Epitaxially grown monocrystalline garnet cathode‐ray tube phosphor screens
`J. M. Robertson and M. W. van Tol
`
`Citation: Applied Physics Letters 37, 471 (1980); doi: 10.1063/1.91728
`View online: http://dx.doi.org/10.1063/1.91728
`View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/37/5?ver=pdfcov
`Published by the AIP Publishing
`
`Articles you may be interested in
`The Cathode‐Ray Tube: Technology, History, and Applications
`Am. J. Phys. 61, 669 (1993); 10.1119/1.17190
`
`The fine‐beam cathode‐ray tube
`Phys. Teach. 22, 80 (1984); 10.1119/1.2341472
`
`Measurement of the Screen Potential of Cathode‐Ray Tubes
`Rev. Sci. Instrum. 26, 1201 (1955); 10.1063/1.1715221
`
`Method of Measuring the Screen Potential of Cathode‐Ray Tubes
`Rev. Sci. Instrum. 26, 647 (1955); 10.1063/1.1715276
`
`Illumination of Cathode‐Ray Oscillograph Screen for Photography
`Rev. Sci. Instrum. 19, 271 (1948); 10.1063/1.1741244
`
`
`
` Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. Download to IP: 198.65.204.101 On: Wed, 23 Nov
`2016 16:41:12
`
`VIZIO 1015
`
`

`
`Epitaxially grown monocrystalline garnet cathode-ray tube phosphor
`screens
`J. M. Robertson and M. W. van Tol
`Eindhoven Research Laboratories. Eindhoven. The Netherlands
`
`(Received 10 March 1980; accepted for publication 17 June 1980)
`
`The technique of liquid phase epitaxy has been used to produce single-crystal garnet phosphor
`screens for cathode-ray tubes. The films were rare-earth-doped yttrium aluminum garnets
`(Y AG's) grown onto Y AG substrates using a PbO:B20 3 flux. We have studied the light output of
`these layers as a function of film growth temperature, rare-earth activator concentration, and
`incident power density. With Ce:Y AG layers it was possible to use beam power densities up to 108
`W 1m2 and this produced a radiance of over lOS W Icm2 sr.
`PACS numbers: 68.55. + b, 78.60.Hk
`
`One of the limiting factors in obtaining a bright cathode
`luminescent device is the heat conductivity ofthe phosphor
`screen. As the incident electron beam density increases, ther(cid:173)
`mal quenching occurs when the phosphor is heated to a tem(cid:173)
`perature at which the efficiency decreases. Eventually the
`phosphors "burn out" and become permanently degraded.
`The actual temperature of the phosphor layer depends not
`only on the incident beam energy, but also on the manner in
`which the phosphor screen was fabricated. Normally the lu(cid:173)
`minescent powder is deposited on a glass substrate. Such a
`structure suffers from both a low heat conductivity of the
`powder and a poor thermal contact between phosphor and
`the substrate. The present work was undertaken to over(cid:173)
`come these drawbacks in order to obtain higher-brightness
`cathode ray tubes (CRT's). I We have chosen yttrium alumi(cid:173)
`num garnet (Y 3AIsOd as a substrate material because it is a
`good host material for various luminescent ions and because
`of its high thermal conductivity.
`We used the method ofliquid phase epitaxy (LPE)2-4 to
`grow Y 3AIsO 12 layers onto Y AG substrates, because this
`ensured a perfect heat contact between phosphor and the
`substrate. The solvent was PbO:B20 3, in which the constitu(cid:173)
`ent garnet oxides are dissolved. A typical melt composition
`used for the growth ofYAG was PbO:B203:Y203:Alz03
`:Re20 3 = 450.0:11.65:3.85:6.52:0.02 (g). This was premelt(cid:173)
`ed into a 90-ml Pt/5% Au crucible using a rf generator and
`then placed into a standard LPE furnace. 3 The activator ox(cid:173)
`ides used in this study were Th40 7, Eu20 3, Pr 203' Tm20 3,
`and ce02. The substrates were Czochralski-grown YAG,
`usually [111] and 2.54 cm in diameter.
`The efficiency of the layers was determined in a de(cid:173)
`mountable cathode ray tube. The samples were given an Al
`coating of about 0.08 pm in order to define the energy of the
`incident electron beam and to carry off the current. The sam(cid:173)
`ples were excited by a continuous current of 10 pA at 10k V
`in a stationary spot of 4 mm diameter. The light output was
`measured in transmission. After correction for the energy
`absorbed in the Al layer (3 k V) and for the refractive index of
`the phosphor (n = 1.83), the measurements yield the inter(cid:173)
`nal energy efficiency 1]. This is only true for films of good
`quality, i.e., in the absence of surface scattering. In our case
`this was ensured, because the lattice parameter mismatch
`between film and substrate was always less than 0.018 A and
`
`no cracking or facetting occurred. 5
`The chemical composition of the layers was determined
`by an electron-probe microanalysis where the x-ray line in(cid:173)
`tensities were compared with those for standard materials.
`The energy efficiency 1] can be correlated to the chemi(cid:173)
`cal composition of the layers. It appears that both the Pb ions
`which are incorporated in the garnet lattice from the flux as
`well as the Re activator ions in the layers influence the effi(cid:173)
`ciency. It is well known from the LPE of iron gl\rnets that the
`amount ofPb incorporated from the flux increases when the
`growth temperature is lowered. 6 This holds to a lesser extent
`for the noniron garnets,7 and in the present case this is also
`observed.
`The influence of the activator ion concentration is
`strongly dependent on the Re ion actually used. In Fig. 1 the
`energy efficiency 1] is plotted as a function of the concentra(cid:173)
`tion per formula unit for [111] Y AG films doped with Ce3 + ,
`Eu3 + , Th3 + , and Tm3
`'I- • It is clear that Tm-doped Y AG
`has its maximum at a lower concentration than Eu or Th.
`This is in accordance with other work on the concentration
`9 Europium and
`quenching ofTm in different host lattices. 8
`.
`terbium show an almost identical behavior, except that the
`overall efficiency ofTh is higher. The cerium ion is a special
`case. The efficiency rises steeply with concentration, which
`shows that the Ce ion is a very efficient energy absorber. It
`
`%
`10
`
`5
`
`o~------------~--------------~~o
`o x-
`01
`0.2
`
`FIG. I. Light output as a function of the activator ion concentration. The
`stationary exciting electron beam of 10 J.iA 10 kV had a diameter of 4 mm.
`
`471
`
`Appl. Phys. Lett. 37(5), 1 September 1980
`
`0003-6951/80/170471-02$00.50
`
`© 1980 American Institute of Physics
`
`471
`
` Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. Download to IP: 198.65.204.101 On: Wed, 23 Nov
`2016 16:41:12
`
`VIZIO 1015
`
`

`
`300
`
`~
`Q.
`
`~200
`:c
`'e'
`#
`"
`
`100
`
`/.-·-..... ..... <111>Eu
`I'
`
`o-o~
`o o ..... <111>Tm
`
`950
`
`1100'C
`Tgrowth
`
`FIG. 2. Light output asa function of growth temperatureof[lll] and [110]
`Y AG films doped with Ce, Th, Eu, and Tm. The samples were excited by a
`stationary electron beam oflow energy density.
`
`seems to have a better coupling to the lattice than the other
`rare-earth ions. The highest measured energy efficiency is
`higher than that reported by Blasse and Brilto for Ce: Y AG
`powder (3.5%). It proved to be impossible to increase the
`cerium concentration above x = 0.03, because at higher con(cid:173)
`centrations..a second phase ofCe02 appeared in the melt and
`the incorporation of cerium into the layer did not increase
`further. A similar effect was observed by Gibbons et al., II
`who found that the maximum amount of cerium accepted in
`a garnet powder amounts to 2 mol. % when the powder is
`prepared under oxidizing conditions.
`In Fig. 2 we show the light output as a function of the
`growth temperature for Ce: Th: Eu:, and Tm:Y AG films. It
`is clear that the films grown on [Ill] substrates are more
`efficient than those grown on [110] substrates for the Ce and
`Th activator ions. In the case of [Ill] films the maxima may
`be explained by the fact that at low growth temperatures the
`increase in Pb ion concentration quenches the efficiency,
`while at the higher growth temperatures the activator ion
`concentration decreases. These effects are borne out by
`chemical analysis of the layers.
`The lower values of the [110] films shown in Fig. 2 are
`explained by concentration quenching. Analysis shows that
`the activator ion concentrations are much higher in these
`layers than for [111] layers. Thus the maximum in the light
`output versus growth temperature is shifted to higher values
`of growth temperature where less activator ion is incorporat(cid:173)
`ed. The melt for [110] growth was not further optimized
`because hillock growth occurs on [110] substrates, analo(cid:173)
`gous to the case of iron garnets explained in great detail by
`van Erk et al. 12
`In Fig. 3 we show the radiance as a function of the
`incident power density. At high powers the behavior de(cid:173)
`pends upon the activator ion. The Ce:YAG layers show an
`almost-linear behavior up to 108 W Icm 2
`, while the Eu:Y AG
`and the Th: Y AG show saturation effects around 104
`
`10'
`
`10'
`
`10'
`
`10]
`
`10
`
`"~0~~'OT2~~--~'rf'-~"--~6--~7--~'"
`10
`10
`10
`10
`---- nCldent Power DenSIty W/m 2
`
`FIG. 3. Light output as a function of incident power density. The 0 refer to
`measurements in the demountable apparatus and 0 refer to measurements
`in the electron microprobe.
`
`W Icm 2
`• The temperature of the phosphor at the maximum
`power density of 108 W 1m2 can be calculated 13 to be less
`than 50 .c. From the known quench temperatures of the
`phosphors, thermal quenching can be ruled out for Ce, Eu,
`and Th. Thulium, with a quench temperature of about 90 ·C,
`remains an intermediate case in this respect. A more detailed
`explanationofthese effects will be given by van der Weg. 14
`In conclusion, we have shown that it is possible to grow
`cathode luminescent single-crystal thin films by the method
`ofliquid phase epitaxy. These materials have excellent ther(cid:173)
`mal properties so that they can be used with electron beam of
`high power density without thermal quenching. In particu(cid:173)
`lar, the epitaxial layers of Ce:Y AG can be used up to power
`densities of 108 W 1m2 and are capable of emitting light with
`a radiance of over 105 W 1m2 Sr.
`The authors wish to thank J. P. H. Heynen and W. H.
`Smits for their help in the growth and measurement of the
`films. P. A. Paans, M. Klerk, and Y. Tamminga are thanked
`for their analytical work. W. F. van der Weg, T. J. A. Popma,
`and A. T. Vink are acknowledged for useful discussions dur(cid:173)
`ing this study.
`
`'M. W. van Tol and J. van Esdonk, IEEE Trans. Electron. Devices (to be
`published).
`'H. J. Levenstein, S. Licht, R. W. Landorf, and S. L. Blank, Appl. Phys.
`Lett. 19, 486 (1971).
`'J. M. Robertson, W. Tolksdorf, and H. D. Jonker, J. Cryst. Growth 27,
`241 (1974).
`4R. Ghez and E. A. Giess, J. Cryst. Growth 27, 221 (1974).
`'So L. Blank and J. W. Nielsen, J. Cryst. Growth 17, 302 (1972).
`"J. M. Robertson, M. J. G. van Hout, J. C. Verplanke, and J. C. Brice,
`Mater. Res. Bull. 9,555 (1974).
`'B. Andlauer, J. Schneider, and W. Tolksdorf, Phys. Rev. B 8 I, 1 (1973).
`HR. A. Buchanan, K. A. Wickersheim, J. L. Weaver, and E. E. Anderson, J.
`Appl. Phys. 39, 4342 (1968).
`9G. Blasse, J. Luminescence 1,2 766 (1970).
`lOG. Blasse and A. Bri1, J. Chern'. Phys. 47,5139 (1967).
`liE. F. Gibbons, T. Y. Tien, R. G. Delosh, P. J. Zacmanidis, and H. L.
`Stadler, J. E1ectrochem. Soc. 120, 835 (1973).
`I2W. van Erk, H. J. G. J. van Hoek-Martens, and G. Bartels, J. Cryst.
`Growth (to be pubslished).
`"L. G. Pittaway, Brit. J. Appl. Phys. 15 (1964) 967.
`I·W. F. van der Weg and M. W. van Tol (unpublished).
`
`472
`
`Appl. Phys. Lett., Vol. 37, No.5, 1 September 1980
`
`J. M. Robertson and M. W. van Tol
`
`472
`
` Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. Download to IP: 198.65.204.101 On: Wed, 23 Nov
`2016 16:41:12
`
`VIZIO 1015