search for


Unexpected Chemical and Thermal Stability of Surface Oxynitride of Anatase TiO2 Nanocrystals Prepared in the Afterglow of N2 Plasma
Applied Science and Convergence Technology 2017;26:62-65
Published online July 31, 2017;
© 2017 The Korean Vacuum Society.

Byungwook Jeona, Ansoon Kimb, and Yu Kwon Kima,*

aDepartment of Chemistry and Department of Energy Systems Research, Ajou University, Suwon 16499, South Korea, bKorea Research Institute of Standards and Science (KRISS), Gajeong-ro 267, Yuseong-gu, Daejeon 34113, South Korea
Correspondence to: E-mail:
Received June 13, 2017; Accepted July 3, 2017.
cc This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Passivation of surface defects by the formation of chemically inert structure at the surface of TiO2 nanocrystals can be potentially useful in enhancing their photocatalytic activity. In this regard, we have studied the surface chemical states of TiO2 surfaces prepared by a treatment in the afterglow of N2 microwave plasma using X-ray photoemission spectroscopy (XPS). We find that nitrogen is incorporated into the surface after the treatment up to a few atomic percent. Interestingly, the surface oxynitride layer is found to be chemically stable when it’s in contact with water at room temperature (RT). The surface nitrogen species were also found to be thermally stable upon annealing up to 150°C in the atmospheric pressure. Thus, we conclude that the treatment of oxide materials such as TiO2 in the afterglow of N2 plasma can be effective way to passivate the surface with nitrogen species.

Keywords : microwave plasma, afterglow, surface oxynitride, N2, TiO2
I. Introduction

As a chemically stable, nontoxic material, TiO2 has been studied for various applications including photocatalysis [1], photovoltaics [2] and sensors [3]. In such diverse applications, the functionality of TiO2, the ultimate performance of TiO2 is largely determined by the surface chemical structure as can be seen from the enhanced photocatalytic activity of surface-modified TiO2-based photocatalysts [4,5].

As a means of controlling the surface chemical structure, plasma treatments have been shown to be quite effective in engineering the photoresponse of TiO2 [6,7]. Especially, N2 plasmas may induce the formation of chemically modified structures such as the reduced phases (e.g., TiO2-x) [7] and the nitride phases [8,9] for enhanced absorption of visible light. However, the facile diffusion of nitrogen species into the bulk of TiO2 under typical treatment conditions at elevated temperatures may pose a technological limit in ‘surface-selective’ modification. Here, we developed a new method of surface modification of TiO2 nanocrystals in the afterglow or the remote plasma of N2 microwave discharges, which is the post-discharge region of N2 plasmas with relative low density of ionic species and low gas temperatures. In the N2 afterglows, the substrate temperature rise is limited to about 350 K and the formation of nitride layers is limited to the surface due to the suppression of thermal diffusion into the bulk.

In addition, the active nitrogen species in the N2 plasmas and their afterglows can be controlled by changing the process parameters such as pressure, flow rate and the applied power [1017]. Various excited states of neutral N2, N2+ ions and atomic N species are generated in the N2 plasmas and in their afterglows under moderate operating conditions [10,11,15,16,18] and their widely different lifetimes make the densities of the excited species in the afterglow region change along the downstream of the gas flow. In general, neutral species such as N atoms, N2(X,ν > 13) and N2(A) are more populated over ionic N2+ species as the gas flows away from the discharge region in the afterglows [11,16,19]. Such high densities of neutral active species in the afterglows can be potentially advantageous in a damage-free selective surface chemical modification with nitrogen.

In this study, we performed a surface modification of TiO2 nanocrystals in the afterglows of N2 microwave plasma for a potential application in photocatalysis in mind. After the surface treatment in the afterglow, we find that the nitrogen species left on the surface of TiO2 are chemically inert and stable under thermal treatment. It was unexpected since the surface nitrogen species could be labile toward desorption by formation of volatile species such as NO and NH3 in the ambient condition with molecular oxygen and moisture. To the contrary, we find chemically inert nitrogen species on the surface are present even after several months in the air. By analyzing N 1s core level spectra, we identified the chemical bonding states of the nitrogen species at the surface.

II. Experimental Details

Anatase TiO2 nanocrystals have been synthesized by a hydrothermal method starting from tetrabutyl titanate as a precursor and HF as a shape-controlling agent as described elsewhere [5]. The as-synthesized TiO2 powder is then dispersed in a solution of deionized water and ethanol for a spin-coating on a clean Si substrate. The SEM image of the film of the TiO2 nanocrystals on Si shown in the SEM image of Fig. 1(a) indicates that the film is made of aggregates of thin square-shaped TiO2 nanocrystals (nanosheets) randomly dispersed on the Si substrate up to about 10 μm.

Fig. 1(b) shows a schematic drawing of our experimental setup for the surface treatment of TiO2 in the afterglow of N2 microwave plasmas. A discharge quartz tube with an inner diameter (ID) of 5 mm is inserted to the surfatron cavity for the microwave (2.45 GHz) plasma generation. Then the discharge tube is connected to a quartz tube with a larger ID of 18 mm for the generation of N2 afterglows [20]. With a rotary pumping and a control of gas flow (0.5–2.0 slm or standard liter per minute) using a mass flow controller (MFC), a steady flow of pure N2 gas through the tube is generated with a constant pressure in the range of 1–15 Torr. For the present set of experiment, the N2 plasma was generated at the N2 flow of 1.8 slm, the steady operating pressure of 8 Torr and the microwave power of 150 W. At this operating condition, the active species in the surface treatment region are characterized by a rather high N2+ density (1.3 × 1011 cm−3) as characterized by the emission of 391 nm (N2+ 1st neg.) [21] in addition to N atoms (~1 × 1015 cm−3). During the surface treatment (for about 8 hours), the substrate temperature was raised up to 350 K.

The X-ray photoelectron spectroscopy (XPS) measurements were carried out with a PHI5000 Versa Probe II (Ulvac-PHI) using a monochromatic Al Kα source operating at a base pressure below 4 × 10−10 Torr. The instrument is well calibrated with the peak positions of Au 4f7/2, Ag 3d5/2 and Cu 2p3/2 core levels of sputter-cleaned Au, Ag and Cu films and provides a small X-ray beam diameter (200 μm) for a space-resolved analysis. A dual type charge neutralizer is used to minimize an undesirable charging effect. The binding energy is calibrated by referring the C 1s peak from ubiquitous hydrocarbon contamination to 284.8 eV.

III. Results and Discussion

After the surface treatment in the post-discharge region, the sample was stored in the air prior to be introduced into the UHV chamber for the XPS analysis. Here we describe our XPS analysis on the chemical bonding structures of the nitrogen species generated by the treatment.

Fig. 2 shows N 1s core level spectra taken from the plasma-treated TiO2 sample for 8 hours. One can easily observe characteristic N 1s features (N1–N3) located around 400 eV from the as-prepared TiO2 which is treated in the afterglow of the N2 microwave plasma. Detailed fitting analysis reveals that the peak at 400 eV (N2) can be resolved into additional N components at 402 eV (N1) and 398.2 eV (N3); the presence of N1 and N3 is evident from the spectral shape of the raw spectra. In addition, we can find another peak at 396 eV which is assigned to substitutional N species (Ns) which is N incorporated into the lattice oxygen sites by displacing the oxygen atom [2224]. Despite the diffusion of N into the bulk is suppressed by the limited temperature rise under our exposure conditions, surface N species may be incorporated into the subsurface lattice during the prolonged exposure in the afterglows.

Interestingly enough, all the N species present at or near the surface of our surface-modified TiO2 seem to be stable under the subsequent post treatments such as rinsing with deionized water and annealing up to 150°C. We find a little change in the spectral shape in the N 1s core level spectra after the treatments. Only the intensity at 400 eV decreases after the rinsing in water, but the decrease in intensity is only about 20% or less. The change in intensity after the annealing is even small (~10%). Ns is even more stable and its intensity seems to be enhanced after the annealing.

In Fig. 2, a mild Ar+ sputtering is shown to induce a removal of nearly all the N 1s features from the sample. The trace amount of remanent N 1s features is attributed to the inherent morphology of the film which is composed of aggregates of TiO2 nanocrystals. This observation again corroborates the fact that all the N species induced by the plasma treatment are suggested to be present at or near the surface.

The atomic percentage of nitrogen species is estimated to be about 2.5 at%. Compared to the atomic percentage of Ti, the N/Ti ratio is obtained to be 0.04. Thus, the net concentration of N species is low. The low N content at the surface can be in part due to the aging effect in the air which may eliminate any labile surface nitrogen species.

The surface oxynitride layer with a few N atomic percentage in the N2 afterglows is proposed to be formed by the following processes. The most dominant active species in the afterglow is the atomic nitrogen (~x1015/cm2). The collision of N atoms with the TiO2 surface induce the formation of a direct N-O bond with the surface oxygen. Then, the surface-bound NO may desorb from the surface under the reaction condition at RT due to the low binding energies of NO on TiO2 surface [2527]. Desorption of NO leaves oxygen vacancy at the surface, which then accepts another N atom from the gas phase to form a direct Ti-N bond. This surface N species may further react at the surface to transform into stable bonding configurations to produce N species at around 400 eV (N1–N3) in the photoemission spectra in Fig. 2. The bonding structures of N1–N3 are generally assigned to bonding configurations such as O-Ti-N linkages [28,29], Ti-O-N-O and Ti-O-N-N [30]. We suggest that such bonding species can be formed at or near the surface of TiO2 due to the low substrate temperature (~350 K) of our treatment condition.

Fig. 3 shows our XPS spectra of Ti 2p and O 1s core levels from our TiO2 samples. A stoichiometric TiO2 is usually characterized by the Ti 2p3/2 peak at 459 eV and the O 1s peak at 530 eV. Our as-treated TiO2 sample also shows the same spectral shape of the Ti 2p3/2 peak at 459 eV, indicating that bulk Ti4+ of TiO2 remains to be the main component of the Ti 2p core level even after the surface modification in the afterglows. It can be attributed to the fact that only the surface or subsurface layer is chemically modified with nitrogen without any significant reduction of the bulk TiO2 under the treatment in the afterglows. O 1s core level is also characterized by a single O 1s peak at 530 eV. It represents a bulk oxygen component of TiO2. The peak usually shows an asymmetric shape due to a tail toward higher binding energies that are typically assigned to the surface oxygenates such as OH and H2O which are readily formed on oxide surfaces under the ambient condition. All such characteristic features of Ti 2p and O 1s core levels show negligible changes upon the post-treatments of the rinsing in water and the annealing. Even after the mild Ar+ sputtering, we cannot find any noticeable change in the spectral shape and intensity of Ti 2p and O 1s core levels, which confirms again that the sputtering condition is mild enough not to induce any reduction of bulk Ti. This fact in turn infers that the surface oxynitride layer is confined to the surface or the subsurface layer of TiO2.

IV. Conclusions

As a means of surface passivation technology, a surface treatment of TiO2 nanocrystals in the afterglow of N2 microwave plasmas has been performed and the surface chemical states are carefully examined by XPS. We find that nitrogen is incorporated into the surface or the subsurface of TiO2 after the treatment up to a few atomic percent, suggesting that the treatments of oxide materials in the afterglows are promising in preparing selective oxynitride layers only at the surface. Even though the atomic percentage of nitrogen is low, the surface oxynitride layer is found to be chemically stable even in the water at room temperature (RT) and is thermally stable upon annealing up to 150°C in the atmospheric pressure. Thus, we conclude that the afterglows of N2 plasmas are very effective in selective surface passivation of oxide materials with nitrogen species.


This research was supported by the International Research & Development Program of the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (NRF-2015K1A3A1A21000248).

Fig. 1. (a) A SEM image of anatase TiO2 nanocrystals spin-coated on a Si wafer and (b) the experimental scheme used in the present study for the surface treatment of TiO2. In the pictures of our sample before and after the plasma treatment, we also find a color change after the treatment for a few hours.
Fig. 2. N 1s core level spectra taken from the surface-modified TiO2 sample. Compared to the as-prepared TiO2, the spectral shape and the intensity of the N 1s core level features remain to be affected little by the post-treatments such as rinsing in water at RT and annealing up to 150°C, but are significantly changed after a mild Ar+ sputtering.
Fig. 3. O 1s and Ti 2p core level spectra taken from the surface-.modified TiO2 nanocrystals before and after the post-treatments such as rinsing in water at RT, annealing up to 150°C and a mild Ar+ sputtering for 2 minutes.
  1. Schneider, J, Matsuoka, M, Takeuchi, M, Zhang, J, Horiuchi, Y, Anpo, M, and Bahnemann, DW (2014). Chem Rev. 114, 9919-9986.
    Pubmed CrossRef
  2. Wei, Q, Hirota, K, Tajima, K, and Hashimoto, K (2006). Chem Mater. 18, 5080-5087.
  3. Nisar, J, Topalian, Z, De Sarkar, A, Österlund, L, and Ahuja, R (2013). ACS Appl Mater Interfaces. 5, 8516-8522.
    Pubmed CrossRef
  4. Leshuk, T, Parviz, R, Everett, P, Krishnakumar, H, Varin, RA, and Gu, F (2013). ACS Appl Mater Interfaces. 5, 1892-1895.
    Pubmed CrossRef
  5. Yu, X, Jeon, B, and Kim, YK (2015). ACS Catal. 5, 3316-3322.
  6. Islam, SZ, Reed, A, Kim, DY, and Rankin, SE (2016). Microporous Mesoporous Mater. 220, 120-128.
  7. Li, B, Zhao, Z, Zhou, Q, Meng, B, Meng, X, and Qiu, J (2014). Chem Eur J. 20, 14763-14770.
  8. Liu, C, Ma, Z, Li, J, and Wang, W (2006). Plasma Sci Technol. 8, 311.
  9. Islam, SZ, Reed, A, Wanninayake, N, Kim, DY, and Rankin, SE (2016). J Phys Chem Solids. 120, 14069-14081.
  10. Ricard, A, Sarrette, J-P, Oh, S-G, and Kim, YK (2016). Plasma Chem Plasma Process, 1-12.
  11. Ricard, A, Oh, SG, Jang, J, and Kim, YK (2015). Curr Appl Phys. 15, 1453-1462.
  12. Zerrouki, H, Ricard, A, and Sarrette, JP (2014). Contrib Plasma Phys. 54, 827-837.
  13. Ricard, A, and Oh, SG (2014). Plasma Sources Sci Technol. 23, 045009.
  14. Afonso Ferreira, J, Stafford, L, Leonelli, R, and Ricard, A (2014). J Appl Phys. 115, 163303.
  15. Zerrouki, H, Ricard, A, and Sarrette, JP (2013). Contrib Plasma Phys. 53, 599-604.
  16. Ricard, A, Oh, SG, and Guerra, V (2013). Plasma Sources Sci Technol. 22, 035009.
  17. Kang, N, Lee, M, Ricard, A, and Oh, S-g (2012). Curr Appl Phys. 12, 1448-1453.
  18. Ricard, A, Zerrouki, H, and Sarrette, J (2015). J Anal Sci Methods and Instrumentation. 5, 59-65.
  19. Boudam, MK, Saoudi, B, Moisan, M, and Ricard, A (2007). J Phys D: Appl Phys. 40, 1694-1711.
  20. Zerrouki, H, Ricard, A, and Sarrette, JP (2014). J Phys Conf Ser. 550, 012045.
  21. Ricard, A, Sarrette, J-P, Jeon, B, and Kim, YK (2017). Curr Appl Phys. 17, 945-950.
  22. Kim, YK, Park, S, Kim, K-J, and Kim, B (2011). J Phys Chem C. 115, 18618-18624.
  23. Batzill, M, Morales, EH, and Diebold, U (2006). Phys Rev Lett. 96, 026103.
  24. Wang, J, Tafen, DN, Lewis, JP, Hong, Z, Manivannan, A, Zhi, M, Li, M, Wu, N, and Am, J (2009). Chem Soc. 131, 12290-12297.
  25. Kim, B, Dohnálek, Z, Szanyi, J, Kay, BD, and Kim, YK (2016). Surf Sci. 652, 148-155.
  26. Sorescu, DC, Rusu, CN, and Yates, JT (2000). J Phys Chem B. 104, 4408-4417.
  27. Rusu, CN, and Yates, JT (2000). J Phys Chem B. 104, 1729-1737.
  28. Sathish, M, Viswanathan, B, Viswanath, RP, and Gopinath, CS (2005). Chem Mater. 17, 6349-6353.
  29. Chen, X, and Burda, C (2004). J Phys Chem B. 108, 15446-15449.
  30. Peng, F, Cai, L, Huang, L, Yu, H, and Wang, H (2008). J Phys Chem Solids. 69, 1657-1664.

July 2017, 26 (4)