Efib

NPa

b

c

d

Ae

f

g

a

ARRAA

KMCNLD

(T

(

0h

Journal of Chromatography A, 1286 (2013) 216– 221

Contents lists available at SciVerse ScienceDirect

Journal of Chromatography A

j our na l ho me p ag e: www.elsev ier .com/ locate /chroma

xploring chip-capillary electrophoresis-laser-induced fluorescenceeld-deployable platform flexibility: Separations of fluorescent dyesy chip-based non-aqueous capillary electrophoresis

antana Nuchtavorna,b, Petr Smejkalb,c,d, Michael C. Breadmored, Rosanne M. Guijte,hilip Doble f, Fritz Bekg, Frantisek Foretc, Leena Suntornsuka, Mirek Mackab,d,∗

Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Mahidol University, 447 Sri-Ayudhaya Rd., Rajathevee, Bangkok 10400, ThailandIrish Separation Science Cluster and School of Chemical Sciences, Dublin City University, Dublin 9, IrelandInstitute of Analytical Chemistry of the ASCR, v.v.i., Veverí 97, 60200 Brno, Czech RepublicAustralian Centre for Research on Separation Science (ACROSS) and School of Chemistry, University of Tasmania, Private Bag 75, Hobart, TAS 7001,ustraliaSchool of Pharmacy, University of Tasmania, Private Bag 26, Hobart, TAS 7001, AustraliaSchool of Chemistry and Forensic Science, University of Technology Sydney, PO Box 123, Broadway, NSW 2007, AustraliaAgilent Technologies, P.O. Box 1280, 76337 Waldbronn, Germany

r t i c l e i n f o

rticle history:eceived 30 August 2012eceived in revised form 13 February 2013ccepted 14 February 2013vailable online 26 February 2013

eywords:icrofluidic chip CE

apillary electrophoresisACEIF detectionyes, Agilent Bioanalyzer

a b s t r a c t

Microfluidic chip electrophoresis (chip-CE) is a separation method that is compatible with portable andon-site analysis, however, only few commercial chip-CE systems with laser-induced fluorescence (LIF)and light emitting diode (LED) fluorescence detection are available. They are established for several appli-cation tailored methods limited to specific biopolymers (DNA, RNA and proteins), and correspondinglythe range of their applications has been limited. In this work we address the lack of commercially avail-able research-type flexible chip-CE platforms by exploring the limits of using an application-tailoredsystem equipped with chips and methods designed for DNA separations as a generic chip-CE platform –this is a very significant issue that has not been widely studied. In the investigated Agilent Bioanalyzerchip-CE system, the fixed components are the Agilent chips and the detection (LIF at 635 nm and LEDIF at470 nm), while the chemistry (electrolyte) and the programming of all the high voltages are flexible. Usingstandard DNA chips, we show that a generic CE function of the system is easily possible and we demon-strate an extension of the applicability to non-aqueous CE (NACE). We studied the chip compatibility withorganic solvents (i.e. MeOH, ACN, DMF and DMSO) and demonstrated the chip compatibility with DMSOas a non-volatile and non-hazardous solvent with satisfactory stability of migration times over 50 h. Thegeneric CE capability is illustrated with separations of fluorescent basic blue dyes methylene blue (MB),toluidine blue (TB), nile blue (NB) and brilliant cresyl blue (BC). Further, the effects of the composition of

the background electrolyte (BGE) on the separation were studied, including the contents of water (0–30%)and buffer composition. In background electrolytes containing typically 80 mmol/L ammonium acetateand 870 mmol/L acetic acid in 100% DMSO baseline separation of the dyes were achieved in 40 s. Linearitywas documented in the range of 5–28 �mol/L, 10–100 �mol/L, 1.56–50 nmol/L and 5–75 nmol/L (r2 val-ues in the range 0.974–0.999), and limit of detection (LOD) values were 90 nmol/L, 1 �mol/L 1.4 nmol/L,and 2 nmol/L for MB, TB, NB and BC, respectively.

∗ Corresponding author at: Australian Centre for Research on Separation ScienceACROSS) and School of Chemistry, University of Tasmania, Private Bag 75, Hobart,AS 7001, Australia. Tel.: +61 3 62266670; fax: +61 3 62262858.

E-mail addresses: mirek.macka@utas.edu.au, mirek.macka@gmail.comM. Macka).

021-9673/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.chroma.2013.02.060

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

The emergence of microfluidics and lab-on-a-chip technologyhas generated much interest in analytical chemistry over recentyears. Moving from CE to the microfluidic platform as microflu-

idic chip electrophoresis (chip-CE) brought the advantages of fasterseparations that could be achieved in tens of seconds rather thanin minutes to tens of minutes as in conventional CE thus offeringa higher throughput and better compatibility with portable and

mato

ot

tsccrmafpt(MfbiCdiia2rlMoeplflDlsbi[

tcdt[tacaptab

2

2

FAb(fU

by separations of NB in BGE containing 26 mmol/L ammoniumacetate and 870 mmol/L acetic acid in DMSO [28] using injectionand separation voltage of 1400 V with the chip thermostated at25 ◦C.

1 The Script Mode is not a standard feature of the Bioanalyzer but is provided

N. Nuchtavorn et al. / J. Chro

n-site analysis. There are now a number of examples illustratinghe benefits in moving to the microchip platform [1–6].

To conduct research in microfluidics, most researchers fabricateheir own microchips and piece together instrumentation to make aeparation and detection system [5–14]. While most of the researcharried out around the world uses research-lab chips or equipment,ommercial suppliers present an alternative route to mainstreamesearch. There are now a number of commercial avenues to sourceicrochips, with both standard and customized designs available,

s well as those that also provide the necessary equipment to per-orm simple electrophoretic separations. Many of these microchiproducts are aimed specifically at the research market, such ashose produced by Micralyne (Edmonton, Canada [15]), MicronitEnschede, Holland [16]), Dolomite (Royston, UK [17]), and the

icrofluidic Chip Shop (Jena, Germany [18]). The chips used varyrom relatively simple designs such as the majority of those offeredy the companies listed above, to more complex task-specific chips

n end-user oriented products from Agilent, Biorad, Shimadzu andaliper Lifesciences, with the last featuring a sipper attachment toirectly sample from 96 and 384 well plates. While the vast major-

ty of chip-CE research has been conducted in research laboratorynstrumentation, only few commercial chip-CE systems with LIFnd LED fluorescence detection are available: Agilent Bioanalyzer100 launched in 1999 (Waldbronn, Germany [19]), Biorad Expe-ion launched in late 2004 (Hercules, CA, USA [20]), Caliper Labchipaunched in early 2004 (Hopkinton, MA, USA [21]) and Shimadzu

CE®-202 MultiNA launched in 2007 (Kyoto, Japan [22]). Theyffer end-user-oriented complete analytical solutions and are wellstablished for several application tailored methods (DNA, RNA androtein analysis), therefore the range of their applications has been

imited [23,24]. While research-grade instruments offer flexibilityor the user, the end-user-oriented applied analysis chip-CE ana-yzers are available with kits including task-specific chips (such asNA, RNA, protein analysis) and fixed ready-to-use methods with

ittle method development flexibility, thus making it appear pos-ible to use them as platforms for generic chip-CE analysis of aroader range of different analytes. This will open up opportunities

n numerous areas including point-of-care and portable analysis25,26], forensic analysis [27] and many others.

All the analytical systems presented so far for this platform inhe applications of commercial methods [28,29] and as a generichip platform in a preliminary investigation into profiling APTSerivatized glycans [30] and recent works showing MEKC separa-ions of FITC derivatized amphetamines for analysis of street drugs31], separations of stained microorganism cells [32] and for iso-achophoresis using in-house made chips [33,34]. In this work ourim was to explore the principal issue of suitability of a commer-ial end-user-oriented chip-CE, Agilent Bioanalyzer provided withccess to script editor in its research mode, as a generic chip-CElatform and to evaluate its performance in combination with theask-specific DNA chips for separations by non-aqueous CE (NACE),nd we demonstrate this with separations of four fluorescent basiclue dyes.

. Experimental

.1. Chemicals

The fluorescent basic blue dyes used in this work are shown inig. 1. Methylene blue (MB) and nile blue (NB) were purchased fromldrich (Milwaukee, WI, USA), toluidine blue (TB), brilliant cresyl

lue (BC) and lithium chloride (LiCl) were purchased from FlukaBuchs, Switzerland). Dimethyl sulfoxide (DMSO) and dimethyl-ormamide (DMF) were purchased from Sigma (St. Louis, MO,SA). Methanol (MeOH), acetonitrile (ACN), glacial acetic acid,

gr. A 1286 (2013) 216– 221 217

ammonium acetate (NH4OAc), sodium acetate (NaOAc) and cal-cium nitrate (Ca(NO3)2) were purchased from Sigma–Aldrich(Dublin, Ireland). Water was treated with a Millipore (Bedford,MA, USA) Milli-Q water purification system. Electrolytes were pre-pared by dissolving NH4OAc and metal salts (i.e., NaOAc, LiCl andCa(NO3)2) in glacial acetic acid, followed by dilution with theappropriate solvent. Stock solutions of dyes were prepared at a con-centration of 1 mmol/L in acetonitrile and diluted as necessary inthe appropriate solvent.

2.2. Instrumentation

Chip-CE separations were performed on the Agilent 2100 Bio-analyzer (Agilent Technologies, Waldbronn, Germany) using DNAchips as originally supplied, with detailed descriptions of the DNAchips and Bioanalyzer provided by the manufacturer and quotedin the literature [19,29,28]. The design for the DNA chip used inthis study is shown in Fig. 2. In brief, each chip is microfabri-cated from glass and encased in black PMMA polymer encasementcontaining 16 wells: 3 for loading of the background electrolyte(A4, B4, C4), and 13 for samples, with chip wells in Fig. 2 thencorresponding to A1–D4. The chip architectures consist of inter-connected microchannels, in which separations can be performedthrough the microchannel from the injection cross on the channelleading from A4 to the detection point before the C4 well. The move-ment of fluids (electrolyte and analytes) through the microchannelsis affected by 16 platinum electrodes reaching from above the chipinto the plastic well reservoirs, which create electrokinetic forces(primarily electrophoresis) moving the analytes. Programmed reg-ulation of voltage on each of the 16 electrodes controls the speedand direction of the analyte zone movements. The microchip fea-tures a separation channel with a total length of approximately40 mm, with 14 mm from injection to the detector. Microchipswere filled and flushed with buffer by manually pipetting into thebuffer reservoir and applying vacuum to the buffer waste reser-voir using a home-made device consisting of a 3 mL plastic syringeand apiece of pipette tip that fitted tightly into the reservoirs onthe top of the chip. For chips used with DMSO electrolytes, beforeuse each chip was flushed with DMSO and left overnight to condi-tion. Microchips were typically flushed with electrolyte for 5 minand the sample reservoirs were filled with 7 �L of sample, while thebuffer reservoirs filled were with 10 �L of buffer. Programmed highvoltage control for the 16 independent platinum electrodes of themicrochip was achieved by developing a new script via the assayand script developer mode.1 Filling the injection intersection wasachieved by application of 100 V to the sample waste and 1500 Vto the sample. Separation was performed by application of 100 Vto the buffer waste and 1600 V to the buffer. Detection was via thein-built LIF detector.

2.3. Methods

Applicability of the DNA chip as a generic CE chip was performed

by Agilent Technologies to prospective users for exploring new applications inMicrofluidic Capillary Electrophoresis under a collaboration agreement. Methodsdeveloped in the script mode are compliant with the Bioanalyzer and can be usedwithout Script Mode access. Agilent Technologies does not accept liability for suchmethods.

218 N. Nuchtavorn et al. / J. Chromatogr. A 1286 (2013) 216– 221

he inv

ioyp

Bt2assr

3

3

aaci

Fow

Fig. 1. Structures of t

The compatibility with organic solvents was examined by fill-ng the acrylate polymer plastic reservoirs of the chip with variousrganic solvents (i.e. MeOH, ACN, DMF and DMSO). Then, the anal-sis repeatability was performed for three different chips over aeriod of 50 h.

NACE conditions were optimized by evaluating effects ofGE solvents (i.e. MeOH, ACN and DMSO), types and concentra-ion of BGE (i.e. 26–90 mmol/L NH4OAc, 26–70 mmol/L NaOAc,6–70 mmol/L LiCl and 15–40 mmol/L Ca(NO3)2 with 870 mmol/Lcetic acid in DMSO) and water contents in BGE (10–30%). Theamples were electrokinetically injected at 1400 V for 50 s. Theeparating voltage and temperature were set at 1500 V and 25 ◦C,espectively.

. Results and discussion

.1. Initial considerations

The principal question at the outset of this work was how can

commercial end-user-oriented chip-CE, in this case Agilent Bio-nalyzer, used with task-specific DNA chips, be used as a generichip-CE platform when provided with access to the script editor ints research mode, and what will be its performance and limitations.

ig. 2. Photographs of the DNA chips used in this work. The photograph of the glass chip (rf the chip as it would appear if viewed looking through the plastic cover from the top.

ells A4 to D4. A4 is the inlet buffer well, B4 and D4 are the sample waste, C4 is the buffe

estigated basic dyes.

Applicability of the chip-CE to solutions of real-world analyticalproblems has to be considered as well. With the short separationchannel length of 14 mm from injection to the detection point, thestrength of the chip-CE analytical system compared to a standard CEis the short analysis time, as shown for instance for APTS deriva-tized glycans [30] comparing separations between CE an chip-CEfor an identical sample and electrolyte, with CE in a 0.5 m capil-lary clearly offering a higher efficiency, but at an expense of over10× longer separation times. Another important consideration isthat fluorescence detection is the only available method of detec-tion, with the bioanalyzer featuring simultaneously operating LIFdetection with a 635 nm red laser diode (10 mW optical power,emission collected above 685 nm), and a LEDIF detection with a475 nm blue LED (2 mW optical power, emission collected above525 nm). Therefore to utilize direct fluorescence detection, any ana-lytes has to be fluorescent or be labeled with suitable fluorescenttags, such as fluorescein or cyanine dyes [28] for the detection.For the majority of this work a group of fluorescent blue dyeswas selected matching the red LIF detection while the applica-

bility of the blue LED detection has been previously documentedin a preliminary investigation in aqueous solutions for APTS-tagged oligosaccharides [30] and of FITC derivatized amphetamines[31].

ight) has been transposed horizontally to resemble the microchannel design layoutWells A1–3, B1–3, C1–3 and D1–3 are the sample reservoirs. Buffer was placed inr waste.

N. Nuchtavorn et al. / J. Chromatogr. A 1286 (2013) 216– 221 219

125

130

135

140

145

150

0 10 20 30 40 50 60

Mig

ratio

n tim

e o

f N

B [

s]

Time [h ]

Chip 1

Chip 2

Chip 3

Fig. 3. Long-term stability of migration times of NB using three individual chips.Electrolyte: 26 mmol/L ammonium acetate and 870 mmol/L acetic acid in 100%Dt

3

ctosodfe(a

v(lDsttDtmNs

3

ssudatfatmsuer

45

65

85

105

125

145

0 20 40 60 80 100

Flu

ore

scence

Time [s]

MB, TB

NB

BC

MB, TB

NBBC

MB

NBBC

100% DMSO

60% MeO H

60% ACN

TB

Fig. 4. Separation of basic blue dyes using microchip CE. Electrolytes: 80 mmol/Lammonium acetate and 870 mmol/L acetic acid in 100% DMSO, 60% MeOHand 60% ACN. Conditions: separation length 14 mm, injection/separation voltage

MSO. Conditions: separation length 14 mm, injection/separation voltage 1400 V,emperature at 25 ◦C and LIF detection with 635 nm excitation and 670 nm emission.

.2. Chip-solvent compatibility

When considered as a generic chip-CE separation platform,ompatibility with organic solvents such as for NACE is impor-ant and therefore was investigated here. Initially, the compatibilityf the acrylate polymer encasement part of the microchips withelected organic solvents was visually observed. The plastic partf the chip was visibly deformed in DMF, while MeOH and ACNid not attack it visibly but evaporated from the reservoir within aew minutes thus making it impossible to run an analysis. Also sev-ral other commonly available solvents including higher alcoholsincluding ethanol, propanol, butanol) showed one or both of thebove two problems.

Somewhat surprisingly when exposed to DMSO, although theisual appearance of the acrylate polymer chip reservoir surfaceoriginally black) appeared slightly white, when examining the ana-ytical performance DMSO proved to be very compatible with theNA chip as whole and the system. Experiments examining analy-

is repeatability with three different chips showed that after 50 h,he migration time of NB remained unchanged within experimen-al error (%RSD = 1.4–1.9%, Fig. 3) and no extra peaks were observed.MSO as a non-toxic, non-hazardous, non-volatile solvent seems

o present an option for chip-CE using the existing DNA chips inixed aqueous-organic electrolytes and for NACE, and its use inACE has been shown [35,36]. Therefore, DMSO was selected as

olvent for NACE for the demonstrated separation of basic dyes.

.3. BGE solvent

DMSO dissolves a wide range of organic and inorganic sub-tances and is miscible with water and most common organicolvents. Its high boiling point and low vapor pressure enable these of small volumes without evaporation. It offers considerablyifferent selectivity to other solvents due, particular for weak acidsnd bases. pKa* values of weak acids (HA) are higher in DMSO, withhe difference in pKa* values between DMSO and water rangingrom 10.5 to 21.5 units. The difference is less pronounced for cationcids (HB+) [37,38]. Further the effect of non-aqueous media onhe detection because intensified fluorescence has been utilized in

any NACE studies. Increased viscosity, decreased polarity, bettertability of compounds, and lowered quenching due to the molec-

lar oxygen present in organic solvents are the reasons for thenhancement of the fluorescence intensity in non-aqueous mediaelative to that in water [37].

1400/1500 V, temperature at 25 ◦C and LIF detection with 635 nm excitation and670 nm emission. Other conditions are described in the text.

In our previous work, conventional CE using 26 mmol/L NH4OAcand 870 mmol/L acetic acid in various organic solvent (e.g. MeOH,ACN, DMF and DMSO) was examined for the separation of 4dyes [39]. Transferring the method to microchip platform pro-vided mixed results. In DMSO, excellent separations of the dyeswere obtained. Four peaks could be achieved within 40 s withRs = 1.05–1.50. No peaks were observed when using MeOH andACN as the wells became devoid of liquid after several minutes,which was attributed to the low volumes (ca. 7 �L) and high volatil-ity of these solvents. This problem was not observed with DMSObecause of its much lower volatility. To circumvent this issue, waterwas added to the electrolyte to decrease its overall volatility. CEexperiments with MeOH and ACN electrolytes containing up to 40%water showed that the efficiency of the separation of the dyes didnot decrease, indicating that this amount of water was sufficientto completely solubilize the dye and prevent hydrophobic inter-action with the capillary wall. However, electrolytes containinghigher amounts of water resulted in significant peak tailing. Trans-ferring the separations from the capillary system to the microchipwas relatively straightforward, although some optimization of theinjection conditions was required. However, the injection volumewas primarily defined by the microchip channel geometry, par-ticularly the width of the injection channel at the intersection. Incomparison to the conventional CE, the separation is much faster inthe microchips, while the resolution slightly decreased due to theconsiderably shorter separation channel. Further experiments onthe uses of MeOH and ACN as the solvent were performed in elec-trolytes containing 40% (v/v) water and 60% (v/v) solvent (Fig. 4).It can be seen that separations in aqueous ACN were the fastest(tm < 30 s), while those in MeOH and DMSO required more time(tm < 40 s). However, the separation in DMSO was superior, possi-bly due to the longer migration times [40,41]. A sufficient resolutionfor the 4 dyes (Rs = 1.05–1.50) could only be achieved using DMSO,whereas separations in MeOH and ACN resulted in co-migratingpeaks (Fig. 4).

The influence of water contents in DMSO was also investigated.Water might be present as water of hydration or as physicallyadsorbed water in the electrolytes added to organic solvents. More-over, organic solvents might absorb water from the environment[40]. Water content in the BGE can influence CE separation of thedyes. The results showed that additions of water (10–30%v/v) into80 mmol/L NH OAc and 870 mmol/L acetic acid in DMSO decreased

4both the resolution and signal intensity of the dyes (Fig. 5), changesthat are well documented in NACE [38]. Increasing the amount ofwater indicated that the dyes were not solubilized completely and

220 N. Nuchtavorn et al. / J. Chromatogr. A 1286 (2013) 216– 221

45

65

85

105

125

145

0 20 40 60 80 100

Time [s ]

Flu

ore

scence

BC

MB

NB

TB

MB

NB

TBBC

TBMB

NB

BC

MBNB

TB BC

100% DMSO

90% DMSO (10% H2O)

80% DMSO (20% H2O)

70% DMSO (30% H2O)

Fig. 5. Separation of basic blue dyes in electrolytes with varying water contents.EDO

tcci

3

oaoEab

colaiaN8o2i

Fba8F

Table 1Linearity LOD and LOQ data.

Dyea Linearityb LODc (nmol/L) LOQc (nmol/L)

MB y = 142.06x + 530.71, r2 = 0.9739 90 3.0 × 102

TB y = 63.856x − 183.8, r2 = 0.9989 1.0 × 103 3.2 × 103

NB y = 22.798x + 48.894, r2 = 0.9988 1.4 4.5BC y = 1.4022x + 1.8989, r2 = 0.9769 2.0 6.2

a MB = methylene blue; TB = toluidine blue; NB = nile blue; BC = brilliant cresolblue.

b Linearity: 5 values of concentration were used, each point measured in triplicate.c LOD = limit of detection; LOQ = limit of quantitation.

Table 2Precision of analysis for MB, TB, NB and BC (%RSD, n = 10).

Dye Intra-day Inter-day Injection

tm Area tm Area tm Area

MB 1.7 1.5 1.8 2.2 2.5 1.3TB 2.1 2.2 2.1 2.1 2.6 1.9

lectrolytes: 80 mmol/L ammonium acetate and 870 mmol/L acetic acid in 70%MSO (30% H2O), 80% DMSO (20% H2O), 90% DMSO (10% H2O) and 100% DMSO.ther conditions are the same as described in Fig. 4.

heir hydrophobicities caused the hydrophobic interaction with thehip surface. This leads to surface adsorption effects, which in turnontribute to significantly broad peaks and decreased reproducibil-ty due to disruption of the electro-osmotic flow (EOF) [42].

.4. Type and concentration of BGE buffer

The composition of the electrolyte has a very significant impactn the separation. Both the ionic strength and pH can influence thenalysis time and the separation selectivity due to either acid/baser ion-association equilibria, as well as through influences on theOF [39]. In NACE the most common buffering systems consist ofcids and their ammonium salts, of which Ac and NH4OAc haveeen the most frequent.

Since non-aqueous solvents were employed and pH is diffi-ult to measure accurately in many in these solvents, the conceptf “relative acidity/basicity” was adopted in which various alka-ine metal salts were added to the electrolyte containing constantmounts of acetic acid to obtain the difference of acidity or basic-ty. The separations of dyes were investigated in NH4OAc, mono-nd divalent metal salts (i.e. 26–90 mmol/L NH4OAc, 26–70 mmol/LaOAc, 26–70 mmol/L LiCl and 15–40 mmol/L Ca(NO3)2) with

70 mmol/L acetic acid in DMSO (Fig. 6). Baseline separationsf the dyes were not observed in the initial condition using6 mmol/L NH4OAc as the electrolyte. The separation efficiency was

mproved with the higher amounts of NH4OAc and the optimum

45

65

85

105

125

145

165

185

0 20 40 60 80 10 0

26 mM calcium nitra te

60 mM lithium chloride

60 mM sodium acetate

80 mM am monium acetate

Time [s ]

Flu

ore

scence

BC

MB

TB NB

BC

MB

NB

TB

BCMB NB

TB

BCMB

NB

TB

ig. 6. Comparison of the separation of basic blue dyes using different electrolyteuffer cation. Electrolytes: (a) 80 mmol/L ammonium acetate, (b) 60 mmol/L sodiumcetate, (c) 60 mmol/L lithium chloride, (d) 26 mmol/L calcium nitrate containing70 mmol/L acetic acid in DMSO. Other conditions are the same as described inig. 4.

NB 1.6 1.6 2.0 2.0 2.4 1.6BC 1.7 2.3 1.7 2.7 2.2 2.3

condition was achieved at 80 mmol/L NH4OAc. At low concentra-tions (26–50 mmol/L) of NaOAc, overlapped peaks of TB and NBwere obtained however, baseline resolution could be achieved at60–70 mmol/L. A broad split peak for NB was observed in 70 mmol/LNaOAc. Another monovalent metal salt, LiCl, could not provide thebaseline separation of TB and NB but offered a sufficient resolutionof MB/TB and NB/BC when used at a concentration of 60 mmol/L.At 60–70 mmol/L LiCl electrolytes, split peaks of BC were observed.We reasoned that additional peaks of either NB in NaOAc or BC inLiCl was a structurally related isomer formed during the synthe-sis of the dye, which remained after subsequent purification steps[39] and that they could be detected as split peaks at high pH due tochanges in ionization as a result of their slightly different pKa. Forthe divalent metal salts, the dyes showed multiple peaks with nobaseline separation in 15 mmol/L Ca(NO3)2. Although the baselineseparation was improved at higher concentration of Ca(NO3)2, TBand NB were still not completely separated. It was concluded thatNH4OAc electrolyte showed the best separation of the dyes in termof the peak shapes, resolutions and intensities.

3.5. Method performance characteristics

Calibration curves of MB, TB, NB, and BC were estab-lished in ranges of 5–28 �mol/L, 10–100 �mol/L, 1.56–50 nmol/Land 5–75 nmol/L, respectively. Regression data calculated frompeak area provided correlation coefficients (r2) in ranges of0.9739–0.9988 (Table 1). The precision values (%RSDs, n = 10) forboth the migration times and Rs and the peak areas (Table 2) ofall the analytes fell within the range of 1.5% and 2.7%. The LODsand LOQs are shown in Table 1, with low-nmol/L LOD values forNB and BC. Analytical performance characteristics document thatthe method is applicable for the quantitative analysis of the dyesby NACE.

4. Conclusions

We demonstrated that a commercial task-specific chip-CE plat-form with application-specific chips, such as in this case AgilentBioanalyzer and DNA-chips, can be successfully used as genericchip-CE system, when the research mode using script editor is used.

Moreover, it was shown that the applicability of the platform couldbe extended into the area of NACE. DMSO has been found compat-ible with the DNA chips and could be used over a period of severaldays. This presents an opportunity to use DMSO as a non-toxic,

mato

nm

rbttCodsNbTasp

srpd

A

tPta(vwfpaoa

R

[

[

[[

[[[[[[

[[

[

[[

[[[[

[

[

[[

[

[

[[

[[

[39] A.R. Fakhari, M.C. Breadmore, M. Macka, P.R. Haddad, Anal. Chim. Acta 580

N. Nuchtavorn et al. / J. Chro

on-hazardous, non-volatile solvent as an option for chip-CE inixed aqueous-organic electrolytes and for NACE.A simple and rapid NACE separation of several structurally

elated basic dyes could be achieved in NH4OAc and acetic aciduffer in DMSO with excellent separation in less than 40 s. Whilehe resolution among the single separated analytes is lower inhe 14 mm length separation channel of the chip compared to theE method [28], the separation times are generally shorter by anrder of magnitude. Detection using a red LIF facilitated sensitiveetection of these dyes, especially NB, which was the most sen-itive dye with LOD and LOQ at the low nmol/L level. ThereforeB could be a potential fluorescent derivatization reagent for car-oxylic acid groups containing bioanalytes and possibly for cells.he developed method on the commercially available microchip CEnd DNA chips show promise for this and other similar type of task-pecific instruments for the other applications as general chip-CElatforms.

Most importantly, the demonstrated possibility of opening apecific task-dedicated chip-CE platform for generic chip-CE sepa-ations is potentially a widely available commercial instrumentallatform for research and even more for point-of-care and fieldeployable analysis.

cknowledgments

Financial support from the Thailand Research Fundhrough the Royal Golden Jubilee Ph.D. Program (Grant No.HD/0144/2548) to Nantana Nuchtavorn and Leena Sun-ornsuk is gratefully acknowledged. M.M. would like tocknowledge the Marie Curie Excellence Grants and FundingMEXT-CT-2004-014361, 2006–2010) and from 2010 the Uni-ersity of Tasmania the New Stars start-up funding. M.C.B.ould like to acknowledge the Australian Research Council

or a QEII fellowship (DP0984745). F.F. acknowledges sup-ort from GACR P20612G014. The support of this projectnd specifically the technical support for the script devel-per research mode by Agilent Technologies is gratefullycknowledged.

eferences

[1] T. Vilkner, D. Janasek, A. Manz, Anal. Chem. 76 (2004) 3373.

[2] P.S. Dittrich, K. Tachikawa, A. Manz, Anal. Chem. 78 (2006) 3887.[3] D.R. Reyes, D. Iossifidis, P.-A. Auroux, A. Manz, Anal. Chem. 74 (2002) 2623.[4] P.-A. Auroux, D. Iossifidis, D.R. Reyes, A. Manz, Anal. Chem. 74 (2002) 2637.[5] J.P. Landers, Handbook of Capillary and Microchip Electrophoresis and Associ-

ated Microtechniques, first ed., CRC Press, Florida, 2008.

[[[

gr. A 1286 (2013) 216– 221 221

[6] A. Arora, G. Simone, G.B. Salieb-Beugelaar, J.T. Kim, A. Manz, Anal. Chem. 82(2010) 4830.

[7] D.C. Duffy, J.C. McDonald, O.J.A. Schueller, G.M. Whitesides, Anal. Chem. 70(1998) 4974.

[8] J.C. McDonald, D.C. Duffy, J.R. Anderson, D.T. Chiu, H. Wu, O.J.A. Schueller, G.M.Whitesides, Electrophoresis 21 (2000) 27.

[9] K. Seiler, D.J. Harrison, A. Manz, Anal. Chem. 65 (1993) 1481.10] D.J. Harrison, K. Fluri, K. Seiler, Z. Fan, C.S. Effenhauser, A. Manz, Science 261

(1993) 895.11] D.J. Harrison, A. Manz, Z. Fan, H. Luedi, H.M. Widmer, Anal. Chem. 64 (1992)

1926.12] C.S. Effenhauser, A. Manz, H.M. Widmer, Anal. Chem. 65 (1993) 2637.13] S. Van Overmeire, H. Ottevaere, G. Desmet, H. Thienpont, IEEE J. Sel Topics

Quantum Electron 14 (2008) 140.14] M. Yu, H.-Y. Wang, A.T. Woolley, Electrophoresis 30 (2009) 4230.15] http://www.micralyne.com, viewed on 1 December 2012.16] http://www.micronit.com, viewed on 1 December 2012.17] http://www.dolomite-microfluidics.com, viewed on 1 December 2012.18] http://www.microfluidicchipshop.com, viewed on 1 December 2012.19] http://www.chem.agilent.com/en-US/Products-Services/Instruments-

Systems/Automated-Electrophoresis/Pages/default.aspx,http://www.genomics.agilent.com/CollectionOverview.aspx?PageType=Application&SubPageType=ApplicationOverview&PageID=275, viewed on 1December 2012.

20] http://www.biorad.com, viewed on 1 December 2012.21] http://www.caliperls.com/products/labchip-systems/, viewed on 1 December

2012.22] http://www.shimadzu-biotech.net/pages/products/2/multina.php, viewed on

1 December 2012.23] P. Smejkal, F. Foret, Chem. Listy 106 (2012) 104.24] S. Fanali, P. Haddad, D. Lloyd, C.F. Poole, P.J. Schoenmakers, Handbooks in Sepa-

ration Science: Liquid Chromatography, first ed., Elsevier, Massachusetts, 2012.25] S. Choi, M. Goryll, L. Sin, P. Wong, J. Chae, Microfluid. Nanofluid. 10 (2011) 231.26] C. Rivet, H. Lee, A. Hirsch, S. Hamilton, H. Lu, Chem. Eng. Sci. 66 (2011) 1490.27] J.P. Pascali, F. Bortolotti, F. Tagliaro, Electrophoresis 33 (2012) 117.28] L. Bousse, S. Mouradian, A. Minalla, H. Yee, K. Williams, R. Dubrow, Anal. Chem.

73 (2001) 1207.29] A.W. Chow, in: C.S. Henry (Ed.), Microchip Capillary Electrophoresis: Methods

and Protocols, Humana Press, New Jersey, 2006, p. 129.30] P. Smejkal, Á. Szekrényes, M. Ryvolová, F. Foret, A. Guttman, F. Bek, M. Macka,

Electrophoresis 31 (2010) 3783.31] A. Lloyd, L. Blanes, A. Beavis, C. Roux, P. Doble, Anal. Method 3 (2011) 1535.32] N. Nuchtavorn, F. Bek, M. Macka, W. Suntornsuk, L. Suntornsuk, Electrophoresis

33 (2012) 1421.33] P. Smejkal, M.C. Breadmore, R.M. Guijt, F. Foret, F. Bek, M. Macka, Electrophore-

sis 33 (2012) 3166.34] P. Smejkal, M.C. Breadmore, R.M. Guijt, F. Foret, F. Bek, M. Macka, Anal. Chim.

Acta 755 (2012) 115.35] J. Wang, M. Pumera, Anal. Chem. 75 (2002) 341.36] S.J.O. Varjo, M. Ludwig, D. Belder, M.-L. Riekkola, Electrophoresis 25 (2004)

1901.37] M.-L. Riekkola, Electrophoresis 23 (2002) 3865.38] M.-L. Riekkola, M. Jussila, S.P. Porras, I.E. Valkó, J. Chromatogr. A 892 (2000)

155.

(2006) 188.40] J. Tjørnelund, S. Hansen, Chromatographia 44 (1997) 5.41] L. Geiser, J.-L. Veuthey, Electrophoresis 28 (2007) 45.42] C. Vogt, J. Vogt, A. Becker, E. Rohde, J. Chromatogr. A 781 (1997) 391.