Materials Transactions, Vol. 48, No. 11 (2007) pp.
3021 to 3023
#2007 The Japan Institute of Metals
A Theoretical Approach of the Heat Transfer in Nanofluids
P. Vizureanu1 and M. Agop1;2;3; *
1
Faculty of Materials Science and Engineering, Technical ‘‘Gh. Asachi’’ University, Blvd. Mangeron, Iasi, 700029, Romania
2
Department of Physics, University of Athens, Athens 15771, Greece
3
Department of Physics, Technical ‘‘Gh.Asachi’’ University, Blvd. Mangeron, Iasi, 700029, Romania
Using the fractal space-time theory (scale relativity theory), the dynamics of the fluid/nano-particle interface was analyzed. In the general
case, the heat transfer through the interface reproduces a d.c. or an a.c. Josephson effects of thermal type, while in the linear approximation, the
standard form of heat transfer is given. Consequently, a negative differential thermal conductance appears and an increase in of heat transfer in
nanofluids results. [doi:10.2320/matertrans.MRP2007118]
(Received May 23, 2007; Accepted July 27, 2007; Published October 25, 2007)
Keywords: nanofluids, anomaly of the heat transfer, thermal conductance
1. Introduction
Experiments over the past decade have revealed that
thermal conductivity of the suspension can be significantly
higher than that of the base medium.1–4) Keblinski et al.4)
have examined four possible mechanisms for the anomalous
enhancement observed in nanofluids. The first is the transport
of thermal energy by Brownian motion of the nano-particles.
An order-of-magnitude calculation shows that the Brownian
effects are not strong enough to explain the observed
anomaly. A similar conclusion was also reached by Wang
and Xu1) with a somewhat different set of assumptions, and
also later confirmed by a molecular dynamics simulation.4)
On the other hand, recent reports continue to attribute the
enhanced conductivity to Brownian motion of the nano-
particles.5–7) Another mechanism, the formation of liquid
layers around the particles, is considered by Keblinski et al.4)
The basic idea is that liquid molecules can form layers around
the solid particles, thereby enhancing the local ordering.
Since phonon transfer in crystalline solid is very effective,
such local ordering in the liquid can lead to enhanced heat
transport. A recent molecular dynamic simulation by Xue et
al.8) confirmed the presence of short-ranged ordering of
liquid molecules, but surprisingly observed little or no effect
on the thermal conductivity. The third mechanism is related
to the nature of heat transport in nano-particles. It is pointed
out4) that the generally accepted diffusive transport mecha-
nism is not valid at the nano-scale. Instead of this, the
ballistic transport is more realistic: if the ballistic ‘phonons’
initiated in a particle can persist in the liquid and get
transmitted to another solid particle, the heat transport can
significantly increase. The phonon mean free path in the
liquid is typically small, because local ordering is limited to
few atomic diameters. Since the particles have a constant
Brownian motion, the coherent phonon transfer is possible
even for low particle concentrations. Prasher et al.7) shows
that the crystalline-like phonon modes account only for a
fraction of the increased thermal conductivity observed in the
experiments. Lastly, Keblinski et al.4) explore the possibility
of nano-particles forming clusters and their effects on the
thermal conductivity. This type of model may be appropriate
*Corresponding author, E-mail: m.agop@yahoo.com
RAPID PUBLICATION
if the particles are in the form of nanotubes or if they are not
finely dispersed.9) In contrast to the assessments of Keblinski
et al.4) and Wang and Xu,1) other authors as Jang and Choi,5)
Kumar et al.6) and Prasher et al.7) hypothesized that
Brownian motion of the nano-particles plays a leading role
in energy transport of in nanofluids. While these models
agree on the overall mechanism, they differ appreciably in
how Brownian motion affects the thermal conductivity of the
nanofluid.
Recently, the increasing of the heat transfer in nanofluids
was related to the fractalic effects.10) Moreover, Wang et
al.11) reported that the modified fractal model agreed well
with the experimental data obtained for the SiO2 /ethanol
nanofluid. This model has some disadvantages as difficult
calculations and measurements on the nanoparticles cluster
space-distribution. In such conjecture, the fractal space-time
theory (scale relativity theory) is a new approach to under-
stand quantum mechanics, and moreover physical domains
involving scale laws, such as the nanosystems.12,13) It is based
on a generalization of Einstein’s principle of relativity to
scale transformations. Namely, one redefines space-time
resolutions as characterizing the state of scale of reference
systems, in the same way as velocity characterizes their state
of motion. Then one requires that the laws of physics apply
whatever the state of the reference system, of motion
(principle of motion-relativity) and of scale (principle of
scale-relativity). The principle of scale-relativity is mathe-
matically achieved by the principle of scale-covariance,
requiring that the equations of physics keep their simplest
form under transformations of resolution. For example,
considering that the motion of micro-particles take place on
continuous but non-differentiable curves, i.e. on fractals,12–14)
it was demonstrated that, in the topological dimension14)
DT ¼ 2, the geodesics of the fractal space-time are given by a
Schrödinger’s type equation.
In the present paper, using the scale relativity theory, we
propose a new mechanism capable to explain the exper-
imentally observed enhanced thermal conductivity of nano-
fluid.
2. Mathematical Model
Let us consider the interaction between two fractal
3022 P. Vizureanu and M. Agop

structures, e.g. the fluid and the nano-particle,10,11) and the J Q ,C

corresponding interface. According with the scale relativity
model, the interface dynamics is described by the coupled
equations set (for details see Appendix A and12,13)),
2iD@t
f ¼ hf
f þ 
np
2iD@t
np ¼ hnp
np þ 
f ; ð1a, bÞ T0

with
f ,
np the wave functions, hf , hnp the specific

‘‘Hamiltonians’’ (per mass unit) on either side of the
interface,  a coupling constant characterizing the interface,
D the fractal/non-fractal transition coefficient,12,13) and the
indices ( f , np) refers to the fluid and nanoparticle, respec- Fig. 1 The negative differential thermal conductance, by means of the
tively. Generally, the ‘‘Hamiltonians’’ Hi , i ¼ 1; 2 correspond negative slope of the Bessel J3 function.
to the free energy of a particle with the rest mass m0 localized
in i region. In the Appendix A it is exemplified such a
situation for D ¼ h =2m, i.e. for the standard Josephson (9a, b) admits locally a Fourier discrete decomposition.16)
effect.14) This means that the previous results are of maximum
Expliciting the wave functions by the following relations: generality.
pffiffiffiffiffi if pffiffiffiffiffiffiffi Substituting the relation (9a, b) into (7a, b) and integrating
f ¼
f e ; np ¼
np einp ð2a, bÞ
it, we obtain the time dependence of the phase difference:
and separating in (1a, b) the real parts from the imaginary
T0 T 0
ones, we obtain:  ¼ 0 þ tþ sinðt þ ’0 Þ ð10Þ
 pffiffiffiffiffiffiffiffiffiffiffi D D
@t
f ¼
@t
np ¼
f
np sinðnp
f Þ With relation (10), the expression (5) of the heat flux
D
rffiffiffiffiffiffiffi becomes
hf 
np
@ t f ¼

cosðnp
f Þ  
2D 2D
f T0 T 0
rffiffiffiffiffiffiffi JQ ðtÞ ¼ JQ;M sin 0 þ tþ sinðt þ ’0 Þ
hnp 
f D D
@t np ¼

cosðnp
p Þ ð3a{cÞ     
2D 2D
np X
þ1
T 0 T0
n
pffiffiffiffiffi pffiffiffiffiffiffiffi ¼ JQ;M ð
1Þ Jn sin
n t þ 0
where
f ,
np are the amplitudes and f , np the phases.
n¼
1 D D
From here, with4,8)
ð11Þ

n ¼
np ¼
; hf ¼ Tf ; hnp ¼ Tnp ; 16)
where Jn is the n-order Bessel function and 0 a constant of
T ¼ Tf
Tnp ;  ¼ np
f ;  ¼ const. ð4a{fÞ integration.
it results the heat flux: When the pulsation n ¼ T0 =D satisfies the relation
n ¼ n, n ¼ 1; 2; 3; . . ., the time-average of JQ  hJQ ðtÞi
JQ ¼ "ð@t
f
@t
np Þ ¼ JQ;M sin  ð5Þ
differs from zero, i.e. there is a continuous component of the
of amplitude JQ;M heat flux of the form:
 
2" n T 0
JQ;M ¼
ð6Þ JQ;C ¼ ð
1Þ JQ;M Jn sin 0 ð12Þ
D D
and phase difference  From relation (12) it results peaks of the continuous heat
Z flux for

 ¼ 0 þ Tdt; 0 ¼ const. ð7a, bÞ
D Dn D
Tn ¼ nTS ; Tn ¼ ; TS ¼ ; n ¼ 1; 2; 3; . . . ð13a{dÞ
with Tf , Tnp the absolute temperatures, T temperature  
difference on the interface, " the elementary amount of and consequently a negative differential thermical conduct-
energy transferred through the interface.9) ance (dJQ;C =dT 0 < 0) – see Fig. 1. Moreover, from eq. (12)
For T ¼ 0, relation (7a, b) reproduces a d.c. Josephson the heat flux of the pick n can varies continuously in the range
effect of thermal type, while for T 6¼ 0 an a.c. Josephson [
JQ;M Jn ðT 0 =DÞ, þJQ;M Jn ðT 0 =DÞ] at constant temper-
effect of thermal type, i.e. oscillations of the heat flux with ature Tn , and the phase varies in the range [
=2, þ=2].
the pulsation, This means that in the interface, the heat can be generated or
absorbed.
T
!¼ ð8Þ
D 3. Conclusions
In this last case, let us consider the dependency T ¼ TðtÞ in
the form A mathematical model of the heat transfer in nanofluids, is
established. Then, the main conclusions are as follow:
T ¼ T0 þ T 0 cosðt þ ’0 Þ; ’0 ¼ const. ð9a, bÞ
i) Using the scale relativity theory, the dynamics of the fluid/
We notice that any time-dependent signal, for exemple nano-particle interface was analyzed. Then, the heat transfer
 A Theoretical Approach of the Heat Transfer in Nanofluids
through the interface reproduces a d.c. or an a.c. Josephson
effects of thermal type;
ii) In the linear approximation (sin   ), the standard form
of heat transfer results,

JQ  GT; G ¼ JQ;M
D ‘‘Hamiltonians’’ m0 hi ¼ Hi , i ¼ 1; 2 correspond to the free
with G the thermal conductance and  a characteristic
time,1–4) while in the general cases,4) the heat transfer is non-
linear;
iii) The interface is self-structuring and consequently, a
negative differential thermal conductance appears. Since
through self-structuring the interface generates heat to the
environment, it explains the anomaly of the heat transfer in
nanofluids.
Our opinion are in according to Kebliski et al.4) results
given by means of molecular level simulations.
Appendix
The theory of fractal space-time is the scale relativity (SR)
theory.12,13) A non-differentiable continuum is necessarily
fractal and the trajectories in such a space (or space-time)
own (at least) the following three properties: i) The test
particle can follow an infinity of potential trajectories: this
leads one to use a fluid-like description; ii) The geometry of
each trajectory is fractal15) (of dimension 2). Each elementary
displacement is then described in terms of the sum,
dX ¼ dx þ d, of a mean classical displacement dx ¼ vdt
and of a fractal fluctuation d, whose behavior satisfies the
principle of SR (in its simplest Galilean version). It is such
that hdi ¼ 0 and hd2 i ¼ 2Ddt. The existence of this
fluctuation implies introducing new second order terms in
the differential equation of motion; iii) Time reversibility is
broken at the infinitesimal level: this can be described in
terms of a two-valuedness of the velocity vector for which we
use a complex representation, V ¼ ðvþ þ v
Þ=2
iðvþ þ
v
Þ=2 (we denoted by vþ the ‘‘forward’’ speed and by v
the
‘‘backward’’ speed). These three effects can be combined to
construct a complex time-derivative operator,
@t ¼ @t þ V  r
iD ðA:1Þ
with V the complex speed, V ¼
2iDr ln . Using the
Newton’s equation in its covariant form @t V ¼
m
1
0 r, we
obtain the Schrödinger type equation
2m0 D2  þ 2im0 D@t
ðm0 Þ
1  ¼ 0 ðA:2Þ
In the general case,12–14) the Schrödinger type equation
take the standard form 2im0 D@t ¼ H , with H the
‘‘Hamiltonians’’ of the particle. If, in such conjecture, we
consider the ‘‘interaction’’ between two physical systems
which have a fractal structure, then the dynamics of the
interface is described by the coupled equations set (for details
see12,13))
3023
2iD@t
1 ¼ h1
1 þ 
2 ; 2iD@t
2 ¼ h2
2 þ 
1 ðA:3Þ
with 1 , 2 the wave function, h1 , h2 the specific ‘‘Hamil-
tonians’’ (per mass unit) on either side of the interface and  a
coupling constant characterizing the interface. Generally, the
energy of a particle with the rest mass m0 localized in i
region. Particularly, for D ¼ h =2m, with h the Planck
reduced constant and m the effective mass of the Cooper
pair (for details see14)), i.e. at the quantum scale, the system
(A3), with substitutions f ;np ¼ 1;2 , mhf ;np ¼ H1;2 ¼ 1;2
take the standard form of the Josephson effect:
@ 1
ih ¼ 1 1 þ 2
@t
@ 2
ih ¼ 2 2 þ 1 ðA:4Þ
@t
In these relations, H1;2 represent the free energy 1;2 of a
Cooper pair in the mentioned region, and the difference
 ¼ 2
1 ¼ qV, with q the effective charge of the
Cooper pair gives the potential V applied on the interface.
REFERENCES
1) X. Wang and X. Xu: J. Thermophys. and Heat Transfer 13 (1999) 474–
480.
2) J. A. Eastman, S. U. S. Choi, S. Li, W. Yu and L. J. Thompson: Appl.
Phys. Lett. 78 (2001) 718–720.
3) H. E. Patel, S. K. Das, T Sundararajan, A. S. Nair, B. George and
T. Pradeep: App. Phys. Lett. 83 (2003) 2931–2933.
4) P. Keblinski, S. R. Phillpot, S. U. S. Choi and J. A. Eastman: Int. J. Heat
and Mass Transfer 45 (2002) 855–863.
5) S. P. Jang and S. U. S. Choi: Appl. Phys. Lett. 84 (2004) 4316–4318.
6) D. H. Kumar, H. E. Patel, V. R. R Kumar, T. Sundararajan, T. Pradeep
and S. K. Das: Phys. Rev. Lett. 93 (2004) 144301.
7) R. Prasher, P. Bhattacharya and P. E. Phelan: Phys. Rev. Lett. 94 (2005)
025901.
8) L. Xue, P. Keblinski, S. R. Phillpot, S. U. S. Choi and J. A. Eastman:
Int. J. Heat and Mass Transfer 47 (2004) 4277–4284.
9) V. Chiroiu, P. Stiuca, L. Munteanu and St. Donescu: Introduction in
nanomechanics (Romanian Academy Publishing House, Bucuresti),
(2005).
10) S. H. Mohoreanu and M. Agop: Journal of Optoelectronics and
Advanced Materials 8 (2006) 1741–1743.
11) B.-W. Wang, L.-R. Zhou and X.-F. Peng: Int. J. Heat Mass Transfer 46
(2003) 2665.
12) L. Nottale: Fractal Space-Time and Microphysics: Towards a Theory
of Scale Relativity (World Scientific, Singapore) (1993).
13) L. Nottale in Eds. M. S. El Naschie, O. E. Rösler and I. Prigogine:
Quantum Mechanics, Diffusion and Chaotic Fractals (Elsevier,
Oxford), (1995).
14) C. P. Poole, H. A. Farach and R. J. Creswick: Superconductivity,
(Academic Press, San Diego) (1995).
15) B. Madelbrot: The fractal geometry of nature, (Freeman, San
Francisco) (1982).
16) E. A. Jackson: Perspectives in nonlinear dynamics, Vol. I and II,
(Cambridge, Cambridge University Press) (1991).