Volume 6/Number 2
`
`of the Illuminating Engineering Society
`
`Lighting in the logging and sawmill industries
`Technical Committee Report
`
`Ultraviolet radiation-considerations In interior lighting design
`R. E. Levin, G. R. Spears, G. W. Clark and E. D. Bickford
`
`Improved color rendition In high pressure mercury vapor lamps
`Mary V. Hoffman
`
`The design of safety-colors
`W. A. Thornton
`
`Tungsten-halogen lamp dosed with tin tetralodide
`A. D. Kulkarni, J. Martin and H. G. Sell
`
`An automated high-speed photometer
`L. C. Snyder and P. E. Westlake
`
`Recommended practice for the specification of an ESI rating in
`interior spaces when specific task locations are unknown
`Technical Committee Report
`
`67
`
`80
`
`89
`
`92
`
`100
`
`105
`
`111
`
`Discussion of previously published papers
`
`124
`
`all:ili~9d quarterly in the United States of America by the Illuminating Engineering Society of North America, 345
`47th St., New York, N.Y. 10017. Copyright, Illuminating Engineering Society of North America, 1977. Sec(cid:173)
`~J-clal>s postage paid at New York, N.Y. and additional mailing offices. This publication is indexed regularly
`Engineering Index. Inc., and is available on microfilm from University Microfilm, Ann Arbor, Mich. 48106.
`•~Jscriotk)l'l rates: $25.00 for fOU' annual issues, plus extra postage to all countries in which second-class postage
`do not apply. Single copies: $2.50 to members; $6.25 to nonmembers. Editorial Office, 345 East 47th St.,
`York, N.Y. 10017. Printing Office, Easton, Pa.
`
`Clifford L. Forbes Jr. Editor
`
`Jack F. Christensen Associate Editor
`
`John E. Kaufman Technical Director
`
`Howard Haynes Engineering Assistant
`c. L. Crouch Director of Research IERI
`
`VIZIO 1008
`
`

`
`Improved color rendition
`in high ·pressure mercury vapor lamps
`
`Mary V. Hoffman
`
`The addition of Y 3AI50 12:Ce to the group of phosphors suitable for color
`correcting the mercury discharge lamp provides for improving the color
`rendering Index without a lumen loss. By altering the cerium content in the
`formulation, the absorption and brightness can be adjusted to its optimum
`value In the lamp.
`
`The ideal in color rendition of a high pressure mer(cid:173)
`cury discharge lamp has been the subject of several
`recent papers, in which simulated spectral distribu(cid:173)
`tions have been created by the addition or subtrac(cid:173)
`tion of spectral energy.1•2 This can be ddne to obtain
`lamps of selected color temperatures, with various
`combinations of real and simulated phosphor emis(cid:173)
`sions. The optimization of the luminous efficiency
`can be calculated in the same manner. One study has
`shown that the addition of emission at 620 nanome(cid:173)
`ters to the Hg discharge is near the optimum for color
`rendition, but that for lamps with color temperature
`of 4000 to 3000 kelvins, it is necessary to remove some
`of the Hg discharge in the blue. For luminosity, ad(cid:173)
`dition of emission near 590 nanometers is need(cid:173)
`ed.2,3
`Among the existing phosphors meeting the tem(cid:173)
`perature and stability requirements of the mercury
`discharge lamps, only Y V04:Eu and Y(VP)04:Eu
`phosphors (YVP) supply emission at the necessary
`spectral region, with the main peak at 618 nanome(cid:173)
`ters. They do not provide absorption of the blue Hg
`lines in sufficient &mount or emission in the high
`luminous region. Filtering of the blue can be obtained
`by the addition of a pigment or of a phosphor which
`also acts as a pigment. The phosphor magnesium
`fluogermanate activated with manganese (MG) is the
`only suitable red emit~r with appreciable absorption
`of the blue lines, but since the emission is at 650 to
`660 nanometers, the luminosity is not enhanced by
`A paper presented at the Ann.ual IES Conference, August 29
`through September 2, 1976, Cleveland, Ohio. AUTHOR: General
`Electric Company, Cleveland, Ohio.
`
`JOURNAL OF JES I JANUARY 1977
`
`this substitution. Combinations of these two phos(cid:173)
`phors can be used, as described by Rokosz, et al. 4
`We have found another phosphor, cerium-acti(cid:173)
`vated yttrium aluminate garnet (YAG:Ce) which is
`useful in improving color rendition by absorbing the
`blue Hg radiation and also adds to the total emission
`of the lamp by converting this blue radiation into
`emission centered at 560 nanometers. It can be
`combined with the Y(VP)04:Eu emission, effectively
`changing the color of the lamp.
`Lamps prepared with blends of YAG:Ce and YVP
`phosphors show the shift in color points from below
`the black body to, on, or above the locus, depending
`on the proportions used. Typical color points are
`shown in Fig. 1 for a warm color corrected lamp. The
`
`Figure 1. Color shifts with increasing YAG:Ce lri Y(VP)
`0 4:Eu.
`
`X
`
`89
`
`VIZIO 1008
`
`

`
`color shift is due largely to the absorption of the 436
`nanometer Hg line, which is filtered by about 40
`percent with 15 percent Y AG in the blend. The lu(cid:173)
`minosity of the lamps remains about the same for
`four lamp colors shown. This is due to the efficient
`conversion of the 436 Hg line into visible emission.
`These data are summarized in Table I.
`The actual color points obtained can be varied by
`the amount of phosphor used on the lamp as well as
`the proportion of YAG:Ce in the blend. It can also be
`
`varied by using the blend over a pre-coat, as de(cid:173)
`scribed by Rokosz, et al. 4 The pre-coat acts as a re(cid:173)
`flector for both ultraviolet and the blue radiation,
`allowing better utilization of the Hg radiation by the
`phosphors. With the Y AG:Ce present, the reflection
`of the blue radiation increases its effect both as a
`pigment and as a phosphor.
`The YAG:Ce phosphor, which was developed for
`its emission characteristics under cathode ray exci(cid:173)
`tation5·6 is strongly absorbing in the blue, as shown
`
`11.1
`0 z
`j!
`~
`...1
`LL.
`11.1 a::
`11.1
`({)
`:::>
`LL.
`
`LL. c
`
`>-
`t:
`.(/) z
`11.1
`.....
`~
`11.1 >
`i=
`<(
`...1
`11.1 a::
`
`300
`
`700
`
`WAVELENGTH, n m
`Figure 2. Diffuse Reflectance curve, (Y2.925 Ceo.o7sh
`Als012·
`
`440 450 460 470 480
`WAVELENGTH,n m
`Figure 3. Excitation curve: detected wavelength-511
`nanometers.
`
`330
`
`nm
`
`Figure 4. Emission curve: excitation wavelength-436
`nanometers.
`
`-at 25"C
`
`11.1
`
`t .(/) z
`~
`11.1 >
`ti ...1
`UJ a::
`
`Figure 5. Emission intensity vs cerium content, measured
`at 25° C and 300° C. Excitation wavelength-436 nm.
`,--(cid:173)
`l--ot 25"C
`I
`I
`I
`f,
`1/
`II
`If
`
`>-
`1-
`(/) z
`UJ
`1-z
`
`f
`
`r
`r
`t
`
`t
`p
`t
`t:
`n
`tl
`
`p
`p
`0
`ci
`sl
`A
`cc
`Ol
`
`R
`1.
`ch
`irr.
`ci<
`2.
`tri
`ch
`3.
`ret
`Ar.
`
`520 560 600 640 680 720
`WAVELENGTH, n m
`
`X
`
`90
`
`JOURNAL OF IES I JANUARY 1977
`
`JOl
`
`VIZIO 1008
`
`

`
`Table 1-lampdata:YAG:Ce with Y(VPIO.:Eu
`
`Percent 100-hour
`lm/w
`YAG:Ce
`
`0
`5
`10
`15
`
`51.5
`52.1
`50.7
`51.3
`
`X
`
`.408
`.411
`.420
`.426
`
`y
`
`.384
`.394
`. 403
`.411
`
`CRI
`
`45
`
`51
`
`by the diffuse reflectance curve (see Fig. 2), and is
`also excited by this radiation. This absorption is into
`the 5d state of the cerium, which in this crystal
`structure, lies in an unusual position. Cerium usually
`absorbs in the ultraviolet and emits in the near ul(cid:173)
`traviolet or in the blue. The Y AG:Ce phosphor is not
`excited by 254 nanometers and o:r:tlY weakly by 365
`nanometers from the Hg arc. The excitation spectra
`(See Fig. 3) shows the very strong dependence of-the
`emission intensity on the 436-nm Hg radiation.
`The emission of the phosphor is in a band peaking
`at about 540 nanometers at 25 ° C shifting to about
`560 nanometers at 300° C (See Fig. 4). This is at the
`maximum eye sensitivity and close to the calculated
`peak of 590 nm for good lumen output in the sys(cid:173)
`tem.
`As with most phosphors, the specific conditions of
`the lamp determine its applicability, and the com(cid:173)
`position can often be altered according to its use. In
`the YAG:Ce phosphor, the cerium concentration is
`the critical factor. Both the absorption at .436 na(cid:173)
`nometers and the emission intensity increase with
`the cerium concentration with the brightness
`reaching a maximum at about one- to two-atom
`percent and the absorption at about three-atom
`percent when measured at 25° C. At the temperature
`of operation of the lamp, the phosphor is less effi(cid:173)
`cient, and brightness measurements made at 300° C
`show that lower cerium concentrations are desirable.
`As shown in Fig. 5, a rather narrow range of cerium
`content must be maintained for the best conversion
`of 436 radiation to emission at 560 nm in the lamp.
`References
`1. J. J. Opstelten, D. Radielovic and W. L. Wanmaker, "The
`choice and evaluation of phosphors for application to lamps with
`improved color rendition," Journal of the Electrochemical So-.
`ciety, Vol. 120, No. 10, October 1973, p. 1400.
`2. T. F. Soules and M.A. Maier, "Optimized spectral power dis(cid:173)
`tributions for mercury vapor lamps," Journal of the Electro(cid:173)
`chemical Society, Vol. 121, No.3, March 1974, p. 407.
`3. H. Ivey, "Color and efficiency of fluorescent and fluoro(cid:173)
`rescent-mercury lamps," Journal of the Optical Society of
`America, Vol. 62, 1972, p. 814.
`
`4. F. Rokosz, J. W. Sausville, and J. Van Broekhoven, "Incan(cid:173)
`descent lamp color with high-intensity discharge lamps," JOUR·
`NAL OF THE ILLUMINATING ENGINEERING SOCIETY, Vol. 3, No.
`I, October 1973, p. 95.
`5. G. Blasse and A. Bril, "A new phosphor for flying-spot cath(cid:173)
`ode-ray tubes for color television: yellow-emitting Y aAl50 12:Ce,"
`Applied Physics Letters, VoL II, No.2, July 15, 1967, p. 53 .
`6. W. W. Holloway, Jr. and M. Kestigian, "Optical properties of
`cerium-activated garnet crystals," Journal of the Optical Society
`of America, Vol. 59, No. 1, January 1969, p. 60.
`
`DISCUSSION
`
`W. A. THORNTON:* The high pressure mercury vapor lamp is
`in need of all the kindly attention it can get. This welcome paper
`brings better lamp performance closer. It performs a service by
`bringing another useful phosphor into the very limited line-up of
`luminescent materials that can endure the 9J5erating conditions
`in this lamp. The new phosphor has the wieful property of con(cid:173)
`verting undesired blue-violet arc emission to desired green light,
`which tends to improve both color rendering and output. Does the
`increase in Color Rendering Index (CRI), as 15 percent of the new
`phosphor is added, represent a real visual improvement in color
`rendering of real objects and scenes? As to the proposed addition
`of yellow wavelengths near 590 nanoeters "for good lumen output
`in the lamp system," we believe that such addition would harm
`color rendering more than it would improve lumen output. The
`usefulness of the new YAG:Ce phosphor is, we believe, due to the
`fact (see Fig. 4) that it contributes green light and red light; that
`it also contributes.yellow light is a disadvantage. The Soules and
`Maier study (see Reference 2) makes clear that peak lumen output
`by use of yellow light is accompanied by color rendering index of
`-20. Reference 1 ascribes poor color rendering to yellow emitting
`phosphors. Has the author any prospect of an efficient phosphor
`emitting a narrow band near 490 nanometers, which, according
`to the same references, can help raise CRI further?
`
`AUTHOR: t The true test of improved color rendition always rests
`on a subjective visual evaluation, especially in a system such as this
`one, with about ten percent calculated change in CRI. No sys(cid:173)
`tematic evaluation of this lamp system has been made, but the
`general appearances have been favorable.
`The emission contribution of YAG:Ce (if present as the only
`phosphor) would definitely result in a lower CRI. The Soules and
`Maier paper points to a system that closely approximates a pure
`Y AG:Ce lamp. It has a CRI of 23. The usefulness of the YAG:Ce
`lies in its ability to absorb the 486-nm Hg line to the extent suffi.
`cient to improve the CRI, and supply enough emission to con(cid:173)
`tribute to the lumen level, at a relatively low weight percent.
`The Soules and Maier paper ascribes the CRI of -20 to a system
`containing a narrow band emission at 580 nanometers, when
`combined with bluer line emission, but shows that when Eu+3
`emission is present the CRI cannot fall that low.
`* Westinghouse Electric Corporation, Bloomfield, New Jersey.
`t The author wishes to acknowledge the help of Dr. T. F. Soules
`for numerous discussions and E. Homonnay for the testing in
`lamps.
`
`Information for authors
`
`The JoURNAL OF THE IES is published quarterly by the Illuminating Engi(cid:173)
`neering Society of North America. Manuscripts should be sent to the editor,
`345 East 47th Street, New York, N. Y. 10017. All manuscripts are subject
`to the review of the Papers CommiHee Of the IES.
`
`'1
`
`JOURNAL OF IES I JANUARY 1977
`
`91
`
`VIZIO 1008