Conformation, and Charge Tunneling through Molecules in SAMs

Article abstract of DOI:10.1021/jacs.0c12571

This paper demonstrates that the molecular conformation (in addition to the composition and structure) of molecules making up self-assembled monolayers (SAMs) influences the rates of charge tunneling (CT) through them, in molecular junctions of the form AuTS/S(CH2)2CONR1R2//Ga2O3/EGaIn, where R1 and R2 are alkyl chains of different length. The lengths of chains R1 and R2 were selected to influence the conformations and conformational homogeneity of the molecules in the monolayer. The conformations of the molecules influence the thickness of the monolayer (i.e. tunneling barrier width) and their rectification ratios at ±1.0 V. When R1 = H, the molecules are well ordered and exist predominantly in trans-extended conformations. When R1 is an alkyl group (e.g., R1 H), however, their conformations can no longer be all-trans-extended, and the molecules adopt more gauche dihedral angles. This change in the type of conformation decreases the conformational order and influences the rates of tunneling. When R1 = R2, the rates of CT decrease (up to 6.3×), relative to rates of CT observed through SAMs having the same total chain lengths, or thicknesses, when R1 = H. When R1 H R2, there is a weaker correlation (relative to that when R1 = H or R1 = R2) between current density and chain length or monolayer thickness, and in some cases the rates of CT through SAMs made from molecules with different R2 groups are different, even when the thicknesses of the SAMs (as determined by XPS) are the same. These results indicate that the thickness of a monolayer composed of insulating, amide-containing alkanethiols does not solely determine the rate of CT, and rates of charge tunneling are influenced by the conformation of the molecules making up the junction.

Full text of DOI:10.1021/jacs.0c12571

pubs.acs.org/JACS
Article
Conformation, and Charge Tunneling through Molecules in SAMs
Lee Belding,§ Samuel E. Root,§ Yuan Li, Junwoo Park, Mostafa Baghbanzadeh, Edwin Rojas,
Priscilla F. Pieters, Hyo Jae Yoon, and George M. Whitesides*
Cite This: J. Am. Chem. Soc. 2021, 143, 3481−3493
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: This paper demonstrates that the molecular conformation
(in addition to the composition and structure) of molecules making up
self-assembled monolayers (SAMs) influences the rates of charge
tunneling (CT) through them, in molecular junctions of the form
AuTS/S(CH2)2CONR1R2//Ga2O3/EGaIn, where R1 and R2 are alkyl
chains of different length. The lengths of chains R1 and R2 were selected
to influence the conformations and conformational homogeneity of the
molecules in the monolayer. The conformations of the molecules influence the thickness of the monolayer (i.e. tunneling barrier
width) and their rectification ratios at 1.0 V. When R1 = H, the molecules are well ordered and exist predominantly in trans-
extended conformations. When R1 is an alkyl group (e.g., R1 ≠ H), however, their conformations can no longer be all-trans-extended,
and the molecules adopt more gauche dihedral angles. This change in the type of conformation decreases the conformational order
and influences the rates of tunneling. When R1 = R2, the rates of CT decrease (up to 6.3×), relative to rates of CT observed through
SAMs having the same total chain lengths, or thicknesses, when R1 = H. When R1 ≠ H ≠ R2, there is a weaker correlation (relative to
that when R1 = H or R1 = R2) between current density and chain length or monolayer thickness, and in some cases the rates of CT
through SAMs made from molecules with different R2 groups are different, even when the thicknesses of the SAMs (as determined
by XPS) are the same. These results indicate that the thickness of a monolayer composed of insulating, amide-containing
alkanethiols does not solely determine the rate of CT, and rates of charge tunneling are influenced by the conformation of the
molecules making up the junction.
illustrates the assumptions underlying this design (and
research) in their simplest form. When the group R1 on
nitrogen is a hydrogen (R1 = H, R2 = n-alkyl), the organic
molecules making up the SAM can condense to a relatively
high surface density through intermolecular hydrogen bonds.
When R1 = CH3 or larger alkyl chain, steric interactions
prevent this condensation, the molecules in the SAM are more
separated, and there is an opportunity (in fact, the require-
ment) for the larger chain to bend into structures that have
gauche conformations in order to increase the density of
packing of the organic layer. Thus, introducing larger R2
groups also introduces gauche conformations. This simple
argumentwithout defining the details of these conforma-
tionsis the immediate justification of the paper. Molecular
dynamics (MD) simulations reinforced our initial assumptions
that (i) secondary amides have well-ordered, trans-extended
geometries, (ii) symmetrical tertiary amides form SAMs that
are not fully trans extended, and (iii) the unsymmetrical
tertiary amides form SAMs with more disordered conforma-
INTRODUCTION
■
When electrical charge tunnels through electrically insulating
organic molecules between electrically conducting electrodes,
the rate of tunneling follows (or is influenced by) the
molecular and electronic structure of these molecules.1 For
junctions based on self-assembled monolayers (SAMs) of
saturated organic groups, a relevant question is “is the
conformation of the molecule important in determining the
rate of charge tunneling through it, or is it simply the number
of, and nature of, bondsindependent of their conformation
that matter?” If the former is correct, the conformation can be
used to influence the pathway (and rate) of charge tunneling
(CT). If it is not important, then the molecular structure
influences tunneling by some mechanism (e.g., perturbation of
the electric field within the tunneling barrier) and thus the
structure, width, average height, and shape of this barrier are
independent of molecular conformation. Here, we study the
simplest form of this question: is the rate of CT from one
electrode to a second different or the same though trans and
gauche fragments of a polymethylene (CH2)n chain with the
same number of carbon atoms?
Received: December 2, 2020
Published: February 23, 2021
To address this question, we synthesized dialkyl disulfides
having the general structure (S-(CH2)2-CON(R1)(R2))2
(where R1 is an alkyl chain and R2 is an alkyl group or a
hydrogen atom) and made SAMs of them by adsorption onto
evaporated, template-stripped gold surfaces (AuTS). Figure 1
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493
© 2021 American Chemical Society
3481

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 1. SAMs having the general structure S-(CH2)2-CON(R1)(R2) (where R1 is an alkyl chain and R2 is an alkyl group or a hydrogen atom).
When R2 = H, adjacent amide groups can undergo hydrogen bonding and the molecules can adopt uniformly trans extended conformations. These
ordered conformations allow the molecules in the SAM to fall into van der Waals contact and to pack close together, thus creating a more densely
packed SAM. When R2 = n-alkyl, the amide groups can no longer accommodate hydrogen bonding, and steric repulsion forces a larger distance
between adjacent molecules, thus decreasing the packing density within the SAM. To avoid empty space, the alkyl chains R1 and R2 adopt a mixture
of gauche and trans conformations.2−4
tions in comparison to secondary amides or symmetrical
tertiary amides.
available and thus enable studies correlating the structure with
the rates of tunneling. (iii) SAMs (as opposed to single
molecules or a solution) are among the plausible physical
systems with which to fabricate molecular electronic devices.27
Efforts to control charge tunneling through SAMs have
focused on the electronic structure of the molecules: the nature
of the molecule−metal contact,28,29 conjugation,30−32 dipole
moments,33,34 donor and acceptor systems,35 and asymmetri-
cally positioned chromophores.36−38 The conformations of the
molecules making up a SAMrather than simply their bond
connectivitymight also influence the rates of charge
transport, by allowing longer-range orbital overlaps, by
influencing the electronic structure, by changing the width,
mean height, or energetic topography of the tunneling barrier,
or by influencing intermolecular electronic interactions in the
SAM. By a comparison of the thicknesses, surface coverages,
rates of charge tunneling, and rectification ratios between well-
ordered SAMs and structurally disordered SAMs (Figure 2), in
this paper, we infer that the conformation does influence
tunneling and that shorter distances between electrodes do not
always imply larger tunneling rates (even in simple aliphatic
molecules). We synthesized four series of dialkyl disulfides
containing one amide group (Figure 2) having the general
structure (S-(CH2)2-CON(R1)(R2))2 (where R1 is an alkyl
chain and R2 is an alkyl group or a hydrogen atom) and made
SAMs of them by adsorption onto evaporated, template-
stripped gold surfaces (AuTS).39
The question of how the molecular conformation affects
tunneling is important for the ability to design or rationalize
phenomena in at least three areas. (i) Molecular electronics: The
transfer of charge by tunneling becomes more important in
modern technology as electronic components approach the
molecular scale5−7 (for example, in capacitors, where organic
monolayers have been explored to decrease leakage currents
and thus reduce power consumption in complementary logic
circuits based on organic field effect transistors8,9). (ii)
Biochemistry: Charge tunneling is integral to the function of
a number of important biological systems such as redox
systems (both single-enzyme and membrane-embedded, multi-
enzyme redox systems), DNA repair enzymes, and enzymes of
nitrogen fixation.10−15 (iii) Theory: The question of the
relationship between HOMOs and LUMOs (both are
important to tunneling) and the conformations of organic
molecules (as opposed to their composition and structure)
have been studied only briefly.16,17
The conformation of molecules may also be important in
other areas, such as the activity of photocatalysts,18 an
understanding of the properties (and mechanism) of charge-
transfer-based fluorophores,19,20 charge transfer at an elec-
trode/electrolyte interface,21 and the development of corrosion
inhibitors.22 Here, a comparison of the rates of CT through
organic molecules of similar structures but different con-
formations suggests that CT is influenced by the conformation
of the molecule(s).
The assumptions underlying this design (and research) is
that molecules with secondary amides can adopt linear (trans-
extended) conformations, but molecules with tertiary amides
must adopt geometries that deviate from linearity by
incorporating gauche conformations.
In tertiary amides (we supposed) larger differences between
the lengths of the chain R1 and R2 would cause the molecules
in the SAM to adopt conformations that were increasingly
Self-assembled monolayers (SAMs) are model systems used
(inter alia) to study charge transport through organic
molecules23−26 for the following reasons. (i) SAMs formed
from n-alkanethiols have been extensively studied, have well-
characterized, trans-extended conformations, and are ordered.
(ii) Organic thiols of a wide range of structures are readily
3482
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 2. Structures of the secondary amides 1−5, with the total number of atoms in the chain (n) ranging from 8 to 16. Structures of the
symmetrical tertiary amides, 6−10, where the length of chains a and b are the same. Structures of the series I amides: 5 and 7−13, where the total
number of carbon atoms in chain a and b is 12. Structures of the series II amides: 5, 10, and 14−18, where the length of chain a is kept at 12
carbons, while the length of chain b is varied.
more disordered. If the two branching chains R1 and R2 have
unequal lengths, they cannot both be linearly extended and
maintain a uniform density in the SAM; there must be a
change in conformation to eliminate void space in the SAM.
The objective of this work was to determine if disordered
conformations and ordered conformations of homologous
molecules differed significantly in the rates of tunneling
through them at a common voltage (+0.5 V).
We chose to use 3-mercaptopropionamides (S-(CH2)2-
CON-R1R2; Figures 1 and 2) to synthesize conformationally
ordered and disordered SAMs, because (i) they are syntheti-
cally simple to prepare, (ii) the amide nitrogen can be
monoalkylated to form linear structures or dialkylated to form
branched structures, (iii) the short anchoring group permits a
large range of R1 and R2 (i.e. lengths of chains a and b), and
(iv) our previous work has shown that molecules with amide
groups in this position exhibit rectification of tunneling current
and that this rectification is sensitive to the conformational
order within the SAMs.34
assigned to each molecule in this study, “a” is the number of
carbon atoms in chain R1, and “b” is the number of carbon
atoms in R2 (Figure 2). We also use n to represent the number
of atoms making up the longest path in the molecules,
including the -(CH2)2-CON− anchoring group. The R1 group
(of length a) is an n-alkyl chain, and the R2 substituent (of
length b) is either an n-alkyl chain or a hydrogen atom.
We used a combination of molecular dynamics (MD)
simulations and X-ray photoelectron spectroscopy (XPS) to
characterize the thicknesses, surface coverages, and degrees of
disorder within the SAMs. We ran all tunneling measurements
using the EGaIn junction at a maximum applied bias of 1.0 V.
Amide-containing alkanethiols rectify current at 1.0 V and not
at 0.5 V; therefore, we compare the rates of tunneling at +0.5 V
and the rectification ratios at 1.0 V. We define the
rectification ratio (r) as the absolute value of the larger
current density divided by the absolute value of the lower
current density and include an indication of polarity: r+ = |
J(+V)|/|J(−V)| or r− = |J(−V)|/|J(+V)|. In this definition r is
always ≥1. We note herein our sign conventionthat at
positive bias (+V) the EGaIn electrode is oxidizing relative to
For the molecules with the form S-(CH2)2-CON(R1)(R2),
we use the abbreviation X(a,b), where X is the index number
3483
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
the Au electrode and at negative bias (−V) the EGaIn
the bonds (favoring trans geometries).60−63 More recent work,
however, based primarily (but not entirely64) on experimental
observations and theoretical models65 of temperature-depend-
ent tunneling rates66−68which are interpreted as evidence of
conformationally dependent (and thus thermally activated)
tunnelingsuggest that charge tunneling takes place, at least
in part, via intramolecular69−71 or intermolecular71 through-
space pathways.
The conformations of molecules within a SAM are
influenced byin addition to their chemical structuresthe
topography of the bottom electrode. Grain boundaries in the
substrate are associated with “thin-area defects” because they
cause localized disorder in the SAM.40,72,73 It is generally
assumed that thin-area defects lead to tunneling currents
through molecular junctions that are higher than expected,
given the chain lengths of the molecules, because the
orientations of these molecules are disordered and leave a
shorter distance between the bottom and top electrodes.40−42
This rationale has been used to explain the observation that the
rates of charge tunneling through SAMs made on template-
stripped (TS) substrates (which have fewer exposed grain
boundaries in comparison to substrates made from direct
evaporation72,73) are lower than rates of tunneling through
SAMs with the bottom electrodes made by direct evapo-
ration.72
Rectification in SAMs Containing Amides. We have
recently reported that SAMs with polar functional groups
(amides, ureas, and thioureas) located close to a bottom AgTS
electrode rectify current (r+ ≈ 20) at an applied bias of greater
than ∼0.6 V34where tunneling takes place predominantly by
a Fowler−Nordheim (F-N) mechanism. Rectification in these
systems is related to the magnitude (and direction) of the net
dipole moment perpendicular to (and away from) the bottom
electrode.74 Changing the orientation of the amide group
(from −CONH− to −NHCO−) or disrupting the structure of
the SAM (by N-methylation) eliminated rectification in these
systems, presumably because these physical changes reduce the
net dipole moment perpendicular to the bottom electrode.
electrode is reducing relative to the Au electrode.36
BACKGROUND
■
Liquid-like and Crystalline SAMs. SAMs made from n-
alkanethiols exhibit length-dependent physical properties. For
example, relative to SAMs with shorter (n < 12) chain lengths,
those with longer chain lengths have larger water contact
angles4 and are more prone to leakage currents.40−42
Alkanethiol-based SAMs have four states (or phases), based
on their chain length.4 Below C10 (an n-alkanethiol SAM with
10 carbon atoms), SAMs exist in a “liquid-like” state, where the
entropic free energy associated with vibrational, translational,
and rotational motion is larger than the enthalpic free energy
associated with intermolecular van der Waals interactions.4
This liquid-like regime is associated with less conformational
order and more gauche defects in comparison to those
observed for SAMs with higher chain lengths.4 Between
C9/10 and C14, there is a transition from liquid-like to
crystalline-like SAMsa “wax-like” statewith physical
properties between those of shorter and longer chain lengths.
At or above C14, the alkanethiols are well-ordered and form
rigid, crystalline SAMs that have all-trans conformations.4,43
Such length-dependent thermophysical transitions also occur
with SAMs of linear alkyl phosphonates on aluminum oxide:
junctions formed with top electrodes of thermally evaporated
gold41 and droplets of mercury42 both exhibited increased
current density for longer chains (n > 12) than would have
been expected on the basis of thickness considerations alone.
These reductions in the attenuation of current (i.e., leakage
currents) were attributed to the presence of voids and
thickness inhomogeneities at grain boundaries.
Amide-Containing SAMs. Previous studies have con-
cluded that SAMs made from alkanethiols containing
secondary amides (HS(CH2)nCONHR) form well-ordered
monolayers with strong intermolecular hydrogen bonding
between neighboring amide groups throughout the back-
bone.44−48 SAMs made from amide-containing alkanethiols,
like those made from simple alkanethiols, become more
crystalline with increasing chain length.49,50 The different
physical states of amide-containing SAMs, however, occur at
different chain lengths (n ≈ 10−12) in comparison to n-
alkanethiols (n ≈ 12−14) and depend on the position of the
amide group in the molecule.49,50
Influence of Branching Chains on the Physical
Properties of SAMs. Spectroscopic and microscopic surface
characterization by Lee et al.51−53 and the study of
spiroalkanedithiols (2,2-dialkylpropanedithiols) by Carmichael
et al.,54 using bis-substituted phosphinic acids, and by us,55
using dialkyl sulfides, have shown that SAMs made from
molecules that have two alkyl chains of equal length, diverging
from a single anchoring group (symmetrical branching), form
densely packed, well-ordered SAMs, while those made from
molecules with two alkyl chains of unequal length (asym-
metrical branching) form densely packed but disordered
SAMs.
Pathway of Charge Tunneling through SAMs. Early
work on charge transport through small, nonaromatic, organic
molecules (e.g. alkanethiols,56,57 thiol-terminated norborny-
logs58,59) largely agreed on a mechanism for tunneling that was
“through-bond”, in which charge tunnels through the σ-orbital
framework (that is, largely the chain of C−C bonds) of the
molecules, with rates that depended on the dihedral angle of
EXPERIMENTAL DESIGN
■
We used secondary amides that form well-ordered SAMs (Figure 2)
as a control group and introduced changes in the geometry and
conformation of the molecules by using tertiary amides. Replacing the
hydrogen atom in secondary amides with another alkyl chain prevents
the molecules from adopting fully trans extended geometries and
eliminates the potential for hydrogen bonding between adjacent
amide groups in the SAM. This H-bonding helps order the
conformations of secondary amides.44
We prepared three other series of amides, one that has two alkyl
chains of equal length (symmetrical tertiary amides) and two that
have alkyl chains with varying lengths, to dissect the competing effects
of thickness and conformational disorder. The symmetrical tertiary
amides were designed to introduce a deviation from linear (trans-
extended) geometries while retaining the perpendicular orientation of
the chain lengths. Series I and series II amides (Figure 2) were
designed to cause the molecules in the SAM to adopt different
conformations (and thus have a higher population of gauche bonds in
comparison to the longer and better-ordered analogues). In series I
amides, the total number of carbon atoms in both chains is constant,
and in series II amides the length of the longest alkyl chain remained
constant.
EXPERIMENTAL SECTION
We, and others, have described the electrical measurements using the
EGaIn junction, which we include in the section S2 in the Supporting
■
3484
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 3. (top) Snapshots from the equilibrated MD simulation trajectories. (middle) Histograms from the MD simulations representing the
probability density function obtained from all C−C−C−C dihedral angles within a simulated SAM, which are used to compute ftrans. (bottom)
Selected, individual molecules from the MD simulations that depict branches pointing “up” and “down”, alongside the probability density functions
of the terminal methyl with respect to the z position, which are used to calculate fdown. Hydrogen atoms are excluded from the molecules for clarity.
Elements are labeled according to color as indicated on the bottom left. Examples from (A) the secondary amide control group (5(12,H)), (B) the
symmetrical tertiary amides (10(12,12)), (C) the series I amides (11(8,4)), and (D) the series II amides (16(12,6)).
Information. SAMs were prepared using standard published protocols
(see section S3 in the Supporting Information). Section S11 in the
Supporting Information also details the synthesis and characterization
of all molecules and intermediates used for the formation of SAMs. It
also includes characterization of the SAMs, including detailed
procedures for angle-resolved X-ray photoelectron spectroscopy
(ARXPS) (section S4 in the Supporting Information).
The fraction of C−C−C−C dihedral angles (θ) in the trans
orientation was defined as
N
Ntotal
trans
ftrans
=
(1)
where Ntrans is the average number of dihedral angles with θ ranging
from 135° to 225° and Ntotal is the total number of dihedral angles.
This cutoff was chosen on the basis of an inspection of the dihedral
angle probability distributions shown in the middle of Figure 3A−D,
to ensure that the entire peak associated with thermal fluctuations
about the trans-orientation minimum energy state was included in the
calculation. We also computed the N−C−C−C dihedral angle and
included the data in the Supporting Information but found that the
trends for the dihedral angles qualitatively match, and so for
simplicity, we restrict our discussion to the C−C−C−C dihedral
angle. Thus, ftrans does not account for any conformational disorder
COMPUTATIONAL DETAILS
■
To characterize the thickness, packing, and conformational structure
of the SAMs, we performed MD simulations with the OPLS-2005
force field75 implemented in LAMMPS76 and visualized with
OVITO77 (Figure 3). The complete computational methodology
and details are summarized in section S5 in the Supporting
Information.
From the equilibrated simulations, we define the height of each
molecule in the monolayer as the vertical distance between the sulfur
atom and the terminal hydrogen atom in the molecule with the
maximum z position. We then approximated the thickness of the
simulated monolayer by averaging the heights of all the molecules in
the SAM over a duration of 5 ns and the roughness of the monolayer
by using the standard deviation of the molecular height distribution
(as indicated by the error bars in the plots).
We chose to compute two statistical metrics from the simulations
to characterize the conformational disorder present in the SAM
both of which were calculated by taking ensemble averages over all of
the molecules in the equilibrated SAM at 300 K over a duration of 5
ns: (1) the fraction of C−C−C−C dihedral angles in the trans
orientation, ftrans, and (2) the fraction of chains that were not oriented
normal to the surface, i.e. pointing down or to the side, fdown (Figure
3).
when a chain is one or two carbons in length (i.e. chain b in 12(10,2)
,
13(11,1), or 14(12,2)), because there is no C−C−C−C dihedral angle.
The states of C−N−C−C and the N−C−C--C dihedral angles (and
the rest of the dihedral angles in the alkyl chain), however, couple
cooperatively to determine whether the shorter alkyl chain in the
branched molecule is oriented normal to the surface or pointing
“down”, as quantified by our alternative statistical metric fdown
.
One observation from our simulations was that, in the conforma-
tionally disordered SAMs, there were a significant fraction of
branched alkyl chains that were not oriented normal to the surface
(Figure 3B−D, bottom). Specifically, we observed that the
equilibrium probability distribution of the z position of the terminal
methyl groups, P(z), could contain a single peak (secondary amides),
two peaks (symmetrical tertiary amides), or up to three peaks (series I
and II amides). By integrating the areas of the respective peaks, we
were able to calculate the fraction of chains pointed down, fdown, i.e.
3485
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 4. (A) Plot of the measured (filled circles) and simulated (empty circles) thicknesses of SAMs made from secondary amides 1 - 5, as a
function of their chain length. The dashed line represents the corresponding data (from the literature80) for alkanethiols. (B) Plot of rectification
ratios (r+) measured at 1.0 V as a function of the conformational order (defined by ftrans) of the SAM. (C) Tunneling rates measured for the
secondary amides, as a function of their chain length, with the measured rates of alkanethiols80 for comparison. (D) Selected snapshots from the
MD simulations of SAMs 2 and 5 (1.5 nm cross sections), depicting their thickness and conformation. Hydrogen atoms have been removed for
clarity.
with a terminal CH3 group that was not at or near the farthest
possible z position in the SAM for that particular alkyl chain
must conclude that changes in packing density and/or the
conformations of the molecules in the SAM influence
tunneling. In our discussions involving thickness, we refer to
the experimentally determined thickness (as determined by
ARXPS, as opposed to the thickness obtained from the MD
simulations), unless specifically specified otherwise. We note
here that the calculation of monolayer thickness (and packing
density) from ARXPS experiments involves the assumption of
a fixed inelastic mean free path of photoelectrons (section S4
in the Supporting Information), which is a rough approx-
imation: the attenuation of photoelectrons is expected to be
directly correlated with packing density.
Packing Density. The packing density of a monolayer is
defined as the number of molecules m (in moles) per unit of
surface area (m/cm2). The tertiary amides used in this work
result in structures that are branched (two alkyl chains for each
sulfur anchoring group) and will thus have larger intermo-
lecular distances within the monolayer in comparison to
secondary amides or alkanethiols and should thus form SAMs
with lower packing densities. If changes in packing density (i.e.,
number of molecules per area) determine the rate of tunneling
by changing the number of electrical conduits, the rate of
charge tunneling should scale linearly with packing density.
That is, when only uniformly structured SAMs are considered,
a SAM with a packing density of 0.5 × 10−9 mol/cm2 should
have half the number of conduits per unit area as a SAM with a
packing density of 1.0 × 10−9 mol/cm2. In this work, we use
ARXPS (see section S4 in the Supporting Information for
details) to measure the packing densities of SAMs and employ
decanethiol (reported23 on Au as 1.0 × 10−9 mol/cm2) as a
standard.
fdown
=
z* P(z) dz
∫
(2)
0
where z* is a cutoff height, chosen by visual inspection, to include
only the bottom peak in the probability distribution (Figure 3A−D,
bottom). A similar limitation to ftrans exists for fdown: that is, for 13(11,1)
the lone CH3 group of chain b” will always have the same orientation/
conformation.
RESULTS AND DISCUSSION
■
Thickness, Packing Density, and Conformation of
SAMs Containing Amides. We expected the differences in
structure among the molecules we studied to influence three
related physical characteristics of the SAMs: thickness of the
monolayer, their packing density, and conformation of the
molecules. To determine the effect of conformation on charge
tunneling, we must also understand the influence of thickness
and packing density. Sections S4−7 in the Supporting
Information contain details of our measurements of thickness
and packing density by XPS and by MD simulations and a
discussion of these techniques in the context of this work.
Thickness. For n-alkanethiols, the rate of charge tunneling
correlates exponentially with the thickness of the SAM,
according to the simplified Simmons equation J(V) = J0(V)
e−βd 33,78,79 where d is the thickness of the SAM. In this work, if
,
the rate of tunneling correlates directly with the thickness of
the SAM, even as the conformation of the molecules change,
then charge tunneling is likely a “through-space” phenomenon
that depends primarily on the distance between the two
electrodes and the electrostatic field generated by the
molecules making up the SAM. If, however, the rate of charge
transport is not a direct function of monolayer thickness, we
Conformation. We used the two disorder parameters from
our MD simulations (described above in the Computational
3486
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 5. (A) Plot of the measured (filled squares) and simulated (empty squares) thicknesses of SAMs made from symmetrical tertiary amides 6−
10 (Figure 2), as a function of their chain length. The dashed line represents the corresponding data (from the literature80) for alkanethiols. (B)
Plot of rectification ratios (r+) measured at 1.0 V as a function of the conformational order (defined by ftrans) of the SAM. (C) Tunneling rates
measured for the symmetrical tertiary amides, as a function of their chain length, with the measured rates of secondary amides for comparison. (D)
Selected snapshots from the MD simulations of SAMs 2 and 5 (1.5 nm cross sections), depicting their thickness and conformation. Hydrogen
atoms have been removed for clarity.
Details), ftrans (the fraction of C−C−C−C dihedral angles that
are trans-extended) and fdown (the fraction of alkyl chains that
are oriented downward), to represent the conformational order
(and disorder) of the molecules making up each SAM.
Secondary Amides Are Well-Ordered and Behave
Like Alkanethiols. Our analysis of secondary amides
supports previous assertions44,46 that this class of thiols
forms well-ordered SAMs that have packing densities,
conformational structures (i.e, trans-extended geometries),
and tunneling characteristics similar to those of alkanethiols.
Figure 4A shows a linear correlation between the thickness
(determined by ARXPS) of amides 1−5 and the chain length
of the molecules (including the amide group), which suggests
that these SAMs are well-ordered. The surface coverages
(determined by XPS) were nearly indistinguishable from those
of alkanethiols (between 0.9 × 10−9 and 1.0 × 10−9 mol/cm2;
see Table S2 in the Supporting Information). This correlation
also supports the assertion that these SAMs are well-ordered.
The MD simulations, in agreement with the ARXPS results,
indicated that the thicknesses of SAMs 1−5 increase
monotonically with chain length (Figure 4A). These
simulations also revealed that their conformations are
predominantly trans extended (ftrans > 85) and are all oriented
perpendicular to the AuTS surface (i.e., fdown = 0) (see Table
S14 in the Supporting Information for the full list of values of
ftrans and fdown). The value of ftrans increased with increasing
chain length, which correlates with the well-known transition
in alkanethiol SAMs from a more liquid-like (disordered) state
to a more crystalline-like (ordered) state, as the chain length
increases.50
(Au) electrode.34 In this work, our investigation of tunneling
rates used J(V) at +0.5 V, to avoid complications associated
with rectification, which in these molecules takes place only
above ∼0.6 V (in the Fowler−Nordheim (F-N) tunneling
regime).34 We observed that the rectification ratio (r+) was
larger in SAMs with longer chain lengths, presumably because
the net dipole perpendicular to the mean plane of the metal
surface (which seems, in this series, to determine rectifica-
tion34) is larger when the SAMs are more crystalline-like in
comparison to that when the SAMs are more liquid-like. This
assertion is supported by an approximately linear relationship
between the measured rectification (r+) and ftrans shown in
Figure 4B. This observation was important, because it
suggested that we might be able to use rectification as an
indirect experimental metric to compare to the conformational
order calculated from the MD simulations.
The measured current densities (log |J|) for the linear
amides (on a AuTS surface, at an applied bias of +0.5 V) show
that the rate of tunneling decreases exponentially with
increasing chain length, as it does with alkanethiols (Figure
4C). Within the limits of our measurement, introducing an
amide bond at the third carbon from the sulfur atom (i.e., S-
(CH2)2-CONH-R) does not significantly change (relative to
those data obtained with alkanethiols) the value of β (β = 0.87
0.03) or the value of log |J0| (log |J0| = 3.0
0.3).80
Alkanethiols containing amide groups have previously been
shown to have no significant effect on the values of log |J|
measured at +0.5 V, relative to alkanethiols without amide
groups,81 and if amide-containing alkanethiols form SAMs with
the same trans-extended, linear orientation of the n-alkyl group
as alkanethiols, the similarity between their β plots is
qualitatively expected.
All of the linear amides rectified current at an applied bias of
1.0 V (r+ > 5), possibly because of the proximity and direction
of the embedded dipole (from the amide group) to the bottom
3487
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Symmetrical Tertiary Amides. SAMs made from sym-
metrically branched, tertiary amides (Figure 2) have structures
different from those of secondary amides and alkanethiols. The
surface coverages measured by XPS (Table S3 in the
Supporting Information) for the symmetrical tertiary amides
were between 0.74 × 10−9 and 0.84 × 10−9 mol/cm2, which is
∼16−26% lower than those for secondary amides ((0.90−
1.00) × 10−9 mol/cm2; see Table S1 in the Supporting
Information). We attribute this difference in packing density to
the increased space occupied by the additional (branching)
alkane chain. Figure 5A shows that, for symmetrical tertiary
amides 6−10 (where R1 = R2), the thickness of the SAM
increases with increasing chain length. This relationship,
however, is not entirely linear (as is observed for linear amides
or alkanethiols): that is, Figure 5A shows a slight discontinuity
between 7(6,6) and 8(8,8). We hypothesize that this divergence
from linearity may be a consequence of the transition from
SAMs that are liquid crystalline like, to more crystalline like
between compounds 7(6,6) and 8(8,8)when the enthalpy
(dominated by intermolecular van der Waals forces) begins to
overcome the entropy of the system.50
through secondary amides. For instance, the rate of CT in 6(4,4)
is 6.3 times lower than that through 1(4,H), despite having the
same thickness (measured by ARXPS) and the same chain
length. Likewise, CT through 7(6,6) is 2.6 times lower than that
through 2(6,H), which has the same chain length and thickness
(determined by XPS). The rate of tunneling is also lower
through these symmetrical tertiary amides than that through
secondary amides when log |J(V)| is plotted as a function of the
ARXPS measured monolayer thickness (Figure 6), suggesting
The MD simulations of symmetrically branched amides
reveal that, as in the case for linear amides, the values of ftrans
increase with the length of the alkyl chain (Table S14 in the
Supporting Information). The same trend is observed for the
rectification ratio at 1.0 V (Table S8 in the Supporting
Information), which is also larger for symmetrical tertiary
amides with longer chain lengths. Figure 5B shows a positive
correlation between r+ and ftrans, further supporting the
relationship among conformational order, chain length (i.e.,
SAM crystallinity), and rectification ratio. The values of ftrans
for the symmetrically branched amides (between 0.70 and
0.88, Table S14 in the Supporting Information) are lower than
those for linear amides (0.87−0.99, Table S14 in the
Supporting Information). We attribute the increase in gauche
conformations to the geometrical constraints imposed by the
tertiary amide, which prevents an all-trans conformation and
forces the alkyl chains to adopt a “kink” in their geometry after
the point of branching. We note that symmetrical tertiary
amides have smaller values of ftrans in comparison to linear
amides but have comparable values of r+ (see plots in Figures
4B and 5B). Our interpretation of this result is that r+ is only
indirectly correlated with conformational disorder and that
ftrans is an indirect (and imperfect) metric for disorder. Thus,
while these two metrics are correlated, the absolute value of
ftrans (derived from simulation) does not determine the
absolute value of r+. In addition to a decrease in ftrans, relative
to secondary amides, symmetrical tertiary amides have values
of fdown > 0 (Table S14 in the Supporting Information),
suggesting that there is significantly more conformational
disorder in SAMs formed from symmetrical tertiary amides in
comparison to SAMs formed from linear amides. Moreover,
Figure 6. Tunneling rates measured for the linear amides (empty
circles) and symmetrically branched amides (solid squares) as a
function of their experimentally determined (by XPS) monolayer
thickness.
that tunneling rates through tertiary amides are lower than
those through secondary amides of equivalent chain length,
even when they form monolayers of equivalent (or smaller)
thickness.
By excluding a variation in thickness and packing density as
the origin of this difference in rate of tunneling between
secondary and symmetrical tertiary amides, we concludeby
eliminationthat differences in conformation influence the
rates of charge tunneling. Because the differences in
conformation between secondary amides and symmetrical
tertiary amides is a combination of gauche defects (defined by
f
trans) and different chain orientations (defined by fdown), we
conclude that charge tunneling is more rapid through alkyl
chains with trans-extended geometries in comparison to those
with gauche defects and assert that, in these systems, the
conformation does influence the rate of charge tunneling. This
conclusion is consistent with previous measurements of
temperature-dependent electrical transport through linear
alkyl monolayers on silicon surfaces, where an inverse
relationship between tunneling rate and temperature was
attributed to cooling-induced ordering of the monolayer (and
thus increased intramolecular electronic coupling) on the basis
of vibrational and photoelectron spectroscopy.82
the same trend is observed for fdown as was observed for ftrans
:
that is, symmetrical tertiary amides with longer chain lengths
have more conformational order in comparison to those with
shorter chain lengths.
Figure 5C shows that the rate of CT through symmetrical
tertiary amides decreases exponentially as a function of chain
length, in a manner similar to that observed with alkanethiols
and secondary amides.76 A direct comparison between
symmetric tertiary amides and secondary amides with
equivalent chain lengths, however, indicates that the rate of
CT through symmetrical tertiary amides is lower than that
Series I Amides (Figure 2; X(n,m), n + m = 12). The set of
amides in series I was designed to test whether the total
number of carbons (regardless of their position in the chain or
the conformation of individual bonds) determines the rate of
tunneling. This series includes the secondary amide 5(12,H) and
the symmetrical tertiary amide 7(6,6). In principle, tunneling
might be determined (regardless of mechanism) by the sum of
3488
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 7. (A) Plot of the measured (filled triangles) and simulated (empty triangles) thicknesses of SAMs made from asymmetrically branched
amides 5, 7, and 11−13, as a function of their longest chain length. The dashed line represents the corresponding data (from the literature80) for
alkanethiols. (B) Tunneling rates measured for the asymmetrically branched amides (disordered series I) as a function of their chain length, with
the measured rates of linear amides for comparison. (C) Tunneling rates measured for the asymmetrically branched amides (disordered series I) as
a function of their measured thickness. (D) Selected snapshots from the MD simulated SAMs (1.5 nm cross sections), depicting their thickness and
conformation. Hydrogen atoms have been removed for clarity.
Figure 8. (A) Plot of the measured (filled diamonds) and simulated (empty diamonds) thicknesses of SAMs made from asymmetrically branched
amides 5, 10, and 14−18, as a function of their longest chain length. (B) Tunneling rates measured for the asymmetrically branched amides
(disordered series II) as a function of their chain length, with the measured rates of linear amides for comparison. (C) Tunneling rates measured for
the asymmetrically branched amides (disordered series II) as a function of their measured thickness. (D) Selected snapshots from the MD
simulated SAMs (1.5 nm cross sections), depicting their thickness and conformation. Hydrogen atoms have been removed for clarity.
the carbon atoms in the monolayer per unit area. Our results
(Figure 7B), however, show that asymmetry in branching
within SAMsnot just the total number of carbon atoms
leads to differences in tunneling rates, despite the fact that the
SAMs have the same number and type of atoms per unit area.
ARXPS measurements indicated that the asymmetrical
tertiary amides within series I (i.e., 11(8,4), 12(10,2), and
13(11,1)) had lower packing densities in comparison to
secondary amides but larger packing densities in comparison
to symmetrical tertiary amides. The thickness (measured by
ARXPS) of these asymmetrical tertiary amides increased as the
length of chain a increased, suggesting that the longest chain
length dominates the thickness of these SAMs.
3489
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Our simulations suggest that, like the symmetrical tertiary
amides, the presence of a second alkyl chain causes a kink in
the conformation, which leads to more gauche conformations
than in linear amides (Table S14 in the Supporting
Information). In agreement with the observed trends in
rectification ratios, all of the asymmetrically branched amides
had r+ < 7.0. This result indicated that the amides are less
ordered in their orientation with respect to the surface of the
gold in comparison to linear amides and thus that these SAMs
of tertiary amides are probably in more disordered
conformations. This conclusion was supported by our MD
simulations, particularly for SAMs made from amide 11(8,4)
(ftrans = 0.78, fdown = 0.58) and SAMs made from amide 12(10,2)
(ftrans = 0.91, fdown = 0.35), which indicate substantial
conformational disorder.
Unlike the secondary amides and symmetrical tertiary
amides, the current density measurements for series I amides
did not correlate linearly with chain length (Figure 7A, R2 =
0.65). A positive correlation was observed, however, between
the current density and the thickness (determined by ARXPS)
of the SAMs (Figure 7C, R2 = 0.83). While this result implies
that the rate of tunneling through this set of SAMseven
when they have disordered conformationsstill correlates
with the thickness of the monolayer, the value of β (i.e., the
slope) is larger for series I amides than for secondary amides.
This difference in β indicates that the relationship between
thickness and J(V) may be different for ordered than for
disordered SAMs.
Series II Amides (Figure 2; X(n,m), n = 12). Series II
amides, which include the asymmetrical tertiary amides 14−
18, the secondary amide 5(12,H) and the symmetrical tertiary
amide 10(12,12), was designed to test how the asymmetry in
branching (which we believe correlates with conformational
disorder) influences tunneling. Measurements of packing
density by XPS indicated that, like series I amides, the density
of asymmetrical tertiary amides (i.e., 14−18) was greater than
that of symmetrical tertiary amides but less than that of
secondary amides. This result suggests that asymmetrical
branching forces the longer chain to adopt a bent
conformation, in order to maximize their intermolecular
contacts and reduce the volume they occupy on the surface.
The thickness of the monolayers (as measured by ARXPS)
across the series remained approximately the same (Figure
8A), comparable to the case for the linear amide 5(12,H). A
small decrease in thickness was observed as the length of chain
b decreased, but chain awith its greater number of carbon
atomsappeared to dominate the thickness of the monolayer.
The simulations also show a relatively constant thickness of the
monolayer for this series of amides (Figure 8A).
|J(V)| against monolayer thickness in Figure 8C. Unlike the
case with series I amides, we observed no linear relationship
between log |J(V)| and monolayer thickness for series II
amides, indicating that in this series the rate of charge
tunneling does not correlate directly with thickness. Moreover,
the discrepancy between J(V) and thickness of 14−18, relative
to 10(12,12) or 5(12,H), cannot be accounted for by changes in
packing density within the SAM (as 5 has a higher packing
density than any of 14−18 and 10 has a lower packing density
than any of 14−18). We thus conclude, again by elimination,
that the conformation of these SAMs must be influencing the
rate of tunneling and that conformation does influence
tunneling rates in molecular junctions.
Unlike symmetrical tertiary amides, the unsymmetrical
tertiary amides in series II have tunneling rates that are higher
than those of secondary amides (or alkanethiols) of equivalent
thickness (or chain length). We believe that this result is a
consequence of the asymmetry in the lengths of chains a and b,
which causes more extensive disorder in asymmetrical tertiary
amides than in symmetric tertiary amides. This difference in
disorder could result in a poorly defined interface between the
top EGaIn electrode and the SAM and may cause electrical
contacts to be made farther down the chain than the ARXPS-
determined thickness would suggest.
CONCLUSION
■
By analyzing molecules that are similar electronically but that
differ in their conformation, we studied the effects of
conformation, thickness, and packing density of molecules in
a SAM on the rate of charge transport (CT) through them. We
observe differences in tunneling characteristics that cannot be
accounted for by changes in thickness or packing density and
conclude by elimination that charge tunneling through SAMs is
influenced by the conformations of the molecules and is not
solely determined by the distance between the two electrodes.
By comparing symmetric tertiary amides to secondary
amides, we conclude that the rate of CT is higher through
alkane chains with trans conformations than through those
with gauche conformations. By analyzing Series I amides, which
have the same number and type of atoms but are distributed
differently among chains a and b, we conclude that the spatial
arrangement of atoms matters. That is, tunneling is not solely
dependent on the density of atoms of a particular type within
the monolayer. These results suggest that CT may be a
through-bond process.
We conclude from our analysis of series II amides that the
conformation (or conformational disorder) influences the rates
of CT. This series revealed that a change in thickness does not
always correlate with an equivalent change in J(V), even for
simple alkanethiols. We believe that the results of this series of
amides indicate that conformational disorder can create a
poorly defined electrode−molecule interface, which can
artificially increase J(V).
Finally, we observed a clear relationship between conforma-
tional order and the dipole-induced rectification in these
systems. Specifically, SAMs that were more conformationally
ordered yielded molecular junctions with larger rectification
ratios in comparison to more conformationally disordered
SAMs. This result strongly suggests that conformation, and not
solely molecular structure, should be considered in the design
and analysis of molecular junctions.
The rectification ratios at 1.0 V for amides (r+ < 7.5; Table
S12 in the Supporting Information) are all lower than those for
the well-ordered SAMs in the series, 5(12,H) (r+ = 10.42) and
10(12,12) (r+ = 11.22), which suggests that amides 14−18 form
disordered SAMs. This interpretation was supported by the
MD simulations (Table S14 in the Supporting Information),
which show a significant number of gauche conformations
(ftrans ranges from 0.82 to 0.93) and varied chain orientations
(fdown ranges from 0.15 to 0.6).
The tunneling rates of series II amides (Figure 8B) are all
higher than that of a corresponding secondary amide of length
n = 16 (i.e, 5(12,H)). From the results of series I amides,
however, we did not expect that the rates of tunneling would
necessarily correlate with chain length, and we thus plotted log
3490
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
National Science Foundation (ECS-0335765). This work used
the Extreme Science and Engineering Discovery Environment
(XSEDE) Comet supercomputer at the San Diego Super-
computer Center through allocation DMR180105, which is
supported by National Science Foundation grant number ACI-
1548562.83 H.J.Y. acknowledges support from the NRF of
K o r e a ( N R F - 2 0 1 9 R 1 A 2 C 2 0 1 1 0 0 3 , N R F -
2019R1A6A1A11044070) and the Future Research Grant
(FRG) of Korea University.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c12571.
■
sı
Materials, junction measurements using “selected”
conical tips of EGaIn, EGaIn measurement protocol,
angular-dependent X-ray photoelectron spectroscopy,
computational methods, comparison between exper-
imental and simulated monolayer thickness, summary of
XPS data, summary of tunneling data, summary of
simulation data, syntheses, and contact angle data
(PDF)
REFERENCES
■
(1) Venkataraman, L.; Klare, J. E.; Nuckolls, C.; Hybertsen, M. S.;
Steigerwald, M. L. Dependence of single-molecule junction
conductance on molecular conformation. Nature 2006, 442, 904−
907.
AUTHOR INFORMATION
Corresponding Author
George M. Whitesides − Department of Chemistry and
Chemical Biology, Harvard University, Cambridge,
Massachusetts 02138, United States; orcid.org/0000-
0001-9451-2442; Email: gwhitesides@
gmwgroup.harvard.edu
■
(2) Kong, G. D.; Jin, J.; Thuo, M.; Song, H.; Joung, J. F.; Park, S.;
Yoon, H. J. Elucidating the role of molecule−electrode interfacial
defects in charge tunneling characteristics of large-area junctions. J.
Am. Chem. Soc. 2018, 140, 12303−12307.
(3) Kong, G. D.; Byeon, S. E.; Park, S.; Song, H.; Kim, S.-Y.; Yoon,
H. J. Mixed molecular electronics: tunneling behaviors and
applications of mixed self-assembled monolayers. Adv. Electron.
Mater. 2020, 6, 1901157−1901176.
Authors
(4) Chen, J.; Chang, B.; Oyola-Reynoso, S.; Wang, Z.; Thuo, M.
Quantifying gauche defects and phase evolution in self-assembled
monolayers through sessile drops. ACS Omega 2017, 2, 2072−2084.
(5) Justeena, A. N.; Nirmal, D.; Gracia, D. In Design and analysis of
tunnel FET using high K dielectric materials. International Conference
on Innovations in Electrical, Electronics, Instrumentation and Media
Technology (ICEEIMT) 2017, 2017, 177.
(6) Anghel, C.; Chilagani, P.; Amara, A.; Vladimirescu, A. Tunnel
field effect transistor with increased ON current, low-k spacer and
high-k dielectric. Appl. Phys. Lett. 2010, 96, 122104−122104−3.
(7) Editorial. Molecular electronics under the microscope. Nat.
Chem. 2015, 7, 181.
Lee Belding − Department of Chemistry and Chemical
Biology, Harvard University, Cambridge, Massachusetts
02138, United States
Samuel E. Root − Department of Chemistry and Chemical
Biology, Harvard University, Cambridge, Massachusetts
02138, United States
Yuan Li − Department of Chemistry and Chemical Biology,
Harvard University, Cambridge, Massachusetts 02138,
United States
Junwoo Park − Department of Chemistry and Chemical
Biology, Harvard University, Cambridge, Massachusetts
02138, United States
Mostafa Baghbanzadeh − Department of Chemistry and
Chemical Biology, Harvard University, Cambridge,
Massachusetts 02138, United States; orcid.org/0000-
0001-7678-1681
Edwin Rojas − Department of Chemistry and Chemical
Biology, Harvard University, Cambridge, Massachusetts
02138, United States
Priscilla F. Pieters − Department of Chemistry and Chemical
Biology, Harvard University, Cambridge, Massachusetts
02138, United States
(8) Martínez Hardigree, J. F.; Dawidczyk, T. J.; Ireland, R. M.;
̈
́
Johns, G. L.; Jung, B.-J.; Nyman, M.; Osterbacka, R.; Markovic, N.;
Katz, H. E. Reducing leakage currents in n-channel organic field-effect
transistors using molecular dipole monolayers on nanoscale oxides.
ACS Appl. Mater. Interfaces 2013, 5, 7025−7032.
(9) Klauk, H.; Zschieschang, U.; Pflaum, J.; Halik, M. Ultralow-
power organic complementary circuits. Nature 2007, 445, 745.
̈
(10) Szent-Gyorgyi, A. Towards a new biochemistry? Science 1941,
93, 609−611.
(11) Gray, H. B.; Halpern, J. Distant charge transport. Proc. Natl.
Acad. Sci. U. S. A. 2005, 102, 3533−3534.
(12) Boal, A. K.; Yavin, E.; Lukianova, O. A.; O'Shea, V. L.; David, S.
S.; Barton, J. K. DNA-Bound Redox Activity of DNA Repair
Glycosylases Containing [4Fe-4S] Clusters. Biochemistry 2005, 44,
8397−8407.
Hyo Jae Yoon − Department of Chemistry, Korea University,
Seoul 02841, Korea
(13) Yavin, E.; Boal, A. K.; Stemp, E. D. A.; Boon, E. M.; Livingston,
A. L.; Shea, V. L.; David, S. S.; Barton, J. K. Protein−DNA charge
transport: Redox activation of a DNA repair protein by guanine
radical. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 3546−3551.
(14) Sjulstok, E.; Olsen, J. M. H.; Solov’yov, I. A. Quantifying
electron transfer reactions in biological systems: what interactions
play the major role? Sci. Rep. 2016, 5, 18446−18457.
(15) Bothe, H.; Schmitz, O.; Yates, M. G.; Newton, W. E. Nitrogen
fixation and hydrogen metabolism in cyanobacteria. Microbiol. Mol.
Biol. Rev. 2010, 74, 529−551.
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c12571
Author Contributions
§L.B. and S.E.R. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(16) Mishchenko, A.; Vonlanthen, D.; Meded, V.; Burkle, M.; Li, C.;
̈
This work was supported by NSF award CHE-1808361. L.B.
acknowledges fellowship support from NSERC Canada. E.R.
acknowledges the Harvard REU program under NSF award
DMR-1420570. Sample characterization was performed in part
at the Center for Nanoscale Systems (CNS) at Harvard
University, a member of the National Nanotechnology
Infrastructure Network (NNIN), which is supported by the
Pobelov, I. V.; Bagrets, A.; Viljas, J. K.; Pauly, F.; Evers, F.; Mayor, M.;
Wandlowski, T. Influence of conformation on conductance of
biphenyl-dithiol single-molecule contacts. Nano Lett. 2010, 10,
156−163.
̂
(17) Xie, Z.; Baldea, I.; Frisbie, C. D. Determination of energy-level
alignment in molecular tunnel junctions by transport and spectros-
copy: self-consistency for the case of oligophenylene thiols and
3491
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
dithiols on Ag, Au, and Pt electrodes. J. Am. Chem. Soc. 2019, 141,
3670−3681.
(18) Li, G.; Xie, J.; Wang, J.; Xia, L.; Li, Y.; Hu, W. Nanoscale
surface disorder for enhanced solar absorption and superior visible-
light photocatalytic property in Ti-rich BaTiO3 nanocrystals. ACS
Omega 2019, 4, 9673−9679.
junctions: 2,2′-bipyridyl-terminated n-alkanethiolates. J. Am. Chem.
Soc. 2014, 136, 17155−17162.
(37) Nijhuis, C. A.; Reus, W. F.; Whitesides, G. M. Mechanism of
rectification in tunneling junctions based on molecules with
asymmetric potential drops. J. Am. Chem. Soc. 2010, 132, 18386−
18401.
(19) Haberhauer, G.; Gleiter, R.; Burkhart, C. Planarized Intra-
molecular Charge Transfer: A Concept for Fluorophores with both
Large Stokes Shifts and High Fluorescence Quantum Yields. Chem. -
Eur. J. 2016, 22, 971−978.
(38) Cho, S. J.; Kong, G. D.; Park, S.; Park, J.; Byeon, S. E.; Kim, T.;
Yoon, H. J. Molecularly controlled stark effect induces significant
rectification in polycyclic-aromatic-hydrocarbon-terminated n-alkane-
thiolates. Nano Lett. 2019, 19, 545−553.
(20) Belding, L.; Guest, M.; Le Sueur, R.; Dudding, T. Fluorescence
of Cyclopropenium Ion Derivatives. J. Org. Chem. 2018, 83, 6489−
6497.
(39) Hegner, M.; Wagner, P.; Semenza, G. Ultralarge atomically flat
template-stripped Au surfaces for scanning probe microscopy. Surf.
Sci. 1993, 291, 39−46.
(21) Nikitina, V. A.; Rudnev, A. V.; Nazmutdinov, R. R.; Tsirlina, G.
A.; Wandlowski, T. Solvent effect on electron transfer through
alkanethiols. J. Elect. Chem. 2018, 819, 58−64.
(40) Jiang, L.; Sangeeth, C. S. S.; Yuan, L.; Thompson, D.; Nijhuis,
C. A. One-nanometer thin monolayers remove the deleterious effect
of substrate defects in molecular tunnel junctions. Nano Lett. 2015,
15, 6643−6649.
́
(22) Juarez, E. G.; Mena-Cervantes, V. Y.; Vazquez-Arenas, J.;
̈
Flores, G. P.; Hernandez-Altamirano, R. Inhibition of CO2 corrosion
via sustainable geminal zwitterionic compounds: effect of the length
of the hydrocarbon chain from amines. ACS Sustainable Chem. Eng.
2018, 6, 17230−17238.
(41) Novak, M.; Jager, C. M.; Rumpel, A.; Kropp, H.; Peukert, W.;
Clark, T.; Halik, M. The morphology of integrated self-assembled
monolayers and their impact on devices − A computational and
experimental approach. Org. Electron. 2010, 11, 1476.
(23) Vilan, A.; Aswal, D.; Cahen, D. Large-area, ensemble molecular
electronics: motivation and challenges. Chem. Rev. 2017, 117, 4248−
4286.
(24) Liu, D.; Miao, Q. Recent progress in interface engineering of
organic thin film transistors with self-assembled monolayers. Mater.
Chem. Front. 2018, 2, 11−21.
(42) Levine, I.; Weber, S. M.; Feldman, Y.; Bendikov, T.; Cohen, H.;
Cahen, D.; Vilan, A. Molecular Length, Monolayer Density, and
Charge Transport: Lessons from Al−AlOx/Alkyl−Phosphonate/Hg
Junctions. Langmuir 2012, 28, 404.
(43) Nishi, N.; Hobara, D.; Yamamoto, M.; Kakiuchi, T. Chain-
length-dependent change in the structure of self-assembled mono-
layers of n-alkanethiols on Au(111) probed by broad-bandwidth sum
frequency generation spectroscopy. J. Chem. Phys. 2003, 118, 1904−
1911.
(44) Lewis, P. A.; Smith, R. K.; Kelly, K. F.; Bumm, L. A.; Reed, S.
M.; Clegg, R. S.; Gunderson, J. D.; Hutchison, J. E.; Weiss, P. S. The
Role of Buried Hydrogen Bonds in Self-Assembled Mixed
Composition Thiols on Au{111}. J. Phys. Chem. B 2001, 105,
10630−10636.
(45) Smith, R. K.; Reed, S. M.; Lewis, P. A.; Monnell, J. D.; Clegg, R.
S.; Kelly, K. F.; Bumm, L. A.; Hutchison, J. E.; Weiss, P. S. Phase
separation within a binary self-assembled monolayer on Au{111}
driven by an amide-containing alkanethiol. J. Phys. Chem. B 2001, 105,
1119−1122.
(46) Thomas, J. C.; Goronzy, D. P.; Dragomiretskiy, K.; Zosso, D.;
Gilles, J.; Osher, S. J.; Bertozzi, A. L.; Weiss, P. S. Mapping buried
hydrogen-bonding networks. ACS Nano 2016, 10, 5446−5451.
(47) Sek, S.; Palys, B.; Bilewicz, R. Contribution of intermolecular
interactions to electron transfer through monolayers of alkanethiols
containing amide groups. J. Phys. Chem. B 2002, 106, 5907−5914.
(48) Angelova, P. N.; Hinrichs, K.; Philipova, I. L.; Kostova, K. V.;
Tsankov, D. T. Monolayer orientation of ω-substituted amide-bridged
alkanethiols on gold. J. Phys. Chem. C 2010, 114, 1253−1259.
(49) Clegg, R. S.; Hutchison, J. E. Control of Monolayer Assembly
Structure by Hydrogen Bonding Rather Than by Adsorbate−
Substrate Templating. J. Am. Chem. Soc. 1999, 121, 5319−5327.
(50) Clegg, R. S.; Reed, S. M.; Smith, R. K.; Barron, B. L.; Rear, J. A.;
Hutchison, J. E. The interplay of lateral and tiered interactions in
stratified self-organized molecular assemblies. Langmuir 1999, 15,
8876−8883.
(25) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.;
Whitesides, G. M. Self-assembled monolayers of thiolates on metals as
a form of nanotechnology. Chem. Rev. 2005, 105, 1103−1170.
(26) Jeong, H.; Kim, D.; Xiang, D.; Lee, T. High-yield functional
molecular electronic devices. ACS Nano 2017, 11, 6511−6548.
(27) Lindsey, J. S.; Bocian, D. F. Molecules for charge-based
information storage. Acc. Chem. Res. 2011, 44, 638−650.
(28) Beebe, J. M.; Engelkes, V. B.; Miller, L. L.; Frisbie, C. D.
Contact resistance in metal−molecule−metal junctions based on
aliphatic sams: effects of surface linker and metal work function. J. Am.
Chem. Soc. 2002, 124, 11268−11269.
(29) Bowers, C. M.; Liao, K.-C.; Yoon, H. J.; Rappoport, D.;
Baghbanzadeh, M.; Simeone, F. C.; Whitesides, G. M. Introducing
ionic and/or hydrogen bonds into the SAM//Ga2O3 top-interface of
AgTS/S(CH2)nT//Ga2O3/EGaIn junctions. Nano Lett. 2014, 14,
3521−3526.
(30) Liao, K.-C.; Yoon, H. J.; Bowers, C. M.; Simeone, F. C.;
Whitesides, G. M. Replacing AgTSSCH2-R with AgTSO2C-R in EGaIn-
based tunneling junctions does not significantly change rates of charge
transport. Angew. Chem., Int. Ed. 2014, 53, 3889−3893.
(31) Yoon, H. J.; Shapiro, N. D.; Park, K. M.; Thuo, M. M.; Soh, S.;
Whitesides, G. M. The rate of charge tunneling through self-
assembled monolayers is insensitive to many functional group
substitutions. Angew. Chem., Int. Ed. 2012, 51, 4658−4661.
(32) Cafferty, B. J.; Yuan, L.; Baghbanzadeh, M.; Rappoport, D.;
Beyzavi, M. H.; Whitesides, G. M. Charge transport through self-
assembled monolayers of monoterpenoids. Angew. Chem., Int. Ed.
2019, 58, 8097−8102.
(33) Yoon, H. J.; Shapiro, N. D.; Park, K. M.; Thuo, M. M.; Soh, S.;
Whitesides, G. M. The rate of charge tunneling through self-
assembled monolayers is insensitive to many functional group
substitutions. Angew. Chem., Int. Ed. 2012, 51, 4658−4661.
(34) Baghbanzadeh, M.; Belding, L.; Yuan, L.; Park, J.; Al-Sayah, M.
H.; Bowers, C. M.; Whitesides, G. M. Dipole-induced rectification
across AgTS/SAM//Ga2O3/EGaIn junctions. J. Am. Chem. Soc. 2019,
141, 8969−8980.
(35) Lenfant, S.; Krzeminski, C.; Delerue, C.; Allan, G.; Vuillaume,
D. Molecular rectifying diodes from self-assembly on silicon. Nano
Lett. 2003, 3, 741−746.
(36) Yoon, H. J.; Liao, K.-C.; Lockett, M. R.; Kwok, S. W.;
Baghbanzadeh, M.; Whitesides, G. M. Rectification in tunneling
(51) Shon, Y.-S.; Lee, T. R. Chelating self-assembled monolayers on
gold generated from spiroalkanedithiols. Langmuir 1999, 15, 1136−
1140.
(52) Shon, Y.-S.; Lee, S.; Perry, S. S.; Lee, T. R. The adsorption of
unsymmetrical spiroalkanedithiols onto gold affords multi-component
interfaces that are homogeneously mixed at the molecular level. J. Am.
Chem. Soc. 2000, 122, 1278−1283.
(53) Park, J.-S.; Smith, A. C.; Lee, T. R. Loosely packed self-
assembled monolayers on gold generated from 2-Alkyl-2-methyl-
propane-1,3-dithiols. Langmuir 2004, 20, 5829−5839.
(54) San Juan, R. R.; Carmichael, T. B. Formation of self-assembled
monolayers with homogeneously mixed, loosely packed alkyl groups
3492
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
using unsymmetrical dialkyldithiophosphinic acids. Langmuir 2012,
28, 17701−17708.
topography of the bottom electrodes. Angew. Chem., Int. Ed. 2014, 53,
3377−3381.
(73) Weiss, E. A.; Chiechi, R. C.; Kaufman, G. K.; Kriebel, J. K.; Li,
Z.; Duati, M.; Rampi, M. A.; Whitesides, G. M. Influence of defects on
the electrical characteristics of mercury-drop junctions: self-assembled
monolayers of n-alkanethiolates on rough and smooth silver. J. Am.
Chem. Soc. 2007, 129, 4336−4349.
(74) Kovalchuk, A.; Abu-Husein, T.; Fracasso, D.; Egger, D. A.;
Zojer, E.; Zharnikov, M.; Terfort, A.; Chiechi, R. C. Transition
voltages respond to synthetic reorientation of embedded dipoles in
self-assembled monolayers. Chem. Sci. 2016, 7, 781−787.
(75) Banks, J. L.; Beard, H. S.; Cao, Y.; Cho, A. E.; Damm, W.;
Farid, R.; Felts, A. K.; Halgren, T. A.; Mainz, D. T.; Maple, J. R.;
Murphy, R.; Philipp, D. M.; Repasky, M. P.; Zhang, L. Y.; Berne, B. J.;
Friesner, R. A.; Gallicchio, E.; Levy, R. M. Integrated modeling
program, applied chemical theory (IMPACT). J. Comput. Chem.
2005, 26, 1752−1780.
(76) Plimpton, S. Fast parallel algorithms for short-range molecular
dynamics. J. Comput. Phys. 1995, 117, 1−19.
(77) Stukowski, A. Visualization and analysis of atomistic simulation
data with OVITO−the Open Visualization Tool. Modell. Simul. Mater.
Sci. Eng. 2010, 18, 015012−015033.
(78) Simeone, F. C.; Yoon, H. J.; Thuo, M. M.; Barber, J. R.; Smith,
B.; Whitesides, G. M. Defining the value of injection current and
effective electrical contact area for EGaIn-based molecular tunneling
junctions. J. Am. Chem. Soc. 2013, 135, 18131−18144.
(79) Simmons, J. G. Electric tunnel effect between dissimilar
electrodes separated by a thin insulating film. J. Appl. Phys. 1963, 34,
2581−2587.
(80) Baghbanzadeh, M.; Simeone, F. C.; Bowers, C. M.; Liao, K.-C.;
Thuo, M.; Baghbanzadeh, M.; Miller, M. S.; Carmichael, T. B.;
Whitesides, G. M. Odd−even effects in charge transport across n-
alkanethiolate-based SAMs. J. Am. Chem. Soc. 2014, 136, 16919−
16925.
(81) Thuo, M. M.; Reus, W. F.; Simeone, F. C.; Kim, C.; Schulz, M.
D.; Yoon, H. J.; Whitesides, G. M. Replacing − CH2CH2− with −
CONH− does not significantly change rates of charge transport
through AgTS-SAM//Ga2O3/EGaIn junctions. J. Am. Chem. Soc. 2012,
134, 10876−10884.
(82) Shpaisman, H.; Seitz, O.; Yaffe, O.; Roodenko, K.; Scheres, L.;
Zuilhof, H.; Chabal, Y. J.; Sueyoshi, T.; Kera, S.; Ueno, N.; Vilan, A.;
Cahen, D. Structure Matters: Correlating temperature dependent
electrical transport through alkyl monolayers with vibrational and
photoelectron spectoscopies. Chemical Science 2012, 3, 851−862.
(83) Towns, J.; Cockerill, T.; Dahan, M.; Foster, I.; Gaither, K.;
Grimshaw, A.; Hazlewood, V.; Lathrop, S.; Lifka, D.; Peterson, G. D.;
Roskies, R.; Scott, J. R.; Wilkins-Diehr, N. XSEDE: Accelerating
Scientific Discovery. Comput. Sci. Eng. 2014, 16, 62−74.
(55) Troughton, E. B.; Bain, C. D.; Whitesides, G. M.; Nuzzo, R. G.;
Allara, D. L.; Porter, M. D. Monolayer films prepared by the
spontaneous self-assembly of symmetrical and unsymmetrical dialkyl
sulfides from solution onto gold substrates: structure, properties, and
reactivity of constituent functional groups. Langmuir 1988, 4, 365−
385.
(56) Slowinski, K.; Chamberlain, R. V.; Miller, C. J.; Majda, M.
Through-bond and chain-to-chain coupling. two pathways in electron
tunneling through liquid alkanethiol monolayers on mercury electro-
des. J. Am. Chem. Soc. 1997, 119, 11910−11919.
(57) Slowinski, K.; Fong, H. K. Y.; Majda, M. Mercury−mercury
tunneling junctions. 1. electron tunneling across symmetric and
asymmetric alkanethiolate bilayers. J. Am. Chem. Soc. 1999, 121,
7257−72261.
(58) Beebe, J. M.; Engelkes, V. B.; Liu, J.; Gooding, J. J.; Eggers, P.
K.; Jun, Y.; Zhu, X.; Paddon-Row, M. N.; Frisbie, C. D. Length
dependence of charge transport in nanoscopic molecular junctions
incorporating a series of rigid thiol-terminated norbornylogs. J. Phys.
Chem. B 2005, 109, 5207−5216.
(59) Yang, W. R.; Jones, M. W.; Li, X.; Eggers, P. K.; Tao, N.;
Gooding, J. J.; Paddon-Row, M. N. Single molecule conductance
through rigid norbornylogous bridges with zero average curvature. J.
Phys. Chem. C 2008, 112, 9072−9080.
(60) Haran, A.; Waldeck, D. H.; Naaman, R.; Moons, E.; Cahen, D.
The dependence of electron transfer efficiency on the conformational
order in organic monolayers. Science 1994, 263, 948−950.
(61) Wang, Y.-F.; Yang, G.; Liu, C.-B. Electron transfers in proteins:
Investigations with a modified through-bond coupling model.
Phys.Rev. E 2009, 80, 021927−021927−8.
(62) Berstis, L.; Beckham, G. T.; Crowley, M. F. Electronic coupling
through natural amino acids. J. Chem. Phys. 2015, 143, 225102−
225102−10.
(63) Salomon, A.; Cahen, D.; Lindsay, S.; Tomfohr, J.; Engelkes, V.
B.; Frisbie, C. D. Comparison of electronic transport measurements
on organic molecules. Adv. Mater. 2003, 15, 1881−1890.
(64) Pathak, A.; Bora, A.; Liao, K.-C.; Schmolke, H.; Jung, A.;
Klages, C.-P.; Shwartz; Tornow, M. Disorder-derived, strong
tunneling attenuation in bis-phosphonate monolayers. J. Phys.:
Condens. Matter 2016, 28, 094008−094020.
(65) Jones, D. R.; Troisi, A. Single molecule conductance of linear
dithioalkanes in the liquid phase: apparently activated transport due
to conformational flexibility. J. Phys. Chem. C 2007, 111, 14567−
14573.
(66) Salomon, A.; Shpaisman, H.; Seitz, O.; Boecking, T.; Cahen, D.
Temperature-dependent electronic transport through alkyl chain
monolayers: evidence for a molecular signature. J. Phys. Chem. C
2008, 112, 3969−3974.
(67) Haiss, W.; van Zalinge, H.; Bethell, D.; Ulstrup, J.; Schiffrin, D.
J.; Nichols, R. J. Thermal gating of the single molecule conductance of
alkanedithiols. Faraday Discuss. 2006, 131, 253−264.
(68) Martín, S.; Giustiniano, F.; Haiss, W.; Higgins, S. J.; Whitby, R.
J.; Nichols, R. J. Influence of conformational flexibility on single-
molecule conductance in nano-electrical junctions. J. Phys. Chem. C
2009, 113, 18884−18890.
(69) Frederiksen, T.; Munuera, C.; Ocal, C.; Brandbyge, M.;
Paulsson, M.; Sanchez-Portal, D.; Arnau, A. Exploring the tilt-angle
dependence of electron tunneling across molecular junctions of self-
assembled alkanethiols. ACS Nano 2009, 3, 2073−2080.
(70) Wang, G.; Kim, T.-W.; Jo, G.; Lee, T. Enhancement of field
emission transport by molecular tilt configuration in metal−
molecule−metal junctions. J. Am. Chem. Soc. 2009, 131, 5980−5981.
(71) Song, H.; Lee, H.; Lee, T. Intermolecular chain-to-chain
tunneling in metal−alkanethiol−metal junctions. J. Am. Chem. Soc.
2007, 129, 3806−3807.
(72) Yuan, L.; Jiang, L.; Zhang, B.; Nijhuis, C. A. Dependency of the
tunneling decay coefficient in molecular tunneling junctions on the
3493
https://dx.doi.org/10.1021/jacs.0c12571
J. Am. Chem. Soc. 2021, 143, 3481−3493