85–88,140–145. In other words, theoverall PET reactions in such systems mainly occurwithin those D-A pairs that are already preexistingwithin the reaction sphere and thus acting as the in-stantaneous encounter complexes immediately afterthe photoexcitation process. Therefore, in these cases,the bimolecular PET reactions effectively behave likethe intramolecular PET reactions that occur betweenthe bound donor and acceptor moieties in a single

molecule. Thus alike the intramolecular PET reac-tions, where Marcus Inversion behavior is easily ob-served, a similar situation is also expected for thebimolecular PET reactions in the self-assembled mi-cro-heterogeneous media to display the Marcus In-version behavior easily, as really have been observedin number of studies reported by various groups in-cluding ours10,69,77–90,140–163.

In conventional low viscosity homogeneous sol-vents, as the rate of the solvent relaxation is extremelyfast, the reactant state (R) involved in a PET reactionalways maintains a thermal equilibrium along the sol-vent relaxation coordinate, which is the effective re-action coordinate (X) in these cases (cf. Fig. 1). Inthese media, therefore, the PET reaction follows aconventional one dimensional ET (1DET) model, asrepresented in Fig. 1, and in this case the intrinsic ETrate constant is represented by eq. (8), which in moreexplicit form can be represented by eq. (18) under anon-adiabatic condition (Vel << 100 cm–1) and byeq. (19) under an adiabatic condition (Vel >> 100cm–1)1–16,45–67,91–101, as discussed earlier with ref-erence to eqs. (1) and (2).2 0 2

elet

BB

V ( G )2k (NA) exp

4 k T4 k T

(18)

0 2

ets B B

( G )1k (ad) exp

16 k T 4 k T

(19)

In the micro-heterogeneous media, as the overall sol-vent relaxation rate is unexpectedly slow164–175, it isexpected that unlike in the 1DET model, the R statein these system cannot maintain a thermal equilib-rium along the solvent relaxation coordinate duringthe ET reaction. For example, in the PET reactions,the R state, which will be initially produced with annon-equilibrium solvent reorganization followingphotoexcitation (because the dipolar character in theexcited state is different than in the ground state),cannot attend the thermal equilibrium along the sol-vent coordinate X in a micro-heterogeneous media,because the solvent relaxation in these media occursvery slowly. Accordingly, in these cases, the solvent


1323

coordinate (X) cannot be considered as the effectivereaction coordinate for the ET reactions. Unlike thenon-equilibrium situation that persists in these casesalong the solvent coordinate X, no such non-equilib-rium condition is imposed to the R (or P) state alongits intramolecular coordinate q, as the intramolecularreorganization involving low frequency vibrationalmodes of the reacting systems occur very fast176–188.Accordingly, the R state would immediately get ridof any unstable situation introduced along q coordi-nate during its formation through photoexcitation andquickly reestablish the thermal equilibrium along thiscoordinate and will maintain this equilibrium through-out the progress of the ET reaction176–188. There-fore, in these cases, instead of the solvent X coordi-nate, the intramolecular q coordinate would effec-tively be considered as the reaction coordinate for theET process. In other words, it is expected that a PETreaction in a constrained micro-heterogeneous me-dium would effectively occur following a 2DET

model176–187, which can be presented schematicallyby the use of contour diagrams as shown in Fig. 6and in terms of actual free energy diagrams of the Rand P states as shown in Fig. 7.

Fig. 6. The contours for the reactant and product free ener-gies with the changing X and q coordinates as can beconceptually presented for the 2DET model. The lineCC in the contour diagram represents the curve pass-ing through the crossing points of the reactant andproduct free energy surfaces and thus corresponds tothat of the free energy of activations for the ET reac-tions along q for different X values.

Fig. 7. Schematic presentation of 2DET following the free energies of the R and P states. Initial R is at X = –Xg (Free energycurves for excited state at the right part) following photoexcitation of the ground state. Neglecting solvent relaxationthis R state continues to oscillate along X axis (cf. double sided curved arrow in the excited state) keeping the excessenergy of solvent destabilization (sXg

2) in the R state unchanged, while the ET reaction proceeds along the intramo-lecular coordinate q (reaction coordinate; cf. upper blue curve in the left part). If the ET reaction involved a fastrelaxation solvent, the system would have relaxed completely along X to X=0 (cf. lower blue curve in the left part) andthe reaction would have occurred following the conventional 1DET route.


1324

According to the 2DET model, followingphotoexcitation, the R state for the PET reaction wouldinvariably be formed with a non-equilibrium X coor-dinate (i.e. X 0) and the system would maintain thisnon-equilibrium situation all along the ET reaction,because relaxation along the X coordinate is very slow.Thus, following 2DET model, the ET reaction is com-pelled to proceed along the fast relaxing q coordi-nate, keeping the excess energy of the R state due toits non-equilibrium salvation along X coordinate (X 0) unchanged during the progress of the ET reaction.With this model, thus, it is evident that even if thesolvent relaxation is completely frozen, the ET reac-tion can still occur involving the fast relaxing q coor-dinate to assist the ET reaction. In this 2DET model,since both q and X coordinates are explicitly used tomodel the ET reaction, the free energies of the R andP states would also be given explicitly in terms of thetwo coordinates and considering these coordinates intheir normalized forms the X and q dependent GRand GP for the reactant and product states can beexpressed as176–187,

GR = sX2 + iq

2 (20)

GP = s (1–X)2 + i (1–q)2 +G0 (21)

Further, since the ET reaction effectively occurs in-dependently along the fast relaxing q coordinate forall the possible non-equilibrium distributions of R statealong X coordinate (cf. Figs. 6 and 7), the effectiveET rate (ket,eff) will be the sum of all the X-depen-dent ET rates weighted by the non-equilibrium distri-

bution () along X, i.e. ket,eff = ket (X)(X). Inthese situations, considering a non-adiabatic condi-tion (Vel << 100 cm–1) to be applicable for the ETreaction along q, the rate constant ket (X) at any Xcoordinate would be expressed as176–187,

2el

eti B

V2k (NA,X) exp

4 k T

0 2i s s

i B

( G 2 |X|)4 k T

(22)

It is quite understandable from the above consider-ations that with the non-equilibrium distributions of R

state along X coordinate, as is the situation for the2DET reaction, the effective ET kinetics for the sys-tem would be strongly non-exponential in nature176–187.As clearly evident from eq. (22), for any nonzero Xvalue, say X = Xg 0, the effective free energy ofactivation for the ET reaction that occurs along qcoordinate (cf. Figs. 6 and 7) would be given as,

(G0 +i + s – 2s|Xg|)2

G* (Xg) = ——————————————— (23)4i

To be noted that this expression for G*(Xg) follow-ing 2DET model is distinctly different than the G*

expression (cf. eq. (7)) applicable for the ET rategiven by eq. (18) following 1DET model, where thetotal reorganization term is explicitly given as =i + s. To be noted here that in eq. (23) we havedeliberately designated the non-equilibrium solventcoordinate as Xg, considering that the R state for thePET reaction would be produced with this non-equi-librium X coordinate immediately after photoexcitation,because the corresponding ground state D-A encoun-ter pair before its excitation was in the solvent equili-brated condition with its free energy minimum at X= Xg.

Following the 2DET model, since the R state can-not attain and maintain a thermal equilibrium along Xduring the ET reaction along q (cf. Figs. 6 and 7; eq.(23)), the effective free energy of activation G*(Xg)involved in these cases incorporate only a part of s,which can be considered as the effective solvent reor-ganization energy, s

eff, and would be expressed as,

seff = s – 2s|Xg|, (24)

It is evident from eq. (24) that the sff for the 2DET

reaction would be much lower than the total s in-volved in determining the G* for a typical 1DETreaction (cf. eqs. (1), (7), (18) or (19)). For the 2DETreaction176–188, since the onset of the Marcus Inver-sion is expected to start at an exergonicity of –G0 =i +s

eff, while for the 1DET reactions1–16,45–67,91–101

the Inversion is expected to start at –G0 = i +s,it is evident that for the similar D-A systems the MarcusInversion for 2DET reaction mechanism (e.g. in con-


1325

strained reaction media) would appear at a relativelylower exergonicity as compared to that happens un-der the 1DET reaction model (e.g. in fast relaxinghomogeneous solvent media)10,69,77–80,85–88,140–145.Therefore, in the 2DET reaction model, since theexergonicity for the onset of Marcus Inversion isshifted favorably towards lower exergonicity (cf. eqs.(23) and (24)), it is quite understandable that the con-strained micro-heterogeneous media effectively pro-vides a conducive reaction environment for the do-nor-acceptor systems to realize the Marcus invertedregion quite easily than in the fast relaxing conven-tional homogeneous solvents, because in the lattercase one needs to have much higher exergonicity forthe reaction to achieve the Marcus inversion behav-ior10,69,77–80,85–88,140–145.

In the context of the aforementioned aspects, ourfirst study on bimolecular PET reactions in constrainedmedia was carried out in sodium dodecyl sulphate(SDS) micelles involving a series of coumarin dyesas the electron acceptors and N,N-dimethylaniline(DMAN) as electron donor and thereby we convinc-ingly demonstrated the appearance of Marcus Inver-sion behavior140, though for the similar coumarin-DMAN systems in conventional solvent like acetoni-trile the observed ET rates followed the typical Rehm-Weller behavior131–133. Following this initial workfrom our group, quite extensive follow-up studies havebeen reported in the literature on bimolecular PETreactions in microheterogeneous media and invari-ably in all the cases the appearance of Marcus Inver-sion behavior has been documented quite convinc-ingly140–156. Interesting to mention that in all thesestudies the overall findings are very similar, thoughin rationalizing the observed facts different researchgroups have put forward somewhat different explana-tions. Understandably thus, a consolidated reasoningfor the easy appearance of Marcus Inversion behav-ior for bimolecular PET reactions in constrained mi-cro-heterogeneous media is yet to be arrived at, tojustify all the intricacies involved in such ET reac-tions.

Many of the initial investigations on bimolecularPET reactions in microheterogeneous media were

carried out following simple steady-state (SS) fluo-rescence intensity quenching and time-resolved (TR)fluorescence quenching kinetics studies77–90,140–151.All these studies have invariantly found out that theSS fluorescence intensity quenching always lead tothe non-linear Stern-Volmer plots, displaying a posi-tive deviation from the expected linearity from eq.(16). Typical of such plots are shown in Fig. 8. Tobe mentioned here that for the Stern-Volmer plots inFig. 8 the quencher concentration has been consid-ered as [Q]eff, which is the calculated quencher con-centration in the Stern layer or palisade layers of themicellar system, assuming that the fluorophores andquenchers are mainly solubilized within the Stern/palisade layers of the micelles. Accordingly, the [Q]effvalues were estimated from all the known param-eters77–80,85–88,140–145 as given by the following equa-tion.

agg t

efft

N [Q][Q]

VL [S] cmc

(25)

where Nagg is the aggregation number of the surfac-tant (S), [S]t is the total surfactant concentration used,cmc is its critical micellar concentration for the sur-factant solution, VSL/PL is the volume of the micellar

Fig. 8. Typical SS fluorescence quenching results showingpositive deviations in the Stern-Volmer plots for bi-molecular ET reactions in a micro-heterogeneousmedia. Data in this figure correspond to the bimo-lecular PET studies in Triton-X100 micelles involv-ing N,N-diethyl-p-toluidine (DMPT) as the electrondonor and (i) C151, (ii) C481 and (iii) C153 as theelectron acceptors141.


1326

Stern/palisade layer per mole of the micelle and [Q]tis the total quencher (amine) concentration as esti-mated/used for the whole surfactant solution.

From these SS fluorescence intensity quenchingresults it is quite evident that the typical diffusionalquenching scheme, as represented earlier by Scheme2 for bimolecular ET reactions in conventional ho-mogeneous solvents, is not quite applicable for thereactions in the micro-heterogeneous media. It isclearly indicated from these results that the constrainednature of the micro-heterogeneous reaction mediaimposes the PET reactions to occur with the involve-ment of significant extent of the transient quenchingeffect, which is otherwise not the case for the reac-tions in conventional low viscosity homogeneous sol-vent media128–139. In the concerned micro-heteroge-neous media, it is expected that the diffusion of thereactants would be largely retarded due to entangle-ment of the reactants with surfactant chains. Thus, inthese cases the PET reactions are logically to be con-sidered to occur effectively under a non-diffusive re-action condition. Therefore, in these cases, the bimo-lecular PET reactions are expected to occur mainlythrough the photoexcitation of the preexisting reason-ably close-proximity F-Q pairs for which the centreto centre separations between the F and Q moietiesare within the dimension of the possible reaction spherefor the ET system, making these PET reactions quiteequivalent to the intramolecular (or unimolecular) PETreactions that occur in the chemically bound D-A sys-tem (e.g. Schemes 3 and 4). As one can thus visua-lize, for the PET reactions in constrained micro-het-erogeneous media, a distribution of the preexistingD-A pairs, differing in regard to the separations be-tween the interacting D and A moieties, but still withinthe dimension of the reaction sphere, will participatein the ET reactions following the photoexcitation pro-cess, without involving any mutual diffusion of theinteracting D and A units in the timescale of the ETreactions. Obviously the preexisting D-A pairs withsmaller D-A separations will undergo a relatively fasterET reaction than those with a comparatively largerD-A separations. Therefore, considering the fact that

the population of the D-A pairs having smaller sepa-rations will gradually increase on increasing thequencher concentration in the solution, it is quite ex-pected that the fluorescence intensity quenching willgradually become more efficient with the increasingquencher concentration in the micro-heterogeneousmedia, leading to the quenching process to show apositive deviation in the Stern-Volmer plots, as typi-cally shown in Fig. 8.

In regard to the TR fluorescence quenching kinet-ics studies also, the bimolecular ET reactions in mi-cro-heterogeneous media display quite different char-acteristics than in the homogeneous solutions. Thus,it is interestingly observed that in micro-heteroge-neous media the fluorescence quenching kineticsalways show a non-exponential kinetic behavi-or77–80,85–88,140–145, which is quite different than thesingle-exponential quenching kinetics usually observedfor the similar PET reactions in low viscosity homo-geneous solvents (cf. eq. (14)). Typical non-expo-nential fluorescence quenching kinetics as observedfor bimolecular PET reactions in constrained micro-heterogeneous media following either time-correlatedsingle photon counting (TCSPC) measurements withsub-nanosecond time-resolution or femtosecond fluo-rescence up-conversion measurements with sub-pico-second resolution are shown in Figs. 9A and B, re-spectively, for their quick visualization.

Following the aforementioned discussions, it isevident that the observed non-exponential kinetics forbimolecular PET reactions in constrained micro-het-erogeneous media arises mainly due to the unusuallyslow or non-diffusion nature of the concerned reac-tions such that the positions of the interacting D andA units are hardly changed during the progress of thereaction77–80,85–88,140–145. In these situations, as dis-tribution of the interacting D-A pairs for which theD-A separations are within the reaction sphere willundergo the ET reactions, and since the rates of thesereactions are a strong function of the D-A separations(cf. eqs. (9), (10) and (22)), the observed PET kinet-ics in these cases would necessarily show a highly


1327

non-exponential behavior. Obviously thus, a compre-hensive analysis of the kinetic traces under the presentsituation is a very complicated task. Considering allthese, in these cases, the experimentally observed fluo-rescence kinetic traces are in general fitted followinga multi-exponential decay function and the decay pa-rameters thus obtained are used to discuss the corre-lations between the kinetics and energetic of such ETreactions in constrained micro-heterogeneous media.

For the TR fluorescence studies as carried out usingthe TCSPC measurements having typically about sub-nanosecond time resolution, the kinetic traces for mostPET reactions in constrained micro-heterogeneousmedia are in general found to fit quite satisfactorilyfollowing a bi-exponential function77–90,140–151 andthe decay parameters thus estimated are often used toestimate the average lifetime (av) values for the fluoro-phore in the presence of varying quencher concen-trations using the following relation77–80,85–88,140–145,

av i i1

B100

(26)

where Bi is the percentage contribution and i is thelifetime of the i-th decay component of the fitted de-cay. Subsequently, the reduction in the av valueswith the quencher concentration used are correlatedfollowing the standard Stern-Volmer relationship,which for the present situations can be expressedas77–90,140-151,

0—— = 1 + kq[Q]eff (27)av

The purpose of such Stern-Volmer correlation in thepresent cases is to estimate the average values of theobserved reaction rates in the form of the bimolecularquenching constants (i.e. kobs = kq), which are con-sidered as the measure of the effective rate constantsfor the observed PET reactions in the micro-hetero-geneous media.

Important to mention here that the above Stern-Volmer type of analysis of the av values estimatedfrom the TR studies using TCSPC measurements withthe typical sun-nanosecond time resolution is almostinvariably found to follow a linear correlation77–80,85–

88,140–145, as defined by eq. (27). Typical such linearo/av versus [Q]eff plots are shown in Fig. 10, forwhich the SS fluorescence quenching data invariablyshowed a positive deviation from the expected Stern-Volmer linearity as discussed before (cf. Fig. 8). Asthe o/av versus [Q]eff plots in the present cases showquite linear Stern-Volmer correlation, it is thus pos-sible to estimate the average reaction rate constantskq (= kobs) from the slopes of these plots quite sat-

Fig. 9. (A) Typical non-exponential fluorescence kineticsobserved in TCSPC measurements for PET reactionsin micro-heterogeneous media. The fluorescence de-cay kinetics for the dye (coumarin-522, C522) is single-exponential in the absence of quencher (N,N-diethyl-p-toluidine; DMPT) but distinctly non-exponential inthe presence of the quencher141. (B) Typical non-exponential fluorescence kinetics as recorded fromup-conversion measurements for PET reactions inmicro-heterogeneous media. The kinetic trace of thedye (coumarin-153, C153) in the absence of quencher(N,N-dimethylaniline; DMAN) is just flat as expectedbut the kinetic trace becomes highly non-exponentialin the presence of the quencher85.


1328

isfactorily such that these kq values can be consideredas the effective measures of the observed ET ratesfor their subsequent correlation with the exergonicityof the concerned ET reactions, to search of the MarcusInversion behavior77–90,140–151. In these cases, how-ever, when the TR studies are carried out usingultrafast techniques like fluorescence up-conversionmeasurements having sub-picosecond time resolutions,obviously the simplified relation as given by eq. (27)would no longer remain applicable. In these cases,however, the inverse of the estimated ultrafast decaytime constants (fast) as obtained from multi-exponen-tial analysis of the kinetic traces can directly be usedas the measures of the absolute ET rates10,69–72,77,83–86,152,153, simply as, ket = fast–1.

As discussed earlier, in low viscosity conventionalhomogeneous solvents the Marcus Inversion behav-ior for bimolecular PET reactions mostly remainsobscured due to the influence of the reactant diffu-sions on the observed reaction rates, which effec-tively leads to the appearance of the typical Rehm-Weller behavior for the kobs versus G0 plots1–

16,32,35,36,53,113–138. Accordingly, in such homoge-

neous solutions, the Marcus inversion behavior forbimolecular PET reactions not only reported for lim-ited cases, but also in most of these reports the ob-served inversion behavior is not found to be veryconvincing113–120. In constrained micro-heterogeneoussystems like micelles, reverse micelles, vesicles, etc.,however, the Marcus Inversion behavior is in generalfound to be displayed for most of the bimolecularPET systems reported in the literature10,69,77–90,140–163.This is undoubtedly thus a very striking feature forthe bimolecular PET reactions in the constrained re-action media, considering that the similar Marcus In-version behavior is extremely rare to be observed forthe similar bimolecular PET systems in the fast relax-ing homogeneous solvent media1–16,32,35,36,53,113–138.Fig. 11 shows two typical such results showing clearMarcus Inversion behavior for bimolecular PET re-actions, the first one (Fig. 11A) observed in aqueousP123 triblock copolymer micelles using coumarin dyesas the electron acceptors and N,N-dimethylaniline(DMAN) as the electron donor79 and the second one(Fig. 11B) in aqueous sodium dodecyl sulphate (SDS)micelles using hydroxy and amino substituted an-thraquinone dyes as the electron acceptors and num-ber of aromatic amines as the electron donors80.

An important aspect to be noticed from the kobsversus G0 correlation plots for the bimolecular PETreactions in constrained micro-heterogeneous media,as is also indicated from the plots in Fig. 11, that theonset of the Marcus Inversion appears at anexergonicity (–G0) which is clearly much lower thanthe total reorganization energy (= s + i) ex-pected for such ET systems. Thus, for the represen-tative Marcus correlation plots shown in Fig. 11, theonset of the Marcus Inversion is seen to appear at anexergonicity of ~0.7 eV, whereas in the studied mi-cellar palisade/Stern layer the total reorganizationenergy (= s + i) is supposed to be about 1.2 eVor more79,80, because in these cases the solvent reor-ganization energy s is estimated to be about 1.0 eVon the basis of the measured dielectric constants intheir palisade/Stern layers, and the intramolecularreorganization energy i involving the studied elec-

Fig. 10. Typical time-resolved (TR) fluorescence quenchingresults obtained using TCSPC measurements for bi-molecular ET reactions in micro-heterogeneous me-dia. The results show quite good linearity followingStern-Volmer analysis. Data in this figure correspondsto bimolecular ET reactions in Triton-X100 micellesinvolving (i) C481, (ii) C151 and (iii) C153 dyes asthe acceptors and N,N-diethyl-p-toluidine (DMPT)as the donor141.


1329

tron acceptor-donor systems is expected to be typi-cally in the range of about 0.2–0.3 eV. That the MarcusInversion appears at a much lower exergonicity thanthe total reorganization energy () involved in the ETreactions is also found to be a quite common observa-tion for bimolecular ET reactions in the other con-strained micro-heterogeneous media reported10,69,77–

80,85–88,140–145. Such a shift in the Marcus Inversiontowards a lower exergonicity region is in fact real-ized to be due to the partial contribution of the sol-vent reorganization energy s towards the free en-ergy of activation (G*) for the ET reaction, becausein these cases the ET reactions occur following a

2DET mechanistic model (due to exceedingly slowsolvent relaxation), as we have discussed before (cf.Fig. 6 and 7, eq. (22), (23) and (24)).

From the discussions made so far, it is evidentthat in constrained micro-heterogeneous media due tothe non-significant (or slow) diffusion of the reac-tants within the timescales of the ET reactions (whichmakes the bimolecular ET reactions quite equivalentto the intramolecular ET reactions) and the excep-tionally slow solvent relaxation dynamics in thesemedia (which enforces only a partial contribution ofs towards G*; cf. eqs. (22), (23) and (24)) theMarcus Inversion behavior becomes quite easy to beobserved even for the bimolecular PET reactions. Thearguments made henceforth in regard to the bimo-lecular PET reactions in constrained micro-heteroge-neous media can logically also be applied for otherhigh viscosity slow relaxing solvents, e.g. the ionicliquid (IL) solvents, for which due to high viscosityof the solvents the diffusion of the reactants and alsothe solvent relaxation dynamics become inherentlyvery slow. In fact, following the unique Marcus In-version behavior reported for the bimolecularPET reactions in various micro-heterogeneous me-dia10,77–80,85–88,140–145, a number of follow up stud-ies on bimolecular PET reactions have also been car-ried out in different ionic liquid solvents and as ex-pected most of these studies also found an easy ap-pearance of the Marcus Inversion behavi-or69,84,89,90,188–191, though such a behavior is rarelybeen reported for the bimolecular PET reactions in fastrelaxing homogeneous solvents1–16,32,35,36,53,113–138.Therefore, the reported results on bimolecular PETreactions carried out in diverse constrained media haveestablished beyond doubted that under such situationsthe PET reaction not only occur under a non-diffu-sive condition but also the reaction follows the unique2DET mechanistic model such that the Marcus Inver-sion behavior is easily observed for the studied bimo-lecular PET reactions in these media, as the inversionregion is significantly shifted towards lowerexergonicity region as compared to that expected inthe fast relaxing homogeneous solvents where the ETreaction in effect occurs following the conventional1DET mechanistic model.

Fig. 11. The Marcus correlation plots for bimolecular PETreactions in (A) P123 triblock copolymer micelles79

and (B) in conventional SDS micelles80. Marcus In-version behavior is clearly indicated in both the cases.Interestingly, the onset of Marcus Inversion in thesecases appear at an exergonicity (–G0) of ~0.7 eV,which is evidently much lower than the expected to-tal reorganization energy (= s+-i) of about 1.2eV or more79,80.

JICS-10


1330

Conclusion

In conclusion, it has been convincingly establishedfrom extensive studies that the PET reactions in mi-cro-heterogeneous media as well in other constrainedenvironments like in IL solvents effectively occursfollowing a 2DET model than the conventional 1DETmodel application to fast relaxing conventional ho-mogeneous media. For PET reactions in constrainedmedia as discuss in this article, the ET reaction effec-tively proceeds along the intramolecular coordinateq, which is reaction coordinate following 2DET model,keeping a non-equilibrium distribution of the reactantstate along the solvent coordinate X during the progressof the reaction. This happens because the solvent re-laxation being slow in these media, its reorganizationcannot compete with the ET reaction. Due to non-equilibrium solvent reorganization during the courseof the ET reaction, which is the basis of the 2DETmodel, in these cases the free energy of activationG* for the ET reaction can only involve a partialcontribution from s and consequently the onset ofMarcus Inversion in these cases get largely shiftedtowards the lower exergonicity region as comparedto the 1DET cases that occur in fast relaxing conven-tional solvents. This shift of the onset of Marcus In-version towards the lower exergonicity effectivelyhelps in observing the Marcus Inversion behavior easilyfor the bimolecular PET reactions, as have been re-ported for bimolecular PET reaction in various con-strained reaction media. The high viscosity of con-strained reaction media also resists the free diffusionof the reactants, making the PET reactions to occureffectively only in those donor-acceptor pairs that arepreexisting within the dimension of the reaction sphereat the moment of their photoexcitation process andthus making these bimolecular ET reactions effec-tively very similar to the intramolecular PET reac-tions. Such a situation also in effect helps the MarcusInversion behavior to be observed easily for bimo-lecular PET reactions in the studied constrained reac-tion media. In brief, it is established beyond doubtthat the combination of the slow diffusion of the reac-tants and slow solvent relaxation effectively assist

Marcus Inversion behavior to be observed easily forbimolecular PET reactions in the constrained reac-tion environments like micro-heterogeneous media andin high viscosity IL solvents. These intriguing resultsfor bimolecular PET reactions in constrained reactionmedia are undoubtedly very distinct observations be-cause such behaviors for bimolecular PET reactionsare hardly observed when reactions are carried out infast relaxing conventional homogeneous solvents.

References

1. "Special Issue on Electron Transfer. Chem. Rev.", Vol.92, pp. 365-490, ed. A. M. Fox, Washington, DC, 1992.

2. J. R. Bolton, N. Mataga and G. L. McLendon, "Elec-tron transfers in inorganic, organic and biological sys-tems", American Chemical Society, Washington, DC,1991.

3. R. A. Rossi, A. B. Pierini and A. B. Peñéñory, Chem.Rev., 2003, 103, 71.

4. A. I. Ivanov and V. A. Mikhailova, Russ. Chem. Rev.,2011, 79, 1047.

5. V. Balzani, "Electron transfer in chemistry", Vol. I-V,Wiley-VCH Publishers Inc., New York, 2001.

6. M. A. Fox and M. Chanon, "Photoinduced electrontransfer", Elsevier, Amsterdam, 1988.

7. T. Kumpulainen, B. Lang, A. Rosspeintner and E.Vauthey, Chem. Rev., 2017, 117, 10826.

8. S. J. Formosinho, L. G. Arnaut and R. Fausto, Prog.Reac. Kin., 1998, 23, 1.

9. H. B. Gray and J. R. Winkler, PNAS, 2005, 102, 3534.

10. M. Kumbhakar and H. Pal, Sci. Lett. J., 2015,4:139, 1.

11. A. M. Kuznetsov and J. Ulstrup, "Electron trans-fer in chemistry and biology : An introduction tothe theory", Wiley, New York, 1998.

12. S. Dadashi-Silab, S. Doran and Y. Yagci, Chem.Rev., 2016, 116, 10212.

13. B. M. Rosen and V. Percec, Chem. Rev., 2009,109, 5069.

14. A. Ito and T. J. Meyer, Phys. Chem. Chem.Phys., 2012, 14, 13731.

15. C. J. Gagliardi, B. C. Westlake, C. A. Kent, J. J.Paul, J. M. Papanikolas and T. J. Meyer, Coord.Chem. Rev., 2010, 254, 2459.

16. S. Doose, H. Neuweiler and M. Sauer, Chem.Phys. Chem., 2009, 10, 1389.

17. M. R. Wasielewski, Chem. Rev., 1992, 92, 435.

18. "Electron transfer from isolated molecules tobiomolecules", Adv. Chem. Phys., Vol. 106 &


1331

107, eds. J. Jortner and M. Bixon, Wiley, NewYork, 1999.

19. J. Barber, Chem. Soc. Rev., 2009, 38, 185.

20. M. Hambourger, G. F. Moore, D. M. Kramer, D.Gust, A. L. Moore and T. A. Moore, Chem. Soc.Rev., 2009, 38, 25.

21. F. Sun, Q. Zhou, X. Pang, Y. Xu and Z. Rao,Curr. Opin. Str. Biol., 2013, 23, 526.

22. M. Cordes and B. Giese, Chem. Soc. Rev., 2009,38, 892.

23. S. K. Chamorovsky, D. A. Cherepanov, C. S.Chamorovsky and A. Y. Semenov, Biochim.Biophys. Acta, 2007, 1767, 441.

24. D. Noy, C. C. Moser and P. L. Dutton, Biochim.Biophys. Acta, 2006, 1757, 90.

25. S. Saen-Oon, M. F. Lucasa and V. Guallar, Phys.Chem. Chem. Phys., 2013, 15, 15271.

26. A. Listorti, B. O’Regan and J. R. Durrant, Chem.Mat., 2011, 23, 3381.

27. H. Petek and J. Zhao, Chem. Rev., 2010, 110,7082.

28. R. D. J. Miller, G. L. McLendon, A. J. Nozik,W. Schmickler and F. Willig, "Surface electrontransfer processes", VCH Publishers Inc., NewYork, 1995.

29. D. Gust, T. A. Moore and A. L. Moore, Acc.Chem. Res., 2001, 34, 40.

30 S. A. Patil, C. Hägerhäll and L. Gorton, Bioanal.Rev., 2012, 4, 159.

31. M. Gratzel, "Heterogeneous photochemical elec-tron transfer", CRC Press, Boca Raton, FL, 1989.

32. G. J. Kavarnos, "Fundamentals of photoinducedelectron transfer", VCH Publishers, New York,1993.

33. K. D. Jordan and M. N. Paddon-Row, Chem.Rev., 1992, 92, 395.

34. A. Arrigo, A. Santoro, M. T. Indelli, M. Natali,F. Scandola and S. Campagna, Phys. Chem.Chem. Phys., 2014, 16, 818.

35. S. A. Angel and K. S. Peters, J. Phys. Chem.,1989, 93, 713.

36. J. Kroon, H. Oevering, J. W. Verhoeven, J. M.Warman, A. M. Oliver and M. N. Paddon-Row,J. Phys. Chem., 1993, 97, 5065.

37. O. S. Wenger, Acc. Chem. Res., 2011, 44, 25.

38. V. Gladkikh, A. I. Burshtein, G. Angulo, S.Pages, B. Lang and E. Vauthey, J. Phys. Chem.(A), 2004, 108, 6667.

39. X. Dang and J. T. Hupp, J. Am. Chem. Soc.,1999, 121, 8399.

40. M. Gratzel, Coord. Chem. Rev., 1991, 111, 167.

41. M. Gratzel, J. Photochem. Photobiol. (C), 2003,4, 145.

42. D. Gust, T. A. Moore and A. L. Moore, Acc.Chem. Res., 2009, 42, 1890.

43. B. Giese, M. Graber and M. Cordes, Cur. Opin.Chem. Biol., 2008, 12, 755.

44. F. Boussicault and M. Robert, Chem. Rev., 2008,108, 2622.

45. R. A. Marcus, J. Chem. Phys., 1956, 24, 966.

46. R. A. Marcus, J. Chem. Phys., 1956, 24, 979.

47. R. A. Marcus, Annu. Rev. Phys. Chem., 1964,15, 155.

48. R. A. Marcus, Angew. Chem. Int. Ed., 1993, 32,1111.

49 P. F. Barbara and W. Jarzeba, Adv. Photochem.,1990, 15, 1.

50. D. F. Calef and P. G. Wolynes, J. Phys. Chem.,1983, 87, 3387.

51. R. A. Marcus and N. Sutin, Biochem. Biophys.Acta., 1985, 811, 265.

52. D. R. Weinberg, C. J. Gagliardi, J. F. Hull, C.F. Murphy, C. A. Kent, B. C. Westlake, A. Paul,D. H. Ess, D. G. McCafferty and T. J. Meyer,Chem. Rev., 2012, 112, 4016.

53. H. Heitele, Angew. Chem. Int. Ed., 1993, 32,359.

54. K. Wynne and R. M. Hochstrasser, in "Advancesin chemical physics : Electron transfer – from iso-lated molecules to biomolecules", eds. I. Prigogineand S. A. Rice, Part 2, Volume 107, John Wiley& Sons, NJ, USA, 2007.

55. L. D. Zusman, Chem. Phys., 1980, 49, 295.


57. M. D. Newton and N Sutin, Ann. Rev. Phys.Chem., 1984, 35, 437.


59. I. Rips and J. Jortner, J. Chem. Phys., 1987, 87,2090.

60. I. Rips and J. Jortner, J. Chem. Phys., 1987, 87,6513.

61. J. Jortner and M. Bixon, J. Chem. Phys., 1988,88, 167.

62. M. Bixon and J. Jortner, Chem. Phys., 1993,176, 467.

63. M. D. Newton, Adv. Chem. Phys., 1999, 107,303.

64. R. A. Marcus and N. Sutin, Biochim. Biophys.Acta, 1985, 811, 265.

65. M. D. Newton, Chem. Rev., 1991, 91, 767.

66. C. Costentin, Chem. Rev., 2008, 108, 2145.


1332

67. P. Chen and T. J. Meyer, Chem. Rev., 1998, 98,1439.

68. P. Verma and H. Pal, Phys. Chem. Chem. Phys.,2015, 17, 15400.

69. M. Kumbhakar, A. Manna, M. Sayed, A. Kumarand H. Pal, J. Phys. Chem. (B), 2014, 118, 10704.

70. P. Mukherjee, S. Biswas and P. Sen, J. Phys.Chem. (B), 2015, 119, 11253.

71. P. Mukherjee, A. Das, A. Sengupta and P. Sen,J. Phys. Chem. (B), 2017, 121, 1610.

72. Y. Venkatesh, V. Munisamy, B. Ramakrishna, P.H. Kumar, H. Mandal and P. R. Bangal, Phys.Chem. Chem. Phys., 2017, 19, 5658.

73. B. Wu, M. Liang, M. Maroncelli and E. W.Castner (Jr.), J. Phys. Chem. (B), 2015, 119,14790.

74. M. Liang, A. Kaintz, G. A. Baker and M.Maroncelli, J. Phys. Chem. (B), 2012, 116, 1370.

75. A. Rosspeintner, G. Angulo and E. Vauthey, J.Am. Chem. Soc., 2014, 136, 2026.

76. M. Koch, A. Rosspeintner, G. Angulo and E.Vauthey, J. Am. Chem. Soc., 2012, 134, 3729.

77. M. Kumbhakar, S. Nath, T. Mukherjee and H.Pal, J. Indian Chem. Soc., 2010, 87, 173.

78. M. Kumbhakar, S. Dey, P. K. Singh, S. Nath, A.K. Satpati, R. Gangully, V. K. Aswal and H. Pal,J. Phys. Chem. (B), 2011, 115, 1638.

79. A. K. Satpati, M. Kumbhakar, S. Nath and H.Pal, J. Phys. Chem. (B), 2007, 111, 7550.

80. M. Kumbhakar, S. Nath, T. Mukherjee and H.Pal, J. Chem. Phys., 2005, 122, a.n. 084512.

81. C. Tablet, I. Matei and M. Hillebrand, J. Mol.Liq., 2011, 160, 57.

82. C. Ghatak, V. G. Rao, S. Mandal and N. Sarkar,Phys. Chem. Chem. Phys., 2012, 14, 8925.

83. U. Mandal, S. Ghosh, S. Dey, A. Adhikari andK. Bhattacharyya, J. Chem. Phys., 2008, 128,a.n. 164505.

84. A. K. Das, T. Mondal, S. Sen Mojumdar and K.Bhattacharyya, J. Phys. Chem. (B), 2011, 115,4680.

85. M. Kumbhakar, P. K. Singh, S. Nath, A. C.Bhasikuttan and H. Pal, J. Phys. Chem. (B),2008, 112, 6646.

86. M. Kumbhakar, P. K. Singh, A. K. Satpati, S.Nath and H. Pal, J. Phys. Chem. (B), 2010, 114,10057.

87. S. Dutta Choudhury, M. Kumbhakar, S. Nath, S.K. Sarkar, T. Mukherjee and H. Pal, J. Phys.Chem. (B), 2007, 111, 8842.

88. S. Dutta Choudhury, M. Kumbhakar, S. Nath andH. Pal, J. Chem. Phys., 2007, 127, a.n. 194901.

89. S. Sarkar, R. Pramanik, C. Ghatak, V. G. Raoand N. Sarkar, J. Chem. Phys., 2011, 134, a.n.074507.

90. S. Sarkar, S. Mandal, R. Pramanik, C. Ghatak,V. G. Rao and N. Sarkar, J. Phys. Chem. (B),2011, 115, 6100.

91. R. A. Marcus, Rev. Modern Phys., 1993, 65,599.

92. J. Zhu and J. C. Rasaiah, J. Chem. Phys., 1991,95, 3325.





97. S. Lee and J. T. Hynes, J. Chem. Phys., 1988,88, 6853.

98. S. Chaudhury and B. J. Cherayil, J. Chem.Phys., 2007, 127, a.n. 105103.

99. M. Tachiya, Y. Tabata and K. Ohshima, J. Phys.Chem., 1973, 77, 263.

100. M. Tachiya, J. Phys. Chem., 1993, 97, 5911.

101. M. Tachiya, J. Chem. Phys., 2008, 129, a.n.066102.

102. J. R. Miller, L. T. Calcaterra and G. L. Closs, J.Am. Chem. Soc., 1984, 106, 3047.

103. J. R. Miller, J. V. Beitz and R. K. Huddleston, J.Am. Chem. Soc., 1984, 106, 5057.

104. M. R. Wasielewski, N. P. Niemczyk, W. A. Svecand E. B. Pewitt, J. Am. Chem. Soc., 1985, 107,1080.

105. G. L. Closs, L. T. Calcaterra, N. J. Green, K.W. Penfield and J. R. Miller, J. Phys. Chem.,1986, 90, 3673.

106. I. R. Gould, D. Ege, S. L. Mattes and S. Farid,J. Am. Chem. Soc., 1987, 109, 3794.

107. P. Chen, R. Duesing, D. K. Graff and T. J.Meyer, J. Phys. Chem., 1991, 95, 5850.

108. J. M. Chen, T. I. Ho and C. Y. Mou, J. Phys.Chem., 1990, 94, 2889.

109. S. L. Larson, L. F. Cooley, C. M. Elliott and D.F. Kelley, J. Am. Chem. Soc., 1992, 114, 9504.

110. N. Mataga, H. Chosrowjan, Y. Shibata, N.Yoshida, A. Osuka, T. Kikuzawa and T. Okada,J. Am. Chem. Soc., 2001, 123, 12422.

111. E. Vauthey, J. Phys. Chem. (A), 2001, 105, 340.

112. M. Murakami, K. Ohkubo, T. Nanjo, K. Souma,N. Suzuki and S. Fukuzumi, Chem. Phys. Chem.,2010, 11, 2594.


1333

113. I. R. Gould and S. Farid, Acc. Chem. Res., 1996,29, 522.

114. D. M. Guldi and K. D. Asmus, J. Am. Chem.Soc., 1997, 119, 5744.

115. M. Tachiya and S. Murata, J. Phys. Chem.,1992, 96, 8441.

116. S. Murata, M. Nishimura, S. Y. Matsuzaki andM. Tachiya, Chem. Phys. Lett., 1994, 219, 200.

117. M. Hilczer and M. Tachiya, J. Mol. Liq., 1995,64, 113.

118. S. Murata and M. Tachiya, J. Phys. Chem.,1996, 100, 4064.

119. L. Burel, M. Mostafavi, S. Murata and M.Tachiya, J. Phys. Chem. (A), 1999, 103, 5882.

120. H. B Gray, J. R. Winkler and D. Wiedenfeld,Coord. Chem. Rev., 2000, 200-202, 875.

121. B. Legros, P. Vandereecken and J. P. Soumillion,J. Phys. Chem., 1991, 95, 4752.

122. G. Grampp, Angew. Chem. Int. Ed., 1993, 32,691.

123. E. Prasad and K. R. Gopidas, J. Am. Chem.Soc., 2000, 122, 3191.

124. S. Fukuzumi, Y. Yoshida, T. Urano, T. Suenobuand H. Imahori, J. Am. Chem. Soc., 2001, 123,11331.

125. S. Fukuzumi, K. Ohkubo, H. Imahori and D. M.Guldi, Chem. Eur. J., 2003, 9, 1585.

126. S. Fukuzumi, M. Nishimine, K. Ohkubo, N. V.Tkachenko and H. Lemmetyinen, J. Phys. Chem.(B), 2003, 107, 12511.

127. D. Rehm and A. Weller, Israel J. Chem., 1970,8, 259.

128. A. K. Satpati, S. Nath, M. Kumbhakar, D. K.Maity, S. Senthilkumar and H. Pal, J. Mol. Str.,2008, 878, 84.

129. M. Kumbhakar, S. Nath, M. C. Rath, T. Mukherjeeand H. Pal, Photochem. Photobiol., 2004, 79, 1.

130. J. Mohanty, H. Pal and A. V. Sapre, Photochem.Photobiol., 2003, 78, 153.

131. S. Nad and H. Pal, J. Chem. Phys., 2002, 116,1658.

132. S. Nad and H. Pal, J. Photochem. Photobiol. A :Chem., 2000, 134, 9.

133. S. Nad and H. Pal, J. Phys. Chem. (A), 2000,104, 673.

134. M. K. Singh, H. Pal and A. V. Sapre, Photochem.Photobiol., 2000, 71, 300.

135. H. N. Ghosh, H. Pal, D. K. Palit, T. Mukherjeeand J. P. Mittal, J. Photochem. Photobiol. A :Chem., 1993, 73, 17.

136. S. Nad and H. Pal, J. Phys. Chem. (A), 2000,

104, 673.

137. J. Mohanty, H. Pal, S. K. Nayak S. Chattopadhyayand A. V. Sapre, J. Chem. Phys., 2002, 117,10744.

138. J. Mohanty, H. Pal and A. V. Sapre, Photochem.Photobiol., 2003, 78, 153.

139. J. R. Lakowicz, "Principle of Fluorescence Spec-troscopy", Plenum Press, Springer, New York,2006.

140. M. Kumbhakar, S. Nath, H. Pal, A. V. Sapre andT. Mukherjee, J. Chem. Phys., 2003, 119, 388.

141. M. Kumbhakar, S. Nath, T. Mukherjee and H.Pal, J. Chem. Phys., 2004, 120, 2824.

142. M. Kumbhakar, T. Mukherjee and H. Pal, Chem.Phys. Lett., 2005, 410, 94.

143. M. Kumbhakar, S. Nath, T. Mukherjee and H.Pal, J. Chem. Phys., 2005, 123, a.n. 034705.

144. M. Kumbhakar, S. Nath, T. Mukherjee and H.Pal, J. Photochem. Photobiol. A : Chem., 2006,182, 7.

145. P. Verma, S. Nath, P. K. Singh, M. Kumbhakarand H. Pal, J. Phys. Chem. (B), 2008, 112,6363.

146. A. Chakraborty, D. Chakrabarty, P. Hazra, D.Seth and N. Sarkar, Chem. Phys. Lett., 2003,382, 508.

147. A. Chakraborty, D. Seth, D. Chakrabarty, P.Hazra and N. Sarkar, Chem. Phys. Lett., 2005,405, 18.

148. A. Chakraborty, D. Chakrabarty, D. Seth, P.Hazra and N. Sarkar, Spectrochim. Acta (A),2006, 63, 594.

149. A. Chakraborty, D. Seth, P. Seth and N. Sarkar,J. Phys. Chem. (B), 2006, 110, 16607.

150. A. Chakraborty, D. Seth, P. Setua and N. Sarkar,J. Chem. Phys., 2006, 124, a.n. 074512.

151. A. Chakraborty, D. Seth, P. Setua and N. Sarkar,J. Chem. Phys., 2008, 128, a.n. 204510.

152. S. Ghosh, K. Sahu, S. K. Mondal, P. Sen and K.Bhattacharyya, J. Chem. Phys., 2006, 125, a.n.054509.

153. S. Ghosh, S. K. Mondal, K. Sahu and K.Bhattacharyya, J. Chem. Phys., 2007, 126, a.n.204708.

154. M. Marchena and F. Sanchez, Prog. Reac. Kin.Mech., 2010, 35, 27.

155. M. Lopez-Lopez, F. Sanchez and M. Marchena,Prog. Reac. Kin. Mech., 2012, 37, 203.

156. F. Sanchez, A. Barrios, M. Lopez-Lopez, P.Lopez-Cornejo, E. Bernal, B. Sarrion, J. A.Lebron and M. Marchena, Prog. Reac. Kin.Mech., 2014, 39, 151.


1334

157. M. Tachiya, in "Kinetics of non-homogeneousprocesses", ed. G. R. Freeman, Wiley, NewYork, 1987.

158. M. Tachiya, Chem. Phys. Lett., 1975, 33, 289.

159. M. Tachiya, Can. J. Phys., 1990, 68, 979.

160. J. H. Fendler, "Membrane Mimetic Chemistry",Wiley, New York, 1982.

161. M. Gratzel, J. J. Kozak and J. K. Thomas, J.Chem. Phys., 1975, 62, 1632.

162. L. K. Patterson and M. Gratzel, J. Chem. Phys.,1975, 79, 956.

163. D. J. W. Barber, D. A. N. Morris and J. K. Tho-mas, Chem. Phys. Lett., 1976, 37, 481.

164. M. Kumbhakar, T. Goel, S. Nath, T. Mukherjeeand H. Pal, J. Phys. Chem. (B), 2006, 110,25646.

165. M. Kumbhakar, S. Nath, T. Mukherjee and H.Pal, J. Chem. Phys., 2004, 121, 6026.

166. P. K. Singh, M. Kumbhakar, H. Pal and S. Nath,J. Phys. Chem. (B), 2008, 112, 7771.

167. D. Chakrabarty, A. Chakraborty, D. Seth, P.Hazra and N. Sarkar, J. Chem. Phys., 2005, 122,a.n. 184516.

168. P. Hazra, D. Chakrabarty and N. Sarkar, Langmuir, 2002, 18, 7872.

169. K. Bhattacharyya, Chem. Commun., 2008, 2848.

170. K. Bhattacharyya and B. Bagchi, J. Phys. Chem.(A), 2000, 104, 10603.

171. N. Nandi, K. Bhattacharyya and B. Bagchi, Chem.Rev., 2000, 100, 2013.

172. P. Sen, S. Ghosh, K. Sahu, S. K. Mondal, D.Roy and K. Bhattacharyya, J. Chem. Phys., 2006,124, a.n. 204905.

173. S. Ghosh, A. Adhikari, U. Mandal, S. Dey andK. Bhattacharyya, J. Phys. Chem. (C), 2007,111, 8775.

174. M. Kumbhakar, T. Goel, T. Mukherjee and H.Pal, J. Phys. Chem. (B), 2004, 108, 19246.

175. M. Kumbhakar, T. Goel, T. Mukherjee and H.Pal, J. Phys. Chem. (B), 2005, 109, 14168.

176. H. Sumi and R. A. Marcus, J. Chem. Phys.,1986, 84, 4894.

177. H. Kandori, K. Kemnitz and K. Yoshihara, J.Phys. Chem., 1992, 96, 8042.

178. Y. Nagasawa, A. P. Yartsev, K. Tominaga, A. E.Johnson, and K. Yoshihara, J. Am. Chem. Soc.,1993, 115, 7922.

179. Y. Nagasawa, A. P. Yartsev, K. Tominaga, A. E.Johnson and K. Yoshihara, J. Chem. Phys., 1994,101, 5717.

180. Y. Nagasawa, A. P. Yartsev, K. Tominaga, P. B.Bisht, A. E. Johnson and K. Yoshihara, J. Phys.Chem., 1995, 99, 653.

181. H. Pal, H. Shirota, K. Tominaga and K. Yoshihara,J. Chem. Phys., 1999, 110, 11454.

182. H. Shirota, H. Pal, K. Tominaga and K. Yoshihara,Chem. Phys., 1998, 236, 355.

183. H. Shirota, H. Pal, K. Tominaga and K. Yoshihara,J. Phys. Chem. (A), 1998, 102, 3089.

184. H. Pal, Y. Nagasawa, K. Tominaga and K.Yoshihara, J. Phys. Chem., 1996, 100, 11964.

185. P. K. Singh, S. Nath, A. C. Bhasikuttan, M.Kumbhakar, J. Mohanty, S. K. Sarkar, T.Mukherjee and H. Pal, J. Chem. Phys., 2008,129, 114504.

186. C. Wang, B. Akhremitchev and G. C. Walker, J.Phys. Chem. (A), 1997, 101, 2735.

187. C. Wang, B. Akhremitchev and G. C. Walker, J.Phys. Chem. (A), 1997, 101, 2735.

188. M. Kumbhakar and H. Pal, J. Phys. Chem. (B),2016, 120, 9804.

189. S. Yuan, Y. Zhang, R. Lu and A. Yu, J.Photochem. Photobiol. A : Chem., 2013, 260,39.

190. S. Sarkar, S. Mandal, C. Ghatak, V. G. Rao, S.Ghosh and N. Sarkar, J. Phys. Chem. (B), 2012,116, 1335.

191. S. Sarkar, R. Pramanik, C. Ghatak, V. G. Raoand N. Sarkar, Chem. Phys. Lett., 2011, 506,211.

1335


Effective delivery of doxorubicin using biologically synthesized goldnanoparticles for cancer therapy†

Ayan Kumar Baruia, Sourav Dasa, Preetika Soanpeta, Bojja Sreedharb and Chitta Ranjan Patraa#*aChemical Biology Division, CSIR-Indian Institute of Chemical Technology, Uppal Road, Tarnaka,Hyderabad-500 007, India

E-mail : [email protected], [email protected] and Physical Chemistry Division, CSIR-Indian Institute of Chemical Technology,Uppal Road, Tarnaka, Hyderabad-500 007, India#Academy of Scientific and Innovative Research (AcSIR), Training and Development Complex,CSIR Campus, CSIR Road, Taramani, Chennai-600 113, India


Abstract : We synthesized gold nanoparticles (AuHS) by a single step reduction of chloroauric acid solutionusing aqueous leaf extract of Hibiscus sabdariffa (HS) plant through an environmental-friendly bio-green ap-proach. The as-synthesized AuHS nanoparticles were thoroughly characterized by several analytical techniquesincluding UV-Visible spectroscopy, DLS, XRD, TEM, XPS and ICP-OES. Cell viability assay revealed thatthe nanoparticles are highly biocompatible to various normal cells (EA.hy926 and CHO) as well as cancercells (HCT-15, HeLa, PC3 and A549). The chick embryonic angiogenesis (CEA) assay also supported thebiocompatible nature of AuHS nanoparticles. The nanoparticles were found to be stable in different physio-logical buffer and other solutions, indicating the feasibility of their biological applications. Considering thebiocompatibility of AuHS nanoparticles, FDA approved anti-cancer drug doxorubicin (DOX) was conjugatedwith the nanoparticles to design a drug delivery system (DDS : AuHS-DOX). The administration of AuHS-DOX to different cancer cells (HCT-15, HeLa and PC3) manifested more inhibition of cell proliferation com-pared to pristine DOX, suggesting the therapeutic efficacy of the DDS. The better therapeutic response ofthe DDS could be attributed to the enhanced cellular uptake of DOX present in AuHS-DOX compared tofree DOX as observed by confocal microscopy. We strongly believe that the biocompatible AuHS nanoparticlescould be employed as an effective delivery vehicle for cancer therapy through nanomedicine approach in im-minent future.

Keywords : Gold nanoparticles, biosynthesis, green chemistry approach, drug delivery system, nanomedicine,cancer therapy.

Introduction

Nanotechnology, one of the emerging technolo-gies of modern world, exhibits its enormous potentialfor various biomedical applications including diagno-sis and therapy of different diseases such as cancer,cardiovascular diseases, neurodegenerative diseases,diabetes, microbial infection etc.1–8. Since past de-

cades, among various nanomaterials, gold nano-particles got more attention to the researchers for theirdiverse applications in varied field including sensor,catalysis, electronics as well as disease diagnosis andtherapeutics due to several advantages e.g. (i) simplesynthesis method, (ii) easier characterization becauseof their surface plasmon resonance (SPR) band, (iii)

†Dr. D. S. Bhakuni Award Lecture (2016).


1336

easier surface modification, (iv) biocompatibility, (v)long historical use of gold in medicine etc.2,4,9–12.Owing to the numerous applications of gold nano-particles in today’s world, it is highly required todesign and develop alternative sustainable, eco-friendly, cost-effective process for the fabrication ofgold nanoparticles12–14. Therefore, scientists are pres-ently highly engaged toward the advancement of bio-green method for the synthesis of low cost bio-compatible gold nanoparticles that could be employedfor different applications15–21. The bio-green routesfor the fabrication of gold nanoparticles possess sev-eral advantages compared to conventional chemicalmethods such as single step, fast reaction, environ-mentally benign, cost effective etc. making it moreconvenient approach22,23.

In this context, we have developed biologicallysynthesized highly stable gold nanoparticles (AuHS)through the reduction of chloroauric acid (HAuCl4)by aqueous leaf extract of Hibiscus sabdariffa (HS),followed by their detailed characterization. In vitrocell viability assay and in vivo chick embryo angio-genesis (CEA) assay revealed the biocompatible na-ture of as-synthesized nanoparticles which compelledus to design an AuHS nanoparticles based drug deli-very system (DDS : AuHS-DOX) containing anti-cancer drug doxorubicin (DOX). The application ofthe DDS to different cancer cells exhibited more inhi-bition of cancer cell proliferation compared to freedrug, indicating the efficacy of AuHS nanoparticlesas a delivery vehicle for cancer therapy. It could bespeculated that the biologically synthesized environ-ment-friendly biocompatible AuHS nanoparticleswould be a promising candidate for cancer therapeu-tic applications in forthcoming future.

Experimental

Materials

Tetrachloroauric(III) acid (HAuCl4.3H2O), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bro-mide (MTT reagent), doxorubicin (DOX), Dulbecco’sModified Eagle Medium (DMEM), Dulbecco’s Phos-phate Buffer Saline (DPBS) and fetal bovine serum

(FBS) were purchased from Sigma-Aldrich, USA andused without further purifications. Hibiscus sabdariffaleaves were collected from the CSIR-IICT Campus.Different cancer cell lines (HeLa : human cervicalcancer cells; PC3 : human prostate cancer cells; A549 :human lung cancer cells) and normal cell line (CHO:Chinese hamster ovarian cell line) were purchasedfrom the American Type Culture Collection (ATCC),USA. Human collateral adenocarcinoma cell line(HCT-15) was obtained from NCCS, Pune, India.Additionally, human endothelial somatic hybrid cellline (EA.hy926) were a kind gift from S. Oglesbee,Tissue Culture Facility, University of North Caro-lina, Lineberger Comprehensive Cancer Center, NC,USA and S. Chatterjee, Anna University – K. B.Chandrasekhar, Chennai, India. The fertilized chickeneggs were purchased from the government poultrystation (Directorate of Poultry Research, Hyderabad,Telangana).

Preparation of stock solutions

Initially, 10–2 M HAuCl4 solution was preparedusing autoclaved MiliQ water and this stock solutionwas further used to synthesize gold nanoparticles.Additionally, 1 mg/mL stock solution of doxorubicin(DOX) was prepared in autoclaved MiliQ water. Thestock solution of DOX was further diluted for experi-mental work.

Preparation of HS leaf extract

At first, the green leaves of HS were thoroughlywashed using MiliQ water. In order to prepare thestock solution, 105 g of grounded HS leaves weredissolved in 420 mL autoclaved MiliQ water in abeaker. The crude extract was then heated occasion-ally at 600 W up to 3 h and stirred for overnight at amagnetic stirrer. The crude extract was finally centri-fuged at 4000 rpm for 1 h at 20 ºC. Finally, thesupernatant (150 mL) was collected and stored at–20 ºC for synthesis of AuHS nanoparticles.

Bio-synthesis of AuHS nanoparticles

We have fabricated AuHS nanoparticles throughthe reduction of HAuCl4 solution using aqueous HS

Barui et al. : Effective delivery of doxorubicin using biologically synthesized gold nanoparticles etc.

1337

leaf extract through a modified bio-green approach24.Mishra et al. earlier synthesized gold nanoparticlesusing highly concentrated stem and leaf extracts ofHS and showed that these nanoparticles were moder-ate to highly toxic to 293 as well as U87 glioblastomacell lines, respectively even at very low dose (2.5 ng/mL). This could be attributed to the presence of HSextract containing toxic phytochemicals (e.g. proto-catechuic acid, delphinidin-3-sambubioside etc.) onthe surface of nanoparticles at higher concentration.The authors also proposed that the cytotoxicity mightbe due to the synergistic effect of plant extract andgold nanoparticles. It is highly essential for any nano-particle/biomaterial/drug to be non-toxic for its dif-ferent biomedical applications. Therefore, to obtainbiocompatible gold nanomaterials, we have synthe-sized AuHS nanoparticles in a modified way employ-ing very diluted HS leaf extract, so that the corre-sponding phytochemicals content become optimum forthe reduction of HAuCl4 as well as stabilization ofnanoparticles, without exerting any nanomaterialsmediated toxic effect even at higher doses (g/mLrange). In brief, HS extract (200–400 L) was dropwisely added to the solution of HAuCl4 (200 L,10–2 M), keeping the total volume of the reactionmixture fixed at 5 mL and stirred it for overnightusing a magnetic stirrer under an ambient conditionto obtain the red colour AuHS nanoparticles. The as-synthesized nanoparticles were designated as AuHS-200, AuHS-300 and AuHS-400 where the numericalvalue indicated the volume of HS extract (in L) usedfor their synthesis. The method for the formation ofgold nanoparticles using the leaf extract is environ-mental-friendly and cost-effective compared to the con-ventional chemical methods. The as-synthesizednanoparticles were purified by removing the unreactedgold ions and HS extract through ultracentrifugationand the loose pellet was used for different characte-rizations as well as biological studies to avoid anyunwanted toxic effect.

Characterizations of AuHS nanoparticles

The as-synthesized AuHS nanoparticles were thor-oughly characterized employing different physico-

chemical techniques. The loose pellet (intense red) ofAuHS nanoparticles were collected after centrifuga-tion of 20 mL reaction mixture at 15,000 rpm, 20 ºCfor 45 min using an ultracentrifuge machine (ThermoScientific, Sorvall-WX ultra 100) and stored at 4 ºCfor characterizations as well as other experimentaluse.

UV-Visible spectroscopy

The absorbance of as-synthesized AuHS-200,AuHS-300, AuHS-400 nanoparticles as well as HSextract was measured using UV-Visible spectroscopy(JASCO spectrophotometer) using quartz cuvette.

Dynamic light scattering (DLS)

DLS is a technique to determine the size and sur-face charge of nanoparticles/drug molecules. Thepellets of as-synthesized AuHS-200, AuHS-300 andAuHS-400 nanoparticles (10 L) were mixed with990 L of MiliQ water and the solutions were ana-lyzed using a DLS instrument (Anton Paar) to deter-mine the size and zeta-potential (surface charge) ofnanoparticles.

X-Ray diffraction (XRD)

AuHS-400 pellet was coated on a glass slide andafter evaporation of the solvent it was submitted forXRD analysis. The phase purity and crystalline na-ture of AuHS nanoparticles was determined employ-ing X-ray diffractometer (Bruker AXS D8 Advance)with CuK= 1.5406 Å radiation.

Transmission electron microscopy (TEM)

TEM study was carried out to measure the sizeand morphology of the as-synthesized AuHS-400nanoparticles obtained after 24 h of reaction condi-tion. The loose pellet of nanoparticles was diluted inMiliQ water (1 : 3 ratio) and a drop of it was pouredon the carbon grid followed by the evaporation ofsolvent. The measurement of size and morphologywas evaluated using TEM (FEI Tecnai F12; PhilipsElectron Optics, Holland; operating condition 100 kV).The selected area electron diffraction (SAED) patternwas also obtained using the microscope.

JICS-11


1338

Fourier transformed infrared spectroscopy (FT-IR)

FT-IR was used to determine the functional groupspresent in HS extract as well as AuHS-400nanoparticles. The HS leaf extract and loose pellet ofnanoparticles were drop-casted on the glass slides,dried at room temperature and submitted for FT-IRanalysis employing Thermo Nicolet Nexus spectrom-eter with resolution of 4 cm–1 on KBr pellets.

X-Ray photoelectron spectroscopy (XPS)

The surface property of the AuHS-400 nano-particles was determined through XPS analysis. Aglass slide was coated with the loose pellet ofnanoparticles and submitted for XPS after evaporat-ing the solvent. The instrument is relied on a KRATOSAXIS 165 with a dual anode (Mg and Al) apparatusemploying Mg K anode.

In vitro stability studies

To check the stability of the AuHS-400nanoparticles in various solutions (e.g. H2O, 5% NaCl,10% NaCl, DPBS and buffer solutions of pH 6–8),200 L nanoparticles were incubated with each solu-tion (1800 L) for different time points (24 h, 7 daysand 14 days) and measured their absorbance usingUV-Visible spectroscopy to evaluate their change inmax.

Conjugation of DOX with AuHS nanoparticles andevaluation of % binding of DOX

DOX (100 L; 1 mg/mL) was drop wisely pouredinto the as-synthesized AuHS-400 nanoparticles (20mL) and stirred for 2 h at room temperature to pre-pare AuHS-DOX drug delivery system (DDS). AuHS-DOX was then centrifuged at 15,000 rpm, 20 ºC for45 min to obtain AuHS-DOX pellet that was used fordifferent experiments. The % attachment of DOX inAuHS-DOX DDS was calculated taking the superna-tant of AuHS-DOX (as unknown) using spectro-fluorimetry technique [ex = 480 nm, em = 600 nmfor DOX], after making a standard curve of DOX(0.125–5 g/mL) using supernatant of AuHS-400nanoparticles.

Cell culture studies

Different normal cells (CHO and EA.hy926) aswell as cancer cells (HCT-15, HeLa, PC3 and A549)were cultured in DMEM media supplemented with10% fetal bovine serum (FBS) and 0.01% antibiotics(penicillin, streptomycin, kanamycin and ciprofloxacin)in a humidified 5% CO2 incubator at 37 ºC for all invitro experiments. HS, AuHS-400 and AuHS-DOXwere kept under UV light for 15 min prior to any invitro experiment.

Cell viability assay

Cell viability assay was carried out employing MTTreagent following our earlier literature25,26. Initially,1×104 cells/well (non-cancerous : EA.hy926 andCHO; cancerous : HCT-15, HeLa, PC3 and A549)were seeded in 96-well plate and kept for 24 h in37 ºC incubator. The cells were incubated with HSextract (0.5–5 L) and AuHS-400 pellet (0.5–5 L)for 24 h and 48 h for normal cells and cancer cells,respectively to check their cell viability. Addition-ally, HCT-15, HeLa and PC3 cells were further in-cubated with AuHS-DOX pellet (0.03–0.25 Mw.r.to DOX), AuHS-400 pellet (0.03–0.25 M w.r.toDOX) and free DOX (0.03–0.25 M) for 48 h toassess the therapeutic efficacy of the DDS. After theincubation periods, 100 L MTT solution (0.5 mg/mL) was added to each well and the plate was kept at37 ºC for 4 h. The media in each well was replacedby freshly prepared 100 L of DMSO : MeOH solu-tion (1 : 1, v/v) and absorbance was measured atmax = 570 nm using multimode reader.

Confocal microscopy for cellular uptake study

To check the intracellular uptake, 1×105 HCT-15cells were grown on embedded cover slips in eachwell of 6-well culture dishes for 24 h. The cells wereincubated with AuHS-DOX pellet (0.12 M w.r.toDOX), AuHS-400 pellet (0.12 M w.r.to DOX) andfree DOX (0.12 M) for 6 h. The cells were thor-oughly washed with DPBS and fixed with 3.7% form-aldehyde solution for 10 min. After washing withDPBS, the cover slips containing cells were mounted


1339

using DAPI vectashield on glass slides and observedunder confocal microscope (laser used, 404.2, 561nm in Nikon Ti Eclipse Confocal Microscope)27.

CEA assay

CEA assay was carried out with AuHS-400nanoparticles following our earlier literature25,26,28.Before four days of experimental study, fertilizedchicken eggs were incubated at egg incubator (~70%humidity, 37 ºC). On the experimental day, a smallwindow was carefully created on the top of the eggcell using a forcep and AuHS-400 pellet (5 L) wasgently administered on the blood vessels area ofchicken embryo. Finally, images were captured at 0h and 4 h post-treatments using an LEICA camera(LEICA MC120-HD) attached to a LEICA stereomi-croscope (LEICA S8AP0) at 10 megapixel magnifi-cations.


The bio-green synthesis of AuHS nanoparticlesusing aqueous HS leaf extract (acting as a reducingagent) was carried out under ambient conditions. Toobtain the desired stable AuHS nanoparticles a seriesof reactions were performed using different amount(200–400 L) of leaf extract as mentioned in Table1. Expt. No. #3 in Table 1 takes minimum time (45min) for the appearance of red coloration (generationof gold nanoparticles), indicating this reaction condi-tion (HS extract : 400 L) as the optimized one. Theresult also shows that the rate of the reaction (in termsof appearance of red coloration) is faster with in-creasing concentration of HS extract. The as-synthe-sized nanoparticles were characterized employing sev-

UV-Visible spectroscopy

The absorbance of the as-synthesized AuHS-200,AuHS-300 and AuHS-400 nanoparticles using diffe-rent concentration of HS extract was measured usingUV-Visible spectroscopy. Fig. 1A reveals that sur-face plasmon resonance (SPR) bands of thesenanoparticles after 24 h of reaction appeared at max~535–540 nm, indicating the formation of goldnanoparticles. The absorbance intensity of nano-particles was enhanced with increasing concentrationof HS extract used for their synthesis. The maximumabsorbance was observed for AuHS-400 nanoparticleshaving ruby red color (inset picture of Fig. 1A), sug-gesting their higher stability compared to AuHS-200and AuHS-300 nanoparticles. Moreover, the appear-ance of red coloration for AuHS-400 nanoparticlescame in a shorter time in comparison with that ofAuHS-200 and AuHS-300 nanoparticles (Table 1).Based on these factors of higher stability and higher

Table 1. Reaction conditions for the synthesis of goldnanoparticles (AuHS) using Hibiscus sabdariffa (HS) plant

leaf extract

Expt. HS extract HAuCl4 Water Total volume Timea

No. (L) (L) (mL) (mL) (min)

1. 200 200 4.6 5 85

2. 300 200 4.5 5 60

3. 400 200 4.4 5 45aTime required to observe the red coloration indicating theformation of AuHS.

Scheme 1. Overall schematic diagram for synthesis, characteri-zation of biocompatible AuHS nanoparticles and theirapplication as delivery vehicle for cancer therapy.

eral analytical techniques and used for different invitro and in vivo assays. The overall synthesis andbiological applications of AuHS nanoparticles are de-scribed in Scheme 1.

Colour


1340

rate of reaction AuHS-400 nanoparticles were con-sidered as the optimized gold nanoparticles for fur-ther studies (thorough characterizations and biologi-cal experiments).

We have also measured the time dependent absor-bance of optimized AuHS-400 nanoparticles (gener-ally termed as AuHS nanoparticles) through UV-Vi-sible spectroscopy (Fig. 1B). The absorbance inten-sity of nanoparticles was found to increase in a timedependant manner up to 24 h. However, after 48 h ofreaction the absorbance intensity of nanoparticles fallsslightly. Therefore, 24 h reaction time point was se-lected as optimized condition to synthesize AuHS-400 nanoparticles for their detailed characterizationsand biological studies.

Dynamic light scattering (DLS)

DLS was employed to determine the size and zetapotential () of as-synthesized nanoparticles. The hy-drodynamic diameter was found to be AuHS-200,AuHS-300 and AuHS-400 nanoparticles 294.2,350.11, 174.85 nm, respectively (Fig. 2a-c). The zetapotential provides an essential idea regarding the sur-face charge and stability of nanoparticles22. DLS studyexhibits the negative zeta potential/surface charge ofAuHS-200 (–6.7±0.3 mV), AuHS-300 (–4.6±0.3mV) and AuHS-400 (–6.5±0.3 mV) nanoparticles(Fig. 2d-f), suggesting the long term stability anddispersity of nanoparticles without any aggregation insolution, due to negative-negative repulsive force.

X-Ray diffraction (XRD) analysis and transmis-sion electron microscopy (TEM)

XRD analysis was employed to analyze the crystalstructure of as-synthesized AuHS nanoparticles. Thediffraction peaks appeared at hkl planes (111), (200),(220) and (311) are consistent with standard data fileshaving JCPDS card no : 04-0784 (Fig. 3a)29. As perthe Bragg’s reflections, AuHS nanoparticles are dis-tinctly indexed to FCC (face centered cubic) crystalstructure. The absence of extra peak in X-ray diffrac-tion pattern indicates the high purity of crystallineAuHS nanoparticles. The inset picture of Fig. 3a rep-resents the SAED pattern of AuHS nanoparticles in-dicating their face centered cubic (FCC) crystal struc-ture which corroborate with the XRD result.

In order to comprehend the size and morphologyof the synthesized AuHS nanoparticles, TEM analy-sis was employed. Fig. 3b represents the TEM imageexhibiting that AuHS nanoparticles consisted of mostlyspherical shaped particles (diameter : ~5–30 nm) aswell as some triangle particles. The generation of thelarger triangle particles might be the result of slowgrowth process due to weak reducing agents presentin HS extract30.

Fourier transformed infrared spectroscopy (FT-IR)

To understand the formation of AuHS nanoparticlesusing HS leaf extract, FT-IR analysis was carried

Fig. 1. (A) UV-Visible spectra of AuHS nanoparticles usingdifferent volumes (200–400 L) of HS extract after24 h of reaction. The numerical values denote thecorresponding volume of HS extract used in L. Theinset optical pictures indicate the colors of as-synthe-sized : (a) AuHS-200, (b) AuHS-300 and (c) AuHS-400 nanoparticles as well as (d) HS extract. (B) Timedependent UV-Visible spectra of optimized AuHS-400 nanoparticles.

Colour


1341

out. Fig. 4 exhibits various peaks of HS extract andAuHS nanoparticles, which could be attributed to thedifferent phytochemicals present in these systems. Thepeak at 3430.21 cm–1 for HS extract can be desig-nated to -OH stretching, which was shifted to lowerfrequency (3424.39 cm–1) and became sharper forAuHS nanoparticles. This result indicates the majorrole of polyphenolic/alcoholic compound for the re-duction of HAuCl4 to form AuHS nanoparticles17,31.The peak appeared at 2931.29 cm–1 for HS corre-sponds to the -C-H stretching of aliphatic group which

was also shifted to 2919.20 cm–1 for AuHSnanoparticles32. The three peaks at 1795.53, 1745.2,1629.79 cm–1 could be assigned for keto group ofacid anhydride, ester and amide I compounds, res-pectively present in HS extract and all of the corre-sponding peaks (1784.45, 1723.87, 1626.25 cm–1)are also present in AuHS nanoparticles, suggestingthe attachment of these phytochemicals on the surfaceof nanoparticles for their stabilization33. While, thepeaks appeared at 1407.97 cm–1 (HS) and 1403.17cm–1 (AuHS) could be referred to the -NH group in

Fig. 2. DLS analysis for the measurement of size and zeta potential of as-synthesized nanoparticles. Size : (a) AuHS-200, (b)AuHS-300 and (c) AuHS-400 nanoparticles; Zeta potential : (d) AuHS-200, (e) AuHS-300 and (f) AuHS-400 nanoparticles.


1342

protein amide linkages34, peaks observed at 1096.55cm–1 (HS) and 1020.32 cm–1 might be assigned tothe -C-N stretching of aliphatic amines. The peaksappeared at 529.41 and 522.51 cm–1 for HS and forAuHS, respectively might be designated to the C-Hbending vibrations35. The FT-IR analysis altogethersuggests that different phytochemicals such as phe-nolic/alcoholic compounds, amides, proteins etc.present in HS extract could play an important role forthe formation as well as stabilization of AuHSnanoparticles.

X-Ray photoelectron spectroscopy (XPS)

XPS analysis was employed to understand the sur-face property of the synthesized AuHS nanoparticles.Fig. 5 exhibits the binding energy (BE) peak at 87.8eV for [Au(4f)], indicating the presence of Au0 inAuHS nanoparticles. The other peaks at 285 eV, 400eV and 533 eV can be assigned to [C(1s)], [N(1s)]and [O(1s)], respectively.

Fig. 3. (a) X-Ray diffraction pattern of as-synthesized AuHSnanoparticles indicating the FCC crystalline structureof nanoparticles. The inset figure represents the SAEDpattern of the nanoparticles. (b) TEM image ofbiosynthesized AuHS nanoparticles suggesting theirspherical morphology.

Fig. 4. FT-IR of HS extract and as-synthesized AuHSnanoparticles indicating the role of different compo-nents of plant leaf extract for the synthesis of AuHSnanoparticles.

Fig. 5. XPS analysis of as-synthesized AuHS nanoparticles.


1343

In vitro stability studies of AuHS nanoparticles

The stability of the nanoparticles is one of thefundamental criteria for their biological applicationswhich compel us to perform in vitro stability studiesof as-synthesized AuHS nanoparticles in differentphysiological buffer (pH 6, 7 and 8) and other solu-tions (H2O, 5% NaCl, 10% NaCl and DPBS)22,36.The result demonstrated only a slight change in plas-mon wavelength (max) and plasmon bandwidth ()was <5 nm for AuHS nanoparticles up to 14 days,indicating their high in vitro stability in those solu-tions (Fig. 6). It could be speculated that differentproteins, carbohydrates, phenolic compounds etc.present in HS extract might support the higher stabi-lity of AuHS nanoparticles29.

sults was also observed with the treatment of aqueousHS leaf extract. Prior to the treatment of AuHSnanoparticles, the gold concentration of pellet wasfound to be 1.77 g/L, as calculated employing ICP-OES technique. Therefore, 0.5, 1, 2.5 and 5 L AuHSnanoparticles pellet treatments for cell viability assaycorrespond to gold concentrations of 8.85, 17.7, 44.25and 88.5 g/mL, respectively.

CEA assay

To confirm the biocompatibility of AuHSnanoparticles, in vivo CEA assay was further em-ployed. Fig. 9 exhibits that AuHS treatment did notaffect the vascular sprouting in chick embryo in termsof blood vessel length, junction and size even after 4h of incubation period as compared to untreated con-trol experiments, indicating the biocompatibility ofnanoparticles. This result corroborates with the cellviability data, suggesting the feasibility of bio-

Fig. 6. In vitro stability studies of AuHS nanoparticles indifferent buffer and other solutions.

Cell viability assay

The cell viability assay using MTT reagent is oneof the fundamental assays for any nanoparticle/drugprior to its biomedical applications37. Therefore, cellviability assay of AuHS nanoparticles was performedin different non-cancerous cells (EA.hy926 : Fig. 7aand CHO : Fig. 7b) as well as cancerous cells (HCT-15 : Fig. 8a, HeLa : Fig. 8b, PC3 : Fig. 8c andA549 : Fig. 8d). The result reveals that AuHSnanoparticles treatment (0.5–5 L) did not affect theviability of both these normal and cancer cell lines,suggesting their biocompatible nature. Similar, re-

Fig. 7. In vitro cell viability assay in different normal cells[(a) EA.hy926 and (b) CHO]. The result exhibits thebiocompatible nature of HS extract (0.5–5 L; 24 h)and AuHS nanoparticles (0.5–5 L; 24 h). The nu-merical values indicate the volume in L.


1344

Design of AuHS-DOX and quantification of % bind-ing of DOX

Considering the biocompatible nature, we havedesigned an AuHS nanoparticles based DDS (AuHS-DOX), containing DOX as an FDA approved anti-cancer drug and hypothesized that AuHS nanoparticlescould be an effective delivery vehicles for the treat-ment of cancers. Prior to investigate the therapeuticefficacy of the DDS, it was highly essential to checkthe % attachment of DOX in AuHS-DOX DDS. Thespectrofluorimetry analysis using supernatant of AuHS-DOX and standard curve of pristine DOX (Fig. 10)reveals that % binding of DOX in the DDS was~41%. It seems that the attachment of DOX withAuHS nanoparticles might be due to the weak elec-trostatic attraction, as these nanoparticles possess nega-tive surface charge while DOX has positive zeta po-tential according to earlier literature17,38,39.

Therapeutic efficacy of AuHS-DOX DDS

To confirm our hypothesis that AuHS could be aneffective delivery vehicle of DOX for cancer treat-

Fig. 8. In vitro cell viability assay in differnet cancer cells[(a) HCT-15, (b) HeLa, (c) PC3 and (d) A549]. Theresult exhibits the biocompatibility of both the HSextract (0.5–5 L; 24 h) and AuHS nanoparticles (0.5–5 L; 24 h). The numerical values indicate the vol-ume in L.

Fig. 9. In vivo CEA assay employing AuHS nanoparticles (5 L, 4 h). The result illustrates that AuHS nanoparticles did notaffect the blood vessel growth in terms of vessel length, junction and size, supporting their biocompatible nature.

compatible AuHS nanopaticles for different biomedi-cal applications.

Colour


1345

ment, cell viability assay was again performed em-ploying the DDS in different cancer cells (HCT-15 :Fig. 11a, HeLa : Fig. 11b and PC-3 : Fig. 11c). Theresult exhibits that the administration of AuHS-DOXsignificantly inhibited the cancer cell proliferation com-pared to pristine DOX in a dose dependent mannerirrespective of cancer cell line, suggesting the thera-peutic potential of the DDS. Considering thebiocompatible nature of AuHS nanoparticles and thebetter antiproliferative effect of AuHS-DOX to can-cer cells, it could be speculated that the DDS couldbe useful for the anticancer therapy in forthcomingfuture.

Cellular uptake studies using confocal microscopy

The higher therapeutic potential of DDS comparedto free drug might often be due to the better cellularuptake of the DDS40. To confirm this fact, confocalmicroscopy was employed for comprehending theuptake of AuHS-DOX in HCT-15 cells in terms offluorescence intensity of DOX, present in the DDS.Fig. 12 and Fig. 13 reveals that untreated controlcells and cell treated with AuHS nanoparticles did notexhibit any fluorescence. However, cells incubatedwith AuHS-DOX showed red fluorescence with higherintensity as compared to cells treated with free DOX,

Fig. 10. Standard curve of doxorubicin (DOX) for evaluationof % binding of DOX in AuHS-DOX DDS usingspectrofluorimetry technique.

Fig. 11. Therapeutic potential of AuHS-DOX DDS in differ-ent cancer cells [(a) HCT-15, (b) HeLa and (c) PC3].The result exhibits that AuHS-DOX (0.03–0.25 Mw.r.to DOX; 48 h) treatment significantly inhibitsthe proliferation of cancer cells as compared to freeDOX (0.03–0.25 M; 48 h) in a dose dependent man-ner.

indicating the better cellular uptake of the DDS thanthat of pristine drug. The enhanced uptake of AuHS-DOX compared to free DOX might be due to theEPR effect in cancer cells41. The results altogethercorroborate with the better anti-proliferative efficacyof AuHS-DOX compared to free DOX to cancer cells.

Plausible mechanism for synthes is of AuHSnanoparticles

There are numerous reports for the synthesis ofgold nanoparticles using different plant extracts. How-ever, precise mechanism for the fabrication of goldnanoparticles using this strategy still remains ambigu-ous. As per the earlier literature, plant extract often

JICS-12


1346

(Au0)47. Moreover, Newman et al. demonstrated theformation of gold nanoparticles through the reductionof the HAuCl4 to elemental gold (Au0) by amine/amide compounds present in the plant extract. Thisredox reaction might take place through an electrontransfer mechanism between amine to the metal ion46.Earlier reports also suggested that low (10–22 kDa)as well as high molecular weight proteins (100–200kDa) present in leaf extract might be one of the rea-son of the reducing properties of the extract17,29.However, the reducing potential of these proteins mayvary depending on the nature of the plant. FT-IRanalysis also supports that HS extract possess diffe-rent phytochemicals (e.g. polyphenol compounds,alcohols, amines etc.) which are responsible for thesynthesis and stabilization of AuHS nanoparticles.Scheme 2 represents the overall mechanism for thesynthesis and stabilization of AuHS nanoparticles us-ing HS leaf extract.

Fig. 12. Cellular uptake study in HCT-15 cells using confocalmicroscope with 40x objective with zoom-2.331. Theresult reveals the enhanced uptake of DOX present inAuHS-DOX (0.12 M w.r.to DOX) as compared tofree DOX (0.12 M) at same treatment dose. Scalebar : 10 m.

Fig. 13. Cellular uptake study in HCT-15 cells using confocalmicroscope with 40x objective (without any zoom-ing). The result reveals the enhanced uptake of DOXpresent in AuHS-DOX (0.12 M w.r.to DOX) ascompared to free DOX (0.12 M) at same treatmentdose. Scale bar : 50 m.

augments the reduction of HAuCl4 for the formationas well as stabilization of gold nanoparticles17,22,30,42.In the present study, we have synthesized AuHSnanoparticles using aqueous HS leaf extract, whichserved as both reducing and stabilizing agent. HScontains several polyphenolic/alcoholic compounds(e.g. ellacgic acids, catechin, hydroxycitric acid, -sitosteryl--D-galactoside etc.), amide/amine com-pounds, aromatic ketones, aldehydes etc. which mightbe responsible for the reduction of Au3+ (HAuCl4)to Au0 (AuHS : gold nanoparticles) as well as theirstabilization43–46. The standard reduction potential forAu3+/Au0 system (E0 Au3+/Au0) is 1.50 eV whichis higher compared to that of acid/aldehyde, alde-hyde/alcohol system (E0 ROH/RCO/RCHO : 0.80eV) and so polyphenolic/alcoholic phytochemicalspresent in HS extract could easily augment the reduc-tion of HAuCl4 (Au3+) to AuHS nanoparticles

Colour


1347

Conclusions

Through the use of a bio-green method, goldnanoparticles (AuHS) were fabricated under an ambi-ent conditions using HS leaf extract, which acts as areducing as well as stabilizing agent. The as-synthe-sized AuHS nanoparticles were characterized by sev-eral analytical methods. AuHS nanoparticles werefound to be stable in different solutions/physiologicalbuffers indicating their feasibility for biological ap-plications. Cell viability assay and CEA assay indi-cate the biocompatible nature of AuHS nanoparticles.Considering the biocompatibility of AuHS nano-particles, AuHS-DOX DDS was designed where AuHSnanoparticles act as delivery vehicle and DOX as anFDA approved anti-cancer drug. The administrationof the DDS exhibited more inhibition of cancer cellproliferation compared to free DOX, suggesting theefficacy of the DDS. Additionally, confocal micros-copy study reveals the enhanced cellular uptake ofAuHS-DOX DDS compared to free DOX, support-ing the therapeutic potential of the DDS. This studyprovides the basis for the development of otherbiocompatible gold nanoparticles that might be usedfor effective drug delivery vehicle for the treatmentof cancers in near future.

AcknowledgementsCRP is grateful to DST, New Delhi for generous

financial support by NanoMission project (SR/NM/NS-1252/2013; GAP0570). AKB and SD are gratefulto UGC, New Delhi for their senior research fellow-ships. The authors are thankful to Analytical Divi-sion, CSIR-IICT for performing ICPOES analysis(AARF project, CSIR 12th FYP) in detection of metalcontent in gold nanoparticles samples.

References

1. P. Alivisatos, Nature Biotechnology, 2004, 22, 47.

2. E. Boisselier and D. Astruc, Chemical Society Reviews,2009, 38, 1759.

3. X. Gao, Y. Cui, R. M. Levenson, L. W. K. Chung andS. Nie, Nature Biotechnology, 2004, 22, 969.

4. C. R. Patra, R. Bhattacharya, E. Wang, A. Katarya, J.S. Lau, S. Dutta, M. Muders, S. Wang, S. A. Buhrow,S. L. Safgren, M. J. Yaszemski, J. M. Reid, M. M.Ames, P. Mukherjee and D. Mukhopadhyay, CancerRes., 2008, 68, 1970.

5. C. R. Patra, R. Bhattacharya, S. Patra, N. E. Vlahakis,A. Gabashvili, Y. Koltypin, A. Gedanken, P. Mukherjeeand D. Mukhopadhyay, Advanced Materials, 2008, 20,753.

6. W. J. Mulder and Z. A. Fayad, Arterioscler. Thromb.Vasc. Biol., 2008, 28, 801.

7. P. F. Wei, L. Zhang, S. K. Nethi, A. K. Barui, J. Lin,

Scheme 2. The plausible mechanism for the generation and stabilization of AuHS nanoparticles using Hibiscus sabdariffa plant extract.

Colour


1348

W. Zhou, Y. Shen, N. Man, Y. J. Zhang, J. Xu, C. R.Patra and L. P. Wen, Biomaterials, 2014, 35, 899.

8. H. W. Sung, K. Sonaje and S. S. Feng, Nanomedicine,2011, 6, 1297.

9. C. R. Patra, R. Bhattacharya, D. Mukhopadhyay andP. Mukherjee, Adv. Drug Deliv. Rev., 2010, 62, 346.

10. M. C. Daniel and D. Astruc, Chem. Rev., 2004,104, 293.

11. R. Arvizo, R. Bhattacharya and P. Mukherjee,Expert Opinion on Drug Delivery, 2010, 7, 753.

12. D. A. Giljohann, D. S. Seferos, W. L. Daniel,M. D. Massich, P. C. Patel and C. A. Mirkin,Angew. Chem. Int. Ed., 2010, 49, 3280.

13. K. J. Savage, M. M. Hawkeye, R. Esteban, A. G.Borisov, J. Aizpurua and J. J. Baumberg, Nature,2012, 491, 574.

14. L. Lu, G. Sun, H. Zhang, H. Wang, S. Xi, J.Hu, Z. Tian and R. Chen, J. Mater. Chem.,2004, 14, 1005.

15. K. G. Lee, J. Hong, K. W. Wang, N. S. Heo, H.Kim do, S. Y. Lee, S. J. Lee and T. J. Park, ACSNano, 2012, 6, 6998.

16. S. D. Haveli, P. Walter, G. Patriarche, J. Ayache,J. Castaing, E. V. Elslande, G. Tsoucaris, P. A.Wang and H. B. Kagan, Nano Lett., 2012, 12,6212.

17. S. Mukherjee, V. Sushma, S. Patra, A. K. Barui,M. P. Bhadra, B. Sreedhar and C. R. Patra,Nanotechnology, 2012, 23, 455103.

18. R. Mie, M. W. Samsudin and L. B. Din, Nano-synthesis and Nanodevice, 2013, 667, 251.

19. C. Liu, Y. G. Wang, J. Q. Geng, Z. Y. Jiang andD. Yang, Progress in Chemistry, 2011, 23, 2510.

20. S. Iravani, Green Chemistry, 2011, 13, 2638.

21. S. Mukherjee, B. Vinothkumar, S. Prashanthi, P.R. Bangal, B. Sreedhar and C. R. Patra, RSCAdv., 2013, 3, 2318.

22. R. Shukla, K. S. Nune, N. Chanda, K. Katti, S.Mekapothula, R. R. Kulkarni, V. W. Welshons,R. Kannan and V. K. Katti, Small, 2008, 4, 1425.

23. S. Mukherjee, D. Chowdhury, R. Kotcherlakota,S. Patra, V. B. M. P. Bhadra, B. Sreedhar and C.R. Patra, Theranostics, 2014, 4, 316.

24. P. Mishra, S. Ray, S. Sinha, B. Das, M. I. Khan,S. K. Behera, S.-I. Yun, S. K. Tripathy and A.Mishra, Biochemical Engineering Journal, 2016,105, 264.

25. A. K. Barui, S. K. Nethi and C. R. Patra, J.Mater. Chem. (B), 2017, 5, 3391.

26. P. Nagababu, A. K. Barui, B. Thulasiram, C. S.Devi, S. Satyanarayana, C. R. Patra and B.Sreedhar, J. Med. Chem., 2015, 58, 5226.

27. R. Kumar, S. Das, S. Mukherjee, R. S. Bhosale,C. R. Patra and R. Narayan, Mater. Sci. Eng.(C), 2017, 81, 580.

28. A. K. Barui, V. Veeriah, S. Mukherjee, J. Manna,A. K. Patel, S. Patra, K. Pal, S. Murali, R. K.Rana, S. Chatterjee and C. R. Patra, Nanoscale,2012, 4, 7861.

29. S. Patra, S. Mukherjee, A. K. Barui, A. Ganguly,B. Sreedhar and C. R. Patra, Mater. Sci. Eng.(C), 2015, 53, 298.

30. J. Lee, H. Y. Kim, H. Zhou, S. Hwang, K. Koh,D. W. Han and J. Lee, J. Mater. Chem., 2011,21, 13316.

31. Z. Krpetic, G. Scari, C. Caneva, G. Speranza andF. Porta, Langmuir, 2009, 25, 7217.

32. G. Jiang, L. Wang and W. Chen, Materials Let-ters, 2007, 61, 278.

33. L. Rastogi and J. Arunachalam, Materials Chemis-try and Physics, 2011, 129, 558.

34. A. M. Awwad, N. M. Salem and A. O. Abdeen,Int. J. Ind. Chem., 2013, 4, 29.

35. R. K. Das, N. Gogoi and U. Bora, Bioprocessand Biosystems Engineering, 2011, 34, 615.

36. C. R. Patra, R. Bhattacharya and P. Mukherjee,J. Mater. Chem., 2010, 20, 547.

37. J. C. Stockert, A. Blázquez-Castro, M. Cañete,R. W. Horobin and Á. Villanueva, Acta Histo-chemica, 2012, 114, 785.

38. A. Z. Mirza and H. Shamshad, European Journalof Medicinal Chemistry, 2011, 46 1857.

39. A. Modak, A. K. Barui, C. Ranjan Patra and A.Bhaumik, Chem. Commun., 2013, 49, 7644.

40. H. C. Arora, M. P. Jensen, Y. Yuan, A. Wu, S.Vogt, T. Paunesku and G. E. Woloschak, CancerRes., 2012, 72, 769.

41. Y. Nakamura, A. Mochida, P. L. Choyke and H.Kobayashi, Bioconjug. Chem., 2016, 27, 2225.

42. S. K. Nethi, S. Mukherjee, V. Veeriah, A. K.Barui, S. Chatterjee and C. R. Patra, Chem.Commun., 2014, 50, 14367.

43. I. Da-Costa-Rocha, B. Bonnlaender, H. Sievers, I.Pischel and M. Heinrich, Food Chemistry, 2014,165, 424.

44. H.-H. Lin, K.-C. Chan, J.-Y. Sheu, S.-W. Hsuan,C.-J. Wang and J.-H. Chen, Food Chemistry,2012, 132, 880.

45. A. Ismail, E. H. K. Ikram and H. S. M. Nazri,Food, 2008, 2, 1.

46. S. D. J. Newman and J. G. Blanchard, Langmuir,2006, 22, 5882.

47. S. A. Korchev, S. T. Shulyak, L. B. Slaten, F.W. Gale and G. Mills, J. Phys. Chem. (B), 2005,109, 7733.

1349


Study of molecular interactions of copper oxide nanoparticles with 10% aqueousethylene glycol at different temperatures : Using physicochemical method†

Mukesh Kumar, Neha Sawhney, Samriti Sharma, Anjali Anand, Amit Kumar Sharma andMeena Sharma*

Department of Chemistry, University of Jammu, Jammu-180 006, Jammu & Kashmir, India



Abstract : In this investigation copper oxide (CuO) nanoparticles were synthesized by precipitation methodusing [Cu(CH3COO)2.2H2O] as a starting material. The synthesized nanoparticles were characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscopy (SEM) and en-ergy dispersive X-ray spectroscopy (EDX). The synthesized nanoparticles were dispersed in an aqueous solu-tion of ethylene glycol in various concentrations with the help of ultrasonicator to obtain nanofluids of differ-ent concentrations. Density (), ultrasonic velocity (u) and viscosity () for these nanofluids were experimen-tally measured as a function of temperatures (T = 303.15 K, 308.15 K and 313.15 K). Using these experi-mentally determined values various acoustic and thermodynamic parameters were evaluated.

Keywords : CuO nanoparticles, base fluid, ultrasonic velocity, density, viscosity, acoustical parameters.

Introduction

In the past decades, metal oxide nanoparticles areextensively used in wide range of technological ap-plications. Transition metal oxide nanoparticles playa key role in many areas of research (i.e. in chemis-try, physics and material science1). The coinage metaloxide nano-particles exhibit unique chemical proper-ties because of their small size and high density2.Nanoparticles have wide range of applications in mis-cellaneous fields, including chemical manufacturing,energy conversion and storage, biological applicationsand environmental technology3,4.

Lot of work on nanohybrids deals with solid statematerials. In 1995 Choi, gave concept of nanofluids5.Nanofluids containing small amount of metal or metaloxide nanoparticles reveal considerably better ther-mal conductivity compared to the base fluids6. Ther-mal conductivity of nanofluid plays imperative rolein the development of energy-efficient heat transfer

†Professor R. S. Varma Memorial Lecture (2017).

appliances. The thermal conductivities of various flu-ids like water, ethylene glycol and engine oil are com-paratively lower than those of solid particles. Now aday, there is strong need of development of superiorheat transfer fluids with higher thermal conductivity.When a small amount of nanoparticle is disperseduniformly in a host fluid (e.g. heat transfer fluids likewater, ethylene glycol, etc.) this results spectacularenhancement in the thermal properties of host fluids.The major problem arises in this suspension is theinstability of nanofluids, this is due to the agglomera-tion of nanoparticles dispersed in base fluids. Thus inorder to prepare stable nano-suspension ultrasonicationplays a very important role to disperse aggregatednanoparticles7. In our present work CuO nanoparticleswere synthesized and dispersed in base fluids. Amongtransition metal oxides, CuO nanoparticles were cho-sen because of the wide range of applications in hightemperature superconductors, metallurgy and in heat


1350

transfer applications8–10. CuO is a p-type semicon-ductor like CuS and can be used in various electronicdevices11,12. Various methods reported in literaturefor the synthesis of CuO NPs like electrochemicalmethod, one-step solid state reaction method, sol-geltechnique, sonochemical method, thermal decompo-sition of precursors, etc.

Ultrasonic velocity, density and rheological stud-ies play an imperative role in understanding solute-solvent, solvent-solvent and solute-solute interactionsin aqueous and non-aqueous medium13,14.

Although the thermal conductivity is the most in-teresting property for practical purposes, less atten-tion has been paid so far to other properties such asdensity and viscosity of nanofluids. Cheng et al.pointed out that these properties should be determinedfor their effects on nanofluid flow and heat transfercharacteristics15. A comparison between the experi-mental viscosity ratios and results predicted byBatchelor and Wang models for magnetite nanofluidswas reported by Toghraie et al. and they observedthat, theoretical models are unable to predict visco-sity of the magnetite nanofluids16. The dynamic vis-cosity of alumina-engine oil nanofluid in different solidvolume fractions and temperature was experimentallyinvestigated by Esfe et al.17. Esfe et al. performedZnO-EG nanofluids thermal conductivity using MLPneural networks. They showed that neural networkscan estimate the experimental results with a high pre-cision18. An examination of viscosity of MWCNTs/ZnO-SAE40 hybrid nano-lubricants under varioustemperatures and solid volume fractions is presentedby Hemmat Esfe et al. Viscosity measurements showedthat the viscosity decreases with increasing tempera-ture and increases with an enhancement in the solidvolume fraction19.

Close examination of the literature indicates thatonly some authors studied the ultrasonic properties ofnanofluids20,21. Despite recent advances, much moreworks involving theoretical, experimental and numeri-cal research are necessary to solve the mysteries ofnanofluids.

In recent years, many researchers have created

new kinds of fluids by dispersion of nanosized par-ticles in pure water or ethylene glycol. However, fewstudies have been reported about nanofluids whosebase fluid is ethylene glycol and water mixture22–27.Similarly very few studies are available about thenanofluids of propylene glycol or hexylene glycol andwater mixtures as base fluids28.

In this paper study on the response of CuOnanoparticles in basefluid to the ultrasonic wave propa-gation for the basic understanding how thesenanoparticles behave in fluids and how they interactwith each other and with fluid was carried out. Stablesuspensions of nanoparticles were prepared usingultrasonication and were used for attaining a deeperunderstanding of interparticle (particle-particle) andparticle-fluid interactions as functions of concentra-tion and temperature.

Experimental

Synthesis of nanoparticles :

For the synthesis of copper oxide nanoparticles,cupric acetate dehydrate [Cu(CH3COO)2.2H2O], gla-cial acetic acid (CH3COOH), sodium hydroxide(NaOH), acetone and ethanol are procured from SigmaAldrich. All the chemicals are of GR-grade and usedwithout any further purification. The synthesis ofcopper oxide nanoparticles is carried out by precipi-tating copper salt in alkaline medium29. The coppersalt solution used is freshly prepared 0.2 M Cu(CH3COO)2.2H2O. The salt solution is mixed with1 ml glacial acetic acid and the resultant solution isheated at a constant stirring speed of 1000 rpm untilrequired temperature of 60 ºC is reached. Highertemperature is favoured because of higher reactionrates, which produces large amounts of nuclei to formin a short time, and the aggregation of crystals isinhibited. Glacial acetic acid is used to prevent thehydrolysis of the copper acetate solution. On vigor-ous stirring, the above solution pH is increased rap-idly to 10.5 by adding NaOH pellets where a blackprecipitate of CuO is formed instantly. At the samepH, temperature and stirring speed, the solution iskept at a digestion time of 30 min. Overall chemicalreaction can be written as

Kumar et al. : Study of molecular interactions of copper oxide nanoparticles with 10% aqueous etc.

1351

Cu(CH3COO)2 + 2NaOH CuO + 2CH3COONa+ H2O

After cooling to room temperature, particles are sepa-rated from the dispersion. The top water layer withexcess salts is discarded. The particles are washedwith water, ethanol and acetone. They are separatedfrom dispersion by centrifugation at 2500 rpm for 30min and dried at room temperature in an inert atmo-sphere.

Preparation of nanofluids :

The nanopowder of CuO thus obtained was dis-persed in base fluid, an aqueous solution of ethyleneglycol, to obtain nanofluids of different concentra-tions. Weighing of nanoparticles powder was deter-mined using electronic single pan four digit MettlerToledo balance (Model ML204) with an accuracy of0.1 mg. To achieve uniform dispersion of the par-ticles ultrasonic wave of frequency 4 MHz was passedthrough the nanofluid for 2 h with the help ofultrasonicator. It is evident that the ultrasonic treat-ment to the fluids increases the settling time of thesuspension.

Characterization :

The synthesized nanoparticles were characterizedby using various techniques like X-ray diffraction(XRD), transmission electron microscopy (TEM),scanning electron microscopy (SEM) and energy dis-persive X-ray spectroscopy (EDX). In the presentstudy, Bruker AXS D8 Advance X-ray diffractometerwas used to obtain X-ray diffraction patterns of thesamples. JEOL JEM-2100 LaB6 was used for TEMstudies.

To examine the morphology of the synthesizednanoparticles, SEM analysis was carried out on theJEOL JSM-6390LV SEM fitted with secondary elec-tron detector, and equipped with an attachment forthe energy dispersive X-ray spectroscopy (EDX) toenable elemental composition analysis.

Dynamic light scattering (DLS) was used to findthe average hydrodynamic size of the synthesized CuOnanoparticles.


The characterization of nanoparticles was done byvarious techniques as described above and thesenanoparticles were dispersed in base fluids. The val-ues of ultrasonic velocity, density and viscosity ofnanofluids were experimental determined. These val-ues help us to evaluate various acoustical parameters.

Characterization :

The XRD pattern of synthesized nanoparticles isdepicted in Fig. 1.

Fig. 1. XRD pattern of CuO nanoparticles.

All diffraction peaks can be indexed as the mono-clinic crystal structure of CuO (space group C2/c) bycomparison with data from JCPDS file no. 48-1548with lattice constants a = 4.6837 Å, b = 3.4226 Å,c = 5.1288 Å and = 90°, = 99.54°, = 90°no characteristic peaks of any other impurity wereobserved. The sharp peaks indicate that the productwas well crystallized. Debye-Scherrer equation wasused for calculating the crystallite size. The crystal-lite size corresponding to most intense peak was foundto be 14.692(0.005) nm corresponding to 2 value of38.7832 and hkl (111). TEM micrograph of the syn-thesized nanoparticles is depicted in Fig. 2.

Average diameter of the nanoparticles was 15-20nm evaluated by ImageJ software. SEM and EDX ofthe synthesized nanoparticles are depicted in Fig. 3and Fig. 4, respectively. SEM image showed that theobtained nanoparticles were agglomerated to a largeextent and the particles have flower like morphology.

The stoichiometry of sample was examined by EDX


1352

spectrum. Only Cu and O signals have been detected,suggesting that the nanoparticles were only made upof Cu and O. Weight percentage of Cu was found tobe 78.95 and that of O was 19.68. Thus the atomicratio of Cu and O was 1 : 1.

ences the movement of the particle in the medium.Thus the hydrodynamic diameter gives us informa-tion of the inorganic core along with any coatingmaterial and the solvent layer attached to the particleas it moves under the influence of Brownian motion.While estimating size by TEM, this hydration layer isnot present hence, we get information only about theinorganic core. Size distribution by intensity of dif-ferent concentrations of CuO nanoparticles dispersedin 10% aqueous solution of ethylene glycol are shownin Figs. 5–9. In case of nanofluids having concentra-tion of CuO between 0.02–0.08 wt% the result qual-ity was found to be good which means the particleswere well dispersed and not agglomerated as shownin Figs. 5–8. But in case of nanofluids having 0.10

Fig. 2. TEM micrograph of CuO nanoparticles.

Fig. 3. SEM image of CuO nanoparticles.

Dynamic light scattering was used to find the av-erage hydrodynamic size of the synthesized CuOnanoparticles. The size determined by DLS was foundto be much larger than the size determined by XRDand TEM. By DLS we get the hydrodynamic radiusof the particle whereas by TEM we get an estimationof the projected area diameter. When a dispersedparticle moves through a liquid medium, a thin layerof the solvent adheres to its surface. This layer influ-

Fig. 4. EDX spectrum of CuO nanoparticles.

Fig. 5. Size distribution by intensity of 0.02 wt% CuOnanoparticles in 10% aqueous ethylene glycol.


1353

wt% concentration of CuO the result quality was notfound to be good and z-average was abruptly increasesto 1061 nm which indicates agglomeration ofnanoparticles.

Acoustical studies of nanofluids :

An ultrasonic interferometer provided by MittalEnterprises, New Delhi was used to evaluate ultra-sonic velocity of nanofluids. Ultrasonic velocity wasmeasured at a frequency of 5 MHz. Viscosity wasmeasured by using Ostwald’s viscometer. The den-sity of various nanofluids was automatically measuredusing Anton Paar densimeter (DMA 5000 M, Aus-

tria). The temperature was automatically controlledto ±1×10–3 K by a built-in Peltier device. All thesemeasurements were performed for the nanofluids ofdifferent concentrations at three different tempera-tures 303.15 K, 308.15 K and 313.15 K. By usingthese experimentally determined values various acous-tical parameters were evaluated, that gives informa-tion regarding type of interactions present in our sys-tem. Table 1 shows the experimentally determinedvalues of ultrasonic velocity (u), density () and vis-cosity () of nanofluids. From the data recorded inTable 1, it has been observed that there occurs an





JICS-13


1354

increase in ultrasonic velocity (Fig. 10) and viscosity(Fig. 12) upto some concentration of CuOnanoparticles after that their values start decreasingand finally a little increase occurs in these values.

By using these experimentally determined valuesvarious acoustic parameters were evaluated like adia-batic compressibility (ad), intermolecular free length(Lf), relaxation time () and attenuation coefficient(/f2). Adiabatic compressibility (ad) of the sampleswas calculated from the speed of sound (u) and thedensity of the medium () using the Newton Laplacerelation30,31.

ad = 1/u2 (1)

The intermolecular free length (Lf) was calculated usingthe following formula given by Jacobson32,33

Lf = KTad1/2 (2)

where, KT is Jacobson’s constant.

The relaxation time ( ) can be calculated from therelation

= (4/3)ad (3)

Attenuation coefficient (/f 2) was evaluated from thevalues of speed of sound and relaxation time by usingStokes relation34

/f 2 = 4n2 /2u (4)

Table 2 shows the values of adiabatic compressibi-lity, intermolecular free length, relaxation time andattenuation coefficient.

The increase in ultrasonic velocity and viscosity atlower concentration of CuO nanoparticles is due tointermolecular hydrogen bonding among CuOnanoparticles, ethylene glycol and water molecules

Table 1. Ultrasonic velocity (u), density () and viscosity() of nanofluids having different concentrations of CuO in

10% aq. ethylene glycol at three different temperatures

Temp. Conc. u×10–3 ×10–3 ×10–3

(K) (wt%) (m s–1) (kg m–3) (kg m–1s–1)

303.15 0 1.54909 1.00911 1.14316

0.02 1.55145 1.00962 1.17486

0.04 1.55373 1.00972 1.19975

0.06 1.55138 1.01015 1.17785

0.08 1.54918 1.01073 1.15719

0.10 1.55150 1.01046 1.17405

308.15 0 1.55528 1.00726 1.05367

0.02 1 .55736 1.00779 1.08517

0.04 1.55989 1.00789 1.11345

0.06 1.55796 1.00831 1.09189

0.08 1.55582 1.00891 1.07273

0.10 1.55797 1.00861 1.09291

313.15 0 1.56180 1.00523 0.98043

0.02 1.56362 1.00576 1.01597

0.04 1.56582 1.00587 1.03555

0.06 1.56336 1.00628 1.01469

0.08 1.56178 1.00693 0.99701

0.10 1.56364 1.00658 1.01053

Fig. 11. Plots of density of various nanofluids of CuO in 10%aqueous ethylene glycol at different temperatures.

Fig. 10. Plots of ultrasonic velocity of various nanofluids ofCuO in 10% aqueous ethylene glycol at different tem-peratures.


1355

interactions with glycol (CuO-EG) and water mol-ecules (CuO-H2O) increases which in turn start de-creasing the hydrogen bonding between glycol andwater molecules (EG-H2O) as a result overall com-

Fig. 12. Plots of viscosity of various nanofluids of CuO in10% aqueous ethylene glycol at different tempera-tures.

Table 2. Adiabatic compressibility (ad), intermolecular free length (Lf), relaxation time ( ), and attenuation coefficient(/f2) of nanofluids having different concentrations of CuO in 10% aq. ethylene glycol at three different temperatures

Temp. Conc. ad×1010 Lf×1011 ×1013 /f2× 10–14

(K) (wt%) (m kg–1 s) (m) (s) (s2 m–1)

303.15 0 4.12961 4.21670 6.29441 8.01248

0.02 4.11497 4.20922 6.44602 8.19300

0.04 4.10250 4.20283 6.56263 8.32898

0.06 4.11320 4.20831 6.45964 8.21068

0.08 4.12250 4.21307 6.36069 8.09640

0.10 4.11130 4.20734 6.43583 8.17978

308.15 0 4.10433 4.24176 5.76612 7.31079

0.02 4.09121 4.23497 5.91955 7.49529

0.04 4.07755 4.22790 6.05354 7.65252

0.06 4.08596 4.23225 5.94856 7.52914

0.08 4.09478 4.23682 5.85676 7.42314

0.10 4.08469 4.23160 5.95226 7.53377

313.15 0 4.07836 4.26618 5.33138 6.73137

0.02 4.06671 4.26009 5.50887 6.94738

0.04 4.05487 4.25388 5.59870 7.05075

0.06 4.06594 4.25968 5.50092 6.93851

0.08 4.07156 4.26263 5.41252 6.83390

0.10 4.06329 4.25829 5.47476 6.90427

(CuO-EG-H2O) as a result of this bonding overallcompressibility of the medium decreases and henceultrasonic velocity and viscosity increases. But abovea particular concentration of CuO nanoparticles their

Fig. 13. Plots of adiabatic compressibility of various nanofluidsof CuO in 10% aqueous ethylene glycol at differenttemperatures.


1356

pressibility of the medium increases (Fig. 10) andhence ultrasonic velocity and viscosity decreases. Atsome higher concentration (0.10 wt%) of CuOnanoparticles they start aggregating due to inter par-ticle interactions (CuO-CuO) and hence their interac-tions with glycol (CuO-EG) and water molecules(CuO-H2O) decrease. Due to this the hydrogen bond-ing between glycol and water molecules (EG-H2O)increases and hence a little increase occurs in thevalues of ultrasonic velocity and viscosity. There oc-curs an increase in density (Fig. 11) with increase inconcentration of CuO nanoparticles. The variation of

ultrasonic velocity in any solution depends upon theincrease or decrease of intermolecular free lengthsafter mixing the component. As the free length de-creases (Fig. 14) due to the increase in concentra-tions of CuO nanoparticles upto some concentration,ultrasonic velocity increases, it indicates significantinteractions between the solute-solvent molecules sug-gesting a structure promoting tendency of the addedsolute. But at some higher concentration of CuOnanoparticles, free length starts increasing due to de-creasing interactions between glycol and water mol-ecules hence ultrasonic velocity start decreasing. Atstill higher concentration of nanoparticles due to interparticle interactions the glycol water interactions in-creases indicated by decrease in intermolecular freelength and increase in ultrasonic velocity. Adiabaticcompressibility also shows the same trend as that ofintermolecular free length with increase in concentra-tion of nanoparticles. Density shows an increase withincrease in concentration of nanoparticles. The trendsof variation of relaxation time (Fig. 15) and attenua-tion coefficient (Fig. 16) with increase in concentra-tion of nanoparticles are same as that of ultrasonicvelocity and viscosity.

ConclusionUltrasonic velocity, density, viscosity, relaxation

time and attenuation coefficient showed an increase

Fig. 14. Plots of intermolecular free length of variousnanofluids of CuO in 10% aqueous ethylene glycol atdifferent temperatures.

Fig. 15. Plots of relaxation time of various nanofluids of CuOin 10% aqueous ethylene glycol at different tempera-tures.

Fig. 16. Plots of attenuation coefficient of various nanofluidsof CuO in 10% aqueous ethylene glycol at differenttemperatures.


1357

with increase in concentration of CuO nanoparticlesupto 0.04 wt% because of increasing molecular inter-actions among CuO, water and ethylene glycol (CuO-EG-H2O). After that these parameters showed a de-crease with increase in concentration of CuOnanoparticles upto 0.08 wt% because of increasingmolecular interactions of CuO with water (CuO-H2O)and ethylene glycol (CuO-EG) and decreasing mo-lecular interactions between water and ethylene gly-col (H2O-EG). On further increasing the concentra-tion of CuO nanoparticles these parameters start in-creasing because of agglomeration of nanoparticleswhich decreases the molecular interactions of CuOwith water (CuO-H2O) and ethylene glycol (CuO-EG). Adiabatic compressibility and intermolecular freelength showed an opposite trend with increase in con-centration of CuO nanoparticles. It has also been ob-served that with rise in temperature there occurs anincrease in ultrasonic velocity and intermolecular freelength while density, viscosity, adiabatic compres-sibility, relaxation time and attenuation coefficientshowed a decreasing trend.

Thus such interactions of nanoparticles with basefluids (particle-fluid interactions) are very helpful inunderstanding the reasons behind the abnormal en-hancements in physical properties and to understandthe mechanism of transport of nanoparticles in fluids.The knowledge to these particle-fluid interactions willhelp to improve the performance of these nanofluidsin MEMS and other practical applications.

AcknowledgementsThe authors are grateful to SAIF Cochin for pro-

viding XRD, SEM-EDX and TEM analysis facilityand Department of Chemistry, University of Jammufor providing all other required facilities to carry outthis work.

References

1. M. Fernandez-Garcia, A. Martinez-Arias, J. C. Hansonand J. A. Rodriguez, Chem. Rev., 2004, 104, 4063.

2. T. Tsoncheva, J. Rooggenbuck, D. Paneva, M. Dimitrov,I. Mitov and M. Frobab, Appl. Surf. Sci., 2010, 257,523.

3. S. Laurent, D. Forge, M. Port, A. Roch, C. Robic, L.Vander Elst and R. N. Muller, Chem. Rev., 2008, 108,

2064.

4. M. B. Gawande, P. S. Branco, K. Parghi, J. J.Shrikhande, R. K. Pandey, C. A. A. Ghumman, N.Bundaleski, O. Teodoro and R. V. Jayaram, Sci.Technol., 2011, 1, 1653.

5. S. U. S. Choi, ASME Fed 231/MD, 1995, 66, 99.

6. J. A. Eastman, S. U. S. Choi, S. Li, W. Yu and L.Thompson, J. Appl. Phys. Lett., 2001, 78, 718.

7. B. Ruan and A. N. Jacobi, Nano Scale Res. Lett., 2012,7, 127.

8. P. O. Larsson and A. Andersson, J. Catal., 1998, 179,72.

9. C. P. Poole, T. Datta, H. A. Farach, M. M. Rigneyand C. R. Sanders, "Copper Oxide Superconductors",John Wiley & Sons, New York, 1988.

10. V. Chikan, A. Molnar and K. Balazsik, J. Catal.,1999, 184, 134.

11. O. Zabihi and S. Ghasemlou, J. Poly. Analy.Characterization, 2012, 17, 108.

12. C. Xu, Z. Zhang, Q. Ye and X. Liu, Chem.Lett., 2003, 32, 198.

13. S. Balija and S. Oza, Fluid Phase Equilibria,2002, 200, 11.

14. M. K. Rawat and Sangeeta, Indian J. Pure Appl.Phys., 2008, 46, 187.

15. L. Cheng, F. Bandarra, P. Enio and J. R. Thome,J. Nanosci. Nanotechnol., 2008, 8, 3315.

16. D. Toghraie, S. M. Alempour and M. J. Afrand,Magn. Mater., 2016, 417, 243.

17. M. H. Esfe, M. Afrand, D. Toghraie and S. H.Rostamian, Int Commun. Heat Mass Transf.,http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.05.013

18. M. H. Esfe, M. Afrand, S. Wongwises, A. Naderi,A. Asadi, S. H. Rostamin and M. Akbari, Int.Commun. Heat Mass Transf., 2015, 67, 46.

19. M. H. Esfe, M. Afrand, S. H. Rostamian and D.Toghraie, Exp. Therm. Fluid Sci., 2017, 80, 384.

20. R. R. Yadav, G. Mishra, P. K. Yadawa, S. K.Kor, A. K. Gupta, B. Raj and T. Jayakumar, Ul-trasonics, 2008, 48, 591.

21. B. Tajik, A. Abbassi, M. Saffar-Avval and M. A.Najafabadi, Powder Technology, 2012, 217, 171.

22. R. S. Vajjha, D. K. Das and B. M. Mahagaonkar,Pet. Sci. Technol., 2009, 27, 612.

23. M. P. Beck, Y. Yuan, P. Warrier and A. S. Teja,J. Nanopart. Res., 2010, 12, 1469.

24. T. Regueira, L. Lugo and J. Fernandez, J. Chem.Thermodyn., 2012, 48, 213.

25. R. S. Vajjha and D. K. Das, J. Heat Transfer,2009, 131, 071601.


1358

26. M. Kumar, N. Sawhney, A. K. Sharma and M.Sharma, Phys. Chem. Liq., 2017, 55, 463.

27. H. Xie, W. Yu, Y. Li and L. Chen, NanoscaleRes. lett., 2011, 6, 124.

28. M. Kumar, N. Sawhney, S. Sharma, A. K. Sharmaand M. Sharma, J. Indian Chem. Soc., 2017, 94,151.

29. N. R. Karthikeyan, J. Philip and B. Raj, Mater.Chem. Phys., 2008, 109, 50.

30. J. S. Rowlinson and F. L. Swinton, "Liquid and

liquid mixtures", 3rd ed., Butterworths, London,1982.

31. M. J. W. Povey, "Ultrasonic techniques for fluidscharacterization", Academic Press, USA, 1997.

32. B. Jacobson, Acta Chemica Scand., 1951, 5,1214.

33. B. J. Jacobson, Chem. Phys., 1952, 20, 927.

34. P. M. Morse and K. U. Ingard, "Theoreticalacoustics", Princeton University Press, New Jer-sey, 1986.

1359


Functionalized triarylamines for applications in organic electronics†

Rajesh M. Kamble

Department of Chemistry, University of Mumbai, Santacruz (East), Mumbai-400 098, India


Manuscript received 28 November 2017, accepted 01 December 2017

Abstract : Over the last few decades significant progress has been made in the field of organic electronicswhich includes Organic light emitting diodes (OLEDs), Organic field effect transistors (OFETs) and Dye sen-sitized solar cells (DSSCs). Small organic -conjugated molecules with a donor-acceptor (D-A) architecturehave attracted significant attention as compared to polymers. In organic electronics, typically an OLED con-sists of a hole-transporting materials (HTMs) (p-type materials) and electron-transporting material (ETMs) (n-type materials). The efficiency and durability of organic electronic devices depend upon the stability of boththe hole and electron-transporting materials. Triarylamine derivatives have been extensively utilized as bothhole transporter and emissive material in opto-electronics, since they exhibit low ionization potential, revers-ible redox potential and bright fluorescence and very few reports are there for triarylamines as n-type mate-rials. In this paper, triarylamines based D-A molecules are discussed for their possible applications as n-typeand/or p-type and ambipolar materials in organic electronics.

Keywords : Organic electronics, Triarylamine derivatives, hole transportation, ambipolar materials.

n-Type materials

Tremendous success have been achieved in thedevelopment of p-type materials, but development ofn-type of materials is still lagging far behind than p-type of materials, hence there is a need for the deve-lopment of high mobility and environmentally stableelectron transporting materials. The n-type materialsplay the role of facilitating electron injection from thecathode into the organic layer. Materials which havehigh electron affinities together with high ionizationpotentials usually function as electron transportingmaterials i.e. materials with electron accepting prop-erties are serve as n-type materials1. Common ap-proach in the development of electron transportingmaterials is the introduction of electron withdrawinggroups such as carbonyl, cyano and per-fluorinatedalkyl chains to p-type of materials. Another approachis the substitution with a more electro-negative atomof a carbon in an extended -system of organic mol-

†Invited Lecture.

ecules. The low LUMO energy levels (LUMO valuebetween –3.0 to –4.0 eV)1 are essential for the effi-ciency and stability of electron transporting (n-type)materials1. Fig. 1 shows commonly used n-type ma-terials.

Recently we have synthesized certain triarylaminesbased on naphtho[2,3-f]quinoxaline-7,12-dione coreas donor-acceptors for n-type materials (Fig. 2)2,3.Absorption spectra of all the synthesized moleculesshowed an intramolecular charge transfer (ICT) tran-sitions in the range of 405–561 nm. Electrochemicalproperties of the compounds were studied by cyclicvoltammetry and differential pulse voltammetry. TheHOMO and LUMO energy levels are in the range of–4.88 to –5.57 eV and –3.37 to –3.48 eV respec-tively.

Low-lying LUMO energy levels of these moleculesare comparable to well-known electron transporter


1360

materials and make them potential candidates as n-type materials and are thermally stable.

p-Type materials

Materials which have low ionization potentials to-gether with low electron affinities usually function asHole Transporting Materials (HTMs) (p-type of ma-terials). The hole transport layer in a multilayer de-

vice facilitates hole injection from the anode and trans-port the injected holes to the emitting layer. There-fore, an ideal hole-transporting material should un-dergo reversible anodic oxidation to form stable cat-ion radicals and should posses high hole mobilities.Most widely used HTMs of the triarylamine family inOLEDs are TPD, -NPD and spiro-OMe-TAD (Fig.3). However, TPD and -NPD, which have low glass

Fig. 1. A sample of n-type materials with high electron affinity for organic devices.

Fig. 2. Molecular structure of triarylamines based on naphtho[2,3-f]quinoxaline-7,12-diones.

Kamble : Functionalized triarylamines for applications in organic electronics

1361

transition temperature (Tg) of 65ºC and 95ºC, re-spectively, tend to crystallize or expand during de-vice fabrication and hence there is a need for newthermally stable HTMs.

Our group recently synthesized triarylamines basedon phenazine derivatives with good thermal stabi-lity4,5

(Fig. 4). The HOMO and LUMO energy levels ofphenazine derivatives are in the range of –5.03 to –

Fig. 3. Commonly used p-type of materials.

Fig. 4. Molecular structures of dibenzo[a,c]phenazine and tribenzo[a,c,i]phenazine derivatives.

5.35 eV and –2.75 to –3.17 eV respectively withelectrochemical bandgap within the range of 2.12–2.45 eV. HOMO energy levels of synthesized com-pounds are comparable with most commonly usedhole transporting materials such as TPD, -NPD andspiro-OMe-TAD etc. and thus make them potentialcandidates for the hole transporting material in or-ganic electronics.

Ambipolar materials

Organic molecules that display both hole and elec-tron transporting properties are called as ambipolarorganic semiconductors. This may be achieved throughthe use of a bilayer or blend of semiconductor mate-rials, one p-type and one n-type6. Ambipolar trans-

port prevails when molecules have HOMO levels >–5.6 eV and LUMO levels < –3.15 eV7. Very re-cently (2017) we have synthesized AnthraquinoneAmine-Based D-A derivatives (Fig. 5) as an ambipolarmaterial8. The HOMO and LUMO energy levels ofsynthesized compounds are found in the range of–5.18 to –5.79 eV and –3.22 to –3.53 eV respec-tively as calculated by CV.

JICS-14


1362

References

1. A. P. Kulkarni, C. J. Tonzola, A. Babel and S. A.Jenekhe, Chem. Mater., 2004, 16, 4556.

2. A. M. Shaikh, B. K. Sharma, Sajeev Chacko and R. M.Kamble, RSC Adv., 2016, 6, 60084.

3. A. M. Shaikh, B. K. Sharma and R. M. Kamble, Chem.Hetero. Compds., 2016, 52(2), 110.

4. A. M. Shaikh, B. K. Sharma, Sajeev Chacko and R. M.

Fig. 5. Molecular structures of anthraquinone amine-based D-A derivatives.

Kamble, RSC Adv., 2016, 6, 94218.

5. A. M. Shaikh, B. K. Sharma, Sajeev Chacko and R. M.Kamble, New J. Chem., 2017, 41, 628.

6. J. Zaumseil and H. Sirringhaus, Chem. Rev., 2007,107, 1296.

7. K. Zhou, H. Dong, H.-L. Zhang and W. Hu, Phys.Chem. Chem. Phys., 2014, 16, 22448.

8. A. M. Shaikh, Sajeev Chacko and R. M. Kamble, Chem-istry Select, 2017, 2, 7620.

1363


Synthesis of ruthenium(II) complex with a hexadentate N2O2S2 donor azo-thioether ligand : X-Ray structure, electrochemistry and DFT calculation

Subrata Jana, Ajoy Kumar Pramanik, Chandan Kumar Manna and Tapan Kumar Mondal*

Inorganic Chemistry Section, Department of Chemistry, Jadavpur University,Kolkata-700 032, India



Abstract : A ruthenium(II) complex (Ru-L) with hexadentate thioether containing N2O2S2 donor azo func-tional ligand (H2L) is synthesized and characterized by several spectroscopic techniques. Structure of the com-plex is confirmed by single crystal X-ray diffraction studies. The complex exhibits two successive RuII/RuIII

and RuIII/RuIV oxidation couples along with a ligand based reduction in cyclic voltammetric study. The elec-tronic structure and redox property of the complex are interpreted by DFT calculation. The electronic transi-tions, calculated by TDDFT/CPCM method are used to assign the UV-Vis absorption bands.

Keywords : Ruthenium(II) complex, N2O2S2 donor thioether ligand, X-ray and electronic structure,electrochemistry, DFT and TDDFT calculations.

Introduction

There is considerable interest in the coordina-tion chemistry of polydentate macrocyclic ligandsand in the recent years enormous progress hasbeen made in macrocyclic chemistry1. The mac-rocyclic ligands having N, S and O donor atomsare designed with enhanced ability to encapsulategiven metal ions selectively2 which can influencethe course of many complicated reactions occur-ring during metabolic activity in living organisms3.The coordination chemistry of transition metalswith multidentate ligands having N and S donoratoms have gained augmented research interest inrecent years owing to their diverse potential appli-cations in biomedicine, biological imaging, catalysisand solar cell4. The study of compounds contain-ing N and S atoms has also greater attention dueto their significant antifungal, antibacterial andanticancer activities5.

The coordination chemistry of transition metalswith thioether ligands is a field of interest for their

†Invited Lecture.

stability, chemical, electrochemical and biologicalactivities6. Thioether containing azo ligands canstabilize metal ions in unusual oxidation states andwith uncommon coordination numbers7. Since ni-trogen atom is at borderline while sulphur atombelongs to soft base in the theory of soft-hard acidand base, multidentate ligands containing these twodonor atoms bind to a wide range of transitionmetals8. Thioether containing azo ligands withacetylacetone and their transition metal complexesare reported9. In continuation to synthesis ofthioarylazo ligands and their transition metal com-plexes10, herein we have synthesized ruthenium(II)complex (Ru-L) with the reported hexadentateN2O2S2 donor azo functional ligand (H2L)11. Theoctahedral geometry of the ruthenium(II) complexis confirmed by single crystal X-ray diffractionmethod. Electrochemical property of the complexis investigated by cyclic voltammetric study. Elec-tronic structure and solution spectra of the com-plex are interpreted by DFT and TDDFT calcula-tions.


1364

Experimental

Materials

All the reagents and solvents were purchasedfrom commercial sources and used as received.Acetyl acetone and inorganic metal salts were avail-able from E. Merck, India. RuCl3.3H2O, 2-aminothiophenol and 1,2-dibromoethane were pur-chased from Sigma Aldrich. 1,2-Bis(2-amino-phenylthio)ethane and ligand H2L were preparedfollowing the published procedure11,12.

Physical measurements

Microanalyses (C, H, N) were performed us-ing a Perkin-Elmer CHN-2400 elemental analyzer.Electronic spectra were measured on Lambda 750Perkin-Elmer spectrophotometer in acetonitrile so-lution. IR spectra were recorded on RX-1 Perkin-Elmer spectrophotometer in the spectral range4000–400 cm–1 with the samples in the form ofKBr pellets. 1H NMR spectra were recorded inCDCl3 on Bruker 300 MHz FT-NMR spectrom-eters in presence of TMS as internal standard.Cyclic voltammetric measurements were carriedout using a CHI Electrochemical workstation. Aplatinum wire working electrode, a platinum wireauxiliary electrode and Ag/AgCl reference elec-trode were used in a standard three-electrode con-figuration. [nBu4N][ClO4] was used as the sup-porting electrolyte and the scan rate used was 50mV s–1 in acetonitrile under argon atmosphere.

Synthesis of ruthenium(II) complex (Ru-L)

RuCl3.3H2O (55 mg, 0.201 mmol), was dis-solved in 15 mL of hot acetonitrile with constantstirring. The mixture was then heated to reflux for2 h. The deep brown solution changed to darkgreen. To it 15 mL acetonitrile solution of H2L(100 mg, 0.201 mmol) was added drop wise andfinally refluxed for 12 h to yield a deep red solu-tion. The reaction mixture was cooled and solventwas removed under reduced pressure. The redcrude solid product was then dissolved in mini-mum volume of CH2Cl2 and purified by using asilica gel column. With CH2Cl2-CH3CN (3 : 1), adeep red compound corresponding to the com-

plex was eluted. The desired red product corre-sponds to ruthenium(II) complex, Ru-L was col-lected upon evaporation of the solvent. Yield was73 mg (61%).

Anal. Calcd. for C24H24N4O4S2Ru (Ru-L) : C,48.23; H, 4.05; N, 9.37. Found (%) : C, 48.1; H,4.0; N, 9.3%; IR data (KBr disc) (cm–1) : 2937(C-H), 1673 (C=O), 1586 (C=N), 1405(N=N); 1H NMR data in CDCl3 (, ppm) : 7.91(2H, d, J 8.2 Hz), 7.42–7.56 (4H, m), 7.16 (2H,t, J 7.0 Hz), 3.12 (4H, s), 2.63 (6H, s), 2.55 (6H,s); ESI-mass in acetonitrile : m/z, [M+Na]+ =620.2; max (, M–1 cm–1) in acetonitrile : 505(4887), 450 (sh.), 360 (10646), 307 (18426); E1/2(RuII/RuIII) : 0.72 V (E = 75 mV), 1.44 (E =146 mV); E1/2 (L/L–) : –1.12 V (E = 110 mV).

Computational details

Full geometry optimization of Ru-L complexwas carried out using the density functional theorymethod at the B3LYP level13. All elements exceptruthenium were assigned the 6-31G(d) basis set.The LanL2DZ basis set with effective core poten-tial was employed for the ruthenium atom14. Thevibrational frequency calculation was performedto ensure that the optimized geometry representsthe local minima and there are only positive eigen-values. All calculations were performed withGaussian09 program package15. Vertical electronicexcitations based on the optimized geometry werecomputed using the time-dependent density func-tional theory (TDDFT) formalism16 in acetonitrileusing conductor-like polarizable continuum model(CPCM)17. GaussSum18 was used to calculate thefractional contributions of various groups to eachmolecular orbital.

Crystal structure determination and refinement

The crystals of Ru-L were grown by slow dif-fusion of dichloromethane solution of the com-plex into n-hexane. Single crystal data collectionswere performed with an automated Bruker AXSKappa smart Apex-II diffractometer equipped withan Apex-II CCD area detector using a fine focussealed tube as the radiation source of graphite mono-

Jana et al. : Synthesis of ruthenium(II) complex with a hexadentate N2O2S2 donor etc.

1365

chromator Mo K radiation ( = 0.71073 Å).Details of crystal analyses, data collection and struc-ture refinement data are given in Table 1. Reflec-tion data were recorded using the scan tech-nique. Unit cell parameters were determined from

least-squares refinement of setting angles withina definite range respectively. The absorption cor-rections were done by the multi-scan technique.The data were reduced and integrated using theSAINT19 program. A semi-empirical multi-scanabsorption correction was made with SADABS20.The structures were solved by direct methods em-ploying WinGX21 and SHELXS-9722 programs.The non-hydrogen atoms were refined anisotropi-cally by full-matrix least-squares method withSHELXL-9722. Hydrogen atoms were generatedin the refinement process as per the riding modelwith thermal parameters equal to 1.2 times that ofassociated C atoms, and participated in the calcu-lation of the final R-indices. Figures of the struc-ture was drawn with ORTEP-3223 program with35% ellipsoidal probability.


Synthesis and spectral characterization

New ruthenium(II) complex with thioether con-taining hexadentate N2O2S2 donor azo functionalligand (H2L) is synthesized by the reaction ofRuCl3.3H2O and H2L in 1 : 1 ratio under reflux-ing condition in acetonitrile (Scheme 1). The com-plex is thoroughly characterized by several spec-troscopic techniques. The 1H NMR signals inCDCl3 well support the proposed structure of thecomplex. Two sharp singlets for -CH3 protons ap-peared at 2.63 and 2.55 ppm. The methylene pro-tons of S-alkyl part appeared at 3.12 ppm as sharpsinglet. The aromatic protons appeared at 7.16–7.91 ppm. IR spectrum of the complex exhibitscharacteristic (C=O), (C=N) and (N=N)

Scheme 1. Synthesis of Ru-L complex using H2L ligand.

Table 1. Crystallographic data of ruthenium(II) complex, Ru-L

Formula C24H24N4O4S2Ru

Formula weight 597.68

Crystal color Red

Crystal system Monoclinic

Space group C2/c

a, b, c (Å) 23.984(5), 11.123(5), 21.013(5)

(°) 115.615(5)

V (Å3) 5055(3)

Z 8

D (Calcd.) (g/cm3) 1.571

Mu (MoK) (mm) 0.824

F(000) 2432

Temperature (K) 293(2)

Radiation (Å) 0.71073

(Min-Max) (º) 2.15–26.41

Data set (h; k; l) –30 to 30; –12 to 13; –23 to 26

Crystal size (mm3) 0.21×0.15×0.13

Total, Unique Data, R(int) 19218 , 5130 , 0.0545

Observed data (I > 2(I)) 3328

Nref, Npar 5130 , 316

R1a, wR2

b 0.0799 , 0.2050

Residual Density (e/Å3) –1.171 and 1.019

Goodness of fit (GOF)c 1.069aR1 = |(|Fo|–|Fc|)|/|Fo|bwR2 = {[w (Fo

2–Fc2)2]/[w(Fo

2)2]}1/2,

w = 1/[2(Fo2) + (0.0.952P)2 + 50.1234P], where P = (Fo

2 +2Fc

2)/3cGOF = {[w(Fo

2–Fc2)2]/(n–p)}1/2, where n = number of mea-

sured data and p = number of parameters.


1366

X-Ray structure of Ru-L

The ORTEP plot with atom numbering schemeof ruthenium(II) complex (Ru-L) is shown in Fig.2. The coordination sphere around ruthenium isdistorted octahedral as revealed from the metricparameters (Table 2). The ligand H2L binds to themetal using its entire six coordination site (O, azo-N and thioether-S). In the octahedral geometrythe two thioether-S donors and two O-donors arecis to each other, while azo-N donor pair dis-placed trans to each other in the complex. TheN2-Ru1-S1 and N3-Ru1-S2 chelate bite angles85.72(14)º and 84.71(14)º are significantly devi-ates from 90º and that’s led to the distortion in theoctahedral geometry. Other chelate bite angles are

Table 2. Selected bond distances (Å) and bond angles (º) ofRu-L

Bonds (Å) X-Ray Calcd.

Ru1-N2 2.101(5) 2.115

Ru1-N3 2.101(5) 2.116

Ru1-O1 2.047(4) 2.071

Ru1-O3 2.033(4) 2.071

Ru1-S1 2.4099(18) 2.425

Ru1-S2 2.4128(18) 2.425

N1-N2 1.305(6) 1.285

N3-N4 1.294(6) 1.285

Angles (°)

N2-Ru1-N3 176.01(19) 178.0

N2-Ru1-O1 86.81(18) 88.27

N2-Ru1-O3 87.03(18) 89.27

N2-Ru1-S1 85.72(14) 84.91

N2-Ru1-S2 94.70(15) 95.51

N3-Ru1-O1 97.17(18) 93.25

N3-Ru1-O3 87.03(18) 89.27

N3-Ru1-S1 90.31(14) 93.51

N3-Ru1-S2 84.71(14) 84.91

O1-Ru1-O3 86.23(19) 86.97

O1-Ru1-S1 172.52(13) 174.2

O1-Ru1-S2 94.37(14) 92.12

O3-Ru1-S1 93.91(14) 92.13

O3-Ru1-S2 171.73(13) 174.1

S1-Ru1-S2 86.57(7) 88.34

Fig. 1. UV-Vis spectrum of Ru-L in acetonitrile.

stretching at 1673, 1586, and 1405 cm–1 respec-tively. The (N=N) stretching in the complex sig-nificantly reduced compare to free ligand value(1432 cm–1)12 suggesting the coordination of azo-N to ruthenium. The UV-Vis spectrum in acetoni-trile exhibits moderate intense low energy band at505 nm (, 4887 M–1 cm–1) along with a shoulderat 450 nm. In addition two sharp peaks at 360 nm(, 10646 M–1 cm–1) and 307 nm (, 18426 M–1

cm–1) were observed (Fig. 1). ESI mass spectrumof the complex exhibits a peak with m/z at 600corresponds to [M+Na]+ species.

Fig. 2. ORTEP plot with 35% ellipsoidal probability of Ru-L.

also slightly deviated from 90º in the complex(Table 2). The Ru1-N2 and Ru1-N3 bond distances(2.101(5) Å and 2.101(5) Å) are found to be asexpected10. The Ru-S(thioether) bond distances are


1367

well correlated with the reported Ru-S bond dis-tances9.

DFT and TDDFT calculations

To get detailed insight into the electronic struc-ture of Ru-L, DFT calculations were performedusing DFT/B3LYP method. The optimized bonddistances and bond angles are summarized in Table2. The calculated Ru-O, Ru-N(azo) and Ru-S(thioether) bond distances are well correlated withthe X-ray distances.

The energies along with compositions of someselected molecular orbitals are given in Table 3.The HOMO, HOMO-1 and HOMO-3 have signifi-cant contribution of d(Ru) orbitals (39–55%)along with the contribution of (L) (45–61%) or-bitals (Fig. 3). HOMO-2 has (L) character (97%)while HOMO-4 to HOMO-7 have mixed charac-ter of (L) and d(Ru) orbitals. The LUMO andLUMO+1 have *(L) character with significantcontribution of N=N function. The HOMO-LUMOenergy gap in the complex is found to be 3.15eV.

Table 3. Energy and compositions of selected molecular orbitalsof Ru-L

MO Energy % Composition

Ru L

LUMO+5 –0.36 01 99

LUMO+4 –0.56 24 76

LUMO+3 –0.93 06 94

LUMO+2 –1.01 03 97

LUMO+1 –1.91 04 96 (N=N, 35)

LUMO –1.96 03 97 (N=N, 37)

HOMO –5.11 48 52

HOMO-1 –5.30 39 61

HOMO-2 –6.00 03 97

HOMO-3 –6.02 55 45

HOMO-4 –6.16 25 75

HOMO-5 –6.33 26 74

HOMO-6 –6.33 14 86

HOMO-7 –6.69 15 85 (N=N, 52)

HOMO-8 –7.23 01 99

HOMO-9 –7.34 03 97

HOMO-10 –7.44 16 84

Fig. 3. Contour plots of selected molecular orbitals of Ru-L.

HOMO HOMO-1 HOMO-2

LUMO LUMO+1 LUMO+2

To get detailed insight into the electronic tran-sitions of the complex TDDFT calculation on theoptimized geometry of Ru-L was performed. Some


1368

selected calculated vertical electronic transitionsare summarized in Table 4. The peak at 505 nmof the complex corresponds to HOMOLUMO( = 537 nm, f = 0.018) and HOMOLUMO+1( = 524 nm, f = 0.014) transitions having mixedmetal to ligand charge transfer (MLCT) and intra-ligand charge transfer (ILCT) in character. Theshoulder at 450 nm also has mixed MLCT andILCT character. The high energy band at 360 nmcorresponds to HOMO-2LUMO transition ( =385 nm, f = 0.253) and has ILCT character (Table4).

Electrochemistry

The electrochemical behavior of the complexwas investigated by cyclic voltammetry (CV) inpresence of [nBu4N][ClO4] in acetonitrile at scanrate 50 mV s–1. The complex exhibits one rever-sible and one quasi-reversible oxidative peaks at0.72 V (E = 75 mV) and 1.44 V (E = 146mV) respectively. When scanned in the negativepotential range a quasi-reversible reduction peakat –1.12 V (E = 110 mV) was appeared (Fig.4). The successive two oxidation peaks are as-signed as the oxidation of RuII to RuIII and RuIII

to RuIV respectively as the HOMO of the complexhas 48% d(Ru) character. Similarly, the quasi-

Table 4. Vertical electron excitation calculated by TDDFT/CPCM method of Ru-L

(nm) E (eV) Osc. Key excitations Character Expt. (nm)

strength (f) (, M–1 cm–1)

536.9 2.3092 0.0176 (90%)HOMO LUMO d(Ru)/(L)*(L) 505 (4887)

MLCT/ILCT

523.6 2.3681 0.0135 (89%)HOMO LUMO+1 d(Ru)/(L)*(L)

MLCT/ILCT

472.9 2.6219 0.0618 (51%)HOMO-1 LUMO+1 d(Ru)/(L)*(L) 450 (sh.)

(44%)HOMO-1 LUMO MLCT/ILCT

384.8 3.2220 0.0948 (87%)HOMO-2 LUMO (L)*(L) 360 (10646)

ILCT

331.3 3.7427 0.2528 (74%)HOMO-6 LUMO (L)*(L) 307 (18426)

ILCT

312.5 3.9674 0.1035 (85%)HOMO-6 LUMO+2 (L)*(L)

ILCT

Fig. 4. Cyclic voltammogram of Ru-L in acetonitrile.

reversible reduction peak is assigned as the re-duction coordinated ligand because LUMO has 97%(L) character.

Conclusion

New ruthenium(II) complex, Ru-L with thioethercontaining hexadentate N2O2S2 donor azo ligand(H2L) is synthesized and characterized. The pseudooctahedral geometry of the complex is confirmedby single crystal X-ray structure analysis. The com-plex exhibits two successive RuII/RuIII and RuIII/


1369

RuIV oxidation couples along with a ligand basedreduction in cyclic voltammetric study. The elec-tronic structure of the complex is well interpretedby DFT calculation. The spin allowed electronictransitions computed by TDDFT method success-fully interpret the UV-Vis spectrum of the com-plex in acetonitrile.

AcknowledgementFinancial support received from the Science and

Engineering Research Board (SERB), New Delhi,India (YSS/2015/001533) is gratefully acknowl-edged.

Supplementary materials

Crystallographic data for the structure Ru-L hasbeen deposited with the Cambridge Crystallo-graphic Data Center, CCDC No. 1036745. Copiesof this information may be obtained free of chargefrom the Director, CCDC, 12 Union Road, Cam-bridge CB2 1EZ, UK (E-mail : [email protected] www:htpp://www.ccdc.cam.ac.uk).

References

1. (a) H. Zhang, R. Zou and Y. Zhao, Coord. Chem. Rev.,2015, 292, 74; (b) H. Li, Z.-J. Yao, D. Liu and G.-X. Jin,Coord. Chem. Rev., 2015, 293-294, 139; (c) H. Okawa,H. Furutachi and D. E. Fenton, Coord. Chem. Rev., 1998,174, 51.

2. (a) D. W. Cho, U. C. Yoo and P. S. Mariano, Acc. Chem.Res., 2011, 44, 204; (b) S. C. Chang, V. V. Karambelkar,R. D. Sommer, A. L. Rheingold and D. P. Goldberg, Inorg.Chem., 2002, 41, 239; (c) A. Dolega, K. Baranowska, D.Gudat, A. Herman, J. Stangret, A. Konitz, M. Smiechowskiand S. Godlewska, Eur. J. Inorg. Chem., 2009, 3644; (d)S. Pal, S. N. Poddar, S. Ghosh and G. Mukherjee, Polyhe-dron, 1995, 14, 3023.

3. D. M. Rudkevich, J. D. Mercer-Chlmer, W. Verboom, R.Ungaro, F. Dejong and D. N. Reinhoudt, J. Am. Chem.Soc., 1995, 117, 6124.

4. (a) Y. Nakamura, M. Yonemura, K. Arimura, N. Usuki, M.Ohba and H. Okawa, Inorg. Chem., 2001, 40, 3739; (b)P. Guerriero, S. Tamburini and P. A. Vigato, Chem. Rev.,1995, 95, 273; (c) B. Kersting, G. Steinfeld and J. Hausmann,Eur. J. Inorg. Chem., 1999, 179.

5. (a) A. Garoufis, S. K. Hadjikakou and N. Hadjiliadis, Coord.Chem. Rev., 2009, 253, 1384; (b) X.-X. Xie, H. Li, J.Wang, S. Mao, M.-H. Xin, S.-M. Lu, Q.-B., Mei and S.-Q. Zhang, Bioorg. Med. Chem., 2015, 23, 6477; (c) H.Li, X. M. Wang, J. Wang, T. Shao, Y.-P. Li, Q.-B. Mei,

S.-M. Lu and S.-Q. Zhang, Bioorg. Med. Chem., 2014,22, 3739; (d) R. L. Richard, Coord. Chem. Rev., 1996,154, 83; (e) P. J. Blower, J. S. Lewis and J. Zweit, Nucl.Med. Biol., 1996, 23, 957.

6. (a) R. L. Richard, Coord. Chem. Rev., 1996, 154, 83; (b)S. Dhar and A. R. Chakraborty, Inorg. Chem., 2003, 42,2483; (c) A. K. Singh and R. N. Mukherjee, Dalton Trans.,2005, 2886; (d) R. A. Allred, S. A. Hufner, K. Rudzka, A.M. Arif and L. M. Berreau, Dalton Trans., 2007, 351; (e)A. K. Pramanik, M. S. Jana and T. K. Mondal, J. Coord.Chem., 2013, 66, 4067; (f) B. Adhikary, S. Liu and C. R.Lucas, Inorg. Chem., 1993, 32, 5957.

7. (a) A. P. Koley, S. Purohit, L. S. Prasad, S. Ghosh and G.T. Manoharan, Inorg. Chem., 1992, 31, 305; (b) S.Mukhopadhyay and D. Ray, J. Chem. Soc., Dalton Trans.,1995, 265; (c) S. Karmakar, S. B. Choudhury, D. Ray andA. Chakravorty, Polyhedron, 1993, 12, 2325; (d) E.Ambundo, L. A. Ochrymowycz and D. B. Rorabacher, Inorg.Chem., 2001, 40, 5133; (e) U. S. Ray, D. Banerjee, G.Mostafa, T.-H. Lu and C. Sinha, New J. Chem., 2004, 28,1432.

8. (a) S. Y. Shaban, Inorg. Chim. Acta, 2011, 367, 212; (b)W. Rammal, H. Jamet, C. Philouze, J. L. Pierre, E. Saint-Aman and C. Belle, Inorg. Chim. Acta, 2009, 362, 2321;(c) M. Shukla, N. Srivastava, S. Saha, T. R. Rao and S.Sunkari, Polyhedron, 2011, 30, 754; (d) J. A. Zhang, M.Pan, R. Yang, Z. G. She, W. Kaim, Z. J. Fan and C. Y.Su, Polyhedron, 2010, 29, 581.

9. (a) M. K. Paira, T. K. Mondal, E. López-Torres, J. Ribasand C. Sinha, Polyhedron, 2010, 29, 3147; (b) M. K.Paira, T. K. Mondal, D. Ojha, A. M. Z. Slawin, E. R. T.Tiekink, A. Samanta and C. Sinha, Inorg. Chim. Acta,2011, 370, 175; (c) S. Jana, M. S. Jana, D. Sarkar, M. K.Paira and T. K. Mondal, J. Mol. Struct., 2013, 1054-1055, 83.

10. (a) T. K. Mondal, J.-S. Wu, T.-H. Lu, Sk.Jasimuddin and C. Sinha, J. Organomet. Chem.,2009, 694, 3518; (b) T. K. Mondal, P. Raghavaiah,A. K. Patra and C. Sinha, Inorg. Chem. Commun.,2010, 13, 273; (c) M. S. Jana, A. K. Pramanik, S.Kundu, D. Sarkar, S. Jana and T. K. Mondal, Inorg.Chim. Acta, 2013, 394, 583; (d) M. S. Jana, A. K.Pramanik, S. Kundu and T. K. Mondal, Polyhedron,2012, 40, 46.

11. A. K. Pramanik, D. Sarkar, S. Biswas and T. K.Mondal, Transition Met. Chem., 2016, 41, 849.

12. C. G. Densmore, H. Wheeler, R. Cohenour, T. W.Robison, D. Hasam, B. J. Cordova, P. C. Stark, E.N. Fuller, C. J. Cook and H. A. Weber, Org. Pro-cess Res. Dev., 2007, 11, 996.

13. (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648;(b) C. Lee, W. Yang and R. G. Parr, Phys. Rev.(B), 1988, 37, 785.

14. (a) P. J. Hay and W. R. Wadt, J. Chem. Phys.,


1370

1985, 82, 270; (b) W. R. Wadt and P. J. Hay, J.Chem. Phys., 1985, 82, 284; (c) P. J. Hay and W.R. Wadt, J. Chem. Phys., 1985, 82, 299.

15. Gaussian 09, Revision D.01, M. J. Frisch, G. W.Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,J. R. Cheeseman, G. Scalmani, V. Barone, B.Mennucci, G. A. Petersson, H. Nakatsuji, M.Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J.Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M.Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M.Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,T. Vreven, J. A. Montgomery (Jr.), J. E. Peralta, F.Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N.Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,K. Raghavachari, A. Rendell, J. C. Burant, S. S.Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.Millam, M. Klene, J. E. Knox, J. B. Cross, V.Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C.Pomelli, J. W. Ochterski, R. L. Martin, K.Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salva-dor, J. J. Dannenberg, S. Dapprich, A. D. Daniels,Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski,and D. J. Fox, Gaussian, Inc., Wallingford CT,2009.

16. (a) R. Bauernschmitt and R. Ahlrichs, Chem. Phys.Lett., 1996, 256, 454; (b) R. E. Stratmann, G. E.Scuseria and M. J. Frisch, J. Chem. Phys., 1998,109, 8218; (c) M. E. Casida, C. Jamorski, K. C.Casida and D. R. Salahub, J. Chem. Phys., 1998,108, 4439.

17. (a) V. Barone and M. Cossi, J. Phys. Chem. (A),1998, 102, 1995; (b) M. Cossi and V. Barone, J.Chem. Phys., 2001, 115, 4708; (c) M. Cossi, N.Rega, G. Scalmani and V. Barone, J. Comput.Chem., 2003, 24, 669.

18. N. M. O’Boyle, A. L. Tenderholt and K. M.Langner, J. Comput. Chem., 2008, 29, 839.

19. SAINT Plus, Data Reduction and Correction Program,v. 6.01., Bruker AXS, Madison, Wisconsin, USA,1998.

20. Bruker APEX2, SAINT and SADABS. Bruker AXSInc. (2010) Madison, Wisconsin, USA.

21. L. J. Farrugia, J. Appl. Cryst., 2012, 45, 849.

22. G. M. Sheldrick, Acta Crysallogr, Sect. A, 2008,64, 112.

23. L. J. Farrugia, ORTEP-3 for windows, J. Appl.Cryst., 2012, 45, 849.

1371


Synthesis of 3-hydroxy 2-oxindoles via deacylative oxygenation (DaO) :Total synthesis of CPC-1†

Nivesh Kumar, Vipin R. Gavit, Manvendra Singh and Alakesh Bisai*

Department of Chemistry, Indian Institute of Science Education and Research (IISER), Bhopal-462 023,Madhya Pradesh, India



Abstract : We report an efficient deacylative oxygenation strategy for the synthesis of a variety of 3-hydroxy2-oxindoles that takes advantage of deacylative oxidation using oxygen gas as green source of oxidants via aretro-Claisen condensation. A wide variety of products could be synthesized under mild reaction conditions.The aforementioned methodology has been utilized for total synthesis of CPC-1.

Keywords : Deacylative oxygenation, retro-Claisen activation, oxygen gas, allylalkoxide, 3-hydroxy 2-oxindoles.

Introduction

Since they have coevolved with their putative bio-logical targets, natural products intersect biologicalspace effectively and perturb its function in a highlycontrolled manner1. It is not surprising that naturalproducts have endured as promising leads for drugdiscovery. A rapid access to small molecules that areguided by natural products appears to be quintessen-tial for the success of a chemical genetics/genomics-based program2. In this regard, synthesis of 2-oxindoleframeworks have been experiencing a dramatic ex-pansion and becoming a hot research area in contem-porary organic chemistry due to their unique reacti-vity and capacity to construct complex chiral frame-works3. They are widely distributed in natural prod-ucts3,4 and have broad applications in the discoveryof pharmaceutically important molecules5. Amongvarious 2-oxindole natural products, donaxaridine (1a)and donaxarine (1b) were isolated from the giant reedArundodonax in 1976 (Fig. 1)6a-b. ConvolutamydineA-B (2a-b) and convolutamydine E (2c) were iso-lated from the Floridian marine bryozoanAmathiaconvoluta by Kamano and co-workers6c. Anew dimeric alkaloid arundaphine (3a), was isolated

†The work is dedicated to Late Professor Asima Chatterjee.Invited Lecture.

from roots and rhizomes of Arundodonax (Poaceaefamily)6d. In the culture broth of the marine Strepto-myces strain B 9173 two closely related, maremycinA (3b) and B (3c), were isolated in 19956e. 2-Oxindolealkaloids, paratunamides A (3d) and D (3e), contain-ing a secologanin unit, were isolated fromCinnamodendronaxillare (Nees at Mart.), belongingto Canellaceae family (local name paratude). In Bra-zil “paratude” is used as a stomachic and a treatmentfor tonsillitis6f.

Further, pyrrolidinoindoline-type alkaloids such asalline (4a) and CPC-1 (4b) were also isolated fromvarious sources6g. Madindolines A (5a) and B (5b)were isolated from the fermentation broth of Strepto-myces nitrosporeus K93-0711, which are selective in-hibitors of interleukin-66h consisting of two portionssuch as a 3a-hydroxy furoindoline fragment and asubstituted cyclopentenedione moiety. In this regard,2-oxindoles intermediates have been utilized in thesynthesis of many bioactive natural pro-ducts sharingpyrroloindoline scaffolds, such asalline (4a), CPC-1(4b), flustraminol B (6)7, and furoindoline scaffolds,such as madindolines A (5a) and B (5b) (Fig. 2)8.


1372

Further, 2-oxindoles have also served as potentialintermediates for the synthesis of various 2-oxindolederivatives with C-3 quaternary center. Towards this,our group have explored Lewis acid catalyzed C-Cbond forming reactions via Friedel-Crafts alkylationsof 3-substituted 3-hydroxy 2-oxindole with variouselectron-rich aromatics9, terminal alkynes10,allylsilanes11, acetophenones12. These reactions pre-sumably go through in situ generation of reactive in-termediate 2H-indol-2-one13. Owing to their thera-peutic value, considerable efforts have been devotedto develop efficient methodology for the synthesis of2-oxindoles14 and indeed some elegant methologieshave been reported to directly construct the 3-functionalized-3-hydroxy-2-oxindole framework15.

Literature reports on the synthesis of 3-hydroxy2-oxindoles include aldol reactions of ketones andaldehydes with isatins16, metal-mediated 1,2-additionsof carbon nucleophiles/equivalents17. Henry reactionof isatins with alkanes18, the oxidation of 3-substi-

tuted indoles19a-c and 2-oxindoles20, amino-oxygen-ation of 2-oxindoles21. There have been very fewmethods that can construct a 3-hydroxy 2-oxindolescaffold and bear a wide spectrum of functional groupsat the C-3 position, to the best of our knowledge.


Recently, we envisioned that an allylic alkoxidemay induce a retro-Claisen condensation of an appro-priately substituted 2-oxindole 7 to form allylmethylcarbonate 11 and an enolate 10b as active intermedi-ate via carbanion 10a (Scheme 1)22. This enolate 10bwould then react with a PdII--allyl complex gener-ated in situ by the reaction of allyl acetate 11 and Pd0

to furnish various 2-oxindoles 9 with a quaternarycenter. A number of 2-oxindole scaffolds with a C-3quaternary center have been synthesized utilizing afore-mentioned methodology (Scheme 1).

While working on this area, we have found that anefficient deacylative oxidation of 3-acyl-2-oxindoles

Fig. 1. Selected biologically active 3-hydroxy 2-oxindoles (1-3).

Fig. 2. Pyrroloindoline (4a-b and 6) and furoindoline alkaloid 5.

Kumar et al. : Synthesis of 3-hydroxy 2-oxindoles via deacylative oxygenation (DaO) etc.

1373

can be performed under O2 atmosphere. We arguedthat, an allylalkoxide may induce a retro-Claisen con-densation of an appropriately substituted 2-oxindole 7to form enolate 10b as active intermediate via car-banion 10a (Scheme 2). This enolate 10b would thenreact with oxygen gas as green source of oxidants toafford a number of 3-hydroxy 2-oxindoles. Herein,we report mild synthesis of 3-substituted 3-hydroxy2-oxindole framework that takes advantage of deacyla-tive oxidation using oxygen gas as green source ofoxidants via a retro-Claisen condensation (Scheme 2).

Initially, optimization of deacylative oxygenation

(DaO) was carried out with 1 equivalent of methyl 3-

methyl N-methyl 2-oxindole 3-carboxylate 7a (0.25

mmol) with 1.2 equivalent of allylalcohol (0.30 mmol)

in the presence of 1.2 equivalent of different bases

(0.30 mmol) under oxygen atmosphere (1 atm.) to

afford 3-hydroxy N-methyl 2-oxindole 12a (Table 1).

It was found that we could isolate product 12a, how-

ever, in combination with protonated 3-methyl N-

methyl 2-oxindole 13a. Clearly, compound 13a was

Scheme 1. Our report on Pd0-catalyzed deacylative allylations for the synthesis of 2-oxindoles sharing C-3 quaternary center.

Scheme 2. Deacylative oxygenation via retro-Claisen activation.


1374

observed because of incomplete oxidation, whereasprotonation also encountered as byproduct. There-fore, various bases such as sodium carbonate, potas-sium carbonate, cesium carbonate, DBU, sodiumhydride, and potassium tert-butoxide were employedfor the oxidation via retro-Claisen condensation (en-tries 1–6).

Following exhaustive optimization, we observedthat deacylative oxygenation (DaO) can be carriedout with 7a (0.25 mmol, 1 equiv.) in the presence ofNaH or KOtBu22 (1.2 equiv.) under oxygen atmo-sphere (1 atm.) to produce the desired 2-oxindole12a in 85% and 79% yields, respectively, at roomtemperature (entries 9–10). Further optimization us-ing different solvents revealed that, toluene is supe-rior over other solvents such as tetrahydrofuran, dim-

ethyl sulfoxide, N,N-dimethyl formamide, acetonitrile,diethyl ether, dichloromethane, and chloroform (en-tries 11–16). Therefore, based on our optimization,1.2 equivalent of NaH in toluene at room tempera-ture was chosen as standard condition. This standardcondition of deacylative alkylation (DaA) was appliedto a variety of substrates.

Under the optimized condition a variety of 3-me-thyl N-alkyl 2-oxindole 3-carboxylates such as N-methyl (7a), N-p-methoxybenzyl (7b), N-allyl (7c),and N-prenyl (7d) were utilized to afford a range of3-hydroxy 2-oxindoles bearing hydroxyl group at thepseudobenzylic position (12a-d) in good yields(Scheme 3).

Further, a variety of methyl 3-benzyl N-methyl 2-oxindole 3-carboxylates such as 3-benzyl (7e), 3-(o-

Table 1. Optimization of deacylative oxygenation (DaO) using oxygen gas as green source of oxidant

Sr. Base Solvent Temp. Time % Yield % Yield

No. (1.2 equiv.) (ºC) (h) (1a) (13a)

1. Na2CO3 THF 25 2 23 63

2. K2CO3 THF 25 3 15 62

3. Cs2CO3 THF 25 3 18 71

4. DBU THF 25 4 27 52

5. NaH THF 25 3 38 44

6. KOtBu THF 25 3 29 41c

7. NaH THF 60 2 35 38

8. KOtBu THF 60 2 30 40

9. NaH PhMe 25 2 85 —

10. KOtBu PhMe 25 2 79 15

11. NaH DMSO 25 3 21 51

12. NaH DMF 25 4 32 49

13. NaH MeCN 25 3 70 17

14. NaH Et2O 25 4 48 21

15. NaH CH2Cl2 25 3 42 18c

16. NaH CHCl3 25 3 34 23c

aReactions were carried out using 0.5 mmol of 1a (1 equiv.) with 0.6 mmol of base (1.2 equiv.) in 2 mL solvent under 1 atm.pressure of oxygen gas. bIsolated yields after column chromatography. cDecomposition of the rest of the mass balance.


1375

methyl)benzyl (7f), 3-(o-bromo)benzyl (7g), and 3-(o-methoxy)benzyl (7h), were utilized to afford prod-ucts 12e-h in good yields (Scheme 4). Gratifyingly,challenging substrates like methyl 3-allyl/prenyl N-methyl 2-oxindole 3-carboxylates such as 7i-j fur-nished products 12i-j in 69–78% isolated yields(Scheme 5). Compounds 7i-j are challenging in a sensethat these are prone to undergo oxidation at the allylicposition.

Mechanistically23, an alkoxide may induce a retro-Claisen condensation of an appropriately substituted2-oxindole 7 to form enolate 10b as active intermedi-ate via carbanion 10a (see, Scheme 1). This enolate10b would then react with oxygen gas via a single

aReactions were carried out using 0.5 mmol of 7a-d (1 equiv.) with 0.6 mmol of NaH (1.2 equiv.) in 2 mL of toluene under 1 atm.pressure of oxygen gas. bIsolated yields after column chromatography.

Scheme 3. Substrate scope of deacylative oxygenation of 7a-d.

electron transfer (SET) mechanism to afford a num-ber of 3-hydroxy 2-oxindoles. Initially, a single elec-tron transfer (SET) from carbanion 10b could gener-ate a 3º radical 14b, which then stabilized as vinyloxyradical 14a. This 3º radical would then react withsinglet oxygen to generate a peroxy radical 15, whichcould form peroxide 16 (Scheme 6). The later couldrelease hydroxyl radical to generate N-alkyl 3-substi-tuted 3-alkoxy radical 17, which then afford product12 by combination of hydrogen radical.

To illustrate the broad synthetic utility of this meth-odology, we undertook total synthesis of medicinallyimportant compound CPC-1 (Fig. 2). Compound 12iwas methylated with methyl iodide in the presence of

aReactions were carried out using 0.5 mmol of 7e-h (1 equiv.) with 0.6 mmol of NaH (1.2 equiv.) in 2 mL of toluene under 1 atm.pressure of oxygen gas. bIsolated yields after column chromatography.

Scheme 4. Further scope of deacylative oxygenation using 7e-h.


1376

sodium hydride to afford 18 in 90% yield. Compound18 was reacted with NMO in the presence of catalyticOsO4 to afford diol, which was directly reacted withsodium meta periodate (NaIO4) to affect oxidativedegradation to furnish aldehyde 19 in 84% yields over2 steps (Scheme 7). This aldehyde 19 was then re-acted with methylamine hydrochloride salt in tetrahy-drofuran to generate imine, which was reduced withlithium aluminum hydride to afford CPC-1 4b in 76%yields over 2 steps (Scheme 7).

In summary, we have developed efficientdeacylative oxidation (DaO) for the synthesis of avariety of 3-substituted 3-hydroxy 2-oxyindoles 12inspired by the “medicinal” scaffold of 3-substituted3-hydroxy 2-oxindoles. This method relies on adeacylative oxidation (DaO) via a retro-Claisen

condensation using oxygen gas as green source ofoxidants. We have demonstrated the broad syntheticutility of our catalytic protocol in the efficient assem-bly of pharmaceutical important 3-hydroxy 2-oxindoles. As an application of this methodology, wehave shown total synthesis of medicinally importantcompound CPC-1 (4b). Further applications of thiscatalytic strategy are underway and will be reportedin due course.

Experimental

General procedure for 2-oxindole synthesis :

In a flame-dried seal tube allyl alcohol (0.55 mmol)was taken in toluene (2 mL) at room temperature. Tothis solution, NaH (60%, 0.60 mmol) was added atonce, followed by 2-oxindole (0.5 mmol) under O2

aReactions were carried out using 0.5 mmol of 7i-j (1 equiv.) with 0.6 mmol of NaH (1.2 equiv.) in 2 mL of toluene under 1 atm.pressure of oxygen gas. bIsolated yields after column chromatography. cDecomposition of the rest of the mass balance.

Scheme 5. Deacylative oxygenation for the synthesis of 7i-j.

Scheme 6. Proposed mechanism of deacylative oxygenation via a Single Electron Transfer (SET).


1377

aAll reactions were carried out using 0.2 mmol of starting material under nitrogen atmosphere. bIsolated yields after column chroma-tography.

Scheme 7. Total synthesis of naturally occurring pyrroloindoline alkaloid, CPC-1 (4b).

(balloon) 1 atm. The reaction mixture was stirred for10 min. Upon completion the reaction (as judged byTLC analysis), it was quenched by adding few dropsof water and extracted with EtOAc twice (6 mL×2).All organic extracts were collected and dried overMgSO4 and concentrated under vacuo. The crudeproduct was purified by flash chromatography on silicagel (30–40% EtOAc in n-hexane as eluent) to affordthe desired product (±)-12a-j.

AcknowledgementAB thanks the Council of Scientific and Industrial

Research (CSIR) (Sanction No. 02(0295)/17/EMR-II) and SERB, Department of Science and Technol-ogy (DST) (Sanction No. EMR/2016/000214), Gov-ernment of India for generous research grants. NKand VRG thank the CSIR for Research Fellowships(SRF and JRF, respectively). We sincerely thank theDepartment of Chemistry, IISER Bhopal for infra-structure. AB sincerely thanks Professor Vinod K.Singh, Director, IISER Bhopal for excellent researchfacilities.

Supporting InformationGeneral experimental procedures, characterization

data including 1H NMR, 13C NMR spectra of se-lected compounds. This material is available free ofcharge.

References

1. (a) K. C. Nicolaou and T. Montagnon, "Molecules ThatChanged the World : A Brief History of the Art andScience of Synthesis and Its Impact on Society", Wiley,Weinheim, 2008; (b) P. M. Dewick, "Medicinal Natu-ral Products", Wiley, West Sussex, 2009.

2. (a) S. L. Schreiber, Bioorg. Med. Chem., 1998, 6,1127; (b) S. L. Schreiber, Science, 2000, 287, 1964;(c) C. M. Crew, Curr. Opin. Chem. Biol., 2000, 4, 47;(d) L. Weber, Curr. Opin. Chem. Biol., 2000, 4, 295.

3. (a) L. T. Vassilev, B. T. Vu, B. Graves, D. Carvajal,F. Podlaski, Z. Filipovic, N. Kong, U. Kammlott, C.Lukacs, C. Klein, N. Fotouhi and E. A. Liu, Science,2004, 303, 844; (b) A. Deiters, M. Pettersson and S. F.Martin, J. Org. Chem., 2006, 71, 6547; (c) X. Zhou,T. Xiao, Y. Iwama and Y. Qin, Angew. Chem. Int. Ed.Engl., 2012, 51, 4909.

4. (a) A. Fensome, W. R. Adams, A. L. Adams, T. J.Berrodin, J. Cohen, C. Huselton, A. Illenberger, J. C.Kern, V. A. Hudak, M. A. Marella, E. G. Melenski,C. C. McComas, C. A. Mugford, O. D. Slayden, M.Yudt, Z. Zhang, P. Zhang, Y. Zhu, R. C. Winnekerand J. E. Wrobel, J. Med. Chem., 2008, 51, 1861; (b)V. V. Vintonyak, K. Warburg, H. Kruse, S. Grimme,K. Hubel, D. Rauh and H. Waldmann, Angew. Chem.Int. Ed. Engl., 2010, 49, 5902.

5. (a) M. Rottmann, C. McNamara, B. K. Yeung, M. C.Lee, B. Zou, B. Russell, P. Seitz, D. M. Plouffe, N. V.Dharia, J. Tan, S. B. Cohen, K. R. Spencer, G. E.Gonzalez-Paez, S. B. Lakshminarayana, A. Goh, R.Suwanarusk, T. Jegla, E. K. Schmitt, H. P. Beck, R.Brun, F. Nosten, L. Renia, V. Dartois, T. H. Keller,D. A. Fidock, E. A. Winzeler and T. T. Diagana, Sci-


1378

ence, 2010, 329, 1175; (b) B. Yu, D. Q. Yu and H. M.Liu, Eur. J. Med. Chem., 2015, 97, 673; (c) N. Ye, H.Chen, E. A. Wold, P.-Y. Shi and J. Zhou, ACS Infect.Dis., 2016, 2, 382.

6. (a) K. A. Ubaidullaev, R. Shakirov and S. Y. Yunosov,Khim. Prir. Soedin., 1976, 12, 553; (b) H. B. Rasmussenand J. K. MacLeod, J. Nat. Prod., 1997, 60, 1152; (c)Y. Kamano, A. Kotake, H. Hashima, I. Hayakawa, H.Hiraide, H.-P. Zhang, H. Kizu, K. Komiyama, M.Hayashi and G. R. Pettit, Collect. Czech. Chem.Commun., 1999, 64, 1147; (d) V. U. Khuzhaev, I.Zhalolov, K. K. Turgunov, B. Tashkhodzhaev, M. G.Levkovich, S. F. Aripova and A. S. Shashkov, Chem.Nat. Compounds, 2004, 40, 269; (e) W. Balk-Bindseil,E. Helmke, H. Weyland and H. Laatsch, Leibigs Ann.,1995, 1291; (f) T. Kagata, S. Saito, H. Shigemori, A.Ohsaki, H. Ishiyama, T. Kubota and J. Kobayashi, J.Nat. Prod., 2006, 69, 1517; (g) M. Kitajima, I. Mori,K. Arai, N. Kogure and H. Takayama, TetrahedronLett., 2006, 47, 3199; (h) M. Hayashi, M.-C. Rho, A.Enomoto, A. Fukami, Y.-P. Kim, Y. Kikuchi, T.Sunazuka, T. Hirose, K. Komiyama and S. Omura, Proc.Natl. Acad. Sci. USA, 2002, 99, 14728.

7. (a) C. V. Galliford and K. A. Scheidt, Angew. Chem.,Int. Ed., 2007, 46, 8748; (b) J. I. Jimeìnez, U. Huber,R. E. Moore and G. M. L. Patterson, J. Nat. Prod.1999, 62, 569; (c) M. Suchy, P. Kutschy, K. Monde,H. Goto, N. Harada, M. Takasugi, M. Dzurilla and E.Balentova, J. Org. Chem., 2001, 66, 3940.

8. (a) R. R. Goehring, Y. P. Sachdeva, J. S. Pisipati, M.C. Sleevi and J. F. Wolfe, J. Am. Chem. Soc., 1985,107, 435; (b) J. Kohno, Y. Koguchi, M. Nishio, K.Nakao, M. Juroda, R. Shimizu, T. Ohnuki and S.Komatsubara, J. Org. Chem., 2000, 65, 990.

9. (a) S. Ghosh, L. K. Kinthada, S. Bhunia and A. Bisai,Chem. Commun., 2012, 48, 10132; (b) L. K. Kinthada,S. Ghosh, K. N. Babu, M. Sharique, S. Biswas and A.Bisai, Org. Biomol. Chem., 2014, 12, 8152.

10. (a) L. K. Kinthada, S. R. Medisetty, K. N. Babu,A. Parida and A. Bisai, J. Org. Chem., 2017, 82,8548; (b) S. Ghosh, S. Chaudhuri and A. Bisai,Org. Lett., 2015, 17, 1373.

11. (a) L. K. Kinthada, K. N. Babu, D. Padhi and A.Bisai, Eur. J. Org. Chem., 2017, 3078; (b) K. N.Babu, L. K. Kinthada, S. Ghosh and A. Bisai,Org. Biomol. Chem., 2015, 13, 10641.

12. L. K. Kinthada, S. Ghosh, S. De, S. Bhunia, D.Dey and A. Bisai, Org. Biomol. Chem., 2013,11, 6984.

13. (a) J. R. Fuchs and R. L. Funk, J. Am. Chem.Soc., 2004, 126, 5068; (b) J. R. Fuchs and R. L.Funk, Org. Lett., 2005, 7, 677; (c) J. Belmer andR. L. Funk, J. Am. Chem. Soc., 2012, 134,16941; (d) C. Menozzi, P. I. Dalko and J. Cossy,

Chem. Commun., 2006, 4638; (e) S. Ma, X. Han,S. Krishnan, S. C. Virgil and B. M. Stoltz,Angew. Chem., Int. Ed., 2009, 48, 8037.

14. For reviews, see : (a) J. J. Badillo, N. V. Hanhanand A. K. Franz, Curr. Opin. Drug Discov.Devel., 2010, 13, 758; (b) R. Dalpozzo, G.Bartoli and G. Bencivenni, Chem. Soc. Rev.,2012, 41, 7247; (c) L. Hong and R. Wang, Adv.Synth. Catal., 2013, 355, 1023; (d) P. Chauhanand S. S. Chimni, Tetrahedron : Asymmetry,2013, 24, 343; (e) G. Koutoulogenis, N.Kaplaneris and C. G. Kokotos, Beilstein J. Org.Chem., 2016, 12, 462.

15. (a) P. Chauhan, S. Mahajan and D. Enders,Chem. Rev., 2014, 114, 8807; (b) B. R. Beno, K.S. Yeung, M. D. Bartberger, L. D. Penningtonand N. A. Meanwell, J. Med. Chem., 2015, 58,4383.

16. (a) Q. Guo, M. Bhanushali and C.-G. Zhao,Angew. Chem., Int. Ed., 2010, 49, 9460; (b) G.Luppi, P. G. Cozzi, M. Monari, B. Kaptein, Q.B. Broxterman and C. Tomasini, J. Org. Chem.,2005, 70, 7418; (c) F. Braude and H. G. Lindwall,J. Am. Chem. Soc., 1933, 55, 325.

17. (a) S. E. Denmark and J. Fu, Chem. Rev., 2003,103, 2763; (b) J. A. Marshall, Chem. Rev., 2000,100, 3163; (c) R. Shintani, M. Inoue and T.Hayashi, Angew. Chem., Int. Ed., 2006, 45,3353.

18. L. Liu, S. Zhang, F. Xue, G. Lou, H. Zhang, S.Ma, W. Duan and W. Wang, Chem.-Eur. J.2011, 17, 7791.

19. (a) T.Ishimaru, N. Shibata, J. Nagai, S.Nakamura, T. Toru and S. Kanemasa, J. Am.Chem. Soc., 2006, 128, 16488; (b) D. Sano, K.Nagata and T. Itoh, Org. Lett., 2008, 10, 1593;(c) S. Miah, C. J. Moody, I. C. Richards and A.M. Z. Slawin, J. Chem. Soc., Perkin Trans. 1,1997, 2405.

20. M. B. Chaudhuri, Y. Sutar, S. Malpathak, A.Hazra and B. Gnanaprakasam, Org. Lett., 2017,19, 3628.

21. T. Bui, N. R. Candeias and C. F. Barbas III, J.Am. Chem. Soc., 2010, 132, 5574.

22. KOtBu was superior for synthesis of 2-oxindoleswith C-3 quaternary centers, see : (a) S. Ghosh,S. De, B. N. Kakde, S. Bhunia, A. Adhikary andA. Bisai, Org. Lett., 2012, 14, 5864; (b) S.Bhunia, S. Ghosh, D. Dey and A. Bisai, Org.Lett., 2013, 15, 2426.

23. A similar type mechanism for oxidation of 3-alkylhomophthalimide derivatives, see : H.Heaney and M. O. Taha, ARKIVOC, 2000, iii,343.

1379


A short overview on fluorescent nanoclusters and its application in sensing ofmetal ions†

Shilpa Bothra and Suban K. Sahoo*

Department of Applied Chemistry, S. V. National Institute of Technology (SVNIT), Surat-395 007,Gujarat, India

E-mail : [email protected], [email protected]

Manuscript received 30 November 2017, accepted 01 December 2017

Abstract : The fluorescence of noble metal nanoclusters (NCs) has attracted burgeoning interest among thechemists working in the field of sensors and biosensors development due to their unique properties such asexcellent photostability, low toxicity, bio-compatibility and size-dependent fluorescence. The size of noble metalNCs is between metal atoms and nanoparticles, and comparable to the Fermi wavelength of electrons, whichresulting in the molecule-like properties and discrete energy levels to show size-dependent fluorescence. Be-cause of the excellent fluorogenic properties, the noble metal NCs are widely used for the development offluorescent probes for various applications. This short review was narrated to summarize some selected ap-plications of noble metal NCs (mainly of gold and silver) in the detection of metal ions.

Keywords : Fluorescent nanoclusters, fluorescent probes, metal ions, gold nanoclusters, silver nanoclusters.

Introduction

There is an increasing demand for the detection ofionic analytes in the chemical environments, both in-side the living systems and surrounding environment.Research on the recognition, sensing and extractionof metal ions of biological and environmental impor-tance has keenly attracted scientists owing to the ubiq-uitous distribution in the environmental, industrial andbiological processes. Metal ions are involved in cel-lular and subcellular functions1 and can effectivelycontrol the enzyme-catalyzed reactions2. Thus, theintake of essential (Cu2+, Zn2+, Mg2+, Fe3+ etc.)and non-essential (Cd2+, Hg2+, Al3+ etc.) metal ionsin proper amount in the body is necessary as theirdisproportion may lead to toxicity and cause a threatto human health through cellular toxicity, liver andkidney damage, and neurodegenerative diseases andso on3.

Owing to these increasing detrimental effects onthe environment and living systems, there has been agrowing interest in the development of analytical

†Invited Lecture.

methods that allows rapid, selective and inexpensivedetermination of metal ions in aqueous medium. There-fore, the finding of new metal ions selective recep-tors is an important goal which involves sensors de-velopment for environmental remediation, selectiveseparation and extraction of chemical species. Vari-ous methods adopted for the detection of these ionsare atomic absorption spectrometry (AAS), ion chro-matography, electrochemistry, stripping voltammetryand inductively coupled plasma atomic emission spec-trometry (ICP-AES) etc.4–6. However, these tech-niques are expensive, requires significant and sophis-ticated instrumentation, non-portable and complicatedsample pre-treatment methods. Thus, the selective andsensitive sensing of metal ions by optical chemosensorsin the proximity of supramolecular chemistry haveconsiderably gained research interest in terms of theirpotential applications7. As described in Fig. 1, thechemosensor generally possess two important com-ponents : a receptor for the selective recognition oftarget analyte, a light-emitting group (fluorophore)


1380

that provides detectable optical response upon analyterecognition. In the recent few decades, many opti-cally active chemosensors were developed for theselective detection of metal ions by using various or-ganic fluorophores and receptors like azo-derivatives8,anthrylacetamide9, crown-ethers10, dithiocarbamate11,naphthalimides12, rhodamine13, anthraquinone14, fluo-rescein15, etc. However, many of these reported or-ganic based chemosensors works in organic solventsor mixed organic-aqueous medium and thus, lacksselectivity and sensitivity in on-site monitoring ofanalyte in aqueous biological and environmental sys-tems.

suitable organic molecules for target specific detec-tion of analytes from the aqueous medium.

Fluorescent metal nanoclusters

Recently, innovations to produce non-toxicnanoparticles smaller than ~2 nm in diameter of metals(Au/Ag) have been made successful where the sizeapproaches the Fermi wavelength of an electron (~2nm), which is the De Broglie wavelength of an elec-tron at the Fermi-level. These ultra-small nanoparticleshave been classified as the new class of nanomaterialscalled nanoclusters (NCs) that exhibit properties thatfundamentally differ from plasmonic counterpartsowing to the quantum size effects and extremely highsurface-to-volume ratio. Precisely, noble metalnanoclusters (e.g. Au, Ag) have been attracting at-tention for their unique role in bridging the “missinglink” between atomic and nanoparticle behaviour17.These sub-nanometre particles demonstrate molecu-lar-like electronic transitions18 between HOMO-LUMO energy levels due to their ultra-small finitecluster size19. As a result, the energy transitions canbe rationalized according to the jellium model (Efermi/N1/3)20. Although, the exact photoluminescence (PL)mechanism of metal NCs remains unknown and un-explored, it has been suggested that their lumines-cence is highly dependent on the size of the metalcores and surface ligands21. These metal NCs canproduce multitudinous measurable signals, such asluminescence and chemiluminescence, and have somedistinctive features such as ultrafine size with narrowsize distribution, and good photo-stability andbiocompatibility.

The ultra-small size of the NCs also allows theligand shell and the surface of the metal core to sig-nificantly contribute to their physical and chemicalproperties. The two possible mechanisms that couldbe used to explain the observed emission from thesesmall metal clusters intra-band (sp/conduction band)transition and inter-band (d-sp) transition. The smalland tuneable core-shell structure of the NCs can fa-cilitate their interaction with the analytes, especiallysmall molecules and ions resulting in improved sen-sor performance. Additionally, the properties of themetal core and the ligand shell in the NCs can be

Fig. 1. General approaches for the designing of opticalchemosensors.

From the last two decades, the nanotechnology-derived optically active materials such as semicon-ductor quantum dots, polymer dots, carbon dots, noblemetal nanoparticles/nanoclusters (NPs/NCs) providesan alternate to organic based chemosensors and be-comes an emerging candidate to increase portability,enhance stability, selectivity, sensitivity of sensorsand analytical measurement technologies. Variationin the properties of nanomaterials to that of bulk dif-fers in the redox properties, band gap, enhancementin toughness and strength, anomalous melting pointsand unusual crystal structures (in metals) which re-sults from smaller sizes, including the quantum sizeeffect on photochemistry, non-linear optical proper-ties of semiconductor or the emergence of metallicproperties with the size of the particles16. Among thevarious nanomaterials, the noble metal nanoparticles/nanoclusters (NPs/NCs) are widely applied in the fieldof sensor and biosensors development because of theone-pot easy synthesis and surface modification with

Bothra et al. : A short overview on fluorescent nanoclusters and its application in sensing of metal ions

1381

independently tailored during the synthesis or throughpost-synthetic modifications by using different metalsand organic ligands19. Moreover, the luminescent NCsoften have well-defined compositions (specific mo-lecular formulas) and structures22 that could facilitatethe investigations into their sensing mechanisms be-cause the products generated by the interaction be-tween the NCs and the analytes could be readilyanalysed at the molecular level by using advancedseparation and characterization techniques (e.g. high-resolution electrophoresis, mass spectrometry, and X-ray absorption spectroscopy). Thus, the fluorescenceemission of NCs could be readily adjusted from thevisible to near-infrared region leading to a wide rangeof applications, including chemical and biological sens-ing23, cellular and animal imaging24, as well as can-cer therapy25.

Synthesis of fluorescent metal nanoclusters

A broad range of luminescent metal NCs com-posed of gold (Au)23,26, silver (Ag)26,27, copper(Cu)26,28, platinum (Pt)26,29, bismuth (Bi)30, molyb-denum (Mo)31 or mixed metals32 have been success-fully fabricated using diverse top-down or bottom-upapproaches. Recently, many different synthetic ap-proaches including radiolytic, chemical, sonochemical,photochemical and microwave methods have beendeveloped to prepare water-soluble fluorescentNCs33,34 and some recent reviews appeared that dis-cussed the synthesis of NCs in details35. As the baremetal NCs are not stable for a longer period in thesolution, they are typically protected by organicligands. Thiolated, carboxylic and amine terminatedligands such as peptides, polymers, dendrimers andproteins are routinely used as protecting and stabilisingagent for the synthesis and stabilization of highly fluo-rescent NCs due to their unique and strong interac-tion with noble metals, which could lead to an ex-traordinary stability and distinct properties of photo-luminescence. This labelling of the biomolecule offluorescent metal NCs is the key basis for their fur-ther application to specific biosensing and bioimagingwhich is commonly based on passive adsorption, mul-tivalent chelation and covalent-bond formation. Im-portant parameters for controlling the size, structure,

oxidation state and surface properties of metal NCsinclude the species and concentration of chemicals(ligands) or templates, the concentration of metal ions,the species and concentration of reducing agents, pHof the solution, as well as the reaction temperatureand time. Various proteins, such as bovine serumalbumin (BSA)35,36, human serum albumin (HSA)37,insulin38, horseradish peroxidase39, pepsin40,lactotransferrin41, as well as lysozyme42,43 have beenemployed as templates for the preparation of the fluo-rescent metal NCs. Protein-directed synthesis is par-ticularly very attractive because proteins serve as en-vironmentally-benign reducing and stabilizing mol-ecules, require only mild reaction conditions, and offergreater water solubility and natural bio-compatibility(Fig. 2). Furthermore, the 3D complexed structuresof proteins can withstand a wide range of pH and canbe easily conjugated with other systems.

Fig. 2. Protein template based synthesis of nanoclusters.

Detect ion metal ions with f luorescent metalnanoclusters

A strong luminescence or high QY of the NCs iscrucial for realizing a good sensitivity of the fluores-cent sensors. The luminescence properties (e.g. emis-sion intensity and wavelength) of the NCs are highlysensitive to the local environment, the size and struc-ture of the NCs that provide an excellent response forsignalling their interaction with analytes. The fluo-rescence “turn-off”, “turn-on” and ratiometric detec-tion are three common schemes in the NCs-basedoptical sensors based on analyte-induced aggregationof NCs and other mechanisms based on energy/charge/electron transfer between the analytes and NCs44.Various AuNCs and AgNCs stabilized by proteinssuch as e.g. lysozyme42, BSA45,46, GSH47 etc. andpolymers such as polyethyleneimine (PEI) have beensuccessfully proposed for the construction of fluores-


1382

cent recognition of metal ions in aqueous medium. Insuch systems, three primary interactions between themetal (Au/Ag) core and analytes have been reported :metallophilic interactions, analyte deposition on themetal core surface, and analyte-induced metal coredecomposition. Highly luminescent AuNCs andAgNCs have been used to construct fluorescent sen-sors for a variety of chemical and biological ana-lytes49–53.

Recognition of heavy metal ions has gained sig-nificant importance due to their high toxicity can bindto various cellular components, leading to the dys-function, harm human health. Many fluorescent NCsas sensitive sensing probes have been used to detectvarious heavy metal ions, such as Zn2+, Ag+, As3+,Cd2+, Cr3+, Cr6+, Cu2+, Fe3+, Hg2+ etc. basedon fluorescence quenching or enhancement. For in-stance, the red fluorescent BSA or lysozyme templatedAuNCs were used as recognition probes for Hg2+

based on fluorescent quenching witha very low limitof detection46. This fluorescence quenching could bedue to metallophilic interactions between Hg2+ andAu+ on the Au cluster surface. By taking advantageof this 5d10–5d10 metallophilic interaction betweenHg2+ and Au+ that alter electronic structures ofAuNCs, many sensitive and selective sensing sys-tems have been developed for the detection ofHg2+42,48. Various nanoclusters for sensing of Cu2+

have been developed, for example, GSH-AuNCs pos-sess carboxylic group that can readily complex withCu2+ based on analyte induced fluorescence quench-ing with LOD was measured to be 86 nM at a signal-to-noise (S/N) ratio of 3, which is much lower thanthe U.S. EPA limit for Cu2+ in drinking water (20M)49. Although, turn-off assays may compromisespecificity since other quenchers or environmentalstimulus might also lead to fluorescence quenchingand report “false positive” results, thus, design anddevelopment of turn-on sensor is still receiving atten-tion. Recently, Chang and co-workers50 had demon-strated a novel, turn-on fluorescent assay for usingDNA-templated AgNCs for Cu2+, wherein additionof copper ions leads to formation of DNA-Cu/AgNCs with more complete protection from DNA tem-plates and thus, enhances the fluorescence.The HSA-

AgNCs were synthesized by tuning the procedure inorder to make them toggle between blue-emitting (Ag9:HSA) and red-emitting (Ag14 : HSA) nanoclusters.This novel probe acts as a dual sensor and can serveas luminescent turn “on” and “off” metal switchesfor Co2+ and Zn2+ ions, as shown in Fig. 351. AuNCsfunctionalised with dicysteine shows fluorescenceenhancement upon interaction with As3+ because theelectrons can flow from the electron rich AuNCs tothe electron deficient As3+, resulting in an increasein the radiative decay rate of the AuNCs. The devel-oped system provides LOD of 53.7 nM that is lowerthan the MAL (133 nM) of arsenic in drinking waterset by the U.S. EPA52. Recently, some interestingnanosensors for metal ions were developed using fluo-rescent nanoclusters are summarized in Table 1, whichindicate their potential to detect down to nanomolarlevel or less.

Fig. 3. Schematic representation showing the dual sensor swit-chable probe for Co2+ and Zn2+ using metalnanoclusters (Adapted from Ref. 51).

As described in Fig. 1, the fluorescent nanoclustersare directly applied for the sensing of analytes orapplied after further covalently conjugated with suit-able organic molecules. In our approaches, we havedeveloped some nanoclusters based nanosensors byconjugating with vitamin B6 cofactors like pyridoxaland pyridoxal 5-phosphate. The vitamin B6 cofactorplays crucial roles in enzymatically catalysed tran-saminations to form an -keto acid and pyridoxamine5-phosphate as well as in many other biosyntheticprocesses60,61. Pyridoxal containing enzymes are cen-tral to numerous metabolic pathways such as decar-boxylations of amino acids62, racemization of aminoacids63 and aldol type addition of the pyridoxal sta-bilized glycine carbanion to formaldehyde oracetaldehyde64. Considering the ability to detect metal


1383

ions by the Schiff base derivatives of vitamin B6derivatives65,66, we have designed and developed vi-tamin B6 cofactor (pyridoxal/pyridoxal-5-phosphate)conjugated fluorescent nanoclusters such as AuNCs/AgNCs templated with proteins (BSA/Lysozyme) orpolymers (PEI) for turn-on or turn-off sensing of metalions like Hg2+, Zn2+ and Cd2+ in aqueous medium67–69.The reported luminescent gold nanoclusters was first

synthesized using BSA as a template and then conju-gated with the vitamin B6 cofactor pyridoxal (Py)67.The free amines present in the coated BSA react withthe aldehyde group of pyridoxal and formed the pyri-doxal conjugated BSA-AuNCs. The developed pyri-doxal conjugated BSA-AuNCs system was success-fully applied for the selective fluorescent detection ofHg2+ in aqueous medium with low limit of detectionin comparison with the reported papers53. For realon-site detection, the paper-based analytical technol-ogy was developed for the sensing of Hg2+ in watervia covalent anchoring of the nanosensor onto thecellulose paper as shown in Fig. 4. The paper-basedsensors offer portability and operational simplicity byphysisorption of molecular sensors and can be ap-plied in many applications such as bio-diagnosis, waterand seafood control as well as environmentalmarkers. Similar to the Py_BSA-AuNCs approach,red fluorescent gold nanoclusters (Lyso-AuNCs) us-ing lysozyme and them conjugated with the vitaminB6 cofactor pyridoxal-5-phosphate (PLP)68. Uponaddition of PLP, the red fluorescence of Lyso-AuNCschanged to yellow due to the formation of a Schiffbase between the -CHO group of PLP and the free-NH2 present in the lysozyme (Fig. 5). Once the PLP

Table 1. Summary of the few reported AgNCs and AuNCsbased sensors for metal ions

Ions Functionalized NCs LOD Ref.

BSA-AuNCs 80 nM 53

BSA-AgNCs 48.7 nM 54

GSH-AgNCs 0.1 nM 55

GSH-AuNCs 0.2 nM 56

Hg2+ Lysozyme Type VI-AuNCs 3 pM 48

AuNCs 10 nM 42

Lipoic Acid-AuNCs 0.5 nM 57

Lipoic Acid-AgNCs 0.l nM 58

Lysozyme-AuNCs 86 nM 49

Cu2+ GSH and cyclodextrin AuNCs 1.3 ppm 59

PEI-AgNCs 48.7 nM 55

Zn2+ Salicylaldehyde-BSA-AuNCs 0.1 nM 56

Cr6+ Polyethyleneimine-AgNCs 0.2 nM 57

As3+ Cysteine-AuNCs 53.7 nM 52

Fig. 4. (a) The paper-based device prepared by the covalent modification with the pyridoxal conjugated BSA-AuNCs, (b)fluorescence color changes visualized on test paper strips of Py_BSA-AuNCs upon interaction with different concen-trations of Hg2+ (1 mM to 1 nM) observed under UV light at 365 nm and (c) the XPS spectrum of Py_BSA-AuNCsbefore and after addition of Hg2+ (Inset showing the XPS spectrum of the Au 4f band and Hg 4f band) (Adapted fromRef. 67).


1384

was conjugated with the Lyso-AuNCs, the nano-as-sembly PLP_Lyso-AuNCs was applied for the detec-tion of metal ions. The yellow fluorescence ofPLP_Lyso-AuNCs turned to bluish-green fluorescencegiving a significant turn-on fluorescence at 475 nmselectively in the presence of Zn2+ due to the com-plexation-induced aggregation of nanoclusters. Withthis system, the Zn2+ can be detected down to39.2 nM. Further, the nanoprobe was validated forthe detection of Zn2+ in various real samples. Inaddition, the system was applied to monitor the intra-cellular Zn2+ in live HeLa cells.

Polyelectrolytes such as polyethyleneimine (PEI)is a positively charged hyper-branched polyamine re-garded as an ideal template to modify and stabilizenanoparticles based on several advantages : firstly, PEIcontains primary, secondary and tertiary amine groups,which could efficiently chelate with metallic nucleus

and stabilize these particles against flocculation bythe strong columbic interaction due to the inherenthigh cationic density of the polymer. Thus, takingadvantage of the nature of PEI, we have recentlydeveloped a nano-assembly, where the vitamin B6cofactor PLP was conjugated over the surface of thefluorescent PEI-AgNCs and then applied for the fluo-rescent turn-on sensing of Zn2+ and Cd2+69. ThePLP conjugation was achieved due to the formationof a Schiff base upon interaction with the aldehydegroup of PLP and the free amines present in the PEI-AgNCs (Fig. 6). This PLP conjugated nanoprobeshowed a complexation-induce turn-on fluorescentresponse in the presence of Zn2+ and Cd2+ with thedetection limit down to 50.8×10–8 M for Zn2+ and58.0×10–8 M, respectively. This reversible fluores-cent nano-assembly with remarkable sensitivity havebeen successfully employed for the real samples analy-

Fig. 5. Schematic representation showing interaction of Zn2+ ions with conjugated Lyso-AuNCs and its intracellular detection(Adapted from Ref. 68).

Fig. 6. (a) Schematic representation showing the mechanism for interaction of Zn2+ and Cd2+ with developed conjugatednanoprobes PLP_PEI-AgNCs, (b) fluorescence emission spectra and (c) bar graph at 475 nm representing sensorreversibility upon interaction with EDTA. (d) The photographic images of the vials showing the reversibility in solutionand DVS modified cellulosic strips (Adapted from Ref. 69).


1385

ses in the quantitative detection of Zn2+ and Cd2+ invarious real samples69. In addition, using the DVS,the chemically-modified cellulosic strips were devel-oped using the PLP_PEI-AgNCs and applied for thesensing of Zn2+ and Cd2+.

Conclusions and perspectives

The unique physicochemical properties of noblemetal fluorescent NCs have gained wide applicationsin the diversified field including sensors and biosensorsdevelopment. This short review was narrated to presentan overview of the properties, synthesis and theirapplications for the detection of metal ions. Varioussmall molecules, proteins, polyelectrolytes etc. areused for the direct one-pot synthesis of noble metalfluorescent nanoclusters and applied for the sensingof metal ions. The direct interaction of metal ions onthe surface of nanoclusters with the coated moleculesand/or the metal core mostly leading to the quenchingof the fluorescence by various mechanisms. Consid-ering the advantages of fluorescent ‘turn-on’ sensor,the further conjugation of nanoclusters with suitableorganic molecules may provide new directions forthe design of turn-on sensors. Such conjugation mayalso provide extra-stability to the nanoclusters. Withthis review, we observed that despite many reportson the synthesis and applications of fluorescent NCs,there is a need of deeper research for the develop-ment of common and efficient route for the synthesisof different types of NCs with improved stability andquantum yield. Further research is also required tounderstand the signal-generation mechanisms from theNCs upon interaction with analytes. With the furtherresearch and development, the noble metal NCs willprovide the novel analytical approaches for themonitoring of metal ions in environmental and bio-logical samples as well as create new directions forunderstanding the various medical and environmentalcomplexity raised due to the metal ions toxicity.

References

1. J. Anastassopoulou and T. Theophanides, "The Role ofMetal Ions in Biological Systems and Medicine", in :D. P. Kessissoglou (Ed.), ‘Bioinorganic Chemistry : AnInorganic Perspective of Life’, Springer Netherlands,Dordrecht, 1995, pp. 209-218.

2. S. Jjrfd and R. Williams, Clarendon Press, Oxford, 1991.

3. S. K. Sahoo, D. Sharma, R. K. Bera, G. Crisponi andJ. F. Callan, Chem. Soc. Rev., 2012, 41, 7195.

4. M. H. Mashhadizadeh and M. Amoli-Diva, J. Anal.Atom. Spectrom., 2013, 28, 251.

5. Z. Huang, Z. Zhu, Q. Subhani, W. Yan, W. Guo andY. Zhu, J. Chromatog. (A), 2012, 1251, 154.

6. A. M. Ashrafi and K. Vytr as, Talanta, 2011, 85, 2700.

7. S. Bothra, J. N. Solanki and S. K. Sahoo, Sensors andActuators B : Chem., 2013, 188, 937.

8. Y. Cheng, M. Zhang, H. Yang, F. Li, T. Yi and C.Huang, Dyes and Pigments, 2008, 76, 775.

9. K.-C. Song, M. H. Kim, H. J. Kim and S.-K. Chang,Tetrahedron Lett., 2007, 48, 7464.

10. G. W. Gokel, W. M. Leevy and M. E. Weber,Chem. Rev., 2004, 104, 2723.

11. S. Kamata, H. Murata, Y. Kubo and A. Bhale,Analyst, 1989, 114, 1029.

12. M. Lee, S. Jo, D. Lee, Z. Xu and J. Yoon, Dyesand Pigments, 2015, 120, 288.

13. J. W. Jeong, B. A. Rao and Y.-A. Son, Sensorsand Actuators B : Chem., 2015, 208, 75.

14. D. Jiménez, R. Martìnez-Máñez, F. Sancenón andJ. Soto, Tetrahedron Lett., 2002, 43, 2823.

15. P. Carol, S. Sreejith and A. Ajayaghosh, Chemis-try–An Asian Journal, 2007, 2, 338.

16. C. Feldmann, Advanced Functional Materials,2003, 13, 101.

17. J. Zheng, C. Zhang and R. M. Dickson, Phys.Rev. Lett., 2004, 93, 077402.

18. R. Jin, Nanoscale, 2015, 7, 1549.

19. X. Yuan, Z. Luo, Y. Yu, Q. Yao and J. Xie,Chemistry–An Asian Journal, 2013, 8, 858.

20. J. Zheng, P. R. Nicovich and R. M. Dickson,Annu. Rev. Phys. Chem., 2007, 58, 409.

21. J. Zheng, C. Zhou, M. Yu and J. Liu, Nanoscale,2012, 4, 4073.

22. C. Zeng, H. Qian, T. Li, G. Li, N. L. Rosi, B.Yoon, R. N. Barnett, R. L. Whetten, U. Landmanand R. Jin, Angew. Chem., 2012, 124, 13291.

23. L. Shang, S. Dong and G. U. Nienhaus, Nano To-day, 2011, 6, 401.

24. P. Zhang, X. X. Yang, Y. Wang, N. W. Zhao, Z.H. Xiong and C. Z. Huang, Nanoscale, 2014, 6,2261.

25. Y. Wang, J. Chen and J. Irudayaraj, ACS Nano,2011, 5, 9718.

26. X. Yuan, Z. Luo, Q. Zhang, X. Zhang, Y. Zheng,J. Y. Lee and J. Xie, ACS Nano, 2011, 5, 8800.

27. X. Yang, L. Gan, L. Han, E. Wang and J. Wang,

JICS-17


1386

Angew. Chem. Int. Ed., 2013, 52, 2022.

28. J. Chen, J. Liu, Z. Fang and L. Zeng, Chem.Commun., 2012, 48, 1057.

29. S. I. Tanaka, J. Miyazaki, D. K. Tiwari, T. Jinand Y. Inouye, Angew. Chem., 2011, 123, 451.

30. H.-T. Sun, Y. Matsushita, Y. Sakka, N. Shirahata,M. Tanaka, Y. Katsuya, H. Gao and K. Kobayashi,J. Am. Chem. Soc., 2012, 134, 2918.

31. Y. Molard, C. Labbé, J. Cardin and S. Cordier,Adv. Funct. Mater., 2013, 23, 4821.

32. C. M. Andolina, A. C. Dewar, A. M. Smith, L.E. Marbella, M. J. Hartmann and J. E. Millstone,J. Am. Chem. Soc., 2013, 135, 5266.

33. H. Xu and K. S. Suslick, Adv. Mater., 2010, 22,1078.

34. L.-Y. Chen, C.-W. Wang, Z. Yuan and H.-T.Chang, Anal. Chem., 2015, 87, 216.

35. J. Xie, Y. Zheng and J. Y. Ying, J. Am. Chem.Soc., 2009, 131, 888.

36. Y. Wang, Y. Wang, F. Zhou, P. Kim and Y. Xia,Small, 2012, 8, 3769.

37. P.-H. Chan and Y.-C. Chen, Anal. Chem., 2012,84, 8952.

38. C.-L. Liu, H.-T. Wu, Y.-H. Hsiao, C.-W. Lai,C.-W. Shih, Y.-K. Peng, K.-C. Tang, H.-W.Chang, Y.-C. Chien, J.-K. Hsiao, J.-T. Cheng,P.-T. Chou, Angew. Chem. Int. Ed., 2011, 50,7056.

39. F. Wen, Y. Dong, L. Feng, S. Wang, S. Zhangand X. Zhang, Anal. Chem., 2011, 83, 1193.

40. H. Kawasaki, K. Hamaguchi, I. Osaka and R.Arakawa, Adv. Funct. Mater., 2011, 21, 3508.

41. P. L. Xavier, K. Chaudhari, P. K. Verma, S. K.Pal and T. Pradeep, Nanoscale, 2010, 2, 2769.

42. H. Wei, Z. Wang, L. Yang, S. Tian, C. Hou andY. Lu, Analyst, 2010, 135, 1406.

43. J. Das and S. O. Kelley, Anal. Chem., 2013, 85,7333.

44. Z. Xu, X. Chen, H. N. Kim and J. Yoon, Chem.Soc. Rev., 2010, 39, 127.

45. J. Xie, Y. Zheng and J. Y. Ying, Chem. Commun.,2010, 46, 961.

46. C. Zhang, Z. Guo, G. Chen, G. Zeng, M. Yan,Q. Niu, L. Liu, Y. Zuo, Z. Huang and Q. Tan,New J. Chem., 2016, 40, 1175.

47. G. Zhang, Y. Li, J. Xu, C. Zhang, S. Shuang, C.Dong and M. M. F. Choi, Sensors and ActuatorsB : Chem., 2013, 183, 583.

48. Y.-H. Lin and W.-L. Tseng, Anal. Chem., 2010,82, 9194.

49. G. Zhang, Y. Li, J. Xu, C. Zhang, S. Shuang, C.Dong and M. M. Choi, Sensors and Actuators B :

Chem., 2013, 183, 583.

50. G.-Y. Lan, C.-C. Huang and H.-T. Chang, Chem.Commun., 2010, 46, 1257.

51. S. Ghosh, U. Anand and S. Mukherjee, Anal.Chem., 2014, 86, 3188.

52. S. Roy, G. Palui and A. Banerjee, Nanoscale,2012, 4, 2734.

53. D. Hu, Z. Sheng, P. Gong, P. Zhang and L. Cai,Analyst, 2010, 135, 1411.

54. D. Lu, C. Zhang, L. Fan, H. Wu, S. Shuang andC. Dong, Anal. Methods, 2013, 5, 5522.

55. C. Wang, L. Xu, Y. Wang, D. Zhang, X. Shi, F.Dong, K. Yu, Q. Lin and B. Yang, Chemistry–AnAsian Journal, 2012, 7, 1652.

56. Q. Zhao, S. Chen, L. Zhang, H. Huang, Y. Zengand F. Liu, Analytica Chimica Acta, 2014, 852,236.

57. L. Shang, L. Yang, F. Stockmar, R. Popescu, V.Trouillet, M. Bruns, D. Gerthsen and G. U.Nienhaus, Nanoscale, 2012, 4, 4155.

58. B. Adhikari and A. Banerjee, Chem. Materials,2010, 22, 4364.

59. A. George, E. Shibu, S. M. Maliyekkal, M.Bootharaju and T. Pradeep, ACS Appl. Mater. In-ter., 2012, 4, 639.

60. W. Liu, P. E. Peterson, J. A. Langston, X. Jin,X. Zhou, A. J. Fisher and M. D. Toney, Bio-chemistry, 2005, 44, 2982.

61. H. Hayashi, H. Mizuguchi, I. Miyahara, M. M.Islam, H. Ikushiro, Y. Nakajima, K. Hirotsu andH. Kagamiyama, Biochimica et Biophysica Acta(BBA)–Proteins and Proteomics, 2003, 1647,103.

62. E. J. Fogle, W. Liu, S.-T. Woon, J. W. Kellerand M. D. Toney, Biochemistry, 2005, 44, 16392.

63. S. Sun and M. D. Toney, Biochemistry, 1999, 38,4058.

64. T. Mukherjee, J.o. Costa Pessoa, A. Kumar andA. R. Sarkar, Inorg. Chem., 2011, 50, 4349.

65. S. Bothra, R. Kumar and S. K. Sahoo, RSC Adv.,2015, 5, 97690.

66. D. Sharma, A. Moirangthem, R. Kumar, S. K.Ashok Kumar, A. Kuwar, J. F. Callan, A. Basuand S. K. Sahoo, RSC Adv., 2015, 5, 50741.

67. S. Bothra, Y. Upadhyay, R. Kumar, S. K. AshokKumar, S. K. Sahoo, Biosensors and Bioelectron-ics, 2017, 90, 329.

68. S. Bothra, L. T. Babu, P. Paira, S. Ashok Kumar,R. Kumar and S. K. Sahoo, Anal. Bioanal. Chem.,(2017) https://doi.org/10.1007/s00216-017-0710-2.

69. S. Bothra, P. Paira, A. Kumar, R. Kumar and S.K. Sahoo, Chemistry Select, 2017, 2, 6023.

1387


Structure, antimicrobial activity and molecular docking study of novelo-vaniline based sulfamethoxazolyl derivatives†

Nilima Sahua, Sudipa Mondala, Kaushik Naskara, Suvroma Guptab, D. Kaleeswaranc andChittaranjan Sinha*a

aDepartment of Chemistry, Jadavpur University, Kolkata-700 032, India

E-mail : [email protected] of Biotechnology, Haldia Institute of Technology, Haldia, Purba Medinipur-721 657,West Bengal, IndiacDepartment of Chemistry, Indian Institute of Technology Bombay, Powai,Mumbai-400 076, India

Manuscript received 05 December 2017, accepted 06 December 2017

Abstract : Four sulfamethoxazolyl-azo Schiff bases, SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4–R)(Sulfamethoxazole, SMX; R = H (2a), -CH3 (2b), -Cl (2c), -OCH3 (2d)) have been prepared by the condensa-tion of SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CHO) (1) with R-C6H4-NH2. The compounds are characterized byspectroscopic studies (FT-IR, UV-Vis, Mass, NMR). The structural confirmation has been carried out by singlecrystal X-ray diffraction studies in one case, SMX-N=N-C6H2(p-OH)(m-OCH3)-C=N-C6H4-p-CH3) (2b). Thesupramolecular 1D chain is constituted by inter- and intra-molecular hydrogen bonds and also by --- inter-action of aromatic rings. The antibacterial properties have been evaluated against Gram-positive bacteria. Mostfavored binding mode of the drugs with the active site residues of DHPS (dihydropteroate synthetase) is es-tablished by in silico Molecular Docking using Discovery studio 3.5 software.

Keywords : Sulfamethoxazolyl-azo-imine derivatives, o-vanillin, X-ray structure, DFT, antimicrobial activity,molecular docking.

Introduction

Antibiotics are used as life-saving drugs sincethe discovery of penicillin in 19281. Recently, an-tibiotic resistance is a common incident2. Conse-quently, design of newer antibiotics derived fromconventional antibiotics by chemical functiona-lization is becoming extremely urgent3,4. There-fore, search of more effective and low toxic drugshas taken fantastic impetus in the area of syntheticand pharmaceutical chemistry5. Azo compoundsshow a variety of interesting biological activitiessuch as, antineoplastics, antidiabetics, antiseptics,anti-HIV and other useful chemotherapeutic prop-erties6. Schiff bases are one of the most exten-

†Invited Lecture.

sively used organic compounds and divulge a widecollection of biological activities7,8 such as, anti-fungal9,10, antibacterial11,12, antiproliferative, an-timalarial13,14 and antipyretic. Therefore, sulfona-mides consisting of both azo (-N=N-) and imine(-C=N-) functions may serve as challenging mol-ecules in the medicinal and pharmaceuticalfields15,16.

Sulfonamide (-SO2NH-), an antibacterial, anti-tumor, anti-viral, anti-fungal, diuretic, antithyroid,hypoglycaemic etc. hinder the synthesis of folicacid. Folate, useful agent for synthesis and regu-lation of nucleic acids (DNA, RNA) in the cells, issynthesized by direct participation of DHPS in a


1388

catalytic cycle17–19. Due to structural analogy ofsulfonamides, C6H4(COOH)(p-NH2) (p-aminoben-zoic acid, PABA) inhibits dihydropteroatesynthatase (DHPS), hence regulates proliferationand growth of bacteria. However, prolong admi-nistration of sulfonamide causes many undesirableeffects including liver disease, kidney injury, skinrashes, drug fever, lung infection and hemolyticanemia20–22. These toxicity and antibiotic resis-tance can be regulated by the functionalization ofsulfonamides. In continuation of our research,herein, we have designed a new series of sulfona-mides consisting of both azo and imine func-tions23–27. o-Vaniline, a natural aldehyde found inAndropogen nardus28 and used for the treatmentof sickle cell disease (SCD)29, is coupled with dia-zonium salt of sulphamethoxazole (SMX-N=N-+)to synthesize SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CHO) (1) which is then condensed with p-R-C6H4-NH2 to isolate azo-imine functionalized sulfon-amides, SMX-N=N-C2H2(p-OH)(m-OCH3)(m-CH=N-C6H4-R) (R = H (2a), -CH3 (2b), -Cl (2c),-OCH3 (2d)). Following different spectroscopictechniques the molecules 2, have been characte-rized along with single crystal X-ray structure ofone of the derivatives. The DFT computation hasbeen performed to explain the electronic structureand spectra. Antimicrobial activity of the com-pounds has been evaluated against several stan-dard Gram-positive and Gram-negative bacterialstain. Docking of these molecules with DHPS pro-tein has been focused to investigate the most pre-ferred binding mode and hence the plausiblemechanism of antimicrobial activity.

Experimental

Materials and methods

Sulfamethoxazole was purchased from Hi-Me-dia. Sodium nitrite, sodium hydroxide, were avail-able from S.D. Fine Chem. Ltd., o-vaniline, aniline,toluedine, anicidine, p-chloroaniline were pur-chased from Aldrich, and other chemicals and sol-vents were reagent grade purified by standard pro-cedure30.

Physical measurements

Microanalytical data (C, H, and N) were col-lected on Perkin-Elmer 2400 CHNS/O elementalanalyzer. UV-Vis spectra were collected fromPerkin-Elmer UV-Vis spectrophotometer modelLambda 25; FTIR spectra (KBr disk, 4000–400cm–1) were assembled from Perkin-Elmer FT-IRspectrophotometer model RX-1; the 1H NMR spec-tra were obtained from Bruker (AC) 300/500 MHzFTNMR spectrometer. ESI mass spectra were re-corded on a micro mass Q-TOF mass spectrom-eter (Serial no. YA 263). Antimicrobial activity isperformed by the Gram-positive bacteria strain.

Synthesis of 4-((3-formyl-4-hydroxy-5-methoxy-phynyl)diazenyl)-N-(5-methyl isoxazol-3-yl)benzene-sulfonamide (1)

Dropwise addition of ice cold aqueous solution(0–5 ºC) of sodium nitrite (0.272 g, 3.95 mmol)to cold 3(N) 18 ml HCl solution of sulphamethoxa-zole (SMX, 1.0 g, 3.95 mmol) formed sulfa-methoxazolyl diazonium (SMX-N=N-+) salt andit was kept for 25 min. Then the diazotized prod-uct was added dropwise to the cold sodium hy-droxide (2.4 g) solution of o-vaniline (0.60 g, 3.95mmol) with continuous stirring; yellow precipitateappeared at pH 7 (Scheme 1) which was filteredand dried. Then it was re-crystallized from aque-ous methanol solution by slow evaporation. Forfurther purification the column chromatographywas performed (in silica gel, 60–120 mesh) andthe desired product was eluted with 1 : 10 (v/v),petroleum ether (60–80)-ethylacetate eluent. Theisolated yield was 68%.

SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CHO), 1,C18H16N4O6S (M.W. 416.4) : m.p, 218 ºC Mi-croanalytical data : Calcd. C, 51.92; H, 3.87; N,13.45; Found : C, 51.84; H, 3.93; N, 13.38%; IRmax (cm–1); (C-O), 1084; (N=N),1453; (C-N), 1616; (O-H), 3438; (S-O), 1176. (Supple-mentary Materials, Fig. S1); 1H NMR (CDCl3) (17-CH3)(oxazole) 2.38 (s); (16-H), 8.02 (s); (O-H), 11.66 (s); (N-H), 6.27 (s); (CH-N) 10.05(s); (-OCH3)(vaniline) 4.04(s); (5-H), 7.72 (s);

Sahu et al. : Structure, antimicrobial activity and molecular docking study of novel o-vaniline etc.

1389

(7-H), 7.75(s); (10,12-H), 7.99 (d, J=3.48);(9,13-H), 7.95 (d, J=2.13) ppm (SupplementaryMaterials, Fig. S2). Mass (m/z), (M+Na)+

(439.06) (Supplementary Materials, Fig. S3). UV-Vis spectroscopic data in MeOH (max(nm)(10–4

(dm3 mol–1 cm–1))) : 265 (2.67), 388 (2.19).

Azo sulfonamide Schiff bases, SMX-N=N-C6H3(p-OH)(m-OCH3)(m-CH=N-Ar), 2 (Ar =-C6H5 (2a), -C6H4-p-CH3 (2b), C6H4-p-Cl (2c),-C6H4-p-OCH3 (2d)

To hot ethanol solution of 1 (0.1 g, 0.24 mmol)was refluxed with Ar-NH2 (0.30 mmol) and re-fluxed for 5 to 10 h. The solution was then al-lowed to evaporate slowly in air and orange yel-low crystals were deposited on glass wall. Thecrystals were collected and TLC test was performedto check purity. The product, SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-Ar), 2 (Ar = -C6H5 (2a),-C6H4-p-CH3 (2b), C6H4-p-Cl (2c), -C6H4-p-OCH3(2d)), was recrystallized from methanol solution.The yield was 63–72%.

SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H5), C24H21N5O5S (2a) (M.W. 491.5) (yield,68%), m.p. 240 ºC. Calcd. : C, 58.64; H, 4.30;N, 14.25; Found : C, 58.73; H, 4.25; N, 14.21%.IR max (cm–1) : 1128, (C-O); 1493, (N=N);1618, (C=N); 3068, (O-H); 1165, (SO2));(Supplementary Materials, Fig. S4); 1H NMR(DMSO-d6) (17-CH3)(oxazole) 2.29 (s); (O-H),10.37 (s); (N-H), 6.16 (s); (CH-N) 9.21 (s); (-OCH3)(vaniline) 3.95(s); (16-H), 7.99 (s); (5H,phenyl protons of aniline, merged) 8.00 ppm, (2H,phenyl protons) 7.69 ppm; (4H, phenyl protonsof benzene ring of SMX) 7.53 ppm (Supplemen-tary Materials, Fig. S5). UV-Vis spectroscopic datain MeOH (max(nm)(10–4 (dm3 mol–1 cm–1))) :226 (6.73), 381 (2.42). Mass (m/z), (M+Na)+

(414.12) (Supplementary Materials, Fig. S6).

SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-CH3), C25H23N5O5S (2b) (M.W 505.53)(yield, 72%),: m.p. 222 ºC. Calcd. C, 59.39; H,4.58; N, 13.85; Found : C, 60.56; H, 4.40; N,14.01%. IR max (cm–1) : 1128, (C-O); 1466,

(N=N); 1622, (C=N); 3445, (O-H); 1173, (S-O) (Supplementary Materials, Fig. S7). 1H NMR(DMSO-d6) (17-CH3)(oxazole) 2.29 (s); (21-CH3)(toluedine) 2.34; (16-H), 7.99 (s); (O-H),12.12 (s); (N-H), 6.15 (s); (CH-N) 9.20 (s);(-OCH3)(vaniline) 3.94(s) (6H, phenyl protons)7.99 ppm, (3H, phenyl protons) 7.44–7.50 ppm(2H, phenyl protons) 7.32 ppm (SupplementaryMaterials, Fig. S8). UV-Vis spectroscopic data inMeOH (max(nm)(10–4 (dm3 mol–1 cm–1))) : 226(7.66), 379 (2.00). Mass (m/z), (M+Na)+ (528.10)(Supplementary Materials, Fig. S9).

SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-Cl), C24H20N5O5SCl (2c) (M.W. 525.89)(yield, 67%) : m.p. 230 ºC. Calcd. : C, 59.39; H,4.58; N, 13.85; Found : C, 59.56; H, 4.49; N,14.01%. IR max (cm–1) : 1128, (C-O); 1464,(N=N); 1617, (C=N); 3450, (O-H); 1172, (S-O) (Supplementary Materials, Fig. S10). 1H NMR(DMSO-d6) (17-CH3)(oxazole) 2.29 (s); (16-H),7.99 (s); (O-H), 12.14(s); (N-H), 6.16 (s); (CH-N) 9.18 (s); (-OCH3)(vaniline) 3.94(s) (6H, phe-nyl protons) 8.00 ppm, (4H, phenyl protons) 7.50ppm (Supplementary Materials, Fig. S11). UV-Vis spectroscopic data in MeOH (max(nm)(10–4 (dm3 mol–1 cm–1))) : 226 (11.18), 381 (3.05).Mass (m/z), (M+Na)+ (549.12) (SupplementaryMaterials, Fig. S12).

SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-OCH3), C25H23N5O6S (2d) (M.W. 521.52)(yield, 63%), : m.p. 248 ºC. Calcd. : C, 57.57;H, 4.83; N, 13.43; Found : C, 57.45; H, 4.75;N, 13.40% by wt. IR max (cm

–1) : 1128, (C-O);1466, (N=N); 1617, (C=N); 3417, (O-H);1172, (S-O). (Supplementary Materials, Fig. 13).1H NMR (DMSO-d6) (17-CH3)(oxazole) 2.29 (s);(16-H), 7.98 (s); (O-H), 10.27 (s); (N-H), 6.22(s); (CH-N) 8.67 (s); (-OCH3) (vaniline) 4.00(s); (-OCH3)(toluidine) 3.85(s); (16-H), 7.73 (s);(4H, 19,20,22,23 phenyl protons) 7.98 ppm;(10,12-H), 7.35 (d, J=7.88); (9,13-H), 7.00(d, J=8.07) ppm; (7-H) 7.73 (s) ppm; (5-H)7.56 (s) ppm (Supplementary Materials, Fig. S14).


1390

UV-Vis spectroscopic data in MeOH (max(nm)(10–

1 (dm3 mol–1 cm–1))) : 233 (10.86), 378 (2.00).Mass (m/z), (M+Na)+ (544.15) (SupplementaryMaterials, Fig. S15).

X-Ray crystal structure analysis of SMX-N=N-C6H3(p-OH)(m-OCH3)(m-CH=N-C6H4-p-CH3), (2b)

The crystal of 2b was formed by the slow evapo-ration of methanol solution (0.12×0.10×0.02mm3) for a week. Data were collected (Table 1)by Rigaku Saturn 724+ CCD diffractometer witha Mo-K; radiation source (Wavelength = 0.71075Å) at 150 K under continuous flow of cooled ni-trogen gas with in the range 2.363 24.999and h k l range : –18 h 18, –23 k 23, –19 l 17. The data collection, integration and in-dexing were performed using Crystalclear-SM soft-ware and a numerical method was employed tocorrect for absorption. All the calculations werecarried out using the programs in WinGX moduleand the structure solution was solved by directmethods using SIR-9231. The final refinement ofthe structure was carried out using full least-squaremethods on F2 using SHELX-1432. All non-hy-drogen atoms were refined anisotropically. Thehydrogen atoms were refined isotropically as rigidatoms in their idealized locations. Crystallographicrefinement data are collected in Table 1. Residualminimum and maximum electron densities are–0.350 and 0.440.

Antimicrobial activity

[SMX-N=N-C6H2-(p-OH)(m-OCH3)(m-CHO)](1) and its four derivatives SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H5) (2a), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-CH3)(2b), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-Cl) (2c), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-OCH3) (2d) weretested against Gram-positive, B. subtillis (ATCC6633). The solution of the compounds were pre-pared by HPLC grade DMSO; final % of DMSOduring assay was varied from 0.5–0.8% althoughthe main stocks solution of the drugs were pre-

pared using spectroscopic grade DMSO and storedat –20 ºC. All the bacterial strains were inocu-lated in a freshly prepared autoclaved LB brothfrom a 24 h old LA slant and kept in shaker forovernight. The final bacterial count was 1×104

cells/ml with sterile LB by dilution and the over-night culture. The diluted culture was distributedin a number of tubes and incubated in absenceand in presence of test compounds. Now the tubeswere incubated for 16–18 h at 37 ºC with con-tinuous shaking. The OD600 had been measuredin a UV-Visible spectrophotometer (Shimadzu).The 100% growth of bacterial species had beenconsidered in absence of test ligands. The OD at600 nm were measured for each concentrationincubated under similar experimental condition and

Table 1. Crystal data and structure refinement of SMX-N=N-C6H3(p-OH)(m-OCH3)(m-CH=N-C6H4-p-CH3) (2b)

2bEmpirical formula C50H46N10O10S2

Formula weight 1011.09

Temperature (K) 150 (2)

Crystal system Monoclinic

Space group P 21/c

a (Å) 15.806(3)

b (Å) 19.565(3)

c (Å) 16.462(3)

(º) 90

(º) 106.932(3)

(º) 90

V (Å)3 4870.1(15)

Z 4

(Mo-K (mm–1) 0.180

range 2.363–24.999

Dcalc (mg m–3) 1.379

Refine parameters 657

Total reflections 36488

Unique reflections 8557

R1a [I > 2 (I)] 0.0522 (8159)

wR2b 0.1333 (8557)

GOFc 1.124

Difference between peak and hole(e Å–3) 0.711, –0.503aR = |F0 –Fc|/ F0.

bwR = [ w(F02 – Fc

2)/ w F04]1/2

are general but w are different, w = 1/[2 (Fo2) + (0.0508P)2

+1.7997P] where P = (Fo2 + 2Fc

2)/3.


1391

compared to control the relative degree of bacte-rial growth inhibition had been calculated. Theconcentration of test compound required to inhibitthe growth of bacteria by 50% had been calcu-lated from the % reduction of bacterial growth incomparison to control (The minimum inhibitoryconcentrations i.e. IC50).

Molecular docking studies of the ligand with DHPS andADMET

The structures of proteins of DHPS(Dihydroptorate Synthase of Versinia pestis, PDBID 3TZF) were downloaded from the RCSB pro-tein data bank (http://www.pdb.org) and used fordocking. Sulfamethoxazole, 6-hydroxymethyl-pterin-diphosphate and magnesium ion were co-crystallized with enzyme. Using CDOCKER Re-ceptor-Ligand interactions protocol section of Dis-covery Studio client 3.533 in the in silico dockingmodule all the calculations were performed. Inter-action energies between ligands and receptor wererecorded. The optimized structures of the com-pounds (1, 2a-d) have been used for docking study.Ligand preparation was done using prepare ligandmodule and protein preparation was done underPrepare Protein module in Receptor-Ligand inter-actions tool of Discovery studio 3.5 and preparedligand was used for docking. The active site wasselected based on the ligand binding domain ofsulfamethoxazole then the pre-existing ligand wasremoved and prepared ligands, were placed. TheCDOCKER protocol was used for docking calcu-lations and a grid box centered at the geometricalcenter of co-crystallized ligand was used. The co-ordinates x, y, z for the center of grid box were41.78 Å, 8.05 Å, 2.24 Å and –77.56 Å, 85.05 Å,93.95 Å respectively. Minimum free energy ofprotein-ligand complex module was used to selectthe most favorable docked pose and analyzed toinvestigate the interaction. Absorption, distribu-tion, metabolism, excretion and toxicity (ADMET)prediction were done in ADMET descriptor mod-ule of small molecules protocol of Discovery stu-dio client 3.5. Druglikeness of the prepared ligands

were checked following Lipinski’s rule of five34,35.Using ADMET module of small molecule proto-col of Discovery studio 3.5 software ADMET prop-erties and toxicity of the compounds were checked.

Computational studies

The geometry optimization of 1 and 2a-d werecarried out using Density Functional Theory (DFT)at the B3LYP level36. All calculations were car-ried out using the Gaussian 09 program package37

with the aid of the GaussView visualization pro-gram38. For C, H, N, O the 6-31G(d) basis setwere assigned and the calculations for molecularorbitals were carried out as before38–40 with theaid of the GaussView visualization program41,42.The vibrational frequency calculations were per-formed to ensure that the optimized geometriesrepresent the local minima and there are only posi-tive eigen values. The electronic excitations werecomputed using the time-dependent density func-tional theory (TD-DFT) formalism in methanolusing conductor-like polarizable continuum model(CPCM)43–45 GaussSum was used to calculate thefractional contributions of various groups to eachmolecular orbital46.


Synthesis and spectroscopic characterisation

Sulfamethoxazolyl diazonium ion (SMX-N=N-+) is coupled with o-vaniline at pH 7 to prepare4- ( (3- formyl-4-hydroxy-5-methoxyphynyl )diazenyl)-N-(5-methylisoxazol-3-yl) benzene sul-fonamide [SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CHO)] (1) which undergoes condensation reac-tion with aromatic amines p-R-C6H4-NH2 to syn-thesize SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-R)) (R = H (2a), -CH3 (2b), -Cl(2c), -OCH3 (2d)) (Scheme 1). In infra-red spec-trum of 1 the presence of strong stretch at 1653cm–1 confirms (CHO) which disappears andemergences a new stretch at 1610–1625 cm–1 in 2which refers to (C=N) and supports the conden-sation reaction. Other significant stretches are(N=N), 1470–1480; (C-O), 1090–1110; (SO2),

-


1392

1170–1180 cm–1 in support of the synthesis of thecompounds (Supplementary Materials, Figs. S4,S7, S10, S13). All the compounds show absorp-tion band at 370–390 nm and 220–270 nm due ton-* and -* transitions, respectively (Fig. 1).The 1H NMR spectra show very complex patternin aromatic zone due to large number of protonsignals from aromatic rings (atom numbering isgiven in the structures, Scheme 1); however sig-nificant signals such as, oxazolyl-CH3 (2.45–2.50ppm), (OH) (10.00–12.50 ppm), (NH) (6.10–6.20 ppm); (CH=N), 8.10–9.20 ppm, (-OCH3)(vaniline) 4.00–3.90 ppm, oxazolyl-H (16-H, sin-glet) 8.29–8.31 ppm support the structure of the

compounds. Besides, sulfonamide-phenyl-Hs(9,10,12,13-H) become visible as two doublets at7.35–7.45 ppm. Azophenolato-H such as 7-H is asinglet at 7.10–7.15 ppm while 4,5-H show dou-blet signal at 7.10–7.30 ppm. The terminal sixmember ring, arylamine has four or five-Hs andhave shown noteworthy signal movement due tosubstituent effect; electron donating -CH3 and -OCH3 influence 21,23-H to shift to upfiled sidewhile electron withdrawing groups -Cl influencethem to move to downfield side (SupplementaryMaterials; Figs. S2, S5, S8, S11, S14). The singlecrystal structure of 1b confirms the proposal fromother spectroscopic information.

Scheme 1. Synthesis of SMX-azo-imine derivatives; SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CHO) (1), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H5) (2a), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-CH3) (2b), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-Cl) (2c), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-OCH3) (2d).


1393

Crystal structure description of 2b

The compound 2b has been crystallized inmonoclinic crystal system and P 21/c space group.The asymmetric unit of the crystal lattice is consti-tuted of two molecules and these are structurallycomparable. A view of the diad including the atomnumbering scheme is shown (Supplementary Ma-terials, Fig. S16). A single molecule is shown inFig. 2 and selected bond parameters are listed inTable 2. In terms of bond distances, angles, aswell as gross geometry, these two molecules areclosely akin to one another.

The four part of the structure are iminephenolato (A-ring), aryl-sulfonamide (B-ring),

oxazolyl (C-ring), methyl substitute phenyl (D-ring).The C-ring is linked to B-ring by the sulfonamide(-NH-SO2-) group in sulfamethoxazole part andthe -N=N- group of B-ring is connected to A-ringand A-ring is attached to D-ring by -HC=N- group.The -C6H3(OH)(OCH3)- (A-ring) and -C6H4(SO2)-(B-ring) are bonded by -N=CH- and make dihe-dral 15.88(13)º; oxazolyl ring (C-ring) is inclinedat 88.16(17)º with B-ring. The N=N azo length is1.260(3) Å and the C=N imine length is 1.309(3)Å who are closer to the reported data23–27. Theresults indicate that the bond strength between azo-N and substituted phenol and imine bond is stron-ger than that of azo-N and sulfonamide-phenyl. In2b the supramolecular aggregation takes placethrough hydrogen bonding (NH...O) and non-clas-sical interaction S-O...CH. (Fig. 3). There are threedifferent mode of CH... interaction in 2b whichis viewed in Fig. 3. The CH... distances in thethree different forms are respectively C-H(25)C...Ar 2.774 Å (form 1); C-H(1B)...Ar 2.903Å (form 2); C-H(1C)...Ar, 2.420 Å (form 2); C-H(50B)...Ar 3.428 Å (form 3) and C-H(42B)...Ar2.880 Å (form 3). The selective bond lengths areS1-N2, 1.654 Å; N3-N4, 1.271 Å; N8-N9, 1.260Å; C18-N5, 1.310 Å and C43-N10, 1.309 Å (Table2).

Antimicrobial activity

The compound [SMX-N=N-C6H2-(p-OH)(m-OCH3)(m-CHO)] (1) and its four Schiff bases SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H5)(2a), SMX-N=N-C6H2(p-OH)(m -OCH3)(m-CH=N-C6H4-p-CH3) (2b), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-Cl) (2c), SMX-N=N-C6H2(p-OH)(m-OCH3)(m-CH=N-C6H4-p-OCH3) (2d) have been tested against Gram-posi-tive, B. subtillis (ATCC 6633). The IC50 i.e. theconcentration of test compound required to inhibitthe growth of bacteria by 50% for 1 (44.3 g/ml)and 2 (2a, 47.8 g/ml; 2b, 36.1 g/ml; 2c, 59.7g/ml; 2d, 49.2 g/ml) have been tested against

Fig. 1. UV-Vis Spectra of 1, 2a-d in MeOH.

Fig. 2. ORTEP view of the crystal structure of 2b (Black C; RedO; Blue N; Yellow S and Pink H) (A, B, C, D are ringabbreviation for brevity).

JICS-18


1394

Fig. 3. Supramolecular construction in 2b derivative by hydrogen bonding (NH...O) and non-classical interaction S-O...CH.

Table 2. Selected bond lengths and bond angles of 2b

SMX-N=N-C6H2(p-OH)(m-OCH3)(m-HC=N-C6H4(p-CH3)) (2b)

Bond length (Å) Bond angle (°)

X-Ray DFT X-Ray DFT

N(8)-N(9) 1.260(3) 1.281 N(8)-N(9)-C(36) 114.6(19) 116.4

C(33)-N(8) 1.419(3) 1.420 C(33)-N(9)-N(8) 113.3(19) 115.1

C(36)-N(9) 1.402(3) 1.413 N(9)-C(36)-C(37) 124.0(2) 124.5

C(26)-N(29) 1.309(3) 1.300 N(9)-C(36)-C(41) 115.7 (2) 115.3

N(10)-C(44) 1.402(3) 1.419 C(44)-N(10)-C(43) 128.6(19) 123.7

C(30)-S(2) 1.751(2) 1.870 N(10)-C(44)-C(45) 115.6(2) 117.5

C(43)-H(43) 0.950(3) 1.091 O(7)-S(2)-C(30) 108.9(10) 108.3

B. subtillis (ATCC 6633) and the results are muchbelow the value of parent SMX (111.9 g/ml).The IC50 of 2a-d show that 2b is more potent thanthat of the others. On the other hand -CHO ismore polar and less space demanding than -CH=N-Ar and may also be easily permeable in the cellmembrane. The inhibition profile is shown in Fig.5. From the result it is evident that the drug mol-

ecules produce a concentration dependent decreasein the growth of both Gram-positive B. subtillisbacteria.

Theoretical interpretation by DFT and docking studies

The best pose docked phase interaction has beenevaluated by docking score, binding energy andlog P data. To account druglikeness property of


1395

Fig. 4. Three diferent fashion of (CH...) interactions in 2b.

Form 1 Form 2 Form 3

Fig. 5. Anti-bacillus activity of SMX, 1, 2a-d on Bacillus subtillis.

the ligands Lipinski’s rule of five34,35, have beenexamined and toxicity has been evaluated byADMET studies. The calculated data show the goodsolubility level, moderate absorption stage, sevenor eight H-bond acceptors, two H-bond donorsand follow Lipinski’s filter. This implies that presentseries of azo-sulfonamids/azo-imino sulfonamidsinteract more strongly than that of SMX to DHPS

protein (Table 3). The hydrophilic and hydropho-bic parts of the derivatives bind through hydro-gen bonding and different electrostatic interac-tions33.

Docking analyses indicate that the C-DOCKERinteraction energy of the drugs with 3TZF followthe ordering 2a > 2c > 2d > 2b > 1. The stepsof best interaction of the drug-to-protein are the


1396

communication to surface interaction which fol-lows the binding to penetration into the cavity orpocket and finally the chemical and physiologicaltransformation through metabolic process34. HenceC-DOCKER interaction energy trend does not re-flect its binding strength; it is the binding energywhich will reflect the strength of binding. The drug2a binds to the protein (Supplementary Materials,Table 1), 3TZF@2a and shows highest stability(–133.53 kcal/mol). The interaction proceeds

mainly through electrostatic path with 3TZF. Thedrug 1 interacts with 3TZF (3TZF@1) and theamino acid residues involve in the process areGln149, Lys221, Ser61, Arg63. The amino acidresidues Gln189, Lys221 for 2a; Lys221, Thr62for 2b; Gly189 Leu194, Gln149 for 2c; andArg255, Lys221, Thr62 for 2d are involved inthe binding cavity of the protein (Table 3). Inter-action analysis suggests that ligand 1 binds withDHPS of E. coli through oxygen atom of -SO2,

Table 3. Details of interactions in most stable protein-ligand complex for 1 and 2a-d

Compd. Hydrogen bonds -bondNo. of End 1 End 2 Bond Angle Bond 19

H-bond distance D-H-A distance

(Å) (º) (Å)

3TZF@1 3 Gln149 (NH) Carbonyl O of 2.13 149 2.66 Thr62 NH to

o-vaniline oxazole nucleus

Ser61 (OH) N of oxazole 2.32 159

Arg63 (NH) N of oxazole 2.96 115

Lys221 (HC) O of SO2 2.70 123

3TZF@2a 1 Gly189 (O=C) HO of 2.21 91 4.58 Pro64 to benzene

o-vaniline ring of o-vaniline

Lys221 (CH) HO of SO2 2.68 118 3.76 Ser61 NH to

oxazole nucleus

3TZF@2b 3 Lys221 (CH) O of oxazole 2.09 145 3.97 Pro64 to benzene

ring of SMX

Thr62 (OH) SO2 of SMX 1.96 151

Thr62 (NH) SO2 of SMX 2.14 150

3TZF@2c 3 Gly189 (O) OH of o-vaniline 2.00 98 4.86 Phe28 nucleus to

moiety o-vaniline ring

Leu194 (NH) O of oxazole 2.84 113 4.40 Pro64 nucleus to

moiety o-vaniline ring

Gln149 (NH) N of oxazole 3.38 144

moiety

3TZF@2d 6 Arg255 (NH) O of SO2 moiety 2.93 140

Arg255 (NH) N of oxazole 2.72 102

moiety

Arg255 (NH) O of oxazole 3.00 90

moiety

Lys221 (NH) O of oxazole 2.11 150

moiety

Thr62 (OH) O of SO2 moiety 2.04 145

Thr62 (NH) O of SO2 moiety 2.04 153


1397

oxazolyl moiety, azo-N and CHO moiety form rec-ognizable H-bonds (Fig. 6). Gln149 NH atom in-teracts with -C=O of o-vaniline moiety: H-O 2.33Å and N-H---O, 149º; Ser61 OH atom interactswith N (oxazole moiety) : O-N, 2.32 Å and O---H-N, 159º; Arg63 NH---N (oxazole) : H---N, 2.96Å; N-H---N, 115º; Lys221(CH---O of (SO2) :H---O, 2.21 Å; C-H---O, 123º and, Thr61 NHnucleus involves --- interaction with oxazole ring

(2.66º) for 1 (Fig. 6) and Gly189 -C=O---H---Oof o-vaniline, 2.21 Å; O-H---C, 91º; Lys221CH interacts with H of SO2 moity of SMX, O---H,2.68 Å; H---O---S 135º and Ser61 CH atom in-teracts with N (oxazole moiety), H---O, 2.38; C-H---N, 164º, Thr62 (OH) interact with SO2 ofSMX moiety, H---O, 1.96 Å; O-H---O, 145º,Ser61 NH nucleus involves --- interaction withoxazole ring (3.76º) for 2a (Fig. 7); Lys221 CH

Fig. 6 . Best fit binding of 1 in the DHPS (PDB id 3TZF) cavity; (a) 2D interaction and (b) 3D interaction.

Fig. 7. Best fit binding of 2a in the DHPS (PDB id 3TZF) cavity; (a) 2D interaction and (b) 3D interaction.


1398

interacts with O of oxazole moiety : H---O, 2.09º;C---H---O 145º, Thr62 OH atom to O of SO2moiety of SMX H---O, 1.96 Å; O-H---O, 151º;Thr62 NH atom to SO2 moiety of SMX : H---O,2.14 Å and N---H---O, 150º; Pro64 NH nucleusinvolves --- interaction with benzene ring ofSMX (3.97º) for 2b (Fig. 8); Gly189 O atom in-teracts with OH of o-vaniline moiety : O---O, 2.00

Å; O---O-H, 89º, Leu194 NH atom interacts withO of oxazole moiety : H---O, 2.84 Å and N-H---O, 113º), Gln149 NH atom interacts with N ofoxazole moiety : H---N, 3.38 Å and N-H---N,144°) and Pro64 NH nucleus involves --- inter-action with benzene ring of o-vaniline (4.40º) for2c (Fig. 9); Arg255 NH atom interacts with O ofSO2 moiety, N of oxazolyl moiety, O of oxazole

Fig. 8. Best fit binding of 2b in the DHPS (PDB id 3TZF) cavity; (a) 2D interaction and (b) 3D interaction.

Fig. 9. Best fit binding of 2c in the DHPS (PDB id 3TZF) cavity; (a) 2D interaction and (b) 3D interaction.


1399

moiety : H---O, 2.93 Å, H---N, 2.72 Å, H---O,3.00 Å, and N-H---O, 140°; N-H---N, 102ºN-H---O, 90°), Lys221 NH atom interact with Oof oxazole moiety : H---O, 2.11 Å and N-H---O,150º), Thr62 OH atom to O of SO2 moiety ofSMX H---O, 2.04 Å; O-H---O, 145º; Thr62 NHatom to SO2 moiety of SMX : H---O, 2.04 Å andN---H---O, 153º; for 2d (Fig. 10).

Theoretical interpretation of electronic spectra

The DFT computation method is used to opti-mize the structure of the ligands. The bond dis-tances and angles have been verified by the com-paring between the DFT optimized and X-ray de-termined structures (Table 2). To simplify analy-sis of DFT computation result the structure of HLhas been divided into azo, benzene sulfonamide(BSN), methyloxazolyl (MOX) and salicylaldehydeand aniline, p-toluidine, p-chloroaniline, p-methoxyaniline (Supplementary Materials, Figs.17–21). Theoretically generated structures of theligands are used to calculate various data and en-ergy of the functions. The electronic properties ofthese molecules have also been verified by DFT

data (Supplementary Materials, Tables 3–12). TheHOMO of 1 is constituted by o-vaniline (79%),azo (7%), benzsulfonamide (13%) and oxazole(1%), where as 2a is constituted by aniline (45%),azo (3%), benzsulfonamide (5%), o-vaniline(47%); 2b is constituted by toluidine (9%), azo(7%), benzsulfonamide (08%), o-vaniline (76%);2c is constituted by p-chloroaniline (4%), azo (7%),benzsulfonamide (9%), o-vaniline (80%); 2d isconstituted by p-methoxyaniline (37%), azo (5%),benzsulfonamide (7%), o-vaniline (51%).

The electronic transitions have been explainedby the population of occupied MOs and unoccu-pied MOs. The intensity of these transitions hasbeen assessed from oscillator strength (f). Theligands show transitions between different molecu-lar fragments, viz. HOMO LUMO, 423.10 nm(f, 0.5044); HOMO-2 LUMO, 336.5 nm (f,0.5961) are considered as admixture of o-vanilineto benzsulfonamide, azo charge transitions, whereasmost of the accepted transitions (SupplementaryMaterials, Tables 3–12) are intra-ligand chargetransfer transitions.

Fig. 10. Best fit binding of 2d in the DHPS (PDB id 3TZF) cavity; (a) 2D interaction and (b) 3D interaction.


1400

Conclusion

One of the wonder discoveries of the 20th cen-tury is Antibiotics. This is true, but the real won-der is the rise of antibiotic resistance in hospitals,communities, and the environment concomitantwith their use. A series of novel azo-imine deriva-tives of sulfamethoxazole has been characterized.The microbial activity against B. subtillis showsthat their efficiency is much better than parentSMX. The molecular docking study computationallyproves the proficiency of these drugs based onfitting in the DHPS cavity of the protein.

Acknowledgements

Financial support is thankfully acknowledgedto the Council of Scientific and Industrial Research,New Delhi (Grant No. 01(2894)/17/EMR-II). Allthe authors are thankful to Professor R. Murugavel,IITB for allowing us to take the facility of singlecrystal X-ray diffraction study established througha DAE-SRC outstanding investigator award.

Supporting Information

Crystallographic data for the structure havebeen deposited to the Cambridge CrystallographicData center, CCDC No. 1589015. These data canbe obtained free of charge Via http://www.ccdc.cam.ac.uk/conts/retrieving.html. orfrom the Cambridge Crystallographic Data Cen-tre, 12 Union Road, Cambridge CB2 lEZ, UK;Fax : (+44) 1223-336-033; or E-mail :[email protected]

References

1. “Discovery and Development of Penicillin”, AmericanChemical Society, 2015.

2. Vanessa M. D’Costa, Christine E. King, Lindsay Kalan,Mariya Morar, Wilson W. L. Sung, Carsten Schwarz, DuaneFroese, Grant Zazula, Fabrice Calmels, Regis Debruyne, G.Brian Golding,Hendrik N. Poinar and Gerard D. Wright,Nature, 2011, 477, 457 .

3. Ten most dangerous antibiotic resistance wave page :longitudeprize.org/blog-post/10.

4. F. Kavanagh, A. Hervey and W. J. Robbins, Proc. Natl.Acad. Sci. USA, 1951, 37, 570.

5. P. J. L. Daniels, D. F. Rane, S. W. McCombe, R. T.Testa, J. J. Wright and T. L. Nagabhushan, ACS Sympo-

sium Series, 1980, 125, 371.

6. W.-T. Ewelina and G. Aukasz, CHEMIK, 2012, 66, 1298.

7. Naredla R. Reddy and D. A. Klumpp, Tetrahedron Lett.,2013, 54, 5945.

8. (a) P. Rajakumar and R. Padmanabhan, Tetrahedron, 2011,67, 9669; (b) P. Przybylski, A. Huczynski, K. Pyta, B.Brzezinski and F. Bartl, Curr. Org. Chem., 2009, 13, 124.

9. M. S. Karthikeyan, D. J. Prasad, B. Poojary, K. S.Bhat, B. S. Holla and N. S. Kumari, Bioorg. Med.Chem., 2006, 14, 7482.

10. A. Echevarria, M. G. Nascimento, V. Geronimo, J.Miller and A. Giesbrecht, J. Braz. Chem. Soc.,1999, 10, 60.

11. L. Shi, H. M. Ge, S. H. Tan, H. Q. Li, Y. C. Songand H. L. Zhu, Eur. J. Med. Chem., 2007, 42,558.

12. S. N. Pandeya, D. Sriram, G. Nath and E. de Clercq.Farmaco, 1999, 54, 624.

13. G. Bringmann, M. Dreyer, J. H. Faber, P. W.Dalsgaard, D. Staerk and J. W. Jaroszewski, J. Nat.Prod., 2004, 67, 743.

14. P. Rathelot, P. Vanelle, M. Gasquet, F. Delmas, M.P. Crozet and P. Timon-David, Eur. J. Med. Chem.,1995, 30, 503.

15. A. Halve and A. Goyal, Orient. J. Chem., 1996, 12,87.

16. P. Phatak, V. S. Jolly and K. P. Sharma, Orient. J.Chem., 2000, 16, 493.

17. X. L. Wang, K. Wan and C. H. Zhou, Eur. J. Med.Chem., 2010, 45, 4631.

18. M. E. Epstein, M. Amodio-Groton and N. S. Sadick.J. Am. Acad. Dermatol., 1937, 37, 365.

19. J. D. Smilack. Mayo Clin. Proc., 1999, 74, 730.

20. G. C. Slatore and A. S. Tilles, Immunol. AllergyClin. North. Am., 2004, 24, 477.

21. G. Choquet-Kastylevsky, T. Vial and J. Descotes,Curr. Allergy Asthma Rep., 2002, 2, 16.

22. S. N. Lavergne, H. Wang, H. E. Callan, B. K. Park,and D. J. Naisbitt, J. Pharmacol. Exp. Ther., 2009,331, 372.

23. S. Mondal, S. M. Mandal, T. K. Mondal and C.Sinha, Spectrochim. Acta, Part A, 2015, 150, 268.

24. S. Mondal, A. K. Bhanja, D. Ojha, T. K. Mondal, D.Chattopadhyay and C. Sinha, RSC Adv., 2015, 5,73626.

25. D. Das, N. Sahu, S. Mondal, S. Roy, P. Dutta, S.Gupta, T. K. Mondal and C. Sinha, Polyhedron,2015, 99, 77.

26. N. Sahu, D. Das, S. Mondal, S. Roy, P. Dutta, N.Sepay, S. Gupta, E. Lopez-Torres and C. Sinha, NewJ. Chem., 2016, 40, 5019.


1401

27. N. Sahu, S. Mondal, N. Sepay, S. Gupta, E. Torres-Lopez, S. Tanaka, T. Akitsu and C. Sinha, J. Biol.Inorg. Chem., 2017, 22, 833.

28. A. Venturini and J. J. J. Gonzalez, Org. Chem.,2002, 67, 9089.

29. A. E. Taggi, A. M. Hafez, H. Wack, B. Young, D.Ferraris and T. J. J. Lectka, J. Am. Chem. Soc.,2002, 124, 6626.

30. A. I. Vogel, A. R. Tatchell, B. S. Furnis and A. J.Hannaford, "Vogel’s Textbook of Practical OrganicChemistry" 5th ed., Longman, 1996.

31. A. Altomare, G. Cascarano, C. Giacovazzo and A.Guagliardi, J. Appl. Crystallogr., 1993, 26, 343.

32. G. M. Sheldrick, Acta Crystallogr., Sect. C :Struct. Chem., 2015, 71, 3.

33. Discovery Studio 3.5 is a product of Accelrys Inc,San Diego, CA, USA.

34. C. A. Lipinski, F. Lombardo, B. W. Dominy and P.J. Feeney, Adv. Drug. Del. Rev., 2001, 46, 3.

35. D. F. Veber, S. R. Johnson, H. Y. Cheng, B. R.Smith, K. W. Ward, K. D. Kopple, J. Med. Chem.,2002, 45, 2615.

36. P. J. Hay, W. R. J. Wadt, J. Chem. Phys., 1985,82, 299.

37. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.Scuseria, M. A. Robb, J. R. Cheeseman, G.Scalmani, V. Barone, B. Mennucci, G. A. Petersson,H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg,M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.Kitao, H. Nakai, T. Vreven, J. A. Montgomery (Jr),J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E.

Brothers, K. N. Kudin, V. N. Staroverov, R.Kobayashi, J. Normand, K. Raghavachari, A.Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M.Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-tin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P.Salvador, J. J. Dannenberg, S. Dapprich, A. D.Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J.Cioslowski, D. J. Fox, GAUSSIAN 09, RevisionD.01, Gaussian Inc., Wallingford, CT, 2009.

38. A. Frisch, A. B. Nielson and A. J. Holder,GAUSSVIEW User Manual, Gaussian Inc, Pittsburgh,PA, 2000.

39. A.-N. M. A. Alaghaz, H. A. Bayoumi, Y. A. Ammarand S. A. Aldhlmani, J. Mol. Struct., 2013, 1035,383.

40. P. K. Ojha and K. Roy, Eur. J. Med. Chem., 2010,45, 4645.

41. S. Tabassum, M. Afzal and F. Arjmand, Eur. J.Med. Chem., 2014, 74, 694.

42. D. J. Cotton, V. J. Gill, D. J. Marshall, J. Gress, M.Thaler and P. A. Pizzo, J. Clin. Microb., 1987, 25,672.

43. M. E. Casida, C. Jamorski, K. C. Casida and D. R.Salahub, J. Chem. Phys., 1998, 108, 4439.

44. V. Barone and M. Cossi, J. Phys. Chem. (A), 1998,102, 1995.

45. M. Cossi and V. Barone, J. Chem. Phys., 2001,115, 4708.

46. M. Cossi, N. Rega, G. Scalmani and V. Barone, J.Comput. Chem., 2003, 24, 669.


1402

the michaelis-menten kinetics and errors in estimated reaction constants … · 2018-09-04 ·...

Documents