Forces between nitrogen-containing self-assembled monolayers (SAMs) and zirconia particles in aqueous solutions

Article abstract of DOI:10.1039/b405788k

An atomic force microscope was used to measure the interaction forces between siloxane-anchored organic self-assembled monolayers (SAMs) with nitrogen-containing surface groups on silicon and micron-sized zirconia spheres in aqueous electrolyte solutions. The surface groups studied were alkylammonium salt and amines (primary, secondary, and tertiary). The pH of the solutions ranged from 2.5 to 9 (controlled by adding NaOH or HCl), and their ionic strength was held constant at 0.022 m (by adding NaCl). The results indicated that the alkylammonium salt surface exhibits a standard Debye double layer, whereas all of the amine surfaces exhibited a Stern layer, i.e. a charge inversion layer. The observed interactions were similar for SAMs formed from surfactants of two different alkyl chain lengths (11 and 16). The corresponding electrostatic interactions comprise the dominant part of the DLVO interactions under the conditions studied. The observed forces were consistent with film deposition results: vanadia films, but not titania films, form on alkylammonium salt surfaces; whereas both titania and vanadia films deposit on amine surfaces.

Full text of DOI:10.1039/b405788k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Forces between nitrogen-containing self-assembled monolayers (SAMs) and
zirconia particles in aqueous solutions
Jing-Jong Shyue, Yin Tang and Mark R. De Guire*
Received 19th April 2004, Accepted 21st October 2004
First published as an Advance Article on the web 26th November 2004
DOI: 10.1039/b405788k
An atomic force microscope was used to measure the interaction forces between siloxane-
anchored organic self-assembled monolayers (SAMs) with nitrogen-containing surface groups on
silicon and micron-sized zirconia spheres in aqueous electrolyte solutions. The surface groups
studied were alkylammonium salt and amines (primary, secondary, and tertiary). The pH of the
solutions ranged from 2.5 to 9 (controlled by adding NaOH or HCl), and their ionic strength was
held constant at 0.022 m (by adding NaCl). The results indicated that the alkylammonium salt
surface exhibits a standard Debye double layer, whereas all of the amine surfaces exhibited a Stern
layer, i.e. a charge inversion layer. The observed interactions were similar for SAMs formed from
surfactants of two different alkyl chain lengths (11 and 16). The corresponding electrostatic
interactions comprise the dominant part of the DLVO interactions under the conditions studied.
The observed forces were consistent with film deposition results: vanadia films, but not titania
films, form on alkylammonium salt surfaces; whereas both titania and vanadia films deposit on
amine surfaces.
has a negative surface charge (negative zeta potential) at
2.5 , pH , 3.0.8a,b Therefore, electrostatic interactions
would be expected to attract vanadium oxide particles or
Introduction
Forces between oxide particles and organic (or organically
modified) surfaces play important roles in fields such as
complexes to amine-terminated surfaces in an acidic
ceramics processing, paint and pigments, printing, and
environment. The deposition of V2O5?nH2O films on amine
phosphor technology.1 Interactions between organic and
SAMs from aqueous solutions of vanadic acid has been
interpreted as being consistent with this expectation.9–13
inorganic moieties in liquid solutions are increasingly being
exploited to synthesize materials with new compositions,
In contrast, titania is positively charged in environments
with pH below its isoelectric point (IEP) of y6.14–16
Nevertheless, titania films also deposit on amine SAMs
morphologies, or microstructures, such as meso- and nanos-
tructured materials2 and bioinspired materials.3
from acidic solutions,9–11 contrary to expectations based
Therefore, it becomes increasingly desirable to understand
and control forces between organic and inorganic surfaces in
on electrostatic interactions between the oxide and the
SAM.
liquid media.
One area of recent interest is the use of organic self-assembled
monolayers (SAMs) to promote deposition of oxide and non-
oxide thin films from aqueous media at low temperatures (for a
review, see ref. 4). One of the advantages of SAMs is the ability
to create surfaces with chemical characteristics ranging e.g. from
acidic to basic, hydrophobic to hydrophilic, or polar to non-
polar. The selectivity of certain surface functionalities toward
promoting or inhibiting deposition of various oxide films has
been attributed to electrostatic and van der Waals forces between
the SAM and oxide particles in the forming film,5 and to the
dependence of these forces on and concentration of the source
solution. For example, measurements of the zeta potential of
amine SAMs6 indicate that amine-terminated SAMs are
protonated to some degree, and therefore positively charged,
in acidic environments, consistent with their basic nature
(pKa y 10). The dominant species in aqueous vanadium-
containing solutions for 0 , pH , 14 at all but the lowest
The surface charge of amine surfaces is expected to vary
significantly with pH of the surrounding solution. In
contrast, the alkylammonium functionality behaves more
like a salt, with the counteranion readily dissociating in
+
aqueous solutions (–NR3X A –NR3 + X2, with R 5 CH3
here). Therefore, alkylammonium surfaces are expected to
be more strongly and uniformly positively charged over a wide
range of pH.
Ducker et al.17,18 attached a silica sphere to the tip of an
atomic force microscope (AFM) and measured the forces
between it and an adjacent silica glass surfaces across aqueous
media of varying pH and concentration. Using this approach,
our group19 measured the force between zirconia spheres and
single-crystal Si substrates, either with or without SO3H-
terminated SAMs. The observed forces were consistent with
the existence of a large negative surface charge density on the
sulfonate SAM providing the main contribution to the total
interaction.
42
vanadium concentrations are anions (e.g., V10O26(OH)2
and V10O27(OH)52 for 2 , pH , 5.8).7 Colloidal vanadia
In the present work, this methodology is applied to ZrO2
spheres and (100) Si wafers with either primary-, secondary-,
tertiary-amine- or alkylammonium salt-terminated SAMs, in
*mrd2@cwru.edu
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 323–330 | 323

View Article Online
aqueous electrolytes of varying pH. In addition, the effect of
SAM chain length has been studied.
system. XPS spectra were fit using MultiPak software. Peak
positions were referenced to the Si 2p peak of (100) single
crystal silicon at 99.7 eV. Unless otherwise noted, all SAM-
coated substrates were washed with ethanol and dried in a
stream of argon, and then thoroughly cleaned twice manually
with fresh chloroform-soaked tissues until the surfaces are
perfectly mirror-like. All the substrates were kept under
argon under opaque aluminium foil until use. Dry THF and
ether were distilled over sodium metal. DMF was dried
with anhydrous MgSO4. All other chemicals were used as
received.
Diffuse double layer and stern layer
When a solid surface undergoes an acidic or basic reaction
with an electrolyte solution, the electrostatic charge of the
bound species on the solid surface is balanced by a local excess
of oppositely charged ions in the region of the solution
adjacent to the surface, called the diffuse double layer. The
concentration of counterions in the double layer typically
varies as an exponential decay with distance from the surface
(Fig. 1a).
Synthesis of surfactant. The trichlorosilane group is chemi-
cally reactive to electrophiles like amines. Therefore, it was
necessary to start with a surfactant whose distal group was
unreactive toward trichlorosilane, such as bromide (–Br). The
bromide group was then replaced in situ after the initial
bromide-terminated surface was established.23
For some surfaces, the counterions form a dense layer near the
interface, sufficient to overcompensate the bound charge
(Fig. 1b).20 The region over which the sign of the surface charge
becomes inverted is called the Stern layer, and its thickness is
denoted SL. At distances greater than SL, the concentration of
counterions obeys an exponential dependence, as with a diffuse
double layer, but with the opposite sign.
The procedure began with synthesis of v-hexadecenyl
bromide from v-undecenyl bromide (11.6 g, 50 mmol) via a
Grignard reaction. The v-hexadecenyl bromide was purified by
distillation at 100–110 uC at 0.1 Torr: yield 7 g (46%); 1H NMR
(200 MHz, CDCl3) d 1.14–1.48 (m, 22H), 1.80 (m, 2H), 1.98 (m,
2H), 3.36 (t, J 5 7 Hz, 2H), 4.88–5.04 (m, 2H), 5.80–5.92 (m,
1H). To add the trichlorosilane bonding group, v-hexadecenyl
bromide (1 g, 3.3 mmol), HSiCl3 (5 mL), and 20 mL of a 4%
solution of H2PtCl6 in i-PrOH were added to a pressure tube
under argon atmosphere in a dry box. The tube was sealed and
held in an oil bath at 60 uC overnight. Excess HSiCl3 was
removed by distillation. The resulting 1-bromo-16-(trichlorosi-
lyl)hexadecane was purified by Kugelrohr distillation at 110–
Evidence of Stern layers on synthetic amine-terminated
organic monolayers has been reported.21,22 X-Ray reflection
on an eicosylamine (C20H43–NH2) Langmuir film formed
between phosphotungstate anions (PW12O4032) in a solution
of pH 5 3 and air showed a Stern-like profile, with SL varying
˚
from 8 to 14 ¡ 3 A as the concentration of amine groups
increased from 2 to 3.6 nm22
.
Experimental
Substrate preparation and SAM deposition
1
120 uC at 0.01 Torr: yield 1.15 g (70%); H NMR (200 MHz,
General. Nuclear magnetic resonance (NMR) spectra are
reported in units of d and were recorded on a Varian XL-200
200 MHz spectrometer in CDCl3 solvent. 1H NMR spectra
were referenced to CHCl3 at 7.24 ppm. Contact angles of water
on the various surfaces were measured using a Rame Hart
goniometer. X-Ray photoelectron spectroscopy (XPS) peak
positions were recorded on a PHI Model 5600 MultiTechnique
CDCl3) d 1.20–1.87 (m, 30H), 3.39 (t, J 5 7 Hz, 2H).
The procedure for synthesis of 1-bromo-11-(trichlorosilyl)-
undecane was similar to that for 1-bromo-16-(trichlorosilyl)-
hexadecane, starting with v-undecenyl bromide (1 g, 4.3 mmol)
but omitting the Grignard reaction. 1-Bromo-11-(trichlorosi-
lyl)undecane was purified by Kugelrohr distillation at 120–
125 uC under 0.05 Torr: yield 1.20 g (76%); 1H NMR
(300 MHz, CDCl3) d 1.18–1.90 (m, 20H), 3.38 (t, J 5 7 Hz,
2H).
Preparation of SAMs. Substrates were p-type single-crystal
(100) silicon wafers, polished on one side, cut to 1 cm by 1 cm.
About 40 pieces were immersed in 100 mL of piranha solution
(70 mL concentrated sulfuric acid (H2SO4) and 30 mL aqueous
30% hydrogen peroxide (H2O2) solution), then heated to 80 uC
for 20 min. Thus oxidized and hydrolyzed, the substrates were
rinsed three times and kept in distilled water until further use.
Typical values of the contact angle of water on these surfaces
were 35 ¡ 10u advancing, 15 ¡ 10u receding. The position of
the Si 2p peak was 103.3 eV as measured using XPS.
Hexadecyl bromide-functionalized SAM surfaces (–C16–Br)
were prepared as reported elsewhere.23 The procedure for
undecyl bromide-functionalized SAM surfaces (–C11–Br) was
identical to that for –C16–Br, except the surfactant (1-bromo-
11-(trichlorosilyl)undecane) was dissolved in dried toluene,
which had been distilled over sodium metal. The –Cn–Br
surfaces were transformed to alkylammonium salt and amine
Fig. 1 (a) Normal diffuse dielectric layer and (b) Stern dielectric
layer.
324 | J. Mater. Chem., 2005, 15, 323–330
This journal is ß The Royal Society of Chemistry 2005

View Article Online
surfaces (–Cn–NH2, –Cn–NH(CH3), –Cn–N(CH3)2) as reported
elsewhere.6,24 XPS and contact-angle measurements confirmed
that the transformations were complete.
Deposition of titania films. Following previous studies,28–32
boric acid (0.015 mol) and ammonium hexafluorotitanate(IV)
(0.0050 mol) were dissolved in 80 mL distilled water. The pH
was adjusted to 2.88 by adding 10% HCl aqueous solution
dropwise. The whole solution was diluted to 100 mL. SAM-
coated substrates were immersed in this deposition solution.
The dish was covered and then placed in an oil bath at 50 uC
for 22 h. The coated samples were then rinsed with distilled
water and dried with a stream of dry argon. The samples were
then ultrasonically cleaned in absolute ethanol for 20 min.
Solutions. The solutions for AFM measurements were
prepared with ionic strength of 0.022 m to match the ionic
strength of the solutions used previously for vanadia deposi-
tion.13 The ionic strength was calculated using I 5 2KSzi2mi,
where mi is the molality of ion i and zi is its valence.25 For the
acidic solution, NaCl (0.0200 mol) was dissolved in 900 mL of
distilled water, and 10% HCl aqueous solution was added until
the pH was around 2.5. For the basic solution, NaCl
(0.0221 mol) was dissolved in 900 mL of distilled water. 1%
NaOH aqueous solution was added until the pH was around 9.
The acidic and basic solutions were then diluted to 1000 mL
with distilled water. For the near-neutral solution, NaCl
(0.0221 mol) was dissolved in 1000 mL of distilled water
(pH 5 5.5). Thus in all solutions used for AFM measurements,
only H+, OH2, Na+, and Cl2 were present.
Results and discussion
The force curves shown below are typical of the measured
curves. Tabulated values of the thickness of the diffuse layers
and Stern layers are averages from 10–30 measurements for
each surface and pH. The standard deviations in these values
ranged from 10 to 17% of the layer thickness.
Interactions between zirconia and different functional groups
Preparation of cantilever. Standard V-shaped silicon nitride
AFM cantilevers (Nanoprobe2 SPM tips, type-DNP, Digital
Instruments) were used. A zirconia sphere was attached to the
tip of the cantilever using low-vapor-pressure resin (Varian
Vacuum Products) under optical microscopy. Details of the tip
preparation were reported previously.19
Alkylammonium SAMs. For alkylammonium salt SAMs
(Fig. 2), at pH 5 9.7, attractive interactions were observed at
all separations. At pH 5 5.5 and pH 5 2.5, repulsive forces
were observed at all separations, larger at the lower pH. The
signs and magnitudes of the forces were qualitatively
consistent with expectations based on DLVO theory, assuming
that the electrostatic interaction dominates over the van der
Waals interaction. That is, the alkylammonium salt surface
was positively charged at all values of pH studied here. At
pH 5 9.7, i.e. above the IEP of zirconia, the particle has a
negative surface potential (235 mV33) and the electrostatic
interaction would be attractive, as was the observed interac-
tion. At pH 5 5.5, the zirconia particle is positively charged
(+35 mV33) and both the expected electrostatic interaction and
the observed force were repulsive. The magnitude of the
interaction increased at the lower pH of 2.5.
AFM force measurement and data processing. Surface force
measurements were performed using a Nanoscope2 III
Multimode AFM (Digital Instruments, Santa Barbara, CA)
in contact force mode with a fluid cell attachment. Details of
the measurements were reported previously.19 At least 10, and
as many as 30, measurements were made on a randomly
chosen area of each type of surface at each value of pH
studied. For primary amine and alkylammonium SAMs, two
independent substrates were studied.
Additional corrections to the raw data were made as
follows. Low-amplitude, periodic fluctuations that were
observable in individual force curves at zero deflection (due
to interference effects in the laser beam) were fit with a sine
function and subtracted from the data. Also, to compensate
for different positions of the laser probe for individual data
sets, it was necessary to multiply the translated tip deflection
by a coefficient so that the slope of deflection versus
displacement during contact of the tip with the surface equaled
21. This coefficient (typically 1 ¡ 0.05) was only needed if,
during acquisition of data, the manually adjusted detection
sensitivity did not yield the proper slope. This range of tip–
surface contact, known as the regime of constant compliance,
was also used to calculate the zero point of the displacement
from the intercept of this regime with the zero in tip
deflection.17,26
At pH 5 2.5, the particle is positively charged and the
alkylammonium surface is also positively charged. The force
To compare force profiles for different surfaces taken with
different particles, the force was divided by the particle radius.
The spring constant of the same model of cantilever from the
same wafer varies about 10% from the average value given by
the manufacturer.27 Therefore, the absolute force cannot be
calculated with accuracy better than 10%.
Fig. 2 Force vs. separation between ZrO2 and alkylammonium
SAMs at various pHs.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 323–330 | 325

View Article Online
curve shows only repulsive forces between particle and surface,
as expected from the electrostatic interaction. The diffuse layer
appears to be about 10 nm thick.
At near-neutral condition, the particle is positively charged
and the alkylammonium surface is still positively charged. The
force curve shows a weaker repulsive force than at pH 5 2.5,
as expected.
At pH 5 9.7, on the other hand, the particle is weakly
negatively charged and the alkylammonium surface is still
positively charged. The force curve shows an attractive force
between particle and surface at all separations.
Primary amine SAMs. For amine surfaces (Fig. 3), the force
profiles were strikingly and consistently different from those
for the alkylammonium salt SAMs. Regardless of pH, the
sense of the interaction between the zirconia particle and the
amine SAM surfaces exhibited a reversal from attraction to
repulsion at some value of separation. At pH 5 2 and pH 5 5,
the force profile exhibited long-range attraction and short-
range repulsion. At pH 5 9.7, in contrast, the force profile
exhibited long-range repulsion and short-range attraction.
This reversal in the sense of the interaction with separation
was not expected on the basis of the Debye double-layer
model. Under the present conditions, amine surfaces can be
only positively charged (–NH2 + H+ A –NH3+). Below the IEP
of zirconia (i.e., pH , 7), zirconia is positively charged, so the
electrostatic interaction should be repulsive regardless of
distance. Conversely, at pH 5 9.7 the interaction should be
attractive.
The reversal in interaction was a general result for amine
surfaces, being observed for primary, secondary, and tertiary
amines. As the number of methyl groups substituted for
hydrogen on the amine increased from zero to two, the surface
forces decreased in magnitude and in range at all values of pH
(Fig. 4).
Fig. 4 AFM force profiles of SAMs with different surface functional
groups under different pHs: (a) secondary and (b) tertiary amine.
A long-range electrostatic repulsion superimposed on a
short-range van der Waals attraction could produce a reversal
such as that seen at pH 5 9.7. However, the electrostatic
interaction expected here is instead attractive. Therefore, an
interplay between the van der Waals and electrostatic
interactions cannot account for the present results.
Table 1 summarizes the signs and relative magnitudes of the
surface charges, and the expected interactions with the zirconia
particle, assuming the existence of a Stern layer on the amine
surfaces. At all values of pH and separation for amine
surfaces, the observed trends (Fig. 3) are consistent with the
existence of a Stern layer.
Table 1 The observed forces (Fobs) between zirconia and amine
SAMs at various pH values (Fig. 3) agree in sign and trend with those
expected in the case of a Stern layer existing on the amine surfaces. yp:
surface potential of zirconia particle; yamine: surface potential of amine
SAM. The number of the + and 2 signs qualitatively indicates the
magnitude of the effect
Amine surface with Stern layer
yamine
pH yp/mV33 (far)
yamine
(near) Fobs (near)
Fobs (far)
2.5 +45
5.5 +35
9.7 235
2 2 2 strong attraction + + +
strong repulsion
repulsion
weak attraction
2 2
2
attraction
weak repulsion
+ +
+
Fig. 3 Force vs. separation between ZrO2 and amine SAMs at
various pHs.
326 | J. Mater. Chem., 2005, 15, 323–330
This journal is ß The Royal Society of Chemistry 2005

View Article Online
At pH 5 2.5, the particle is positively charged. The amine
surface is also positively charged (y44% of amines are
protonated.)6,24 At some distance from the surface beyond
the Stern layer, the environment surrounding the surface
would be negatively charged. Consistent with this picture, the
force curve shows attractive forces between particle and
surface at 1.9–17.4 nm from the surface. This suggests that
the diffuse layer is about 15 nm thick. For separations from 0
to 1.9 nm, the curve clearly shows repulsive forces between the
particle and the surface. This suggests the existence of a Stern
layer of 1.9 nm thick formed by counter-ions from the solution
near the primary amine SAM.
the dominant anion species that forms the Stern layer in the
deposition solution of vanadia is V10O26(OH)242.7 That is, the
Stern layer would consist predominantly of vanadium-
containing polyanions. Thus, it is inferred that the vanadia
films grew via condensation of the Stern layer.
Secondary- and tertiary-amine SAMs. The behavior of both
secondary and tertiary amine SAMs (Fig. 4) was qualitatively
similar to that of the primary amine (Fig. 3). The major
differences were the decreasing magnitude of the forces and the
thicknesses of the Stern layer as well as of the diffuse layer. The
reason for this change is that as the number of methyl groups
increases, protonation is initially discouraged due to the steric
hindrance. This effect should be even more pronounced for
amines closely packed in a SAM, compared with amines in
dilute solution. In dilute solutions, the alkylamine (which is
analogous to an amine SAM) is known to be slightly
more basic than ammonia because the positive charge can
be distributed to the methyl group helping to stabilize the
protonated form (pKb(NH3) 5 4.76 vs. pKb(MeNH2) 5 3.38).35
The effect continues slightly for two C–N bonds in dialkyla-
mine (pKb(Me2NH) 5 3.27). However, as the number of
methyl groups increases to 3 C–N bonds, the basicity drops
(pKb(Me3N) 5 4.21) as steric hindrance of methyl groups
overcomes the benefit of charge distribution to the methyl
groups.35 For the close-packed amines on SAMs, the steric
hindrance would be much higher than in dilute solution.
Therefore, one might expect this steric hindrance to cancel the
charge distribution effect. As a result, the surface charge of
amine SAMs would decrease as the number of methyl groups
increased, leading to the observed decrease in both force and
interaction distances.
Assuming the siloxane-anchored SAM is a well-ordered,
two-dimensional hexagonal array with a characteristic inter-
molecular spacing of 0.45 nm,34 the density of amine groups is
about 5.5 nm22. Cuvillier et al.21,22 observed that the thickness
of the Stern layer increases with increasing density of amine
groups and surface charge. Compared to their system, the
density of amine groups here is higher (5.5 vs. 2–3.6 groups nm22
)
and the degree of protonation is higher due tonm in ref. 21,22).
At near-neutral condition (pH 5 5.5), the particle is weakly
positively charged and the amine surface is positively charged
(y29% of amines are protonated).6,24 The force curve shows
attraction between particle and surface at 0.7–12.4 nm from
the surface. This suggests that the diffuse layer is 11.7 nm thick
and the Stern layer is 0.7 nm thick. The thinner Stern layer at
pH 5 5.5 vs. pH 5 2.5 is consistent with the lower surface
charge due to the lower degree of protonation.
At pH 5 9.7, the particle is negatively charged and the
amine surface would be positively charged (y13% of amines
are protonated),6,24 whereas beyond the Stern layer the
environment is negatively charged. Consistent with this
picture, the force curve shows a weak repulsive force between
particle and surface at 2.4–15.2 nm from the surface, indicating
that the diffuse layer is 12.8 nm thick. At 2.4 nm separation,
jump-in occurred due to an attractive force. The apparent
thickness of the Stern layer, 2.4 nm, is thicker than those at
pH 5 2.5 and 5.5. The appearance of the force reversal at
higher separation at pH 5 9.7 may result from the expected
additional van der Waals attraction. At lower pH, the van der
Waals interaction may have been masked by the stronger
electrostatic forces (higher amine surface charge).
Summary
Table 2 summarizes the numerical results from Fig. 2–4. The
thickness of the Stern layer (SL) on amine surfaces was
calculated using the distance from the surface to the
point where zero force occurred. The thickness of the diffuse
layer (DL) on the alkylammonium surface was determined by
the distance from the surface to the point of zero force,
whereas for the amine surfaces the double layer was
measured from the outer surface of the Stern layer to the
point of zero force. Fig. 2 and 3 illustrate this determination
for pH 5 2.5.
Considering that the Stern layer makes the amine SAM
surface appear to have a negatively charged barrier, it might be
expected to prevent the film formation of negatively charged
colloids like vanadia. However, it has already been reported
that vanadia films can be deposited on amine SAMs.10,11,13a,b
This apparent contradiction is removed when one considers
Note that k21, the distance from the surface where the
potential is considered to be 1/e times the surface potential, as
calculated from the ionic strength of these solutions,36 was
Table 2 Thickness of Stern layer (SL*, in nm), diffuse layer (DL{, in nm), and magnitude of surface force (SF{, absolute value, in 1023 N m21) for
different nitrogen-containing surfaces under various pH values. For these solutions, k21 5 2.2 nm
–C16–NH2
SL*
–C16–NHR
–C16–NR2
SL*
–C16–NR3
DL{
pH
DL{
SF{
SL*
DL{
SF{
DL{
SF{
SF{
2.5
5.5
9.7a
a
1.9 ¡ 0.2
0.7 ¡ 0.1
2.4
15.5 ¡ 2
11.7 ¡ 2
12.8
0.37
0.11
0.30
0.4
0.1
1.6
7.9
6.6
6.4
0.18
0.11
0.18
0.3
0.1
1.6
6.6
5.7
3.8
0.15
0.07
0.20
10.0 ¡ 1
10.0 ¡ 1
5.0
0.80
0.41
0.56
These values are subject to uncertainty, because at pH 5 9.7 the system exhibited jump-in, i.e. it was mechanically unstable.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 323–330 | 327

View Article Online
Table 3 Expected surface forces of primary amine SAMs based on
the surface forces of alkylammonium SAMs and observed surface
forces of primary amine SAMs under various pH values. All the forces
were in units of 1023 N m21
–C16–NH2
–C16–NR3
Surface
force at
pH contact
Degree of
protonation (%) (calculated)
Surface force Surface
force (actual)
2.5
5.5
9.7a
a
0.80
0.41
0.56
44
29
13
0.35
0.12
0.08
0.37
0.11
0.30
These values are subject to uncertainty, because at pH 5 9.7, both
electrostatic and van der Waals interactions would be attractive, and
the system exhibited jump-in, i.e. it was mechanically unstable.
2.2 nm. We expect k21 to be smaller than the thickness of the
diffuse layer (Fig. 1). The data (Table 2) are generally
consistent with these expectations.
The surface charge of amine SAMs under various pH was
previously determined using XPS by our group.6 From those
results, the degrees of protonation (surface charge) of primary
amine can be interpolated to be 44%, 29%, and 13% for
pH 5 2.5, 5.5, and 9.7, respectively. Since alkylammonium salt
is expected to be an ionic surface and strongly dissociated
(100% charged) at any pH, the surface forces on primary
amine and alkylammonium salt can be further compared
(Table 3). The surface force of primary amine is expected to be
the surface force of alkylammonium salt times the degree of
protonation, assuming that the densities of amine and
alkylammonium salt are the same.
The actual surface forces of primary amines at pH 5 2.5 and
5.5 agree well with the calculated values. The differences are
less than 10%, while the spring constant itself contains 10%
error.27 The surface force at pH 5 9.7 was much higher than
the calculated value. However, at pH 5 9.7 the electrostatic
and van der Waals components are both attractive, increasing
the observed interaction. Since the jump-in occurs under
mechanically unstable conditions, this value of force may
contain a deviation of 30%.19 Overall, the surface forces
obtained via different techniques (XPS-based calculation and
AFM measurement) are quite consistent with each other.
Comparisons of the observed forces to those expected from
DLVO theory were not made here. DLVO theory presumes
the existence of a simple diffuse dielectric double layer, which
would not apply to the amine surfaces. The high surface
potential of alkylammonium salt surfaces (.100 mV, as
inferred from zeta potential measurements6), exceeds that
allowed by the Derjaguin approximation (limited to |y| ,
50 mV).37
Fig. 5 AFM force profiles for SAMs of two different carbon chain
lengths at various pHs: (a) amine SAMs, (b) alkylammonium salt
SAMs.
forces, here dominated by electrostatic interactions, determine
the ability of SAM-functionalized surfaces to promote the
formation of oxide films from aqueous media.
Deposition of titania on SAMs
Due to the Stern layer, amine SAMs behave as if they were
negatively charged at distances more than y2 nm from the
surface (with the exact value depending on pH and degree of
methylation). On the other hand, the alkylammonium salt
surface only showed a positive charge.
Difference between sixteen-carbon and eleven-carbon chains
Fig.
5 shows the AFM force curves for amine and
alkylammonium SAMs with sixteen-carbon and eleven-carbon
chains. There were no significant differences between sixteen-
and eleven-carbon chains for the same functionality under the
same pH. This is consistent with the lack of performance
difference for C16 and C11 SAMs with regard to the
deposition of vanadium oxide films.12 These two observations
thus indirectly support the inference that substrate–particle
Fig. 6 SEM images of titania on amine SAMs (continuous film, left)
and alkylammonium SAMs (isolated particles, right).
328 | J. Mater. Chem., 2005, 15, 323–330
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Positively charged particles would be attracted toward the
amine surface due to the negatively charged Stern layer. Once
attracted to the Stern surface, the particle would have a chance
to reach the amine surface via Brownian jumps over the
electrostatic barrier, due to the van der Waals attraction and
possible non-DLVO adhesive forces. In contrast, the alky-
lammonium salt surface would repulse positively charged
Conclusion
In the present work, the force profiles of nitrogen-containing
SAMs in aqueous solutions were investigated using AFM.
Evidence for a Stern layer (charge inversion layer) was
observed on primary, secondary, and tertiary amine
surfaces. The thickness of the Stern layer ranged from 0.1
to 2.4 nm depending on the pH of the solution and the
degree of methylation of the amine. In contrast, the force
profiles exhibited by alkylammonium SAMs indicated a
positively charged surface with a standard ionic double layer.
This interpretation was corroborated by the formation of
TiO2 as well as V2O5 films on amine surfaces but only
V2O5 films (and not TiO2 films) on alkylammonium salt
surfaces.
particles. Therefore,
a positively charged colloid like
titanium(IV) oxide might collect on amine surfaces but not
on alkylammonium salt surfaces, accounting for the growth of
titania films on amine SAMs reported previously.9–11
The film deposited here on amine SAMs (Fig. 6 and 7, left)
was 0.5 mm thick after 22 h of growth. On the other hand, on
alkylammonium salt SAMs, only small isolated clusters of
titania were observed (Fig. 6, right.) Under the same
conditions, a film 1.0 mm thick grew on a sulfonate SAM
(Fig. 7, right), similar to that reported by Pizem et al. (0.92 ¡
0.15 mm)38 and coinciding with that surface’s higher negative
surface charge density. The topography of the films was
similar on amine and sulfonate SAMs (Fig. 7). Both contained
anatase as the only detectable crystalline phase, with sig-
nificant c-axis orientation (XRD data not shown). The cracks
evident in both films in Fig. 7 occur in films thicker than a few
hundreds of nanometers38 due to shrinkage while drying. As
expected, these cracks are more evident in the thicker film on
the sulfonate surface.
SAMs formed from surfactants with two different chain
lengths were studied. There were no significant differences in
forces observed between chains with sixteen carbons and
eleven carbons. Whereas the C16 SAMs were deposited in
bicyclohexyl, the C11 SAMs were deposited in less expensive
dried toluene. Ultimately, the surface functionality decided the
surface force, and neither chain length nor solvent played an
important role.
Titania films were deposited on amine SAMs with prefer-
ential orientation of anatase (00,). The properties like
topography and crystallography of the titania films on amine
SAMs were identical with those of films deposited on sulfonate
SAMs, but the films on amine SAMs were only about half as
thick for equivalent deposition conditions and time.
The partial c-axis orientation of these films has been
attributed38 to the polar nature of the c-axis in anatase. Note
that, if the film formed by ion-by-ion growth, the isoelectric
point of the particles and the pH of the solution would be
irrelevant. Any polar surface would be able to induce
formation of a film. Aligned anatase (00,) would grow on
an alkylammonium salt surface, starting with the oxygen
layers. As this did not occur, it may be taken as an indication
that ion-by-ion growth did not occur on the amine surfaces
either.
Acknowledgements
This sponsorship of the U. S. National Science Foundation
through grant nos. DMR 9803851 and DMR 0203655 is
gratefully acknowledged. The AFM measurements were
performed in the Center for Cardiovascular Biomaterials,
CWRU, Prof. R. Marchant, director. The authors thank
Prof. Lennart Bergstrom (Institute for Surface Chemistry,
Stockholm, Sweden) for donating the spherical zirconia
particles.
Ion-by-ion growth is just one type of surface-nucleated
growth. Growth could occur in principle by condensation of
complexes. Considering that cationic complexes predominate
in the titanium solution at 2 , pH , 3,7 it might be that the
anatase films were formed via ‘‘complex-by-complex’’ growth,
followed by transformation to the oxide. In this case, no film
would be formed on alkylammonium salt surfaces (as observed
here), while the Stern layer on amine surfaces could induce film
formation. In other words, growth via surface nucleation (as
opposed to particle attachment) cannot be ruled out by the
present results.
Jing-Jong Shyue, Yin Tang and Mark R. De Guire*
Department of Materials Science & Engineering, Case Western Reserve
University, 10900 Euclid Avenue, Cleveland, Ohio 44106-7204, USA.
E-mail: mrd2@cwru.edu; Fax: +1(216)368-3209; Tel: +1(216)368-4221
References
1 A. W. Adamson and A. P. Gast, Physical Chemistry of Surfaces,
6th edn., John Wiley & Sons, Inc., New York, 1997.
2 Organic-Inorganic Hybrid Materials II, ed. L. C. Klein, L. F. Francis,
M. R. De Guire and J. E. Mark, MRS Symposium Proceedings,
Materials Research Society, Warrenville, Pennsylvania, 1999,
vol. 576.
3 S. Mann, in Biomineralization: Principles and Concepts in
Bioinorganic Materials Chemistry, ed. S. Mann, Oxford
University Press, Oxford, 2001, pp. 156–193.
4 T. P. Niesen and M. R. De Guire, J. Electroceram., 2001, 6,
169–207.
5 H. Shin, M. Agarwal, M. R. DeGuire and A. H. Heuer, Acta
Mater., 1998, 46, 801–815.
Fig. 7 SEM images of titania on amine SAMs (left) and sulfonate
6 J.-J. Shyue, M. R. De Guire, T. Nakanishi, Y. Masuda,
K. Koumoto and C. N. Sukenik, Langmuir, 2004, 20, 8693–8698.
SAMs (right).
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 323–330 | 329

View Article Online
7 C. F. J. Baes and R. E. Mesmer, in The Hydrolysis of Cations,
Robert E. Krieger Publishing Company, Inc., Malabar, FL, 1976,
pp. 197–210.
8 (a) J. Livage, Chem. Mater., 1991, 3, 578–593; (b) S. Kittaka,
N. Fukuhara and H. Sumida, J. Chem. Soc., Faraday Trans., 1993,
89, 20, 3827–3832.
21 N. Cuvillier, M. Bonnier, F. Rondelez, D. Paranjape, M. Sastry
and P. Ganguly, Progr. Colloid Polym. Sci., 1997, 105, 118–125.
22 N. Cuvillier and F. Rondelez, Thin Solid Films, 1998, 327–329,
19–23.
23 N. Balachander and C. N. Sukenik, Langmuir, 1990, 6, 1621–1627.
24 J.-J. Shyue and M. R. De Guire, Trans. MRS J., 2004, 29, 5,
2383–2386.
25 P. W. Atkins, Physical Chemistry, 5th edn., Oxford University
Press, Oxford, 1994, p. 321.
26 A. C. Hillier, S. Kim and A. J. Bard, J. Phys. Chem., 1996, 100,
18808–18817.
27 T. J. Senden and W. A. Ducker, Langmuir, 1994, 10, 1003–1004.
28 S. Deki, Y. Aoi, Y. Asaoka, A. Kajinami and M. Mizuhata,
J. Mater. Chem., 1997, 7, 733–736.
29 K. Koumoto, S. Seo, T. Sugiyama and W. S. Seo, Chem. Mater.,
1999, 11, 2305–2309.
30 S. Deki, Y. Aoi, H. Yanagimoto, K. Ishii, K. Akamatsu,
M. Mitzuhata and A. Kajinami, J. Mater. Chem., 1996, 6,
1879–1882.
31 H. Nagayama, H. Honda and H. Kawahara, J. Electrochem. Soc.,
1988, 135, 2013–2016.
32 H. Kishimoto, K. Takahama, N. Hashimoto, Y. Aoi and S. Deki,
J. Mater. Chem., 1998, 8, 2019–2024.
33 M. Prica, S. Biggs, F. Grieser and W. Healy, Colloids Surf. A—
Physicochem. Eng. Asp., 1996, 119, 205–213.
34 H. Shin, M. Agarwal, M. R. De Guire and A. H. Heuer, Acta
Mater., 1998, 46, 801–815.
35 K. Peter, C. Vollhardt and N. E. Schore, Organic Chemistry, 2nd
edn., W. H. Freeman and Company, New York, 1994, pp. 832–
833.
36 R. J. Hunter, Foundations of Colloid Science; Clarendon Press,
Oxford, 1987, vol. 1.
37 R. Hogg, T. W. Healy and D. W. Fuerstenau, Trans. Faraday Soc.,
1966, 62, 1638.
9 T. P. Niesen, J. Bill and F. Aldinger, Chem. Mater., 2001, 13,
1552–1559.
10 T. P. Niesen, J. Wolff, J. Bill, M. R. De Guire and F. Aldinger, in
Organic-Inorganic Hybrid Materials II, ed. L. C. Klein, L. F.
Francis, M. R. De Guire and J. E. Mark, Materials Research
Society, Warrendale, Pennsylvania, 1999, vol. 576, pp. 197–202.
11 T. P. Niesen, J. Wolff, J. Bill, T. Wagner and F. Aldinger, in 9th
Cimtec World Forum on New Materials, ed. P. Vincenzini, 1999,
Techina S. R. L., Faenza, Italy, pp. 27–34.
12 J.-J. Shyue, Deposition of Vanadium (V) Oxide Thin Films on
Nitrogen-Containing Self-Assembled Monolayers, thesis in
Department of Materials Science and Engineering, Case Western
Reserve University, Cleveland, OH, 2002.
13 (a) J.-J. Shyue and M. R. De Guire, in CIMTEC 2002 (10th
International Ceramics Congress and 3rd Forum on New Materials),
ed. P. Vincenzini, 2002, Techina S. R. L., Faenza, Italy, p 469; (b)
J.-J. Shyue and M. R. De Guire, Chem. Mater., 2005, DOI:
10.1021/cm048501x.
14 I. Larson, J. Am. Chem. Soc., 1993, 115, 11885–11890.
15 G. R. Wiese and T. W. Healy, J. Colloid Interface Sci., 1975, 51,
427–433.
16 H. Yotsumoto and R. H. Yoon, J. Colloid Interface Sci., 1993, 157,
426–433.
17 W. A. Ducker, Nature, 1991, 353, 239–241.
18 W. A. Ducker and T. J. Senden, Langmuir, 1992, 8, 1831–1836.
19 Y. Tang and M. R. De Guire, J. Mater. Chem., 2004, 14,
1173–1179.
20 D. J. Shaw, in Introduction to Colloid and Surface Chemistry, 3rd
edn.,1992, Butterworths, Boston, pp. 148–162.
38 H. Pizem, C. N. Sukenik, U. Sampathkumaran and M. R. De Guire,
Chem. Mater., 2002, 14, 2476–2485.
330 | J. Mater. Chem., 2005, 15, 323–330
This journal is ß The Royal Society of Chemistry 2005