Elsevier

Thin Solid Films

Volume 519, Issue 20, 1 August 2011, Pages 7100-7105
Thin Solid Films

Influence of coil current modulation on TiO2 nanoparticle synthesis using pulse-modulated induction thermal plasmas

https://doi.org/10.1016/j.tsf.2010.11.063Get rights and content

Abstract

This paper describes the relation between the size of synthesized nanoparticles and the temperature variation of the surrounding plasma in TiO2 nanoparticle synthesis using pulse-modulated induction thermal plasmas (PMITP). The Ar–O2 PMITP at 20 kW was used to obtain repetitional temperature fields at pressure of 200 Torr. The TiO2 nanoparticles were synthesized by direct injection of Ti powder with mean diameter of 45 μm. Dependence of the nanoparticle size on modulation parameters such as the duty factor and the shimmer current level of the PMITP were investigated experimentally. Instantaneous temperature evolution was evaluated through spectroscopic observation. Experimental results show that the temperature decay rate in the PMITP reached 105106 K/s, and that the mean diameter of synthesized particles decreased with the temperature decay rate.

Introduction

Inductively coupled thermal plasmas (ICTPs) operating at several tens of kilowatts are widely used for various material processes such as syntheses of diamond films [1], thermal barrier coatings [2], syntheses of fullerene [3], materials surface modifications [4], and nanoparticle syntheses [5], [6], [7]. For example, titanium dioxide (TiO2) nanoparticles have attracted much interest for various applications such as photocatalysts [8], photonic crystals [9], photovoltaic cells [10], and gas sensors [11]. The synthesis of TiO2 nanoparticles has therefore been investigated in many research efforts using a steady-state type of ICTP [6], [7].

We have developed some high-power modulated induction thermal plasma systems, known as PMITP and AMITP [12], [13]. The pulse-modulated induction thermal plasma (PMITP) system can modulate the coil current sustaining the induction thermal plasma into a rectangular waveform. It can change the temperature and radical densities as well as the gas flow field in the thermal plasma in a time domain [12]. Some studies have tried to adopt the PMITP for surface modification of materials using good controllability of the temperature and radical densities [14], [15], [16]. A high-power arbitrary-waveform-modulated induction thermal plasma (AMITP) system has also been developed for additional enhancement of the degrees of freedom in thermal plasma control [13]. The AMITP system can modulate the coil current not only into a rectangular waveform but also into an externally given waveform, which provides precise time-domain control of the temperature of thermal plasmas [13].

We have started studying nanoparticle synthesis using a modulated induction thermal plasma [17]. The two following main effects can be expected in nanoparticle syntheses using modulated thermal plasmas. One is rapid and complete evaporation of injected raw materials during high-power input time duration, with subsequent rapid cooling of the evaporated material during the low-power input period. In particular, the low-power input period in the modulation might provide rapid cooling of the evaporated material, which can be described as the temperature decay rate |  T/∂ t|. This temperature decay rate |  T/∂ t| is expected to be of 105–106 K/s order in the PMITP, in addition to the temperature history of the evaporated material by convection |u   T|, where T is the temperature, t is the time, and u is the gas flow vector. In other words, we anticipate that the temperature history of evaporated material |dT/dt| is |  T/∂ t + u   T|. The rapid temperature history of 105–106 K/s for evaporated material is known to be useful to create nanoparticle synthesis [18], [19]. The other is the enhancement of the time-averaged temperature-gradient, which also causes cooling of evaporated material during transfer to the downstream chamber. It is related with |u   Tave|, where Tave is the average temperature. This enhancement of the time-averaged temperature gradient has been observed also in other applications of the PMITP in experiments and numerical simulations conducted in our previous works [15], [20]. Actually, our previous study related to nanoparticle synthesis using PMITP [17], revealed that the higher degree of the coil current modulation for the PMITP can provide smaller TiO2 nanoparticles at a fixed duty factor (DF) of 80%, and that the weight fraction of anatase-TiO2 nanoparticles to the synthesized particles were 80–90%, almost irrespective of the modulation degree. In addition, our estimation of the time-averaged temperature in the reaction chamber revealed that the time-averaged temperature was decreased by the modulation of the coil current even at the same input power of 20 kW used in our previous study. However, the effect of current modulation for different duty factors and the instantaneous temperature variation have been not yet investigated.

This paper describes the influence of modulation conditions such as the duty factor (DF) and the shimmer current level (SCL) on TiO2 nanoparticle synthesis using Ar–O2 PMITP at 20 kW. For this purpose, 15 combinations of DF and SCL were experimentally studied to assess their effects on the mean particle size and the weight fraction of synthesized nanoparticles. Furthermore, the time evolution in the temperature of the thermal plasma in the reaction chamber was determined using time resolution spectroscopic observation to investigate rapid changes in the temperature. Based on the temperature measurement results, the relation between the temperature decay rate during the low input-power duration and the mean diameter of synthesized nanoparticles was obtained to investigate instantaneous rapid cooling effects of evaporated material. The main contribution of this paper is the experimental confirmation of the pulse modulation effect on nanoparticle synthesis.

Section snippets

Experimental arrangements

Fig. 1 depicts the experimental setup and the details of this setup can be found in our previous paper [17]. We used a radio frequency (rf) power supply with a metal-oxide semiconductor field emission transistor (MOSFET). It can also modulate electric current of several hundreds of amperes for supplying the coil current. The amplitude of the coil current sustaining induction thermal plasmas was modulated into a rectangular waveform in this work. The modulated coil current has modulation

SEM images and mean diameters

Fig. 4 presents examples of FE-SEM micrographs of particles synthesized using induction thermal plasma with and without the coil current modulation. Panel (a) portrays the image for DF = 100%, i.e. a non-modulation condition, whereas panels (b)–(e) portray images for different DFs. For all images, the SCL was fixed at 80%. As might be readily apparent, reducing DF from 100% to 67% roughly increases the particle size. In contrast, reducing DF from 67% to 60% seems to increase the particle size

Conclusions

In summary, a 20 kW class of Ar–O2 pulse-modulated induction thermal plasma (PMITP) was used for synthesis of TiO2 nanoparticles for different modulation conditions including those of the shimmer current level and duty factor. Experimental results show that a modulation condition of 80%DF–65%SCL provides smaller nanoparticles among 15 combinations of DF and SCL, and that the weight fraction of anatase TiO2 in the synthesized particles was estimated as 80%–90%, almost independent of the

References (21)

  • W.R. Chen et al.

    Surf. Coat. Technol.

    (2005)
  • C. Wang et al.

    Thin Solid Films

    (2003)
  • H. Tanaka et al.

    Thin Solid Films

    (2004)
  • M. Shigeta et al.

    Thin Solid Films

    (2004)
  • J.E. Lee et al.

    Thin Solid Films

    (2004)
  • S. Malato et al.

    Catal. Today

    (2007)
  • J. Li et al.

    Anal. Chem. Acta

    (2008)
  • K.P. Biju et al.

    Thin Solid Films

    (2008)
  • N. Ohashi et al.

    J. Cryst. Growth

    (2007)
  • Y. Li et al.

    J. Cryst. Growth

    (2002)
There are more references available in the full text version of this article.

Cited by (16)

View all citing articles on Scopus
View full text