Advertisement

Nanomanufacturing and Metrology

, Volume 2, Issue 3, pp 168–176 | Cite as

Etching Characteristics of Quartz Crystal Wafers Using Argon-Based Atmospheric Pressure CF4 Plasma Stabilized by Ethanol Addition

  • Rongyan Sun
  • Xu Yang
  • Keiichiro Watanabe
  • Shiro Miyazaki
  • Toru Fukano
  • Masanobu Kitada
  • Kenta Arima
  • Kentaro Kawai
  • Kazuya YamamuraEmail author
Original Articles
  • 114 Downloads

Abstract

Atmospheric-pressure plasma chemical vaporization machining (AP-PCVM) using helium as a carrier gas to achieve uniform thickness of quartz crystal wafers has been practically applied to the mass production of high-performance crystal units. However, because of the high cost and limited supply of helium, the proposed use of argon instead of helium has garnered interest. Argon plasma at atmospheric pressure tends to become unstable, but it has been reported that the addition of a small amount of ethanol makes it possible to generate a stable and uniformly spread glow discharge atmospheric-pressure argon plasma. In this study, an AP-PCVM experiment was conducted on quartz crystal wafers using an argon-ethanol mixture gas as the carrier gas and CF4 and O2 as process gases. A uniform, stable glow discharge plasma was generated by applying the ethanol-containing process gas, and satisfactory removal characteristics for correcting the thickness deviation of a quartz crystal wafer were obtained.

Keywords

Atmospheric-pressure plasma chemical vaporization machining (AP-PCVM) Quartz crystal wafer Ethanol-added argon-based CF4 plasma Glow discharge plasma 

1 Introduction

Quartz crystal units have good frequency–temperature characteristics at around room temperature. In addition, the crystal units are very stable both physically and chemically, so the change in frequency with aging is small. Because of these excellent properties, quartz crystal units have been widely used in communication devices such as mobile phones and in consumer devices such as digital cameras and computers in order to provide accurate reference signals [1, 2, 3]. To increase the communication speed, the resonance frequency of the quartz crystal unit should be increased by reducing the thickness of the quartz crystal wafer. In the recent commercial production of quartz resonators, a wafer process was intensively developed to improve the productivity of resonators. To achieve improved productivity by reducing the processing time for frequency adjustment, a uniform thickness is essential for the quartz crystal wafer. However, commercially available quartz crystal wafers, which are formed by conventional mechanical fabrication processes such as cutting with a wire saw, lapping, ion beam figuring, and polishing, typically have a thickness deviation of ± 0.1%. This value is about 100-fold larger than the tolerance required for a commercial product. Furthermore, owing to the poor parallelism and the existence of subsurface damage, many spurious peaks, which deteriorate the resonance characteristics, are observed in a resonance curve [4]. As a highly efficient and damage-free thickness correction technique, atmospheric-pressure plasma chemical vaporization machining (AP-PCVM) has been applied to improve the thickness uniformity of quartz crystal wafers [5].

AP-PCVM is an ultraprecise figuring technique that uses fluorine radicals generated by atmospheric-pressure plasma to change the surface atoms of a substrate into volatile reaction products to form the desired shape. Since AP-PCVM is a noncontact chemical figuring technique that does not apply a mechanical load to substrates, the breakage of thin brittle materials is prevented and no subsurface damage (SSD) layer is formed during the chemical removal process [6, 7]. Moreover, as AP-PCVM is an atmospheric-pressure process, no vacuum chamber is required. Until now, the most widely used methods for plasma generation at atmospheric pressure have mainly included inductively coupled plasma (ICP) [8, 9], capacitively coupled plasma (CCP) [7], and microwave plasma (plasma jet) [6, 10]. Since ICP and microwave plasma produce high temperatures exceeding 200 °C, the possibility of both crystal twinning and breaking of the quartz crystal wafer increases. Therefore, CCP is used in AP-PCVM to process the crystal wafers in this study. In our previous research, the thickness uniformity of a commercially available quartz crystal wafer was decreased from 250 to 50 nm by a single correction process without the formation of SSD [5]. KYOCERA has developed the world’s smallest crystal unit (1.0 × 0.8 mm) for smartphones, wearables, and other electronic devices by AP-PCVM. However, helium gas has been used as the carrier gas in AP-PCVM until now. The helium gas used in industry is mostly produced from natural gas. Problems such as the depletion of natural resources and high cost are of wide concern. To solve these problems, researchers have proposed the use of argon gas instead of helium gas as the carrier gas in the generation of atmospheric-pressure plasma, as argon gas can be industrially produced by the fractional distillation of liquid air. Although this will reduce operational cost, the breakdown voltage for argon is much higher than that for helium [11]. As is well known, the higher breakdown voltage may cause the rapid multiplication of electrons after breakdown, leading to the formation of filamentary arc streamers [12]. As reported by Sun et al. [13], the addition of ethanol to argon has been proven to be very useful for generating an atmospheric-pressure glow discharge plasma. In our study, we used argon with a small amount of ethanol instead of helium as the carrier gas in AP-PCVM. Experiments were conducted to investigate the etching characteristics of quartz crystal wafers by AP-PCVM using ethanol-added argon-based atmospheric-pressure CF4 plasma.

2 Experimental Setup

Figure 1a shows the experimental setup of AP-PCVM used in this study. Four flow paths exist, one for the carrier gas, one for the ethanol, and two for the process gases (CF4 and O2). The flow rates of the carrier gas and process gases were controlled using mass flow controllers (MFCs). Ethanol vapor was introduced using the gas–liquid mixture vaporization method. The carrier gas (Ar) and liquid ethanol (concentration 99.5%), where the latter was controlled using liquid mass flow controllers (LMFCs), were heated and mixed together in the vaporizer. The mixture gas was supplied from the center of the electrode and flowed through the space between the aluminum alloy electrode and the alumina ceramic cover arranged coaxially with the electrode. The size of the AT-cut quartz crystal wafers used in the experiment was 54 mm × 50 mm × 90 μm. All the AP-PCVM experiments in this paper were conducted at room temperature (23.3 °C) without substrate heating. The quartz crystal wafer was held on a worktable by a vacuum chuck. The distance between the tip of the electrode and the quartz crystal wafer was 3.5 mm. The diameter of the powered electrode was 3 mm. Atmospheric-pressure Ar-based CF4 plasma was generated by applying 13.56 MHz radio frequency (RF) power between the electrode and the worktable. Fluorine radicals, which are considered to be the reactive species that etch the quartz crystal, were generated by the dissociation of CF4 in the atmospheric-pressure argon plasma. Figure 1b shows a photograph of argon plasma generated on a quartz crystal wafer without adding ethanol. The high breakdown voltage of argon led to the formation of filamentary arc streamers, and the quartz crystal wafer was easily broken because of the high temperature due to the localized arc discharge. In addition, since the state of the localized arc discharge was unstable, the reproducibility of the removal spot formation was low. Figure 1c shows a photograph of ethanol-added argon plasma generated on a quartz crystal wafer. By adding ethanol, a stable glow discharge plasma without filamentary arc streamers was formed and the quartz crystal wafer was not broken. The reproducibility of the removal spot formation was also improved. Thus, it is possible to precisely correct the thickness distribution of quartz crystal wafers by numerically controlled AP-PCVM using ethanol-added argon instead of helium as the carrier gas.
Fig. 1

a Schematic of the experimental setup used for AP-PCVM. b Image of AP-PCVM on a quartz crystal wafer without adding ethanol. c Image of AP-PCVM on a quartz crystal wafer using ethanol-added argon

3 Results and Discussion

3.1 Role of Oxygen in AP-PCVM

To investigate the AP-PCVM etching characteristics, an experiment was conducted on quartz crystal wafers in which different oxygen flow rates were applied. The RF power was 60 W and the ethanol flow was 0.003 g/min. The carrier gas was Ar (600 sccm) and the process gases were CF4 (10 sccm) and O2 (0–8 sccm). Figure 2 shows photographs of removal spots on quartz crystal wafers formed by AP-PCVM using ethanol-added argon as the carrier gas, optical emission spectroscopy (OES) spectra of the plasma used in AP-PCVM, and X-ray photoelectron spectroscopy (XPS) measurement results at various oxygen flow rates. First, no oxygen was added to the process gas. Under this condition, a concentrated black deposit was formed in the area surrounding the removal spot as shown in Fig. 2a. From the OES spectrum of the plasma shown in Fig. 2f, a strong emission of C2 Swan band peaks was observed [14]. In addition, as the plasma was generated at atmospheric pressure, N2 mixed into the plasma from the ambient was also dissociated and excited, so emissions from N2 were observed in the OES spectrum [15, 16]. Next, the area with the black deposit was measured using XPS (Fig. 2k). A strong peak was observed in the C1s spectrum obtained from the area with the black deposit. To elucidate the composition of the black deposit, the deconvolution spectra of C1s were evaluated from the chemical shifts relative to the C–C peak at 284.8 eV [17]. Four peaks in addition to the C–C peak were observed, which were attributed to C–OH (285.3 eV), C–O–C (286.0 eV), C=O (287.9 eV), and π–π* (290.4 eV) [17, 18]. The shake-up satellite peak (π–π*, 290.4 eV) assigned to π-electrons was delocalized at the aromatic network in graphite and disappeared with increasing oxidation [19]. Jimenez et al. [20] also observed the C2 Swan band from ethanol-added argon plasma in addition to a carbon deposit. Research on the production of graphene by ethanol chemical vapor deposition was also reported by Faggio et al. [21]. Thus, it is considered that the black deposit originated from incompletely decomposed carbon from ethanol. Next, 1 sccm of oxygen was added to the process gas. A carbon deposit still existed, but its color was lighter and the deposition area became smaller. Compared with the OES spectrum of the plasma without adding oxygen, the emission of the C2 Swan band was weaker. The XPS measurement result showed that the C–C peak became weaker. It was considered that through the addition of oxygen to the process gas, part of the incompletely decomposed carbon was oxidized. Thus, it is predicted that the addition of oxygen can suppress the formation of the carbon deposit on the quartz crystal surface. As shown in the OES spectra in Fig. 2, by increasing the oxygen flow rate from 0 to 4 sccm, the optical emission from atomic oxygen became stronger and the emission from the C2 Swan band became weaker. When the oxygen flow rate was increased to 4 sccm, the optical emission from the C2 Swan band of the plasma almost disappeared. Also, there was no C peak in the XPS spectrum (the same as the sample before the plasma treatment, as shown in Fig. 3). These results proved that the addition of oxygen to the process gas can suppress the formation of the carbon deposit on a quartz crystal wafer. The relationship between the etching rate of the quartz crystal wafer and the oxygen flow rate is shown in Fig. 4. The carbon deposit disappeared with the addition of 4 sccm O2. However, the etching rate decreased as the oxygen flow increased. It is considered that as the oxygen itself did not contribute to the etching reaction of the quartz crystal, and the generation of O and F radicals both require energy from argon in the active state, the competition between O2 and CF4 led to a decrease in the etching rate at high oxygen fractions.
Fig. 2

Photographs ae of removal spots formed by AP-PCVM on quartz crystal wafers using ethanol-added argon as the carrier gas, optical emission spectroscopy (OES) spectra of the plasma used in AP-PCVM (fj), and X-ray photoelectron spectroscopy (XPS) measurements (ko)

Fig. 3

X-ray photoelectron spectroscopy (XPS) measurements of the sample surface before the plasma treatment

Fig. 4

Removal volume of quartz crystal wafer in 60 s with different O2 fractions

3.2 Comparison Between He and Ar-Based Plasmas

From the above results, it has been confirmed that a stable glow discharge of argon-based atmospheric-pressure CF4 plasma can be used to process quartz crystal wafers. In our previous research, helium gas was used as the carrier gas in AP-PCVM [5]. To investigate the difference between He and Ar-based CF4 plasmas, a comparative experiment was conducted. The RF power was 60 W and the ethanol flow rate was 0.003 g/min. The carrier gas was He/Ar (600 sccm) and the process gases were CF4 (10 sccm) and O2 (4 sccm). The processing time was 15–90 s. Figure 5a, b shows the cross-sectional profiles of removal spots formed on the quartz crystal wafers by AP-PCVM in different processing times using ethanol-added argon and helium as the carrier gas. As an example, scanning white-light interferometer (SWLI) images of the removal spots formed by etching for 45 s using helium and ethanol-added argon as carrier gases as well as comparison of their cross-sectional profiles were shown in Fig. 5c–e. The diameter of the removal spot processed using ethanol-added argon as the carrier gas was larger than that of the spot processed using He, but the maximum depth was lower. The relationship between the removal volume of the removal spots and the processing time is shown in Fig. 6. The removal volumes using He and ethanol-added argon were almost the same for the same processing time. In addition, the removal volume was proportional to the processing time in both cases. The results above show that, by using ethanol-added argon as the carrier gas in AP-PCVM instead of helium, the same etching rate is obtained and the operational cost is reduced.
Fig. 5

Scanning white-light interferometer (SWLI) data of removal spots formed by AP-PCVM on quartz crystal wafers in different etching times. a Cross-sectional profiles of the removal spots in different processing time using ethanol-added argon as the carrier gas. b Cross-sectional profiles of the removal spots in different processing time using helium as the carrier gas. c 2D-image of the removal spot formed by applying ethanol-added argon as the carrier gas. d 2D-image of the removal spot formed by applying helium as the carrier gas. e Comparison of cross-sectional profiles of removal spots formed by etching for 45 s

Fig. 6

Relationship between processing time and removal volume of the removal spot formed by AP-PCVM (RF power: 60 W, ethanol: 0.003 g/min, carrier He/Ar: 600 sccm, CF4: 10 sccm, O2: 4 sccm)

3.3 Relationship Between Scan Speed and Removal Volume

From the above, we have used ethanol-added argon as the carrier gas instead of helium to generate a stable glow plasma and obtained the same etching rate. In the actual production process, to realize quartz crystal wafers with high thickness uniformity, we use plasma for raster scan processing by controlling the dwell time rather than stationary point processing. Thus, it is necessary to define the relationship between the removal volume and the scanning speed of the plasma. Line scan experiments were conducted on a quartz crystal wafer. The RF power was 60 W and the ethanol flow rate was 0.003 g/min. The carrier gas was He/Ar (600 sccm) and the process gases were CF4 (10 sccm) and O2 (4 sccm). The scanning speed was 3, 6, 9, and 12 mm/min. Figure 7a, b show the cross-sectional profiles of the grooves formed on the quartz crystal wafer during AP-PCVM in different scanning speed using ethanol-added argon and helium as the carrier gas. A comparison of cross-sectional profiles of the grooves formed by etching with the scanning speed of 12 mm/min using ethanol-added argon and helium as carrier gases is shown in Fig. 7c. In the case of formation of the removal spot using ethanol-added argon, the central cross-sectional shape of the removal spot was complicated, as shown in Fig. 5. However, in the case of line-scan experiment, the cross-sectional shape of the grooves formed by AP-PCVM using ethanol-added argon and helium both became smooth due to the integration of the removal spot shape along the scanning direction. The removal rate was evaluated by calculation of the cross-sectional area of the groove. Figure 8a shows the relationship between the scanning speed and the cross-sectional area of the groove formed on the quartz crystal wafer by a line scan during AP-PCVM. This result shows that the cross-sectional area correlates with the scanning speed. Since the reciprocal of the scanning speed is considered to be the dwell time of plasma, the relationship between the dwell time and the cross-sectional area of the groove is shown in Fig. 8b. The cross-sectional area of the groove was proportional to the dwell time of plasma. In our previous research, we confirmed that AP-PCVM using helium as a carrier gas was able to obtain quartz crystal wafers with high thickness uniformity [5]. From the above results, a similar cross-sectional shape of grooves was obtained during AP-PCVM using helium and ethanol-added argon as the carrier gas. Therefore, in AP-PCVM using ethanol-added argon as substitute for helium as the carrier gas, it is possible to realize quartz crystal wafers with high thickness uniformity by controlling the scanning speed in accordance with this relationship.
Fig. 7

Cross-sectional profiles of grooves formed by AP-PCVM on quartz crystal wafers in different scanning speeds. a Ethanol-added argon as the carrier gas. b Helium as the carrier gas. c Comparison of cross-sectional profiles of the grooves formed by etching with the scanning speed of 12 mm/min

Fig. 8

Relationship between scanning speed (a) or dwell time (b) and cross-sectional area of the etched groove formed by AP-PCVM (RF power: 60 W, ethanol: 0.003 g/min, carrier He/Ar: 600 sccm, CF4: 10 sccm, O2: 4 sccm)

4 Conclusions

As the high breakdown voltage for argon easily leads to the formation of filamentary arc streamers, which can break a quartz crystal wafer, argon has not until now been used as carrier gas in AP-PCVM to correct the thickness distribution of a quartz crystal wafer. In this paper, the etching characteristics of a quartz crystal wafer obtained by AP-PCVM using ethanol-added argon-based atmospheric-pressure CF4 plasma were investigated. The addition of a small fraction of ethanol to argon can generate atmospheric-pressure plasma with a stable glow discharge state instead of arc streamers. To avoid carbon deposition on quartz crystal wafers during etching, O2 was added to the process gas. An O2 flow rate of 4 sccm is considered suitable because a high etching rate can be obtained without carbon deposition. Using argon instead of helium as the carrier gas in AP-PCVM will effectively solve problems such as the depletion of natural resources and high cost.

Notes

Acknowledgements

This study was supported by JST A-STEP Grant Number JPMJTS1623, Japan. This study was also supported by Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Number JP19J20167.

References

  1. 1.
    Manbachi A, Cobbold RSC (2011) Development and application of piezoelectric materials for ultrasound generation and detection. Ultrasound 19:187–196.  https://doi.org/10.1258/ult.2011.011027 CrossRefGoogle Scholar
  2. 2.
    Pan L, Krim J (2013) Scanning tunneling microscope-quartz crystal microbalance study of temperature gradients at an asperity contact. Rev Sci Instrum 84:014901–1–014901–9.  https://doi.org/10.1063/1.4767239 CrossRefGoogle Scholar
  3. 3.
    Nakazawa M, Kozima S (1982) A study of GT-Type quartz crystal plates. IEEE Trans Sonics Ultrason 29:121–127.  https://doi.org/10.1109/T-SU.1982.31321 CrossRefGoogle Scholar
  4. 4.
    Nagaura Y, Yokomizo S (1999) Manufacturing method of high frequency quartz oscillators over 1 GHz. In: Proceedings of the 1999 IEEE international frequency control symposium, pp 425–428.  https://doi.org/10.1109/FREQ.1999.840797
  5. 5.
    Yamamura K, Shimada S, Mori Y (2008) Damage-free improvement of thickness uniformity of quartz crystal wafer by plasma chemical vaporization machining. Ann. CIRP 57:567–570.  https://doi.org/10.1016/j.cirp.2008.03.132 CrossRefGoogle Scholar
  6. 6.
    Sun RY, Yang X, Ohkubo Y, Endo K, Yamamura K (2018) Optimization of gas composition used in plasma chemical vaporization machining for figuring of reaction-sintered silicon carbide with low surface roughness. Sci Rep 8:2376-1–2376-9.  https://doi.org/10.1038/s41598-018-20849-5 Google Scholar
  7. 7.
    Mori Y, Yamauchi K, Yamamura K, Sano Y (2000) Development of plasma chemical vaporization machining. Rev Sci Instrum 71:4627–4632.  https://doi.org/10.1063/1.1322581 CrossRefGoogle Scholar
  8. 8.
    Yu N, Jourdain R, Gourma M, Shore P (2016) Analysis of De-Laval nozzle designs employed for plasma figuring of surfaces. Int J Adv Manuf Technol 87:735–745.  https://doi.org/10.1007/s00170-016-8502-y CrossRefGoogle Scholar
  9. 9.
    Castelli M, Jourdain R, Morantz P, Shore P (2012) Rapid optical surface figuring using reactive atom plasma. Precis Eng 36:467–476.  https://doi.org/10.1016/j.precisioneng.2012.02.005 CrossRefGoogle Scholar
  10. 10.
    Piechulla P, Bauer J, Boehm Paetzelt H, Arnold T (2016) Etch mechanism and temperature regimes of an atmospheric pressure chlorine-based plasma jet process. Plasma Process Polym 13:1128–1135.  https://doi.org/10.1002/ppap.201600071 CrossRefGoogle Scholar
  11. 11.
    Schutze A, Jeong JY, Babayan SE, Park J, Selwyn GS, Hicks RF (1998) The atmospheric-pressure plasma jet: a review and comparison to other plasma sources. IEEE Trans Plasma Sci 26:1685–1694.  https://doi.org/10.1109/27.747887 CrossRefGoogle Scholar
  12. 12.
    Park J, Henins I, Herrmann HW, Selwyn GS, Hick RF (2001) Discharge phenomena of an atmospheric pressure radio-frequency capacitive plasma source. J Appl Phys 89:20–28.  https://doi.org/10.1063/1.1323753 CrossRefGoogle Scholar
  13. 13.
    Sun WT, Li G, Li HP, Bao CY, Wang HB, Zeng S, Gao X, Luo HY (2007) Characteristics of atmospheric-pressure, radio-frequency glow discharges operated with argon added ethanol. J Appl Phys 101:123302–1–123302–6.  https://doi.org/10.1063/1.2748430 Google Scholar
  14. 14.
    Vandevelde T, Wu TD, Quaeyhaegens C, Vlekken J, D’Olieslaeger M, Stals L (1999) Correlation between the OES plasma composition and the diamond film properties during microwave PA-CVD with nitrogen addition. Thin Solid Films 340:159–163.  https://doi.org/10.1016/S0040-6090(98)01410-2 CrossRefGoogle Scholar
  15. 15.
    Flores O, Castillo F, Martinez H, Villa M, Villalobos S, Reyes PG (2014) Characterization of direct current He-N2 mixture plasma using optical emission spectroscopy and mass spectrometry. Phys Plasmas 21:053502–1–053502–6.  https://doi.org/10.1063/1.4875343 CrossRefGoogle Scholar
  16. 16.
    Cvelbar U, Krstulovic N, Milosevic S, Mozetic M (2008) Inductively coupled RF oxygen plasma characterization by optical emission spectroscopy. Vacuum 82:224–227.  https://doi.org/10.1016/j.vacuum.2007.07.016 CrossRefGoogle Scholar
  17. 17.
    Teng CC, Ma CCM, Lu CH, Yang SY, Lee SH, Hsiao MC, Yen MY, Chiou KC, Lee TM (2011) Thermal conductivity and structure of non-covalent functionalized graphene/epoxy composites. Carbon 49:5107–5116.  https://doi.org/10.1016/j.carbon.2011.06.095 CrossRefGoogle Scholar
  18. 18.
    Ren PG, Yan DX, Ji X, Chen T, Li ZM (2011) Temperature dependence of graphene oxide reduced by hydrazine hydrate. Nanotechnology 22:055705–1–055705–8.  https://doi.org/10.1088/0957-4484/22/5/055705 Google Scholar
  19. 19.
    Hontoria-Lucas C, Lopez-Peinado AJ, Lopez-Gonzalez D, Rojas-Cervantes ML, Martin-Aranda RM (1995) Study of oxygen-containing groups in a series of graphite oxides: physical and chemical characterization. Carbon 33:1585–1592.  https://doi.org/10.1016/0008-6223(95)00120-3 CrossRefGoogle Scholar
  20. 20.
    Jimenez M, Rincon R, Marinas A, Calzada MD (2013) Hydrogen production from ethanol decomposition by a microwave plasma: influence of the plasma gas flow. Int J Hydrogen Energy 38:8708–8719.  https://doi.org/10.1016/j.ijhydene.2013.05.004 CrossRefGoogle Scholar
  21. 21.
    Faggio G, Capasso A, Messina G, Santangelo S, Dikonimos Th, Gagliardi S, Giorgi R, Morandi V, Ortolani L, Lisi N (2013) High-temperature growth of graphene films on copper foils by ethanol chemical vapor deposition. J Phys Chem C 117:21569–21576.  https://doi.org/10.1021/jp407013y CrossRefGoogle Scholar

Copyright information

© International Society for Nanomanufacturing and Tianjin University and Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • Rongyan Sun
    • 1
  • Xu Yang
    • 1
  • Keiichiro Watanabe
    • 2
  • Shiro Miyazaki
    • 3
  • Toru Fukano
    • 3
  • Masanobu Kitada
    • 4
  • Kenta Arima
    • 1
  • Kentaro Kawai
    • 1
  • Kazuya Yamamura
    • 1
    Email author
  1. 1.Department of Precision Science & Technology, Graduate School of EngineeringOsaka UniversityOsakaJapan
  2. 2.Keihanna Research Center, Corp R & D GroupKYOCERA CorporationKyotoJapan
  3. 3.Crystal Components Division, Corporate Electronic Components GroupKYOCERA CorporationKyotoJapan
  4. 4.Solar Energy Development Division, Corporate Solar Energy GroupKYOCERA CorporationKyotoJapan

Personalised recommendations