Formation of polytetrafluoroethylene thin films by ...

Formation of polytetrafluoroethylene thin films by using CO2 laser evaporation and XeCl laser ablation Muneto Inayoshi, Masaru Hori,a) and Toshio Goto Department of Quantum Engineering, School of Engineering, Nagoya University, Chikusa-ku, Nagoya 464-01, Japan

Mineo Hiramatsu and Masahito Nawata Department of Electrical and Electronic Engineering, Faculty of Science and Technology, Meijo University, Tempaku-ku, Nagoya 468, Japan

Shuzo Hattori Nagoya Industrial Science Research Institute, Naka-ku, Nagoya 460, Japan

~Received 4 January 1996; accepted 5 April 1996! Laser evaporation and laser ablation methods were applied to the preparation of polytetrafluoroethylene ~PTFE! thin films. In the case of the laser evaporation method, PTFE targets were evaporated by a continuous wave ~cw! CO2 laser ~10.6 mm!, and fluorocarbon thin films were formed at a deposition rate of as high as 2 mm/min for a laser power of 10 W. The chemical composition and structure of the deposited film corresponded to those of a PTFE target, which was confirmed by x-ray photoelectron spectroscopy and Fourier transform infrared absorption spectroscopy analyses. In the laser ablation method, PTFE targets were ablated by a XeCl excimer laser ~308 nm!. It is found that the deposited films contained a small amount of fluorine atoms on the surface. From these experiments, the successful formation of PTFE thin films was demonstrated for the first time using cw CO2 laser evaporation method. © 1996 American Vacuum Society.

I. INTRODUCTION Fluorocarbon materials have excellent properties such as highly electrical resistivity, a low coefficient of friction, nonwetting property, etc., and have been used widely in many industrial applications. Polytetrafluoroethylene ~PTFE! is one of these materials which has been used for protective films of hydrophilic crystals and antireflection films. It can also be applied as a protective and insulating film in microelectronics devices. So far, PTFE thin films have been formed by plasma polymerization employing fluorocarbon monomer gases,1 vacuum evaporation of PTFE using a resistive heating source2 or an electron beam,3 rf sputtering of PTFE,4 – 6 and laser ablation of PTFE using an ultraviolet ~UV! laser beam.7,8 In general, the films prepared by rf sputtering or vacuum evaporation methods are found to be fluorine deficient. Moreover, using the plasma polymerization method, many impurity materials are found to remain in the deposited films. Therefore, the deposited films are not stoichiometric. On the other hand, thin films deposited by the laser ablation method using an UV laser beam are reported to be stoichiometric.8 The formation of thin films by the laser deposition technique can be driven at a high rate. The required setup is very simple, which consists of three parts, i.e., a laser as an external energy source to vaporize the materials, a reaction chamber, and a pumping system. The decoupling facility of the vacuum hardware and the evaporation power source enables this technique to be quite flexible. Consequently, a variety of operational modes can be easily adaptable without the constraint imposed by the use of internally powered evaporation a!

Electronic mail: [email protected]

1981

J. Vac. Sci. Technol. A 14(4), Jul/Aug 1996

sources. Since only a bulk material and a substrate are placed in the reaction chamber and a laser beam is used for evaporation of the bulk material, thin films are deposited in a relatively cleaner environment than in the case of plasma polymerization or rf sputtering methods. Therefore, high-quality films with less impurities can be synthesized in a high vacuum. Moreover, since the source material irradiated from a laser beam of high power is sublimed momentarily, the deposited films have the same chemical component as that of the source material. The laser deposition method has excellent characteristics and has been utilized for evaporation of organic and inorganic substances, but to the best of our knowledge there has been no report on the formation of PTFE thin films using infrared lasers such as the CO2 laser which is preferred owing to its stability, ease of handling, and lesser cost. On the other hand, ablation of PTFE using an excimer laser has been performed by Kuper and Stuke,9 but their concentration was to clarify the mechanism of PTFE ablation by an excimer laser. Also there has been no report on PTFE films deposited by XeCl excimer laser ablation. In this study, we have employed a continuous wave ~cw! CO2 laser and XeCl excimer laser for evaporation and ablation of PTFE bulk targets, respectively. The properties of deposited films by using these methods are compared with that of the PTFE bulk target. It is found that the CO2 laser evaporation method produced high-quality films of PTFE. II. EXPERIMENT A schematic diagram of the experimental equipment for excimer laser ablation is shown in Fig. 1. A XeCl excimer laser ~308 nm! was used for ablation of a PTFE bulk target.

0734-2101/96/14(4)/1981/5/$10.00

©1996 American Vacuum Society

1981

1982

Inayoshi et al.: Formation of PTFE thin films

1982

FIG. 1. Schematic diagram of the experimental equipment for excimer laser ablation.

The XeCl excimer laser beam was focused by a quartz lens and irradiated the surface of the PTFE target in the reaction chamber through a quartz window. A Si~100! single crystal was used as the substrate, which was placed at a distance of 25 mm from the PTFE target. The energy density of the XeCl excimer laser beam at the surface of the PTFE target was maintained at about 1.5 J/cm2 with a repetition rate of 5 Hz. The laser pulse duration was 20 ns, and the ablation of the PTFE target was performed for 5 h. The typical thickness of the films deposited by using the XeCl excimer laser ablation was 2.4 mm. The experimental equipment for the evaporation of the PTFE bulk target by CO2 laser beam irradiation is shown in Fig. 2. A cw CO2 laser ~10.6 mm! was used for evaporation of the PTFE bulk target. The CO2 laser beam was introduced into the reaction chamber through a ZnSe window and irradiated the surface of the PTFE bulk target directly. Films were deposited on the substrate of single crystal Si~100!, which was placed at a distance of 60 mm above the PTFE bulk target. The CO2 laser power was changed from 5 to 15 W. Before starting the ablation and the evaporation, the re-

FIG. 2. Schematic diagram of the experimental equipment for CO2 laser evaporation. J. Vac. Sci. Technol. A, Vol. 14, No. 4, Jul/Aug 1996

FIG. 3. XPS spectrum of the film formed using the CO2 laser evaporation method at a CO2 laser power of 10 W.

action chamber was pumped down to a pressure of about 1.3 Pa with a rotary pump. The ablation and evaporation of the PTFE target were performed at room temperature. The films deposited using XeCl laser ablation and CO2 laser evaporation along with PTFE bulk targets were analyzed with x-ray photoelectron spectroscopy ~XPS!, Fourier transform infrared absorption spectroscopy ~FT-IR!, and scanning electron microscopy ~SEM!. III. RESULTS AND DISCUSSION The cross sections of deposited films were observed with SEM and the thickness of thin films, i.e., the distance from the surface of the substrates to the tops of the thin films, were measured from the SEM image. Since the surface of the deposited film was rough within 100 nm, the thickness of the thin film was calculated by the average value of a few points. The typical thickness of the thin film was ;10 mm. The deposition rate of the XeCl laser ablated film was typically 0.008 mm/min. Figure 3 shows the deposition rate of the CO2 laser evaporated film as a function of CO2 laser power. It is found that the PTFE thin film deposition started when the CO2 laser power was more than 4 W. This indicated that PTFE evaporation started when the CO2 laser power was more than 4 W, i.e., the surface temperature of the PTFE target reached its melting point ~325 °C! at a CO2 laser power of around 4 W. The deposition rate increased linearly with increasing CO2 laser power. A high deposition rate of more than 1 mm/min was attained using the laser evaporation method at a CO2 laser power of 7 W. By using the CO2 laser evaporation method, a PTFE thin film was formed at a much higher rate, by 2 orders of magnitude, than the case of the excimer laser ablation method. Furthermore, the deposition rate of a few mm/min is the highest rate of PTFE thin films formed by a variety of techniques. It was confirmed that the CO2 laser evaporation method was suitable for high rate deposition of PTFE thin films. The XeCl laser ablated films and the CO2 laser evaporated films together with the PTFE target were investigated by

1983


1983

FIG. 4. FT-IR spectra of the PTFE target, the CO2 laser evaporated film, and the XeCl laser ablated film.

XPS analysis. Figure 4 shows a typical XPS spectrum of the film formed using the CO2 laser evaporation method at a CO2 laser power of 10 W. There are a variety of peaks in the XPS spectrum which are attributed to fluorine ~F!, carbon ~C!, and nickel ~Ni!. Similar peaks were observed in the XPS spectrum for the case of the XeCl laser ablated films. Ni peaks may be due to the sample holder for the XPS analysis, which is made of Ni. Therefore, the surface of deposited films is considered to be composed of F and C atoms. F/C ratios of the PTFE target and the deposited films were calculated from the ratio of the intensity of F(1s) and C(1s) peaks of XPS spectra. Table I summarizes the F/C ratios of the PTFE target, the CO2 laser evaporated film, and the XeCl laser ablated film. From Table I, the F/C ratios were 1.3 and 1.5 for the PTFE target and the CO2 laser evaporated film, respectively. It was evident that the chemical components of the surface of the CO2 laser evaporated film was similar to those of the PTFE target. In the case of the XeCl laser ablation method, on the other hand, the F/C ratio of the deposited film was estimated to be 0.01. It was found that the XeCl laser ablated film was fluorine-deficient fluorocarbon film. A slice of the PTFE target and the deposited films were characterized by using a FT-IR spectrometer in transmission mode. Figure 5 shows FT-IR spectra of the PTFE target, the CO2 laser evaporated film, and the XeCl laser ablated film. A summary of the absorption peaks is shown in Table II. As shown in Fig. 5, a variety of peaks attributed to CF2 groups were observed around 500–1300 cm21 in the absorption spectra of the XeCl laser ablated film and the CO2 laser evaporated film. Features of the absorption spectra of the films were similar to that of the PTFE target for the most part, although the bands of 500–700 cm21 in the films were TABLE I. Comparison of fluorine-to-carbon ratios of the PTFE target, the CO2 laser evaporated film, and the XeCl laser ablated film obtained from the intensities of XPS F(1s) and C(1s) signals. Materials PTFE target CO2 laser evaporated film XeCl excimer laser ablated film

JVST A - Vacuum, Surfaces, and Films

F/C 1.3 1.5 0.01

FIG. 5. Deposition rate of the CO2 laser evaporated films as a function of CO2 laser power.

weak as compared with those of the PTFE target. The bands in the region of 1600–1800 cm21 and around 2360 cm21 on FT-IR spectra of the deposited films can be attributed to H2O and CO2 in air, respectively. It was found, therefore, that the CO2 laser evaporated film has the same chemical composition as the PTFE target. In the case of the XeCl laser ablated film, on the other hand, the peaks which did not appear in the FT-IR spectrum of the PTFE target were observed around 3000 cm21. This peaks can be assigned to the CH, CH2 , and CH3 stretching vibration from the left side, respectively. Accordingly, H atoms are considered to be contained at the states of CH, CH2 , and CH3 in the XeCl laser ablated film. Since the ablation of the PTFE target was performed in low vacuum ~1.3 Pa!, water vapor as an impurity still remaining

TABLE II. Summary of FT-IR absorption peaks. Wave number ~cm21! 2900–3000

2361 1500–1800 1200 1141 730–760 635 553 505

Assignment CH stretch CH2 stretch CH3 stretch OvCvO stretch H2O CF2 asym. stretch CF2 sym. stretch Amorphous PTFE CF2 wagging CF2 bending CF2 rocking

1984


in the reaction chamber gave rise to a gas-phase reaction in a plasma plume which was produced from the surface of the PTFE target by the irradiation of the intense UV laser beam. In the case of laser ablation of materials using a high power pulsed laser with a short pulse and short wavelength such as the excimer laser, after the laser radiation is absorbed by a solid surface, photon energy is converted first into electronic excitation and then into thermal, chemical, and mechanical energy to cause evaporation, ablation, excitation, plasma formation, and exfoliation. Evaporants from a plasma plume consist of a mixture of energetic species including electrons, atoms, radicals, ions, and clusters. The excited evaporants will react easily with any kind of gas in the environment. When water vapor exists in the region of plasma plume, active H atoms will be created by the plasma plume. As mentioned above, H atoms generated from water vapor would react with fluorocarbon radicals and clusters in the gas phase as well as on the growing surface, which presumably result in the existence of hydrocarbon in the film formed by using the XeCl laser ablation method. After the laser irradiation, a number of fluorocarbon radicals such as CF, CF2 , and CF3 created by the ablation of the PTFE target expand into the vacuum from the target surface. Moreover, H atoms will act as a scavenger of F atoms. Furthermore, the highly energetic species produced from the plume bombard the surface of the film and then the fluorocarbon radical and F atom will be generated from the film. As a result, films prepared by the XeCl laser ablation method were deficient in fluorine and the chemical composition of the films was different from that of the PTFE target as listed in Table I. However, if film formation will be carried out in high vacuum or in a gas atmosphere such as Ar, which is nonreactive, the deposited films will not contain impurity materials. The laser ablation method is very attractive. However, the disadvantage of the laser ablation method is the narrow forward angular distribution that makes large-area uniform film formation a difficult task. On the other hand, the CO2 laser evaporation method can form films in a large area uniformly because a number of fragments ejected from the surface of the target after the irradiation of the CO2 laser beam diffuse forward in a wide angular distribution. Moreover, in the case of CO2 laser evaporation of polymer materials, most of the photon energy of the CO2 laser beam absorbed by a polymer surface would be converted into thermal energy, and fragments of large molecular weight as oligomers would be ejected into the vacuum and repolymerized at the substrate surface. As a result, the chemical composition and structure of films prepared by the CO2 laser evaporation method corresponded to those of the PTFE target as shown in Table I and Fig. 5. Figures 6 and 7 show SEM images of the XeCl laser ablated film and the CO2 laser evaporated film, respectively. It was found from Fig. 6 that the XeCl laser ablated film had a smooth surface. As shown in Fig. 7, on the other hand, the surface of the CO2 laser evaporated film was very rough. According to Goodwin and Otis, most of the major neutral components created from the surface of the PTFE target irJ. Vac. Sci. Technol. A, Vol. 14, No. 4, Jul/Aug 1996

1984

FIG. 6. SEM image of the film formed using the XeCl laser ablation method.

radiated by the XeCl laser ~308 nm! was C2F4 .10 Therefore, it was expected that a number of small molecules such as C2F4 arrived at the substrate and the film with the smooth surface was formed using the XeCl laser ablation method. On the other hand, in the case of the CO2 laser evaporation method, a number of fragments with large molecule weight would be produced from the surface of PTFE irradiated by the CO2 laser ~10.6 mm!. Since the repolymerization of the large fragments cannot proceed well on the substrate at low substrate temperature, the surface of the deposited film became very rough. PTFE film formation was performed at a substrate temperature of 100 °C using the CO2 laser evaporation method for the purpose of improving the film quality. The result is shown in Fig. 8. It was found from a comparison between Figs. 8 and 7 that the surface of the film deposited at a substrate temperature of 100 °C was smoother than that of the film deposited at room temperature. Large fragments produced from the PTFE target by the CO2 laser evaporation method would be repolymerized effectively on the substrate by heating. In this case, the chemical composition and structure of the deposited film corresponded to those of the PTFE target from XPS and FT-IR analyses. Although the deposition rate in this case was 1 order of magnitude smaller than that in the case of film formation without substrate heating, the deposition rate was still 1 order of magnitude higher than that by XeCl laser ablation and the same or still higher compared with those by a variety of techniques.

1985


1985

FIG. 7. SEM image of the film formed using the CO2 laser evaporation method.

FIG. 8. SEM image of the film formed at a substrate temperature of 100 °C using the CO2 laser evaporation method.

IV. CONCLUSION

ACKNOWLEDGMENTS

Formation of PTFE thin films was carried out by using the CO2 laser evaporation method and the XeCl laser ablation method. The chemical composition and structure of films prepared by the CO2 laser evaporation method were similar to those of the PTFE target. Moreover, the CO2 laser evaporation method was found to form PTFE thin films at a high rate and the film with the smooth surface by substrate heating. It was confirmed that the CO2 laser evaporation method was an excellent technique to form PTFE thin films. On the other hand, in the case of the XeCl laser ablation method, the chemical composition of the deposited films was different from that of the PTFE target. It was confirmed that the XeCl laser ablation method was not suitable for formation of PTFE thin films at present.

The authors thank Hajime Kawamura at Meijo University for the use of FT-IR. Etsuko Mizuno at Meijo University for FT-IR analysis. Also the Instrument Center, Institute for Molecular Science, for the use of XPS.

JVST A - Vacuum, Surfaces, and Films

1

L. Holland, H. Biederman, and S. M. Ojha, Thin Solid Films 35, L19 ~1976!. 2 P. P. Luff and M. White, Vacuum 18, 437 ~1968!. 3 W. De Wilde, Thin Solid Films 24, 101 ~1974!. 4 R. Harrop and P. J. Harrop, Thin Solid Films 3, 109 ~1969!. 5 D. T. Morrison and T. Robertson, Thin Solid Films 15, 87 ~1973!. 6 I. H. Pratt and T. C. Lausman, Thin Solid Films 10, 151 ~1972!. 7 G. B. Blanchet, Appl. Phys. Lett. 62, 479 ~1993!. 8 G. B. Blanchet and S. Ismat Shah, Appl. Phys. Lett. 62, 1026 ~1993!. 9 S. Kuper and M. Stuke, Appl. Phys. Lett. 54, 4 ~1989!. 10 P. M. Goodwin and C. E. Otis, J. Appl. Phys. 69, 2584 ~1991!.