Journal of Non-Crystalline Solids 423–424 (2015) 1–8

Contents lists available at ScienceDirect

Journal of Non-Crystalline Solids

journal homepage: www.elsevier.com/ locate/ jnoncrysol

Temperature and frequency dependent conductivity and electric

modulus formulation of manganese modified bismuth silicate glasses
Meenakshi Dult a, R.S. Kundu a, J. Hooda a, S. Murugavel b, R. Punia a,⁎, N. Kishore a
a
Department of Applied Physics, Guru Jambheshwar University of Science & Technology, Hisar 125001, India
b
Department of Physics and Astro-physics, University of Delhi, Delhi 110007, India

a r t i c l e i n f o a b s t r a c t

Article history: The ac conductivity and electric modulus formalism of bismuth silicate glasses containing manganese with com-
Received 26 February 2015 positions xMnO2·(60 − x)Bi2O3·40SiO2, (x = 0, 5, 10, 15 and 20) have been studied in the frequency range
Received in revised form 4 May 2015 10−1 Hz to 1 MHz and temperature range 583 K to 703 K. The ac conductivity is found to increase with an in-
Accepted 10 May 2015
crease in MnO2 content. The frequency dependent conductivity for manganese doped bismuth silicate glassy sys-
Available online 15 May 2015
tem shows a single plateau for samples with x = 0 and 5, but for glasses with x = 10, 15 and 20, the temperature
Keywords:
and frequency dependent conductivity shows a double plateau in the whole temperature range 583 K to 703 K.
Oxide glasses; The observed dispersive behavior of ac conductivity of bismuth silicate glasses containing manganese is
AC conductivity; discussed in the framework of Jonscher's universal power law fitted to the experimental data of samples showing
Jonscher's universal power law; single as well as two plateau regions. The results show that the ac conductivity increases with the increase in the
Small polarons; frequency of applied field. Further, it is observed that small polaron conduction mechanism is more or less suit-
Electric modulus formulation able to explain the conduction in the present glass samples with x = 10, 15 and 20 in the high frequency region.
However, the temperature dependent behavior of the frequency exponent s for glass samples with x = 0 and 5
seems to follow CBH conduction mechanism. The electrical relaxation is rationalized using the electric modulus
formulation in the presently studied glasses at all the temperatures. The values of activation energy for dc con-
duction (EA) and electric modulus (ER) were estimated. For glass sample with x = 0 and 5, the activation energy
associated with electric modulus is almost equal to that of the dc conductivity at the low frequency range, where-
as for glass samples with x = 10, 15 and 20, the values of ER are equal to the activation energy due to dc conduc-
tion in the high frequency region.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction overview of the electrical properties like ac conductivity and dc conduc-

tivity along with electric modulus formulation in manganese modified
Glasses containing transition metal oxides (TMO) show semicon- bismuth silicate glasses as a function of frequency and temperature.
ducting properties due to the presence of more than one valance state
of the transition metal ions [1–9]. These glasses are technologically im- 2. Material and methods
portant due to their remarkable applications in switching and electro-
chromatic devices, fuel cells, solid state batteries, chemical sensors, Glass samples having compositions xMnO2·(60 − x)Bi2O3·40SiO2,
etc. [10–16]. The charge transport in these glasses is believed to take (x = 0, 5, 10, 15 and 20) were prepared using standard melt-
place by hopping of electrons from the low valence state to high valence quenching technique. The appropriate amounts of these chemicals
state of transition metal ions [1,2], for example, in glasses containing were thoroughly mixed in an agate pestle–mortar. Silica crucible con-
manganese, the electronic conduction occurs due to the hopping of a taining the mixture was put in an electrically heated muffle furnace
3d electron from the Mn2+ to Mn3+ ion. The introduction of TMI, like and the temperature was raised slowly to a temperature between
manganese, to the bismuth silicate glasses is expected to improve 1000 and 1150 °C depending on the composition. The temperature
their electrical and dielectric features. The conductivity of bismuth sili- was maintained for 1 h and the melt was shaken frequently to ensure
cate [17,18] and manganese borate glasses [19,20] has been reported proper mixing and homogeneity. The melt was then poured onto a
by many researchers, but there is hardly any report available in litera- stainless steel block and was immediately pressed by another similar
ture on electronic transport of manganese bismuth silicate glass system. block, at room temperature. The amorphous nature of samples was con-
The objective of the present study is to analyze a comprehensive firmed by the absence of any sharp peak in the X-ray diffraction (XRD)
patterns of the synthesized glass samples, recorded by using Rigaku
⁎ Corresponding author. Table-Top X-Ray Diffractometer. The prepared glass samples were cut
E-mail address: rajeshpoonia13@gmail.com (R. Punia). into small pieces of rectangular shapes with thickness ~ 1 mm and

http://dx.doi.org/10.1016/j.jnoncrysol.2015.05.021
0022-3093/© 2015 Elsevier B.V. All rights reserved.
2 M. Dult et al. / Journal of Non-Crystalline Solids 423–424 (2015) 1–8

their surfaces were smoothed and polished. For electrical measure- glass samples with x = 0 and 5 are observed to obey single Jonscher's
ments, both sides of the samples were coated with the silver paint so power law at all the temperatures. The typical fitting of Eq. (1) with ex-
that they serve as electrodes. Conductivity measurements were carried perimental data for glass samples with x = 5 is shown in Fig. 2. For glass
out using Alpha-A High Resolution Dielectric, Conductivity, Impedance samples with x = 10, 15 and 20, where an additional plateau is also ob-
and Gain Phase Modular Measurement System (Novo control Technol- served at intermediate higher frequencies, the ac conductivity cannot be
ogies GmbH & Co. KG) in the frequency range of 10− 1 Hz to 1 MHz accounted for by Eq. (1) only, rather two such formulae have been used
and temperature ranging from 583 K to 703 K. The fitting of experimen- in different frequency regions to explain the presence of two relaxation
tal data was done using linear and non-linear curve fitting modules of mechanisms [27]. The ac conductivity curve is, therefore, bifurcated into
origin Pro 8.6 software and errors in different parameters have been es- two parts and the experimental data of both parts are independently
timated by fitting. fitted with Jonscher's power law. In both regions the experimental
data of conductivity show good fitting with the Eq. (1). Thus single
3. Results and discussion Jonscher's power law is fitted to the experimental ac conductivity data
of the present oxide glasses in both bifurcated regions of low as well
3.1. Alternating current (AC) conductivity as high frequencies for glass samples with x = 10 and 15. The values
of σdc, ωH, and s have been obtained for low and high frequency plateau
The total frequency dependent conductivity (σ′(ω)), of different regions independently, wherever observed.
glass compositions 40SiO2–(60 − x)Bi2O3–xMnO2 with x = 0, 5, 10, The fitting of single Jonscher's power law with experimental data in
15 and 20, measured in temperature range 583 K to 703 K and frequen- low and high frequency regions for glass samples with x = 10 and 15
cy range 10− 1 Hz to 1 MHz, is found to increase with an increase in were similar and typical curves for a glass sample with x = 10 are
MnO2 content. The frequency dependent conductivity of glass samples, shown in Fig. 3. For experimental data of sample with x = 20, as
presented in Fig. 1 at any particular temperature (683 K), is character- shown in Fig. 4, the first plateau does not seem to be complete in the
ized by two regions: (i) a plateau region and (ii) dispersion region for studied frequency range, rather it appears that might appear at still
glass samples with x = 0 and 5. But for higher manganese content i.e. lower frequencies and hence for this sample, single Jonscher's power
x = 10, 15 and 20, it shows an additional plateau region at intermediate law was applied for high frequency plateau region. The appearance of
frequencies as shown in Fig. 1. The appearance of this additional plateau the second plateau in the intermediate/high frequency range in glass
occurs not only at 683 K but at all the studied temperatures for glass compositions with x = 10, 15 and 20 exhibiting heterogeneous struc-
samples with x = 10, 15 and 20. The width of the additional plateau is tures with different phases, may be due to the Maxwell–Wagner effect.
observed to decrease with increase in MnO2 content. It is not an unex- The Maxwell–Wagner effect is generally observed in the heterogeneous
pected result in heterogeneous glasses like the present ones, as this sec- system where the difference in conductivity of the two phases is very
ond plateau is due to the heterogeneity (difference) in conductivities of high [21,28]. In the present system, the MnO2 is added in the bismuth
manganese and silicate networks. This type of behavior of conductivity silicate matrix. Generally, silicate glasses are insulators but MnO2 is
is also reported for other ionic glasses and ceramics having different semiconducting and hence there is a large difference of conductivity
phases [21]. Electrical conductivity in the present glasses increases in manganese and silicate networks/phases [29,30]. So, the second pla-
with the increase in temperature as a result of the increase in drift mo- teau is observed in the glass samples with higher manganese concentra-
bility of the thermally activated charge carriers [22,23]. tion, wherein MnO2 rich phase is expected to be formed.
In general, the ac conductivity of hopping charges for many semi- If the conduction occurs via a defect mechanism then at a given in-
conducting amorphous materials exhibit Jonscher's power law [6,24, stant only a fraction (n) of the total charge carriers (N) are mobile,
25]: which is given by [6,31]

s 
ω



σ 0 ðωÞ ¼ σ dc 1 þ ð1Þ Gf Sf Hf
ωH n ¼ N exp − ¼ N exp  exp − ð2Þ
kT k kT
where σdc is dc conductivity, ωH is crossover frequency separating dc re-
gime (plateau region) from the dispersive conduction and s is frequency
exponent that lies between 0.5 and 1 [25,26]. The experimental data of

Fig. 2. Measured total ac conductivity (σ′(ω)) versus frequency curves for

xMnO 2 ·(60 − x)Bi2 O 3 ·40SiO 2 with x = 5 at thirteen different temperatures. The
Fig. 1. Compositional variation of total ac conductivity (σ′(ω)) of xMnO 2 · solid lines in the figure are the best fits obtained from fitting of experimental data
(60 − x)Bi2 O 3 ·40SiO 2 glass system at 683 K. with Jonscher's power law.
 M. Dult et al. / Journal of Non-Crystalline Solids 423–424 (2015) 1–8 3

are shown in Fig. 5. The curves for high frequency region are shown in
Fig. 6. The values of Hf + Hm and Hm as calculated from the slopes of
plots of log (σdcT) and log ωH versus 1000/T using linear fitting for all
the compositions of 40SiO2–(60 − x)Bi2O3–xMnO2 glass system in the
low and high frequency region are presented in Table 1. The corre-
sponding plots of log ωH versus 1000/T for low and high frequency re-
gions are shown in Figs. 7 and 8 respectively.
The data presented in Table 1, shows that Hf N 0, for all glass samples,
whereas its value increases with increase in MnO2 content and there-
fore, on addition of MnO2 to bismuth silicate glass system, it is found
that the number of mobile charge carriers at any instant becomes less
than total number of charge carriers [6]. The conductivity of glass sam-
ples is observed to increase with increase in MnO2 content (as shown in
Fig. 1). The increase in conductivity may be either due to increase in mo-
bility of charge carriers or increase in the number of mobile charge
carriers [22,23].

3.2. Variation of frequency exponent (s)

Fig. 3. Total ac conductivity (σ′(ω)) of xMnO2·(60 − x)Bi2O3·40SiO2 in glass system for
x = 10. The solid lines in the figure are the best fits obtained from fitting of experimental Several models based on classical hopping of charge carriers over
data with independent Jonscher's power law in low and high frequency regions. barrier and quantum mechanical tunneling and the overlapping large-
polaron tunneling, [2,32–36] have been proposed to investigate the
where Sf, Hf and Gf are entropy, enthalpy and Gibb's free energy, respec- conduction mechanism on the basis of variation of frequency exponent
tively, to dissociate the cation from its original site next to a charge com- with temperature and frequency. However, for the presently studied
pensating center. The crossover frequency (ωH) and dc conductivity glass system, in the studied frequency range, no frequency dependence
(σdc) are given [31] by of s is observed, and so temperature dependence of s plays a key role in


 estimation of conduction mechanism. The variation of frequency expo-
Sm Hm nent (s) with temperature at low and high frequency regions [obtained
ωH ¼ ωo exp  exp − ð3Þ
k kT from fitting of experimental data to Eq. (1)] are shown in Figs. 9 and 10
respectively. It is noted that s decreases with an increase in temperature
and in the low frequency region (for a glass sample with x = 0, 5, 10 and 15),
whereas s increases with an increase in temperature in the high fre-
2


NZ 2 e2 d γωo S f þ Sm H f þ Hm quency region (for samples with x = 10, 15 and 20). The variation of fre-
σ dc ¼ exp  exp − ð4Þ
kT k kT quency exponent with temperature correlates with the ac conductivity
mechanism:
where ωo is the so called attempt frequency, Sm and Hm are entropy and
enthalpy of migration, respectively, γ is a constant including geometric (1) The frequency exponent depends upon frequency but is inde-
and correlation factors and d is the jump distance. Eqs. (3) and (4) sug- pendent of temperature, then the conduction mechanism can
gest linear relationship for log ωH versus 1000/T and log (σdcT) versus be explained by the quantum mechanical electron tunneling
1000/T curves. The experimental data when fitted with the above equa- theory [6,24].
tions exhibit the good linearity indicating that the conduction in present (2) If the frequency exponent increases with an increase in temper-
glass occurs through defect mechanism. The plots of log (σdcT) versus ature, then the conduction process can be explained with the
1000/T for samples with x = 0, 5, 10 and 15 in low frequency region small polaron quantum mechanical tunneling theory whereas if

Fig. 4. Total ac conductivity log (σ′(ω)) of xMnO2·(60 − x)Bi2O3·40SiO2 in the glass sys- Fig. 5. Linear fit of log (σdcT) versus 1000/T for 40SiO2–(60 − x)Bi2O3–xMnO2 glass system,
tem for x = 20. The solid lines in the figure are the best fits obtained from fitting of exper- where, σdc is dc conductivity and T is absolute temperature for glass samples with x = 0, 5,
imental data with independent Jonscher's power law only at the high frequency region. 10 and 15 in the low frequency region.
4 M. Dult et al. / Journal of Non-Crystalline Solids 423–424 (2015) 1–8

Fig. 6. Linear fit of log (σdcT) versus 1000/T for glass samples with x = 10, 15 and 20 in the Fig. 7. Linear fit of log ωH versus 1000/T for 40SiO2–(60 − x)Bi2O3–xMnO2 glass system,
high frequency region. where, ωH is crossover frequency and T is absolute temperature for glass samples with
x = 0, 5, 10 and 15 in the low frequency region.

frequency exponent first decreases, reaches a minimum and in- 3.3. Direct current (dc) conductivity
creases thereafter with an increase in temperature then it can
be explained by the large polaron quantum mechanical tunnel- The values of the dc conductivity have been obtained by fitting of ex-
ing model [6,24]. perimental data of conductivity (σ′(ω)) with Eq. (1) at different tem-
(3) If frequency exponent decreases with temperature, then it peratures for all the five glass compositions in the low and high
follows the correlated barrier hopping (CBH) conduction frequency regions [7,24]. It is observed that it lies in the range from
mechanism [6,24]. 10−9 to 10−5 S cm−1.
The temperature variation of dc conductivity for the studied glasses
follows the Arrhenius equation [22,24]
So, in the glass samples with x = 0, 5, 10 and 15, the conduction may
occur due to CBH conduction mechanism, whereas for samples with σ dc ðT Þ ¼ σ 0 exp½‐ðEA =kT Þ ð6Þ
x = 10, 15 and 20 in the high frequency region, the conduction process
can be explained with small polarons quantum mechanical tunneling
theory. If small polarons are formed, then according to tunneling where σ0 is the pre-exponential parameter which depends on semicon-
model [6] frequency exponent (s) becomes ductor nature, EA denotes the thermal activation energy of electrical
conduction, and k is Boltzmann's constant. The activation energy (EA)
is calculated from the slope of linear fit of log σdc versus 1000/T plot in
4 the low and high frequency region as shown in Figs. 11 and 12. The
s ¼ 1−
 ð5Þ
1 ωH values of activation energies (EA) in the low as well as high frequency
ln −
ωτ0 KBT regions are presented in Table 1 is observed to decrease with the in-
crease in concentration of MnO2 indicating thereby the increase in
conductivity.
Eq. (5) suggests that s increases with increase in temperature. In
present samples small polaron conduction mechanism is predominant
at high frequencies. Similar behavior has been reported by several au-
thors for TMO glasses [6,37–40]. However, conduction mechanism at
low frequencies in glass sample with x = 0, 5, 10 and 15 may be due
to CBH conduction.

Table 1
Enthalpy of migration (Hm), enthalpy to dissociate the cation from its original site next to a
charge compensating center (Hf), activation energy of dc conduction (EA) and activation
energy of electric modulus (ER) of 40SiO2·(60 − x)Bi2O3·xMnO2 glasses with different
values of x in the low frequency region (LFR) and high frequency region (HFR).

Parameters x=0 x=5 x = 10 x = 15 x = 20

Hf + Hm (eV) LFR 1.381 1.057 0.707 0.511 –

(±0.001) HFR – – 0.844 0.797 0.731
Hm (eV) LFR 1.302 0.879 0.412 0.426 –
(±0.001) HFR – – 0.757 0.608 0.520
Hf (eV) LFR 0.079 0.178 0.295 0.085 –
(±0.001) HFR – – 0.087 0.189 0.211
EA (eV) LFR 1.325 1.002 0.652 0.455 –
(±0.001) HFR – – 0.789 0.742 0.676
ER (eV) 1.324 1.063 0.793 0.695 0.675
Fig. 8. Linear fit of log ωH versus 1000/T for glass samples with x = 10, 15 and 20 in the
(±0.001)
high frequency region.
 M. Dult et al. / Journal of Non-Crystalline Solids 423–424 (2015) 1–8 5

Fig. 9. Variation of frequency exponent (s) as a function of temperature (T) for

Fig. 11. Linear fit of log (σdc) versus 1000/T for 40SiO2–(60 − x)Bi2O3–xMnO2 glass system,
xMnO2·(60 − x)Bi2O3·40SiO2 glass compositions with x = 0, 5, 10 and 15 in the low fre-
where, σdc is dc conductivity and T is the absolute temperature for the glass sample with
quency region.
x = 0, 5, 10 and 15 in the low frequency region.

The concentration of manganese ions (N) present in the glass matrix

has been estimated by using the following relation [7] hopping near the Fermi level is calculated by the relation [7]
 
W MnO2
N ¼ 2dNA ð7Þ 3
MMnO2 NðE F Þ ¼ ð10Þ
4πR3 EA

where d is the density, MMnO2 is molecular weight of MnO2, WMnO2 is

weight fraction of MnO2 and NA is the Avogadro's number. The relation- and its values come out to be of the order of 1021 eV−1 cm−3 as given in
ship between N and mean spacing between any two Mn-ions (R) is gen- Table 2, which are reasonable for localized states [41,42].
erally described as [7]
3.4. Electric modulus formulation
 13
1
R¼ : ð8Þ The electric modulus was originally introduced by Macedo et al. [43]
N
to study the space charge relaxation phenomena. Physically, the electric
modulus corresponds to the relaxation of the electric field in the mate-
The polaron radius (rp) is related to R by the following relation [7] rial when the electric displacement remains constant, so that the com-
plex electric modulus which is the reciprocal of the permittivity and
1 hπ i3
1

rp ¼ R: ð9Þ represents the real dielectric relaxation process, can be expressed as

2 6 [43]

The calculated values of N, R and rp are included in Table 2. The data

presented in Table 2, shows that R decreases with an increase in MnO2 M ¼ ðεÞ−1 ð11Þ
content. The density of states, N(EF), for thermally activated electron

Fig. 10. Variation of frequency exponent (s) for glass samples with x = 10, 15 and 20 in the Fig. 12. Linear fit of log (σdc) versus 1000/T for glass samples with x = 10, 15 and 20 in the
high frequency region. high frequency region.
6 M. Dult et al. / Journal of Non-Crystalline Solids 423–424 (2015) 1–8

Table 2 presently studied glass system, M′ shows the general trend as observed
Concentration of Mn-ions (N), Mean spacing between Mn-ions (R), polaron radius for many other disordered materials [21,44,45] i.e. for each tempera-
(rp), density of states, N(EF) near the Fermi level in the low frequency region (LFR)
and high frequency region (HFR), and relaxation time of electric modulus (τ M″) of
ture, M′ reaches a constant value at high frequencies. At low frequen-
40SiO2·(60 − x)Bi2O3·xMnO2 glasses with different values of x. Fig. 1. Compositional cies, M′ approaches zero, confirming the presence of an appreciable
variation of total ac conductivity (σ′ (ω)) of xMnO2·(60 − x)Bi2O3·40SiO2 glass sys- electrode and/or ionic polarization in the studied temperature range
tem at 683 K. and M′ increases with increase in frequency and attains a maximum
Physical parameters x=0 x=5 x = 10 x = 15 x = 20 value, M∞, at higher frequencies due to relaxation processes. It is also ob-
served that the value of M∞ decreases with an increase in temperature.
N (×1021 cc) (±0.001) – 0.678 1.405 2.256 3.247
R(Å) (±0.1) – 11.4 8.9 7.6 6.8 The imaginary part of electric modulus (M″) represents the energy loss
rp (Å) (±0.1) – 4.6 3.6 3.1 2.7 under electric field. It (M″) may be expressed as Fourier transform of a
N(EF) (×1021 eV−1 cm−3) LFR – 0.16 0.52 1.20 – relaxation function Φ (t) [44]
(±0.01) HFR – – 0.43 0.73 1.12
τM″ (μs) at 683 K (±0.001) 2 3
47.729 16.342 1.378 0.225 0.064 Z∞

dφ
M ¼ M ∞ 41− expð−ωt Þ

dt 5 ð13Þ
dt
0
and
where the function φ(t) is the time evolution of the electric field within
 0 ″ ε0 ε″ the material and is usually taken as the Kohlrausch–Williams–Watts
M ¼ M þ iM ¼ 2
þi 2
ð12Þ
ðε0 Þ2 þ ðε″ Þ ðε0 Þ2 þ ðε″ Þ (KWW) function [46,47]
"
 #
where M′, M″ and ε′, ε″ are the real and imaginary parts of the electric t β
φðt Þ ¼ exp − ð14Þ
modulus and dielectric constants, respectively. Fig. 13(a) and (b) τm
show the real and imaginary parts of electric modulus, M′ and M″, re-
spectively, as a function of frequency at various temperatures. For the
where τm is the conductivity relaxation time and the exponent β indi-
cates the deviation from Debye type relaxation. The electric modulus
behavior of the presently studied glass system is rationalized by invok-
ing the modified KWW function suggested by Bergman [48]. According
to this, the imaginary part of electric modulus may be represented as

M″
M″ ¼ "
max 
β # ð15Þ
β f f
ð1−βÞ þ β max þ
1þβ f f max

where M″max is the peak value of M″ and fmax is the corresponding

frequency.
As shown in Fig. 13(b), the above Eq. (15) is theoretically fitted to
the experimental data for present samples gives very good fitting and
shows that the peak of M″ shifts to higher frequencies with increase in
temperature. This is an evidence of temperature dependent relaxation
process in the presently studied glasses. The frequency region below
the M″ peak indicates the frequency range in which polarons drift to
long distances and in the frequency range above the peak, the polarons
are confined to potential wells and free to move within the wells. So, the
peak frequency represents the transition from long range to short range

Fig. 13. (a). Real part of electric modulus (M′) as a function of frequency for
5MnO2·55Bi2O3·40SiO2 glass samples at different temperatures. (b). Fitting of imaginary
part of electric modulus (M″) as a function of frequency with Eq. (15) at different temper- Fig. 14. Variation of exponent (β) with temperature for xMnO2·(60 − x)Bi2O3·40SiO2
atures for 15MnO2·45Bi2O3·40SiO2 glass samples. with x = 0, 5, 10, 15 and 20. The solid line is a guide to the eye.
 M. Dult et al. / Journal of Non-Crystalline Solids 423–424 (2015) 1–8 7

mobilities. It is observed that with increase in manganese content, the 4. Conclusions

value of peak frequency (fmax) increases.
The values of β, obtained from fitting, are presented in Fig. 14 and are The ac conductivity of manganese doped bismuth silicate glasses has
found to increase with increase in temperature for all the glass compo- been investigated in the frequency range 10− 1 Hz to 1 MHz and the
sitions. The increase in β with rise in temperature indicates that as tem- temperature range 583 K to 703 K. The ac conductivity is found to in-
perature is increased, the glass network gets loosened and thereby crease with increase in MnO2 content. The observed dispersive behavior
weakening the interactions between polarons and surrounding matrix. of ac conductivity of bismuth silicate glasses containing manganese is
In the present study, the value of β b 1 which suggests that the non- studied in the framework of Jonscher's universal power law which
Debye type of relaxation prevails in these samples [49]. when fitted to the experimental data exhibit two plateau regions. Fur-
The characteristic frequency at which M″ is maximum (M″max), cor- ther, it is observed that small polaron tunneling model is by and large
responds to relaxation frequency (fmax), which has been determined suitable to describe the conduction mechanism in the glass samples
from the fitting of M″ versus frequency experimental data with with x = 10, 15 and 20, whereas, the conduction in glass sample with
Eq. (15). The relaxation time τM″ has been calculated by the relation; x = 0 and 5 obeys the CBH mechanism. It is also observed that both
the dc activation energy (EA) and the electric modulus activation energy
1 (ER) show a similar trend of decrease in their values with increase in
τ M″ ¼ ð16Þ manganese content (x). For glass samples with x = 0 and 5, the activa-
2π f max
tion energy associated with electric modulus is almost equal to that of
the dc conductivity at the low frequency range, whereas for glass sam-
and it is found to lie in the range 2.020 ms to 0.045 μs. The relaxation
ples with x = 10, 15 and 20, the values of ER are equal to the activation
time shows a decreasing trend with increase in manganese content
energy due to dc conduction in the high frequency region, indicating
(relaxation time of various compositions at 683 K is presented in
that the charge carriers have to overcome the same energy barrier dur-
Table 2). This is an expected trend as the conductivity of the present sys-
ing conduction as well as relaxation processes. The value of the expo-
tem increases with increase in manganese content. The activation ener-
nent parameter β b 1 reveals the presence of non-Debye type of
gy (ER) involved in the relaxation process could be obtained from the
relaxation in the present samples.
temperature dependent relaxation time as

 Acknowledgments
ER
τM″ ¼ τo exp ð17Þ
kT
Authors are thankful to University Grants Commission (UGC), New
Delhi (Major Project No. 41-862/2012(SR)) for providing financial
where ER is the activation energy associated with the relaxation process, assistance.
τo is pre exponential factor, k is Boltzmann constant, and T is tempera-
ture in Kelvin.
References
Fig. 15 shows plots between log τM″ and 1/T for different glass sys-
tems along with their linear fitting. The estimated values of activation [1] N.F. Mott, J. Non-Cryst. Solids 1 (1968) 260.
energy of electric modulus are presented in Table 1. The data presented [2] I.G. Austin, N.F. Mott, Adv. Phys. 18 (1969) 41.
[3] L. Murawski, C.H. Chung, J.D. Mackenzie, J. Non-Cryst. Solids 32 (1979) 91.
in Table 1, shows that both the dc activation energy (EA) and the electric [4] A. Ghosh, S. Bhattacharya, A. Ghosh, J. Phys. Condens. Matter 21 (2009) 145802.
modulus activation energy (ER) exhibit a decrease in their values with [5] A. Ghosh, S. Bhattacharya, D.P. Bhattacharya, A. Ghosh, J. Phys. Condens. Matter 20
an increase in manganese content (x). For glass samples with x = 0 (2008) 035203.
[6] R. Punia, R.S. Kundu, M. Dult, S. Murugavel, N. Kishore, J. Appl. Phys. 112 (2012)
and 5, the activation energy associated with electric modulus is almost 083701.
equal to that of the dc conductivity, whereas for glass samples with [7] R. Punia, R.S. Kundu, S. Murugavel, N. Kishore, J. Appl. Phys. 112 (2012) 113716.
x = 10, 15 and 20, the values of ER are equal to activation energy due [8] A. Ghosh, Phys. Rev. B 42 (1990) 5665.
[9] M.L. Dinesha, G.D. Prasanna, C.S. Naveen, H.S. Jayanna, Indian J. Phys. 87 (2013) 147.
to dc conduction in the high frequency region, indicating that the charge [10] A. Ghosh, J. Appl. Phys. 64 (1988) 2652.
carriers have to overcome the same energy barrier during conduction as [11] J. Livage, J.P. Jollivet, E. Tronc, J. Non-Cryst. Solids 121 (1990) 35.
well as relaxation processes. [12] Y. Sakuri, J. Yamaki, J. Electrochem. Soc. 132 (1985) 512.
[13] B. Peng, Z. Fan, X. Qui, L. Jaing, H.D. Ford, W. Huang, Adv. Mater. 17 (2005) 857.
[14] L.F. Maia, A.C.M. Rodrigues, Solid State Ionics 168 (2004) 87–92.
[15] R.K. Brow, C.A. Click, T.M. Alam, J. Non-Cryst. Solids 274 (2000) 9–16.
[16] V.C.V. Gowda, R.V. Anavekar, Solid State Ionics 176 (2005) 1393–1401.
[17] K. Trzebiatowski, J. Gackowska, W. Lizak, L. Muarwski, Opt. Appl. 35 (2005) 869.
[18] S.K. Despande, V.K. Shrikhande, M.S. Jogad, P.S. Goyal, G.P. Kothiyal, Bull. Mater. Sci.
30 (2007) 497.
[19] W.J. Gawande, S.S. Yawale, S.P. Yawale, Int. Res. J. Sci. Eng. 2 (4) (2014) 145–149.
[20] M.K. Murthy, K.S.N. Murthy, N. Veeraiah, Bull. Mater. Sci. 23 (4) (2000) 285–293.
[21] M. Pant, D.K. Kanchan, Nirali Gondaliya, Mater. Chem. Phys. 115 (2009) 98–104.
[22] M. Gabr, K.A. Ali, A.G. El-Din Mostafa, Turk. J. Phys. 31 (2007) 31–39.
[23] P.V. Rao, M.S. Reddy, V.R. Kumar, Y. Gandhi, N. Veeraiah, Turk. J. Phys. 32 (2008)
341–356.
[24] M. Dult, R.S. Kundu, S. Murugavel, R. Punia, N. Kishore, Phys. B 452 (2014) 102–107.
[25] A.K. Jonscher, Dielectric Relaxation in Solids, Chelsea Dielectrics, London, 1983.
[26] J.C. Dyre, T.B. Schoder, Rev. Mod. Phys. 72 (2000) 873.
[27] L. Zhang, Appl. Phys. Lett. 87 (2005) 022907.
[28] E. Barsoukov, J. Ross Macdonald Book, Impedance Spectroscopy Theory, Experiment,
and Applications, 2nd Edition John Wiley & Sons, Inc., Hoboken, New Jersey, 2005.
[29] J.C. Maxwell, Electricity and Magnetism, 1Oxford University Press, Oxford, England,
1873. (sec 328).
[30] K.W. Wagner, Ann. Phys. (Leipzing) 40 (1913) 53.
[31] H. Jain, J.N. Mundy, J. Non-Cryst. Solids 91 (1987) 315.
[32] S.R. Elliot, Philos. Mag. 36 (1977) 1291.
[33] S.R. Elliot, Philos. Mag. B 37 (1978) 553.
[34] S.R. Elliot, Adv. Phys. 36 (1987) 135.
[35] N.F. Mott, E.A. Davis, Electronic Processes in Non-Crystalline Materials, Oxford Uni-
Fig. 15. Variation of log (τM″) versus 1/T for 40SiO2–(60 − x)Bi2O3–xMnO2 glass system. versity Press, Oxford, 1979.
Here symbols represent experimental data and solid lines represent the linear fit. [36] A.R. Long, Adv. Phys. 31 (1982) 553.
8 M. Dult et al. / Journal of Non-Crystalline Solids 423–424 (2015) 1–8

[37] A. Ghosh, J. Appl. Phys. 66 (1989) 2425. [45] D.B. Dhwajam, M. Buchi Suresh, U.S. Hareesh, J.K. Thomas, S. Solomon, Annamma
[38] A. Ghosh, Phys. Rev. B 42 (1990) 1388. John, J. Mater. Sci. Mater. Electron. 23 (2012) 653.
[39] L. Murawski, Philos. Mag. B 50 (1984) L69. [46] R. Kohlrausch, Pogg. Ann. Phys. 91 (1854) 179.
[40] A. Ghosh, S. Bhattacharya, A. Ghosh, J. Phys. Condens. Matter 21 (2009) 145802. [47] G. Williams, D.C. Watt, Trans. Faraday Soc. 66 (1970) 80.
[41] K.V. Ramesh, D.L. Sastry, Phys. B 387 (2007) 45–51. [48] R. Bergman, J. Appl. Phys. 88 (2000) 1356.
[42] P.N. Rao, D.K. Rao, N. Veeraiah, Indian J. Eng. Mater. Sci. 13 (2006) 69–74. [49] N. Ponpandian, A. Narayanasamy, J. Appl. Phys. 92 (2002) 2770–2778.
[43] P.B. Macedo, C.T. Moynihan, N.L. Laberge, Phys. Chem. Glasses 14 (1973) 122.
[44] Rahul Vaish, K.B.R. Varma, J. Appl. Phys. 106 (2009) 064106.