Transition metal mediated surface modification of porous silicon

Article abstract of DOI:10.1016/S0040-4020(01)00366-0

Porous silicon has received considerable attention lately due to its unique optical properties and use in sensor applications. The utility of this material can be further advanced with the aid of new surface chemical modification methods. Here we describe two transition metal mediated reactions that allow for the mild and chemoselective functionalization of porous silicon. The first method utilizes rhodium carbene chemistry to modify porous silicon surfaces with esters and amides. The second technique exploits a pt-based Karstedt's catalyst to affect the hydrosilylation of porous silicon with alkenes. The mildness and selectivity of the hydrosilylation is highlighted by modification of n-type porous silicon matrix with an 8-residue cyclic peptide.

Full text of DOI:10.1016/S0040-4020(01)00366-0

TETRAHEDRON
Pergamon
Tetrahedron 57 :2001) 5131±5136
Transition metal mediated surface modi®cation of porous silicon
Alan Saghatelian, Jillian Buriak, Victor S. Y. Lin and M. Reza Ghadirip
Departments of Chemistry and Molecular Biology and the Skaggs Institute for Chemical Biology,
The Scripps Research Institute, La Jolla, CA 92037, USA
Dedicated to Professor BarryM. Trost on the occasion of his 60th birthday
Received 5 February2001; revised 12 March 2001; accepted 13 March 2001
AbstractÐPorous silicon has received considerable attention latelydue to its unique optical properties and use in sensor applications. The
utilityof this material can be further advanced with the aid of new surface chemical modi®cation methods. Here we describe two transition
metal mediated reactions that allow for the mild and chemoselective functionalization of porous silicon. The ®rst method utilizes rhodium
carbene chemistryto modifyporous silicon surfaces with esters and amides. The second technique exploits a pt-based Karstedt's catalyst to
affect the hydrosilylation of porous silicon with alkenes. The mildness and selectivity of the hydrosilylation is highlighted by modi®cation of
n-type porous silicon matrix with an 8-residue cyclic peptide. q 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
mechanism byattacking a surface exposed silicon atom
and breaking a Si±Si bond in the process. While effective
in functionalizing the surface of porous silicon these
methods do have some drawbacks. In reactions with alco-
hols, an alkylsilylether moiety is created which is moisture
sensitive and not stable in aqueous environment. On the
other hand, reactions of alkyllithiums and Grignard reagents
with freshly etched p-Si yields alkylsilanes that are typically
stable under a varietyof conditions. However, these hard
Porous silicon, prepared from the electrochemical etching of
single-crystals of p- or n-type silicon, is a versatile material
with utilityin optical, electronic, and sensor applications. 1±5
Recent advances in the design of porous silicon-based inter-
ferometric biosensors and desorption ionization :DIOS)
mass spectrometryhave highlighted the need for selective
surface chemical modi®cation methods that can meet the
needs of a given application.6,7 Here we describe two tran-
sition metal mediated surface functionalization schemes
that are able to modifythe porous silicon surface through
mild and highlychemoselective processes.
Under certain conditions electrochemical etching of single-
crystal p- or n- type silicon wafers produces a microporous
material dubbed porous silicon :p-Si). The p-Si surface is a
hydride-terminated material composed of three types of
silicon hydride groups, Si±Hx :xˆ1, 2 or 3) as identi®ed
bycharacteristic IR absorbances :Fig. 1:a)). 8,9 Modi®cation
of the porous silicon surface with organic compounds can be
monitored bychanges in the IR spectrum due to the sepa-
ration of the porous silicon spectrum from that of most
common organic compounds. The chemistryused to intro-
duce molecules onto the surface of porous silicon depends
heavilyon the class of molecules employed and can exploit
the electrophilicityof Si±Si
9±11
centers as well as the
reactivityof Si±H bonds.5,12±14 Alcohols, Grignard, and
x
alkyllithium reagents are the most commonly employed
nucleophilic reagents for surface modi®cation of porous
silicon.9±11 These reagents react through an Sn2 type
Keywords: porous silicon; rhodium mediated carbene insertion; platinum
hydrosilylation; surface modi®cation.
p
Corresponding author. Tel.: 11-858-784-2700; fax: 11-858-784-2798;
e-mail: ghadiri@scripps.edu
Figure 1.
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020:01)00366-0

5132
A. Saghatelian et al. / Tetrahedron 57 ,2001) 5131±5136
reactions with the C±H, C±C, or Si±C bonds of the triethyl-
silane moiety. This favorable chemoselectivity makes this
reaction an attractive candidate for developing a new
method for surface modi®cation of porous silicon.
A freshlyetched p-type porous silicon wafer :Fig. 1:a))
placed in a benzene solution containing catalytic quantities
of Rh2:OAc)4 was exposed to excess ethyldiazoacetate.
Immediate evolution of N2 ensued signifying formation of
the metallocarbene intermediate :Scheme 1). The reaction
mixture was then heated until gas evolution ceased
:~10 min). The porous silicon chip was then removed
from the reaction mixture, washed, and analyzed by FT-
IR spectroscopy. The IR spectrum of the p-Si matrix showed
appearance of new peaks characteristic of ethylester func-
tionality:peaks at 2900 cm 21 belonging to the aliphatic CH
stretching and a strong absorbance at 1760 cm21 corre-
sponding to the carbonyl stretch) :Fig. 1:b)). Reactions
performed in the absence of catalyst or in the absence of
the diazo compound displayed no surface modi®cation. The
modi®ed wafer was then subjected to sonication in THF for
30 min with no change in the IR spectrum thus ruling
out physisorbtion and supporting covalent modi®cation of
the p-Si surface.
Scheme 1.
nucleophiles place a limit on the types of functionality that
can be incorporated onto the p-Si surfaces. For example,
functional groups such as esters and amides are not com-
patible with manyof these reagents. The modi®cation
of porous silicon through the Si±Hx bonds provides an
orthogonal method for surface modi®cation. This strategy
has been demonstrated in photo-induced hydrosilylation of
alkenes to give patterned surfaces.14 Other studies highlight
the use of Si±H mediated reactions in providing stable
surface modi®cations with manydifferent classes of
molecules.5,12,13
Although disappearance of Si±Hx bands in the IR spectrum
of the reacted p-Si matrix is indicative of insertion into the
silicon hydride bonds, a priori additional insertions into
surface exposed Si±Si bonds can not be ruled out. The
work of Hatanaka and coworkers, provides some important
experimental precedent regarding the relative reaction rates
of the C±H vs Si±Si bond insertion during a rhodium
carbene insertion reactions.17 In their studyvarious sila-
cycloalkanes were probed in Rh2:OAc)4 catalyzed reactions
with ethyl- or tert-butyl diazoacetate. Two substrates used
in their studyare of particular interest because of the
presence of a Si±Si bond within a ®ve-member ring.
When these disilacyclopentanes were reacted with a diazo
compound in the presence of Rh2:OAc)4 the onlyproduct
detected was that of C±H insertion. This suggests that inser-
tion into a C±H bond is more favorable than insertion into a
Si±Si bond.17 On the other hand, Doyle has shown that
insertion into the Si±H bond is kineticallymore favorable
than insertion into C±H bond.16 Collectivelythese examples
indicate that carbene insertion into the Si±H bond is much
more favorable than reactions with Si±Si bond. Therefore,
byanalogyin the above rhodium-catalyzed p-Si modi®-
cation reactions, Si±Hx insertions are expected to be
the dominant reaction pathway. To further support this
hypothesis an oxidized porous silicon surface lacking
exposed Si±Si bonds was employed in a similar modi®-
cation reaction. Surface oxidized p-Si wafer was produced
by®rst exposing a freshlyetched p-Si substrate to bromine
gas followed byexposure to air. This method produces a
thin layer of silicon oxide at the p-Si surface by reacting
with and thus removing all surface exposed Si±Si bonds
while still retaining silicon hydride bonds in the form of
±OSi±Hx groups.7 Exposure of the oxidized p-Si wafers to
In this studywe document the utilityof two transition metal
mediated processes for the chemical modi®cation of hy-
dride-terminated porous silicon matrices. The ®rst reaction
exploits rhodium catalyzed metallocarbene insertion reac-
tions of diazo compounds into exposed silicon hydride
bonds at the porous silicon surface. The second method
employs a platinum based hydrosilylation reaction with
alkenes. Both methods create stable carbon silicon bonds
and are highlychemoselective. These mild chemical modi-
®cation methods allow new functionalityto easilybe placed
on the p-Si surface as highlighted bythe incorporation of a
peptide onto the silicon surface.
2. Results and discussion
2.1. Rhodium mediated carbene insertions
In solution, carbene insertion reactions provide an ef®cient
and selective method for modi®cation of silanes. Exposure
of a diazo compound to Rh2:OAc)4 produces rhodium
carbene species that can insert into the Si±H bond to give
a new silicon-carbon bond.15 Doyle demonstrated that in the
presence of Rh2:OAc)4 and triethylsilane, ethyldiazoacetate
is converted cleanly into 2-:triethylsilyl)-ethylacetate in
94% yield.16 This highlyef®cient reaction is also very
selective with there being no side products stemming from

A. Saghatelian et al. / Tetrahedron 57 ,2001) 5131±5136
5133
Scheme 2.
the Rh-catalyzed carbene insertion reactions gave a product
with characteristic IR bands at 2985 cm21 corresponding to
the alkyl group of the ester as well as the carbonyl stretch at
1743 cm21 :Fig. 1:c)). Therefore, the similar reactivityof
the oxidized substrates lacking surface exposed Si±Si bonds
lends further support to the Si±H insertion as the only
mechanism in surface derivatization.
alkenes as starting reagents. Out of a number of different
catalysts known to promote hydrosilylation reactions in
solution, onlyKarstedt's catalyst gave the desired reactivity
pattern for surface modi®cation of p-Si :Scheme 3).19,20
Karstedt's catalyst is a highly active colloidal Pt-based
species that is commonlyused in mild and chemoselective
hydrosilylation reactions. The utility of this method in
modi®cation of freshlyetched porous silicon was studied
using a 3.5% Pt solution of Karstedt's catalyst in the
presence of 1-octene. The IR spectra of porous silicon
surfaces, after the reaction in benzene at 458C for 20 h
clearlyindicated incorporation of octly moieties. New
peaks corresponding to the C±H stretching frequencies
for octyl methylenes and methyl group now appeared at
2850±2960 cm21 as well as alkane bending modes at
1466 and 1387 cm21 :Fig. 2:a)). The ole®nic stretch
of 1-octene, which appears at 1670 cm21, is not present,
indicating that the alkene is not simplyphysisorbed on
the surface. Control experiments demonstrated that the
surface remains essentiallyunchanged in absence of added
catalyst.
The scope and utilityof the above method was demonstrated
using different diazoacetamide starting reagents prepared
according to the methods of Ouihia and coworkers :Scheme
2).18 For example, the reaction between n-butyldiazo-
acetamide and p-Si provided the expected surface bound
butyl amide moiety as evidenced by IR spectroscopy
:appearance of new stretching frequencies at 2964 cm21
belonging to the butyl group, as well as amide I and II
stretches at 1659 and 1539 cm21) :Fig. 1:d)). Therefore,
transition metal-mediated carbene insertion provides a
facile route to modifyp-Si matrices with amide and ester
functionalitythat is largelyinaccessible using most of the
previouslyavailable methods.
Concomitant with hydrosilylation was a substantial amount
of oxidation as evidenced bythe increase in intensityof the
broad Si±O absorption centered at 1070 cm21 :Fig. 2:a)).
Interestingly, if hydrosilylation was carried out in strict
absence of air under an N2 atmosphere, the porous silicon
surface remained unchanged as neither oxidation nor octyl
incorporation was noted, indicating that molecular oxygen
was essential for the reaction to proceed. The rate of surface
oxidation was dramaticallyaccelerated in the presence of
the Pt catalyst as has also been noted for copper salts.21
Oxidation of the surface silicon atoms to form OSi±H
species, increases the reaction rate because hydrosilylation
of the oxygen backbonded silicon hydrides occurs preferen-
tiallyover nonoxgyen bound Si±H functionalities. This
trend is consistent with what has been seen with homoge-
neous silanes where oxygen back bonded silanes :i.e.
HSi:OEt)3) are much more reactive than C- or Si-bound
silanes :i.e. HSiMe3).20
2.2. Platinum mediated hydrosilylation
The second method developed for the modi®cation of n-type
p-Si utilized Pt-mediated hydrosilylation reaction using
The high reactivityof oxygen back-bonded SiH function-
alities was veri®ed upon thermal oxidation of the porous
silicon surface. Surface oxidation for 5 days at a 1008C in
air results in the quantitative transformation of silicon bound
SiH groups to OSiH surface species as determined byIR.
When this surface was submitted to Pt catalyzed hydrosilyl-
ation conditions using 1-octene in benzene with Karstedt's
catalyst solution, either in the presence or absence of air :N2
Scheme 3.

5134
A. Saghatelian et al. / Tetrahedron 57 ,2001) 5131±5136
Figure 2.
atmosphere) at 458C, octyl incorporation occurred and
selective disappearance of one of the oxygen back-bonded
OSi±H absorption at 2202 cm21 was observed over a 3 day
period. The oxygen back-bonded OSi±H absorption at
2254 cm21 remains essentiallyunchanged. The band at
2254 cm21 has been assigned to O3Si±H :silicon hydride
back-bonded to three oxygen atoms) while the absorption
at 2202 cm21 has been assigned to O2SiSi±H in which the
silicon hydride is back-bonded to one silicon and two
in a less stericallyhindered position on the porous surface
than O3Si±H, allowing convergence of alkene, Pt-colloid
and Si±H. Molecular O2 has been shown to be important
in stabilizing Pt colloids20 but in these reactions on porous
silicon, it is not essential if the surface is previously
oxidized.
Similar to its molecular equivalent, hydrosilylation of
surface bound OSi±H species should involve formation of
silicon-carbon bonds. Unfortunately, the Si±C vibrational
mode could not be seen in the IR due to an intense Si±O
band which buried all other stretching frequencies in that
area. Since the proposed silicon-carbon bond could not be
identi®ed spectroscopically, the chemical stability of the
surface-bound species was examined. Si±C bonds should
be highly stable to hydrolysis and alcoholysis, as compared
with SiO±C bonds that undergo rapid exchange. Indeed, the
octyl monolayers on porous silicon are stable inde®nitely in
air and the octyl-decorated surface remains unchanged for at
least 72 h in neat MeOH at room temperature. In contrast, a
porous silicon surface coated with SiO:CH2)7CH3 groups
resulted in total exchange of octyl groups for methyl groups
in 10 min of soaking in methanol. While the silicon±carbon
bond was not observed through direct methods, onlygroups
covalentlyattached to silicon or oxidized silicon surfaces
through Si±C bonds have been shown to have such
stability.24
oxygen atoms.22,23 It maybe that the O SiSi±H groups are
2
In an attempt to improve the reaction we postulated that
precoating the oxidized surface with more accessible silicon
hydrides should allow the catalyst to function more
ef®ciently. This process began by preparing thermally
oxidized wafers byheating wafers at 100 8C for 5 days and
then soaking them for 16 h in neat dimethylethoxysilane,
which coated the surface with a layer of H±SiMe2OSi
groups.
Figure 3.

A. Saghatelian et al. / Tetrahedron 57 ,2001) 5131±5136
5135
the silicon hydride bonds present on the surface of porous
silicon. The silicon carbon bonds created through these
reactions render the functionalized surface highlystable.
Furthermore, the mild conditions applied permit the use of
entirelynew classes of functional groups. Together these
two attributes are expected to allow the newlyfunctional-
ized porous silicon surfaces to be used in novel ways open-
ing up the door for its applications in analytical and
materials settings.
3. Experimental
3.1. General
FT-IR spectra were collected on a Nicolet Mangna Series
550 spectrometer in transmission mode. A 4 cm21 reso-
lution was used with 128 scans collected for each spectrum.
All reagents were purchased from Aldrich and used without
further puri®cation. For the carbene insertion porous silicon
was made from the electrochemical etch of p-type silicon
wafers :3 V-cm) under a current of 33 mA/cm2 for 20 min
in a solution of HF/H2O/Ethanol :1/1/2) yielding a surface
area of about 50 cm2. For hydrosilylation reactions n-type
silicon wafers :0.5 V-cm) were etched under a current of
88 mA/cm2 for 3 min in a solution of HF/H2O/Ethanol
The infrared spectrum showed new silicon hydride absorp-
tion bands at 2151 cm21 stemming from the H±SiMe2OSi
group :Fig. 2:b)). Submission of this hydride-coated porous
silicon surface to hydrosilylation conditions in the presence
of 1-octene for 3 days resulted in octyl incorporation and
disappearance the H±SiMe2OSi groups as evidenced byIR
spectroscopy. Additionally, the hydride absorption at
2202 cm21, from the O2SiSi±H species, is also consumed
during the reaction :Fig. 2:c)). Thus, higher level of octyl
group incorporation can be achieved on surfaces that have
silicon hydride groups added subsequent to the HF etch.
Additionally, a second porous silicon surface devoid of
anynative Si±H bonds was prepared byheating at 600 8C
for 24 h in air. This treatment eliminates all native silicon
hydrides and leaves a fairly thick oxide layer. Deposition of
H±SiMe2OEt onto this porous silicon wafer results in only
one Si±H absorption at 2163 cm21 and a corresponding
Si±H bending mode at 910 cm21 in the IR spectrum. Hydro-
silylation of this surface with 1-octene was more ef®cient
with the reaction being essentiallycomplete in less than
24 h as indicated bythe disappearance of the Si±H absorp-
tion and the appearance of peaks corresponding to the C±H
stretching and C±H bending vibrations of the octyl group.
2
:1/1/2) while being illuminated bya 30 mW/cm white
light to afford a surface area of about 50 cm2. These
processes yielded a 1.2 cm2 circular area of etched porous
silicon. Karstedt's catalyst was purchased as a 3±3.5% solu-
tion in xylenes :Petrarch).
3.2. Rhodium carbene insertion reactions
A freshlyetched wafer was placed into a round bottom ¯ask
with benzene :10 mL) and Rh2:OAc)4 :0.4 mg). The reac-
tion was initiated byaddition of the diazo compound {ethyl
diazoacetate :1 mL, 9.5 mM)} to the solution. The reaction
was detected bythe evolution of nitrogen. The reaction
mixture was then heated for about 10 min or until gas evo-
lution ceases. The wafer was then immediatelyremoved
washed with ethanol and methylene chloride and dried
under a gentle stream of nitrogen prior to analysis by IR.
With the reaction conditions optimized with 1-octene we
looked to test the chemoselectivity of the hydrosilylation.
Hydrosilylation of the cyclic peptide cyclo[:-Nallyl-l-Ala-d-
Phe-.l-Ala-d-Phe)2-]25 :Fig. 3) which contains 2 allyl
functionalities proceeds rapidlyand is essentiallycomplete
in 24 h in air. The amide I and II stretching vibrations are
observed at 1661 and 1518 cm21 respectivelyalong with
the NH absorption at 3317 cm21 :Fig. 2:d)). The surface
coverage of the peptide is approximately15±20% based
on the absorbance of the amide I and II stretching vibrations.
The peptide cannot be washed off with excess volumes of
anysolvents in which the peptide is soluble such as DMF,
CH2Cl2, EtOH or MeOH. Only24% HF in EtOH wipes the
surface clean since the surface oxide is unstable to HF. This
substrate shows that the chemoselectivityof this reaction
allows the modi®cation of the surface with completely
new classes of molecules. Furthermore, this modi®cation
of the surface with this cyclic peptide opens up the possi-
bilityof self-assembling peptide nanotubes onto the silicon
surface for ion sensing applications.26±29
3.3. Oxidation of the porous silicon substrate
A freshlyetched wafer was placed under an inert atmos-
phere of Argon. Bromine :100 mL) was added into the ¯ask
and the ¯ask sealed. After 1 h, the wafer was removed and
exposed to ambient air.
3.3.1. Preparation of diazoacetamides. Succinimidyl
diazoacetate :92 mg, 0.5 mmol, 1 equiv.) was dissolved in
THF :5 mL). Neat butyl amine :98 mL, 1 mmol, 2 equiv.)
was added to the solution via syringe in one portion. After
1 h, the solution was concentrated in vacuo. The crude
material was puri®ed by¯ash chromatography:EtOAc) to
yield 49 mg :69%) of a pale yellow solid. IR :CDCl3): 3305,
3099, 2962, 2933, 2875, 2104, 1726, 1618, 1552, 1524,
1
1383, 916 cm21. H NMR :200 MHz, CDCl3) d: 0.847
In summary, we have described two transition metal
mediated reactions that take advantage of the reactivityof
:3H, t, Jˆ5.736 Hz), 1.31 :m, 4H), 3.19 :2H, m), 6.34
:1H, s). 13C NMR :95 MHz, CDCl3) d: 13.60, 19.89,

5136
A. Saghatelian et al. / Tetrahedron 57 ,2001) 5131±5136
3. Doan, V. V.; Sailor, M. J. Appl. Phys. Lett. 1992, 60, 619±620.
4. Harper, J.; Sailor, M. J. Anal. Chem. 1996, 68, 3713±3717.
5. Stewart, M. P.; Buriak, J. M. Adv. Mater. 2000, 12, 859±869.
6. Lin, V. S.-Y.; Motesharei, K.; Dancil, K.-P. S.; Sailor, M. J.;
Ghadiri, M. R. Science ,Washington, DC) 1997, 278, 840±
843.
25.41, 31.83, 39.60, 46.72, 165.93. Mass calcd for
C6H11ON3: 142.0980; Found: 142.0986, 4.2 ppm.
3.3.2. n-Dodecyldiazoacetamide. IR :CDCl3): 3300, 3093,
1
2954, 2920, 2850, 2104, 1709, 1612, 1549, 1396 cm21. H
NMR:200 MHz, CDCl3) d: 0.838 :3H, t, Jˆ5.16 Hz), 1.21
:18H, m), 1.46 :2H, m), 3.21 :1H, d, 4.37 Hz), 4.77 :1H, s),
5.61 :1H, s). 13C NMR :95 MHz, CDCl3) d: 13.98, 22.57,
26.83, 29.24, 29.51, 29.80, 31.80, 39.95, 46.71, 165.83.
Mass calcd for C14H27ON3: 254.2232; Found: 254.2223,
3.5 ppm.
7. Wei, J.; Buriak, J. M.; Siuzdak, G. Nature ,London) 1999, 399,
243±246.
8. Lee, E. J.; Ha, J. S.; Sailor, M. J. J. Am. Chem. Soc. 1995, 117,
8295±8296.
9. Kim, N. Y.; Laibinis, P. E. J. Am. Chem. Soc. 1997, 119,
2297±2298.
3.3.3. n-Benzyldiazoacetamide. IR :CDCl3): 3294, 3099,
10. Kim, N. Y.; Laibinis, P. E. J. Am. Chem. Soc. 1998, 120,
4516±4517.
2927, 2110, 1610, 1558, 1389, 1358 cm21 1H NMR
.
:200 MHz, CDCl3) d: 4.35 :2H, d, Jˆ4.61 Hz), 4.71 :1H,
s), 5.70 :1H, m), 7.22 :5H, m). 13C NMR :95 MHz, CDCl3)
d: 43.73, 47.06, 126.25, 127.37, 127.45, 128.59, 166.04.
Mass calcd for C9H9ON3: 176.0824; Found: 176.0831,
4 ppm.
11. Song, J. H.; Sailor, M. J. J. Am. Chem. Soc. 1998, 120, 2376±
2381.
12. Gurtner, C.; Wun, A. W.; Sailor, M. J. Angew. Chem., Int. Ed.
1999, 38, 1966±1968.
13. Bateman, J. E.; Eagling, R. D.; Worrall, D. R.; Horrocks, B. R.;
Houlton, A. Angew. Chem., Int. Ed. 1998, 37, 2683±2685.
14. Buriak, J. M.; Allen, M. J. J. Am. Chem. Soc. 1998, 120,
1339±1340.
3.3.4. n-Morpholinediazoacetamide. IR :CDCl3): 3481,
3082, 2970, 2900, 2860, 2110, 1603, 1432, 1240, 1117,
1
1039 cm21. H NMR:200 MHz, CDCl3) d: 3.43 :4H, m),
15. Landais, Y.; Planchenault, D. Tetrahedron Lett. 1994, 35,
4565±4568.
3.59 :4H, m), 4.96 :1H, s). 13C NMR :95 MHz, CDCl3) d:
44.00, 46.37, 66.51, 164.99. Mass calcd for C6H9O2N3:
156.0773; Found: 156.0766, 4.5 ppm.
16. Bagheri, V.; Doyle, M. P.; Taunton, J.; Claxton, E. E. J. Org.
Chem. 1988, 53, 6158±6160.
17. Hatanaka, Y.; Watanabe, M.; Onozawa, S.-y.; Tanaka, M.;
Sakurai, H. J. Org. Chem. 1998, 63, 422±423.
18. Ouihia, A.; Rene, L.; Guilhem, J.; Pascard, C.; Badet, B.
J. Org. Chem. 1993, 58, 1641±1642.
3.4. Platinum catalyzed hydrosilylation reactions
A freshlyetched porous silicon wafer was washed
thoroughlywith EtOH, CH Cl2 and dried under a stream
2
19. Cai, G. P.; Weber, W. P. Macromolecules 2000, 33, 8976±
8982.
of N2. To this wafer was added 1 mL of benzene, 20 mL
of 1-octene and 1 mL of Karstedt's catalyst solution
:3±3.5% solution in xylenes). This mixture was then heated
at 458C until the reaction was complete as determined byIR
spectroscopy.
20. Lewis, L. N. J. Am. Chem. Soc. 1990, 112, 5998±6004.
21. Graef, D.; Grundner, M.; Muehlhoff, L.; Dellith, M. J. Appl.
Phys. 1991, 69, 7620±7626.
22. Dubin, V. M.; Ozanam, F.; Chazalviel, J.-N. Thin Solid Films
1995, 255, 87±91.
23. Tsu, D. V.; Lucovsky, G.; Davidson, B. N. Phys. Rev. B 1989,
40, 1795.
Acknowledgements
24. Pomerantz, M.; Segmueller, A.; Netzer, L.; Sagiv, J. Thin
Solid Films 1985, 132, 153±162.
Support from the Of®ce of Naval Research :MURI95:
N00014-95-1-1293) is gratefullyacknowledged. J. M. B.
thanks the Natural Science and Engineering Research
Council :NSERC) of Canada for a post-doctoral fellowship.
We would like to thank our colleague Professor M. J. Sailor
for helpful assistance and stimulating discussions.
25. Clark, T. D.; Buriak, J. M.; Kobayashi, K.; Isler, M. P.;
McRee, D. E.; Ghadiri, M. R. J. Am. Chem. Soc. 1998, 120,
8949±8962.
26. Hartgerink, J. D.; Granja, J. R.; Milligan, R. A.; Ghadiri, M. R.
J. Am. Chem. Soc. 1996, 118, 43±50.
27. Hartgerink, J. D.; Clark, T. D.; Ghadiri, M. R. Chem.-Eur. J.
1998, 4, 1367±1372.
References
28. Ghadiri, M. R.; Kobayashi, K.; Granja, J. R.; Chadha, R. K.;
McRee, D. E. Angew. Chem., Int. Ed. Engl. 1995, 34, 93±95.
29. Ghadiri, M. R.; Granja, J. R.; Milligan, R. A.; McRee, D. E.;
Khazanovich, N. Nature ,London) 1993, 366, 324±327.
1. Canham, L. T. Appl. Phys. Lett. 1990, 57, 1046±1048.
2. Canham, L. T. Appl. Phys. Lett. 1993, 63, 337±339.