1. INTRODUCTION
The optical diagnostics of confined fusion plasma plays an important role to understand the performance of Fusion devices (e.g., International Thermonuclear Experimental Reactor) (Marot et al., Reference Marot, Temmerman, Thommen, Mathys and Oelhafen2008). The plasma facing mirrors in the optical diagnostic system of these fusion devices are known as first mirrors (FMs). FMs are one of the most critical elements of optical diagnostics systems of fusion device. FMs have to withstand the harsh environment arising from electromagnetic radiation and bombardment of particles (neutrons, charge exchange atoms) from high temperature plasma (Zhou et al., Reference Zhou, Gao, Jiao, Deng, Tang, Yi, Tian, Ding and Liu2006). The long term exposure of the mirrors under such harsh environment deteriorates its surface quality. The main causes for the degradation of FMs under long term diagnostic operation are: (1) erosion and re-deposition processes due to bombardment of charge exchange atoms, (2) volumetric swelling due to neutron flux, and (3) mirror surface heating due to X-ray and ultraviolet (UV) radiation (Voitsenya et al., Reference Voitsenya, Konovalov, Shtan, Solodovchenko, Becker, Bardamid, Yakimov, Gritsyna and Orlinskij1999, Reference Voitsenya, Bardamid, Bondarenko, Jacob, Konovvalov, Masuzaki, Motojima, Orilinsikij, Poperenko, Ryzhokov, Sagara, Shtan, Solodovchenko and Vinnichenko2001). It is reported that even a 10–20 nm thick contaminated impurities onto the mirrors can drastically change the optical property of metal mirrors (Voitsenya et al., Reference Voitsenya, Bardamid, Bondarenko, Jacob, Konovvalov, Masuzaki, Motojima, Orilinsikij, Poperenko, Ryzhokov, Sagara, Shtan, Solodovchenko and Vinnichenko2001). It has been shown that mirrors of mono-crystalline refractory metals (Mo, W, Rh) can have sufficiently long life time as a FM in fusion devices (Voitsenya et al., Reference Voitsenya, Bardamid, Bondarenko, Jacob, Konovvalov, Masuzaki, Motojima, Orilinsikij, Poperenko, Ryzhokov, Sagara, Shtan, Solodovchenko and Vinnichenko2001). The growth of the bulk crystal of heavy elements such as Mo, W, Rh, etc., with good polished surface is a difficult task. Therefore, fabrication and testing of mirrors made from thin films of these metals as an alternative have generated intensive research interest (Voitsenya et al., Reference Voitsenya, Konovalov, Shtan, Solodovchenko, Becker, Bardamid, Yakimov, Gritsyna and Orlinskij1999, Reference Voitsenya, Costley, Bandourko, Bardamid, Bandourko, Hirooka, Kasai, Klassen, Konovalov, Nagatsu, Orlinskij, Orsitto, Poperenko, Solodovchenko, Stan, Sugie, Taniguchi, Vinnichenko, Vukolov and Zvonkov2001; Marot et al., Reference Marot, Temmerman, Thommen, Mathys and Oelhafen2008). The reflectivity of these films is comparable to that of polished bulk crystal (Voitsenya et al., Reference Voitsenya, Costley, Bandourko, Bardamid, Bandourko, Hirooka, Kasai, Klassen, Konovalov, Nagatsu, Orlinskij, Orsitto, Poperenko, Solodovchenko, Stan, Sugie, Taniguchi, Vinnichenko, Vukolov and Zvonkov2001). These thin metal films also possess high electrical conductivity, excellent mechanical strength, high melting point, good metal barrier performance, and fine pattern ability (Shen et al., Reference Shena, Mai, Zhang, Mckenzie, Mcfall and Mcbride2000; Djerdj et al., Reference Djerdj, Tonejc, Tonejc and Radic2005). Due to high reflectivity in the far infra red (FIR) and UV-visible range, Molybdenum (Mo) mirror is an attractive candidate for FMs in fusion devices (Lipa et al., Reference Lipa, Schunke, Gil, Bucalossi, Voitsenya, Konovalov, Vukolov, Balden, Temmerman, Oelhafen, Litnovsky and Wienhold2006; Voitsenya et al., Reference Voitsenya, Costley, Bandourko, Bardamid, Bandourko, Hirooka, Kasai, Klassen, Konovalov, Nagatsu, Orlinskij, Orsitto, Poperenko, Solodovchenko, Stan, Sugie, Taniguchi, Vinnichenko, Vukolov and Zvonkov2001). Besides this, Mo is widely used as an alloying addition in stainless steels (SS) to facilitate the formation of the passive film and to improve resistance to pitting attack (Tomachuk et al., 2003). Also, Mo thin films are used for NO gas detection, the back contact in thin film solar cells, micro-electronics and semiconductor industries (Tomachuk et al., 2003). Mo thin films prepared by direct current and radiofrequency magnetron sputtering technique (Hirata et al., Reference Hirata and Saito1990; Khatri et al., Reference Khatri and Marsillac2008) and chemical vapor deposition (Gesheva et al., Reference Gesheva and Abrosimova1992; Juppo et al., Reference Juppo, Vehkamaki, Ritala and Leskela1998) are well documented in literature. There are few reports on the Mo thin films fabricated via pulsed laser deposition (PLD) technique (Fruchart et al., Reference Fruchart, Jaren and Rothman1998). PLD is highly flexible technique that can be applied to any material (Shukla et al., 2010; Wolowski et al., Reference Wolowski, Badziak, Czarnecka, Parys, Pisarek, Rosinski, Turan and Yerci2007; Lam et al., Reference Lam, Tran and Zheng2007). Besides this, pulsed laser ablation has potential application toward intense heavy ion generation and acceleration (Orlov et al., Reference Orlov, Denisov, Rosmej, Schafer, Nisius, Wilhein, Zhidkov, Kunin, Suslov, Pinegin, Vatulin and Zhao2011; Roth et al., Reference Roth, Brambrink, Audebert, Blazevic, Clarake, Cobble, Cowan, Fernandez, Fuchs, Geissel, Habs, Hegelich, Karsch, Ledingham, Neely, Ruhl, Schlegel and Schreiber2005), nanoparticle synthesis (Nath et al., Reference Nath and Khare2011; Wang et al., Reference Wang, Chen, Ding, Chu, Deng, Liang, Chen and Fu2011), and laser ablation lithography (Kamlesh et al., Reference Kamlesh and Khare2006).
In this work, PLD technique was used to deposit mirror like Mo thin film on SS substrate. The effect of target-substrate distance and background Helium gas pressure on surface morphology, FIR and UV-visible reflectivity of the mirror like PLD thin films of Mo is reported.
2. EXPERIMENTAL SETUP
The experimental set up used for PLD of Mo thin films is shown in Figure 1. The high power laser beam (second harmonic of Q-switched Nd:YAG laser, Model: Quanta systems-HYL101, 450 mJ/pulse in fundamental with about 10 ns pulse duration and 10 Hz repetition rate) was steered with a high damaged threshold right angled prism into the deposition chamber and loosely focused by a lens of 35 cm focal length onto Mo target. The Mo target in the form of a strip of size 50 mm × 10 mm and thickness 1 mm, purity 99.95%, was mounted inside the ablation chamber with a motorized translated stage. Focusing of high power laser leads to luminous high temperature plasma formation of Mo. This plasma expands hemi-spherically (Kamlesh et al., Reference Kamlesh and Khare2005) and perpendicularly to the surface of the target in vacuum, cools down and gets deposited onto the SS substrate placed parallel to and few cm apart from the target. The continuous translation of the target during the deposition process provides the fresh target surface on shot to shot basis of the laser beam. The SS substrates of size 10 mm × 10 mm were polished by abrasive high precession polishing machine and were cleaned by usual substrate cleaning protocols before deposition. Prior to PLD, the ablation chamber and substrate heater during evacuation was baked (about 12 h) for removal of residual oxygen and water vapor to prevent the oxides formation of Mo. The Mo films were deposited at elevated substrate temperature of 773° K to improve the film-substrate adhesion and to reduce root mean square (RMS) roughness. The effect of target-substrate distance onto the PLD thin films of Mo studied under vacuum (about 10−3 Pascal). From this, the optimized target-substrate distance was obtained best quality thin film. The effect of Helium gas pressure onto the quality of PLD Mo thin films at this optimized distance was studied. For this, Mo thin films were fabricated at five different Helium gas pressures; 5, 10, 50, 100, and 200 Pascal.
(Color online) Schematic diagram of experimental PLD set up.
All the Mo thin films were deposited under loosely focused condition of the laser beam so as to minimize the formation of liquid droplets. The prime reasons for the formation of liquid droplet and microstructure with very high fluence of laser are due to the tight focusing of the laser beam onto the target. Tight focusing of the laser beam leads to three type of instabilities during ablation and plasma formation process: (1) Instabilities of the plane front of liquid evaporation due to spatial modulation of the pressure in the near surface plasma layer, followed by melt outflow from pits to humps and subsequent solidification, (2) Rayleigh-Taylor type instability at liquid-vapor interface due to multi-pulse relief formation, and (3) Kelvin-Helmholtz type instability when vapor velocity is much higher than the liquid layer velocity (Brailovsky et al., Reference Brailovsky, Gapnov and Luchin1995). The loosely focused laser beam results into the lower fluence and hence the evaporation of the target material extended to larger laser spot area which results to a low energetic plasma plume (Kamlesh et al., Reference Kamlesh and Khare2005, Reference Kamlesh and Khare2006). The single shot laser beam spot size is shown in Figure 2. The major and minor diameters of laser spot are 1808.6 µm and 1315.3 µm, respectively. This corresponds to the laser fluence of 21 kJ/m2.
(Color online) Micrograph of laser beam Spot size.
Thickness of Mo thin films was measured by stylus profilometer with a 12.5 µm diamond tip. For this, the films were partially deposited by placing a copper mask onto the SS substrate. Thus, a substrate-film step was formed on it. In the profilometer set-up, the substrate-film step was scanned linearly at three different places. Surface morphology of the Mo thin films was studied with scanning electron microscope (SEM). Film composition was identified using energy dispersive X-ray (EDX). The crystal structure of mirror like Mo thin film was analyzed using X-ray diffraction (XRD) pattern. The specular reflectivity of the Mo thin films were recorded at 17.5° incidence angle with Fourier transforms infra red (FTIR) spectrometer. The UV-visible reflectivity was recorded at 45° incidence angle. The fringe visibilities of the SS substrates and Mo thin films were tested by inteferometric technique (Hernandez et al., Reference Hernandez, Juarez and Hernandez1999). For this, one of the mirrors of the Michelson interferometer was replaced with PLD thin film of Mo. Interference pattern was recorded onto a CCD (PixelFly, PCO, 230 XS 1839) for the measurement of fringe visibility.
3. RESULTS AND DISCUSSION
The sample codes assigned to PLD of Mo thin films, at various target-substrate distances are listed in Table 1.
Sample code, target-substrate distance (DTS), FWHM of (110) peak of Mo, Radius of curvature (r), average grain size, RMS roughness at 2 µm × 2 µm (q), Reflectivity (% R), and Fringe Visibility (V)
3.1. Effect of Target-Substrate Distance onto the Quality of the Mo Thin Films
3.1.1. Surface Morphology
The SEM images at about 5 kX magnifications of mirror like PLD thin films of Mo deposited for 3 h on polished SS substrate for samples Mo1, Mo2, Mo3, and Mo4 are shown in Figure 3. The corresponding EDX spectra are shown in Figure 4. The average size of grain formed due to the deposition of liquid droplets obtained from the SEM image analysis is listed in Table 1. Mo1 film shows formation of large grains and microstructure (Fig. 3a). The average grain size was about 1.8 µm. This particular film was deposited at target-substrate distance (DTS) 0.02 m. The smallest grain size due to the deposition of liquid droplet was observed for Mo4 thin film, which was deposited at DTS: 0.05 m. The average size of the liquid droplet for this film was 0.5 µm. With the increase of target-substrate distances, Figure 3 the grain size and microstructure decreases due to the fragmentation into smaller droplet before deposition onto the surface at large target-substrate distances. The EDXs of Mo1, Mo2, Mo3, and Mo4 confirm the presence of Mo and absence of substrate constituent elements (Fe, Cr) and impurities: carbon, oxygen. The EDX of Mo thin film deposited at DTS = 0.06 m showed the substrate constituent elements and beyond this distance, Mo could not be detected, as the longitudinal plasma density falls down nearly exponentially with the increase of target-substrate distance. Thus the deposition of the Mo thin film at DTS = 0.06 m is very low and beyond the detection limit of the probing electron of EDX. Therefore, the upper limit of the target-substrate distance was limited to 0.05 m in the present studies.
SEM image of the samples (a) Mo1, (b) Mo2, (c) Mo3, and (d) Mo4, respectively.
EDX of the samples (a) Mo1, (b) Mo2, (c) Mo3, and (d) Mo4, respectively.
The AFM images of Mo1, Mo2, Mo3, and Mo4 are shown in Figure 5. The RMS roughness of the corresponding films averaged over three different scan areas of size 2 µm × 2 µm are listed in Table 1. The Mo1 sample shows maximum (49 nm) RMS roughness while that of Mo4 shows minimum (9 nm) RMS roughness. The RMS roughness data presented in Table 1 clearly shows that the surface morphology of the films improves with the increase of target-substrate distance. Due to the expansion, the arriving Mo plasma plume substrate is more uniformly distributed at larger target-substrate distance and hence the RMS roughness reduces.
(Color online) AFM image of the samples (a) Mo1, (b) Mo2, (c) Mo3, and (d) Mo4, respectively.
3.1.2. Structural Characterization
The XRD pattern of Mo1, Mo2, Mo3, Mo4, and polished bulk Mo mirror is shown in Figure 6. The Mo1 shows no prominent crystal orientation except that of the substrate. The Mo2, Mo3, and Mo4 thin films show prominent crystal orientation of (110) plane of Mo along with small peak of (200) orientation. The inset in Figure 6 shows the enlarge view of the XRD pattern for (110) plane of Mo from 2θ = 38° to 42°. The full width at half maximum (FWHM) of these samples are listed in Table 1. The FWHM decreases with the increase of target-substrate distance. It was 1.4° for Mo2 and 0.7° for that of Mo4, respectively. The FWHM of polished Mo target was 0.2°. The broadening of XRD peak in the PLD thin films is due to the cumulative effect of stress present in the film, formation of nanomicro structures and the amorphous nature of the film (Culity, Reference Culity1956). The decrease in FWHM with increase of target-substrate distance is due to cumulative effect of the lowering of stress present in the film and decrease of size of the liquid droplets with the increase of target-substrate distance. The kinetic energy of the ablated species in the plasma plume decreases with the increase of target-substrate distance. Therefore, the ablated species striking the substrate with low kinetic energy at large target-substrate distance and hence reduces the stress in the film. Moreover, the ejected liquid droplet from the target gets enough time for fragmentation that results in the decrease of the size of the liquid droplet. The XRD pattern as shown in Figure 6 further confirms the absence of Molybdenum oxide formation.
(Color online) XRD patterns of the samples Mo1, Mo2, Mo3, Mo4, and bulk Mo mirror.
3.1.3. FIR and UV-visible Reflectivity
The specular FIR reflectivity in the spectral range of 1.4 µm to 25 µm for all the samples and polished bulk Mo mirror are shown in Figure 7a. The UV-visible reflectivity from λ = 350 nm to λ = 1150 nm of these samples are shown in Figure 7b. The reflectivity at λ = 20 µm and at λ = 840 nm is listed in Table 1. It confirms that the reflectivity increases with target-substrate distance. The good surface morphology (Figs. 3 and 5) is the prime factor for the increase in mirror reflectivity. The reflectivity of Mo2 and Mo3 film approaches to that of SS substrate and the mirror made of polished bulk Mo. The Mo4 thin film and polished bulk Mo mirror shows maximum reflectivity (about 95% at λ = 20 µm). This particular film was deposited at DTS = 0.05 m and has lowest RMS roughness. The Mo1 thin film deposited at DTS = 0.02 m shows minimum reflectivity in FIR and UV-visible range as its surface is very rough due to the deposition of micron sized liquid droplets. The sharp dip in specular reflectivity at λ = 4.258 µm is due the substrate only as SS has strong absorption around this wave length.
(Color online) (a) The specular FTIR reflectivity of the samples Mo1, Mo2, Mo3, Mo4, SS-substrate and Bulk Mo mirror (b) The corresponding UV-visible reflectivity.
3.1.4. Fringe Visibility
The interference patterns of PLD Mo mirror (after deposition) and corresponding SS substrate (before deposition) are shown in Figure 8 for samples Mo1–Mo4, respectively. The radius of curvature and the fringe visibility of the PLD mirrors of Mo obtained from respective interference patterns are listed in Table 1. The curved fringes are due to the curvature present in the corresponding substrate, initially, before deposition. The marginal changes in radius of curvature of the Mo thin films to its corresponding substrate confirmed the uniform deposition within substrate area. The distinct interference pattern of the thin films confirms that PLD Mo thin film possess mirror like quality. The fringe visibility of the interference pattern obtained from the mirror like Mo thin films, standard mirror and polished SS substrate is listed in Table 1. The fringe visibility of the Mo4 thin film is 0.80 and close to that obtained by the standard mirror 0.84. The data presented in Table 1 shows fringe visibility of the thin film increases with the target-substrate distance. It is again due to the improvement of mirror like quality (smooth surface morphology) with the increase of target-substrate distance.
Interference pattern of the samples (a) Mo1, (b) Mo2, (c) Mo3, and (d) Mo4 before and after deposition, respectively.
3.2. Effect of Background Helium onto the Quality of Mo Thin Films
The presence of inert gas during the laser ablation helps in plasma confinement and cooling. Therefore, the kinetic energy of the ablated atoms propagating toward the substrate reduces. This prohibits the sputtering of the film deposited on shot to shot basis and hence the surface roughness reduces. As the films deposited under vacuum at DTS: 0.03 and 0.04 m were of better quality, therefore to study the effect of background helium gas pressure onto the quality of the thin films, a systematic variation of gas pressure 5, 10, 50, 100, and 200 Pascal at DTS: 0.03 and 0.04 m only was performed. The detail deposition parameters of these samples are listed in Table 2. The deposition time for all these films was maintained at one hour.
Sample code, target-substrate distance (DTS), Helium ambient (Bp), thickness (t), FWHM of XRD peak (110) of Mo, Reflectivity (% R)
3.2.1. Thickness
The measured thickness (averaged over three different locations) of the Mo5–Mo9 deposited at DTS: 0.03 m and Mo10–Mo14 at DTS: 0.04 m thin films are listed in Table 2. The variation of thickness with deposition background gas pressure for the films Mo5–Mo9 and Mo10–Mo14 are shown in Figure 9. It was observed that the thickness of the Mo thin films increases initially with Helium pressure to 50 Pascal gas pressure and then slowly decreased with the further increase of gas pressure up to 200 Pascal. This behavior is due to the confinement of the laser induced plasma with the increase of background pressure. Initially, when the background pressure is low, the confinement taking place within a hemispherical region was comparable to that of the target-substrate distance. Therefore, the plasma density increased with the confinement of the plasma which increased the deposition rate on the substrate and hence increase in thickness. With the further increase of background pressure, the plasma was confined to smaller region, than the target-substrate distance. Thus, there is a significant decrease in the laser ablated species reaching towards substrate which decreased the deposition rate and hence thickness of the film. The maximum thickness, 386 nm was observed for Mo7 thin films deposited at DTS: 0.03 m. The deposition rate of this thin film, deposited around 50 Pascal, is about 6 nm/min where as that of the films deposited under vacuum is about 4 nm/min at 0.03 m target-substrate distance. The thickness of the films at DTS: 0.03 m was more under same background pressure than that of the films deposited at DTS: 0.04 m.
Variation of thickness of PLD Mo thin film as a function of pressure (a) DTS 0.03m, (b) DTS 0.04m.
3.2.2. Structural Characterization
The XRD pattern of Mo5–Mo9 and Mo10–Mo14 are shown in Figure 10. The prominent Mo (110) crystal orientation along with small peak of Mo (200) orientation and two substrate peaks for Mo5–Mo8 and Mo10–Mo13 were confirmed from their respective XRD patterns. The FWHM of Mo (110) XRD peak for Mo5–Mo8 and Mo10–Mo13 are listed in Table 2. The FWHM data presented in Tables 1 and Table 2 confirms the overall improvement in the Mo (110) crystal orientation with Helium pressure in the range of 5 and 50 Pascal. The FWHM of Mo (110) XRD peak for Mo13 is 1.2°. The poor crystalline nature of this thin film could be due to the low kinetic energy of the laser ablated species of Mo at DTS: 4 cm under 100 Pascal helium ambient, which is not sufficient to attain supper-saturation temperature for pure crystalline growth towards Mo (110) plane. XRD pattern of Mo9 and Mo14 films clearly showed MoO2 (101) and MoO2 (220) crystal orientation. This is due to the oxygen impurity which becomes prominent at high pressure of 200 Pascal.
(Color online) XRD patterns of samples (a) Mo5 to Mo9 and (b) Mo10 to Mo14.
3.2.3. FIR and UV-visible Reflectivity
The specular FIR reflectivity of the Mo5–Mo9 and Mo10–Mo14 are shown in Figure 11. The specular UV-visible reflectivities of the Mo5–Mo9, and Mo10–Mo14, respectively, are shown in Figure 12. The FIR reflectivity at λ = 20 µm and UV-visible reflectivity at λ = 840 nm of Mo5–Mo9 and Mo10–Mo14 are listed in Table 2. The film deposited at 50 Pascal background pressure had maximum FIR reflectivity, at λ = 20 µm about 98%, for both 0.03 m and 0.04 m target-substrate distance. It was observed that the film deposited at 5 Pascal background pressure had poor reflectivity for Mo5 and Mo10 thin films. The maximum recorded specular reflectivity at λ = 840 nm, was about 84% for Mo7–Mo8 and about 87% for Mo12 thin films, respectively.
(Color online) The specular FTIR reflectivity of the samples (a) Mo5 to Mo9 and (b) Mo10 to Mo14.
(Color online) UV-visible reflectivity of the samples (a) Mo5 to Mo9 and (b) Mo10 to Mo14.
4. CONCLUSION
The detailed characterization of mirror like Mo thin films fabricated by PLD technique has been presented. The PLD Mo thin films were free from impurities: oxygen, carbon, and substrate elements. The formation of liquid droplets and hence large micro-structure were observed in the Mo thin films deposited at DTS ≤ 0.02 m. The surface morphology of the Mo mirrors improved with the increase of the target-substrate distance. The Mo films were predominately orientated in (110) plane at large target-substrate distance (≥ 0.03 m). The Mo4 mirror showing reflectivity 95 % at λ = 20 µm and 70% at λ = 840 nm is close to that of the polished bulk Mo mirror. The fringe visibility was observed to be improved with the increase of target-substrate distance. The crystallinity and the reflectivity were of the thin films was found to improved in presence of helium ambient. The thickness of the Mo thin films were observed to increase initially with the increase of background Helium pressure up to 50 Pascal and then falls down with further increase of background pressure to 200 Pascal. Mo7 and Mo12 were the best thin films having reflectivity about 98% at λ = 20 µm. At λ = 840 nm Mo7 and Mo12 showed about 83% and about 87% reflectivity, respectively. These films also posses minimum FWHM for Mo (110) plane. Thus it can be concluded that the optimized parameters for obtaining mirror like quality PLD thin films of Mo having sufficient thickness about 300 nm for FM application are; 0.03–0.04 m target substrate distance, 50 Pascal Helium ambient, 500° C substrate temperature.
ACKNOWLEDGEMENT
This work is partially supported by BRFST, NFP (India), Project No. NFP/DIAG/01.
