Hindawi Publishing Corporation

Journal of Nanomaterials
Volume 2007, Article ID 30389, 9 pages
doi:10.1155/2007/30389

Review Article
The AC and DC Conductivity of Nanocomposites

David S. McLachlan1, 2 and Godfrey Sauti1, 3

1 Department of Chemistry and Polymer Science, University of Stellenbosch, Private Bag X1, Matieland 7602, South Africa
2 Materials Physics Research Institute, School of Physics, University of the Witwatersrand, Private Bag 3, Wits 2050, South Africa
3 National Institute of Aerospace, 100 Exploration Way, Hampton, VA 23666, USA

Received 25 April 2007; Accepted 9 August 2007

Recommended by Christian Brosseau

The microstructures of binary (conductor-insulator) composites, containing nanoparticles, will usually have one of two basic
structures. The first is the matrix structure where the nanoparticles (granules) are embedded in and always coated by the matrix
material and there are no particle-particle contacts. The AC and DC conductivity of this microstructure is usually described by
the Maxwell-Wagner/Hashin-Shtrikman or Bricklayer model. The second is a percolation structure, which can be thought to be
made up by randomly packing the two types of granules (not necessarily the same size) together. In percolation systems, there
exits a critical volume fraction below which the electrical properties are dominated by the insulating component and above which
the conducting component dominates. Such percolation systems are best analyzed using the two-exponent phenomenological
percolation equation (TEPPE). This paper discusses all of the above and addresses the problem of how to distinguish among the
microstructures using electrical measurements.

Copyright © 2007 D. S. McLachlan and G. Sauti. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.

1. INTRODUCTION per bound (conductor host) [2, 3] and will be referred to

here as the Maxwell-Wagner/Hashin-Shtrikman (MW/HS)
As it is all but impossible to measure the electrical properties model. For matrix dominated media, the Bricklayer model
of a single particle, which is three-dimensionally a nanopar- (BLM) [4] is also useful. The second nanostructure is where
ticle, its electrical properties will usually have to be deduced the conducting particles, in a two-phase material, make elec-
from a compact or a composite containing the nanoparti- trical contact with each other, when the volume fraction of
cles. A measurement of the electrical properties is also im- the conducting particles ϕ reaches a certain critical ϕc . At this
portant as it is usually not possible to deduce the con- point, a critical or spanning cluster is formed and there is an
nectivity of a composite system from SEM and TEM mi- abrupt (usually many orders of magnitude) change in the DC
crographs alone. This paper will show how to analyze the conductivity. The complex electrical conductivity of these
AC and DC conductivity, or complex dielectric constant, systems is best [5, 6] described by the two-exponent phe-
of binary (conductor and insulator) nanocomposites (me- nomenological percolation equation (TEPPE) (also known
dia, compacts). Most binary and some tertiary media (with as the general effective medium (GEM) equation) [7–17].
some ingenuity, the analysis can be applied to tertiary me- The three standard percolation equations are reviewed by
dia), which are found in practice to at least approximate Clerc et al. [18], Bergman and Stroud [19], and Nan [20], but
one of the two basic nanostructures, are discussed in this it is specifically shown by McLachlan et al. [5, 6] that, while
paper. The first are matrix media, where the matrix phase the single TEPPE reduces to the three standard equations in
surrounds the granular (particle) phase at all volume frac- the appropriate limits, it is far superior in the second-order
tions and the distance between the conducting particles is terms (i.e., the imaginary dielectric constant εmi in the con-
greater than the tunneling distance for electrons. This is usu- ducting media above ϕc and the dielectric loss εii (or σ mr )
ally best described using the Maxwell-Wagner effective me- below ϕc ). The theory given here applies to nano-, micro-,
dia equation (also known as the Maxwell-Garnet equation). and macromedia. Note that the word nanostructure is used
The Maxwell-Wagner model is formally equivalent to the throughout this paper although most of the existing experi-
Hashin-Shtrikman [1] lower bound (insulator host) and up- mental results are for microstructures. Section 2.1 deals with
2 Journal of Nanomaterials

the MW/HS equations, the specific nano- (micro-) structure arc—the peak frequency of which is determined by the prop-
pertaining to the MW/HS equations [2, 21], as well as ob- erties of the components and the volume fraction of each
servations regarding its applicability to nanostructures. The [24]. When the insulating phase forms the matrix, there are
“rectangular” Bricklayer model [4], widely used in microce- two arcs, each characterizing the contribution of one compo-
ramics, which could be appropriate for rectilinear nanostruc- nent. The peak frequencies for the multiple component arc
tures, is not discussed in this paper. Section 2.2 describes per- (above ϕc ) and the two single component arcs (below ϕc ) are
colation media and some of the nanostructures which give given by
rise to ϕc ’s between 0.0005 [22] and 0.56 [23]. The TEPPE,
σ mr (ϕ, 0)  
which is widely used to describe percolation systems, is in- fp = ϕ < ϕc ,
troduced in Section 2.3. This section is widely illustrated 2πε0 εmr (ϕ, 0)
by experimental results. Section 2.4 gives a brief discussion σ cr (0)  
of what features in the nanostructure determine the values fp = ϕ < ϕc ; high frequency , (2)
2πε0 εcr (0)
of the percolation exponents, s and t, as well as discussing
charging effects. Ways of differentiating between effective σ ir (0)  
fp = ϕ < ϕc ; low frequency ,
media and percolation media are discussed in Section 3. 2πε0 εir (0)
respectively. The subscript “m” refers to the properties of the
2. THEORY AND RESULTS composite, and “c” and “i” to those of the components. The
εxr (0) are the static dielectric constants. Above ϕc , the peak
The AC conductivity of the media (composites) (σ m ) is the frequency (ωp ) increases with ϕ. Note that in order to observe
sum of the real and imaginary conductivities, which is given all these arcs, the experimental apparatus must adequately
by σ m = σ mr + iσ mi . The conductivity of the more conduct- cover all ω values defined by (2).
ing component is given by σ c = σ cr + iσ ci or simply σ c = σ cr The MW/HS equations ([2] and the references therein),
if ideal conductivity (σ cr  σ ci ) is assumed. For the insu- derived directly from Maxwell’s equations, for the spherical
lating component, the conductivity is σ i = σ ir + iσ ii , where isotropic microstructure shown in Figure 1, are
σ ii = ωε0 εir . σ i is often approximated as σ i = iωε0 εir (i.e,  
σ ir  iσ ii ). In practice, σ ir incorporates both, a usually very σm − σc (1 − ϕ) σ i − σ c
small, DC conductivity and the dielectric polarization loss = , (3a)
σ m + 2σ c σ i + 2σ c
term (ωε0 εii ). The expressions for σ c and σ i can be disper-  
σm − σi ϕ σc − σi
sive and/or temperature-dependant. Although, to the best of = . (3b)
σ m + 2σ i σ c + 2σ i
the authors’ knowledge, it has never been attempted for an
experimental system, Nan [20] showed how a granular com- The MW/HS media can be visualized as built up out of a
ponent can be modeled as consisting of both a core and a space-filling array of coated spheres, with an infinite range of
coating. The theory given in this paper definitely applies to sizes, as illustrated in Figure 1. In (3a), σ c is the conductivity
micro- and macromedia and most nanostructures, but there of the coating or matrix material, while σ i is the conductiv-
are some special features regarding nanostructures which ity of the spherical core. In (3b), σ c is for the core while σ i is
must be considered. In nanostructures, the coating matrix for the coating (matrix). Note that the model has been widely
component thickness and the interparticle distances will of- and successfully applied to systems which differ from the ide-
ten be 10 nm thick or less. From 10 nm or less, the coatings alized microstructures shown in Figure 1, for instance, the
are below the tunneling distance for electrons, which would electrical conductivity results and the complex impedance
mean that the bulk conductivity (σ ir ) of the insulating phase plane plots for sintered polycrystalline yttria stabilized zir-
cannot be substituted into the expressions given below, but conia. With values of ϕ close to one (ϕ ≈ 0.999), micron-
should be obtainable by fitting the results. sized grains (σ c ), and grain boundaries (σ i ), a few nanome-
ters thick (3b) and the BLM are found to successfully fit ex-
2.1. Matrix media perimental data provided that suitable dispersive expressions
are substituted for σ c (ω) and σ i (ω) [25]. Suitable dispersive
In matrix media, the conductivity results are often presented expressions include the universal dielectric response function
using both the complex impedance (Z ∗ ) and modulus (M ∗ ) [26] and an expression due to León et al. [27].
representations. The notations adopted for Z ∗ and M ∗ are Other effective media theories are discussed in the fea-
Z ∗ = Z  − iZ  and M ∗ = M  + iM  , where M ∗ = iωZ ∗ , ture article by McLachlan et al. [2] and the review article
and Z  and Z  are the real and imaginary impedances, re- by Meredith and Tobias [28] which gives expressions for the
spectively, while M  and M  are the corresponding modulus electrical conductivity for a wide variety of other structures.
parameters. This leads to the following equations:
2.2. Percolation media: critical volume fraction
M  = ωZ  , M  = ωZ  . (1)
If conducting spheres (discs), of just sufficient size to touch
Note that the results are often presented as complex plane their nearest neighbors, are placed at random on the sites of
(Nyquist) plots (Z  − Z  and M  − M  ) or Z  -frequency 3D (2D) Bravais lattices, it is found that ϕc is 0.16 ± 0.02
and M  -frequency plots. When the more conducting phase (0.50 ± 0.02) [2, 29]. If equally sized conducting and noncon-
forms the “dominant” matrix, these plots show only a single ducting spheres are placed at random in a container, ϕc is also
D. S. McLachlan and G. Sauti 3

to intergranular channels (see Figure 2(c)). At a ϕ value of

0.03 (ϕc ), the channels become isolated. In the sintering of
ceramics, it is well known that closed porosity is usually
achieved at a value of about 3%. The next model is for cel-
lular structures, which is illustrated in Figure 3. Here, a fine
conducting powder has been compacted with a coarse insu-
lating one. Calculations, based on there being a percolation
path on the surface of the insulating spheres, have been made
by Kusy [30]. For instance, when the ratio of the radii of the
components is 30, ϕc ≈ 0.03. A series of such cellular systems,
which give ϕc ’s between 0.012 and 0.065, has been examined
in Chiteme and McLachlan [12]. For particles with irregular
geometries, especially extended ones (rods, discs), one must
use the excluded volume model of Balberg et al. [31]. This
has been used in explaining the results for single and multiple
wall carbon nanotubes (SWCNT and MWCNT), modeled as
random sticks in 3D (McLachlan et al. [22] and the references
therein). Figure 4 shows random sticks in 2D and Figure 5
shows ϕc as a function of the aspect ratio of the sticks. Ac-
cording to Figure 5, the SWCNT bundles should have a ϕc
higher than that observed [22]. Reasons for this are given by
McLachlan et al. [22].

2.3. Percolation media: the TEPPE and

experimental results
Conductor Conductor
Insulator Insulator Note that no experimental details will be given here; for these
(a) (b)
the reader should consult the original papers. As previously
stated, a recent series of papers [7–13, 16, 32] has shown that
Figure 1: The microstructure described by the (a) insulator-host the TEPPE, which is
and (b) conductor-host Maxwell-Wagner effective media (MW/HS)
theory. σ 1/s
i − σm
1/s
c − σm
σ 1/t 1/t
(1 − ϕ) 1/s + ϕ = 0, (4)
σ i + Aσ 1/s
m σ c + Aσ 1/t
1/t
m

with A = (1 − ϕc )/ϕc and the exponents s and t, best describes

experimental results for percolation systems, especially the
found to be about 0.16. This value of ϕc is often taken to char-
second-order terms [10, 11]. When s = t = 1 and ϕc = 1/3,
acterize 3D random media but other values of ϕc are permit-
the equation is equivalent to the Bruggeman symmetric me-
ted [2, 21, 29]. The case, where ϕc > 0.16, is best illustrated
dia equation [2, 21]. Equation (4) yields the two limits
using the grain consolidation model illustrated in Figures
 s  s
2(a) and 2(b). The model starts with the random nucleation   ϕc ϕc
σ c  −→ ∞ : σ m = σ i or εmr = εir ,
of conducting (white) spheres which grow to form a ran- ϕc − ϕ ϕc − ϕ
dom close-packed lattice of touching spheres at ϕ = 0.636.
As the radii of the spheres increase further, the interfacial ϕ < ϕc ,
contact area grows, as illustrated in Figure 2(b), which shows (5)
thin layers of insulating material between all the conduct-  t  
  ϕ − ϕc t
ing grains. Examining Figure 2(b) shows that the sample be- σ i  −→ 0 : σ m = σ c ϕ − ϕc or εmr = εir ,
comes conducting when a sufficient number of the barriers, 1 − ϕc 1 − ϕc
between the conducting grains, have been removed at ran- ϕ > ϕc .
dom for a percolation path to form, which, for thin barriers, (6)
can give a high value of ϕc . In the case of nanostructures, the
barriers will almost certainly be below the tunneling length These equations are the normalized standard percolation
so that a conducting critical path, which passes through a results [18, 19] and characterize the exponents s and t.
In the crossover region, where ϕ = ∼ ϕ , and which lies
series of sufficiently low tunnel barriers, will be formed at c
ϕc . Below ϕc , the electrons follow a “pseudopercolation” path between
through a series of barriers with the lowest resistances.  1/t+s  1/t+s
σi ωε0 εir
For ϕc < 0.16, three models can be considered. The first ϕc − or ϕc −
σc σ cr
is the grain consolidation model illustrated in Figures 2(c)  1/t+s  1/t+s (7)
and 2(d), where insulating spheres have now nucleated and σi ωε ε
< ϕ < ϕc + or ϕc + 0 ir ,
grown sufficiently to confine the conducting material (white) σc σ cr
4 Journal of Nanomaterials

(a) ϕ = 0.636 (b) ϕ = 0.800

(c) ϕ = 0.10 (d) ϕ = 0.03

Figure 2: The grain consolidation model.

Conductor
Figure 4: Percolation network of random sticks in two dimensions.
Insulator

Figure 3: Cellular micro-/nanostructure. ϕc = 0.03, for a radius

ratio of about 30.
portional to ωt/(s+t) and the dielectric constant εmr (recall that
εxy = σ xy /ωε0 ) is proportional to ωs/(s+t) . Note that (5) and
(4) gives (6) are not valid in the crossover region.
  The relationship between the DC exponents, s and t, and
t/(s+t)
σ m ≈ σ it/(s+t) σ cs/(s+t) or ωε0 εir s/(s+t)
σ cr , (8) the high frequency slopes, ut = t/(s + t), vt = s/(s + t) and
ut + vt = 1, in the crossover region, is a basic premise for uni-
which is in agreement with the theory given in Clerc et al. versal behavior, upon which the three standard percolation
[18] and Bergman and Stroud [19]. Equation (8) shows that, equations are based ([18–20] and the references therein). In
in the crossover region, the conductivity σ mr should be pro- this respect, all AC experimental results ([7–13, 16] and the
D. S. McLachlan and G. Sauti 5

10−1
Bundle: L/W = 20 100

10−4

σ(φ, 0) (Ω cm)−1
φc (random orientation)

Random sticks
10−2 in 3 dimensions
10−8

10−12
10−3

10−16
0 0.2 0.4 0.6 0.8
SWNT: L/W = 2000
Volume fraction φ
10−4
101 102 103 104 Figure 6: DC percolation results for Graphite-hBN discs. ϕc =
Length/width 0.150 ± 0.001, s = 1.01 ± 0.05, and t = 2.63 ± 0.07.

Figure 5: The percolation thresholds for random sticks.

references therein) are nonuniversal, with the results in Wu

and McLachlan [7, 8] being the closest to universal. There- 10−4
fore, these results are used to illustrate the first-order results,
σ mr above ϕc and εmr below ϕc , and the second-order re-
10−6
sults, εmr above ϕc and σ mr below ϕc . Arguably, the best DC
conductivity results, for a percolation system, are those mea-
σmr

sured on compressed Graphite-hBN discs, which are given in 10−8

Figure 6, from which ϕc , s, and t can easily be determined us-
ing (4). Figure 7 shows the AC conductivity plotted as a func-
tion of frequency for a loosely packed Graphite-hBN mix- 10−10
ture, and Figure 8 shows the dielectric constant as a function
of frequency. These results are presented as they are the only 10−12
ones where the results below ϕc can be scaled. Note how one 101 103 105 107
can clearly distinguish between samples above and below ϕc ω/2π (Hz)
in Figure 7. Close to ϕc , the slope in the high-frequency re-
gion for a percolation system should be t/(s + t) with s and Figure 7: AC conductivity plotted against frequency for a loosely
packed Graphite-hBN mixture. The range of ϕ values is between
t being determined from the DC results. Not too much can
0.112 and 0.133. A gap occurs between the low-frequency results
be determined about the dielectric constant from these plots,
above and below ϕc .
except that the linear slope, which should be s/(s + t), is only
observed for higher frequencies and very close to ϕc , as it is
to be expected from (5). Universality requires that the exper-
imental slopes of such plots be ue = t/(s+t) and ve = s/(s+t), dielectric loss) and the percolation clusters and their sum are
using the s and t determined from DC measurements. Note shown by the solid line in the figure (see Wu and McLachlan
that whereas the slopes in the graphite boron nitride system [8] for further details). To date, the percolation contribution
are close to the universal values, in most systems they are has been only clearly seen in this system, where the insula-
not [12, 16] and ue + ve < 1. The dielectric constant results tor is mainly dry air and a little dry BN. In most systems,
shown in Figure 8 are not even qualitatively in accordance the percolation contribution is swamped by the contribution
with the predictions of the standard equations. Results of from σ ii .
this nature can be qualitatively fitted using (4) [10]. However,
a more complex algorithm is needed for quantitative fitting 2.4. The exponents s and t
[11]. Note that the dielectric constant is frequency dependant
below ϕc [12]—something which has never been previously The universal value of the exponent t (tun ), which is usu-
taken into account theoretically, using (4) or any other ex- ally observed in computer simulations and for “ideal” sys-
pression. tems, is 2 [18, 19] while sun ≈ 0.87. However, values in the
Figure 9 shows a further plot of the conductivity as ranges of 0.37 < s < 1.28 and 1 < t < 6.27 can be found
a function of frequency, from a special experiment using in the literature ([15] and the references therein). Kogut and
loosely packed Graphite-hBN in ultra dry air [10]. The con- Straley [33] showed, in computer simulations, that if the
tributions to the conductivity from the insulator (DC and low-conductance bonds in the percolation network, or the
6 Journal of Nanomaterials

100 2

0
50
−2

log (σm ) ((Ωm)−1 )

−4
εmr

20

−6

10
−8

5 −10
101 103 105 107
ω/2π (Hz) −12
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 8: Dielectric constant plotted against frequency for a loosely Conducting component volume fraction (φ)
packed Graphite-hBN mixture. The range of ϕ values is between
0.112 and 0.123. The top curve is for a sample marginally above ϕc . Conductor-host MW/HS
TEPPE
Insulator-host MW/HS

Figure 10: DC conductivities for the conductor-host Maxwell-

10−4 Wagner/Hashin-Shtrikman equation, the TEPPE, and the insulator-
10−5 host Maxwell-Wagner/Hashin-Shtrikman equation. In the simula-
tion, σ cr = 102 (Ωm)−1 and σ ir = 10−12 (Ωm)−1 . For the TEPPE, ϕc =
10−6
0.16, s = 0.87, and t = 2.
Conductivity (ohms/cm)

10−7
10−8
10−9
10−10
neous (nongranular) conducting phase. A model for granu-
10−11
lar conducting systems which gives rise to t > tun was pro-
10−12 posed by Balberg [38]. This model is based on the dominant
10−13 resistances in the current carrying links and blobs (now con-
10−14 sisting of a granular conductor) being due to a large range
10−15 of interparticle tunneling contacts. Due to the characteris-
10−1 100 101 102 103 104 105 106 107 tic dimensions of nanostructures, the intergranular distances
Frequency (Hz) will often be 10 nm thick or less, that is, below the charac-
teristic tunneling distance. This means that the Balberg [38]
Figure 9: AC conductivity of loosely packed Graphite-hBN in ul-
tra dry air. The straight line is the calculated contribution from the
model will probably be the most appropriate for nanostruc-
percolation clusters and the lowest plot shows the total dielectric tures, where values of t greater than two are to be expected.
loss. One might expect the charging effect or coulomb block-
age in nanostructures because, for a 100 nm radius (R)
sphere carrying charge of one excess electron, the charging
energy, which is e2 /C or e2 /2πε0 R, is equal to the thermal
intergranular conductances of the conducting component in energy kT at 300 K. However, this will not affect the conduc-
a continuum system, have a very wide distribution, then t tivity of a bulk sample because, well below ϕc , there will be a
can be larger than tun . This distribution can be due to a large large number of hopping conductivity paths, with a number
range of effective geometrical resistivity factors in a continu- of different hopping lengths in each. This, combined with the
ous homogeneous conducting phase, as occurred in the Swiss fact that e2 /C will vary from site to site, will average out any
Cheese (random void (RV)) model and the inverse Swiss observable effects. Closer to ϕc , the hopping occurs between
Cheese (inverse random void (IRV)) model [32, 34]. In the random clusters which have larger and more varied C values,
Swiss Cheese model, a range of very thin and highly resistive which will again lead to a smearing out of charging effects. It
threads of the conductor (Cheese), between the large over- should also be noted that it has been shown by McLachlan et
lapping voids (air), give rise to a wide distribution in the al. [13] that the scaling results for percolation systems and
conductances between the more extended or bulky regions universal conductivity systems, which have been observed
of conductor. These models give values for t in the range of in amorphous conductors, ionic conductors, lightly doped
2–2.5. An extension to the RV model allowing still higher semiconductors, and other disordered systems, are very sim-
(apparently unlimited) values of t is given berg [35]. The ilar. As in some, but not all, universal conductivity systems,
links, blobs, and nodes model [36, 37] gives an upper limit coulomb blocking occurs; the presence of coulomb blocking
of 2.35. Note that all of the above models assume a homoge- does not necessarily invalidate the TEPPE.
D. S. McLachlan and G. Sauti 7

−2 1

−4
0.8

−6

−Z  / − Zmax
0.6


log σ ((Ωcm)−1 )

−8 Above φc

0.4
0.00035 < φc < 0.00075
−10

0.2
−12

−14 Below φc 0
−8 −6 −4 −2 0 2 4 6 8 10 12 14
−16 log(ω) (rad s−1 )
−3 −2 −1 0 1 2 3 4 5 6 7 φ=0 φ = 0.9
log f (Hz) φ = 0.1 φ = 0.9995
φ = 0.5 φ=1
φ=0 φ = 0.002
φ = 0.0002 φ = 0.005 Figure 12: The normalized impedance for an insulator-host
φ = 0.00035 φ = 0.01
Maxwell-Wagner/Hashin-Shtrikman system. For the ratio of com-
φ = 0.00075 φ = 0.02
φ = 0.001 φ = 0.05 ponent conductivities in the simulations, no peaks are observed be-
tween those of the insulating and conducting components. The sys-
Figure 11: The AC conductivity of SWCNT-polymer composites at tem is dominated either by the insulating or the conducting com-
various ϕ values. Note the big jump in the conductivity around ϕc ponent. For the simulations, σ cr = 102 (Ωm)−1 , εcr = 10, σ ir = 10−16 ,
[22]. and εir = 4 (For lower values of σ cr /σ ir two peaks are observed in
the spectrum).

fluenced not only by the structure of the composites, but also

3. DIFFERENTIATING BETWEEN A MATRIX AND by the dispersive properties of their components [25].
A PERCOLATION NANOSTRUCTURE Where AC conductivity data are available for vari-
ous compositions, there are various ways of distinguishing
Fully fitting DC conductivity measurements, over a wide between percolation and effective media systems. Percola-
range of compositions, is the easiest and clearest way of dis- tion and effective media systems can be differentiated by the
tinguishing nanostructures. Figure 10 shows the DC conduc- big changes (usually several orders of magnitude) that oc-
tivities for the conductor-host MW/HS equation, TEPPE, cur in the AC conductivities at ϕc for the percolation sys-
and the insulator-host MW/HS equation. It can be clearly tem. Figure 11 shows the AC conductivities, at various ϕ’s
seen that over a wide enough range of ϕ’s (on either side of for a SWCNT-polymer composite. The samples clearly lie in
ϕc for the percolation system), the three models predict vastly two groups: one below the percolation threshold and another
different conductivities. above. Such a big jump in the AC conductivity, for a relatively
Equally good, but involving far more work, is a full fit of small change in ϕ, is not observed for effective media systems,
the AC conductivity data. Youngs [14, 15] describes the use except where ϕ ≈ 0 or 1.
of a genetic algorithm that enables robust fitting of AC con- The changes in the positioning of peaks in Z  -frequency
ductivity data to the TEPPE (GEM). With sufficient care, and (or Z  − Z  ) plots can also be used to distinguish between
the use of both the AC and DC data, satisfactory results can systems described by the different models. For the insulator-
be obtained, from the TEPPE, for systems with data only on host MW/HS, the position of the impedance peak does not
one side of ϕc [39]1 . Various algorithms can be used to fit change much with increasing ϕ but instead remains under
data to the MW/HS equations [25, 39]. Unfortunately, it is or very close to the ϕ = 0 peak, until ϕ ≈ 1, as shown in
not possible to differentiate between the different nanostruc- Figure 12.
tures from features of the AC conductivity data, such as the For a system described by the conductor-host MW/HS,
high frequency slope of the conductivity-frequency curves, the peaks in Z  , as shown in Figure 13, remain close to the
for a sample of a single composition. These features are in- ϕ = 1 peak until ϕ ≈ 0.
For a percolation system, Z  peaks are seen at all fre-
quencies between the characteristic frequencies of the com-
1 Copies of the fitting and simulation algorithms implemented in MATH- ponents as shown in Figure 14. Up to ϕc , the impedance is
EMATICA can be obtained from G. S. dominated by the insulating component and the Z  peaks
8 Journal of Nanomaterials

1 1
0.9
0.8 0.8
0.7
−Z  / − Zmax

−Z  / − Zmax
0.6 0.6



0.5
0.4 0.4
0.3
0.2 0.2
0.1
0 0
−8 −6 −4 −2 0 2 4 6 8 10 12 14 −8 −6 −4 −2 0 2 4 6 8 10 12 14
log(ω) (rad s−1 ) log(ω) (rad s−1 )
φ=0 φ = 0.9 φ=0 φ = 0.2
φ = 0.1 φ = 0.9995 φ = 0.15 φ = 0.5
φ = 0.5 φ=1 φ = 0.16(φc ) φ=1
φ = 0.165

Figure 13: Normalized impedance for the conductor-host Figure 14: The normalized impedance for a percolation system.
Maxwell-Wagner/Hashin-Shtrikman model. Note that the single Note the peaks at frequencies in between those of the insulating
peak starts close to the ϕ = 1 peak, even for very low ϕ and moves and conducting components. For the simulations, σ cr = 102 (Ωm)−1 ,
to the right, closer to ϕ = 1 with increasing ϕ. For the simulations, εcr = 10, σ ir = 10−16 (Ωm)−1 , εir = 4, s = 1, t = 2, and ϕc = 0.16.
σ cr = 102 (Ωm)−1 , εcr = 10, σ ir = 10−16 (Ωm)−1 , and εir = 4.

are all superimposed onto the ϕ = 0 peak. However, just results for the composite using the MW/HS equations or the
above ϕc , the peak starts to move towards the ϕ = 1 peak TEPPE together with the best expressions for the dispersive
with increasing ϕ. conductivity of the other component.
Similar behavior as seen in Figures 12, 13, and 14 is
observed for smaller σ i /σ c with the only difference being REFERENCES
that the insulator and conductor peaks lie closer together. If
σ i /σ c > 10−3 , the peaks will start to overlap which may cause [1] Z. Hashin and S. Shtrikman, “A variational approach to the
problems. theory of the effective magnetic permeability of multiphase
materials,” Journal of Applied Physics, vol. 33, no. 10, pp. 3125–
3131, 1962.
4. CONCLUSIONS
[2] D. S. McLachlan, M. Blaszkiewicz, and R. E. Newnham, “Elec-
This paper has shown how the AC and DC conductivity (di- trical resistivity of composites,” Journal of the American Ce-
electric constant) results, for binary composites, with one ramic Society, vol. 73, no. 8, pp. 2187–2203, 1990.
or more of which having nanodimensions, can be analyzed [3] M. Campo, L. Woo, T. Mason, and E. Garboczi, “Frequency-
in terms of either (i) a matrix structure using the Maxwell- dependent electrical mixing law behavior in spherical particle
composites,” Journal of Electroceramics, vol. 9, no. 1, pp. 49–56,
Wagner/Hashin-Shtrikman (or Bricklayer) model or (ii) a
2002.
percolation structure using the two-exponent phenomeno-
[4] J.-H. Hwang, D. S. Mclachlan, and T. O. Mason, “Brick layer
logical percolation equation (TEPPE). It is also shown how
model analysis of nanoscale-to-microscale cerium dioxide,”
the two structures may be distinguished using electrical mea- Journal of Electroceramics, vol. 3, no. 1, pp. 7–16, 1999.
surements. As the actual complex conductivities appear in
[5] D. S. McLachlan, C. Chiteme, W. D. Heiss, and J. Wu, “The
expressions, it is theoretically possible to determine the dis-
correct modelling of the second order terms of the complex
persive conductivities for both components and, for perco- AC conductivity results for continuum percolation media, us-
lation systems, the critical volume fraction as well as the ex- ing a single phenomenological equation,” Physica B, vol. 338,
ponents s and t. Unfortunately, even when a large amount of no. 1–4, pp. 256–260, 2003.
data, on both sides of the critical volume fraction, are avail- [6] D. S. McLachlan, C. Chiteme, W. D. Heiss, and J. Wu, “Fitting
able, this is a formidable task due to the large number of un- the DC conductivity and first order AC conductivity results for
known parameters (a dispersive medium has to have two or continuum percolation media, using percolation theory and a
more). In practice, it will probably be necessary to know an single phenomenological equation,” Physica B, vol. 338, no. 1–
expression to directly measure the conductivity of one of the 4, pp. 261–265, 2003.
components. If only one component is nanosized, it should [7] J. Wu and D. S. McLachlan, “Percolation exponents and
be possible to deduce its dispersive conductivity by fitting the thresholds obtained from the nearly ideal continuum
D. S. McLachlan and G. Sauti 9

percolation system graphite-boron nitride,” Physical Review B, [25] G. Sauti, Electrical Conductivity and Permittivity of Ceramics
vol. 56, no. 3, pp. 1236–1248, 1997. and other Composites, Ph.D. thesis, Physics, University of the
[8] J. Wu and D. S. McLachlan, “Scaling behaviour of the com- Witwatersrand, Johannesburg, South Africa, 2005.
plex conductivity of graphite-boron nitride systems,” Journal [26] A. K. Jonscher, “The ‘universal’ dielectric response,” Nature,
of Physical Review B, vol. 58, no. 22, pp. 14880–14887, 1998. vol. 267, no. 5613, pp. 673–679, 1977.
[9] D. S. McLachlan and M. B. Heaney, “Complex ac conductivity [27] C. León, M. L. Lucı́a, and J. Santamarı́a, “Correlated ion hop-
of a carbon black composite as a function of frequency, com- ping in single-crystal yttria-stabilized zirconia,” Physical Re-
position, and temperature,” Physical Review B, vol. 60, no. 18, view B, vol. 55, no. 2, pp. 882–887, 1997.
pp. 12746–12751, 1999. [28] R. E. Meredith and C. W. Tobias, “Conduction in heteroge-
[10] D. S. McLachlan, W. D. Heiss, C. Chiteme, and J. Wu, “Ana- nous systems,” in Advances in Electrochemistry and Electro-
lytic scaling functions applicable to dispersion measurements chemical Engineering, C. Tobias, Ed., vol. 2, pp. 15–47, John
in percolative metal-insulator systems,” Physical Review B, Wiley & Sons, New York, NY, USA, 1962.
vol. 58, no. 20, pp. 13558–13564, 1998. [29] R. Zallen, The Physics of Amorphous Solids, John Wiley & Sons,
[11] W. D. Heiss, D. S. McLachlan, and C. Chiteme, “Higher-order New York, NY, USA, 1983.
effects in the dielectric constant of percolative metal-insulator [30] R. P. Kusy, “Influence of particle size ratio on the continuity
systems above the critical point,” Physical Review B, vol. 62, of aggregates,” Journal of Applied Physics, vol. 48, no. 12, pp.
no. 7, pp. 4196–4199, 2000. 5301–5305, 1977.
[12] C. Chiteme and D. S. McLachlan, “AC and DC conductiv- [31] I. Balberg, C. H. Anderson, S. Alexander, and N. Wagner, “Ex-
ity, magnetoresistance, and scaling in cellular percolation sys- cluded volume and its relation to the onset of percolation,”
tems,” Physical Review B, vol. 67, no. 2, Article ID 024206, 18 Physical Review B, vol. 30, no. 7, pp. 3933–3943, 1984.
pages, 2003. [32] B. I. Halperin, S. Feng, and P. N. Sen, “Differences be-
[13] D. S. McLachlan, G. Sauti, and C. Chiteme, “Static dielectric tween Lattice and continuum percolation transport expo-
function and scaling of the ac conductivity for universal and nents,” Physical Review Letters, vol. 54, no. 22, pp. 2391–2394,
nonuniversal percolation systems,” Physical Review B, vol. 76, 1985.
no. 1, Article ID 014201, 13 pages, 2007. [33] P. M. Kogut and J. P. Straley, “Distribution-induced non-
[14] I. J. Youngs, Electrical percolation and the design of functional universality of the percolation conductivity exponents,” Jour-
electromagnetic materials, Ph.D. thesis, Electronic and Electri- nal of Physics C, vol. 12, no. 11, pp. 2151–2159, 1979.
cal Engineering, University College, London, UK, 2001. [34] S. Feng, B. I. Halperin, and P. N. Sen, “Transport properties of
[15] I. J. Youngs, “Exploring the universal nature of electrical per- continuum systems near the percolation threshold,” Physical
colation exponents by genetic algorithm fitting with general Review B, vol. 35, no. 1, pp. 197–214, 1987.
effective medium theory,” Journal of Physics D, vol. 35, no. 23, [35] I. Balberg, “Limits on the continuum-percolation transport
pp. 3127–3137, 2002. exponents,” Physical Review B, vol. 57, no. 21, pp. 13351–
[16] C. Chiteme, D. S. McLachlan, and G. Sauti, “AC and DC per- 13354, 1998.
colative conductivity of magnetite-cellulose acetate compos- [36] H. E. Stanley, “Cluster shapes at the percolation threshold:
ites,” Physical Review B, vol. 75, no. 9, Article ID 094202, 13 and effective cluster dimensionality and its connection with
pages, 2007. critical-point exponents,” Journal of Physics A, vol. 10, no. 11,
[17] C. Brosseau, “Generalized effective medium theory and dielec- pp. L211–L220, 1977.
tric relaxation in particle-filled polymeric resins,” Journal of [37] A. Coniglio, “Cluster structure near the percolation thresh-
Applied Physics, vol. 91, no. 5, pp. 3197–3204, 2002. old,” Journal of Physics A, vol. 15, no. 12, pp. 3829–3844, 1982.
[18] J. P. Clerc, G. Giraud, J. M. Laugier, and J. M. Luck, “The [38] I. Balberg, “Tunneling and nonuniversal conductivity in com-
electrical conductivity of binary disordered systems, percola- posite materials,” Physical Review Letters, vol. 59, no. 12, pp.
tion clusters, fractals and related models,” Advances in Physics, 1305–1308, 1987.
vol. 39, no. 3, pp. 191–309, 1990. [39] G. Sauti and D. S. McLachlan, “Impedance and modulus spec-
[19] D. J. Bergman and D. Stroud, “The physical properties of tra of the percolation system silicon-polyester resin and their
macroscopically inhomogeneous media,” Solid State Physics, analysis using the two exponent phenomenological percola-
vol. 46, pp. 148–270, 1992. tion equation,” Journal of Materials Science, vol. 42, no. 16, pp.
[20] C.-W. Nan, “Physics of inhomogeneous inorganic materials,” 6477–6488, 2007.
Progress in Materials Science, vol. 37, pp. 1–116, 1993.
[21] D. S. Mclachlan, “Analytical functions for the DC and AC con-
ductivity of conductor-insulator composites,” Journal of Elec-
troceramics, vol. 5, no. 2, pp. 93–110, 2000.
[22] D. S. McLachlan, C. Chiteme, C. Park, et al., “AC and DC per-
colative conductivity of single wall carbon nanotube polymer
composites,” Journal of Polymer Science Part B, vol. 43, no. 22,
pp. 3273–3287, 2005.
[23] D. S. McLachlan, R. Rosenbaum, A. Albers, et al., “The tem-
perature and volume fraction dependence of the resistivity
of granular Al-Ge near the percolation threshold,” Journal of
Physics: Condensed Matter, vol. 5, no. 27, pp. 4829–4842, 1993.
[24] D. S. McLachlan, J.-H. Hwang, and T. O. Mason, “Evaluating
dielectric impedance spectra using effective media theories,”
Journal of Electroceramics, vol. 5, no. 1, pp. 37–51, 2000.
Journal of International Journal of International Journal of Smart Materials Journal of
Nanotechnology
Hindawi Publishing Corporation
Corrosion
Hindawi Publishing Corporation
Polymer Science
Hindawi Publishing Corporation
Research
Hindawi Publishing Corporation
Composites
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of
Metallurgy

BioMed
Research International
Hindawi Publishing Corporation Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Nanomaterials
Submit your manuscripts at
http://www.hindawi.com

Journal of Journal of
Materials
Hindawi Publishing Corporation
Nanoparticles
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Nanomaterials
Journal of

Advances in The Scientific International Journal of

Materials Science and Engineering
Hindawi Publishing Corporation
Scientifica
Hindawi Publishing Corporation Hindawi Publishing Corporation
World Journal
Hindawi Publishing Corporation
Biomaterials
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of Journal of Journal of Journal of Journal of

Nanoscience
Hindawi Publishing Corporation
Coatings
Hindawi Publishing Corporation
Crystallography
Hindawi Publishing Corporation
Ceramics
Hindawi Publishing Corporation
Textiles
Hindawi Publishing Corporation
http://www.hindawi.com
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Volume 2014