A variety of elements were implanted into ALE SrS, SrS:Ce, and SrS:Cu thin films in order to study their influence on the luminescence properties and to find possible dopants and codopants for improving SrS based blue EL phosphors. After necessary annealing, the luminescence of the ion beam modified samples was studied.
The selection of elements for implanting in SrS:Ce was based on their anticipated function in the phosphor [V]. For example, Na, K, P, and newly discovered Ag may be substitutional cations, while F and Cl are interstitial anions. These ions are expected to compensate the charge mismatch, influence the crystal field, or act as co-activators in the SrS:Ce matrix. Gallium, on the other hand, is known to be a flux agent that improves the crystallinity and thereby the luminescence.
Compared with the reference ALE SrS:Ce, implantation of F resulted in a clear emission blue shift without decrease in the PL intensity when annealing was at above 500 °C. The blue shift increased to a maximum of about 10 nm with increasing F concentration. However, excessive F doses (0.5 at.% F as compared to ~0.2 at.% Ce) and high temperature annealing (>700 °C) reduced the decay time (SN value(*) of 9.5 ns compared to the reference ALE SrS:Ce, which usually has a value of > 12 ns) and quenched the luminescence. A NRB study showed that there was severe F loss upon annealing and only traces (< 100 ppm) remained in the SrS:Ce bulk after annealing at 900 °C. In contrast to SrS:Ce,F, no clear spectrum blue shift was observed with implantation of the isoelectronic Cl and the emission intensity of SrS:Ce,Cl was less than that of the reference samples.
CeCl3 and CeF3 are frequently used source materials in the deposition of SrS:Ce by evaporation methods and sputtering.(22)-(24) Codoping of F or Cl has also been attempted by ALE.(25),(26) It is possible that the cerium ions are not entirely in a pure sulfide matrix but at least partially in a halide environment. Studies on Ce-doped alkaline earth chlorides showed that the Ce luminescence lies in the UV region and these hosts are able to incorporate rather large amount of Ce dopant (more than 5 mol.%) without concentration quenching [II]. But the UV emission from SrCl2-type environment has not been observed in, for example ALE SrS:Ce,Cl (SiCl4 as precursor) or in reactively evaporated SrS:Ce,Cl. It is likely that F and Cl may incorporate as charge compensators in the host. Some early work has suggested that the optimum Ce to halide ratio may be close to one.(27),(28) In reactively evaporated SrS:Ce,Mn,Cl, loss of Cl was observed when higher temperature was used during both film deposition and post annealing as compared to that at optimized deposition condition, and as a consequence, the EL decreased.(29)ß
A number of studies have shown that codoping with F results in either a more pronounced blue emission shift or better EL performance as compared to codoping with Cl.(22),(30) The implantation results seem to support such findings. It may be that F- is more electronegative than Cl- ion and hence results in higher excited energy levels of the Ce3+ ion. Furthermore, since the ionic radius of Cl- is comparable to that of S2-, while the F- ion is much smaller, it is possible that F- would locate at interstitial sites, while Cl- would prefer substitutional sites, and the influence of the two ions on the luminescence of SrS:Ce might then be different. Thirdly, the finding of only traces of implanted F in high temperature annealed SrS:Ce,F suggests that codoped F may also act as a flux agent for better crystallization. On the other hand, since CeF3 is thermally rather stable, it is understandable that CeCl3 may be a better choice for the evaporation process, despite its hygroscopic nature. CeCl3 is used in reactive evaporated SrS:Ce,Mn,Cl, and the EL performances are good.(31)
Unlike the implanted SrS:Ce,F samples whose PL was quenched after annealing at 700 °C, implantation of K into SrS:Ce resulted in an improved PL after annealing at above 700 °C compared to the reference SrS:Ce. Almost two-fold enhancement of the luminescence was achieved with SrS:Ce,K having optimum K doses and after annealing at 800 °C. Decay time studies showed that the SN values of most of the SrS:Ce,K samples were comparable to those of the reference sample. A further study was conducted by implanting the same amount of Ce and K ions into undoped SrS film with partially overlapping implantation areas. As shown in Fig. 6, the codoped SrS:Ce,K not only had higher PL intensity than the singly doped SrS:Ce, but quenching of the samples did not occur in SrS:Ce,K at high annealing temperature and dopant concentration.
Fig. 6. Dependence of PL intensity on the annealing temperature of Ce,K co-implanted SrS with various Ce and K concentration. The solid lines represent SrS:Ce,K and the dotted lines singly implanted SrS:Ce.
Sodium is used as a charge compensator in SrS:Ce,Na powders.(17) However, the beneficial effects of K implantation were not observed in Na implanted SrS:Ce. It may be relevant that the ionic radius of Na+ (1.02 Å) is smaller than that of Sr2+ (1.18 Å) and K+ (1.38 Å). In the case of Ce doped alkaline earth chlorides [II], even though codoping of Na+ or K+ ions enhances the luminescence of both CaCl2:Ce and SrCl2:Ce, but CaCl2:Ce,Na shows more intense luminescence than CaCl2:Ce,K, while SrCl2:Ce,Na shows weaker luminescence than SrCl2:Ce,K. A similar trend may well be at work in SrS:Ce. On the other hand, ion beam analysis revealed that the presence of Na at the interface between the phosphor and insulator may reduce the EL performances [III]. A mobility study indicated that Na diffuses rapidly in SrS:Ce even when annealing is at 600 °C.(32)
Preliminary results showed P and Ga implantation to quench the luminescence of ALE SrS:Ce films.
Since reactively evaporated SrS:Ce,Ag,Mn,Cl has the best reported luminance (170 cd/m2) among blue-green phosphors, study was made of the possibility of Ag codoping in ALE SrS:Ce thin films by ion implantation. Surprisingly, the implanted SrS:Ce,Ag showed significant PL already after annealing at 500 °C. The PL intensity of samples containing 0.5 at. % Ag and annealed at 800 °C was more than double that of reference samples. Furthermore, decay time at room temperature (RT) had an SN value of 22 ns, which is the best value recorded for an ALE SrS:Ce based thin film phosphor. Figure 7 shows the plot of SN values vs emission wavelength of an ALE SrS:Ce and an implanted SrS:Ce,Ag which were annealed at 800 °C, and a reference ALE SrS:Ce which was annealed at 500 °C immediately after deposition. The improved PL properties suggest that codoping with Ag+ is advantageous in ALE SrS:Ce without the need for other codopants such as used in reactively evaporated SrS:Ce,Ag,Mn,Cl.
Fig. 7. Comparison of SN values vs. wavelength for (a) implanted SrS:Ce,Ag (0.5 at.% 800 °C), (b) ALE SrS:Ce (800 °C), and (c) reference ALE SrS:Ce (510 °C). Open squares and circles represent the SN values after extraction of the fast decay component.
Several test EL devices were fabricated from the implanted SrS:Ce,Ag thin films and studied for their luminance.(33) When driving at 60 Hz, most of the pixels studied in the neighboring non-implanted area broke down before the turn-on (< 1 cd/m2), while all the SrS:Ce,Ag pixels exhibited luminance of greater than 10 cd/m2, despite the samples were treated at 800 °C for 2 hours. Furthermore, the threshold voltage of SrS:Ce,Ag TFEL devices was about 40 V lower than that of the reference ALE SrS:Ce devices. Preliminary studies demonstrate promising results of implanted SrS:Ce,Ag for TFEL applications.
The influence of codopants on the luminescence of SrS:Cu was examined by modifying ALE SrS:Cu thin films with Ag, Al, Ga, B, F, and Cl in a dose range of 0.05 to 0.8 at.%.(34) The PL studies were carried out after necessary thermal annealing at temperature of 500 to 800 °C.
In comparison with the reference ALE SrS:Cu, implantation of Ag and Al did not improve the PL intensity, and high doses tended to quench the Cu emission. However, blue emission from Ag centers was observed since, as will be discussed in chapter 5, Ag participates in the luminescence of SrS:Ag,Cu. Interestingly, a number of the implanted SrS:Cu,Ag samples showed only reduced Cu emission.
Implantation of F, Cl, and B improved the PL intensity of SrS:Cu. In particular, implantation of F and especially Cl into green SrS:Cu caused a luminescence shift towards blue. It is not yet clear whether the luminescence of implanted SrS:Cu,Cl is due to emission from the Cu+ center or to an emission of donor-acceptor pair type after the incorporation of Cl. In the case of implanted SrS:Cu,Ga, the luminescence intensity remained unchanged regardless of the implantation dose. It is worth mentioning that a set of Cu and O co-implanted samples showed codoping of O in SrS:Cu to improve the luminescence intensity.(35)
It is quite possible that the trivalent Ce3+ creates Sr vacancy in the SrS host, while monovalent Cu+ creates S vacancy. Positively charged ions such as K+ and Ag+ with right size would be more easily incorporated into SrS:Ce than would negatively charged ions such as F- and Cl-; and vice versa, the negative F- and Cl- ions might be preferred to the positive Ag+ and Al3+ ions in SrS:Cu. This does not of course exclude the beneficial effects of those less easily incorporated codopants such as Ag in the case of SrS:Ag,Cu. Nevertheless, ion beam modification of SrS based blue TFEL phosphors has demonstrated the potential of implantation in studying the luminescence properties of dopants and the effects of various codopants. There are a few reports on EL devices prepared by ion implantation,(36) but the modest results have not attracted much attention. However, there seem to be no fundamental reasons why a successful EL device could not be obtained. The full capacity of ion implantation in TFEL applications has yet to be explored.
* The decays of most of the SrS:Ce thin films are non-exponential as compared to the powder, which has a single decay time of t 0=27 ns. The areas under the normalized decay curves are integrated, therefore, to obtain the SN values for a more precise comparison of the decay properties. The SN/t 0 ratio gives the internal luminescence efficiency.
22. T.A. Oberacker and H.W. Schock, J. Cryst. Growth 159 (1996) 935.
23. W. Tong, T. Yang, W. Park, M. Chaichimansour, S. Schön, B.K. Wagner, and C.J. Summers, J. Electr. Mater. 26 (1997) 728.
24. K. Ohmi, T. Hirose, M. Harada, S. Tanaka, and H. Kobayashi, Jpn. J. Appl. Phys. 36 (1997) L33. 25. P. Soininen, L. Niinistö, E. Nykänen, and M. Leskelä, Appl. Surf. Sci. 75 (1994) 99.
26. E. Nykänen, P. Soininen, L. Niinistö, M. Leskelä, and E. Rauhala, in Proc. 7th Inter. Workshop Electroluminescence, Beijing, 1994 (Science, Beijing, 1996), pp. 97.
27. S. Tanda, A. Miyakoshi, and T. Nire, in Electroluminescence, in Springer Proc. Phys. 38 (eds. S. Shionoya and H. Kobayashi), Springer-Verlag, Berlin, 1989, pp. 180.
28. T. Tohda, M. Okajima, M. Yamamoto, and T. Matsuoka, Jpn. J. Appl. Phys. 30 (1991) 2786.
29. U. Troppenz, G. Bilger, W. Bohne, G. Gers, J. Kreissl, R.H. Mauch, K. Sieber, and K.O. Velthaus, in Inorganic and Organic Electroluminescence, EL 96 Berlin, (ed. R.H. Mauch and H.-E. Gumlich), Wissenschaft und Technik, Berlin, 1996, pp. 181.
30. D. Poelman, R. Vercaemst, R.L. Van Meirhaeghe, W.H. Laflere, and F. Cardon, Jpn. J. Appl. Phys. 32 (1993) 3477.
31. B. Hüttle, K. Troppenz, U, Herrmann, and R.H. Mauch, J. Cryst. Growth 159 (1996) 943.
32. R. Lappalainen, M. Karjalainen, R. Serimaa, S. Sevanto, and M. Leskelä, unpublished results.
33. W.-M. Li, M. Ritala, M. Leskelä, R. Lappalainen, E. Soininen, and L. Niinistö, 5th IUMRS Inter. Conf. Adv. Mater., Abstracts, 1999, pp. 346.
34. W.-M. Li, M. Ritala, M. Leskelä, R. Lappalainen, E. Soininen, L. Niinistö, C. Barthou, P. Benalloul, and J. Benoit, 4th Inter. Conf. Sci. Tech. Display Phosphors, Extended Abstracts, Bend, OR, 1998, pp. 259.
35. T. Sajavaara, R. Lappalainen, K. Arstila, W.-M. Li, M. Ritala, M. Leskelä, and E. Soininen, Nucl. Instr. Meth. Phys. Res. B 148 (1999) 715.
36. T. Parodos, H.P. Maruska, W. Halverson, R.A. Budzilek, D. Monarchie, and E. Schlam, SID 1993 Digest 24, pp. 777.