1 |
\documentclass[journal = jpccck, manuscript = article]{achemso} |
2 |
\setkeys{acs}{usetitle = true} |
3 |
|
4 |
\usepackage{caption} |
5 |
\usepackage{geometry} |
6 |
\usepackage{natbib} |
7 |
\usepackage{setspace} |
8 |
\usepackage{xkeyval} |
9 |
%%%%%%%%%%%%%%%%%%%%%%% |
10 |
\usepackage{amsmath} |
11 |
\usepackage{amssymb} |
12 |
\usepackage{times} |
13 |
\usepackage{mathptm} |
14 |
\usepackage{caption} |
15 |
\usepackage{tabularx} |
16 |
\usepackage{longtable} |
17 |
\usepackage{graphicx} |
18 |
\usepackage{achemso} |
19 |
\usepackage{wrapfig} |
20 |
\usepackage[version=3]{mhchem} % this is a great package for formatting chemical reactions |
21 |
\usepackage{url} |
22 |
|
23 |
\title{Interfacial Thermal Conductance of Alkanethiolate-Protected Gold |
24 |
Nanospheres} |
25 |
|
26 |
\author{Kelsey M. Stocker} |
27 |
\author{J. Daniel Gezelter} |
28 |
\email{gezelter@nd.edu} |
29 |
\affiliation[University of Notre Dame]{251 Nieuwland Science Hall\\ |
30 |
Department of Chemistry and Biochemistry\\ |
31 |
University of Notre Dame\\ |
32 |
Notre Dame, Indiana 46556} |
33 |
|
34 |
|
35 |
\keywords{Nanoparticles, interfaces, thermal conductance} |
36 |
|
37 |
\begin{document} |
38 |
|
39 |
\begin{tocentry} |
40 |
\center\includegraphics[width=3.9cm]{figures/TOC} |
41 |
\end{tocentry} |
42 |
|
43 |
\newcolumntype{A}{p{1.5in}} |
44 |
\newcolumntype{B}{p{0.75in}} |
45 |
|
46 |
|
47 |
\begin{abstract} |
48 |
Molecular dynamics simulations of alkanethiolate-protected and |
49 |
solvated gold nanoparticles were carried out in the presence of a |
50 |
non-equilibrium heat flux between the solvent and the core of the |
51 |
particle. The interfacial thermal conductance ($G$) was computed for |
52 |
these interfaces, and the behavior of the thermal conductance was |
53 |
studied as a function of particle size and ligand chain length. In |
54 |
all cases, thermal conductance of the ligand-protected particles was |
55 |
higher than the bare metal--solvent interface. A number of |
56 |
mechanisms for the enhanced conductance were investigated, including |
57 |
thiolate-driven corrugation of the metal surface, solvent mobility |
58 |
and ordering at the interface, and ligand ordering relative to the |
59 |
particle surface. The shortest and least flexible ligand, butanethiolate, |
60 |
exhibited the highest interfacial thermal conductance and was the |
61 |
least likely to trap solvent molecules within the ligand layer. At |
62 |
the 50\% coverage levels studied, heat transfer into the solvent |
63 |
relies primarily on convective motion of the solvent molecules from |
64 |
the surface of the particle into the bulk. This mode of heat |
65 |
transfer is reduced by slow solvent escape rates, and this effect was |
66 |
observed to lower the interfacial conductance for the longer-chain ligands. |
67 |
\end{abstract} |
68 |
|
69 |
\newpage |
70 |
|
71 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
72 |
% INTRODUCTION |
73 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
74 |
\section{Introduction} |
75 |
|
76 |
Heat transport across various nanostructured interfaces has been |
77 |
the subject of intense experimental |
78 |
interest,\cite{Wilson:2002uq,PhysRevB.67.054302,doi:10.1021/jp048375k,PhysRevLett.96.186101,Wang10082007,doi:10.1021/jp8051888,PhysRevB.80.195406,doi:10.1021/la904855s} |
79 |
and the interfacial thermal conductance, $G$, is the principal quantity of |
80 |
interest for understanding interfacial heat |
81 |
transport.\cite{cahill:793} Because nanoparticles have a significant |
82 |
fraction of their atoms at the particle / solvent interface, the |
83 |
chemical details of these interfaces govern the thermal transport |
84 |
properties. |
85 |
|
86 |
Previously, reverse nonequilibrium molecular dynamics (RNEMD) methods |
87 |
have been applied to calculate the interfacial thermal conductance at |
88 |
flat (111) metal / organic solvent interfaces that had been chemically |
89 |
protected by varying coverages of alkanethiolate groups.\cite{kuang:AuThl} |
90 |
These simulations suggested an explanation for the increased thermal |
91 |
conductivity at alkanethiol-capped metal surfaces compared with bare |
92 |
metal interfaces. Specifically, the chemical bond between the metal |
93 |
and the ligand introduces a vibrational overlap that is not present |
94 |
without the protecting group, and the overlap between the vibrational |
95 |
spectra (metal to ligand, ligand to solvent) provides a mechanism for |
96 |
rapid thermal transport across the interface. The simulations also |
97 |
suggested that this phenomenon is a non-monotonic function of the |
98 |
fractional coverage of the surface, as moderate coverages allow |
99 |
diffusive heat transport of solvent molecules that come into close |
100 |
contact with the ligands. |
101 |
|
102 |
Simulations of {\it mixed-chain} alkylthiolate surfaces showed that |
103 |
solvent trapped close to the interface can be efficient at moving |
104 |
thermal energy away from the surface.\cite{Stocker:2013cl} Trapped |
105 |
solvent molecules that were aligned with nearby |
106 |
ligands (but which were less able to diffuse into the bulk) were able |
107 |
to increase the thermal conductance of the interface. This indicates |
108 |
that the ligand-to-solvent vibrational energy transfer is a key |
109 |
feature for increasing particle-to-solvent thermal conductance. |
110 |
|
111 |
Recently, we extended RNEMD methods for use in non-periodic geometries |
112 |
by creating scaling/shearing moves between concentric regions of a |
113 |
simulation.\cite{Stocker:2014qq} In this work, we apply this |
114 |
non-periodic variant of RNEMD to investigate the role that {\it |
115 |
curved} nanoparticle surfaces play in heat and mass transport. On |
116 |
planar surfaces, we discovered that orientational ordering of surface |
117 |
protecting ligands had a large effect on the heat conduction from the |
118 |
metal to the solvent. Smaller nanoparticles have high surface |
119 |
curvature that creates gaps in well-ordered self-assembled monolayers, |
120 |
and the effect of those gaps on the thermal conductance is unknown. |
121 |
|
122 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
123 |
% INTERFACIAL THERMAL CONDUCTANCE OF METALLIC NANOPARTICLES |
124 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
125 |
%\section{Interfacial Thermal Conductance of Metallic Nanoparticles} |
126 |
|
127 |
For a solvated nanoparticle, it is possible to define a critical value |
128 |
for the interfacial thermal conductance, |
129 |
\begin{equation} |
130 |
G_c = \frac{3 C_s \Lambda_s}{R C_p} |
131 |
\end{equation} |
132 |
which depends on the solvent heat capacity, $C_s$, solvent thermal |
133 |
conductivity, $\Lambda_s$, particle radius, $R$, and nanoparticle heat |
134 |
capacity, $C_p$.\cite{Wilson:2002uq} In the limit of infinite |
135 |
interfacial thermal conductance, $G \gg G_c$, cooling of the |
136 |
nanoparticle is limited by the solvent properties, $C_s$ and |
137 |
$\Lambda_s$. In the opposite limit, $G \ll G_c$, the heat dissipation |
138 |
is controlled by the thermal conductance of the particle / fluid |
139 |
interface. It is this regime with which we are concerned, where |
140 |
properties of ligands and the particle surface may be tuned to |
141 |
manipulate the rate of cooling for solvated nanoparticles. Based on |
142 |
estimates of $G$ from previous simulations as well as experimental |
143 |
results for solvated nanostructures, gold nanoparticles solvated in |
144 |
hexane are in the $G \ll G_c$ regime for radii smaller than 40 nm. The |
145 |
particles included in this study are more than an order of magnitude |
146 |
smaller than this critical radius, so the heat dissipation should be |
147 |
controlled entirely by the surface features of the particle / ligand / |
148 |
solvent interface. |
149 |
|
150 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
151 |
% STRUCTURE OF SELF-ASSEMBLED MONOLAYERS ON NANOPARTICLES |
152 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
153 |
\subsection{Structures of Self-Assembled Monolayers on Nanoparticles} |
154 |
|
155 |
Though the ligand packing on planar surfaces has been characterized for many |
156 |
different ligands and surface facets, it is not obvious \emph{a |
157 |
priori} how the same ligands will behave on the highly curved |
158 |
surfaces of spherical nanoparticles. Thus, as new applications of |
159 |
ligand-stabilized nanostructures have been proposed, the structure |
160 |
and dynamics of ligands on metallic nanoparticles have been studied |
161 |
using molecular simulation,\cite{Henz2007,Henz:2008qf} NMR, XPS, FTIR, calorimetry, and surface microscopies.\cite{Badia1996:2,Badia1996,Badia1997:2,Badia1997,Badia2000} |
162 |
Badia, \textit{et al.} used transmission electron microscopy to |
163 |
determine that alkanethiol ligands on gold nanoparticles pack |
164 |
approximately 30\% more densely than on planar Au(111) |
165 |
surfaces.\cite{Badia1996:2} Subsequent experiments demonstrated that |
166 |
even at full coverages, surface curvature creates voids between linear |
167 |
ligand chains that can be filled via interdigitation of ligands on |
168 |
neighboring particles.\cite{Badia1996} The molecular dynamics |
169 |
simulations of Henz, \textit{et al.} indicate that at low coverages, |
170 |
the thiolate alkane chains will lie flat on the nanoparticle |
171 |
surface\cite{Henz2007,Henz:2008qf} Above 90\% coverage, the ligands stand upright |
172 |
and recover the rigidity and tilt angle displayed on planar |
173 |
facets. Their simulations also indicate a high degree of mixing |
174 |
between the thiolate sulfur atoms and surface gold atoms at high |
175 |
coverages. |
176 |
|
177 |
In this work, thiolated gold nanospheres were modeled using a united atom force field and non-equilibrium molecular dynamics. Gold nanoparticles |
178 |
with radii ranging from 10 - 25 \AA\ were created from a bulk fcc |
179 |
lattice. These particles were passivated |
180 |
with a 50\% coverage -- based on coverage densities reported by Badia \textit{et al.} -- of a selection of alkyl thiolates of varying |
181 |
chain lengths. The passivated particles were then solvated in hexane. |
182 |
Details of the models and simulation protocol follow in the next |
183 |
section. |
184 |
|
185 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
186 |
% COMPUTATIONAL DETAILS |
187 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
188 |
\section{Computational Details} |
189 |
|
190 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
191 |
% NON-PERIODIC VSS-RNEMD METHODOLOGY |
192 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
193 |
\subsection{Creating a thermal flux between particles and solvent} |
194 |
|
195 |
The non-periodic variant of VSS-RNEMD\cite{Stocker:2014qq} applies a |
196 |
series of velocity scaling and shearing moves at regular intervals to |
197 |
impose a flux between two concentric spherical regions. To impose a |
198 |
thermal flux between the shells (without an accompanying angular |
199 |
shear), we solve for scaling coefficients $a$ and $b$, |
200 |
\begin{eqnarray} |
201 |
a = \sqrt{1 - \frac{q_r \Delta t}{K_a - K_a^\mathrm{rot}}}\\ \nonumber\\ |
202 |
b = \sqrt{1 + \frac{q_r \Delta t}{K_b - K_b^\mathrm{rot}}} |
203 |
\end{eqnarray} |
204 |
at each time interval. These scaling coefficients conserve total |
205 |
kinetic energy and angular momentum subject to an imposed heat rate, |
206 |
$q_r$. The coefficients also depend on the instantaneous kinetic |
207 |
energy, $K_{\{a,b\}}$, and the total rotational kinetic energy of each |
208 |
shell, $K_{\{a,b\}}^\mathrm{rot} = \sum_i m_i \left( \mathbf{v}_i |
209 |
\times \mathbf{r}_i \right)^2 / 2$. |
210 |
|
211 |
The scaling coefficients are determined and the velocity changes are |
212 |
applied at regular intervals, |
213 |
\begin{eqnarray} |
214 |
\mathbf{v}_i \leftarrow a \left ( \mathbf{v}_i - \left < \omega_a \right > \times \mathbf{r}_i \right ) + \left < \omega_a \right > \times \mathbf{r}_i~~\:\\ |
215 |
\mathbf{v}_j \leftarrow b \left ( \mathbf{v}_j - \left < \omega_b \right > \times \mathbf{r}_j \right ) + \left < \omega_b \right > \times \mathbf{r}_j. |
216 |
\end{eqnarray} |
217 |
Here $\left < \omega_a \right > \times \mathbf{r}_i$ is the |
218 |
contribution to the velocity of particle $i$ due to the overall |
219 |
angular velocity of the $a$ shell. In the absence of an angular |
220 |
momentum flux, the angular velocity $\left < \omega_a \right >$ of the |
221 |
shell is nearly 0 and the resultant particle velocity is a nearly |
222 |
linear scaling of the initial velocity by the coefficient $a$ or $b$. |
223 |
|
224 |
Repeated application of this thermal energy exchange yields a radial |
225 |
temperature profile for the solvated nanoparticles that depends |
226 |
linearly on the applied heat rate, $q_r$. Similar to the behavior in |
227 |
the slab geometries, the temperature profiles have discontinuities at |
228 |
the interfaces between dissimilar materials. The size of the |
229 |
discontinuity depends on the interfacial thermal conductance, which is |
230 |
the primary quantity of interest. |
231 |
|
232 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
233 |
% CALCULATING TRANSPORT PROPERTIES |
234 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
235 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
236 |
% INTERFACIAL THERMAL CONDUCTANCE |
237 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
238 |
\subsection{Interfacial Thermal Conductance} |
239 |
|
240 |
As described in earlier work,\cite{Stocker:2014qq} the thermal |
241 |
conductance of each spherical shell may be defined as the inverse |
242 |
Kapitza resistance of the shell. To describe the thermal conductance |
243 |
of an interface of considerable thickness -- such as the ligand layers |
244 |
shown here -- we can sum the individual thermal resistances of each |
245 |
concentric spherical shell to arrive at the inverse of the total |
246 |
interfacial thermal conductance. In slab geometries, the intermediate |
247 |
temperatures cancel, but for concentric spherical shells, the |
248 |
intermediate temperatures and surface areas remain in the final sum, |
249 |
requiring the use of a series of individual resistance terms: |
250 |
|
251 |
\begin{equation} |
252 |
\frac{1}{G} = R_\mathrm{total} = \frac{1}{q_r} \sum_i \left(T_{i+i} - |
253 |
T_i\right) 4 \pi r_i^2. |
254 |
\end{equation} |
255 |
|
256 |
The longest ligand considered here is in excess of 15 \AA\ in length, |
257 |
and we use 10 concentric spherical shells to describe the total |
258 |
interfacial thermal conductance of the ligand layer. |
259 |
|
260 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
261 |
% FORCE FIELDS |
262 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
263 |
\subsection{Force Fields} |
264 |
|
265 |
Throughout this work, gold -- gold interactions are described by the |
266 |
quantum Sutton-Chen (QSC) model.\cite{PhysRevB.59.3527} Previous work\cite{kuang:AuThl} has demonstrated that the electronic contributions to heat conduction (which are missing from the QSC model) across heterogeneous metal / non-metal interfaces are negligible compared to phonon excitation, which is captured by the classical model. The hexane |
267 |
solvent is described by the TraPPE united atom |
268 |
model,\cite{TraPPE-UA.alkanes} where sites are located at the carbon |
269 |
centers for alkyl groups. The TraPPE-UA model for hexane provides both |
270 |
computational efficiency and reasonable accuracy for bulk thermal |
271 |
conductivity values. Bonding interactions were used for |
272 |
intra-molecular sites closer than 3 bonds. Effective Lennard-Jones |
273 |
potentials were used for non-bonded interactions. |
274 |
|
275 |
To describe the interactions between metal (Au) and non-metal atoms, |
276 |
potential energy terms were adapted from an adsorption study of alkyl |
277 |
thiols on gold surfaces by Vlugt, \textit{et |
278 |
al.}\cite{vlugt:cpc2007154} They fit an effective pair-wise |
279 |
Lennard-Jones form of potential parameters for the interaction between |
280 |
Au and pseudo-atoms CH$_x$ and S based on a well-established and |
281 |
widely-used effective potential of Hautman and Klein for the Au(111) |
282 |
surface.\cite{hautman:4994} |
283 |
|
284 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
285 |
% SIMULATION PROTOCOL |
286 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
287 |
\subsection{Simulation Protocol} |
288 |
|
289 |
Gold nanospheres with radii ranging from 10 - 25 \AA\ were created |
290 |
from a bulk fcc lattice and were thermally equilibrated prior to the |
291 |
addition of ligands. A 50\% coverage of ligands (based on coverages |
292 |
reported by Badia, \textit{et al.}\cite{Badia1996:2}) were placed on |
293 |
the surface of the equilibrated nanoparticles using |
294 |
Packmol\cite{packmol}. We have chosen three lengths of ligands: butanethiolate ($C_4$), octanethiolate ($C_8$), and dodecanethiolate ($C_{12}$). The nanoparticle / ligand complexes were |
295 |
thermally equilibrated to allow for ligand conformational flexibility. Packmol was then used to solvate the |
296 |
structures inside a spherical droplet of hexane. The thickness of the |
297 |
solvent layer was chosen to be at least 1.5$\times$ the combined |
298 |
radius of the nanoparticle / ligand structure. The fully solvated |
299 |
system was equilibrated for at least 1 ns using the Langevin Hull to |
300 |
apply 50 atm of pressure and a target temperature of 250 |
301 |
K.\cite{Vardeman2011} Typical system sizes ranged from 18,310 united |
302 |
atom sites for the 10 \AA\ particles with $C_4$ ligands to 89,490 sites |
303 |
for the 25 \AA\ particles with $C_{12}$ ligands. Figure |
304 |
\ref{fig:NP25_C12h1} shows one of the solvated 25 \AA\ nanoparticles |
305 |
passivated with the $C_{12}$ ligands. |
306 |
|
307 |
\begin{figure} |
308 |
\includegraphics[width=\linewidth]{figures/NP25_C12h1} |
309 |
\caption{A 25 \AA\ radius gold nanoparticle protected with a |
310 |
half-monolayer of TraPPE-UA dodecanethiolate (C$_{12}$) |
311 |
ligands and solvated in TraPPE-UA hexane. The interfacial |
312 |
thermal conductance is computed by applying a kinetic energy |
313 |
flux between the nanoparticle and an outer shell of |
314 |
solvent.} |
315 |
\label{fig:NP25_C12h1} |
316 |
\end{figure} |
317 |
|
318 |
Once equilibrated, thermal fluxes were applied for 1 ns, until stable |
319 |
temperature gradients had developed. Systems were run under moderate |
320 |
pressure (50 atm) with an average temperature (250K) that maintained a |
321 |
compact solvent cluster and avoided formation of a vapor layer near |
322 |
the heated metal surface. Pressure was applied to the system via the |
323 |
non-periodic Langevin Hull.\cite{Vardeman2011} However, thermal |
324 |
coupling to the external temperature bath was removed to avoid |
325 |
interference with the imposed RNEMD flux. |
326 |
|
327 |
\begin{figure} |
328 |
\includegraphics[width=\linewidth]{figures/temp_profile} |
329 |
\caption{Radial temperature profile for a 25 \AA\ radius particle protected with a 50\% coverage of TraPPE-UA butanethiolate (C$_4$) ligands and solvated in TraPPE-UA hexane. A kinetic energy flux is applied between RNEMD region A and RNEMD region B. The size of the temperature discontinuity at the interface is governed by the interfacial thermal conductance.} |
330 |
\label{fig:temp_profile} |
331 |
\end{figure} |
332 |
|
333 |
Because the method conserves \emph{total} angular momentum and energy, |
334 |
systems which contain a metal nanoparticle embedded in a significant |
335 |
volume of solvent will still experience nanoparticle diffusion inside |
336 |
the solvent droplet. To aid in measuring an accurate temperature |
337 |
profile for these systems, a single gold atom at the origin of the |
338 |
coordinate system was assigned a mass $10,000 \times$ its original |
339 |
mass. The bonded and nonbonded interactions for this atom remain |
340 |
unchanged and the heavy atom is excluded from the RNEMD velocity |
341 |
scaling. The only effect of this gold atom is to effectively pin the |
342 |
nanoparticle at the origin of the coordinate system, thereby |
343 |
preventing translational diffusion of the nanoparticle due to Brownian |
344 |
motion. |
345 |
|
346 |
To provide statistical independence, five separate configurations were |
347 |
simulated for each particle radius and ligand length. The |
348 |
structures were unique, starting at the point of ligand placement, |
349 |
in order to sample multiple surface-ligand configurations. |
350 |
|
351 |
|
352 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
353 |
% EFFECT OF PARTICLE SIZE |
354 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
355 |
\section{Results} |
356 |
|
357 |
We modeled four sizes of nanoparticles ($R =$ 10, 15, 20, and 25 |
358 |
\AA). The smallest particle size produces the lowest interfacial |
359 |
thermal conductance values for most of the of protecting groups |
360 |
(Fig. \ref{fig:NPthiols_G}). Between the other three sizes of |
361 |
nanoparticles, there is no discernible dependence of the interfacial |
362 |
thermal conductance on the nanoparticle size. It is likely that the |
363 |
differences in local curvature of the nanoparticle sizes studied here |
364 |
do not disrupt the ligand packing and behavior in drastically |
365 |
different ways. |
366 |
|
367 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
368 |
% EFFECT OF LIGAND CHAIN LENGTH |
369 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
370 |
|
371 |
We have also utilized half-monolayers of three lengths of |
372 |
alkanethiolate ligands -- S(CH$_2$)$_3$CH$_3$, S(CH$_2$)$_7$CH$_3$, |
373 |
and S(CH$_2$)$_{11}$CH$_3$ -- referred to as C$_4$, C$_8$, and |
374 |
C$_{12}$ respectively, in this study. Unlike our previous study of varying thiolate ligand chain lengths on |
375 |
planar Au(111) surfaces, the interfacial thermal conductance of |
376 |
ligand-protected nanospheres exhibits a distinct dependence on the |
377 |
ligand length. For the three largest particle sizes, a half-monolayer |
378 |
coverage of $C_4$ yields the highest interfacial thermal conductance |
379 |
and the next-longest ligand, $C_8$, provides a similar boost. The |
380 |
longest ligand, $C_{12}$, offers only a nominal ($\sim$ 10 \%) |
381 |
increase in the interfacial thermal conductance over the bare |
382 |
nanoparticles. |
383 |
|
384 |
\begin{figure} |
385 |
\includegraphics[width=\linewidth]{figures/NPthiols_G} |
386 |
\caption{Interfacial thermal conductance ($G$) values for 4 |
387 |
sizes of solvated nanoparticles that are bare or protected |
388 |
with a 50\% coverage of C$_{4}$, C$_{8}$, or C$_{12}$ |
389 |
alkanethiolate ligands.} |
390 |
\label{fig:NPthiols_G} |
391 |
\end{figure} |
392 |
|
393 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
394 |
% HEAT TRANSFER MECHANISMS |
395 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
396 |
\section{Mechanisms for Ligand-Enhanced Heat Transfer} |
397 |
|
398 |
corrugation |
399 |
|
400 |
escape rate |
401 |
|
402 |
orientation of ligand |
403 |
|
404 |
orientation of solvent |
405 |
|
406 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
407 |
% CORRUGATION OF PARTICLE SURFACE |
408 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
409 |
\subsection{Corrugation of Particle Surface} |
410 |
|
411 |
The bonding sites for thiols on gold surfaces have been studied |
412 |
extensively and include configurations beyond the traditional atop, |
413 |
bridge, and hollow sites found on planar surfaces. In particular, the |
414 |
deep potential well between the gold atoms and the thiolate sulfurs |
415 |
leads to insertion of the sulfur into the gold lattice and |
416 |
displacement of interfacial gold atoms. The degree of ligand-induced |
417 |
surface restructuring may have an impact on the interfacial thermal |
418 |
conductance and is an important phenomenon to quantify. |
419 |
|
420 |
Henz, \textit{et al.}\cite{Henz2007,Henz:2008qf} used the metal density as a |
421 |
function of radius to measure the degree of mixing between the thiol |
422 |
sulfurs and surface gold atoms at the edge of a nanoparticle. Although |
423 |
metal density is important, disruption of the local crystalline |
424 |
ordering would also have a large effect on the phonon spectrum in the |
425 |
particles. To measure this effect, we use the fraction of gold atoms |
426 |
exhibiting local fcc ordering as a function of radius to describe the |
427 |
ligand-induced disruption of the nanoparticle surface. |
428 |
|
429 |
The local bond orientational order can be described using the method |
430 |
of Steinhardt \textit{et al.}\cite{Steinhardt1983} The local bonding environment, $\bar{q}_{\ell m}$, for each |
431 |
atom in the system is determined by averaging over the spherical |
432 |
harmonics between that atom and each of its neighbors, |
433 |
\begin{equation} |
434 |
\bar{q}_{\ell m} = \sum_i Y_\ell^m(\theta_i, \phi_i) |
435 |
\end{equation} |
436 |
where $\theta_i$ and $\phi_i$ are the relative angular coordinates of |
437 |
neighbor $i$ in the laboratory frame. A global average orientational |
438 |
bond order parameter, $\bar{Q}_{\ell m}$, is the average over each |
439 |
$\bar{q}_{\ell m}$ for all atoms in the system. To remove the |
440 |
dependence on the laboratory coordinate frame, the third order |
441 |
rotationally invariant combination of $\bar{Q}_{\ell m}$, |
442 |
$\hat{w}_\ell$, is utilized here.\cite{Steinhardt1983,Vardeman:2008fk} |
443 |
|
444 |
For $\ell=4$, the ideal face-centered cubic (fcc), body-centered cubic |
445 |
(bcc), hexagonally close-packed (hcp), and simple cubic (sc) local |
446 |
structures exhibit $\hat{w}_4$ values of -0.159, 0.134, 0.159, and |
447 |
0.159, respectively. Because $\hat{w}_4$ exhibits an extreme value for |
448 |
fcc structures, it is ideal for measuring local fcc |
449 |
ordering. The spatial distribution of $\hat{w}_4$ local bond |
450 |
orientational order parameters, $p(\hat{w}_4 , r)$, can provide |
451 |
information about the location of individual atoms that are central to |
452 |
local fcc ordering. |
453 |
|
454 |
The fraction of fcc-ordered gold atoms at a given radius in the |
455 |
nanoparticle, |
456 |
\begin{equation} |
457 |
f_\mathrm{fcc}(r) = \int_{-\infty}^{w_c} p(\hat{w}_4, r) d \hat{w}_4 |
458 |
\end{equation} |
459 |
is described by the distribution of the local bond orientational order |
460 |
parameters, $p(\hat{w}_4, r)$, and $w_c$, a cutoff for the peak |
461 |
$\hat{w}_4$ value displayed by fcc structures. A $w_c$ value of -0.12 |
462 |
was chosen to isolate the fcc peak in $\hat{w}_4$. |
463 |
|
464 |
As illustrated in Figure \ref{fig:Corrugation}, the presence of |
465 |
ligands decreases the fcc ordering of the gold atoms at the |
466 |
nanoparticle surface. For the smaller nanoparticles, this disruption |
467 |
extends into the core of the nanoparticle, indicating widespread |
468 |
disruption of the lattice. |
469 |
|
470 |
\begin{figure} |
471 |
\includegraphics[width=\linewidth]{figures/NP10_fcc} |
472 |
\caption{Fraction of gold atoms with fcc ordering as a |
473 |
function of radius for a 10 \AA\ radius nanoparticle. The |
474 |
decreased fraction of fcc-ordered atoms in ligand-protected |
475 |
nanoparticles relative to bare particles indicates |
476 |
restructuring of the nanoparticle surface by the thiolate |
477 |
sulfur atoms.} |
478 |
\label{fig:Corrugation} |
479 |
\end{figure} |
480 |
|
481 |
We may describe the thickness of the disrupted nanoparticle surface by |
482 |
defining a corrugation factor, $c$, as the ratio of the radius at |
483 |
which the fraction of gold atoms with fcc ordering is 0.9 and the |
484 |
radius at which the fraction is 0.5. |
485 |
|
486 |
\begin{equation} |
487 |
c = 1 - \frac{r(f_\mathrm{fcc} = 0.9)}{r(f_\mathrm{fcc} = 0.5)} |
488 |
\end{equation} |
489 |
|
490 |
A sharp interface will have a steep drop in $f_\mathrm{fcc}$ at the |
491 |
edge of the particle ($c \rightarrow$ 0). In the opposite limit where |
492 |
the entire nanoparticle surface is restructured by ligands, the radius |
493 |
at which there is a high probability of fcc ordering moves |
494 |
dramatically inward ($c \rightarrow$ 1). |
495 |
|
496 |
The computed corrugation factors are shown in Figure |
497 |
\ref{fig:NPthiols_combo} for bare nanoparticles and for |
498 |
ligand-protected particles as a function of ligand chain length. The |
499 |
largest nanoparticles are only slightly restructured by the presence |
500 |
of ligands on the surface, while the smallest particle ($r$ = 10 \AA) |
501 |
exhibits significant disruption of the original fcc ordering when |
502 |
covered with a half-monolayer of thiol ligands. |
503 |
|
504 |
Because the thiolate ligands do not significantly alter the larger |
505 |
particle crystallinity, the surface corrugation does not seem to be a |
506 |
likely candidate to explain the large increase in thermal conductance |
507 |
at the interface when ligands are added. |
508 |
|
509 |
% \begin{equation} |
510 |
% C = \frac{r_{bare}(\rho_{\scriptscriptstyle{0.85}}) - r_{capped}(\rho_{\scriptscriptstyle{0.85}})}{r_{bare}(\rho_{\scriptscriptstyle{0.85}})}. |
511 |
% \end{equation} |
512 |
% |
513 |
% Here, $r_{bare}(\rho_{\scriptscriptstyle{0.85}})$ is the radius of a bare nanoparticle at which the density is $85\%$ the bulk value and $r_{capped}(\rho_{\scriptscriptstyle{0.85}})$ is the corresponding radius for a particle of the same size with a layer of ligands. $C$ has a value of 0 for a bare particle and approaches $1$ as the degree of surface atom mixing increases. |
514 |
|
515 |
|
516 |
|
517 |
\begin{figure} |
518 |
\includegraphics[width=\linewidth]{figures/NPthiols_combo} |
519 |
\caption{Computed corrugation values, solvent escape rates, |
520 |
ligand orientational $P_2$ values, and interfacial solvent |
521 |
orientational $P_2$ values for 4 sizes of solvated |
522 |
nanoparticles that are bare or protected with a 50\% |
523 |
coverage of C$_{4}$, C$_{8}$, or C$_{12}$ alkanethiolate |
524 |
ligands.} |
525 |
\label{fig:NPthiols_combo} |
526 |
\end{figure} |
527 |
|
528 |
|
529 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
530 |
% MOBILITY OF INTERFACIAL SOLVENT |
531 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
532 |
\subsection{Mobility of Interfacial Solvent} |
533 |
|
534 |
Another possible mechanism for increasing interfacial conductance is |
535 |
the mobility of the interfacial solvent. We used a survival |
536 |
correlation function, $C(t)$, to measure the residence time of a |
537 |
solvent molecule in the nanoparticle thiolate |
538 |
layer.\cite{Stocker:2013cl} This function correlates the identity of |
539 |
all hexane molecules within the radial range of the thiolate layer at |
540 |
two separate times. If the solvent molecule is present at both times, |
541 |
the configuration contributes a $1$, while the absence of the molecule |
542 |
at the later time indicates that the solvent molecule has migrated |
543 |
into the bulk, and this configuration contributes a $0$. A steep decay |
544 |
in $C(t)$ indicates a high turnover rate of solvent molecules from the |
545 |
chain region to the bulk. We may define the escape rate for trapped |
546 |
solvent molecules at the interface as |
547 |
\begin{equation} |
548 |
k_\mathrm{escape} = \left( \int_0^T C(t) dt \right)^{-1} |
549 |
\label{eq:mobility} |
550 |
\end{equation} |
551 |
where T is the length of the simulation. This is a direct measure of |
552 |
the rate at which solvent molecules initially entangled in the |
553 |
thiolate layer can escape into the bulk. When $k_\mathrm{escape} |
554 |
\rightarrow 0$, the solvent becomes permanently trapped in the |
555 |
interfacial region. |
556 |
|
557 |
The solvent escape rates for bare and ligand-protected nanoparticles |
558 |
are shown in Figure \ref{fig:NPthiols_combo}. As the ligand chain |
559 |
becomes longer and more flexible, interfacial solvent molecules become |
560 |
trapped in the ligand layer and the solvent escape rate decreases. |
561 |
This mechanism contributes a partial explanation as to why the longer |
562 |
ligands have significantly lower thermal conductance. |
563 |
|
564 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
565 |
% ORIENTATION OF LIGAND CHAINS |
566 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
567 |
\subsection{Orientation of Ligand Chains} |
568 |
|
569 |
As the ligand chain length increases in length, it exhibits |
570 |
significantly more conformational flexibility. Thus, different lengths |
571 |
of ligands should favor different chain orientations on the surface of |
572 |
the nanoparticle. To determine the distribution of ligand orientations |
573 |
relative to the particle surface we examine the probability of |
574 |
finding a ligand with a particular orientation relative to the surface |
575 |
normal of the nanoparticle, |
576 |
\begin{equation} |
577 |
\cos{(\theta)}=\frac{\vec{r}_i\cdot\hat{u}_i}{|\vec{r}_i||\hat{u}_i|} |
578 |
\end{equation} |
579 |
where $\vec{r}_{i}$ is the vector between the cluster center of mass |
580 |
and the sulfur atom on ligand molecule {\it i}, and $\hat{u}_{i}$ is |
581 |
the vector between the sulfur atom and \ce{CH3} pseudo-atom on ligand |
582 |
molecule {\it i}. As depicted in Figure \ref{fig:NP_pAngle}, $\theta |
583 |
\rightarrow 180^{\circ}$ for a ligand chain standing upright on the |
584 |
particle ($\cos{(\theta)} \rightarrow -1$) and $\theta \rightarrow |
585 |
90^{\circ}$ for a ligand chain lying down on the surface |
586 |
($\cos{(\theta)} \rightarrow 0$). As the thiolate alkane chain |
587 |
increases in length and becomes more flexible, the ligands are more |
588 |
willing to lie down on the nanoparticle surface and exhibit increased |
589 |
population at $\cos{(\theta)} = 0$. |
590 |
|
591 |
\begin{figure} |
592 |
\includegraphics[width=\linewidth]{figures/NP_pAngle} |
593 |
\caption{The two extreme cases of ligand orientation relative |
594 |
to the nanoparticle surface: the ligand completely |
595 |
outstretched ($\cos{(\theta)} = -1$) and the ligand fully |
596 |
lying down on the particle surface ($\cos{(\theta)} = 0$).} |
597 |
\label{fig:NP_pAngle} |
598 |
\end{figure} |
599 |
|
600 |
% \begin{figure} |
601 |
% \includegraphics[width=\linewidth]{figures/thiol_pAngle} |
602 |
% \caption{} |
603 |
% \label{fig:thiol_pAngle} |
604 |
% \end{figure} |
605 |
|
606 |
An order parameter describing the average ligand chain orientation relative to |
607 |
the nanoparticle surface is available using the second order Legendre |
608 |
parameter, |
609 |
\begin{equation} |
610 |
P_2 = \left< \frac{1}{2} \left(3\cos^2(\theta) - 1 \right) \right> |
611 |
\end{equation} |
612 |
|
613 |
Ligand populations that are perpendicular to the particle surface have |
614 |
$P_2$ values of 1, while ligand populations lying flat on the |
615 |
nanoparticle surface have $P_2$ values of $-0.5$. Disordered ligand |
616 |
layers will exhibit mean $P_2$ values of 0. As shown in Figure |
617 |
\ref{fig:NPthiols_combo} the ligand $P_2$ values approaches 0 as |
618 |
ligand chain length -- and ligand flexibility -- increases. |
619 |
|
620 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
621 |
% ORIENTATION OF INTERFACIAL SOLVENT |
622 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
623 |
\subsection{Orientation of Interfacial Solvent} |
624 |
|
625 |
Similarly, we examined the distribution of \emph{hexane} molecule |
626 |
orientations relative to the particle surface using the same angular |
627 |
analysis utilized for the ligand chain orientations. In this case, |
628 |
$\vec{r}_i$ is the vector between the particle center of mass and one |
629 |
of the \ce{CH2} pseudo-atoms in the middle of hexane molecule $i$ and |
630 |
$\hat{u}_i$ is the vector between the two \ce{CH3} pseudo-atoms on |
631 |
molecule $i$. Since we are only interested in the orientation of |
632 |
solvent molecules near the ligand layer, we select only the hexane |
633 |
molecules within a specific $r$-range, between the edge of the |
634 |
particle and the end of the ligand chains. A large population of |
635 |
hexane molecules with $\cos{(\theta)} \sim \pm 1$ would indicate |
636 |
interdigitation of the solvent molecules between the upright ligand |
637 |
chains. A more random distribution of $\cos{(\theta)}$ values |
638 |
indicates a disordered arrangement of solvent molecules near the particle |
639 |
surface. Again, $P_2$ order parameter values provide a population |
640 |
analysis for the solvent that is close to the particle surface. |
641 |
|
642 |
The average orientation of the interfacial solvent molecules is |
643 |
notably flat on the \emph{bare} nanoparticle surfaces. This blanket of |
644 |
hexane molecules on the particle surface may act as an insulating |
645 |
layer, increasing the interfacial thermal resistance. As the length |
646 |
(and flexibility) of the ligand increases, the average interfacial |
647 |
solvent P$_2$ value approaches 0, indicating a more random orientation |
648 |
of the ligand chains. The average orientation of solvent within the |
649 |
$C_8$ and $C_{12}$ ligand layers is essentially random. Solvent |
650 |
molecules in the interfacial region of $C_4$ ligand-protected |
651 |
nanoparticles do not lie as flat on the surface as in the case of the |
652 |
bare particles, but are not as randomly oriented as the longer ligand |
653 |
lengths. |
654 |
|
655 |
These results are particularly interesting in light of our previous |
656 |
results\cite{Stocker:2013cl}, where solvent molecules readily filled |
657 |
the vertical gaps between neighboring ligand chains and there was a |
658 |
strong correlation between ligand and solvent molecular |
659 |
orientations. It appears that the introduction of surface curvature |
660 |
and a lower ligand packing density creates a disordered ligand layer |
661 |
that lacks well-formed channels for the solvent molecules to occupy. |
662 |
|
663 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
664 |
% SOLVENT PENETRATION OF LIGAND LAYER |
665 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
666 |
\subsection{Solvent Penetration of Ligand Layer} |
667 |
|
668 |
We may also determine the extent of ligand -- solvent interaction by |
669 |
calculating the hexane density as a function of radius. Figure |
670 |
\ref{fig:hex_density} shows representative radial hexane density |
671 |
profiles for a solvated 25 \AA\ radius nanoparticle with no ligands, |
672 |
and 50\% coverage of C$_{4}$, C$_{8}$, and C$_{12}$ thiolates. |
673 |
|
674 |
\begin{figure} |
675 |
\includegraphics[width=\linewidth]{figures/hex_density} |
676 |
\caption{Radial hexane density profiles for 25 \AA\ radius |
677 |
nanoparticles with no ligands (circles), C$_{4}$ ligands |
678 |
(squares), C$_{8}$ ligands (triangles), and C$_{12}$ ligands |
679 |
(diamonds). As ligand chain length increases, the nearby |
680 |
solvent is excluded from the ligand layer. Some solvent is |
681 |
present inside the particle $r_{max}$ location due to |
682 |
faceting of the nanoparticle surface.} |
683 |
\label{fig:hex_density} |
684 |
\end{figure} |
685 |
|
686 |
The differences between the radii at which the hexane surrounding the |
687 |
ligand-covered particles reaches bulk density correspond nearly |
688 |
exactly to the differences between the lengths of the ligand |
689 |
chains. Beyond the edge of the ligand layer, the solvent reaches its |
690 |
bulk density within a few angstroms. The differing shapes of the |
691 |
density curves indicate that the solvent is increasingly excluded from |
692 |
the ligand layer as the chain length increases. |
693 |
|
694 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
695 |
% DISCUSSION |
696 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
697 |
\section{Discussion} |
698 |
|
699 |
The chemical bond between the metal and the ligand introduces |
700 |
vibrational overlap that is not present between the bare metal surface |
701 |
and solvent. Thus, regardless of ligand chain length, the presence of |
702 |
a half-monolayer ligand coverage yields a higher interfacial thermal |
703 |
conductance value than the bare nanoparticle. The shortest and least |
704 |
flexible ligand ($C_4$), which exhibits the highest interfacial |
705 |
thermal conductance value, has a smaller range of available angles relative to |
706 |
the surface normal. The longer $C_8$ and $C_{12}$ ligands have |
707 |
increasingly disordered orientations and correspondingly lower solvent |
708 |
escape rates. When the ligands are less tightly packed, the cooperative |
709 |
orientational ordering between the ligand and solvent decreases |
710 |
dramatically. |
711 |
|
712 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
713 |
% **ACKNOWLEDGMENTS** |
714 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% |
715 |
\begin{acknowledgement} |
716 |
Support for this project was provided by the National Science Foundation |
717 |
under grant CHE-1362211. Computational time was provided by the |
718 |
Center for Research Computing (CRC) at the University of Notre Dame. |
719 |
\end{acknowledgement} |
720 |
|
721 |
|
722 |
\newpage |
723 |
|
724 |
\bibliography{NPthiols} |
725 |
|
726 |
\end{document} |