Modification and chemical transformation of Si(1 1 1) surface

Article abstract of DOI:10.1016/j.jorganchem.2006.09.048

Modification of hydrogen-terminated Si(1 1 1) surfaces by hydrosilylation of activated alkenes and further chemical transformation of the modified surfaces is reported. A Si(1 1 1)-H surface was reacted with activated alkenes such as acrylate esters, acry

Full text of DOI:10.1016/j.jorganchem.2006.09.048

Journal of Organometallic Chemistry 691 (2006) 5809–5824
www.elsevier.com/locate/jorganchem
Modification and chemical transformation of Si(111) surface
a,b
Yang Liu , Shoko Yamazaki a,b,*, Suguru Izuhara a
a
Department of Chemistry, Nara University of Education, Takabatake-cho, Nara 630-8528, Japan
The Core Research for Evolutional Science and Technology, The Japan Science and Technology Agency (JST), Kawaguchi, Saitama, Japan
b
Received 26 July 2006; received in revised form 15 September 2006; accepted 18 September 2006
Available online 30 September 2006
Abstract
Modification of hydrogen-terminated Si(111) surfaces by hydrosilylation of activated alkenes and further chemical transformation of
the modified surfaces is reported. A Si(111)–H surface was reacted with activated alkenes such as acrylate esters, acrylonitrile, and maleic
anhydride under mild conditions to give modified surfaces with terminal functional groups. A modified surface with a terminal ester
group was reduced by LiAlH4 to give a hydroxy-terminated surface, and the hydroxy-terminated surface was transformed to a
2 2 2
bromo-terminated surface. XPS analysis revealed that the brominated surface („Si(111)–CH CH CH Br) had 32% coverage with
the 3-bromopropyl group. Ester and amide formation reactions were carried out on hydroxy- and carboxy-terminated Si surfaces by
reaction with tert-butoxycarbonyl glycine, glycine tert-butyl ester, 2,2,2-trifluoroethanol and 4-trifluoromethylbenzyl alcohol in the pres-
ence of carbodiimide. XPS characterization indicated that the esters and amide were successfully formed with coverage ranging from 16%
to 58%. Coverage ratios of octadecyl ester modified surfaces were also estimated by combination of surface reduction and gas chroma-
tography analysis to be 25–35%.
ꢀ
2006 Elsevier B.V. All rights reserved.
Keywords: Hydrosilylation; Silicon surface; Chemical transformation of modified surfaces; Acrylate esters; Propiolate esters
1
. Introduction
Grignard or organolithium reagents [3]. Significant
improvement of Si surface modification under milder con-
ditions suitable for binding labile or bioactive functional
groups onto the silicon surface has been achieved. Zuilhof
et al. reported a mild attachment of saccharides onto a
hydrogen terminated silicon surface by visible light [7].
We have recently reported a mild modification of hydrogen
terminated Si(111) surfaces by activated alkynes at room
temperature without addition of catalyst [8].
In this study, we investigated the modification of a
hydrogen-terminated Si(111) surface with various acti-
vated alkenes and further transformation of the surface
functionalities of the organic monolayer modified surfaces.
By further chemical transformation, surfaces with new
reactive functional terminal groups were obtained without
breaking the Si–C bond linkage or the Si–Si back-bond.
Although modification techniques have been rapidly
developed, quantitative analysis especially estimation of
the monolayer coverage of organic groups bonded on sili-
con surfaces is still a crucial subject. Quantitative infrared
Organic modification of silicon surfaces through forma-
tion of Si–C covalent bonds has been intensively studied
1]. Because the modified surfaces usually possess enhanced
[
oxidative stability and novel functionalities and properties
compared with a hydrogen-terminated Si surface [2,3], they
are considered to be potential semiconducting substrates
for fabrication of electronic devices or biosensors [4]. Var-
ious modification methods based on wet-chemistry have
been explored to form Si–C bonded organic monolayers
on silicon surfaces, such as hydrosilylation of alkenes or
alkynes initiated by a free-radical initiator [2], hydrosilyla-
tion activated by thermal or UV irradiation [2,5], hydrosi-
lylation mediated by Lewis acids [6], and alkylation with
*
Corresponding author. Address: Department of Chemistry, Nara
University of Education, Takabatake-cho, Nara 630-8528, Japan. Tel.:
+
81 742 9193; fax: +81 742 27 9289.
E-mail address: yamazaks@nara-edu.ac.jp (S. Yamazaki).
0
022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.09.048

5810
Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
spectroscopy (IR) [6,9] and X-ray photoelectron spectros-
copy (XPS) [5,10] characterization are most often used to
estimate the monolayer coverage of modified surfaces.
However, the accuracy of XPS analysis in estimation of
coverage of an organic monolayer is usually decreased by
surface contaminated carbon. In order to prevent compli-
cation by surface contamination, we have previously used
fluorine or nitrogen containing alkynes to modify the Si
surface and estimated the monolayer coverage by quantita-
tive analysis of the relative intensities of fluorine or nitro-
gen photoelectrons to silicon photoelectrons [8].
Determination of surface coverage by a stepwise procedure
including detachment of fluorescence probes from the sur-
face monolayer and quantitative fluorescence analysis was
recently reported [11]. In this report, we have attempted
determination of the coverage ratios of ester modified sur-
faces by quantitative analysis of the corresponding alcohols
reductively liberated from ester monolayers, using gas
chromatography.
In this study, we have investigated the reactions of the
model tris(trimethylsilyl)silane (1) and the Si(111)–H sur-
face with various activated alkenes such as acrylate esters,
acrylonitrile and maleic anhydride (4). Reactions of silane
1 and alkenes 2 and 4 gave b-silylated addition products
3 and 5 (Scheme 1). Compared to the corresponding propi-
olate esters and propionitrile [8], acrylate esters and acrylo-
nitrile showed relatively lower reactivity with silane 1.
Reactions of alkenes 2b–c and 2e–f gave moderate yields
of hydrosilylation products, and reaction 2b and 2d–f
required a slightly higher reaction temperature and pro-
longed reaction times (Table 1).
This hydrosilylation is considered to proceed through a
radical chain mechanism [12]. The initiation step included
an autoxygenation reaction of silane group by adventitious
in situ oxygen and a hydrogen abstraction reaction to gen-
erate a silicon radical. In general, alkynes are less reactive
than alkenes in radical addition reactions [13]. Rate con-
stants for the reaction of Et Si radicals to alkynes com-
3
pared to alkenes also shows that alkynes are slightly less
reactive [14]. Possible radical intermediates A and B were
calculated by UB3LYP/6-31G* calculations [15] (Scheme
2 and Fig. 1). A has a quasi-linear (\C@C–C is 158ꢁ)
and slightly more compact structure than B. The highly
bulky (Me Si) Si group [16] may destabilize the slightly ste-
2
. Results and discussion
2.1. Modification of a hydrogen-terminated Si(111) surface
with activated alkenes
3
3
We have reported that a hydrogen-terminated Si(111)
surface was easily modified by activated alkynes such as
propiolate esters and propionitrile. This surface modifica-
tion was directed by model reactions of activated alkynes
and tris(trimethylsilyl)silane (1), a molecular model of the
Si–H group on hydrogen terminated Si(111) surface [8,12].
rically hindered radical B compared to radical A. The
explanation is supported by the calculated free energy dif-
ferences DG298 of A (ꢀ7.8 kcal/mol) [17] and B (ꢀ0.5 kcal/
mol) referred to (Me Si) Si and methyl propiolate
3
3
CO Me
2
(
Me Si) Si
3 3
CO Me
Si(SiMe3)3
+
2
H
neat
(
Me Si) Si
A
(
Me Si) Si
H
H
+
X
3
3
3
3
X
[
Δ
G = 0 kcal/mol]
[
Δ
G = -7.8 kcal/mol]
1
2
3
For X, see Table 1
H
Si(SiMe3)3
+
CO Me
(Me
3
Si) Si
3
2
CO Me
(
Me Si) Si
2
3
3
CH Cl
2
2
H
H
(
Me Si) Si
+
2a
Δ
[ G = 0 kcal/mol]
3
3
O
O
O
O
O
B
O
5
1
4
[ G = -0.5 kcal/mol]
Δ
Scheme 1.
Scheme 2.
Table 1
Reaction of Tris(trimethylsilyl)silane (1) and alkenes 2
b
Entry
Substrate
X
Reaction conditions
Product (yield , %)
a
1
2
3
4
5
6
7
a
2a
2b
2c
2d
2e
2f
4
X = CO
X = CO
X = CO
X = CO
X = CO
2
Me
CH
CH
r.t. , 20 h
3a (79)
3b (52)
3c (62)
3d (87)
3e (57)
3f (50)
5 (87)
2
2
2
2
2
2
CH
CF
17CH
–C
2
OH
r.t. ꢁ40ꢁC, 48 h
3
r.t., 20 h
(CH
CH
2
)
3
r.t. ꢁ40ꢁC, 72 h
r.t. ꢁ40 ꢁC, 48 h
r.t., 72 h
2
6
H
4
ꢀ p–CF
3
X = CN
r.t., 20 h
Room temperature ꢁ25–30 ꢁC
b
Isolated yield.

Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
5811
Fig. 1. B3LYP/6-31G*-optimized geometries of A and B in Scheme 2.
(
„–CO Me) or 2a. Even so, comparatively mild reaction
[18,19]. Slightly higher temperature was used because more
steric hindrance with alkenes than alkynes is considered to
exist also on the surface.
2
conditions (r.t. to 40 ꢁC) are enough to promote the reac-
tion of (Me Si) Si and activated alkenes because of stabil-
ꢀ
3
3
ization of the intermediate radical by spin delocalization
The various alkene-modified surfaces were characterized
by XPS. XPS spectra of modified surfaces S1e and S2 are
shown in Fig. 2. Peak of fluorine 1s photoelectrons at a
binding energy of 688.9 eV for trifluoromethylbenzyl ester
modified surface S1e indicates that the organic monolayers
were successfully formed on the silicon surface. In the high
with the CO R group.
2
The hydrosilylation of activated alkenes was applied to
Si(111)–H surface modification. Hydrogen-terminated
Si(111) surface was prepared by etching oxidized Si surface
in HF and NH F solutions, as described previously [8]. The
4
freshly prepared Si(111)–H surface was immersed in neat
alkene or in a dichloromethane solution and the reaction
mixture was kept at 30–35 ꢁC for 40 h to give the modified
surface with Si–C bonded organic monolayer (Scheme 3)
resolution carbon 1s spectra, peaks of CF carbon and car-
3
bonyl (C@O) carbon were also identifiable [8,20]. CF3
group on S1e gave the typical C 1s peak at high binding
energy of 293.6 eV, and carbonyl group showed C 1s peaks
at 289.8 eV [21]. Peaks of C–Si (around 284–5 eV) [10a] and
C–C species in the C 1s spectra are overlapped and not
resolved in this measurement condition. Si 2p spectrum
of XPS for S2 shows that the Si 2p peak at ca. 103 eV cor-
responding to Si oxidation [22] is small (Equiv.ml of SiOx
a
H
H
H
H
Si
Si
Si
Si
+
Si (111)
X
2a~f
Si(111)-H
[
23] 0.23).
X
X
X
Monolayer coverage of modified surfaces containing
fluorine and nitrogen atoms were estimated by calculation
of the relative intensities of F 1s peaks and N 1s peaks with
respect to the Si 2p peaks according to the method previ-
ously reported [8,24]. The estimated coverage ratios of
S1c and S1e–f range from 19% to 39% (Table 2) [25].
The mechanism of hydrosilylation of activated alkenes
with a Si(111)–H surface may be similar to that of alkynes
previously reported [8,26]. The silicon radical would react
with alkene (2 or 4) to form Si–C bond, leading to a carbon
radical. The carbon radical abstracts a hydrogen from an
adjacent Si–H group to regenerate a surface Si radical to
propagate the chain reaction [27].
neat or in CH Cl2
2
o
3
0~35 C, 40h
Si
Si
Si
Si
Si (111)
S1a: X = CO Me
S1d: X = CO (CH ) CH3
2
2
2 17
S1b: X = CO CH CH OH S1e: X = CO CH -C H -p-CF
2
2
2
2
2
6
4
3
S1c: X = CO CH CF
S1f: X = CN
2
2
3
H
H
H
H
b
Si
Si
Si
Si
+
O
O
Si (111)
O
Si(111)-H
4
O
O
O
O
O
O
O
O
CH Cl2
2
O
2.2. Chemical transformation of surface functional groups
o
3
0~35 C, 40h
Si
Si
Si
Si
Further chemical transformations of free carboxyl, ester
and hydroxy-terminated Si surfaces were carried out, in
order to prepare various functionalized surfaces which
may be difficult to prepare by hydrosilylation reactions.
Si (111)
S2
Scheme 3.

5812
Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
Fig. 2. (a) XPS spectra of modified Si(111) surface S1e. (b) XPS spectra of surface S2.

Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
5813
(
1) Ester reduction and bromination: Ester-terminated
surface was reduced to a hydroxy-terminated surface
Model reactions
Me Si) Si
CO Me
LiAlH / Et O
4
2
(
(Me
3
Si)
3
Si
OH
3
3
2
by LiAlH [28]. Furthermore, the hydroxy terminal
4
r.t., 20h
94%
group was readily transformed to a bromoalkyl-ter-
minated surface by bromination reaction.
3a
6
The transformation was first investigated with model
compounds. Methyl ester 3a was reduced in diethyl
CBr /PPh /CH Cl
4
3
2
2
(
Me Si) Si
Br
3
3
r.t., 2h
90%
ether by a suspension of LiAlH to give alcohol 6,
4
7
and 6 was brominated by CBr and PPh to give 7
4
3
Surface transformations
(Scheme 4). Both 6 and 7 were obtained in good
yields without cleavage of the Si–Si bond. Next,
methyl acrylate modified surface S1a was examined
by immersing in an ether suspension of LiAlH at
4
room temperature overnight to give the reduced sur-
face S3. The proof of successful reduction of the
methyl ester group was shown by XPS analysis. Com-
paring the carbon 1s spectra of S1a and the reduced
derivative S3 (Fig. 3a), disappearance of the C 1s
peak for carbonyl carbon at 289.7 eV indicated that
the ester reduction was almost complete. Moreover,
the water contact angle changed from 66ꢁ (S1a) to
LiAlH / Et O
4
2
Si
Si Si Si
Si
Si Si
Si
r.t., 20h
Si (111)
Si (111)
S1a
S3
CBr /PPh /CH Cl
4
8ꢁ (S3) after reduction (Tables 3 and 5). This may
4
3
2
2
Si
Si Si
Si
arise from higher hydrophilicity of the hydroxy group
than ester group. This hydroxy-terminated surface
was then treated with a mixture of CBr , PPh in
r.t., 16h
Si (111)
S4
4
3
dichloromethane to transform the hydroxy group to
a bromo group. XPS analysis of the obtained surface
S4 verified the existence of bromine. Br 3d photoelec-
trons showed two peaks at 70 and 71 eV, which are
Scheme 4. Ester reduction and bromination.
5
/2
3/2
the spin-splitting peaks of Br 3d
and 3d
lines
strands or proteins [4a,4b,4c,31] towards biotechno-
logical applications. Ester and amide formations are
the most basic reactions for covalent immobilization
of biomolecules. Esterification of a hydroxy-termi-
nated surface has been reported by reaction with ace-
tic anhydride [28a] or acetyl chloride [28b] to give
acetate termination. In this work, a mild esterification
using 1-ethyl-3-(3-dimethylaminopropyl)-carbodii-
mide hydrochloride (EDCI) and 4-N,N-dimethylami-
nopyridine (DMAP) was carried out. Model
compounds 3b, 6 and 8 react with tert-butoxycar-
bonyl (Boc) glycine in the presence of EDCI/DMAP.
Esters 9a–c were obtained in high yields after over-
night reactions at room temperature (Scheme 5a).
[
29]. In the C 1s spectrum of S4, a peak at 286.9 eV
with a chemical shift of 1.2 eV from the carbon cen-
terline was assigned to –CH Br (Fig. 3b) [30]. In a
comparison, the C 1s peak assigned to –CH OH of
2
2
S3 appeared at 287.0 eV which had a chemical shift
of 1.7 eV from the centerline. The monolayer cover-
age calculated from the relative intensities of Br 3d
to Si 2p photoelectrons was 32% (Table 4). Thus,
these results indicated that the transformation of
hydroxy group to bromine proceeded smoothly.
2) Ester and amide formation: Silicon surfaces with ter-
minal functionalities such as hydroxyl, carboxy or
amine are of interest for immobilization of DNA
(
Table 2
Surface hydrosilylation and the estimated monolayer coverages by XPS
Surface
Surface group
Coverage (%)
Equiv.ml
x
of SiO [23]
a
S1c
S1e
S1f
„Si–CH
„Si–CH
„Si–CH
2
2
2
CH
CH
CH
2
2
2
CO
CO
CN
2
CH
CH
2
CF
–C
3
39
1.07
0.38
1.62
a
2
2
6
H
4
ꢀ p–CF
3
19
b
38
O
c
S2
Si
O
14
0.23
a
Determined by F 1s/Si 2p peaks.
Determined by N 1s /Si 2p peaks.
Determined by C 1s (C@O) /Si 2p peaks.
b
c

5814
Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
1
x 1 0
2
a
x 1 0
b
Si 2p
3
3
2
5
0
5
Wide
2
2
1
1
5
0
5
0
C 1s
S1a
C O
O 1s
O
C O
20
C 1s
1
5
0
1
Br 3d
5
0
5
0
1000
x 1 01
800
600
400
200
0
1
0
S3
Br 3d 5/2
4
0
0
0
0
0
Br 3d
Br 3d 3/2
8
6
4
2
0
3
2
1
-
CH OH
2
7
8
x 1 02
76
74
72
70
68
66
64
3
3
2
2
1
1
5
0
5
0
5
0
S4
Si 2p
2
5
2
1
0
5
-
CH Br
2
10
5
5
0
SiOx
0
300
290
280
108
106
104
102
100
98
96
Binding Energy (eV)
Binding Energy (eV)
Fig. 3. (a) C 1s XPS region spectra of surfaces S1a and S3–4. (b) XPS survey scan spectrum and region spectra (Br 3d and Si 2p) of S4.
Table 3
dichloromethane solution of Boc-glycine, EDCI and
DMAP at room temperature for 20 h, the obtained
surfaces were characterized by XPS and contact angle
measurement. XPS analysis of surface S6a derived
from S3 proved that nitrogen was contained in the
monolayer while the N 1s photoelectrons showed its
peaks at 401.3 eV. In the C 1s spectrum, the peak
for the carbonyl carbon appeared at 290.3 eV is fur-
ther proof of the successful introduction to Boc-gly-
cine on the hydroxy-terminated surface through
ester formation. Similarly, XPS spectra of surfaces
S6b and S6c confirmed the formation of an ester
linked monolayer of Boc-glycine (for surface S6c,
see Fig. 4). Based on quantitative XPS results, mono-
layer coverages of S6a–c calculated from relative
intensities of N 1s peak to Si 2p peak were in a range
of 34–57%. Measured water contact angles of S6a–c
Static water contact angles of modified surfaces
Surface Surface group
Contact
angle (ꢁ) error (ꢁ)
Standard
S1a
S1b
S1c
S1d
S1e
S1f
„Si–CH
„Si–CH
„Si–CH
„Si–CH
„Si–CH
„Si–CH
2
2
2
2
2
2
CH
CH
CH
CH
CH
2
2
2
2
2
2
CO
CO
CO
CO
CO
2
2
2
2
2
CH
CH
CH
3
2
2
66
53
69
86
72
64
0.3
0.4
0.5
0.9
0.5
0.4
CH
CF
17CH
–C
2
OH
3
(CH
CH
2
)
3
2
6
H
4
ꢀ p–CF
3
CH
O
CN
S2
Si
O
52
0.7
Next, this esterification method was applied to
hydroxy-terminated Si surfaces, S3, S1b and S5
(
Scheme 5b). After immersion in a mixture of a

Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
5815
Table 4
XPS estimated monolayer coverages of Si surfaces after chemical transformation
Surface
Surface group
„Si–CH CH
Coverage (%)
x
Equiv.ml of SiO [23]
S4
2
2
CH
2
Br
32
42
35
58
16
22
27
0.54
2.03
0.41
0.43
0.87
0.88
0.51
t
S6a
S6b
S6c
S9a
S9b
S9c
BSi–CH2CH2CH2OCðOÞCH2NH–CO2 Bu
t
BSi–CH2CH2CO2CH2CH2OCðOÞCH2NH–CO2 Bu
t
BSi–CH@CHCO2CH2CH2OCðOÞCH2NH–CO2 Bu
„Si–CH@CHCO
„Si–CH@CHCO
2
CH
CH
2
CF
–C
3
2
2
6
4 3
H –p-CF
t
BSi–CH@CHCðOÞNH–CH2CO2 Bu
Table 5
Static water contact angles of modified surfaces
Surface
Surface group
Contact angle (ꢁ)
Standard error (ꢁ)
S3
S4
„Si–CH
„Si–CH
2
CH
CH
2
CH
CH
2
OH
Br
48
52
62
57
58
47
60
59
0.4
0.4
0.3
0.3
0.3
0.4
0.4
0.5
2
2
2
t
S6a
S6b
S6c
S9a
S9b
S9c
BSi–CH2CH2CH2OCðOÞCH2NH–CO2 Bu
t
BSi–CH2CH2CO2CH2CH2OCðOÞCH2NH–CO2 Bu
t
BSi–CH@CHCO2CH2CH2OCðOÞCH2NH–CO2 Bu
„Si–CH@CHCO
„Si–CH@CHCO
2
CH
CH
2
CF
–C
3
2
2
6
4 3
H –p-CF
t
BSi–CH@CHCðOÞNH–CH2CO2 Bu
ranged from 57ꢁ to 62ꢁ, which were a little higher
than them of the hydroxy-terminated surfaces (44ꢁ–
Model reactions
Me Si) Si R1 OH + HO C
EDCI/DMAP/
t
CO Bu
2
CH Cl2
(
N
H
2
3
3
2
5
3ꢁ for S1b, S3 and S5).
r.t., 20h
1
6
3
8
: R = CH CH CH
2 2 2
1
b: R = CH CH CO CH CH
It has been reported that esterification of free carboxylic
acid-terminated Si surfaces was carried out in boiling acidic
propanol [28a]. Ester formation reaction using EDCI/
DMAP was applied to a carboxy-terminated surface
28b]. As shown in Scheme 6, ester and amide formation
of carboxyl group with EDCI were investigated with both
model compounds and Si surface. Acid model compound
2 2 2 2 2
1
: R = CH=CHCO CH CH
2
2
2
t
CO Bu
Me Si) Si R1 O C
N
2
(
3
3
2
H
1
9
9
9
a: R = CH CH CH , 94%
2 2 2
[
1
b: R = CH CH CO CH CH , 92%
2 2 2 2 2
1
c: R = CH=CHCO CH CH , 91%
2
2
2
1
0 reacted with 2,2,2-trifluoroethyl alcohol (11a) and 4-tri-
Surface transformations
OH OH OH
R2 R2 R2
Si
fluoromethylbenzyl alcohol (11b) in the presence of EDCI
and DMAP to give corresponding esters 12a and 12b in
9
3% and 78% yields, respectively. For amide formation,
EDCI/DMAP/
Si Si
Si
t
CH Cl2
reaction of 10 and glycine tert-butyl ester (11c) was carried
out in the presence of EDCI and 1-hydroxybenzothiazole
CO Bu
2
2
+
HO C
N
H
Si (111)
2
r.t., 20h
2
(
HOBT). Amide 12c was obtained in 83% yield after 20 h
S3: R = CH CH CH
2
2
2
2
S1b: R = CH CH CO CH CH
reaction at room temperature. Then, model reactions were
applied to carboxy-terminated Si surface. Free carboxylic
acid-terminated surface S8 was prepared by two-step pro-
cedure in order to avoid possibility to react with Si–H
group to form silyl ester (Si–OC(O)) bond [28a,31,32]. Sur-
face S7 was prepared first by modification of Si(111)–H
surface with tert-butyl propiolate. S7 was then treated with
trifluoroacetic acid (TFA)/dichloromethane solution to
cleave the tert-butyl group to give free carboxy-terminated
surface S8 (Fig. 5) [33,8]. Thereafter, ester and amide for-
mation reactions were carried out on surface S8 with alco-
hols 11a,b and amine 11c under the same condition of
model reactions. Surface S9a obtained from esterification
with trifluoroethanol showed F 1s peak and the typical C
2
2
2
2
2
2
S5: R = CH=CHCO CH CH
2
2
2
3
3
O CR3
2
R2 R2
R CO
R CO
2
2
R2
Si
Si Si Si
Si (111)
2
3
t
S6a: R = CH CH CH ; R = CH NHCO Bu
2
2
2
2
2
2
3
tBu
S6b: R = CH
CH
CO
CH
CH
; R = CH
NHCO
2
2
2
2
2
2
2
t
2
3
S6c: R = CH=CHCO CH CH ; R = CH NHCO Bu
2
2
2
2
2
Scheme 5.
Monolayer coverage of trifluoroethyl ester estimated by
XPS was 16% (Table 4). The contact angle of S9a was
47ꢁ (Table 5) which is lower than that of the same surface
1
s peak for CF3 group in the XPS spectra (Fig. 6).

5816
Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
Fig. 4. XPS spectra of surface S6c.
prepared directly from propiolate ester (contact angle 75ꢁ,
coverage 56%) [8]. The left unreacted carboxy groups and
higher degree of the surface oxidation [34] may be respon-
sible to this low contact angle. Similarly, S9b containing
fluorine has a coverage of trifluoromethylbenzyl ester
group estimated to be 22% by XPS. Formation of amide
group on surface S9c was also proved by quantitative anal-
ysis of nitrogen composition by XPS. From Fig. 7, it can be
found that the N 1s peak clearly appeared at 400.8 eV. The
coverage of amide group was estimated to be 27% based on
N 1s XPS data, and the water contact angle was measured
to be 59ꢁ (Table 5).
functional groups and inclusion of heteroatom such as fluo-
rine or nitrogen. In order to explore the complementary
methods to the well known techniques, we have attempted
determination of the coverage ratios by combination of
chemical transformation and gas chromatography.
In the former section, the reduction of surface ester
group with LiAlH transforms an ester-terminated surface
4
to a hydroxy-terminated surface. In the reduction of S1a,
methanol is liberated. When the surface ester group is com-
pletely reduced, it is possible to detect the liberated alcohol
quantitatively and calculate the absolute coverage ratio of
the starting ester group. The concentration of Si–H group
ꢀ
9
2
The chemical transformation presented in this work
sometimes leads to significant oxidation of the silicon sur-
face, as estimated using fractional monolayer coverage
on an ideal Si(111) surface is about 1.3 · 10 mol/cm
[2,36]. For detection of the alcohol liberated from Si sur-
face, the gas chromatography (GC) with flame-ionization
detectors (FID) was used.
(
Equiv. ) of SiO by XPS (Tables 2 and 4) [23]. Such oxide
ml x
formation may be a serious issue so as to render the inter-
face useless for electronic applications. Study to solve the
problem is under investigation [35].
The ester modified Si surfaces, surfaces S1d and S10
(„Si(111)–CH@CH–CO (CH ) CH ) [37] were prepared
2
2 17
3
for estimation of monolayer coverage by combination of
ester reduction and GC analysis. The ester modified surface
was then reduced by LiAlH in Et O at room temperature
2.3. Estimation of monolayer coverage by ester reduction and
GC analysis
4
2
for 20 h. The liberated alcohol, octadecyl alcohol was
extracted and concentrated to a certain volume of n-propa-
nol solution, and this solution was used for quantitative
GC analysis to determine the concentration (Scheme 7).
IR and XPS spectra for estimation of the coverage ratio
of modified surfaces have limited applicability to selected

Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
5817
Model reactions
Me Si) Si CO H
functional terminal groups. The further chemical transfor-
mation of the surface functional groups was also carried
out based on the reactions such as ester reduction, bromin-
ation, esterification and amide formation. Developments of
modification and chemical transformation on silicon sur-
face are of interest and important to tailor the surface
structures and properties for a variety of applications.
We have investigated the combination of ester reduction
and GC analysis to estimate the monolayer coverage of
ester modified Si surfaces. This method can be used as a
complement for XPS and ATR-IR spectroscopies.
(
i) EDCI/ DMAP
or (ii) EDCI/ HOBT
(
+
H Y
3
3
2
CH Cl , r.t., 20h
10
11
2
2
1
1
1
1a: Y= OCH CF
2 3
1b: Y= OCH -C H -
p
-CF3
2
6 4
t
1c: Y= NH-CH -CO Bu
2
2
(
Me Si) Si
COY
3
3
Condition: (i) 12a: Y= OCH CF , 93%
2
3
(
i) 12b: Y= OCH -C H -p-CF , 78%
2 6 4 3
ii) 12c: Y= NH-CH -CO Bu, 83%
t
(
2
2
Surface transformations
3. Experimental
t
t
t
BuO C BuO C
CO Bu
HO C
2
HO C
2
CO H
2
2
2
2
3.1. Model reactions
TFA/
CH Cl
Si
Si Si Si
Si
Si Si Si
2
2
General methods: Melting points are uncorrected. IR
+
H Y
Si (111)
1
Si (111)
r.t., 1h
spectra were recorded in the FT-mode. H NMR spectra
1
1
1
3
S7
S8
were recorded at 400 MHz.
C NMR spectra were
1
1
1
1
1a:Y = OCH CF
recorded at 100.6 MHz. H chemical shifts are reported
in ppm relative to Me Si. C chemical shifts are reported
2 3
1
3
1b: Y = OCH -C H -
p
-CF3
2
6
4
4
t
1c: Y = NH-CH -CO Bu
2
2
in ppm relative to CDCl (77.1 ppm). Mass spectra were
3
recorded at an ionizing voltage of 70 eV by EI or FAB.
COY
COY COY
All reactions were carried out under
atmosphere.
a
nitrogen
(
i) EDCI/ DMAP
or (ii) EDCI/ HOBT Si
Si Si Si
Materials: Substrates 1, 2a–d,f, 4, 11a–c were purchased
and used without further purification. 8 and 10 were pre-
pared according to the previous report [8] Spectral data of
3a, 3f, 5, 7a and7b were in accord with reported data [40].
4-Trifluoromethylbenzyl acrylate (2e): To a mixture of
acrylic acid (360 mg, 5.0 mmol) and 4-(trifluorom-
ethyl)benzyl alcohol (1.05 g, 6.0 mmol) in 5.0 ml dichloro-
methane was added 4-(dimethylamino)pyridine (122 mg,
Si (111)
S9
CH Cl , r.t., 20h
2
2
Condition: (i) S9a: Y = OCH CF
2
3
(
(
i) S9b: Y = OCH -C H -p
-CF3
2 6 4
ii) S9c: Y = NH-CH -CO Bu
t
2
2
Scheme 6.
1.0 mmol) and 1-ethyl-3-(3-dimethylaminopropyl)carbodii-
GC diagram of the sample and the reference solutions is
mide hydrochloride (1.15 g, 6.0 mmol). After stirring at
room temperature for 20 h, the reaction mixture was
diluted with dichloromethane and washed with water.
shown in Fig. 8. The concentration of the sample solutions
obtained from surface reduction was determined by inte-
gration of the peak areas [38]. Using the determined con-
centration, the monolayer coverage was calculated as
following equation:
The organic phase was dried (MgSO ) and evaporated in
4
vacuo. The residue was purified by column chromatogra-
phy over silica gel eluting with hexane–ether (10:1) to give
ꢀ
ꢁ
the title compound (593 mg, 52%) (R = 0.2). Colorless
ꢀ
9
f
Coverage ¼ ðC ꢂ V Þ= ð1:3 ꢃ 10 Þ ꢂ Ssurface
1
liquid; H NMR (400 MHz, CDCl ) d 5.25 (s, 2H), 5.89
3
where (C (mol/L) is the concentration of the sample solu-
(dd, 1H, J = 10.4, 1.4 Hz), 6.18 (dd, 1H, J = 17.2,
tion; V (L) is the volume of the sample solution; S
10.4 Hz), 6.48 (dd, 1H, J = 17.2, 1.4 Hz), 7.49 (d, 2H,
surface
2
13
(
cm ) is the total surface area of the Si wafer.)
J = 8.1 Hz), 7.63 (d, 2H, J = 8.1 Hz); NMR
(100.6 MHz, CDCl ) d 65.4, 124.1 (q, –CF , JF–C =
3 3
C
The estimated coverages of surfaces S1d and S10 were
2
5% and 35%, respectively (Table 6) [39]. Coverages of sur-
272.0 Hz), 125.6 (q, JF–C = 3.8 Hz), 128.0, 128.2, 130.5
(q, JF–C = 32.5 Hz), 131.7, 139.9, 165.9 (–CO –); IR (neat)
1734, 1635, 1622, 1408, 1327, 1296, 1269, 1170, 1125, 1067,
faces S1d and S10 by GC are comparable to those values
estimated by C 1s (C@O) spectra of XPS. Because XPS
determination of C 1s spectra may be not very accurate
by the contaminated carbons, GC method can be suitable
for octadecyl ester modified surfaces.
In summary, a mild modification of hydrogen-termi-
nated silicon surface by hydrosilylation of activated
alkenes was investigated. This surface modification resulted
in the formation of Si–C bonded organic monolayer with
2
ꢀ1
+
1020, 984, 822, 809 cm . MS (EI) m/z 230 (M ), 211
+
+
+
(M ꢀF), 175 (M ꢀCH
CHCO –); exact mass
, 230.0555).
Hydrosilylation of alkenes with tris(trimethylsilyl)silane
@CHC(O)–), 159 (M ꢀCH
@
2
2
+
M
230.0554 (calcd for
2
C H F O
11 9 3 2
1 was carried out through a similar procedure reported for
alkynes [12]. Reaction of 2b,d,e with 1 was performed at

5818
Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
Fig. 5. XPS spectra of surface S8.
room temperature overnight and followed by warming to
0 ꢁC until the high conversion of the alkenes was
reached.
-Hydroxyethyl
3b): (52%) Column chromatography: hexane–ether = 2/1,
Octadecyl 3-[Tris(trimethylsilyl)silyl]propionate (3d):
(87%) Column chromatography: hexane, R = 0.3. Color-
4
f
1
less oil; H NMR (400 MHz, CDCl ) d 0.17 (s, 27H),
3
2
3-[tris(trimethylsilyl)silyl]propionate
0.88 (t, 3H, 1.11 m, 2H), 1.25–1.31 (broad, 30H), 1.76
1
3
(
(m, 2H), 2.31 (m, 2H), 4.05 (t, 2H, J = 6.8 Hz);
C
1
R = 0.1. Colorless crystals; m.p. = 98–99 ꢁC; H NMR
NMR (100.6 MHz, CDCl3) d 1.2 (–Si(CH ) ), 3.0, 14.2,
f
3 3
(
2
400 MHz, CDCl ) d 0.18 (s, 27H), 1.11 (m, 2H), 2.38 (m,
H), 3.84 (m, 2H), 4.22 (m, 2H); C NMR (100.6 MHz,
22.8, 26.0, 28.7, 29.4, 29.5, 29.6, 29.7, 29.8, 32.0, 33.4,
3
1
3
64.7, 174.9 (–CO –); IR (neat) 2925, 2854, 1739, 1467,
2
ꢀ
1
CDCl ) d 1.2 (–SiCH ), 2.9, 33.2, 61.5, 66.3, 175.2 (–COO–
1337, 1245, 1198, 1148, 836, 688, 622 cm . MS (FAB)
3
3
+
+
)
8
; IR (KBr) 3349, 2947, 2892, 1736, 1246, 1202, 1156, 1081,
m/z 571 (M ꢀH); exact mass M ꢀH 571.4219 (calcd for
ꢀ1
+
34, 688, 623 cm . MS (FAB) m/z 387 (M +Na); exact
C H O Si , 571.4218).
3
0
67
2
4
+
mass M +Na 387.1614 (calcd for C H NaO Si ,
4-Trifluoromethylbenzyl 3-[tris(trimethylsilyl)silyl]pro-
pionate (3e): (57%) Column chromatography: hexane–
1
4
36
3
4
3
87.1639).
,2,2-Trifluoroethyl 3-[tris(trimethylsilyl)silyl]propio-
1
2
ether = 20:1, R = 0.15. Colorless oil;
H
NMR
f
nate (3c): (62%) Column chromatography: hexane/ethyl
(400 MHz, CDCl ) d 0.15 (s, 27H), 1.10 (m, 2H), 2.37
3
1
acetate = 40/1, R = 0.25. Colorless oil; H NMR
(m, 2H), 5.13 (s, 2H), 7.44 (d, 2H, J = 8.1 Hz), 7.60 (d,
f
1
3
(
400 MHz, CDCl ) d 0.18 (s, 27H), 1.12 (m, 2H), 2.42
2H, J = 8.1 Hz); C NMR (100.6 MHz, CDCl ) d 1.1
3
3
1
3
(
m, 2H), 4.46 (q, 2H, JF–H = 8.4 Hz);
C
NMR
(–Si(CH ) ), 3.0 (Si–CH CH –), 33.2 (Si–CH CH –), 65.4
3 3
2
2
2
2
(
100.6 MHz, CDCl ) d 1.1 (–SiCH ), 2.8 (Si–CH –CH –),
(–CH –C H – CF ), 124.1 (q, J
= 272.4 Hz), 125.6 (q,
F–C
3
3
2
2
2
6
4
3
3
3
2
6
2.6 (Si–CH –CH –), 60.4 (q, –CH –CF , JF–C
=
JF–C = 3.8 Hz), 128.3, 130.4 (q, JF–C = 32.3 Hz), 140.2,
2
2
2
3
6.6 Hz), 123.1 (q, –CH –CF , J
= 277.0 Hz); IR (neat)
174.4 (–CO –); IR (neat) 2950, 2894, 1744, 1623, 1420,
2
3
F–C
2
950, 2895, 1764, 1411, 1282, 1246, 1170, 1130, 834, 689,
1326,1246, 1167, 1131, 1068, 1020, 836, 747, 689,
ꢀ
1
+
+
ꢀ1
+ + +
+
+
22 cm . MS (EI) m/z 403 (M +1), 387 (M –CH ), 329
623 cm . MS (EI) m/z 477 (M ꢀH), 463 (M ꢀCH ),
3
3
+
+
(
M –Si(CH ) ), 73 ((CH ) Si ); Anal. Calc. for
405 (M ꢀSi(CH ) ), 159 ( CH C H CF ), 73 ((CH ) Si );
3
3
3 3
3 3
2
6
4
3
3 3
+
C H F O Si : C, 41.75; H, 8.26. Found: C, 41.55; H,
exact mass M ꢀH 477.1745 (calcd for C H O F Si ,
1
4
33
3
2
4
20 36
2
3
4
8
.31%.
477.1744).

Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
5819
Fig. 6. XPS spectra of surface S9a.
3
-[Tris(trimethylsilyl)silyl]propanol (6): LiAlH in sat-
10 min, the mixture was extracted by dichloromethane
4
urated ether solution (ca. 1 N, 8.0 mL) was transferred to a
5 ml flask containing 3a (320 mg, 1.0 mmol). After stirring
(15 mL · 3). The extract was washed by brine, dried
2
(MgSO ) and evaporated in vacuo. The residue was puri-
4
at room temperature for 20 h, the mixture was carefully
poured into cold aqueous solution of hydrogen chloride
fied by column chromatography over silica gel eluting with
hexane to give the title compound (134 mg, 90%)
1
(
(
1 N) and then extracted by ether. The extract was dried
(R = 0.65). Colorless wax; H NMR (400 MHz, CDCl )
f
3
MgSO ) and evaporated in vacuo. The residue was puri-
d 0.17 (s, 27H), 0.87 (m, 2H), 1.91 (m, 2H), 3.38 (t, 2H,
4
1
3
fied by column chromatography over silica gel eluting with
J = 6.9 Hz); C NMR (100.6 MHz, CDCl ) d 1.2 (–
3
hexane–ether (5:1) to give the title compound (288 mg,
SiCH ), 6.7 (–SiCH –), 32.7 (–SiCH CH –), 37.2 (–CH –
Br); IR (KBr) 2948, 2893, 1431, 1396, 1492, 1245, 831,
3
2
2
2
2
1
9
4%) (R = 0.1). Colorless crystal, m.p. = 49–51 ꢁC;
H
f
ꢀ
1
+
NMR (400 MHz, CDCl ) d 0.16 (s, 27H), 0.76 (m, 2H),
742, 687, 622, 562 cm . MS (EI) m/z 370, 368 (M ),
3
1
3
+
+
1
.62 (m, 2H), 3.57 (t, 2H, J = 6.7Hz);
C NMR
355, 353 (M ꢀCH ), 297, 295 (M ꢀSi(CH ) ), 73
3
+ +
3 3
(
100.6 MHz, CDCl ) d 1.2 (–SiCH ), 3.3 (–SiCH –), 32.3
((CH ) Si ); exact mass M 370.0817, 368.0846 (calcd for
3 3
3
3
2
8
1
79
12 33 4
(
2
3
–SiCH CH –), 66.1 (–CH –OH); IR (KBr) 3348, 2947,
C H
12 33
BrSi , 370.0822; C H BrSi , 368.0843).
3-[Tris(trimethylsilyl)silyl]propyl(tert-butoxycarbon-
2
2
2
4
ꢀ1
892, 1243, 1052, 835, 746, 686, 623 cm . MS (EI) m/z
+
+
+
05 (M ꢀH), 291 (M ꢀCH ), 233 (M ꢀSi(CH ) ), 73
ylamino)acetate (9a): (94%) Column chromatography:
3
3 3
+
+
(
(CH ) Si ); exact mass
M
306.1679 (calcd for
hexane–ether = 2:1,
m.p. = 48–50ꢁC. H NMR (400 MHz, CDCl ) d 0.16
(s, 27H), 0.75 (m, 2H), 1.45 (s, 9H), 1.70 (m, 2H), 3.92
R = 0.3.
Colorless
crystal,
3
3
f
1
C H OSi , 306.1687). Anal. Calc. for C H OSi : C,
1
2
34
4
12 34
4
3
4
7.02; H, 11.18. Found: C, 46.99; H, 11.17%.
-Bromo-3-[tris(trimethylsilyl)silyl]propane (7): CBr4
166 mg, 0.5 mmol) and PPh3 (131 mg, 0.5 mmol) was
1
(d, 2H, J = 5.5 Hz), 4.07 (t, 2H, J = 6.9 Hz), 5.01 (broad,
1
3
(
1H); C NMR (100.6 MHz, CDCl ) d 1.2 (–SiCH ), 3.5
3 3
added to a solution of 6 in dichloromethane (1.0 mL).
(–SiCH –), 28.0 (–SiCH CH –), 28.4 (–C(CH ) ), 42.5
2 2 2 3 3
The reaction mixture was stirred at room temperature for
(–CH NH–), 68.1 (CH CH O–), 80.0 (–C(CH ) ), 155.7
2 2 2 3 3
2
h. After addition of 2 mL of water and stirring for
(–NH–CO –), 170.5 (–CH CO –); IR (KBr) 3308, 2948,
2 2 2

5820
Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
Fig. 7. XPS spectra of surface S9c.
1
8
6
4
2
0
000
00
00
00
00
LiAlH / Et O
4
2
Si
Si Si
Si
Si
Si Si
Si
r.t., 20h
Si (111)
Si (111)
S1d or S10
R = -(CH2)17CH3)
+
Reference
(
R
OH
Sample
6
.2
6.4
6.6
6.8
7.0
7.2
GC analysis
Retention Time [min]
Scheme 7.
Fig. 8. Typical GC diagrams of a sample solution from surface reaction
and a reference solution (1-octadecanol (concentration of reference
solution, 1.0 · 10 mol/L)).
2
1
892, 1758, 1689, 1540, 1417, 1392, 1365, 1301, 1244,
206, 1167, 1052, 980, 834, 688, 622 cm . MS (FAB)
m/z 486 (M +Na); exact mass M +Na 486.2320 (calcd
ꢀ4
ꢀ
1
+
+
(m, 2H), 1.44 (s, 9H), 2.34 (m, 2H), 3.93 (d, 2H,
for C H NO Si Na, 486.2323). Anal. Calc. for
J = 5.7 Hz), 4.28 (m, 2H), 4.34 (m, 2H), 5.02 (broad,
1H); C NMR (100.6 MHz, CDCl ) d 1.1 (–SiCH ), 2.7
3 3
1
9
45
4
4
1
3
C H NO Si : C, 49.19; H, 9.78; N, 3.02. Found: C,
1
9
45
4
4
4
9.15; H, 9.90; N, 3.16%.
-[(2-tert-Butoxycarbonylamino)acetoxy]ethyl
tris(trimethylsilyl)silyl]propionate (9b): (92%) Column
(Si–CH CH –), 28.3 (–C(CH ) ), 33.0 (Si–CH CH –),
2 2 3 3 2 2
42.3 (–NHCH –), 62.0 and 63.1 (–O–CH CH –O–), 80.1
2 2 2
(–C(CH ) ), 155.7 (–NH–CO –), 170.3 (–NHCH CO –),
3 3 2 2 2
2
3-
[
chromatography: hexane–ether = 2:1, R = 0.3. Colorless
174.5 (–CH CH CO –); IR (neat) 3384, 2949, 2894, 1742,
2 2 2
1723, 1513, 1367, 1246, 1168, 1057, 836, 689, 622 cm .
f
1
ꢀ1
wax; H NMR (400 MHz, CDCl ) d 0.16 (s, 27H), 1.09
3

Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
5821
Table 6
431.2164). Anal. Calc. for C H O Si : C, 50.06; H, 9.57;
16 38 2 4
N, 3.24. Found: C, 50.17; H, 9.66; N, 3.29%.
Monolayer coverages of ester modified Si surfaces by reduction and GC
analysis
Surface
Surface group
Coverage
estimated
Coverage
estimated
3.2. Si(111) surface modification
by GC (%) by XPS
a
Single-crystal n-Si(111) wafers (resistivity of 0.5–
(
%)
1
.5 X cm and thickness of 500 ± 25 lm), polished on one
S1d
S10
a
„Si–CH
„Si–CH@CHCO
2
CH
2
CO
2
(CH
(CH
2
)
17CH
17CH
3
3
25
35
38
31
side or both sides, were used in all surface reactions. Prep-
aration of hydrogen-terminated Si(111) surface was fol-
lowed by the previously reported method [8]. All surface
modifications were carried out under nitrogen atmosphere,
and all solutions used for surface modification were pre-
deoxygenated by nitrogen bubbling for 30 min. XPS spectra
were obtained with a KRATOS-AXIS-165 spectrometer,
with a Mg Ka line (1253.6 eV) or a monochromated Al
Ka line (1486.6 eV) used as the X-ray source. An analyzer
pass energy of 80 eV was used for survey scans, and that
of 40 eV was used for fine scans of specific elements. Static
water contact angle was measured by an ERMA G-1 con-
tact angle meter. A drop of pure water (ꢁ1 lL) was put
on the surface using a micro-syringe. Five measurements
were made at different spots for each sample surface. GC
analysis was performed on a Yanaco G-6800 gas chroma-
tography using a capillary column (30 m · 0.25 mm, supe-
lcowax 10) with flame-ionization detectors.
2
2
)
Determined by C 1s (C@O)/Si 2p peaks.
+
+
MS (EI) m/z 506 (M ꢀCH ), 448 (M ꢀSi(CH ) ), 247
3
3 3
+
+
+
(
(TMS) Si ), 73 ((CH ) Si ); exact mass M 521.2485
3 3 3
(
calcd for C H O NSi , 521.2480).
2
1
47
6
4
[
(2-tert-Butoxycarbonylamino)acetoxy]ethyl
3-
[
tris(trimethylsilyl)silyl]acrylate (9c): (91%) Column
chromatography: hexane–ether = 2:1, R = 0.3. Colorless
f
1
crystal, m.p. = 60–61 ꢁC; H NMR (400 MHz, CDCl ) d
3
0
.17 (s, 27H), 1.45 (s, 9H), 3.94 (d, 2H,J = 5.7 Hz), 4.32
(
m, 2H), 4.36 (m, 2H), 4.99 (broad, 1H), 6.59 (d, 1H,
1
3
J = 13.7 Hz), 6.81 (d, 1H, J = 13.7 Hz);
C NMR
(
100.6 MHz, CDCl3) d 1.4 (–SiCH ), 28.4 (–C(CH ) ),
3 3 3
4
2.4 (–HCH –), 61.4 and 63.3 (–O–CH CH –O–), 80.2
2 2 2
(
–C(CH ) ), 133.8 (Si–CH@CH–), 150.5 (Si–CH@CH–),
3
3
1
55.7 (–NH–CO –), 166.2 (–CH@CH–CO –), 170.4
2
2
(
1
6
–NHCH CO –); IR (KBr) 3364, 2978, 2950, 2893, 1743,
3.2.1. Hydrosilylation of alkene on Si(111)–H surface
A piece of freshly prepared Si(111)–H wafer was
immersed in neat alkene (2a–c,f) or a dichloromethane
solution of alkene (2d,e or 4, 1.0 mol/L) in a reaction tube.
The reaction mixture was gently warmed and kept at 30–
35 ꢁC for 40 h. The immersed wafer was successively rinsed
with dichloromethane, ethanol (for 4), sonicated in dichlo-
romethane, and dried under vacuum. Preparation of S5
was reported in the previous paper [8].
2
2
709, 1522, 1380, 1369, 1247, 1193, 1172, 1050, 836,
ꢀ
1
+
87 cm . MS (FAB) m/z 542 (M +Na); exact mass
+
M +Na 542.2221 (calcd for C H O NSi Na, 542.2222).
Anal. Calc. for C H NO Si : C, 48.51; H, 8.72; N, 2.69.
Found: C, 48.47; H, 8.83; N, 2.67%.
2
1
45
6
4
2
1
45
6
4
tert-Butyl
[[(Z)-3-tris(trimethylsilyl)silyl]acryloyla-
mino]acetate (12c): 11c (84 mg, 0.5 mmol), HOBT
(
68 mg, 0.5 mmol), 1-ethyl-3-(3-dimethylaminopropyl)
carbodiimide hydrochloride (EDCI, 115 mg, 0.6 mmol)
and triethylamine (36 mg, 0.5 mmol) was added to a solu-
tion of 10 (160 mg, 0.5 mmol), in dichloromethane
3
.2.2. Reduction by LiAlH4
A freshly prepared ester-modified surface was immersed
(
1.0 mL). After stirring at room temperature for 20 h, the
in ether solution of LiAlH at room temperature for 20 h.
4
reaction mixture was diluted with dichloromethane and
The obtained surface was successively washed with 1 N
aqueous solution of HCl and Milli-Q water, rinsed with
dichloromethane, sonicated in dichloromethane, and dried
under vacuum.
washed with saturated aqueous solution of NaHCO and
3
water. The organic phase was dried (MgSO ) and evapo-
4
rated in vacuo. The residue was purified by column chro-
matography over silica gel eluting with hexane–ether (2:1)
to give the title compound (178 mg, 83%) (R = 0.5). Color-
f
1
less crystal, m.p. = 92–94 ꢁC; H NMR (400 MHz, CDCl )
3.2.3. Bromination of S3
Freshly prepared surface S3 was immersed in dichloro-
3
d 0.16 (s, 27H), 1.47 (s, 9H), 3.97 (d, 2H, J = 4.9 Hz), 5.87
(
br, 1H), 6.51 (d, 1H, J = 13.4 Hz), 6.59 (d, 1H,
methane solution of CBr
for 20 h (the concentration of CBr
4
and PPh
3
at room temperature
and PPh is 0.25 mol/
1
3
J = 13.4 Hz);
C NMR (100.6 MHz, CDCl ) d 1.4
4
3
3
(
(
–SiCH ), 28.1 (–C(CH ) ), 42.0 (–NH–CH –), 82.2
L, respectively). The obtained surface was successively
washed with Milli-Q water, rinsed with ethanol and dichlo-
romethane, sonicated in dichloromethane, and dried under
vacuum to give S4.
3
3 3
2
–C(CH ) ), 136.0 (Si–CH@CH–C(O)–), 144.0 (Si–
3
3
CH@CH–), 165.8 (–C(O)NH–), 169.6 (–COO–); IR
(
KBr) 3421, 2947, 2892, 1735, 1670, 1515, 1392, 1240,
ꢀ
1
+
1
4
3
167, 837, 755, 687, 623 cm . MS (EI) m/z 431 (M ),
+
+
+
16 (M ꢀCH ), 374 (M ꢀC(CH ) ), 358 (M ꢀSi(CH ) ),
3.2.4. Ester and amide formation on surface
A freshly prepared hydroxy- or carboxy-terminated
surface was immersed in dichloromethane solution of
3
3 3
3 3
+
+
30 (M ꢀCO C(CH ) ), 302 (M ꢀSi(CH ) ꢀCH @
2
3 3
+
3 3
2
C(CH ) ); exact mass M 431.2155 (calcd for C H O Si ,
3
2
16 38
2
4

5822
Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
Boc-glycine or 11 (1 equiv.), EDCI (1.2 equiv.) and DMAP
0.2 equiv. for ester formation) or HOBT (1.0 equiv. for
every two-piece wafers a clean Teflon O’ring in a proper size
was inserted to keep an interval and enable the surface con-
tacting with the reaction solution. Modification of these Si
wafers by propiolate or acrylate esters and followed reduc-
(
amide formation) at room temperature for 20 h (the con-
centration of Boc-glycine or 11 is 0.5 mol/L). The obtained
surface was successively washed with Milli-Q water, rinsed
with ethanol and dichloromethane, sonicated in dichloro-
methane, and dried under vacuum.
tion by LiAlH were performed by the procedure described
4
above. After surface reduction, the reaction mixture was trea-
ted with 1 N HCl solution and extracted with diethyl ether.
The extract was concentrated and transferred to 500 ll vessel.
Ether was evaporated and 200 ll of n-propanol was added to
the vessel. The obtained propanol solution was analyzed by
GC to estimate the concentration of liberated alcohol.
GC calibration: Six n-propanol solutions of octadecyl
alcohol and 4-(trifluoromethyl)benzyl alcohol were pre-
pared, respectively. The concentration of each solution
3
.2.5. Estimation of monolayer coverage by XPS
Coverage ratio was calculated according to the follow-
ing equations as an example for a CF -functionalized sam-
3
ple S1c shown below [8]
Coverage Ratio
ꢀ
5
ꢂ
ꢃꢄꢂI
ꢃ
were different within
a
range of 4.0 · 10 –5.0 ·
IF1s
RSFF1s
Si2p ꢂ Intensity RatioSið1 1 1Þ surface
ꢀ4
¼
ð1Þ
ð2Þ
10 mol/L. Each solution was analyzed by GC for five
times. The mean intensities (peak integral) of the alcohol
peak were used to draw the intensity vs. concentration
3
AF ꢂ RSF
ml
Si2p
AFmlðAttenuation FactormonolayerÞ ¼ eꢀ
dml=ðkmlꢂcos hÞ
(I–C) plot. A linear fitting of the plot gave a simple linear
Intensity Ratio
Sið1 1 1Þ surface
R
0
relationship between the GC intensity and the solution
concentration of each alcohol compound. The linear rela-
tionship is presented as: I = k Æ C. For octadecyl alcohol,
d
0
ꢀ
dml=ðkmlꢂcos hÞ
1 1 1 ꢀz=ðkSiꢂcos hÞ
e
ꢂ
e
dðzÞ
ꢀd0 1 1=ðkSiꢂcos hÞ
1
¼
R
¼ 1 ꢀ e
1
ꢀ
dml=ðkmlꢂcos hÞ
ꢀz=ðkSiꢂcos hÞdðzÞ
e
ꢂ
e
0
7
k = 1.652 · 10 ; for 4-trifluoromethylbenzyl alcohol,
ð3Þ
7
k = 2.704 · 10 .
0
˚
Average layer spacing d
= 2d1 1 1 = 1.568 A [24a]
1
1 1
were used. Relative sensitivity factors (RSFF1s = 1.000,
RSFN1s = 0.505, RSFSi2p = 0.371 for Mg Ka X-rays;
RSFN1s = 0.477, RSFSi2p = 0.328 for monochromated Al
Ka X-rays) in KRATOS-AXIS-165 data library and the
Acknowledgements
We are grateful to Prof. K. Kakiuchi, Mr. Y. Okajima
and Prof. Y. Uraoka (Nara Institute of Science and Tech-
nology) for measurement of XPS spectra.
˚
inelastic mean free path (IMFP) kSi (21.88 A for Si photo-
electrons of kinetic energy = 1154 eV for Mg Ka X-rays;
˚
2
5.0 A for kinetic energy = 1387 eV for monochromated
References
Al Ka X-rays) were used [24b]. h represents takeoff angle
in XPS measurement. dml was calculated by PM3 calcula-
tions of distances of monolayer using model compounds
[
[
1] (a) J.M. Buriak, Chem. Rev. 102 (2002) 1271;
(
(
b) D.D.M. Wayner, R.A. Wolkow, J. Chem. Soc., Perkin Trans. 2
2002) 23.
(
for example, (Me Si) Si–CH –CH –CO –CH –CF ).
3 3 2 2 2 2 3
2] M.R. Linford, P. Fenter, P.M. Eisenberger, C.E.D. Chidsey, J. Am.
Chem. Soc. 117 (1995) 3145.
IMFP of organic monolayer was calculated according to
the literature [24c]. In the coverage estimation, total peak
area for Si 2p (including oxidized Si) was used.
[3] (a) A. Bansal, N.S. Lewis, J. Phys. Chem. B 102 (1998) 1067;
b) S. Takabayashi, R. Nakamura, Y. Nakato, J. Photochem.
Photobiol. A: Chem. 166 (2004) 107;
c) S. Takabayashi, M. Ohashi, K. Mashima, Y. Liu, S. Yamazaki,
(
(
CF3
CF3
CF3
Y. Nakato, Langmuir 21 (2005) 8832.
[
4] (a) T. Strother, W. Cai, X. Zhao, R.J. Hamers, L.M. Smith, J. Am.
Chem. Soc. 122 (2000) 1205;
O
O
O
O
O
O
dml
(
b) A.R. Pike, L.H. Lie, R.A. Eagling, L.C. Ryder, S.N. Patole, B.A.
Connolly, B.R. Horrocks, A. Houlton, Angew. Chem., Int. Ed. 41
2002) 615;
Sisurf.
(
Si Si Si Si Si Si Si Si Si Si
(c) T.L. Lasseter, B. Clare, N.L. Abbott, R.J. Hamers, J. Am. Chem.
Soc. 126 (2004) 10220;
d111
Si
(
d) Y. Coffinier, C. Olivier, A. Perzyna, B. Grandidier, X. Wallart,
Si(111) Surface S1c
J.-O. Durand, O. Melnyk, D. Sti e´ venard, Langmuir 21 (2005) 1489.
5] R.L. Cicero, M.R. Linford, C.E.D. Chidsey, Langmuir 16 (2000)
[
[
5
688.
6] (a) R. Boukherroub, S. Morin, F. Bensebaa, D.D.M. Wayner,
Langmuir 15 (1999) 3831;
(
b) J.M. Buriak, M.P. Stewart, T.W. Geders, M.J. Allen, H.C. Choi,
J. Smith, D. Raftery, L.T. Canham, J. Am. Chem. Soc. 121 (1999)
1491.
3.2.6. Estimation of monolayer coverage by ester reduction
and GC analysis
1
n-Si(111) wafers polished on both sides were used. They
[
7] L.C.P.M. de Smet, G.A. Stock, G.H.F. Hurenkamp, Q.Y. Sun, H.
Topal, P.J.E. Vronen, A.B. Sieval, A. Wright, G.M. Visser, H.
Zuilhof, E.J.R. Sudh o¨ lter, J. Am. Chem. Soc. 125 (2003) 13916.
2
were cut to several pieces with a same size (1.5 · 1.5 cm ).
These wafers were put into a reaction tube, and between

Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
5823
[
[
8] Y. Liu, S. Yamazaki, S. Yamabe, Y. Nakato, J. Mater. Chem. 15
2005) 4906.
9] (a) A. Faucheux, A.C. Gouget-Laemmel, C.H. de Villeneuve, R.
Boukherroub, F. Ozanam, P. Allongue, J.–N. Chazalviel, Langmuir
is calculated as the following equation. Equiv.ml = vSiOx/Inten-
sity_RatioSi(1 1 1)_surf, vSiOx = Peak area of SiO /total peak area of Si
2p ratios. For Intensity_RatioSi(1 1 1)_surf, Eq. (3) in Section 3 is used.
The equivalent fractional monolayer coverage for SiO may include
(
x
x
2
(
2 (2006) 153;
b) Y.-J. Liu, N.M. Navasero, H.-Z. Yu, Langmuir 20 (2004) 4039.
10] (a) X. Wallart, C.H. de Villeneuve, P. Allongue, J. Am. Chem. Soc.
27 (2005) 7871;
silicon oxidation both on the surface and under the surface
(formation of silicon oxide layer in partial area).
[24] (a) F.J. Himpsel, F.R. McFeely, A. Taleb-Ibrahimi, J.A. Yarmoff,
Phys. Rev. B 38 (1988) 6084;
[
[
1
(
(
b) L.J. Webb, N.S. Lewis, J. Phys. Chem. B 107 (2003) 5404;
c) L.J. Webb, E.J. Nemanick, J.S. Biteen, D.W. Knapp, D.J.
(b) M.P. Seah, W.A. Dench, Surf. Inter. Anal. 1 (1979) 2;
(c) P.J. Cumpson, Surf. Inter. Anal. 31 (2001) 23.
[25] The coverage ratios of modified surfaces by alkenes, S1c and S1e–f
are lower than those modified by the corresponding alkynes (Si(111)–
Michalak, M.C. Traub, A.S.Y. Chan, B.S. Brunschwig, N.S. Lewis,
J. Phys. Chem. B 109 (2005) 3930.
11] F. Cattaruzza, A. Cricenti, A. Flamini, M. Girasole, G. Longo, A.
Mezzi, T. Prosperi, J. Mater. Chem. 14 (2004) 1461.
CH@CHCO
2
CH
2
CF
3
,
2 2 6 4
56%; Si(111)–CH@CHCO CH –C H -p-
CF , 56%; Si(111)–CH@CHCN, 51%) [8]. This possibly indicates
3
[
[
[
[
12] Y. Liu, S. Yamazaki, S. Yamabe, J. Org. Chem. 70 (2005) 556.
13] B. Giese, Angew. Chem., Int. Ed. Engl. 22 (1983) 753.
14] C. Chatgilialoglu, Chem. Rev. 95 (1995) 1229.
that the reactivity of alkenes in this hydrosilylation is slightly lower
than the activated alkynes, which was also observed for the model
reactions. Other factors such as the packing ability may be related to
the surface coverage. The reaction of Si(111)–H with a nonconju-
gated alkene (H C@CH(CH ) CO CH CF ) leading to 14% coverage
2 2 3 2 2 3
ratio was reported [8]. The comparable low coverage for S2 may also
15] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C.
Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B.
Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene,
X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin,
R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma,
G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S.
Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G.
Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T.
Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challa-
combe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C.
Gonzalez, J.A. Pople, GAUSSIAN 03, Revision C.02, Gaussian Inc.,
Wallingford, CT, 2004.
arise from the surface steric factor.
[26] The surface reaction does not require an explicit radical initiator or
high reaction temperature, similar to the reaction of Si(111) with
propiolate esters or tris(trimethylsilyl)silane with propiolate esters
[12,8]. According to the proposed mechanism for the reaction of
tris(trimethylsilyl)silane with propiolate esters [8] autoxidation of
surface Si–H group by contaminated molecular oxygen to generate a
silicon radical on the surface may be the speculative initial step of the
total radical chain reaction.
[27] In order to examine the role of charge carriers and obtain some
insight into the reaction mechanisms, the reaction of p-type Si(111)–
H with trifluoroethyl propiolate was carried out. The reaction gave a
46% coverage ratio by F1s XPS spectra. We believe the value is
similar to that of n-Si(111) within the experimental error and the
effect of charge carriers may be small [42].
[
16] (a) J.B. Lambert, S. Zhang, S.M. Ciro, Organometallics 13 (1994)
[28] (a) A.B. Sieval, A.L. Damirel, J.W.M. Nissink, M.R. Linford, J.H.
van der Maas, W.H. de Jeu, H. Zuilhof, E.J.R. Sudh o¨ lter, Langmuir
14 (1998) 1759;
(b) R. Boukherroub, D.D.M. Wayner, J. Am. Chem. Soc. 121 (1999)
11513.
[29] (a) Handbook of X-ray Photoelectron Spectroscopy; J. Chastain, ed.;
Perkin-Elmer Corp., Physical Electronics Division: Eden Prairie,
MN, 1992;
2430;
(
(
(
(
(
b) F.H. Elsner, H-G. Woo, T.D. Tilley, J. Am. Chem. Soc. 110
1988) 313;
c) H. Bock, J. Meuret, K. Ruppert, Angew. Chem., Int. Ed. Engl. 32
1993) 414;
d) H. Bock, J. Meuret, R. Baur, K. Ruppert, J. Organomet. Chem.
446 (1993) 113;
(
e) J. Frey, E. Schottland, Z. Rappoport, D. Bravo-Zhivotovskii, M.
(b) H. Jin, C.R. Kinser, P.A. Bertin, D.E. Kramer, J.A. Libera,
M.C. Hersam, S.T. Nguyen, M.J. Bedzyk, Langmuir 20 (2004)
6252.
Nakash, M. Botoshansky, M. Kaftory, Y. Apeloig, J. Chem. Soc.,
Perkin Trans. 2 (1994) 2555.
[
[
17] DG298 value of A was previously reported as –6.1 kcal/mol [12]. This
time, slightly more stable C–O single bond rotamer was obtained.
18] When CH Cl was used as a solvent, no formation of Si(111)–Cl was
2 2
[30] (a) U. Gelius, P.F. Heden, J. Hedman, B.J. Lindberg, R. Manne, R.
Nordberg, C. Nordling, K. Siegbahn, Phys. Scripta 2 (1970) 70;
(b) X.L. Zhou, F. Solymosi, P.M. Blass, K.C. Cannon, J.M. White,
Surf. Sci. 219 (1989) 294.
[31] R. Voicu, R. Boukherroub, V. Bartzoka, T. Ward, J.T.C. Wojtyk,
D.D.M. Wayner, Langmuir 20 (2004) 11713.
[32] (a) H. Asanuma, G.P. Lopinski, H.-Z. Yu, Langmuir 21 (2005) 5013;
(b) R. Boukherroub, J.T.C. Wojtyk, D.D.M. Wayner, D.J. Lock-
wood, J. Electrochem. Soc. 149 (2002) H59.
[33] C. Gurtner, A.W. Wun, M.J. Sailor, Angew. Chem. Int. Ed. 38
(1999) 1966.
detected by XPS spectra (Cl 2p (200 eV) and Cl 2s (270 eV)) of the
surfaces S1d, S1e and S2.
[
19] For the reaction of Si(111)–H with 2b, formation of the surface S1b
was assumed from the model reaction result. The reaction of 1 and 2b
gave 3b as major product and no O-silylated product was identified.
However, because of the reported results of competing reactions of
Si(111)–H with C@C and OH groups [4c] and formation of Si–OR by
thermal reaction of Si(111)–H and alcohols, [41] a close analysis of
the surface will be needed to rule out the reaction of the OH groups
with the Si–H bonds.
[34] Compared with surface S8 (Fig. 4, Equiv.ml of SiO
intensity of oxidized Si peak in Si 2p XPS spectrum in S9a increased
(Fig. 5, Equiv.ml of SiO = 0.87).
x
[23] = 0.37), the
[
[
20] J. Zhang, C.Q. Cui, T.B. Lim, E.-T. Kang, K.G. Neoh, S.L. Lim,
K.L. Tan, Chem. Mater. 11 (1999) 1061.
x
[35] The possibility to increase the surface quality after chemical
21] The conflicting intensity of C–C and C@O in C 1s peak of XPS
spectra for surface S2 may arise from contaminated carbon. The
coverage ratio for S2 was estimated by C 1s (C@O) peak.
22] E.J. Lee, T.W. Bitner, J.S. Ha, M.J. Shane, M.J. Sailor, J. Am.
Chem. Soc. 118 (1996) 5375.
transformation was examined. Retreatment of Si(111)–CH@
CHCO
2
CH
2
CF
3
surface (22% coverage) with 5% HF (1 min) and
subsequent reaction with alkyne HC„CCO
2
CH CF gave the surface
2
3
[
[
with 38% coverage by XPS. Further study on application to various
surfaces is planned in due course.
23] The silicon oxidation was estimated from the equivalent fractional
[36] X. Zhou, M. Ishida, A. Imanishi, Y. Nakato, J. Phys. Chem. B 105
(2001) 156.
x x
monolayer coverage (Equiv.ml)of SiO by XPS [10b]. Equiv.ml of SiO

5824
Y. Liu et al. / Journal of Organometallic Chemistry 691 (2006) 5809–5824
[
[
37] Surface S10 was prepared from reactions of Si(111)–H surface with
octadecyl propiolate at room temperature for 40 h [8].
ing the concentration of the solution might cause the decrease of the
coverage ratio. Investigation of other methods to preconcentrate
other than evaporation (e.g. solid phase extraction) is required for
volatile alcohols.
38] A serial of propanol solution of octadecyl alcohol in the concentra-
ꢀ5
ꢀ4
tion range of about 4.0 · 10 ꢁ 5.0 · 10 mol/L was prepared and
measured by GC to draw the calibration plots. A linear relationship
between the GC intensity (peak integral) and the solution concentra-
tion was obtained.
[40] B. Kopping, C. Chatgilialoglu, M. Zehnder, B. Giese, J. Org. Chem.
57 (1992) 3994.
[41] R. Boukherroub, S. Morin, P. Sharpe, D.D.M. Wayner, P. Allongue,
Langmuir 16 (2000) 7429.
[
2 2 6 4 3
39] Surface S11 („Si–CH@CH–CO CH –C H -p-CF ) was also pre-
pared from reaction of Si(111)–H surface with 4-trifluoromethylb-
enzyl propiolate [8] and subsequent reduction and GC analysis were
performed. Coverage of surface S11 was estimated to be 24%, which
is quite lower than that of 56% previously reported by XPS [8]. The
possible loss of more volatile trifluoromethylbenzyl alcohol (b.p. 78–
[42] (a) Q.Y. Sun, L.C.P.M. de Smet, B. van Lagen, M. Giesbers, P.C.
Th u¨ ne, J. van Engelenburg, F.A. de Wolf, H. Zuilhof, E.J.R.
Sudh o¨ lter, J. Am. Chem. Soc. 127 (2005) 2514;
(b) Q.-Y. Sun, L.C.P.M. de Smet, B. van Lagen, A. Wright, H.
Zuilhof, E.J.R. Sudh o¨ lter, Angew. Chem., Int. Ed. 43 (2004) 1352;
(c) M.P. Stewart, J.M. Buriak, J. Am. Chem. Soc. 123 (2001) 7821.
80 ꢁC/4 mmHg) than 1-octadecanol (b.p. 170–171 ꢁC/2 mmHg) dur-