Characterization of streptavidin binding to biotinylated, binary self-assembled thiol monolayers - Influence of component ratio and solvent

Article abstract of DOI:10.1021/la904087s

Many biosensor applications are based on streptavidin (SA) binding to partially biotinylated self-assembled thiol monolayers (SAMs). In our study, binary SAMs on gold were prepared from solutions containing 16-mercapto-1- hexadecanol (thiol I) and N-(8-biotinyl-3,6-dioxa-octanamidyl)-16- mercaptohexadecanamide (thiol II) in varying component ratios. Either chloroform or ethanol was used as solvent. After 24 h thiol incubation, SA was immobilized on the resulting SAMs using the strong SA-biotin interaction. The SA binding process was monitored by QCM-D (quartz crystal microbalance monitoring dissipation factor). It is shown that the Sauerbrey equation is valid to calculate the mass quantities of the immobilized SA layers. Under the chosen incubation conditions, marginal fractions of the biotinylated component II in chloroform ((n_I/n_II)_solution ≈ 1000) lead to SAMs which ensure a maximal SA binding quantity of m_{Sauerbrey SA} ≈ 400 ngcm^-2, being equivalent to a SA single-layer arrangement on the SAM surface. In case of incubations from ethanolic solutions, a complete SA layer formation needs significantly higher amounts of the biotinylated component II during SAM preparation ((n_I/n_II) _solution ≈ 50). X-ray photoelectron spectroscopy data show that the fraction of biotinylated thiol II in the SAM determines the amount of surface-bound SA. The SAM thiol ratio ((n_I/n_II) _SAM) not only depends on the corresponding component ratio in the incubation solution, but is also strongly influenced by the solvent. Using chloroform as solvent during SAM preparation significantly increased the fraction of biotinylated thiol II in the SAMs compared to ethanol.

Full text of DOI:10.1021/la904087s

pubs.acs.org/Langmuir
©
2010 American Chemical Society
Characterization of Streptavidin Binding to Biotinylated, Binary Self-
Assembled Thiol Monolayers;Influence of Component Ratio and Solvent
†
‡
,†
Michael Seifert, Matthias T. Rinke, and Hans-Joachim Galla*
†
Institute of Biochemistry, University of Muenster, Wilhelm-Klemm-Strasse 2, 48149 Muenster, Germany, and
Institute of Physical Chemistry, University of Muenster, Corrensstrasse 28/30, 48149 Muenster, Germany
‡
Received October 27, 2009. Revised Manuscript Received January 28, 2010
Many biosensor applications are based on streptavidin (SA) binding to partially biotinylated self-assembled thiol
monolayers (SAMs). In our study, binary SAMs on gold were prepared from solutions containing 16-mercapto-1-
hexadecanol (thiol I) and N-(8-biotinyl-3,6-dioxa-octanamidyl)-16-mercaptohexadecanamide (thiol II) in varying
component ratios. Either chloroform or ethanol was used as solvent. After 24 h thiol incubation, SA was immobilized
on the resulting SAMs using the strong SA-biotin interaction. The SA binding process was monitored by QCM-D
(
quartz crystal microbalance monitoring dissipation factor). It is shown that the Sauerbrey equation is valid to calculate
the mass quantities of the immobilized SA layers. Under the chosen incubation conditions, marginal fractions of the
biotinylated component II in chloroform ((n
I
/nII
)
solution ≈ 1000) lead to SAMs which ensure a maximal SA binding
-
2
quantity of mSauerbrey SA ≈ 400 ng cm , being equivalent to a SA single-layer arrangement on the SAM surface. In case
3
of incubations from ethanolic solutions, a complete SA layer formation needs significantly higher amounts of the
biotinylated component II during SAM preparation ((n
that the fraction of biotinylated thiol II in the SAM determines the amount of surface-bound SA. The SAM thiol ratio
(n /nII)SAM) not only depends on the corresponding component ratio in the incubation solution, but is also strongly
I
/nII
)
solution ≈ 50). X-ray photoelectron spectroscopy data show
(
I
influenced by the solvent. Using chloroform as solvent during SAM preparation significantly increased the fraction of
biotinylated thiol II in the SAMs compared to ethanol.
Introduction
biomolecules are immobilized in non-native, unknown conforma-
1
3
tional orientation, often unspecifically bound. Such a disor-
dered arrangement may lead to a decreased biological function
and altered structural properties.
One strategy to obtain compatibility of inorganic and biologi-
cal components is the use of the well-known streptavidin-biotin
(SA-biotin) interaction exhibiting an extremely high binding
Within current biotechnology research, one major emphasis
lies in adequate biochip development for monitoring interactions
and functions of biologically active molecules in a high-through-
1-3
put and reproducible manner.
Nowadays, these systems gain
4
broad application in analysis of environmental pollutants, expre-
5
6
ssion profiling, or medical diagnostics.
1
3
-1 14
7
8
constant (Ka ≈ 10
M ). The SA protein with its four
A general feature of such biosensors are noble metal or silica
surfaces as solid supports to connect the corresponding bio-
6
equivalent binding pockets for biotin can be used as connecting
entity between a prefunctionalized, biotinylated transducer sur-
face and the (also prebiotinylated) biological component of
interest. In many cases, this assembly can be established without
9
molecule to a transducer based on, e.g., piezoelectric, spectro-
1
0
11
12
scopic, electrochemical, or optical techniques. The ap-
plication of all these surface-sensitive analytical tools demands an
interface between an inorganic substrate and a biological system.
Without appropriate modification of the transducer surface, the
1
5
noticeable impact on function and activity of the biomolecule.
In the case of noble metal transducers (e.g., gold substrates),
the biotinylation of the inorganic surface is often accomplished by
constitution of self assembled monolayers (SAMs) composed of
alkanethiols, partially functionalized with biotin. A great benefit
of these SAMs in comparison to, e.g., surface functionalization by
spreading of biotin doped liposomes, is the relatively easy
*
Corresponding author. Hans-Joachim Galla. Mailing address: Institute of
Biochemistry, Wilhelm-Klemm-Strasse 2, 48149 Muenster, Germany. Phone:
049/251-833200. Fax: 0049/251-8333206. E-mail: gallah@uni-muenster.de.
1) Tanaka, M.; Sackmann, E. Nature 2005, 437, 656.
2) Schaferling, M.; Riepl, M.; Pavlickova, P.; Paul, H.; Kambhampati, D.;
0
(
(
7
preparation and long-term stability/functionality. In the case
Liedberg, B. Microchim. Acta 2003, 142, 193.
of rather long alkyl chains, hydrophobic forces among the
chemisorbed thiols lead to highly organized monolayers, which
(
3) Su, X. D.; Wu, Y. J.; Robelek, R.; Knoll, W. Langmuir 2005, 21, 348.
(
4) Helali, S.; Ben Fredj, H.; Cherif, K.; Abdelghani, A.; Martelet, C.; Jaffrezic-
1
6
Renault, N. Mater. Sci. Eng. C 2008, 28, 588.
5) Schena, M.; Shalon, D.; Heller, R.; Chai, A.; Brown, P. O.; Davis, R. W.
Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 10614.
completely shield the hydrophilic metal surface. The physico-
chemical character of thesurfaceofthe sensor system is exclusively
determined by the SAM molecules and their different headgroup
(
(
6) Grunwald, C. Z. Phys. Chem. 2008, 222, 789.
1
7,18
(
7) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M.
modifications.
SAMs for subsequent SA immobilization
Chem. Rev. 2005, 105, 1103.
8) Mehne, J.; Markovic, G.; Proll, F.; Schweizer, N.; Zorn, S.; Schreiber, F.;
Gauglitz, G. Anal. Bioanal. Chem. 2008, 391, 1783.
(
(13) Ladd, J.; Boozer, C.; Yu, Q. M.; Chen, S. F.; Homola, J.; Jiang, S. Langmuir
2004, 20, 8090.
(
9) Larsson, C.; Rodahl, M.; Hook, F. Anal. Chem. 2003, 75, 5080.
(
10) Pradier, C. M.; Salmain, M.; Liu, Z.; Methivier, C. Surf. Interface Anal.
(14) Green, N. M. Adv. Protein Chem. 1975, 29, 85.
(15) Jung, L. S.; Nelson, K. E.; Stayton, P. S.; Campbell, C. T. Langmuir 2000,
16, 9421.
(16) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.;
Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321.
2
1
7
002, 34, 67.
11) Hays, H. C. W.; Millner, P. A.; Prodromidis, M. I. Sens. Actuators B 2006,
14, 1064.
12) Reimhult, E.; Larsson, C.; Kasemo, B.; Hook, F. Anal. Chem. 2004, 76,
211.
(
(
(17) Bain, C. D.; Evall, J.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111, 7155.
6
386 DOI: 10.1021/la904087s
Published on Web 02/16/2010
Langmuir 2010, 26(9), 6386–6393

Seifert et al.
Article
are often composed ofbinary thiol mixtures, one biotinylated, one
terminated with a so-called spacer or diluent function like hydro-
1
xyl
9-22
15,23,24
or oligoethylene glycol
groups. The biotin func-
tions protrude from the surface being accessible for specific SA
2
2
linkage, whereas the diluent function prevents unspecific bind-
2
0,25
ing.
The protrusion effect is enhanced by insertion of oli-
goethylene glycol spacers between the alkyl chain and the biotin
function, ensuring flexibility of the linker unit. The precise
fraction of biotinylated thiol needed for maximal SA binding
seems to depend on structural properties like alkyl chain length
1
5,20,24,26
and/or headgroup character.
Even marginal structural
variations might be responsible for changes in SAM surface
2
6
texture, resulting in differing SA immobilization quantities. In
addition, the thiol ratio in the SAMs often deviates from the
component ratio in the incubation solution, since chemisorp-
tion efficiency depends on both the structural properties of the
corresponding constituent and the solvent used during SAM
Figure 1. Schematic view of a binary thiol SAM (thiol compo-
nents I and II) on gold. SA may bind to one or two protruding
biotin moieties. (SA crystal structure was taken from RCSB
Protein Data Bank (www.rcsb.org)).
1
7,24,27
formation.
In the present study, we focus on preparation and analysis of
partially biotinylated binary SAMs composed of a hydroxyl
functionalized diluent component (16-mercapto-1-hexadecanol,
(0.02 M tris(hydroxymethyl)aminomethane (tris base) (Sigma/
GER), 0.1 M sodium chloride (NaCl) (AppliChem) in water) was
adjusted to pH 7.4 by hydrochloric acid (p.a.). Streptavidin (MSA
thiol I) and
a biotinylated one (N-(8-biotinyl-3,6-dioxa-
≈
55 kDa) (Rockland/USA) was received in the form of lyoph-
octanamidyl)-16-mercaptohexadecanamide, thiol II) (Figure 1).
For SAM formation, we used chloroform and ethanol as incuba-
ilized powder (from 0.15 M NaCl in water). Storage was con-
ducted at -20 °C after restoring in water (1 mg/mL, 0.1 mL
aliquots. Thawed stock solutions were stored at 4 °C for several
weeks without further freeze/thaw cycles. Diluted Mucasol
tion solvent. Thiol SAM ratios ((n /nII)SAM) and subsequent SA
I
immobilization capacities were determined by QCM-D (quartz
3
,9,28,29
crystal microbalance monitoring dissipation factor
and
(
Merz/GER) (5 mL/L water) was used as detergent for gold
XPS (X-ray photoelectron spectroscopy) analysis. We will show
that the SA adsorption efficiency depends on the thiol SAM ratio
substrate cleaning. Gold surfaces applied for XPS measurements
were prepared with Tempax glass slides (Rettberg/GER), chro-
mium (Bal Tec/FL), and gold (Degussa/GER). QCM-D analysis
was accomplished by using commercially available AT-cut quartz
crystals (QSX 301-Standard Gold from Q-Sense/Sweden).
Synthesis of Thiol II. N-(8-Biotinyl-3,6-dioxa-octanamidyl)-
(
n /n )
and that (n /n )
is strongly influenced by the
I
II SAM
I
II SAM
choice of thiol solvent during SAM formation.
Experimental Section
16-mercaptohexadecanamide (Figure 1) was synthesized by a
combination of former developed synthesis pathways in a three
Materials. 16-Mercaptohexadecanoic acid/90% (Sigma-Al-
ꢀ
drich/GER), 2,2-(ethylenedioxy)bis-(ethylamine) (Aldrich/
26,30,31
step synthesis.
DIPEA and directly coupled to dropwise added 2,2 -(ethylene-
dioxy)-bis(ethylamine) (step 1). The carboxylic acid function of
Biotin was in situ activated by TSTU/
0
0
0
GER), biotin (Sigma/GER), O-(N-succinimidyl)-N,N,N ,N -tet-
ramethyl-uroniumtetrafluoroborate (TSTU) (Fluka/GER), N,N-
diisopropylethylenamine (DIPEA) (Sigma-Aldrich/GER), N-
16-mercaptohexadecanoic acid was activated for amine coupling
0
hydroxysuccinimide (NHS) (Aldrich/GER), and N,N -dicyclo-
by NHS/DCC (step 2). A crucial advantage of the gentle NHS/
DCC method in comparison to TSTU coupling procedure lies in
preserving the free thiol function. Using the very fast and effi-
hexylcarbodiimide (DCC) (Aldrich/GER) were used for synthesis
of N-(8-biotinyl-3,6-dioxa-octanamidyl)-16-mercaptohexadeca-
namide (thiol II). 16-Mercapto-1-hexadecanol/99% (thiol I)
30
cient (and thus less selective) TSTU pathway resulted in irre-
versible bonding between the free thiol function and the
carbodiimide derivative fragment of TSTU.
(
Frontier Scientific Europe Ltd., UK) was used without further
purification. Chloroform and ethanol in HPLC grade (Merck/
GER) were applied as organic solvents for thiol incubations.
Ammoniac solution (25%) and hydroxyperoxide (30%) for gold
substrate cleaning were purchased from Merck/GER. Water for
SAM rinsing and buffer preparation was purified and deionized
by a Millipore multicartridge system (Billerica/USA). TBS buffer
Using DCC, previous protection (andfinal deprotection) of the
thiol function could beavoided, leading toa morestraightforward
synthesis pathway. In a last step (step 3), the NHS-activated 16-
mercaptohexadecanoic acid was coupled to the biotinylated
amine resulting in thiol II for SAM formation.
SAM Preparation. Thiol SAMs were constructed on either
QSX 301-Standard Gold AT-cut quartz crystals (QCM-D
measurements) or self-made gold supports (XPS analysis). The
in-house fabricated gold layers were arranged by consecutive
deposition of a 2 nm chromium adhesion layer and a 100-nm-
(
(
18) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111, 7164.
19) Yao, D. F.; Kim, J.; Yu, F.; Nielsen, P. E.; Sinner, E. K.; Knoll, W. Biophys.
J. 2005, 88, 2745.
20) Perez-Luna, V. H.; O’Brien, M. J.; Opperman, K. A.; Hampton, P. D.;
Lopez, G. P.; Klumb, L. A.; Stayton, P. S. J. Am. Chem. Soc. 1999, 121, 6469.
(
2
(
(
(
21) Stengel, G.; Hook, F.; Knoll, W. Anal. Chem. 2005, 77, 3709.
22) Azzaroni, O.; Mir, M.; Knoll, W. J. Phys. Chem. B 2007, 111, 13499.
23) Jung, L. S.; Nelson, K. E.; Campbell, C. T.; Stayton, P. S.; Yee, S. S.;
thick gold layer on extensively purified glass slides of (0.5-1) cm
3
2
size. A detailed description can be found elsewhere. All gold
supports were precleaned in 5/1/1 water/ammoniac solution
Perez-Luna, V.; Lopez, G. P. Sens. Actuators B 1999, 54, 137.
24) Nelson, K. E.; Gamble, L.; Jung, L. S.; Boeckl, M. S.; Naeemi, E.; Golledge,
S. L.; Sasaki, T.; Castner, D. G.; Campbell, C. T.; Stayton, P. S. Langmuir 2001, 17,
807.
25) Ostuni, E.; Chapman, R. G.; Holmlin, R. E.; Takayama, S.; Whitesides,
G. M. Langmuir 2001, 17, 5605.
26) Riepl, M.; Enander, K.; Liedberg, B.; Schaferling, M.; Kruschina, M.;
Ortigao, F. Langmuir 2002, 18, 7016.
27) Xing, Y. F.; Li, S. F. Y.; Lau, A. K. H.; O Shea, S. J. J. Electroanal. Chem.
005, 583, 124.
(
25%)/hydroxyperoxide (30%) (volume ratio) for 5 min at
(
80 °C. Accurate purification was reached by consecutive washing
withdetergent solution(Mucasol) and waterat50°C for 15min in
each case. After drying in a nitrogen stream, the substrates were
2
(
(
(30) Bannwarth, W.; Knorr, R. Tetrahedron Lett. 1991, 32, 1157.
(31) Charvet, N.; Reiss, P.; Roget, A.; Dupuis, A.; Grunwald, D.; Carayon, S.;
Chandezon, F.; Livache, T. J. Mater. Chem. 2004, 14, 2638.
(32) Bourdos, N.; Kollmer, F.; Benninghoven, A.; Ross, M.; Sieber, M.; Galla,
H. J. Biophys. J. 2000, 79, 357.
0
(
2
(
(
28) Rodahl, M.; Hook, F.; Kasemo, B. Anal. Chem. 1996, 68, 2219.
29) Rodahl, M.; Kasemo, B. Rev. Sci. Instrum. 1996, 67, 3238.
Langmuir 2010, 26(9), 6386–6393
DOI: 10.1021/la904087s 6387

Article
Seifert et al.
stored until utilization. Directly before immersion in thiol solu-
tions, the gold substrates were treated with argon plasma (plasma
cleaner, PDC 32G-2, Harrick/USA) for 5 min. All incubations
were done from thiol solutions in chloroform or ethanol at room
temperature with an overall component concentration of 0.1 mM
Fairley/www.casaxps.com). Each experiment was repeated three
times from independent sample preparations.
Evaluation of QCM-D data;Theoretical basis. QCM-D
measures the acoustical response of a shear oscillating, piezo-
electric quartz crystal by temporal shifts in the intrinsic quartz
crystal shear oscillation resonant frequency (Δf) and changes in
(cThiol = 0.1 mM) and a total volume of 2 mL (QSX 301-standard
1
2
gold quartz crystals, QCM-D measurements) or 4 mL (self-made
supports, XPS analysis), respectively. Adequate stock solutions
were prepared in deoxygenated chloroform and ethanol; storage
was conducted at -20 °C. All incubations were done in weighing
bottles to avoid any solvent vaporization. After 24 h thiol
incubation time, the gold supports were removed from the thiol
solutions, rinsed successively with 10 mL chloroform or 10 mL
ethanol and 30 mL water, dried under a nitrogen stream, and
directly used for XPS/QCM-D analysis.
dissipation factor (ΔD). The latter represents the inverse of the
29
more common Q-factor and displays the ratio of the energy
dissipated in one period of oscillation (Edissipated) and the total
energy stored in the system (Estored).
In the case of sufficiently thin layers adsorbed to the oscillating
quartz plate, frequency shifts are often directly converted to the
mass of material bound to the sensor surface. Hereby, quantifica-
3
4
tion is based on the well-known Sauerbrey relation, which
linearly relates frequency shift of the nth overtone (Δf ) to the
n
All analyzed SAMs densely covered the gold substrates. This
was proven by impedance spectroscopy (data not shown). All
immobilized mass (ΔmSauerbrey)
Δf
n
-2
SAMs had capacitances of ≈1 μF cm , which agrees with
ΔmSauerbrey ¼ Ffilm
h
film
ð1Þ
3
n
reported values for tightly packed n-alkane SAMs with an alkyl
3
3
chain length of ≈16 carbon atoms.
where CQCM denotes the mass sensitivity constant
(17.7 ng cm Hz for 5 MHz quartz crystals) and n is the
-2
-1
QCM-D Analysis. QCM-D measurements were performed
using a Q-Sense E1 system (Q-Sense/Sweden), connected to a
peristaltic flow module. The QCM-D device allows time-resolved
and simultaneous analysis of resonance frequency shifts (Δf) and
dissipation factor changes (ΔD) during mass deposition on shear
oscillating quartz crystal surfaces covered with gold electrodes.
The Δf and ΔD changes of the fundamental resonance frequency
of the quartz shear oscillation (n = 1, 5 MHz for QSX 301-
Standard Gold quartz crystals, normally omitted due to its spatial
sensitivity across the wafer), as well as for harmonics with n = 3,
3
3
accordant overtone number. However, if the quartz plate is
immersed in a Newtonian bulk liquid this relation for thin layers
only holds if the adsorbed film can be regarded as an ideally rigid
mass layer, homogeneously distributed and with insignificant
3
5,36
viscoelastic properties.
in dissipation factor induced by the attached film are marginal and
Under these circumstances, the changes
12
exhibit negligible influence on the mass calculation.
In the case of adsorption of soft layers, internal viscous energy
dissipative) lossesmight occur duringshear oscillation. The shear
(
acoustic wave propagating into and through the film underlies
explicit damping, originating from, e.g., hydrodynamically
coupled water to the macromolecules constituting the adsorbed
5
,
, 13 were monitored.
Data sets were acquired using QSoft 401 software (Q-Sense/
3
3 3
Sweden). For measurements, SAM-functionalized quartz crystals
were mounted in the measurement chamber and constantly over-
flowed with TBS buffer, pH 7.4, in flow-through. All measure-
ments were accomplished under virtually nonperturbing flow
conditions, provided by the Q-Sense flow module with a flow
velocity of 0.2 mL liquid per minute at 20 °C. For monitoring
SA immobilization, TBS buffer was replaced by SA solution
3
,9,35-37
layer.
This can lead to a theoretical description more
complex than the Sauerbrey relation. In QCM-D analysis, the
Voight-Voinova representation is used to describe the viscoelastic
properties of the layer. The adsorbed film is described by a
complex shear modulus leading to analytical expressions for Δf
and ΔD changes containing the mass (mfilm = Ffilmhfilm) response
-1
as well as the elastic shear modulus (μ) and shear viscosity (η) of
the layer. The Voight representation accounts for a frequency
(
5 μg mL SA in TBS, pH 7.4 (1 mL), circular flow). SA
3
38
immobilizations were followed by rinsing with TBS buffer,
ensuring disposal of loosely bound SA from the gold surface. Each
SA immobilization experiment (specifiable by its SAM prefunctio-
nalization component ratio and type of solvent) was repeated at
least three times from independent sample preparations.
(
overtone) dependence of normalized Δf shifts as well as for
significant changes in ΔD during layer deposition; the effective
mass of the film (including surface coupled water) can be calcu-
lated in sufficient approximation.
In the present study, SA immobilization experiments moni-
tored by QCM-D were quantified by using the Sauerbrey relation
(eq 1) and by modulating the measured Δf and ΔD changes with
XPS Analysis. XPS spectra were obtained using an AXIS-
ULTRA spectrometer (KRATOS, Manchester, UK) in ultrahigh
-7
n
n
R
vacuum (<10 Pa). Monochromatic Al K radiation (hν =
486.6 eV) was used with a 15 kV accelerating voltage and 10 mA
the Voight-based representation using Q-Tools 3.0.5.198 software
Q-Sense/Sweden). The SA layers are specified by their calculated
mass and thickness quantities.
1
(
filament current (≈500 μm spot diameter). The charge neutralizer
was run with a filament current of 1.8 A, a charge balance of 2.3 V,
and a filament bias of 1.0 V. For wide survey scans, a pass energy
of 160 eV was applied, while high-resolution scans for core-level
spectra were acquired with a pass energy of 20 eV. For each
sample, XPS spectra were collected by employing a takeoff angle
of θ = 90°. For each sample, 3 single measurements were taken at
different spots on the sample surface. Between 10 and 20 scans
were acquired to get smooth lines and minimize analysis errors.
Further data analysis was done using the software CasaXPS v
Results
QCM-D Analysis of SA Binding to SAMs. SA binding
capacities to thiol SAMs were analyzed via QCM-D. SAMs were
formed on gold supports from incubation solutions with varying
ratios of thiol I and II (Figure 1). Either chloroform or ethanol
was used as solvent. Incubations lasted 24 h to guarantee a tight,
7
2.2.0 (Neal Fairley/www.casaxps.com). The C1s peak was used as
an internal reference, and the binding energy of the signal of the
ideal thiol packing density.
Figure 2 displays frequency (Δf ) and dissipation factor (ΔD)
exclusively aliphatic C-C bond was assumed to be 285.0 eV
changes as a function of time resulting from SA application to
(
referenced to its Fermi level). The raw spectra were subsequently
smoothed and fitted using mixed Lorentzian-Gaussian line
shapes. Quantification was done using the survey spectra and
sensitivity factors from the CasaXPS element library (Neal
(
34) Sauerbrey, G. Z. Phys. 1959, 155, 206.
(35) Cho, N. J.; Kanazawa, K. K.; Glenn, J. S.; Frank, C. W. Anal. Chem. 2007,
9, 7027.
7
(
36) Hook, F.; Kasemo, B.; Nylander, T.; Fant, C.; Sott, K.; Elwing, H. Anal.
Chem. 2001, 73, 5796.
37) Patel, A. R.; Frank, C. W. Langmuir 2006, 22, 7587.
(38) Voinova, M. V.; Rodahl, M.; Jonson, M.; Kasemo, B. Phys. Scr. 1999, 59,
391.
(
(
33) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem.
Soc. 1987, 109, 3559.
6
388 DOI: 10.1021/la904087s
Langmuir 2010, 26(9), 6386–6393

Seifert et al.
Article
Table 1. Layer Mass (m) and Thickness (h) Resulting from SA
Deposition on Different Biotinylated SAMs (N = 3), Specified by
a
I solution
Their Component Solution Ratio ((n /nII) )
Sauerbrey
Voight
-2
-2
m
SA (ng cm ) hSA (nm) mSA (ng cm ) hSA (nm)
3
3
(
(
n
n
I
/nII
)
solution = 50
413 ( 16
199 ( 10
3.6 ( 0.2
1.7 ( 0.1
415 ( 16
202 ( 15
3.6 ( 0.1
1.8 ( 0.2
Calculations were done via Sauerbrey and Voight method. An
I
/nII
)
solution = 10 000
a
effective SA layer density of Ffilm = 1150 kg m- was assumed.
339
Figure 3. Immobilized massquantitiesmSauerbrey SA resulting from
-
1
SA application (5 μg mL ) to SAMs formed with different
3
I
solution ratios (n /nII)solution (24 h incubation, cThiol = 0.1 mM).
The SAMs were formed in chloroform or ethanol.
5 5
Figure 2. Δf and ΔD data versus time (5th harmonic) resulting
1
-
from SA application (5 μg mL ) to SAMs prepared with varying
3
shifts upon SA exposure were negligible (Δf ≈ 1 Hz, ΔD ≈ 0,
5
5
molar thiolratiosinsolution(24h incubation). (A)(n
50, (B) (n solution = 10 000. Incubations were done in chloro-
I solution
/nII) =
data not shown).
I
/nII)
SA layer mass quantities resulting from SA application to
biotinylated SAMs as shown in Figure 2 were calculated both via
form (cThiol = 0.1 mM).
3
4
SAM functionalized Standard Gold quartz crystals. SAMs were
formed from thiol solutions in chloroform composed of two
Sauerbrey (cf. eq 1, Δf (n = 5, i.e., 25 MHz)) and Voight
5
38
method (the data of the fifth, seventh, and ninth overtone were
used). The data of three independent measurements were used
(N = 3). Additionally, the layer thicknesses were calculated,
different molar ratios of I and II (solution ratio (n /n )
). In
II solution
I
each case, Δf and ΔD data of the fifth overtone, normalized by the
-3 39
overtone number, are shown (Δf5 and ΔD ). For all other
assuming an effective SA film density of Ffilm = 1150 kg m .
3
5
harmonics, the normalized Δf and ΔD changes display con-
Table 1 lists the corresponding data. Both calculations yield the
same mass quantities, thus denoting that the linear proportion-
ality approximation between frequency shift and adsorbed mass
holds in the present case. Thereby, SA applications to SAMs
n
n
gruent traces (data not shown). The moment of SA addition
(
point “t ”) and the start of buffer rinsing after the completed
1
process (point “t ”) are indicated in the plots.
2
The QCM-D data in Figure 2 display differing characteristics,
primary observable from the varying shifts of resonance fre-
formed from a molar solution ratio of (n /n )
significantly higher SA mass/layer thickness immobilized on the
SAM surface than the exposure to the SAMs formed from
= 50 lead to a
I
II solution
quency. While the incubation with a solution ratio of (n /n )
=
I
II solution
5
0 (Figure 2A) gives a normalized frequency shift of Δf ≈ -25
incubations with less amount of thiol II ((n /n )
A more detailed view on the SA binding behavior of binary
SAMs is given in Figure 3. SAMs were formed from 24 h
= 10000).
5
I
II solution
Hz, the SA application to a SAM formed from a mixture of
n /n ) = 10 000 (Figure 2B), shows less of a frequency
(
I
II solution
shift (Δf ≈ -12 Hz). 80-90% of Δf changes are achieved after
incubations with varying thiol solution ratios (n /n )
Either chloroform or ethanol served as thiol solvent. Figure 3
shows the accumulated masses (mSauerbrey SA) onto the SAMs
.
5
I
II solution
only 3-5 min.
Examination of the dissipation factor (ΔD ) traces again
5
demonstrates a differing behavior of the two systems. For
specified by (n /nII)solution.
I
the SAM formed from a solution ratio of (n /n )
= 50
In the case of chloroform as thiol solvent, a SA layer mass
I
II solution
-
6
-2
(
Figure 2A), a rapid increase to ΔD ≈ 0.2 ꢀ 10 can be observed
of mSauerbrey SA ≈ 400 ng cm is observed over a broad range of
5
3
during the first 2 min. The initial rise is followed by a decrease for
another 5 min, which then passes into an almost constant plateau,
only influenced by a small drift (ΔD j 0.1 ꢀ 10 within 30 min).
thiol solution ratios (2 e (n /n )
e 1000). Higher fractions
I
II solution
(e.g., one-component system thiol II ((n /n )
= 0)) as well
I
II solution
-6
5
as lower fractions of thiol II ((n /n ) > 1000) display a
II solution
I
The ΔD trace resulting from SA application to the SAM formed
reduced amount of immobilized mass.
from a solution ratio of (n /n )
= 10000 (Figure 2B) shows
For ethanol, SAMs were formed from incubation solutions of
different representative component ratios (Figure 3). Comparison
of SA layer mass quantities resulting on SAMs formed in ethanol
to SAMs prepared in chloroform shows characteristic differences.
I
II solution
-6
an initial increase to ΔD ≈ 0.2 ꢀ 10 (≈ first 2 min), which
5
is followed by the already mentioned drift. No intermediate
decrease is observed.
If SA was applied to a SAM formed from a one-component
incubation solution exclusively containing thiol I, Δf and ΔD
(39) Hook, F.; Ray, A.; Nordn, B.; Kasemo, B. Langmuir 2001, 17, 8305.
Langmuir 2010, 26(9), 6386–6393
DOI: 10.1021/la904087s 6389

Article
Seifert et al.
a
Table 2. Elemental Percentages of SAM I and SAM II as Revealed by Wide-Scan Survey XPS Analysis
peak position SAM I SAM II
Calculated (%)
Measured (%)
Calculated (%)
Measured (%)
C (C1s, BE = 285.0 eV)
N (N1s, BE = 400 ( 1.0 eV)
S (S2p, BE = 162.5 ( 1.0 eV)
88.89
-
5.56
5.56
85.4 ( 0.2
-
1.9 ( 0.3
12.7 ( 0.5
74.42
9.30
4.65
66.8 ( 1.1
11.9 ( 0.1
3.4 ( 0.3
O (O1s, BE = 532.3 ( 1.0 eV)
11.63
17.9 ( 1.3
a
The theoretical values of the elemental percentages (chemical structure) and the respective electron binding energies (BEs) are also depicted.
While a molar solution ratio of (n /n )
=10 in ethanol led
I
II solution
-2
to a SA layer mass of mSauerbrey SA ≈ 400 ng cm (similar to
3
chloroform), a solution ratio of (n /n )
=100 caused an
= 1000 in
II solution
I
II solution
already decreased binding capacity. For (n /n )
I
ethanol, significantly less material was immobilized (m
=
Sauerbrey SA
-2
(
151 ( 15) ng cm ). Regarding the lowest amount of thiol II
3
in the incubation solution ((n /n )
= 10 000), SA addition
I
II solution
resulted in only marginal mass deposition (m
=
Sauerbrey SA
-2
(
(
27 ( 7) ng cm ).
3
XPS Study of SAMs. SAMs formed from thiol solutions
cThiol = 0.1 mM, 24 h) in chloroform and ethanol were
investigated by XPS. Wide-scan spectra (electron binding energy
(
BE) range 0-1200 eV) and high-resolution scans for the core-
level spectra of carbon (C1s) were collected for the one-
component SAMs (SAM I and SAM II) as well as for SAMs
formed from binary thiol mixtures with varying thiol solution
ratios (n /nII)solution.
I
Chloroform, One-Component SAMs (SAM I and SAM
II). Table 2 summarizes the elemental percentages of SAM I and
SAM II formed in chloroform obtained by wide scan survey
analysis (peak intensity analysis of C1s, N1s, O1s, and S2p
signals; spectra not shown). Additionally, the calculated mole-
cular percentages (chemical structure) are given. For all elements
of the one-component SAMs, only moderate agreement of
calculated and measured data was obtained. In particular, the
measured fractions of O deviate to larger values for both SAMs.
The reverse was observed for the percentage of measured S.
Any chlorine peak intensities originating from residual solvent
were absent. In all cases, signals from the gold substrate (Au4f,
Au4d, Au4p) could be identified (data not shown).
Figure 4 shows the C1s core-level spectra taken from SAM I
and SAM II. For both thiols, a pronounced superposition
resulting from the varying chemical environments of the corres-
ponding carbon atoms is visible. For thiol I, the C1s spectrum was
decomposed into three different carbon positions with varying
intensities (CPS - counts per second). The corresponding spec-
trum of thiol II was deconvoluted into six individual peaks. The
different signals used for the deconvolution were adapted from
NMR spectra of the pure components (data not shown); all
electron binding energies (BEs) of the different types of carbon
peak positions were derived from the literature. The positions
C-O and C-N were unified since rather similar BEs must be
expected (for detailed information about the deconvolution
procedure, see Supporting Information).
Table 3 lists the carbon BE positions used in the deconvolution
procedure as well as the percentages of the different C species
determined from the peak intensities of the C1s core-level spectra
of SAM I and SAM II (cf. Figure 4). The theoretical percentages
following from the chemical structure were added. Similar to the
wide scan analysis, exact agreement of calculated and measured
data could not be achieved. However, the percentages of the
Figure 4. C1s core-level spectra of SAM I (A) and SAM II (B).
The deconvolution into the different carbon peaks and the sum of
the single carbon peaks are also depicted. For peak positions and
the respective results of the peak intensity analysis, see Table 3.
4
0
different C species approximated by the signal deconvolution
resemble the calculated values.
Chloroform, Binary SAM Systems. SAMs formed by 24 h
incubations in chloroform with varying thiol solution ratios (1 e
(
n /n )
e 500) were analyzed in the same way as the one-
I
II solution
component SAMs. For all SAMs, the elemental percentages were
derived from the wide scan survey spectra. The C1s core-level
spectra were used to identify the percentages of the different C
species.
From these data, the thiol ratios in the SAMs, (n /n )
),
I
II SAM
were calculated in two different ways: Calculation method A: The
measured ratios of N and C percentages (N/C ratios), derived
from the wide scan spectra, were used. Since N is unique for thiol
(
40) Beamson, G.; Briggs, D. High Resolution XPS of Organic Polymers; Wiley:
Chichester (England), 1992.
6
390 DOI: 10.1021/la904087s
Langmuir 2010, 26(9), 6386–6393

Seifert et al.
Article
a
Table 3. Percentages of the Different Carbon Species of SAM I and SAM II as Revealed by C1s Core-Level Analysis
C1s peak position SAM I SAM II
calculated (%)
measured (%)
calculated (%)
measured (%)
C-C (BE = 285.0 eV)
C-C (β-carbonyl) (BE = 285.7 ( 0.2 eV)
C-S (BE = 286.3 ( 0.2 eV)
88.90
-
5.55
5.55
-
-
91.1 ( 0.2
50
50.4 ( 0.7
6.4 ( 0.2
7.1 ( 0.3
25.4 ( 0.4
7.3 ( 0.5
3.4 ( 0.2
-
6.25
9.38
25
6.25
3.13
4.0 ( 0.2
4.9 ( 0.1
C-N/C-O (BE = 286.7 ( 0.2 eV)
N-CdO (BE = 288.3 ( 0.2 eV)
-
-
N) -CdO (BE = 289.2 ( 0.2 eV)
2
a
The theoretical values (elemental formula) and the respective BEs are also depicted.
II, the measured N percentage in binary systems reflects the
fraction of thiol II in the SAM. The deviations between measured
and calculated elemental percentages identified for the one-
component SAMs were considered in the binary systems (for a
detailed description, see Supporting Information).
Calculation method B: The measured ratios of the aliphatic
(
C-C, C-H) carbon percentages (BE = 285.0 eV) and the sum of
all residual (nonaliphatic) carbon percentages (Caliphatic/Cnonaliphatic
ratios) were identified from the C1s core-level spectra. Thereby,
the percentages of the individual C species were approximated
by a deconvolution of the corresponding C1s signals (cf.
Figure 4, Table 3). Since the percentage of non-aliphatic carbon
peak intensities is significantly higher in thiol II compared to
thiol I, the corresponding percentage in binary systems reflects
the fraction of thiol II in the SAM (a graphic evolution of the
C1s signal intensity with BE > 285.0 eV due to the enrichment of
thiol II in the system can be found in the Supporting In-
formation). The Caliphatic/Cnonaliphatic ratios were used to identify
Figure 5. Thiol SAM ratios (n
corresponding ratios in incubation solution ((n
ratios (n SAM were derived from the XPS wide-scan survey
spectra(N/C ratio analysis, “CalculationmethodA”) and from the
C1s core-level spectra (Caliphatic/Cnonaliphatic ratio analysis, “Calcu-
I
/nII
)
SAM as a function of the
I
/nII)solution). The
I
/nII)
the SAM ratios (n /nII)SAM. Similar to “Calculation method A”,
I
lation method B”). Incubations were done in chloroform (cThiol
.1 mM, 24 h).
=
the SAM ratios were determined on the basis of the data obtained
for the one-component systems (SAM I and SAM II, cf. Table 3;
for a detailed description, see Supporting Information).
0
Figure5 shows the SAM ratios (n /n )
of all analyzedSAM
I
II SAM
substrates as
a
function of the corresponding solution
ratio (n /n )
in chloroform. The parameter (n /n )
II SAM
I
II solution
I
was identified using “Calculation method A” (N/C ratio) and
Calculationmethod B”(Caliphatic/Cnonaliphatic ratio), respectively.
In all cases, large differences between (n /n ) and (n /n )
“
I
II SAM
I
II solution
are demonstrated. The fractions of biotinylated thiol II in the
SAMs are distinctly increased if compared to the corresponding
fractions within the solution. For example, a solution ratio of
(
n /n )
= 100 leads to a 37 fold lower surface ratio of
I
II solution
(
n /n )
= 2.7 ( 0.2.
I
II SAM
For solution ratios of (n /n )
e 100, an exact agreement
I
II solution
between both calculation methods can be observed. For
n /n ) > 100 the consistency decreases. The reason is that
(
I
II solution
Figure 6. Comparison ofSAM ratios(n
from different solution ratios (n
ethanol were used as thiol solvent (cThiol = 0.1 mM, 24 h). The
(n /n ) were calculated via N/C ratio analysis (Calculation method
I
/nII
)
SAM ofSAMs formed
only marginal N1s signal intensities were detected above the
inelastic scattering background, meaning that the XPS detection
limit was reached. Accordingly, the C1s signal deconvolution
must be considered to be increased erroneously. Nevertheless,
both calculation methods agree quantitatively for the major part
of the analyzed thiol systems.
I
/nII)solution. Either chloroform or
I
II SAM
A). For solution ratios of (n
1
I
/nII
)
solution = 50 and (n
I
/nII
)
solution
=
00 in ethanol, the percentages of nitrogen were below the
detection limit of XPS (N D ꢁ Not Detectable).
Ethanol. Different SAMs formed from ethanolic incubation
solutions (cThiol = 0.1 mM, 24 h) were investigated via wide-scan
survey ((BE) range 0-1200 eV) analysis (data not shown). Similar
to the measurements in chloroform, deviations of elemental
fractions in the one-component SAMs were analyzed and com-
pared to the corresponding calculated values (data not shown).
Again, the N/C ratios measured for SAMs formed from binary
Figure 6 shows the calculated SAM ratios (n /n )
for
I
II SAM
different solution ratios (n /nII)solution. The corresponding results
I
obtained for incubations in chloroform are also depicted (cf.
Figure 5). Significant differences are visible: In ethanol, SAMs
with distinctly smaller fractions of thiol II were obtained. For
example, SAMs formed with solution ratios of (n /n )
= 20
I
II solution
in ethanol yield a surface ratio of (n /n )
corresponding incubations in chloroform led to (n /n )
= 7.5 ( 0.2, while
I
II SAM
solutions (specified by (n /nII)solution) were used to calculate the
=
I
I
II SAM
SAM ratios (n /n )
tion method A”).
(see Supporting Information, “Calcula-
1.5 ( 0.1. Ethanolic incubations with solution ratios of
I
II SAM
(n /n )
g 50 led to SAMs with fractions of thiol II below
I
II solution
Langmuir 2010, 26(9), 6386–6393
DOI: 10.1021/la904087s 6391

Article
Seifert et al.
the detection limit of XPS. No N1s signal could be detected in the
thiol I, and contribute to the SAM thickness. For example, for
one-component SAMs of (N-(8-biotinyl-3,6-dioxa-octanamidyl)-
functionalized thiols a contribution of approximately 1.5-
wide-scan spectra. Thus, the SAM ratio (n /n )
could not be
I
II SAM
determined.
2
4,47,48
2
.0 nm was reported.
After SA deposition, a fraction of
Discussion
biotin groups is buried in the SA binding pockets (insertion depth
4
3
QCM-D Analysis of SA Binding to SAMs. In the present
study, SAbinding characteristics to biotinylatedSAMs composed
of thiols I and II (Figure 1) were analyzed by QCM-D. The Δf and
ΔD dependencies obtained for SA binding on a SAM formed
≈ 1.4 nm), while residual biotin moieties are located in possible
SA layer vacancies. Thus, the biotin groups have to be regarded as
a part of the SA surface film, not of the underneath SAM.
Furthermore, the calculated film thickness implies a homo-
geneous SA layer, where surface defects and porosity of the real
system are neglected. Other uncertainties originate from the
estimated water content of the SA layers. Considering these
criteria, the approximated SA layer thickness of ∼3.6 nm appears
reasonable with respect to the molecular size of SA.
from a solution ratio of (n /n )
= 50 show a region of
I
II solution
simultaneous decrease in resonance frequency and dissipation
factor, e.g., from ≈ 2 min onward in Figure 2A. Such a temporal
decrease of both Δf and ΔD has also been demonstrated in recent
QCM-D studies, where SA layers were arranged on biotinylated
9
,39
22
lipid bilayers or biotinylated SAMs. In our case, the absolute
ΔD changes detected are of low magnitude and effected by a
typical ΔD drift vs time (cf. Figure 2). However, the temporal
decrease can be reliably identified from the ΔD trace.
In the case of SA layer formation on SAMs formed from a
solution ratio of (n /nII)solution = 10000 (Figure 2B), two impor-
I
tant differences can be observed. First, mass (thickness) calcula-
tions of deposited SA show less immobilized material on the
transducer surface compared to the complete 2-D crystalline
coverage (cf. Table 1). Again, the thickness is calculated using
Surface films causing low dissipation exhibit a high rigidity. A
decrease in ΔD must be attributed to a film transformation, i.e., a
9
,22,39
-3
stiffening process of the layer to a more rigid state.
In the
an effective density of Flayer = 1150 kg m . Note that this value
case of SA surface films, the stiffening is ascribed to the formation
of lateral SA-SA interactions, mediated by specific contact
regions between the individual protein molecules. SA is able to
form crystal-like structures of monomolecular thickness at bioti-
nylated interfaces (2-D crystallization). Within intact 2-D
crystal regions, the SA molecules cover approximately 50-60%
of the surface, thus forming structures with rather high water
must be classified as an approximated value. The intrinsic fraction
of trapped water (i.e., the protein/water mass ratio) within SA
films showing only partial coverage might be different from that
1
2,37
of a complete, crystalline SA layer,
leading to variations in the
effective density. However, the overall trend of a reduced SA
binding capacity to the SAMs can be concluded.
Second, the Δf and ΔD traces do not show the indication for
2-D crystallization. The film stiffening process is less pronounced,
emphasizing the observation originating from the mass determi-
nation: SA layers, which do not show a maximal protein surface
density, naturally have fewer SA-SA interactions within the
surface film and exhibit less compact structures.
It should be mentioned that one-component SAM I inhibits SA
binding. The SA binding detected in the binary systems can be
ascribed to the specific SA-biotin interaction.
4
1-45
content.
Former QCM-D studies quantified the mass of
solid supported SA crystalline arrangements (including hydro-
dynamically coupled water) via Sauerbrey method to approxi-
mately 450-550 ng cm-
.
2 9,22,39
The mass of SA layers on SAMs formed from a molar solution
-
2
ratio of (n /n )
= 50 in chloroform (mlayer ≈ 415 ng cm
34
,
I
II solution
Table 1) is in good agreement with the mentioned QCM-D
9
,22,39
studies.
Application of the Sauerbrey relation and the
3
8
Voight representation gave consistent results, showing that the
Sauerbrey approximation for rigid layers is sufficient for mass
calculation. Considering the maximal SA surface capacity in a
The data shown in Figure 3 confirm the dependence between
the SA layer formation quantity and the solution ratio
(n /nII)solution. High fractions of biotinylated thiol II in the
I
4
3,44
monolayer arrangement (2-D crystalline ≈ 50-60%)
and the
incubation solution lead to a reduced SA binding capacity due
2
,15,20,24,26
molecular weight of SA (M ≈ 55 kDa), a molecule surface
to sterical hindrance between adjacent biotin groups.
SA
1
2
-2
density of ≈2.5 ꢀ 10 molecules cm can be determined for a
Regarding the SA binding behavior on SAMs formed from
solutions with low fractions of thiol II, a more detailed inter-
pretation has to be given. For a thiol SAM on gold exhibiting a
3
2
complete SA layer. From this, a projection area of ≈25 nm
4
3
results for a single SA molecule, in line with reported values.
√
√
-3
14
Using an effective density of Flayer = 1150 kg m (approxi-
(
3ꢀ 3)R30° periodicity, a molecule density of ≈4.5 ꢀ 10
-3
-2
7
mated value, based on a 1/1 mass ratio of SA (F ≈ 1300 kg m
)
molecules cm can be calculated. Above a molecule surface
3
-3
39
12
-2
and water (F ≈ 1000 kg m )) results in a layer thickness of
≈ 3.6 ( 0.14 nm. This agrees with former studies based on,
density of ≈2.5 ꢀ 10 molecules cm was calculated for a
3
h
SA
complete SA layer. Assuming an ideal SA packing on the SAM
where each biotin binds a SA molecule, a SAM ratio of
4
6,47
e.g., ellipsometry.
Considering the dimensions of a SA mole-
3
43
cule (≈5.4 ꢀ 5.8 ꢀ 4.8 nm ), the calculated thickness seems to
be underestimated. However, it is important to note that the
calculated layer thickness is related to the SAM. Before SA
immobilization, the biotin moieties and ethylene glycol spacers
of thiol II protrude from the hydroxyl-terminated background of
n
diluent thiol/nbiotinylated thiol ≈ 180 would be sufficient for the
assembly of a complete SA 2-D crystal. The optimal ratio is
decreased by various effects which influencethe described optimal
interaction efficiency:
7
,49
(1) SA exhibits two binding sites for
biotin on either side of the molecule. Enhanced fractions of biotin
thiol in the monolayer lead to increased probability of SA binding
via both binding pockets. (2) A nonrandom or heterogeneous
thiol distribution induces less biotin accessibility for the binding
pockets. (3) Microscopic and nanoscopic gold surface irregula-
rities (e.g., grain structures, etch pits, or pinholes) lead to
mismatched thiol assemblies in the SAM.
(
41) Scheuring, S.; Muller, D. J.; Ringler, P.; Heymann, J. B.; Engel, A.
J. Microsc. (Oxford) 1999, 193, 28.
42) Schief, W. R.; Edwards, T.; Frey, W.; Koppenol, S.; Stayton, P. S.; Vogel,
V. Biomol. Eng. 1999, 16, 29.
43) Hendrickson, W. A.; Pahler, A.; Smith, J. L.; Satow, Y.; Merritt, E. A.;
Phizackerley, R. P. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 2190.
44) Darst, S. A.; Ahlers, M.; Meller, P. H.; Kubalek, E. W.; Blankenburg, R.;
Ribi, H. O.; Ringsdorf, H.; Kornberg, R. D. Biophys. J. 1991, 59, 387.
45) Coussaert, T.; Volkel, A. R.; Noolandi, J.; Gast, A. P. Biophys. J. 2001, 80,
004.
(
(
(
(
2
(48) Schmidt, A.; Spinke, J.; Bayerl, T.; Sackmann, E.; Knoll, W. Biophys. J.
1992, 63, 1385.
(49) Yang, G. H.; Liu, G. Y. J. Phys. Chem. B 2003, 107, 8746.
(
(
46) Reiter, R.; Motschmann, H.; Knoll, W. Langmuir 1993, 9, 2430.
47) Kim, D. J.; Koo, K. K. J. Ind. Eng. Chem. 2004, 10, 920.
6
392 DOI: 10.1021/la904087s
Langmuir 2010, 26(9), 6386–6393

Seifert et al.
Article
5
1
Taking these arguments into account, the QCM-D data of SA
binding events to SAMs formed from incubations in ethanol and
chloroform can be considered (Figure 3). In the case of chloro-
form, the complete SA binding capacity is preserved for
incubation solutions exhibiting remarkably low fractions of thiol
II. In fact, the thiol solution ratio being sufficient for maximal SA
earlier, all SAMs contained residual ethanol and thus C1s signal
intensities not belonging to the thiol molecules. The trapped
solvent leads to C1s signal fine structures, which could not be
reasonably decomposed in separated C1s contributions.
The XPS results emphasize the interpretation originating from
the QCM-D data. Incubations in chloroform lead to an elevated
adsorption efficiency of thiol II leading to enhanced fractions of
biotinylated thiol II on the gold surfaces compared to the
corresponding portions in solution. This effect ensures a complete
SA coverage of the thiol monolayers even if only marginal
amounts of thiol II are present during the 24 h incubations for
SAM formation. In the case of ethanolic incubations, the SAMs
contained distinctly fewer fractions of thiol II. The XPS results
demonstrate in line with the QCM-D study that the thiol system
used here is subjected to a strong solvent effect. For ethanolic
SAM formations, moderately enhanced adsorption efficiencies of
biotinylated thiols in similar thiol systems are described in the
immobilization quantity ((n /n )
= 1000) ranges two
I
II solution
decades below the optimal SAM ratio stated in the literature for
1
5,20
similar thiol systems ((ndiluent thiol/n
)
≈ 10).
biotinylated thiol SAM
In addition, a thiol solvent effect is revealed. SAMs formed of
binary thiol solutions in chloroform demonstrate higher SA binding
capacities compared to SAMs prepared from the same solution ratio
(
n /n )
in ethanol. Former studies on binary SAMs based on,
I
II solution
e.g., XPS or impedance spectroscopy illustrated varying adsorption
affinities for different thiols on gold depending on structural
characteristics (e.g., alkyl chain length, headgroup character) and
17,24,27
nature of incubation solvent.
Considering our QCM-D data,
2
0,24
an elevated adsorption efficiency of thiol II compared to thiol I was
observed, and the effect is more pronounced in chloroform.
XPS Study of SAMs. The focus of the XPS study is the
correlation between the used thiol solution ratios for SAM
literature:
In this study, we show that the adsorption effi-
ciency of thiol II is strongly enhanced in chloroform.
Deviations between thiol ratiosin solution (n /n )
gold surfaces (n /n )
and on
I
II solution
after long incubations (g1 day) can be
I
II SAM
formation ((n /n )
)
and the resulting SAM ratios
after 24 h incubation.
II SAM
explained in terms of thermodynamic aspects of the adsorption
I
II solution
1
6-18,27
(
n /n )
process.
For ethanol, the moderately enhanced adsorption
I
Concerning the wide-scan analysis of the one-component
efficiency of thiol II can be attributed to a lower solubility of the
biotinylated component compared to thiol I, i.e., a thermody-
SAMs, deviations of calculated and measured data must be
expected due to the oriented thiol assembly in the monolayer.
The influence of inelastic photoelectron scattering (inelastic
2
0
namic control. The significantly increased effect in chloroform
may be a result from a higher solubility difference of thiol I and
thiol II in chloroform. These conditions would elevate the
(thermodynamic) absorption affinity of thiol II to the gold and
explain the strong solvent effect observed in our study.
5
0
electron mean free paths) and the mean element specific distance
2
4
from the SAM surface lead to the attenuation of S2p signals. In
particular, this effect was observed for SAM I, which solely
contains covalently linked sulfur at the Au interface. The elevated
percentage of oxygen is due to physisorbed water on top of the
SAM, which is often reported to remain on hydrophilic SAM
Conclusion
In our study, SA layers were prepared on binary SAMs
consisting of 16-mercapto-1-hexadecanol (thiol I) and N-(8-
biotinyl-3,6-dioxa-octanamidyl)-16-mercaptohexadecanamide
5
1-53
surfaces.
For this reason, the calculation of (n /n )
of
I
II SAM
SAM formed from binary solution was done by using the carbon
and nitrogen signal intensities which are less influenced by the
described effects.
(
thiol II). Our findings demonstrate differing protein binding
behavior depending on the fraction of biotin moieties within the
preformed SAM. The SA layer surface density correlates with the
fraction of thiol II in the system. For appropriate amounts of
biotin, the QCM-D data indicate the formation of complete SA
monolayers. SA film mass quantities could be calculated via the
The gold signals detected in all wide-scan spectra confirm the
fact known in literature that the organic thiol layer exhibits
2
6,54
surface defects.
No signal intensities originating from chlorine could be de-
tected in the spectra of SAMs prepared from chloroform, mean-
ing that no residual solvent was existent. This fact enabled the
rathercomplex deconvolution of the C1s core spectra presented in
Figure 4. The total C1s signal intensity can be exclusively assigned
to the thiol molecules, since no contribution from chloroform
must be expected. We do not claim our signal decomposition to be
the only one possible. However, in the present case and under the
constraints of our procedure (see Supporting Information) the
decomposed signal intensities resemble the calculated expecta-
tions (Table 3). Furthermore, application of “Calculation method
3
4
Sauerbrey method. During protein immobilization, a film
stiffening, originating from a (partially) 2-D crystalline SA
arrangement, was observed. Excessive or insufficient fractions
of thiol II lead to SA layers with noncomplete coverage. These
layers feature fewer intermolecular contact regions and fewer
indications of film stiffening. The fraction of thiol II in the SAM
showed a dependence on two major aspects: (1) the portion of
thiol II in the incubation solution used for SAM formation,
and (2) the distinct influence of the choice of solvent. The
adsorption efficiency of thiol II is elevated, and the effect is
significantly more pronounced in chloroform compared to
ethanol. Considering these aspects, the SAM surface composition
can be controlled.
B” for the identification of (n /n )
in binary SAM systems led
I
II SAM
to similar results as “Calculation method A” (Figure 5), giving
evidence that the chosen deconvolution is applicable at least in the
present case.
For ethanolic incubations, a thiol surface ratio calculation
based on core-level spectra analysis was omitted. As reported
Acknowledgment. Prof. Dr. H. Eckert is gratefully thanked
for providing the XPS device being essential for the presented
study. We further thank The“Fond der Chemischen Industrie”,
the “NRW Graduate School of Chemistry” and the “Deutsche
Forschungsgemeinschaft” (DFG) for financial support.
(
50) Powell, C. J.; Jablonski, A. J. Phys. Chem. Ref. Data 1999, 28, 19.
(
51) Sprik, M.; Delamarche, E.; Michel, B.; Rothlisberger, U.; Klein, M. L.;
Wolf, H.; Ringsdorf, H. Langmuir 1994, 10, 4116.
(
52) Wirde, M.; Gelius, U.; Nyholm, L. Langmuir 1999, 15, 6370.
(
53) Tielens, F.; Costa, D.; Humblot, V.; Pradier, C. M. J. Phys. Chem. C 2008,
Supporting Information Available: Additional informa-
tion as described in the text. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
12, 182.
(
54) Voicu, R.; Boukherroub, R.; Bartzoka, V.; Ward, T.; Wojtyk, J. T. C.;
Wayner, D. D. M. Langmuir 2004, 20, 11713.
Langmuir 2010, 26(9), 6386–6393
DOI: 10.1021/la904087s 6393