D.J. Benford 1. California Institute of Technology. Mail Code , Pasadena, CA M.C. Gaidis. Mail Stop , Pasadena, CA 91109

Transcription

1 TRANSMISSION PROPERTIES OF ZITEX IN THE INFRARED TO SUBMILLIMETER D.J. Benford California Institute of Technology Mail Code 32-47, Pasadena, CA 925 M.C. Gaidis NASA-Jet Propulsion Laboratory Mail Stop 68-34, Pasadena, CA 99 J.W. Kooi California Institute of Technology Mail Code 32-47, Pasadena, CA 925 contact: ABSTRACT The results of measurements of the refractive index and power absorption coeæcient of Zitex at 29K and 77K in the spectral region from to microns are presented. Zitex is a porous Teæon sheet with a ælling factor of ç 5è, and is manufactured in several varieties as a ælter paper. Zitex is found to be an eæective IR block, with thin è2çmè sheets transmitting less than è in the -5çm range while absorbing çéè at wavelengths longer than 2çm. Some variation in the cutoæ wavelength is seen, tending to be a shorter wavelength cutoæ for a smaller pore size. Additionally, the thermal conductivity of Zitex at cryogenic temperatures has been measured, and is found to be roughly one-half that of bulk Teæon.. INTRODUCTION To reduce the loading on cold optical elements operating in the far infrared, room temperature radiation must be blocked eæciently while allowing the desired wavelengths to pass unattenuated ë; 2ë. Commonly used materials include blackened polyethylene and Quartz. Unfortunately, the transmittance of black polyethylene is dependent on the size, concentration, and form of the carbon used to blacken it, and varies substantially in its far-infrared properties ë3ë. Quartz is a low-loss material when a suitable antireæection coating like Teæon is applied, but this is difæcult and restricts the wavelengths over which it can be used as a highly transmissive element ë4ë.teæon itself is a good IR block, but transmits power in the 5-çm range and longward of 5çm, limiting its usefulness. Several more absorbing materials, such as Fluorogold and Fluorosint, have been used for low frequency applications, but their slow spectral cut-oæ characteristics are not ideal for receivers operating near THz ë4; 5ë. Zitex is a sintered Teæon material with voids of -6çm and a ælling factor of 5è. Several diæerent varieties are Present address: Code 685, NASA-Goddard Space Flight Center, Greenbelt, MD 277

2 available, divided into two categories by manufacturing process. Zitex A ë6ë is designed to reproduce ælter paper, and so has many narrow linear paths through it and is a rough but soft sheet. It is available in grades with eæective pore sizes ranging from 3çm to45çm and in thicknesses from.3mm è.5"è to 4mm è.25"è. Zitex G is made of sintered Teæon spheres of small sizes, resulting in a denser, smoother material. Available in 5 grades, the pore sizes range from.5çm to 5.5çm and is available in standard thicknesses of.mm è.4"è to.38mm è.5"è, although larger thicknesses are available. Zitex is similar in geometry to glass bead ælters, in which dielectric spheres are embedded into a suspending material with a diæerent index of refraction. A single sphere of radius a will scatter strongly for wavelengths ç ç éçaèn,èë7ë. Thus, for Teæon èn =:44 ë8ë è, a sphere of radius çm produces a shadow for wavelengths shortward of 5çm. At short wavelengths, then, a perfectly scattering screen will redistribute the optical power in an incident beam equally in all directions, resulting in a large loss for wellcollimated beams. 2. MEASUREMENTS Because of the large wavelength range involved, three FTS instruments were used to characterize Zitex. For the nearto mid-infrared è - 8 çm; -25 cm, è a commercially available Nicolet 6SX spectrometer ë9ë was used. The farinfrared è5-2 çm; 2-5 cm, è measurements were made on a Bruker interferometer at JPL. The submillimeter data è2- çm; 5- cm, è was obtained on an FTS at Caltech ëë. The focal ratio of the spectrometers was roughly f=4. A perfectly scattering surface would yield a transmission of roughly :4è in this case. Table lists the samples we measured. 3. TEFLON In order that we might characterize qualitatively the diæerence between bulk Teæon and Zitex sheets, we measured one thin è5mmè and one thick è.75mmè sample of plane-parallel Teæon sheet. Figure shows the results of a measurement of the far-infrared transmission of the thick sample near the cut-on region at 5-çm. The sample was measured at room temperature and liquid nitrogen temperature, showing a slight improvement in the transmission when the sample is cold. Figure 2 shows the mid-infrared transmission of the thinner sample, which highlights the fairly narrow regions near - 2çm where the absorption is large. 4. ZITEX G4, G6, G8, G, G5, A55 The samples of G4 and G6 were measured in the near- to mid-infrared to derive a transmission and an eæective absorption coeæcient, as shown in ægure 3. The absorption coeæcient æ for a sheet of thickness h is calculated from the transmission T as,lnèt è=h. Since some wavelength-dependent fraction of the loss è, T è is from scattering and some from absorption, the absorption coeæcient cannot be used to estimate the transmission of arbitrary thicknesses. It does, however, provide a useful means of comparison with other, more purely absorptive, materials. Combining sets of data in the nearthrough far-infrared for samples of G8 and G allows us to build a more complete picture of the proæle of the cut-on

3 Table. Zitex samples measured. Grade Pore Size Thickness Filling NIR FIR Submm èçmè èmmè Factor æ data data data G X G X G X X G -2 5 X X G5-2 X G25 ç ç.5 X X A X æ Relative density of Teæon..2 Teflon Warm Teflon Cold - Fig.. Far-Infrared transmission of the thick slab of Teæon at 3K and 77K... - Fig. 2. Mid-Infrared transmission of a 5mm thick Teæon sheet, highlighting the regions of good absorption.

4 .2 G4 Absorption G6 Absorption G4 G6 Absorption Coefficient (cm - ) - Fig. 3. and absorption coeæcient in nepersècm of Zitex G4 and G6. of Zitex near çm, as shown in ægure 4. Measurements of G5 and A55 are shown in ægure 5. A marked shortening of the cut-on wavelength can be seen in the A55 sample, presumably as a result of its diæerent structure. 5. ZITEX G25 As the thickest of all the samples, the G25 sheet of Zitex was used for the longest wavelengths, covering 4 and 6 GHz è88 and 75çmè. Even with a 3.5mm thick slab, the loss was small enough to be below detectability at longer wavelengths. The sample was cooled to 2K in order to determine its suitability as a mid-infrared blocking ælter for helium-cooled cryostats. The transmission and eæective absorption coeæcient are shown in ægure 6. The 4 GHz absorption feature is known as an absorption band seen in cold Teæon ë8ë. Combining the data on G25 near THz with data at shorter wavelengths allows us to determine a broadband transmission and absorption coeæcient, as shown in ægure 7, for the range 2-çm è3 GHz-5 THzè. The effective absorption coeæcient aswell æt by æ = 23 expëèç çm=37è :77 ë nepersècm. The transmission, neglecting the absorption band, follows T = expè,7ç,:8 çm è. 6. MULTIPLE LAYERS A helium-cooled receiver is likely to have several layers of infrared-blocking æltration in the optical path. As a result, it is natural in the case of a scattering material like Zitex to question its efæcacy in a multilayer application. Layering single-, double-, and triple-ply sheets of Zitex in close proximity èlimited only by the natural wavy contours of the thin sheetsè yields the transmission measurements shown in ægure 8. Because the transmission drops more slowly that for

5 .2 G8 G Absorption Coefficient G8 Absorption G Absorption - Fig. 4. and absorption coeæcient in nepersècm of Zitex G8 and G using near-, mid-, and far-infrared data..2 G5 A55 G5 Absorption A55 Absorption Absorption Coefficient - Fig. 5. of Zitex G5 & A55 in the near- to mid-infrared.

6 .2. Absorption Coefficient (cm - ) Abs. Coeff. α~5ν Frequency (GHz) Fig. 6. and absorption coeæcient in nepersècm of Zitex G25 between 4 and 6 GHz è88 and 75çmè..2 Absorption Coefficient. Absorption Coefficient -. Fig. 7. and absorption of Zitex G25.

7 2% 5% % Layer 2 Layers 3 Layers 5% % -5% -% Fig. 8. of single, double, and triple layers of Zitex in close proximity. The transmission drops more slowly than for a pure absorbing medium, implying strong scattering. a pure absorbing medium èi.e. T 3 = T 3 and T 2 = T 2 è in the mid-infrared, we can infer than scattering is the dominant loss mechanism and that multiple sheets are not substantially more eæective than single sheets. If, however, we separate the layers slightly and look at longer wavelengths, the picture changes. Using Zitex A55 sheets spaced by roughly 5mm, we ænd ænd the transmission shown in ægure 9. At mid-infrared wavelengths, the Zitex still appears to be dominated by scattering, whereas at far-infrared wavelengths, the transmission appears to be increasingly determined by absorption èpresumably in the bulk of the Teæonè. 7. TEMPERATURE VARIATION In the case of many materials èe.g., quartzè, the absorption of mid-infrared radiation is known to vary as the temperature changes ëë. To determine if there was any temperature eæect in the transmission of Zitex, we measured the transmission of samples of G at 3K and 77K in the far-infrared near the cut-on region. No signiæcant variation in the transmission was seen upon cooling èægure è, which is to be expected for dielectric scattering. 8. REFRACTIVE INDEX Measuring the refractive index of a material of low dielectric constant is diæcult near THz for a nondispersive FTS. Only the thick sample of G25 could be measured, via a determination of the fringe spacing in the 3.5mm thick slab. The fringe spacing, averaged between 3 and 45 cm, è2-8 çm or 4-35 GHzè, was.8 cm,. This yields a refractive index for Zitex of is n = :2 æ :7 at a temperature of 2K. This

8 .2 One Layer Absorption Two Layers Absorption Two Layers One Layer Absorption Coefficient (cm - ) - Fig. 9. of single and double layers of Zitex in an f=4 beam with cm spacing. At mid-ir wavelengths, the Zitex acts as two scattering surfaces; at far-ir wavelengths, like an absorber..2. Warm Cold. - Fig.. Far-Infrared transmission of Zitex at 3K and 77K. No substantial variation is seen.

9 can be compared to Teæon, which has n =:44 ë8ë ; with a ælling factor of ç 5è, the expected refractive index is n =:22, exactly as measured. 9. THERMAL CONDUCTIVITY The thermal conductance of a thick slab of Zitex was measured in the direction along the sheet using an apparatus developed for the purpose ë2ë. At cryogenic temperatures èt ç 5Kè, the conductivity of Zitex is found to be well æt by KèT è=: T :58 WK, m,. This value is half the bulk conductivity of Teæon, indicating that the porous nature of Zitex does not substantially affect its thermal conductance beyond the geometric reduction. However, since Zitex sheets tend to be thin èç.3mmè, even the small power incident on Zitex when used as a near-infrared blocking ælter will raise its temperature by a signiæcant fraction. This suggests the use of two layers for good blocking, whereas one Teæon layer might have been suæcient to handle the optical loading. However, since the loss in Zitex at long wavelengths is so low, this solution is likely to be more eæcient than the use of Teæon.. CONCLUSIONS We have measured the refractive index and power absorption coeæcient of Zitex at 29K and 77K in the spectral region from 2 to 2 microns. Its absorption at wavelengths longer than this is found to be extremely small, and with an index of refraction of n =:2, there is very little reæection loss. Zitex is an eæective IR block when used at wavelengths longward of 2çm, having lower absorptionèreæection losses than black polyethylene or quartz and better IR blocking characteristics than Teæon. REFERENCES ëë Hunter, T.R., Benford, D.J. & Serabyn, E. 996, PASP, 8, 42 ë2ë Kooi, J.W., Chan, M., Bumble, B., LeDuc, H.G., Schaæer, P.L. & Phillips, T.G. 995, Infrared & Millimeter Waves, v.6, n.2, 249 ë3ë Blea, J.M., Parks, W.F., Ade, P.A.R. & Bell, R.J. 97, J. Opt. Soc. America, v.6, n.5, 63 ë4ë Benford, D.J., Kooi, J.W. & Serabyn, E. 998, Proc. Ninth Int. Symp. on Space Terahertz Technology, p. 45 ë5ë Lamb, J.W. 996, Int. J. IR MM Waves, v.7, n.2, pp.997í234 ë6ë Norton Performance Plastics, Wayne, New Jersey. è2è ë7ë Sato, S., Hayakawa, S., Matsumoto, T., Matsuo, H., Murakami, H., Sakai, K., Lange, A.E. & Richards, P.L. 989, Applied Optics, v.28, n.2, p ë8ë Kawamura, J., Paine, S. & Papa, D.C. 996, Proc. Seventh Int. Symp. on Space Terahertz Technology, p. 349 ë9ë Nicolet 6SX spectrometer, Nicolet Instruments, 5225 Verona Rd., Madison, WI 537; è8è ëë Bin, M., Gaidis, M.C., Benford, D.J., Bíuttgenbach, T.H., Zmuidzinas, J. & Phillips, T.G., 999, Int. J. IR MM Waves, v.2, n.3, in press. ëë Brçehat, F. & Wyncke, B. 997 Int. J. IR MM Waves, v.8, n.9, pp.663í679 ë2ë Benford, D.J., Powers, T.J. & Moseley, S.H. 999, Cryogenics, in press