Applied Physics A (2019) 125:73

[Link]

XRD analysis, Raman, AC conductivity and dielectric properties of Co

and Mn co-doped ­SnO2 nanoparticles
N. Bhakta1 · P. K. Chakrabarti1

Received: 24 September 2018 / Accepted: 24 December 2018 / Published online: 5 January 2019
© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract
The dielectric behavior, AC conductivity, and optical properties of Mn- and Co co-doping ­SnO2 nanoparticles were inves-
tigated in this paper. Co-doped S­ n0.98−xMnxCo0.02O2 (x = 0.01, 0.02, and 0.03) and undoped S
­ nO2 nanoparticles were syn-
thesized by sol–gel method. Pure crystallographic phases of undoped and co-doped samples were confirmed by Rietveld
analyses of X-ray diffraction (XRD) patterns. From XRD patterns, it was also confirmed that no impurity phase is present in
co-doped and undoped samples. The crystallite size of each sample was calculated by two ways, namely, Scherrer’s formula
and Williamson–Hall (W–H) method. The crystallographic strain of the co-doped samples was estimated from W–H method
and the values were compared with that of the undoped one. Raman spectra recorded at room temperature indicates that
oxygen vacancy in co-doped samples was enhanced compared to that of ­SnO2. In addition, dielectric properties of the co-
doped samples were improved compared to that of ­SnO2 and the AC conductivity of all the co-doped samples was lowered
due to replacement of Sn ion by Mn and Co ions which indicate that the resistivity of co-doped sample increases compared
to that of the pristine sample. However, the AC conductivity of undoped and all co-doped samples was increased with the
increase of frequency due to charge-hopping mechanism.

1 Introduction better electrical conductivity due to increase of charge car-

rier concentration [10, 11]. Material having low loss and
Research on dilute magnetic semiconductors (DMS) is an high dielectric constant are used in microwave integrated
emerging field, where both charge and spin degrees of free- circuits, microwave telecommunications, dynamic random
dom of an electron are utilized to design spintronic devices access memory, and multilayer capacitor [12]. Doping in
[1]. Transition-metal (TM)-doped S ­ nO2, ZnO, and T­ iO2 have ­SnO2 with different 3d/4f ions creates defects in the host
drawn great attention due to their potential application in lattice like oxygen vacancy (Vo), Sn interstitial (Sni), change
optical and carrier controlling [2]. Among the various DMS of bond length, etc., which are helpful for the enhance-
materials, tin oxide (­ SnO2) is an important candidate having ment of the properties of doped sample. These defects are
wide bandgap ~ 3.6 eV at 300 K and interesting magnetic, responsible for the inhomogeneous dielectric structure and
optical, and electrical properties [3]. ­SnO2 has tetragonal hopping mechanism of electrons [13]. In this regard, the
structure with space group P42/mnm and point group ­D4h14 choice of dopant and the preparation method also play an
in which one Sn ion is surrounded by six oxygen anions [4]. important role. Metal ions doped in S ­ nO2 nanoparticles can
Nano-size ­SnO2 has various applications in gas sensors [5], be prepared by different synthesis methods such as sol–gel
lithium-ion batteries [6], solar cell devices [7], optoelec- [14, 15], mechanical alloying [16], co-precipitation [11, 17],
tronic devices [8], dye-base solar cells [9], etc. Nanoparticles and chemical vapour deposition [18]. To achieve the suit-
of ­SnO2 with suitable choice of doping ions from transi- able textural and surface properties of metal oxides, sol–gel
tion-metal group makes them potential candidates for tech- method is the most effective method [19] for this, here in the
nological applications due to enhance dielectric response, present paper, Co–Mn co-doped S ­ nO2 nanoparticles were
prepared by sol–gel method. In addition, in case of nanopar-
* P. K. Chakrabarti ticles, surface to volume ratio increases and grain boundaries
pabitra_c@[Link] developed during the growth of nanoparticles introduce the
polarization in the sample in the presence of external applied
1
Solid State Research Laboratory, Department of Physics, field [20]. There are several works which were reported on
Burdwan University, Burdwan, WB 713104, India

13
Vol.:(0123456789)
73 Page 2 of 11 N. Bhakta, P. K. Chakrabarti

the dielectric, optical, and electrical properties of ­SnO2. Die- 3 Results and discussion
lectric response and AC electrical conductivity of Zn-doped
­SnO2 nanoparticles were studied and reported by Sahay et al. 3.1 XRD and Rietveld analysis
[11] and Mazumder et al., reported Mg doped ­SnO2 [21]. In
addition, Mehraj reported Fe–Co co-doped S ­ nO2 [15] and The observed peaks in XRD patterns of all the samples
Sb-doped ­SnO2 was reported by Skoromets et al. [14]. In confirm the formation of rutile-type (tetragonal) structure
addition, dielectric response of Zn–Co co-doped ­SnO2 was having space group P42/mnm of S ­ nO2 (JCPDS file No.
reported by Khan et al. [22]. Mn–Co co-doped S ­ nO2 showed 41-1445). All the positions of peaks in the XRD patterns
ferromagnetic response at room temperature (RT), and this of Co–Mn co-doped S ­ nO2 nanoparticles are matches well
work was reported by Inpasalini et al. [23] and Ahmad et al. with those of undoped S ­ nO2 which indicate the successful
reported the result of optical and dc resistivity of Mn–Co co- substitution of Co and Mn ions in the rutile lattice. The lat-
doped ­SnO2 [24]. However, the dielectric response and AC tice planes corresponding to different diffraction peaks are
electrical conductivity of Co–Mn co-doped ­SnO2 were not assigned in Fig. 1. The average crystallite size was estimated
yet investigated. Here, in this paper, good dielectric response using Scherrer’s equation [26]:
and AC electrical conductivity are found and this fact would
promote the quality of applications of Co–Mn co-doped k𝜆
⟨D⟩(110) = , (1)
­SnO2. 𝛽 cos 𝜃
where <D>(110) is the crystallite size corresponding to high-
est intense (110) peak, k is a constant (= 0.89), 𝜆 represents
2 Experimental the wavelength of the incident X-ray (0.154184 nm), 𝜃 is
Bragg diffraction angle, and full-width at half-maximum
Samples of Mn–Co co-doped in ­SnO2, ­Sn0.98−xMnxCo0.02O2 (FWHM) of (110) peak denoted by 𝛽 . The obtained aver-
(x = 0.01, 0.02, and 0.03, denoted by S1, S2, and S3, respec- age crystallite sizes of S0, S1, S2, and S3 are 27 (± 1.01),
tively) and undoped ­SnO2 (S0) nanoparticles were synthe- 20 (± 1.12), 15 (± 1.52), and 11 (± 1.06), respectively. It is
sized by sol–gel method [15, 25]. To prepare the Mn–Co co- observed that the diffraction peaks of co-doped ­SnO2 sam-
doped samples, required amount of stannous (II) chloride ples are shifted toward a lower diffraction angle with respect
­(SnCl4, ­2H2O), cobalt chloride hexahydrate ­(CoCl2, ­6H2O), to undoped S ­ nO2 suggest the incorporation of C ­ o2+ and
manganese chloride tetrahydrate ­(MnCl2, ­4H2O), and citric 3+ 4+
­Mn ions in ­Sn sites of ­SnO2 host lattice [27]. This shift
acid monohydrate (­ C6H8O7, ­H2O) (99.9%) were taken in a is quite likely due to mismatch of ionic diameters between
beaker and the aqueous solution was prepared. Citric acid ­Co2+ and M ­ n2+ with that of S­ n4+. In addition, these shifting
and ethylene glycol were added dropwise into the mixed of each peak indicates the change of lattice parameter and
solution of precursor material and ethanol. Few drops of the presence of oxygen vacancy in different samples viz., S1,
hydrochloric acid (HCl) were added in the mixed solution S2, and S3 [15]. In addition, the crystallite sizes of all the
to dissolve ­SnCl4. The solution was then stirred for 12 h at samples are calculated by Williamson–Hall (W–H) plot as
80 °C and it was continued until the preparation of the gel. well. We used two methods (Scherrer’s and W–H plot) for
The prepared gel was then annealed at 200 °C for 6 h in the the calculation of crystallite size to find the effect of lattice
air to remove the unwanted organic compound which served strain on the crystallite size, since Scherrer’s equation does
as binder of the precursor salt during the method of prepa- not consider lattice strain, while W–H plot duly considers
ration. The as-dried sample was finally annealed at 550 °C the lattice strain. The required equation for W–H plot is [28]
for 3 h in a furnace with heating rate 7 °C per min. The
crystallographic phase of all the samples were obtained by 𝛽 cos 𝜃 =
K𝜆
+ 2𝜀 sin 𝜃, (2)
analysing the X-ray diffractograms (XRD) recorded at room D
temperature in Brukers Advanced D8 diffractometer with Cu where 𝜀 denote internal lattice strain and the other param-
­Kα radiation (λ = 0.154184 nm) in the range of 2θ from 20° eters are already defined in Eq. 1. 𝛽 cos 𝜃 vs. 2 sin 𝜃 plot of
to 80°. High-resolution transmission electron microscopy S0, S1, S2, and S3 are shown in Fig. 2. The lattice strain
(HRTEM) of some selected samples (S1 and S2) was car- and average nanocrystallite sizes of all samples are obtained
ried out by JEOL JEM-2100F. Using the excitation source from the slope and ordinate intercept of 𝛽 cos 𝜃 vs. 2 sin 𝜃
of wavelength 532 nm, Raman spectra of the samples were plot, respectively. The obtained value of lattice strain from
recorded at RT in the range of wave number starting from W–H plot of S0, S1, S2, and S3 is 4.61 × 10−3, 5.02 × 10−3,
200 to 1000 cm−1 using T64000 Raman spectrophotometer 5.53 × 10−3, and 6.23 × 10−3, respectively. In addition, the
(J.Y. HORIBA). The frequency variation of dielectric con- calculated nanocrystallite size of S0, S1, S2, and S3 is 30,
stant and conductivity of different samples, S0, S1, S2, and 22, 17, and 14 nm, respectively. Thus, due to increase of
S3, were recorded in HIOKI 3532-50 LCR HiTESTER.

13
XRD analysis, Raman, AC conductivity and dielectric properties of Co and Mn co-doped S­ nO2… Page 3 of 11 73

Fig. 1  Fitted pattern of XRD

along with observed patterns
of a S0, b S1, c S2 and d S3
nanoparticles obtained from the
Rietveld analysis

in each sample is shown in Fig. 1. The lattice parameters (a

and c), volume of unit cell, particle size, etc. in each sample
are also calculated from the fitted pattern. The accuracy of
profile fitting is assessed by the low values of goodness of fit
(𝜒GOF) together( with) the reliability parameters, i.e., weighted
( )
profile factor Rwp , expected
( ) weighted profile factor Rexp ,
and the Bragg factor Rb , which are as follows [26]:

Rwp
𝜒GOF = , (3)
Rexp

�∑ �1
Wi (Jio − Jic )2 2

Rwp = ∑ , (4)
Wi Jio 2
Fig. 2  Williamson–Hall plot of S0, S1, S2, and S3 nanoparticles
∑� �
�Jmo − Jmc � ,
doping percentage, the strain of different co-doped samples RB = ∑ (5)
Jko
increases, but the size decreases. The obtained value of lat-
tice strain also indicates that the bond length and bond angle
� �
of ­SnO2 lattice decrease due to replacement of S ­ n4+ ions
2+ 3+ S−U
by ­Co and ­Mn ions [29]. Instead of 2 + and 4 + val- Rexp = ∑ , (6)
Wi Jio 2
ance state, valance of Mn ion is most probably 3 + [30]. In
case of 2% Fe and 2% Co-co doped S ­ nO2 samples [15], the ( )
reported value of strain is lower (~ 1.1665 × 10−3) than the where Wi = 1
Jio
is the weight factor, Jio and Jic are the
present case of S2 (~ 5.53 × 10−3), since the ionic radius of observed and calculated intensities at the ith step, Jm repre-
­Fe3+ (~ 0.60 Å) is smaller than M ­ n3+ (~ 0.65 Å). To deter- sent the intensities assigned to the mth Bragg reflection at
mine the crystallographic phases of all the samples, Rietveld the end of the refinement cycles, and (S–U) is the number of
analysis is carried out to fit the experimental X-ray data and degrees of freedom. The result of Rietveld refinement of all
the corresponding fitted pattern along with the observed one the samples is summarized in Table 1. The value of 𝜒GOF for

13
73 Page 4 of 11 N. Bhakta, P. K. Chakrabarti

Table 1  Rietveld refinement result of undoped ­SnO2 and ­Sn1−xMnxCo0.02O2 (x = 0.01, 0.02, and 0.03)
Composition GOF Rwp (%) Rb (%) Rexp (%) Lattice parameter (Å) Volume of unit Particle size (nm)
cell (Å)3
a c Scherrer’s Rietveld
method refinement

SnO2 1.054 7.522 6.48 3.11 4.7406 3.1875 71.634 30 31.4

Sn0.97Mn0.01Co0.02O2 1.287 9.670 7.07 9.10 4.7384 3.1861 71.536 22 21.1
Sn0.96Mn0.02Co0.02O2 1.105 9.697 7.7 8.61 4.7381 3.1857 71.518 17 18.2
Sn0.95Mn0.03Co0.02O2 1.096 9.229 7.5 8.47 4.7378 3.1852 71.497 14 16.1

3.2 HRTEM analysis

Micrographs, selected area electron diffraction (SAED), and

nanocrystalline fringe patterns of S1, S2, and S3 are cap-
tured during HRTEM observations to study the nanostruc-
ture and the morphology of the samples. Details of prepara-
tion of TEM grid is discussed in our earlier paper [27]. Some
representative pictures of micrographs of S2 and S3 are dis-
played in Fig. 4a and b, respectively, and it is found from
the pictures that most of the particles are agglomerating in
nature. The histograms shown in Fig. 4c and d represent the
particle size distribution of S2 and S3 samples, respectively.
The average particle size of S2 and S3 samples is determined
by fitting the histogram of sizes using log-normal function
and the obtained values of particle size are, respectively,
16.77 (± 0.159) nm for S2 and 12.61 (± 0.399) nm for S3.
Fig. 3  Theoretically generated unit cell structure of S3 nanoparticles The separation between consecutive fringes of sample S1
denoted by two parallel line is shown in Fig. 4e and the
obtained value of spacing is 3.38 Å which represents (110)
S0, S1, S2, and S3 is 1.37, 1.33, 1.38, and 1.29, respectively. lattice planes of ­SnO2. SAED patterns of sample S1 show
The obtained value of 𝜒GOF suggests that the fitting is quite well-defined lattice planes in Fig. 4f and the assigned lattice
good. Extracted values of oxygen vacancy (Vo) from the planes of S­ nO2 are (110), (101), (200), (211), and (301). The
Rietveld analysis of different samples S0, S1, S2, and S3 are fringe and SAED patterns also revealed that no impurity is
0.0001, 0.0126, 0.0138, and 0.0142, respectively. Thus, present in the prepared co-doped samples and the doping
Co–Mn co-doping in ­SnO2 enhances the oxygen vacancy or ions (Co, Mn) successfully replaced the fraction of ­Sn4+ in
defect in S
­ nO2 lattice compared to undoped S ­ nO2. The crys- ­SnO2 host lattice. In addition, X-ray fluorescence spectra
tallographic structure of S3 is also determined from the of S3 sample is shown in Fig. 5, and from the quantitative
theoretically fitted pattern, as shown in Fig. 3 which gives analysis, it is seen that the doping percentage of Co and Mn
the different bond lengths like Sn–O (~ 2.051 Å), O–O ions is ~ 1.9 and ~ 2.78, respectively. Thus, the calculated
(~ 2.90 Å) and bond angle Sn–O–Sn (~ 100.67°). The struc- doping percentage is in good agreement with the chosen
ture shown in Fig. 3 is drawn using the VESTA program stoichiometry of the samples. In addition, in XRF spectra,
[31]. The obtained value of Sn–O bond length of S0, S1, and no impurity is found which confirm the pure phase formation
S2 samples is 2.112, 2.081, and 2.072 Å, respectively. In of co-doped sample.
addition, the obtained value of Sn–O–Sn bond angle of S0,
S1, and S2 samples is 101.51, 101.02, and 100.91, respec- 3.3 Raman spectroscopic behavior
tively. The reported values of bond length (Sn–O) and bond
angle (Sn–O–Sn) of undoped S ­ nO2 and Ti-doped S ­ nO2 are Raman spectra of all the prepared samples are recorded at
0.214 Å, 101.463°, and 2.048 Å, 100.374°, respectively [29]. RT to confirm the formation of desired crystallographic
The doping ions, i.e., M­ n3+ (~ 0.65 Å) and C
­ o2+ (~ 0.58 Å), phase, presence of impurity, etc. of all samples like S0, S1,
4+
have smaller ionic radius than S ­ n (~ 0.69 Å), and for this S2, and S3. The phase purity of samples can be verified by
reason, both Sn–O bond length and Sn–O–Sn bond angle of the standard band positions of samples which are normally
co-doped samples decrease as compared to these of S0. found in such system. However, the presence of impurity

13
XRD analysis, Raman, AC conductivity and dielectric properties of Co and Mn co-doped S­ nO2… Page 5 of 11 73

Fig. 4  a and b Micrographs of S2 and S3 nanoparticles, respectively; c and d histogram of particle size distribution with log-normal fitted of S2
and S3 samples, respectively; e fringe pattern of S1 nanoparticles and f SEAD pattern of S1 nanoparticles

Fig. 5  XRF spectra of S3 nanoparticles

13
73 Page 6 of 11 N. Bhakta, P. K. Chakrabarti

normally deviates the expected spectra. The Raman spectra respectively [4]. The other peaks at 582 and 690 cm−1 are
of S0, S1, S2, and S3 samples recorded at RT in the wave- attributed to B1u and A2u (LO) modes, respectively [15].
number range of 200–850 cm−1 are shown in Fig. 6. Accord- Here, LO is the longitudinal optical phonon mode. IR active
ing to group theory, there are 18 vibrational modes of rutile Eu (TO) and A2u (LO) modes are also observed by Luo et al.
­SnO2 in the first Brillouin zone which is given by [32] [35]. Eg mode is generated from the vibration of two oxygen
atoms opposite to each other (but parallel to c-axis) [4]. The
Γ = A1g + A2g + B1g + B2g + Eg + 2A2u + 2B1u + 4Eu . (7)
Eg mode also represents Vo, since it is highly sensitive to
Among these 18 modes, three non-degenerate A1g, B1g, oxygen vacancies [36]. The increase in sharpness of peak
B2g modes and a degenerate Eg mode, i.e., four are Raman corresponding to Eg mode with the increase of Mn doping
active modes. A2g and B1u modes are neither infrared active suggests that defects due to oxygen vacancy (Vo) increase in
(IR) nor Raman active, i.e., these are silent modes. Among the co-doped sample compared to undoped S ­ nO2. Modes of
the two IR active modes, one is triply degenerate Eu mode B1u and A2u (LO) are assigned as surface phonon mode [34].
and the other is single A2u mode. The peaks for S0 are The position of A1g mode of S1, S2, and S3 samples are at
observed at 248, 320, 496, 582, 630, 690, and 769 cm−1. 493.8, 492, and 490 cm−1, respectively. Thus, it is observed
In addition, the observed Raman peaks of S1, S2, and S3 that with increase of Mn-doping concentration, the mode
samples with corresponding vibrational modes are shown corresponds to A1g mode is shifted to lower wavenumber
in Table 2. The two small peaks at 248 and 320 cm−1 are region which suggests the dispersion of optical phonon [15].
assigned by Eu (TO) mode, and TO is the transverse opti- A1g mode is originated due to the expanding vibration of
cal phonon modes [33]. This Eu (TO) mode is attributed Sn–O bond and the B2g mode is generated from the con-
to the formation of ­SnO2 particle in nano range [34]. The tracting vibration of Sn–O bonds in a plane perpendicular
three fundamental peaks of S0 at 496, 630, and 769 cm−1 to c-axis [37].
are also assinged by the Eg, A1g and B2g vibrational modes,

Fig. 6  Room temperature

Raman spectra of a S0, b S1, c
S2 and d S3 nanoparticles

Table 2  Raman spectra of Composition Mode of vibration ­(cm−1)

­Sn1−xMnxCo0.02O2 (x = 0.01,
0.02, 0.03) and S
­ nO2 Eu (TO) Eg B1u A1g A2u (LO) B2g
nanoparticles
SnO2 248, 320 496 582 630 690 769
Sn0.97Mn0.01Co0.02O2 219, 267 493.8 576 626 692 761
Sn0.96Mn0.02Co0.02O2 265 492 – 610 684 754
Sn0.95Mn0.03Co0.02O2 253, 310 490 – 608 677 748

13
XRD analysis, Raman, AC conductivity and dielectric properties of Co and Mn co-doped S­ nO2… Page 7 of 11 73

3.4 Dielectric properties the field reversal in the high-frequency region but ionic, dipo-
lar, and electronic polarization contribute in the dielectric
The variation of real part of dielectric constant (ε′) with fre- response. For this reason, ε′ and dielectric loss factor (shown
quency of all the prepared samples S0, S1, S2, and S3 are in Fig. 9) decrease with the measure of frequency and become
recorded at different temperature (i.e., at 300, 325, 350, 375, nearly constant above a certain value of frequency. In addi-
400, 425, and 450 K), as shown in Fig. 7a–d, respectively. tion, the hopping between cations (­ Sn4+, ­Mn3+, ­Co2+) cannot
The measured values of dielectric constant show high value follow the field reversal and the dielectric constant becomes
at low-frequency region and become nearly constant in high- independent of frequency after a certain value of frequency
frequency region. Mainly four types of polarization, i.e., ionic, [40]. This behavior of ε′ can be explained by Maxwell–Wag-
electronic, dipolar, and space charge polarizations, contribute ner model [38] and Koop’s theory [39] of dielectric behavior.
in dielectric properties. In low-frequency region, all the types Dielectric response of the sample depends on various factors
of polarization can follow the time reversal of field and high such as temperature, applied electric field, the sample prepara-
value of dielectric constant is observed. According to Max- tion method, grain structure, etc. For dielectric measurements
well–Wagner model [38], the highly resistive grain bounda- in parallel plate capacitor geometry, powder samples were pel-
ries separate the one conducting grain from other conducting letized and coated with carbon paste on both the upper and
grains and moving charge carriers are accumulated at grain lower surfaces of the pellet. Frequency-dependent real part of
boundaries in the presence of applied field. This accumulation the dielectric constant can be calculated using the relation [41]:
of charge carriers creates polarization. In addition, according
Cd
to Koops theory [39], in low-frequency region, high value of 𝜀� = , (8)
𝜀0 A
dielectric constant is observed due to inhomogeneous medium
(like the presence of oxygen vacancy) of dielectric structure where 𝜀0 is the free space permittivity (8.85 × 10− 12 F m−1),
and thin width of grain boundaries having small conductiv- capacitance of the sample is denoted by C, d is the thickness,
ity which has the contribution of space charge relaxation at and A is the area of the pellet. Due to doping of Co and Mn
the interface [10]. Space charge polarization cannot follow ions in ­SnO2 system, free charge carrier is generated in ­SnO2

Fig. 7  Frequency dependence of real part of dielectric constant of a S0 b S1 c S2 and d S3 nanoparticles

13
73 Page 8 of 11 N. Bhakta, P. K. Chakrabarti

to maintain charge neutrality. These free charge carriers are where the permittivity in high and low frequency limits is
trapped at defect zones, and as a result, nano-dipoles, i.e., denoted by 𝜀(0) and 𝜀(∞), respectively. 𝜀(0) − 𝜀(∞) is known
nanoparticles with dipole moment, are created. In the pres- as the dielectric relaxation strength. 𝜔 is the angular fre-
ence of externally applied electric field, these nano-dipoles quency, 𝜏cc represents relaxation time, and 𝛼 represents the
create polarization [42, 43]. At low frequency, the high value shape parameter. To avoid the electrode polarization effect
of ε′ is observed, since the oscillating nano-dipoles can eas- for the determination of the values of Δ𝜀 and 𝜏cc using Eq. 9,
ily orient themselves along the direction of the applied elec- we have considered 𝜀′ data in high frequency. Experimental
tric field. Thus, in low-frequency region due to space charge data recorded at 300 K along with Cole–Cole-fitted curve of
polarization, high value of ε′ is observed [44]. It is seen from S0, S1, S2, and S3 samples are shown in Fig. 8. It is found
Fig. 7b–d that with the increase of Mn-doping percentage, that doped samples were fitted well by Cole–Cole model
the value of ε′ increases in the co-doped samples. The varia- compared with undoped S ­ nO2. The values of Δ𝜀 and 𝜏cc
tion of dielectric constant (ε′) with angular frequency of S1, of S1, S2, and S3 samples extracted from the Cole–Cole
S2, and S3 samples is found to follow the Cole–Cole relaxa- fitting are ~ 98, ~ 7.1 × 104; ~ 91, ~ 4.72 × 104; and ~ 94,
tion in the high-frequency region. We fitted the experimental ~ 9.88 × 10−5, respectively. The variation of dielectric loss
ε′ vs. 𝜔 curve in the frequency range approximately 3.6 × 104 factor (tan 𝛿) of S0, S1, S2, and S3 samples as a function of
to 1.0 × 107 rad s−1 using the following relation [45]: frequency is shown in Fig. 9a–d. Dielectric loss factor of
[ [ ]] co-doped samples (i.e., S1, S2, and S3) sharply decreases
(𝜀(0) − 𝜀(∞)) from undoped ­SnO2. Imaginary part of dielectric constant
𝜀� (𝜔) = Re 𝜀(∞) +
[1 + (i𝜔𝜏cc )𝛼 ] (ε″) represents the amount of energy dissipation and tan 𝛿
[ ]
(𝜀(0) − 𝜀(∞)) 1 + (𝜔𝜏cc )1−𝛼 sin (𝛼𝜋/2) is proportional to ε″. From Fig. 9a–d, it is observed that
= 𝜀(∞) + , dielectric loss tangent of co-doped sample decreases from
1 + 2(𝜔𝜏cc )1−𝛼 sin (𝛼𝜋/2) + (𝜔𝜏cc )2(1−𝛼)
undoped ­SnO2; thus, co-doped sample will be better in case
(9)

Fig. 8  Frequency dependence of dielectric constant recorded at 300 K along with Cole–Cole fit of a S0, b S1, c S2 and d S3 nanoparticles

13
XRD analysis, Raman, AC conductivity and dielectric properties of Co and Mn co-doped S­ nO2… Page 9 of 11 73

Fig. 9  Frequency dependence of dielectric loss factor (tan 𝛿) of a S0, b S1, c S2 and d S3 nanoparticles

S1, S2, and S3 shows single semicircle which indicates that

material interface effect and domain wall do not have any
contribution to the electrical response of the prepared sam-
ples [46].

3.5 AC electrical conductivity

The variation of AC electrical conductivity (𝜎 � ) with

frequency is calculated by the following relation,
𝜎 � = 𝜀0 𝜔𝜀� tan 𝛿 . Here, 𝜀0 is the permittivity of free space
(~ 8.854 × 10−12 F m−1). Frequency-dependent AC con-
ductivity (𝜎 � ) of different samples of S0, S1, S2, and S3
recorded at different temperatures are shown in Fig. 11a–d.
It is observed from Fig. 11a–d that the value of 𝜎 ′ of all the
samples increases with the increases of frequency. In addi-
Fig. 10  M″ vs. M′ curve of S1, S2, and S3 nanoparticles at 300 K; tion, it is observed that 𝜎 ′ increases weakly in low-frequency
Inset shows the M″ vs. M′ curve of S0 at 300 K region and increases rapidly in high-frequency region. The
value of 𝜎 ′ changes with frequency by following the law
[47], i.e., 𝜎(𝜔) = A𝜔s ; 𝜔 is the angular frequency of the AC
of dielectric application. The ratio of imaginary part to real signal, A is complex proportionality constant, and S repre-
part of electric modulus can also be represented by dielectric sents temperature dependent parameter. The value of S is
loss tangent, i.e., tan 𝛿 = M �� /M �. Imaginary part of electric less than unity. There are different conduction mechanisms
modulus (M″) vs. real part of electric modulus (M′) plot for the explanation of the behavior of 𝜎 ′, i.e., hopping model,
(shown in Fig. 10) in the measured frequency region of S0, tunnelling, and free band model [11]. According to hopping

13
73 Page 10 of 11 N. Bhakta, P. K. Chakrabarti

Fig. 11  Frequency dependence of AC conductivity at different temperatures of a S0, b S1, c S2 and d S3 nanoparticles

model, in localized states, the conduction arises due to hop- study confirm that all the samples possess tetragonal struc-
ping of carriers between occupied and unoccupied localized ture with space group P42/mnm. The absence of impurity
states [48]. In the nanoparticle system, grain boundaries are phase is confirmed from the Rietveld analysis of XRD pat-
in disordered state and electrons cannot easily move from terns of all samples. Both the transition-metal ions are suc-
one grain to another grain because of these grain boundaries cessfully entered in the respective and desired sites of the
[47]. Due to the large grain boundaries and small grain size, host lattice. Due to the incorporation of Mn and Co ions in
Fermi level of nanoparticles is localized and the electronic place of Sn ion, the X-ray diffraction peaks are shifted in
state nearer to Fermi level is also localized. In addition, with lower diffraction angle compare to these of undoped S ­ nO2.
the increase of frequency of applied field charges which are From the structural analysis, it is concluded that C ­ o2+ and
trapped gets liberated from the deep traps. Due to the defects ­Mn3+ ions successfully occupied the site of ­Sn4+ in host
such as dangling bonds, micro porosities, oxygen vacancies, lattice. Raman analysis also confirms that oxygen vacancy
etc., trap states are created [11]. In general, with increase of of each co-doped sample is higher than undoped S ­ nO2 and
frequency of applied field, the electrons gain energy from oxygen vacancy increases with increase of Mn-doping per-
the applied field and conductivity increases [21]. The elec- centage. Dielectric measurements of all the co-doped sam-
tron hopping together with liberated charges is the origin ples indicate that dielectric response of the pristine sample
of conduction mechanism. Since the AC electrical con- is improved due to co-doping and this is due to space charge
ductivity decreases due to co-doping which confirms that polarization effect. AC conductivity depends on trapping
the co-doped sample can be used in a device for dielectric centre concentration and internal lattice strain across the
application. microstructure and increases with the increase of frequency.
The structural and electrical behavior is quite interesting as
observed from different investigations and this fact may be
4 Conclusions fruitful for applications of such potential materials.

Undoped ­SnO2 and Co–Mn co-doped ­SnO2 nanoparticles Acknowledgements Authors wish to acknowledge WBDST,
Govt. of W.B. for financial assistance (Memo: 292(SANC)/ST/P/
have been successfully prepared by sol–gel method. The S&T/16G-28/2017, Dated: 28.03.2018). Authors also wish to
Rietveld analyses of XRD patterns and Raman spectroscopic

13
XRD analysis, Raman, AC conductivity and dielectric properties of Co and Mn co-doped S­ nO2… Page 11 of 11 73

acknowledge UGC, Govt. of India for CAS-II (No: F.530/20/CAS- 24. M.S. Inpasalini, R. Choubey, S. Mukherjee, J. Electron. Mater.
II/2018 (SAP-I), Dated: 25.07.2018). 45, 3562–3569 (2016)
25. J. Wang, W. Zhou, P. Wu, J. Nanopart. Res. 16, 2573 (2014)
26. N. Bandyopadhyay, S. Bhakta, B.J. Sutradhar, A.K. Sarkar, S.
Deb, K. Kobayashi, Yoshimura, P.K. Chakrabarti, RSC Adv. 6,
References 101818–101826 (2016)
27. N. Bhakta, T. Inamori, R. Shirakami, Y. Tanioku, K. Yoshimura,
1. K.-C. Zhang, Y.-F. Li, Y. Liu, Y. Zhu, J. Appl. Phys. 112, 043705 P.K. Chakrabarti, Mater. Res. Bull. 104, 6–14 (2018)
(2012) 28. A. Mitra, A.S. Mahapatra, A. Mallick, A. Shaw, N. Bhakta, P.K.
2. J.M.D. Coey, M. Venkatesan, C.B. Fitzgerald, Nat. Mater. 4, Chakrabarti, Ceram. Int. 44, 4442–4449 (2018)
173–179 (2005) 29. J.K. Yang, H.L. Zhao, Y. Zhu, L.P. Zhao, J. Li, Adv. Mater. Res.
3. A. Parveen, S.A. Ahmad, S. Agrawal, A. Azam, Mater. Today 634–638, 2545–2549 (2013)
Proc. 4, 9429–9433 (2017) 30. S.A. Ahmed, S.H. Mohamed, J. Magn. Magn. Mater. 324, 812–
4. P. Chetri, B. Saikia, A. Choudhury, J. Appl. Phys. 113, 233514 817 (2012)
(2013) 31. K. Momma, F. Izumi, J. Appl. Crystallogr. 44, 1272–1276 (2011)
5. G. Ansari, D. Boroojerdian, S.R. Sainker, R.N. Karekar, R.C. 32. S.H. Sun, G.W. Meng, G.X. Zhang, T. Gao, B.Y. Geng, L.D.
Aiyer, S.K. Kulkarni, Thin Solid Films 295, 271–276 (1997) Zhang, J. Zuo, Chem. Phys. Lett. 376, 103–107 (2003)
6. S. Ding, J.S. Chen, X.W. Lou, Adv. Funct. Mater. 21, 4120–4125 33. K. Bouras, J.-L. Rehspringer, G. Schmerber, H. Rinnert, S. Colis,
(2011) G. Ferblantier, M. Balestrieri, D. Ihiawakrim, A. Dinia, A. Slaoui,
7. E. Ramasamy, J. Lee, J. Phys. Chem. C 114, 22032–22037 (2010) J. Mater. Chem. C 2, 8235–8243 (2014)
8. S. Pan, G. Li, Recent Pat. Nanotechnol. 5, 138–161 (2011) 34. G. Singh, N. Kohli, R.C. Singh, J. Mater. Sci. Mater. Electron. 28,
9. S. Ferrere, A. Zaban, B.A. Gsegg, J. Phys. Chem. B 101, 4490– 2257–2266 (2017)
4493 (1997) 35. S. Luo, P.K. Chu, Z. Di, M. Zhang, W. Liu, C. Lin, J. Fan, X. Wu,
10. Y. Zhang, A. Kolmakov, S. Chretien, H. Metiu, M. Moskovits, Appl. Phys. Lett. 88, 013109 (2006)
Nano Lett. 4, 403–407 (2004) 36. K. Vijayarangamuthu, S. Rath, J. Alloy. Compd. 610, 706–712
11. P.P. Sahay, R.K. Mishra, S.N. Pandey, S. Jha, M. Shamsuddin, (2014)
Curr. Appl. Phys. 13, 479–486 (2013) 37. K. Srinivas, M. Vithal, B. Sreedhar, M. Manivel Raja, P. Venugo-
12. T.M. Shaw, S.T. McKinstry, P.C. McIntyre, Annu. Rev. Mater. pal Reddy, J. Phys. Chem. C 113, 3543–3552 (2009)
Sci. 30, 263–298 (2000) 38. J. Maxwell, Electric and Magnetism, vol. 2 (Oxford University
13. L.A.K. Dominik, R.K. Maccrone, Phys. Rev. 156, 910–913 (1967) Press, New York, 1973)
14. V. Skoromets, H. Nemec, J. Kopecek, P. Kuzel, J. Phys. Chem. C 39. N.K. Divya, P.U. Aparna, P.P. Pradyumnan, Adv. Mater. Phys.
119, 19485–19495 (2015) Chem. 5, 287–294 (2015)
15. S. Mehraj, M.S. Ansari, Phys. E 65, 84–92 (2015) 40. C.G. Koops, Phys. Rev. 83, 121 (1951)
16. A.S. Albuquerque, J.D. Ardisson, W.A.A. Macedo, T.S. Plivelic, 41. K. Rajwali, F. Ming-Hu, Chin. Phys. B 24(12), 127803 (2015)
I.L. Torrianib, J. Larrea, E.B. Saitovitch, J. Magn. Magn. Mater. 42. S. Ramo, J. Winnery, T.V. Duzer, Fields and waves in communica-
272–276, 2211–2213 (2004) tion electronics (Wiley, New York, 1965)
17. K. Subramanyam, N. Sreelekha, D. Amaranatha Reddy, G. Murali, 43. M. Ashokkumar, S. Muthukumaran, J. Magn. Magn. Mater. 374,
R.P. Vijayalakshmi, Superlattices Microstruct. 82, 207–218 (2015) 61–66 (2015)
18. M.Y. Huh, S.H. Kim, J.P. Ahn, J.K. Park, B.K. Kim, Nanostruct. 44. A. Tabib, N. Sdiri, H. Elhouichet, M. Ferid, J. Alloys Compd. 622,
Mater. 11, 211–220 (1999) 687–694 (2015)
19. L.M. Fang, X.T. Zu, Z.J. Li, S. Zhu, C.M. Liu, W.L. Zhou, L.M. 45. A.S. Mahapatra, A. Mitra, A. Mallick, A. Shaw, J.M. Greneche,
Wang, J. Alloy. Compd. 454, 261–267 (2008) P.K. Chakrabarti, J. Alloy. Compd. 743, 274–282 (2018)
20. M. Jayalakshmi, M. Mohan Rao, K.-B. Kim, Int. J. Electrochem. 46. R. James, S. Priya, Uchino, Jpn. J. Appl. Phys. 41, 5272–5276
Sci. 1, 324–333 (2006) (2002)
21. N. Mazumder, A. Bharati, S. Saha, D. Sen, K.K. Chattopadhyay, 47. S. Mehraj, M.S. Ansari, J. Nanoeng. Nanomanufact. 3, 229–236
Curr. Appl. Phys. 12, 975–982 (2012) (2013)
22. K. Rajwali, F. Ming-Hu, Chin. Phys. B 24, 127803 (2015) 48. S. Sagadevan, J. Podder, Soft Nanosci. Lett. 5, 55–64 (2015)
23. N. Ahmad, S. Khan, J. Alloy. Compd. 720, 502–509 (2017)

13