Effect of boron doping on optical properties of sol-gel based nanostructured zinc oxide films on glass

Article abstract of DOI:10.1016/j.materresbull.2011.08.038

Boron doped zinc oxide thin films (～80 nm) were deposited onto pure silica glass by sol-gel dip coating technique from the precursor sol/solution of 4.0 wt.% equivalent oxide content. The boron concentration was varied from 0 to 2 at.% w.r.t. Zn using cry

Full text of DOI:10.1016/j.materresbull.2011.08.038

Materials Research Bulletin 46 (2011) 2392–2397
Contents lists available at SciVerse ScienceDirect
Materials Research Bulletin
journal homepage: www.elsevier.com/locate/matresbu
Effect of boron doping on optical properties of sol–gel based nanostructured zinc
oxide films on glass
b
a
b
a
Sunirmal Jana a, , Angela Surca Vuk , Aparajita Mallick , Boris Orel , Prasanta Kumar Biswas
*
a Sol–Gel Division, CSIR-Central Glass & Ceramic Research Institute, (Council of Scientific & Industrial Research, CSIR, India), 196 Raja S. C. Mullick Road, Kolkata 700032, India
b National Institute of Chemistry, Hajdrihova 19, SI-1000 Ljubljana, Slovenia
A R T I C L E I N F O
A B S T R A C T
Article history:
Boron doped zinc oxide thin films (ꢁ80 nm) were deposited onto pure silica glass by sol–gel dip coating
technique from the precursor sol/solution of 4.0 wt.% equivalent oxide content. The boron concentration
was varied from 0 to 2 at.% w.r.t. Zn using crystalline boric acid. The nanostructured feature of the films
was visualized by FESEM images and the largest cluster size of ZnO was found in 1 at.% boron doped film
(B1ZO). The presence of mixed crystal phases with hexagonal as major phase was identified from XRD
reflections of the films. Particle size, optical band gap, visible specular reflection, room temperature
photoluminescence (PL) emissions (3.24–2.28 eV), infra-red (IR) and Raman active longitudinal optical
(LO) phonon vibration were found to be dependent on dopant concentration. For the first time, we report
the room temperature fine structured PL emissions as phonon replicas originated from the LO phonon
(both IR and Raman active) in 1 at.% boron doped zinc oxide film.
Received 27 April 2011
Received in revised form 29 July 2011
Accepted 24 August 2011
Available online 31 August 2011
Keywords:
A. Thin films
A. Nanostructures
A. Semiconductors
D. Optical properties
D. Luminescence
ß 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Among the optical properties, the presence of longitudinal
optical (LO) phonon vibrations in pure and doped ZnO is one of the
Zinc oxide films have always attracted material researchers for
[1–7] their novel optical, electrical and electronic properties and
widespread applications. Optical properties of sol–gel based films
depend on several factors [3,8–10] such as preparative method,
precursor and substrate materials, annealing treatment, etc.
Alternatively, the properties can be tailored by chemical doping
in the precursor sol/solution especially with the elements of group
IIIA (e.g. B, Al, Ga) [11–14]. Among the group IIIA elements, boron
doping could be most effective [12] for tailoring the properties
because the element has lowest cationic radius, highest electro-
negativity and Lewis acid strength along with the highest Z/r2 (Z,
empirical ionic radius and r, charge number of the atomic core)
value. In this regard, it is sometimes observed that the starting
material [12,15] of dopant can also be a predominant factor. As for
example, no distinct deviation was observed in the optical band
gap [12] of zinc oxide thin films derived from the precursors
containing up to 1.4 at.% of boron if boron tri-i-propoxide be used
as a starting material. On the other hand, use of trimethyl borate,
even in a very low concentration of boron (0.2 at.%) showed the
prominent effect [15] in optical band gap property. The reason for
the effect is not yet clear.
important characteristics of exciton formation in nanoclusters. The
vibrations are generally observed as the phonon replicas [2,4] in
low temperature photoluminescence (PL) emissions. In general,
the Raman spectral study [1,5–7,16,17] are mostly used to
characterize the LO phonon vibrations of ZnO.
This report focuses on the effect of boron concentration upon
the optical properties of sol–gel based boron doped nanostruc-
tured zinc oxide thin films using crystalline boric acid as a starting
material of boron. In this work, for the first time, we are reporting
the room temperature fine structured PL emissions as phonon
replicas originated from both IR and Raman active LO phonon in
1 at.% boron doped zinc oxide film by Near Grazing Incidence Angle
(NGIA) IR, Raman and photoluminescence studies.
2. Experimental
2.1. Preparation of precursor sols/solutions and thin films
All the precursor sols/solutions (4.0 wt.% equivalent oxides)
were prepared by varying boron dopant concentrations from 0 to
2 at.% w.r.t. Zn. Zinc acetate dihydrate (GR, 99.5%; Sarabhai M.
Chemicals, India) and crystalline boric acid (99.8%, Fluka) were
used as the sources of ZnO and B, respectively while triethano-
lamine (TEA, 97%, E. Merck, India) was used as precursor sol/
solution stabilizer. Iso-propanol (99%, E. Merck, India) and doubly
distilled water mixture (ꢀ2.3 times higher than water by weight)
* Corresponding author. Tel.: +91 33 2473 3476x3405; fax: +91 33 2473 0957.
E-mail address: sjana@cgcri.res.in (S. Jana).
0025-5408/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.materresbull.2011.08.038

S. Jana et al. / Materials Research Bulletin 46 (2011) 2392–2397
2393
were used as the dispersion medium of precursor sol/solution.
Initially, requisite amount of zinc acetate and boric acid was
dissolved in water at ꢁ60 8C. Then iso-propanol and TEA mixture
was added slowly to the aliquot in cold condition. The dipping
technique with a lifting speed, 4 cm/min was used to coat (Dip
Master 200, Chemat Technology Inc., USA) on the pre-cleaned pure
silica glass (Suprasil grade, Heraus, Germany, dimensions:
20 mm ꢂ 15 mm ꢂ 0.25 mm) substrate. Adherence of freshly
prepared precursors on glass surface was poor while ꢁ72 h ageing
produced good adherence to the substrate. The as-coated samples
were baked in an air oven at 500 ꢃ 5 8C for 60 min. The films were
designated as B0ZO, B0.5ZO, B1ZO and B2ZO derived from the
precursor sols/solutions having the dopant concentrations, 0.0, 0.5,
1.0 and 2.0 at.%, respectively.
were measured using Thermo-Nicolet (model 5700, wavenumber
accuracy: 4 cmꢄ1), USA from 4000 to 400 cmꢄ1 wavenumber
region. Near Grazing Incidence Angle (NGIA) reflection-absorption
(RA) IR spectra were recorded using Bruker spectrometer (model
IFS 66/S). The NGIA IR spectra were obtained with the help of a
Monolayer Grazing Angle Specular Reflectance Accessory (Specac)
equipped with a standard polarizer for obtaining p-polarized
infrared light. For this measurement, the films were deposited on
fluorine doped tin oxide (FTO) glass (Pilkington) by sol–gel dip
coating process from the aged (72 h) precursors maintaining the
same curing conditions as done on silica glass substrate and the
background was corrected from the FTO glass substrate. Room
temperature substrate corrected Raman spectral study (micro-
Raman, Renishaw inVia Raman microscope) was done using argon-
ion laser with an incident wavelength of 514 nm as the excitation
source.
2.2. Characterization of films
Gaertner make autogain Ellipsometer (L116B, 708 angle of
incidence, He–Ne laser beam wavelength: 632.8 nm) was used to
measure thickness and refractive index (RI) of the films. Variation
of RI with dopant concentration was observed and the lowest RI
(1.69) was found in B1ZO film. FESEM (ZEISS, SUPRATM 35VP) was
used to characterize the surface morphology of films and cluster
size of ZnO was calculated from the FESEM images. Crystallinity of
3. Results and discussion
FESEM images (Fig. 1) reveal the nanostructured surface feature
of the films. A uniform and dense microstructure was observed in
B1ZO film. The average cluster size of ZnO calculated from the
FESEM images of 1 at.% doped (B1ZO) film was 41 ꢃ 4 nm whereas
in 0.5 at.% (B0.5ZO) and 2 at.% (B2ZO) boron doped films, the average
clustered sizes were 30 ꢃ 4 and 22 ꢃ 2 nm, respectively. However,
the average cluster size (21 ꢃ 2 nm) of undoped ZnO was almost
equivalent to that of 2 at.% boron doped film. Hence, 1 at.% doped film
(B1ZO) generated the largest size of ZnO cluster.
the films was checked by XRD (Philips PW-1730, Ni-filtered CuK
a
radiation). Shimadzu make UV-vis-NIR spectrophotometer (model
UV3101PC) and Perkin-Elmer (LS55) spectrofluorimeter were
employed to measure the optical (absorption and photolumines-
cence) properties of the films at room temperature. FTIR spectra
Fig. 1. FESEM images of pure and doped ZnO thin films on pure silica glass: (a) B0ZO, (b) B0.5ZO, (c) B1ZO and (d) B2ZO.

2394
S. Jana et al. / Materials Research Bulletin 46 (2011) 2392–2397
19 ꢃ 1 nm of B0ZO, B0.5ZO, B1ZO and B2ZO films, respectively.
Hence, the trend of change of particle size was same as calculated
from FESEM images.
ZnO is a direct transition semiconductor and its absorption co-
efficient (
a) and optical band gap energy (Eg) are interrelated
[13,19–22]. The band gap energy of the films was calculated from
2
Eq. (1). The (ahn) versus hn (energy) is plotted. The plots are
parabolic in nature and the number of inflexions reveals the
number of transitions. At inflexion, the tangential extrapolation to
X-axis (energy) gives the band gap energy (Eg).
1=2
a
hn
¼ Aðh
n
ꢄ EgÞ
(1)
where A is a constant.
Two Eg values were obtained (Fig. 3(i)–(iii)) as two inflexions
were obtained for each film; one Eg was at 3.31 eV (ꢁ375 nm) for
all films which would correspond to the bulk Eg [15] of zinc oxide
and another Eg were found at 4.13 eV (ꢁ300 nm), 3.74 eV
(ꢁ332 nm) (figure not shown here), 3.65 eV (ꢁ340 nm) and
3.80 eV (ꢁ326 nm) for B0ZO, B0.5ZO, B1ZO and B2ZO films,
respectively. This was due to [22] the size effect of nanostructured
ZnO semiconductor.
Single surface visible specular reflection (R) spectra (Fig. 3(iv))
of the films show that %R decreased with doping and it was lowest
in 1 at.% doped film (B1ZO). We have also found that B1ZO
possessed the lowest refractive index (1.69) which confirms with
its visible reflection spectrum. This indicated the presence of
relatively large amount of oxygen vacancy [22] in the film network.
However, the low RI is contrary to the uniform and dense
microstructure (Fig. 1(c)) of the film.
Fig. 2. XRD reflections of pure and 0, 1.0, 2.0 at.% B-doped ZnO thin films on pure
silica glass. Inset shows the XRD reflections of 0.5 at.% B-doped ZnO thin films on
pure silica glass.
The XRD reflections (Fig. 2) of the films confirmed the presence
of mixed crystal phases of ZnO with hexagonal as major phase [16]
along with a minor new phase, similar to that reported by Guo et al.
[18]. From the XRD reflections, it is clear that the crystallinity is
better in B1ZO film. The crystallite size was also calculated from
the h(1 0 0) plane reflection using Scherer’s equation [19] and the
average crystallite sizes were 15 ꢃ 2 nm, 22 ꢃ 2 nm, 25 ꢃ 2 nm and
The FTIR spectra of the films are shown in Fig. 4(a). In B1ZO film,
four distinct vibrations appeared at 407 cmꢄ1, 434 cmꢄ1, 472 cmꢄ1
BOZO(500)
a = 3.31 eV
a = 3.31 eV
b = 3.65 eV
B1ZO (500)
b = 4.13 eV
4
1.5
1.0
0.5
0.0
3
2
1
(ii)
a
b
(i)
a
b
0
2
3
4
5
6
(eV)
7
8
9
10
2
3
4
5
6
(eV)
7
8
9
10
hυ
hυ
a = 3.31 eV
b = 3.80 eV
B2ZO (500)
1.6
1.2
0.8
0.4
0.0
12
10
8
0.0 at% B
1.0 at% B
0.5 at% B
2.0 at% B
6
4
(iii)
(iv)
a
b
2
3
4
5
6
(eV)
7
8
9
10
400
500
600
700
800
hυ
Wavelength (nm)
Fig. 3. Plots of (
ah
n
)
2 versus h
n for direct band gap determinations of (i) B0ZO, (ii) B1ZO and (iii) B2ZO films on pure silica glass. Single surface visible specular reflection
spectra (iv) of pure and doped ZnO thin films on pure silica glass.

S. Jana et al. / Materials Research Bulletin 46 (2011) 2392–2397
2395
472
(a)
515
434
407
2 at% B
1 at% B
0 at% B
600 575 550 525 500 475 450 425 400
Wavenumber (cm-1)
1.0 at% B
0.5 at% B
Fig. 5. Raman spectra (substrate corrected) of pure and doped ZnO films.
566
2.0 at% B
ZnO films. The Raman spectra (corrected from substrate) of the
films are shown in Fig. 5. A strong intensity peak at ꢁ436 cmꢄ1
along with
a
medium intensity peak at ꢁ408 cmꢄ1 (most
0.0 at% B
prominent in case of B1ZO film) observed for all the films. The
appearance of strong peak at ꢁ436 cmꢄ1 confirmed the presence of
wurtzite ZnO crystal and assigned as E2 (high) mode [5–7] whereas
the peak at ꢁ408 cmꢄ1 corresponded to E1(TO) component [5] of
ZnO in the thin film form. A weak vibration appeared at ꢁ380 cmꢄ1
which is most prominent in B1ZO film indicating the presence of
A1(TO) component [6]. A prominent peak was found at 581 cmꢄ1 in
undoped ZnO film which shifted towards lower energy at 575, 570
and 566 cmꢄ1 in B2ZO, B0.5ZO and B1ZO films, respectively. This
peak corresponded to E1(LO) component of ZnO. It is interesting to
note that the trend of peak shifting is similar to that of NGIA IR peak
shifting (Fig. 4). The E1(LO) component is generally associated [1]
with lattice defects, such as oxygen vacancies in ZnO nanocrystals
and it generates photoluminescence (PL) emissions. The peak
shifting of E1(LO) towards lower energy in both IR and Raman
spectral studies was due to increase of ZnO grain size [1] as
evidenced from FESEM (Fig. 1(c)) and XRD (Fig. 2) studies. Two
additional peaks at ꢁ492 and 540 cmꢄ1 was also prominently
observed in B1ZO films. The former might be associated with the
‘‘defect band’’ [25] and the later corresponded to A1(LO) compo-
nent [6] of ZnO.
570
572
(b)
573
1200
1000
800
600
400
Wavenumber (cm-1)
Fig. 4. Substrate corrected FTIR (a) and Near Grazing Incidence Angle (NGIA) IR
spectra (b) of pure and doped ZnO thin films.
and 515 cmꢄ1. The peak appeared at 472 cmꢄ1 may be assigned as
the stretching vibration of Zn–O [23]. The presence of wurtzite
crystal structure (hexagonal) was also evident from the appear-
ance of FTIR peaks at 407, 434 and 515 cmꢄ1 (related to oxygen
vacancy) [5,23,24]. The IR active longitudinal optical (LO) phonon
vibration could be present in the film but it was not identified
distinctly in the FTIR spectra. Hence, the NGIA IR spectral
measurement were done to reveal the LO phonon vibration. A
strong vibration band at 573 cmꢄ1 (Fig. 4(b)) was observed in
undoped (B0ZO) film while in B0.5ZO, B1ZO and B2ZO films, it was
566, 570 and 572 cmꢄ1, respectively. The vibrations of hexagonal
ZnO as found in FTIR spectra (Fig. 4(a)) were possibly overlapped
within the broad intensity vibration of NGIA IR spectra. The
observed vibrational peak energy matched with the LO phonon
energy of nanostructured ZnO films and designated as E1(LO)
The phonon replicas (Fig. 5(A)) of the films have also been well
understood from the PL spectral study at room temperature. The PL
spectrum of B1ZO film showed fine structured emissions with
relatively high intensity (inset (b), Fig. 6(A)) in the visible region
and a broad tail extended towards the ultraviolet (UV) region when
the film was excited at 340 nm (inset (c), Fig. 6(A), photolumines-
cence excitation (PLE) spectrum of B1ZO film, fixing the emission
peak at 426 nm). The fine structured PL emissions appeared
prominently at 401, 406.5, 426, 436.5, 447.5, 459, 471 (shoulder),
485, 498 (shoulder), 512, 530 and 544 nm in B1ZO film. The peak
intensity of the other films was very low. We have plotted (inset
(b), Fig. 6(A)) the intensity of PL peak at 426 nm with dopant
concentration to compare the change of PL intensity among the
doped and undoped films. Although, part of the spectrum below
400 nm is not well resolved but its second derivative spectrum
(inset (a), Fig. 6(A)) shows three distinct emissions at 382.5, 391
and 397 nm. The origins of the fine structured PL visible emissions
are already reported by several authors [1–6,14,26–32] and these
have been assigned mostly due to various types of oxygen
component [4]. The LO phonons at the
wurtzite ZnO crystal belong to the following irreducible represen-
tation (Eq. (2)) [16],
G-point of hexagonal
G
¼ A1 þ 2B1 þ E1 þ 2E2
(2)
In Eq. (2), the A1 and E1 modes [5,16] are both IR and Raman active
which split into LO and transverse optical (TO) components but the
E2 mode is only Raman active. The B1 mode is forbidden. As Raman
spectral study is very much useful to know the phonon modes [5–
7,17] of ZnO, the study was performed on the undoped and doped

2396
S. Jana et al. / Materials Research Bulletin 46 (2011) 2392–2397
observed in PL study at very low temperature (10 K). This was for
the first time, we have observed the phonon replicas in PL
emissions at room temperature of sol–gel based boron doped
nanostructured ZnO thin film. The large amount of defect
concentration in the B1ZO film might be the reason for generating
fine PL emissions as [3] larger the grain size, smaller is the non-
radiative relaxation rates.
4. Conclusions
Boron doped nanostructured ZnO thin films of mixed crystal
phases with hexagonal as major phase were prepared by sol–gel
technique using crystalline boric acid as the source of boron. The
presence of both IR and Raman active longitudinal optical (LO)
phonon vibration was observed by NGIA IR and Raman spectral
studies. Boron doping controls the LO phonon energy, refractive
index (RI), band gap and grain size of ZnO in nanostructured ZnO
thin films. The variation of RI due to boron doping has been
observed in the specular reflectance. For the first time, we observed
the room temperature fine structured PL emissions in the UV–vis
regions (3.24–2.28 eV) as phonon replicas originated from the LO
phonon (both IR and Raman active) in 1 at.% boron doped film.
Acknowledgements
Authors are thankful to the director, CSIRꢄCGCRI, Kolkata, India
for his kind permission to publish this paper. They are also thankful
to Electron Microscopy, XRD Sections and Nanostructured
Materials Division (NSMD), CSIRꢄCGCRI, Kolkata for sample
characterizations. This work has been done under Bilateral Indo-
Slovenian Collaborative Project [Sanction No. DST/INT/SLOV/P-7/
05, dated 21st November, 2007] sponsored by Department of
Science & Technology (DST), New Delhi, Government of India.
References
[1] H.C. Hsu, C.S. Cheng, C.C. Chang, S. Yang, C.S. Chang, W.F. Hsieh, Nanotechnology
16 (2005) 297–301.
[2] S. Singh, D. Nakamura, K. Sakai, O. Tatsuo, M.S.R. Rao, New J. Phys. 12 (2010),
023007 (1–11).
[3] X. Teng, H. Fan, S. Pan, C. Ye, G. Li, Mater. Lett. 61 (2007) 201–204.
[4] A.N. Gruzintsev, E.E. Yakimov, Inorg. Mater. 41 (2005) 725–729.
[5] X. Sui, Y. Liu, C. Shao, Y. Lui, C. Xu, Chem. Phys. Lett. 424 (2006) 340–344.
[6] B. Zhang, X.T. Zhang, H.C. Gong, Z.S. Wu, S.M. Zhou, Z.L. Du, Phys. Lett. A 372 (2008)
2300–2303.
Fig. 6. (A) Characteristics of room temperature photoluminescence (PL) emissions
of pure and doped ZnO thin films on silica glass. Insets: (a) 2nd derivative PL
spectrum of B1ZO film, (b) change of PL intensity of 426 nm emission peak with
dopant concentration of the films and (c) photoluminescence excitation (PLE)
spectrum of B1ZO film fixing PL emission at 426 nm. (B) Excitation dependent PL
spectra of 1 at.% boron doped ZnO thin film (B1ZO) on pure silica glass show the
emissions of free and bound excitons.
[7] A.A. Khan, A.F. Vladimir, A.B. Alexander, Appl. Phys. Lett. 86 (2005), 053103 (1–3).
[8] A. Sakthivelu, V. Saravanan, M. Anusuya, J.J. Prince, J. Ovon. Res. 7 (2011) 1–7.
[9] M. Caglar, S. Ilican, Y. Caglar, F. Yakuphanoglu, J. Alloys Compd. 509 (2011)
3177–3182.
[10] T. Ivanova, A. Harizanova, T. Koutzarova, B. Vetruyen, Mater. Lett. 64 (2010) 1147–
1149.
[11] S.H. Jeong, J.H. Boo, Thin Solid Films 447–448 (2004) 105–111.
[12] R.B.H. Tahar, N.B.H. Tahar, J. Mater. Sci. 40 (2005) 5285–5289.
[13] G. Kim, J. Bang, Y. Kim, S.K. Rout, S.I. Woo, Appl. Phys. A 97 (2009) 821–828.
[14] M. Wang, K.E. Lee, S.H. Hahn, E.J. Kim, S. Kim, J.S. Chung, E.W. Shin, C. Park, Mater.
Lett. 61 (2007) 1118–1121.
[15] L.P. Purohit, V. Kumar, R. Kumar, J. Nat. Phys. Sci. 23 (2009) 7–12.
[16] J. Wang, S. Zhang, J. You, H. Yan, Z. Li, X. Jing, M. Zhang, Bull. Mater. Sci. 31 (2008)
597–601.
[17] P.Y. Emelie, J.D. Phillips, B. Buller, U.D. Venkateswaran, J. Electron. Mater. 35
(2006) 525–529.
[18] L. Guo, J.X. Cheng, X.Y. Li, J. Yan, S.H. Yang, C.L. Yang, J.N. Wang, W.K. Ge, Mater. Sci.
Eng. C 16 (2001) 123–127.
[19] S. Jana, I.C. Baek, S.l. Seok, J. Colloid Interface Sci. 322 (2008) 473–477.
[20] S. Kundu, P.K. Biswas, Chem. Phys. Lett. 414 (2005) 107–110.
[21] S. Jana, P.K. Biswas, Mater. Lett. 30 (1997) 53–58.
[22] S. Jana, P.K. Biswas, Mater. Chem. Phys. 117 (2009) 511–516.
[23] R.N. Gayan, K. Sarkar, S. Hussain, R. Bhar, A.K. Pal, Ind. J. Pure Appl. Phys. 49 (2011)
470–477.
vacancies, oxygen interstial, free holes, etc. The PL emissions below
the 391 nm are fully dependent on excitation energy [22] which is
evident from the excitation dependent PL spectra (Fig. 6(B)) of the
B1ZO film. Hence, on the basis of excitation dependent PL
emissions, we may divide the observed emission peaks into two
main groups that appeared in the 3.24–2.28 eV. The emission
peaks appeared up to 391 nm could be considered due to near band
edge (NBE) exciton transitions of wide band gap ZnO [26,27] while
the emission peaks appeared above 391 nm attributed to bound
exciton recombination [22] which trapped in the potential well of
the defect centres such as oxygen vacancy of nanostructured ZnO
film. The emission peaks appeared at 382.5 nm and 391 nm were
considered as the phonon replicas generated from both IR and
Raman active LO phonon of the film because the presence of LO
phonon vibration was confirmed by NGIA IR and Raman spectral
studies (Fig. 4(b)). On the other hand, the fine structure PL
emissions observed above 391 nm in the visible region could also
be considered as phonon replicas [2] as these were considered as
bound excitons [22]. The emission peak appeared at 397 nm may
be considered as zero-phonon transition for the visible region. It is
already reported [2,4] that the phonon replicas of ZnO was
[24] G. Xiong, U. Pal, J.G. Serrano, K.B. Ucer, R.T. Williams, Phys. Stat. Sol. C 3 (2006)
3577–3581.
[25] M.F. Paul, R.L. Richard Jr., Am. Miner. 71 (1986) 772–778.
[26] Y.U. DongQi, H.U. LiZhang, L. Liao, H.U. Hao, Z. HeQiu, L.S.-E. Andy, C. Xi, F.U. Qiang,
Q. ShuangShuang, Sci. Chin. Ser. G: Phys. Mech. Astron. 52 (2009) 207–211.

S. Jana et al. / Materials Research Bulletin 46 (2011) 2392–2397
2397
[27] N. Boukos, C. Chandrinou, K. Giannakopoulos, G. Pistolis, A. Travlos, Appl. Phys. A
88 (2007) 35–39.
[30] Y. Gu, I.L. Kuskovsky, M. Yin, S. O’Brien, G.F. Neumark, Appl. Phys. Lett. 85 (2004)
3833–3835.
[28] K. Vanheusden, W.L. Warren, C.H. Seager, D.R. Tallant, J.A. Voigt, B.E. Gnade, J.
Appl. Phys. 79 (1996) 7983–7990.
[31] L.W. Yin, M.S. Li, Y. Bando, D. Golberg, X.L. Yuan, T. Sekiguchi, Adv. Funct. Mater. 17
(2007) 270–276.
[29] D. Wang, C. Song, Z. Hu, Cryst. Res. Technol. 43 (2008) 55–60.
[32] D.H. Zhang, Z.Y. Xue, Q.P. Wang, J. Phys. D: Appl. Phys. 35 (2002) 2837–2840.