Journal Information
Vol. 8. Issue 3.
Pages 2481-3388 (May - June 2019)
Download PDF
More article options
Vol. 8. Issue 3.
Pages 2481-3388 (May - June 2019)
Original Article
DOI: 10.1016/j.jmrt.2018.12.027
Open Access
New approaches in lowering the gas-phase synthesis temperature of TiO2 nanoparticles by H2O-assisted atmospheric pressure CVS process
Mostafa Rahiminezhad-Soltania,
Corresponding author

Corresponding author.
, Kamal Saberyanb, Abdolreza Simchic, Christoph Gammerd
a Young Researchers and Elites Club, Saveh Branch, Islamic Azad University, P.O. Box: 39187-366, Saveh, Iran
b Materials and Nuclear Fuel Research School, NSTRI, P.O. Box: 11365-8486, Tehran, Iran
c Department of Materials Science and Engineering, Institute for Nanoscience and Nanotechnology, Sharif University of Technology, Azadi Avenue, P.O. Box: 11365-9466, Tehran, Iran
d Erich Schmid Institute of Materials Science, Austrian Academy of Sciences, Jahnstrasse 12, 8700 Leoben, Austria
This item has received

Under a Creative Commons license
Article information
Full Text
Download PDF
Figures (8)
Show moreShow less
Tables (2)
Table 1. Process parameters used for the synthesis of TiO2 nanoparticles by the H2O-assisted APCVS route.
Table 2. Analysis of the SAED pattern.
Show moreShow less

H2O-assisted atmospheric pressure chemical vapor synthesis is a modern economical process for the gas-phase synthesis of TiO2 nanoparticles. In the present work, the influence of synthesis temperatures (100–800°C) on the phase structure, nanoparticle size, morphology, and agglomeration is investigated by transmission electron microscopy, selected area electron diffraction, X-ray diffraction, thermogravimetry, and differential thermal analysis. Down to 400°C, crystalline TiO2 nanoparticles are synthesized and at 200°C amorphous nanoparticles are formed. Therefore, a decrease in minimum synthesis temperature by more than 500°C is achieved. In addition, the paper investigates the hypothesis that the high heat capacity of the H2O particles is responsible for the achieved decrease in synthesis temperature and for the dramatic decrease in size, coalescence, coagulation, and agglomeration of the nanoparticles. It is shown that the nanoparticles size is considerably higher for nanoparticles produced with gas-phase H2O particles in comparison to those produced with liquid-phase H2O particles, (average size 41 and 13nm, respectively), because of the lower heat capacity of gas-phase H2O particles, thus confirming the hypothesis.

Chemical vapor synthesis
TiO2 nanoparticles
Gas phase reaction
Low-temperature synthesis
Full Text

Nanomaterials with different morphologies and structures can be made by gas phase and liquid phase technologies [1]. The research on process engineering and scale-up is important for the commercial production and application of nanomaterials, because the properties and the structure of nanomaterials are not only determined by the nucleation and growth process, but also strongly affected by processing parameters, such as the method of mixing, the heat transfer or the temperature distribution [1,2].

Metals, metal oxides, metal carbides, metal nitrides, and their composite nanoparticles can be manufactured by chemical vapor synthesis (CVS) [3–5], microwave plasma [6] and radio frequency plasma enhanced chemical vapor deposition [7]. In conventional CVS processes, faster pyrolysis at higher temperatures results in smaller particle size, but it enhances subsequent sintering and particle size growth [8]. On the other hand, slower decomposition rates at lower temperatures result in larger primary particles caused by surface reaction mechanisms [8,9]. Therefore, conventional CVS processes still face many problems. By using the H2O-assisted atmospheric pressure CVS (APCVS) process, TiO2 nanoparticles can be produced at lower temperatures without dramatic coalescence or coagulation, thanks to the present primary and secondary H2O particles [10].

Lowering the cost of the CVS process by lowering the synthesis temperature without compromising the quality of the synthesized materials will widen the scope of the applications of the CVS process [11]. Unlike high-temperature gas-phase processes such as oxidation and combustion synthesis routes [12,13], the low-temperature H2O-assisted APCVS process allows to control the product powders, yielding TiO2 powders with small size, narrow size distribution, and weak agglomeration. In addition, a decrease in energy consumption, retarded corrosion of the reactor and a reduction in operation problems can be expected for the low-temperature H2O-assisted APCVS process.

In addition to lowering the cost of CVS, the decrease in synthesis temperature by the H2O-assisted APCVS process has considerable influence on the TiO2 nanoparticle characteristics. Kuo et al. [14] showed that the visible-light activity of TiO2 photocatalysts increased as the synthesis temperature decreased. In other words, the minimum synthesis temperature for TiO2 nanoparticles is the best CVS reaction temperature for producing nano-TiO2 with high visible-light-responsive photocatalytic activity. In addition, Kuo et al. [14] suggested that the synthesis temperature should not increase because of the appearance of the rutile phase at higher temperatures.

In conventional CVS processes, the heat loss caused by the collisions with gas molecules only depends on the heat capacity of gas molecules (cg) [15]. In the presence of H2O particles, TiO2 nanoparticles have collisions either with H2O or gas particles. Since the specific heat capacity of H2O is high, the heat loss caused by the collisions increases. The nanoparticles’ temperature significantly decreases leading to a lower coalescence and particle size growth. Therefore, much smaller nanoparticles are obtained.

In the present work, the minimum synthesis temperature for TiO2 nanoparticles is examined experimentally and theoretically. Our hypothesis that the high heat capacity of the H2O particles leads to a decrease in synthesis temperature is experimentally investigated.

Xia et al. [16] stated that although the hydrolysis of TiCl4 can be linked to oxidation, it mainly acts as a nucleation agent; however, not much attention was paid on the hydrolysis as an independent preparation route and therefore detailed information is still limited. Hence, in the present work, the effect of hydrolysis on lowering the gas-phase synthesis temperature of TiO2 nanoparticles at atmospheric pressure is studied.


The raw materials used in this study were TiCl4 (purity>99.99%, Merck), oxygen (99.999%) and H2O (99.99%). The carrier gas was highly pure argon (99.999%). Fig. 1 shows a schematic illustration of the experimental apparatus. A hot-wall, atmospheric pressure, horizontal quartz reactor with an inner diameter of 80mm and a length of 800mm was used. The liquid precursor (TiCl4) was vaporized in a vertical bubbler in an oil bath. The precursor was introduced into a horizontal quartz tube reactor by bubbling argon through a precursor container. In addition, H2O was introduced into the reactor by bubbling argon through an H2O container.

Fig. 1.

Schematic presentation of the H2O-assisted APCVS apparatus used for the synthesis of TiO2 nanoparticles.


The vapor concentration of the precursor was calculated, assuming that the carrier gas through the bubbler was completely saturated. A mass flow meter was used for the measurement of the TiCl4 flow rate. The precursor was introduced into the reactor at a flow rate of 50g/h. Since the bubblers were scaled, the reductions of precursors during each test were measured with respect to the total gas flow rate (2.1L/min) and sampling time. Pure argon was introduced directly into the reactor at a flow rate of 0.5L/min under atmospheric temperature and pressure. For TiCl4 oxidation, oxygen gas was introduced at a flow rate of 0.5L/min under atmospheric temperature and pressure. In the final step, the H2O was introduced directly to the reactor at a flow rate of 0.5L/min. The total gas flow rate was maintained at 2.1L/min, resulting in equal residence time in the reactor for all runs. The furnace consisted of a fully recrystallized quartz reactor with an isothermal hot zone (Effective Reaction (ER) zone) and an adjustable accurate temperature controller. While passing through the reactor, the generated monomers underwent coalescence, coagulation, agglomeration, and sintering, finally ending up as TiO2 nanoparticles. The produced nanoparticles were collected in the cold trap which consisted of a chamber, with inlet and outlet lines. The chamber was maintained in an ice-water bath.

All experiments were carried out 3 times and no large differences (1–5nm) in particles diameter were observed. Before all runs, the internal surface of the quartz reactor was cleaned by laboratory alcohol (Ethanol, 96%). After cleaning the internal surface, the reactor was purged with pure argon at a flow rate of 0.5L/min for 10min. After that, sampling was carried out for 30min at the constant above-mentioned flow rates.

To examine the effects of the ER zone temperature and find the minimum temperature for generating anatase TiO2 nanoparticles by the H2O-assisted APCVS process, experiments were carried out at different ER zone temperatures of 800, 700, 600, 400, 200 and 100°C, respectively.

To study the effects of the heat capacity of the H2O particles on the minimum synthesis temperature, two experiments were carried out. Firstly, liquid-phase H2O particles with a high heat capacity of about 4.18J/(gK) were introduced directly into the reactor. Secondly, H2O particles were introduced into the gas phase. The heat capacity of H2O particles in the gas phase is about 2.08J/(gK) and therefore is significantly lower than the liquid phase.

Transmission electron microscopy (TEM), selected area electron diffraction (SAED), X-ray diffraction methods (XRD), thermogravimetry (TGA) and differential thermal analysis (DTA), were used to characterize the produced TiO2 nanoparticles. The transmission electron microscopy (TEM) images and SAED patterns for the TiO2 nanoparticles were obtained on a Philips-EM-208S instrument with a tungsten filament at an accelerating voltage of 100kV and a vacuum system which operates in the 10−9Torr range. The crystal structure of the TiO2 nanoparticles was characterized by X-ray diffraction (XRD). Patterns were recorded on a Philips-PW1800 diffractometer using Cu Kα1 radiation (λ=1.54Å) and a scanning angle between 10° and 90°. The average crystallite size was determined from the XRD measurements using the Scherrer equation [17]:

where D is the crystallite size (nm), λ the wavelength of X-ray used (1.542Å for Cu Kα1), θ half of the diffraction peak angle, and β the full width at half maximum (FWHM) of the peaks (radians).

20mg from each of the synthesized prepared TiO2 nanopowders from all experiments underwent thermogravimetry and differential thermal analysis (TG-DTA) in an argon atmosphere using a Rheometric Scientific STA-1500 machine. Thermogravimetric analysis of the all samples was performed using an Al2O3 sample holder in a nitrogen atmosphere at a flow rate of 50mLmin−1 with a heating rate of 10°C/min from 20 to 1000°C.

3Results and discussion3.1Effect of the ER zone temperature

The synthesis conditions are summarized in Table 1. The synthesis temperature was varied from 800 to 100°C. By H2O-assisted APCVS process the TiO2 nanoparticles were achieved at synthesis temperatures down to 200°C, but at 100°C, the production rate of TiO2 nanoparticles was too low to yield a significant number of nanoparticles.

Table 1.

Process parameters used for the synthesis of TiO2 nanoparticles by the H2O-assisted APCVS route.

Sample  Argon flow rate (sccm)  Oxygen flow rate (sccm)  Precursor concentration (g/h)  ER zone temperature (°C)  Crystal structure (XRD)  Titania size (nm) (XRD) 
500  500  50  800  Anatase  13 
500  500  50  700  Anatase  40 
500  500  50  600  Anatase  35 
500  500  50  400  Anatase  10 
500  500  50  200  Amorphous  Not specified 
500  500  50  100  Not specified  – 

Fig. 2 shows X-ray diffraction patterns (XRDs) from the TiO2 nanopowders synthesized by H2O-assisted APCVS at different synthesis temperatures (samples 1–5 in Table 1). The particles synthesized at 400°C and above show crystalline peaks that can be clearly identified as anatase TiO2. On the other hand, the nanoparticles synthesized at 200°C show no crystalline peaks but a broad hump indicating that only amorphous particles are produced.

Fig. 2.

XRD patterns of TiO2 nanoparticles prepared at different synthesis temperatures: (a) 800°C, (b) 700°C, (c) 600°C, (d) 400°C, and (e) 200°C.


In previous studies [18–21] comprehensive experimental, thermodynamics and kinetics studies of simultaneous oxidation/hydrolysis showed that nanoparticles synthesized above 700°C were the result of either the oxidation or the hydrolysis reactions. Rahiminezhad-Soltani [22] showed that both oxidation and hydrolysis of TiCl4 have negative free standard energies, ΔG0, in the temperature range between 298 and 1200K, but the oxidation reaction has higher activation energy and thus the oxidation reaction does not occur up to 700°C at atmospheric pressure. According to Refs. [23,24] hydrolysis of TiCl4 dominates the synthesis process of TiO2 nanoparticles but oxidation of TiCl4 dominates at high temperatures.

As the synthesis temperature increases from 200 to 600°C, the thermal velocity, the kinetic energy and the reaction rate increase due to higher mobility and collisions and thus more crystalline TiO2 nanoparticles are synthesized, leading to an increase in the height of the peaks in the XRD profiles (cf. Fig. 2c–e). When the synthesis temperature is further increased to 700°C (cf. Fig. 2b), the oxidation reaction starts to occur partly and less crystalline TiO2 nanoparticles are obtained. Therefore, as shown in Fig. 2 by increasing the temperature from 600°C up to 700°C the peaks of (103), (112) and (213) disappear. At a synthesis temperature of 800°C, the oxidation reaction dominates the hydrolysis reaction and the intensity of the crystalline peaks in the XRD profile increases again, indicating more crystalline TiO2 nanoparticles are generated (cf. Fig. 2a).

It should be pointed out that an identical sample setup and an identical amount of powder was used for all XRD measurements. Therefore, a lower intensity of the crystalline peaks is not only an indication that lower crystalline nanoparticles were synthesized but also an indication of a higher volume fraction of amorphous nanoparticles. Also, the synthesis time held constant and this would mean the height of the peaks indicates how much volume crystalline powder was generated.

Fig. 3(a and b) shows TEM images and selected area electron diffraction (SAED) patterns comparing the TiO2 nanoparticles obtained at synthesis temperatures of 800 and 700°C, respectively. To determine the crystal structure for the nanoparticles synthesized at 800°C from the SAED patterns, the PASAD-tools software was used [25]. A profile was deduced from the SAED pattern by azimuthal integration and the positions of the crystalline peaks were determined by peak fitting. The peaks indexed in Table 2 confirm the TiO2 anatase crystal structure (JCPDS number 21-1272). A comparison of the SAED pattern in Fig. 3a and b, reveals more intense and continuous rings for the specimen synthesized at 800°C, indicating that more crystalline TiO2 nanoparticles are produced and that their size is smaller. This is also confirmed in the TEM bright-field images in Fig. 3. Both, amorphous and crystalline nanoparticles can be observed at a synthesis temperature of 700°C. The amorphous particles are marked by "A" in the TEM bright-field image in Fig. 3b.

Fig. 3.

TEM images and SAED patterns of TiO2 nanoparticles synthesized at: (a) 800°C and (b) 700°C.

Table 2.

Analysis of the SAED pattern.

Number  Reflection  Position [nm−1d [Å]  Simulated intensity 
1  100.284  3.520  100 
2  000.420  2.378  20 
3  200.528  1.892  35 
4100.588  1.699  20 
210.600  1.666  20 
5200.675  1.480  14 
110.733  1.364 
6  220.747  1.337 
7  210.790  1.264  10 
8  220.857  1.166 
9  320.956  1.043 
10  211.105  0.866 

Nakaso et al. [9] showed that in solitary oxidation synthesis processes the surface reactions dominate the gas phase reactions at lower temperatures. Due to surface reactions, the condensed clusters of TiOClx serve as seed nuclei at the exit of the ER zone, where the rest of the unreacted precursor condenses out forming partially oxidized Ti oxychlorides. As the temperature drops, they transform into TiO2 upon thermal decomposition [9]. This results in large amorphous particles. Therefore, it can be concluded that the TiO2 nanoparticles synthesized at a temperature of 700°C are the result of incomplete oxidation reactions. Both, crystalline nanoparticles produced by hydrolysis reactions and large amorphous nanoparticles produced by oxidation reactions are obtained at the synthesis temperature of 700°C. This explains the observed decrease in crystalline nanoparticles at 700°C. At the synthesis temperature of 800°C, because of a decrease in surface reactions and thus the number of oxidized Ti oxychlorides, the produced crystalline TiO2 nanoparticles increases again.

To quantify the effect of the synthesis temperature on the average nanoparticle size, the Scherrer equation was used [17]. At a synthesis temperature of 200°C, only amorphous nanoparticles are produced. At synthesis temperatures of 400, 600, 700 and 800°C, the particles show a crystalline structure with average nanoparticle sizes of 10, 35, 40 and 13nm, respectively.

There are two main reasons why the average TiO2 particle size decreases again at a synthesis temperature of 800°C: (1) as the hydrolysis of TiCl4 is sometimes linked to the oxidation route, it acts as a nucleation agent and therefore this phenomenon causes nanoparticles size reduction [16]; (2) in the gas-phase processes, a greater equilibrium constant (Kp) favors the formation of smaller particles [16]. Therefore, at a synthesis temperature of 800°C, where complete oxidation reactions dominate the synthesis process, the size of the TiO2 nanoparticles decreases because the equilibrium constant (Kp) is larger [16]. At 700°C, on the other hand, due to dominating surface reactions and incomplete oxidation reactions [9], larger nanoparticles are reached.

Fig. 4 shows the equilibrium constants of oxidation and hydrolysis reactions for TiCl4. At a temperature of 800°C mainly complete oxidation reactions occur, and the oxidation reactions dominate over the hydrolysis reactions because the Kp value of oxidation is kept above the Kp value of hydrolysis. Therefore, smaller nanoparticles will be obtained [16]. Xia et al. [16] stated that a Kp value that exceeds 102–3 is necessary for preparing nanoscale particles. As one can easily observe in Fig. 4, the Kp value of oxidation is above 104 up to a temperature of 1700°C and the Kp value of hydrolysis is above 102 up to 1100°C.

Fig. 4.

Equilibrium constants for oxidation and hydrolysis reactions of TiCl4.


The particle diffusion coefficient, Γ, which is affected by the temperature is denoted by Γ=Γt+ΓB, where ΓB is the Brownian diffusivity and Γt is the turbulent diffusivity [26]. The Brownian diffusivity can be expressed as:

where Cc is the Cunningham correction factor, kb the Boltzmann constant, μ the viscosity of compound gas and da the mean agglomerated particle diameter. Therefore, according to Eq. (2), Brownian diffusivity decreases at lower synthesis temperatures and thus smaller nanoparticles and clusters are obtained.

Eq. (3) defines turbulent diffusivity, which can be expressed as [27]:

where μt is the turbulent viscosity and σϕ is the turbulent Schmidt number. The turbulent viscosity is defined by [27]:
where Cμ is a modeling coefficient in a standard two-equation κɛ turbulence model and is ascribed the value 0.09, 〈ρ〉 is the density, κ is the turbulent kinetic energy and ɛ is the dissipation rate of turbulent kinetic energy [27].

The turbulent kinetic energy, k, is defined as [28,29]:

where U is the velocity magnitude and Tu is the turbulent intensity [28,29].

The turbulent Schmidt number, σϕ, is defined by [30]:

where νt is the turbulent kinematic viscosity and Dt is the turbulent diffusion coefficient [30]. Valero et al. [28] showed that as far as turbulence arises, it is reasonable to assume that turbulence transport will prevail over molecular diffusion; hence Dt≈D, where D is the total diffusion (molecular and turbulent) coefficient [28].

In practical situations, small particles suspended in a flowing fluid are simultaneously subjected to Brownian diffusion and turbulent dispersion. For submicron particles, the Brownian diffusion effects may become quite important [31]. Hence the Brownian effect may become important only for very small particles. For particles larger than 5μm the Brownian effects are negligible and turbulent diffusion is dominant. Ounis et al. [31] showed that turbulence effects generally dominate Brownian effects.

According to Eq. (5), by decreasing the temperature in an isotropic flow field, the turbulent kinetic energy, k, decreases due to lower velocity and lower turbulent intensity. Therefore, according to Eqs. (4) and (3) by decreasing the turbulent kinetic energy, the turbulent viscosity, μt, and finally the turbulent diffusivity, Γt, decreases. Also, by decreasing the temperature the total diffusion decreases and due to D≈Dt[28] the turbulent diffusion coefficient, Dt, decreases and therefore the turbulent Schmidt number increases. By increasing the turbulent Schmidt number, σϕ, the turbulent diffusivity, Γt, decreases. Finally, by decreasing the synthesis temperature either the Brownian diffusivity ΓB or the turbulent diffusivity, Γt, decreases and therefore the particle diffusion coefficient, Γ, decreases. By decreasing the particle diffusion coefficient, Γ, smaller nanoparticles would be obtained.

The influence of the temperature on the morphology, coagulation, and agglomeration of the prepared TiO2 nanoparticles was investigated by TEM. Fig. 5(a and b) shows TEM images of the TiO2 nanoparticles synthesized at temperatures of 400 and 700°C, respectively. In accordance with the theoretical prediction, larger particles are obtained at 700°C.

Fig. 5.

TEM images of TiO2 nanoparticles formed due to simultaneous oxidation and hydrolysis reactions synthesized at (a) 400°C and (b) 700°C.


The characteristic time for coagulation of TiO2 nanoparticles, τc (s) is [32]:

where β is the collision frequency for Brownian coagulation in the transition regime, ρg the density of the carrier gas and N the particle number [32]. The thermal velocity increases with increasing temperature [10,32]. Therefore, the collision frequency of TiO2 particles, β, and the particle number, N, increase and cause the reduction of τc and increment of coagulation. Therefore, lowering the synthesis temperature leads to a reduction of the coagulation. This is confirmed in Fig. 5, showing no coagulation at 400°C.

Lehtinen and Zachariah [15] reported that the characteristic coalescence time, τf, for volume diffusion is:

where k is the Stefan-Boltzmann constant, T the temperature, N the number of TiO2 molecules for each particle, σ the surface tension and D the diffusion coefficient. Hence, we can expect a decrease in nanoparticles size with decreasing temperature, which is confirmed by the experimental results as shown in the TEM images in Fig. 5.

The characteristic sintering time based on a surface diffusion neck growth model, τsin, was proposed as (SI units) [33]:

where dp is the primary particle diameter and T the temperature. According to Eq. (9) in conventional CVS processes the sintering rate increases with increasing synthesis temperature. In the case of the H2O-assisted APCVS process, the primary and secondary H2O particles significantly decrease the sintering neck and surface diffusion [22]. As shown in the TEM images in Figs. 3 and 5, no noticeable sintering necks were observed between the nanoparticles produced using the H2O-assisted APCVS process.

Furthermore, theoretical studies [9,10,34] suggest that the secondary H2O particles eliminate unreacted TiCl4 particles which would otherwise condense on the TiO2 nanoparticles and thus result in nanoparticles with unclear boundaries [9]. This is experimentally confirmed by the TEM images showing that the TiO2 nanoparticles synthesized by the H2O-assisted APCVS process are round with clear boundaries (cf. Figs. 3 and 5). It can be concluded, that due to the low activation energy of the TiCl4 hydrolysis reaction and the low reaction temperature between TiCl4 and H2O [10,22], the synthesis temperature for TiO2 nanoparticles decreases dramatically and pure round nanoparticles can be achieved.

While lowering the synthesis temperature has many advantages, a lower synthesis temperature has various problems in the case of the conventional CVS process. Therefore, it is important to point out that the present work demonstrates that the H2O-assisted APCVS process enables to lower the synthesis temperature without facing those problems. In conventional CVS processes, the synthesis temperature cannot be reduced since no oxidation reaction occurs before thermal decomposition of TiCl4[19]. Johannessen et al. [35] showed that TiCl4 oxidation is assumed to occur instantaneously once the gas temperature exceeds the decomposition temperature for the precursor, TD. Nakaso et al. [9] showed that before 800°C, no TiO2 nanoparticles are produced and Rahiminezhad-Soltani et al. [10] showed that at atmospheric pressure the oxidation reaction does not occur before 700°C. In addition, Nakaso et al. [9] showed that at low reactor temperatures only 5% of the precursor reacts, resulting in large and amorphous particles [9].

The particle diameter, D, of TiO2 nanoparticles synthesized by gas-phase reactions is proportional to the number of nuclei, N, as shown in Eq. (10)[36]:

where C0 is the concentration of metal halide and N the number of nuclei. According to Eq. (10), the particle size of TiO2 decreases with increasing nuclei. Therefore, in conventional CVS processes, a decrease in synthesis temperature leads to an increase in particle diameter for the TiO2 nanoparticles, due to the lower number of nuclei, N.

As an estimate of the degree of agglomeration in a suspension, the average agglomeration number (AAN) was calculated. AAN is the average number of primary particles contained within an agglomerate. AAN can be calculated as [37]:

where ɛ is the estimated fractional porosity of the agglomerates, approximated as 0.4; dsvDLS the surface-volume diameter from particle size distribution of dilute dispersions measured with the DLS method, dsvBET is the surface-volume diameter calculated from BET analysis of nitrogen adsorption isotherms.

The dsvDLS can be calculated using particle size analysis data as follows [37]:

where x is the particle size and N is the number of particles [37]. According to Eqs. (7)–(9) by increasing the synthesis temperature, the particles size increases. Hence, according to Eqs. (11) and (12) by increasing the particles size, the average agglomeration number (AAN) as an estimate of the degree of agglomeration increases. Therefore, by increasing the synthesis temperature, the agglomeration of nanoparticles increases.

The pathway of particle formation in high-temperature processes can often be characterized by nucleation of condensable material and subsequent particle growth by coagulation. Depending on the temperature, particles formed in this way may have different morphological structures. Spherical particles are produced when the colliding particles coalesce. However, when coalescence is quenched, agglomerates are formed. The size of the primary particles composing the agglomerates depends on the temperature history of the particle-formation process and on the temperature dependence of the material properties determining particle coalescence [38].

Fig. 6 shows the TG-DTA diagram of the samples synthesized at different synthesis temperatures. TG-DTA was carried out to assess the purity of the H2O-assisted APCVS-synthesized nanoparticles and their thermal stability. A small endothermic peak was observed in the DTA curves at a temperature of around 100°C. This peak is attributed to dehydration of the titanium hydrates [16]. The weight loss below 200°C is caused by the adsorbed water as the result of the exposure of the samples to air [10,22]. As one can easily observe, the DTA curve of synthesized TiO2 at 200°C, shows a small exothermic peak at 325°C corresponding to the crystallization of the amorphous nanoparticles.

Fig. 6.

TG-DTA results for TiO2 nanoparticles formed due to simultaneous oxidation and hydrolysis reactions at different synthesis temperatures.


A very small exothermic peak around 337°C can be observed also for the sample synthesized at 700°C, confirming the presence of some amorphous nanoparticles in this specimen. Apart from this small peak, no exothermic peak can be observed in the DTA curves of the samples produced in the range of 400–800°C, showing that no crystallization or phase transition occurs in this temperature range.

3.2Effect of the heat capacity of the H2O particles

In previous studies [10,22] we showed that H2O particles have extraordinary effects on the characteristics of the TiO2 nanoparticles synthesized through the gas-phase APCVS process [10,22]. Our hypothesis is that the high heat capacity of H2O particles is the major factor affecting the nanoparticle. This hypothesis is based on two observations: Firstly, a decrease in the size of the nanoparticles is observed when H2O particles are introduced [10]. Therefore, H2O particles have decreased the temperature of TiO2 monomers and primary particles. Secondly, introducing the liquid-phase H2O particles reduced the coagulation, coalescence, agglomeration, and sintering of the nanoparticles [10]. Since the specific heat capacity of water is high, the heat loss due to the collisions with water molecules increases and thus the nanoparticles’ temperature decreases significantly, leading to lower coalescence and particle size growth. Therefore, much smaller nanoparticles are obtained. In addition, decreasing the nanoparticles’ temperature reduces agglomeration, coagulation, and sintering of the nanoparticles. Therefore, it seems reasonable to assume that the change in nanoparticles characteristics is caused by the heat capacity of H2O particles. However, Akhtar et al. [34] and Jang [39] stated that the changes in particle characteristics in the presence of H2O particles are still not well understood because of the complex collisions and reactions occurring during synthesis.

The only way to confirm our hypothesis is to carry out experiments comparing H2O particles with different heat capacities. Therefore, H2O particles were introduced to the reactor in the liquid and in the gas phase. The heat capacity of gas-phase H2O particles is about half of the heat capacity of liquid-phase H2O particles. Based on our hypothesis, using gas-phase H2O particles should lead to an increase in size, coagulation, agglomeration, and coalescence of the TiO2 nanoparticles as compared to using liquid phase H2O particles.

Fig. 7(a and b) shows a comparison of TEM images of TiO2 nanoparticles synthesized using gas-phase and liquid-phase H2O particles, respectively. The nanoparticles synthesized using gas-phase H2O particles have larger sizes than those using liquid-phase H2O particles. Furthermore, coagulation, coalescence, and agglomeration of the nanoparticles increase significantly, when using gas-phase H2O particles. Fig. 8(a and b) shows XRD profiles of the TiO2 nanoparticles synthesized with liquid-phase H2O particles and gas-phase H2O particles, respectively. Both XRD patterns clearly show the TiO2 anatase phase, showing that the heat capacity of the H2O particles has no effect on the phase structure of the synthesized TiO2 nanoparticles. In addition, the average TiO2 nanoparticles size was calculated from the XRD profiles using the Scherrer equation (cf., Eq. (1)) [17]. For gas-phase H2O particles the average TiO2 nanoparticle size was 41nm and for liquid-phase H2O particles 13nm.

Fig. 7.

TEM images of TiO2 nanoparticles synthesized using: (a) liquid-phase and (b) gas-phase H2O particles.

Fig. 8.

XRD patterns of TiO2 nanoparticles prepared using: (a) liquid-phase and (b) gas-phase H2O particles.


As resulted in theoretical studies, introducing H2O particles in the gas phase leads to an increase in the average nanoparticle size. The experimental results thus confirm the predicted hypothesis: the heat capacity of the H2O particles used in the H2O-assisted APCVS process is the major factor affecting the synthesis and the characteristics of the TiO2 nanoparticles. Increasing the heat capacity leads to a decrease in size, coagulation, coalescence, agglomeration, and aggregation because it lowers the temperature of the TiO2 nanoparticles and monomers during synthesis.


The H2O-assisted APCVS method allows to decrease the synthesis temperature dramatically. The minimum synthesis temperature for TiO2 nanoparticles was reduced to 200°C, resulting in amorphous nanoparticles. The minimum synthesis temperature for crystalline anatase TiO2 nanoparticles was reduced to 400°C in atmospheric pressure. The synthesis temperature of 400°C is the best temperature for the gas-phase synthesis of anatase TiO2 nanoparticles by H2O-assisted APCVS process. In addition, the present work demonstrates that the formation of TiO2 nanoparticles is determined by a temperature reduction mechanism caused by the high heat capacity of the H2O particles. The importance of high heat capacity of H2O particles has not been shown experimentally to date. Therefore, experiments comparing H2O particles with different heat capacities were carried out. The results support the hypothesis that the formation and characteristics of the TiO2 nanoparticles are linked to the ability of H2O particles to reduce the temperature of the TiO2 nanoparticles. The size of the TiO2 nanoparticles is significantly reduced and coagulation, coalescence, agglomeration, and aggregations decrease with increasing heat capacity of the H2O particles.

Conflicts of interest

The authors declare no conflicts of interest.


We would like to acknowledge the support of the Iranian Nanotechnology Initiative Council (INIC), Tehran, Iran. Also, one of the authors, M. Rahiminezhad-Soltani, would like to express his deepest gratitude to his late father, Asadollah Rahiminezhad-Soltani for his love, support and abundant kindness.

A. Simchi.
Introduction to nanoparticles: properties, processing and applications.
first ed., Institute of Scientific Publications of Sharif University of Technology, (2008),
C. Li.
Structure controlling and process scale-up in the fabrication of nanomaterials.
Front Chem Sci Eng, 4 (2010), pp. 18-25
F. Mirhoseini, A. Bateni, S. Firoozi.
Gas phase synthesis of Ni3Fe nanoparticles by magnesium reduction of metal chlorides.
Powder Technol, 228 (2012), pp. 158-162
G. Akgul, F.A. Akgul, K. Attenkofer, M. Winterer.
Structural properties of zinc oxide and titanium dioxide nanoparticles prepared by chemical vapor synthesis.
J Alloys Compd, 554 (2013), pp. 177-181
W. Jin, I. Lee, A. Kompch, U. Dörfler, M. Winterer.
Chemical vapor synthesis and characterization of chromium doped zinc oxide nanoparticles.
J Eur Ceram Soc, 27 (2007), pp. 4333-4337
P. Huang, M. Wong.
Nanostructures of mixed-phase boron nitride via biased microwave plasma-assisted CVD.
Vacuum, 100 (2014), pp. 66-70
Y. Zhou, D. Probst, A. Thissen, E. Kroke, R. Riedel, R. Hauser, et al.
Hard silicon carbonitride films obtained by RF–plasma–enhanced chemical vapour deposition using the single–source precursor bis(trimethylsilyl)carbodiimide.
J Eur Ceram Soc, 26 (2006), pp. 1325-1335
I. Ahmad Md., S.S. Bhattacharya.
Effect of process parameters on the chemical vapour synthesis of nanocrystalline titania.
J Phys D Appl Phys, 41 (2008), pp. 155313-155319
K. Nakaso, K. Okuyama, M. Shimada, S.E. Pratsinis.
Effect of reaction temperature on CVD-made TiO2 primary particle diameter.
Chem Eng Sci, 58 (2003), pp. 3327-3335
M. Rahiminezhad-Soltani, K. Saberyan, F. Shahri, A. Simchi.
Formation mechanism of TiO2 nanoparticles in H2O-assisted atmospheric pressure CVS process.
Powder Technol, 209 (2011), pp. 15-24
K.L. Choy.
Chemical vapour deposition of coatings.
Prog Mater Sci, 48 (2003), pp. 57-170
H. Lee, M.Y. Song, J. Jurng, Y. Park.
The synthesis and coating process of TiO2 nanoparticles using CVD process.
Powder Technol, 214 (2011), pp. 64-68
H. Ma, H. Yang.
Combustion synthesis of titania nanoparticles in a premixed methane flame.
J Alloys Compd, 504 (2010), pp. 115-122
C. Kuo, Y. Tseng, C. Huang, Y. Li.
Carbon-containing nano-titania prepared by chemical vapor deposition and its visible-light-responsive photocatalytic activity.
J Mol Catal A Chem, 270 (2007), pp. 93-100
K.E.J. Lehtinen, M.R. Zachariah.
Energy accumulation in nanoparticle collision and coalescence processes.
J Aerosol Sci, 33 (2002), pp. 357-368
B. Xia, W. Li, B. Zhang, Y. Xie.
Low temperature vapor-phase preparation of TiO2 nanopowders.
J Mater Sci, 34 (1999), pp. 3505-3511
B.D. Culity.
Elements of X-ray diffraction.
second ed., Addison-Wesley Publishing Company Press, (1978),
T.H. Wang, A.M. Navarrete-Lopez, S. Li, D.A. Dixon.
Hydrolysis of TiCl4: initial steps in the production of TiO2.
J Phys Chem A, 114 (2010), pp. 7561-7570
R. Raghavan.
Measurement of the high-temperature kinetics of titanium tetrachloride (TiCl4) reactions in a rapid compression machine [Ph.D. thesis].
Case Western Reserve University, (2001),
F. Teyssandier, M.D. Allendorf.
Thermodynamics and kinetics of gas-phase reactions in the Ti–Cl–H system.
J Electrochem Soc, 145 (1998), pp. 2167-2178
R.H. West.
Modelling the chloride process for titanium dioxide synthesis [Ph.D. thesis].
University of Cambridge, (2008),
M. Rahiminezhad-Soltani.
Study of H2O effects on chemical vapor synthesis (CVS) of TiO2 nanoparticles from TiCl4 precursor [M.Sc. thesis].
Islamic Azad University, (2011),
A.K. John, S. Savithri, K.P. Vijayalakshmi, C.H. Suresh.
Density functional theory study of gas phase hydrolysis of titanium tetrachloride.
Bull Chem Soc Jpn, 83 (2010), pp. 1030-1036
J.K. Ani, S. Savithri, G.D. Surender.
Characteristics of titania nanoparticles synthesized through low temperature aerosol process.
Aerosol Air Qual Res, 5 (2005), pp. 1-13
C. Gammer, C. Mangler, C. Rentenberger, H.P. Karnthaler.
Quantitative local profile analysis of nanomaterials by electron diffraction.
Scr Mater, 63 (2010), pp. 312-315
M. Yua, J. Lin, T. Chanc.
Effect of precursor loading on non-spherical TiO2 nanoparticle synthesis.
Chem Eng Sci, 63 (2008), pp. 2317-2329
S.B. Pope.
PDF methods for turbulent reactive flows.
Prog Energy Combust Sci, 11 (1985), pp. 119-192
D. Valero, D.B. Bung.
Sensitivity of turbulent Schmidt number and turbulence model to simulations of jets in crossflow.
Environ Model Softw, 82 (2016), pp. 218-228
S.M. Murman.
A scalar anisotropy model for turbulent eddy viscosity.
Int J Heat Fluid Flow, 42 (2013), pp. 115-130
A.N. Colli, J.M. Bisang.
A CFD study with analytical and experimental validation of laminar and turbulent mass-transfer in electrochemical reactors.
J Electrochem Soc, 165 (2018), pp. E81-E88
H. Ounis, G. Ahmadi.
A comparison of Brownian and turbulent diffusion.
Aerosol Sci Technol, 13 (1990), pp. 47-53
R.N. Grass, S. Tsantilis, S.E. Pratsinis.
Design of high-temperature, gas-phase synthesis of hard or soft TiO2 agglomerates.
Am Inst Chem Eng J, 52 (2006), pp. 1318-1325
M.C. Heine, S.E. Pratsinis.
Polydispersity of primary particles in agglomerates made by coagulation and sintering.
J Aerosol Sci, 38 (2007), pp. 17-38
M.K. Akhtar, S. Vemury, S.E. Pratsinis.
Competition between TiCI4 hydrolysis and oxidation and its effect on product TiO2 powder.
Am Inst Chem Eng J, 40 (1994), pp. 1183-1192
T. Johannessen, S.E. Pratsinis, H. Livbjerg.
Computational analysis of coagulation and coalescence in the flame synthesis of titania particles.
Powder Technol, 118 (2001), pp. 242-250
Z. Wang, Z. Yuan, E. Zhou.
Influence of temperature schedules on particle size and crystallinity of titania synthesized by vapor-phase oxidation route.
Powder Technol, 170 (2006), pp. 135-142
C. Chen, C.S. Oakes, K. Byrappa, R.E. Riman, K. Brown, K.S. TenHuisen, V.F. Janas.
Synthesis, characterization, and dispersion properties of hydroxyapatite prepared by mechanochemical-hydrothermal methods.
J Mater Chem, 14 (2004), pp. 2425-2432
W. Koch, S.K. Friedlander.
Particle growth by coalescence and agglomeration.
Part Part Syst Charact, 8 (1991), pp. 86-89
H.D. Jang.
Effects of H2O on the particle size in the vapor-phase synthesis of TiO2.
Am Inst Chem Eng J, 43 (1997), pp. 2704-2709
Copyright © 2019. The Authors
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.