ViewVC Help
View File | Revision Log | Show Annotations | View Changeset | Root Listing
root/group/trunk/nanoglass/analysis.tex
Revision: 3247
Committed: Thu Oct 4 21:11:58 2007 UTC (16 years, 11 months ago) by gezelter
Content type: application/x-tex
File size: 13373 byte(s)
Log Message:
new stuff

File Contents

# User Rev Content
1 chuckv 3226 %!TEX root = /Users/charles/Desktop/nanoglass/nanoglass.tex
2    
3 chuckv 3213 \section{Analysis}
4    
5 gezelter 3242 Frank first proposed local icosahedral ordering of atoms as an
6     explanation for supercooled atomic (specifically metallic) liquids,
7     and further showed that a 13-atom icosahedral cluster has a 8.4\%
8     higher binding energy the either a face centered cubic ({\sc fcc}) or
9     hexagonal close-packed ({\sc hcp}) crystal structures.\cite{19521106}
10     Icosahedra also have six five-fold symmetry axes that cannot be
11     extended indefinitely in three dimensions, which makes them long-range
12     translational order incommensurate with local icosahedral ordering.
13     This does not preclude icosahedral clusters from possessing long-range
14     {\it orientational} order. The ``frustrated'' packing of these
15     icosahedral structures into dense clusters has been proposed as a
16     model for glass formation.\cite{19871127} The size of the icosahedral
17     clusters is thought to increase until frustration prevents any further
18     growth.\cite{HOARE:1976fk} Molecular dynamics simulations of a
19     two-component Lennard-Jones glass showed that clusters of face-sharing
20     icosahedra are distributed throughout the
21 gezelter 3230 material.\cite{PhysRevLett.60.2295} Simulations of freezing of single
22     component metalic nanoclusters have shown a tendency for icosohedral
23     structure formation particularly at the surfaces of these
24     clusters.\cite{Gafner:2004bg,PhysRevLett.89.275502,Ascencio:2000qy,Chen:2004ec}
25     Experimentally, the splitting (or shoulder) on the second peak of the
26     X-ray structure factor in binary metallic glasses has been attributed
27     to the formation of tetrahedra that share faces of adjoining
28     icosahedra.\cite{Waal:1995lr}
29 chuckv 3226
30 gezelter 3230 Various structural probes have been used to characterize structural
31 gezelter 3242 order in molecular systems including: common neighbor analysis,
32     Voronoi tesselations, and orientational bond-order
33 gezelter 3230 parameters.\cite{HoneycuttJ.Dana_j100303a014,Iwamatsu:2007lr,hsu:4974,nose:1803}
34 gezelter 3242 The method that has been used most extensively for determining local
35     and extended orientational symmetry in condensed phases is the
36     bond-orientational analysis formulated by Steinhart {\it et
37     al.}\cite{Steinhardt:1983mo} In this model, a set of spherical
38 gezelter 3230 harmonics is associated with each of the near neighbors of a central
39     atom. Neighbors (or ``bonds'') are defined as having a distance from
40     the central atom that is within the first peak in the radial
41     distribution function. The spherical harmonic between a central atom
42     $i$ and a neighboring atom $j$ is
43 chuckv 3213 \begin{equation}
44 gezelter 3230 Y_{lm}\left(\vec{r}_{ij}\right)=Y_{lm}\left(\theta(\vec{r}_{ij}),\phi(\vec{r}_{ij})\right)
45     \label{eq:spharm}
46 chuckv 3213 \end{equation}
47 gezelter 3230 where, $\{ Y_{lm}(\theta,\phi)\}$ are spherical harmonics, and
48     $\theta(\vec{r})$ and $\phi(\vec{r})$ are the polar and azimuthal
49     angles made by the bond vector $\vec{r}$ with respect to a reference
50     coordinate system. We chose for simplicity the origin as defined by
51     the coordinates for our nanoparticle. (Only even-$l$ spherical
52     harmonics are considered since permutation of a pair of identical
53     particles should not affect the bond-order parameter.) The local
54     environment surrounding atom $i$ can be defined by
55     the average over all neighbors, $N_b(i)$, surrounding that atom,
56 chuckv 3213 \begin{equation}
57 gezelter 3230 \bar{q}_{lm}(i) = \frac{1}{N_{b}(i)}\sum_{j=1}^{N_b(i)} Y_{lm}(\vec{r}_{ij}).
58     \label{eq:local_avg_bo}
59 chuckv 3213 \end{equation}
60 gezelter 3230 We can further define a global average orientational-bond order over
61     all $\bar{q}_{lm}$ by calculating the average of $\bar{q}_{lm}(i)$
62     over all $N$ particles
63 chuckv 3213 \begin{equation}
64 gezelter 3230 \bar{Q}_{lm} = \frac{\sum^{N}_{i=1}N_{b}(i)\bar{q}_{lm}(i)}{\sum^{N}_{i=1}N_{b}(i)}
65     \label{eq:sys_avg_bo}
66 chuckv 3213 \end{equation}
67 gezelter 3233 The $\bar{Q}_{lm}$ contained in equation \ref{eq:sys_avg_bo} is not
68     necessarily invariant under rotations of the arbitrary reference
69     coordinate system. Second- and third-order rotationally invariant
70     combinations, $Q_l$ and $W_l$, can be taken by summing over $m$ values
71     of $\bar{Q}_{lm}$,
72 chuckv 3222 \begin{equation}
73 gezelter 3230 Q_{l} = \left( \frac{4\pi}{2 l+1} \sum^{l}_{m=-l} \left| \bar{Q}_{lm} \right|^2\right)^{1/2}
74     \label{eq:sec_ord_inv}
75 chuckv 3222 \end{equation}
76     and
77     \begin{equation}
78     \hat{W}_{l} = \frac{W_{l}}{\left( \sum^{l}_{m=-l} \left| \bar{Q}_{lm} \right|^2\right)^{3/2}}
79     \label{eq:third_ord_inv}
80     \end{equation}
81 gezelter 3230 where
82 chuckv 3222 \begin{equation}
83 gezelter 3230 W_{l} = \sum_{\substack{m_1,m_2,m_3 \\m_1 + m_2 + m_3 = 0}} \left( \begin{array}{ccc} l & l & l \\ m_1 & m_2 & m_3 \end{array}\right) \\ \times \bar{Q}_{lm_1}\bar{Q}_{lm_2}\bar{Q}_{lm_3}.
84 chuckv 3222 \label{eq:third_inv}
85     \end{equation}
86 gezelter 3230 The factor in parentheses in Eq. \ref{eq:third_inv} is the Wigner-3$j$
87 gezelter 3233 symbol, and the sum is over all valid ($|m| \leq l$) values of $m_1$,
88     $m_2$, and $m_3$ which sum to zero.
89 chuckv 3226
90     \begin{table}
91 gezelter 3230 \caption{Values of bond orientational order parameters for
92     simple structures cooresponding to $l=4$ and $l=6$ spherical harmonic
93     functions.\cite{wolde:9932} $Q_4$ and $\hat{W}_4$ have values for {\it
94     individual} icosahedral clusters, but these values are not invariant
95     under rotations of the reference coordinate systems. Similar behavior
96     is observed in the bond-orientational order parameters for individual
97     liquid-like structures.}
98 chuckv 3226 \begin{center}
99     \begin{tabular}{ccccc}
100     \hline
101     \hline
102     & $Q_4$ & $Q_6$ & $\hat{W}_4$ & $\hat{W}_6$\\
103    
104     fcc & 0.191 & 0.575 & -0.159 & -0.013\\
105    
106     hcp & 0.097 & 0.485 & 0.134 & -0.012\\
107    
108     bcc & 0.036 & 0.511 & 0.159 & 0.013\\
109    
110     sc & 0.764 & 0.354 & 0.159 & 0.013\\
111    
112 gezelter 3230 Icosahedral & - & 0.663 & - & -0.170\\
113 chuckv 3226
114 gezelter 3230 (liquid) & - & - & - & -\\
115 chuckv 3226 \hline
116     \end{tabular}
117     \end{center}
118     \label{table:bopval}
119     \end{table}
120    
121 gezelter 3230 For ideal face-centered cubic ({\sc fcc}), body-centered cubic ({\sc
122     bcc}) and simple cubic ({\sc sc}) as well as hexagonally close-packed
123     ({\sc hcp}) structures, these rotationally invariant bond order
124     parameters have fixed values independent of the choice of coordinate
125     reference frames. For ideal icosahedral structures, the $l=6$
126     invariants, $Q_6$ and $\hat{W}_6$ are also independent of the
127     coordinate system. $Q_4$ and $\hat{W}_4$ will have non-vanishing
128     values for {\it individual} icosahedral clusters, but these values are
129     not invariant under rotations of the reference coordinate systems.
130     Similar behavior is observed in the bond-orientational order
131     parameters for individual liquid-like structures.
132    
133 gezelter 3233 Additionally, both $Q_6$ and $\hat{W}_6$ are thought to have extreme
134     values for the icosahedral clusters.\cite{Steinhardt:1983mo} This
135     makes the $l=6$ bond-orientational order parameters particularly
136     useful in identifying the extent of local icosahedral ordering in
137     condensed phases. For example, a local structure which exhibits
138     $\hat{W}_6$ values near -0.17 is easily identified as an icosahedral
139     cluster and cannot be mistaken for distorted cubic or liquid-like
140     structures.
141 gezelter 3230
142 gezelter 3233 One may use these bond orientational order parameters as an averaged
143     property to obtain the extent of icosahedral ordering in a supercooled
144     liquid or cluster. It is also possible to accumulate information
145     about the {\it distributions} of local bond orientational order
146     parameters, $p(\hat{W}_6)$ and $p(Q_6)$, which provide information
147     about individual atomic sites that are central to local icosahedral
148     structures.
149 gezelter 3230
150 gezelter 3233 The distributions of atomic $Q_6$ and $\hat{W}_6$ values are plotted
151     as a function of temperature for our nanoparticles in figures
152     \ref{fig:q6} and \ref{fig:w6}. At high temperatures, the
153     distributions are unstructured and are broadly distributed across the
154     entire range of values. As the particles are cooled, however, there
155     is a dramatic increase in the fraction of atomic sites which have
156     local icosahedral ordering around them. (This corresponds to the
157     sharp peak appearing in figure \ref{fig:w6} at $\hat{W}_6=-0.17$ and
158     to the broad shoulder appearing in figure \ref{fig:q6} at $Q_6 =
159     0.663$.)
160    
161     \begin{figure}[htbp]
162     \centering
163 gezelter 3242 \includegraphics[width=5in]{images/w6_stacked_plot.pdf}
164 gezelter 3233 \caption{Distributions of the bond orientational order parameter
165     ($\hat{W}_6$) at different temperatures. The upper, middle, and lower
166     panels are for 20, 30, and 40 \AA\ particles, respectively. The
167     left-hand column used cooling rates commensurate with a low
168     interfacial conductance ($87.5 \times 10^{6}$
169     $\mathrm{Wm^{-2}K^{-1}}$), while the right-hand column used a more
170     physically reasonable value of $117 \times 10^{6}$
171     $\mathrm{Wm^{-2}K^{-1}}$. The peak at $\hat{W}_6 \approx -0.17$ is
172 gezelter 3239 due to local icosahedral structures. The different curves in each of
173     the panels indicate the distribution of $\hat{W}_6$ values for samples
174     taken at different times along the cooling trajectory. The initial
175     and final temperatures (in K) are indicated on the plots adjacent to
176     their respective distributions.}
177 gezelter 3233 \label{fig:w6}
178     \end{figure}
179    
180     \begin{figure}[htbp]
181     \centering
182 gezelter 3242 \includegraphics[width=5in]{images/q6_stacked_plot.pdf}
183 gezelter 3233 \caption{Distributions of the bond orientational order parameter
184     ($Q_6$) at different temperatures. The curves in the six panels in
185     this figure were computed at identical conditions to the same panels in
186     figure \ref{fig:w6}.}
187     \label{fig:q6}
188     \end{figure}
189    
190 gezelter 3242 We have also calculated the fraction of atomic centers which have
191     strong local icosahedral order:
192 gezelter 3233 \begin{equation}
193     f_\textrm{icos} = \int_{-\infty}^{w_i} p(\hat{W}_6) d \hat{W}_6
194     \label{eq:ficos}
195     \end{equation}
196     where $w_i$ is a cutoff value in $\hat{W}_6$ for atomic centers that
197     are displaying icosahedral environments. We have chosen a (somewhat
198     arbitrary) value of $w_i= -0.15$ for the purposes of this work. A
199     plot of $f_\textrm{icos}(T)$ as a function of temperature of the
200     particles is given in figure \ref{fig:ficos}. As the particles cool,
201     the fraction of local icosahedral ordering rises smoothly to a plateau
202     value. The larger particles (particularly the ones that were cooled
203 gezelter 3242 in a lower viscosity solvent) show a slightly smaller tendency towards
204     icosahedral ordering.
205 gezelter 3233
206     \begin{figure}[htbp]
207     \centering
208 gezelter 3242 \includegraphics[width=5in]{images/fraction_icos.pdf}
209 gezelter 3233 \caption{Temperautre dependence of the fraction of atoms with local
210     icosahedral ordering, $f_\textrm{icos}(T)$ for 20, 30, and 40 \AA\
211     particles cooled at two different values of the interfacial
212     conductance.}
213 gezelter 3242 \label{fig:ficos}
214 gezelter 3233 \end{figure}
215    
216     Since we have atomic-level resolution of the local bond-orientational
217     ordering information, we can also look at the local ordering as a
218     function of the identities of the central atoms. In figure
219     \ref{fig:AgVsCu} we display the distributions of $\hat{W}_6$ values
220     for both the silver and copper atoms, and we note a strong
221 gezelter 3242 predilection for the copper atoms to be central to icosahedra. This
222     is probably due to local packing competition of the larger silver
223     atoms around the copper, which would tend to favor icosahedral
224     structures over the more densely packed cubic structures.
225 gezelter 3233
226     \begin{figure}[htbp]
227     \centering
228 gezelter 3242 \includegraphics[width=5in]{images/w6_stacked_bytype_plot.pdf}
229 gezelter 3233 \caption{Distributions of the bond orientational order parameter
230     ($\hat{W}_6$) for the two different elements present in the
231     nanoparticles. This distribution was taken from the fully-cooled 40
232     \AA\ nanoparticle. Local icosahedral ordering around copper atoms is
233     much more prevalent than around silver atoms.}
234 gezelter 3242 \label{fig:AgVsCu}
235 gezelter 3233 \end{figure}
236 gezelter 3242
237     Additionally, we have observed that those silver atoms that {\it do}
238     form local icosahedral structures are usually on the surface of the
239     nanoparticle, while the copper atoms which have local icosahedral
240     ordering are distributed more evenly throughout the nanoparticles.
241 gezelter 3247 Figure \ref{fig:Surface} shows this tendency as a function of distance
242     from the center of the nanoparticle. Silver, since it has a lower
243     surface free energy than copper, tends to coat the skins of the mixed
244     particles.\cite{Zhu:1997lr} This is true even for bimetallic particles
245     that have been prepared in the Ag (core) / Cu (shell) configuration.
246     Upon forming a liquid droplet, approximately 1 monolayer of Ag atoms
247     will rise to the surface of the particles. This can be seen visually
248     in figure \ref{fig:cross_sections} as well as in the density plots in
249     the bottom panel of figure \ref{fig:Surface}. This observation is
250     consistent with previous experimental and theoretical studies on
251     bimetallic alloys composed of noble
252     metals.\cite{MainardiD.S._la0014306,HuangS.-P._jp0204206,Ramirez-Caballero:2006lr}
253     Bond order parameters for surface atoms are averaged only over the
254     neighboring atoms, so packing constraints that may prevent icosahedral
255     ordering around silver in the bulk are removed near the surface. It
256     would certainly be interesting to see if the relative tendency of
257     silver and copper to form local icosahedral structures in a bulk glass
258     differs from our observations on nanoparticles.
259    
260     \begin{figure}[htbp]
261     \centering
262     \includegraphics[width=5in]{images/dens_fracr_stacked_plot.pdf}
263     \caption{Appearance of icosahedral clusters around central silver atoms
264     is largely due to the presence of these silver atoms at or near the
265     surface of the nanoparticle. The upper panel shows the fraction of
266     icosahedral atoms ($f_\textrm{icos}(r)$ for each of the two metallic
267     atoms as a function of distance from the center of the nanoparticle
268     ($r$). The lower panel shows the radial density of the two
269     constituent metals (relative to the overall density of the
270     nanoparticle). Icosahedral clustering around copper atoms are more
271     evenly distributed throughout the particle, while icosahedral
272     clustering around silver is largely confined to the silver atoms at
273     the surface.}
274     \label{fig:Surface}
275     \end{figure}