Synthesis of carbon-incorporated titanium oxide nanocrystals by pulsed solution plasma: electrical, optical investigation and nanocrystals analysis

Abstract

TiO_x/carbon composite nanosheets were synthesized by a “pulsed solution plasma” technique in a Ti-contained solution. The discharge characteristics of the plasma were investigated via the resistance of the plasma channel from current and voltage profiles as a function of capacitor energy. Furthermore, the Ti-based oxides were characterized in detail using high-resolution transmission electron microscopy and selected area electron diffraction. Finally, the formation mechanism of the product was discussed through optical emission spectroscopy. The analyses revealed that the products were composed of spherical nanoparticles of TiO, TiO₂ rutile phases and polycrystalline anatase TiO₂/carbon composite nanosheets, which were formed by plasma or plasma-induced reactions between species in Ti-contained solution.

---

Introduction

The synthesis of oxide nanomaterials is a common and significant interest because the nanomaterials have been utilized scientifically and industrially in various fields over the past two decades.^1–4 Furthermore, with increasing production and consumption amounts of nanomaterials, the offer of competitive price became important.⁵ Therefore, a considerably more attractive issue is to develop the synthesis method of nanomaterials, which is cost-effective and capable of easy scale-up. Diverse liquid-phase synthesis methods have been used for the synthesis of the oxide nanomaterials such as co-precipitation,⁶ polymerizable complex method,⁷ hydrothermal,⁸ and sol–gel processing.⁹ However, most of these methods lead to high electric energy consumption in the aspect of processing to maintain high temperature or pressure for a long time. On the other hand, the pulsed solution plasma method can be a promising candidate for cost-saving due to the plasma characteristics of high energy density resulting from the concentration of energy during tens of μsec.^10,11

Pulsed solution plasma technique has been already utilized as a general method for the synthesis of nanomaterials. For example, the nanomaterials of Ag,¹² Al₂O₃,¹³ Cu,¹⁴ Cu₂O,¹⁴ CuO,¹⁴ and ZrO₂ (ref. 15) were synthesized by the arc discharge of a metal electrode in aqueous solution. Moreover, carbon-based nanocomposites or carbon-based nanomaterials, such as carbon-encapsulated Co, Fe NPs,^16,17 carbon onions,¹⁸ and carbon nanotubes,¹⁹ were synthesized by the arc discharge of a metal electrode in ethanol or carbon electrode in water. Furthermore, arc discharge using a liquid precursor as a metal source to synthesize Au NPs and Fe₂O₃, FeO(OH) NPs was reported.^20,21 Nevertheless, the synthesis of oxide nanomaterials and its systematic investigation with a liquid precursor have almost not been studied. In this study, we, with a home-made solution plasma system, synthesized titanium oxide nanocrystals–carbon composites using a liquid precursor and accomplished the research on their formation process. This solution plasma technique is expected to provide the general synthesis method for metal oxide–carbon composite nanomaterials in various areas such as lithium ion battery applications.

Experimental

Precursor preparation

All of the chemicals were of analytical grade and were used without further purification. A Ti-contained sol was prepared as follows. Titanium tetraisopropoxide (TTIP, Aldrich, 97%) was dissolved in ethanol (OCI company Ltd., 99.9%), and the resulting solution was hydrolyzed and stabilized by adding a mixture of distilled water, nitric acid, and ethanol. The molar ratio of TTIP/H₂O/HNO₃ was 1 : 2 : 0.08.

Arc discharge setup

Fig. 1 shows a schematic diagram of the pulsed solution plasma system, which consists of a high-voltage power supply, high-voltage capacitor and discharge chamber. The spherical discharge chamber was filled with the Ti-contained sol and the tungsten electrodes of needle to plane geometry were immersed in the solution. The radius of the anode was 1.5 mm and the electrode gap distance was 5 mm. Five capacitors with capacitance of 0.1 μF (Maxwell Laboratories Inc.) were connected in parallel, and the capacitors with equivalent capacitance of 0.5 μF were charged by a positive polarity DC power supply (Glassman PK sires, 50 kV, 80 mA) with maximum charging voltage of 19 kV. After the discharge switch was closed, a few kA discharge current flowed through the electrode gap in the reaction chamber, generating a current channel accompanied by light and explosive sound. In the experiment, the number of discharge was fixed at 100 times, and the stored energy per pulse in capacitor was changed from 10 J to 90 J.

Fig. 1 A schematic diagram of the pulsed solution plasma system.

The inter-electrode voltage was measured using a 1000 : 1 high voltage probe (Tektronix, P6015A) attached to the anode and cathode. The discharge current was measured using a Pearson current probe (Model 301X) with sensitivity 0.01 V/A, which was connected to attenuators (Tektronix, 011-0069-02, 011-0060-02) with total attenuation factor 1/10. The voltage and current signal were transmitted to a computer through a GPIB cable from the oscilloscope (Tektronix, TDS 3014). The pressure of the shock wave caused by a high current electrical discharge was measured utilizing a piezoelectric pressure gauge (PCB, S113B23) at the distance of 55 mm from the center of the discharge chamber.

Optical emission spectroscopy (OES)

The optical fiber was installed at about 130 mm from a quartz window of 5 mm thickness and 28 mm diameter. The optical fiber was connected to a monochromator (Princeton instruments, Acton standard series SP2500), which has a CCD resolution of 0.09 nm and scan wavelength range of 300–1400 nm. CCD exposing time was fixed to 1 second for collecting emitted light completely. In addition, a neutral density filter was installed in front of the optical fiber for reducing the intensity of the light. All of experiments were conducted in 300 g mm⁻¹ gratings.

Nanocrystals characterization

After a hundred times of discharge was completed, the solution in the chamber was collected, centrifuged, washed several times with ethanol and the resultant black precipitate was finally freeze-dried. The crystal structures of the prepared powders were identified using an X-ray diffractometer (XRD, D8-advance, Bruker). The morphologies and micro-structures were investigated with field-emission scanning electron microscopy (FESEM, SUPRA 55VP, Carl Zeiss). The high-resolution transmission electron microscopy (HRTEM) images and selected area electron diffraction (SAED) patterns were taken with a JEOL JEM-3000F microscope at an accelerating voltage of 300 kV.

Results and discussion

In this experiment, for capacitor energies of <10 J, streamer discharge was mainly generated in between a pin-to-plate electrode system in liquid, which is proved from discharge images (Fig. S1†). In the corresponding images, no conductive path was formed. On the other hand, for capacitor energies of ≥10 J, the occurrence of arc discharge was identified, showing that the bright conductive path reached the plate electrode (Fig. 2 and S2†). Furthermore, the formation of the damped sinusoid waveforms of current and voltage with a sharp voltage drop after a breakdown also proved the arc discharge (Fig. 3).²² When the arc channel is formed inside the electrode gap, the high-voltage and ground electrode are electrically connected with the ionized particle in the plasma and the transferred energy from the capacitor is concentrated locally inside the arc channel. Therefore, the capacitor energies for the synthesis of nanocrystals were set to ≥10 J.

Fig. 2 The resistance of the plasma channel in solution as a function of capacitor energy. The inset is the discharge images with an exposure time of 1 s.

Fig. 3 Current and voltage waveforms of arc discharge for capacitor energy of 30 J.

The resistance of the plasma channel in solution as a function of capacitor energy was measured to identify discharge behavior as shown in Fig. 2. Moreover, the inset figures represent the discharge images for capacitor energies of 10 J and 30 J. The voltage measured between high-voltage and ground electrodes is expressed into the sum of inductance and resistance components as eqn (1).²³

V(t) = L(t)(di/dt) + R(t)i(t) (1)

where V(t) and i(t) describe the measured voltage and current between the electrodes, L(t) and R(t) are inductance and resistance associated with electrodes and arc channel, respectively. However, at the condition of maximum current (di/dt = 0), the inductance component can be eliminated. Moreover, the resistance of the tungsten electrode can be neglected relative to the resistance of the plasma channel due to high conductivity. Therefore, the resistance of the plasma channel was calculated simply by dividing voltage by current, especially using the first peak of the damped oscillation.

The resistance value was approximately 0.25 Ω and 0.1 Ω for capacitor energies of 10 J and ≥30 J, showing a drastic decrease between 10 J and 30 J. Considering that the resistance means the opposition to the passage of electric current through the channel, the capacitor energies of ≥30 J are thought to be enough to overcome the opposite barrier completely. This trend corresponded with the experimental observation where the color of the liquid precursor changed into black after 100 times of discharge in the experiments of only 30–90 J except for 10 J. Therefore, it is assumed that high energy electrons, which make the precipitation reaction possible, were generated for the capacitor energies of ≥30 J through arc discharge. Moreover, the channel size was fairly widened as the capacitor energy increased from 10 J to 30 J (Fig. 2), which is attributable to the higher plasma energy density of the channel, followed by introducing higher work against the surrounding medium and a change in the internal energy of plasma.²⁴ Fig. S3† shows that the shock wave pressure increases as the capacitor energy is increased. Higher pressure will be helpful for crystallization of precipitates. As a result, high temperature and pressure caused by these effects can help the precipitation reactions to proceed.

Fig. 4a–d show the typical TEM image, SAED pattern, HRTEM image and EDS mapping image, respectively, of TiO₂/carbon composite nanosheets synthesized by pulsed solution plasma at capacitor energy of 90 J. As shown in Fig. 4a, the thin, rippled and dented sheets are randomly scrolled and entangled. The SAED pattern (Fig. 4b) of the nanosheets displays sets of diffuse rings, which could be completely assigned to all of the reflection peaks of TiO₂ anatase phase, representing the poly-crystalline nature of the anatase phase. In addition, the innermost diffuse ring was considered to be a superposition of two sets of reflections: anatase (101) and graphitic disordered carbon (002) planes (d₁₀₁ of anatase = 0.352 nm, d₀₀₂ of graphite = 0.338 nm).²⁵ The existence of carbon was also revealed from the EDS mapping image (Fig. 4d). The polycrystalline nature of the sheet could be identified in Fig. 4c from the result that grains having anatase (101) planes of d-spacing 0.35 nm were distributed with various directions.

Fig. 4 (a) TEM image, (b) SAED patterns, (c) HRTEM image and (d) EDS mapping image of TiO₂ anatase (A)/carbon composite nanosheets synthesized by pulsed solution plasma.

However, aside from the nanosheets, there also exist spherical nanoparticles in all of the experimental samples. The nanoparticles were proved to be a mixture of TiO and TiO₂ rutile phase (mainly TiO) with a mean size of ∼7 nm from the HRTEM, SAED, and EDS mapping analysis (Fig. 5). The TiO nanoparticles are considered to be formed under the condition of temperature quenching due to the nature of solution plasma discharge.²⁶

Fig. 5 (a) TEM image, (b) SAED patterns, (c) HRTEM image and (d) EDS mapping image of TiO (T), TiO₂ rutile (R) nanoparticles synthesized by pulsed solution plasma. The inset in Fig. 5c is the FFT patterns corresponding to the observed nanoparticles.

To investigate the effect of discharge number on products, the number of arc discharges increased from 100 to 500 times. Fig. 6 represents XRD patterns and FESEM images of the synthesized products after the pulsed plasma discharge of 100, 300, and 500 times for the capacitor energy of 90 J. The XRD patterns (Fig. 6a) show that all products are composed of TiO, TiO₂ anatase and TiO₂ rutile phases. As the number of discharge is increased, the amount of TiO₂ anatase and rutile crystallites is increased relatively compared to TiO crystalline. No trace of secondary oxides from tungsten electrodes was observed in the final products. It is reasonable in that continuous temperature and pressure stimuli prefer to generate a more stable phase rather than meta-stable TiO. Moreover, as shown in Fig. 6b–d, 300 and 500 times discharged samples have abundant and huge nanosheets compared to the 100 times discharged sample. As a result, the products form appropriate aggregations of nanoparticles and nanosheets. Consequently, we could obtain Ti oxide/carbon composites by controlling the number of pulsed plasma discharges.

Fig. 6 (a) XRD patterns, (b–d) FESEM images of products after the pulsed plasma discharge of 100, 300, and 500 times for the capacitor energy of 90 J. Symbols A, T and R indicate anatase, TiO and rutile phases, respectively.

To discuss the formation mechanism of the product, the optical emission spectra were measured during pulsed plasma discharge. Fig. 7a shows the normalized spectra discharged for the capacitor energies of 10–90 J, and the inset represents the magnified view in the range of 410–590 nm to identify species in detail. For all energy, the spectra showed no sharp emission lines but broad peak distribution. This phenomenon was attributed to short discharge on-time, which causes high plasma density and the merging of spectral lines.²⁷ Therefore, it is difficult to assign the spectra with a high resolution, but it can be possible to identify the existence of specific elements qualitatively with distinct heads of merging lines. The spectral lines and emission bands were identified using the reference data.^28,29 The spectrum of 10 J is similar to the spectrum of water arc discharge, which can be explained by the dissociation of the water molecule, by hydrogen, oxygen molecular and atomic excitation.²¹ Moreover, excited hydrogen was dominant in the spectrum because less energy was required to be excited compared to oxygen.³⁰ As the charged energy in the capacitor was increased from 10 J to 30–90 J, the spectrum changed drastically. Specifically, the intensity of the H₂O molecule emission band (λ = 692.2, 716.4, 809.7 nm) and O emission lines (λ = 777.2, 844.6 nm) became larger compared to the intensity of the H_α emission line (λ = 656.3 nm). This is meaningful as the reactive oxygen can be a source for the formation of oxide materials.²¹ Furthermore, the marked change was observed from the magnified view of the spectra in the inset as the capacitor energy increased. The spectrum for the capacitor energy of 10 J was composed of broadened H emission lines (λ = 486.1, 434.0 nm), swan bands system of the C₂ molecule (λ = 473.7, 516.5, 558.5 nm) and G → B system of the H₂ molecule (λ = 453–464 nm). However, the broadened H emission lines were dominant, whereas C₂ and H₂ emission bands were weak. As the capacitor energy increased, a shallow hole near the wavelength of 465 nm was filled as a result of the increase of the intensity of the C₂ and H₂ emission bands. These changes were attributed to the decomposition of ethanol molecules by plasma arc discharge. The presence of Ti species in the plasma channel was also detected via the comparison between the spectra of 10 J and 30–90 J, showing broadened Ti(IV) emission, but almost no Ti(I) emission.

Fig. 7 (a) Optical emission spectra from the pulsed plasma for different capacitor energies (10–90 J). The inset represents the magnified view in the range of 410–590 nm. (b) Optical emission spectra from the repeated pulsed plasma for the capacitor energy of 70 J.

The role of the reactive oxygen was also proved from the repeated pulsed plasma for the capacitor energy of 70 J, as shown in Fig. 7b. The optical emission spectra were normalized to identify the ratio between O and H_α emission lines. Although the blackening of solution distorted the spectrum of the visible range, the intensity of the H_α emission line became noticeable relatively with increasing discharge number. On the other hand, the relative emission intensity of O decreases compared to that of H_α, representing that reactive oxygen functioned as a source for the formation of oxide, and the amount decreased as oxygen was consumed in the reaction.

In conclusion, it is thought that the excited Ti ions reacted with the reactive oxygen produced by the decomposition of ethanol molecules and Ti-complex to form the TiO phase. The sequential process of plasma plume formation, expansion under the presence of a liquid compression and quenching of the plume by arc discharge can provide the extreme non-equilibrium conditions for the formation of the disordered phase.²⁶ Moreover, the TiO₂/carbon composite nanosheets are considered to be obtained via the formation of carbon sheets by the decomposition and quenching of ethanol, followed by the condensation reaction of Ti complex onto carbon sheets.^31,32 It is reasonable as the homogenous precipitation of TiO₂ is prone to lead to the formation of spherical particles. In addition, the decomposition of ethanol molecules in plasma and quenching at the plasma–liquid interface surrounded by gas bubbles can provide carbon sheets.³³

Conclusions

TiO_x/carbon composite nanosheets were synthesized by pulsed solution plasma in Ti-contained sol. The channel resistance and discharge images of the plasma, which were investigated as a function of capacitor energy, revealed that the capacitor energy more than the specific value is required to induce the precipitation reactions resulting from the generation of high plasma energy density and high energy electron. Moreover, the HRTEM and SAED analysis revealed that the products were composed of spherical nanoparticles of TiO, TiO₂ rutile phases and anatase TiO₂/carbon composite nanosheets. However, appropriate Ti oxide/carbon composite nanomaterials could be obtained by controlling the number of plasma discharges. Finally, the OES analysis displayed that the Ti-based oxides were formed by the reaction between the excited Ti ions and reactive oxygen, which were dissociated products of ionized Ti species and ethanol molecules, respectively. In particular, the formation of the nanosheets was attributed to the solidification of carbon sheets from the decomposition of ethanol molecules and forced the condensation of Ti-complex on the carbon sheets.

Acknowledgements

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and future Planning (2012R1A2A2A01045382, 2008-0061900).

Notes and references

1. M. M. Khin, A. S. Nair, V. J. Bahu, R. Murugan and S. Ramakrishna, Energy Environ. Sci., 2012, 5, 8075.
2. G. A. Ozin and L. Cademartiri, Small, 2009, 5, 1240.
3. G. R. Patzke, Y. Zhou, R. Kontic and F. Conrad, Angew. Chem., Int. Ed., 2011, 50, 826.
4. C. Zhao, L. He, S. Z. Qiao and A. P. J. Middelberg, Chem. Eng. Sci., 2011, 66, 1463.
5. N. Osterwalder, C. Capello, K. Hungerbuhler and W. J. Stark, J. Nanopart. Res., 2006, 8, 1.
6. H. Wei, Y. Wu, N. Lun and F. Zhao, J. Mater. Sci., 2004, 39, 1305.
7. K. Iizuka, T. Wato, Y. Miseki, K. Saito and A. Kudo, J. Am. Chem. Soc., 2011, 133, 20863.
8. Z. Y. Yuan and B. L. Su, Colloids Surf., A, 2004, 241, 173.
9. S. W. Lee, I.-S. Cho, J. H. Lee, D. H. Kim, D. W. Kim, J. Y. Kim, H. Shin, J.-K. Lee, H. S. Jung, N.-G. Park, K. Kim, M. J. Ko and K. S. Hong, Chem. Mater., 2010, 22, 1958.
10. F. Fang, J. Kennedy, E. Manikandan, J. Futter and A. Markwitz, Chem. Phys. Lett., 2012, 521, 86.
11. W. G. Graham and K. R. Stalder, J. Phys. D: Appl. Phys., 2011, 44, 174037.
12. C. H. Lo, T. T. Tsung and H. M. Lin, J. Alloys Compd., 2007, 434−435, 659.
13. D. Delaportas, P. Svarnas, I. Alexandrou, A. Siokou, K. Black and J. W. Bradley, J. Phys. D: Appl. Phys., 2009, 42, 245204.
14. W. T. Yao, S. H. Yu, Y. Zhou, J. Jiang, Q. S. Wu, L. Zhang and J. Jiang, J. Phys. Chem. B, 2005, 109, 14011.
15. L. Chen, T. Mashimo, E. Omurzak, H. Okudera, C. Iwamoto and A. Yoshiasa, J. Phys. Chem. C, 2011, 115, 9370.
16. R. Sergiienko, E. Shibata, Z. Akase, D. Shindo, T. Nakamura and G. Qin, Acta Mater., 2007, 55, 3671.
17. R. Sergiienko, E. Shibata, Z. Akase, H. Suwa, D. Shindo and T. Nakamura, J. Mater. Res., 2006, 21, 2525.
18. N. Sano, H. Wang, I. Alexandrou, M. Chhowalla, K. B. K. Teo and G. A. J. Amaratunga, J. Appl. Phys., 2002, 92, 2783.
19. I. Alexandrou, H. Wang, N. Sano and G. A. J. Amaratunga, J. Chem. Phys., 2004, 120, 1055.
20. J. Hieda, N. Saito and O. Takai, J. Vac. Sci. Technol., A, 2008, 26, 854.
21. M. A. Bratescu, N. Saito and O. Takai, Curr. Appl. Phys., 2011, 11, 530.
22. P. Vanraes, A. Nikiforov and C. Leys, J. Phys. D: Appl. Phys., 2012, 45, 245206.
23. M. J. Given, I. V. Timoshkin, M. P. Wilson and S. J. Macgregor, 17th IEEE International Pulsed Power Conference, 2009, p. 965.
24. A. I. Ioffe, J. Appl. Mech. Tech. Phys., 1966, 7, 69.
25. D. Babonneau, J. Briatico, F. Petroff, T. Cabioc’h and A. Naudon, J. Appl. Phys., 2000, 87, 3432.
26. N. G. Semaltianos, S. Logothetidis, N. Frangis, I. Tsiaoussis, W. Perrie, G. Dearden and K. G. Watkins, Chem. Phys. Lett., 2010, 496, 113.
27. A. Descoeudres, Ch. Hollenstein, G. Walder, R. Demellayer and R. Perez, Plasma Sources Sci. Technol., 2008, 17, 024008.
28. R. W. B. Pearse and A. G. Gaydon, The identification of molecular spectra, Chapman & Hall, A Halsted Press Book, John Wiley & Sons, New York, 4th edn, 1976.
29. Y. Ralchenko, A. E. Kramida and J. Reader, NIST Atomic Spectra Database, Version 4, National Institute of Standards and Technology, Gaithersburg, 2010.
30. Y. K. Heo and S. Y. Lee, Met. Mater. Int., 2011, 17, 943.
31. T. Sugimoto, X. Zhou and A. Muramatsu, J. Colloid Interface Sci., 2002, 252, 339.
32. H. Wang, J. T. Robinson, G. Diankov and H. Dai, J. Am. Chem. Soc., 2010, 132, 3270.
33. A. Dato, V. Radmilovic, Z. Lee, J. Phillips and M. Frenklach, Nano Lett., 2008, 8, 2012.

---

Footnote

† Electronic supplementary information (ESI) available: Discharge images for capacitor energy of <10 J, 50 J, 70 J and 90 J. Pressure of shock wave as a function of capacitor energy. See DOI: 10.1039/c4ra15251d

---