Available online at www.sciencedirect.

com

Materials Chemistry and Physics 107 (2008) 28–32

Influence of solvent on the structural and morphological properties

of AgI particles prepared using ultrasonic spray pyrolysis
Ivana Lj.Validžić a , Vukoman Jokanović a , Dragan P. Uskoković b , Jovan M. Nedeljković a,∗
a Vinča Institute of Nuclear Sciences, P.O. Box 522, 11001 Belgrade, Serbia
b Institute of Technical Sciences of the Serbian Academy of Sciences and Arts, Knez Mihailova 35/IV, 11001 Belgrade, Serbia
Received 5 April 2006; received in revised form 29 May 2007; accepted 20 June 2007

Abstract
Silver iodide particles were synthesized in the process of ultrasonic spray pyrolysis using aqueous solutions of thermodynamically stable silver
iodide clusters as precursor. The AgI particles were collected in solvents of different polarities such as water, 2-propanol and toluene. In order to
study influence of solvent and aging time on the morphological and structural properties of the AgI particles ultra-filtration was employed to isolate
solid material from solution. The scanning electron microscopy showed that morphology of the AgI particles is different for different solvents. The
AgI particles with hexagonal/triangular shape were obtained in water and toluene, while in 2-propanol hollow spheres were formed. The X-ray
diffraction analysis, besides wurtzite hexagonal and zinc blende cubic phases, revealed appearance of tetragonal high-pressure AgI modification
for sample collected in toluene and for sample aged in 2-propanol. The aging in toluene induced transformation of all three modifications into
single phase with one preferred orientation.
© 2007 Elsevier B.V. All rights reserved.

Keywords: Semiconductor; Aging; Scanning electron microscopy (SEM); X-ray diffraction

1. Introduction USP. In order to study influence of solvent (water, 2-propanol

Development of synthetic procedures for preparation of AgI
particles have been an area of active research mainly due to its
photosensitivity and superionic conductivity [1–5]. The growth
mechanism is particularly important since the sensitivity of the
photographic material depends on its grain size and crystal struc-
ture. Recently, much attention has been paid to the preparation
and characterization of AgI nanoparticles [6–14], as well as
structural changes of halides during the aging process [15].
On the other hand, very recently, it has been shown that small
AgI clusters are in equilibrium with insoluble silver iodide if
sufficient excess of iodide ions is present [16,17].
Ultrasonic spray pyrolysis (USP) has been used for prepa-
ration of particles of desired morphology on a large scale.
Recently, colloidal nanoparticles have been used as building
blocks in order to obtain hollow submicronic spheres in the
process of USP [18–21].
In this paper, we use aqueous solution containing clusters
with up to 10 AgI pairs as starting material in the process of
∗ Corresponding author. Tel.: +381 112458222; fax: +381 112447382.
E-mail address: jovned@vin.bg.ac.yu (J.M. Nedeljković).
and toluene) and aging time on the morphological and struc-
tural properties of the AgI particles prepared by the process
of USP, we took advantage of the fact that all the AgI parti-
cles can be prepared with exactly the same history before being
collected in solvent. We have strong evidence that morphology
of the AgI particles is solvent dependent. More important, we
observed tetragonal AgI modification in 2-propanol and toluene
which is known as a high-pressure modification. Also, phase
transformation of three coexisting phases (hexagonal, cubic and
tetragonal) into single phase was observed during the aging
process in toluene.
2. Experimental
All chemicals (silver nitrate, potassium iodide, 2-propanol and toluene) were
purchased from Merck and they were of the highest purity available and used
without further purification. Aqueous solutions of thermodynamically stable
AgI clusters, used as a starting material in the process of USP, were prepared as
described in literature [16]. Briefly, 1.25 M AgNO3 was combined with 5 M KI.
After precipitation of AgI, supernatant solution containing thermodynamically
stable AgI clusters was separated and used in further work.
The AgI powders were obtained in the process of USP using solutions
containing AgI clusters as a starting material. Briefly, laboratory setup for
USP consists of ultrasonic atomizer (GAPUSOL-RBI-91-012) operating at a

0254-0584/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.matchemphys.2007.06.035
 I. Lj.Validžić et al. / Materials Chemistry and Physics 107 (2008) 28–32 29

frequency of 1.7 MHz for aerosol generation, and horizontal electric furnace
with the quartz tube and a vessel for particle collection. The effective heating
length of reactor tube was 1.25 m with the maximum temperature of 300 ◦ C in
the middle of the furnace. The flow rate of air was 30 l h−1 . The flow rate of
aerosol droplets was assumed to be equal to the flow rate of gas carrier, and
residence time of aerosol droplets in the furnace was found to be 1 min.
The obtained AgI powders were collected in different solvents such as water,
2-propanol and toluene. In the first series of experiments AgI particles were
separated from solvents containing excess of KI immediately after synthesis,
while in the second set of experiments aging time of AgI particles was 2 weeks.
The separation process was performed using ultra-filtration through a Milli-
pore membrane of 0.01 ␮m pore size, and solid AgI was characterized using
optical and structural techniques. All samples were prepared under the same
experimental condition.
Absorption spectra of the precursor solution, as well as solutions obtained
after the process of USP were measured using a Perkin-Elmer Lambda 35
UV–vis spectrophotometer. Diffuse reflectance measurements of solid AgI were
performed on the same instrument using the Labsphere RSA-PE-20 accessory.
The scanning electron microscopy (SEM) measurements were performed
using JEOL JSM-6460LV instrument. The solid AgI samples on a Millipore
membrane were coated with thin layer of gold deposited by sputtering process.
The thickness of the gold film was up to 40–50 nm.
The X-ray diffraction (XRD) measurements were carried out on a Philips
PW-1050 automatic diffractometer, using Cu K␣ radiation. Measurements were
performed in the 2θ range from 10◦ to 120◦ with scanning step width of 0.02◦
and time of 10 s per step.

3. Results and discussion

Fig. 1. (A) Absorption spectra of precursor (a) and AgI particles collected in
water after the process of USP (b). (B) Absorption spectra of 2-propanol (a) and
Recently, it has been shown that clusters containing up to 10 toluene (b) solution in equilibrium with AgI particles obtained by USP.
AgI pairs in aqueous solution are in equilibrium with the insolu-
ble silver iodide if sufficient excess of iodide ions is present [16].
Clusters of such a small agglomeration number were, to the best some complex species are in equilibrium with bulk AgI phase.
of our knowledge, the first time used as starting material in the Most likely polynuclear complexes are formed in which the Ag
process of USP. So far, either ionic species or uniform nanopar- atoms are joined by iodide bridges. Anyway, the structure of
ticles were used as a precursor. During the process of USP the complex species is not subject of this paper and we will further
sprayed droplets containing AgI clusters were transformed into concentrate on morphology and structure of bulk AgI obtained
particles by different processes, including solvent evaporation by the process of USP.
and in this case evaporation of NO3 − ions. The obtained AgI SEM image of the AgI sample aged 2 weeks in water is shown
particles were collected in different solvents such as water, 2- in Fig. 2. Most of the particles are in micrometer size range
propanol and toluene. We performed ultra-filtration immediately and they have hexagonal/triangular habits. Sample obtained
after the process of USP and after 2 weeks in order to under- immediately after the process of USP contained mostly spher-
stand influence of solvent and aging on the morphological and ical particles smaller then 100 nm (image not shown). In this
structural properties of AgI.
Absorption spectra of the precursor solution and dispersion
in water obtained after the process of USP are shown in Fig. 1A.
Featureless absorption spectrum of AgI clusters (curve a) is the
same as already shown spectrum in literature [16]. Absorption
spectrum of dispersion in water obtained after the process of USP
(curve b) showed typical features of bulk AgI with an excitonic
peak around 420 nm [11]. Long wavelength tail is a consequence
of light scattering due to large size of AgI particles. Reflection
spectra (not shown) of powders obtained by ultra-filtration of
samples aged in 2-propanol and toluene showed the same opti-
cal behaviour as AgI aged in water. However, absorption spectra
of filtrates showed different spectral properties for different sol-
vents (Fig. 1B). No absorption was observed in filtrate from AgI
aged in water, except in UV region due to the presence of iodide
ions (spectrum not shown). Absorption spectra of filtrates of AgI
aged in 2-propanol (yellow solution with peak at 360 nm) and
toluene (orange-red solution with peak at 495 nm) indicated that Fig. 2. SEM image of AgI particles aged 2 weeks in water.
30 I. Lj.Validžić et al. / Materials Chemistry and Physics 107 (2008) 28–32
Fig. 3. SEM images of AgI particles collected in 2-propanol immediately after
the process of USP.
case particle size distribution could not be precisely determined
because before the SEM analysis sample was coated with thin
layer of gold whose thickness is comparable to the size of AgI
particles. The disappearance of small spherical particles as well
as very broad size distribution of aged sample indicates growth
of large particles on the expense of small ones, i.e. Ostwald
ripening growth mechanism.
SEM images of AgI particles collected in 2-propanol imme-
diately after the process of USP are shown in Fig. 3. It can
be noticed that hollow AgI spheres are constituted of smaller
subunits, suggesting that an aggregation mechanism could be
responsible of particle growth [18]. No significant difference
in shape or size distribution of particles was observed for sam-
ples aged 2 weeks in 2-propanol. Although samples were coated
with gold layer we estimate that thickness of wall of hollow AgI
spheres must be smaller than 100 nm. Our attempt to perform
transmission electron microscopy measurements failed because
the hollow AgI spheres are extremely fragile and change when
exposed to electron beam. To the best of our knowledge hollow
AgI particles never been synthesized before. Submicronic TiO2
and SiO2 particles of the similar shape have been synthesized
by the process of USP using nanoparticles as precursors with or
without the presence of latex particles as a template [19–23].
SEM images of AgI particles collected in toluene imme-
diately after the process of USP and after 2 weeks of aging
are shown in Fig. 4(A and B). The most of the particles have
hexagonal habits and change of shape of the particles was not
observed with the aging. On the other hand, we observed sig-
nificant change of the particle size distribution with the aging
time (see Fig. 4C). We took as a measure of particle size mean
value of the smallest and largest distance in plane. Opposite to
the aging process in water, the average size decreased from 2.3
to 0.8 ␮m followed with the narrower size distribution of AgI
particles. This result is in agreement with experimental data con-
cerning Au, Ag and PbS nanocrystals synthesized in organic
solvents in the presence of excess ligand where the growth inhi-
bition or even disintegration of larger nanocrystals took place
in the process named digestive ripening [24]. Also, we believe
that decrease of the mean particle size might be connected with
disappearance of random orientation of the crystal grains and
promotion of preferential alignment during the aging in toluene
(detailed structural analysis is given further in the text) [25].
The change of solvent polarity and aging caused two main
effects in AgI particles obtained by USP process: (i) change of
the growth mechanism and (ii) change of particle shape. The
fact that all AgI particles are formed under the same experimen-
tal conditions before being collected in solvent gave us unique
opportunity to study influence of different solvents on the aging
process of AgI. It is obvious that aging in water and 2-propanol
follow Ostwald ripening and aggregation growth mechanism,
respectively, while aging in toluene induced decrease of the size
of AgI particles and narrower size distribution. On the other
hand, significant effect of solvent composition on the shape and
size of the particles is similar to already reported data in literature
[26–28].
The AgI particles collected in different solvents (water, 2-
propanol and toluene) were isolated as solid material by ultra-
filtration of the colloidal dispersion at different aging times and
characterized by the XRD measurements. The XRD analysis
clearly showed that ultra-filter is amorphous which aloud its
usage as substrate in structural study of AgI particles. After
subtraction of the signal from ultra-filter, the well-defined XRD
patterns indicated high degree of crystallinity of AgI samples.
The XRD patterns of the AgI particles collected in water
immediately after the process of USP and after 2 weeks of aging
are shown in Fig. 5(A and B). It can be noticed that two AgI
phases coexist, hexagonal wurtzite and cubic zinc blende struc-
ture. Hexagonal phase dominates in the XRD pattern of AgI
obtained immediately after the process of USP (Fig. 5A). The
XRD pattern of AgI after 2 weeks of aging (Fig. 5B) indi-
cated increase of the content of cubic phase (peaks attributed
only to the cubic phase at 56.7 and 62.3 are more pronounced),
although the hexagonal phase is still dominant. Transformation
of hexagonal into cubic AgI modification as a function of aging
is in agreement with recent theoretical calculations indicating
slightly higher thermodynamic stability of cubic phase [29].
The XRD patterns of the AgI particles collected in 2-propanol
immediately after the process of USP and after 2 weeks of aging
are shown in Fig. 5(C and D). Two AgI phases, hexagonal and
cubic, coexist in the sample with no aging. However, the aging
 I. Lj.Validžić et al. / Materials Chemistry and Physics 107 (2008) 28–32 31

Fig. 4. SEM images of AgI particles collected in toluene: (A) immediately after the process of USP and (B) after 2 weeks of aging. (C) Particle size distribution
(PSD) of AgI particles collected in toluene: (a) immediately after the process of USP and (b) after 2 weeks of aging.

of AgI in 2-propanol induced appearance of new tetragonal
phase (peaks at 14.6◦ , 29.3◦ and 34.5◦ can be only attributed to
tetragonal phase). Tetragonal modification of AgI is known as a
high-pressure modification which can exist at room temperature
only if pressure is between 2.8 and 3.8 kbar [30]. Appearance of
unusual structures as a function of aging time has been observed
with some other halides [15]. For example, yellow orthorhom-
bic form of HgI2 that spontaneously converts to the tetragonal
modification at temperature above 204 K was isolated during the
aging process at room temperature.
The XRD patterns of the AgI particles collected in toluene
immediately after the process of USP and after aging of 2 weeks
are shown in Fig. 5(E and F). All three AgI phases (hexagonal,
cubic and tetragonal) coexist in sample obtained immediately
after the process of USP. The XRD pattern of the AgI sample
aged 2 weeks in toluene clearly showed presence of only one
diffraction peak at 24.1◦ . It is obvious that during the aging pro-
cess in toluene hexagonal, cubic and tetragonal crystal phases
transformed into single phase with one preferred orientation.
According to the JCPDS data positions of (0 0 2) hexagonal,
(1 1 1) cubic, as well as (1 1 3) and (2 2 2) tetragonal crystal
planes are very close (23.706◦ , 23.726◦ , 23.986◦ and 24.185◦ ,
respectively). Because of that we are not certain which one
of three crystal phases survived aging in toluene. However,
tetragonal AgI phase is our favourite because experimentally
determined position of the XRD peak (24.1◦ ) is just in between
positions of very close to each other (1 1 3) and (2 2 2) tetragonal
crystal planes (23.986◦ and 24.185◦ , respectively). Also, noticed
tendency that decrease of solvent polarity promotes appearance
of tetragonal AgI phase supports above mentioned statement.
We believe that this is the first example of AgI particles with
single crystal phase synthesized under standard experimental
conditions. A similar situation has been observed for a different
material (BaTiO3 ), where the cubic phase is registered at room
temperature for samples prepared under special conditions [31].
The stabilization at room temperature of the cubic phase in this
material was attributed to small crystallite size.
In conclusion, we showed that it is possible to synthesize AgI
particles in wide size range by using thermodynamically sta-
ble silver iodide clusters as precursor and the USP as method.
These particles, collected and aged in different solvents, exhibit
extraordinary reach behaviour in the structural and morpholog-
ical sense. Proper choice of solvent and aging can lead to the
formation of AgI particles with desired morphology or even
32 I. Lj.Validžić et al. / Materials Chemistry and Physics 107 (2008) 28–32

References

[1] R.H. Otterwill, R.F. Woodbridge, J. Colloid Sci. 16 (1961) 581.

[2] C.R. Berry, Phys. Rev. 161 (1967) 848.
[3] T. Tanaka, M. Iwasaki, J. Imaging Sci. 29 (1985) 86.
[4] D. Hayes, K.H. Schmidt, D. Miesel, J. Phys. Chem. 93 (1989) 6100.
[5] P. Mulvaney, Colloid Surf. A 81 (1993) 231.
[6] M.I. Vučemilović, O.I. Mićić, Radiat. Phys. Chem. 32 (1988) 79.
[7] O.I. Mićić, M.T. Nenadović, T. Rajh, J.M. Nedeljković, M.I. Vučemilović,
Prog. Colloid Polym. Sci. 76 (1988) 20.
[8] K.H. Schmidt, R. Patel, D. Meisel, J. Am. Chem. Soc. 110 (1988) 4882.
[9] A. Henglein, M. Gutierrez, H. Weller, A. Fojtik, J. Jirkovsky, Ber. Bunsen-
ges Phys. Chem. 93 (1989) 593.
[10] O.I. Mićić, M. Meglić, D. Lawless, D.K. Sharma, N. Serpone, Langmuir 6
(1990) 487.
[11] S.H. Chen, T. Ida, K. Klimura, J. Phys. Chem. B 102 (1998) 6169.
[12] M.C. Brelle, J.Z. Zhang, J. Chem. Phys. 108 (1998) 3119.
[13] T. Ida, K. Kimura, Solid State Ionics 107 (1998) 313.
[14] Y.H. Wang, J.M. Mo, W.L. Cai, L.Z. Yao, J. Mater. Res. 16 (2001) 990.
[15] J.M. Nedeljković, R. Herak, O.I. Mićić, Langmuir 8 (1992) 299.
[16] I.Lj. Mladenović, W.K. Kegel, P. Bomas, P.M. Freederik, J. Phys. Chem.
B 107 (2003) 5717.
[17] I.Lj. Validžić, W.K. Kegel, J. Colloid Interface Sci. 275 (2004) 197.
[18] E. Matijević, Colloid J. 69 (2007) 29.
[19] J.M. Nedeljković, Z.V. Šaponjić, Z. Rakočević, D.P. Uskoković, Nanos-
truct. Mater. 9 (1997) 125.
[20] Z.V. Šaponjić, Z. Rakočević, N.M. Dimitrijević, J.M. Nedeljković, V.
Fig. 5. XRD patterns of AgI particles obtained by the process of USP and Jokanović, D.P. Uskoković, Nanostruct. Mater. 10 (1998) 333.
collected in different solvents (water, 2-propanol and toluene) as a function of [21] V. Jokanović, B. Jokanović, J. Nedeljković, O. Milošević, Colloids Surface
the aging time. A: Physicochem. Eng. Aspects 249 (2004) 111.
[22] V. Jokanović, A.M. Spasić, D. Uskoković, J. Colloid Interface Sci. 278
(2004) 342.
transformation of different crystal modifications into single one. [23] C. Song, G. Gu, Y. Lin, H. Wang, Y. Guo, X. Fu, Z. Hu, Mater. Res. Bull.
We believe that correlation between structural changes of bulk 38 (2003) 917.
AgI and the structures of the small complex precursor species [24] X.M. Lin, C.M. Sorensen, K.J. Klabunde, J. Nano Res. 2 (2000) 157.
exists, and development of theoretical model is currently under- [25] S. Mossa, F. Sciortino, P. Tartaglia, E. Zaccarelli, Langmuir 20 (2004)
10756.
way in our laboratory. [26] S. Hamada, E. Matijević, J. Chem. Faraday Trans. 178 (1982) 2147.
[27] W. Wang, B. Gu, L. Liang, W.A. Hamilton, D.J. Wesolowski, J. Phys.
Acknowledgments Chem. B 108 (2004) 14789.
[28] A.C.S. Samia, K. Hyzer, J.A. Schlueter, C.J. Qin, J.S. Jiang, S.D. Bader,
The authors would like to thank Dr. Miodrag Mitrić for his X.M. Lin, J. Am. Chem. Soc. 127 (2005) 4126.
[29] V. Söhnel, H. Hermann, P. Schwerdtfeger, J. Phys. Chem. B 109 (2005)
help and useful comments concerning the XRD measurements. 526.
Financial support for this study was granted by the Ministry of [30] D.A. Keen, S. Hull, J. Phys.: Condens Matter 5 (1993) 23.
Science and Environmental Protection of the Republic of Serbia [31] L.A. Pérez-Maqueda, M.J. Diánez, F.J. Gator, M.J. Sayagués, C. Real, J.M.
(Project No. 142066). Criado, J. Mater. Chem. 13 (2003) 2234.