AD-AI02 112 MASSACHUSETTS INST OF TECH CAMBRIDGE DEPT OF CHEMISTRY F/6 7/3PHO0TOCHEMISTRY OF SURFACE-CONFINED ORGANOMETALLICS: PHOTOCHEMIC--EC (U)

JUN 81 C L REICHEL, M S WRIGHTON NOOO475C-080

UNCLASSIFIED TR-28 NL

SE U IY LA i l_-T10 -F -. I PAG S.he ToPs Fod ILT& EI6 OS~

S DEFO REPGORNZTIOCU4MEANTATIOES PAGE BEORE EL6'EME '; Ft AS

eprtent2 ofV Cheisry Rrn. 6-335 AREA&*ORINM8ES

j I CI RLL IN OF I E NA E A O AD RES 1 . T P fREPO RT ATEI 6 OV A

Office of Naval ResearchaJune 8, 198Department PotohemiNavy Rela. ofSE aF SufAe-n-ESAingdCoto Virna 22217st 48FRIdOFG E

7 . I..C~ * 5 EC A IF TON N GP RAO!NCrlL Ri-1 n Mr S.. Wrigton N EDULE5C-8

S5 DISRRIB IORNI AT IO AEADADESt.PORMEEMENT rot;E= 'A5 Kpee

Apparved ofo puirelae rerduto is. pe6ite3fr5npurpocsetfthe nitdStes Genment; ditibto unlimited

Cheical ociavl ety rhue,18

D3 eY rtm Sn ofanu th 'e av i3. NUMBERaf OFd PAGES y lcknmh

Suintnrfa-cnfine cataysis oraoe48cphtctss

Approve o introdelas fntoait olwdb reardction withmtedfr npups funto nied wits overnment by isttratin thnslidsithed. )

17 1STIO TO T TMN o Dervate suracenhae bn Bcharctdered by inraedspctosop

andcopared tod solutdon analogueston conr the porence of thei Aean

Surface-confinAd cataAysis OFaoetli Thtcaay ils PG WhnD

phtceity derivat- , /surfaces20.~~~~~~~~~~~ ABTRC (Cniu-nrvread ~oee n dniyb lc ubr

InorganiLcJ oxd sufcsii,2ad403,bern O ucJoaiyhv

V t

.K -. iCg, C2 zroucs ,:, -:'e sjrface. -e surcaae-conr, ec -So ' : 'rer-es-

:ncoreactons that oein ;, oss ,2 subsecuent to ootica" excitatocn 4n -irear-ultraviolet. The cho-c err s try Cl/ :araIe1s the :enav;or o 0 1.:1.

-o'2,., anaIocues; rar :e -'crosuostituted by (Ph ,r . t .e s r-ace-conineo iCo(O) (P(CPh' species is etectabie ty infrared soectroscopy.

:rradiation of the oxide powders bearing -SiCo(CO)a suspended in :t'SiH solutcrs

results in the release of the -Co(CCO) into solution as E SiCo(CO)4 concurrent

with the regeneration of surface - SiH functionality. Irradiation of the cowders

Co(CO)r), in the Presence of 1-pentene /-eiks rojCO)f-in solution and

surface Si(pentenyl) <groups. in the presence of EtfiSiH/l-pentene photo-

activated catalysis by the derivatized powders (7SiCo(CO)4) occurs to give

isomerization of the alkene, hydrosilation to give Et Si(n-pentyl), and small

amounts of n-pentane. Reaction under H2 improves the relative yield of n-pentane4

while H?/CO mixtures yield no hydroformylation products and lower the observed

C...* rate due to CO competing for the coordinatively unsaturated species. The.

surface --SiH groups also add to 1-pentene to aive oowders bearing

"Si(n-pentyl) functionality. The use of powders functionalized with

-.SiCojCOY4(P(OPh)r') also gives catalysis uoon ohotoexcitation, but the

.roduct distribution differs~siqnificant.y and includes at least two isomers of

Et3Si(pentenyl). in all cases the bulk of the catalysis appears to result3

Iron Co-carbonyl fragments photoreleased from the oxide support. The initial

rate of catalysis appears to depend on the initial rate at which the fragments

are released into homogeneous solution. The heterogeneous photocatalysts thus

give the same product distribution as appropriate homogeneous precursors, but

the oxide supported -SiCo(CO)g is more easily isolated and handled and more

durable than R SiCo(CO)4.

UNCLASSI FIEDSRCURIT CLASSIICATION OF THIS PAGItwon oe. Entered)

OFFICE OF NAVAL RESEARCH

CONTRACT NOO014- 75-C-0880

Task No. NR 051-579

TECHNICAL REPORT NO. 28

"PHOTOCHEMISTRY OF SURFACE-CONFINED ORGANOMETALLICS: PHOTOCHEMICAL

RELEASE OF A SURFACE-CONFINED COBALT CARBONYL CATALYST"

by

Carol L, Reichel and Mark S. Wrighton

Department of ChemistryMassachusetts Institute of Technology

Cambridge, Massachusetts 02139

Prepared for publication in the Journal of the American Chemical Society

June 8. 1981

Reproduction in whole or in part is permitted for anypurpose of the United States Government

This document has been approved for public release andsale; its distribution is unlimited.

NTIS GRA&I

DTIC TABUnannouncedJustification

By -Distribution/

Availabil ity CodesiAvail and/or

Dist pecia

P hot och em is tr v of S ur ac e-C rn ed ra rce t a cs: ho t ccernica I

Release of a Sur ce-C'cnf ired Cotal t Car! orvli Catalvst+

Carol L. Reichel and Mark S. '.riahton

Depart-ment of Chemristry';assachusetts Institute of TechnologyCambridge, Massachusetts 02139

This paper is dedicated to George S. Hammond in honor of his 60th birthday.

Abstract: Tnorganic oxide surfaces, SiO 2 and Al203, bearinq -OH functioralitylhave ceen

functionalized with -Co(CO)byfirst treaing tha sclid with (Et.SiN. 141SiH, orCl-SiH

to introduce "SiH functionality followed by reaction with Co2(CO)9 .

Derivatized surfaces have been characterized by infrared spectroscopy

and compared to solution analogues to confirm the presence of SiH and

SiCo(CO) 4 groups on the surface. The surface-confined :SiCo(CO)4

undergoes photoreactions that begin with loss of CO subsequent to optical

excitation in the near-ultraviolet. The photochemistry closely parallels

the behavior of solution R3SiCo(CO)4 analogues; CO can be photosubstituted

by P(OPh)3 and the surface-confined 1SiCo(CO)3(P(OPh)3) species is detectable

by infrared spectroscopy. Irradiation of the oxide powders bearing

--SiCo(CO)4 suspended in Et SiH solutions results in the release of the

-Co(CO) 4 into solution as Et3SiCo(CO) 4 concurrent with the regeneration

of surface Z--SiH functionality. Irradiation of the powders ( SiCo(CO))" 4)

in the presence of l-pentene yields Co4 (CO) 12 in solution and surface

-Si(pentenyl) groups. In the presence of Et SiH/l-pentene photoactivated

catalysis by the derivatized powders ( SiCo(CO)4) occurs to give isomerization

of the alkene, hydrosilation to give Et Si(n-pentyl), and small amounts of

n-pentane. Reaction under H2 improves the relative yield of n-pentane,

while H2/CO mixtures yield no hydroformylation products and lower the

observed rate due to CO competing for the coordinatively unsaturated

species. The surface --SiH groups also add to 1-pentene to give powders

bearing tSi(n-pentyl) functionality. The use of powders functionalized

with 1SiCo(CO)3(P(OPh)3)also gives catalysis upon photoexcitation, but

the product distribution differs significantly and includes at least two

isomers of Et3Si(pentenyl). In all cases the bulk of the catalysis

appears to result from Co-carbonyl fragments photoreleased from the oxide

support. The initial rate of catalysis appears to depend on the initial

rate at which the fragments are released into homogeneous solution. The

heterogeneous photocatalysts thus give the same product distribution as'

appropriate homogeneous precursors, but the oxide supported -7SiCo(CO)4is more easily isolated and hatdled and more durable than R3SiCo(CO) 4.

-2-

Chemistry of surfaces modified with molecular reagents may have use

in stoichiometric and catalytic syntheses, in adhesion and other surface

physical tailoring, in analysis and separations, and in chemical energy

conversion. 1 10 Successful application of molecular modification depends

on characterization of both the structure and reactivity of the surface-confined

species. One aim of research in.this laboratory has been to initiate

catalysis by optical excitation of thermally inert, surface-confined

organometallic reagents. Thermally inert, surface-confined -Fe(CO)n

reagents have been demonstrated to be photochemical precursors to catalysts

for olefin isomerization and hydrosilation. 11 The anchoring surface for

the -Fe(CO) n was a phosphinated polystyrene polymer, - where

the organic polymer can be regarded as an insoluble, though solvent

swellable, triarylphosphine ligand. For the (-Fe(CO) n system we found

that the Fe-P anchor bond is sufficiently photoinert that olefin

isomerization could be concluded to occur when the Fe-P bond is still

intact. However, the polymer appeared to be sufficiently flexible that

photogenerated coordinative unsaturation does not persist for extended

periods. Further, it is likely that the(D system would be reactive to

extensively coordinatively unsaturated Fe(O) species. Consequently, we

have now turned attention to inorganic oxides as surfaces to which

photochemical catalyst precursors can be attached. Inorganic oxides

such as SiO 2 and Al203 should be inert, can be functionalized, and are

transparent to ultraviolet and visible light over a wider range than for

the & system.

The system described in this article is the surface-confined

-SiCo(CO)4 SiCo(CO) , where the & is Al203 or SiO The photochemistry

and photocatalytic activity of R3SiCo(CO)4 (R = Et, Ph) have recently been

described. 12 The Q SCo(CO)4 materials are air stable at 250C for

-, __ __ _ _ _ _

-3-

prolonged periods in contrast to Et3SiCo(CO) 4 , but the Q "SiCo(CO) 4 is

very pnotosensitive. The photoactivity can be used to release catalytically

active species into solution, but the primary photoprocess is CO release,

not Co-Si bond cleavage. A preliminary communication has established that

photochemical Si-Co bond cleavage does not occur; irradiation under i3CO

yields )iCo(CO)n(1CO) 4 -n' not the release of Co(CO)4 to form

C02(CO)n(1 3 C0)8 n 13

Studies of the surfaces of insulating metal oxides are not easily probed

by surface sensitive electron spectroscopies. We have made use of Fourier

transform transmission infrared to characterize the surfaces functionalized

with 3SiH and ZSiCo(CO)4. This technique affords a molecular specific

probe of structure and allows monitoring of the dilute catalytic sites of

interest. As shown below, the surfaces of Al203 and SiO 2 that we have used

for photocatalysis have many surface functionalities present. As in

solution, we exploit the fact that a chromophore, -SiCo(CO) 4, present in

dilute amounts can be selectively activated by light. This represents

one of the major advantages for optical vs. thermal activation of catalysis;

in principle, the activation energy can be deposited selectively into the sites

that are active.

-4-

:xoerirmental

Ohvsical 'easurements arc naivm.. Infrared snectra of surfaces ,ire taken as KBr

eeliets (--1- AlO, or SiCQ) on a Perkin-Elmer Model ISO discersion or a

Nicolet 7199 Fourier transform spectrometer. Solutions were analyzed in

matched 1.0 mm or 0.1 mm oathlength sealed NaCl cells. Electronic spectra

of solutions were obtained on a Cary 17 uv-vis-nir spectrophotometer in 1.00 cm

quartz or OO cm Pyrex cells. Gas chromatographic analyses were performed on

a Varian Model 1440 or 2440 gas chromatograph equipped with flame ionization

detectors and interfaced with a Hewlett-Packard model 3370S recording

integrator. Hydrocarbon products were analyzed at 200C on columns of 20%

propylene carbonate on Chromosorb P (Johns-Manville, 1/8" x 30') against

internal standard,n-hexane. Hydrosilation products were analyzed at 60'C on

columns of 200% s,.'-oxydipropionitrile (Applied Science Laboratories, 1/8" x 30'"

against an internal standard, n-decane. Authentic samples of all products were

available for calibration of the detector response and for retention time

comparisons.

Reaction Conditions. All manipulations of 02- and H20-sensitive materials

were performed in a Vacuum Atmospheres He-43-6 Dri-Lab glove box with an

attached HE-493 Dri-Train or in conventional Schlenk glassware under an N2

atmosphere. Samples for irradiation were freeze-pump-thaw degassed at

least 4 times in 13 mm o.d. Pyrex ampules containing 2 mm x 7 mm Teflon-coated

magnetic stirring bars, and sealed hermetically under vacuum (10 5 torn).

Reactions under H2 were performed in 13 mm o.d. Pyrex tubes fused to

ground-glass joints, which were connected to a source of vacuum and of H2.

The system was flushed at least 4 times with H2 before H2 gas was admitted

to the sample under vacuum. Irradiations were performed with a medium-pressure

Hg arc lamp (Hanovia, 450 W) fitted with a supplementary H20 filter to remove

infrared radiation; or at lower intensities with 2 General Electric #FlST8/BLB

31ack Lites ( 255 20 nm- 2 x - ein/min as determined by

ferrioxalate actinometry).

.Laterials. Chromatocrapnic ,-race SiD Kavison :herical Corrcany, C-CC

mesh) was used as received. Chromatographic grade natural A1203 ('.ioelm grade I)

was activated for 24 hr at 4500 C before use and stored in the dry box. High

surface area y-Al2 0 (200 m2/g, Strem) was received as pellets and usea as

received or pulverized before use in some experiments. Powrerea, nigh surface area

SiO 2 (400 m2/g, Alfa) was usea as received. 1-pentene (99',,

Phillips Chemical Company) and HSiEt3 (Petrarch) were passed through activated

Al203 prior to use and stored in amber bottles at -VC. Triphenylsilane

(Petrarch) was recrystallized from n-pentane prior to use. Triethoxysilane

(Petrarch) was freeze-pump-thaw degassed upon receipt and stored in thie dry box.

CeC%)8 (Strem), n-hexane (M ,drd&), n-decane (A,,c, Si C, r "idr- cr,","i

were used as received. RSiCo(Ct?)R = t, Ph) were prepared as previously described

and stored at ". under N-.12 CO!o)0) was prepared Dv neating Co2(C0) in

isooctane at 80-90'C for 12 hr. The product was filtered off, recrystallized

from benzene/isooctane and stored at 0"C under N,. Hydrocarbon solvents were

distilled from CaH 2 under N2; THF was distilled from Na/benzophenone under N2.

Jerivatization of Oxide Supports. A typical procedure is outlined. 1.6 ml HSi(OEt)1

in 10 ml isooctane under N2 was added to 4 g of powdered A1203 in a Schlenk tube.

In some cases, 0.12 ml distilled H20 was added via

syringe. The suspension was allowed to stand at 250C under slow N2 purge for 12-36 hr

with periodic agitation, after which time excess solution was drained off.

In some cases, the -SiH derivatized surface was !,,ashed with hexanes and

isolated at this stage. Under N2, a solution of 1.4 g Co2(CO)8 in 10 ml

isooctane was added by syringe. After 12 hrat 25C under slow N2 purpe with

intermittent agitation, the tube was opened in the dark. The defr:i d powderwas

washed with a copious quantity of hexanes or pentane until the rinsings were

colorless. The derivatized powder was dried under vacuum at 25°C.

unctionaiization usina ',IeClSiH was carrieo out in a manner similar :o

that Oor CI SiH or t--tOC.SiH except that '.r/ .,as adde, A tvoical croceaure

was to react 2 g o' .12 03 or SiC 2 with 5 1 .eClSiH and 4 -! of Et. in -0 <

of isooctane -For 21 h at 200c. The excess .e....iH, ... nd soocane .ere

stripped off under vacuum. The solid was washed with iscoctane leaving the

& SiMe 2 H and the Et-N'HCl. The Et il was removed by ,qashino with

CH2CI2 and again with isooctane. To furtner fuinctionalize to 7orm

SiMe2 Co(CO)4 the& SiMe H was reacted with excess

isooctane under N2 for 22 h at 20'C. The solid was washed with isooctane and

dried under vacuum.

The surface -SiH and-SiCo(CC are infrared detectable, Table

Typical analyses were performed by making measurements of KBr pellets

containing iH and/or j SiCo(CO), (-10% by weight). The pellets

are made in as reproducible a fashion as possible. 200 mg of cry, infrared

grade K3r and 20 mg of iH or SiCo(CO) 4 are first grouna together&/

with a mortar and pestle until a homogeneous powder results. The pellet is

then pressed in a KSr die at 20,000 psi for 5 min to yield a pellet of 1.00 cm

diameter and 0.C5 cm thickness. Absorptivity of model compounds Ph3SiH

(wSi.H = 2118 cm-1 in K r) and Ph3 SiCo(CO) 4 (vC_ 0 = 2092, 2030, 1990 cm- 1 in

KBr) was determined in KBr and reproducibility was determined to be +20%.

Errors for samples including the SiO 2 or A1203 are even larger. The semi-

quantitative infrared measurements show (&)SiCo(C0)4 to be air-stable at

25'C for at least several weeks. The relative absorptivities in Table I,

relative to Ph3Si-H (vSi-H = 2118 cm-1 ) are proportional to concentration and

can thus be used to determine concentrations of the S iH or iCo(CO) 4.

The relative absorbance of 1.0 for the 2118 cm l Si-H stretch in Ph3SiH

corresponds to an absolute absorptivity of 1.0 O.D. units for a 2.54 tolOOweight

ratio of Ph3SiH to KBr in our 1.0 cm diameter, 0.05 cm thick, pellets prepared as

described above. Tius, the relatix absorbance of 5.5 farv CO= 2092 cm of Ph3SiC CO). meansthat the absortivity is 5.5 tirmes that of the 2118 cm "I .i-H 5.r tch nf Ph.iH for the same

number of moles in a given mass of KSr.

-7-

c insure that <Br is not reacti4ve wnen sec as a nDellet -7aterial

,.e nave recorded the infrar-ea spectrum of S'Si~oCO1 ) and 7 -

'ujoi mulls. The relative aosorpri.,ities are tne same as 4;n ',3'

ellets. Also, the infrared absoroances in .-'r are proportlonai to -tne

signals from Fourier transform infrared photoacoustic spectroscoov of the

-~ 13S '-SiCo(CO), powaers.

Results anc iscssion

a. Surface 'eri v at'z i on and Characteri zation. -le s,/rtes ,s --

deends on t.eo crevious synthnetic orccedures eo'a: rs a' t -e/

-t2,SiH + surface---i srface-- -- - C.I

/

2H-SiR C2(C --- 2 HSiCo(CO)+ H

and SiO 2 generaily have variable amourt of surface -CH depending on prior history -,f the sai;Des.

We have synthesized (aSiCo(CO)4 by exploiting the chemistry represented by

equations (1) and (2) and have characterized the surfaces by transmission infra-

red spectroscopy. epresentative infrared spectra for a typical

05/ "SiCo(o )40 ,Q = Al203 , synthesis are shown in Figure 1. -he Ai203

is featureless in the Si-H and C-0 stretching region. -reat-et .41i1h

(EtO) 3SiH results in a prominent spectral feature at -2250 cmi ..,e associate

this feature with the Si-H stretch for the surface-confined 7rrss. Peact;cn

of thei S-SiH powder with Co2(CO) 8 results in a powder that has infrareo

absorption at -2110, -2050, and -2020 cm in accord with a S ico(ooyi

system. Note that while we generally find some dimunition in the absorption

associated with the Si-H stretch,it is not completely absent even after

exhaustive reaction with Co2(CO) 8 . Table I lists infrared absorptions for

(2J-SiH; SiCo(CO)4 ; and various other systems of relevance to results

described below. Included in Table I are data from experiments using CI3SiH

or Me2ClSiH, rather than (EtO) 3SiH, to introduce :SiH functionality on the surface.

Infrared spectroscopy, Table I, reveals the presence, but not the

uniformity of coverage, of '-SiH and -SiCo(CO) on A103 and SiO. TheI. 4 2 3 2'

characteristic absorption associated with the SiH is in the range 2134-2257 cn-I ,

depending on the oxide (A1203 or SiO 2 ), the starting silane ((EtO) 3SiH, C13SiH,

or Me2ClSiH), and the derivatization conditions (H20 added or not added). A

large range of Si-H stretching frequencies is also found for the various silanes

measured in alkane solution. Generally, the Si-H stretching frequency would be

e adectc to ':e a : -nc-- r t' "-e "r_ ree!- aninc 3rcur s zrre.- to -:e

are c_ z-table " -e a t2S

-2 e ".',o a~/~. :rucs on toe surface S4 .hen ;sinc 'ceC]- whereas tre -

or -tC.SiH 'ikely result in only oxygen linkaoes to -re surface Si. -- _ '_ezor colvmeric material 'rom the reac "c'r ocF .t? " i " C>Cf- rcc-r.

. .. .. C r. ., r r.. -_r. IH .. .

.-,el vitn the surfaces derivatized with tnese reagents. :ssentiali> co a~e.e

nyarolysis of the (EtO) 3SiH is inferred from e very T6 signas f-or C-H stretc"inn

absorptions (from EtO- groups) for A120, or SiO 2 functionalized with

\.tO),SiH. For Me.ClSiH the retention of the metnyl groups is direct.y

confirmed by the observation of a C-H stretching absorption at 2985 an for

-ne -erivat zec surfaces; 'e CSiH shcws a C-H stretcn at 2965cm in .

s.jcon. -'he value of i or ..e-CSiH, ,'H, _.CSi. nc t e r --

cor a-incnc -aterials is consistent with the retention ol the tv, e -et.'frl -rcuos

.-,hen isinc Me.,'lSiH to functionalize the surface.

Spectral features attributable to the -Co(CO> are also demendent on

the nature of the group bonded to the Co. The variation in the Co stretching

frequencies for (EtO) 3SiCo(CO) 4, Et3SiCo(CO)a, and Ph3SiCo(CO)4 accord well

with the variation for & /SiCo(CO)4 depending on whether Me2CISiH vs.

CI3SiH or (EtO) 3SiH is used as the Si-H source. The two alkyl groups on the

Si release electron density (compared to three oxygen groups) such that the

C 0 is moved to lower energy presumably reflecting greater 7-back bonding to CC from

the more electron-rich Co. Thus, for both .SiH and -Co(CO)4 some molecular

level information can be deduced from the infrared spectroscopy. But small

differences in band position should not be overinterpreted, since the bands are

relatively (compared to solutions) broad. Even for the same sample, some

variation (+3 cm"1) in band maxima can be found from KBr pellet to KBr pellet.

)

-u--1, measurin tne absoroance associated wit h - 2250 cm' ana

/7.\(CD) , . 2C25 cr"' for KBr pelle.s .-.e nave estimated the amount o7 i an c confined to the oxide surfaces. Resuts from various

3 3,, onfnedto he xid sufaes. :est froiariuvsiia

Dreoararjion ;roceaures using various sources of ., ye .uali:ativei s4i'ar

results, Table :I. Note that !SiH oersists in every case; some of this

functionality may be either structurally or mechanistically inaccessible to the

Co2C0),. The data for the high surface area (200 m2/g) ".-l2C3 De~lets, sample

=i in Table 1I, allow a calculation of average coverage of and iCo(.Oa

assuming all of the surface area to be accessible. WJe calculate an average

coverage of lO 12 mole of ;SiCo(cO)4 and -3 x 1011 mole of ;SiH oer m2 of

surface. Similar coverages are found for the high surface area '00 M2 /g) SiC 2,

sample =8 of Table Ii. Taking lOlO nol/cm 2 to be a monolayer, we find mono-

layer to sub-monolayer coverage to be a typical result. The coverages from

infrared spectroscopy are very approximate, perhaps correct within a factor of

three. There are many sources of error including the fact that absorptivities

for the Si-H stretch vary considerably, Table I, depending on the environment

around the Si and the medium in which the --Si-H is dispersed or dissolved. But/

the qualitative finding of monolayer or sub-monolayer coverage is in accord with

determinations of coverage of other functionality, albeit larger groups, using

hydrolytically unstable silane reagents. 16It is noteworthy that -OH absorption

in the infrared does not disappear upon functionalization of the oxide powders

with either (EtO) 3SiH or the more reactive C13SiH. Some of the -OH after

derivatization could be due to -Si(H)(OH)x from the hydrolysis of the (EtO)3 SiH

or Cl3SiH. Further, it is unlikely that all -OH associated with the hydrated

oxides is accessible.

While it may be attractive to conclude that (EtO) 3SiH or Cl3SiH functionalize

the oxide with a uniform monolayer or less of --'SiH, such is unlikely. The

(EtO) 3SiH and C13SiH can polymerize and such is known to occur when using

(RO) 3SiR' or CI3SiR' reagents for modifying chromatographic supports or electrode

materialbs.' 9 Indeed, unless H20 can be completely removed the polymerization

is probably unavoidable. Removing all of the H20 would also mean

,envc r.-e t n e ,-_ i ,es an2 c-erage tre on _ re .nnicrous oxi-ies

o vi4es szre *a:iec - ra ccncerru'o -r eo '-eriz:cn s'nce

"eCISiH nave ar a ouflt o --.- , zasec -n infrared a: scr.ion, :,?.a" is

not cualitatively zi4=erent than that ;rcm cerivatization with CiSiH or

(EtO) 3SiH. This likely neans oolirerizatinn from CI3SiH or (EtO),SiH is not

eyTensive. ..t least, we nd ; r', e . esu! ts are consistent :ii r-her

uniform, low coverage of -Si-H inceoendent of the source of -Si-H. iowever,

as indicated above, the Me2Si-H is electronically different from the-- Si-H

.nere the three groups bonded :o tne Si are oxygen.

Another complication in -ne attachment of -Co(CO) 4 to the oxide supports is

17-at Co2(CO) 8 reacts rapidly with Le.5is bases to aive redox products. 1 or

example, we find that Al3 3 reacts witn Co2 (C0), to give a bluish oowder. This

bluish color also appears when the 420 has been first treated with

E'tc)SiH to introduce -S'H sur4ace zrouos prior to reaction on

,ie attribute the bluish color to Co on the surface; this was confirmed byeacting the Co2(CO)8 treatec 123 with aqueous 1 M HCl. The resulting pink

solution showed optical absorption spectral features identical to those for an

autneitic Co2+ sample in aqueous 1 M HCI % max: 1250, 510, 463 sh no. 1 The

Co2(CO) 8-treated A1203 shows no C-0 infrared stretches attributable to surface

confined metal carbonyls. We thus conclude that Co2(CO) 8 reacts with A1203 to

yield a Co-oxide/hydroxide. Thus, Co-oxide formation appears as a side-reaction

accompanying the attachment of -Co(CO) 4 by reaction of Co2(CO) 8 with the

QS iH groups.

Elemental analyses for some of the A1203 samples are included in Table I.

Non-derivatized samples show little or no detectable Co or Si and samples

derivatized with (EtO) 3SiH but not with Co2(CO)8 show the presence of Si but

not Co, as expected. The comparison of the elemental analyses with the infrared

analyses is revealing. There is apparently far more Si present than can be accounted

for by the -SiH and SiCTC 14 . Further, there is far more Co present that can be

-12-

aczounted For by ':.SiCo(CO). T e infrared determined coverages of =SiCo'C O

are :onsistent >2O) with the amounts of Et3S iCO(CC) 4 that can be detected

in solution wnen S3iCo( CO) 4 is irradiated in isooctane as a suspension in

the presence of EtviH, vide infra. Thus, the excess Co From the analyses

must be due to the Co-oxide/hydroxide formed from the decomposition of Co2 (CC),

the derivatization procedure. Though we do nt hae a check of :SiH as we do/

for TSiCoCO)4, it would seem that the error in Si from infrared

detectable _' SiH cannot be as large as would be necessary to account for the

s Si from elemental analyses. We propose that decomposition of SiH

occurs, perhaps by acid-base reactions, to give silicate on the surface.

Reaction with Co2 (CO) 8 could lead to decomposition of SiH, but Even surfaces

that have not been treated with Co2(CO) 8 give more Si than can be detected

as -SiH by infrared. A major conclusion from these results is that neither

elemental analyses or infrared alone can provide quantitative

information concerning the surface functionality present.

-om our studies, we conclude that the derivatized oxide particles have a

large number of chemical functionalities present. The representation of a

derivatized alumina particle in Figure 2 seems justifiable in view of the

available infrared data and known reactions. Despite the large number of

chemical functionalities, light-activated reactions of the-SiCo(CO)4/

can still be studied, owing to the fact that this group can absorb incident

light and reactions of this metal carbonyl fragment can be monitored in thc

C-O stretching region of the infrared without interference from other

groups. Photoexcitation of -S'iCo(CO) 4, though it is present in dilute

-13-

amounts, is analogous to optical excitation of suostances in solution at lcw

concentration, explciting one of tne 'key advantaces in optical vs. thermal

activation ci chemical systems. The absorption onset of $ i," os i

expected to occur in the near-uv, based on solution analogues, and

consequently, photochemical studies have been carried out using near-uv

300 nm) light.

b. rnotocnemistry of QSi ko(O in the Presence of P(OPh). rraiation

of deoxygenated 3.0 ml isooctane suspensions of 0.045 g of iCo(CO)4 ,

sample i3 in Table II, in the presence of 0.01 M P(OPh) 3 yields

reactions accorting to equation 3. Infrared analysis of the sucernatant

\near-u(3S - IC°C) 0.0 -. ((~h D iCo( CO), kP(Oh)- 10 (3)Si' (C) isoo-tane O

liauid shows no metal carbonyl species. The formation of Co2(C0)6 ((OPh))2

woulk be possible if light-induced cleavage oftheSi-Co hond resulted followed

by substitution of the released Co(CO)4 radical by P(OPh3 ) to yield Co(CO)3-

(P(OPh)3 ) radicals that then couple to form Co2(CO) 6(P(OP 3)2. 9 Infrared

analysis of the powder after various irradiation times, Figure 3, shows the

dimunition of absorptions attributable to the (&)-SiCo(CO)4 and the growth

of features that are assigned to S-SiCo(CO)JP(OPh3,Table I. Thus, it

appears that CO can be photochemically ejected from Q-S iCo(CO)4 and Si-Co

bond cleavage is apparently unimportant.

Irradiation of Q /SiCo(CO)4 in isooctane solution containing no

deliberately added entering groups results in the dimunition of metal

carbonyl absorptions in the C-O stretching region. An unidentifiable

absorption feature does grow in at 2073 cm"1 as a shoulder at intermediate

stages of reaction, but the surface would appear to be eventually denuded of

metal carbonyls by the near-uv irradiation. The isooctane supernatant

exhibits infrared absorptions at positions characteristic of Co4 (CO) 12, Table I.

The yield of this species appears to be -15%. Both these data and data for

-14-

photochemistry in the :resence of ?(OPh) 3 parallel the -hoi:.chemical data :or

R3SiCo(CC)' in solution.12

3\

c. Photochemistry of i in *he .resence ( t 7. t- S

conditions wrere ,:' s otolabilized, 4r-adiatior of sujsence ,4

in the presence of EtSiH results in the detachment of the -CC). from the

surface. For example, degassed, 3.0 ml isooctane solutions containinc

0.045 g of S iCo(CO),, sample =3 of Table I, and 0.5 ' Et SiH

under near-uv irraciation at 25'C yield Et3SiCo(CO) 4 in the

supernatant isooctane solution, Figure 4. Analysis of the irradiated powder

reveals the dimunition of absorptions due to metal carbonyl fragments,

Figure 5. The only feature that grows is one at -2250 cm-i attributable

to the regeneration of S-iH. infrared analysis shows that

>80% of the surface-confined -Co(CO) 4 fragments can be detected as

Et3SiCo(CO)4 insolution. SuIrh data provide confirmation of the coverages of

surface-confined Co(C0)4 groups given in Table II. The photochemistry again

can be rationalized as resulting from photoinduced loss of CO as the primary

photoprocess according to equations (4)-(7), paralleling the solution photo-

~ - iCo(co) 3 +CO 4

OD- SiCo(CO)4 4- (4)H

SiCo(CO) EtSiH Si 3 (5)

H SiEt3

(-SiCo(CO)3 - ()SiH + Et SiCo(CO)3 6

Si Et3

Et3SiCo(CO) 3 + CO - Et3SiCo(CO)4 (7)

chemical behavior of R3SiCo(CO) 4 .12

d. Photochemistry of -SiCo(CO) 4 Related to Alkene Catalysis. We have

previously reported that R3SiCo(CO)4 is a photochemical precursor to

c-lailyticail actie species, cesnite- -e fact that Si-Co cond cleavage

c: the pri7ary pnotoprocess. "iear-uv :rradiaticn o, f S i':0o < -c r

4scoctane solutions .f I-oentene leads to the zeneration of ro,

solltirn. :n~rarec analysis of the irradiateG powcer snows lo.s f

groups and the growth of weak absorptions in the ^-H stret:hing recion tnat

could be due to Q.Si (pentenyl. Photolysis of Et3SiCo'CO), in the

presence of l-pentene yields Et3Si(pentenyl) as the only Si-containing

product.12 The irradiateo powcer also snows consumption of ( 4 1

functionality. Apparently, the Si-H groups not originally reacted with

Co2(CO)8 can react with photoreleased Co-carbonyl fragments and subsequently

with 1-pentene to yield SJ Si;,n-pentyl). that is, the resizua7 S iH(D-/

reacts with the Co-carbonyl/l-pentene to hydrosilate the I-pentene. Thus,

reaction with alkene and S.SiCo(CO)4/ S SiH can be induced pnoto-

chemically. However, there is no accumulation of catalysis oroducts in

solution. Even alkene isomerization activity (1-pentene - :is- or trans-

2-pentene) is nil.

Irradiation of OS iCo(CO) 4 in the presence of 1/1 mixtures of

l-pentene/Et 3SiH does result in catalytic chemistry. -or example, near-uv

irradiation of 0.045 g of S SiCo(CO)4, sample #3 in Table II,

in the presence of 0.39 M Et3SiH/O.39 M 1-pentene in isooctane rapidly

yields infrared detectable metal carbonyls in the supernatant solution.

Initially, Et3SiCo(CO) 4 is formed, and

irradiation also yields infrared absorptions in the

supernatant at 1960 cmn l and 1966 cm-l (sh) that have been previously

observed, but not unambiguously assigned, 12'15 in solutions initially

containing Co2(CO) 8/l-pentene/Et3SiH. Gas chromatographic analysis of the

solution after 48 min of irradiation shows the following composition (% based

on initial 1-pentene concentration): n-pentane (4.2); 1-pentene (1.3);

trans-2-pentene (46.3); cis-2-pentene (15.5); (n-pentyl)SiEt3 (32.7). Thus,

photocatalysis is effected, equation (8). On the timescale of this experiment

-16-

e En SiH Et(Co )4 n-pentane, cis-2-oee:ene,

penene + 3 near-uvtrans-2-pentene,( -pe nty l 'S i .t3 f,

there is no therral catalysis. Further, Al2 03, Al203 treated only with

Co2 (CO) 8 , or A!203 treated only with (Eto)3SiH gives neither thermal or

photoinduced catalysis under the same conditions. Thus, the S SiCo(CoD

would appear to be a photochemical precursor to catalysis. in view of the

photogeneration of Et3SiCo(CO)4 from Q SiCo(CO)4 , equations (4)-(7),

in the presence of Et3SiH or in the presence of Et3SiH/l-pentene, catalysis

should result if for no other reason than Et3SiCo(CO)4 yields catalysis

when irradiated in the Et1SiH/l-pentene mixture.12

Irradiation of Q-SiCo(CO) 4 in the presence of neat 1/l

Et3SiH/l-pentene results in catalytic chemistry. Data for a typical

photocatalysis run are included in Table II. No dark reaction occurs on the

timescale of this experiment. Infrared analysis of the (-iCo(CO) 4

shows >95% loss of -Co(CO)4 from the Al2C3 surface at -30 min irradiation

time. Analysis of the solution shows the growth of the absorptions at

-1960 cm-1 found in thermal reactions of Co2(CO)8 with Et3SiH/l-pentene1

and phocoreactions of Et3SiCo(CO)4 with Et3SiH/l-pentene.12 At longer

irradiation times infrared bands for Et3SiCo(C0) 4 become apparent and the

1960 cm 1 feature intensifies somewhat. After 375 min, there are no detectable

Co-carbonyl fragments on the surface of the Al203. The mol % of Si-H in the

Al203 decreases during the irradiation, likely reflecting the

generation of (S Si(n-pentyl).

The main conclusion regarding catalysis from the irradiations of

Q-6iCo(CO)4 in the presence of Et3SiH/l-pentene is that the bulk of the

catalysis occurs from the photoreleased Co-carbonyl fragments. For example,

in the particular run detailed in Table III, most (>951) of the Co-carbonyl is

in homogeneous solution after the -30 min irradiaticn time. To orovide

-17-

r:tona' confirratior S ' :r\',/

samcle =3 in -acle , .as i -acia-ec in 1 - o: neat E. iH

nt~l 2C- -1i rnn irrad ati or tre -Co:CC , ias oresent 4r solution as

e s, aa a~n - ~u-,eO nt 7-en ee t o zrinc tre

EtSiH'/l-centene .o a 1/1 7ole -atio. Irradiation of toe Et3SiCo(CO) /'Et i/

1-pentene gave ohotocatalv/tic action with a rate and product distribution

verv similar to that Then the , SiH S S -Co( CO ase was not

removed. The catalytic procucts and their initial ratios accord well with

previous findings in this laboratory from thermal catalysis using Co2 (C0) 8

or photocatalysis using Thus, we firmly conclude that

photocatalytic action from y'S-.:SiCo(C0) 4 results from reactive Co-carbonyl

fragments released into homogeneous solution. The active species appears

to be the same as that involved in Co2(C0)8 or Et.SiCo(CO), catalysis.

e. Ohotocatalvsis Results Under Variable Conditions. Photocatalysis of

Et3SiH/l-pentene mixtures has been carried out using various preparations

of SiCo(CO), and under variable conditions. Taole 1V details results

for various oreparaticns of /SiCo(CO, 4 where the najcr variable would

appear to be the amount of -Co(CO)4 actually added when 0.045 g/ml of

(D-siCo(CO), is added to the mixture. Note that the irradiation source

(355 + 20 nm, 2 x 10-6 ein/min) for the experiments summarized in Table IV

has lower intensity than for the Pyrex filtered 450 W Hg-arc used for the

experimerts summarized in Table III. Accordingly, the reaction times are

longer. Variations in the activity of the -SiCo(CO) 4 system crudely

parallel the expectations for rate of release of -Co(CO)4 . Higher coverages

and large surface areas provide for a higher rate of appearance of active

material in solution. All of the catalyst precursors yield the same

Si-containing'product, namely Et3Si(n-pentyl) and rather small amounts of

n.-pentane. In cases where there is significant dark reaction at 25'C on

the 69 h timescale we also find very rapid photocatalytic action. The thermal

activity in the dark is attributable to a slow thermal process again involving

;ss 31 is 'ne " 4e 4:4:na step. However, jncer if' concicirs tne

".,t-acci'ate, :zacas-s croceecs rore raciciy even at zre icw light

s ... :' -jen 4r - 'e T ,e , -arc: source

a sec, E r, :re!.a! cot s sho cr no reaction or t'!e

timescale of the exoeriments summarized.

Table I'l includes aata for reactions carried out in neat 1/1

_.SiH/1-pentene mixtures exoosed to H or H2,'CO -i:,tures. The data show at

the product distribution under 4 psig H2 shifts markedly to higher yields

of n-pentane. However, irradiation of Q- SiCo(CO)4 in the presence of

neat l-pentane and 4 psig H2 yields little isomerization or hydrogenation.

These results are consistent with our previous findings for photoactivation

of alkene catalysis using Co-carbonyl species: H2 as an oxidative addition

substrate to create hydrides for isomerization or reduction does not seem

to be effective.2 The H2 /CO atmosphere suppresses the rate of catalysis

even at the 4 psig CO pressure used. The resulting product mixture contains

no detectable hydroformylazion products. The suppression of catalysis is

associated with the capture of coordinatively unsaturated sites by the CO.

The effects of H2 and H2 /CO parallel those found earlier for

photocatalysis of alkene reactions using Co-carbonyl species. 12,20

The ability to functionalize SSiCo(CO)4 by photosubstitution affords

an opportunity to examine the photocatalytic activity of

QSio(CO)(P(OPh)3 ). Irradiation of in the presence of-, 3 Yi0SCo(CO) 4 peec

P(OPh)3 followed by decanting and washing yields a mixture of

SiCo(CO) 3(P(OPh)3) and S SiCo(CO)4 '. Use of this mixture as a

catalyst precursor compared to pure S iCo(CO) under the same

conditions yields an important difference in the catalytic products, Table V.

The formation of at least two isomers of Et3Si(pentenyl) accompanies the

usual distribution of products. Such products have previously been found

from the thermal and photocatalyzed reaction

of Et3SiH/1-pentene mixtures using Fe(CO) 521,22

-19-

3r~~ 1 -,s' -nother o r ~ :

s:r t r uon is zre lower initial trans,'cis ratic o- 'e .-eftee

*scrers comoarea to the catalist having no aPhs :eset. - halst3

of te /iCo(CC)4 SiCo(CC),((JPh cowoer snows loss of

:o-carbonyl scecies, again indicating that the catalys4 s occurs via Onoto-

•?'easeo :o-carbonyl fragments. The very different --ocuc, ratio ,4ith the

zomolex indicates that the actual catalyst is dillerent than that -or

io(O.It is attractive to conclude that CO loss is again the\ -S'Co(CO)4 g

cr'mary ohotoreaction leaving P(OPh) 3 in the coordination sphere to influence

:roduct ratios by steric and electronic effects. The catalytic activity

aooears to be somewhat lower than for the unsubstituted S-SiCo(CO)4.

-his is very likely due to lower quantum yields for :he primary process

3 loss from the P(CPh)3 complex.

Cornoarison of Catalytic Activity of S -SiCo(C^, .,4trh Co CC) 3

Photoexcitation of. SiCo(CQ)4 in the presence of l-oentene/HSiEt,

,nixtures apparently yields the same catalyst species that results from the

irradiation of Et3SiCo(CO)4.12 Further, we have previously concluded that

the irradiation of Et3SiCo(CO)4 leads to the active species formed

thermally from Co2(CO)8.12 '15 Thus, comparison of the catalytic activity

of Co2 (CO)8 and Et3SiCo(CO)4 is appropriate. We have examined the initial

thermal activity of Co2 (CO)8 at 200C for isomerization and hydrosilation

of 1-pentene under the same conditions where we can photoactivate the

" SS iCo(CO)4 system. Data in Table VI are representative of the activity

of the two different precursor systems. In both systems the isomerization

products are formed at a faster rate than the hydrosilation products and it

would appear that the rate of formation of hydrosilation products declines

as the l-pentene is converted to the 2-pentenes. This is reasonable, since

it is likely that the hydrosilation of 1-pentene is more rapid than the same

-20-

,'eact*;on of :he in-:e~nal isomers. The interesting -4rcing is that the

:hotoactivated system is in fact quite similar in turnover rate to the

C o"' ther-,al system. F-,'rther, since the ooserved rate of procuct fornatlcn

depends on excitation rate (light intensity) the turnover rate for the Photo-

activated system could be even higher at higher light intensity. In both

systems it would aopear that the isomerization turnover number is similar;

m' e rrj aj. the numrer of product molecules formed is generally

limited by substrate concentration not catalyst lifetime. For ydrosilation,

we have already demonstrated that illumination of EtSiCo(CO), allows more

turnovers for (n-pentyl)SiEt3 formation, since the Co2 (CO) 8 thermal system

ultimately ceases activity due to formation of Et3SiCo(CO), that is not

thermally active.

Irradiation of soecies such as RSiCo(CO), does result ir ersistent

thermal activity in the dark at 202C. However, tne activity is relatively

low and short-lived. For example, a flash excitation of Ph.,5iCo(CO), leads

to .50% dark isomerization of 1 M I-pentene in the presence of 1 M HSiEt 3

during a 4 h period. Continuous irradiation leads to nearly complete

equilibration of the linear pentenes under the same conditions in <4 h.

Thus, considerable activity can persist in the dark, but irradiation

accelerates the catalytic rate even further.

u.- a...-21-

Summrary

The =ata acecuarev/ ,:erorstrae trat -Co Q nc'ore: :o xe s races

via reaction if "c-. ,.Sit ... - iH exhibits :;notccre-:7,rv ta:

arallels that for solution R3SiCo(CO)4 analogues. The phctoinouced loss

of CO is the principal result of photoexcitation of either R3siCo. Th, in

solution 2 er iCo(C); s~spenced in solutions. In neither :ase :C

,4e find evi-cence for prompt scission of the Co-Si bond, but in each -ase

reaction with R3SiH or alkenes leads to chemically efficient loss of the

original Co-Si bond. For eSo(CO) 4 this means that the Co-carbonyl

soecies does not remain anchored to the surface. Thus, with this system we

have not achieved the capability of "matrix isolating" photogenerated,

coordinativel unsaturated inter-nediates. However, theSSiC (^'"

system as a zotal/st precursor is easily handled and is :ess sens-:i',e

than is the solution s:eciesEt 3SiCC(CC). The concept of releasInc a caal /s-

ohotochemically at a rate controlled by light intensity is one that zcuic

be exploited. 4n zrinciple, in catalytic reactions.

Acknowledgements. We thank the Office of Naval Research for partial support

of this research. Support from the Dow Chemical Company is also gratefully

acknowledged. Support for the Fourier transform infrared spectrometer was

supplied by the National Institutes of Health, grant number GM 27551.

K ___ ___

;e erences

?.) Z'k'e , i.,_ che . h . " " , * ) ..- en3 , 12 . e cr, 9 , ' ,Ka r'e-, 2.. ese. ,a r~ 2,2

3 ai ey, -. ar.ger, S.. :.hem. Rev., , 2, ¢9

5 Hartley, F.R.; Vezey, P.. Aav. Organo'et. , em., 1977, 7, 139.

6. Grusnka, E.; Kikta, E.J. Anal. Chem., 1977, a9. 100aA.

7. Coliman, J.?. ; Denisevich, P.; <onai, Y., 'arroccc, M.: <cvai. C.;Anson, F.C. J. Am. Chem. Soc., 1980, 102, 6027.

S. Degrand, C.; Miller, L.L. J. Am. Chem. Soc., 1980, 102, 5728.

9. Bolts, J.M.; Bocarsly, A.B.; Palazzotto, M.C.; Walton, E.G.; Lewis, N•S.;Wrighton, M.S. J. Am. Chem. Soc., 1979, 101, 1378.

10. Bookbinder, D.C.; Bruce, J.A.; Doniney, R.N.; Lewis, >I.S.; Irightin, M.S.Proc. Natl. Acad. Sci., U.S.A., 1980, 77, 6280.

11. Sanner, R.D. ; Austin, R.G.; Wrichton, M.S. ; Honnick , 4.. Pittman, C.U Jr.Inorg. Chem., 1979, 18, 928.

12. Reichel, C.L.; Arighton, M.S. Inorg. Chem., 1980. 19, 3858.

13. Kinney, J.8.; Staley, R.H. ; Reichel , C.L. ; ,.rihton, M.S.J. Am. Chem. Soc., 1981, 103, 0000.

ld. Hatchard, C.G.; Parker, C.A. Proc. Roy. Soc. 4, 1956, 235, 518.

15. Chalk, A.J.; Harrod, J.F. J. Am. Chem. Soc., 1967, 39, 1640.

16. Fischer, A.B.; Kinney, J.B.; Staley, R.H.; Wrighton, M.S.J. Am. Chem. Soc., 1979, 101, 6501.

17. Wender, I.; Sternberg, H.W.; Orchin, M. J. Am. Chem. Soc., 1952, 74, 1216.

1g. Figgis, B.N. "Introduction to Ligand Fields", Wiley: New York, 1966,p. 223.

19. Brown, T.L.; Absi-Halabi, M. J. Am. Chem. Soc., 1977, 99, 2982.

20. Reichel, C.L.; Wrighton, M.S. J. Am. Chem. Soc., 1979, 101, 6769.

21. Nesmeyanov, A.N.; Friedlina, R.K.; Chukovskaya, E.C.; Petrova, R.G.;Belyavksy, A.B. Tetrahedron, 1961, 17, 61.

22. Schroeder, M.A.; Wrighton, M.S. J. Organometal. Chem., 1977, 128, 345.

23. Austin, R.G.; Paonessa, R.S.; Giordano, P.J.; Wrighton, M.S.Adv. Chem. Ser., 1978, 168, 189.

Tab e I. Tnf'arec Szec,:ra of S race-Confirec -:eces :": .e-e.n-

Species for Comparison.

Species medium Key IR 3an(s), c- or ,el. abs.,

;EtO) 3SiCo(C0)4 heotane 2105(0.29), 2CdC0.'1), 2025(0.36), 2010 1.SI'

Et3SiCo(CO)4 ,:yclohexane 2089(2740), 2C26(4390), 1995(3580,1992sh (30CC)

Ph3SiCo(CO)4 cyclohexane 2093(3250), 2032(3720), 2003(7000,

KBrb 209 2(5.5)b, 20 30(5.5)b, 1990 (7 .4,7.7)b

Ph3SiCo(CO) 3(P(OPh) 3)c cyclohexane 2049(0.11), 1980(0.32), 1973(I.00)

CEt3SiCo(Ch)3(P(OPh)3 ) nexane 2035(0.35), 1963(0.98), 1963(1.00)

(EtO) 3SiCo(CO) 3(P(OPh) 3)c isooctane 1987(1.00), 1974(0.78)

Co4( CO) 2 cyclohexane 2062(12550), 2053(la,000), 2037(930),2025(1590), 2020sh(930), 1864(4890)

;h 3Si-H ycl ohexane 2129(204)

isooctane 2126(21")

KBrb 2118(!1.05

(EtO) 3Si-H isooctane 2196(135)

neat 2195(--)

Et3Si-H isooctane 2100(149)

C13Si-.4 heptane 2250()Me2CISi-H isooctane 2165(223)

[03Si-H I neat 2246

(from (EtO) 3SiH+H 20) Nujol mull 2253

KBr 2256

[0 3Si-H]n(from CI3SiH + H20) Nujol mull 2252

D -- i-H (from (EtO) 3SiH) KBr

A1203 2257 (H20 added); 2239(dry)

SiO 2 2248 (dry)

QSi Co(CO)4( from (EtC)3SiH) KBr

Al 203(dry) 2109(0.27), 2051(0.58), 2023(1.00)

A AI203(H20 added) 2111(0.40), 2054(0.66), 2024(1.00)

5 S02 2108(0.29), 2048(0.51), 2020(1.00)

_____ ,cznt:. uec)

SDecies 'eciuir .Ke, 3anc, - - .r re. >3.

iC( (? 3KBr

S3 2060(0.21), i95 ., ) 19 85, 0 . ,

2 2C 59 Co) 7. : 0)'~Si -H 'from HSiCl3Q , r

'S = AI203 2253(H20 added/; 2257;drv

-iCo(CO) (from HSiCI )C KBr

S=Al 23 2114(0.30), 2054(g.:9), 2025(7.30)

Si Ie-H from Nle24./ ,<3r

S = AI2 03 2134

=: SiO 2 2148

3-- SiMe2 Co(CO) 4

(from Me2 CSiH) K3 r

S= SiO 2 2100(0.25), 2C39(0.118), 2O9(1.00'

aGenerated in situ by reaction of Co. CO) 8 with (EtO).SiH in heotane.

Underlined values in Parentheses are for "mocel" complexes in <Br

relative to the 2118 cm i' Si-H stretch in PhSiH. O.D. .,as determined to be

proportional to concentration in K3r; cf. Experimental. These

absorptivities are absolute and allow the coverage calculations for

-SiCo(CO)4 and SiH assuming similar absorptivities for the various

bands, Table II.CGenerated in situ by 355 nm irradiation of appropriate R3SiCo(C0)4 in the

presence of 0.01 M P(OPh) 3; cf. ref. 12.

dValues in parentheses are relative absorbance (relative to most intense

band for the substance = 1.00), extinction coefficient in solutions, or

quantitative absorbance relative to Ph3SiH, cf. Experimental and note b

above.

Li 6 LO mil I Ot- 0 C'.~ -

0 0

0~' 0

0 0 N

- 1 4-; 0 '

.- o.~c'~.,~ ~ Cj ~

C.)

-V -i

-n 4-

C"-

A Z

o 'a-~ C\4 - ~ >

- n4-' 04 c

C Z

-z ~ 0

A ~ I 4- 4a) 3j

r- LO - 0 *o '

') 4-1~ 4-

> 0 -'j

A A W

o J 0 m Cc.4-) A 0 1 =I

Aj A

A3 40 MC~0 '

c s- 'Vcu L'V ~ -~W - S-

0b x C- V

o .4-i A -1z -

-w w ea w 0

AO . S-* w~ c

-~u (a - -

-'1 C

- - -o 1- , N 0 - -JZ\J -m. -O i "Z .~ ~ \ \

LL % 4- - *-1

0 xM- C

4-) 4 -J to

o~L -r- >

4-4-1

4-

0S' (V cu .

0 vi 0 -) 0r 0r - to 0 - -

Q~ 0:

to- 0

4C 0. 3~ 0.i

W '(~ I~C-0.. CI (A Aj

LO Lo 'n m 0o Q '0 -a0(n %. k i LOc = cm

~n 01L

4-

4-) 4-

S..J

In

m - r. 0CO 0 -M oMU) Dm 0 N L

. . .- . .

- S4-

Li,~ CAO..

oc C~ CD a

44- CL- L

U-) I-n (aL mL

0 ~ w

to I. *0 * *. .

CU CD S Q -L

0 .0tzCL e

~~. cu %I- -W0~ I ~~I ~ ~ ~to 0 4C0 CL 4-(n 0 . V4

0 4-

.4-m

bd (

.

L.4-) F

V x

rD* CD, 0000

0 O

-oj

0.r ~ 0 Lfl C.i c

C~~( CD CDJ 0 00 0 % .- ~

c 3--v

4-01 - c

Ln - - -. %.0*o CD C)

n 3 4-

L o '.0 r-4 I z.

4-J 0 0 0- 0 0 0 0z

a)0. L a C 4-o e

I .- to

ea .00 *,- el o

M. r- 0*'

.0 013

CL eO 4-U

4...0 I~- L L. -. 1 to - -

m t C). C'. CDJ CDl LA..I .n 00 CO

&- LA

*r Ir 1. e n (Z4- ~ ~~ ~ ~ -j t -4- L - (

= a.- 4j I 0 -C4- .J a'L).0 cn m-

In. .0 (AC U

cu, cu0 J 0 ?aU cc *c* C )(a 4.J.e. *-) 01 - 1 11 40 -- 0 L. e

M 4 r= C C JU- 4- = 0o3'U ~m. CLJ4 4V-.. M fC C~ o

C) CCi

CC\

0) - ) vi VI V)

ci)

4-1 4-1 *a

C-C ULO C -'DJ U.I W~co co VI V n ~ ~ I- - -

0) 0.0. C)7 C)T 0. a)i eu vi 4-1 4=

0.0.>1 vi Z. a . C -

C\1 .0O fl .- en 4-'. 'C (A I- 'a

cc. 41 .6-'4-J 4- cAN ( .O - ' - rU4''

0 L Nj (A C O - -. - N ' - N N C C L-

(a ;Zr- 4-- (A 0-W

4- .41 4'1 4-J cu -4 Ja0

(v =l C)i w--

CL) C. .- M

.4-' C)40A

'a i '4o C)) 4-A '

00f.. q j'o - 'o l

Figure Caotions

Figure 1. infrared spectral changes accompanyinc t'e :ro~ress-ve ,er~yat'Bt. on

of a naked alumina surface. Curve A re-ers to -ne tr- ¢e e7-re reac:'or-.

curve 3 after reaction with HSi~qEt) 3; curve ' after reaction with

HSi(OEt)3 and subsequently Co2(CN)8. ote that left scale is 'or A and B

and right scale is for C.

Figure 2. Representation of Al203"xH2 0 at various staces in zte der-

vatization to confine Co(CO) 4 to the surface.

Figure 3. Infrared spectral changes during the irradiation of a derivatized

alumina surface in the presence of P(OPh)3 . The only CO-containing oroduct-NS

is -SiCo(CO)3(P(OPh) 3, 1995 vs, 1985 vs.

Fiqure 4. Infrared spectral changes in solution during the irradiation of

S iCo(CO) ,, = Al2 in the presence of HSiEt3 . The only detectable

photoproduct in solution is Et3SiCo(CO) 4 , 2027. 1996, 1991 sh cm

Figure 5. Infrared spectral changes of a derivatized AlO, surface

accompanying irradiation in the presence of HSiEt3. -SiCo(CO) 4 (2110, 2050,

2020 cm-l) is removed from the surface while the -SiH concentration

(2250 cm l) increases.

Figure 6. Infrared spectral changes in solution during the irradiation of

a derivatized alumina surface in the presence of l-pentene/HSiEt3. The

solution products are the same as those observed when homogeneous Et3SiCo(CO) 4

(photochemical) or Co2 (CO) 8 (thermal) is used as the cobalt source. The bands

at 2025, 1995, 1991 sh are due to Et3SiCo(CO)4.

A)c.)-OH8BEFORE REACTION

8)©D-OH+HSi(OEt)3 sooctae "'®j-Sl H + 3EtOH

C)©a-SiH+Co2 (CO)IS isooctan e-. ®-S IC (00)4± -t H2

H2 2

0 1. z0

H B

r) c)

(/ 734 2102

8.7- 6.5

8.1 2256 2019 -5.7

2245 2133 2020 1910

WAVENUMBERS (cm'1)

Si Si,

(O)C) C Si ~

si SiSH

H0

®-S ICo (CO) 4 +P(O Ph) 3 otn

(0.01 M)

z

00

155

155

2245 2133 2020 1910

WAVENUMBERS (cm'1)

(D-S~~~~~ I O(O4 ~ t sooctane-1.2- hSC(C)zSIt

9961.0 - 1991 sh

75min hzu

wL 30min hzu~0.8-

0 0.6-

< 20270.475 11lmin hu

30

0.2-

0.02050 2000 1950 1900

WAVENUMBERS (cm'1)

Q - SCo(CO)4+ HSIEt130.S 5 r~

0. ~Iisooctane 0- mm In -

0

0.32 4LUI

Z 75

CO 0.24-A

11 ii

0t

75i 75 minh

0.08

75

2245 2133 2020 - 11WAVE NUMBER .cm

S'6 ©SCO(CO)4 + I-pentene/H Si Et 3 -

1964sh

1.4 1958

,,375min hi)1.2

LUJ G3minhu

1995 -- 32min hu08 1991 sh

M 375

0.6-

0.42025

072 63 0min hi)0.2- 63

32 320 0

0.02000 1950 1900

WAVENUMBERS (crr-')

Sz%472-3/A1 472 7 1:'6,:d;c78u472-606

TECHNICAL REPORT DISTR:DUTION LIST, GEN

No. o.Copies Copies

ffice of 'Nava. Research U.S. Army Research OfficeA:rn: Code g" ,2 Attn: CRD-AA-IP,'C ':orth tuincy Street P.O. Box 1211:-rington, Virginia 22217 Research Triangle Park, N.C. 27709

0'N Branch Office Naval Ocean Systems CenterAttn: D r. George Sandoz Attn: Mr. Joe 'IcCartneyf3A S. Clark S:ree: San Diego, California 92152Chicago, 1llinois 60605 1

Naval Weapons Centera,. C~ffi Attn: Dr. A. B. Amster,

*Chemistry DivisionChina Lake, California 93555

Naval Civil Engineering Laboratory307 ;:estern Regional Office Attn: Dr. R. W. Drisko1C30 East Green Street Port Hueneme, California 93401Pasadena, California 91106

Department of Physics & Chemistry

.R Zast'ern/Central Regional Office Naval Postgraduate SchoolA-:n: Dr. L. H. Peebles Monterey, California 93940

..... ion D

66 Suz=er Street Dr. A. L. SlafkoskyBostzn, assachusa::s 02_ 0o I Scientific Advisor

Commandant of the Marine CorpsDirector, N aval Resear:h Laboratory (Code RD-I)t::!: Co'e A130 Washington, D.C. Z0380

20390Office of Naval Research

The Assis:ant Secretary Attn: Dr. Richard S. Iillerof the Navy (RE&S) 800 N. Quincy Street

Department of the Navy Arlington, Virginia 22217Room _-_736, Pentagon;ashington, ID.C. 20350 1 Naval Ship Research and Development

CenterCommander, Naval Air Systems Command Attn: Dr. G. Bosmajian, AppliedAttn: Code 310C (H. Rosenwasser) Chemistry Division

Department of the Navy Annapolis, Maryland 21401.ashington, D.C. 20360 1

Naval Ocean Systems CenterDefense Technical Information Center Attn: Dr. S. Yamamoto, MarineBuilding 5, Cameron Station Sciences DivisionAlexandria, Virginia 22314 12 San Diego, California 91232

Dr. Fred Saalfeld Kr. John BoyleChemistry Division, Code 6100 Materials BranchNaval &esearch Laboratory Naval Ship Engineering Centerwashington, D.C. 20375 1 Philadelphia, Pennsylvania 19112

T:-CHiNICAL ?~O7DIS7RIBC7ON LIST, GEN

No.Copie s

-";dooh '.arcus- ..ce c- 'aval Research

sz~eftific Liaison. CroupAer~cafl 7=bassy

APO Sani Francisco 96503 1

X'r. J ames KelleyY) :RDC Code 2RC3.A~rapolis, M!aryl'ard 214021

2

A-.0

A' ' : C : 717 8u 4 7 2-608

TFCFNICAL REPORT DISTRIBUTION LIST, 359

No. No.Co-'es Co, 71es

. Pau" Delahav Dr. P. J. HendraChemistry Department of Chemistrv

';ew Yot"'. -niversitv University of SouthhamptorNe- ".'Yor ', Net- York 10003 Southhamoton S09 5NH

United Kingdomnr. F. YeazerPenart"ner.: o' Chemistry Dr. Sam PeroneCase .;es:R.r. -,serve "'niversity Department of ChemistryCleveland, Ohio 4I1106 Purdue University

West Lafayette, Indiana 47907nr. D. \. Sennion

Deoarr.ent of rhemical Ens.ineering Dr. Royce W. MurrayBrihar Youre 'niversitv Department of Chemistry'rovo, Utah B4602 1 University of North Carolina

Chapel Hill, North Carolina 27514r. R. A. Marcus

Denarteien: o4 Chemistry Naval Ocean Systems Center_aifori.a :ns:iute of Technolosv Attn: Technical Library

?asadena, ,.z.ifornia 91125 1 San Diego, California 92152

Or. -. j. Auhorn Dr. C. E. Mueller-e'!- Labera:ories The Electrochemistry Branchv.rray u-'£!, New Jersey 0797k Materials Division, Research

& Technologv DepartmentDr. Ada.- Teler Naval Surface Weapons Center3e! a-cra:ries Ahite Oak Laboratory"'rrav - F1, Yew Jersey 07974 Silver Spring, Maryland 20910

.. . Katan Dr. G. GoodmanLockheed Yissiles & Space Globe-Union Incorporated

Co, Inc. 5757 North Green Bay Avenuep.O. Pox 50& Milwaukee, Wisconsin 53201.unnyvale, California 94088

Dr. J. BoechlerDr. Joseph Sin2er, Code 302-1 Electrochimica CorporationYAA-Lewis Attention: Technical Lihrarv?!Oh" UrookDark Road 2485 Charleston PoadCleveland, Ohio 44135 Mountain View, California 944O

Dr. S. ?rummer Dr. P. P. Schmidt

FIC Incoroorated Department of Chemistry5, Chapel Street Oakland Universityvewton, "fassachusetts 02158 1 Rochester, Michigan 48063

Librar. Dr. F. Richtol. R. Mallory and Company, Inc. Chemistry Department

Northweqt Industrial Park Rensselaer Polytechnic InstituteSurlincron, Massachusetts 01803 Troy, New York 12181

7gu 4 72-608

TFCRNICAL RFPnPT PZTRZRL'TITN LTST, 350

No. No.Copies Copies

Dr. A. ,l, is Dr. R. P. Van DuvneThet-.:r-.: 7enarvtent Deoartment of Chemistryhvers" -'5 "iOnsi Northwestern University

Yadiso,, Wisconsin 53706 1 Evanston, Illinois 60201

"' ViDr. B. Stanley PonsChem. rl v Department of Cheniistrv:as.qa- se'! -'Sr tue University of Alberta

ch nolav Edmonton, Albertaride, uassacnuserts 02139 1 CANADA T6C* 2G2

Larry 7. Plew Dr. Michael J. WeaverNaval "':eaDons Su-oort Center Department of ChemistryCede 30736, Buildin2 29n6 Michigan State UniversityCrane, :ndiana a7522 I East Lansing, Michigan 48824

. 9uv Dr. R. David Rauh

DJ 5T1R) EIC Corporationr- Sreet 55 Chapel Street

t~ashinzgtor, D.C. '0Oia5 1 Newton, Massachusetts 02158

Dr. Aaron c Dr. J. David Margerun'Zo-wn "niversi.-v Research Laboratories D)ivisionDeoar:men: of Che.istrv Hughes Aircraft ComoanvProviden:e, Ahcde island 02192 1 3011 'alibu Canyon Road

Malibu, California 90265Dr. R. C. ChueacekY:,ra.-'ison Conrany Dr. Martin FleischmannHdison Ba:tery Division Department of ChemistryPost Office Box 28 University of SouthamptonBlooefield, New Jersey 07003 1 Southampton 509 5NV Eneland

Dr. A. J. gard Dr. Janet OstervoungUniversi:v of Texas Department of ChemistryDepartment of Chemistrv State University of NewAusti., Texas 78712 1 York at Buffalo

Buffalo, New York 14214Dr. Y. Y.. N/icholson

Electronics ?esearch Center Dr. R. A. Osteryoun.Rockwell International Department of Chemistry337n Miraloma Avenue State University of NewAnahei", California 1 York at Ruffalo

Buffalo, New York 14214Dr. Donald W. ErnstNaval S urface W¢eapons Center Mr. Jares R. ModenCo~e R-33 Naval Underwater SystemsVhite Oak Laboratory CenterSilver qnring, Marvland 20Q10 I Code 3632

Newport, Rhode Island 02840

2

SP472-3/AI5 472:GAN:716:1dc78u472-608

TECHNICAL REPORT DISTRIBUTION LIST, 359

No. No.Co pies Copies

Dr. R. Nowak Dr. John KincaidNaval Research Laboratory Department of the NavyCode 6130 Stategic Systems Project OfficeWashington, D.C. 20375 1 Room 901

Washington, DC 20376Dr. John F. HoulihanShenango Valley Campus M. L. RobertsonPennsylvania State University Manager, ElectrochemicalSharon, Pennsylvania 16146 1 Power Sonices Division

Naval Weapons Support CenterDr. M. G. Sceats Crane, Indiana 47522Department of ChemistryUniversity of Rochester Dr. Elton CairnsRochester, New York 14627 1 Energy & Environment Division

Lawrence Berkeley LaboratoryDr. D. F. Shriver University of CaliforniaDepartment of Chemistry Berkeley, California 94720Northwestern UniversityEva nston, Illinois 60201 1 Dr. Bernard Spielvogel

U.S. Army Research OfficeDr. D. H. Whitmore P.O. Box 12211Department of Materials Science Research Triangle Park, NC 27709 1Northwestern UniversityEvanston, Illinois 60201 Dr. Denton Elliott

Air Force Office ofDr. Alan Bewick Scientific ResearchDeoartment of Chemistry Bldg. 104The University Boiling AFBSouthamocon, S09 5NH England 1 Washington, DC 20332

Dr. A. HimyNAVSEA-5433NC #42541 Jefferson Davis HighwayArlington, Virginia 20362 1

3

r

7Su 72-6,8

Copies Conines

Dr. Y. A. El-Saved Dr. M. Rauhut

Department of Chemistry Chemical Research DivisionUniversity of California, American Cyanamid Company

Los Angeles Bound Brook, New Jersey 08805!.os Angeles, California 90024 1

Dr. J. I. ZinkDr. E. R. Bernstein Department of ChemistryDepa!rment of Cbenis:ry University of California,Colorado State University Los AngelesFort Collins, Colorado 80521 * Los Angeles, California 90024

Dr. C. A. Heller Dr. D. HaarerNaval Weapons Center IBM:ode 6059 San Jose Research CenterChina Lake, California 93555 5600 Cottle Road

San Jose, California 95143Dr. J. R. MacDonaldThemistrv Division Dr. John CooperNaval Research Laboratory Code 6130rode 6110 Naval Research LaboratoryXashington, D.C. 20375 1 Washington, J.C. 20375

Zr. G. B. Schuster Dr. William M. JacksonChemistry eoar:menz Department of Chemistry7ni':arsity of Illinois Howard Universitv

65ana, llinois 6C Washington, DC 20C59

Ar . ',:damson Dr. George Z. Walraffenee;a_:sent of ChemIsty Department of Chemistry

University of Southern Howard UniversityCalifornia Washington, DC 20059

Los Angeles, California 90007 1

M. S." Wrighton

Massachush istitute of

Cidge, Massachuset '11139 1

4V

~DAT