Polysilane bearing "sulfide tripod" terminus: Preparation and selective chemisorption on gold surface [3]

Article abstract of DOI:10.1021/ma0212036

Polysilane with a 'sulfide tripod' terminus was synthesized and the selective chemisorption of the polysilanes on a gold surface was demonstrated. Ultraviolet absorption spectroscopy and atomic force microscopy were used for the analysis. A combination of

Full text of DOI:10.1021/ma0212036

Macromolecules 2003, 36, 9-11
9
Ch a r t 1. Syn th esis Rou te of P olysila n e Bea r in g
Su lfid e Tr ip od Ter m in u s
P olysila n e Bea r in g “Su lfid e Tr ip od ”
Ter m in u s: P r ep a r a tion a n d Selective
Ch em isor p tion on Gold Su r fa ce
Ka zu a k i F u r u k a w a ,* Keisu k e Eba ta ,
Hir osh i Na k a sh im a , Yosh ia k i Ka sh im u r a , a n d
Keiich i Tor im itsu
NTT Basic Research Laboratories, NTT Corporation,
3-1 Morinosato Wakamiya, Atsugi,
Kanagawa 243-0198, J apan
Received J uly 31, 2002
Revised Manuscript Received November 11, 2002
Research on surface-grafted polymer has been draw-
ing much attention both experimentally and theoreti-
cally for a couple of decades.1 The grafted polymer in a
high grafting density regime is often called polymer
brush and well studied for the purpose of, for instance,
surface modification. On the other hand, the grafted
polymer in a low density regime is useful for observing
structures and properties of the polymers at the single
molecule level. We have proposed the end-graft tech-
nique for fixing single polymer chains individually on a
solid surface and have demonstrated the usefulness of
this technique using polysilanes.2 We successfully vis-
ualized a variety of single molecular and supramolecular
structures of polysilane molecules by using our end-
grafted polysilane samples.3 Polysilanes have a semicon-
ducting nature based on the σ-conjugation4 and exhibit
such features as hole mobility5 and electrolumines-
cence.6 Thus, the observation of those physical proper-
ties by using the single polysilane molecule would be
also of great interest.7
We have achieved the syntheses of end-grafted poly-
silanes on solid surfaces with the hydroxyl (-OH) and
hydrosilyl (Si-H) groups. Our aim in this paper is to
synthesize end-grafted polysilane on metal surface.
Increasing the variety of surfaces is important for
research on grafted polymer. It would be also important
for studying electronic interactions between such func-
tional polymers of polysilane and the conductive metal
surface. We chose gold as the metal, because chemi-
sorption between gold and thiol has been extensively
studied and is useful for bonding molecules to the gold
surface. A self-assembled monolayer (SAM) on a gold
surface is one of the most important systems.8 The
chemisorption is also quite selective9 so that the one-
pot synthesis of a gold-thiol bond and a carboxyl acid-
indium tin oxide bond has been proposed and demon-
strated.9a
sorbed polysilanes are also confirmed by atomic force
microscopy (AFM) that clearly visualizes single molec-
ular structures of the polysilane on a gold surface.
We used two polysilanes, poly(methyl-n-propylsilane)
(1) and poly(isobutyl-n-octadecylsilane) (2). Both are
alkyl-substituted polysilanes, but their backbone rigidi-
ties are significantly different from each other.11 The
strategy for synthesizing 1b and 2b is to modify the
reaction that we have proposed (Chart 1). We use the
reaction between polysilane with a silyl anion in its
terminus, 1a and 2a , and primary alkyl bromide with
a sulfide tripod, 3. We synthesized 1a by the living
anionic polymerization of the corresponding masked
disilene2a,12 and 2a by the scission reaction of 2 with
lithium 4,4′-di-tert-butylbiphenylide.3b
We examined the chemisorption of 1b and 2b on a
gold surface and confirmed it by means of UV absorption
spectroscopy. The UV absorption band, which is as-
signed to the σ-σ* transition of the silicon backbone of
1,4a was observed only from the combination of 1b with
a gold surface as shown in Figure 1a. We did not see
any absorption band from the other two samples. This
indicates that the interaction, including physisorption
and chemisorption, between 1b and an SiO2 surface and
between 1 and a gold surface or an SiO2 surface is weak
and cannot withstand the washing processes we em-
ployed. The results clearly show that chemisorption
occurred only between the sulfide terminus of 1b and
the gold surface. Identical results were also obtained
from 2 and 2b as shown in Figure 1b, which further
confirms the selective chemisorption of the sulfide tripod
on a gold surface.
In this paper, we synthesize polysilane with a “sulfide
tripod” terminus and demonstrate the selective chemi-
sorption of the polysilane on a gold surface. As for a
similar class of linker molecule, a thiol tripod linker with
an amino functional group has been prepared and used
for an initiator of polypeptide synthesis on a gold
surface.10 The linker we designed here has a haloge-
nated alkyl functional group. This enables us to intro-
duce the sulfide tripod terminus to a semiconducting
polymer of polysilane. We show that the chemisorption
on a gold surface occurs at the sulfide tripod terminus
by means of UV absorption spectroscopy. The chemi-
Although the sulfide group is known to chemisorb on
a gold surface,13 the process seems to be slower and less
effective than with the thiol group.14 Our present results
indicate that the bond remains stable during our solvent
washing processes at room temperature. Three sulfides
on a terminus might help the stabilization of chemi-
sorbed bonds.15
Figure 2 shows AFM topographic images of a gold
surface with 1b and 2b chemisorbed on a gold surface
formed on an Si(111) wafer. With the 1b sample, we
* Corresponding author: e-mail furukawa@will.brl.ntt.co.jp,
phone +81-46-240-3551, fax +81-46-270-2364.
10.1021/ma0212036 CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/13/2002

10 Communications to the Editor
Macromolecules, Vol. 36, No. 1, 2003
be convenient for further study of conductivity at a
single molecular level.
In summary, we synthesized polysilane with a sulfide
tripod terminus. We demonstrated the selective chemi-
sorption of this terminus on a gold surface, which was
confirmed by UV absorption spectroscopy and, more
obviously, by AFM. This method enables us to connect
a semiconducting polymer terminus selectively to a gold
surface. A combination of this technique with submi-
crometer-scale gold electrodes would contribute to the
understanding of electronic properties at the single
polymer level. Furthermore, the strategy of terminating
an organic polymer chain with a sulfide tripod is general
and could be utilized in other systems.
Exp er im en ta l Section . a . Gen er a l. UV absorption
spectra were recorded using a Hitachi U3500 spectro-
photometer. AFM topographies were recorded on a Seiko
Instruments SPI 3800N atomic force microscope. The
molecular weights of the polysilanes were determined
with a Shimadzu LC-10 size exclusion chromatography
(SEC) system using THF eluent at 30 °C, calibrated by
using polystyrene standard. NMR spectra were recorded
on a Varian Unity 300 NMR spectrometer. The semi-
transparent gold film was fabricated on one side of a
quartz substrate by the conventional sputter deposition
technique. Titanium was at first deposited on the quartz
substrate, and gold was deposited on top of it. For the
AFM observations, we used NH4F-treated Si(111) pieces,
on which gold was deposited in the same manner.
b. Syn th esis. 1, 1a , 1b. The condensation of 1,2-
dichloro-1,2-dimethyl-1,2-di-n-propyldisilane (12 g, 0.05
mol) with lithium biphenylide, prepared by the reaction
between biphenyl (16 g, 0.10 mol) and Li (0.68 g, 0.10
mol) in THF (200 mL) at -78 °C, yielded the monomer
for 1, 1-phenyl-7,8-dimethyl-7,8-di-n-propyl-7,8-disila-
bicyclo[2.2.2]octa-2,5-diene (colorless oil, 12.5 g, 78%
yield, bp 120 °C/0.06 Torr). 29Si NMR (CDCl3, δ):
-18.424, -18.806 ppm. Polymer 1a was prepared by
the living anionic polymerization of the monomer (2.0
g, 6.1 mmol) by using a drop of n-BuLi as an initiator
in THF (15 mL). Adding a drop of 3 to the 1a solution
yielded 1b (colorless solid, 620 mg, yield 58%, Mw ) 6.8
× 104, Mn ) 5.5 × 104, Mw/Mn ) 1.2).16 Adding ethanol
instead of 3 quenched 1a to give 1 (colorless solid, 670
mg, yield 63%, Mw ) 2.5 × 104, Mn ) 2.2 × 104, Mw/Mn
) 1.2).
F igu r e 1. UV absorption spectra of sample substrates: (a)
poly(methyl-n-propylsilane), (b) poly(isobutyl-n-octadecylsi-
lane). Au/SiO2 indicates a quartz substrate with a gold surface
on one side.
F igu r e 2. AFM topographic images (1000 nm × 1000 nm) of
a gold surface (deposited on a flat NH4F-treated Si(111)
surface) with 1b (left) and 2b (right) chemisorbed. The images
were recorded in air at room temperature, using the tapping
mode.
2. The monomer for 2, i-C4H9-n-C18H37SiCl2, was
prepared by the condensation of n-C18H37MgCl Grignard
reagent with i-C4H9SiCl3 (colorless liquid, 69% yield, bp
140-150 °C/0.1 Torr). 29Si NMR (CDCl3, δ): -28.941
ppm. Polysilane 2 was synthesized by the sodium-
mediated (1.7 g, 0.073 mol) condensation of the mono-
mer (10 g, 0.024 mol) in refluxing n-octane (30 mL) with
a small piece of 18-crown-6. The synthesized polymer
had a bimodal molecular weight distribution, and
fractionation yielded 2 with a monomodal distribution
(colorless solid, 280 mg, yield 3.4%, Mw ) 1.3 × 107, Mn
) 2.2 × 106, Mw/Mn ) 5.8).
observed dots about 5 nm high and several tens of
nanometers in diameter that covered the surface densely
and homogeneously. This AFM image is clearly distin-
guished from those of a bare gold surface and a gold
surface dipped in a solution of 1 (without the chemi-
sorbed polysilane), both of which were observed as much
flatter surfaces. Each dot corresponds to the collapsed
structure of a 1b single molecule chemisorbed on the
surface at the terminus. The collapsed structure is a
common structure for an end-grafted flexible polymer
under poor solvent conditions including the air condi-
tion.1,3a By contrast, 2b was observed as ropes on the
gold surface. Polysilane 2 is a semiflexible polysilane,
whose main chain is very rigid.3b,11 The rope images
therefore represent the rigidity of 2. The nature of 2,
namely that it is stretched over the substrate, would
2a , 2b. Lithium 4,4′-di-tert-butylbiphenylide was
added to a hexane solution of 2 (ca. 50 mg/20 mL) until
the solution became pale yellow, yielding 2a . One drop
of 3 was added to this solution to yield 2b (colorless
solid, 36 mg, yield 70%, Mw ) 8.5 × 105, Mn ) 1.2 ×
105, Mw/Mn ) 7.3).
3. Methylthiomethyllithium, CH3SCH2Li (ca. 0.08
mol), prepared by the previously reported reaction,17
was added to a hexane (50 mL) solution of 1-bromoun-

Macromolecules, Vol. 36, No. 1, 2003
Communications to the Editor 11
E.; Yuh, H. J .; Weagley, R. J .; Stolka, M. Solid State
Commun. 1987, 62, 547.
(6) (a) Suzuki, H. Adv. Mater. 1996, 8, 657. (b) Yuan, C.-H.;
Hoshino, S.; Toyoda, S.; Suzuki, H.; Fujiki, M.; Matsumoto,
N. Appl. Phys. Lett. 1997, 71, 3326. (c) Suzuki, H.; Hoshino,
S.; Furukawa, K.; Ebata, K.; Yuan, C.-H.; Bleyl, I. Polym.
Adv. Technol. 2000, 11, 460.
(7) Furukawa, K.; Ebata, K.; Fujiki, M. Mater. Res. Soc. Proc.
Ser. 2000, 582, H10.2.1.
(8) For a review, see: Ulman, A. Chem. Rev. 1996, 96, 1533.
(9) (a) Gardner, T. J .; Frisbie, C. D.; Wrighton, M. S. J . Am.
Chem. Soc. 1995, 117, 6927. (b) Hickman, J . J .; Laibinis, P.
E.; Auerbach, D. I.; Zou, C.-F.; Gardner, T. J .; Whitesides,
G. M.; Wrighton, M. S. Langmuir 1992, 8, 357. (c) Laibinis,
P. E.; Hickman, J . J .; Wrighton, M. S.; Whitesides, G. M.
Science 1989, 245, 845.
(10) (a) Whitesell, J . K.; Chang, H.-K. Science 1993, 261, 73. (b)
Whitesell, J . K.; Chang, H.-K.; Whitesell, C. S. Angew.
Chem., Int. Ed. Engl. 1994, 33, 871.
(11) The chain rigidity of polysilanes can be estimated from their
molar absorption coefficients. Polysilane 1 has been deter-
mined as a flexible polymer. See: Fujiki, M. J . Am. Chem.
Soc. 1996, 118, 7424. The molar absorption coefficient of 2
was 48 000 cm-1 dL3 mol-1(Si unit)-1 in THF at room
temperature, from which we estimate 2 to be a semiflexible
polysilane, according to the rule discussed in the reference.
(12) (a) Sakamoto, K.; Obata, K.; Hirata, H.; Nakajima, M.;
Sakurai, H. J . Am. Chem. Soc. 1989, 111, 7641. (b) Saka-
moto, K.; Yoshida, M.; Sakurai, H. Macromolecules 1990,
23, 4494. (c) Sakamoto, K.; Yoshida, M.; Sakurai, H.
Macromolecules 1994, 27, 881. (d) Sakamoto, K.; Yoshida,
M.; Sakurai, H. Polymer 1994, 35, 4990.
decyltrichlorosilane, Br(CH2)11SiCl3 (10 g, 0.027 mol),
within 40 min. This was an exothermic reaction yielding
a colorless suspension, which was further stirred for 1
h. Then the solution was dehydrated with diluted
aqueous HCl, extracted from ethyl ether, washed with
water, neutralized with saturated NaHCO3, dried with
saturated NaCl and Na2SO4, and evaporated to yield 3
1
(pale yellow oil, 8.2 g, 68% yield). H NMR (C6D6, δ):
0.8-1.7 (22H), 1.907 (s, 9H), 1.956 (s, 6H) ppm. 13C
NMR (C6D6, δ): 12.388, 17.208, 20.266, 23.985, 28.378,
29.032, 29.613, 29.753, 29.913 (2C), 33.051, 33.639,
33.966 ppm. 29Si NMR (C6D6, δ): -3.221 ppm.
UV Stu d ies. The quartz substrates with gold on one
side were immersed in a 5 × 10-2 wt % toluene solution
of 1 and 1b for 5 min. A quartz substrate without gold
was also immersed in a solution of 1b as a reference.
The substrates were removed from the solutions, washed
repeatedly, and then immersed in fresh toluene for 1 h.
The same procedures were employed, using hexane
solutions of 2 and 2b. The background absorption,
mainly due to the deposited gold, was between 0.7 and
1.2 and possessed a smoothly curved shape. The sub-
traction of the background absorption successfully
extracted the absorption spectrum with such small
absorbance (∼0.01) of chemisorbed polysilanes on gold
surfaces.
(13) Troughton, E. B.; Bain, C. D.; Whitesides, G. M.; Nuzzo, R.
Ack n ow led gm en t. The author (K.F.) thanks Drs.
Yuzuru Iwasaki and Osamu Niwa for the semitranspar-
ent gold film fabrication. We thank Drs. Michiya Fujiki,
Masao Morita, and Hideaki Takayanagi for their en-
couragement.
G.; Allara, D. L.; Porter, M. D. Langmuir 1988, 4, 365.
(14) We attempted to introduce the thiol group at the initial stage
of our research. However, it was difficult to synthesize the
counterpart of 3 with the thiol group. The reaction yielded
a cyclic compound rather than a linear compound because
of the intramolecular reaction that occurred due to the high
reactivity of the primary alkyl bromide and thiol groups.
Refer en ces a n d Notes
(15) A question for the stability of the chemisorbed polysilane
molecules is still open. We should prepare not only 3 but
also the linkers with one or two sulfide groups for compari-
son. The washing conditions, especially the time and tem-
perature, must be considered as well for further discussions
of the stability.
(1) For a review, see: (a) Halperin, A.; Tirrell, M.; Lodge, T. P.
Adv. Polym. Sci. 1992, 100, 31. (b) Szleifer, I.; Carignano,
M. A. Adv. Chem. Phys. 1996, 94, 165.
(2) (a) Ebata, K.; Furukawa, K.; Matsumoto, N. J . Am. Chem.
Soc. 1998, 120, 7367. (b) Ebata, K.; Furukawa, K.; Matsu-
moto, N.; Fujiki, M. Polym. Prepr. 1999, 40 (2), 157.
(3) (a) Furukawa, K.; Ebata, K.; Matsumoto, N. Appl. Phys. Lett.
1999, 75, 781. (b) Furukawa, K.; Ebata, K.; Fujiki, M. Adv.
Mater. 2000, 12, 1033. (c) Furukawa, K.; Ebata, K. Appl.
Phys. Lett. 2000, 77, 4289.
(4) For a review, see: (a) Miller, R. D.; Michl, J . Chem. Rev.
1989, 89, 1359. (b) Matsumoto, N. J pn. J . Appl. Phys. 1998,
37, 5425. (c) Michl, J .; West, R. Silicon-Containing Polymers;
J ones, R. G., Ando, W., Chojnowski, J ., Eds.; Kluwer
Academic Publishers: Dordrecht, The Netherlands, 2000;
Chapter 18.
(16) 1H NMR of 1b also indicated the existence of a sulfide tripod
1
terminus. The two H NMR peaks (C6D6, δ) were observed
at 1.917 (s, 9H) and 1.966 (s, 6H), together with 0-0.7, 0.7-
1.3, and 1.3-1.8. From the observed integral areas of these
peaks, we can estimate the ratio of the sulfide tripod/
monomer unit to be 1:650 on average. This corresponds to
a molecular weight of 5.6 × 104, which is in good agreement
with the weight-average molecular weight determined by
size exclusion chromatography using the polystyrene stan-
dard.
(17) Peterson, D. J . J . Org. Chem. 1967, 32, 1717.
(5) (a) Stolka, M.; Yuh, H. J .; McGrane, K.; Pai, D. M. Chem.
Phys. Lett. 1987, 136, 451. (b) Abkowitz, M. A.; Knier, F.
MA0212036