Crucial role of reactive pulse-gas on a sputtered Zn₃N₂ thin film formation

Abstract

Herein, we demonstrate a powerful technique, known as reactive gas-timing (RGT) rf magnetron sputtering, to fabricate high quality Zn₃N₂ thin films at room temperature without applying any additional energy sources. A single phase of Zn₃N₂ film formation can only be obtained when a reactive pulse-gas of N₂ is utilized. We find that selecting a small atomic mass of sputtered reactive gas coupled with the pulse-gas technique is very crucial to adjust the number of sputtered atoms obtained from the target and enrich the forming energy of the sputtered Zn₃N₂ films during the deposition process. Our results highlight that the RGT technique is a promising method to fabricate high quality sputtered compound thin films that can be applied in flexible devices. A simplified model of the materials system at the surface region of the de-nitride Zn₃N₂ during ion bombardment is also presented.

---

Introduction

Zn₃N₂ thin film is a very attractive group II–V compound with a high potentiality for photonic and electronic devices due to its high mobility coupled with direct band gap.^1–5 Likewise, an interesting property of Zn₃N₂ is the possibility of transforming Zn₃N₂ into p-type ZnO upon thermal annealing,^6–10 which will have major implications for the semiconductor industry. However, the difficulties in preparing high quality Zn₃N₂ are still challenging issues because the stoichiometric nitride is a provision of nitrogen having high chemical reactivity.¹¹ Such controllability and reproducibility of the chemical composition for Zn₃N₂ formation requires a high substrate temperature (>300 °C).^2,5,12,13 Furthermore, it is difficult to control the re-evaporation of metallic Zn atoms and/or clusters in such high temperature processes, which directly affects the thickness of the films due to their high vapor pressure characteristic.^13,14 Although several research groups have demonstrated low temperature methods (i.e. substrate temperature ∼ 150 °C) to fabricate Zn₃N₂ thin films using reactive rf magnetron sputtering, which is available to fabricate the Zn₃N₂ thin films via the provision of active nitrogen species from plasma.¹⁵ Such substrate heater sources enhanced the forming energy of Zn₃N₂ thin films increasing the process cost and cannot be used to fabricate sputtered Zn₃N₂ on flexible substrates. Recently, our group established a special technique, known as gas-timing (GT) rf magnetron sputtering, to control the texture orientation of Ag thin films at room temperature without applying any additional energy sources.¹⁶ The on–off sputtered gas sequence can be attributed to deposition mode and energy-assisted mode. Such alternated sequence not only allows us to adjust sputtered atoms and/or clusters from the target source but also to enhance the sputtered energy during thin film growth,¹⁶ which is possible to be used for sputtered compound materials such as oxide, nitride and oxynitride.^17–29 Herein, we demonstrate the utilization of reactive gas-timing (RGT) rf magnetron sputtering to fabricate Zn₃N₂ thin films at room temperature. By comparing with N₂–Ar mixing gas, Ar–N₂ reactive gas-timing (Ar–N₂-RGT) and continued N₂ gas flow, a single phase of Zn₃N₂ thin film can only be obtained when the reactive pulse-gas of N₂ is utilized. We found that the RGT technique not only enabled us to enrich the forming energy of thin film formation but also provides the ability to adjust the number of sputtered atoms from the target through the selection of a small atomic mass of sputtered reactive gas. In addition, the chemical composition, electrical and optical properties of the Zn₃N₂ films were investigated. Thus, our results highlight that the RGT technique is a promising method to fabricate high quality sputtered compound thin films at room temperature (RT) without applying any additional energy sources, which are applicable for flexible electronic devices.

Experimental

Fabrication of zinc nitride thin films

Zinc nitride (Zn₃N₂) thin films were fabricated on Si (100) and glass slides using reactive gas-timing (RGT) rf magnetron sputtering (ATC 200-F, AJA International, Inc.). A 2-inch diameter with a 0.25-inch thick zinc target (99.99%) was used as the sputtering target. The Si (100) and glass slide substrates were cleaned via an alcohol process in which the Si (100) substrates were ultrasonically cleaned with acetone, ethyl alcohol and deionized water for 10 min, respectively, and then dried under a flow of nitrogen. The samples were then transferred to a high vacuum chamber for thin film deposition. The distance between the target-to-substrate was set at 70 mm. The substrates were mounted on a rotational holder substrate, which was driven by a motor at a rotation speed of 10 rpm. High purity nitrogen (99.999%) was supplied as the sputtering gas. The flow rate of nitrogen gas was controlled with mass flow meters (MKS), while the pressure in the chamber was measured with Pirani and Penning gauges. When a base pressure reached 1.0 × 10⁻⁶ mbar, a constant flow of nitrogen was introduced to the chamber at 20 sccm. The rf power and working pressure for sputtering were 100 watts and 8.5 × 10⁻³ mbar, respectively. During the film fabrication, the RGT technique was utilized to fabricate Zn₃N₂ thin films via controlling the on–off sequences of nitrogen gas at specific temporal intervals in order to operate alternate gas flows into the vacuum chamber as shown in Fig. 1. In this work, the turn-on sequence of nitrogen gas was changed from 5 s to 90 s whereas the turn-off sequence was fixed at 5 s. The thickness of the Zn₃N₂ thin films was 300 nm. The fabricated samples were then investigated for the changes in structure, morphology, chemical composition, electrical and optical properties based on the gas-timing technique. It should be noted that the deposition processes were carried out at room temperature, i.e. the substrate was not heated during or after deposition. Details of all the conditions are summarized in Fig. 1.

Fig. 1 A schematic of reactive N₂ gas-timing rf magnetron sputtering and conditions for Zn₃N₂ thin film growth.

Characterization of the zinc nitride thin film

The crystal orientation of the Zn₃N₂ thin films was examined by X-ray diffraction (GIXRD, TTRAX III, Rigaku) using Cu Kα radiation. Atomic force microscopy (AFM, SEIKO SPA400) was employed to investigate the morphology of the Zn₃N₂ thin films. The thickness of the Zn₃N₂ thin films was measured using a step profiler (MITUTOYO SURFTEST). X-ray photoemission spectroscopy (XPS) was used to characterize the composition near the surface region. XPS spectra were measured at the Synchrotron Light Research Institute (Public Organization, Thailand) using BL3.2Ua (VG Scienta R4000, 600 eV of photon energy) for preliminary inspection and SUT-NOATEC-SLRI XPS (PHI 5000 VersaProbe II, monochromatized Al Kα radiation at 1486.6 eV, 100 micron beam diameter) for detailed analysis presented in this paper, respectively. The XPS depth profile was performed by repeated cycles of 1 min argon ion etching on the sample surface (3 × 3 mm² raster etching mode, 4 keV of ion energy, 7 mA of electron emission current) followed by XPS measurement. It is worth noting that ion etching was performed using an ion gun equipped with a differential pumping system. Thus, the vacuum pressure could be maintained in the 10⁻⁸ mbar range during ion etching. The electrical properties of the Zn₃N₂ thin films were obtained using Hall effect measurements, which utilized the four-probe Van der Pauw technique. Au electrodes were deposited at the four corners of each sample using the sputtering technique. The contacts presented ohmic characteristics as demonstrated by the linear I–V curves obtained between every pair of contacts. Transmission measurements were performed using a UV/Vis spectrophotometer (Lambda 650, Perkin Elmer) in the 0.8–3.2 eV energy range.

Results and discussion

Fig. 2(a) shows the XRD patterns of the sputtered Zn₃N₂ thin films fabricated via N₂–Ar mixing gas (pink line), Ar–N₂ reactive gas-timing (Ar–N₂-RGT, purple line), continued N₂ gas flow (blue line) and N₂ reactive gas-timing technique (N₂-RGT, green), respectively. Note that the utilization of the Ar–N₂ RGT technique to fabricate sputtered thin films can be found elsewhere.^20–24 The film thickness was set to 300 nm. The results show that a single phase of Zn₃N₂ at the peak of 400 and 440 can only be obtained when the N₂-RGT is employed. Note that an on–off sequence of N₂ of 10 : 5 was utilized. In contrast, the other techniques demonstrate the mixed phase between Zn₃N₂ and small 100 and 110 peaks of metallic Zn. It should be noted that the Zn₃N₂ and metallic Zn diffraction peaks are correlated to the Joint Committee in Powder Diffraction Standard (JCPDS) files (01-088-0618 Zn₃N₂ and 00-004-0831 Zn). To understand the roles of the N₂-RGT technique, which enabled us to obtain a single phase of Zn₃N₂ thin film, herein we have investigated the dependence of the turn-on N₂ sequence on the structure, deposition rate and surface grain boundary of the Zn₃N₂ thin films as shown in Fig. 2(b)–(d). Fig. 2(b) shows the XRD patterns of the sputtered Zn₃N₂ thin films fabricated via the N₂-RGT technique as a function of the turn-on N₂ sequence. The turn-off sequence was set at 5 s. It should be noted that the turn-off sequence could not be longer than 5 s due to the limitation of the equipment. It can be seen that a single phase of Zn₃N₂ can be obtained when the turn-on sequence was below 60 s. Above a turn-on timing of 60 s, the Zn₃N₂–Zn mixed film emerges, indicating that there is an upper limit for the turn-on sequence to obtain a single phase of Zn₃N₂. Fig. 2(c) demonstrates the dependence of the turn-on N₂ timing sequence on the FWHM of (400). The FWHM of (400) slightly decreases from 0.59 to 0.56 when the turn-on N₂ sequence increases from 5 to 60 s. On the other hand, the FWHM of (400) rapidly increases to 2.3 when the turn-on N₂ sequence was 90 s, indicating that the crystallinity of the Zn₃N₂ thin films decreased due to the mixed zinc in the film texture.⁴ In addition, we also found that the (400) peak was shifted to a higher angle when the turn-on N₂ sequence was decreased as shown in the inset of Fig. 2(c). This red shift of the (400) peak should be from the emergence of the compressive stress during the deposition process.^20,21 Fig. 2(d) shows the deposition rate and surface grain boundary of the Zn₃N₂ thin films as a function of the turn-on N₂ timing sequence. Note that the surface grain boundary of the Zn₃N₂ films have been collected using AFM measurements. The results exhibit that the growth rate of the thin film decreases when the turn-on timing of N₂ increases. On the other hand, the surface grain boundary of Zn₃N₂ increases upon increasing the N₂ timing and then decreases above the turn-on N₂ timing of 60 s. The variation of the morphology of the Zn₃N₂ thin films as a function of turn-on timing N₂ were also confirmed by FE-SEM as shown in the inset of Fig. 2(d). In regard the reactive sputtering process, when the sputtered ion bombarded on the target materials, the energy and momentum transferred to the atoms on the target surface (energy per atom) can hit some of these atoms off the target surface and/or can change surface condition of the target, caused by plasma bombardment and reactions.³⁰ For the former event, the sputtered atom ejected from the surface of the target will react with the reactive gas during moving to the substrate surface and then, the formation of the compound will be gained via the kinetic energy of the sputtered atom impinged on the surface of thin films.³¹ On the other hand, the deposition rate of the sputtered compound thin film may be obstructed by the latter event, especially when the reactive gas acts as the sputtered gas.³¹ Therefore, the degree of crystallization (e.g. surface grain boundary) and the deposition rate related to the sputtering yield of the Zn₃N₂ thin films should be reliant on the ability to enhance the energy per atom for sputtering and/or to suppress the formation of compound on the surface of the target, respectively.^20,21 When compared with the RGT technique, the turn-off timing of N₂ can be attributed to the reduction of the working pressure.³² Such a rapid decrease in the working pressures with constant rf power will drastically enrich the energy and momentum of particles bombarded on the target surface, resulting in the high energetic sputtered atoms/cluster ejected from the target being gained.³² Thus, the decrease in the surface grain boundary of the Zn₂N₃ thin films with the shorter turn-on timing of N₂ may be caused by the atomic peening effect, which is directly related to the sputtered energy.^33,34 On the other hand, the turn-on timing of N₂ can be considered as the conventional reactive sputtering process. Since the deposition rate of the thin film compound grown via reactive sputtering decreases with the long-time deposition because of the formation of the compound on the target surface;³⁵ a shorter turn-on timing of N₂ coupled with the high-energy sequence due to the turn-off timing may allow us to eliminate the formation of the compound on the surface of the target, which can actually be seen in a special technique as called high-power impulse (HIPIMS) magnetron sputtering.³⁶ Generally, it has been known that the fabrication of Zn₃N₂ thin films grown via conventional reactive sputtering can be performed using reactive Ar + N₂ mixing gases.^{6,8,12,13,15,37} The Ar gas mainly acts as the sputtered ion, which bombards the target surface whereas the N₂ gas is the reactive gas, which reacts with the sputtered atoms from the target to form the compound thin films. However, heating the substrate during reactive sputtered thin film growth was required to enhance the activation energy and/or for controlling the amount of Zn from the target to obtain a single phase of Zn₃N₂ thin films.³⁸ Such requirement for a heating process can be confirmed by the results of the mixed phase between Zn₃N₂ and Zn as shown in Fig. 2(a) when the RGT technique and substrate heating were not provided. Although the ability to control the amount of sputtered atoms from the target can be accomplished using N₂ as the only sputtered ion gas, which has a lower atomic mass than that of Ar (Ar = 39.948 amu, N₂ = 14.00 amu), a single phase of Zn₃N₂ cannot be achieved due to insufficient forming energy.³⁸ In addition, our results also demonstrate that utilizing the RGT technique with Ar–N₂ cannot be used to fabricate a single phase of Zn₃N₂ films, confirming that the excess of the number of sputtered Zn atoms from the target are produced due to the bombardment of large atomic mass Ar ions. Thus, our results highlight that the RGT technique not only enabled us to enhance the forming energy of Zn₃N₂ thin film formation without applying any external source (e.g. heater and/or rf bias) but also to adjust the suitable number of sputtered atoms from the target using small atomic mass N₂ gas. Fig. 3(a)–(d) exhibits the XPS analysis of the Zn₃N₂ thin films prepared via using a turn-on N₂ sequence of 30 s. It should be noted that 200 nm of Ta_xO_y thin film was utilized to protect the air moisture absorbed on the surface of Zn₃N₂ thin films before characterization as shown in the inset of Fig. 3(a). Fig. 3(a) shows the elemental composition as a function of ion etching time. It should be noted that the Ar ion etching was 1 min per cycle. The Ta_xO_y/Zn₃N₂ and Zn₃N₂/Si interfaces were, respectively, be reached at the etching time of 9 min and 18.5 min, indicating that the etching rate for the Ta_xO_y layer was ∼22 nm min⁻¹ whereas the etching rate for the Zn₃N₂ layer was ∼35 nm min⁻¹. The increase in O composition at the Zn₃N₂/Si interface may be the native SiO_x layer on the Si substrate. Fig. 3(b)–(d) are the electron spectra after 15 min etching time. Fig. 3(b) shows the well-defined Zn 2p doublet photoelectrons. The peaks located at a binding energy (BE) of 1021.6 eV and 1045.02 correspond to the Zn 2p_3/2 and Zn 2p_1/2 peaks, respectively.^37,38 A weak shoulder appearing between the Zn 2p_3/2 peak and Zn 2p_1/2 peak should be plasmon related.³⁷ Since it has been known that the Auger Zn L₃M_4,5M_4,5 region of the spectra is more sensitive to differences in the chemical environment rather than that of Zn 2p photoelectron peaks,^15,39 we have investigated the Auger Zn L₃M_4,5M_4,5 region of the spectra as shown in Fig. 3(c). It is clearly suggested that the Zn L₃M_4,5M_4,5 Auger electrons come from two different Zn species; i.e. from the metallic Zn and Zn that bonded to N. The doublet metallic Zn peaks are located at a BE of 494 eV and 490.6 eV. The Zn bonded to N appears as a peak at the shoulder located at a BE of 496.6 eV. Fig. 3(d) is the XPS spectra of the N 1s photoelectrons, showing the two peaks that originate from the two different N species. The lower BE peak at 396.4 eV was attributed to the N–Zn bonds reported at 395.8 eV,^15,37–39 whereas the higher BE peak at 404.8 eV relates to molecular nitrogen or another possible bonding configuration.^15,37–39 It should also be noted that a very small amount of O was found. Here we have identified the quantitative evaluation of Zn₃N₂ thin film using the XPS results. Since XPS is a surface sensitive technique, which is limited by the very short electron inelastic mean free path (IMFP) in solids, quantitative analysis of the XPS results can be delivered only for the elemental composition near the surface region. Thus, the Zn/N ratio in the bulk of the thin film could not be deduced from the XPS results. After ion etching for 15 min, the main compositions were found to be 75.7 at% Zn and 17.7 at% N (nitrogen atoms that are bonded to Zn). In addition, there was ∼3.4 at% molecular nitrogen (N₂), ∼1.6 at% O, 1.2 at% Ta and 0.4 at% Si. The amount of O is in the order of the systematic error. O, as well as Ta and Si, may not originally exist in the Zn₃N₂ film, but was introduced during the ion etching and measurement process. The XPS results show a non-stoichiometric value of Zn : N (4 : 1) ratio suggests that de-nitridation and oxidation occurs at the surface vicinity of the Zn₃N₂ film due to Ar-ion sputtering, which is in agreement with Jiang et al.³⁷ It is interesting to elucidate the material structure at the surface vicinity of Zn₃N₂ since such information may provide the information for further investigations on the thermal instability of Zn₃N₂. Based on the obtained results, a simplified material system at the surface vicinity of the measured Zn₃N₂ films may be depicted as in shown in Fig. 4(a). A zinc-rich layer was formed on the surface vicinity of the Zn₃N₂ thin film by the de-nitridation induced by the ion sputtering process. Most of the by-product nitrogen escapes from the film surface and some forms molecules that are trapped in the zinc-rich layer. ZnO could then be formed on the surface of the zinc-rich layer as there has been a report that oxidation occurs on a clean zinc surface, even in an ultra high vacuum environment and at room temperature.⁴⁰ The amount of ZnO and the thickness of the zinc-rich layer, d, can be calculated using the measured compositions and IMFP.

Fig. 2 (a) The XRD patterns obtained for the sputtered Zn₃N₂ thin films fabricated via the N₂–Ar mixing gas (pink line), Ar–N₂ reactive gas-timing (Ar–N₂-RGT, purple line), continued N₂ gas flow (blue line) and N₂ reactive gas-timing technique (N₂-RGT, green), respectively. (b) The XRD patterns obtained for the sputtered Zn₃N₂ thin films fabricated via the N₂-RGT technique as a function of turn-on N₂ sequence. (c) The dependence of the turn-on N₂ timing sequence on the FWHM of (400). Inset of (c): the 2-theta of the (400) peak as a function of the turn-on N₂ timing sequence. (d) The deposition rate and surface grain boundary of the Zn₃N₂ thin films as a function of the turn-on N₂ timing sequence. Inset of (d): SEM images of the Zn₃N₂ thin films prepared with different turn-on N₂ timing sequences.

Fig. 3 The XPS analysis: (a) the normalized peak area of the Zn 2p line, O 1s line, N 1s, Ta 4f line and Si 2p line as a function of sputtering time, (b) the Zn 2p line, (c) the Zn Auger L₃M_4,5M_4,5 line and (d) the N 1s line. Note that the XPS spectra of Zn₃N₂ were investigated at a sputtering time of 15 min.

Fig. 4 (a) A schematic of the simplified material system at the surface region of the measured Zn₃N₂ films during sputtering. (b) The dependence of the turn-on N₂ timing sequence on the resistivity of the Zn₃N₂ thin films. (c) The mobility and carrier concentration of Zn₃N₂ as a function of turn-on N₂ timing sequence. (d) The optical band gap of the Zn₃N₂ thin films at various turn-on N₂ timing sequences. Inset of (d): the dependence of the absorption coefficient on the photon energy for the Zn₃N₂ films as a function of turn-on N₂ timing sequence using the band structure equation.

The IMFP, λ, is given by λ (Å) = 1430/E² + 0.5E^1/2, where the kinetic energy of the electron E (eV) is relative to the Fermi level.⁴¹ The IMFP for Zn 2p_3/2, λ_Zn, and for N 1s, λ_N, are 10.7 Å and 16.4 Å, respectively. From the XPS measurements, oxygen was found to be about 1.6 at%, corresponding to a ZnO layer with about 0.2 monolayers coverage. The atomic ratio between all the Zn atoms and N bonded to Zn may be approximated to be

where θ is the angle between the sample surface plane and the electron energy analyser. With the Zn to N ratio of 75.7/17.7 and θ = 45°, the thickness d was found to be 12.2 Å.

Fig. 4(b) and (c) demonstrate the electrical properties of the Zn₃N₂ thin films. It should be noted that all the films are n-type semiconductors. The resistivity of the thin films as a function of the turn-on N₂ timing increases when a longer turn-on time is operated as shown in Fig. 4(b). The lowest resistivity was 0.461 Ω cm at a turn-on timing of 5 s. On the other hand, the carrier concentration increases from 4.03 × 10¹⁶ to 2.60 × 10¹⁷ cm⁻³ whereas the mobility increases from 30.29 to 54.13 cm² V⁻¹ s⁻¹ when a shorter turn-on N₂ sequence was utilized as shown in Fig. 4(c). In addition, the resistivity, carrier concentration and mobility of the Zn–Zn₃N₂ mixed phase prepared by continued N₂ reactive sputtering were 0.061 Ω cm, 1.03 × 10²⁰ cm⁻³ and 1.06 cm² V⁻¹ s⁻¹, respectively. Although the resistivity value of the Zn₃N₂ thin films in our study were higher than that of the previous reports, which mostly exhibited resistivity values in the range of 10⁻² Ω cm,^13,15,37,38 the numerical values of the carrier concentration and mobility in our study are in agreement with the published data (i.e. carrier concentration ≈ 10¹⁵ to 10²⁰ cm⁻³ and mobility ≈ 30–100 cm² V⁻¹ s⁻¹).^13,15,37,38 In polycrystalline semiconductors, the mobility is affected by impurity scattering and grain boundary scattering.^13,15,37 Since the mobility of Zn₃N₂ (in Fig. 4(c)) tends to increase with an increasing carrier concentration, the grain boundary scattering effect should be responsible for mobility of the thin films. Recently, Jiang et al. suggested that the increase in carrier concentration and mobility of the Zn₃N₂ films may be due to the increased nitrogen incorporation together with the improvement of crystallinity in the films,³⁷ indicating that grain boundary scattering may responsible for the increase in mobility. By considering the role of the RGT technique, the sputtered kinetic energy rapidly increases during the turn-off gas sequence. The frequency of such kinetic energy enhancement as to the turn-off of N₂ sequence should repeatedly elevate the forming energy of the Zn₃N₂ compound,^20–24 leading to an improvement in the forming energy of N and Zn sputtered atoms reaction. In addition, our previous work showed that the atomic peening effect caused by the GT technique could improve the texture orientation of a thin film, which enhanced the conductivity of the thin film.^16,18 Fig. 4(d) shows the optical band gap (E_g) of the Zn₃N₂ thin films as a function of the turn-on N₂ timing sequence. It should be noted that the direct band gap of the Zn₃N₂ films was identified by the band structure equation as shown in the inset of Fig. 4(d). The fundamental equations used for the analysis can be found elsewhere.^2–5 It can be seen that the optical band gap slightly increases from 1.24 eV to 1.35 eV when the turn-on N₂ timing increased from 5 s to 60 s. It is known that the E_g value of Zn₃N₂ varies from 1.23 eV to 3.2 eV (ref. 2–5, 8, 10, 12, 15, 38 and 39) depending on the fabrication method and/or the deformation of films after as-deposition.^{2–5,8,10,12,15,38,39} This indicates that our E_g values are in broad agreement with the previous studies.^{2–5,8,10,12,15,38,39} Thus, our results highlight that the RGT technique is a promising method to fabricate high quality sputtered compound thin films at RT without applying any additional energy sources.

Conclusions

Herein we have demonstrated a powerful technique, known as reactive gas timing (RGT) rf magnetron sputtering, which allowed us to fabricate high quality Zn₃N₂ thin films at room temperature without applying any additional energy sources. When compared with N₂–Ar mixing gas, Ar–N₂ reactive gas-timing (Ar–N₂-RGT) and continued N₂ gas flow, a single phase of Zn₃N₂ thin film can be obtained only when N₂-RGT was utilized. We found that the RGT technique not only enables us to enrich the forming energy of thin film formation but also provides the ability to adjust the number of sputtered atoms from the target using the small atomic mass sputtered gas. The Zn₃N₂ thin films grown via the RGT technique show a mobility of about 54.13 cm² V⁻¹ s⁻¹ whereas a direct gap of 1.24 eV can be obtained. Thus, our results highlight that controlling the amount of sputtered atoms from the target and enhancing the forming energy during growth are crucial for sputter compound thin film formation, which can be achieved by utilizing the RGT technique. In addition, the elemental composition at the surface vicinity of the Zn₃N₂ thin film after de-nitridation induced by Ar-ion sputtering was investigated. With an ion energy of 4 keV, a zinc-rich layer of ∼12 Å was formed at the surface of the film. A monolayer of ZnO with coverage of ∼0.2 was formed on the surface.

Acknowledgements

This work was partially supported by the Research and Development of White Light Emitting Diode based on Zinc Oxide Optoelectronics Material; Phase-1: Method of ZnO-Substrate Fabrication project (P1450015) from the National Science and Technology Development Agency, Thailand.

References

1. E. Aperathitis, V. Kambilafka and M. Modreanu, Thin Solid Films, 2009, 518, 1036–1039.
2. R. Ayouchi, C. Casteleiro, L. Santos and R. Schwarz, Phys. Status Solidi C, 2010, 7, 2294–2297.
3. C. García Núñez, J. L. Pau, E. Ruíz and J. Piqueras, Appl. Phys. Lett., 2012, 101, 253501.
4. W. Jiangyan, Y. Jinliang, Y. Wei and L. Ting, J. Semicond., 2012, 33, 043001.
5. S. Sinha, D. Choudhury, G. Rajaraman and S. K. Sarkar, RSC Adv., 2015, 5, 22712–22717.
6. E. Kaminska, A. Piotrowska, J. Kossut, A. Barcz, R. Butkute, W. Dobrowolski, E. Dynowska, R. Jakiela, E. Przezdziecka, R. Lukasiewicz, M. Aleszkiewicz, P. Wojnar and E. Kowalczyk, Solid State Commun., 2005, 135, 11–15.
7. V. Kambilafka, P. Voulgaropoulou, S. Dounis, E. Iliopoulos, M. Androulidaki, V. Šály, M. Ružinský and E. Aperathitis, Superlattices Microstruct., 2007, 42, 55–61.
8. V. Kambilafka, P. Voulgaropoulou, S. Dounis, E. Iliopoulos, M. Androulidaki, K. Tsagaraki, V. Šály, M. Ružinský, P. Prokein and E. Aperathitis, Thin Solid Films, 2007, 515, 8573–8576.
9. J. P. Zhang, L. D. Zhang, L. Q. Zhu, Y. Zhang, M. Liu, X. J. Wang and G. He, J. Appl. Phys., 2007, 102, 114903.
10. N. H. Erdogan, K. Kara, H. Ozdamar, R. Esen and H. Kavak, Appl. Surf. Sci., 2013, 271, 70–76.
11. I. Volintiru, M. Creatore, W. H. van Helvoort, J. L. Linden and M. C. M. van de Sanden, Appl. Phys. Lett., 2006, 89, 022110.
12. C. G. Núñez, J. L. Pau, M. J. Hernández, M. Cervera, E. Ruiz and J. Piqueras, Thin Solid Films, 2012, 520, 1924–1929.
13. N. Jiang, D. G. Georgiev, T. Wen and A. H. Jayatissa, Thin Solid Films, 2012, 520, 1698–1704.
14. M. Neirynck, W. Samaey and L. Van Poucke, J. Vac. Sci. Technol., 1974, 11, 647.
15. M. Futsuhara, K. Yoshioka and O. Takai, Thin Solid Films, 1998, 322, 274–281.
16. P. Ukahapunyakul, N. Gridsadanurak, C. Sapcharoenkun, A. Treetong, P. Kasamechonchung, P. Khemthong, M. Horprathum, S. Porntheeraphat, W. Wongwiriyapan, J. Nukeaw and A. Klamchuen, RSC Adv., 2016, 6, 7661–7667.
17. A. Klamchuen, N. Pornteeraphat and J. Nukeaw, e-J. Surf. Sci. Nanotechnol., 2005, 3, 272–275.
18. P. Prajuabwan, S. Porntheeraphat, A. Klamchuen and J. Nukeaw, Presented in part at the 2^nd IEEE international Conference on Nano/Micro Engineered and Molecular Systems (NEMS 07), IEEE, Bangkok, Thailand, 2007.
19. D. Klaitabtim, S. Pratontep and J. Nukeaw, Ceram. Int., 2008, 34, 1103–1107.
20. N. Kietipaisalsophon, W. Bunjongpru and J. Nukeaw, Int. J. Mod. Phys. B, 2002, 16, 4418–4422.
21. N. Kietipaisalsophon, W. Bunjongpru, W. Techitdheera and J. Nukeaw, Presented in part at the 2002 IEEE International Conference on Industrial Technology: productivity reincarnation through robotics & automation, IEEE, Asian Institute of Technology, 2002.
22. S. Porntheeraphat and J. Nukeaw, Appl. Surf. Sci., 2008, 254, 7851–7854.
23. W. Techitdheera, C. Thassana, W. Pecharapa and J. Nukaew, Presented in part at the 2010 3rd International Nanoelectronics Conference: (INEC 2010), IEEE, City University of Hong Kong, 2010.
24. N. Khemasiri, N. Paleeya, D. Sae-tang Phromyothin, M. Horprathum, S. Porntheeraphat, A. Sungthong, J. Nukeaw and S. Pratontep, Adv. Mater. Lett., 2015, 6, 554–559.
25. W. Bunjongpru, S. Porntheeraphat, N. Somwang, P. Khomdet, C. Hruanun, A. Poyai and J. Nukeaw, Adv. Mater. Res., 2008, 55–57, 573–576.
26. A. Poyai, W. Bunjongpru, N. Klunngien, S. Porntheerapat, C. Hruanan, S. Sopitpan and J. Nukeaw, Mater. Sci. Semicond. Process., 2008, 11, 319–323.
27. A. Sungthong, S. Porntheeraphat, A. Poyai and J. Nukeaw, Appl. Surf. Sci., 2008, 254, 7950–7954.
28. J. T-Thienprasert, J. Nukeaw, A. Sungthong, S. Porntheeraphat, S. Singkarat, D. Onkaw, S. Rujirawat and S. Limpijumnong, Appl. Phys. Lett., 2008, 93, 051903.
29. A. Sungthong, P. Khomdet, S. Porntheeraphat, C. Hruanun, A. Poyai and J. Nukeaw, Adv. Mater. Res., 2010, 93–94, 443–446.
30. G. A. Bleykher, V. P. Krivobokov, A. V. Yurjeva and I. Sadykova, Vacuum, 2016, 124, 11–17.
31. S. M. Rossnagel, J. J. Cuomo and W. D. Westwood, Handbook of plasma processing technology: fundamentals, etching, deposition, and surface interactions, Noyes Publicatons/William Andrew Publishing, LLC 13 Eaton Avenue, Norwich, New York 13815, 1989.
32. L. Gao, S. Jiang and R. Li, Thin Solid Films, 2016, 603, 391–394.
33. T. Kim and K. Lee, J. Korean Phys. Soc., 2010, 56, 805.
34. T. Kim, J. Korean Phys. Soc., 2011, 58, 787.
35. W. Tao, J. Ya-Dong, Y. He, W. Zhi-Ming and Z. He-Nan, Chin. Phys. B, 2011, 20, 038101.
36. E. Wallin and U. Helmersson, Thin Solid Films, 2008, 516, 6398–6401.
37. N. Jiang, D. G. Georgiev, A. H. Jayatissa, R. W. Collins, J. Chen and E. McCullen, J. Phys. D: Appl. Phys., 2012, 45, 135101.
38. G. Z. Xing, D. D. Wang, B. Yao, L. F. N. A. Qune, T. Yang, Q. He, J. H. Yang and L. L. Yang, J. Appl. Phys., 2010, 108, 083710.
39. X. Cao, A. Sato, Y. Ninomiya and N. Yamada, J. Phys. Chem. C, 2015, 119, 5327–5333.
40. S. Noothongkaew, H. Nakajima, A. Tong-on, W. Meevasana and P. Songsiriritthigul, Appl. Surf. Sci., 2012, 258, 1955–1957.
41. M. P. Seah and W. A. Dench, Surf. Interface Anal., 1979, 1, 2–11.

---