Self-assembled monolayers of double-chain disulfides of adenine on Au: An IR-UV sum-frequency generation spectroscopic study

Article abstract of DOI:10.1021/la9021992

We have synthesized double-chain disulfides of adenine with different chain lengths (n = 2,4,5,9, and 10) and studied their self-assembled monolayers (SAM) on gold surface by IR-UV doubly resonant sum frequency generation spectroscopy with the help of DFT calculation. A versatile way to investigate the orientation angle of functional groups of SAMs and their surface coverage has been demonstrated. It was revealed that the IR dipoles of the band at around 1630 cm^-1, which were almost parallel to the long molecular axis of the adenine ring, were less tilted with respect to the substrate surface in the SAMs with longer chains (n=9 and 10) in comparison to those with shorter chains (n=2,4, and 5).

Full text of DOI:10.1021/la9021992

pubs.acs.org/Langmuir
©
2009 American Chemical Society
Self-Assembled Monolayers of Double-Chain Disulfides of Adenine on Au:
An IR-UV Sum-Frequency Generation Spectroscopic Study
‡
Tetsuhiko Nagahara, Takumi Suemasu, Misako Aida, and Taka-aki Ishibashi*
Center for Quantum Life Sciences and Graduate School of Science, Hiroshima University, Kagamiyama,
Higashi-Hiroshima, 739-8530, Japan. Present address: Department of Chemistry and Materials Technology,
Kyoto Institute of Technology, Matsugasaki, Kyoto 606-8585, Japan.
‡
Received June 18, 2009. Revised Manuscript Received September 2, 2009
We have synthesized double-chain disulfides of adenine with different chain lengths (n=2, 4, 5, 9, and 10) and studied
their self-assembled monolayers (SAM) on gold surface by IR-UV doubly resonant sum frequency generation
spectroscopy with the help of DFT calculation. A versatile way to investigate the orientation angle of functional groups
of SAMs and their surface coverage has been demonstrated. It was revealed that the IR dipoles of the band at around
-
1
1
630 cm , which were almost parallel to the long molecular axis of the adenine ring, were less tilted with respect to the
substrate surface in the SAMs with longer chains (n=9 and 10) in comparison to those with shorter chains (n=2, 4, and 5).
Introduction
disulfides of adenine (A) derivatives of various alkyl chain lengths
[
compounds 3, A-(CH ) -S-S-(CH ) -A; n=2, 4, 5, 9 and 10]
2
n
2 n
Molecular recognition and hydrogen bond formation between
nucleic acid bases have been extensively studied in connection to
Watson-Crick base pairing in DNA or to molecular recognition,
for example, enzymes and cofactors. Interactions between bases
on the surface of self-assembled monolayers (SAMs) and com-
plementary bases in a solution or gas phase were also studied in
and examined their SAMs on gold surfaces.
Orientations and conformations of SAM forming molecules
can bestudied bysurface specific vibrational spectroscopies which
probe vibrational resonances of functional groups in the molecule
on the surface, such as reflection absorption IR spectroscopy
(RAIRS) and vibrationally resonant sum-frequency generation
spectroscopy (SFG), since these spectroscopies utilize polarized
1-4
this regard.
The attractive interactions between complemen-
tary bases are applicable to chemical force microscopy or adhe-
sion map AFM, which may enable the reading of sequences of a
DNA directly in the near future by forminga SAM of a baseon an
6-9
light.
We attempted to examine SAMs of alkylated adenines
by RAIRS at first, but the spectra obtained were poor in quality
for determination of molecular orientation because of a high and
sloping background that could not be subtracted in the finger-
print region.
In contrast, we obtained vibrational spectra of the SAMs,
which were in high quality and were affected little by the back-
ground, by DR-SFG spectroscopy which utilizes both IR and UV
polarized lasers and which enabled us to probe the bases in
SAMs selectively with high sensitivity rather than to probe the
whole SAM forming molecules because of the electronic resonant
5
AFM tip.
On the basis of these viewpoints, SAMs of adenine derivatives
on gold substrates have been studied. It is known that optimal
surface density of adenine is crucial for the specific binding of
uracil, which is an analogue of complementary Watson-Crick
counterpart of thymine, to adenine since adenine-adenine homo-
interactions and interactions between adenine and hydrophobic
3
,4
groups in the SAM may hinder the adsorption of uracil. It is
also known that there exists some non-Watson-Crick binding
modes (e.g., Hoogsteen binding mode), in which different sets of
hydrogen-bonding sites function and which may alter the binding
specificities of the bases. That is to say, the binding sites of the
SAM should be directed to the binding species. Thus, it is
important to estimate orientations of the hydrogen-bonding sites
or the adenine ring in the SAMs. The purpose of this study is to
demonstrate the effectiveness of IR-UV doubly resonant sum-
frequency generation (DR-SFG) spectroscopy as a tool to in-
vestigate the orientation angle of aromatic functional groups,
especially DNA bases, in SAMs. We synthesized double-chain
1
0-13
effect.
We investigated the orientation of adenine moieties at
the top of the SAMs of different lengths of alkyl chains by
analyzing the DR-SFG spectra with the help of vibrational
analyses using quantum chemical calculations.
Experimental Section
General. Water was purified to a resistivity of ∼18 MΩ cm
by a water purification system (Arium 611UV, Sartorius). All
organic solvents and reagents were used without further purifica-
tion unless otherwise noted. H NMR spectra of dimethylsulf-
oxide (DMSO) d or CDCl solutions were recorded on a JEOL
1
6
3
*
To whom correspondence should be addressed Fax: þ81-82-424-0727.
E-mail: taib@hiroshima-u.ac.jp.
1) Kurihara, K.; Ohto, K.; Honda, Y.; Kunitake, T. J. Am. Chem. Soc. 1991,
13, 5077–5079.
2) Calvo, E. J.; Rothacher, M. S.; Bonazzola, C.; Wheeldon, I. R.; Salvarezza,
R. C.; Vela, M. E.; Benitez, G. Langmuir 2005, 21, 7907–7911.
(6) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 52–66.
(7) Ihs, A.; Uvdal, K.; Liedberg, B. Langmuir 1993, 9, 733–739.
(8) Wang, J.; Chen, C.; Buck, S. M.; Chen, Z. J. Phys. Chem. B 2001, 105, 12118–
(
1
12125.
(
(9) Zhuang, X.; Miranda, P. B.; Kim, D.; Shen, Y. R. Phys. Rev. B 1999, 59,
12632–12640.
(
(
3) Matsuura, K.; Ebara, Y.; Okahata, Y. Langmuir 1997, 13, 814–820.
(10) Raschke, M. B.; Hayashi, M.; Lin, S. H.; Shen, Y. R. Chem. Phys. Lett.
2002, 359, 367–372.
4) Weisser, M.; K €a shammer, J.; Menges, B.; Matsumoto, J.; Nakamura, F.;
Ijiro, K.; Shimomura, M.; Mittler, S. J. Am. Chem. Soc. 2000, 122, 87–95.
5) Sunami, H.; Ijiro, K.; Karthaus, O.; Kraemer, S.; Mittler, S.; Knoll, W.;
Shimomura, M. Mol. Cryst. Liq. Cryst. 2001, 371, 151–154.
(11) Ishibashi, T.; Onishi, H. Appl. Phys. Lett. 2002, 81, 1338–1340.
(12) Ishibashi, T.; Onishi, H. Appl. Spectrosc. 2002, 56, 1298–1302.
(13) Maeda, T.; Ishibashi, T. Appl. Spectrosc. 2007, 61, 459–464.
(
Langmuir 2010, 26(1), 389–396
Published on Web 09/22/2009
DOI: 10.1021/la9021992 389

Article
Nagahara et al.
2 n
-S-S-(CH ) -A; n = 2, 4, 5, 9, and 10]
Scheme 1. Synthesis of Double-Chain Adenine Disulfides [A-(CH
2
)
n
GSX500 (500 MHz) spectrometer. Chemical shifts were reported
in parts per million relative to tetramethylsilane. IR spectra of
adenine disulfides were observed as KBr pellets or solutions of
DMSO on a Nicolet 6700 FT-IR spectrometer (Thermoelectron)
equipped with an LN-cooled MCT detector. Reflection absorp-
tion IR spectra (RAIRS) of the adenine SAMs on gold substrates
were recorded with the FT-IR spectrometer equipped with
a reflectance accessory (Specac). The incident angle of the
P-polarized probe beam was set at 80 degrees from the surface
normal. Raman spectra of the adenine disulfides were observed as
powder on a Raman spectrometer (NRI-1866M, JASCO)
angles of UV and IR beams were 70 and 50 degrees relative to the
surface normal, respectively.
Cyclic Voltammetry. Reductive desorption of SAMs on the
substrates, which were mounted at a hole of an electrochemical
cell by an elastic O-ring, were measured using a potentiostat
(Hokuto Denko). The surface area of the electrode was calcu-
2
lated to be 0.5 cm from the diameter of the O-ring. A basic
solution of aqueous KOH (0.5 M) in the cell was degassed with
Ar for 15 min prior to the measurements. The potential (E) is
referred to an Ag|AgCl electrode. Scan rate was set to a value of
-1
5
0 mV s
.
þ
equipped with an Ar ion laser (514.5 nm, Spectra Physics). UV
Materials. Double-chain adenine disulfides were synthesized
absorption spectra of adenine disulfide solutions in a 1-mm
cuvette were recorded by an absorption spectrometer (V-570,
JASCO). Vibrational analyses using quantum chemical calcula-
tions of model molecules were carried out by the B3LYP method
with the basis set 6-31G* using the Gaussian 03 program pack-
by alkylation of adenine followed by subsequent introduction of
thiosulfate to form disulfides (see below) as shown in Scheme 1.
SAMs of the alkylated adenines were formed on gold-on-mica
substrates (Picosubstrate, Molecular Imaging) by immersing
the substrates in ∼1 mM ethanol solutions of the disulfides
1
4
age. The harmonic wavenumbers, the IR transition dipoles, and
the preresonance Raman tensors excited at 260 nm were calcu-
lated on an energy-optimized geometry of the molecule. The
harmonic wavenumbers were scaled by the wavenumber-linear
(
compounds 3) for about 12 h unless otherwise stated. It is
known that molecules having disulfide groups form chemisorbed
To minimize contam-
1
6,17
SAMs on gold as thiolates efficiently.
1
5
ination, the gold substrates were cleaned by soaking in a chromic
acid solution (Wako Pure Chemical), which strongly oxidizes
organic contaminants, for about 12 h prior to use and then
rinsed with ultrapure water and ethanol. Immediately after the
cleaning, they were immersed into the ethanol solutions of
adenine disulfides for deposition. After removal from the solu-
tion, the substrate was rinsed thoroughly with copious ethanol
to remove physisorbed molecules and was brown dried with
nitrogen gas.
scaling (WLS) method.
SFG Apparatus. The details of our SFG spectrometer using
wavelength-tunable optical parametric amplifiers (OPAs) and
1
1-13
multiplex detection scheme can be found elsewhere.
our light sources were based on an amplified Ti:Sapphire laser
2.5 mJ, 1 kHz, TITAN-II, Quantronix). Femtosecond broad-
Briefly,
(
-1
-1
band IR probe pulses (∼2 μJ at ∼1600 cm , ∼200 cm fwhm)
-
1
and picosecond narrow-band UV pulses (∼0.1 μJ, ∼8 cm
fwhm) were generated by using femtosecond and picosecond
OPAs (TOPAS, Light Conversion), respectively. It should be
noted that an SFG spectrometer that is both IR and UV
wavelength tunable is indispensable to satisfy the doubly resonant
condition of the adenine moiety. The two input probe pulses were
overlapped spatially and temporally on the sample, and the sum-
frequency (SF) signal generated was analyzed with a spectrograph
Synthesis of Compounds 1. Alkylhalides of adenine [com-
pounds 1, A-(CH
2 n
) -Br; n = 2, 4, 5, 9, and 10] were synthe-
18,19
sized as described in the literature.
Adenine (10 mmol)
was alkylated with R,ω-dibromoalkanes (45 mmol) in DMF
(75 mL) in the presence of potassium carbonate (30 mmol) under
nitrogen atmosphere. The solutions were stirred at room
temperature for 24-48 h. Compounds 1 were extracted with
chloroform and washed with saturated NaCl aq. After evapora-
tion of the solvent, the products [A-(CH ) -Br; n=2, 4, 5, 9,
[asymmetric double spectrograph consisting of a prism premono-
chromator stage (CT-25UV, JASCO) and a grating polychroma-
tor stage (TRIAX550, Horiba Jobin Yvon)]/LN-cooled CCD
2
n
(Roper Scientific) combination. A GaAs(110) wafer was used as a
reference of the SFG signal. The polarization directions of SF,
UV, and IR beams were P, P, and P, respectively. The incidence
and10] were purified by a preparative silica gel column (eluted by
ethyl acetate-methanol mixture=85:15) and by crystallization
from methanol.
1
6
n=2:. 82% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.17
1H, s, CH), 8.14(1H, s, CH), 7.25(2H, s, NH ), 4.57(2H, t, CH ),
(
2
2
(
14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
3
.95 (2H, t, CH
n=4:. 41% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.15
1H, s, CH), 8.14(1H, s, CH), 7.20(2H, s, NH ), 4.18(2H, t, CH ),
), 1.75 (2H, q, CH ).
n=5:. 72% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.15
1H, s, CH), 8.12(1H, s, CH), 7.18(2H, s, NH ), 4.14(2H, t, CH ),
.51 (2H, t, CH ), 1.82 (4H, m, CH ), 1.35 (2H, q, CH ).
2
).
Cheeseman, J. R.; Montgomery, J.A. J.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski,
V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul,
A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Gonzalez, C.; Pople, J. A., Gaussian 03, revision D.01; Gaussian:
Wallingford, CT, 2004.
1
6
(
2
2
3.55 (2H, t, CH
2
), 1.93 (2H, q, CH
2
2
1
6
(
3
2
2
2
2
2
(16) Taniguchi, I.; Toyosawa, K.; Yamaguchi, H.; Yasukouchi, K. J. Chem.
Soc., Chem. Commun. 1982, 18, 1032–1033.
(17) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481–4483.
(18) Fkyerat, A.; Demeunynck, M.; Constant, J.-F.; Michon, P.; Lhomme, J.
J. Am. Chem. Soc. 1993, 115, 9952–9959.
(
15) Yoshida, H.; Takeda, K.; Okamura, J.; Ehara, A.; Matsuura, H. J. Phys.
Chem. A 2002, 106, 3580–3586.
(19) Itahara, T. J. Chem. Soc., Perkin. Trans. 1998, 2, 1455–1462.
3
90 DOI: 10.1021/la9021992
Langmuir 2010, 26(1), 389–396

Nagahara et al.
Article
Figure 1. UV absorption spectrum of a representative double-chain
-
5
disulfide of adenine (n=2) in ethanol (7.8ꢀ10 M). The spectra of
unsubstituted adenine and other adenine derivatives with different
chain lengths (not shown) were almost identical in this spectral
region.
1
n=9:. 67% yield; H NMR (500 MHz, CDCl
3
, δ ppm) 8.37
(
1H, s, CH), 7.80(1H, s, CH), 5.59(2H, s, NH ), 4.19(2H, t, CH ),
2 2
3
.40 (2H, t, CH
2
), 1.91-1.26 (14H, m, CH
2
).
1
n=10:. 73% yield; H NMR (500 MHz, CDCl , δ ppm) 8.38
3
Figure 2. DR-SFG spectra of double-chain adenine disulfides
[A-(CH ) -S-S-(CH ) -A; n = 2, 4, 5, 9 and 10] SAMs on
gold substrates.
(
1H, s, CH), 7.79(1H, s, CH), 5.59(2H, s, NH ), 4.19(2H, t, CH ),
2 2
2
n
2 n
3
.40 (2H, t, CH
Synthesis of Compounds 2. Thiosulfate compounds
compounds 2, A-(CH -S H; n=2, 4, 5, 9, and 10] were
synthesized as described in the literature.
4 mmol) and sodium thiosulfate pentahydrate (4 mmol) in
5 mL solvent (water for n=2, 4 and 5; water-methanol mix-
2
), 1.91-1.26 (16H, m, CH
2
).
[
)
2 n
2 3
O
-
1
2
0,21
Table 1. Band Positions (in cm ) and Band Intensities (Normalized
by the Reference Signal, Shown in Parentheses) Obtained for Double-
Compounds 1
(
3
Chain Disulfides of Adenine [A-(CH
2
)
n
-S-S-(CH
2
)
n
-A] SAMs
on Gold from SFG Experiments
ture = 1:1 for n = 9 and 10) were refluxed for 2.5 h. After
completion of the reaction, aqueous hydrochloric acid was added
to the reaction mixtures at room temperature, and the mixtures
were cooled on ice to obtain solid products. The solid products
n = 2
n = 4
n = 5
n = 9
n = 10
1629 (0.017) 1636 (0.024) 1637 (0.020)
1
1
581 (0.085) 1582 (0.11) 1580 (0.029) 1577 (0.0051) 1578 (0.010)
511 (0.013) 1512 (0.026) 1509 (0.014) 1504 (0.0048) 1513 (0.0049)
(
compounds 2) were washed thoroughly with ice-cold water and
dried in vacuo.
n=2:. 95% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.46
1H, s, CH), 8.43(1H, s, CH), 4.60 (2H, t, CH ), 3.29 (2H, t, CH ).
n=4:. 81% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.51
1
6
1472 (0.049) 1477 (0.010) 1481 (0.022) 1477 (0.050) 1479 (0.018)
1
1
430 (0.029) 1436 (0.062) 1435 (0.0067) 1441 (0.025) 1433 (0.065)
405 (0.0014) 1432 (0.061) 1420 (0.0069) 1413 (0.039) 1413 (0.046)
(
(
2
2
1
6
1374 (0.00088) 1374 (0.0011)
1366 (0.0085) 1363 (0.026)
1H, s, CH), 8.43(1H, s, CH), 4.25 (2H, t, CH ), 2.86 (2H, t, CH
2
2
),
1332 (0.016) 1333 (0.062) 1330 (0.067) 1334 (0.027) 1332 (0.063)
1291 (0.068) 1289 (0.044) 1301 (0.0019) 1290 (0.0091) 1293 (0.016)
1
.92 (2H, q, CH ), 1.64 (2H, q, CH ).
2
2
1
6
1249 (0.019) 1250 (0.0035)
1247 (0.0066) 1249 (0.016)
n=5:. 91% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.52
(
1
1H, s, CH), 8.45(1H, s, CH), 4.24 (2H, t, CH ), 2.51 (2H, t, CH ),
2
2
1
6
.83 (2H, q, CH
2
), 1.68 (2H, q, CH ), 1.30 (2H, q, CH
2
2
).
n=5:. 71% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.13
1
6
n=9:. 65% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.48
(2H, s, CH), 8.13(2H, s, CH), 7.14(4H, s, NH
2.63 (4H, t, CH ), 1.81 (4H, q, CH ), 1.59 (4H, q, CH
n=9:. 51% yield; H NMR (500 MHz, CDCl , δ ppm) 8.37
(2H, s, CH), 7.79(2H, s, CH), 5.74(4H, s, NH ), 4.19(4H, t, CH
2.66 (4H, t, CH ), 1.91-1.26 (30H, m, CH ).
n=10:. 58% yield; H NMR (500 MHz, CDCl
(2H, s, CH), 7.79(2H, s, CH), 5.79(4H, s, NH ), 4.19(4H, t, CH
2.66 (4H, t, CH ), 1.91-1.26 (32H, m, CH ).
2
), 4.13(4H, t, CH
2
),
(
1
1H, s, CH), 8.43(1H, s, CH), 4.23 (2H, t, CH ), 2.66 (2H, t, CH
2
2
),
2
2
2
).
1
.81 (2H, q, CH ), 1.58 (2H, q, CH
2
2
), 1.33-1.20 (10H, m, CH
2
).
3
1
6
n=10:. 57% yield; H NMR (500 MHz, DMSO d , δ ppm)
.51 (1H, s, CH), 8.45 (1H, s, CH), 4.22 (2H, t, CH ), 2.65
), 1.81 (2H, q, CH ), 1.58 (2H, q, CH ), 1.33-1.20
12H, m, CH2).
2
2
),
8
2
2
2
1
(
(
2H, t, CH
2
2
2
3
, δ ppm) 8.37
),
2
2
Synthesis of Compounds 3. Disulfide compounds [compounds
, A-(CH ) -S-S-(CH ) -A; n = 2, 4, 5, 9, and 10]
2
2
3
2
n
2 n
2
1
were synthesized as described in the literature. Aqueous sod-
ium borohydride (10%) was poured into solutions of com-
pounds 2 (0.7 mmol) in 5 mL solvent (water for n=2, 4 and 5;
water-methanol mixture=1:1 for n=9 and 10), and the mixtures
were stirred for 6 h. The products (compounds 3) were filtered,
washed thoroughly with ice-cold water and purified by crystal-
lization from chloroform.
Results and Discussion
UV absorption spectrum of double-chain disulfide of adenine
compounds 3, n = 2) in ethanol was measured as shown in
Figure 1. The spectrum maximized at 261 nm and was almost
identical to that of unsubstituted adenine. In the following part of
the SFG experiments, we fixed the UV probe wavelength at
(
1
6
n=2:. 90% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.13
2
71 nm to satisfy an electronic resonant condition of adenine
(
3
1H, s, CH), 8.13(1H, s, CH), 7.21(2H, s, NH
.25 (2H, t, CH2).
2
2
), 4.42(2H, t, CH ),
moiety at a sum-frequency wavelength (∼260 nm).
1
6
As mentioned before, the orientations of functional groups of
molecules in SAMs can be studied, in principle, by RAIRS. We
attempted to do this, but instead, we obtained spectra by RAIRS
with poor quality in the fingerprint region for determination of
adenine orientation because of a high and sloping background
that could not be precisely subtracted.
n=4:. 78% yield; H NMR (500 MHz, DMSO d , δ ppm) 8.14
(
1H, s, CH), 8.13(1H, s, CH), 7.19(2H, s, NH ), 4.16(2H, t, CH ),
2
2
2
2 2 2
.65 (2H, t, CH ), 1.86 (4H, q, CH ), 1.53 (4H, q, CH ).
(
(
20) Distler, H. Angew. Chem., Int. Ed. Engl. 1967, 6, 544–553.
21) Tsalmane, L. V.; Lidak, M. J. Bioorg. Khim. 1990, 16, 976–980.
Langmuir 2010, 26(1), 389–396
DOI: 10.1021/la9021992 391

Article
Nagahara et al.
Table 2. Harmonic Wavenumbers Obtained by DFT Calculations for
Alkylated Adenines
a
9
-methyladenine
9-ethyladenine
9-decyladenine
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
ν
1
1633
1596
1582
1515
1494
1479
1441
1419
1378
1336
1343
1320
1239
1262
1633
1593
1580
1507
1482
1481
1470
1422
1388
1360
1340
1319
1285
1256
1632
1593
1580
1507
1483
1481
1467
1422
1388
1360
1340
1320
1270
1259
1255
1244
1220
1204
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Figure 3. Raman and IR spectra of polycrystalline form of dou-
ble-chain adenine disulfides. Polycrystalline powder and those in
KBr were used in Raman and IR experiments, respectively.
1226
1203
1203
a
Vibrational modes ascribed to almost pure vibrations of alkyl chains
are omitted in this table. The wavenumbers are scaled by the WLS
method.
1
5
that of adenine. On the other hand, spectral features in the 1350-
-1
1
200 cm region change gradually with increasing chain length
up to n=5andthenchangedrasticallytocomplexonesinn=9 and 10.
The change of the spectra in the 1350-1200 cm- region may be
due to change of vibrational modes caused by different alkyl-
chain length and/or by the orientational change of adenine
moieties.
1
To gain insight into the assignments of the observed SFG
vibrational bands, we measured and compared vibrational spec-
tra (IR and Raman) of orientationally averaged (isotropic)
samples, polycrystalline solids, and DMSO solutions of adenine
disulfides of different chain lengths. The spectra of polycrystalline
samples are shown in Figure 3. Since SFG can be considered as IR
excitation followed by anti-Stokes Raman transition, one has to
consider both IR and Raman spectra. The IR spectra of a sample
in KBr and in solution (not shown) were similar.
In contrast to the SFG spectra of the SAMs, all the IR spectra
and also Raman spectra, are quite similar in the 1650-1500 cm
region, regardless of the chain length. This evidence suggests that
-1
Figure 4. (a and b) IR (in DMSO, dashed curves) and Raman
(polycrystalline, solid curves) spectra of randomly oriented sam-
ples (n = 2 and 10) and the DFT calculated wavenumbers (straight
lines). The solvent bands are subtracted in the IR spectra. (c and d)
For comparison, SFG spectra of SAM (the same as that shown
in Figure 2) are also shown. The fine structures in panel d above
the differences observed by the SFG experiments in the
1
1
650-1500 cm- region are because of the orientation changes
of adenine moiety in the SAMs.
It is not surprising that the IR spectral feature above 1650 cm
-
1
-1
1
650 cm are because of the experimental error.
is absent in the SFG spectra shown in Figure 2, since the feature is
known to be ascribed to almost pure NH vibration, which is less
likely to be enhanced by the electronic resonant effect of adenine
2
SFG spectra of adenine SAMs of different chain lengths (n=2, 4, 5,
and 10) obtained in the 1700-1200 cm-1 region are shown in
9
ring in the DR SFG. In addition, it is known that NH deforma-
2
Figure 2. The spectra were acquired several times in a day to confirm
that they are reproducible. The band positions and intensities obtained
-1
tion modes, which may be observed in the 1700-1650 cm
2
2,23
region, are sensitive to the surrounding environments.
The
are summarized in Table 1. The spectral features of SAMs with
shorter chains (n=2, 4, 5) in the 1650-1350 cm-1 region are quite
surrounding environments, in this case, are alkyl chains or
adenine rings of adjacent molecules and the solvent molecules
in polycrystalline solids and in DMSO solutions, respectively. The
different from those with longer chains (n=9, 10): Intensities are of
-1
comparable order for ∼1630 and ∼1510 cm bands for SAMs with
wavenumbers of the NH deformation mode in solutions and in
-1
2
shorter chains, while the intensities at ∼1630 cm are negligibly small
polycrystalline solids were thus slightly different, as shown in
Figures 4a and b.
-1
compared to those at ∼1510 cm for SAMs with longer chains. It is
known that vibrational bands resulting from the NH scissoring and
-1
2
Spectral features in IR and Raman below 1500 cm change
-1
22
ring vibrations of adenine appear in the 1650-1500 cm region.
The orientations of adenine in n=(2, 4, 5) and n=(9, 10) SAMs may
be thus quite different if the bands observed by SFG were ascribed to
substantially with chain length, similar to that shown in the
SFG spectra. This suggests that the SFG spectral changes below
(
23) Wiorkiewicz-Kuczera, J.; Karplus, M. J. Am. Chem. Soc. 1990, 112, 5324–
(
22) Kyogoku, Y.; Lord, R. C.; Rich, A. J. Am. Chem. Soc. 1967, 89, 496–504.
5340.
3
92 DOI: 10.1021/la9021992
Langmuir 2010, 26(1), 389–396

Nagahara et al.
Article
-
1
Table 3. Wavenumbers Obtained by Experiments and Calculations for Representative Adenine Derivatives (n = 2 and 10) in the 1650-1500 cm
region
a
DFT calculation
n = 2) (n = 10)
IR (solution)
(n = 2) (n = 10)
Raman (powder)
(n = 2) (n = 10)
DR SFG (SAM)
Assignments
(DFT calc.)
(
(n = 2)
(n = 10)
ν
1
ν
2
ν
3
ν
4
1633
1593
1580
1507
1632
1593
1580
1507
1649
1597
1573
1510
1648
1597
1573
1510
1674
1609
1576
1513
1659
1603
1573
1515
NH
ring, NH
ring, NH
ring, NH
2
1629
1581
1511
2
2
2
1578
1513
, CH
15
2
a
Assignments based on the DFT calculation are also shown. The wavenumbers obtained by DFT calculations are scaled by the WLS method.
-1
1
500 cm are not necessarily caused byorientation effects, but by
(n = 2, 4 and 5) while the bands of ν and ν are observed in
3 4
changes in vibrational modes because of different chain lengths,
and hence vibrational modes of alkyl chain may be involved in the
SAMs of longer chain lengths (n=9 and 10) by DR-SFG experi-
ments. It should be noted here that all the three vibrational bands
(ν , ν , and ν ) were observed in the spectra of isotropic samples
-1
vibrational bands below the 1500 cm region. In the following
2
3
4
-1
part of this paper, we will focus on the 1650-1500 cm region
that is expected to provide direct information on the orientation
of the adenine group.
(solution and polycrystalline solid), and the spectra were very
similar in their features regardless of the chain length. Conse-
quently, the differences in the observed SFG spectral patterns in
the 1650-1500 cm- region are not caused by the change in the
vibrational modes, but to the change in the orientation of the
adenine ring.
1
Vibrational analyses were conducted with the help of density-
functional theory (DFT) calculation. The harmonic wavenum-
bers obtainedfor 9-methyl, 9-ethyl, and 9-decyladenine, which are
scaled by the WLS method, are shown in Table 2. Vibrational
modes assigned to almost pure vibrations of alkyl chains are
omitted from the table. The agreement of the calculated wave-
numbers between 9-ethyl and 9-decyladenine is better than that
between 9-methyl and 9-ethlyadenine, except for the modes below
To elucidate the orientations of adenine ring in the SAMs from
the observed DR SFG spectra, we used IR transition dipoles and
Raman tensors of 9-ethyladenine calculated by DFT. The SFG
signal intensity (I) is described as follows:
I ꢀ jχð2Þ
j
2
ð1Þ
-
1
1
300 cm , which are quite complex and may be coupled to the
PPP
alkyl chains. This finding suggests that the ethyl group is long
enough to predict the influence of the alkyl chain to adenine
vibrations, especially for those with wavenumbers larger than
(
2)
The effective nonlinear susceptibility of χ
for the PPP
PPP
polarization combination in azimuthally isotropic case is described
8,9
-1
as follows:
1
9
500 cm . In the following part of this paper, we use
-ethyladenine as the model molecule.
ð2Þ
χ
¼
PPP
In Figure 4 and Table 3, spectra and wavenumbers of repre-
sentative adenine disulfides with shorter and longer chains, n=2
and 10, respectively, are compared with the calculated wavenum-
bers. The assignments of the modes are also shown in Table 3.
ð2Þ
-
L
XX ðωSFÞLXX ðωUVÞLZZðωIRÞcos θSF cos θUV sin θIR
χ
XXZ
ð2Þ
-
L
XX ðωSFÞLZZðωUVÞLXX ðωIRÞcos θSF sin θUV cos θIR
χ
-1
XZX
It is clear that the ∼1510 cm bands observed are assigned to
-
1
-1
-1
ð2Þ
ν (1507 cm ) and the ∼1580 cm bands to ν (1580 cm ),
4
3
þLZZðωSF ÞLXX ðωUVÞLXX ðωIRÞsin θSF cos θUV cos θIR
χ
ZXX
respectively, since the wavenumbers obtained by experiments and
calculation coincide very well. The IR spectra of KBr pellets
ð2Þ
þLZZðωSF ÞLZZðωUV ÞLZZðω Þsin θ sin θUVsin θIR
IR
SF
χ
ZZZ
(
Figure 3, right panel) change with increasing chain length, while
ð2Þ
the spectra of DMSO solutions (Figures 4a and b) are very similar
irrespective of the chain length. The change observed in KBr may
be the result of interactions with surrounding alkyl chains.
where the ω and θ values are the respective frequencies and beam
angles from the surface normal, and the L values are Fresnel factors
related to the refractive indices and the angles; a laboratory
Cartesian coordinate system (X, Y, Z) is chosen so that Z is along
the surface normal and X is in the plane of incidence. LXX(ω ) and
LYY(ω ) were evaluated to be negligibly small compared to
-1
The bands in IR/Raman spectra (1648-1674 cm ), which are
absent in the SFG spectra, are assigned to the calculated ν mode
1
-1
(
1633 cm ), since ν is assigned to almost pure vibration of NH
1
2
IR
which may not resonate electronically in SFG experiments. The
difference between the observed and the calculated wavenumbers of
IR
LZZ(ω ) because the substrate is gold. Thus we calculated the
first and forth term in eq 2 in our experimental conditions by
IR
the ν mode can be explained by the different environments around
8,9
1
the NH2 groups as described before. Regarding the different
24
using the refractive index of gold. As a result, we obtained
environments around the NH groups, Kyogoku et al. reported
2
that the scissoring mode of hydrogen bonded 9-ethyladenine was
ð2Þ
¼ -0:266χð2Þ
þ 1:256χð2Þ
χ
ð3Þ
-1
-1 22
PPP
XXXZ
ZZZ
located at 1642 cm , while that of the free form was at 1629 cm
.
-1
We assigned the remaining bands observed at 1629 cm by SFG
(
2)
and observed at 1597-1609 cm-1 by IR/Raman to ν . The
The nonlinear susceptibility, χ , of the laboratory coordinates is
related to a second-order molecular hyperpolarizability tensor of a
molecular frame (l, m, n), β , as follows:
lmn
ijk
2
discrepancies between the observed (in powder and SAMs) and
(2)
the calculated wavenumbers (in a vacuum) of the ν mode, in which
2
NH vibration may be involved in part (see assignments in Table 2),
can also be explained by the different environments around the NH2
2
X
ð2Þ
ð2Þ
lmn
Æð i^ lÞðj m^ Þðk n^ Þæβ
^
^
^
χ
ꢀN
ð4Þ
ijk
3
3
3
groups. The observed shifts of ν band between the polycrystalline
l
,
m
,
n
2
and solution samples (Figure 4a and b) support these findings.
From Figures 2 and 4, it is clear that the vibrational bands
of ν , ν and ν are observed in SAMs of shorter chain lengths
(
24) Lynch, D. W.; Hunter, W. R. In Handbook of Optical Constants of Solids;
Palik, E. D., Eds.; Academic. Press.: San Diego, CA, 1985; pp 286-295.
2
3
4
Langmuir 2010, 26(1), 389–396
DOI: 10.1021/la9021992 393

Article
Nagahara et al.
Figure 6. Calculated SFG intensities of the ν
against tilt angle, θ. Solid (green), dotted (blue), and dashed curves
red), respectively, indicate calculated SFG intensities of the ν , ν
and ν modes against the tilt angle θ.
2 3 4
, ν , and ν modes
(
2
3
,
4
rotation matrix T as follows:
2
3
cos θ
0
1
0
sin θ
0
0
μ
¼
¼
Tμ
6
7
5
;
t
T ¼ 4
0
ð7Þ
Figure 5. Molecular model used to analyze the orientations of
adenine ring from the observed SFG spectra. (a) The standard
orientation of the adenine moiety and molecular Cartesian coordi-
nate system (ξ, η, ζ) are chosen. (b) The adenine group is rotated by
tilt angle (θ) around η axis. The tilt angle (θ) of adenine from the
surface normal, Z=ζ, is defined as shown in the figure. Arrows
0
A
TA T
-sin θ
cos θ
By assuming the adenine plane to be perpendicular to the
substrate surface (see below) and random orientation about Z=ζ
axis (azimuthally isotropic), the nonlinear susceptibilities shown
on the right-hand side of eq 3 are described as follows:
indicate IR transition dipoles of the ν
the IR transition moments of ν and ν
that of ν
2
, ν
3 4
, and ν modes. Note that
3
4
are almost perpendicular to
2
.
ð2Þ
0
0
ζζ
μ
ζ
χ
χ
ꢀ
ꢀ
A
ZZZ
where N is the surface density of the molecules, and the angular
bracket denotes an average over the molecular orientations. The
hyperpolarizability is described as follows:
0
0
ð8Þ
A
þ A ηη
ð2Þ
ξξ
0
μ
XXZ
ζ
2
The schematic illustration of the molecular orientation model
used in this paper is shown in Figure 5b. The rotation angle of the
adenine ring around the C4-C5 axis is fixed in the model. By
using eqs 1 and 3-8 and assuming δ-function distribution of θ,
SFG intensities of the ν , ν , and ν modes are simulated against
ð2Þ
β
¼ Almμn
ð5Þ
lmn
where Alm and μ denote anti-Stokes Raman tensor and infrared
n
2
3
4
8
,9,25
transition dipole, respectively. From eq 5, SFG can be considered
as an infrared excitation of a vibrational mode followed by an anti-
Stokes Raman transition. Therefore, the electronic resonance
effects of DR-SFG at sum-frequency wavelengths arise from the
Raman tensor.
We choose the standard orientation of the adenine moiety and
molecular Cartesian coordinate system (ξ, η, ζ) as shown in
Figure 5a, such that ξ is along the line from the center of positions
θ in a similar manner described in the literature (Figure 6).
By comparing the SFG intensity ratios of the ν and ν modes,
2
3
I /I (and alsoI /I for the ν and ν modes), obtained from the
ν2 ν3
ν4 ν3
4
3
experiment (Table 1) and from the simulation (Figure 6), tilt
angles of 20-30 or 40-45 degrees for n = (2, 4, 5) and 0-10
degrees for n=(9, 10), samples are deduced. Although we cannot
uniquely specify the tilt angle for n=(2, 4, 5) samples, we can
obtain valuable information from them. It has thus been con-
cluded that the tilt angle of adenine with longer chains [0-10
degrees for n=(9, 10)] are less than those with shorter chains
[20-30 or 40-45 for n=(2, 4, 5)]. It should be noted that the IR
1
and 2 to position 8 in the adenine ring, and the ξζ plane is
parallel to the adenine ring. The calculated IR transition dipoles
μ) and the Raman tensors (A) of ν , ν , and ν modes at the
(
2
3
4
standard orientation are represented in the molecular Cartesian
dipole of the ν mode is located almost parallel to the substrate
2
coordinate system as follows:
surface at a tilt angle of 0-10 degrees.
SFG intensities are in proportion to a square of the surface
2
4
2
4
2
4
3
5
3
5
3
5
2
4
3
5
2
(2) 2
coverage (N ), as well as to the component of |β | as shown in
eqs 1 and 4. Therefore, once the orientation angles of adenine ring
are determined, their relative molecular area densities can be
deduced based on the orientation angles and relative SFG
intensities (of a certain mode) of different SAMs.
-
10:0
-0:55
:34
0:04
1:59
51:4
0:58
0:44
-0:15 -0:04
-23:9
v
v
v
2
3
4
:
:
:
,
,
,
1
-23:3 -0:15 9:26
14:5 0:05
0:05 -0:07 -0:31
0:10 -0:31 -7:94
2
3
-
0:10
4
5
ð6Þ
Ideal SFG intensities per molecule in an arbitrary unit, which is
deduced from the DFT calculations and the model (see previous),
were calculated for the obtained tilt angles (θ). The ratios of the
6
:05
2
4
3
-
0:32
4
0:42
-49:1 -0:50 32:9
5
-0:69 0:14
0:48
-23:1
-5
ideal intensities of 0-10 and 20-30 degrees were 3.4 ꢀ 10 -0.1,
:32
33:4 0:69
2
.0-12, and 1.2-74, while those of 0-10 and 40-45 degrees were
When the adenine group is rotated by θ around η, the resultant
IR dipoles (μ ) and Raman tensors (A ) are obtained by using
(
25) Hirose, C.; Akamatsu, N.; Domen, K. Appl. Spectrosc. 1992, 46, 1051–
0
0
1072.
3
94 DOI: 10.1021/la9021992
Langmuir 2010, 26(1), 389–396

Nagahara et al.
Article
Figure 7. RAIRS spectra of an adenine SAM (n = 10, solid curve)
and an undecanethiol (UDT, dotted curve) SAM on gold.
Figure 8. Surface coverage of SAMs at different immersion times
i
(t = 2 s, 1 h, and 12 h) determined by the desorption peaks in the
cyclic voltammograms. Open circles with solid lines and open
squares with dashed lines indicate the surface densities of short-
chain (n = 2) and long-chain (n = 10) SAMs of adenine at the
specified times, respectively. The surface density of the long-chain
Peak position of CH antisymmetric stretching band of the ade-
2
-
1
nine SAM (2928 cm ) is higher than that of the UDT SAM
-
1
2920 cm ).
(
-
5
1
.8 ꢀ 10 -0.67, 2.0-12, and 4.7-560 for the ν , ν , and ν
2
3
4
i
SAM increases with increasing t , while that of the short-chain
SAM remains constant.
modes, respectively. By comparing these values with the experi-
mentally obtained SFG intensity ratios calculated from Table 1,
relative molecular area densities of the SAMs with different chain
lengths were estimated. The estimation is based on the fact that
the quotient of the experimental band intensity ratio to the ideal
intensity ratio is proportional to the squares of the surface
coverage of the SAMs. The estimated surface coverages of
the SAMs revealed that the densities of the SAMs with longer
chains [n=(9, 10)] were less than 60% of those with shorter chains
2
7,28
with respect to the short-chain SAM.
Some interactions other
than the steric hindrance, for example, π-π interaction between
adenines and/or chain-chain interaction, may also be involved in
our adenine SAMs.
It was reported that the SAM forming process, in general,
comprises two steps: The initial step of diffusion controlled
adsorption is followed by the second process of crystallization
2
9
[
n=(2, 4, 5)].
of alkyl chains. We speculate that the chemisorption process
during SAM formation is highly disturbed by the disordered alkyl
chain in long-chain adenine disulfides compared to that in short-
chain ones, and the disordered structures cannot be crystallized to
form a densely packed crystalline-like molecular assembly at the
surface in the second step due to the bulky terminal group, in
contrast to the unsubstituted alkane thiolates.
In general, long-chain alkane thiols form densely packed
crystalline-like structures with extended alkyl chains tilted at an
2
6
angle of 20-30 degrees. This tilt angle of the all-trans alkyl
chain corresponds to 0-10 degrees for the tilt angle of adenine (θ)
in our model, which apparently agrees with that obtained for our
long-chain SAMs. We compared RAIRS spectrum of adenine
SAM with that of undecanethiol (UDT) in the CH stretching
The mechanism we described above was confirmed by compar-
2
region (Figure 7). The peak position of CH2 antisym-
ing the SFG spectra of different immersion times (t ), 1 and 12 h.
i
metric vibration of the adenine SAM (n = 10) was located at
The spectra of short-chain SAMs (n = 2) of different t s were
identical, while all the band intensities increased by a factor of
i
-1
-1
2
928 cm , while that of UDT SAM was at 2920 cm . The peak
position of the short-chain (n=5) adenine SAM was located at
∼1.5 with increasing t in the spectra of long-chain SAMs (n=10).
i
-1
2
927 cm (not shown), while that of hexanethiol was reported to
It was found that the molecular orientations were not affected by
the immersion times in this time regime, since the spectral shapes
of SAMs of respective chain lengths were very similar and
independent of the immersion times. From eqs 1 and 4, the
increase inband intensities corresponds to an increase of∼22% of
the surface density in the duration for the long-chain SAM. Thus
the SAM formation process in the long-chain SAMs should be
highly disturbed and slower compared to the short-chain SAMs.
To make sure that the relative density estimation from the SFG
results are correct, we measured electrochemical desorption of
-1 26
be at 2921 cm
.
It is known that peak position of the
antisymmetric CH vibration in the liquid phase is higher than
2
2
6
that in the crystalline phase; therefore, our observations suggest
that the alkyl chains of the n=10 and n=5 SAMs are more likely
to be in an unorganized liquid-like state than in a crystalline
phase. In this connection, Weisser et al. also reported an
4
unorganized SAM consisting of adenine thiol. Although our
observationssuggest that alkyl chains of the long-chainand short-
chain adenine SAMs form liquid-like rather than crystalline-like
2
6
structure of unsubstituted alkane thiol SAMs, the difference of
the peak positions between the unsubstituted and the adenine
SAMs are smaller in the short-chain SAMs suggesting more dense
structures for the short-chain SAMs. This is qualitatively con-
sistent with the results on the relative surface coverage obtained
from the SFG spectra.
SAMs of different t s (2 s, 1 h, and 12 h) by using cyclic
i
3
0
voltammetry, which can quantify the thiolates. The reductive
desorption of the SAMs of the disulfides were observed at
approximately E = -0.9 and -1.0 V for n = 2 and 10 SAMs,
respectively. The molecular densities of SAMs were calculated
from the area under the peaks in the cyclic voltammograms and
the surface area of the electrodes by assuming completely flat Au
surface (Figure 8). Regardless of immersion times, the surface
One might expect that steric hindrance between the bulky
adenine moieties may be involved in the unorganized liquid-like
alkyl chain in the adenine SAM. However, our results on relative
surface coverage are in considerable contrast to that reported for
the bulky cyclodextrin SAMs with different chain lengths, where
higher density structures were suggested for the long-chain SAM
densities of short-chain SAMs are very similar even at t =2 s. The
i
(27) Weisser, M.; Nelles, G.; Wohlfart, P.; Wenz, G.; Mittler-Neher, S. J. Phys.
Chem. 1996, 100, 17893–17900.
(
(
28) Qian, J.; Hentschke, R.; Knoll, W. Langmuir 1997, 13, 7092–7098.
29) Ulman, A. Chem. Rev. 1996, 96, 1533–1554.
(
26) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem.
(30) Walczak, M. M.; Popenoe, D. D.; Deinhammer, R. S.; Lamp, B. D.;
Soc. 1987, 109, 3359–3568.
Chung, C.; Porter, M. D. Langmuir 1991, 7, 2687–2693.
Langmuir 2010, 26(1), 389–396
DOI: 10.1021/la9021992 395

Article
Nagahara et al.
density of long-chain SAM, however, increases with increasing ti;
a density increase of 24% was observed in the 1-12 h regime. The
relative densities of the short-chain and the long-chain SAMs
obtained by SFG are thus consistent with those obtained by the
quantitative electrochemical measurements.
In conclusion, we have synthesized and studied orientation
angles of adenine ring on SAMs with different alkyl chain lengths
by IR-UV DR-SFG with the help of DFT calculation. Although
the orientation angles of the base obtained may be different if
the adenine SAMs are diluted by other SAM forming molecules
(e.g., alkanethiol without adenine moiety such as UDT), since a
decrease in adenine-adenine interactions and an increase in
chain-chain interactions are expected, we are sure that our
method provides a versatile way to study the orientation angles
of the base. In future work, we would like to extend this study to
the base pairs on SAMs.
We have thus far discussed the tilt angles and the densities of
the adenine ring on the assumption that the plane of the adenine
ring is perpendicular to the substrate surface, where the rotation
angle around the C4-C5 axis (i.e., short molecular axis of the
adenine ring) is fixed. If this is not the case and the adenine group
is rotated substantially out of the ζξ plane around the C4-C5
axis, a significant decrease in the SFG intensity of ν mode would
2
be expected compared to those of ν and ν modes. This is because
the rotation decreases the Z-component of the IR transition
Acknowledgment. The authors thank Prof. Keiichi Ohno and
Dr. Yukiteru Katsumoto of Hiroshima University for FT-IR
measurements, and NBARD of Hiroshima University for NMR
and Raman spectrum measurements. The authors also thank
Prof. Kohji Maeda of Kyoto Institute of Technology for electro-
chemical measurements. This work was supported in part by
Grant-in-Aids for Scientific Research from the Ministry of
Education, Culture, Sports, Science and Technology (MEXT)
of the Japanese Government [No. 19056007, Priority Areas (477)]
and NaBiT program.
3
4
dipole for the ν mode, while it does not for ν and ν whose
2
3
4
dipoles are almost parallel to the C4-C5 axis. Therefore the
assumption may be valid, especially for the short chain SAMs
where the SFG signal of ν mode was observed.
2
For the long-chain SAMs, however, the SFG signals of the
ν2 modes were not observed, and the assumption is thus not
necessarily valid. It should be noted that there is another possible
model for the orientation of the adenine group, which can explain
the SFG spectra of the long-chain SAMs. In this alternative
model, the adenine ring is rotated around the C4-C5 axis to
decrease the μ so that the SFG signal of the ν mode almost
Supporting Information Available. Cyclic voltammograms
of short-chain (n = 2) and long-chain (n = 10) adenine
SAMs on gold-on-mica substrates at different immersion
times (t ). This material is available free of charge via the
i
Internet at http://pubs.acs.org.
z
2
disappears irrespective of the tilt angle (θ). Again the IR dipole of
the ν2 mode may be located almost parallel to the substrate
surface in long-chain SAMs. Thus it should be clear that the
dipoles of the ν mode of the long-chain SAMs are less tilted with
2
respect to the substrate surface than those of the short-chain
SAMs, regardless of the rotation angle around the C4-C5 axis.
The surface densities of adenine in SAMs, however, can not be
obtained from the observed SFG intensities by the latter model
incorporating both the rotation and the tilt.
Note Added after ASAP Publication. This article
was published ASAP on September 22, 2009. Because
of a production error, equation 7 was rendered incor-
rectly. The correct version was published on September
30, 2009.
3
96 DOI: 10.1021/la9021992
Langmuir 2010, 26(1), 389–396