ViewVC Help
View File | Revision Log | Show Annotations | View Changeset | Root Listing
root/group/trunk/chainLength/GoldThiolsPaper.tex
Revision: 3854
Committed: Fri Dec 21 23:22:48 2012 UTC (12 years, 7 months ago) by kstocke1
Content type: application/x-tex
File size: 39503 byte(s)
Log Message:
Done!

File Contents

# User Rev Content
1 kstocke1 3801 \documentclass[11pt]{article}
2     \usepackage{amsmath}
3     \usepackage{amssymb}
4 gezelter 3819 \usepackage{times}
5     \usepackage{mathptm}
6 kstocke1 3801 \usepackage{setspace}
7     \usepackage{endfloat}
8     \usepackage{caption}
9 kstocke1 3830 \usepackage{tabularx}
10     \usepackage{longtable}
11 kstocke1 3801 \usepackage{graphicx}
12     \usepackage{multirow}
13 kstocke1 3830 \usepackage{multicol}
14 gezelter 3822 \usepackage[version=3]{mhchem} % this is a great package for formatting chemical reactions
15 kstocke1 3801 \usepackage[square, comma, sort&compress]{natbib}
16     \usepackage{url}
17     \pagestyle{plain} \pagenumbering{arabic} \oddsidemargin 0.0cm
18     \evensidemargin 0.0cm \topmargin -21pt \headsep 10pt \textheight
19     9.0in \textwidth 6.5in \brokenpenalty=10000
20    
21     % double space list of tables and figures
22     %\AtBeginDelayedFloats{\renewcomand{\baselinestretch}{1.66}}
23     \setlength{\abovecaptionskip}{20 pt}
24     \setlength{\belowcaptionskip}{30 pt}
25    
26     \bibpunct{}{}{,}{s}{}{;}
27 gezelter 3822
28     \citestyle{nature}
29 kstocke1 3801 \bibliographystyle{achemso}
30    
31     \begin{document}
32    
33 kstocke1 3830 \newcolumntype{A}{p{1.5in}}
34     \newcolumntype{B}{p{0.75in}}
35    
36 gezelter 3819 \title{Simulations of heat conduction at thiolate-capped gold
37     surfaces: The role of chain length and solvent penetration}
38 kstocke1 3801
39 gezelter 3803 \author{Kelsey M. Stocker and J. Daniel
40     Gezelter\footnote{Corresponding author. \ Electronic mail:
41     gezelter@nd.edu} \\
42     251 Nieuwland Science Hall, \\
43 kstocke1 3801 Department of Chemistry and Biochemistry,\\
44     University of Notre Dame\\
45     Notre Dame, Indiana 46556}
46    
47     \date{\today}
48    
49     \maketitle
50    
51     \begin{doublespace}
52    
53     \begin{abstract}
54 kstocke1 3815
55 kstocke1 3851 We report on simulations of heat conduction from Au(111) / hexane interfaces in which the surface has been protected by a mix of short and long chain alkanethiolate ligands. A variant of reverse non-equilibrium molecular dynamics (RNEMD) was used to create a thermal flux between the metal and solvent, and thermal conductance was computed using the resulting thermal profiles of the interface. We find a non-monotonic dependence of the interfacial thermal conductance on the fraction of long chains present at the interface, and correlate this behavior to solvent ordering and escape from the thiolate layer immediately in contact with the metal surface.
56     Our results suggest that a mixed vibrational transfer / convection model is
57     necessary to explain the features of heat transfer at this interface. The
58     alignment of the solvent chains with the ordered ligand allows rapid transfer
59 kstocke1 3854 of energy to the trapped solvent and is the dominant feature for ordered
60     ligand layers. Diffusion of the vibrationally excited solvent into the bulk
61 kstocke1 3851 also plays a significant role when the ligands are less tightly packed.
62    
63 kstocke1 3801 \end{abstract}
64    
65     \newpage
66    
67     %\narrowtext
68    
69     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
70     % **INTRODUCTION**
71     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
72     \section{Introduction}
73    
74 gezelter 3848 The structural details of the interfaces of metal nanoparticles
75     determines how energy flows between these particles and their
76     surroundings. Understanding this energy flow is essential in designing
77 kstocke1 3854 and functionalizing metallic nanoparticles for use in plasmonic photothermal
78 gezelter 3848 therapies,\cite{Jain:2007ux,Petrova:2007ad,Gnyawali:2008lp,Mazzaglia:2008to,Huff:2007ye,Larson:2007hw}
79     which rely on the ability of metallic nanoparticles to absorb light in
80     the near-IR, a portion of the spectrum in which living tissue is very
81     nearly transparent. The relevant physical property controlling the
82     transfer of this energy as heat into the surrounding tissue is the
83     interfacial thermal conductance, $G$, which can be somewhat difficult
84     to determine experimentally.\cite{Wilson:2002uq,Plech:2005kx}
85 gezelter 3803
86     Metallic particles have also been proposed for use in highly efficient
87 gezelter 3848 thermal-transfer fluids, although careful experiments by Eapen {\it et
88 gezelter 3850 al.} have shown that metal-particle-based nanofluids have thermal
89     conductivities that match Maxwell predictions.\cite{Eapen:2007th} The
90     likely cause of previously reported non-Maxwell
91 gezelter 3803 behavior\cite{Eastman:2001wb,Keblinski:2002bx,Lee:1999ct,Xue:2003ya,Xue:2004oa}
92     is percolation networks of nanoparticles exchanging energy via the
93 gezelter 3850 solvent,\cite{Eapen:2007mw} so it is important to get a detailed
94     molecular picture of particle-ligand and ligand-solvent interactions
95     in order to understand the thermal behavior of complex fluids. To
96     date, there have been few reported values (either from theory or
97 kstocke1 3854 experiment) of $G$ for ligand-protected nanoparticles embedded in
98 gezelter 3850 liquids, and there is a significant gap in knowledge about how
99     chemically distinct ligands or protecting groups will affect heat
100     transport from the particles.
101 gezelter 3803
102 gezelter 3848 Experimentally, the thermal properties of various nanostructured
103     interfaces have been investigated by a number of groups. Cahill and
104 gezelter 3850 coworkers studied thermal transport from metal nanoparticle/fluid
105     interfaces, epitaxial TiN/single crystal oxide interfaces, and
106     hydrophilic and hydrophobic interfaces between water and solids with
107     different self-assembled
108 gezelter 3819 monolayers.\cite{cahill:793,Wilson:2002uq,PhysRevB.67.054302,doi:10.1021/jp048375k,PhysRevLett.96.186101}
109 gezelter 3803 Wang {\it et al.} studied heat transport through long-chain
110     hydrocarbon monolayers on gold substrate at the individual molecular
111     level,\cite{Wang10082007} Schmidt {\it et al.} studied the role of
112     cetyltrimethylammonium bromide (CTAB) on the thermal transport between
113     gold nanorods and solvent,\cite{doi:10.1021/jp8051888} and Juv\'e {\it
114     et al.} studied the cooling dynamics, which is controlled by thermal
115     interface resistance of glass-embedded metal
116     nanoparticles.\cite{PhysRevB.80.195406} Although interfaces are
117     normally considered barriers for heat transport, Alper {\it et al.}
118 gezelter 3850 have suggested that specific ligands (capping agents) could completely
119 gezelter 3803 eliminate this barrier
120     ($G\rightarrow\infty$).\cite{doi:10.1021/la904855s}
121    
122     In previous simulations, we applied a variant of reverse
123     non-equilibrium molecular dynamics (RNEMD) to calculate the
124     interfacial thermal conductance at a metal / organic solvent interface
125     that had been chemically protected by butanethiolate groups. Our
126 gezelter 3850 calculations suggested an explanation for the very large thermal
127 gezelter 3803 conductivity at alkanethiol-capped metal surfaces when compared with
128     bare metal/solvent interfaces. Specifically, the chemical bond
129     between the metal and the ligand introduces a vibrational overlap that
130     is not present without the protecting group, and the overlap between
131     the vibrational spectra (metal to ligand, ligand to solvent) provides
132     a mechanism for rapid thermal transport across the interface.
133    
134 gezelter 3850 A notable result of the previous simulations was the non-monotonic
135     dependence of $G$ on the fractional coverage of the metal surface by
136     the chemical protecting group. Gaps in surface coverage allow the
137 kstocke1 3854 solvent molecules to come into direct contact with ligands that have been
138 gezelter 3850 heated by the metal surface, absorb thermal energy from the ligands,
139     and then diffuse away. Quantifying the role of surface coverage is
140     difficult as the ligands have lateral mobility on the surface and can
141     aggregate to form domains on the timescale of the simulation.
142 gezelter 3803
143 kstocke1 3854 To isolate this effect while avoiding lateral mobility of the
144 gezelter 3850 surface ligands, the current work involves mixed-chain-length
145     monolayers in which the length mismatch between long and short chains
146 kstocke1 3851 is sufficient to accommodate solvent molecules. These completely
147 gezelter 3850 covered (but mixed-chain) surfaces are significantly less prone to
148     surface domain formation, and allow us to further investigate the
149     mechanism of heat transport to the solvent. A thermal flux is created
150     using velocity shearing and scaling reverse non-equilibrium molecular
151     dynamics (VSS-RNEMD), and the resulting temperature profiles are
152     analyzed to yield information about the interfacial thermal
153     conductance.
154 gezelter 3803
155 gezelter 3848
156 kstocke1 3801 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
157     % **METHODOLOGY**
158     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
159     \section{Methodology}
160    
161 gezelter 3803 There are many ways to compute bulk transport properties from
162     classical molecular dynamics simulations. Equilibrium Molecular
163     Dynamics (EMD) simulations can be used by computing relevant time
164     correlation functions and assuming that linear response theory
165     holds.\cite{PhysRevB.37.5677,MASSOBRIO:1984bl,PhysRev.119.1,Viscardy:2007rp,che:6888,kinaci:014106}
166     For some transport properties (notably thermal conductivity), EMD
167     approaches are subject to noise and poor convergence of the relevant
168     correlation functions. Traditional Non-equilibrium Molecular Dynamics
169     (NEMD) methods impose a gradient (e.g. thermal or momentum) on a
170     simulation.\cite{ASHURST:1975tg,Evans:1982zk,ERPENBECK:1984sp,MAGINN:1993hc,Berthier:2002ij,Evans:2002ai,Schelling:2002dp,PhysRevA.34.1449,JiangHao_jp802942v}
171     However, the resulting flux is often difficult to
172     measure. Furthermore, problems arise for NEMD simulations of
173     heterogeneous systems, such as phase-phase boundaries or interfaces,
174     where the type of gradient to enforce at the boundary between
175     materials is unclear.
176    
177     {\it Reverse} Non-Equilibrium Molecular Dynamics (RNEMD) methods adopt
178     a different approach in that an unphysical {\it flux} is imposed
179     between different regions or ``slabs'' of the simulation
180 gezelter 3850 box.\cite{MullerPlathe:1997xw,ISI:000080382700030,Kuang2010} The
181 gezelter 3803 system responds by developing a temperature or momentum {\it gradient}
182     between the two regions. Since the amount of the applied flux is known
183     exactly, and the measurement of a gradient is generally less
184     complicated, imposed-flux methods typically take shorter simulation
185     times to obtain converged results for transport properties. The
186     corresponding temperature or velocity gradients which develop in
187     response to the applied flux are then related (via linear response
188     theory) to the transport coefficient of interest. These methods are
189     quite efficient, in that they do not need many trajectories to provide
190     information about transport properties. To date, they have been
191     utilized in computing thermal and mechanical transfer of both
192     homogeneous
193     liquids~\cite{MullerPlathe:1997xw,ISI:000080382700030,Maginn:2010} as
194     well as heterogeneous
195     systems.\cite{garde:nl2005,garde:PhysRevLett2009,kuang:AuThl}
196    
197 gezelter 3822 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
198     % VSS-RNEMD
199     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
200     \subsection{VSS-RNEMD}
201 gezelter 3803 The original ``swapping'' approaches by M\"{u}ller-Plathe {\it et
202     al.}\cite{ISI:000080382700030,MullerPlathe:1997xw} can be understood
203     as a sequence of imaginary elastic collisions between particles in
204     regions separated by half of the simulation cell. In each collision,
205     the entire momentum vectors of both particles may be exchanged to
206     generate a thermal flux. Alternatively, a single component of the
207     momentum vectors may be exchanged to generate a shear flux. This
208     algorithm turns out to be quite useful in many simulations of bulk
209     liquids. However, the M\"{u}ller-Plathe swapping approach perturbs the
210 kstocke1 3854 system away from ideal Maxwell-Boltzmann distributions, which has
211 gezelter 3803 undesirable side-effects when the applied flux becomes
212     large.\cite{Maginn:2010}
213    
214 gezelter 3848 The most useful alternative RNEMD approach developed so far utilizes a
215     series of simultaneous velocity shearing and scaling exchanges between
216 gezelter 3850 the two slabs.\cite{Kuang2012} This method provides a set of
217 gezelter 3803 conservation constraints while simultaneously creating a desired flux
218     between the two slabs. Satisfying the constraint equations ensures
219     that the new configurations are sampled from the same NVE ensemble.
220    
221     The VSS moves are applied periodically to scale and shift the particle
222     velocities ($\mathbf{v}_i$ and $\mathbf{v}_j$) in two slabs ($H$ and
223     $C$) which are separated by half of the simulation box,
224     \begin{displaymath}
225     \begin{array}{rclcl}
226    
227     & \underline{\mathrm{shearing}} & &
228     \underline{~~~~~~~~~~~~\mathrm{scaling}~~~~~~~~~~~~} \\ \\
229     \mathbf{v}_i \leftarrow &
230     \mathbf{a}_c\ & + & c\cdot\left(\mathbf{v}_i - \langle\mathbf{v}_c
231     \rangle\right) + \langle\mathbf{v}_c\rangle \\
232     \mathbf{v}_j \leftarrow &
233     \mathbf{a}_h & + & h\cdot\left(\mathbf{v}_j - \langle\mathbf{v}_h
234     \rangle\right) + \langle\mathbf{v}_h\rangle .
235    
236     \end{array}
237     \end{displaymath}
238     Here $\langle\mathbf{v}_c\rangle$ and $\langle\mathbf{v}_h\rangle$ are
239     the center of mass velocities in the $C$ and $H$ slabs, respectively.
240     Within the two slabs, particles receive incremental changes or a
241     ``shear'' to their velocities. The amount of shear is governed by the
242     imposed momentum flux, $j_z(\mathbf{p})$
243     \begin{eqnarray}
244     \mathbf{a}_c & = & - \mathbf{j}_z(\mathbf{p}) \Delta t / M_c \label{vss1}\\
245     \mathbf{a}_h & = & + \mathbf{j}_z(\mathbf{p}) \Delta t / M_h \label{vss2}
246     \end{eqnarray}
247 kstocke1 3854 where $M_{\{c,h\}}$ is the total mass of particles within each slab and $\Delta t$
248 gezelter 3803 is the interval between two separate operations.
249    
250     To simultaneously impose a thermal flux ($J_z$) between the slabs we
251     use energy conservation constraints,
252     \begin{eqnarray}
253     K_c - J_z\Delta t & = & c^2 (K_c - \frac{1}{2}M_c \langle\mathbf{v}_c
254     \rangle^2) + \frac{1}{2}M_c (\langle \mathbf{v}_c \rangle + \mathbf{a}_c)^2 \label{vss3}\\
255     K_h + J_z\Delta t & = & h^2 (K_h - \frac{1}{2}M_h \langle\mathbf{v}_h
256     \rangle^2) + \frac{1}{2}M_h (\langle \mathbf{v}_h \rangle +
257     \mathbf{a}_h)^2 \label{vss4}.
258     \label{constraint}
259     \end{eqnarray}
260     Simultaneous solution of these quadratic formulae for the scaling
261     coefficients, $c$ and $h$, will ensure that the simulation samples from
262     the original microcanonical (NVE) ensemble. Here $K_{\{c,h\}}$ is the
263     instantaneous translational kinetic energy of each slab. At each time
264     interval, it is a simple matter to solve for $c$, $h$, $\mathbf{a}_c$,
265     and $\mathbf{a}_h$, subject to the imposed momentum flux,
266 kstocke1 3854 $j_z(\mathbf{p})$, and thermal flux, $J_z$, values. Since the VSS
267 gezelter 3803 operations do not change the kinetic energy due to orientational
268     degrees of freedom or the potential energy of a system, configurations
269     after the VSS move have exactly the same energy ({\it and} linear
270     momentum) as before the move.
271    
272     As the simulation progresses, the VSS moves are performed on a regular
273     basis, and the system develops a thermal or velocity gradient in
274     response to the applied flux. Using the slope of the temperature or
275 kstocke1 3854 velocity gradient, it is quite simple to obtain the thermal
276 gezelter 3803 conductivity ($\lambda$),
277     \begin{equation}
278     J_z = -\lambda \frac{\partial T}{\partial z},
279     \end{equation}
280     and shear viscosity ($\eta$),
281     \begin{equation}
282     j_z(p_x) = -\eta \frac{\partial \langle v_x \rangle}{\partial z}.
283     \end{equation}
284     Here, the quantities on the left hand side are the imposed flux
285     values, while the slopes are obtained from linear fits to the
286     gradients that develop in the simulated system.
287    
288     The VSS-RNEMD approach is versatile in that it may be used to
289     implement both thermal and shear transport either separately or
290     simultaneously. Perturbations of velocities away from the ideal
291 kstocke1 3854 Maxwell-Boltzmann distributions are minimal, as is thermal anisotropy. This ability to generate simultaneous thermal and
292 gezelter 3803 shear fluxes has been previously utilized to map out the shear
293     viscosity of SPC/E water over a wide range of temperatures (90~K) with
294 gezelter 3850 a single 1 ns simulation.\cite{Kuang2012}
295 gezelter 3803
296 gezelter 3822 \begin{figure}
297     \includegraphics[width=\linewidth]{figures/rnemd}
298     \caption{The VSS-RNEMD approach imposes unphysical transfer of
299     linear momentum or kinetic energy between a ``hot'' slab and a
300     ``cold'' slab in the simulation box. The system responds to this
301     imposed flux by generating velocity or temperature gradients. The
302     slope of the gradients can then be used to compute transport
303     properties (e.g. shear viscosity or thermal conductivity).}
304     \label{fig:rnemd}
305     \end{figure}
306    
307     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
308     % INTERFACIAL CONDUCTANCE
309     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
310     \subsection{Reverse Non-Equilibrium Molecular Dynamics approaches
311 gezelter 3803 to interfacial transport}
312 kstocke1 3801
313 gezelter 3803 Interfaces between dissimilar materials have transport properties
314     which can be defined as derivatives of the standard transport
315     coefficients in a direction normal to the interface. For example, the
316     {\it interfacial} thermal conductance ($G$) can be thought of as the
317     change in the thermal conductivity ($\lambda$) across the boundary
318     between materials:
319     \begin{align}
320     G &= \left(\nabla\lambda \cdot \mathbf{\hat{n}}\right)_{z_0} \\
321     &= J_z\left(\frac{\partial^2 T}{\partial z^2}\right)_{z_0} \Big/
322     \left(\frac{\partial T}{\partial z}\right)_{z_0}^2.
323     \label{derivativeG}
324     \end{align}
325     where $z_0$ is the location of the interface between two materials and
326     $\mathbf{\hat{n}}$ is a unit vector normal to the interface (assumed
327     to be the $z$ direction from here on). RNEMD simulations, and
328     particularly the VSS-RNEMD approach, function by applying a momentum
329     or thermal flux and watching the gradient response of the
330     material. This means that the {\it interfacial} conductance is a
331     second derivative property which is subject to significant noise and
332     therefore requires extensive sampling. We have been able to
333     demonstrate the use of the second derivative approach to compute
334     interfacial conductance at chemically-modified metal / solvent
335     interfaces. However, a definition of the interfacial conductance in
336     terms of a temperature difference ($\Delta T$) measured at $z_0$,
337     \begin{displaymath}
338     G = \frac{J_z}{\Delta T_{z_0}},
339     \end{displaymath}
340     is useful once the RNEMD approach has generated a stable temperature
341     gap across the interface.
342    
343     \begin{figure}
344     \includegraphics[width=\linewidth]{figures/resistor_series}
345 gezelter 3822 \caption{The inverse of the interfacial thermal conductance, $G$, is
346 kstocke1 3854 the Kapitza resistance, $R_K$. Because the gold / thiolate /
347 gezelter 3822 solvent interface extends a significant distance from the metal
348     surface, the interfacial resistance $R_K$ can be computed by
349     summing a series of temperature drops between adjacent temperature
350     bins along the $z$ axis.}
351 gezelter 3803 \label{fig:resistor_series}
352     \end{figure}
353    
354     In the particular case we are studying here, there are two interfaces
355     involved in the transfer of heat from the gold slab to the solvent:
356 kstocke1 3854 the metal / thiolate interface and the thiolate / solvent interface. We
357     can treat the temperature on each side of an interface as discrete,
358 gezelter 3803 making the interfacial conductance the inverse of the Kaptiza
359     resistance, or $G = \frac{1}{R_k}$. To model the total conductance
360     across multiple interfaces, it is useful to think of the interfaces as
361     a set of resistors wired in series. The total resistance is then
362 kstocke1 3854 additive, $R_{total} = \sum_i R_{i}$, and the interfacial conductance
363 gezelter 3803 is the inverse of the total resistance, or $G = \frac{1}{\sum_i
364     R_i}$). In the interfacial region, we treat each bin in the
365     VSS-RNEMD temperature profile as a resistor with resistance
366     $\frac{T_{2}-T_{1}}{J_z}$, $\frac{T_{3}-T_{2}}{J_z}$, etc. The sum of
367 kstocke1 3854 the set of resistors which spans the gold / thiolate interface, thiolate
368     chains, and thiolate / solvent interface simplifies to
369 kstocke1 3801 \begin{equation}
370 gezelter 3825 R_{K} = \frac{T_{n}-T_{1}}{J_z},
371 gezelter 3803 \label{eq:finalG}
372     \end{equation}
373     or the temperature difference between the gold side of the
374 kstocke1 3854 gold / thiolate interface and the solvent side of the thiolate / solvent
375 gezelter 3803 interface over the applied flux.
376 kstocke1 3801
377    
378     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
379     % **COMPUTATIONAL DETAILS**
380     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
381     \section{Computational Details}
382    
383 gezelter 3803 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
384 gezelter 3819 % FORCE-FIELD PARAMETERS
385     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
386     \subsection{Force-Field Parameters}
387 kstocke1 3801
388 gezelter 3819 Our simulations include a number of chemically distinct components.
389     Figure \ref{fig:structures} demonstrates the sites defined for both
390     the {\it n}-hexane and alkanethiolate ligands present in our
391 gezelter 3850 simulations.
392 gezelter 3819
393     \begin{figure}
394     \includegraphics[width=\linewidth]{figures/structures}
395 gezelter 3822 \caption{Topologies of the thiolate capping agents and solvent
396     utilized in the simulations. The chemically-distinct sites (S,
397     \ce{CH2}, and \ce{CH3}) are treated as united atoms. Most
398     parameters are taken from references \bibpunct{}{}{,}{n}{}{,}
399     \protect\cite{TraPPE-UA.alkanes} and
400     \protect\cite{TraPPE-UA.thiols}. Cross-interactions with the Au
401     atoms were adapted from references
402     \protect\cite{landman:1998},~\protect\cite{vlugt:cpc2007154},~and
403     \protect\cite{hautman:4994}.}
404 gezelter 3819 \label{fig:structures}
405     \end{figure}
406    
407 kstocke1 3854 The Au-Au interactions in the metal lattice slab were described by the
408 gezelter 3819 quantum Sutton-Chen (QSC) formulation.\cite{PhysRevB.59.3527} The QSC
409     potentials include zero-point quantum corrections and are
410     reparametrized for accurate surface energies compared to the
411     Sutton-Chen potentials.\cite{Chen90}
412    
413     For the {\it n}-hexane solvent molecules, the TraPPE-UA
414     parameters\cite{TraPPE-UA.alkanes} were utilized. In this model,
415     sites are located at the carbon centers for alkyl groups. Bonding
416     interactions, including bond stretches and bends and torsions, were
417     used for intra-molecular sites closer than 3 bonds. For non-bonded
418 gezelter 3850 interactions, Lennard-Jones potentials were used. We have previously
419 gezelter 3819 utilized both united atom (UA) and all-atom (AA) force fields for
420 gezelter 3850 thermal conductivity,\cite{kuang:AuThl,Kuang2012} and since the united
421 gezelter 3819 atom force fields cannot populate the high-frequency modes that are
422     present in AA force fields, they appear to work better for modeling
423     thermal conductivity. The TraPPE-UA model for alkanes is known to
424     predict a slightly lower boiling point than experimental values. This
425 kstocke1 3854 is one of the reasons we used a lower average temperature (220 K) for
426 gezelter 3850 our simulations.
427 gezelter 3819
428     The TraPPE-UA force field includes parameters for thiol
429     molecules\cite{TraPPE-UA.thiols} which were used for the
430     alkanethiolate molecules in our simulations. To derive suitable
431 kstocke1 3854 parameters for butanethiolate adsorbed on Au(111) surfaces, we adopted
432 gezelter 3819 the S parameters from Luedtke and Landman\cite{landman:1998} and
433     modified the parameters for the CTS atom to maintain charge neutrality
434     in the molecule.
435    
436     To describe the interactions between metal (Au) and non-metal atoms,
437     we refer to an adsorption study of alkyl thiols on gold surfaces by
438     Vlugt {\it et al.}\cite{vlugt:cpc2007154} They fitted an effective
439     Lennard-Jones form of potential parameters for the interaction between
440     Au and pseudo-atoms CH$_x$ and S based on a well-established and
441     widely-used effective potential of Hautman and Klein for the Au(111)
442     surface.\cite{hautman:4994} As our simulations require the gold slab
443     to be flexible to accommodate thermal excitation, the pair-wise form
444 kstocke1 3832 of potentials they developed was used for our study.
445 kstocke1 3821
446     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
447     % SIMULATION PROTOCOL
448     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
449     \subsection{Simulation Protocol}
450    
451 gezelter 3850 We have implemented the VSS-RNEMD algorithm in OpenMD, our group
452     molecular dynamics code.\cite{openmd} An 1188 atom gold slab was
453     equilibrated at 1 atm and 200 K. The periodic box was then expanded
454 kstocke1 3854 in the $z$ direction to expose two Au(111) faces on either side of the
455 gezelter 3850 11-atom thick slab.
456 kstocke1 3821
457 gezelter 3850 A full monolayer of thiolates (1/3 the number of surface gold atoms)
458     was placed on three-fold hollow sites on the gold interfaces. The
459     effect of thiolate binding sites on the thermal conductance was tested
460     by placing thiolates at both fcc and hcp hollow sites. No appreciable
461 kstocke1 3854 difference in the temperature profile due to the location of
462     thiolate binding was noted.
463 kstocke1 3821
464 gezelter 3850 To test the role of thiolate chain length on interfacial thermal
465     conductance, full coverages of each of five chain lengths were tested:
466     butanethiolate (C$_4$), hexanethiolate (C$_6$), octanethiolate
467     (C$_8$), decanethiolate (C$_{10}$), and dodecanethiolate
468     (C$_{12}$). To test the effect of mixed chain lengths, full coverages
469     of C$_4$ / C$_{10}$ and C$_4$ / C$_{12}$ mixtures were created in
470     short/long chain percentages of 25/75, 50/50, 62.5/37.5, 75/25, and
471     87.5/12.5. The short and long chains were placed on the surface hollow
472     sites in a random configuration.
473 kstocke1 3821
474 gezelter 3850 The simulation box $z$-dimension was set to roughly double the length
475 kstocke1 3854 of the gold / thiolate block. Hexane solvent molecules were placed in
476 kstocke1 3851 the vacant portion of the box using the packmol algorithm. Figure \ref{fig:timelapse} shows two of the mixed chain length systems both before and after the RNEMD simulation.
477 kstocke1 3821
478 kstocke1 3851 \begin{figure}
479     \includegraphics[width=\linewidth]{figures/timelapse}
480 kstocke1 3854 \caption{Images of 25\%~C$_4$~/~75\%~C$_{12}$ (top panel) and 75\%~C$_4$~/~25\%~C$_{12}$ (bottom panel) interfaces at the beginning and end of 3 ns simulations. Solvent molecules that were initially present in the thiolate layer are colored light blue. Diffusion of the initially-trapped solvent into the bulk is apparent in the interface with fewer long chains. Trapped solvent is orientationally locked to the ordered ligands (and is less able to diffuse into the bulk) when the fraction of long chains increases.}
481 kstocke1 3851 \label{fig:timelapse}
482     \end{figure}
483    
484 gezelter 3850 The system was equilibrated to 220 K in the canonical (NVT) ensemble,
485     allowing the thiolates and solvent to warm gradually. Pressure
486     correction to 1 atm was done using an isobaric-isothermal (NPT)
487     integrator that allowed expansion or contraction only in the $z$
488     direction, maintaining the crystalline structure of the gold as close
489     to the bulk result as possible. The diagonal elements of the pressure
490     tensor were monitored during the pressure equilibration stage. If the
491     $xx$ and/or $yy$ elements had a mean above zero throughout the
492     simulation -- indicating residual surface tension in the plane of the
493     gold slab -- an additional short NPT equilibration step was performed
494     allowing all box dimensions to change. Once the pressure was stable
495     at 1 atm, a final equilibration stage was performed at constant
496     temperature. All systems were equilibrated in the microcanonical (NVE)
497     ensemble before proceeding with the VSS-RNEMD and data collection
498     stages.
499 kstocke1 3821
500 gezelter 3850 A kinetic energy flux was applied using VSS-RNEMD. The total
501     simulation time was 3 nanoseconds, with velocity scaling occurring
502 kstocke1 3854 every 10 femtoseconds. The ``hot'' slab was centered in the gold and the
503     ``cold'' slab was placed in the center of the solvent region. The entire
504     system had a (time-averaged) temperature of 220 K, with a temperature
505 gezelter 3850 difference between the hot and cold slabs of approximately 30 K. The
506 kstocke1 3854 average temperature and kinetic energy flux were selected to avoid
507     solvent freezing (or glass formation) and to prevent the thiolates from
508     burying in the gold slab. The Au-S interaction has a deep potential
509 gezelter 3850 energy well, which allows sulfur atoms to embed into the gold slab at
510     temperatures above 250 K. Simulation conditions were chosen to avoid
511     both of these undesirable situations.
512 kstocke1 3821
513 gezelter 3850 Temperature profiles of the system were created by dividing the box
514     into $\sim$ 3 \AA \, bins along the $z$ axis and recording the average
515     temperature of each bin.
516 kstocke1 3821
517 gezelter 3850
518 kstocke1 3801
519     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
520     % **RESULTS**
521     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
522     \section{Results}
523    
524 kstocke1 3851 The solvent, hexane, is a straight chain flexible alkane that is structurally
525     similar to the thiolate alkane tails. Previous work has shown that UA models
526     of hexane and butanethiolate have a high degree of vibrational
527     overlap.\cite{kuang:AuThl} This overlap provides a mechanism for thermal
528     energy conduction from the thiolates to the solvent. Indeed, we observe that
529     the interfacial conductance is twice as large with the thiolate monolayers (of
530     all chain lengths) than with the bare metal surface.
531 kstocke1 3801
532 gezelter 3819 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
533     % CHAIN LENGTH
534     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
535     \subsection{Effect of Chain Length}
536    
537 kstocke1 3854 We examined full coverages of five alkyl chain lengths, C$_{4}$, C$_{6}$, C$_{8}$, C$_{10}$, and C$_{12}$. As shown in table \ref{table:chainlengthG}, the interfacial conductance is roughly constant as a function of chain length, indicating that the length of the thiolate alkyl chains does not play a significant role in the transport of heat across the gold/thiolate and thiolate/solvent interfaces. In all cases, the hexane solvent was unable to penetrate into the thiolate layer, leading to a persistent 2-4 \AA \, gap between the solvent region and the thiolates. However, while the identity of the alkyl thiolate capping agent has little effect on the interfacial thermal conductance, the presence of a full monolayer of capping agent provides a two-fold increase in the G value relative to a bare gold surface.
538 kstocke1 3830 \begin{longtable}{p{4cm} p{3cm}}
539 kstocke1 3851 \caption{Computed interfacial thermal conductance ($G$) values for bare gold and 100\% coverages of various thiolate alkyl chain lengths.}
540 kstocke1 3830 \\
541 kstocke1 3854 \centering {\bf Chain Length} & \centering\arraybackslash {\bf G (MW/m$^2$/K)} \\ \hline
542 kstocke1 3830 \endhead
543     \hline
544     \endfoot
545 kstocke1 3851 \centering bare metal & \centering\arraybackslash 30.2 \\
546 kstocke1 3854 $~~~~~~~~~~~~~~~~~~$ C$_{4}$ & \centering\arraybackslash 59.4 \\
547     $~~~~~~~~~~~~~~~~~~$ C$_{6}$ & \centering\arraybackslash 60.2 \\
548     $~~~~~~~~~~~~~~~~~~$ C$_{8}$ & \centering\arraybackslash 61.0 \\
549     $~~~~~~~~~~~~~~~~~~$ C$_{10}$ & \centering\arraybackslash 58.2 \\
550     $~~~~~~~~~~~~~~~~~~$ C$_{12}$ & \centering\arraybackslash 58.8
551 kstocke1 3830 \label{table:chainlengthG}
552     \end{longtable}
553 kstocke1 3801
554 gezelter 3819 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
555     % MIXED CHAINS
556     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
557     \subsection{Effect of Mixed Chain Lengths}
558    
559 kstocke1 3854 Previous simulations have demonstrated non-monotonic behavior for $G$ as a function of the surface coverage. One difficulty with the previous study was the ability of butanethiolate ligands to migrate on the Au(111) surface and to form segregated domains. To simulate the effect of low coverages while preventing thiolate domain formation, we maintain 100\% thiolate coverage while varying the proportions of short (butanethiolate, C$_4$) and long (decanethiolate, C$_{10}$, or dodecanethiolate, C$_{12}$) alkyl chains. Data on the conductance trend as the fraction of long chains was varied is shown in figure \ref{fig:Gstacks}. Note that as in the previous study, $G$ is dependent upon solvent accessibility to thermally excited ligands. Our simulations indicate a similar (but less dramatic) non-monotonic dependence on the fraction of long chains.
560 kstocke1 3829 \begin{figure}
561     \includegraphics[width=\linewidth]{figures/Gstacks}
562 kstocke1 3851 \caption{Interfacial thermal conductivity of mixed-chains has a non-monotonic dependence on the fraction of long chains (lower panels). At low fractions of long chains, the solvent escape rate ($k_{escape}$) dominates the heat transfer process, while the solvent-thiolate orientational ordering dominates in systems with higher fractions of long chains (upper panels).}
563 kstocke1 3829 \label{fig:Gstacks}
564     \end{figure}
565 kstocke1 3815
566 kstocke1 3801 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
567 kstocke1 3854 % **DISCUSSION**
568 kstocke1 3801 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
569 kstocke1 3854 \section{Discussion}
570 gezelter 3819
571 kstocke1 3854 In the mixed chain-length simulations, solvent molecules
572     can become temporarily trapped or entangled with the thiolate chains. Their
573     residence in close proximity to the higher temperature environment close to
574     the surface allows them to carry heat away from the surface quite efficiently.
575     There are two aspects of this behavior that are relevant to thermal
576     conductance of the interface: the residence time of solvent molecules in the
577     thiolate layer, and the alignment of solvent molecules with the ligand alkyl
578     chains as a mechanism for transferring vibrational energy to these entrapped
579     solvent molecules. To quantify these competing effects, we have computed
580     solvent escape rates from the thiolate layer as well as a joint orientational
581     order parameter between the trapped solvent and the thiolate ligands.
582    
583 gezelter 3819 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
584     % RESIDENCE TIME
585     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
586 kstocke1 3841 \subsection{Mobility of solvent in the interfacial layer}
587 gezelter 3819
588 kstocke1 3851 We use a simple survival correlation function, $C(t)$, to measure the
589     residence time of a solvent molecule in the thiolate
590 gezelter 3819 layer. This function correlates the identity of all hexane molecules
591     within the $z$-coordinate range of the thiolate layer at two separate
592     times. If the solvent molecule is present at both times, the
593     configuration contributes a $1$, while the absence of the molecule at
594     the later time indicates that the solvent molecule has migrated into
595     the bulk, and this configuration contributes a $0$. A steep decay in
596     $C(t)$ indicates a high turnover rate of solvent molecules from the
597 kstocke1 3837 chain region to the bulk. We define the escape rate for trapped solvent molecules at the interface as
598 kstocke1 3815 \begin{equation}
599 gezelter 3825 k_{escape} = \left( \int_0^T C(t) dt \right)^{-1}
600 kstocke1 3821 \label{eq:mobility}
601 kstocke1 3815 \end{equation}
602 gezelter 3825 where T is the length of the simulation. This is a direct measure of
603 kstocke1 3851 the rate at which solvent molecules initially entangled in the thiolate layer
604 kstocke1 3843 can escape into the bulk. As $k_{escape} \rightarrow 0$, the
605 kstocke1 3854 solvent becomes permanently trapped in the thiolate layer. In
606 kstocke1 3841 figure \ref{fig:Gstacks} we show that interfacial solvent mobility
607 kstocke1 3854 decreases as the fraction of long thiolate chains increases.
608 gezelter 3819
609     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
610     % ORDER PARAMETER
611     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
612    
613     \subsection{Vibrational coupling via orientational ordering}
614    
615 kstocke1 3851 As the fraction of long-chain thiolates increases, the entrapped
616 gezelter 3819 solvent molecules must find specific orientations relative to the mean
617 kstocke1 3851 orientation of the thiolate chains. This alignment allows for
618 gezelter 3819 efficient thermal energy exchange between the thiolate alkyl chain and
619     the solvent molecules.
620    
621 kstocke1 3854 Once the interfacial solvent molecules have picked up thermal energy from the
622     thiolates, they carry heat away from the gold as they diffuse back
623     into the bulk solvent. When the percentage of long chains decreases,
624     the tails of the long chains are much more disordered and do not
625     provide structured channels for the solvent to fill.
626    
627     To measure this cooperative ordering, we compute the orientational order
628 kstocke1 3851 parameters and director axes for both the thiolate chains and for the
629     entrapped solvent. The director axis can be easily obtained by diagonalization
630     of the order parameter tensor,
631 gezelter 3819 \begin{equation}
632     \mathsf{Q}_{\alpha \beta} = \frac{1}{2 N} \sum_{i=1}^{N} \left( 3 \mathbf{e}_{i
633     \alpha} \mathbf{e}_{i \beta} - \delta_{\alpha \beta} \right)
634 kstocke1 3815 \end{equation}
635 kstocke1 3854 where $\mathbf{e}_{i \alpha}$ is the $\alpha = x,y,z$ component of
636 gezelter 3819 the unit vector $\mathbf{e}_{i}$ along the long axis of molecule $i$.
637 kstocke1 3854 For both the solvent and the ligand, the $\mathbf{e}$ vector is defined using
638     the terminal atoms of the molecule.
639 gezelter 3819
640     The largest eigenvalue of $\overleftrightarrow{\mathsf{Q}}$ is
641 kstocke1 3854 traditionally used to obtain the orientational order parameter, while the
642 gezelter 3819 eigenvector corresponding to the order parameter yields the director
643 kstocke1 3854 axis ($\mathbf{d}(t)$), which defines the average direction of
644 gezelter 3819 molecular alignment at any time $t$. The overlap between the director
645     axes of the thiolates and the entrapped solvent is time-averaged,
646 kstocke1 3815 \begin{equation}
647 kstocke1 3851 \langle d \rangle = \langle \mathbf{d}_{thiolates} \left( t \right) \cdot
648 gezelter 3825 \mathbf{d}_{solvent} \left( t \right) \rangle_t
649 kstocke1 3815 \label{eq:orientation}
650     \end{equation}
651 kstocke1 3854 and reported in figure \ref{fig:Gstacks}. Values of $\langle d \rangle$ range from $0$ (solvent molecules in the ligand layer are perpendicular to the thiolate chains) to $1$ (solvent and ligand chains are aligned parallel to each other).
652 kstocke1 3815
653 kstocke1 3854 C$_4$ / C$_{10}$ mixed monolayers have a peak interfacial conductance with 75\% long chains. At this fraction of long chains, the cooperative orientational ordering of the solvent molecules and chains becomes the dominant effect while the solvent escape rate is quite slow. C$_4$ / C$_{12}$ mixtures have a peak interfacial conductance for 87.5\% long chains. The solvent-thiolate orientational ordering reaches its maximum value at this long chain fraction. Long chain fractions of over $0.5$ for the C$_4$ / C$_{12}$ system are well ordered, but this effect is tempered by the exceptionally slow solvent escape rate ($\sim$ 1 molecule / 2 ns).
654 kstocke1 3851
655 kstocke1 3843 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
656     % **CONCLUSIONS**
657     %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
658     \section{Conclusions}
659 kstocke1 3854 Our results suggest that a mixed vibrational transfer / convection model is necessary to explain the features of heat transfer at this interface. The alignment of the solvent chains with the ordered ligand allows rapid transfer of energy to the trapped solvent and becomes the dominant feature for ordered ligand layers. Diffusion of the vibrationally excited solvent into the bulk also plays a significant role when the ligands are less tightly packed.
660 kstocke1 3843
661 kstocke1 3851 In the language of earlier continuum approaches to interfacial
662     conductance,\cite{RevModPhys.61.605} the alignment of the chains is an
663     important factor in the transfer of phonons from the thiolate layer to the
664     trapped solvent. The aligned solvent and thiolate chains have nearly identical
665     acoustic impedances and the phonons can scatter directly into a solvent
666     molecule that has been forced into alignment. When the entrapped solvent has
667     more configurations available, the likelihood of an impedance mismatch is
668     higher, and the phonon scatters into the solvent with lower
669     efficiency. The fractional coverage of the long chains is therefore a simple
670     way of tuning the acoustic mismatch between the thiolate layer and the hexane
671     solvent.
672 kstocke1 3843
673 kstocke1 3851 Efficient heat transfer also can be accomplished via convective or diffusive motion of vibrationally excited solvent back into the bulk. Once the entrapped solvent becomes too tightly aligned with the ligands, however, the convective avenue of heat transfer is cut off.
674 kstocke1 3843
675 kstocke1 3854 Our simulations suggest a number of routes to make interfaces with high thermal conductance. If it is possible to create an interface which forces the solvent into alignment with a ligand that shares many of the solvent's vibrational modes, while simultaneously preserving the ability of the solvent to diffuse back into the bulk, we would expect a significant jump in the interfacial conductance. One possible way to accomplish this is to use polyene ligands with alternating unsaturated bonds. These are significantly more rigid than long alkanes, and could force solvent alignment (even at low relative coverages) while preserving mobility of solvent molecules within the ligand layer.
676 kstocke1 3851
677 kstocke1 3801 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
678 kstocke1 3851 % **ACKNOWLEDGMENTS**
679 kstocke1 3801 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
680 kstocke1 3851 \section*{Acknowledgments}
681 kstocke1 3801
682 kstocke1 3851 We gratefully acknowledge conversations with Dr. Shenyu Kuang. Support for
683     this project was provided by the National Science Foundation under grant
684     CHE-0848243. Computational time was provided by the Center for Research
685     Computing (CRC) at the University of Notre Dame.
686    
687 kstocke1 3801 \newpage
688    
689     \bibliography{thiolsRNEMD}
690    
691     \end{doublespace}
692     \end{document}
693    
694    
695    
696    
697    
698    
699    
700    

Properties

Name Value
svn:executable *