RSC Advances

View Article Online

PAPER View Journal | View Issue

Band gap engineered zinc oxide nanostructures via

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Cite this: RSC Adv., 2019, 9, 14638

a sol–gel synthesis of solvent driven shape-
controlled crystal growth†
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

Klinton Davis, Ryan Yarbrough, Michael Froeschle, Jamel White

and Hemali Rathnayake *

A reliable sol–gel approach, which combines the formation of ZnO nanocrystals and a solvent driven,
shape-controlled, crystal-growth process to form well-organized ZnO nanostructures at low
temperature is presented. The sol of ZnO nanocrystals showed shape-controlled crystal growth with
respect to the solvent type, resulting in either nanorods, nanoparticles, or nanoslates. The solvothermal
process, along with the solvent polarity facilitate the shape-controlled crystal growth process,
augmenting the concept of a selective adhesion of solvents onto crystal facets and controlling the final
shape of the nanostructures. The XRD traces and XPS spectra support the concept of selective adhesion
of solvents onto crystal facets that leads to yield different ZnO morphologies. The shift in optical
absorption maxima from 332 nm in initial precursor solution, to 347 nm for ZnO nanocrystals sol, and
finally to 375 nm for ZnO nanorods, evidenced the gradual growth and ripening of nanocrystals to
dimensional nanostructures. The engineered optical band gaps of ZnO nanostructures are found to be
ranged from 3.10 eV to 3.37 eV with respect to the ZnO nanostructures formed in different solvent
systems. The theoretical band gaps computed from the experimental XRD spectral traces lie within the
range of the optical band gaps obtained from UV-visible spectra of ZnO nanostructures. The spin-casted
thin film of ZnO nanorods prepared in DMF exhibits the electrical conductivity of 1.14
103 S cm1,
which is nearly one order of magnitude higher than the electrical conductivity of ZnO nanoparticles
formed in hydroquinone and ZnO sols. The possibility of engineering the band gap and electrical
Received 18th March 2019
Accepted 29th April 2019
properties of ZnO at nanoscale utilizing an aqueous-based wet chemical synthesis process presented
here is simple, versatile, and environmentally friendly, and thus may applicable for making other types of
DOI: 10.1039/c9ra02091h
band-gap engineered metal oxide nanostructures with shape-controlled morphologies and
rsc.li/rsc-advances optoelectrical properties.

attention for solar cells,12 lasers,13 spintronics,14 transparent

Introduction conductive oxide,12 catalysis,15 bioimaging,16 and biosensors.17
Zinc oxide nanostructures are functional materials that can Among dimensional ZnO nanomaterials, one-dimensional
tailor their morphology through a variety of synthesis methods (1D) ZnO nanostructures with defect free high crystallinity are
to yield a wide range of morphologies such as nanowires,1,2 a particular interest due to their unique and inherent intrinsic
nanorods,3–5 nanobelts,6 nanocombs,7 nanorings,8 and nanoc- chemical, electrical, physical, and mechanical properties
ages.9 Owing to the lack of a centre-of-symmetry in the wurtzite compared to that of bulk and thin lm counterpart.10 However,
crystal structure, and high exciton binding energy (60 meV), synthesis of defect free 1D ZnO nanostructures, with desired
nanostructured ZnO possess strong piezoelectric10 and pyro- morphology and composition, has been challenging as most
electric11 properties and acts as a wide band gap (3.37 eV) growth techniques involve either high-cost fabrication
semiconductor for short wavelength optoelectronic devices.12 In processes or high temperature wet-chemical syntheses per-
recent research advancements, ZnO has received considerable formed in highly toxic solvents. Among such fabrication tech-
niques, chemical vapor deposition (CVD),18 pulse-laser
deposition (PLD),19 molecular beam epitaxy (MBE),20 and
electro-chemical deposition21 have been utilized to grow ZnO
Department of Nanoscience, Joint School of Nanoscience & Nanoengineering, nanostructures directly onto the substrate. In addition to these
University of North Carolina at Greensboro, Greensboro, NC, 27401, USA. E-mail:
thin lm deposition techniques, sol–gel and solvothermal
[email protected]; Tel: +1-336-285-2860
† Electronic supplementary information (ESI) available: Additional SEM and TEM
methods are two of the most common chemical solution
images along with XRD powder diffraction traces. See DOI: 10.1039/c9ra02091h methods, which show the promise in terms of scalability,

14638 | RSC Adv., 2019, 9, 14638–14648 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Paper RSC Advances

energy efficiency, and cost-effectiveness for the preparation of to make metal oxide nanostructures by adjusting experimental
catalyst-free metal oxide nanostructures with better control over conditions, such as monomer concentration, catalysts, and
the growth conditions and morphology.22–24 Particularly, the solvent type, in combination with a suitable growth tempera-
sol–gel process has been extensively investigated for making ture.36 For example, utilizing the concept of crystal growth-
homogenous, highly stoichiometric, and high-quality metal oriented attachment mechanism, the formation of ZnO nano-
oxide nanostructures such as nanorods,25,26 nanoakes,27 rods from 3 nm nanocrystals upon prolong heating was
nanotubes,25,28 and nanobers.29 In general, the sol–gel process demonstrated.37 Also, the concept of selective adhesion of
involves formation of sol from homogeneously mixed solutions surfactants onto crystal facets has been widely tested for the
of a metal precursor and a base. The sol of metal oxide nano- shaped-controlled synthesis of a variety of metal oxide nano-
This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

crystals can either be deposited onto a substrate to grow structures.38–40 Despite the work describes herein, up to date,
nanostructures or be continued through the polycondensation there has been no efficient, wet-chemical synthesis process
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

process to form gels. The gel can be used to form particles, developed, augmenting the shape-controlled crystal growth by
xerogels, aerogels, glass, and ceramics, depending upon the combining Ostwald ripening and oriented attachment mecha-
nal processing step involved.30 nisms. In our method, the monomer concentration and the
Up to date, the sol–gel method of either aqueous or non- reaction temperature were kept constant in order to minimize
aqueous hydrolysis and condensation process, has been the effect of the crystal growth regime, which is a balance
adapted primarily to make sols of ZnO nanoparticles, followed between the kinetic and thermodynamic growth regimes.
by casting on substrates to grow ZnO nanostructures, upon Herein, we hypothesize that the polarity of the solvent selec-
subjecting to thermal annealing.1,4,5,31,32 However, as per our tively adsorbs onto surfaces of the growing crystallites to
knowledge, there is no records of a sol–gel based green modulate the surface energy of nanocrystals and yields the nal
synthesis method, which combines the formation of sols and shape of the nanostructures.
a solvent-driven shape-controlled crystal growth process in The ZnO nanostructures were prepared by combining the
solution at low temperature to make ZnO nanostructures, with sol–gel chemical process with a solvent-driven nanocrystals
the wurtzite crystal lattice. This method benets to make ZnO growth at low temperature. The typical sol–gel process involved
nanomaterials in powder form for solution processable thin base catalysed hydrolysis of the metal salt to form metal
lms fabrications in large scale. Thus, herein, we demonstrate hydroxide followed by condensation to form nanocrystals of the
a reliable sol–gel approach, which combines the formation of metal oxide sol. Upon subjecting the sol of nanocrystals to an in
a ZnO nanocrystals sol and a solvent driven shape-controlled in situ solvothermal process, ZnO nanostructures with different
situ crystal growth process to form well-organized ZnO nano- morphologies were obtained. The solvothermal process induces
structures at low temperature (<80  C) in solution. This one-pot the nanocrystal elongation along the facet of high energy crystal
synthesis process allows us to make ZnO nanostructures with lattice surface and initiates nanocrystals growth through Ost-
a variety of morphologies, without disturbing their crystalline wald ripening and oriented attachment mechanisms.35 Also, the
structures and compositions. The ZnO nanostructures prepared highest packing efficiency of nanocrystals into well-ordered
in this manner showed excellent improvement in crystallinity nanostructures is feasible via solvent-driven self-assembly of
and optoelectrical properties, with engineered optical band gap. nanocrystals through van der Waals interactions.41–44 Such non-
The method developed here is a “green synthesis” where we covalent interactions enable interacting tiny nanocrystals to
utilized environmentally friendly and benign materials, with form larger crystal aggregates and induce the crystal growth
energy efficient wet-chemical approach. Thus, our method is mechanisms (Ostwald ripening and “oriented attachment”).34–36
rather advantages in terms of scalability, processability, and As a result, nanocrystals with sharing crystallographic orienta-
reliability compared to other wet-chemical and electrochemical tions directly combine together to form larger ones along the
synthesis methods, which typically grow nanostructures directly most preferential crystal lattice axis. The dimensionality and
from thin lm casted sol in environmentally friendly manner or morphology of these crystalline nanostructures were tailored by
high temperature solvothermal methods that utilize highly toxic changing the solvent mixture. The difference in polarity and
hydrazine-based solvents and additives. surface adhesion of each organic solvent controlled the shape
and size of nanocrystal growth. The solvent molecules act as
Results and discussion surfactants that adsorb onto surfaces of the growing crystal-
lites.45 The selective adhesion of solvent molecules onto crystal
In colloidal solutions, shape-controlled nanocrystal growth is lattice facets is governed by the surface energy difference, which
governed by: (1) the classical crystal growth kinetics of Ostwald drives solvent binding onto a selected facet. For example, past
ripening theory and the “oriented attachment” mechanism literature evidences that if a solvent has high binding affinity to
where a sol of nanocrystals with shared crystallographic orien- the {001} facet of the crystal lattice, the growth rate along the
tations directly combine together to form larger ones;33–35 (2) the direction of the [001] plane reduces and consequently, results in
relative surface energy of crystal facets and selective adhesion of the formation of nanorods.37,40,45 Similarly, in our studies, we
solvents/surfactants onto crystal facets; and (3) the crystal speculate that depending on the polarity, chemical function-
growth regime, which depends on the monomer concentration ality, hydrophilicity and hydrophobicity, solvents selectively
and temperature.23,24 Past studies demonstrated that shape- adsorb onto different faces of the nanocrystal lattice, reducing
controlled crystal growth has been successfully implemented

This journal is © The Royal Society of Chemistry 2019 RSC Adv., 2019, 9, 14638–14648 | 14639
 View Article Online

RSC Advances Paper

the growth rate along a particular lattice facet to form a shape- relative to band edges; which is inversely proportional to the
controlled nanostructure.40 square of the particle size and reduced mass.48 Thus, their
energy spectrum is quantized. With the increase in crystallites
Preparation and characterization size, a red shi in the absorption spectra can be observed
because of narrowing the band gap, which leads to the effective
In a typical procedure, the preparation of zinc oxide nano-
band gap smaller than its bulk value. In our case, the absorption
structures was performed by a base-catalysed hydrolysis and
peak shi from 332 nm in initial precursor solution, to 347 nm
condensation of zinc(II)chloride with NaOH in a series of
for ZnO nanocrystals sol, and nally to 375 nm for ZnO nano-
organic solvent–water systems in open atmosphere. The molar
rods (Fig. 1(a) and (c)) evidence the gradual growth and ripening
This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

ratio of the metal precursor to NaOH and the nanocrystals

of nanocrystals to dimensional nanostructures. As a result, we
growth temperature were tailored to achieve highly crystalline
observed a gradual decrease in the band gap from 3.57 eV to
nanostructures. The chemistry of making ZnO nanostructures
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

3.20 eV, conrming the presence of highly conned carriers in

is depicted in Scheme 1.
ZnO nanocrystals compared to that of in solvothermal grown
The sol of ZnO nanocrystals was prepared by mixing 1 : 5
ZnO nanorods. This is in good agreement with the previously
molar ratio of the metal precursor (ZnCl2) and the base (NaOH)
reported results.48
at room temperature with initially stirring for 15 minutes.
The sol prepared in this manner was subjected to in situ
Aer 15 minutes of stirring at room temperature, the sol of
condensation and shape-controlled nanocrystal growth process
nanocrystal formation was monitored at 15 minutes time
by transferring the reaction mixture onto a sand bath with no
intervals up to 60 minutes while homogeneous stirring at 80  C
stirring while maintaining the temperature at 80  C for over-
to yield an opaque solution. The reaction time for the formation
night (24 h). The nanostructures' formation, morphologies, and
of ZnO sol was monitored by acquiring UV-visible spectra of the
crystallinities were characterized by acquiring UV-visible
solution at different time intervals. The time-dependent UV-
spectra, SEM images, and X-ray powder diffraction (XRD)
visible spectra were compared with the absorption traces
traces combined with selective area electron diffraction (SAED)
collected for the initial reactants' solution at room temperature.
patterns. The atomic compositions and binding energies of Zn,
All the spectra were recorded in solution as either a clear or an
O, and C present in ZnO nanostructures with respect to
opaque solution. As depicted in Fig. 1(a), the UV-visible spec-
different morphologies were analysed using X-ray photoelectron
trum, obtained for the reactants' solution aer 15 minutes of
spectroscopy.
stirring at room temperature, exhibits the absorption maximum
at 264 nm with a shoulder peak at 332 nm, characterizing the
Zn2+ absorption. As the reaction progressed, a gradual red shi Morphology analysis
in the shoulder absorption band at 332 nm was observed. The
As shown in Table 1, shape-controlled ZnO nanostructures with
reaction mixture heated for 60 minutes showed a red-shied
different morphologies were obtained by varying the organic
absorption maximum at 347 nm, evidencing the formation of
solvent type in the reaction mixture to six different organic
ZnO nanocrystals (Fig. 1(a)). The transmission electron
solvents (dimethyl formamide, acetonitrile, dimethyl sulfoxide,
microscopy (TEM) images further conrmed that nanocrystals
toluene, hydroquinone, and xylene). The SEM analysis of the
were somewhat spherical in shape with the size ranging from 3–
nal products resulted from different solvent systems exhibit
5 nm (Fig. 1(b)).
three different distinct morphologies: hexagonal nanorods,
The absorption peak in the range of 345–450 nm arises due
slates like structures, and globular shaped nanoparticles (see
to the surface plasma resonance effect of ZnO nanocrystals and
Fig. 2 and S1†). The ZnO hexagonal nanorods, with 1–5 mm in
the peak shi from blue to red is characteristic to the quantum
length and 50 nm to 180 nm in width were resulted in water and
size effect.46–48 The stronger exciton effect is characteristic of the
dimethyl formamide (DMF) mixture whereas irregular slates
quantum connement in semiconducting nanostructures, in
like structures with an average length ranged from 500 nm to 2
which the electrons, holes, and excitons have limited space to
mm were formed in the solvent mixture of water and dimethyl
move; and their movement only is possible for denite values of
sulfoxide (DMSO). Morphologies of ZnO nanostructures, resul-
energies. As a result, the continuum of states in conduction and
ted from solvent systems of xylene, toluene, and acetonitrile
valence bands exhibits discrete states with an energy spacing
mixed with water, were shorter nanorods comparing to those
resulted in from the DMF/water solvent system. Particularly,
comparing to the dimensions of ZnO nanorods formed in
toluene and acetonitrile mixtures, lengths of nanorods formed
in m-xylene were considerably shorter. This may be due to the
differences in solubility of each solvent in water as toluene and
acetonitrile have higher solubility compared to m-xylene. Thus,
m-xylene/water system has high concentration of non-solvated
xylene molecules, which could hinder nanorods growth along
the [001] direction due to solvent molecule adhesion, yielding
Scheme 1 The formation of ZnO nanostructures via sol–gel process shorter and smaller in diameter nanorods compared to the
followed by solvothermal self-assembly. nanorods formed in toluene and acetonitrile. Surprisingly, the

14640 | RSC Adv., 2019, 9, 14638–14648 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Paper RSC Advances

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

Fig. 1 (a) Time-dependent UV-visible absorption spectra acquired during ZnO nanocrystals formation; (b) the TEM image of ZnO nanocrystals
formed after heating for 60 minutes; and (c) UV-visible spectra of ZnO sols and ZnO nanorods taken after 60 min and 24 h at 80  C respectively
(all spectra were acquired in aqueous solutions).

nanostructures prepared in a mixture of hydroquinone and literature.10 We observed a clear difference in the intensities of
water were globular shaped particles with an average diameter [100], [002], and [101] Bragg peaks as well as the peak resolution
ranging from 100 nm to 500 nm. of the [002] reection plane in each case. The diffraction pattern
The formation of ZnO nanorods, nanoparticles, and random obtained for the ZnO nanorods prepared in DMF/water solvent
slates like structures with respect to the solvent type can be system shows a poorly resolved second order reection of [001]
explained by understanding the solvent-driven shape-controlled plane as a shoulder that merges onto the [101] reection plane.
crystal growth process. The polarity and chemical nature of the The XRD traces of ZnO nanorods, prepared in the mixtures of
organic solvent in each reaction mixture may act as a selective acetonitrile, toluene, and xylene with water show a weak
adhesion surfactant to facilitate the shape-controlled crystal intensity peak for the [002] diffraction plane of the rst order
growth. As studied in the past literature, the morphology of the {001} facet, while maintaining the intensity of the [100] peak by
nal nanostructure controls by the surface energy and selective three-fold higher than the intensity of [002] reection plane.
adhesion of solvent molecules onto crystallites facets that However, the intensity ratios of [100] : [101] diffraction planes
modulate the crystal growth direction in the nanocrystal unit for the nanorods prepared in the presence of DMF, acetonitrile,
cell.35,36 Since the crystal growth rate is correlated exponentially and xylene were 1 : 2, which is slightly higher compared to the
to the surface energy, surface-energy differences induce much intensity ratio of the [100] : [101] peak for the nanorods
faster growth of the higher surface energy planes and keep the prepared in the toluene and was found to be 1 : 1.5. The reason
slower growing planes (lower surface energy) as the facets of the for the slight deviation in the intensity ratio could be due to the
product. Comparing the intensities of crystal planes reections dimensional difference between thin hexagonal nanorods and
for the crystallites facets of {100}, {001}, and {101} of XRD traces wide slates; as observed in Fig. 2. Overall, from these observa-
for each solvent system, the preferential crystal growth direction tions, it is evidenced that the presence of poorly resolved second
and the morphology of the nal nanostructure can be realized. order reection planes of {001} facets in the XRD patterns of the
Therefore, we investigated the diffraction patterns for the nal ZnO nanorods, formed in DMF, acetonitrile, xylene, and toluene
products of ZnO nanostructures by acquiring X-ray powder could be preferentially bonded to {001} and {002} facets and
diffraction traces and SAED patterns from the TEM under dark reduced the crystal growth rate along the second order reec-
eld diffraction mode. As shown in Fig. 3(a), the diffraction tion planes in the [001] direction to yield nanorods. Similarly, in
patterns of ZnO nanostructures prepared in ve of the six the past literature, it was shown that TiO2 nanorods was formed
solvent systems indexed to hexagonal phase wurtzite crystal when lauric acid was used and acted as the surfactant that
structure, and is in good agreement with the previous published

Table 1 The morphologies of ZnO nanostructures obtained in different solvent systemsa

Trial # Solvent system (6 : 1 v/v) Nanostructures morphology and dimension description

1 DMF : H2O Hexagonal rods with the average length of 1–5 mm and width of 50–180 nm
2 Acetonitrile : H2O Hexagonal rods with the average length of 1–3 mm and width of 50–180 nm
3 DMSO : H2O Thin slates like structures ranging from 500–2 mm
4 Toluene : H2O A mix of thin hexagonal rods and wide slates. The average length of rods 1–5 mm and width of 100–200 nm
5 Hydroquinone : H2O Globular shaped particle-like structures ranging from 100–500 nm
6 m-Xylene : H2O Very short hexagonal rods with the average length of <300 nm and width of 40–50 nm
a
A molar ratio of the metal precursor (ZnCl2) to NaOH was maintained at 1 : 5 for each trial. In each case, pH was maintained at the range of 12–13.

This journal is © The Royal Society of Chemistry 2019 RSC Adv., 2019, 9, 14638–14648 | 14641
 View Article Online

RSC Advances Paper

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

Fig. 2 SEM images of ZnO nanostructures prepared using different solvent systems; scale bars represent 200 nm.

reduces the growth rate along the [001] direction, binding that DMSO molecules hinder the crystal growth direction along
strongly to the {001} facet.40 both [100] and [001] faces by selectively binding to both surfaces
In contrast to the XRD patterns of ZnO nanorods, the XRD of {100} and {001} facets.
pattern obtained for the ZnO nanoparticles exhibits well- The SAED pattern of a single nanostructure prepared in each
resolved rst and second order reection planes for the Bragg solvent mixture was taken from the TEM under dark eld
peaks of [100], [002], and [101]. The difference in the intensity of diffraction mode by directing the electron beam at 90 angle to
these Bragg peaks conrms that the crystal growth rate along a one of the nanostructure's faces. The SAED patterns collected
the directions of [100], [002], and [101] planes is same. This in this manner are depicted in Fig. 3(b). The SAED patterns
suggests that there is no preferential adsorption of hydroqui- viewed along the [001] axis show defect free and single crystal-
none molecules onto a particular facet over others. In turn, it line nanocrystals growth for the nanostructures prepared in all
exhibits equal binding affinity for all facets of crystal lattice. So, other solvent systems except for the nanostructures prepared in
the crystals grew into nanoparticles, enabling the crystal growth DMSO. The single crystal unit cell diffraction planes are well
along all the directions of crystal facets. In contrast, the XRD aligned with the crystal growth directions, revealed from the
traces obtained for ZnO nanostructures, prepared in the pres- peak intensities of the respective XRD powder spectra. The
ence of DMSO show only the well-resolved [101] diffraction SAED patterns for the ZnO nanorods and nanoparticles, show
plane with a poorly resolved [100] reection peak. This suggests defect free single crystalline pattern with rst, second, and third

Fig. 3(a) The powder XRD patterns, and (b) the SAED patterns along with the TEM images of ZnO nanostructures prepared in different solvent
systems.

14642 | RSC Adv., 2019, 9, 14638–14648 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Paper RSC Advances

order facets of the nanocrystals' ordering along the [100] crystal facets of the lattice structure without any selectivity
diffraction planes. The difference in the well-resolved spacing of towards a particular crystal facet.
higher order reections at [100] inter-diffraction planes The Table S1† summarizes binding energy (BE) for the core
conrms the unit cell packing and unit cell distances of crys- levels Zn 2p, O 1s, C1s and their corresponding full-width at half
talline facets. As observed from the SAED patterns, ZnO nano- maximum (FWHM) along with the respective %atomic
rods prepared in DMF and toluene exhibit larger nanocrystal concentrations for two types of ZnO nanostructures and sol. The
unit cell spacing of higher order reections whereas ZnO C 1s atomic concentration for nanoparticles is twenty times
nanoparticles, formed in the presence of hydroquinone, show higher than that of ZnO sol and ve times higher than that of
the smallest unit cell spacing in the crystal packing. The crystal nanorods. Overall, with the increase in carbon and oxygen
This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

growth pattern and nanocrystal packing during the formation atomic concentrations and a gradual decrease in Zn atomic
of ZnO nanorods in acetonitrile and xylene are alike and exhibit concentration clearly evidence that the presence of adsorbed
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

same unit cell spacing among the higher order reections. The solvent molecules on crystallites facets that affect in modulating
SAED pattern obtained for ZnO nanostructures formed in the crystal growth direction in the nanocrystal unit cell. As we
DMSO (also see Fig. S2†) reveals weak polycrystalline pattern discussed in the morphology analysis section, it is clear that m-
with poorly resolved nanocrystals ordering along [100] plane, xylene prefers binding only onto a particular facet, hindering
conrming truncated crystal facets with polycrystallinity to the crystal growth along the respective facet, yielding rod shape
amorphous. nanostructures. Whereas hydroquinone shows equal binding
affinity onto all the facets of the crystal lattice and adsorb
X-ray photoelectron spectroscopy analysis of different solvent molecules onto all the facets, yielding spherical shape
morphologies of ZnO nanostructures nanostructures.
The noticeable differences in the binding energy spectra of
In order to correlate the selective solvent adhesion onto crystal
Zn 2p, O 1s, and C 1s of ZnO sol, nanorods, and nanoparticles
facets and the shape-controlled crystal growth process that
reveal the chemical environment interaction between solvent
yields ZnO nanostructures with either nanorods or nano-
molecules and the surface atoms of the crystal facets. Fig. 5
particles, X-ray photoelectron spectroscopy (XPS) analysis was
demonstrates the comparison of orbital binding energy states
conducted. The XPS wide survey spectra obtained for ZnO
of high resolution XPS spectra of the Zn 2p region, O 1s core-
nanorods, nanoparticles, and sol, are depicted in Fig. 4. In all
level, and C 1s for ZnO sol, nanorods, and nanoparticles. The
three cases, the atomic orbital bonding states of Zn and O peaks
binding energies of the Zn 2p components of sol are slightly
were detected and conrms the presence of strong Zn–O bonds
lower than the binding energies of the Zn 2p components of
on the nanostructure surface. The Zn peaks observed at 1044 eV
nanorods and nanoparticles (Fig. 5(a)). The low binding energy
(2.10 eV), 1020 eV (2.20 eV), 498 eV (0.10), and 475 eV
of Zn 2p in sol evidences that the chemical environment inter-
(0.15 eV) were attributed to the chemical states of the Zn
action of Zn–O bonding surfaces of sol could be different from
species, which include higher binding energy states for Zn 2p1/2
the chemical environment interactions of nanorods and nano-
and Zn 2p3/2 of Zn–O bonds, and lower binding energy states for
particles. This may also be due to the difference in surface
interstitial zinc (Zni) respectively.49 The carbon 1s peak is only
morphology and crystal size in nanocrystals compared to
noticeable in the XPS survey spectrum of ZnO nanoparticles,
nanorods and nanoparticles.50 Surprisingly, the binding energy
prepared in hydroquinone/water solvent system. The presence
peaks of the Zn 2p components for ZnO nanorods and nano-
of carbon 1s peak in nanoparticles suggests that there is
particles are located at the same positions. This suggests that
a considerable concentration of organic solvent molecules
the nature of chemical interactions in both rods and particles
adsorb onto the nanoparticle surface. This supports our
are attributed to strong Zn–O bonding interactions. In overall,
prediction of adsorbing hydroquinone molecules onto all the
Zn 2p components binding energy spectral traces indicate that
the chemical valence of Zn at the surface of nanostructure
morphologies and sol is Zn2+ oxidation state. The binding
energy difference between the Zn 2p1/2 and Zn 2p3/2 is 23 eV for
ZnO nanostructures and ZnO sol.
The O 1s spectrum for ZnO sol in Fig. 5(b) shows a sharp
peak at 529.1 eV and a shoulder peak at 530.5 eV. The lower
energy peak is attributed to O2 in Zn–O bonding of the wurtzite
structure of ZnO.51 The higher binding energy (530.5 eV) is
usually related to OH group absorbed onto the surface of the
ZnO nanoparticles.50 However in our case, sol usually contains
a mixture of ZnO and Zn(OH)2. It is expected to have OH groups
on nanocrystal surface. The O 1s spectra of nanorods and
nanoparticles, shown in Fig. 5(b), show only O2 in Zn–O
bonding with a shi in binding energy of 2 eV and 2.5 eV,
Fig. 4 XPS wide survey spectra of ZnO sol, nanorods (formed in m- respectively, compared to the binding energy peak of O2 in
xylene), and nanoparticles (formed in hydroquinone). ZnO sol. The peaks positions at 531.1 eV and 531.6 eV in

This journal is © The Royal Society of Chemistry 2019 RSC Adv., 2019, 9, 14638–14648 | 14643
 View Article Online

RSC Advances Paper

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

Fig. 5The comparison graphs of orbital binding energy states of high resolution XPS spectra of: (a) Zn 2p region, (b) oxygen 1s core-level, and (c)
carbon 1s state for ZnO sol, nanorods, and nanoparticles.

nanorods and nanoparticles are in good agreement with the and 3.58 eV respectively. The nanoparticles' optical band gap,
previous work of ZnO nanorods and nanoparticles.50,51 The C 1s calculated from the UV on-set was ranged from 3.33–3.41 eV,
spectra of sols, nanorods, and nanoparticles show three which is considerably higher than the optical band gap range of
different binding energy peaks, which are uniquely character- 3.16–3.24 eV for nanorods made in DMF solvent system. The
istic to each morphology of ZnO nanostructures. The C 1s ZnO nanorods prepared in acetonitrile, toluene, and xylene
spectrum of ZnO sol shows two very weak intensity peaks at solvent systems exhibit the absorption maxima ranging from
283.8 eV and 288.4 eV, reecting C–C/C–H and C–O bonding. In 3.30–3.39 eV with the lowest optical band gap of 3.10 eV for
our case, these weak C–C/C–H and C–O bonding energies are nanorods formed in acetonitrile. The optical band gaps of the
attributed to traces of ethanol molecules adsorbed onto the ZnO nanorods formed in toluene and xylene are 3.22  0.10 and
nanocrystal surface during the washing of centrifuged powder 3.19  0.04 respectively. The changes in the optical band gaps
form of ZnO sol. However, the binding energy peak for C–C/C–H with respect to the ZnO nanostructure morphology further
in nanorods and nanoparticles are clearly visible and slightly conrms that nanostructure crystallinity, crystal growth facets,
shied to yield binding energy of 284.4 eV and 284.8 eV and crystal grain size lead to the effective band gap of nano-
respectively. In nanoparticles, the intensity counts for the C–C/ structured ZnO smaller than its bulk value of 3.37 eV.
C–H bonding binding energy is signicantly higher than The optical band gap of ZnO nanorods formed in DMF
nanorods and conrms the presence of adsorbed organic solvent systems was compared with the theoretical band gap
molecules onto the nanoparticle surface. The higher binding computed from the respective ZnO crystal structure, which was
energy peaks at 289.8 eV (weak peak) and 288.9 eV (strong peak) extracted from a crystallographic data base, and followed by
in nanorods and nanoparticles, respectively, are attributed to adjusting the unit cell parameters comparing the simulated
carbonyl (C]O) bonds. The presence of a strong peak for XRD pattern of the extracted ZnO crystal structure with the
carbonyl bonds in nanoparticles further conrms that the experimental XRD traces of ZnO nanorods. The theoretical band
adsorbed hydroquinone molecules on nanoparticle surface. structure was computed using the plane-wave form of Density
Functional Theory (DFT) implemented in the open source
Quantum ESPRESSO (QE) suite. Ultra-so pseudopotentials
Optical properties and band gap analysis created with low-density approximation (LDA) functions were
The thin lms UV-visible spectra of ZnO nanostructures, used and were obtained from QE. The band gap calculated
prepared in different solvent systems, were collected and is using only DFT was severely underestimated, forcing a change
depicted in Fig. 6(a) and S4.† The optical band gaps were to DFT+U; this scheme uses an additional Hubbard U potential
calculated from the on-set of the spectra in each case. The for each element to correct band overlap discrepancies. The
deviation of the optical band gap with respect to the different potentials for zinc and oxygen to correct the large discrepancy
nanostructures formed in each solvent type were calculated by were based on literature, 12 eV and 6.5 eV respectively. The
collecting UV-visible thin lm spectra of 5–7 samples prepared potentials were applied to the 3d orbitals of the Zn atoms and
from each solvent type. Depending on the nanostructure the 2p orbitals of the O atoms.52,53 With this method, we were
morphology, there is a noticeable shi in the absorption able to compute the corrected band gap to be 3.29 eV (Fig. 6(b)),
maxima. The photo-absorption energy (hv) of the absorption which is in good agreement with the optical band gap of ZnO
maximum is ranged from 3.29 eV to 3.58 eV with respect to the nanorods formed in the DMF solvent system.
ZnO nanostructures formed in different solvent systems.
The comparison in the energy absorption maxima and
optical band gaps of the ZnO nanostructures formed in Electrical properties
different solvent systems are summarized in Table 2. The ZnO ZnO nanorods, formed in DMF/water and ZnO nanoparticles,
nanorods and nanoparticles formed in DMF and hydroquinone formed in hydroquinone/water, along with ZnO sol and sol-
solvent systems show maximum absorption energies at 3.42 eV vothermal grown ZnO nanorods were chosen to evaluate and

14644 | RSC Adv., 2019, 9, 14638–14648 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Paper RSC Advances

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

Fig. 6 (a) Thin film UV-visible spectra of the ZnO nanostructures with respect to the solvent type; (b) the calculated band structure for ZnO
wurtzite hexagonal lattice structure obtained by matching with the simulated XRD spectra and the experimental XRD traces (Fig. S3†). The energy
of the valence-band maximum (VBM) was set to zero and used Hubbard U+DFT hybrid function for the calculation; and (c) respective ZnO crystal
structure extracted from the crystallography data base to compute the band structure.

compare the electrical properties of ZnO at nanoscale. The Fig. S5†). In overall, the electrical conductivities of solution-
electrical conductivities were measured using a four-probe processed thin lms of ZnO nanorods and solvothermal
technique and a device conguration of a fabricated test grown ZnO nanorods are in good agreement with a typical ZnO
device is depicted in Fig. 7(a). The detailed experimental set up eld effect transistor's peak conductivity of 1.25

is described in the Experimental section. The average electrical 103 S cm1.10
conductivity of each morphology of ZnO nanostructures was
calculated from the slopes of the multiple IV curves. The I–V
curves of nanorods, nanoparticles, and sol are shown in Experimental
Fig. 7(b) along with ZnO nanorods directly grown from a ZnO Materials
sol-casted thin lm substrate using the solvothermal process. Sodium hydroxide (98% purity), and zinc chloride (97% purity),
The ZnO nanorods, formed in DMF/water solvent mixture, were obtained from Aldrich Chemicals. Anhydrous dimethyl
exhibits an average electrical conductivity of 1.14
103 S cm1 formamide, anhydrous dimethyl sulfoxide, meta-xylene, aceto-
(0.01), which is almost one order of magnitude higher than nitrile, hydroquinone, and toluene were used as received
the electrical conductivity of ZnO nanoparticles and ZnO sol. without any purication otherwise specied.
The average electrical conductivity of ZnO nanoparticles and sol
were found to be 6.65
104 S cm1 (0.06), and 2.67

104 S cm1 (0.13), respectively. The signicant increase in the Characterization
electrical conductivity of ZnO nanorods prepared in DMF The powder XRD analysis was conducted using Mo Ka radiation
solvent system further agrees with its narrow optical band gap, (40 kV, 40 mA, k ¼ 0.7093 Å) with a speed of 60 s on the X-ray
compared to the optical band gap of ZnO nanoparticles. As diffractometer (XRD, Agilent technologies Gemini). The
shown in Fig. 7(c), the electrical conductivity of ZnO nano- morphology and the size of ZnO nanostructures were analysed
structures gradually increases with respect to nanocrystal size, using transmission electron microscopy (TEM Carl Zeiss Libra
shape, crystallinity, and nanostructure orientation. In the case 120) at 120 keV and scanning electron microscopy (Zeiss Auriga
of sol, we observed the lowest electrical conductivity due to non- FIB/FESEM). The Atomic concentrations and binding energies
ripened nanocrystals. The nanorods grown from thin-lm cas- of all the elements present in ZnO nanostructures with different
ted sols shows the highest electrical conductivity (2.44
103  morphologies were obtained from X-ray photoelectron spec-
0.005 S cm1) as a result of crystal shape, defect free crystal- troscopy (XPS) using XPS-Escalab Xi+ Thermo Scientic electron
linity, and nanorods orientation on the glass substrate compare spectrometer. The binding energies were corrected for the
to the solution-processed ZnO nanorods thin lms (see charge shi using the C 1s peak of graphitic carbon (BE ¼ 284.6

Table 2 The energy absorption maxima and optical band gaps of the ZnO nanostructures formed with respect to solvent systems

Energy (hv)
Solvent system of the absorption maximum (eV) Optical band gap (eV)

DMF : H2O 3.40  0.02 3.20  0.04

Acetonitrile : H2O 3.36  0.03 3.15  0.05
Toluene : H2O 3.33  0.03 3.22  0.10
Hydroquinone : H2O 3.50  0.08 3.33  0.04
m-Xylene : H2O 3.34  0.04 3.19  0.04

This journal is © The Royal Society of Chemistry 2019 RSC Adv., 2019, 9, 14638–14648 | 14645
 View Article Online

RSC Advances Paper

This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

Fig. 7(a) A schematic diagram of a test device used for I–V characterization; (b) I–V graphs of ZnO nanostructures; and (c) the electrical
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

conductivities of solution formed nanorods and nanoparticles comparing with the electrical conductivity of sols and solvothermal grown
nanorods from ZnO sols-casted glass substrate.

eV) as a reference. The optical properties were determined using device fabrication. First, substrates were washed with Isopropyl
ultraviolet-visible spectroscopy (UV-vis spectroscopy, Varian Alcohol in an ultrasonic bath for ten minutes, followed by
Cary 6000i). The electrical properties were determined using washing with soap solution. Substrates were then rinsed in
Keithley source meter controlled by a Photo Emission TEC. INC deionized water and treated with a mixture of ammonium
(PET) I–V test system. E-beam evaporation (Kurt Lesker PVD 75 hydroxide and hydrogen peroxide for 15 minutes at 50–70  C.
e-beam evaporator) was used to deposit Cu (100 nm). Thin lms Aer the above treatment, the plates were again washed with
were prepared by spin coating the sample onto ozone/UV deionized water under ultrasonication for 15 minutes and
treated (Bio Force UV/Ozone Pro Cleaner) glass coverslips and blown dry with dry nitrogen. The cleaned glass substrates were
silicon substrates. subject to UV cleaning for 35 min prior to the deposition of the
sample. A solution of ZnO nanostructures (25 mg in 1 mL NMP)
Typical sol–gel synthesis procedure for the preparation of ZnO was spin coated on the substrate in nitrogen atmosphere at the
nanostructures rate of 1400 rpm for 10 seconds to spread, then 3000 rpm for 50
seconds to dry followed by annealing at 120  C for 5 min. This
In a typical procedure, the synthesis of ZnO nanostructures was
process was repeated six times using 70 mL of solution followed
performed using the wet chemical synthesis in an aqueous-
by vacuum drying under nitrogen atmosphere. The test devices
based solvent mixture by maintaining the molar ratio of zinc
prepared in this manner were subjected to the deposition of the
chloride precursor to sodium hydroxide at 1 : 5 mmol. First,
cathode (100 nm thick Cu layer) on top of the active layer using
zinc chloride (0.21 g) was dissolved in a selected organic solvent
PVD e-beam deposition and transferred into a nitrogen lled
(30 mL) and a solution of NaOH (0.40 g in 5 mL water) was
sealed chamber to prevent any oxidation of the copper layer and
added into the reaction mixture. The solution was initially
transferred into a nitrogen lled sealed chamber to prevent any
stirred at room temperature about 15 minutes and continued
oxidation of the copper layer. I–V measurements were con-
stirring another 60 minutes at 80  C. Then, the reaction ask
ducted immediately aer the deposition of the cathode. A
was sealed and transferred onto a sand bath and continued
schematic diagram of a test device is depicted in Fig. 7(a). For I–
heating at 80  C for 24 hours with no stirring. The white solid
V measurements, the channel length was kept constant at
was collected by centrifugation and repeated washing with
1.7 cm while the active cell area maintained at 4.25 cm2.
water to remove salt, and unreacted base and the precursor. The
Conductance of samples was calculated from the slope of the
white powder of ZnO nanostructures yielded (110 mg) was re-
ohmic region of I–V curves and conductivities were obtained
suspended in water (20 mL) with sonication as necessary. The
from;
suspension was drop casted on a cleaned silicon substrate and
imaged under SEM. The UV-visible spectra of ZnO nano- Gl
r¼
structures were obtained in water/ethanol mixture. The sol- a
vothermal grown ZnO nanorods on a glass substrate was where r is conductivity, G is conductance, and l and a are
obtained by spin coating the solution of sols on a cleaned glass channel length and area respectively. Each sample was tested
substrate followed by placing the substrate face-up in an multiple times to ensure reproducibility of the results.
aqueous mixture (DMF and water) of zinc chloride precursor
and the base at 80  C for 24 hours with no stirring.
The method used for the theoretical band structure
Electrical conductivity measurements calculation
Electrical conductivities of all test devices were measured using The theoretical band gap and the respective band structure were
copper as a common electrode using four-probe method. For computed using the open source Quantum ESPRESSO (QE)
this purpose, spin-coated thin lms of ZnO nanostructures only suite. QE uses plane-waves to nd solutions to a specic density
devices were fabricated. Glass substrates were cleaned prior to functional theory (DFT) setup. The computations were done

14646 | RSC Adv., 2019, 9, 14638–14648 This journal is © The Royal Society of Chemistry 2019
 View Article Online

Paper RSC Advances

using Ultra-so pseudopotentials created with LDA functions. References

Based on an optimization calculation the global cut-off energy
for the calculation was set at 100 Ry. The ZnO wurtzite CIF 1 L. E. Greene, M. Law and D. H. Tan, Nano Lett., 2005, 5, 1231.
structure le used for the computations was from the Materials 2 L. Greene, B. Yuhas and M. Law, Inorg. Chem., 2006, 45, 7535.
Project with the ID: mp-2133. The visualization soware VESTA 3 B. Postels, M. Kreye, H.-H. Wehmann, A. Bakin, N. Boukos,
was used to compute the theoretical XRD from the CIF le. With A. Travlos and A. Wang, Superlattices Microstruct., 2007, 42,
both the experimental and theoretical XRD, the unit cell's a, 425.
b and c values were changed slightly to make sure the computed 4 C. X. Xu, A. Wei, X. W. Sun and Z. L. Dong, J. Phys. D: Appl.
material matched the experimental material. The unit cell Phys., 2006, 39, 1690.
This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

values used were a ¼ b ¼ 3.22 A and c ¼ 5.16 A. 5 S. Baruah and J. Dutta, J. Sol-Gel Sci. Technol., 2009, 50, 456.
6 Z. W. Pan, Z. R. Dai and Z. L. Wang, Science, 2001, 291, 1947.
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

7 Y. K. Mishra and R. Z. Nanowire, Nanotechnology, 2004, 15, 8.

Conclusions 8 X. Y. Kong, Y. Ding, R. Yang and Z. L. Wang, Science, 2004,
In summary, the one-pot wet-chemical method that follows 303, 1348.
a versatile sol–gel synthesis, combined with low-temperature 9 P. X. Gao and Z. L. Wang, J. Am. Chem. Soc., 2003, 21, 11299.
solvothermal process, allows us to make highly crystalline 10 Z. L. Wang, J. Phys.: Condens. Matter, 2004, 16, 829.
ZnO nanostructures with controlled-morphologies. The crystal 11 C. Hsiao and S. Yu, Sensors, 2012, 12, 17007.
growth process driven by the solvent polarity, augments the 12 S. Panigrahi, S. Sarkar and D. Basak, ACS Appl. Mater.
crystal growth-oriented attachment mechanism. As evidenced Interfaces, 2012, 4, 2709.
by XRD and XPS analysis, the growth of crystal facets is gov- 13 M. Willander, O. Nur, Q. X. Zhao, L. L. Yang, M. Lorenz,
erned by the relative surface energy of crystal facets and selec- B. Q. Cao, P. J. Zúñiga, C. Czekalla, G. Zimmermann,
tive adhesion of solvents (act as surfactant in our case) onto M. Grundmann, A. Bakin, A. Behrends, M. Al-Suleiman,
crystal facets. The morphologies of the ZnO nanostructures A. El-Shaer, M. A. Che, B. W. A. Postels, N. Boukos,
formed in different solvent systems were either nanorods, or A. Travlos, H. S. Kwack, J. Guinard and S. D. D. Le,
nanoparticles, or nanoslates. The nanorods and nanoparticles Nanotechnology, 2009, 20, 332001.
show highly ordered, defect free wurtzite crystalline structures. 14 K. Sato and H. Katayama-Yoshida, Phys. B, 2001, 310, 904.
The effective optical band gap of ZnO nanostructures with 15 K. Mondal and A. Sharma, RSC Adv., 2016, 6, 94595.
respect to their morphology was able to tailor within the range 16 A. V. Kachynski, A. N. Kuzmin, M. Nyk, I. Roy and
of 3.37 eV to 3.10 eV by controlling the crystallites size and P. N. Prasad, J. Phys. Chem. C, 2008, 112, 10721.
shape. The enhanced electrical conductivities further suggest 17 H. A. Wahab, A. A. Salama, A. A. El Saeid, M. Willander,
the correlation of optical and electrical properties to crystal O. Nur and I. K. Battisha, Results Phys., 2018, 9, 809.
growth shape, size, and orientation, which easily could be tuned 18 K. Minegishi, Y. Koiwai, Y. Kikuchi, K. Yano, M. Kasuga and
through this synthesis approach of combined sol–gel and sol- A. Shimizu, Jpn. J. Appl. Phys., 1997, 36, L1453.
vothermal crystal growth process. Thus, this sol–gel synthesis, 19 P. Ghosh and A. K. Sharma, Phys. Status Solidi A, 2017, 214,
combined with solvent-driven solvothermal method offers 1600755.
simple, versatile, and environmentally friendly wet-chemical 20 H. Kato, M. Sano, K. Miyamoto and T. Yao, Jpn. J. Appl. Phys.,
method to make band gap-engineered metal oxide nano- 2003, 42, 1002.
structures, with enhanced optical and electrical properties, 21 D. Gal, G. Hodes, D. Lincot and H. W. Schock, Thin Solid
from a wide variety of metal cation precursors. Films, 2000, 362, 79.
22 G. Oskam, J. Sol-Gel Sci. Technol., 2006, 37, 161.
23 A. Iribarren, E. Hernández-Rodrı́guez and L. Maqueira,
Conflicts of interest Mater. Res. Bull., 2014, 60, 376.
24 E. A. Meulenkamp, J. Phys. Chem. B, 1998, 102, 5566.
There are no conicts to declare. 25 M. Zhang, Y. Banko and K. Wada, J. Mater. Sci. Lett., 2001, 20,
167.
Acknowledgements 26 W. Kyoungja, J. L. Ho, A. Jae-Pyoung and Y. S. Park, Adv.
Mater., 2003, 15, 1761.
This work is performed at the Joint School of Nanoscience and 27 M. Kashif, S. M. Usman Ali, M. E. Ali, H. I. Abdulgafour,
Nanoengineering, a member of the South-eastern Nanotech- U. Hashim, M. Willander and Z. Hassan, Phys. Status Solidi
nology Infrastructure Corridor (SENIC) and National Nano- A, 2012, 147, 143.
technology Coordinated Infrastructure (NNCI), supported by 28 J. Li, Y. Wu, M. Yang, Y. Yuan, W. Yin, Q. Peng, Y. Li and
the NSF (Grant ECCS-1542174). Financial support for this work X. H. Yang, J. Am. Ceram. Soc., 2017, 100, 5460.
is provided, in part, from the Joint School of Nanoscience and 29 S. Maensiri, W. Nuansing, J. Klinkaewnarong and P. Laokul,
Nanoengineering and the Office of Research, University of J. Colloid Interface Sci., 2006, 297, 578.
North Carolina at Greensboro. The authors also gratefully 30 L. L. Hench and J. K. West, Chem. Rev., 1990, 90, 33.
acknowledge Dr Kyle Nowlin for SEM and TEM support, Dr
Kristen Dellinger for XRD analysis.

This journal is © The Royal Society of Chemistry 2019 RSC Adv., 2019, 9, 14638–14648 | 14647
 View Article Online

RSC Advances Paper

31 C. B. Tay, S. J. Chua and K. P. Loh, J. Cryst. Growth, 2009, 311, 43 K. Bian, Z. Wang and T. Hanrath, J. Am. Chem. Soc., 2012,
1278. 134, 10787.
32 K. Kim, K. Utashiro, Y. Abe and M. Kawamura, Materials, 44 E. V. Shevchenko, D. V. Talapin, S. O. Brien and C. B. Murray,
2014, 7, 2522. Nature, 2006, 439, 55.
33 P. J. Skrdla, Langmuir, 2012, 28, 4842. 45 X. Peng, L. Manna, W. Yang and J. Wickham, Nature, 2000,
34 H. O. K. Kirchner, Metall. Trans., 1971, 2, 2861. 404, 59.
35 J. Zhang, F. Huang and Z. Lin, Nanoscale, 2010, 2, 18. 46 Z. Wang, X.-F. Qian, J. Yin and Z.-K. Zhu, Langmuir, 2004, 20,
36 C. Pacholski, A. Kornowski and H. Weller, Angew. Chem., Int. 3441.
Ed., 2002, 41, 1188. 47 J.-B. Xia and K. W. Cheah, Phys. Rev. B: Condens. Matter
This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

37 C. Dinh, T. Nguyen, F. Kleitz and T. Do, ACS Nano, 2009, 3, Mater. Phys., 1997, 55, 15688.
3737. 48 A. D. Yoffe, Adv. Phys., 1993, 42, 173.
Open Access Article. Published on 10 May 2019. Downloaded on 1/5/2023 8:30:01 AM.

38 T. Nguyen, C. Dinh and T. Do, Langmuir, 2009, 25, 11142. 49 G. Z. Xing, B.-Q. Yao, C. X. Cong, T. Yang, Y. P. Xie, B. H. Li
39 R. Si, Y. W. Zhang, L. P. You and C. H. Yang, Angew. Chem., and D.-Y. Shen, J. Alloys Compd., 2008, 457, 36.
2005, 117, 3320. 50 H. Zhou and Z. Li, Mater. Chem. Phys., 2005, 89, 326.
40 Y. Jun, M. F. Casula, J. Sim, S. Y. Kim, J. Cheon and 51 J. Das, S. K. Pradhan, D. R. Sahu, D. K. Mishra, S. N. Sarangi,
A. P. Alivisatos, J. Am. Chem. Soc., 2003, 125, 15981. B. B. Nayak, S. Verma and B. K. Roul, Phys. B, 2010, 405,
41 M. I. Bodnarchuk, M. V. Kovalenko, W. Heiss and 2492.
D. V. Talapin, J. Am. Chem. Soc., 2010, 132, 11967. 52 L. George, S. Sappati, P. Ghosh and R. N. Devi, J. Phys. Chem.
42 D. K. Smith, B. Goodfellow, D. Smilgies and B. A. Korgel, J. C, 2015, 119, 3060.
Am. Chem. Soc., 2009, 131, 3281. 53 D. Vogel, P. Krüger and J. Pollmann, Phys. Rev. B: Condens.
Matter Mater. Phys., 1996, 54, 5495.

14648 | RSC Adv., 2019, 9, 14638–14648 This journal is © The Royal Society of Chemistry 2019