- Email: [email protected]
Low temperature effect on the electron beam induced degradation of ZnS:Cu,Al,Au phosphor powders
Applied Surface Science 193 (2002) 77–82
Low temperature effect on the electron beam induced degradation of ZnS:Cu,Al,Au phosphor powders K.T. Hillie, H.C. Swart* Department of Physics, University of the Orange Free State, P.O. Box 339, Bloemfontein 9300, South Africa Received 19 February 2002; received in revised form 27 February 2002; accepted 1 March 2002
Abstract ZnS:Cu,Al,Au phosphor powders were bombarded by an electron beam of 2 keV with the current density of 8.7 mA/cm2 at an oxygen pressure of 2 106 Torr. The studies were carried out at temperatures between 125 and 25 8C. Auger electron spectroscopy (AES) and cathodoluminescence (CL) were used to monitor the changes in the surface chemistry of the phosphor and luminous efficiency during electron bombardment, respectively. Thermoluminescence (TL) glow curves produced by the phosphor warming up from 125 8C were also detected. Degradation was manifested by the non-luminescent ZnO layer that formed on the surface of the phosphor according to the electron stimulated surface chemical reaction (ESSCR) mechanism. A decrease in temperature lead to an increase in the surface stay time of the oxygen molecules on the surface with the increase in the ESSCR probability. The effect of thermal quenching of the initial CL was reduced at lower temperatures. The residual water vapour in the vacuum, however, had a significant effect on the rate of degradation because it was frozen at lower temperatures. The formation of O defects during electron bombardment, which act as electron traps, were confirmed through TL glow curves. # 2002 Published by Elsevier Science B.V. PACS: 68.35D; 78.60K; 85.45F Keywords: Cathodoluminescence; Thermoluminescence; ZnS phosphor; Degradation
1. Introduction The field emission display (FED) technology has been aggressively developed due to the high cathode ray tube (CRT) picture quality with much reduced weight and bulk [1–3]. Its advantage over the market dominating active-matrix liquid-crystal display (AMLCD) is that it creates it own light flux as part of operation, which is achieved by the cathodoluminescent (CL) excitation of phosphors in a similar manner to CRT. This provides fast video response, * Corresponding author. Fax: þ27-51-4013507. E-mail address: [email protected] (H.C. Swart).
wide viewing angle, high contrast ratio and a large range of operating temperatures than the AMLCD [4]. ZnS based CRT phosphors have been intensively investigated for FED applications due to brightness and high efficiency shown in the CRT environment [5]. Sebastian et al. [6] and Swart et al. [7] collected data on the degradation of ZnS based phosphors that culminated in a postulated model for the thin films and powders, respectively. According to this model, a non-luminescent ZnO layer forms on the surface of the ZnS phosphor with a release of volatile SO2 gas in an oxygen gas ambient [8] through an electron stimulated surface chemical reaction (ESSCR) mechanism which resulted in a decrease in luminescence intensity of the
0169-4332/02/$ – see front matter # 2002 Published by Elsevier Science B.V. PII: S 0 1 6 9 - 4 3 3 2 ( 0 2 ) 0 0 2 1 1 - 8
78
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
phosphor. Since the ESSCR is not accountable for the entire loss of efficiency of the phosphor, complimentary point defect theories have been postulated [9,10]. The FED technology has its own impediments because of the high surface to volume ratio resulting from compactness. This means that the pressure of the usual vacuum residual gases is much higher compared to the CRT in operation. The residual gases constantly impinge the surface of the phosphor during operation and therefore play a major role in the degradation of the phosphor [11]. The low voltage used for operation has to be compensated by the increase in the current density to achieve optimum brightness from the phosphor. The ZnS:Cu,Al,Au (P22G) phosphor powder degradation at different current densities was reported in a previous publication [12]. The studies lead to the determination of the role of temperature on the degradation when the P22G phosphor powders were bombarded by an electron beam at temperatures ranging from room temperature up to 300 8C. The results confirmed that the reaction between oxygen and ZnS is due to ESSCR and not to local heating of the surface by the electron beam, and that the surface reaction rate depends on the dissociation cross-section of the oxygen molecule and the time that an arriving molecule spends on the surface [13]. In this letter, the results of electron bombardment of the P22G phosphor powder at temperatures from 125 8C to room temperature are reported. Auger electron spectroscopy (AES) and the CL both excited by the same primary electron beam were utilised to monitor changes in surface composition and luminous efficiency of the phosphor powder during electron bombardment at different temperatures in O2 ambient. Thermoluminescence (TL) glow curves were also monitored using a Spectra Pritchard Photometer.
2. Experimental details Standard P22G (ZnS:Cu,Al,Au) phosphor powder from Osram Sylvania, USA, was subjected to 2 keV electron bombardment at 125 8C and at room temperature in 2 106 Torr O2 pressure backfilled from the vacuum base pressure of <3 109 Torr. The Auger measurements were done in a UHV chamber using a PHI model 549 spectrometer. The electron beam current density was kept constant at 8.7 mA/cm2
and the beam size was stable during each experiment. CL intensity measurements were done at an angle of 908 to the incident electron beam. The CL was collected through a sapphire port using a Spectra Pritchard Photometer. For the low temperature experiments, a stainless steel sample stage was constructed for the introduction of liquid nitrogen to cool down the sample. The samples were degraded by electron bombardment under different experimental conditions. The degradation were done at (a) room temperature, (b) with cooling of the sample stage alone (sample temperature was 125 8C), (c) with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system and (d) at room temperature with only the cryogenic pump cooled off. Degraded and samples that were not degraded were subjected to an electron dose of 10 C/cm2 at a temperature of 125 8C. The electron beam was switch off and TL glow curves were detected as the phosphor was warming up from 125 8C to room temperature. The partial gas pressures were monitored using a residual gas analyser (RGA).
3. Results and discussion Fig. 1 shows the Auger peak to peak heights (APPHs) of S, Zn, C 5 and O 5 together with the CL intensity as function of Coulomb dose taken in 2 106 Torr O2 ambient at room temperature during electron bombardment. The graph depicts the ESSCR mechanism. The initial step is the core level ionisation of Zn through the interatomic Auger cross transition stimulated by the electron beam leaving the S atom with the net positive charge surrounded by positive metal ions, which occurs simultaneously with the electron beam dissociation of O2 to reactive O species. As soon as the adventitious C is depleted from the surface as volatile COx compounds, S is then released as SO2 [8] with an apparent accumulation of O that forms a non-luminescent ZnO layer, hence the correlated decrease in the CL intensity. Fig. 2 shows the RGA spectra between 0 and 50 amu before degradation at room temperature (a), with cooling of the sample stage (b) and with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system (c). At <3 109 Torr, it was apparent that there was still a small amount of
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
79
Fig. 1. The APPHs of S, Zn, C 5 and O 5 together with the CL intensity as a function of Coulomb dose at room temperature.
water vapour present in the vacuum. The presence of water vapour would increase the rate of degradation of the phosphor [6]. When the phosphor is cooled down, the bellowed stainless steel that formed part of the
sample stage inserted into the UHV system, has a vast surface area leading to a large amount of residual water vapour to be frozen on it with the introduction of liquid nitrogen. The water vapour was further reduced
Fig. 2. RGA spectra at <3 109 Torr base pressure between 0 and 50 amu before degradation at room temperature (a), with cooling of the sample stage (b) and with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system (c).
80
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
Fig. 3. Normalised S APPHs as a function of Coulomb dose at room temperature (a), with cooling of the sample stage (b), with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system (c) and the cryogenic pump at room temperature (d).
when the cryogenic pump was used for the whole vacuum system. From the RGA spectra, Fig. 2, the water vapour peak decreased by 31% from the peak at room temperature when the sample was at 125 8C and by 63% when the cryogenic pump was also used. Fig. 3 shows the normalised S APPHs as a function of Coulomb dose at room temperature (a), with cooling of the sample stage (b), with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system (c) and with the cryogenic pump at room temperature (d). According to the studies of ESSCR at temperatures between 45 and 300 8C, the reaction rate is controlled by the mean surface stay time of the adsorbed molecule on the surface of the phosphor [13]. It is then supposed that at low temperatures the rate of reaction will increase, but the influence of the small amount of water vapour at these temperatures is significant. The effect of reduced water vapour competes with the effect of the longer mean surface stay time of the adsorbed molecule at 125 8C (Fig. 3(b)) and hence a moderate rate of reaction that is comparable with the rate at room temperature (Fig. 3(a)) both without the cooling of the system. Fig. 3(c) and (d) are the S APPHs plotted against the Coulomb dose at 125 8C and at room
temperature both with the cryogenic pump on, respectively. Both graphs manifest a slower reaction rate proving that the residual water vapour is significantly reduced when liquid nitrogen is introduced for cryogenic pumping. They also exhibit an increase in the S APPHs during the first 30 C/cm2 of electron bombardment, which is due to C leaving the surface at a slower rate. The reaction rate at 125 8C (Fig. 3(c)) is slightly faster than the reaction rate at room temperature (Fig. 3(d)). Although the residual water vapour was present at room temperature, the reaction was slower due to the shorter surface mean stay time of the oxygen molecule at room temperature compared to 125 8C [13]. This observation substantiates the role of the mean surface stay time of the molecule on the ESSCR, which is longer at lower temperatures and so the relatively increased reaction rate. Fig. 4 shows the CL intensity as a function of temperature before degradation. The CL intensity increases with decreasing temperature, which can be attributed to the decrease of the thermal quenching effect that occurs at elevated temperatures [13]. During electron beam bombardment in an O2 ambient, O atoms form iso-electronic impurities in the ZnS lattice by substituting for the S atoms [6]. The O atoms
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
81
Fig. 4. CL intensity as a function of temperature before degradation.
have a higher electron affinity than the substituted S atoms and the inclination to attract and trap electrons on these sites will increase. Fig. 5 shows the TL glow curves of the reference sample (a) and the degraded sample (b), which were both bombarded with an electron dose of 10 C/cm2 at 125 8C before acquiring the
glow curves. The peak at 58 8C, Fig. 5(b), on the glow curve that was attained after a degraded phosphor powder was cooled down to 125 8C and bombarded with an electron dose of 10 C/cm2 exhibits the effect of the O iso-electronic traps. Sebastian [10] also reported the peak at 60 8C attributed to
Fig. 5. TL glow curves of the reference (a) and the degraded sample (b).
82
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
O iso-electronic traps on the TL glow curves acquired on ZnS:Mn phosphor films. It confirms that during the electron bombardment of the ZnS:Cu,Au,Al phosphor powder, O iso-electronic traps were created and they captured free electrons that were excited during the later bombardment at 125 8C. These electrons were then released to the conduction band and later detected as TL as the sample was warming up to room temperature. The reference sample, Fig. 5(a), represents the sample that was bombarded at –125 8C with an electron dose of 10 C/cm2 without prior degradation. Sebastian [10] observed a peak at 80 8C on the reference ZnS:Mn film, which was attributed to S interstitials. This peak was not observed on the ZnS:Cu,Au,Al phosphor powder, which means that the S interstitial could have probably resulted from the film growth process [14]. Structural defects, such as dislocations and stacking faults formed in the ZnS films have also been reported to seriously affect the CL intensity within a distance of 200 nm from the interface [15]. Although there is a possible contribution to degradation by some other complex defect [16], however, their effect was not readily observed on the TL glow curves.
4. Conclusion The effect of the mean surface stay time of the absorbed molecule together with the significant influence of the residual water vapour during electron bombardment at low temperatures were verified. The
reduction of the thermal quenching effect to the initial CL is observed at lower temperatures. The formation of O defects during electron bombardment, which act as electron traps thereby reducing the luminescence efficiency was also confirmed through TL glow curves. References [1] S.M. Jacobsen, Journal of SID 4 (4) (1996) 331. [2] R. Baptist, F. Bachelet, C. Constacias, J. Vac. Sci. Technol. B 15 (2) (1997) 385. [3] M.M. Slusarczuk, Mat. Res. Soc. Symp. Proc. 424 (1997) 363. [4] P.H. Holloway, J. Sebastian, T. Trottier, H. Swart, R.O. Petersen, Solid State Technol. 8 (1995) 47. [5] T. Hase, T. Kano, E. Nakazawa, H. Yamamoto, Adv. Electron. Electrophys. 79 (1990) 271. [6] J.S. Sebastian, H.C. Swart, T.A. Trottier, S.L. Jones, P.H. Holloway, J. Vac. Sci. Technol. A 15 (4) (1997) 1. [7] H.C. Swart, J.S. Sebastian, T.A. Trottier, S.L. Jones, P.H. Holloway, J. Vac. Sci. Technol. A 14 (3) (1996) 1657. [8] L. Oosthuizen, H.C. Swart, P.E. Viljoen, P.H. Holloway, G.L.P. Berning, Appl. Surf. Sci. 120 (1997) 9. [9] C.W. Wang, T.J. Sheu, M. Yokoyama, Appl. Surf. Sci. 113/ 114 (1997) 709. [10] J.S. Sebastian, Ph.D. Thesis, University of Florida, FL, USA, 1998. [11] H.C. Swart, T.A. Trottier, J.S. Sebastian, S.L. Jones, P.H. Holloway, J. Appl. Phys. 83 (9) (1998) 1. [12] H.C. Swart, K.T. Hillie, Surf. Interf. Anal. 30 (2000) 383. [13] H.C. Swart, K.T. Hillie, A.P. Greeff, Surf. Interf. Anal. 32 (2001) 110. [14] J. Fang, P.H. Holloway, J.E. Yu, K.S. Jones, B. Pathangey, E. Brettscheneider, T.L. Anderson, Appl. Surf. Sci. 70/71 (1993) 701. [15] T. Mitsui, N. Yamamoto, Jpn. J. Appl. Phys. 39 (2000) 1172. [16] Y. Shono, T. Oka, J. Cryst. Growth 210 (2000) 272.
Low temperature effect on the electron beam induced degradation of ZnS:Cu,Al,Au phosphor powders K.T. Hillie, H.C. Swart* Department of Physics, University of the Orange Free State, P.O. Box 339, Bloemfontein 9300, South Africa Received 19 February 2002; received in revised form 27 February 2002; accepted 1 March 2002
Abstract ZnS:Cu,Al,Au phosphor powders were bombarded by an electron beam of 2 keV with the current density of 8.7 mA/cm2 at an oxygen pressure of 2 106 Torr. The studies were carried out at temperatures between 125 and 25 8C. Auger electron spectroscopy (AES) and cathodoluminescence (CL) were used to monitor the changes in the surface chemistry of the phosphor and luminous efficiency during electron bombardment, respectively. Thermoluminescence (TL) glow curves produced by the phosphor warming up from 125 8C were also detected. Degradation was manifested by the non-luminescent ZnO layer that formed on the surface of the phosphor according to the electron stimulated surface chemical reaction (ESSCR) mechanism. A decrease in temperature lead to an increase in the surface stay time of the oxygen molecules on the surface with the increase in the ESSCR probability. The effect of thermal quenching of the initial CL was reduced at lower temperatures. The residual water vapour in the vacuum, however, had a significant effect on the rate of degradation because it was frozen at lower temperatures. The formation of O defects during electron bombardment, which act as electron traps, were confirmed through TL glow curves. # 2002 Published by Elsevier Science B.V. PACS: 68.35D; 78.60K; 85.45F Keywords: Cathodoluminescence; Thermoluminescence; ZnS phosphor; Degradation
1. Introduction The field emission display (FED) technology has been aggressively developed due to the high cathode ray tube (CRT) picture quality with much reduced weight and bulk [1–3]. Its advantage over the market dominating active-matrix liquid-crystal display (AMLCD) is that it creates it own light flux as part of operation, which is achieved by the cathodoluminescent (CL) excitation of phosphors in a similar manner to CRT. This provides fast video response, * Corresponding author. Fax: þ27-51-4013507. E-mail address: [email protected] (H.C. Swart).
wide viewing angle, high contrast ratio and a large range of operating temperatures than the AMLCD [4]. ZnS based CRT phosphors have been intensively investigated for FED applications due to brightness and high efficiency shown in the CRT environment [5]. Sebastian et al. [6] and Swart et al. [7] collected data on the degradation of ZnS based phosphors that culminated in a postulated model for the thin films and powders, respectively. According to this model, a non-luminescent ZnO layer forms on the surface of the ZnS phosphor with a release of volatile SO2 gas in an oxygen gas ambient [8] through an electron stimulated surface chemical reaction (ESSCR) mechanism which resulted in a decrease in luminescence intensity of the
0169-4332/02/$ – see front matter # 2002 Published by Elsevier Science B.V. PII: S 0 1 6 9 - 4 3 3 2 ( 0 2 ) 0 0 2 1 1 - 8
78
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
phosphor. Since the ESSCR is not accountable for the entire loss of efficiency of the phosphor, complimentary point defect theories have been postulated [9,10]. The FED technology has its own impediments because of the high surface to volume ratio resulting from compactness. This means that the pressure of the usual vacuum residual gases is much higher compared to the CRT in operation. The residual gases constantly impinge the surface of the phosphor during operation and therefore play a major role in the degradation of the phosphor [11]. The low voltage used for operation has to be compensated by the increase in the current density to achieve optimum brightness from the phosphor. The ZnS:Cu,Al,Au (P22G) phosphor powder degradation at different current densities was reported in a previous publication [12]. The studies lead to the determination of the role of temperature on the degradation when the P22G phosphor powders were bombarded by an electron beam at temperatures ranging from room temperature up to 300 8C. The results confirmed that the reaction between oxygen and ZnS is due to ESSCR and not to local heating of the surface by the electron beam, and that the surface reaction rate depends on the dissociation cross-section of the oxygen molecule and the time that an arriving molecule spends on the surface [13]. In this letter, the results of electron bombardment of the P22G phosphor powder at temperatures from 125 8C to room temperature are reported. Auger electron spectroscopy (AES) and the CL both excited by the same primary electron beam were utilised to monitor changes in surface composition and luminous efficiency of the phosphor powder during electron bombardment at different temperatures in O2 ambient. Thermoluminescence (TL) glow curves were also monitored using a Spectra Pritchard Photometer.
2. Experimental details Standard P22G (ZnS:Cu,Al,Au) phosphor powder from Osram Sylvania, USA, was subjected to 2 keV electron bombardment at 125 8C and at room temperature in 2 106 Torr O2 pressure backfilled from the vacuum base pressure of <3 109 Torr. The Auger measurements were done in a UHV chamber using a PHI model 549 spectrometer. The electron beam current density was kept constant at 8.7 mA/cm2
and the beam size was stable during each experiment. CL intensity measurements were done at an angle of 908 to the incident electron beam. The CL was collected through a sapphire port using a Spectra Pritchard Photometer. For the low temperature experiments, a stainless steel sample stage was constructed for the introduction of liquid nitrogen to cool down the sample. The samples were degraded by electron bombardment under different experimental conditions. The degradation were done at (a) room temperature, (b) with cooling of the sample stage alone (sample temperature was 125 8C), (c) with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system and (d) at room temperature with only the cryogenic pump cooled off. Degraded and samples that were not degraded were subjected to an electron dose of 10 C/cm2 at a temperature of 125 8C. The electron beam was switch off and TL glow curves were detected as the phosphor was warming up from 125 8C to room temperature. The partial gas pressures were monitored using a residual gas analyser (RGA).
3. Results and discussion Fig. 1 shows the Auger peak to peak heights (APPHs) of S, Zn, C 5 and O 5 together with the CL intensity as function of Coulomb dose taken in 2 106 Torr O2 ambient at room temperature during electron bombardment. The graph depicts the ESSCR mechanism. The initial step is the core level ionisation of Zn through the interatomic Auger cross transition stimulated by the electron beam leaving the S atom with the net positive charge surrounded by positive metal ions, which occurs simultaneously with the electron beam dissociation of O2 to reactive O species. As soon as the adventitious C is depleted from the surface as volatile COx compounds, S is then released as SO2 [8] with an apparent accumulation of O that forms a non-luminescent ZnO layer, hence the correlated decrease in the CL intensity. Fig. 2 shows the RGA spectra between 0 and 50 amu before degradation at room temperature (a), with cooling of the sample stage (b) and with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system (c). At <3 109 Torr, it was apparent that there was still a small amount of
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
79
Fig. 1. The APPHs of S, Zn, C 5 and O 5 together with the CL intensity as a function of Coulomb dose at room temperature.
water vapour present in the vacuum. The presence of water vapour would increase the rate of degradation of the phosphor [6]. When the phosphor is cooled down, the bellowed stainless steel that formed part of the
sample stage inserted into the UHV system, has a vast surface area leading to a large amount of residual water vapour to be frozen on it with the introduction of liquid nitrogen. The water vapour was further reduced
Fig. 2. RGA spectra at <3 109 Torr base pressure between 0 and 50 amu before degradation at room temperature (a), with cooling of the sample stage (b) and with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system (c).
80
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
Fig. 3. Normalised S APPHs as a function of Coulomb dose at room temperature (a), with cooling of the sample stage (b), with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system (c) and the cryogenic pump at room temperature (d).
when the cryogenic pump was used for the whole vacuum system. From the RGA spectra, Fig. 2, the water vapour peak decreased by 31% from the peak at room temperature when the sample was at 125 8C and by 63% when the cryogenic pump was also used. Fig. 3 shows the normalised S APPHs as a function of Coulomb dose at room temperature (a), with cooling of the sample stage (b), with the cooling of the sample stage plus the cryogenic pump for the entire vacuum system (c) and with the cryogenic pump at room temperature (d). According to the studies of ESSCR at temperatures between 45 and 300 8C, the reaction rate is controlled by the mean surface stay time of the adsorbed molecule on the surface of the phosphor [13]. It is then supposed that at low temperatures the rate of reaction will increase, but the influence of the small amount of water vapour at these temperatures is significant. The effect of reduced water vapour competes with the effect of the longer mean surface stay time of the adsorbed molecule at 125 8C (Fig. 3(b)) and hence a moderate rate of reaction that is comparable with the rate at room temperature (Fig. 3(a)) both without the cooling of the system. Fig. 3(c) and (d) are the S APPHs plotted against the Coulomb dose at 125 8C and at room
temperature both with the cryogenic pump on, respectively. Both graphs manifest a slower reaction rate proving that the residual water vapour is significantly reduced when liquid nitrogen is introduced for cryogenic pumping. They also exhibit an increase in the S APPHs during the first 30 C/cm2 of electron bombardment, which is due to C leaving the surface at a slower rate. The reaction rate at 125 8C (Fig. 3(c)) is slightly faster than the reaction rate at room temperature (Fig. 3(d)). Although the residual water vapour was present at room temperature, the reaction was slower due to the shorter surface mean stay time of the oxygen molecule at room temperature compared to 125 8C [13]. This observation substantiates the role of the mean surface stay time of the molecule on the ESSCR, which is longer at lower temperatures and so the relatively increased reaction rate. Fig. 4 shows the CL intensity as a function of temperature before degradation. The CL intensity increases with decreasing temperature, which can be attributed to the decrease of the thermal quenching effect that occurs at elevated temperatures [13]. During electron beam bombardment in an O2 ambient, O atoms form iso-electronic impurities in the ZnS lattice by substituting for the S atoms [6]. The O atoms
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
81
Fig. 4. CL intensity as a function of temperature before degradation.
have a higher electron affinity than the substituted S atoms and the inclination to attract and trap electrons on these sites will increase. Fig. 5 shows the TL glow curves of the reference sample (a) and the degraded sample (b), which were both bombarded with an electron dose of 10 C/cm2 at 125 8C before acquiring the
glow curves. The peak at 58 8C, Fig. 5(b), on the glow curve that was attained after a degraded phosphor powder was cooled down to 125 8C and bombarded with an electron dose of 10 C/cm2 exhibits the effect of the O iso-electronic traps. Sebastian [10] also reported the peak at 60 8C attributed to
Fig. 5. TL glow curves of the reference (a) and the degraded sample (b).
82
K.T. Hillie, H.C. Swart / Applied Surface Science 193 (2002) 77–82
O iso-electronic traps on the TL glow curves acquired on ZnS:Mn phosphor films. It confirms that during the electron bombardment of the ZnS:Cu,Au,Al phosphor powder, O iso-electronic traps were created and they captured free electrons that were excited during the later bombardment at 125 8C. These electrons were then released to the conduction band and later detected as TL as the sample was warming up to room temperature. The reference sample, Fig. 5(a), represents the sample that was bombarded at –125 8C with an electron dose of 10 C/cm2 without prior degradation. Sebastian [10] observed a peak at 80 8C on the reference ZnS:Mn film, which was attributed to S interstitials. This peak was not observed on the ZnS:Cu,Au,Al phosphor powder, which means that the S interstitial could have probably resulted from the film growth process [14]. Structural defects, such as dislocations and stacking faults formed in the ZnS films have also been reported to seriously affect the CL intensity within a distance of 200 nm from the interface [15]. Although there is a possible contribution to degradation by some other complex defect [16], however, their effect was not readily observed on the TL glow curves.
4. Conclusion The effect of the mean surface stay time of the absorbed molecule together with the significant influence of the residual water vapour during electron bombardment at low temperatures were verified. The
reduction of the thermal quenching effect to the initial CL is observed at lower temperatures. The formation of O defects during electron bombardment, which act as electron traps thereby reducing the luminescence efficiency was also confirmed through TL glow curves. References [1] S.M. Jacobsen, Journal of SID 4 (4) (1996) 331. [2] R. Baptist, F. Bachelet, C. Constacias, J. Vac. Sci. Technol. B 15 (2) (1997) 385. [3] M.M. Slusarczuk, Mat. Res. Soc. Symp. Proc. 424 (1997) 363. [4] P.H. Holloway, J. Sebastian, T. Trottier, H. Swart, R.O. Petersen, Solid State Technol. 8 (1995) 47. [5] T. Hase, T. Kano, E. Nakazawa, H. Yamamoto, Adv. Electron. Electrophys. 79 (1990) 271. [6] J.S. Sebastian, H.C. Swart, T.A. Trottier, S.L. Jones, P.H. Holloway, J. Vac. Sci. Technol. A 15 (4) (1997) 1. [7] H.C. Swart, J.S. Sebastian, T.A. Trottier, S.L. Jones, P.H. Holloway, J. Vac. Sci. Technol. A 14 (3) (1996) 1657. [8] L. Oosthuizen, H.C. Swart, P.E. Viljoen, P.H. Holloway, G.L.P. Berning, Appl. Surf. Sci. 120 (1997) 9. [9] C.W. Wang, T.J. Sheu, M. Yokoyama, Appl. Surf. Sci. 113/ 114 (1997) 709. [10] J.S. Sebastian, Ph.D. Thesis, University of Florida, FL, USA, 1998. [11] H.C. Swart, T.A. Trottier, J.S. Sebastian, S.L. Jones, P.H. Holloway, J. Appl. Phys. 83 (9) (1998) 1. [12] H.C. Swart, K.T. Hillie, Surf. Interf. Anal. 30 (2000) 383. [13] H.C. Swart, K.T. Hillie, A.P. Greeff, Surf. Interf. Anal. 32 (2001) 110. [14] J. Fang, P.H. Holloway, J.E. Yu, K.S. Jones, B. Pathangey, E. Brettscheneider, T.L. Anderson, Appl. Surf. Sci. 70/71 (1993) 701. [15] T. Mitsui, N. Yamamoto, Jpn. J. Appl. Phys. 39 (2000) 1172. [16] Y. Shono, T. Oka, J. Cryst. Growth 210 (2000) 272.