lable at ScienceDirect

Journal of Molecular Structure 1178 (2019) 251e260

Contents lists avai

Journal of Molecular Structure

journal homepage: ht tp: / /www.elsevier .com/locate/molstruc

Solvothermal synthesis of zinc oxide nanoparticles: A combinedexperimental and theoretical study

Ankica �Sari�c a, *, Ines Despotovi�c b, Goran �Stefani�c a

a Center of Excellence for Advanced Materials and Sensing Devices, RuCer Bo�skovi�c Institute, Bijeni�cka 54, 10000 Zagreb, Croatiab Division of Physical Chemistry, RuCer Bo�skovi�c Institute, Bijeni�cka 54, 10000 Zagreb, Croatia

a r t i c l e i n f o

Article history:Received 2 August 2018Received in revised form4 October 2018Accepted 8 October 2018Available online 11 October 2018

Keywords:Nanostructured materialsOxide materialsDensity functional calculationsOptical propertiesSEMX-ray diffraction

* Corresponding author.E-mail address: asaric@irb.hr (A. �Sari�c).

https://doi.org/10.1016/j.molstruc.2018.10.0250022-2860/© 2018 Elsevier B.V. All rights reserved.

a b s t r a c t

Zinc oxide particles were synthesized from zinc acetylacetonate in the presence of triethanolamine (TEA)and various alcoholic solvent, ethanol or octanol, at 170 �C. The structural, optical and morphologicalcharacteristics of ZnO particles were monitored using X-ray powder diffraction (XRD), UVeVis and FT-IRspectroscopies and field emission scanning electron microscopy (FE-SEM). The experimental findingswere confirmed by means of DFT calculations which were obtained in nonpolar alcohol 1-octanol, as afollow up to our previous study in polar ethanol [9]. The nucleation and formation mechanism of ZnOnanoparticles is proposed considering the results obtained from a computational study of Gibbs freeenergies of ZnO�TEA molecular interactions (DG*INT) in various solvent system. The calculations revealeddifferent binding affinities which initiated the nucleation processes of ZnO nanoparticles in the presenceof alcohols of different size and polarity. The high chelating efficiency of TEA towards zinc with tetra-hedral geometry is observed. The results of X-ray diffraction size-strain analysis indicate the presence ofsize anisotropy as well as a reduction in the ZnO crystallite size with the change of solvent from ethanolto 1-octanol (

Table 1The notation and experimental conditions used for the preparation of ZnO samples.Each precipitation system contained 0.5 g of Zn(acac)2 [8]. Ageing temperature was170 �C.

Sample [TEA]/[Zn (acac)2] Ethanol/ml Octanol/ml tageing/h

E1 1:2 30 4E2 1:2 30 24E3 1:1 30 4E4 1:1 30 24O1 1:2 30 4O2 1:2 30 24O3a 1:1 30 4O4 1:1 30 24

a There was no precipitation.

A. �Sari�c et al. / Journal of Molecular Structure 1178 (2019) 251e260252

and can be considered as the initiation step of ZnO formation. Thusslowly released water molecules act as promoter of particle for-mation, including both nucleation and ZnO crystal growth. Woj-narowicz et al. [12e14] focused on the investigation of thesignificance of water in the microwave solvothermal synthesis ofundoped and Co-doped ZnO NPs. The authors emphasized that bychanging the water content in the precursor solution of hydratedzinc acetate in ethylene glycol, it is possible to control the size ofboth, undoped [12,14] and Co-doped [13] ZnO NPs. It is confirmedthat an increase in the water content in precursor resulted in theincrease in the size of obtained ZnO NPs, as a consequence of thechange of quantity of the formed crystalline nuclei of ZnO to theremaining unreacted quantity of the zinc acetate [12e14]. Woj-narowicz et al. [12e14] proposed a size control growth mechanismof undoped and Co-doped ZnO nanoparticles.

Generally, nonaqueous routes can be distinguished as surfactantor solvent controlled preparation routes [10]. Many researchers[3,15e17] investigated the influence of a solvent on the ZnO for-mation in surfactant-free nonaqueous synthesis routes. Ambro�zi�cet al. [3] proposed the formation mechanism of ZnO nanoparticlesproduced by starting from zinc acetylacetonate hydrate and 1-butanol or isobutanol. Tonto et al. [15] investigated the reactionmechanism of ZnO nanorods synthesized using the solvothermalesterification between zinc acetate and various alcohols in a seriesfrom 1-butanol to 1-decanol. The authors emphasized the inter-esting relationship between boiling point of the alcohol used andaspect ratio of ZnO nanorods, which increased from 1.7 to 5.6 whenthe alcohol was changed from 1-butanol to 1-decanol. On the otherside, Bilecka et al. [18] studied the formation of ZnO nanoparticlesin the presence of aromatic alcohol via a non-aqueous route by theuse of benzyl alcohol. The formation mechanism of nanocrystallineZnO particles from the reaction of zinc acetylacetonate with NaOHin ethanol was likewise investigated by Inubushi et al. [19].

Salavati-Niasari research group [20e25] investigated the influ-ence of synthesis procedure on the formation and properties of ZnOnanomaterial and composites. Along with other physical andchemical properties the size and morphology of ZnO nanoparticlescan be modified by using different routes of the synthesis. ZnOnanoflowers were synthesized from zinc oxalate without usingadditives by a fast microwave-assisted polyol method [23]. Also,Salavati-Niasariʼs research group reported synthesis of ZnO nano-particles by solid-state reaction [20e22]. Solvent-free syntheseshave recently attracted considerable attention in the field of greenchemistry as novel and simple mechano-chemical solid-solidmethod [22]. Also, the solid-state thermal decomposition of zincacetylacetonate was already utilized in the production of ZnOnanofibers and nanoparticles [24]. Notwithstanding the simplicityof this method, the mechanism of thermal decomposition of zincacetylacetonate obtained by the sublimation process of Zn(acac)2powder is complex. This method can be developed to preparevarious morphologies of metal oxides [24]. ZnO morphology has asignificant influence on the optical properties of ZnO, as well as onthe other semiconductor photocatalyst material [25].

Here we present new results obtained by the theoretical study,by means of quantum chemical calculations at the density func-tional theory level, related to ZnO nanomaterial synthesized in 1-octanol following our previously reported procedure [8]. Thisresearch is a follow up to our previous combined experimental andtheoretical study of ZnO nanomaterial obtained in ethanol [9]. In aprevious paper [9] we studied and compared the impact of a ver-satile family of ethanolamines (MEA and TEA) on the microstruc-tural properties of the ZnO particles prepared in ethanol. The aim ofpresent combined experimental and theoretical study is to betterunderstand on amolecular level themechanismwhich initiated theZnO nucleation process continued by the growth of nanoparticles

in the presence of alcohols of different size and polarity and TEA, aswell as to improve the predictability of formation of nanoparticlesand their aggregation.

2. Experimental

2.1. Materials and synthesis

Zinc acetylacetonate monohydrate (Zn(C5H7O2)2$H2O; AlfaAeser®), triethanolamine (C6H15NO3; Fisher Chemical), absoluteethanol (C2H5OH; J. T. Baker) and 1-octanol (CH3(CH2)7OH; SigmaAldrich), all of analytical purity, were used for the preparation ofsamples. Commercial ZnO was supplied by Ventron.

Sample notation and the experimental conditions used for theirpreparation are given in Table 1 following our previously reportedprocedure [8]. In a typical synthetic procedure triethanolamine(TEA) was first dissolved in alcohol. Alcohols used in this researchwere ethanol (samples E) and 1-octanol (samples O). Then Zn(a-cac)2$H2O was added to this solution and the molar ratio of TEA toZn(acac)2$H2O was adjusted to 1:2 and 1:1 (TEA/alcohol precursorsolutions). Each precipitation system contained 0.5 g of Zn(acac)2.Different amounts of TEA were dissolved in alcohol. Thus preparedtransparent precursor solution was autoclaved at 170 �C for 4 or24 h. A 50mL Teflon-lined stainless steel autoclave was used. Afterautoclaving the precipitates were separated from supernatants bycentrifugation, washed several times with ethanol and dried.

2.2. Measurements and characterization

Field Emission Scanning Electron Microscope (FE-SEM) modelJSM-7000F, manufactured by Jeol Ltd. was used to examine themorphology and size of the particles. UV/Vis absorption spectrawere recorded with a Shimadzu UV/Vis/NIR spectrometer equippedwith an integrated sphere (model UV-3600). FT-IR spectra wererecorded using a Perkin Elmer spectrometer (model 2000). Thestructure of samples was determined at room temperature by X-raydiffraction (XRD) using Italstructures X-ray powder diffractometer(APD 2000, Cu-Ka radiation, graphite monochromator, scintillationdetector).

2.3. Sizeestrain line-broadening analysis

Sizeestrain line-broadening analysis of the ZnO samples wasperformed using the results of Williamson-Hall analysis [26]. Thevalues of volume-averaged domain size (Dv) and the upper-limits ofmicrostrains (e) were estimated from the results of Williamson-Hall analysis by using equation:

A. �Sari�c et al. / Journal of Molecular Structure 1178 (2019) 251e260 253

�b cos q

l

�¼ K

Dv��4e sinq

l

�; [1]

where l is wavelength, b stands for the physical broadening ofdiffraction line and K is constant close to 0.9. The b values wereobtained by convolution-fitting approach (program SHADOW [27])in which the instrumental profile (diffraction lines of well-crystalline zincite powder) is convoluted with a refinable Voigtfunction to fit the observed profile.

2.4. Computational methods

All calculations were performed by means of quantum chemicalcalculations at the density functional theory (DFT) level using theGaussian 09 program (revision D1) [28]. The M05-2X functional[29] designed by the Truhlar's group has been chosen. It providesvery accurate thermodynamic parameters, being particularly suc-cessful in nonbonding interactions treatment [29e31], especially inreproducing geometries, dipole moments and homolytic bond en-ergies in various zinc complexes [32,33]. The 6-31 þ G(d,p) þ LANL2DZ mixed basis set has been utilized. ThePople's 6-31 þ G(d,p) double-x basis set was chosen for N, O, Hatoms and the LANL2DZ basis (LANL2 pseudopotential for innerelectrons and its associated double-x basis set (DZ)) was used forthe transition-metal (Zn) atoms [34]. This gave rise to the M05-2X/6-31þG(d,p)þ LANL2DZmodel utilized for geometry optimizationwhich has been frequently used for studies of transition-metalcontaining systems. The geometric structures of the moleculeswere optimized by minimizing energies with respect to allgeometrical parameters without imposing any molecular symme-try constraints and using a tight convergence condition. The Bernyalgorithm using redundant internal coordinates was employed.Frequency calculations were made under the harmonic approxi-mation on all the optimized structures at the same level of theorywith no scaling in order to confirm that the structures correspondto the true minima, meaning that no imaginary frequencies werepresent, as well as to extract thermal Gibbs free energy corrections.The final single point energies were obtained using a highly flexible6-311þþG(2df,2pd) basis set for the N, O, H atoms, while the sameLANL2DZ ECP type basis set for zinc atoms was employed. The self-consistent field (SCF) calculations were conducted under a tightcondition imposing the threshold value of 10�8 hartree to totalenergy difference during the iteration process. The integration gridin self-consistent field calculations was set to FineGrid having 75radial shells and 302 angular points per shell. The 2-electron in-tegral accuracy was set to 10�13. The FoFCou algorithm withNoSymm option was used. All geometry optimizations, frequencycalculations and single point energy evaluations were conducted bytaking solvent effects into account. To evaluate the bulk solventeffects (1-octanol, ε¼ 9.86; ethanol, ε¼ 24.85) the implicit SMDpolarizable continuum solvation model [35] has been employed. Itrepresents a very practical approach to simulate the solvationenvironment and determine the effect of the medium on thestructure and stability of solutes in a solution.

One of major difficulties in modelling the formation process ofZnOeTEA monomers, as well as their association into oligomers, isthe correct sampling of the conformational space of molecules. Dueto many different monomer structures, as well as their associatestructures, optimization for a large number of different startingstructures had to be performed in order to find the global mini-mum. In practice, different ways are utilized to create such startingstructures. They are usually obtained by a low-level quantum-chemical method, or by an empirical force-field combined withmolecular dynamics or Monte Carlo techniques, or using generic

algorithms. Starting structures are also frequently build fromliterature data (X-ray data), or manually sampled by using chemicalintuition. In this work the latter approach was used to generate thestarting structure of the ZnOeTEA monomer as well as other olig-omers. The conformational space was manually sampled for theZnOeTEA monomer, taking into account various donor andacceptor sites of the ZnO and TEA molecule. The numbers of con-figurations were optimized, then the most stable ones wereselected and further used as a starting point for building the di-mers. The two most stable dimers were then associated directlyinto tetramers.

The Gibbs free energy of interaction,DG*INT, was computed usingthe supramolecular approach according to the simple formula:

DG*INT;AB ¼ G�AB � G�A � G�B (1)

as the difference between the total free energy (G*AB) of the asso-ciates (ZnOeTEA, or their oligomers), and the sum of the total freeenergies (G*A þ G*B) of the associating units A and B. The total freeenergy of the species in the liquid was computed using theexpression

G*X ¼ ETotsoln þ DG*VRT;soln (2)

where the term ETotsoln corresponds to the basic energy of a densityfunctional theory calculation using the SMD model, while the€AG*VRT,soln term encompasses vibrational, rotational and trans-lational contribution to the solution free energy, being computedby applying the ideal gas partition functions to the frequenciescalculated in the dielectric medium and the 1M standard state. Amore negative value of the binding energy implied the more stableformed species. No BSSE correction to the binding energies hasbeen applied.

The topological analysis of the charge density distribution usingthe Bader's quantum theory of atoms in molecules (QTAIM) [36]was performed by employing AIMALL software package [37] usingthe SMD/M05-2X/6-31 þ G(d,p) þ LANL2DZ wave function ob-tained from optimization. Within the QTAIM analysis the electrondensity was analyzed for twomajor characteristics including (a) theexistence of critical points (CPs), which are special points whereelectron density exhibits maximum, minimum, or a saddle point inspace, and (b) for the bond paths [38], the line of maximum elec-tron density connecting two interacting atoms in the energeticminimum structure which indicates that the two atoms arebonded. The point of the minimum value of electron density alongthat line is called the bond critical point (BCP) and the value oftopological parameters, such as electron density r(rc), Lap-lacianV2r(rc), electronic kinetic energy G(rc), electronic potentialenergy density V(rc), total energy density H(rc) at that point explainthe features of interatomic interactions. V2r(rc)< 0 indicates thatthe charge density is locally concentrated, while V2r(rc)> 0 in-dicates that the charge density is locally depleted. The chemicalbond nature can be described qualitatively concerning the signsand values of the electron density Laplacian V2~n(rc) and of theelectron energy density H(rc) at the corresponding bond criticalpoint in accordance with the following criteria. The interactionscharacterized by V2r(rc)< 0 and H(rc)< 0 point to a shared inter-action, i.e., weakly polar and nonpolar covalent bonds. On the otherhand, the interaction characterized by V2r(rc)> 0 and H(rc)> 0suggests the closed shell interactions such as weak hydrogenbonds, van der Waals interactions and ionic bonds. The interme-diate interactions which include strong hydrogen bonds and mostof the coordinate bonds are characterized by V2r(rc)> 0 andH(rc)< 0 [39,40]. A very high negative value of the V2r(rc) is anindication of a strong covalent bond, while a high positive value

Table 2Estimated values of the volume averaged domain sizes (Dv) and upper limits ofmicrostrains (e) and the corresponding values in the directions parallel to the c-ax ofzincite lattice as determined from the results of diffraction line broadening [8](Williamson-Hall analysis).

Sample Dv/nm e� 103 Dvkc-ax/nm e� 103kc-axE1 9 (1) 0.6 (2) 19 (2)

Fig. 2. FE-SEM images of samples: (a) O4, (b) E4.

Fig. 3. X-ray diffraction patterns of samples O2, O4, E2 and E4.

Fig. 4. UVeVis spectra of samples O2, E2, and reference ZnO. Inset images show thecorresponding SEM images at the same magnification (10000� ) and calculated bandgap values.

A. �Sari�c et al. / Journal of Molecular Structure 1178 (2019) 251e260 255

oligomers in the system, which shortens the turbidity time. Fromthe relationship between the dielectric constant and turbidity timeit can be concluded that the reaction participating units are smalloligomers rather than monomers [48]. Sadasivan et al. [49] alsoobserved similar results. The authors concluded that the sterichindrance may be predominant factor for lower alcohols, weakerhydrogen bonding predominant one for the higher alcohols.Namely, a weaker hydrogen bonding between solvent moleculesand nucleophile enhance the rate of hydrolysis, while the sterichindrance of an alkyl group expressed through higher solvent vis-cosity for higher alcohols slow down the rate of hydrolysis [49]. Theauthors concluded that the final particle size depend on the solventpolarity [48,49]. Triethanolamine (TEA) molecules play importantroles as complexing, assembling and structure directing ligands inthe precursor solution probably as coordinated ions [Zn(TEA)x]2þ or[ZnO(TEA)x] (besides remaining (Zn(OH)2, [(Zn(OH)4]2- and Zn2þ)which limits the existence of free zinc ions and controls the ZnOnuclei formation.

3.2. UVeVis and FT-IR spectroscopies

The optical absorbance spectra of ZnO particles are shown inFig. 4. Fig. 4 shows the UVeVis spectra of selected samples and thereference ZnO material, their corresponding particle morphologies(inset) and calculated band gap values. As can be seen, the as

prepared powder, as well as reference ZnO, had no optical ab-sorption in the visible region, but there was good absorption nearlyin thewhole UV region. The UVeVis spectra of ZnO samples, as wellas commercial ZnO, are characterized by broad and intensive ab-sorption between 400 and 200 nm. The band gap values were ob-tained using the procedure described by Dharma and Pisal [50].One can see from Fig. 4 that the as prepared sample in octanol (O2)solution show a UV absorption edge which is blue-shifted withrespect to the absorbance of reference ZnO material. It is attributed

A. �Sari�c et al. / Journal of Molecular Structure 1178 (2019) 251e260256

to the size induced change of ZnO band gap energy. The relation-ship between the particle size and optical properties of ZnOnanoparticles was recently reported [51]. ZnO morphology has asignificant influence on the optical properties of ZnO, as well as onthe other semiconductor photocatalyst material. Among thevarious semiconductor photocatalyst material nanostructuredZn3V2O8 has attracted significant consideration for degradation ofanionic and cationic dyes under UV irradiation in a short time [25].

Fig. S3 shows characteristic parts of the FT-IR spectra of all ZnOsamples. The right panel shows the spectra of samples synthesizedin the presence of octanol (O1, O2 and O4), whereas the left panelshows the FT-IR spectra of samples synthesized in the presence ofethanol (E1, E2, E3 and E4). The FT-IR bands between 540 and397 cm�1 with two poorly resolved bands are characteristic of ZnO.Besides, the FT-IR spectra of ZnO samples are characterized by thebands above 600 cm�1 due to the adsorbed water and organicmolecules. The IR band at ~1630 cm�1 can be attributed to thebending vibration of H2O molecules adsorbed on ZnO particles. TheFT-IR bands at 1582 to 1619 cm�1, as well as the bands at 1385 to1414 cm�1, can be attributed to the stretching vibration of chem-isorbed C]O bands originating from acetylacetonate [8]. The pre-sented FT-IR spectra indicate chemisorptions of carbonyl andcarbonate groups on the surface of ZnO nanoparticles.

3.3. The mechanism of ZnO particle formation

The formation of final ZnO particles can be considered in severalsequential steps: (1) monomer formation, (2) nucleation, (3) crystalgrowth, and (4) possibly aggregation of ZnO nanoparticles. Thepresence of triethanolamine in alcoholic solvent of different sizeand polarity, had a significantly different influence on the all steps,nucleation as well as on the crystal growth and aggregation of ZnOnanoparticles. Triethanolamine combines the properties of aminesas weak bases and alcohols with three hydroxyl chains. Due tospecial ability of TEA to participate in reactions common to bothgroups, the spontaneous nucleation and growth of ZnO nano-particles was limited.

On the basis of both microstructural and theoretical studies, thenucleation and growth mechanism of ZnO nanoparticles is pro-posed. Based on the results obtained by means of quantumchemical calculations at the density functional theory level, it wasestablished that the nucleation kinetics and ZnO nanoparticlesgrowth, in the presence of TEA as a particles growth modifier andalcoholic solvent of different size and polarity, are stronglydependent on the properties of alcohol. The nucleation and for-mationmechanism of ZnO nanoparticles was proposed consideringthe results obtained from a computational study of Gibbs free en-ergies of ZnO�TEAmolecular interactions (DG*INT) in 1-octanol andour previously reported results in ethanol [9]. The calculationsrevealed different binding affinities which initiated the nucleationprocesses of ZnO into nanoparticles in the presence of alcohols ofdifferent size and polarity. These processes involve noncovalentinteractions such as coordinate bonding, hydrogen bonding, vanderWaals forces, electrostatic forces, etc. Here, as a follow-up to ourprevious investigations [9] performed in the presence of polarethanol solvent, we systematically studied and compared all non-covalent interactions involved in the processes of nucleation andZnO nanoparticle growth in the presence of nonpolar octanolsolvent.

It is interesting to relate the structural motif for bothmost stableisolated ZnOeTEAmonomers obtained in the presence of one polar[9] or the other nonpolar alcohol, as well as calculated values ofGibbs free energies of interactions for their formation (Fig. 5,Table 3). It is noticed that the structural motif for ZnOeTEAmonomer (Fig. 5a) obtained in the presence of 1-octanol doesn't

correspond to the motif of ZnOeTEA monomer obtained in ethanol[9] (Fig. 5b). Namely, when a ZnOeTEA monomer was formed, allthree TEA hydroxyl chains could be attached to the ZnOmolecule indifferent ways depending on the type of alcohol used (shown inFig. 5 as ZnOeTEA(octanol) or ZnOeTEA(ethanol)).

To sum up, in themost stable conformation of isolated ZnOeTEAmonomers obtained in both polar ethanol or nonpolar octanol, thestrongest binding was accomplished via a similar coordinate bond,initiated with a characteristic ionicedipolar interaction betweenhydrogen atoms from the OH group in TEA and oxygen in ZnO,which resulted in hydrogen atom transferring from the OH groupon oxygen to ZnO. As a consequence of this process a lone electronpair on the oxygen atom from the OH group was involved in a newexceptionally strong coordinate bondwith zinc (ZnOeTEA(ethanol)[9]: dZneO¼ 1.948 Å, EZneO¼�38.0 kcalmol�1; ZnOeTEA(octanol):dZneO¼ 1.954 Å, EZneO¼�37.5 kcalmol�1). The ZneO bond isattributed to an intermediate type of interaction according toV2r(rc)> 0 and H(rc)< 0. It is more of an ionic type as indicated bythe higher positive value of V2r(rc) compared with a correspondingvalue in the ZneN bond. Namely, in the most stable ZnOeTEAmonomers in both ethanol and 1-octanol a free electron pair on thenitrogen atom was involved in the coordinate bond with zinc(ZnOeTEA(ethanol) [9]: dZneN¼ 2.257 Å, EZneN¼�17.9 kcalmol�1;ZnOeTEA(octanol): dZneN¼ 2.246 Å Å, EZneN¼ - 18.6 kcalmol�1).The critical point of the ZneN bond was characterized by thepositive values of electron density Laplacian V2r(rc)> 0 and bynegative value of electron energy density H(rc)< 0 indicating thistype of interaction also as intermediate which is a feature of thecoordinate bond. However, the most important binding responsiblefor additional stabilization of the most stable ZnOeTEA(octanol)monomer in comparison to ZnOeTEA(ethanol) monomer wasaccomplished via a newly coordinate bond where a free electronpair on the oxygen atom from the second remaining free TEA hy-droxyl group was involved in bonding with zinc (dZneO¼ 2.2231 Å,EZneO¼�16.7 kcalmol�1), beingmore stabilized by strong OeH/Ohydrogen bond between the hydrogen atom from remaining freeTEA hydroxyl group and oxygen in ZnO (dO … H¼ 1.563 Å, EO …H¼�18.9 kcalmol�1, see Fig. 5a). Differently, the most stableZnOeTEA(ethanol) monomer was stabilized by the weaker bifur-cated OeH/O hydrogen bonds between the hydrogen atoms fromboth remaining free TEA hydroxyl groups and oxygen in ZnO (dO …H¼ 1.644 Å, EO … H¼�13.7 kcal mol�1 and dO … H¼ 1.677 Å, EO …H¼�12.1 kcalmol�1, see Fig. 5b [9]). It is important to emphasizethat the most stable ZnOeTEA monomer obtained in the presenceof octanol possess stronger coordination ability which couldreplace weaker hydrogen bonds obtained in ethanol media. Obvi-ously, the synergistic effect of all the above mentioned noncovalentinteractions was present. The formation of such completelyenclosed structure of ZnOeTEA(octanol) monomer in nonpolaralcohol was a consequence of the exceptional flexibility of all threeof TEA hydroxyl chains compared to ethanol media. The calculatedvalue of the Gibbs free energy interaction obtained in nonpolaroctanol suggested that the amount of especially stableZnOeTEA(octanol) monomer with DG*INT¼�62.50 kcalmol�1 wasfound to be the main controlling parameters for both nucleationand growth of ZnO nanoparticles, being 10.5 kcalmol�1 more stablethan ZnOeTEA(ethanol) monomer withDG*INT¼�51.50 kcalmol�1 [9]. Moreover, when all is considered,the tetrahedral geometry in ZnOeTEA(octanol) monomer whichincreases zinc stability, as well as the DG*INT of its formation, onemight conclude that in the presence of nonpolar octanol monomersfound to imply most important parameters which initiated thenucleation processes of ZnO nanoparticles.

Furthermore, the formation of most stable dimer from the as-sociation of most stable monomers provide evidence of coordinate

Fig. 5. The most stable structures of (a) ZnOeTEA monomers in 1-octanol and (b) ZnOeTEA monomers in ethanol (selected values of bond distances in Å and bond energies in kcalmol�1).

Table 3Formation of the most stable ZnOeTEA monomersa and their oligomersb, andZnOeTEAeZnO speciesc in 1-octanol and ethanol [9]. Standard state (1M) free en-ergies of interaction in 1-octanol and ethanol DG*INT computed by using the SMDsolvation model at the M05-2X/6-311þþG(2df,2pd) þ LANL2DZ//M05-2X/6-31 þ G(d,p) þ LANL2DZ level of theory (in kcal mol�1).

species ZnOeTEA(octanol)

DG*INT species ZnOeTEA(ethanol)

DG*INT [9]

ZnOeTEA (octanol) �62.5 ZnOeTEA (ethanol) �51.5(ZnOeTEA)2(octanol) �10.5 (ZnOeTEA)2(ethanol) �11.5ZnOeTEAeZnO(octanol) �63.2 ZnOeTEAeZnO(ethanol) �28.3(ZnOeTEA)4(octanol) 8.2 (ZnOeTEA)4(ethanol) 5.8

a According to the reaction: ZnO þ TEA/ ZnOeTEA.b According to the reaction for dimer: ZnOeTEA þ ZnOeTEA / (ZnOeTEA)2 and

for tetramer: (ZnOeTEA)2 þ (ZnOeTEA)2 / (ZnOeTEA)4.c According to the reaction for ZnOeTEAeZnO species: ZnOeTEA þ ZnO /

ZnOeTEAeZnO.

Fig. 6. The most stable structures of (a) (ZnOeTEA)2 dimers in 1-octanol and (b) (ZnOeTEAmol�1).

A. �Sari�c et al. / Journal of Molecular Structure 1178 (2019) 251e260 257

bonding via free electron pair on oxygen atoms with zinc atoms,having the same structural motif for the both most stable(ZnOeTEA)2(octanol) and (ZnOeTEA)2(ethanol) dimers, havingclosely followed the values of Gibbs energies,DG*INT¼�10.5 kcalmol�1 and DG*INT¼�11.5 kcalmol�1 [9] (Fig. 6,Table 3). Thanks to high chelating efficiency of triethanolaminetowards zinc, a same tetrahedral geometry which increased itsstability was noticed in both (ZnOeTEA)2(octanol) (Fig. 6a) and(ZnOeTEA)2(ethanol) (Fig. 6b, [9]) dimers. The advantages of as-sociation of a ZnOeTEA(ethanol) monomers during the formationof (ZnOeTEA)2(ethanol) dimer should be emphasized in compari-son to the association of a ZnOeTEA(octanol) monomers into dimerdue to higher energetic requirements during the process ofdimerization in octanol media, which involve at first an additionalstep of releasing one of the coordinate bonds in especially stable

)2 dimers in ethanol (selected values of bond distances in Å and bond energies in kcal

A. �Sari�c et al. / Journal of Molecular Structure 1178 (2019) 251e260258

monomeric structure. On the other hand, the possibility of theprocess of releasing one of the coordinate bonds in especially stableZnOeTEA(octanol) monomer, as well as its potential dimerization,is also expected to be dependent on the access of other species withstronger coordination ability, as in the example at the excessamount of zinc salt. Namely, it is important to note the large ca-pacity of the ZnOeTEA(octanol) monomer to attract other ZnOmolecule when it's in excess ([TEA]/[Zn(acac)2]¼ 1:2]), via coor-dinate bonding of free electron pairs on oxygen atoms with atomszinc, which was accompanied by the formation ofZnOeTEAeZnO(octanol) species. The tetrahedral structural motif ofZnOeTEAeZnO(octanol) (Fig. 7a) species corresponds to the motifof the (ZnOeTEA)2(octanol) dimer (Fig. 6a). Interestingly, supple-mentary adding of ZnO molecule to the isolated ZnOeTEA(octanol)monomer increased the stability of the system yielding the espe-cially stable species ZnOeTEAeZnO(octanol) withDG*INT¼�63.2 kcalmol�1, being 1.2 kcalmol�1 even more stablethan the existing ZnOeTEA(octanol) monomer and 52.7 kcalmol�1

more stable than the (ZnOeTEA)2(octanol) dimer withDG*INT¼�10.5 kcal mol�1 (Table 3). However, as shown in Fig. 7,the structural motif for most stable ZnOeTEAeZnO(octanol) spe-cies doesn't correspond to the motif of ZnOeTEAeZnO(ethanol)species, having very big difference in values of Gibbs energies,DG*INT¼�63.2 kcalmol�1 and DG*INT¼�28.3 kcalmol�1 [9]. Thehigher chelating efficiency of TEA towards zinc with tetrahedralgeometry is observed with the change of solvent from ethanol tooctanol. Despite the mentioned differences related to octanol me-dia, the most stable species of ZnOeTEAeZnO(ethanol) withDG*INT¼�28.3 kcalmol�1, being 16.8 kcalmol�1 more stable thanthe existing (ZnOeTEA)2(ethanol) dimer withDG*INT¼�11.5 kcalmol�1 was found to imply important processparameters in controlling the ZnO nanoparticles nucleation at theexcess amount of zinc salt in ethanol media [9]. It is interesting tonote that the precipitation of ZnO was delayed for at least 4 h (itoccurred after 24 h of ageing) at the mole ratio [TEA]/[Zn(a-cac)2]¼ 1:1] solely in octanol media, indicating the strong impactof three TEA chains in such nonpolar media [8]. It seems that it isnot possible to predict mechanisms responsible for nucleation andpreferential growth of ZnO nanoparticles only from the knowledgeof the most thermodynamically stable prenucleation intermediate.

The observed results point to a different formation mechanismof ZnOeTEA intermediateswhich initiated the nucleation processesof ZnO nanoparticles in the presence of alcohols of different size

Fig. 7. The most stable structures of (a) ZnOeTEAeZnO species in 1-octanol and (b) ZnOeTEkcal mol�1).

and polarity and at differentmolar ratio [TEA]/[Zn(acac)2]. Based onthe results obtained by the theoretical study which was in excellentagreement with the experimental microstructural study, it wasestablished that the presence of solvent of different size and po-larity had a strong impact on the ZnO nucleation process as well ascrystal growth. To sum up, the use of 1-octanol as a nonpolar sol-vent leads to ZnO crystallites smaller than those obtained usingpolar ethanol. The results of X-ray diffraction line-broadeninganalysis show that for the same experimental conditions ZnOcrystallites prepared in octanol solutions in the presence of smallamount of triethanolamine ([TEA]/[Zn(acac)2]¼ 1:2]) appeared tobe significantly smaller (~5 nm) than in ethanol (~9 nm) both ob-tained in same experimental conditions [8]. This could be attrib-uted to the spontaneous formation of ZnOeTEA intermediates(DG*INT0). Here, only the free hydroxylgroups from one (ZnOeTEA)2 dimer could be attracted to free hy-droxyl groups of other neighbouring (ZnOeTEA)2 dimers by severalbifurcated OeH/O hydrogen bonds, giving most stable tetramers

AeZnO species in ethanol (selected values of bond distances in Å and bond energies in

A. �Sari�c et al. / Journal of Molecular Structure 1178 (2019) 251e260 259

of (ZnOeTEA)4(ethanol) (Figure S4b) with DG*INT¼ 5.8 kcal mol�1[9], and (ZnOeTEA)4(octanol) (Figure S4a) with DG*INT¼ 8.2 kcalmol�1. Due to hydrogen type of bonding, the advantages of asso-ciation of (ZnOeTEA)2 dimers during the formation of (ZnOeTEA)4tetramers should be emphasized in polar ethanol related tononpolar octanol.

4. Conclusions

In this research paper in a combined experimental and theo-retical study we presented impacts of alcohols of different size andpolarity and different molar ratio [TEA]/[Zn(acac)2] on the nucle-ation and growth mechanism of ZnO nanoparticles. Here, we pre-sent new results obtained by the theoretical study, bymeans of DFTcalculations, related to ZnO nanomaterial synthesized in nonpolaralcohol 1-octanol, as a follow up to our previous study in polarethanol [9]. The main part of the manuscript comprises a theoret-ical study based on DFT calculations which are in agreement withthe experimental microstructural study.

Based on the results of, both microstructural and theoreticalstudies, the nucleation and growth mechanism of ZnO nano-particles is proposed. The results obtained by means of quantumchemical calculations at the density functional theory level are inexcellent agreement with the experimental microstructural study.It was established that the kinetics of nucleation and ZnO nano-particles growth in the presence of TEA as particles growth modi-fier and alcoholic solvent of different size and polarity are stronglydependent on the properties of alcohol. The observed results pointto a different formation mechanism of ZnOeTEA intermediateswhich initiated the nucleation processes of ZnO nanoparticles inthe presence of alcohols of different size and polarity and atdifferent molar ratio [TEA]/[Zn(acac)2]. The calculations revealeddifferent binding affinities which initiated the nucleation processesof ZnO nanoparticles in the presence of alcohols of different sizeand polarity. It was established that the most stable ZnOeTEAmonomer obtained in the presence of octanol possess strongercoordination ability which could replace weaker hydrogen bondsobtained in ethanol media. The higher chelating efficiency of TEAtowards zinc with tetrahedral geometry is observed with thechange of solvent from ethanol to octanol. The calculated value ofthe Gibbs free energy interaction obtained in nonpolar octanolsuggested that the amount of especially stable ZnOeTEA(octanol)monomer with DG*INT¼�62.50 kcalmol�1 was found to be themain controlling parameters for both nucleation and growth of ZnOnanoparticles, being 10.5 kcalmol�1 more stable thanZnOeTEA(ethanol) monomer [9]. The formation of such completelyenclosed structure of ZnOeTEA(octanol) monomer in nonpolaralcohol was a consequence of the exceptional flexibility of all threeof TEA hydroxyl chains compared to ethanol media.

It is interesting to note that the precipitation of ZnOwas delayedfor at least 4 h (it occurred after 24 h of ageing) at the mole ratio[TEA]/[Zn(acac)2]¼ 1:1] solely in octanol media, indicating thestrong impact of three TEA chains in suchmedia. On the other hand,the large capacity of the ZnOeTEA(octanol) monomer to attractother ZnO molecule when it's in excess ([TEA]/[Zn(acac)2]¼ 1:2])was observed. This was accompanied by the formation of theespecially stable ZnOeTEAeZnO(octanol) species withDG*INT¼�63.2 kcalmol�1. The big difference in values of Gibbs freeenergies of ZnOeTEA molecular interactions revealed that the useof octanol as solvent leads to more spontaneous formation ofZnOeTEA monomers, as well as ZnOeTEAeZnO species in com-parison to ethanol solvent, which results in a higher nucleation rateof ZnO nuclei. The use of octanol as a nonpolar solvent leads to ZnOcrystallites smaller than those obtained using polar ethanol(

A. �Sari�c et al. / Journal of Molecular Structure 1178 (2019) 251e260260

Sci. Technol. 37 (2006) 157e160, https://doi.org/10.1007/s10971-005-6620-3.[18] I. Bilecka, P. Elser, M. Niederberger, Kinetic and thermodynamic aspects in the

microwave-assisted synthesis of ZnO nanoparticles in benzyl alcohol, ACSNano 3 (2009) 467e477, https://doi.org/10.1021/nn800842b.

[19] Y. Inubushi, R. Takami, M. Iwasaki, H. Tada, S. Ito, Mechanism of formation ofnanocrystalline ZnO particles through the reaction of [Zn(acac)2] with NaOHin EtOH, J. Colloid Interface Sci. 200 (1998) 220e227, https://doi.org/10.1006/jcis.1997.5354.

[20] M. Salavati-Niasari, F. Davar, Z. Fereshteh, Synthesis and characterization ofZnO nanocrystals from thermolysis of new precursor, Chem. Eng. J. 146 (2009)498e502. https://doi.org/10.1016/j.cej.2008.09.042.

[21] M. Salavati-Niasari, F. Davar, A. Khansari, Nanosphericals and nanobundles ofZnO: synthesis and characterization, J. Alloys Compd. 509 (2011) 61e65.https://doi.org/10.1016/j.jallcom.2010.08.060.

[22] A.K. Babaheydari, M. Salavati-Niasari, A. Khansari, Solvent-less synthesis ofzinc oxide nanostructures from Zn(salen) as precursor and their opticalproperties, Particuology 10 (2012) 759e764. https://doi.org/10.1016/j.partic.2012.03.006.

[23] N. Mir, M. Salavati-Niasari, F. Davar, Preparation of ZnO nanoflowers and Znglycerolate nanoplates using inorganic precursors via a convenient rout andapplication in dye sensitized solar cells, Chem. Eng. J. 181e182 (2012)779e789. https://doi.org/10.1016/j.cej.2011.11.085.

[24] F. Soofivand, M. Salavati-Niasari, F. Mohandes, Novel precursor-assisted syn-thesis and characterization of zinc oxide nanoparticles/nanofibers, Mater. Lett.98 (2013) 55e58. https://doi.org/10.1016/j.matlet.2013.01.129.

[25] F. Mazloom, M. Masjedi-Arani, M. Ghiyasiyan-Arani, M. Salavati-Niasari, Novelsodium dodecyl sulfate-assisted synthesis of Zn3V2O8 nanostructures via asimple route, J. Mol. Liq. 214 (2016) 46e53. https://doi.org/10.1016/j.molliq.2015.11.033.

[26] G.K. Williamson, W.H. Hall, Acta Metall. 1 (1953) 22e31.[27] Modern powder diffraction, in: D.L. Bish, J.E. Post (Eds.), Reviews in Miner-

alogy, Mineralogical society of America, Washington, 1989.[28] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,

J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson,H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino,G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers,K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam,M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski,R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador,J.J. Dannenberg, S. Dapprich, A.D. Daniels, €O. Farkas, J.B. Foresman, J.V. Ortiz,J. Cioslowski, D.J. Fox, Gaussian 09, Revision D.01, Gaussian, Inc., WallingfordCT, 2009.

[29] Y. Zhao, N.E. Schultz, D.G. Truhlar, Design of density functionals by combiningthe method of constraint satisfaction with parametrization for thermo-chemistry, thermochemical kinetics, and noncovalent interactions, J. Chem.Theor. Comput. 2 (2006) 364e382, https://doi.org/10.1021/ct0502763.

[30] Y. Zhao, D.G. Truhlar, Density functionals with broad applicability in chem-istry, Acc. Chem. Res. 41 (2008) 157e167, https://doi.org/10.1021/ar700111a.

[31] Y. Zhao, D.G. Truhlar, Density functional theory for reaction energies: test ofmeta and hybrid meta functionals, range-separated functionals, and otherhigh-performance functionals, J. Chem. Theor. Comput. 7 (2011) 669e676,https://doi.org/10.1021/ct1006604.

[32] E.A. Amin, D.G. Truhlar, Zn coordination chemistry: development of bench-mark suites for geometries, dipole moments, and bond dissociation energiesand their use to test and validate density functionals and molecular orbitaltheory, J. Chem. Theor. Comput. 4 (2008) 75e85, https://doi.org/10.1021/

ct700205n.[33] C.J. Cramer, D.G. Truhlar, Density functional theory for transition metals and

transition metal chemistry, Phys. Chem. Chem. Phys. 11 (2009) 10757e10816,https://doi.org/10.1039/b907148b.

[34] W.R. Wadt, P.J. Hay, Ab initio effective core potentials for molecular calcula-tions. Potentials for main group elements Na to Bi, J. Chem. Phys. 82 (1985)284e298, https://doi.org/10.1063/1.448800.

[35] A.V. Marenich, C.J. Cramer, D.G. Truhlar, J. Phys. Chem. B 113 (2009)6378e6396.

[36] R.F.W. Bader, Atoms in Molecules: a Quantum Theory, Oxford UniversityPress, Oxford, New York, 1994.

[37] T.A. Keith, AIMAll (Version 17.01.25), TK Gristmill Software, Overland Park KS,USA, 2017 (aim.tkgristmill.com).

[38] R.F.W. Bader, A bond path: a universal indicator of bonded interactions,J. Phys. Chem. A 102 (1998) 7314e7323, https://doi.org/10.1021/jp981794v.

[39] R.F.W. Bader, H. Ess�en, The characterization of atomic interactions, J. Chem.Phys. 80 (1984) 1943e1960, https://doi.org/10.1063/1.446956.

[40] D. Cremer, E. Kraka, A description of the chemical bond in terms of localproperties of electron density and energy, Croat. Chem. Acta 57 (1984)1259e1281.

[41] E. Espinosa, E. Molins, C. Lecomte, Hydrogen bond strengths revealed by to-pological analyses of experimentally observed electron densities, Chem. Phys.Lett. 285 (1998) 170e173, https://doi.org/10.1016/S0009-2614(98)00036-0.

[42] E. Espinosa, I. Alkorta, I. Rozas, J. Elguero, E. Molins, About the evaluation ofthe local kinetic, potential and total energy densities in closed-shell in-teractions, Chem. Phys. Lett. 336 (2001) 457e461, https://doi.org/10.1016/S0009-2614(01)00178-6.

[43] A.O. Borissova, M.Y. Antipin, H.A. Karapetyan, A.M. Petrosyan, K.A. Lyssenko,Cooperativity effects of H-bonding and charge transfer in an L-nitroargininecrystal with Z’ > 1, Mendeleev Commun. 20 (2010) 260e262, https://doi.org/10.1016/j.mencom.2010.09.006.

[44] G.V. Baryshnikov, B.F. Minaev, V.A. Minaeva, V.G. Nenajdenko, Single crystalarchitecture and absorption spectra of octathio[8]circulene and sym-tetrase-lenatetrathio[8]circulene: QTAIM and TD-DFT approach, J. Mol. Model. 19(2013) 4511e4519, https://doi.org/10.1007/s00894-013-1962-1.

[45] G.V. Baryshnikov, B.F. Minaev, A.A. Korop, V.A. Minaeva, A.N. Gusev, Structureof zinc complexes with 3-(pyridin-2-yl)-5-(arylideneiminophenyl)-1H-1,2,4-triazoles in different tauto-meric forms: DFT and QTAIM study, Russ. J. Inorg. Chem. 58 (2013) 928e934,https://doi.org/10.1134/S0036023613080032.

[46] F. Shahangi, A.N. Chermahini, H. Farrokhpour, A. Teimouri, Selectivecomplexation of alkaline earth metal ions with nanotubular cyclopeptides:DFT theoretical study, RSC Adv. 5 (2014) 2305e2317, https://doi.org/10.1039/C4RA08302D.

[47] L.N. Puntus, K.A. Lyssenko, M.Y. Antipin, J.-C.G. Bünzli, Role of inner- andouter-sphere bonding in the sensitization of EuIII-luminescence decipheredby combined analysis of experimental electron density distribution functionand photophysical data, Inorg. Chem. 47 (2008) 11095e11107, https://doi.org/10.1021/ic801402u.

[48] J.S. Park, H.J. Hah, S.M. Koo, Y.S. Lee, Effect of alcohol chain length on particlegrowth in a mixed solvent system, J. Ceram. Process. Res. 7 (2006) 83.

[49] S. Sadasivan, A.K. Dubey, Y. Li, D.H. Rasmussen, Alcoholic solvent effect onsilica synthesisdNMR and DLS investigation, J. Sol. Gel Sci. Technol. 12 (1998)5e14, https://doi.org/10.1023/A:1008659708390.

[50] J. Dharma, A. Pisal, Application Note, Simple Method of Measuring the BandGap Energy Value of TiO2 in the Powder Form Using a UV/Vis/NIR Spec-trometer, Perkin-Elmer Inc, Shelton, CT., USA, 2009.

[51] B. Efafi, M. Sasani Ghamsari, M.A. Aberoumand, M.H. Majles Ara, H. Hojati Rad,Highly concentrated ZnO sol with ultra-strong green emission, Mater. Lett.111 (2013) 78e80, https://doi.org/10.1016/j.matlet.2013.08.035.

https://doi.org/10.1007/s10971-005-6620-3https://doi.org/10.1021/nn800842bhttps://doi.org/10.1006/jcis.1997.5354https://doi.org/10.1006/jcis.1997.5354https://doi.org/10.1016/j.cej.2008.09.042https://doi.org/10.1016/j.jallcom.2010.08.060https://doi.org/10.1016/j.partic.2012.03.006https://doi.org/10.1016/j.partic.2012.03.006https://doi.org/10.1016/j.cej.2011.11.085https://doi.org/10.1016/j.matlet.2013.01.129https://doi.org/10.1016/j.molliq.2015.11.033https://doi.org/10.1016/j.molliq.2015.11.033http://refhub.elsevier.com/S0022-2860(18)31208-0/sref26http://refhub.elsevier.com/S0022-2860(18)31208-0/sref26http://refhub.elsevier.com/S0022-2860(18)31208-0/sref27http://refhub.elsevier.com/S0022-2860(18)31208-0/sref27http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28http://refhub.elsevier.com/S0022-2860(18)31208-0/sref28https://doi.org/10.1021/ct0502763https://doi.org/10.1021/ar700111ahttps://doi.org/10.1021/ct1006604https://doi.org/10.1021/ct700205nhttps://doi.org/10.1021/ct700205nhttps://doi.org/10.1039/b907148bhttps://doi.org/10.1063/1.448800http://refhub.elsevier.com/S0022-2860(18)31208-0/sref35http://refhub.elsevier.com/S0022-2860(18)31208-0/sref35http://refhub.elsevier.com/S0022-2860(18)31208-0/sref35http://refhub.elsevier.com/S0022-2860(18)31208-0/sref36http://refhub.elsevier.com/S0022-2860(18)31208-0/sref36http://refhub.elsevier.com/S0022-2860(18)31208-0/sref37http://refhub.elsevier.com/S0022-2860(18)31208-0/sref37https://doi.org/10.1021/jp981794vhttps://doi.org/10.1063/1.446956http://refhub.elsevier.com/S0022-2860(18)31208-0/sref40http://refhub.elsevier.com/S0022-2860(18)31208-0/sref40http://refhub.elsevier.com/S0022-2860(18)31208-0/sref40http://refhub.elsevier.com/S0022-2860(18)31208-0/sref40https://doi.org/10.1016/S0009-2614(98)00036-0https://doi.org/10.1016/S0009-2614(01)00178-6https://doi.org/10.1016/S0009-2614(01)00178-6https://doi.org/10.1016/j.mencom.2010.09.006https://doi.org/10.1016/j.mencom.2010.09.006https://doi.org/10.1007/s00894-013-1962-1https://doi.org/10.1134/S0036023613080032https://doi.org/10.1039/C4RA08302Dhttps://doi.org/10.1039/C4RA08302Dhttps://doi.org/10.1021/ic801402uhttps://doi.org/10.1021/ic801402uhttp://refhub.elsevier.com/S0022-2860(18)31208-0/sref48http://refhub.elsevier.com/S0022-2860(18)31208-0/sref48https://doi.org/10.1023/A:1008659708390http://refhub.elsevier.com/S0022-2860(18)31208-0/sref50http://refhub.elsevier.com/S0022-2860(18)31208-0/sref50http://refhub.elsevier.com/S0022-2860(18)31208-0/sref50https://doi.org/10.1016/j.matlet.2013.08.035

Solvothermal synthesis of zinc oxide nanoparticles: A combined experimental and theoretical study1. Introduction2. Experimental2.1. Materials and synthesis2.2. Measurements and characterization2.3. Size–strain line-broadening analysis2.4. Computational methods

3. Results and discussion3.1. Microstructural analysis3.2. UV–Vis and FT-IR spectroscopies3.3. The mechanism of ZnO particle formation

4. ConclusionsAcknowledgmentsAppendix A. Supplementary dataReferences