Variable density effect of self-assembled polarizable monolayers on the electronic properties of silicon

Article abstract of DOI:10.1021/ja077933g

Electronic structures at the Si/SiO₂/molecule interfaces were studied by Kelvin probe techniques (contact potential difference) and compared to theoretical values derived by the Helmholtz equation. Two parameters influencing the electronic properties of n-type 〈100〉 Si/SiO ₂ substrates were systematically tuned: the molecular dipole of coupling agent molecules comprising the layer and the surface coverage of the chromophoric layer. The first parameter was checked using direct covalent grafting of a series of trichlorosilane-containing coupling agent molecules with various end groups causing a different dipole with the same surface number density. It was found that the change in band bending (ΔBB) clearly indicated a major effect of passivation due to two-dimensional polysiloxane network formation, with minor differences resulting from the differences in the end groups' capacity to act as electron traps . The change in electron affinity (ΔEA) parameter increased upon increasing the dipole of the end group comprising the monolayer, resulting in a range of 600 mV. Moreover, a shielding effect of the aromatic spacer compared with the aliphatic spacer was found and estimated to be about 200 mV. The density effect was examined using the 4-[4-(N,N-dimethylamino phenyl)azo]pyridinium halide chromophore which has a calculated dipole of more than 10 D. It was clearly shown that upon increasing surface chromophoric coverage an increase in the electronic effects on the Si substrate was observed. However, a major consequence of depolarization was also detected while comparing the experimental and calculated values.

Full text of DOI:10.1021/ja077933g

Published on Web 03/04/2008
Variable Density Effect of Self-Assembled Polarizable
Monolayers on the Electronic Properties of Silicon
Naama Peor, Ruthy Sfez, and Shlomo Yitzchaik*
The Institute of Chemistry and the Hebrew UniVersity Center for Nanoscience and
Nanotechnology, The Hebrew UniVersity of Jerusalem, Jerusalem 91904, Israel
Received October 16, 2007; E-mail: sy@cc.huji.ac.il
Abstract: Electronic structures at the Si/SiO2/molecule interfaces were studied by Kelvin probe techniques
(contact potential difference) and compared to theoretical values derived by the Helmholtz equation. Two
parameters influencing the electronic properties of n-type <100> Si/SiO2 substrates were systematically
tuned: the molecular dipole of coupling agent molecules comprising the layer and the surface coverage of
the chromophoric layer. The first parameter was checked using direct covalent grafting of a series of
trichlorosilane-containing coupling agent molecules with various end groups causing a different dipole with
the same surface number density. It was found that the change in band bending (∆BB) clearly indicated
a major effect of passivation due to two-dimensional polysiloxane network formation, with minor differences
resulting from the differences in the end groups’ capacity to act as “electron traps”. The change in electron
affinity (∆EA) parameter increased upon increasing the dipole of the end group comprising the monolayer,
resulting in a range of 600 mV. Moreover, a shielding effect of the aromatic spacer compared with the
aliphatic spacer was found and estimated to be about 200 mV. The density effect was examined using the
4-[4-(N,N-dimethylamino phenyl)azo]pyridinium halide chromophore which has a calculated dipole of more
than 10 D. It was clearly shown that upon increasing surface chromophoric coverage an increase in the
electronic effects on the Si substrate was observed. However, a major consequence of depolarization was
also detected while comparing the experimental and calculated values.
device performances can be tuned.24-32 Polar molecules are
often used to modify electronic properties such as barrier height
values1-6 and work functions (WF).7-10 The chemical func-
tionality directed from the surface up has a crucial influence
on the resulting surface properties, both structurally and
electronically.11 Indeed, it was already shown that organic self-
Introduction
The adsorption of organic molecules on inorganic semicon-
ductors substrates has attracted increasing interest in past
decades due to their ability to control and tune the electronic
properties of semiconductors and metal surfaces;1-23 ergo,
(1) Vilan, A.; Cahen, D. Trends Biotechnol. 2002, 20, 22-29.
(2) Haick, H.; Ambrico, M.; Ligonzo, T.; Tung, R. T.; Cahen, D. J. Am. Chem.
Soc. 2006, 128, 6854-6869.
(15) Cohen, R.; Kronik, L.; Shanzer, A.; Cahen, D.; Liu, A.; Rosenwaks, Y.;
Lorenz, J. K.; Ellis, A. B. J. Am. Chem. Soc. 1999, 121, 10545-10553.
(16) Cohen, R.; Kronik, L.; Vilan, A.; Shanzer, A.; Rosenwaks, Y.; Cahen, D.
AdV. Mater. 2000, 12, 33-37.
(17) Cahen, D.; Kahn, A.; Umbach, E. Mater. Today 2005, 8, 32-41.
(18) Cui, Y.; Lieber, C. M. Science 2001, 291, 851-853.
(19) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. AdV. Mater. 1999, 11, 605-625.
(20) Saito, N.; Hayashi, K.; Sugimura, H.; Takai, O. Langmuir 2003, 19, 10632-
10634.
(3) Vilan, A.; Shanzer, A.; Cahen, D. Nature (London) 2000, 404, 166-168.
(4) Ashkenasy, G.; Cahen, D.; Cohen, R.; Shanzer, A.; Vilan, A. Acc. Chem.
Res. 2002, 35, 121-128.
(5) Zuppiroli, L.; Si-Ahmed, L.; Kamaras, K.; Nu¨sch, F.; Bussac, M. N.; Ades,
D.; Siove, A.; Moons, E.; Gra¨tzel, M. Eur. Phys. J. B 1999, 11, 505.
(6) Crispin, X.; Geskin, V.; Crispin, A.; Cornil, J.; Lazzaroni, R.; Salaneck,
W. R.; Bredas, J. L. J. Am. Chem. Soc. 2002, 124, 8131-8141.
(7) Kru¨ger, J.; Bach, U.; Gra¨tzel, M. AdV. Mater. 2000, 12, 447-451.
(8) Ofir, Y.; Zenou, N.; Goykhman, I.; Yitzchaik, S. J. Phys. Chem. B 2006,
101, 8002-8009.
(21) Reed, M. A.; Tour, J. M. Appl. Phys. Lett. 2000, 77, 1224.
(22) Iozzi, M. F.; Cossi, M. J. Phys. Chem. B 2005, 109, 15383-15390.
(23) Seker, F.; Meeker, K.; Kuech, T. F.; Ellis, A. B. Chem. ReV. 2000, 100,
2505-2536.
(9) Nuesch, F.; Rotzinger, F.; Si-Ahmed, L.; Zuppiroli, L. Chem. Phys. Lett.
1998, 288, 861-867. Ganzorig, C.; Kwak, K.-J.; Yagi, K.; Fujihira, M.
Appl. Phys. Lett. 2001, 79, 272-274. Rudich, Y.; Benjamin, I.; Naaman,
R.; Thomas, E.; Trakhtenberg, S.; Ussyshkin, R. J. Phys. Chem. A 2000,
104, 5238-5245. Linford, M. R.; Fenter, P.; Eisenberger, P. M.; Chidsey,
C. E. D. J. Am. Chem. Soc. 1995, 117, 3145-3155.
(24) Wu, D. G.; Ashkenasy, G.; Shvarts, D.; Ussyshkin, R. V.; Naaman, R.;
Shanzer, A.; Cahen, D. Angew. Chem., Int. Ed. 2000, 39, 4496-4500.
(25) Wu, D. G.; Cahen, D.; Graf, P.; Naaman, R.; Nitzan, A.; Shvarts, D.
Chem.sEur. J. 2001, 7, 1743-1749.
(26) Cahen, D.; Naaman, R.; Vager, Z. AdV. Funct. Mater. 2005, 15, 1571-
1578.
(10) Si-Ahmed, L.; Nu¨esch, F.; Zuppiroli, L.; Franc¸ois, B. Macromol. Chem.
Phys. 1998, 199, 625-632.
(27) Patolsky F.; Zheng, G. F.; Lieber, C. M. Anal. Chem. 2006, 78, 4260-
4269 and references therein.
(11) Ulman, A. Chem. ReV. 1996, 96, 1533-1554. Ulman, A. An Introduction
to Ultrathin Organic Films; Academic Press: Boston, 1991. Kronik, L.;
Shapira, Y. Surf. Sci. Rep. 1999, 37, 1-299.
(28) Niwa, D.; Yamada, Y.; Homma, T.; Osaka, T. J. Phys. Chem. B 2004,
108, 3240-3245.
(29) Rampi, M. A.; Whitesides, G. M. Chem. Phys. 2002, 281, 373-391.
(30) Salomon, A.; Cahen, D.; Lindsay, S.; Tomfohr, J.; Engelkes, V. B.; Frisbie,
C. D. AdV. Mater. 2004, 16, 477.
(12) Cohen, R.; Zenou, N.; Cahen, D.; Yitzchaik, S. Chem. Phys. Lett. 1997,
279, 270-274. Zenou, N.; Zelichenok, A.; Yitzchaik, S.; Cohen, R.; Cahen,
D. ACS Symp. Ser. 1997, 695, 57-66.
(31) Shen, Y.; Hosseini, A. R.; Wong, M. H.; Malliaras, G. G. Chem. Phys.
Chem. 2004, 5, 16-25.
(13) Sfez, R.; Peor, N.; Cohen, S. R.; Cohen, H.; Yitzchaik, S. J. Mater. Chem.
2006, 16, 4044-4050.
(32) Chabinyc, M. L.; Chen, X.; Holmlin, R. E.; Jacobs, H.; Skulason, H.; Frisbie,
C. D.; Mujica, V.; Ratner, M. A.; Rampi, M. A.; Whitesides, G. M. J. Am.
Chem. Soc. 2002, 124, 11730-11736.
(14) Ray, S. G.; Cohen, H.; Naaman, R.; Liu, H.; Waldeck, D. H. J. Phys. Chem.
B 2005, 109, 14064-14073.
9
4158
J. AM. CHEM. SOC. 2008, 130, 4158-4165
10.1021/ja077933g CCC: $40.75 © 2008 American Chemical Society

Effects on the Electronic Properties of Silicon
A R T I C L E S
Scheme 1. Covalent Assembly of Various
assembled monolayers (SAM) can tune the electronic properties
of interfaces by modifying the existing surface dipole and adding
an external dipole layer.12-14 Moreover, the changes in the dipole
of the adsorbed molecules are considered to be a major
controlling factor in novel chemical and biological molecule-
based sensing electronic and opto-electronic devices.24,31 The
change in the surface dipole can be derived experimentally by
Kelvin probe analysis or theoretically by the Helmholtz equation.
The Kelvin probe method is widely used to determine the
relative WF of semiconductors by measuring the contact
potential difference (CPD) between a reference surface and the
sample. The unknown difference between the conductive/
valence band and the Fermi level is eliminated by subtracting
the absolute values of the sample and the reference substrate.
Since the WF of a semiconductor is composed from band
bending (BB) and electron affinity (EA) parameters, it is
possible to determine both of them experimentally by finding
the CPD.12,13 On the other hand, the surface dipole can also be
derived theoretically from the Helmholtz equation (eq 1). From
eq 1 it can be seen that two parameters are of extreme
importance when choosing a specific molecule for tailoring
surface electronic properties, the molecular dipole (magnitude
and orientation) and the density of the adsorbed molecules on
the surface. Eq 1 is
Trichlorosilane-Containing Coupling Agents on Hydroxyl
Terminated Substrates (Step 1)a
a An stands for the free residue, while An stands for the SAM.
electronic surface dipole effect obtained upon molecular as-
sembly. Most of the literature deals with polar and dense
monolayers anchored to metal and bare semiconductors surfaces
without native oxides, in order to increase the influence of the
layer on the electronic properties of the substrate.38,39 Due to
the fact that Si/SiO2 is the most common interface for electronic
devices, there is a great motivation to investigate the influence
of organic monolayers on the electronic properties of this
substrate. However, few researchers have studied oxide-bearing
semiconductors as substrates8,12,13,37,40 or as devices41 or both,42
indicating tuning of the electronic properties of the semiconduc-
tor even through the oxide layer. It was shown12 that the change
in the electron affinity parameter (∆EA) in <1,0,0> Si/SiO2
had a linear correlation with the Hammett parameter of isolated
molecules in the gas phase. Moreover, it was shown that a major
surface passivation occurs upon the formation of a two-
dimensional (2D) polysiloxane-based monolayer on the sub-
strate.13 Upon application on a Si/SiO2-based device, the ability
to control the carrier density in the conduction channel of an
organic field-effect transistor by the use of SAMs with a
different terminal group that is characterized by a different
molecular dipole was also verified.43 It is clear that the hybrid
consisting of Si substrate, native oxide, and adsorbed organic
monolayer has high importance and potential in both device
application and fundamental science.
In this contribution, we systematically examined the molecular
variables which control the electronic properties of <1,0,0>
n-type Si/SiO2 surfaces. In order to do so, two steps of assembly
were conducted on the surface. The first step involved the
assembly of coupling agents baring a trichlorosilane head group
with various tail groups and spacers (Scheme 1). The tail groups
differ mostly by their dipole moments as was obtained from
MOPAC minimization for the single molecule (Table 1). A
linear fit was obtained between the changes in electronic
properties of the semiconductor with what was predicted by
the Helmholtz equation (Figure 3). Two spacers were exam-
ined: aliphatic vs aromatic, giving rise to an important shielding
Nµ
ꢀꢀ0
∆Φs )
cos θ
(1)
where Nµ is the dipole density (in m2/molecule), µ is the dipole
moment (in debye, 1 D ) 3.34 × 10-30 C‚m), θ is the average
tilt angle of the dipole with respect to the surface normal, ꢀ is
the effective dielectric constant of the molecular film, and ꢀ0 is
the permittivity of vacuum; here ∆Φs is expressed in units of
volts.
Usually, there is a factor between the calculated and the
observed values, due to local environmental effects such as
hydration and depolarization.33 The depolarization influences
the obtained layer a lot and should be taken into account
concerning the packaging and ordering of the molecules, when
the molecular dipole is greater than about 5 D.35-37 Although
many experimental and theoretical works have dealt with the
influence of the dipole on the electronic properties of the
substrate, almost no experimental works have been conducted
with the aim to elucidate the influence of the layer’s number
density on the substrate. Recently, a theoretical study was
conducted regarding the depolarization effects arising from
different clusters sizes, aiming to predict the density effect on
modeled etched <1,1,1> Si substrates.35 Experimental work that
dealt with molecular monolayers adsorbed on periodic surfaces
showed a linear correlation between the lattice constant of the
semiconductor and the molecular effect obtained.4 The higher
the density of the binding sites (i.e., smaller lattice constant),
the higher the molecular dipoles’ density, and thus the larger
(38) Hunger, R.; Jaegermann, W.; Merson, A.; Shapira, Y.; Pettenkofer, C.;
Rappich, J. J. Phys. Chem. B 2006, 110, 15432-15441.
(39) Chen, J.; Reed, M. A.; Rawlett, A. M.; Tour, J. M. Science 1999, 286,
1550-1552.
(33) He, T.; He, J.; Lu, M.; Chen, B.; Pang, H.; Reus, W. F.; Nolte, W. M.;
Nackashi, D. P.; Franzon, P. D.; Tour, J. M. J. Am. Chem. Soc. 2006, 128,
14537-14541.
(40) Gershewitz, O.; Grinstein, M.; Sukenik, C.; Regev, K.; Ghabboun, J.; Cahen,
D. J. Phys. Chem. B 2004, 108, 664-672.
(34) El-Abed, A.; Ionov, R.; Goldmann, M.; Fontaine, P.; Billard, J.; Peretti, P.
Europhys. Lett. 2001, 56, 234-240.
(41) Collier, C. P.; Wong, E. W.; Belohradsky, M.; Raymo, F. M.; Stoddart, J.
F.; Kuekes, P. J.; Williams, R. S.; Heath, J. R. Science 1999, 285, 391-
394.
(35) Deutsch, D.; Natan, A.; Shapira, Y.; Kronik, L. J. Am. Chem. Soc. 2007,
129, 2989-2997.
(36) Cornil, D.; Olivier, Y.; Geskin, V.; Cornil, J. AdV. Funct. Mater. 2007, 17,
1143-1148.
(42) Liu, Y.; Yu, H. Chem. Phys. Chem. 2003, 4, 335-342.
(43) Kobayashi, T.; Nishikawa, T.; Takenobu, S.; Mori, T.; Shimoda, T.; Mitani,
H.; Shimonati, N.; Yoshimoto, S.; Ogawa and Iwasa, Y. Nat. Mater. 2004,
3, 317-322.
(37) Gershevitz, O.; Sukenik, C. N.; Ghabboun, J.; Cahen, D. J. Am. Chem.
Soc. 2003, 125, 4730-4731.
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4159

A R T I C L E S
Peor et al.
Table 1. Electronic and Spectroscopic Properties of the Various
Silylated Substrates
cleaned with H2O/H2O2/NH3 (5:1:0.25) solution while sonicating for
15 min at 60 °C. After subsequent washing with TDW, the substrates
were immersed for 5 min in pure acetone and finally dried under a
stream of nitrogen.
A.2. Chemicals. A.2.a. Coupling Agents. Ethyltrichlorosilane,
phenyltrichlorosilane, (chloromethyl)phenyltrichlorosilane, (3-chloro-
propyl)trichlorosilane, and (3-bromopropyl)trichlorosilane [Gelest], were
vacuum distilled before use. The (3-iodopropyl)triiodosilane was
synthesized immediately before use, obtained from (3-bromopropyl)-
trichlorosilane.
A.2.b. Solvents. n-Heptane, hexanes (95% n-hexane ULTRA RESI
analyzed) [J. T. Baker] were distilled on sodium under a nitrogen
atmosphere, acetonitrile was distilled on CaH2, n-propanol was used
after passing alumina column. 2-Propanol, THF, dichloromethane,
ethanol, acetone, chloroform, diethylether anhydrous, ethyl acetate and
methanol [Aldrich] were used as received.
thickness [Å]a
calcda
∆
(CPD
±
10) [mV]a
calcd
dipole [D]
contact
angle [deg]a
ellipsometry
dark
light
SiO2
A1
A2
A3
A4
A5
A6
<5
82 ( 2
74 ( 2
74 ( 2
75 ( 1
76 ( 2
78 ( 2
16 ( 3
0
0
-96
-219
200
379
276
-0.89
-0.98
0.79
1.63
1.60
5.0
6.6
7.9
6.8
7.0
7.2
4.7 ( 0.3
6.3 ( 0.4
8.1 ( 0.6
8.6 ( 1.4
6.7 ( 0.5
6.7 ( 0.5
-246
-349
80
239
146
1.47
142
212
a Statistical error statements are based on at least five different substrates.
factor in the aromatic spacer. The second step consisted of the
construction of various surface coverage at the submonolayer
regime, of a polarizable chromophore via an in-situ SN2 reaction
(Scheme 2). We choose 4-[4-(N,N-dimethylamino phenyl)azo]-
pyridine (MAP) as the desired chromophore precursor due to
the fact that upon quaternization reaction it becomes a polariz-
able molecule with a high calculated dipole moment (∼10 D)
that can be a good candidate to induce electronic changes in
the bulk Si. In the case of the dense and organized monolayer
containing high molecular dipoles (>5 D), molecular layer
depolarization is observed. This depolarization effect might be
also expected in a less dense and organized monolayer, however,
still giving rise to a notable change in the electronic properties
as predicted by theory.44 This establishes the important role
played by coverage, in addition to local chemical properties, in
tailoring surface chemistry via polar molecule adsorption.
The SAMs’ binding onto the surface was verified by X-ray
photoelectron spectroscopy (XPS), atomic force microscopy
(AFM), UV-vis absorption spectra, variable-angle spectroscopic
ellipsometry (VASE), and contact angle measurements. CPD
equipment, using the vibrating Kelvin probe technique, was used
to monitor and determine the molecular effects on the electronic
properties of the silicon. The relevant properties that were
evaluated are the silicon work function (WF), band-bending
(BB), and electron affinity (EA). The influence of the molecular
dipole and type of spacers in a SAM of coupling agent on the
electronic properties of a Si/SiO2 substrate was demonstrated.
It was shown that the larger the molecular dipole, the larger
the ∆EA parameter value, as was predicted by theory. Aromatic
spacers induced a shielding that was estimated to be of about
200 mV in magnitude. The second layer of the MAP+
chromophore has shown an important dependence of the surface
dipole upon increase in chromophore surface number density,
while preserving the shielding effect of the ring from the
coupling agent layer beneath. The depolarization effect was
clearly obtained for these chromophoric layers.
A.2.c. Reagents. Propyl bromide, benzyl chloride, sodium, 4-ami-
nopyridine, N,N-dimethylaniline, tetrafloroburate acid (HBF4 48% in
water), (4-dimethylamino)pyridine, sodium iodide anhydrous, sodium
nitrite, and sodium hydroxide [Aldrich] were used without further
purification.
B. Synthesis of Starting Materials. B.1. Synthesis of 3-Iodopropyl
Triiodosilane. An excess of sodium bromide (50 gr of NaBr) in a three-
neck flask with a mechanical stirrer was left under vacuum at 60 °C
overnight. Portions of 50 mL of dry CH3CN and 3-bromopropyl
triclorosilane (8.8 mmol) were added at room temperature. The mixture
was stirred for 8 h. Then 100 mL of dry heptane was added, and the
mixture was stirred vigorously for 5 min to extract the product. The
product was transferred under ambient conditions for further surface-
anchoring reaction. XPS analysis following surface adsorption showed
the disappearance of the characteristic peak of bromine (binding energy
69.1 eV) and the appearance of the iodide characteristic peak (binding
energy 630 and 617 eV).
B.2. Synthesis of 4-[4-(N,N-Dimethylamino Phenyl)azo]pyridine
(MAP). 4-Aminopyridine (0.021 mol) was dissolved in tetrafluoroborate
acid (48% water) at 0 °C, and a milky solution was then obtained.
Sodium nitrite (0.021mol) was carefully added, keeping the mixture
temperature at 0 °C. Dimethylaniline (0.043 mol) was added, and the
mixture was stirred at room temperature. The adition of a concentrated
solution of NaOH until the neutralization of the solution led to product
precipitation. The residue was filtered, dissolved in minimum DCM,
and then purified by column chromatography (silica gel 60, 0.2 mesh)
using hexane/EtOAc (1:2, respectively) as an eluent to afford the
product in 85-90% yield.
1
The product MAP precursor was obtained as an orange solid. H
NMR: δ 3.13(s, N(CH3)2, 6H), 6.75-6.78(dd, HAr, 2H), 7.63-7.65-
(d, HAr, 2H), 7.89-7.92(d, HAr, 2H), 8.71-8.73(d, HAr, 2H). FTIR
(KBr pellet): 1445, 1519, 1552, 1563, 1581, and 1603 cm-1. UV-
EtOH
max
vis. λ
) 440 nm, ꢀλ (MeOH) ) 25 973 [cm-1 M-1]. Anal.
max
Calcd: 69.00 C, 6.24 H, 24.76 N. Found: 68.70 C, 6.34 H, 24.47 N.
B.3. General Synthetic Procedure for Chromophores’ Models.
A mixture of the precursor (MAP) in an excess of propylbromide or
benzylbromide was heated under reflux for 12 h. The solid was filtered
and washed under cold EtOAc. Finally, the product was recrystallized
from methanol.
1-Propyl[4-(4-)N,N-dimethylamino phenyl)azo)pyridinium bromide
(PrBr-MAP+): 1H NMR: δ 1.04-1.25(t, CH3CH2, 3H), 2.08-2.16-
(sec, CH3CH2, 2H), 3.25(s, N(CH3)2, 6H), 4.86-4.91(t, N + -CH2,
2H), 6.79-6.91(dd, HAr, 2H), 7.96-7.99(d, HAr, 2H), 8.09-8.12(d,
Experimental Section
A. General. A.1. Substrate Cleaning. Quartz (Chemglass), glass
slides (Knittel Glaser), and n-Si <100> (Virginia Semiconductors)
substrates were cleaned in aqueous detergent, rinsed copiously with
triple distilled water (TDW), then dipped in hot (90 °C) piranha solution
for 60 min (3:7 by volume of 30% H2O2 (MOS) and conc. H2SO4 (MOS
“BAK-ANAL” REAG) (Caution: strong oxidizing solution, handle
with care). The substrates were then rinsed with TDW and further
HAr, 2H), 9.22-9.24(d, HAr, 2H). FTIR (KBr pellet): 1500, 1541,
EtOH
1560, 1600, and 1629 cm-1. UV-vis. λmax ) 555 nm, ꢀλ (MeOH)
max
) 50 149 [cm-1 M-1]. Anal. Calcd: 55.02 C, 6.06 H, 16.04 N. Found:
54.55 C, 6.04 H, 15.72 N.
1-Benzyl[4-(4-)N,N-dimethylamino phenyl)azo)pyridinium]chloride
(BzCl-MAP+): 1H NMR: δ 3.26(s, N(CH3)2, 6H), δ 6.25(s, HAr, 1H),
6.79-6.82(dd, HAr, 2H), 7.33-7.42(m, HAr, 2H), 7.61-7.63(dd, HAr,
(44) Natan, A.; Zidon, Y.; Shapira, Y.; Kronik, L. Phys. ReV. B: Condens. Matter
Mater. Phys. 2006, 73, 193310-193314. Demchak, R. J.; Fort, J. R. T. J.
Colloid Interface Sci. 1974, 46, 191-202
9
4160 J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008

Effects on the Electronic Properties of Silicon
A R T I C L E S
Scheme 2 . Synthetic Route for the Assembly of Chromophoric Monolayer. Step 2: Chromophore Anchoring via Quaternization, SN2
Reaction
2H), 7.96-7.98(dd, HAr, 2H), 8.06-8.08(d, HAr, 2H), 9.43-9.45(d,
in the WF of the semiconductor surface and the vibrating Au grid.
The vibrating capacitor leads to a time-dependent variation in capaci-
tance which induces an ac current flow through the circuit. By tuning
the external dc bias, the compensating VCPD value is determined when
the current flow is nullified. Therefore, the change in the semiconduc-
tor’s WF can be determined. Upon illumination of the semiconductor,
electron-hole pairs are generated close to the surface, leading to a
decrease in the BB. This reduction of BB upon illumination will
continue until the bands become almost flat (photosaturation condition).
A quartz tungsten halogen (QTH) lamp, intensity 130 mW cm-2, which
was assumed to lead to complete flattening of the bands, was used.
The BBwas determined by comparing the CPD value in the dark with
the value under intense illumination, where the bands are nearly flat.
The CPD measurements were taken after the few minutes needed for
the signal to stabilize. In order to eliminate molecule decomposition
during photosaturation measurements, we used an optical filter that
blocks wavelengths adsorbed by the molecules (Schott, RG-780).
EtOH
max
HAr, 2H). UV-vis. λ
) 558 nm, ꢀλ (MeOH) ) 54 214 [cm-1
max
M-1]. Anal. Calcd: 68.15 C, 6.01 H, 15.91 N. Found: 66.88 C, 6.17
H, 15.45 N.
C. SAM Preparation. C.1. General Synthetic Procedure for
Silylated SAMs (Scheme 1). Freshly cleaned Si/SiO2 (silicon’s native
oxide) or glass substrates were immersed in a 1% (v/v) coupling agent/
hexane solution for 20 min under inert conditions in a Schlenk-line
system. Upon completion of the reaction, the substrates were washed
three times with dry hexane under inert conditions, sonicated for 1
min in acetone in order to remove any excess of coupling agent, and
allowed to dry in an oven at 110 °C for about 15 min.
C.2. Monolayer Functionalization with Azo-Containing Chro-
mophore, Pyridinium Salt, MAP+ (Scheme 2). The silylated surfaces
with terminal halide groups were dipped in dry acetonitrile ∼0.03 M
MAP solution at 70 °C for 24 h. In order to tune the chromophores’
surface number density, the substrates were monitored at different
reaction times and characterized both electrically and by spectroscopy
after washing with IPA and nitrogen drying.
D. Instrumentation. AFM measurements were carried out with a
Nanoscope IV (DI) in tapping mode using a tapping etched silicon
probe (TESP, DI) with a 30 N/m force constant. All substrates were
imaged in air. UV-vis spectra were acquired on a Shimadzu UV-
3101PC spectrophotometer using glass substrates. Contact angle
measurements were performed using Si substrates with a FTA125 video-
based contact angle meter (first 10 Å). XPS spectra were collected at
ultrahigh vacuum (2.5 × 10-10 Torr) on a 5600 Multi-Technique (AES/
XPS) system (PHI) using an X-ray source of Al K (1486.6 eV). VASE
measurements were carried out on a VB-200 ellipsometer (Woollam
Co.) around the Brewster angle of Si (75°). CPD measurements were
conducted using an Au grid (Kelvin probe S, DeltaPhi Besocke, Ju¨lich,
Germany) that vibrates by a piezoelectric crystal. In order to measure
the VCPD value, a semiconductor sample, with InGa Ohmic back contact,
is placed parallel to the grid, thus creating a closely spaced parallel
plate capacitor. The entire experimental setup was placed in a home-
built Faraday cage in an inert atmosphere. Upon electrical connection,
equilibrium is reached and then the VCPD value is equal to the difference
E. Molecular Modeling. The surface dipole Φs values for the various
coupling agents and for the different chromophore-derived surface
dipole densities were extracted from the Helmholtz equation (eq 1).
The molecular dipole µ was obtained from MOPAC after minimization
(AM1 method on the trimethoxysilanes derivatives as “free” molecule)
which gave the dipole value in debye. N and ꢀ were taken as average
experimental values obtained for halogen-modified aliphatic chains,44-45
and the tilt angle θ was estimated by ellipsometric measurements to
be 20° to the normal.
Results and Disscusion
Coupling-Agent Layers. The various coupling agents dis-
cussed in this study (Scheme 1) were self-assembled covalently
to the native oxide of Si/SiO2 substrates and structurally
characterized by ellipsometry and contact angle along with CPD
electronically (Table 1).
(45) Olivera, O. N.; Taylor, D. M.; Lewis, T. J.; Salvagno, S.; And, Stirling, J.
M. J. Chem. Soc., Faraday Trans. 1 1989, 85 (4), 1009-1018.
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4161

A R T I C L E S
Peor et al.
Figure 2. Change in silicon’s electron affinity (∆EA) as a function of the
various surface functionalities ((10 mV) relative to clean and activated Si
substrate.
Figure 1. Change in silicon’s band-bending (∆BB) as a function of the
various surface functionalities ((10 mV) relative to clean and activated Si
substrate.
interface and from the terminal silanol to siloxane groups ratio
of the oxide layer located at the oxide/adsorbed monolayer
interface. The coupling agent molecules covalently anchored
to the surface affect only the latter due to the fact that the
dangling bonds are completely out of reach for creating a
covalent bond with the coupling agent molecules. Following
surface activation, a maximal density of terminal hydroxyl
groups’ sites (silanol groups) was obtained in the silicon oxide/
air interface, depending strongly on the cleaning and activation
method. Therefore, a high number of local negative charges
exist on the surface of the n-type Si leading to a high value of
BB. The surface condensation reaction that alter the terminal
silanol bonds with 2D polysiloxane network bonds eliminates
the electron traps (acidic silanol) on the surface, thus signifi-
cantly diminishing the BB, although there are still unreacted
hydroxyl sites. The fact that similar passivation is obtained for
various coupling agents indicates that the differences in the
number of the unreacted sites are negligible; i.e., similar surface
number densities can be estimated. The differences between the
∆BB values for the different coupling agents molecules have
to do with their ability to stabilize a negative surface charge on
n-type Si. As expected,12 the smaller effect observed for the
iodide group (A6) is related to the ability of this polarizable
group to act as an “electron trap,” hence, increasing the surface
density of states by half. However, the fact that the A6 differs
may also arise from the large size of the iodine end group which
generates a less dense and organized monolayer, giving a
comparatively lower change in BB.
Two different parameters affecting the WF of Si were
studied: the end group which determines the direction and
strength of the molecular dipole and the character of the spacer
(aromatic vs aliphatic). As known, the change in WF is directly
related to the molecular dipole size and direction arising from
the molecules comprising the layer.46,47 The dipole can be
determined experimentally by CPD and theoretically derived
from the Helmholtz relation. CPD measurements were con-
ducted by the Kelvin probe technique in order to observe the
effect of the different coupling agents on the WF of highly doped
n-type Si substrates. In order to enable the separation of the
different parameters contributing to the WF (i.e., BB, EA), the
sample was illuminated until band flattening was reached. The
BB was calculated by subtracting the measured dark value of
the WF from the value measured under photosaturation condi-
tions. The EA value was determined by subtracting the BB value
from the WF value. We used relative parameters (∆EA, ∆BB)
by subtracting the reference value of the bare Si substrate from
that of the modified Si value.
As can be seen in Figures 1 and 2, upon surface function-
alization, changes in the BB and EA values can be observed.
Regarding the difference in the ∆BB parameter for each
coupling agent, (Figure 1), we observed that the major effect is
surface-state passivation characterized by a decrease in the BB
value, i.e., a negative ∆BB value for all the couplers of about
130 mV (except A6 due to the high polarizability of the end
group, as will be discussed below).
As opposed to the common major passivation effect observed
in all the coupling agents’ monolayers concerning the ∆BB
parameter, two tendencies on the ∆EA parameter can be
observed which differ from one molecule to another and can
account for the wide tunability range of 600 mV which varies
from -220 up to +380 mV for the various coupling agent-
derived monolayers (Figure 2).
It is worth mentioning that upon coupling agent modification,
only surface (chemical) passivation occurs. The general value
of the change in the BB value can be regarded as originating
from the dangling bond that exists at the Si lattice/native oxide
(46) Mo¨nch, W. Semiconductor Surfaces and Interfaces; Springer: Berlin, 1995.
(47) Nicolini, R.; Vanzetti, L.; Mula, G.; Baratina, G.; Sorba, L.; Franciosi, A.;
Peressi, M.; Baroni, S.; Resta, R.; Baldereschi, A.; Angelo, J. E.; Gerberich,
W. W. Phys. ReV. Lett. 1994, 72, 294-297.
The first major effect is attributed to the dipole direction on
the ∆EA parameter, as can be predicted from the Helmholtz
equation (eq 1). Figure 2 shows that the changes in electron
affinity are negative values for donor end groups (A1, A2) and
positive for acceptor end groups (A3-A6). Donor end groups
induce dipole pointing toward the surface, thus facilitating the
(48) Waggner, A. S.; Grinvald, A. Ann. N.Y. Acad. Sci. 1977, 303, 217-241.
(49) Loew, L. M.; Simpson, L. L. Biophysical J. 1981, 34, 353-365.
(50) Roscoe, S.; Yitzchaik, S.; Kakkar, K.; Marks, J. T. Langmuir 1996, 12,
5338-5349.
(51) Yitzchaik, S.; Marks, J. T. Acc. Chem. Res. 1996, 29, 197-202.
9
4162 J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008

Effects on the Electronic Properties of Silicon
A R T I C L E S
Figure 4. Chromophore anchoring reaction followed by absorption at 550
nm O.D 0.013 for full coverage. Reactions conditions are 0.03 M of MAP
in CH3CN, at 70 °C. The derived biexponential first-order kinetic parameters
are R ) 0.314 ( 0.012, k1 ) (7.49 ( 0.80) × 10-3, k2 ) (5.86 ( 0.17) ×
10-5, and R ) 0.9996.
Chromophoric Layers. Our synthetic approach to chro-
mophoric layer assembly utilizes two coupling agent monolayers
with good leaving halogen end groups and similar distances
(ca. 7 to 8 Å) from the surface (A3, A5) which can easily
undergo SN2 reaction on the surface (Scheme 2). Our motivation
for choosing the MAP chromophore was the fact that this
chromophore is known to be a voltage-sensitive dye48,49 and
can be applied in sensors and other devices due to the dipole
flipping ability upon excitation. Scheme 2 describes the MAP
precursor which undergoes quaternization reaction on the surface
from solution while leaving a compensating labile anion upon
this surface-anchoring reaction. As shown in Figure 4, full
coverage was achieved, creating a chromophoric monolayer
assembly which leads to a collective dipole change of the
surface. This dipole change consists of molecular dipole due to
the formation of covalently attached pyridinium cation and an
ionic dipole due to the position of the labile anion relative to
the grafted organic cation. In our case, the density of the
chromophoric layer (∼55 Å2/molecule) enables the anion to be
in the pyridinium plane.50,51 Since CPD measurements can detect
only dipoles which are perpendicular to the surface, in the case
where the anion is in the pyridinium’ plane (Scheme 2), the
major contribution to the dipole density arises from the
molecular dipoles. In the case of higher pyridinium density, the
anion is constrained to move from the pyridinium plane and
the measured CPD value is a summation of both dipoles (this
effect will be discussed elsewhere). Upon quaternization, the
MAP+ molecules undergo a bathochromic shift (Supporting
Information Figure I) accompanied by band broadening, char-
acteristic of intramolecular charge transfer (CT) absorption. The
course of the monolayer assembly was also verified by contact
angle, XPS, and ellipsometric measurements (see the Supporting
Information).
Figure 3. Correlation between experimentally measured change in electron
affinity (∆EA) and surface dipole values (∆Φ) calculated from the
Helmholtz equation for the various coupling layers.
removal of electrons by decreasing the WF. On the other hand,
acceptor end groups induce a dipole pointing away from the
surface, thus causing an increase in the WF. The other minor
effects relate to the differences in the dipoles’ magnitude and
give insight to the shielding of the aromatic spacer ring
compared to the aliphatic spacer. When the dipole magnitude
is decreased from A4 to A6 (Table 1), a decrease in the surface
dipole is obtained as well. It is noteworthy that although there
might be differences in the molecular spatial orientation that
can influence the measured ∆EA value, we consider that the
densities of all the monolayers obtained (apart from A6) are
similar, based on previous studies.11,52 The ring shielding
parameter can be derived by comparison of aromatic and
aliphatic spacers with the same end groups (A3, A4). As can
be seen, the delocalization of the π electrons shields the effect
of the acceptor end group, thus causing a decrease in the surface
dipole which is about 200 mV lower than that of the aliphatic
spacer. This effect is preserved even after the deposition of
another chromophoric layer on the template layer of the coupling
agent (vide infra). Figure 3 shows the correlation between the
experimental surface dipole value obtained from CPD measure-
ments on modified substrates and the theoretical calculations
of couplers’ dipoles by the Helmholtz equation.
A linear dependence was obtained between the calculated
∆Φs values and observed CPD measurements, indicating that
the Helmholtz equation describes accurately the change in
electron affinity due to monolayer assembly of various molecular
dipoles. The Helmholtz equation gives the surface potential drop
through a monolayer and is evaluated by using the experimental
values of the molecular layer number density and tilt angle
including the calculated molecular dipole and dielectric constant
of the isolated molecules. The experimental surface potential
differs from the calculated one by measuring the effective
dielectric constant of the monolayer. The observed deviation
from unity slope exhibits the change in the effective dielectric
constant induced by these polarizable monolayers.
The maximal surface coverage of the chromophoric layer was
found to be 1.8 × 1014 molecules/cm2, based on absorption
spectroscopic analysis at a wavelength of 550 nm (λmax). Figure
4 demonstrates the quaternization reaction kinetics, confirming
full surface coverage with kinetic parameters similar to those
reported for the related stilbazolium chromophore.51
(1 - θ) ) Re-k t[C] + (1 - R)e-k t[C]
(2)
1
2
The successful biexponential fit observed here provides a real
indication that the self-assembling quaternization must be
(52) Malik, A.; Lin, W.; Durbin, M. K.; Marks, T. J.; Dutta, P. J. Chem . Phys.
1997, 107, 645-652.
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4163

A R T I C L E S
Peor et al.
Figure 5. Change in silicon WF relative to the relevant coupling agent-
modified substrate, as a function of MAP+ surface number density on
aromatic A3 (diamonds) and aliphatic A5 (triangles) templates ((10 mV).
The curVes are drawn as guides to the eye. The zero point describes the
WF for zero chromophoric coverage for both coupling agents.
Figure 6. Change in electron affinity (∆EA) and band bending (∆BB)
parameters ((10 mV) relative to the relevant coupling agent-modified
substrate, as a function of the chromophore surface number density on
aromatic (A3) and aliphatic (A5) coupling agents containing SAMs. The
curVes are drawn as guides to the eye. The zero point describes the ∆CPD
value for zero chromophoric coverage for both coupling agents.
described by more than one rate process (“fast” and “slow” rate
constants). The slower rate k′ is likely associated with in-
2
creasing repulsive interactions between adsorbed molecules at
higher coverage. Upon grafting a chromophore on a coupling
agent position, a change in the dipole orientation occurs; i.e.,
dipoles that were pointing out of the surface (coupling agents,
A3, A5) alter their dipole direction. Thus, increasing the
chromophoric monolayer density will lead to more dipoles
pointing toward the surface. In order to reach chromophoric
submonolayer densities, different incubation times in chro-
mophoric solution were measured for the coupling agent-
modified substrate (Scheme 2). In that way we controlled the
chromophore number density on the surface by reacting different
numbers of alky- or benzyl-halide sites on the surface via
solution assembly. It would be expected that the magnitude of
the change in WF would increase upon increasing chromophoric
layer density versus the coupling layer-modified Si (Figure 5).
However, this trend was observed only for the MAP+ monolayer
built on an aliphatic coupling agent (A5-MAP+), while on the
aromatic coupling agent (A3-MAP+) there seems to be no
dependence of the WF on the chromophore number density
(Figure 5, aromatic). In order to explain this fact we isolated
each parameter contributing to the change in the WF (i.e., ∆BB
and ∆EA).
For A3, the aromatic coupling agent, a positive value of ∆EA
(+200 mV) was obtained due to the coupling agent’s surface
modification (Figure 2). Upon grafting mounting densities of
the chromophoric monolayer MAP+ on A3 sites, the surface
dipole density grows, leading to a ∆EA value of -200 mV for
full surface coverage (∼55 Å2/molecule) as can be seen in
Figure 6. In addition, by increasing the chromophoric monolayer
density, more electron traps are introduced on the surface,
causing a depletion layer due to an increase of local negative
charge concentration on the n-type Si surface. This effect is
observed by the change in the ∆BB value from -120 mV upon
coupling agents’ passivation to +200 mV upon full chro-
mophoric monolayer coverage. It can be seen that the net result
of a summation of the ∆BB and ∆EA values is nearly annealed
(Figure 5, aromatic).
Figure 7. Change in calculated (+) and experimentally measured surface
dipole relative to the relevant coupling agent-modified substrate, as a
function of chromophore surface number density on aromatic ([) and
aliphatic (1) spacers containing monolayers. The zero point describes the
∆EA value for zero chromophoric coverage for both coupling agents.
monolayer, a value of -420 mV was obtained for the ∆EA.
The difference in the ∆EA values for the same chromophore
and surface density can be accounted for the shielding parameter
due to the aromatic ring, as was shown for coupling agent
modifications (∼200 mV). A positive value in the ∆BB is
obtained due to an increase in electron traps sites upon grafting
the chromophoric monolayer. The value obtained in this case
is smaller than the one obtained for the aromatic spacer due to
the lack in aromatic rings, which can serve as traps for electrons
as well.
Figure 7 shows the dependence of the change in the calculated
and measured surface dipole as a function of chromophore
surface number density. The change in surface dipole was
calculated by putting the calculated molecular dipole value into
the Helmholtz equation and comparing it with the measured
For maximal surface number density of chromophoric mono-
layer (∼55 Å2/molecule) grafted upon an A5 coupling agent
9
4164 J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008

Effects on the Electronic Properties of Silicon
A R T I C L E S
CPD values. It can be seen that for the calculated values there
is a linear fit between the surface dipole and the coverage, as
expected for increasing chromophore number densities. As
opposed to that result, the measured values give saturated curves,
indicating a global effect of monolayer’ dipole depolarization,
as expected for such a high molecular dipole packed densely.
obtained in the second step (Scheme 2), it was clearly shown
that upon increasing surface chromophoric coverage an increase
in the electronic properties was observed. However, a major
effect of depolarization was observed on both coupling template
layers. On the basis of this observation, it can also be concluded
that in order to get a high surface dipole it is better to use
molecules with lower value molecular dipoles which can form
a more ordered and dense monolayer with low depolarization
affects than to use a molecule with a very high value molecular
dipole that undergoes large depolarization in the layer. In any
case, it is clear that when increasing the surface coverage, the
influence on the surface dipole increases.
Conclusion
We have shown a way to systematically control the surface
and interface electronic properties of an oxide-bearing silicon
by the use of an organic monolayer. Two parameters seem to
be of major importance: the first one is the molecular dipoles
of the surface anchored molecule and the other one is the surface
coverage of the substrate. A secondary effect concerned the
spacer type used for the coupling of the inorganic substrate and
the organic monolayers. It was shown that an aromatic spacer
gives rise to a shielding effect of the ring that can be estimated
to be about 200 mV and which diminishes the electronic effect
when compared with an aliphatic spacer of the same length.
This effect originates from the larger effective dielectric constant
of the aromatic spacer. Due to that observation, it seems to be
more efficient for further applications in devices to use aliphatic
spacers which do not shield the effect of the added monolayers’
dipole on the substrate. Concerning the chromophoric layer
Acknowledgment. We thank Dr. Noemi Zenou for helpful
and illuminating discussions. This work was supported by the
EC through contract FP6-029192 (“DNA-Based Nanode-
vices”).
Supporting Information Available: Model chromophores
UV-vis spectroscopy and monolayers characterization by XPS,
VASE, and UV-vis spectroscopy. This material is available
free of charge via the Internet at http://pubs.acs.org.
JA077933G
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4165