Factors that determine the protein resistance of oligoether self-assembled monolayers - Internal hydrophilicity, terminal hydrophilicity, and lateral packing density

Article abstract of DOI:10.1021/ja034820y

Protein resistance of oligoether self-assembled monolayers (SAMs) on gold and silver surfaces has been investigated systematically to elucidate structural factors that determine whether a SAM will be able to resist protein adsorption. Oligo(ethylene glycol) (OEG)-, oligo(propylene glycol)-, and oligo(trimethylene glycol)-terminated alkanethiols with different chain lengths and alkyl termination were synthesized as monolayer constituents. The packing density and chemical composition of the SAMs were examined by XPS spectroscopy; the terminal hydrophilicity was characterized by contact angle measurements. IRRAS spectroscopy gave information about the chain conformation of specific monolayers; the amount of adsorbed protein as compared to alkanethiol monolayers was determined by ellipsometry. We found several factors that in combination or by themselves suppress the protein resistance of oligoether monolayers. Monolayers with a hydrophobic interior, such as those containing oligo(propylene glycol), show no protein resistance. The lateral compression of oligo(ethylene glycol) monolayers on silver generates more highly ordered monolayers and may cause decreased protein resistance, but does not necessarily lead to an all-trans chain conformation of the OEG moieties. Water contact angles higher than 70° on gold or 65° on silver reduce full protein resistance. We conclude that both internal and terminal hydrophilicity favor the protein resistance of an oligoether monolayer. It is suggested that the penetration of water molecules in the interior of the SAM is a necessary prerequisite for protein resistance. We discuss and summarize the various factors which are critical for the functionality of "inert" organic films.

Full text of DOI:10.1021/ja034820y

Published on Web 07/15/2003
Factors that Determine the Protein Resistance of Oligoether
Self-Assembled Monolayers - Internal Hydrophilicity,
Terminal Hydrophilicity, and Lateral Packing Density
Sascha Herrwerth, Wolfgang Eck,* Sven Reinhardt, and Michael Grunze*
Contribution from the Angewandte Physikalische Chemie, UniVersita¨t Heidelberg,
INF 253, D 69120 Heidelberg, Germany
Received February 23, 2003; E-mail: wolfgang.eck@urz.uni-heidelberg.de; michael.grunze@urz.uni-heidelberg.de
Abstract: Protein resistance of oligoether self-assembled monolayers (SAMs) on gold and silver surfaces
has been investigated systematically to elucidate structural factors that determine whether a SAM will be
able to resist protein adsorption. Oligo(ethylene glycol) (OEG)-, oligo(propylene glycol)-, and oligo-
(trimethylene glycol)-terminated alkanethiols with different chain lengths and alkyl termination were
synthesized as monolayer constituents. The packing density and chemical composition of the SAMs were
examined by XPS spectroscopy; the terminal hydrophilicity was characterized by contact angle measure-
ments. IRRAS spectroscopy gave information about the chain conformation of specific monolayers; the
amount of adsorbed protein as compared to alkanethiol monolayers was determined by ellipsometry. We
found several factors that in combination or by themselves suppress the protein resistance of oligoether
monolayers. Monolayers with a hydrophobic interior, such as those containing oligo(propylene glycol), show
no protein resistance. The lateral compression of oligo(ethylene glycol) monolayers on silver generates
more highly ordered monolayers and may cause decreased protein resistance, but does not necessarily
lead to an all-trans chain conformation of the OEG moieties. Water contact angles higher than 70° on gold
or 65° on silver reduce full protein resistance. We conclude that both internal and terminal hydrophilicity
favor the protein resistance of an oligoether monolayer. It is suggested that the penetration of water
molecules in the interior of the SAM is a necessary prerequisite for protein resistance. We discuss and
summarize the various factors which are critical for the functionality of “inert” organic films.
and cells in buffered aqueous solution.6 The protein resistance
of high molecular weight PEG is well explained by “steric
Introduction
The development of surfaces that inhibit nonspecific protein
adsorption has been an important subject for many medical and
biotechnological applications.1 Protein repelling surfaces are
used, for example, as substrates for cell culture,2 and as coatings
for contact lenses3 or catheters.4 The design and preparation of
surface coatings that suppress nonspecific protein adsorption
with high reliability and reproducibility is of high technological
interest, for example, to ensure specific recognition in biosensor
applications.5 To optimize the protein compatibility of surfaces,
or to develop new and better surface coatings suitable for this
purpose, a fundamental understanding of the underlying mech-
anisms of protein repulsion is required.
repulsion”,7 which is an entropic effect caused by the unfavor-
able change in free energy associated with the dehydration and
confinement of polymer chains with high conformational
freedom. To elucidate the underlying physical mechanisms of
the protein resistance of PEG in more detail, a model system
based on self-assembled monolayers of oligo(ethylene glycol)
(OEG) alkanethiols on gold was introduced by Whitesides
and co-workers as a molecularly defined model suitable for
precise physical measurements.8 As compared to polymeric,
high molecular weight PEG, these monolayers offer the
advantage of monodispersity, that is, a defined length of all
chains and a defined interface to bulk water. Although the
conformational freedom of the OEG end groups is restricted in
these densely packed films as compared to polymeric PEG, they
also show protein repellent properties. It was therefore suggested
Since the early 1980s, poly(ethylene glycol) (PEG) has been
used as a surface coating to prevent the adsorption of proteins
(1) (a) Proteins at Interfaces II, Fundamentals and Applications; Horbett, T.
A., Brash, J. L., Eds.; ACS Symposium Series; American Chemical
Society: Washington, DC, 1995; Vol. 602. (b) Chuang, H. Y. K. In Blood
Compatibility; Williams, D. F., Ed.; CRC Press: Boca Raton, FL, 1987;
Vol. 1, pp 87-102.
(6) (a) Poly(ethylene Glycol) Chemistry: Biotechnical and Biomedical Ap-
plications; Harris, J. M., Ed.; Plenum Press: New York, 1992. (b) Lee, J.
H.; Kopecek, J.; Andrade, J. D. J. Biomed. Mater. Res. 1989, 23, 351-
368. (c) Desai, N. P.; Hubbell, J. A. Biomaterials 1991, 12, 144-153.
(7) Jeon, S. I.; Lee, J. H.; Andrade, J. D.; De Gennes, P. G. J. Colloid Interface
Sci. 1991, 142, 149-158.
(8) (a) Prime, K. L.; Whitesides, G. M. Science 1991, 252, 1164-1167. (b)
Pale-Grosdemagne, C.; Simon, E. S.; Prime, K. L.; Whitesides, G. M. J.
Am. Chem. Soc. 1991, 113, 12-20.
(2) Chen, C. S.; Mrksich, M.; Huang, S.; Whitesides, G. M.; Ingber, D. E.
Science 1997, 276, 1425-1428.
(3) Johnston, E.; Ratner, B. D. In Immobilized Biomolecules in Analysis; Cass,
T., Ligler, F. S., Eds.; Oxford University Press: Oxford, U.K., 1998; pp
79-94.
(4) Stickler, D. J.; McLean, R. J. C. Cells Mater. 1995, 5, 167-182.
(5) Herrwerth, S.; Rosendahl, T.; Feng, C.; Fick, J.; Eck, W.; Himmelhaus,
M.; Dahint, R.; Grunze, M. Langmuir 2003, 19, 1880-1887.
9
10.1021/ja034820y CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 9359-9366
9359

A R T I C L E S
Herrwerth et al.
that other mechanisms than those in the polymeric films may
cause the inertness to protein adsorption in these monolayers.9,10
The efficiency of the protein resistance in these short and
monodisperse films was shown to increase with the length of
the OEG chains, with a minimum of two ethylene glycol (EG)
units necessary for negligible protein adsorption under standard
experimental conditions.11 In SAMs with three EG units and
methoxy termination, the protein resistant properties were found
al. reported electrostatic repulsive forces of up to several tens
of nanometers in range toward an EG3OMe functionalized20 and
a hydrophobic AFM probe in aqueous electrolyte solution.10
These repulsive forces are caused by an effective negative charge
on the EG3OMe and hydrophobic21,22 surfaces, and it has been
suggested10,22 that they might be a major factor that leads to
the repulsion of negatively charged proteins. The negative sur-
face charge on these nonionic SAMs arises from the adsorption
of hydroxide ions from solution, as shown in electrokinetic
measurements on EG3OMe and PEG monolayers on gold and
on glass22 and in a theoretical study based on density functional
calculations.23 The charge densities are comparable to those
estimated from AFM measurements.10 As is explained in ref
23, the negative surface charge on these nonionic monolayers
originates from the adsorption of hydroxide ions which are
present due to the autoionization of bulk water. Although
negative surface potentials can explain the repulsive interactions
between the AFM tip and the EG3OMe films on gold substrates,
we note that many other surfaces exhibit negative charges in
aqueous solutions21,22,24 which are not protein resistant. Hence,
there must be other factors involved in rendering a surface
“inert” in aqueous solutions which we will discuss in this paper.
Except for OEG, other functional tail groups with the ability
to resist nonspecific protein adsorption have been developed
for monodisperse SAMs. Prime et al. showed that a maltose-
terminated SAM suppressed the adsorption of proteins.8 Also,
SAMs with tripropylenesulfoxide25 and zwitterionic26 tailgroups
resisted the adsorption of proteins. In a general search for protein
repelling SAMs, Whitesides and co-workers27 found that
hydrogen bond acceptors appear to be necessary within the SAM
backbone to achieve protein repellent properties, whereas tail
groups with hydrogen bond donors exhibit no protein resistance.
An exception to this general rule was found in monolayers
terminated with mannitol groups prepared by Mrksich and co-
workers.28 The hypothesis that the surface wettability as mea-
sured by the contact angle is a singular criterion that determines
the protein resistance of a SAM has been ruled out in a recent
study.29 It was shown that oligo(ethylene glycol)-terminated
surfaces show far higher protein resistance than other SAMs
with different chemical termination but similar contact angles.
To broaden the understanding of the factors involved in the
protein resistance of oligoether SAMs, we investigated a series
of structural variations in the SAMs. We synthesized a series
of oligoether-terminated alkanethiols with different oligoether
backbones, different lengths, and different alkyl terminations
to depend on the conformation of the oligoether chains.12
A
helical conformation on gold was found to exhibit protein
resistance, whereas on silver a planar all-trans conformation
showed no protein repelling properties. The conformational
transition between the different surfaces is caused by a higher
packing density on silver that leads to lateral compression of
the oligoether units.13 A further study found that SAMs with a
helical or amorphous conformation of the OEG end groups in
general repel proteins.14 More recently, different chain confor-
mations depending on the length of the OEG units have been
found by Valiokas et al.15 for OEG-terminated alkanethiol
amides on gold, but a correlation to protein resistance has not
been established. In these systems, lateral compression of the
OEG chains is effected by hydrogen bonds between the amide
linker groups to the underlying alkanethiol which may lead to
an all-trans chain conformation. An OEG-terminated alkanethiol
amide with six EG units on gold exhibited a temperature-driven
phase transition between a helical phase at lower temperature
and phases with increasing all-trans content at higher temper-
atures.16 On protein resistant polydisperse monolayers with a
PEG molecular weight of 2000, a predominant helical confor-
mation has also been found.17
Earlier theoretical ab initio studies on the interaction of water
with OEG clusters18 suggested that the inertness of the helical
and amorphous conformers might be caused by a tightly bound
interfacial layer of water that prevents direct contact between
the proteins and the surface. However, Monte Carlo simulations
later revealed that the interface between water and methoxy-
terminated monolayers on gold with three EG units (EG3OMe
(5)) is characterized by a reduced water density as compared to
that of bulk water19 and that these surfaces are hydrophobic
and therefore should show hydrophobic attraction.9 The hydro-
phobic interaction is, however, apparently offset by electrostatic
repulsion as was shown by Ha¨hner et al.10,20 On these mono-
layers on gold with three EG units (EG3OMe (5)), Ha¨hner et
(9) Pertsin, A. J.; Hayashi, T.; Grunze, M. J. Phys. Chem. B 2002, 106, 12274-
12281.
(10) (a) Feldman, K.; Ha¨hner, G.; Spencer, N. D.; Harder, P.; Grunze, M. J.
Am. Chem. Soc. 1999, 121, 10134-10141. (b) Dicke, C.; Ha¨hner, G. J.
Phys. Chem. B 2002, 106, 4450-4456.
(21) Schweiss, R.; Welzel, P.; Werner, C.; Knoll, W. Langmuir 2001, 17, 4304-
4311.
(22) Chan, Y.-H. M.; Schweiss, R.; Werner, C.; Grunze, M. Langmuir 2003, in
press.
(23) Kreuzer, H. J.; Wang, R. L. C.; Grunze, M. J. Am. Chem. Soc. 2003, 125,
8384-8389.
(24) Zimmerman, R.; Dukhin, S.; Werner, C. J. Phys. Chem. B 2001, 105, 8544-
8549.
(11) Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1993, 115, 10714-
10721.
(12) Harder, P.; Grunze, M.; Dahint, R.; Whitesides, G. M.; Laibinis, P. E. J.
Phys. Chem. B 1998, 102, 426-436.
(13) Pertsin, A. J.; Grunze, M.; Garbuzova, I. A. J. Phys. Chem. B 1998, 102,
4918-4926.
(14) Harder, P.; Grunze, M.; Waite, J. H. J. Adhes. 2000, 73, 161-177.
(15) (a) Valiokas, R.; Svedhem, S.; O¨ stblom, M.; Svensson, S. C. T.; Liedberg,
B. J. Phys. Chem. B 2001, 105, 5459-5469. (b) Svedhem, S.; Hollander,
C.-A.; Shi, J.; Konradsson, P.; Liedberg, B.; Svensson, S. C. T. J. Org.
Chem. 2001, 66, 4494-4503.
(25) Deng, L.; Mrksich, M.; Whitesides, G. M. J. Am. Chem. Soc. 1996, 118,
5136-5137.
(26) (a) Kane, R. S.; Deschatelets, P.; Whitesides, G. M. Langmuir 2003, 19,
2388-2391. (b) Ostuni, E.; Chapman, R. G.; Liang, M. N.; Meluleni, G.;
Pier, G.; Ingber, D. E.; Whitesides, G. M. Langmuir 2001, 17, 6336-
6343. (c) Holmlin, R. E.; Chen, X.; Chapman, R. G.; Takayama, S.;
Whitesides, G. M. Langmuir 2001, 17, 2841-2850.
(27) (a) Ostuni, E.; Chapman, R. G.; Holmlin, R. E.; Takayama, S.; Whitesides,
G. M. Langmuir 2001, 17, 5605-5620. (b) Chapman, R. G.; Ostuni, E.;
Takayama, S.; Holmlin, R. E.; Yan, L.; Whitesides, G. M. J. Am. Chem.
Soc. 2000, 122, 8303-8304.
(16) Valiokas, R.; O¨ stblom, M.; Svedhem, S.; Svensson, S. C. T.; Liedberg, B.
J. Phys. Chem. B 2000, 104, 7565-7569.
(17) Tokumitsu, S.; Liebich, A.; Herrwerth, S.; Eck, W.; Himmelhaus, M.;
Grunze, M. Langmuir 2002, 18, 8862-8870.
(18) (a) Wang, R. L. C.; Kreuzer, H. J.; Grunze, M. J. Phys. Chem. B 1997,
101, 9767-9773. (b) Wang, R. L. C.; Kreuzer, H. J.; Grunze, M. Phys.
Chem. Chem. Phys. 2000, 2, 3613-3622.
(19) Pertsin, A. J.; Grunze, M. Langmuir 2000, 16, 8829-8841.
(20) Dicke, C.; Feldman, K.; Eck, W.; Herrwerth, S.; Ha¨hner, G. Polym. Prepr.
2000, 41(2), 1444-1445.
(28) Luk, Y.-Y.; Kato, M.; Mrksich, M. Langmuir 2000, 16, 9604-9608.
(29) Sigal, G. B.; Mrksich, M.; Whitesides, G. M. J. Am. Chem. Soc. 1998,
120, 3464-3473.
9
9360 J. AM. CHEM. SOC. VOL. 125, NO. 31, 2003

Oligoether Self-Assembled Monolayers
A R T I C L E S
Chart 1. Overview of the Synthesized Oligoether Alkanethiolsa
a The abbreviations used here and later in the text mean the following: EG3OEt is a tri(ethylene glycol) unit with an ethoxy end group, PRO3OMe is a
tri(propylene glycol) unit with a methoxy end group, and TRI3OH is a trimethylene glycol unit with OH termination. All molecules bear an undecylthiol
chain on one end for anchoring on the gold and silver surfaces.
(Chart 1). The oligoether units used comprise ethylene glycol,
propylene glycol, and trimethylene glycol. SAMs of these com-
pounds were prepared on gold and on silver surfaces and were
characterized by contact angle measurements, XPS, and IRRAS
spectroscopy. Protein adsorption on these SAMs was examined
by ellipsometric measurements using fibrinogen as a test
molecule.
hydrophobicity of the end group, the advancing contact angles
of all methoxy-terminated monolayers are 65-76°. The lowest
contact angles of the methoxy-terminated monolayers are found
for the longer EG3OMe (5) and EG6OMe (10) chains and for
TRI3OMe (17) on gold and silver. The shorter OEG chains
EG1OMe (1) and EG2OMe (3) and the oligo(propylene gly-
col)-terminated SAMs PRO2OMe (13), PRO3OMe (14), and
PRO4OMe (15) have slightly higher contact angles. The strong
increase of the contact angles for the ethoxy-, propoxy-, and
butoxy-terminated tri(ethylene glycol) and hexa(ethylene glycol)
end groups reflects the rising hydrophobicity of the termination.
Results
Contact Angle Measurements. The hydroxy-terminated
SAMs with oligoether end groups on gold and silver have the
lowest advancing contact angles of water (32-36°, Table 1).
The substrate (Au or Ag) has no specific influence on the
advancing or receding contact angle.30 Because of the higher
(30) To check for possible differences of surface reorganization in contact with
water, the receding contact angles of EG3OMe and EG6OMe SAMs have
also been measured on gold and silver. The hysteresis was constant within
experimental errror, being 10-14° in all cases.
9
J. AM. CHEM. SOC. VOL. 125, NO. 31, 2003 9361

A R T I C L E S
Herrwerth et al.
Table 1. Advancing Water Contact Angles of the SAMs on Gold
and Silver
advancing
contact angle
advancing
contact angle
Au
Ag
Au
Ag
EG1OMe (1)
EG2OH (2)
EG2OMe (3)
EG3OH (4)
EG3OMe (5)
EG3OEt (6)
EG3OPr (7)
EG3OBu (8)
EG6OH (9)
71°
33°
69°
34°
65°
84°
94°
107°
32°
71°
33°
71°
35°
65°
84°
95°
104°
36°
EG6OMe (10)
EG6OEt (11)
EG6OPr (12)
PRO2OMe (13)
PRO3OMe (14)
PRO4OMe (15)
TRI3OH (16)
TRI3OMe (17)
67°
87°
97°
73°
72°
71°
34°
67°
65°
85°
96°
76°
75°
73°
32°
65°
Table 2. Calculated and Experimental Ratios of Ether Carbon to
Oxygen XPS Intensities
Figure 1. Attenuation of the substrate photoelectron intensity for alkanethiol
SAMs with 12, 16, 18, and 20 carbon atoms and 100% surface coverage.
The logarithmic Ag 3d and Au 4f intensities decrease linearly with increasing
molecular chain length of the unsubstituted alkanethiols.
C1s ether/O1s ether
calcd. Au Ag
EG1OMe (1) 2/1 1.82/1 1.85/1 EG6OMe (10) 2/1
C1s ether/O1s ether
calcd. Au Ag
Table 3. Packing Density of the Oligoether-Terminated SAMs on
Gold and Silver, Expressed by the Average Number of Molecules
per nm2
2.1/1 2.1/1
2.07/1 1.95/1
2.12/1 2.05/1
2.03/1 1.88/1
1.84/1 1.85/1
2.05/1 1.91/1
EG2OH (2) 1.5/1 1.51/1 1.52/1 EG6OEt (11)
EG2OMe (3) 2/1 1.84/1 1.87/1 EG6OPr (12)
2/1
2/1
EG3OH (4) 1.75/1 1.67/1 1.71/1 PRO2OMe (13) 2/1
molecule density
[nm-2
molecule density
-2
EG3OMe (5) 2/1
EG3OEt (6) 2/1
EG3OPr (7) 2/1
EG3OBu (8) 2/1
1.87/1 1.97/1 PRO3OMe (14) 2/1
2.01/1 1.92/1 PRO4OMe (15) 2/1
]
[nm ]
Au
Ag
Au
Ag
2.09/1 2.11/1 TRI3OH (16)
1.75/1 1.81/1 1.65/1
1.89/1 1.83/1
1.99/1 2.05/1 TRI3OMe (17) 2/1
EG1OMe (1)
EG2OH (2)
EG2OMe (3)
EG3OH (4)
EG3OMe (5)
EG3OEt (6)
EG3OPr (7)
EG3OBu (8)
EG6OH (9)
3.93
3.93
3.83
3.79
3.61
3.69
3.79
3.83
3.46
4.93
4.66
4.66
4.19
4.19
4.14
4.19
4.09
3.62
EG6OMe (10)
EG6OEt (11)
EG6OPr (12)
PRO2OMe (13)
PRO3OMe (14)
PRO4OMe (15)
TRI3OH (16)
TRI3OMe (17)
3.32
3.32
3.32
3.13
2.99
2.71
3.92
3.60
3.83
3.72
3.62
3.72
3.56
3.30
4.19
4.40
EG6OH (9) 1.86/1 1.99/1 1.83/1
X-ray Photoelectron Spectroscopy. To ensure that the
organic layers of the oligoether-terminated alkanethiols on gold
and silver were free of contamination, we calculated the
expected ratio of the intensities of C1s ether carbon to O1s ether
oxygen signals and measured it by XP spectroscopy. Because
only the oligoether units of the monolayers were considered
for this purpose, the signals from both atomic species were
weighted equally. For the O1s emission, a single peak at about
532.4-533 eV (depending on the monolayer and the substrate)
was used to fit the data. For the C1s line, two distinct C1s peaks
at about 284.4-285.1 eV for the alkyl chain and at about 286.4-
287.0 eV for the ether carbon atoms were required. For all
samples, the C1s/O1s ratios were close to the calculated
stoichiometric ratio of ether carbon to oxygen atoms (Table 2).
We therefore conclude that the monolayers were free of
impurities and that the silver substrates had no significant
oxygen contamination within the sensitivity limits of XPS.
The surface coverage and lateral packing density of an
oligoether SAM can be determined by measurement of the
attenuation of the metal substrate photoelectrons (Au 4f and
Ag 3d).12 As a reference system, unsubstituted alkanethiol SAMs
(C12SH, C16SH, C18SH, and C20SH) were used to determine the
attenuation of the Au and Ag substrate signals as a function of
monolayer thickness (Figure 1).
carbon atoms and 4 oxygen atoms is identical to the attenuation
by an unsubstituted C17 alkanethiol film on gold. This implies
that an EG3OMe monolayer has a relative coverage of 77% on
gold as compared to that of an alkanethiol SAM on gold, or an
average lateral packing density of 3.61 molecules/nm2. On silver,
the attenuation of the Ag 3d substrate signal by an EG3OMe
(5) film corresponds to the attenuation of an alkanethiol film
of 17.6 carbon atoms on average. Thus, an EG3OMe monolayer
on silver has a coverage of 80% relative to an alkanethiol on
silver or an average lateral packing density of 4.19 molecules/
nm2. Details of this method to evaluate packing densities have
been described in ref 12.
Table 3 shows that the oligoether-terminated alkanethiol
monolayers generally have a higher packing density on silver
than on gold, which correlates with the behavior of unsubstituted
alkanethiols on these two substrates.
With increasing length of the OEG tail group, lower packing
densities of the OEG-terminated monolayers are obtained. The
monolayers of the methoxy-terminated mono(ethylene glycol)
undecanethiol EG1OMe (1) and the hydroxy- and methoxy-
terminated di(ethylene glycol) undecanethiols EG2OH (2) and
EG2OMe (3) have about 20% higher density on silver than on
gold. For the longer EG3- and EG6-derivatives, the influence
of the substrate becomes smaller, and the difference in packing
densities between gold and silver is only 5-13%. The higher
packing densities of the EG2OMe (3) and EG1OMe (1) mono-
layers, as compared to EG6OMe (10) and EG3OMe (5), are also
reflected in the slightly higher contact angles for EG2OMe (3)
and EG1OMe (1). Because of the shorter OEG chain, the
In an idealized, densely packed, and defect free alkanethiol
SAM with 100% coverage, the area occupied by a single chain
is 21.4 Å2 on Au(111) and 19.1 Å2 on Ag(111).12 This
corresponds to packing densities of 4.67 molecules/nm2 on gold
or 5.24 molecules/nm2 on silver. When the attenuation of the
Au 4f and Ag 3d XPS signals by a given oligoether monolayer
is compared to the attenuation by the alkanethiol reference films
in Figure 1, the lateral packing density of the oligoether SAM
can be calculated.
As an example, the experimental data show that the attenu-
ation of the Au 4f substrate by an EG3OMe (5) film with 18
9
9362 J. AM. CHEM. SOC. VOL. 125, NO. 31, 2003

Oligoether Self-Assembled Monolayers
A R T I C L E S
Figure 2. Ag 3d XP spectra of PRO3OMe (14) and TRI3OMe (17) on
silver. The Ag 3d intensity of the TRI3OMe sample is weaker due to the
higher surface coverage.
Figure 3. IRRAS spectra for EG6OMe (10) monolayers on gold and silver.
influence of the alkyl chain on the packing behavior is more
pronounced. The alkyl termination of the OEG end group has
a much weaker influence on the packing density in comparison
to the chain length of the OEG end group. The densities
measured here for EG6OH (9) and EG3OMe (5) are in good
agreement with the values determined by Harder et al.12 For
EG3OMe (5) on silver, we found identical results with respect
to chain conformation and protein resistance, but slightly lower
packing densities which might be caused by different roughness
or degree of surface oxidation of the silver substrates or the
use of dimethylformamide instead of ethanol as solvent for the
SAM preparation.
The tri(trimethylene glycol)-terminated alkanethiol mono-
layers TRI3OH (16) and TRI3OMe (17) have packing densities
approximately equal to those of the di(ethylene glycol)- and
tri(ethylene glycol)-terminated SAMs. The additional methylene
groups in the tri(trimethylene glycol) tail groups do not hinder
effective packing of the adsorbate molecules. Correspondingly,
the similar contact angles of the tri(trimethylene glycol)- and
the tri(ethylene glycol)-terminated monolayers indicate com-
parable packing densities and order of the monolayers.
On the other hand, the oligo(propylene glycol)-terminated
monolayers PRO2OMe (13), PRO3OMe (14), and PRO4OMe
(15) have lower packing densities as compared to the corre-
sponding oligo(ethylene glycol)- or trimethylene glycol-
terminated SAMs without methyl side groups. The experimental
data (Figure 2) show that the attenuation of the Ag 3d substrate
photoelectrons by a PRO3OMe (14) film is much weaker than
that of a TRI3OMe (17) film with an equal number of atoms.
Obviously, the more bulky nature of the oligo(propylene
glycol) groups hinders denser packing of the adsorbate mol-
ecules. The higher contact angles measured for the oligo-
(propylene glycol) SAMs also imply more disordered, less
densely packed films and point to the significant influence of
the methyl side chain.
Figure 4. Fingerprint region of EG6OMe (10) monolayers on gold and
silver.
1). Therefore, we characterized the monolayers of EG6OMe (10)
on gold and silver in more detail by IR spectroscopy to analyze
possible differences in their molecular order and conformation.
The IR spectra of EG6OMe (10) on gold and silver (Figure
3) reveal structural differences that strongly depend on the
substrate. The EG6OMe (10) spectrum on silver shows very
strong characteristic peaks at 965 cm-1 (CH2 rocking mode),
1117 cm-1 (COC stretching mode), 1243 cm-1 (CH2 twisting
mode), 1348 cm-1 (CH2 wagging mode), and 1463 cm-1 (CH2
scissoring mode) in the fingerprint region and at 2892 cm-1
(CH2 stretching mode) in the CH stretching region. On gold,
the spectral intensities of the peaks in the fingerprint region are
much weaker, and the COC stretching mode at 1117 cm-1 is
broader than on silver. The band at 2892 cm-1 does not
dominate in the CH stretching region of EG6OMe (10) on gold
as it does on silver. Rather, a broad signal is observed in the
CH region on gold.
The detailed spectrum of the fingerprint region (Figure 4) of
EG6OMe (10) on silver shows a characteristic, very strong
skeletal peak at 1117 cm-1 with a weak shoulder appearing on
the high-frequency side at 1146 cm-1. The intensities and
IRRAS Experiments with EG6OMe Monolayers on Gold
and Silver. As will be shown later, we unexpectedly observed
protein adsorption for EG6OMe (10) monolayers on silver, a
system that had not been examined previously. On the other
hand, EG6OMe (10) monolayers on gold are fully resistant to
protein adsorption, which is consistent with earlier results.11 The
EG6OMe (10) monolayers on gold and silver vary in their
relative surface coverage (see Table 3) but show nearly no
difference in their advancing water contact angles (see Table
features of the peaks at 1463, 1348, 1243, and 965 cm-1
,
respectively, as well as the strong and sharp skeletal peak at
1117 cm-1, indicate the presence of a highly oriented and
crystalline helical OEG phase which does, however, show
protein adsorption as described later. On gold, the COC
stretching band has a different shape and is shifted to higher
frequencies than that on silver. The detailed spectrum shows a
9
J. AM. CHEM. SOC. VOL. 125, NO. 31, 2003 9363

A R T I C L E S
Herrwerth et al.
Table 4. Amount of Adsorbed Fibrinogen on SAMs on Gold and
Silver, Measured by the Ellipsometric Thickness of the Protein
Layer and Normalized to the Amount of Fibrinogen Adsorbed on a
Monolayer of Hexadecanethiol (C16SH) on Gold ()100%)
protein
adsorption [%]
protein
adsorption [%]
Au
Ag
Au
Ag
EG1OMe (1)
EG2OH (2)
EG2OMe (3)
EG3OH (4)
EG3OMe (5)
EG3OEt (6)
EG3OPr (7)
EG3OBu (8)
EG6OH (9)
22
0
0
0
0
60
79
77
0
57
0
62
0
37
88
93
98
0
EG6OMe (10)
EG6OEt (11)
EG6OPr (12)
PRO2OMe (13)
PRO3OMe (14)
PRO4OMe (15)
TRI3OH (16)
TRI3OMe (17)
0
51
69
52
49
42
0
35
89
100
87
80
60
0
0
52
Figure 5. CH stretching region of EG6OMe (10) monolayers on gold and
silver.
broad peak around 1126 cm-1 that can be separated into at least
three signals at 1146, 1126, and 1117 cm-1. This and the
significant decrease of the intensity of the signals at 1463, 1348,
1243, and 965 cm-1 indicate that the OEG chains on gold are
less oriented and in a more amorphous conformation which was
found to be protein resistant. On silver, the presence of a highly
ordered and crystalline helical OEG phase with a higher packing
density and less defects than on gold leads to loss of protein
resistance of the monolayers, revealing that - contrary to
previous suggestions - the molecular conformation in the SAM
per se is not a sufficient indicator for protein resistance.
The detailed spectra of the CH stretching bands show quite
similar differences (Figure 5). The EG6OMe (10) monolayer
on silver shows a very strong CH stretching peak at 2892 cm-1
characteristic for the crystalline helical conformation of the OEG
phase.12,16 On gold, a different shape of the peak together with
a strong decrease in intensity is observed. Instead, a broad band
appears from 2840 to 2960 cm-1, which implies an amorphous
and less oriented conformation of the OEG groups on gold.
Protein Adsorption Experiments. Fibrinogen adsorption on
the different oligoether-terminated SAMs was quantified relative
to protein adsorption on reference monolayers of hexadecane
thiol on the respective substrates. An ex-situ ellipsometric
technique similar to the one developed by Whitesides and co-
workers8,11 was used to determine the relative thickness of
adsorbed protein layers that remain on the SAMs after rinsing
with water (see experimental section in the Supporting Informa-
tion). The detection limit of this technique is a few percents of
a monolayer, and it is less sensitive than in-situ ellipsometry or
surface plasmon resonance. However, it gives well reproducible
information on the relative amount of protein that is irreversibly
adsorbed on different surfaces and cannot be removed by rinsing.
The data shown in Table 4 and Figures 6 and 7 indicate that
all hydroxy-terminated SAMs with oligo(ethylene glycol) or tri-
(trimethylene glycol) tail groups show no protein adsorption
on gold and silver. However, the methoxy-terminated SAMs
with OEG tail groups (EG2OMe (3), EG3OMe (5), and
EG6OMe (10)) are only protein resistant on gold. On silver,
where higher packing densities are observed, the methoxy-
terminated SAMs are no longer protein resistant. The mono-
layers of EG1OMe (1) adsorb proteins both on gold and on
silver, which is fully consistent with the previous result that at
least two ethylene glycol groups are necessary for protein
resistance.11 The methoxy-terminated SAMs with tri(trimeth-
ylene glycol) tail groups (TRI3OMe (17)) show behavior similar
Figure 6. Amount of protein adsorption on a given oligoether SAM on
gold normalized to the amount of protein adsorbed on a monolayer of hexa-
decanethiol on gold (100%) versus advancing aqueous contact angle of the
SAM. Symbols: red 2, EG2OH; orange |, EG3OH; green +, EG6OH; blue
[, TRI3OH; blue 9, EG1OMe; green 1, EG2OMe; light blue b, EG3OMe;
red ×, EG6OMe; blue 0, TRI3OMe; red b, PRO2OMe; blue ], PRO3-
OMe; orange +, PRO4OMe; purple 9, EG3OEt; green O, EG6OEt; blue 3,
EG3OPr; green 4, EG6OPr; blue b, EG3OBu.
Figure 7. Amount of protein adsorption on a given oligoether SAM on
silver normalized to the amount of protein adsorbed on a monolayer of
hexadecanethiol on gold (100%) versus advancing aqueous contact angle
of the SAM. Symbols are identical to those in Figure 6.
to that of the EG3OMe and EG6OMe monolayers: on gold, they
are protein resistant, but on silver, they show even stronger
adsorption. In remarkable contrast, all SAMs with oligo(propy-
lene glycol) tail groups adsorb proteins on gold and silver. These
9
9364 J. AM. CHEM. SOC. VOL. 125, NO. 31, 2003

Oligoether Self-Assembled Monolayers
A R T I C L E S
experiments by Harder et al.12 have shown that, in the special
case of EG3OMe (5) SAMs, the protein resistance is coupled
to the molecular conformation of the OEG groups. This
conclusion can, however, not be generalized, because on silver
densely packed EG3OMe (5) monolayers with an all-trans
conformation and EG6OMe (10) SAMs with a highly ordered
crystalline helical structure both adsorb proteins.
The dependence of protein adsorption on lateral density is a
new and important aspect. It revises our previous model that
protein resistance is determined by molecular conformation;
rather, lateral density (which, however, determines molecular
conformation) seems to be the important variable.
Discussion
Our results show that several factors are required to make an
oligoether SAM protein resistant. First, an interior hydrophilic
chemical structure is necessary, because only the oligo(ethylene
glycol) and oligo(trimethylene glycol) SAMs exhibit protein
resistance, but no repelling properties are found on the
hydrophobic oligo(propylene glycol) SAMs. Quite similarly, the
lateral compression of a methoxy-terminated hexa(ethylene
glycol) SAM on silver leads to reduced protein resistance,
although a helical conformation is retained, and the SAM is
highly ordered. This is in excellent agreement with a recent
report by Vanderah et al.34 based on IRRAS data, who found
that a methoxy-terminated hexa(ethylene glycol) SAM on gold
without an alkyl spacer chain showed reduced protein resistance
when a highly ordered crystalline helical conformation was
present. Only when disorder and a larger number of defects were
introduced into the film by assembly from solvents other than
ethanol was full protein resistance obtained.
These results imply that either a relaxed lateral packing
density as found in SAMs on gold surfaces or some disorder or
defects of the monolayer are necessary for high protein
resistance. For the methoxy-terminated monolayers, packing
densities exceeding a well-defined threshold value (see Figure
8) lead to a loss of resistance; for example, EG3OMe-terminated
monolayers with a highly ordered all-trans conformation on
silver show protein adsorption.12 Similarly, SAMs of 11-
mercaptoundecan-1-ol on gold with a highly hydrophilic surface
and contact angles below 15° are not resistant to fibrinogen
adsorption,29 because no hydrophilic interior is present. We
conclude from our experiments that water must be able to
penetrate into the SAM to achieve resistance to protein
adsorption. That water penetration and lateral density are
correlated has been shown in grand canonical Monte Carlo
simulations (GCMC).9,19 The fact that the protein resistance
increases with higher chain length of the OEG units11 also
suggests that a higher water content in the SAM leads to better
protein repulsion. In a very recent study, Vanderah et al. have
shown by electrochemical impedance spectroscopy that oligo-
(ethylene glycol)-terminated SAMs on gold contain water when
they are disordered and protein resistant, but that water
penetration into a non-protein resistant and highly ordered oligo-
(ethylene glycol) SAM is slow.35
Figure 8. Amount of protein adsorption versus lateral packing densities
of the methoxy-terminated SAMs with oligo(ethylene glycol) tail groups
on gold and silver. Red squares are monolayers on gold, and blue triangles
are monolayers on silver.
data show that the character of the oligoether tail group influ-
ences the protein resistance of the SAMs. Only the SAMs with
a hydrophilic interior show protein resistance under the condi-
tions used in our study. It is well known that high molecular
weight poly(propylene glycol) is a water-insoluble polymer,
whereas both poly(ethylene glycol) and poly(trimethylene
glycol) are water-soluble.31 By analogy, both for oligo(ethylene
glcyol) alkyl ethers32 and for poly(trimethylene glycol),33 hy-
drates have been reported, but not for polypropylene glycol.
The SAMs with oligo(ethylene glycol) tail groups and the
longer ethoxy, propoxy, or butoxy termination show strong
protein adsorption due to their more hydrophobic surface
properties. This indicates that besides the inner hydrophilicity
also the terminal hydrophilicity or wettability is an important
factor for protein adsorption. On gold, the oligoether-terminated
monolayers do not show any significant protein adsorption at
advancing contact angles below 70° (Figure 6). For higher
contact angles, an increase of protein adsorption is observed.
On silver (Figure 7), protein adsorption can be detected for
slightly lower contact angles as compared to the films on gold.
For the methoxy-terminated monolayers on gold and on silver,
there is a high variability in protein adsorption that depends on
the length and type of the oligoether chain.
In Figure 8, we show that protein adsorption on the methoxy-
terminated SAMs with oligo(ethylene glycol) tail groups is a
function of the lateral packing density, with a transition from
nonadsorbing to adsorbing above packing densities of about 3.85
molecules/nm2. The packing density in turn depends on the
length of the oligo(ethylene glycol) chain. In contrast to the
methoxy-terminated OEG SAMs, the hydroxy-terminated OEG
SAMs show full protein resistance also for short chain lengths
and high packing densities on silver (Table 4; Figures 6, 7).
Because, in terms of surface wettability, the methoxy-terminated
SAMs are at the borderline of protein resistance, suitable
packing characteristics of their oligoether units are of higher
significance than for hydroxy-terminated monolayers.
That EG6OMe (10) monolayers on gold are resistant to protein
adsorption is fully consistent with previous results.11 Earlier
The terminal hydrophobicity of the films plays an equally
important role. As a general rule, longer alkyl termini and higher
(31) (a) Gagnon, S. D. Encycl. Polym. Sci. Eng. 1986, 6, 273-307. (b) Dreyfuss,
M. P.; Dreyfuss, P. Encycl. Polym. Sci. Eng. 1987, 10, 653-670.
(32) Heusch, R. Ber. Bunsen-Ges. 1979, 83, 834-840.
(34) Vanderah, D. J.; Valincius, G.; Meuse, C. W. Langmuir 2002, 18, 4674-
4680.
(35) Vanderah, D. J.; Arsenault, J.; La, H.; Gates, R. S.; Silin, V.; Meuse, C.
W. Langmuir 2003, 19, 3752-3756.
(33) Yoshida, S.; Sakiyama, M.; Seki, S. Polym. J. (Tokyo) 1970, 1, 573-581.
9
J. AM. CHEM. SOC. VOL. 125, NO. 31, 2003 9365

A R T I C L E S
Herrwerth et al.
terminal hydrophobicity lead to reduced protein repelling
properties. Contact angles up to 70° on gold and 65° on silver
seem to be tolerable for full protein resistance. Hydroxy
termination and therefore the highest hydrophilicity of the
monolayer generate in all cases the highest protein resistance.
This is in excellent agreement with recent results by Vogler36
who reported that long-range attractive forces are found only
between hydrophobic surfaces exhibiting a water contact angle
> 65°, but short-range repulsive forces are detected between
hydrophilic surfaces with contact angles below this value. The
reason for the general observation that a balance of forces is
achieved at a contact angle of about 65° has been discussed
and explained by us in a previous publication.37
Our results indicate that only the combination of several key
factors, the hydrophilicity of the termination, the hydrophilicity
of the internal units, and the lateral packing density, allows the
formation of a SAM that is fully protein resistant. If the ability
of a polyether SAM to coordinate water both in its interior and
on its surface is reduced when one of these factors is unfavorable
or absent, the overall protein resistance decreases.
approaching negatively charged object cannot replace the
negative surface charge and therefore will not adsorb on the
surface. Adsorption of cationic polyelectrolytes onto EG3OH
surfaces was shown to depend on the pH of the solution,38
presumably due to changes in the charge density and sign both
on the macromolecule and on the SAM surface.
Only at high ion concentration or low pH10b,39 is the negative
surface charge screened or neutralized, respectively, and other
forces and effects dominate the adsorption behavior at a distance
of e10 nm in front of the surface. Typically, an attractive
interaction is observed, in agreement with the hydrophobic
nature of perfect (theoretical) SAMs as was revealed in the
GCMC simulations for the EG3OMe SAMs both on Au and on
Ag substrates in contact with water.9 However, real surfaces
may have a less than ideal packing density and hence different
hydration properties that lead to repulsive solvation forces at
close approach,37 in the extreme for long chain and polydisperse
oligoethers leading to a mechanism in which resistance to
dehydration similar to polymeric PEG determines the adsorption
behavior.
As was shown in two other recent studies,22,23 the negative
surface charge inferred from the surface force experiments on
EG3OMe SAMs is due to preferential adsorption of hydroxide
ions from solution. Adsorption involves hydrogen bonding
between the hydroxide ion and the terminal hydrogen atoms in
the SAM surface, where it seems to be of less importance23 if
the terminal function is a methoxy or hydroxy moiety. This is
supported by experiments which show that negative surface
charging in aqueous solution is by no means unique to the
EG3OMe SAMs, but occurs to about the same extent (limited
apparently by the repulsive electrostatic interaction between the
adsorbed ions) on hexadecanethiolate and 11-mercaptoundecan-
1-ol SAMs on gold,22 which are clearly not protein resistant
and are not penetrated by water. As was discussed by Kreuzer
et al. in ref 23, one important role of water inside the SAM is
the formation of an additional hydrogen bond to the hydroxide
ion on the surface, and thereby the stabilization of the adsorbed
ion in absolute terms, but also against lateral displacement.
Hypothetically, protons or hydronium ions solvated within defect
sites of the oligoether monolayer could lead to further im-
mobilization of the adsorbed negative charge layer. Hence, when
the hydroxide ions are stabilized against displacement, an
Conclusions
In this paper, we have studied the importance of internal and
external hydrophilicity of oligoether SAMs for protein resis-
tance. We conclude that only the combination of several factors,
the hydrophilicity of the termination, the hydrophilicity of the
internal units, and the lateral packing density, allows the
formation of a SAM which is fully protein resistant. If the ability
of a polyether SAM to coordinate water both in its interior and
on its surface is reduced when one of these factors is unfavorable
or absent, the overall protein resistance decreases.
Acknowledgment. We thank Roman Glass and Alexander
Ku¨ller for help with the synthesis of several compounds and
Sabina Tatur for help with contact angle measurements. This
work was supported by the Deutsche Forschungsgemeinschaft,
the Office of Naval Research, and the Fonds der Chemischen
Industrie.
Supporting Information Available: Experimental section
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA034820Y
(38) (a) Clark, S. L.; Hammond, P. T. Langmuir 2000, 16, 10206-10214. (b)
Jiang, X.-P.; Clark, S. L.; Hammond, P. T. AdV. Mater. 2001, 13, 1669-
1673. (c) Jiang, X.; Ortiz, C.; Hammond, P. T. Langmuir 2002, 18, 1131-
1143.
(36) (a) Vogler, E. A. J. Biomater. Sci., Polym. Ed. 1999, 10, 1015-1045. (b)
Vogler, E. A. AdV. Colloid Interface Sci. 1998, 74, 69-117.
(37) Hayashi, T.; Pertsin, A. J.; Grunze, M. J. Chem. Phys. 2002, 117, 6271-
6280.
(39) Dicke, C.; Ha¨hner, G. J. Am. Chem. Soc. 2002, 124, 12619-12625.
9
9366 J. AM. CHEM. SOC. VOL. 125, NO. 31, 2003