1: %%%%%%%%%%%%%%%%%%%%%%% file template.tex %%%%%%%%%%%%%%%%%%%%%%%%%
2: %
3: % This is a template file for Granular Matter
4: %
5: % Copy it to a new file with a new name and use it as the basis
6: % for your article
7: %
8: %%%%%%%%%%%%%%%%%%%%%%%% Springer-Verlag %%%%%%%%%%%%%%%%%%%%%%%%%%
9: %
10: \begin{filecontents}{leer.eps}
11: %!PS-Adobe-2.0 EPSF-2.0
12: %%CreationDate: Mon Jul 13 16:51:17 1992
13: %%DocumentFonts: (atend)
14: %%Pages: 0 1
15: %%BoundingBox: 72 31 601 342
16: %%EndComments
17:
18: gsave
19: 72 31 moveto
20: 72 342 lineto
21: 601 342 lineto
22: 601 31 lineto
23: 72 31 lineto
24: showpage
25: grestore
26: %%Trailer
27: %%DocumentFonts: Helvetica
28: \end{filecontents}
29: %
30: %\documentclass[granma,referee]{svjour}
31: \documentclass[granma]{svjour}
32: % Remove option referee for final version
33: %
34: % Remove any % below to load the required packages
35: %\usepackage{latexsym}
36: \usepackage{graphicx}
37: % etc
38: %
39: \begin{document}
40: %
41: \title{A note on the upward and downward intruder segregation in granular media}
42: %\headnote{Short communication}
43: %\subtitle{Do you have a subtitle?\\ If so, write it here}
44: %
45: \author{Leonardo Trujillo$^{1}$ and Hans J. Herrmann$^{1,2}$}
46: %
47: \institute{
48: $^1$Physique et M\'ecanique des Milieux H\'et\'erog\`enes\\
49: \'Ecole Sup\'erieure de Physique et de Chimie Industrielles,\\
50: 10, rue Vauquelin, 75231 Paris Cedex 05, France\\
51: \\
52: $^{2}$Institut f\"ur Computeranwendungen 1\\
53: Universit\"at Stuttgart\\
54: Pfaffenwaldring 27, 70569 Stuttgart, Germany
55: }
56:
57:
58:
59: %
60: \date{Received: date / Revised version: date}
61: % The correct dates will be entered by Springer
62: %
63: \maketitle
64: %
65: \begin{abstract}
66: The intruder segregation dependence on size and density is investigated in the framework of a
67: hydrodynamic theoretical model for vibrated granular media. We propose a segregation mechanism
68: based on the difference of densities between different regions of the granular system, which give origin
69: to a buoyant force that acts on the intruder. From the analytic solution of the segregation velocity we can
70: analyze the transition from the upward to downward intruder's movement.
71: \end{abstract}
72: %
73: %\section{Introduction}
74: %\label{intro}
75: The understanding of the behavior of granular segregation is relevant due to their practical
76: importance in many industries\cite{Herrmann98,deGennes98}.
77: When a granular mixture is subject to vertical vibrations under gravity, the grains
78: tends to segregate with the larger particles at the top of the bed.
79: Many studies have been devoted to this subject, usually referred as the
80: ``Brazil nut effect"\cite{Rippie64,Ahmad73,Rosato87,Jullien92,Duran93,Knight93,Jullien93,Duran94,Poschel95,Gallas96,Cooke96,Vanel97,Lan97,Caglioti98,Shishodia01,Rosato02}.
81: There exists a great controversy concerning the upward to downward segregation in
82: granular materials\cite{Shinbrot98,Shinbrot00,Hong01,Liffman01,Mobius01,Walliser02}.
83: Here we address this problem using a recently
84: proposed model for intruder size segregation in dry granular media\cite{leo}. This model characterizes
85: the rise velocity of a large intruder particle immersed in a medium of monodisperse fluidized small
86: particles. In Ref.\cite{leo} we have proposed a segregation mechanism based on the difference of
87: densities between different regions of the system, which gives origin to a buoyant force that
88: acts on the intruder. This force include an {\it Archimedean buoyancy force} due to the differences
89: between the intruder material density $\rho_I$ and the bed density $\rho_F$,
90: ${\bf f}_A=(\rho_F - \rho_I)V_I{\bf g}$. Here ${\bf g}$ is the gravity field and
91: $V_I = \frac{\Omega_D}{D} r_I^D$ is the $D$--dimensional volume of an intruder with radius
92: $r_I$. The factor $\Omega_D=2\pi^{D/2}/\Gamma(D/2)$ is the surface area of a $D$--dimensional
93: unit sphere.
94: Also, we include a {\it thermal buoyancy force} caused by density variations of the granular fluid
95: which comes from differences in the local ``granular temperature'' $\Delta T_g$. The change
96: in the granular fluid density through the thermal expansion, produced by
97: the difference of temperatures, is $\rho_F'=\rho_F(1-\alpha \Delta T_g)$, where $\alpha$ is the
98: thermal expansion coefficient. The thermal contribution to the buoyancy force is
99: ${\bf f}_T = \Delta \rho_F V_I {\bf g}$, where $\Delta \rho_F = \rho_F' - \rho_F = -\alpha \rho_F \Delta T_g$.
100: The granular temperature $T_g$ is defined proportional to the mean kinetic energy associated
101: to the velocity of each particle. The granular temperature difference $\Delta T_g$ is due to the
102: dissipative nature of the collisions between grains. This difference is due to the fact that the
103: number of collisions on the intruder surface increases with the size, but the local density
104: of dissipated energy diminishes. The region with intruder is {\it hotter} than the region
105: without intruder. This lead to the thermal buoyancy force that contributes to the intruder's
106: upward movement. The intruder also experiences a viscous drag force of the granular fluid.
107: The drag force ${\bf f}_d$ \cite{Zik92,Albert99} is considered to be linear in the velocity of segregation
108: ${\bf u}(t)$, and is like the Stokes' drag force ${\bf f}_d = -6\pi \mu r_I {\bf u}(t)$, where
109: $\mu$ is the coefficient of viscosity.
110: Therefore the equation of motion that governs the segregation process is
111: %\begin{equation}
112: %\frac{\Omega_D}{D}r_I^D\rho_I \frac{d {\bf u}(t)}{dt} = \frac{\Omega_D}{D}r_I^D \left[ \rho_I - \rho_F(1+\alpha \Delta T_g) \right] {\bf g} -6\pi \mu r_I {\bf u}(t).
113: %\label{Eqmov}
114: %\end{equation}
115: \begin{eqnarray}
116: \frac{\Omega_D}{D}r_I^D\rho_I \frac{d {\bf u}(t)}{dt} & = & \frac{\Omega_D}{D}r_I^D \left[ \rho_I - \rho_F(1+\alpha \Delta T_g) \right] {\bf g}
117: \nonumber
118: \\
119: & & -6\pi \mu r_I {\bf u}(t).
120: \label{Eqmov}
121: \end{eqnarray}
122:
123: We take the reference frame positive in the upward vertical direction. Arranging terms in
124: Eq.(\ref{Eqmov}) we find the following differential equation
125: \begin{equation}
126: \frac{du(t)}{dt}=\left[ \rho_F (1+ \alpha \Delta T_g)-\rho_I \right]\frac{g}{\rho_I} - \frac{6\pi D \mu}{\Omega_D \rho_I r_I^{D-1}}u(t),
127: \label{dynamic}
128: \end{equation}
129: and the solution is
130: \begin{equation}
131: u(t) = t_0 g \left[ \frac{\rho_F}{\rho_I}(1+\alpha \Delta T_g)-1\right]\left(1-e^{-t/t_0} \right),
132: \label{velocity}
133: \end{equation}
134: where the time--scale $t_0$ is
135: \begin{equation}
136: t_0\equiv\frac{\Omega_D \rho_I r_I^{D-1}}{6\pi D \mu}.
137: \label{time}
138: \end{equation}
139:
140: The drag force always acts opposite to the intruder velocity. So, the intruder's
141: upward/downward movement is exclusively due to the buoyancy forces. For our theoretical
142: calculations we define the settling velocity $u_s$
143: \begin{equation}
144: u_s = t_0 g \left[ \frac{\rho_F}{\rho_I}(1+\alpha \Delta T_g)-1\right].
145: \label{settling}
146: \end{equation}
147: When $u_s>0$ the resulting movement is upward. On the
148: other hand, if $u_s<0$ the resulting movement is downward.
149: Our analysis reveals that if $(1+\alpha \Delta T_g)\rho_F/\rho_I>1$, the intruder ascends.
150: The opposite occurs when $(1+\alpha \Delta T_g)\rho_F/\rho_I<1$. When $u_s=0$ there is no
151: upward neither downward movement.
152:
153: The intruder's presence modifies the local temperature of the system due to the collision that
154: happen at its surface. In Ref.\cite{leo} we have proposed an analytic procedure to estimate the
155: temperature difference among the granular fluid. We can calculate within a sphere of radius $r_0$
156: the value of the temperature $T_1$ in the granular fluid in presence of the intruder and
157: compare it with the temperature $T_2$ in the granular fluid without intruder.
158: In both cases we calculate the granular temperatures $(T_1, T_2)$ at a distance $r=r_I$ from the
159: center of the sphere of radius $r_0$.
160: See Fig. 1 for
161: a schematic picture of the regions used to calculate the granular temperature. In the model this
162: temperature ratio for two dimensions is given by (see Ref.\cite{leo} for details)
163:
164: \begin{equation}
165: \frac{T_1}{T_2} = \left(\frac{I_0(\lambda_Fr_0)[\Theta_{AB}I_0(\lambda_Fr_I) + K_0(\lambda_Fr_I)]}{I_0(\lambda_Fr_I)[\Theta_{AB}I_0(\lambda_F r _0) + K_0(\lambda_F r_0)]} \right)^2,
166: \label{tau2d}
167: \end{equation}
168: where $I_0(x)$ and $K_0(x)$ are modified Bessel functions, and
169: \begin{equation}
170: \Theta =\frac{\lambda_FI_0(\lambda_Ir_I)K_1(\lambda_Fr_I)+\lambda_II_1(\lambda_Ir_I)K_0(\lambda_Fr_I)}{\lambda_FI_0(\lambda_Ir_I)I_1(\lambda_Fr_I) - \lambda_I I_1(\lambda_Ir_I)I_0(\lambda_Fr_I)}.
171: \end{equation}
172:
173: For three dimensions,
174: \begin{equation}
175: \frac{T_1}{T_2} = \left(\frac{i_0(\lambda_Fr_0)[\Theta_{AB}i_0(\lambda_Fr_I) + k_0(\lambda_Fr_I)]}{i_0(\lambda_Fr_I)[\Theta_{AB}i_0(\lambda_F r _0) + k_0(\lambda_F r_0)]} \right)^2,
176: \label{tau3d}
177: \end{equation}
178: where $i_0(x)$ and $k_0(x)$ are spherical modified Bessel functions, and
179: \begin{equation}
180: \Theta = \frac{\lambda_Fi_0(\lambda_Ir_I)k_1(\lambda_Fr_I)+\lambda_I i_1(\lambda_Ir_I)k_0(\lambda_Fr_I)}{\lambda_F i_0(\lambda_Ir_I)i_1(\lambda_Fr_I) - \lambda_I i_1(\lambda_Ir_I)i_0(\lambda_Fr_I)}.
181: \label{theta3}
182: \end{equation}
183:
184: \begin{figure}
185: \centerline{\includegraphics[width=7.0cm]{Fig1.ps}}
186: \caption{Regions used to calculate the granular temperature. (a) Region around the intruder within a
187: sphere of radius $r_0$ and (b) region without intruder.}
188: \label{fig1}
189: \end{figure}
190:
191: The granular bed is formed of $N$ monodisperse particles of mass $m_F$ and radius $r_F$. The particles
192: are inelastic hard disks $(D=2)$ or spheres $(D=3)$. The inelasticity is specified by a restitution
193: coefficient $e\leq 1$.
194: The factor $\lambda$ couples the coefficient of thermal conductivity $\kappa = \kappa_0 T_g^{1/2}$ and
195: the dissipation rate $\gamma = \xi T_g^{3/2}$, explicitly one has\cite{leo}
196: \begin{equation}
197: \lambda^2 = \frac{\xi}{2\kappa_0},
198: \label{lambda}
199: \end{equation}
200: where $\xi$ and $\kappa_0$ depend, among other things, on the mass of the particles.
201: For the fluid particles this is
202: \begin{equation}
203: \xi_F = \frac{\Omega_D}{2\sqrt{2\pi}}(1-e^2) n^2 g_0 (2r_F)^{D-1}\left( \frac{2}{m_F} \right)^{1/2},
204: \label{xi_F}
205: \end{equation}
206: and for the region near the intruder
207: \begin{equation}
208: \xi_I = \frac{\Omega_D}{2\sqrt{2\pi}}(1-e^2)\frac{n}{V}g_0(r_F+ r_I)^{D-1} \left( \frac{m_I + m_F}{m_I m_F} \right)^{1/2},
209: \label{xi_I}
210: \end{equation}
211: where $n=N/V$, $V=L^D$ is the volume of the system of size $L$, $g_0$ is the pair correlation function for two fluid
212: particles. In $2D$ the pair correlation function is\cite{Verlet82}
213: $g_0 = (1-\frac{7}{16}\nu)/(1-\nu)^2$, with the area fraction $\nu=n\pi r_F^2$.
214: In $3D$ the pair correlation function is\cite{Carnahan69} $g_0 = (2-\nu)/2(1-\nu)^3$, with
215: the volume fraction $\nu=4n\pi r_F^3/3$.
216: Equations (\ref{lambda}), (\ref{xi_F}) and (\ref{xi_I}) define the factors $\lambda_F = \sqrt{\xi_F/2\kappa_0}$ and $\lambda_I = \sqrt{\xi_I/2\kappa_0}$.
217: The prefactor $\kappa_0$ in two dimensions is
218: \begin{equation}
219: \kappa_0 = 3n r_F\left( \frac{\pi}{m_F} \right)^{1/2}\left[ 1 + \frac{1}{3}\frac{1}{G} + \frac{3}{4} \left( 1+\frac{16}{9\pi} \right)G\right],
220: \end{equation}
221: and for three dimensions
222: \begin{equation}
223: \kappa_0 = \frac{15}{8}n r_F\left( \frac{\pi}{m_F} \right)^{1/2}\left[ 1 + \frac{5}{24}\frac{1}{G}+\frac{6}{5}\left( 1 + \frac{32}{9\pi} \right) G \right],
224: \end{equation}
225: where $G=\nu g_0$.
226:
227: For a dense system the pressure is related to the density by the
228: virial equation of state
229: is $p=\frac{1+e}{2}nT_g( 1+\frac{\Omega_D}{2D} n g_0 (2r_F)^D)$.
230: The thermal expansion coefficient is defined as
231: $\alpha\equiv V^{-1} (\partial V /\partial T_g)$.
232: From the equation of state we can calculate the coefficient
233: $\alpha = T_g^{-1}(\nu^2+8)^2/(\nu^3-3\nu^2-8\nu -8)(\nu -1)$ for two dimensions and
234: $\alpha = T_g^{-1}(\nu^3 -\nu^2-\nu -1)(\nu -1)/(\nu^4-4\nu^3 +4\nu^2+4\nu+1)$
235: for three dimensions.
236:
237: The state-dependent viscosity possesses the general form $\mu=\mu_0\sqrt{T_g}$. The
238: prefactor $\mu_0$ is
239: \begin{equation}
240: \mu_0 = \frac{1}{4} n r_F (\pi m_F)^{1/2} \left[ 2 + \frac{1}{G} + \left( 1 + \frac{8}{\pi} \right)G \right],
241: \end{equation}
242: for two dimensions and
243: \begin{equation}
244: \mu_0 = \frac{1}{3}nr_F(\pi m_F)^{1/2}\left[ 1 + \frac{5}{16}\frac{1}{G} + \frac{4}{5}\left( 1 + \frac{12}{\pi} \right)G \right],
245: \end{equation}
246: for three dimensions.
247:
248: The temperatures of the region with intruder and the region
249: without intruder are different.
250: For $T_1/T_2>1$ the thermal buoyancy force favors the upward movement. When $T_1/T_2<1$ the
251: thermal buoyancy force favors the downward movement.
252: The nonlinear form of Equations (\ref{tau2d}) and (\ref{tau3d}) doesn't allow us to calculate analytically
253: the explicit dependence on the mass ratio $m_I/m_F$ and the size ratio $\phi = r_I/r_F$.
254: Let us examine numerically the temperature ratio $T_1/T_2$ as a
255: function of the mass ratio $m_I/m_F$ for different values of $\phi$.
256: We set the number of
257: particles $N = 5\times 10^3$, the volume fraction $\nu = 0.75$, $r_0=L/2$ and the restitution
258: coefficient $e=0.95$.
259: We consider small particles with unitary mass $m_F = 1$, and we vary the intruder's mass like
260: $m_I = x m_F$, which $x\sim 0$ $(m_I<<m_F)$ to $x=2$. Figure 2 shows the temperature ratio dependence
261: as function of the mass ratio for two and three dimensions.
262: \begin{figure}
263: \centerline{\includegraphics[width=6.0cm]{temp_bnp_2d.ps}}
264: \centerline{\includegraphics[width=6.0cm]{temp_bnp_3d.ps}}
265: \caption{Granular temperature ratio $T_1/T_2$ as a function of the mass ratio $m_I/m_F$
266: for (a) two dimensions, and (b) three dimensions.}
267: \label{fig2}
268: \end{figure}
269:
270: From the analysis of Fig. 2 we can conclude that always $T_1>T_2$.
271: and the upward to downward movement is basically controlled
272: by the {\it Archimedean buoyancy force}.
273:
274: Let us concentrate on the dynamic of the segregation process described by the Equation (\ref{dynamic}). From
275: the settling velocity $u_s$ (Eq.(\ref{settling})) we can analyze the dependence on size and density ratio. An
276: uniformly fluidized state can be realized when the granular system is subject to a vertical vibration with
277: amplitude $A_0$ and frequency $\omega_0=2\pi f$. In the experiments the excitation is described by the
278: dimensionless acceleration $\Gamma = A_0 \omega_0/g$. The characteristic velocity of the system is
279: $u_0 = A_0 \omega_0$.
280: The system increases its energy as a result of external driving
281: while its decreases its energy by dissipation.
282: %
283: In our theoretical model we do not consider a sinusoidal excitation.
284: %
285: Analogies with shaken granular systems and the dependency of the granular temperature on the amplitude of
286: vibration has been studied in Ref.\cite{Sunthar99} for dense granular systems in $2D$, in which the
287: following expression relating the global granular temperature to a symmetric vibration with
288: maximum velocity $u_0$ is: $T_g = 2\sqrt{2}m_F L (A_0 \omega_0)^2/2Nr_F(1-e^2)$.
289: %
290: For three dimensions we estimate the granular
291: temperature as $T_g\sim m_F(A_0 \omega_0)^2$. In order to calculate $u_s$ we use the following model
292: parameters: mass particle density $\rho_F = 2.7$ gcm$^{-3}$, $r_F = 0.1$ cm, $e = 0.9$, $\nu = 0.75$,
293: $N = 5\times 10^3$, $g=100$ cms$^{-2}$, $r_0 = L/2$, $A_0 = 2r_F$ cm, $\omega_0 =5.81\sqrt{g/A_0}$
294: for two dimensions, and $\omega_0 =0.7\sqrt{g/A_0}$ for three dimensions.
295: Fig. 3 summarizes the results of our calculations for size ratio $\phi$ from $1$ to $10$ in two dimensions
296: and three dimensions.
297:
298: \begin{figure}[h]
299: %\includegraphics[width=\hsize]{Fig1.ps}
300: \centerline{\includegraphics[width=5.6cm]{vel_bnp_2d.ps}}
301: \centerline{\includegraphics[width=5.6cm]{vel_bnp_3d.ps}}
302: \caption{Dimensionless intruder settling velocity $u_s/(A_0\omega_0)$, as a function of density
303: ratio $\rho_I/\rho_F$ in (a) two dimensions, and (b) three dimensions.}
304: \label{fig3}
305: \end{figure}
306: \begin{figure}[h]
307: %\includegraphics[width=\hsize]{Fig1.ps}
308: \centerline{\includegraphics[width=5.6cm]{diagram2d.ps}}
309: \centerline{\includegraphics[width=5.6cm]{diagram3d.ps}}
310: \caption{Phase diagram determined for (a) two dimensions, and (b) three dimensions.}
311: \label{fig4}
312: \end{figure}
313:
314: From these results we can observe that the segregation is rapid in systems in which small particles
315: are more dense than the intruder. For fixed values of $\phi$, when the density ratio increases the rise
316: velocity diminishes, this implies a diminution of the intruder rise time. Finally, when the intruder's
317: density is bigger than the small particles density the intruder particle sinks $u_s<0$.
318: For a fixed value of the density ratio the segregation
319: rate increases with the size ratio.
320: These result shown the condition for the crossover from the
321: upward to the downward intruder's motion.
322:
323: Recently, the upward/downward transition has been studied in
324: references \cite{Hong01,Both02,Jenkins02}.
325: In Ref.\cite{Hong01} Hong {\it et al.,} performed molecular dynamic simulations, in two and three
326: dimensions, of weakly dissipative particles under gravity that are in global thermal
327: equilibrium with a heat reservoir.
328: They observed that for a binary mixture of granular particles the large particles can rise
329: and the upward/ downward movement depends on mass and diameter ratios.
330: They proposed an explanation based
331: on a competition between the percolation effect and the condensation of hard spheres under
332: gravity (See Refs.\cite{Hong01,Hong99,Quinn00} for a detailed explanation of the
333: {\it percolation--condensation} mechanism).
334: Both and Hong presented a theory based on the variational principle for hard spheres and disks
335: under gravity\cite{Both02}. They also characterized the segregation phenomenon and investigated the
336: crossover between the upward/downward movement for large particles and the dependence on mass
337: and size ratios. In the framework of kinetic theory for a binary granular mixture,
338: Jenkins and Yoon obtained a segregation criteria for spheres and disks that differ in size
339: and/or mass\cite{Jenkins02}. Their mechanism is based on a competition between the inertia of the
340: particles through the ratio of partial pressures. The three different approaches presented in
341: references \cite{Hong01,Both02,Jenkins02} coincide qualitatively among them.
342: In the same spirit we can calculate a phase diagram for the crossover from the upward/downward
343: movement (See Figs. 4(a) and 4(b)).
344: In our model these phase diagrams are derived from the dependence of the segregation velocity on
345: the mass and size ratio at the situation where the intruder tends to rise ($u_s > 0$), or
346: to sink ($u_s<0$) (Figs. 3(a) and 3(b)).
347:
348: In summary, we have studied the segregation dependence on size and density for a single intruder in
349: a fluidized bed.
350: On a qualitative level our model is in agreement with the experimental phenomenology described in
351: Refs.\cite{Rippie64,Ahmad73}: the higher the density of the intruder the lower the tendency to
352: rise. In Fig.(3) we can note that when the density of the particles increases, in the range
353: where $\rho_I/\rho_F < 1$, the segregation velocity diminishes, so the intruder's rise time will
354: be bigger. In general a lighter particle rise more quickly than a heavier particle of the same size
355: in the range where $\rho_I/\rho_F < 1$. In the range where $\rho_I/\rho_F > 1$ the intruder sinks,
356: and in this case the downward segregation velocity increases with the density and size.
357: Our model also agrees with the qualitative behavior reported in Ref.\cite{Shishodia01}, where the
358: intruder height decreases as the density ratio increases. In this case, for the range where
359: $\rho_I/\rho_F < 1$, the height of a lighter particle will be bigger than the height of a
360: heavier particle of the same size at the same time. This is due to the fact that the rise velocity of
361: a lighter particle will be bigger than the rise velocity of a heavier particle of the same size.
362: In agreement with the theoretical results obtained in
363: Refs.\cite{Hong01,Both02,Jenkins02} and the numerical simulations of Refs.\cite{Shishodia01,Hong01},
364: our model predicts that in the case where the density ratio
365: increases, the intruder particle will sink.
366:
367: A direct comparison with experiments performed in Ref.\cite{Mobius01} are not appropriate in this work
368: since they investigate the intruder's density effect in presence of convection and interstitial air.
369: Comparison is also difficult with the experiments performed in Refs.\cite{Shinbrot98,Liffman01},
370: where the role of inertia and interstitial air may play an important role. It is important to note
371: that the experiment performed by Shinbrot and Muzzion\cite{Shinbrot98}, doesn't correspond with
372: the typical experimental conditions of the Brazil nut effect. In these experiments the intruders
373: are placed at the surface of a vibrated bed. They reported that the heavy intruder remained at the
374: surface while the lighter intruder sank. The boundary condition for an intruder in this experimental
375: set--up is more different and complicated than the boundary condition for an intruder immersed in
376: the granular bed. So, in this experiment we can not consider upward/downward
377: transition.
378: In the experiment performed by Liffman {\it et al.,} \cite{Liffman01} they reported that the total rise time
379: of the intruder is inversely proportional to the density of the intruder. This is, a priori, in contrast to our
380: findings. In our case, if the segregation velocity decreases with the intruder's density, then the rise
381: time decreases. This apparent contradictory fact can be understood from the procedure followed
382: by Liffman {\it et al.,} for the intruder's density variation.
383: They studied the motion of disks placed at the base of a vibrated granular bed. They varied
384: the density of the disks with similar size, drilling holes in the disks. The intruder is
385: ``transformed'' from a disk to a ring. The inertia of these two object is different. Also, it is
386: important to note that the energy dissipation is different. Energy is dissipated during
387: collision, among other things, due to the excitation of the internal modes of the object. So,
388: the coefficient of restitution should be different. In terms of our model, based on the
389: concept of granular temperature, the density of dissipated energy should be bigger for a ring than
390: a disk. In this case the disk is ``hotter'' than a ring, and
391: the buoyancy force would favor the upward movement.
392:
393: In order to compare our theory to the experiments one should do measurements varying the intruder's
394: density homogeneously in dry granular materials. Strangely, a systematic study has not get been reported
395: in the literature and since this should be easy to carry out, we hope that some experiment and/or
396: simulation will soon be done.
397:
398: %\newpage
399:
400: \begin{thebibliography}{}
401: %
402: \bibitem{Herrmann98}
403: H. J. Herrmann, J.--P. Holvi and \& S. Luding (eds),
404: Physics of Dry Granular Media
405: (Kluwer, Doldrecht, 1998)
406: %
407: \bibitem{deGennes98}
408: P. G. de Gennes, Physica {\bf A261}, 267 (1998).
409: %
410: \bibitem{Rippie64}
411: E. G. Rippie, J. L. Olsen \& M. D. Faiman,
412: J. Pharm. Sci. {\bf 53}, 1369 (1964).
413: %
414: \bibitem{Ahmad73}
415: K. Ahmad \& I. J. Smalley,
416: Powder Technology {\bf 8}, 69 (1973).
417: %
418: \bibitem{Rosato87}
419: A. Rosato, K. J. Strandburg, F. Prinz \& R. H. Swendsen,
420: %{\it Why the Brazil nuts are on top: Size segregation of particulate matter by shaking},
421: Powder Tech. {\bf 49}, 59 (1986);
422: Phys. Rev. Lett. {\bf 58}, 1038 (1987).
423: %
424: \bibitem{Jullien92}
425: R. Jullien, P. Meakin \& A. Pavlovitch,
426: %{\it Three--dimensional model for particle--size segregation by shaking},
427: Phys. Rev. Lett. {\bf 69}, 640 (1992).
428: %
429: %\bibitem{Rosato86}
430: %A. Rosato, F. Prinz, K. J. Standburg and R. Swendsen, Powder Technology {\bf 49}, 59 (1986).
431: %
432: \bibitem{Duran93}
433: J. Duran, J. Rajchenbach \& E. Cl\'ement,
434: %{\it Arching effect model for particle size segregation},
435: Phys. Rev. Lett. {\bf 70}, 2431 (1993).
436: %
437: \bibitem{Knight93}
438: J. B. Knight, H. M. Jaeger \& S. R. Nagel,
439: %{\it Vibration--induced size separation in granular media: The convection connection},
440: Phys. Rev. Lett. {\bf 70}, 3728 (1993).
441: %
442: \bibitem{Jullien93}
443: R. Jullien, P. Meakin \& A. Pavlovitch,
444: %{\it Particle size segregation by shaking in two--dimensional disc packings},
445: Europhys. Lett. {\bf 22}, 523 (1993).
446: %
447: \bibitem{Duran94}
448: J. Duran, T. Mazoi, E. Cl\'ement \& J. Rajchenbach,
449: %{\it Size segregation in a two--dimensional sandpile: Convection and arching effects},
450: Phys. Rev. {\bf E50}, 5138 (1994).
451: %
452: \bibitem{Poschel95}
453: T. P\"oschel \& H. J. Herrmann,
454: %{\it Size segregation and convection},
455: Europhys. Lett. {\bf 29}, 123 (1995).
456: %
457: \bibitem{Gallas96}
458: J. A. C. Gallas, H. J. Herrmann, T. P\"oschel \& S.Soko\l owski,
459: %{\it Molecular dynamics simulation of size segregation in three dimensions},
460: J. Stat. Phys. {\bf 82}, 443 (1996).
461: %
462: \bibitem{Cooke96}
463: W. Cooke, S. Warr, J. M. Huntley \& R. C. Ball,
464: Phys. Rev. {\bf E53}, 2812 (1996).
465: %
466: \bibitem{Vanel97}
467: L. Vanel, A. D. Rosato \& R.N. Dave,
468: %{\it Rise--time regimes of a large sphere in vibrated bul solids},
469: Phys. Rev. Lett. {\bf 78}, 1255 (1997).
470: %
471: \bibitem{Lan97}
472: Y. Lan \& A. D. Rosato,
473: Phys. Fluids {\bf 9}, 3615 (1997).
474: %
475: \bibitem{Caglioti98}
476: E. Caglioti, A. Coniglio, H. J. Herrmann, V. Loreto \& M. Nicodemi,
477: %{\it Segregation of granular mixtures in the presence of compactation},
478: Europhys. Lett. {\bf 43}, 591 (1998).
479: %
480: \bibitem{Shishodia01}
481: N. Shishodia \& C. R. Wassgren,
482: Phys. Rev. Lett. {\bf 87}, 084302 (2001), {\it ibid.} {\bf 88}, 109901(E) (2002).
483: %
484: \bibitem{Rosato02}
485: A. D. Rostato, D. L. Blackmore, N. Zhang \& Y. Lan,
486: Chem. Enf. Sci. {\bf 57}, 265 (2002).
487: %
488: \bibitem{Shinbrot98}
489: T. Shinbrot \& F.J. Muzzio,
490: %{\it Reverse buoyancy in shaken granular beds},
491: Phys. Rev. Lett. {\bf 81}, 4365 (1998).
492: %
493: \bibitem{Shinbrot00}
494: T. Shinbrot \& F. J. Muzzio, Physics Today, March Issue, 25 (2000).
495: %
496: \bibitem{Hong01}
497: D. C. Hong, P. V. Quinn \& S. Luding,
498: Phys. Rev. Lett. {\bf 86}, 3423 (2001), see also cond-mat/0202384, cond-mat/0202385.
499: %
500: \bibitem{Liffman01}
501: %
502: K. Liffman, K. Muniandy, M. Rhodes, D. Gutteridge \& G. Metcalfe,
503: Granular Matter {\bf 3}, 205 (2001).
504: %
505: \bibitem{Mobius01}
506: M. E. M\"obius, B. E. Lauderdale, S. R. Nagel \& H. M. Jaeger,
507: Nature (London) {\bf 414}, 270 (2001).
508: %
509: \bibitem{Walliser02}
510: H. Walliser, con-mat/0111559.
511: %
512: \bibitem{leo}
513: L. Trujillo \& H. J. Herrmann, cond-mat/0202484.
514: %
515: \bibitem{Zik92}
516: O. Zik, J. Stavans and Y. Rabin,
517: Europhys. Lett. {\bf 17}, 315 (1992).
518: %
519: \bibitem{Albert99}
520: R. Albert, M. A. Pfeifer, A.--L. Barab\'asi and P. Schiffer,
521: Phys. Rev. Lett. {\bf 82}, 205 (1999).
522: %
523: \bibitem{Verlet82}
524: L. Verlet \& D. Levesque,
525: Mol. Phys. {\bf 46}, 969 (1982).
526: %
527: \bibitem{Carnahan69}
528: N. F. Carnahan \& K. E. Starling,
529: J. Chem. Phys. {\bf 51}, 635 (1969)
530: %
531: \bibitem{Sunthar99}
532: P. Sunthar \& V. Kumaran,
533: Phys. Rev. {\bf E60}, 1951 (1999).
534: %
535: %\bibitem{Ohtsuki95}
536: %T. Ohtsuki, D. Kinoshita, Y. Takmoto \& A. Hayashi,
537: %J. Phys. Soc. Jpn. {\bf 64}, 430 (1995).
538: %
539: \bibitem{Both02}
540: J. A. Both \& D. C. Hong,
541: Phys. Rev. Lett. {\bf 88}, 124301 (2002).
542: %
543: \bibitem{Jenkins02}
544: J. T. Jenkins \& D. K. Yoon,
545: Phys. Rev. Lett. {\bf 88}, 194301 (2002).
546: %
547: \bibitem{Hong99}
548: D. C. Hong, Physica A271, 192 (1999).
549: %
550: \bibitem{Quinn00}
551: P. V. Quinn \& D. C. Hong, Phys. Rev. {\bf E62}, 8295 (2000).
552: %
553: \end{thebibliography}
554:
555:
556: %\begin{figure}
557: % Use the relevant command for your figure-insertion program
558: % to insert the figure file.
559: % For example, with the option graphics use
560: %\resizebox{0.75\hsize}{!}{%
561: % \includegraphics{leer.eps}
562: % If not, use
563: %\vspace{5cm} % Give the correct figure height in cm
564: %\caption{Please write your figure caption here}
565: %\label{fig:1} % Give a unique label
566: %\end{figure}
567: %
568: % For two-column wide figures use
569: %\begin{figure*}
570: % Use the relevant command for your figure-insertion program
571: % to insert the figure file. See example above.
572: % If not, use
573: %\vspace*{5cm} % Give the correct figure height in cm
574: %\caption{Please write your figure caption here}
575: %\label{fig:2} % Give a unique label
576: %\end{figure*}
577: %
578: % For tables use
579:
580: % BibTeX users please use
581: % \bibliographystyle{}
582: % \bibliography{}
583: %
584: % Non-BibTeX users please use
585: \end{document}
586:
587: % end of file template.tex
588:
589:
590:
591:
592: