Control of monolayer assembly structure by hydrogen bonding rather than by adsorbate-substrate templating

Article abstract of DOI:10.1021/ja9901011

Stratified amide-containing self-assembled monolayers (SAMs) provide opportunities for investigating the fundamental dependence of supramolecular structure upon molecular constitution. We report a series of amide-containing alkanethiol SAMs (C(n)-1AT/Au, n = 9, 11-16, 18) in which the hydrophobic overlayer thickness is systematically varied and the thickness of the polar region is held constant. The results from X-ray photoelectron spectroscopy, contact angle goniometry, reflective IR spectroscopy, and electrochemical measurements provide a consistent structural picture of the series. The amide underlayers in all the SAMs are well-ordered and extensively hydrogen bonded. However, the alkyl chains are disordered below n = 15. Comparison of the assembly structures shows that the chain length threshold for alkyl ordering is several methylenes higher than in n-alkanethiol SAMs. This indicates that alkyl chains adjacent to an amide underlayer are destabilized as compared to n-alkanethiols and that the amide underlayer destructively interferes with alkyl close packing as compared to the Au(111)-sulfur template. However, the amide regions of the SAMs are all well-ordered, showing that the amide sublayer acts as a 'template' that is independent of alkyl chain length. The amide region dominates over gold-sulfur epitaxy in establishing the structure of these assemblies, and the amide-alkyl boundary provides an example of a 'rigid-elastic' buried organic interface. Implications of these studies for molecular control of bulk properties, lipid-linked protein structure and function, buried organic interfaces in other systems, rationally designed ordered multilayers, and hybrid supramolecular systems are discussed.

Full text of DOI:10.1021/ja9901011

J. Am. Chem. Soc. 1999, 121, 5319-5327
5319
Control of Monolayer Assembly Structure by Hydrogen Bonding
Rather Than by Adsorbate-Substrate Templating
Robert S. Clegg and James E. Hutchison*
Contribution from the Department of Chemistry and Materials Science Institute, UniVersity of Oregon,
Eugene, Oregon 97403-1253
ReceiVed January 11, 1999
Abstract: Stratified amide-containing self-assembled monolayers (SAMs) provide opportunities for investigating
the fundamental dependence of supramolecular structure upon molecular constitution. We report a series of
amide-containing alkanethiol SAMs (Cn-1AT/Au, n ) 9, 11-16, 18) in which the hydrophobic overlayer
thickness is systematically varied and the thickness of the polar region is held constant. The results from
X-ray photoelectron spectroscopy, contact angle goniometry, reflective IR spectroscopy, and electrochemical
measurements provide a consistent structural picture of the series. The amide underlayers in all the SAMs are
well-ordered and extensively hydrogen bonded. However, the alkyl chains are disordered below n ) 15.
Comparison of the assembly structures shows that the chain length threshold for alkyl ordering is several
methylenes higher than in n-alkanethiol SAMs. This indicates that alkyl chains adjacent to an amide underlayer
are destabilized as compared to n-alkanethiols and that the amide underlayer destructively interferes with
alkyl close packing as compared to the Au(111)-sulfur template. However, the amide regions of the SAMs
are all well-ordered, showing that the amide sublayer acts as a “template” that is independent of alkyl chain
length. The amide region dominates over gold-sulfur epitaxy in establishing the structure of these assemblies,
and the amide-alkyl boundary provides an example of a “rigid-elastic” buried organic interface. Implications
of these studies for molecular control of bulk properties, lipid-linked protein structure and function, buried
organic interfaces in other systems, rationally designed ordered multilayers, and hybrid supramolecular systems
are discussed.
Introduction
in vertically architectured films has been achieved by sequential
multilayer growth.4 However, Ulman proposed in 1990 that
monolayer films might be similarly tiered to form vertical
domains with contrasting dielectric or optical properties.5 Such
films are desirable because the adsorption process is one-step
and monolayers tend to be better ordered than multilayers.6 The
introduction of functional groups in alkanethiol SAMs offers
ready access to such stratification. Thus far, well-ordered,
internally stratified alkanethiol-based SAMs are limited to those
in which diynes, ethers, and appropriately placed sulfones and
amides are incorporated in the hydrocarbon backbones of the
precursors.7,8 Often, internal functionalities result in decreased
Self-assembly of ultrathin organic films has found widespread
use in the control of surface properties as well as the
fundamental study of interfacial phenomena. The formation,1a-c
structure,1c-f properties,1f-h and applications2 of self-assembled
monolayers (SAMs) of n-alkanethiols on gold and other coinage
metals has been thoroughly reviewed. Although lithographic
patterning of surface bound films has received much attention,2e,f
vertical differentiation with controlled incorporation of sublay-
ered architectures at the molecular level is relatively unstudied.
Vertical structuring is desired for applications of SAMs for
nonlinear optics,3a electron-transfer catalysis,3b protein3c and
cell3d immobilization, physical3e and chemical3f sensing, artificial
photosynthesis,3g electron-transfer studies,3h biocompatible
surfaces,3i and corrosion passivation.3j The greatest complexity
(3) For reviews of the application of SAMs in these areas of interest,
see: (a) Nie, W. AdV. Mater. 1993, 5, 520-545. (b) Offord, D. A.; Sachs,
S. B.; Ennis, M. S.; Eberspacher, T. A.; Griffin, J. H.; Chidsey, C. E. D.;
Collman, J. P. J. Am. Chem. Soc. 1998, 120, 4478-4487. (c) Madoz, J.;
Kuznetzov, B. A.; Medrano, F. J.; Garcia, J. L.; Fernandez, V. M. J. Am.
Chem. Soc. 1997, 119, 1043-1051. (d) Mrksich, M. Cell. Mol. Life Sci.
1998, 54, 653-662. (e) Pope, J. M.; Tan, Z.; Kimbrell, S.; Buttry, D. A. J.
Am. Chem. Soc. 1992, 114, 10085-10086. (f) Li, S.; Crooks, R. M.
Langmuir 1993, 9, 1775-1780. (g) Hiroshi, I.; Hiroyuki, N.; Shinichiro,
O.; Kiminori, U.; Hiroko, Y.; Takayuki, A.; Koichi, T.; Yoshiteru, S.
Langmuir 1998, 14, 5335-5338. (h) Yu, H.-Z.; Zhang, H.-L.; Liu, Z.-F.
Langmuir 1998, 14, 619-624. (i) Miura, Y.; Kimura, S. Langmuir 1998,
14, 2761-2767. (j) Zamborini, F. P.; Crooks, R. M. Langmuir 1998, 14,
3279-3286.
(1) (a) Poirier, B. E.; Pylant, E. D. Science 1996, 272, 1145-1148. (b)
Karpovich, D. S.; Schessler, H. M.; Blanchard, G. J. Thin Films 1998, 24,
44-81. (c) Ulman, A. Chem. ReV. 1996, 96, 1533-1554. (d) Liu, G.-y.;
Xu, S.; Cruchon-Dupeyrat, S. Thin Films 1998, 24, 82-111. (e) Fenter, P.
Thin Films 1998, 24, 112-149. (f) Dubois, L. H.; Nuzzo, R. G. Annu. ReV.
Phys. Chem. 1992, 43, 437-463. (g) Finklea, H. O. In Electroanalytic
Chemistry: A Series of AdVances; Bard, A. J., Ed.; Dekker: New York,
1996; pp 110-336. (h) Delamarche, E.; Michel, B.; Biebuyck, H. A.;
Gerber, C. AdV. Mater. 1996, 8, 719-729.
(2) (a) Murray, R. W.; Ewing, A. G.; Durst, R. A. Anal. Chem. 1987,
59, 379A-385A. (b) Mrksich, M.; Whitesides, G. M. Annu. ReV. Biophys.
Biomol. Struct. 1996, 25, 55-78. (c) Crooks, R. M.; Ricco, A. J. Acc. Chem.
Res. 1998, 31, 219-227. (d) Liu, J.; Kim, A. Y.; Wang, L. Q.; Palmer, B.
J.; Chen, Y. L.; Bruinsma, P.; Bunker, B. C.; Exarhos, G. J.; Graff, G. L.;
Rieke, P. C.; Fryxell, G. E.; Virden, J. W.; Tarasevich, B. J.; Chick, L. A.
AdV. Colloid Interface Sci. 1996, 69, 131-180. (e) Xia, Y.; Whitesides, G.
M. Angew. Chem., Int. Ed. Engl. 1998, 37, 550-575. (f) Tien, J.; Younan,
X.; Whitesides, G. M. Thin Films 1998, 24, 227-254.
(4) For a review, see: Mallouk, T. E.; Kim, H.-N.; Ollivier, P. J.; Keller,
S. W. In ComprehensiVe Surpramolecular Chemistry, Vol. 7; Alberti, G.,
Bein, T., Eds.; Elsevier: Oxford, U.K., 1996; pp 189-217.
(5) Kuhn, H.; Ulman, A. In Thin Films, Vol. 20; Organic Thin Films
and Surfaces: Directions for the Nineties; Ulman, A., Ed.; Academic: San
Diego, CA, 1995; pp 1-10.
(6) Laibinis, P. E.; Palmer, B. J.; Lee, S.-W.; Jennings, G. K. Thin Films
1998, 24, 1-41.
10.1021/ja9901011 CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/25/1999

5320 J. Am. Chem. Soc., Vol. 121, No. 22, 1999
Clegg and Hutchison
By systematically varying the hydrophobic overlayer thick-
ness while holding constant the thickness of the polar amide
region, we show how the interaction of these two sublayers
influences the order of the entire assembly. The structural
interpretation based on an array of analytical data for a series
of alkanethiol SAMs containing a single, deeply buried amide
group (Cn-1AT/Au; n ) 9, 11-16, 18) is presented. For n g
Figure 1. Possible assembly structures of hybrid amide-alkyl SAMs
in which amide regions are sandwiched between a gold substrate and
an alkyl region. The parallel and nonparallel lines indicate ordered and
disordered regions respectively and do not contain information about
chain orientations. (A) Fully ordered SAM. (B) Fully disordered SAM.
(C) Disordered alkyl overlayer with an ordered amide underlayer. (D)
Ordered alkyl overlayer with a disordered amide underlayer.
15, the films are fully ordered (Figure 1A), but those with shorter
alkyl “tails” show selective disorder of the hydrocarbon region
(Figure 1C). In the n-alkanethiol series, an analogous chain
length ordering threshold of n ) 10 has been attributed to the
summation of hydrophobic interactions.14 The observed trends
among the assembly structures lead to two new concepts: (1)
alkyl chain packing adjacent to an amide underlayer is desta-
bilized as compared to n-alkanethiols, yet very highly ordered
assemblies are formed above a characteristic threshold chain
length, and (2) the buried organic interface in one-amide
assemblies behaves in a “rigid-elastic” manner, with the rigid
amide underlayer templating the alkyl chains analogously to
the current understanding of n-alkanethiols on gold.
order, as with alkenes and alkynes, N- and C-substituted
diglycines, single amide groups, and sulfones.8-11 Obviously,
disorder is incompatible with well-defined stratification and
stability of lateral patterning.5,12 Systematic studies of the effect
of sublayers of contrasting polarizability upon SAM order and
structure have not been reported.
Our approach toward a well-defined vertical architecture
involves replacing methylene units of alkanethiol SAMs with
amide groups. Previously, amide groups have been incorporated
at various levels in monolayers for a number of reasons. Primary
amides have been used as platforms for biomolecule im-
mobilization and cell adhesion.2b Internal (secondary and
tertiary) amides, in addition to providing efficient synthetic
routes,10 have been incorporated with the goals of controlling
molecular orientation11 and improving SAM stability.9,12 We
previously reported well-ordered, extensively hydrogen bonded
SAMs containing one7a and three9a internal amide groups. As
ordered, stratified SAMs, these systems have great potential for
studying the relationship between molecular and supramolecular
structure9,13 and the nature of buried interfaces within organic
assemblies.13 In addition, amide-based stratified films may prove
advantageous for lithographic applications because hydrogen
bonding interactions should promote monolayer stability9a and
hinder adsorbate diffusion across domain boundaries in mixed
SAMs.5,12
The rational design of ordered, stratified, amide-containing
SAMs requires an understanding of how the various material
sublayers affect each other. In these monolayers, two adjacent
regions are immediately obvious: (1) the more polar amide
underlayer lying proximal to the substrate and (2) the overlying
hydrophobic alkyl layer. The interactions within and between
these two sublayers will dictate the overall assembly structure
and order. Several scenarios are possible: the film may be fully
ordered (Figure 1A), completely disordered (Figure 1B), or only
partially ordered (Figure 1C,D).
Experimental Section
Materials and Synthesis. Dichloromethane was distilled from
calcium hydride immediately before use. All other solvents and reagents
were used as received from Aldrich, Bachem, Fluka, or Lancaster as
previously described.7a,9a
As a representative procedure, the synthesis of 3-mercapto-N-n-
nonylpropionamide C9-1AT is described here. In a 250 mL round-
bottom flask fitted with a magnetic stir bar, 0.5518 g (3.730 mmol) of
S-acetyl-3-mercaptopropionic acid,7a,18 0.7011 g (3.694 mmol) of 3-N′-
dimethylaminopropyl-N-ethylcarbodiimide,19 and 0.0447 g (0.369
mmol) of 4-(dimethylamino)pyridine were dissolved in 100 mL of
dichloromethane at room temperature. Upon addition of 0.670 mL (3.66
mmol) of n-nonylamine, the solution remained clear and colorless but
cooled transiently. After being stirred under nitrogen for 2-4 h, the
organic solution was washed with 3 × 100 mL of saturated sodium
bicarbonate, 3 × 100 mL of 0.01 M hydrochloric acid (adding 25 mL
of brine as needed to resolve emulsions), and 1 × 100 mL of distilled
water. The organic portion was separated, dried over sodium sulfate,
and filtered. The solvent was removed in vacuo to yield a white solid.
In an Erlenmeyer flask, 1.24 g of sodium hydroxide was dissolved in
125 mL of absolute methanol. This was added all at once to the white
solid in a 250 mL round-bottom flask. After being stirred under nitrogen
for 0.5-1 h, 40 mL of 5% hydrochloric acid was added all at once.7a,9b
A white precipitate immediately formed, and stirring was continued
for 0.5 h. After addition of 25-50 mL of brine, the product was
extracted with 3 × 80 mL of dichloromethane. The organic portion
was washed with 3 × 100 mL of distilled water (adding 25 mL of
brine as needed to resolve emulsions), dried over sodium sulfate, and
filtered. The solvent was removed in vacuo to yield a white micro-
(7) (a) Clegg, R. S.; Hutchison, J. E. Langmuir 1996, 12, 5239-5243.
(b) Batchelder, D. N.; Evans, S. D.; Freeman, T. L.; Ha¨ussling, L.;
Ringsdorf, H.; Wolf, H. J. Am. Chem. Soc. 1994, 116, 1050-1053.
(8) (a) Sinniah, K.; Cheng, J.; Terrettaz, S.; Reutt-Robey, J. E.; Miller,
C. J. J. Phys. Chem. 1995, 99, 14500-14505. (b) Evans, S. D.; Goppert-
Berarducci, K. E.; Urankar, E.; Gerenser, L. J.; Ulman, A. Langmuir 1991,
7, 2700-2709.
(9) (a) Clegg, R. S.; Reed, S. M.; Hutchison, J. E. J. Am. Chem. Soc.
1998, 120, 2486-2487. (b) Tam-Chang, S.-W.; Biebuyck, H. A.; Whitesides,
G. A.; Jeon, N.; Nuzzo, R. A. Langmuir 1995, 11, 4371-4382.
(10) Chechik, V.; Scho¨nherr, H.; Vancso, G. J.; Stirling, C. J. M.
Langmuir 1998, 14, 3003-3010.
(11) Zhang, J.; Zhang, H. L.; Chen, M.; Zhao, J.; Liu, Z. F.; Li, H. L.
Ber. Bunsen-Ges. Phys. Chem. 1998, 102, 701-703.
(14) (a) Bain, C. D.; Troughton, E. B.; Tao, Y.-T.; Evall, J.; Whitesides,
G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321-335. (b) Porter,
M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc.
1987, 109, 3559-3568.
(15) Dettre, R. H.; Johnson, R. E. J. Phys. Chem. 1965, 69, 1507-1515.
(16) Bar, G.; Rubin, S.; Cutts, R. W.; Taylor, T. N.; Zawadowski, T.
A., Jr. Langmuir 1996, 12, 1172-1179.
(17) Finklea, H. O.; Hanshew, D. D. J. Electroanal. Chem. 1993, 347,
327-340.
(12) Lenk, T. J.; Hallmark, V. M.; Hoffman, C. L.; Rabolt, J. F.; Castner,
D. G.; Erdelen, C.; Ringsdorf, H. Langmuir 1994, 10, 4610-4617.
(13) Clegg, R. S.; Reed, S. M.; Smith, R. K.; Barron, Bridgette L.; Rear,
J. A.; Hutchison, J. E. Submitted for publication.
(18) Daeniker, H. U.; Druey, J. HelV. Chim. Acta 1957, 40, 2148-2156.
(19) (a) Bodanszky, M. Principles of Peptide Synthesis; Springer-
Verlag: New York, 1984; pp 9-58. (b) Sheehan, J. C.; Cruickshank, P.
A.; Boshart, G. I. J. Org. Chem. 1961, 26, 2525-2528.

Control of Monolayer Assembly Structure by H Bonding
J. Am. Chem. Soc., Vol. 121, No. 22, 1999 5321
crystalline solid which was recrystallized from 95% ethanol. Yield (with
respect to n-alkylamine starting material): 87%. Mp: 45.6-47.5 °C.
1H NMR (300 MHz, CDCl3): δ 5.71 (broad, 1H), 3.26 (m, 2H), 2.82
(m, 2H), 2.47 (t, J ) 6.8 Hz, 2H), 1.62 (t, J ) 8.4 Hz, 1H), 1.49 (m,
2H), 1.45-1.24 (broad overlapping resonance, 12H), 0.88 (t, J ) 6.6
Hz, 3H).
3-Mercapto-N-n-undecylpropionamide C11-1AT was synthesized
analogously to C9-1AT with the substitution of n-undecylamine for
n-nonylamine. Yield: 85%. Mp: 57.5-59.0 °C. 1H NMR (300 MHz,
CDCl3): δ 5.55 (broad, 1H), 3.26 (m, 2H), 2.82 (m, 2H), 2.47 (t, J )
6.7 Hz, 2H), 1.61 (t, J ) 8.4 Hz, 1H), 1.50 (m, 2H), 1.41-1.23 (broad
overlapping resonance, 16H), 0.88 (t, J ) 6.6 Hz, 3H).
3-Mercapto-N-n-dodecylpropionamide C12-1AT was synthesized
analogously to C9-1AT with the substitution of n-dodecylamine for
n-nonylamine and with purification by preparative radial thin-layer
chromatography (1 mm rotor, gradient of 25% to 50% ethyl acetate in
n-hexane). Yield: 70%. Mp: 63.0-64.4 °C. 1H NMR (300 MHz,
CDCl3): δ 5.83 (broad, 1H), 3.27 (m, 2H), 2.83 (m, 2H), 2.48 (t, J )
6.8 Hz, 2H), 1.61 (t, J ) 8.3 Hz, 1H), 1.50 (m, 2H), 1.40-1.24 (broad
overlapping resonance, 18H), 0.89 (t, J ) 6.5 Hz, 3H).
3-Mercapto-N-n-tridecylpropionamide C13-1AT was synthesized
analogously to C9-1AT with the substitution of n-tridecylamine for
n-nonylamine and with purification by preparative radial thin-layer
chromatography (1 mm rotor, gradient of 10% to 30% ethyl acetate in
n-hexane). Yield: 44%. Mp: 67.5-68.7 °C. 1H NMR (300 MHz,
CDCl3): δ 5.70 (broad, 1H), 3.27 (m, 2H), 2.83 (m, 2H), 2.47 (t, J )
6.8 Hz, 2H), 1.61 (t, J ) 8.4 Hz, 1H), 1.51 (m, 2H), 1.38-1.23 (broad
overlapping resonance, 20H), 0.89 (t, J ) 6.6 Hz, 3H).
30 s at takeoff angles of 45° and 30°.16 Complete film removal was
ensured by multiplexing at 300-280 eV (the C(1s) region) after
sputtering. The attenuation length λ is 42 Å based on our previous
work.9a
Contact Angle Goniometry. Contact angles were measured using
a captive drop technique14a on a goniometer constructed in our
laboratory.9a Advancing and receding contact angles are defined as the
maximum and minimum angles formed between the film-droplet
interface and the tangent to the probe droplet at its intersection with
the substrate.15
Molecular Modeling. Molecular modeling was performed using
Spartan 5.0 (Wavefunction, Inc.) as previously reported.9a
FTIR-ERS. External reflective IR spectroscopy was performed using
a Nicolet Magna 550 IR with a Spectra-Tech 80 fixed reflectance
accessory and a Cambridge Physical Sciences IR wire grid polarizer.
Spectra were collected as 1024 signal-averaged interferograms as
previously reported.7a To minimize baseline periodicity and gas-phase
water peaks, the amide spectra shown are the Fourier transforms of
the interferogram averages of four to six independent samples. Residual
gas-phase water was subtracted from these spectra, and minimal
automatic baseline correction was applied using the Nicolet software.
All spectra shown are unsmoothed.
Electrochemical Characterization. A connect between the Au
working electrode (ca. 1 cm2) and a Teflon shrink-wrapped 22 g solid
Cu wire was made with Ag paint (Ted Pella). The exposed Ag, Cu,
and Cr surfaces were passivated with epoxy (Dexter). Double-layer
capacitance (Cdl) and electrochemical blocking effect (EBE) were
measured on a BAS 100 electrochemical workstation using a Pt counter
electrode and an SSCE reference electrode.7a The electrolyte was 1.0
M KCl (aq). For EBE, the analyte was 1.0 mM K3Fe(CN)6. The films
were electrochemically annealed by the method of Finklea.17
3-Mercapto-N-n-tetradecylpropionamide C14-1AT was synthesized
analogously to C9-1AT with the substitution of n-tetradecylamine for
n-nonylamine. Yield: 78%. Mp: 68.1-69.2 °C. 1H NMR (300 MHz,
CDCl3): δ 5.75 (broad, 1H), 3.26 (m, 2H), 2.82 (m, 2H), 2.47 (t, J )
6.7 Hz, 2H), 1.62 (t, J ) 8.3 Hz, 1H), 1.51 (m, 2H), 1.36-1.22 (broad
overlapping resonance, 22H), 0.88 (t, J ) 6.5 Hz, 3H).
Results
3-Mercapto-N-n-pentadecylpropionamide C15-1AT was synthesized
analogously to C9-1AT with the substitution of n-pentadecylamine for
n-nonylamine and with purification as for C12-1AT. Yield: 67%. Mp:
69.2-69.7 °C. 1H NMR (300 MHz, CDCl3): δ 5.54 (broad, 1H), 3.27
(m, 2H), 2.83 (m, 2H), 2.48 (t, J ) 6.8 Hz, 2H), 1.62 (t, J ) 8.4 Hz,
1H), 1.51 (m, 2H), 1.35-1.23 (broad overlapping resonance, 24H),
0.88 (t, J ) 6.6 Hz, 3H).
3-Mercapto-N-n-hexadecylpropionamide C16-1AT was synthesized
analogously to C9-1AT with the substitution of n-hexadecylamine for
n-nonylamine and with purification as for C12-1AT. Yield: 39%. Mp:
74.5-75.7 °C. 1H NMR (300 MHz, CDCl3): δ 5.68 (broad, 1H), 3.27
(m, 2H), 2.83 (m, 2H), 2.48 (t, J ) 6.7 Hz, 2H), 1.62 (t, J ) 8.4 Hz,
1H), 1.51 (m, 2H), 1.35-1.22 (broad overlapping resonance, 26H),
0.89 (t, J ) 6.5 Hz, 3H).
3-Mercapto-N-n-octadecylpropionamide C18-1AT was synthesized
analogously to C9-1AT with the substitution of n-octadecylamine for
n-nonylamine and with purification as for C12-1AT. Yield: 45%. Mp:
85.7-87.3 °C. 1H NMR (300 MHz, CDCl3): δ 5.81 (broad, 1H), 3.28
(m, 2H), 2.83 (m, 2H), 2.48 (t, J ) 6.8 Hz, 2H), 1.62 (t, J ) 8.3 Hz,
1H), 1.51 (m, 2H), 1.35-1.20 (broad overlapping resonance, 30H),
0.89 (t, J ) 6.6 Hz, 3H).
Formation of Substrates and SAMs. Au substrates (1500 Å) with
75 Å Cr adhesion layers were formed on glass slides by evaporation
in an Edwards 306A evaporation chamber (base pressure ) 6 × 10-7
mbar). The substrates were stored under absolute ethanol until use.
Glassware for adsorption and glass slides were cleaned with fresh
piranha solution (1:5 30% H2O2 in concentrated H2SO4; caution: reacts
Violently with organic material). Immediately before soaking in a 1
mM ethanolic solution of the desired thiol, substrates were cleaned for
5 min using a high-intensity Hg lamp (UV Clean, Boekel Industries)
in air and rinsed with copious Nanopure (Barnstead) water and absolute
ethanol. At least four independent measurements on separate samples
were obtained for all data sets.
Interpretation of the analytical data leads to a consistent
structural description of the monolayer series Cn-1AT/Au (n )
9, 11-16, 18). Each of the members of the series has an
extensively hydrogen bonded underlayer of amide groups.
Adjacent to the amide layer is either an ordered or a disordered
hydrocarbon overlayer, depending on the length of the alkyl
tails. We will show that the length threshold for alkyl tail
ordering is n ) 15.
X-ray Photoelectron Spectroscopy (XPS) Shows That
Chemisorbed, Anisotropic Monolayers Are Formed. XPS
provides information about the identity, chemical environment,
and depth of the constituent elements of these films.20 The S(2p)
peak positions are consistent with covalent attachment to
gold,14a,21 and the remaining SAM peaks occur at energies
expected for amide and alkyl material.7a,9,22 Due to the attenu-
ation of photoelectrons by overlying material, the relative
intensity of C(1s) is enhanced at the expense of the signals from
the other atoms.23 This effect is exaggerated in the longer chain
SAMs because the hydrocarbon overlayer is thicker.8b,12,24 The
film thicknesses calculated from XPS intensities agree with those
predicted by molecular modeling (Figure 2).9a
(20) For comprehensive reviews, see: (a) Briggs, D.; Riviere, J. C.;
Hofmann, S.; Seah, M. P. In Practical Surface Analysis by Auger X-ray
Photoelectron Spectroscopy; Briggs, D., Seah, M. P., Eds.; Wiley & Sons:
Chichester, U.K., 1983; pp 87-216. (b) Hochella, M. F., Jr. ReV. Mineral.
1988, 18, 573-637.
(21) (a) Dubois, L. H.; Zegarski, B. R.; Nuzzo, R. G. J. Am. Chem. Soc.
1990, 112, 570-579. (b) Nuzzo, R. G.; Dubois, L. H.; Allara, D. L. J. Am.
Chem. Soc. 1990, 112, 558-569.
(22) (a) Bomben, K. D.; Dev, S. D. Anal. Chem. 1988, 60, 1393-1397.
(b) Clark, D. T.; Peeling, J.; Colling, L. Biochim. Biophys. Acta 1976, 453,
533-545.
X-ray Photoelectron Spectroscopy. XPS was performed on a Kratos
HSi spectrometer as previously described,9a with binding energies
referenced to Au (4f7/2) at 84.0 eV. Thickness was calculated by area
integration of the Au (4f7/2) peak before and after Ar+ sputtering for
(23) Cadman, P.; Evans, S.; Scott, J. D.; Thomas, J. M. J. Chem. Soc.,
Faraday Trans. 2 1975, 71, 1777-1784.
(24) Troughton, E. B.; Bain, C. D.; Whitesides, G. M.; Nuzzo, R. G.;
Allara, D. L.; Porter, M. D. Langmuir 1988, 4, 365-385.

5322 J. Am. Chem. Soc., Vol. 121, No. 22, 1999
Clegg and Hutchison
small but significant (up to 6-8 cm-1) red shifts compared to
disordered films.14,21,24,26,27 Full widths at half-maxima of 12
cm-1 for CH2(as) and 8 cm-1 for CH2(sym) correspond to well-
ordered, all-trans chains with a low dispersion of conformational
states.8b,14b In accordance with the IR surface selection rule, peak
intensities are proportional to the dipole components perpen-
dicular to the substrate plane.28 Thus, the relative intensities of
the methylene stretch modes have been used to obtain quantita-
tive information about chain orientation for n-alkanethiols on
gold.14b,21b,26,29 Such an analysis is not applicable for amide-
containing alkanethiols because of uncertainty about the con-
formation of the ethylene linker between the amide and the
sulfur7a as well as its contribution to the intensities of the
methylene peaks.14b However, semiquantitative evaluation of
spectra using the parameters described above8b,14,21b,24,26,27 and
qualitative comparisons to SAMs with known struc-
tures8b,9b,10,12,24,30 are useful in evaluating film order.
Figure 2. Thickness by XPS of SAMs of alkanethiols containing a
single internal amide on gold (the series Cn-1AT/Au). Error bars denote
first standard deviation. Dashed line: calculated thicknesses for the
amide-containing alkanethiol SAMs of different chain lengths.
The Alkyl Region Is Well-Ordered When n g 15, but
Shorter SAMs have Disordered Chains. The reflective IR
spectra (Figure 4) for C15-tailed one-amide precursors show that
a well-ordered hydrocarbon overlayer forms and that the SAMs
are even more highly ordered for n g 16. For the latter
films, the CH2(as) and CH2(sym) frequencies,14,21b,24,26,27
intensities,14b,21b,26,29 and bandwidths8b,14b,21b,26,29 match those
of very highly ordered n-alkanethiol SAMs. As chain length
decreases, the first hint of decreased order comes from the
decreased intensity of the CH2 peaks in C15-1AT/Au. However,
the low peak frequencies and widths, as well as comparison to
C15S/Au26 (which is highly ordered), indicate that C15-1AT/Au
is well-ordered.
In contrast, the alkyl overlayers in Cn-1AT/Au with n ) 11-
14 are disordered. For these chain lengths, peaks due to
methylene stretching modes are weak, broadened, and blue
shifted. Although the weak intensities (Figure 5) are consistent
either with general disorder28 or a gradual loss of linearity
between peak intensity and C-H(str) dipole orientation,14b the
blue shifts14,21b,24,26,27 and broadening8b,14b are consistent only
with disorder.
Figure 3. Contact angles θ for SAMs of the series Cn-1AT/Au:
squares, water; triangles, hexadecane; filled symbols, advancing; open
symbols, receding. The dashed lines show the average contact angles
for n ) 13-18 and are included as guides to the eye. All measurements
are (2°.
The breakdown of quantitative reflective IR for very short
alkyl chains has been previously observed in n-alkanethiol
SAMs and has been attributed to an optoelectronic interaction
between methylene dipoles and metal electrons at a polarizable
interface.14b It is unlikely that the absence of methylene peaks
in the C9-1AT/Au spectra is evidence of close-packed chains
constrained vertically by decreased amide spacing because peaks
due to the amide dipoles are identical throughout the series.
The quantitative use of reflective IR for C-H(as) and C-H(sym)
intensities is not universally reliable;8a,14b indeed the data show
that the method is unreliable for C9-tailed one-amide SAMs.
Thus, a complementary characterization method is required to
test the structure of the hydrocarbon matrix for n ) 9-14. Such
a method is provided by electrochemical techniques (vide infra).
Contact Angle Goniometry Suggests That Order Varies
with Alkyl Chain Length. The hydrophobicity of a surface
reflects the degree of order in the top few atomic layers.21a,24
For Cn-1AT/Au, n g 13, the contact angle θ (Figure 3) is as
high as that for well-ordered n-alkanethiol SAMs,14a,24 suggest-
ing a maximal degree of order in the amide-containing series
above a minimum alkyl chain length. A slight decrease in θ
consistent with greater disorder at shorter chain lengths is noted
for n ) 11 and 12, especially using hexadecane, which is more
sensitive to disorder than water.6 The contact angle is substan-
tially decreased for n ) 9, consistent with disordered methyl
surfaces.6,24 The absence of increase in hysteresis below the
ordering threshold is addressed in the Discussion.
The IR Data Also Shows That the Amide Underlayers in
All of the SAMs Are Well-Ordered and Extensively Hydro-
gen Bonded. In contrast to the two types of alkyl regions, the
spectra indicate that the amide groups of the entire series are
Reflective IR (FTIR-ERS) Provides Both Orientational
and Conformational Information about the Hydrocarbon
Overlayers. In organic assemblies on reflective substrates,
specific information about alkyl chain orientation and order is
contained in the frequencies, widths, and relative intensities of
the methylene stretching bands.25 For example, close-packed
alkyl chains in n-alkanethiol SAMs on gold have CH2(as) and
CH2(sym) at 2918 and 2950 cm-1, respectively, representing
(26) Laibinis, P. E.; Whitesides, G. M.; Allara, D. L.; Tao, Y.-T.; Parikh,
A. N.; Nuzzo, R. G. J. Am. Chem. Soc. 1991, 113, 7152-7167.
(27) Snyder, R. G.; Strauss, H. L.; Elliger, C. A. J. Phys. Chem. 1982,
86, 5145-5150.
(28) (a) Greenleer, R. G. J. Chem. Phys. 1966, 44, 310-316. (b) Francis,
S. A.; Ellison, A. H. J. Opt. Soc. Am. 1959, 49, 131-138. (28) (c) Pearce,
H. A.; Sheppard, N. Surf. Sci. 1976, 59, 205-217.
(29) (a) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 52-65. (b) Parikh,
A. N.; Allara, D. L. J. Chem. Phys. 1992, 96, 927-945.
(25) For comprehensive reviews, see: (a) Crooks, R. M.; Xu, C.; Sun,
L.; Hill, S. L.; Ricco, A. J. Spectroscopy 1993, 8, 29-39. (b) Porter, M. D.
Anal. Chem. 1988, 60, 1143A-1155A.

Control of Monolayer Assembly Structure by H Bonding
J. Am. Chem. Soc., Vol. 121, No. 22, 1999 5323
Figure 4. FTIR-ERS spectra of SAMs of alkanethiols containing a single internal amide on gold: left, high-frequency C-H(str) region; right,
mid-frequency region (amides I, II, and III). (A) C18-1AT/Au. (B) C16-1AT/Au. (C) C15-1AT/Au. (D) C14-1AT/Au. (E) C13-1AT/Au. (F) C12-1AT/
Au. (G) C11-1AT/Au. (H) C9-1AT/Au.
alkyl overlayer) is essentially invariant for the series. The high
frequency of the amide II peak is consistent with a high degree
of hydrogen bonding, as the oscillator frequency follows the
degree of constraint placed by an associated carbonyl on the
N-H(bend) vibration.7a,32a,33
Amide III occurs as a broad peak at ∼1250 cm-1, supporting
a trans conformation as expected for an secondary amide
group.9b,31a,b The primary components of amide III are C-N(str)
and N-H(ipb),31a,b both of which contain large vectors parallel
to the long molecular axis. The extinction coefficient of amide
III is low compared to that of amide II,31a,b accounting for the
decreased intensity of amide III compared to amide II. The
amide IR data thus uniformly support the adsorbate orientations
depicted in Figure 1A,C and are consistent with the XPS and
contact angle data described above.
Microscopic Defects Such as Pinholes and Disordered
Chains Do Not Significantly Contribute to Electron Transfer
across Amide-Containing SAMs. Using the gold substrate and
the associated SAM as a working electrode in an electrochemical
cell provides further structural information about the SAM. Due
to the molecular scale of electron transfer (ET) and charge
storage (capacitive) processes, voltammetric techniques reflect
molecular order and disorder better than any of the previously
discussed methods, enabling discrimination between various
types of SAM defects.1g
In the electrochemical blocking experiment, the ability of a
dielectric layer to block ET between a dissolved redox probe
and the electrode is determined from a cyclic voltammogram
(CV). A systematic qualitative comparison of CVs often enables
differentiation among the three possible mechanisms for ET at
a SAM-covered electrode, which are in turn related to various
types of SAM defects that may be present.1g (1) Radial diffusion-
limited ET yields a sigmoidal increase in current near E°′ to a
plateau at significant overpotentials and is characteristic of
Figure 5. Intensities of FTIR-ERS peaks that are important indicators
of adsorbate orientation and order. Peak intensities are shown as a
function of alkyl chain length n for the series Cn-1AT/Au: filled
symbols, CH2(as); open symbols, amide II. The dashed line through
the data points at n ) 16 and 18 is included as an aid to the eye (see
text).
oriented with CdO and N-H bonds nearly parallel to the
substrate plane. Two diagnostic signals, amide A (N-H(str),
∼3300 cm-1) and amide I (CdO(str), ∼1640 cm-1),31 are absent
or very weak and broad in the SAM spectra. In accordance with
the surface selection rule, these weak intensities are consistent
with CdO and N-H bonds nearly parallel to the substrate
plane.7a,9,10,12,32
The dominant feature in the amide region is the amide II
peak at ∼1560 cm-1, consisting mainly of N-H(ipb).31 The
high intensity of this peak (along with the weak or absent amides
A and I) indicates that this dipole oscillates nearly perpendicular
to the substrate plane.7a,9,12,10,32 Importantly, the amide II
bandwidths (Figure 4) and intensities (Figure 5) are constant
among the spectra, indicating the amide underlayer (unlike the
(32) (a) Sabapathy, R. C.; Bhattacharyya, S.; Leavy, M. C.; Cleland, W.
E., Jr.; Hussey, C. L. Langmuir 1998, 14, 124-136. (b) Strong, A. E.;
Moore, B. D. Chem. Commun. 1998, 473-474. (c) Yamada, N.; Koyama,
E.; Imai, T.; Matsubara, K.; Ishida, S. Chem. Commun. 1996, 2297-2298.
(d) Cha, X.; Ariga, K.; Kunitake, T. Bull. Chem. Soc. Jpn. 1996, 69, 163-
168.
(33) Bamford, C. H.; Elliott, A.; Hanby, W. C. Synthetic Polypeptides:
Preparation, Structure and Properties; Academic: New York, 1956; pp
193-194.
(30) Tsao, M.-W.; Hoffman, C. L.; Rabolt, J. F.; Johnson, H. E.; Castner,
D. G.; Erdelen, C.; Ringsdorf, H. Langmuir 1997, 13, 4317-4322.
(31) (a) Krimm, S.; Bandekar, J. J. AdV. Protein Chem. 1986, 38, 181-
364. (b) Cheam, T. C.; Krimm, S. J. Chem. Phys. 1985, 82, 1631-1641.
(c) Dwivedi, A. M.; Krimm, S. Macromolecules 1982, 15, 177-185. (d)
Dwivedi, A. M.; Krimm, S. Biopolymers 1982, 21, 2377-2397.

5324 J. Am. Chem. Soc., Vol. 121, No. 22, 1999
Clegg and Hutchison
microelectrode arrays (i.e., widely spaced pinholes34 less than
∼1 µm in diameter).36 (2) Linear diffusion-limited ET results
in current decay at potentials beyond the peak potential. Such
a CV is characteristic of a film with large or closely spaced
pinholes34 or that is highly permeable to redox probe14b,37
(corresponding to extreme disorder, hydrophilicity, or poor
coverage34). (3) Kinetically limited ET yields a quasiexponential
current rise with increasing overpotential,38 consistent with a
through-chain mechanism of ET occurring at a nearly defect-
free SAM.14b,37b,39
In a complementary and more quantitative treatment of
electrochemical blocking CVs, the electrochemical blocking
effect (EBE) is calculated as
EBE ) 1 - (i1/2(film)/i1/2(bare)
)
(1)
where i1/2 denotes the Faradaic current (total current less
charging current) for a SAM-covered and bare electrode (each
measured at the half-wave potential recorded for the bare
electrode).40a A value of EBE near unity indicates that kinetically
limited ET predominates, as a significant component of either
Figure 6. Representative cyclic voltammograms from the electro-
chemical blocking and double layer capacitance experiments on SAMs
of alkanethiols containing a single internal amide on gold (scan rate )
100 mV/s; SSCE reference electrode): solid lines, 1.0 mM K3Fe(CN)6
in 1.0 M KCl; dashed lines, 1.0 M KCl. (A) C9-1AT/Au. (B) C18-1AT/
Au.
radial or linear diffusion-limited ET will sharply increase the
40
measured i1/2(film)
.
A more stringent test of the absence of
Table 1. Double-Layer Capacitance and Electrochemical Blocking
Effect for SAMs of Alkanethiols Containing a Single Internal
Amide on Golda
diffusion limited ET is provided by measuring the Faradaic
currents at different potential scan rates. For nearly defect-free
SAMs in which kinetically limited ET predominates, the
Faradaic current should be independent of scan rate.38
All of the SAMs in the series Cn-1AT/Au yield kinetically
limited voltammograms14b,37b,39 (Figure 6), indicating that neither
film permeation by analyte14b,37 nor diffusion of analyte to
pinholes36 plays a significant role in ET. Thus, ET is limited
primarily to through-chain electronic coupling,14b,37b,39 the
density of pinholes is insignificant,14b,36 and the substrate is well-
covered with adsorbate.14b,37 This interpretation is supported by
Cdlb (µF/cm2)
EBEc
C9-1AT/Au
C11-1AT/Au
C12-1AT/Au
C13-1AT/Au
C14-1AT/Au
C15-1AT/Au
C16-1AT/Au
C18-1AT/Au
C18S/Au
13.7 ( 0.8
8.0 ( 0.5
0.9985 ( 0.0003
0.9988 ( 0.0004
0.9989 ( 0.0003
0.9993 ( 0.0002
0.9994 ( 0.0003
0.9996 ( 0.0002
0.9996 ( 0.0002
0.9998 ( 0.0001
0.9995 ( 0.0003
5.9 ( 0.3
4.5 ( 0.2
2.8 ( 0.2
1.03 ( 0.04
1.00 ( 0.05
0.91 ( 0.03
0.97 ( 0.06
(34) For the purpose of this analysis, we adopt three general functional
classes of SAM defects. (i) Pinholes are bare spots large enough for analyte
to participate in unmediated ET at the electrode surface.35 These include
large areas of bare electrode, missing rows of adsorbate, and single missing
adsorbate molecules. At pinholes, analyte can approach the electrode closely
enough to undergo direct ET, so the EBE is sharply reduced if the total
pinhole area is significant.14b,35 (ii) Disordered domains are areas of high
coverage into which electrolyte (and possibly analyte, depending on
coverage) can permeate due to adsorbate disorder.14b ET from analyte
coupling to the electrode nearly across the full thickness of the SAM is far
slower than at pinholes (because the ET rate decreases exponentially with
distance), so if chain density is high enough that analyte neither approximates
the metal surface nor diffuses freely through the film,14b then disorder of
the alkyl overlayer reduces EBE the least of these defect types. (iii)
Intermediate cases include point defects and grain boundaries,1g where
adsorbate molecules in neighboring domains tilt away from each other. If
such defects extend to the level of the sulfur atoms, the effect on
electrochemical blocking behavior will resemble that of pinholes.1g However,
if such defects extend only to the base of the hydrocarbon overlayer, such
that the hydrogen bonding amide layer is not disrupted and electronic
coupling must occur through a significant part of the chain, the effect will
tend toward that of disordered domains.1g
a SSCE reference electrode. Our measurements for octadecanethiol
SAMs on gold (C18S/Au) are included for comparison to literature
b
values (see ref 14b). Cdl ) double-layer capacitance, as defined in eq
3; 1.0 M KCl, scan rate 100 mV/s. c EBE ) electrochemical blocking
effect at E°′, as defined in eq 1 (0 ) bare electrode; unity ) fully
blocking electrode); 1.0 mM K3Fe(CN)6 in 1.0 M KCl, scan rate 50-
500 mV/s.
the measured EBE values, which approach unity for all the
SAMs (Table 1) and indicate that direct ET between analyte
and metal is minimal.40 EBE is independent of scan rate (Table
1), further supporting kinetically limited ET, high adsorbate
coverage, and a low pinhole density.38
The EBE data also indicate that ordered and disordered alkyl
regions are present in the films of different lengths. The slight
but significant decrease in EBE with chain length (Table 1) is
consistent with decreased order for the shorter chains.14b,37 With
low pinhole density and high coverage, the amount of ET
occurring by a through-chain mechanism can increase in
disordered films because the redox probe can access portions
of the chains inside the terminal methyl groups.14b,37 The
observed EBE data are thus consistent with the alkyl IR data
which indicate disordered hydrocarbon chains for n e 14. A
more sensitive indication of the degree of alkyl order in films
known to have low pinhole densities is obtained from double-
layer capacitance measurements, as described next.
(35) Stranick, S. J.; Parikh, A. N.; Allara, D. L.; Weiss, P. S. J. Phys.
Chem. 1994, 98, 11136-11142.
(36) (a) Dayton, M. A.; Brown, J. C.; Stutts, K. J.; Wightman, R. M.
Anal. Chem. 1980, 52, 946-950. (b) Gao, Z.; Siow, K. S. Electrochim.
Acta 1997, 42, 315-321.
(37) (a) Ikeda, T.; Schmehl, R.; Denisevich, P.; Willman, K.; Murray,
R. W. J. Am. Chem. Soc. 1982, 104, 2683-2691. (b) Chidsey, C. E. D.;
Loiacono, D. N. Langmuir 1990, 6, 682-691.
(38) Miller, C.; Gra¨tzel, M. J. Phys. Chem. 1991, 95, 5225-5233.
(39) (a) Cheng, J.; Sa`ghi-Szabo´, G.; Tossell, J. A.; Miller, C. J. J. Am.
Chem. Soc. 1996, 118, 680-684. (b) Slowinski, K.; Chamberlain, R. V.;
Miller, C. J.; Majda, M. J. Am. Chem. Soc. 1997, 119, 11910-11919.
(40) (a) Chen, C.-h.; Hutchison, J. E.; Postlethwaite, T. A.; Richardson,
J. N.; Murray, R. W. Langmuir 1994, 10, 3332-3337. (b) Aslanidis, D.;
Fransaer, J.; Celis, J.-P. J. Electrochem. Soc. 1997, 144, 2352-2357. (c)
Skoog, M.; Kronkvist, K.; Johansson, G. Anal. Chim. Acta 1992, 269, 59-
64. (d) Hrbek, J. J. Phys. Chem. 1986, 90, 6217-6222.
Capacitance Measurements Indicate That the Alkyl Chains
Are Close-Packed in Cn-1AT/Au for n g 15. Another useful
parameter describing SAM characteristics is the double-layer
capacitance (Cdl) because a well-ordered SAM in the absence
of redox probe approximates the behavior of an ideal capacitor.1g

Control of Monolayer Assembly Structure by H Bonding
J. Am. Chem. Soc., Vol. 121, No. 22, 1999 5325
The magnitude of Cdl reflects the extent of the electrical double
layer formed at the electrode surface.41a If the SAM contains
significant defects such as pinholes34 or disordered chains,34
electrolyte should penetrate into the film, resulting in a more
extensive double layer and a high Cdl.14b,38,41 However, if the
SAM is impermeable to electrolyte, Cdl is low. Cdl is more
sensitive than EBE to the presence of disordered chains due to
selectivity of disordered domains for small ions of high charge
density (e.g., K+ and Cl- as used in Cdl measurements) over
large ions (e.g., Fe(CN)63-, the redox probe used in EBE
experiments).14b The sensitivity of Cdl is of course enhanced if
the pinhole area (from EBE) is low.
As predicted by the Helmholtz parallel plate capacitor
model,42 Cdl in well-ordered SAMs varies inversely with film
thickness d as
Figure 7. Plot of reciprocal capacitance as a function of SAM thickness
(scan rate ) 100 mV/s; 1.0 M KCl electrolyte). Error bars denote first
standard deviation. The dashed line is the regression line for n ) 15,
16, and 18.
Cdl-1 ) d(ꢀꢀï)-1
(2)
in which Cdl is determined from the measured charging current
ic according to
close-packed, well-ordered films. For the shorter monolayers
(n e 14), Cdl falls off precipitously as chain length is decreased,
indicating a sharp transition in either chain packing or ꢀ of the
adsorbate. The latter is unreasonable as noted above because
the change in material composition with chain length is gradual.
The deviation from the model indicates that a more extensive
electrical double layer forms than expected for n e 14, consistent
with monolayer permeability to ions or solvent. Therefore, the
electrochemical capacitance measurements and the FTIR-ERS
data both support an alkyl overlayer ordering threshold at n )
15 in Cn-1AT/Au.
Cdl ) ic(νA)-1
(3)
where ν is scan rate and A is electrode surface area. For example,
in ordered n-alkanethiol SAMs (CnS/Au, n g 12), eq 2 yields
a linear plot with slope corresponding to the apparent dielectric
constant ꢀ of the SAM.14b,39b Disordered n-alkanethiol SAMs
(n < 12) exhibit capacitances larger than predicted by the model,
consistent with film permeation by charge carriers.14b
In the amide-based series, a gradual decrease in ꢀ is expected
with a decreasing proportion of amide to alkyl material. Thus,
amide-containing SAMs with short alkyl chains may deviate
from the ideal line for two reasons: (1) permeation of a
disordered alkyl overlayer by electrolyte and/or solvent (adding
a diffuse layer component to the monolayer capacitance) and
(2) an increase in ꢀ with the increase in relative amide
component. At long chain lengths, however, the proportion of
amide material in the films is nearly constant among different
chain lengths. This allows another measure of order in the
SAMs, as the actual monolayer capacitances can be tested
against those predicted by the model for ordered films.
For the amide-containing SAMs with n g 15, Cdl is
comparable to that of C18S/Au (1.0 µF cm-2) (Table 1). The
slope of the regression line for n g 15 yields ꢀ ) 3.3 ( 0.8
(Figure 7), which compares favorably to Nylon-6,6 (ꢀ ) 3.6)43a
and polyethylene (ꢀ ) 2.3),43b as well as to the ordered CnS/
Au series (for which the slope yields ꢀ ) 2.3 using a chain tilt
of 28°).14b The conformity to the model supports impermeability
of the long-chain SAMs to ions and solvent as expected for
Discussion
The adsorption of Cn-1AT (n ) 9, 11-16, 18) onto gold
substrates yields chemisorbed monolayers with the following
structural features: (1) well-ordered amide layers with extensive
hydrogen bonding for all alkyl chain lengths, (2) disordered alkyl
domains in SAMs with short chains (n e 14), and (3) close-
packed, tilted alkyl domains in SAMs with long chains (n g
15). This structural interpretation is consistent with all results
from XPS, contact angle goniometry, FTIR-ERS, and capaci-
tance and electrochemical blocking studies.
The Alkyl Chain Length Ordering Threshold in the One-
Amide SAMs Is Approximately n ) 15. Based on the data,
alkyl chain length clearly determines whether the hydrocarbon
region is ordered or disordered. The results from the various
techniques must be evaluated by an integrated approach, and
appropriate comparisons to structurally related systems are
required.44
(41) (a) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Funda-
mentals and Applications; Wiley & Sons: New York: 1980; pp 488-511.
(b) Becka, A. M.; Miller, C. J. J. Phys. Chem. 1993, 97, 6233-6239.
(42) The empirically measured capacitance is actually the total interfacial
capacitance CT which is treated by Guoy-Chapman-Stern theory as two
series capacitances: (i) the double-layer capacitance Cdl and (ii) the
capacitance of the diffuse layer of electrolyte Cd. In the case of a SAM-
covered electrode, Cdl is equivalent to the monolayer capacitance Cm, the
parameter of interest for examining charge storage at the SAM-electrolyte
interface. CT provides an excellent approximation of Cm because Cd > 200
µF cm-2 in 1.0 M aqueous electrolyte near the potential of zero charge,14b,41b
and
As characterized by reflective IR the chains are well-ordered
at n ) 15 and very highly ordered at n g 16, but according to
Cdl, the chains are ordered at n g 15. This apparent discrepancy
(44) The synergy of the multiple characterization methods used in this
study leads to a stronger case for the structures proposed than if only one
or two of the methods were considered. Errors in this regard are not
uncommon and can lead to unfounded conclusions. We note in particular
a recent study11 in which similar amide-containing sublayers were covalently
bonded to propyl through heptyl tails. The use of limited characterization
techniques and the evaluation of the results without consideration of known
principles of n-alkanethiol and amide-based alkanethiol SAM structure led
to two apparently erroneous conclusions: (i) that amide underlayers in the
heptyl- and propyl-tailed films differ from each other and (ii) that heptyl
chains are ordered whereas propyl ones are not. The first conclusion is not
supported by differences in the amide IR peaks shown in that report,11 and
the second is inconsistent with the Cdl and contact angle data presented
here as well as with evidence that reflective IR does not reliably indicate
orientation for short alkyl chains.14b
-1
CT-1 ) Cm-1 + Cd
It follows that the contribution of Cd to CT is vanishingly small in this
system, with ic measured at +150 mV vs SSCE.
(43) (a) Dasgupta, S.; Hammond, W. B.; Goddard, W. A., III. J. Am.
Chem. Soc. 1996, 118, 12291-12301. (b) Lanza, V. L.; Herrman, D. B. J.
Polym. Sci. 1958, 28, 622-625.

5326 J. Am. Chem. Soc., Vol. 121, No. 22, 1999
Clegg and Hutchison
arises from differences in the two techniques. Cdl measures the
effect of the monolayer on the interaction of electrolyte and
electrode,41a,1g whereas reflective IR directly measures a physical
property of the adsorbate.25 Due to the hydrophobicity of
hydrocarbon chains, a difference of one methylene between Cdl
and FTIR-ERS is reasonable in view of the polarity of the
environments (aqueous electrolyte vs dinitrogen respectively).
constant could affect the hydrocarbon ordering process, as
suggested by recent work by Ulman et al.47
The Amide Region Dominates in Establishing the Struc-
ture of These Monolayers. All the characterization data are
consistent with the presence of amide underlayers that are well-
ordered and extensively hydrogen bonded, independent of
variations in alkyl packing. This directly indicates spacing and
order of the amide “template” are not affected by changes in
hydrocarbon orientation and order. It also suggests that the
hydrophobic interactions exert little influence on the degree of
hydrogen bonding in these films. In this way, alkanethiol SAM
stability may be enhanced by introducing hydrophilic cross-
linking interactions, as has been suggested by other authors.3c,9b,12
We previously reported results supporting this effect.9a Such
stability in short-chain SAMs is desired for improved conduc-
tometric sensor design.48
A likely reason hydrogen bonding dominates over gold-
sulfur templating (as in n-alkanethiol SAMs) in determining
these film structures is that the possible cost in free energy of
a gold-sulfur lattice mismatch is offset by replacing hydro-
phobic interactions (0.6-1.6 kcal/mol14a,45c) with N-H-O
hydrogen bonds (3-5 kcal/mol7a,9b,31a). Two types of gold-
sulfur lattice mismatches are possible. (1) All sulfur may be
bound to 3-fold gold hollow sites via the sp3 mode characteristic
of n-alkanethiol SAMs.45c-e Mismatch between the Au(111)
lattice and the amide lattice would be accommodated by
variations in the conformations of the ethylene linkers7a between
the sulfur and the amide groups (e.g., at an increase in enthalpy
of only ∼0.8 kcal/mol for a change from trans to gauche49)
and by occasional skips in the 3-fold hollow binding pattern.
(2) Multiple sulfur-gold bonding geometries or sites would also
allow the sulfur-gold templating to relax in favor of the
hydrogen bonding lattice parameters. Different modes are
certainly accessible, as both sp3 and sp bonding modes have
been observed in alkanethiol45c-e and aryl thiol50a SAMs,
respectively. Previous studies have calculated a potential energy
difference of 0.4 kcal/mol between these modes,1c and a
spectrum of binding sites and geometries may exist on the Au-
(111) surface.50b The actual binding motif at the organic-metal
interface may contain elements of each of these binding modes.
Elucidation of these details awaits direct structural analysis by
XRD, helium diffraction, or STM.
The threshold for order in the contact angle plot at n ) 13
occurs at shorter chain lengths than indicated by FTIR-ERS and
Cdl. Of the characterization techniques used here, contact angle
goniometry is the least sensitive to disorder. For example, in
n-alkanethiol SAMs the onset of contact angle decrease6,14,21a,24
is two to four methylenes below the ordering threshold
established by diffraction methods.45a-d This comparison ex-
plains the difference in the ordering threshold indicated by
contact angle goniometry compared to the other techniques.
Also, the hysteresis (difference between advancing and receding
contact angles) does not change as expected for measurements
in the vicinity of an ordering threshold.24 For both of these
reasons, we view contact angle goniometry as a useful screening
technique that is less reliable for detailed studies of order and
disorder in monolayer films than FTIR-ERS and measurement
of Cdl.
A Higher Alkyl Chain Length Ordering Threshold Exists
in One-Amide Than in n-Alkanethiol SAMs. In ordered CnS/
Au films, the threshold chain length is known to be n ) 10-
12,45a-d several methylenes shorter than in the amide-based
SAMs. Thus, an amide region destructiVely interferes with close
packing of alkyl chains in one-amide SAMs in comparison to
n-alkanethiol SAMs. In n-alkanethiol SAMs, the structure is
templated by the Au (111) lattice.45 Our results suggest that, in
the amide-containing SAMs studied, other interactions dominate
over Au-S epitaxy in establishing the assembly structure.
The difference in chain length threshold indicates a sensitivity
of hydrocarbon ordering to some difference between the
respective “templates.” Three general differences between an
amide underlayer and a gold-sulfur array may contribute to
the change in chain length threshold. (1) A difference between
the interamide spacing in an amide underlayer and the sulfur-
sulfur spacing in an n-alkanethiol SAM26 likely changes the
steric constraints on the alkyl overlayer. The average sulfur-
sulfur spacing of 4.97 Å in n-alkanethiols45d is replaced by the
∼5.1 Å amide cross section in the amide-based series,9b
assuming similarity of the adsorbate unit cell to that of the
structurally analogous N-methylacetamide.31b Any difference in
mismatch between head group and alkyl chain cross sections
likely changes the stabilization per hydrophobic interaction, as
discussed below.14a,45c,46 (2) As a result of a difference in head
group cross section, the orientation and/or torsional angle of
the N-CR bond in an amide-based SAM is likely different from
that of the S-CR bond in an n-alkanethiol SAM.1c Any change
in the constraints upon the nitrogen-carbon bond at the amide-
alkyl interface obviously alters the geometric considerations for
maximizing the energetics of the hydrophobic interactions.14a,45c,46
(3) Additionally, possible electronic effects arising from dif-
ferences in template polarizability, dipole moment, or dielectric
The Independence of Amide Structure from Hydrocarbon
Packing Demonstrates “Rigid-Elastic” Behavior at the
Amide-Alkyl Interface. A recent grazing incidence XRD study
of CnS/Au assemblies (n ) 10-30) revealed gradual trends in
tilt angle and direction as a function of n. This was explained
by an interplay between variable intermolecular interactions and
fixed substrate-monolayer interactions.45a-b Such a rigid-
elastic interface differs fundamentally from “rigid-rigid”
interfaces as found in epitaxially grown metal-metal films,
which exhibit sharp transitions between invariant regimes.
Here we find that the amide-alkyl interface of one-amide
SAMs proVides an example of a rigid-elastic buried organic
(47) Kang, J. F.; Liao, S.; Jordan, R.; Ulman, A. J. Am. Chem. Soc.
1998, 120, 9662-9667.
(48) Murray, R. W. In Chemical Sensors and Microinstrumentation;
Murray, R. W., Dessy, R. E., Heineman, W. R., Janata, J., Seitz, W. R.,
Eds.; ACS Symposium Series 403; American Chemical Society: Wash-
ington, DC, 1989; pp 2-3.
(45) (a) Camillone, N., III; Chidsey, C. E. D.; Liu, G.; Scoles, G. J.
Chem. Phys. 1993, 98, 3503-3511. (b) Camillone, N., III; Leung, T. Y.
B.; Scoles, G. Surf. Science 1997, 373, 333-349. (c) Fenter, P.; Eberhardt,
A.; Liang, K. S.; Eisenberger, P. J. Chem. Phys. 1997, 106, 1600-1608.
(d) Fenter, P.; Eberhardt, A.; Eisenberger, P. Science 1994, 266, 1216-
1218. (e) Strong, L.; Whitesides, G. M. Langmuir. 1988, 4, 546-558. (f)
Poirier, G. E.; Tarlov, M. J. Langmuir 1994, 10, 2853-2856.
(46) Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. J. Am. Chem. Soc.
1987, 109, 733-740.
(49) Carey, F. A.; Sundberg, R. J. AdVanced Organic Chemistry: Part
A, Structure and Mechanisms, 3rd ed.; Plenum: New York, 1990; pp 121-
122.
(50) (a) Sabatani, E.; Cohen-Boulakia, J.; Bruening, M.; Rubinstein, I.
Langmuir 1993, 9, 2974-2981. (b) Sellers, H.; Ulman, A.; Shnidman, Y.;
Eilers, J. E. J. Am. Chem. Soc. 1993, 115, 9389-9401.

Control of Monolayer Assembly Structure by H Bonding
J. Am. Chem. Soc., Vol. 121, No. 22, 1999 5327
interface. The amide region is ordered in all cases, and the
hydrocarbon overlayer is ordered only for n g 15. Thus, the
amide region acts as a structured “template” upon which the
preattached hydrocarbon chains interact with each other laterally
via van der Waals forces. Analogous to n-alkanethiols, these
chains will become ordered if they are long enough to generate
sufficient hydrophobic interactions. However, a different tem-
plating occurs in the amide-containing SAMs compared to
n-alkanethiol ones, as evidenced by the higher ordering thresh-
old. Above this threshold, each assembly structure presumably
finds a unique energetic minimum for each chain length. Below
the minimum number of hydrophobic interactions, disorder
results, but the rigid amide “template” is essentially unaffected.
The Relationships among Amide-Containing Alkanethiol
SAMs Have Implications for Other Systems in Which
Interactions between Hydrophilic and Hydrophobic Do-
mains Are Important. It should be noted that the rigid-elastic
nature of one-amide-containing SAMs is not common to all
amide-containing alkanethiol SAMs. In similar SAMs with two
amide groups per precursor, the interplay between the amide
and alkyl regions drives the ordering of both regions above a
certain alkyl threshold (corresponding to a transition from Figure
1B to Figure 1A).13 Thus, the two-amide SAMs provide
examples of a previously unobserved elastic-elastic interface.
However, SAMs with three amide groups behave similarly to
those with one amide, except that the amide region assumes a
different rigid conformation.13 Thus, in this general class of
SAMs, a rich higher order structural diversity depends on the
atomic constitution of the precursor molecules, analogous to
the dependence of protein folding upon primary structure51 and
of cell membrane function upon surfactant chain length.51b
Films containing buried organic interfaces and competing
stabilizing interactions are also important as analogues of lipid-
protein arrays52a-b and covalently bonded lipid-linked proteins.52c-e
Lipoproteins and lipid-linked proteins are known to interact with
cell membranes to cause local disturbances in surfactant
packing,52f providing the basis for diverse events such as the
anchoring of viral coat proteins52c and modulation of membrane
permeablility by oligopeptide drugs.52g Further development of
the films reported here could provide the basis for modeling
such biological processes.
assembly process of an adjacent stratum, causing lattice
perturbations that would affect the order or disorder of the
overall assembly or subsequently assembled strata. Such effects
may not be limited to two-dimensional ordering processes and
may be important in the design of ordered heterostructured bulk
materials.54 We are presently investigating the thermal and
solvent stability of these SAMs, as a quantitative understanding
of these subtle inter- and intramolecular interactions is the basis
for predicting structure and function in future systems.
Conclusions
In the series of alkanethiol SAMs Cn-1AT/Au (n ) 9, 11-
16, 18) containing a single internal amide group, the interaction
of a hydrocarbon overlayer with an adjacent amide underlayer
has been probed. Variation of the hydrocarbon overlayer
thickness allows an understanding of the influence of alkyl chain
length on supramolecular structure. FTIR-ERS, supported by
electrochemical studies, contact angle goniometry, and XPS,
shows that the amide underlayers are identical among the entire
series. However, the alkyl overlayers are ordered only above n
g 15, indicating that hydrophobic interactions are of secondary
importance to hydrogen bonding in establishing film order and
stability. The destabilization of alkyl chains adjacent to an amide
underlayer (compared to n-alkanethiols) is probably due to
differences in spacing of the templating head groups. Similarly,
the amide underlayer dominates in determining SAM structure
most likely because hydrogen bonding interactions are signifi-
cantly stronger than (1) the hydrophobic interactions they replace
and (2) the difference between accessible sulfur-substrate
binding modes with which they compete.
This system provides unique opportunities to study the nature
of buried organic interfaces. The one-amide amide-alkyl
interface behaves in a rigid-elastic manner, similar to a recently
proposed model45b of interplay between gold-sulfur templating
and hydrophobic interactions in n-alkanethiol SAMs. This class
of SAMs is relevant to diverse areas of research including
studies of interactions between lipid-linked peptides and lipid
bilayers, control of bulk properties using intermolecular interac-
tions and design of heterostructured monolayers, multilayers and
bulk materials.
The significance of the observed variable alkyl threshold to
the design of new heterostructured assemblies must not be
underestimated. The general principle may apply to any system
in which sublayers with different lateral interactions are to be
incorporated. For example, in Langmuir-Blodgett films53 or
self-assembled multilayers,4 a rigid sublayer may influence the
Acknowledgment. This research was supported by the
National Science Foundation (NMR: CHE-9421182; XPS:
CHE-9512185), the National Science Foundation CAREER
Program (CHE-9702726), the Camille and Henry Dreyfus
Foundation, the Oregon Medical Research Foundation, and the
University Club Foundation of Portland, OR. We thank Kevin
Becraft for technical assistance with the electrochemical experi-
ments and Scott M. Reed and Rachel K. Smith for critical
readings of the manuscript.
(51) (a) Kelly, J. W. Structure 1997, 5, 595-600. (b) Yeagle, P. L. The
Structure of Biological Membranes; CRC: Boca Raton, FL, 1991; pp 539-
602, 721-780.
(52) (a) Kaiser, E. T.; Ke´zdy, F. J. Annu. ReV. Biophys. Biophys. Chem.
1987, 16, 561-581. (b) von Heijne, G. Biochim. Biophys. Acta 1988, 947,
307-333. (c) Sefton, B. M.; Buss, J. E. J. Cell. Biol. 1987, 104, 1449-
1453. (d) Yu, Y.-C.; Tirrell, M.; Fields, G. B. J. Am. Chem. Soc. 1998,
120, 9979-9987. (e) Yu, Y.-C.; Berndt, P.; Tirrell, M.; Fields, G. B. J.
Am. Chem. Soc. 1996, 118, 12515-12520. (f) Batenburg, A. M.; Hibbeln,
J. C. L.; de Kruijff, B. Biochim. Biophys. Acta 1987, 903, 155-165. (g)
Surewicz, W. K.; Hantsch, H. H.; Stahl, G. L.; Epand, R. M. Proc. Natl.
Acad. Sci. U.S.A. 1987, 84, 7028-2030.
JA9901011
(53) Maoz, R.; Yam, R.; Berkovic, G.; Sagiv, J. Thin Films 1995, 20,
41-68.
(54) Krejchi, M. T.; Atkins, E. D. T.; Waddon, A. J.; Fournier, M. J.;
Mason, T. L.; Tirrell, D. A. Science 1994, 265, 1427-1432.