Photoluminescence and Raman Scattering in Ag-doped ZnO Nanoparticles

R. Sánchez Zeferino, M. Barboza Flores, and U. Pal

Citation: J. Appl. Phys. 109, 014308 (2011); doi: 10.1063/1.3530631

View online: http://dx.doi.org/10.1063/1.3530631
View Table of Contents: http://jap.aip.org/resource/1/JAPIAU/v109/i1
Published by the American Institute of Physics.

Additional information on J. Appl. Phys.

Journal Homepage: http://jap.aip.org/
Journal Information: http://jap.aip.org/about/about_the_journal
Top downloads: http://jap.aip.org/features/most_downloaded
Information for Authors: http://jap.aip.org/authors

Downloaded 16 May 2013 to 147.188.128.74. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions
 JOURNAL OF APPLIED PHYSICS 109, 014308 共2011兲

Photoluminescence and Raman Scattering in Ag-doped ZnO Nanoparticles

R. Sánchez Zeferino,1 M. Barboza Flores,2 and U. Pal1,a兲
1
Instituto de Física, Universidad Autónoma de Puebla, Apdo. Postal J-48, Puebla 72570, Mexico
2
Centro de Investigación en Física, Universidad de Sonora, Apdo. Postal 5-088,
Hermosillo, Sonora 83190, Mexico
共Received 23 September 2010; accepted 17 November 2010; published online 7 January 2011兲
Effects of Ag doping on the crystallinity and optical properties of zinc oxide 共ZnO兲 nanoparticles
have been studied by x-ray diffraction, diffuse reflectance spectroscopy, micro-Raman, and
photoluminescence spectroscopy. It has been observed that while Ag-doping at low concentration
improves the optoelectronic properties of ZnO nanostructures, Ag-doping at high concentrations
drastically modify the emission behavior and lattice vibrational characteristics of the nanostructures.
High Ag content in ZnO nanostructures causes lattice deformation, induces silent vibrational modes
in Raman spectra, and reduces excitonic UV emission due to concentration quenching. © 2011
American Institute of Physics. 关doi:10.1063/1.3530631兴

I. INTRODUCTION UV emission in ZnO is relatively well known, origin of the

visible emissions is still considered controversial. It has been
Nanostructured semiconductors are of great interest due reported that both the UV and visible emission in ZnO can
to their extraordinary physicochemical properties, which dif- be tailored through metal doping.12,13
fer from their bulk counterparts. Zinc oxide 共ZnO兲 is a II–VI In the present article, we report on the synthesis of un-
compound semiconductor, promising for applications as gas doped and Ag-doped ZnO nanoparticles of the order of 100
sensors, piezoelectric transducers, and solar cell windows.1,2 nm diameters through nonionic polymer assisted thermoly-
It has also been used for applications in optoelectronics, e.g., sis. Effects of Ag doping on the PL and Raman scattering
short wavelength light emitting diode, photodetector, ultra- behavior of the nanoparticles are studied in details.
violet 共UV兲 laser, etc., due to its wide band gap 共3.36 eV at
room temperature兲 and large excitonic binding energy 共60
meV兲.3 II. EXPERIMENTAL
Fabrication of efficient devices based on semiconductor The procedure for preparing undoped and Ag-doped
nanostructures requires in-depth understanding of their opto- ZnO nanoparticles was as follows. First, a 20 ml aqueous
electronic behaviors, which depend on their shape, size, and solution 共0.02 M兲 of zinc acetate dihydrate
impurity contents. Recently, several research groups have re- 关Zn共CH3COO兲2 · 2H2O, Baker, 99.9%兴 was prepared under
ported that the doping process can influence the lumines- mild magnetic stirring. After stirring this solution for about
cence properties of ZnO; frequently leading to blueshift or 20 min, 5.8 ml of polyethylene glycol-p-isocooctylphenyl
redshift in its near band edge 共NBE兲 emission and drastic ether, commonly known as Triton X100 关共C2H4O兲nC14H22O,
changes in their visible emission.4,5 On the other hand, lumi- Aldrich兴 was slowly added. For Ag doping, different
nescence properties of those nanostructures might also de- amounts 共nominally 0.5, 1.0, and 2.0 mol %兲 of silver nitrate
pend on their morphology, growth technique, synthesis con- 关AgNO3, Aldrich, 99.8%兴 were added to this mixture. The
ditions, and imperfections. For example, silver 共Ag兲 doping final solution was heated in a muffle furnace for 48 h at
has been seen to affect the photocatalytic activity of semi- 260 ° C 共with 2 ° C / min temperature increase兲. Finally, the
conductor nanostructures.6 Role of Ag as a prominent lumi- resulting powder was air annealed in a 100 cm long horizon-
nescent activator in compound semiconductor has been re- tal 共50 mm diameter兲 quartz tube furnace at 550 ° C for 2 h.
ported by Park et al.7 The morphology and size of the nanoparticles were veri-
Due to the vast optoelectronic applications of ZnO nano- fied by a scanning electronic microscope 共SEM, JEOL JSM-
structures, several theoretical and experimental studies have 5600LV兲. Crystallinity of the nanostructures was analyzed
been made on the optical properties of ZnO nanostructures using the Cu K␣ radiation 共␭ = 1.5406 Å兲 of a Rigaku® x-ray
with different morphologies, such as nanoparticles, nano- diffractometer 共XRD兲. Optical properties of the synthesized
wires, nanobelts,8 nanoprisms,9 and nanostructured thin nanoparticles were studied by diffuse reflectance spectros-
films.10 Very often, two emission bands appear in the photo- copy 共DRS, Varian Cary 100 UV-Vis spectrophotometer with
luminescence 共PL兲 spectra of ZnO. One UV band located at DRA-CA-30I diffuse reflectance accessory兲, micro-Raman
about 3.23 eV, commonly attributed to its excitonic transi- spectroscopy 共Horiba JOBIN-YVON spectrophotometer兲
tion, and a broad yellow-green emission in between 2.33 and and PL spectroscopy at different temperatures. For recording
2.48 eV, associated with defect states like oxygen vacancies the PL spectra of the samples, a 0.5 m long Science Tech
共Vo兲 and interstitial zinc atoms 共Zni兲.11 While the origin of monochromator, a Hamamatsu photomultiplier, and a Janis
cryogenic cooling system were used. The samples were ex-
a兲
Electronic mail: upal@sirio.ifuap.buap.mx. cited with the 325 nm emission of a He–Cd laser.

0021-8979/2011/109共1兲/014308/6/$30.00 109, 014308-1 © 2011 American Institute of Physics

Downloaded 16 May 2013 to 147.188.128.74. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions
 014308-2 Zeferino, Flores, and Pal J. Appl. Phys. 109, 014308 共2011兲

!
! #!
TABLE I. Position, FWHM of the main diffraction peaks, along with the
lattice parameter values calculated from the XRD patterns of the undoped
#! and Ag-doped ZnO nanoparticles.

Position of Position of Lattice

"! the 共101兲 the 共002兲 parameters
$! Nominal Ag diffraction peak diffraction peak 共Å兲
concentration and its FWHM and its FWHM
共at. %兲 共deg兲 共deg兲 a c

0 36.23 共0.22兲 34.41 共0.19兲 3.254 5.212

"! $! 0.5 36.25 共0.22兲 34.40 共0.20兲 3.251 5.212
1.0 36.30 共0.23兲 34.45 共0.20兲 3.247 5.206
2.0 36.28 共0.25兲 34.42 共0.21兲 3.248 5.210

Ag+ ions in the ZnO lattice sites, probably substituting Zn2+

ions, appearance of Ag peaks in the diffraction patterns indi-
cates clearly the formation of crystalline silver clusters in the
nanoparticles. It is worth mentioning that Ag+ ions in ZnO
FIG. 1. Typical SEM micrographs of the 共a兲 undoped, 共b兲 0.5%, 共c兲 1.0%,
and 共d兲 2.0% Ag-doped ZnO nanostructures. lattice behave similar to other monovalent dopant ions like
Na+ and K+, which have ability to occupy both the lattice
and interstitial sites.14 Limited incorporation of Ag+ ions into
III. RESULTS AND DISCUSSION
the ZnO lattice through substitution of Zn2+ ions was due to
Figure 1 shows the typical SEM micrographs of the ZnO their higher ionic radius 共1.22 Å兲 in comparison with the
nanoparticles prepared with different nominal concentrations radius of Zn2+ ions 共0.74 Å兲.
of silver nitrate. Formation of quasisphreical ZnO particles in Though the position of the NBE emission of ZnO could
the 100–200 nm size range is clear from the micrographs. No be blueshifted for its nanoparticles of diameters comparable
appreciable change either in morphology or in size of the or smaller than its excitonic diameter 共7.4 nm兲 due to quan-
ZnO particles could be observed on Ag-doping 关Figs. tum size effect,15 doping with appropriate element might
1共b兲–1共d兲兴. As can be seen from their XRD patterns 共Fig. 2兲, cause both blue and redshifts. As the sizes of our undoped
all the samples were well crystalline, and of hexagonal and Ag-doped ZnO particles were bigger than 100 nm, a
wurtzite phase 共JCPDS file no. 36-1451兲. However, the Ag- quantum size effect is not expected to be observed, and any
doped samples revealed some additional diffraction peaks change in the band gap energy 共Eg兲 could be attributed solely
共marked with “ ⴱ”兲 associated with the face-centered-cubic to Ag doping.
共fcc兲 phase of metallic Ag 共JCPDS file no. 04-0783兲. In gen- Diffuse reflectance spectra of all the samples 共Fig. 3兲
eral, the intensity of the diffraction peaks associated with revealed characteristic absorption edge near 380 nm. Un-
ZnO decreases and their full width at half maximum doped ZnO powders revealed high reflectance 共⬃50%兲 in the
共FWHM兲 increases on Ag incorporation 共see Table I兲, indi- visible region. Reflectance of the Ag-doped ZnO samples
cating a decrease in crystallinity of the nanostructures. While decreased with the increase in Ag concentration. Band gap
the decrease of crystallinity and a shift in position of the energy of the samples was estimated through the Kubelka–
main diffraction peaks 共Table I兲 indicate the incorporation of Munk treatment to their DRS spectra, following the method
reported earlier.16 Figure 4 presents the Kubelka–Munk plots
#!!!! for the undoped and Ag-doped samples, used to determine
! their band gap energies. It can be observed that while the
"!!! *
"&$&%

!!!
"&$$%

"&&#%
"$$#%

%!!! #!
"&&$%
"&$#%

"&$'%

"#$&%
0)12)3415 "6-7-%

"#$#%
"#$$%

$!!!
()*
%!!!
!
$!!!
()*+ ! "$-./%
* *
"

%!!!

$!!! &! /(0

()*+ ! "&-$/% * /(012) ,345".
* /(012) ,643".
%!!!
"&&&%

/(012) ,743".
"#$$%

"##$%

$!!!
()*+ ! "#-$/% * * * !
! !! "!! #!! $!! %!!
$! &! %! '! ! (!
#$%&'&()*+ ,(-.
896!! 6)!:2; # "<2!9223%
FIG. 3. 共Color online兲 Diffuse reflectance spectra of the undoped and Ag-
FIG. 2. XRD patterns of the undoped and Ag-doped ZnO nanoparticles. doped ZnO nanoparticles.

Downloaded 16 May 2013 to 147.188.128.74. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions
 014308-3 Zeferino, Flores, and Pal J. Appl. Phys. 109, 014308 共2011兲

% %
)!!!!

BC*
%&'()!

%&'()!D@$A
!" !"-.$ /0'123 %& '.)!
@$A

$%&'()!
$ $
#$%&'() +, #$%&'(1 +, !

/01203415 '6787!
%&'()!D@&'()!

*+,
# # '!!!
$E? D@$.
" "
%!!! "!
(

! !
56/7 38!"

!"-.$ /4'023 !"-.$ /('023

#!!! #!
$ $
#$%&'(& +, #$%&'(& +,
# # !!!
" " 6! >?,
! ! !
!! "!! #!! $!! %!! &!! '!! (!! )!!! ))!!
=&
'(% '(! '(' '(% '(! '(' 96:60 3;4<1 '": !
98:;:! +!+<$= /+,3
FIG. 6. Raman spectra of the ZnO nanoparticles containing: 共a兲 0%, 共b兲 0.
5%, 共c兲 1.0%, and 共d兲 2.0% 共nominal兲 of Ag.
FIG. 4. Kubelka–Munk plots and band gap energy estimation for the un-
doped and Ag-doped ZnO nanoparticles.

of the undoped ZnO sample in the 50– 800 cm−1 spectral

band gap energy value for the sample doped with 0.5% range. The spectrum consists of four peaks located at about
共nominal兲 Ag did not change with respect to the undoped 100, 380, 437, and 580 cm−1, which correspond to the E2L,
one, there occurred a small redshift 共0.02 eV兲 for the samples A1共TO兲, E2H, and A1共LO兲 fundamental phonon modes of
doped with 1.0% and 2.0% Ag. Wang et al.17 have observed hexagonal ZnO, respectively. The Raman peak located at
a similar effect in their Zn1−xCdxO nanorods and nanon- about 204 cm−1 was assigned to the 2E2L second-order pho-
eedles due to Cd substitution in ZnO lattice. Kukreja et al.18 non mode. Other four Raman peaks located at about 331,
also have reported a band gap energy value as low as 2.90 508, 664, and 822 cm−1 could be assigned to the 3E2H
eV for the alloy made of ZnO and CdO. − E2L, E1共TO兲 + E2L, 2共E2H − E2L兲, and A1共TO兲 + E2L mul-
In order to investigate the influence of Ag doping on the tiphonon scattering modes, respectively.
Raman scattering of ZnO nanoparticles, room temperature Raman spectra of the Ag-doped ZnO nanoparticles with
Raman spectra of all samples were measured. It is well different Ag contents are shown in Fig. 6. The A1共TO兲 and
known that wurtzite ZnO has eight sets of characteristic op- A1共LO兲 polar branches appeared at about 380 cm−1 and
tical phonon modes at the center of Brillouin zone 共⌫ point兲, 570 cm−1, respectively, for all the samples. On incorporating
⌫ = 1A1 + 2B1 + 1E1 + 2E2 , 共1兲 Ag in ZnO, the A1共LO兲 peak broadened and shifted about
9 cm−1 toward lower energy. Such a shift and broadening in
where A1 and E1 are the two polar branches, which split into the A1共LO兲 phonon mode can be attributed to the scattering
longitudinal optical 共LO兲 and transversal optical components contributions of the A1共LO兲 branch outside the Brillouin
with different frequencies due to macroscopic electric fields zone center.20 The A1共LO兲 phonon mode is commonly as-
associated with the LO phonons. The E2 modes are nonpolar
signed to the oxygen vacancies,21,22 zinc interstitials,23,24 or
with low frequency mode 共E2L兲 associated to the heavy Zn
defect complexes containing oxygen vacancy and zinc inter-
sublattice and high frequency mode 共E2H兲 involving only
stitial in ZnO.23 Additionally, there appeared a broad Raman
oxygen atoms.19 While the A1, E1, and E2 modes are active
peak at about 486 cm−1 exclusively for the Ag-doped
in Raman and infrared spectra, the B1 modes are generally
samples, which was assigned as the interfacial surface pho-
inactive in Raman spectra and are so-called silent modes.
non mode in the literature.25 This phonon mode cannot be a
Figure 5 presents the room temperature Raman spectrum
local vibrational mode associated to Ag ions as it was also
observed for ZnO doped with Mn 共Ref. 26兲 and Co.27 The
!" Raman peak appeared at about 690 cm−1 has also been re-
=$; ported for Co-doped ZnO.28 However, its origin is not yet
clear. The E2H mode of ZnO appeared at about 437 cm−1 for
#$%&$'(%) +,-.-/

the undoped sample. The intensity of this peak decreased

drastically on increasing the silver concentration in the
"

!7
"
!6+:;/ < !
"

samples. We can recall that the incorporation of Ag in our

96+:;/ < !
+! 75 ! "/
8! 75 !

ZnO nanoparticles reduces their crystallinity 共Fig. 2兲; and

96+:;/

96+";/

such a drastic reduction in intensity of the E2H mode might

be due to the breakdown of translational crystal symmetry by
!"
the incorporated defects and impurity. Raman peaks ap-
! " # $ % & ' ( )
peared at about 754 and 1065 cm−1 could be assigned to the
0,1,$ '2(3% +41 /
56 2A1共TO兲 second-order polar mode and multi-phonon mode,
respectively. Intensity of these Raman peaks increased with
FIG. 5. Typical Raman spectrum of ZnO nanoparticles. the increase of Ag content.

Downloaded 16 May 2013 to 147.188.128.74. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions
 014308-4 Zeferino, Flores, and Pal J. Appl. Phys. 109, 014308 共2011兲

4.0

Normalized PL Intensity (a.u.)

$%&
:63 ;9<=4
3.5

01 2%34%5637 *8,9,.
IVis /IUV
$%&'() *+,#-.
3.0
ZnO
ZnO:Ag (0.5% ) 0.5 1.0 1.5 2.0
ZnO:Ag (1.0% ) Ag concentration (% nominal)
$%&'() */,+-.
ZnO:Ag (2.0% )

#>? #?@
"?@ !"# $%&'() *>,+-.

!! "!! #!! $!! %!!

400 500 600 700 800
A8=4B4%)3C *%D.
Wavelength (nm)
FIG. 8. 共Color online兲 Gaussian deconvolution of the visible PL band for the
FIG. 7. 共Color online兲 Room temperature PL spectra of the ZnO and Ag- ZnO nanoparticles containing: 共a兲 0%, 共b兲 0.5%, 共c兲 1.0%, and 共d兲 2.0%
doped ZnO nanoparticles. The inset shows the variation in relative intensity 共nominal兲 of Ag.
共IVis / IUV兲 with Ag content in the samples.

in to four Gaussian shaped components 共Fig. 8兲 peaked

For the case of ZnO:Ag 共0.5%兲 sample, there appeared a around 583 nm 共⬃2.12 eV兲, 483 nm 共⬃2.56 eV兲, 528 nm
Raman peak at about 275 cm−1 关Fig. 6共c兲兴, which has also 共⬃2.35 eV兲, and 645 nm 共⬃1.92 eV兲. While the component
been reported for ZnO doped with Fe, Sb, Al, and Ga.29 This band close to 2.58 eV has been frequently designated as the
anomalous vibrational mode could be associated to the blue emission33,34 and associated to the intrinsic defects 共va-
lattice-host intrinsic defects created by doping, as it has also cancies or interstitials of O and Zn and their complexes兲 in
been observed for the ZnO samples doped with so many ZnO nanostructures,35 the component band at about 2.35 eV
other elements. Several authors have assigned this anoma- was assigned as the green emission and associated to the
lous mode to the B1共low兲 silent mode due to its closeness to oxygen and zinc vacancies.36,37 On the other hand, the com-
the position of theoretically calculated B1 mode ponent bands appeared at about 2.12 eV and 1.92 eV are
共261 cm−1兲.30 frequently assigned as the yellow and orange emissions in
Room temperature PL spectra of the undoped and Ag- ZnO, respectively, and associated with the excess oxygen
doped ZnO nanostructures are presented in Fig. 7. There ap- 共Oi兲 and interstitial zinc atoms 共Zni兲.38 As can be seen from
peared an intense narrow emission band in the UV spectral the Fig. 8, while the intensity of the blue emission decreases,
range centered at around 380 nm for all the samples. Gener- the intensity of the green emission increases with the in-
ally, this emission has been assigned to the free excitonic crease in Ag concentration in the samples. While the incor-
recombination in ZnO.31 No appreciable change in the posi- poration of Ag does not affect the intensity of the orange
tion of this UV band has been observed on silver doping. emission considerably, the intensity of the yellow emission
Apart from this UV band, there appeared another wider reduces up to 1.0% 共nominal兲 of Ag, and then increases.
emission band in the visible spectral range, centered at To determine the nature of the UV emission in our Ag-
around 580 nm. This multicomponent visible emission in doped and undoped ZnO nanoparticles correctly, their room
ZnO has been frequently assigned to several intrinsic and temperature PL spectra were recorded at different excitation
extrinsic defects. As can be seen in the Fig. 7, the intensity of intensities of the laser beam. Generally, the intensity of the
this visible emission initially decreases for 0.5% 共nominal兲 NBE PL emission varies with the excitation power obeying
silver doping, and then increases for higher Ag concentra- the following relation:39
tions. Such anomalous intensity variation in Ag-doped ZnO
I = ␤ P␣ , 共2兲
nanostructures can be understood considering the Ag incor-
poration process in the nanoparticles. The Ag+ ions can be where I is the PL intensity, ␤ is a constant, P is the excitation
incorporated in to the ZnO nanostructures in two different power density, and ␣ is an exponent. For the excitation en-
ways: substituting Zn+2 ions creating doubly ionized oxygen ergy exceeding the band gap energy, ␣ varies in between 1
vacancies 共VO ··
兲, or incorporating as interstitials 共Agi兲.32 and 2 for excitonic transition, and ␣ ⬍ 1 for free to bound
While for low doping concentrations 共e.g., 0.5%兲, most of and donor-acceptor pair transitions.40 In Fig. 9, the log-log
the Ag+ ions might have been incorporated into the ZnO plots of PL intensity as the function of excitation power are
lattice through substitution, for higher concentrations, the ex- plotted for Ag-doped and undoped ZnO nanoparticles. As
cess amount of Ag+ ions incorporated in to the nanoparticles can be seen, the intensities of the UV emission under various
interstitially, creating larger amount of lattice defects. Pres- excitation power densities could be fitted by the power law
ence of metallic Ag peaks in the XRD patterns of the Ag- with superlinear dependence for all the samples, and the gra-
doped samples supports our argument. From the 共IVis / IUV兲 dients of the plots 共␣兲 were 1.037, 1.090, 1.093, and 1.099
intensity ratio plot against the silver concentration presented for the undoped, 0.5%, 1.0%, and 2.0% 共nominal兲 Ag-doped
as the inset of the Fig. 7, it is very clear that a higher Ag ZnO nanostructures, respectively. These results indicate
concentration in the nanostructures incorporates higher clearly an excitonic origin of the UV emission for all the
amount of structural defects. samples; and the Ag doping does not alter the luminescence
The broad visible emission band could be deconvoluted mechanism of ZnO nanoparticles.

Downloaded 16 May 2013 to 147.188.128.74. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions
 014308-5 Zeferino, Flores, and Pal J. Appl. Phys. 109, 014308 共2011兲

7#8

FX-1LO
a) D0X b)

FX-2LO
*!34$9 :-;<= 3.362 eV 3.362 eV

7#89>? ,;-@A/
3.314 eV
*!34$9 :-;B; 3.238 eV 3.314 eV
3.238 eV

)#($#*"(+ ,%-.-/ 7#89>? ,:-;A/

*!34$9 :-;B< x6
10 K 10 K

PL Intensity (a.u.)
50 K x6 50 K
7#89>? ,C-;A/ x8
100 K x8 100 K
*!34$9 :-;BB x8 200 K x8 200 K
x8 295 K x10 295 K

c) 3.364 eV d)
3.364 eV

!"#$%& '"( 3.238 eV

3.314 eV
3.314 eV
3.238 eV
10 K 10 K
x5 x6
!" !# $!" $!# x8
50 K
100 K x10
50 K
100 K
012"(%("3# 435$& 6$#*"(+ x10 200 K x11 200 K
x10 295 K x10 295 K

FIG. 9. 共Color online兲 Log-log plots of PL intensity and excitation power

3.1 3.2 3.3 3.4 3.5 3.1 3.2 3.3 3.4 3.5
for the Ag-doped and undoped ZnO nanoparticles.
Energy (eV) Energy (eV)

FIG. 10. 共Color online兲 Low temperature 共10 to 295 K兲 PL spectra of ZnO
As can be observed from the Fig. 9, there is a significant nanoparticles dopes with: 共a兲 0.0%, 共b兲 0.5%, and 共c兲 1.0% and 2.0% 共nomi-
increase in the UV emission intensity for the 0.5% Ag-doped nal兲 Ag.
sample in comparison with other samples, for the whole
range of considered excitation power 共2.3 to 0.05 mW兲. That
indicates the efficiency of exciton recombination was high ZnO nanowires, and associated it to the compressive stains
for this sample. However, the UV emission intensity de- generated in the nanostructures due to bending. The small
creased on increasing the Ag-content further. As has been blueshift in the D0X transition in our highly Ag-doped nano-
stated earlier, for Ag-doping at lower concentrations 共e.g., structures could be associated to the compressive strain gen-
0.5%兲, most of the incorporated Ag+ ions occupy ZnO lattice erated through Ag-doping, as has been observed from their
through Zn2+ ion substitution. When the incident UV light XRD results 共Table I兲. Generally a compressive strain causes
共325 nm emission of the He-Cd laser beam兲 excites the car- a blueshift in PL, and tensile strain causes a redshift.45
riers in ZnO nanostructures, the photocarriers can escape On increasing the temperature, all the excitonic peaks
more easily from Ag ions than from the Zn ions, leading a moved toward lower energy and gradually became broader
quicker diffusion of excitons in ZnO. On the other hand, the due to the LO-phonon scattering and increased thermal ion-
probability of excitonic recombination would be high in Ag- ization of excitons.46 Finally, above 200 K, all the excitonic
doped ZnO nanoparticles due to increased exciton emissions were quenched.
concentration.41 For the samples with higher content of Ag+
ions 共e.g., ZnO prepared with 1.0%, and 2.0% of Ag兲, as a
good fraction of Ag remain at interstitial sites of the ZnO
lattice forming small silver clusters, they would act as re-
combination centers, reducing the effective concentration of IV. CONCLUSIONS
free excitons 共FXs兲 and hence the intensity of UV emission.
This effect is commonly known as concentration quenching. Pure and Ag-doped ZnO nanoparticles of about 140 nm
To study the origin of the UV emission in more detail, average size and relatively low size dispersion could be
PL measurements for all the samples were carried out at grown by nonionic polymer assisted thermolysis of zinc ac-
different temperatures in between 10 and 295 K 共Fig. 10兲. etate. For a low concentration of doping, most of the incor-
There appeared three component peaks of the UV emis- porated silver ions occupy ZnO lattice sites through substi-
sion in the low-temperature PL spectra of the undoped and tution of Zn ions, and for high Ag concentrations, most of the
Ag-doped nanostructures. The predominant sharp peak ap- silver ions occupy interstitial sites forming small metallic
peared at about 3.362 eV could be assigned to a neutral clusters. While the incorporation of Ag in the ZnO nano-
donor bound exciton 共D0X兲.42,43 On the other hand, the other structures does not affect their band gap energy substantially,
two less intense peaks appeared at about 3.314 and 3.238 eV it affects their lattice vibration behaviors drastically. Accu-
correspond to the LO phonon replicas of the FX recombina- mulation of Ag at the interstitial sites causes a distortion in
tion, assigned as FX-1LO and FX-2LO transitions. The ob- ZnO lattice, breaks down its translational symmetry, increas-
served energy separation between the two replicas was about ing the intensity of the A1共TO兲 and A1共LO兲 polar optical
76 meV, which is consistent with the reported value 共72 phonon modes. The lattice incorporated Ag ions increase the
meV兲 for the LO phonon energy of ZnO.44 The energy sepa- probability of excitonic transitions and hence the intensity of
ration did not vary with the variation of Ag content in the UV emission in the PL through an increase in the concentra-
nanostructures. However, the position of the D0X transition tion of free excitonic states in the band structure of ZnO. On
slightly blueshifted 共from 3.362 to 3.364 eV兲 for higher Ag the other hand, the interstitial Ag atoms/ions produce lattice
contents 共1.0% and 2.0%兲. Fang et al.45 have also found a distortion, generate electron-phonon coupling, and reduce
similar shift in the position of D0X transition for their bent the intensity of UV emission, acting as PL quencher.

Downloaded 16 May 2013 to 147.188.128.74. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions
 014308-6 Zeferino, Flores, and Pal J. Appl. Phys. 109, 014308 共2011兲

23
ACKNOWLEDGMENTS C. J. Youn, T. S. Jeong, M. S. Han, and J. H. Kim, J. Cryst. Growth 261,
526 共2004兲.
24
The work was partially supported by VIEP-BUAP M. Tzolov, N. Tzenov, D. Dimova-Malinovska, M. Kalitzova, C. Pizzuto,
共Grant No. VIEP-EXC-2010-I3兲. R. Sánchez Zeferino ac- G. Vitali, G. Zollo, and I. Ivanov, Thin Solid Films 379, 28 共2000兲.
25
B. Jusserand and M. Cardona, Light Scattering in Solids V, Topics in
knowledges CONACyT, Mexico for granting the doctoral Applied Physics, edited by M. Cardona and G. Güntherodt 共Springer,
fellowship. Heidelberg, 1989兲, Vol. 66, p. 48.
26
C. Vargas-Hernández, F. N. Jiménez-García, and J. F. Jurado, Rev.
1
T. C. Zhang, Z. X. Mei, Y. Guo, Q. K. Xue, and X. L. Du, J. Phys. D: LatinAm. Metal. Mater. S1, 501 共2009兲.
27
Appl. Phys. 42, 065303 共2009兲. H. Zhou, L. Chen, V. Malik, C. Knies, D. M. Hofmann, K. P. Bhatti, S.
2
Z. L. S. Seow, A. S. W. Wong, V. Thavasi, R. Jose, S. Ramakrishna, and Chaudhary, P. J. Klar, W. Heimbrodt, C. Klingshirn, and H. Kalt, Phys.
G. W. Ho, Nanotechnology 20, 045604 共2009兲. Status Solidi 204, 112 共2007兲.
3
C. Klingshirn, Phys. Status Solidi B 71, 547 共1975兲. 28
N. Hasuike, R. Deguchi, H. Katoh, K. Kisoda, K. Nishio, T. Isshiki, and
4
X. B. Wang, C. Song, K. W. Geng, F. Zeng, and F. Pan, J. Phys. D: Appl. H. Harima, J. Phys.: Condens. Matter 19, 365223 共2007兲.
Phys. 39, 4992 共2006兲. 29
F. J. Manjón, B. Marí, J. Serrano, and A. H. Romero, J. Appl. Phys. 97,
5
Z. F. Liu, F. K. Shan, J. Y. Sohn, S. C. Kim, G. Y. Kim, Y. X. Li, and Y. 053516 共2005兲.
S. Yu, J. Electroceram. 13, 183 共2004兲. 30
J. Serrano, A. H. Romero, F. J. Manjón, R. Lauck, M. Cardona, and A.
6
R. H. Wang, J. H. Z. Xin, Y. Yang, H. F. Liu, L. M. Xu, and J. H. Hu, Rubio, Phys. Rev. B 69, 094306 共2004兲.
Appl. Surf. Sci. 227, 312 共2004兲. 31
C. Ton-That, M. Foley, and M. R. Phillips, Nanotechnology 19, 415606
7
W. Park, T. C. Jone, and C. J. Summers, Appl. Phys. Lett. 74, 1785 共1999兲. 共2008兲.
8 32
X. Wang, Y. Ding, Z. Li, J. Song, and Z. L. Wang, J. Phys. Chem. C 113, S. T. Kuo, W. H. Tuan, J. Shieh, and S. F. Wang, J. Eur. Ceram. Soc. 27,
1791 共2009兲. 4521 共2007兲.
9
D. Wang and C. Song, J. Phys. Chem. B 109, 12697 共2005兲. 33
Z. Fu, B. Lin, G. Liao, and Z. Wu, J. Cryst. Growth 93, 316 共2008兲.
10
L. Zhao, J. Lian, Y. Liu, and Q. Jiang, Appl. Surf. Sci. 252, 8451 共2006兲. 34
L. Dai, X. L. Chen, W. J. Wang, T. Zhou, and B. Q. Hu, J. Phys.: Condens.
11
S. H. Jeong, B. N. Park, S. B. Lee, and J. H. Boo, Surf. Coat. Technol. Matter 15, 2221 共2003兲.
193, 340 共2005兲. 35
R. Dingle, Phys. Rev. Lett. 23, 579 共1969兲.
12 36
S. Ananthakumar, S. Anas, J. Ambily, and R. V. Mangalaraja, J. Ceram. B. Lin, Z. Fu, and Y.Jia, Appl. Phys. Lett. 79, 943 共2001兲.
Proc. Res. 11, 164 共2010兲. 37
Z. L. Wang, J. Phys.: Condens. Matter 16, R829 共2004兲.
13 38
E. Monroy, F. Omnès, and F. Calle, Semicond. Sci. Technol. 18, R33 N. O. Korsunska, L. V. Borkovska, B. M. Bulakh, L. Yu. Khomenkova, V.
共2003兲. I. Kushnirenko, and I. V. Markevich, J. Lumin. 733, 102 共2003兲.
14
J. Fan and R. Freer, J. Appl. Phys. 77, 4795 共1995兲. 39
T. Schmidt, K. Lishka, and W. Zulehner, Phys. Rev. B 45, 8989 共1992兲.
15
H. S. Nalwa, Encyclopedia of Nanoscience and Nanotechnology 共Ameri- 40
Z. C. Feng, A. Mascarenhas, and W. J. Choyke, J. Lumin. 35, 329 共1986兲.
41
can Scientific Publishers, New York, 2004兲, Vol. 8, 336. Y. Zhang, Z. Zhang, B. Lin, Z. Fu, and J. Xu, J. Phys. B 109, 19200
16
A. Escobedo Morales, U. Pal, and M. Herrera Zaldivar, J. Nanosci. Nano- 共2005兲.
technol. 8, 6551 共2008兲. 42
W. W. Liu, B. Yao, Y. F. Li, B. H. Li, Z. Z. Zhang, C. X. Shan, D. X. Zhao,
17
F. Wang, Z. Ye, D. Ma, L. Zhu, and F. Zhuge, J. Cryst. Growth 283, 373 J. Y. Zhang, D. Z. Shen, and X. W. Fan, Thin Solid Films 518, 3923
共2005兲. 共2010兲.
18
L. M. Kukreja, S. Barik, and P. Mishra, J. Cryst. Growth 268, 531 共2004兲. 43
F. Fang, D. Zhao, B. Li, Z. Zhang, J. Zhang, and D. Shen, J. Phys. D:
19
C. X. Xu, X. W. Sun, X. H. Zhang, L. Ke, and S. J. Chu, Nanotechnology Appl. Phys. 42, 135415 共2009兲.
15, 856 共2004兲. 44
T. V. Butkhuzi, T. G. Chelidze, A. N. Georgobiani, D. L. Jashiashvili, T.
20
M. Schumm, ZnO-based semiconductors studied by Raman spectroscopy: G. Khulordava, and B. E. Tsekvava, Phys. Rev. B 58, 10692 共1998兲.
45
semimagnetic alloying, doping and nanostructures, Ph.D. thesis, Julius- F. Fang, D. Zhao, B. Li, Z. Zhang, D. Shen, and X. Wang, J. Phys. Chem.
Maximilians-Universität Würzburg, 2008. C 114, 12477 共2010兲.
21
J. M. Liu, C. K. Ong, and L. C. Lim, Ferroelectrics 231, 223 共1999兲. 46
T. Makino, C. H. Chia, N. T. Tuan, Y. Segawa, M. Kawassaki, A. Ohtomo,
22
G. J. Exarhos and S. K. Sharma, Thin Solid Films 270, 27 共1995兲. K. Tamura, and H. Koinuma, Appl. Phys. Lett. 76, 3549 共2000兲.

Downloaded 16 May 2013 to 147.188.128.74. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions