Soft x-ray transmission of optical blocking filters for...

Soft x-ray transmission of optical blocking filters for the x-ray CCD camera onboard Astro-E2

S. Kitamoto *.a,b, T. Kohmura c,b, N. Yamamotob, H. Takanoa, H. Saitoa, K. Sugaa, H. Sekiguchia,

S. Chibaa, I. Okamotoa, K. Hayashidad, H. Katayamad, T. Enoguchid, Y. Nakashimad, T.

Shiroshoujid, Y. Tawarae, T. Furuzawae, T. Tanakae

a Department of Physics, Rikkyo University, 3-34-1, Nishi-Ikebukuro, Toshima-ku Tokyo, 171-8501, Japan;

b Research Center of the Advanced Measurement, Rikkyo University, 3-34-1, Nishi-Ikebukuro, Toshima-ku Tokyo,

171-8501, Japan; c Physics Department, Kogakuin University, 2665-1, Nakanocho, Hachioji, Tokyo, 192-0015, Japan;

d Department of Earth and Space Science, Graduate School of Science, Osaka University, 1-1, Machikaneyama,

Toyonaka, Osaka 560-0043, Japan; e Department of Physics, Graduate School of Science, Nagoya University, Furocho, Chikusa-ku, Nagoya,

464-8602, Japan

ABSTRACT

We measured optical and soft X-ray transmission of Optical Blocking Filters (OBFs) for Charge Coupled Device (CCD) cameras, which will be launched as focal plane detectors of X-ray telescopes onboard the Japanese 5th X-ray astronomical satellite, Astro-E 2. The filters were made from polyimide coated with Al. The X-ray absorption fine

structures (XAFSs) at the K edges of C, N, O and K and L edges of Al were measured. The depth of the absorption edge of O was deep, compared to the other elements of polyimide. This is evidence of the oxidation of Al. The optical transmission is roughly less than 10-6 except for a peak around the wave length of 550nm. Long term change of the soft X-ray transmission was measured. No significant change of the thickness of the oxidation layer was found during half

year. Keywords: X-ray, CCD, XAFS

1.INTRODUCTION

A Charge Coupled Device (CCD) is widely used as an X-ray detector [1,2,3,4] and has many wonderful advantages.

As well as its good imaging capability with a resolution of less than 10 µm, it has an almost ideal energy resolution of a Si detector if it is used as a photon detector. We are planning to install CCD cameras on the Japanese 5th X-ray astronomical satellite, Astro-E 2, which is now scheduled for launch in early 2005.

Since a CCD has high detection efficiency for optical and ultra violet light, we will use an Optical Blocking Filter (OBF) in front of a CCD chip. The experimental calibration of the soft X-ray transmission of them is very important.

Especially, the X-ray transmission properties around the absorption edges are complex [5,6], which is known as X-ray Absorption Fine Structure (XAFS). Unexpected high transmission of optical light compared to the design value has *Correspondence: [email protected]

SPIE USE, V. 3 5168-40 (p.1 of 10) / Color: No / Format: A4/ AF: A4 / Date: 2003-07-17 18:04:42

Please verify that (1) all pages are present, (2) all figures are acceptable, (3) all fonts and special characters are correct, and (4) all text and figures fit within themargin lines shown on this review document. Return to your MySPIE ToDo list and approve or disapprove this submission.

been reported [7,8]. There is a possibility that the oxidation of the Al makes filters transparent for optical light. Thus, the precise and quantitative confirmation of the oxidation is required. We measured the soft X-ray transmission of flight candidates of the OBFs, paying attention to XAFS and the oxidation of Al. Also we measured the optical transmission

and checked the consistency between them. If the Al had been oxidized, measurement of the long term stability of the performance of the filter is very important, because we do not know when the Al had been oxidized. Thus we measured the soft X-ray transmission and quantitatively estimated the oxidation at two times separating a half year.

2. THE OBFs FOR ASTRO-E 2

The Astro-E 2 satellite is a recovery satellite of Astro-E, which was launched on 10 Feb. 2000 but could not enter a satellite orbit. Thus, the Astro-E 2 satellite was proposed and was scheduled for launch in early 2005. Because of this short time scheduling, the design of Astro-E 2 is almost the same as that of Astro-E. The design of the CCD cameras is also similar. The properties of the CCD camera for Astro-E can be seen in various reports[9,10,11], and some

reports on the OBFs have also been published[8,12]. However, the measurement of soft X-ray transmission of the OBFs for Astro-E was not sufficient. For Astro-E 2, we had an opportunity to measure the details of the soft X-ray transmission using synchrotron radiation. The first result of the X-ray transmission measurement had been published elsewhere[13]

The OBFs, which were made by Luxel Co. LTD, are composed of a thin polyimide (C22H10O4N2) film sandwiched by Al. Four cameras will be installed on the Astro-E 2 satellite and we have six OBFs including spares. Three films were made. Two pieces were cut from each film and were attached to two custom frames. A picture of the OBF is shown in figure 1. The thickness measured by Luxel Co. LTD is summarized in table 1.

Figure 1: A picture of the OBF. The film is attached on a custom frame. The effective area of the filter is 34.5 mm square with

rounded corner.



Table 1: Thickness of the Materials of Three Films Film ID Al(nm) Polyimide(nm) Al(nm) 8913-2 41.4±5 97.8±10 81.2±5 8929-1 40.0±5 106.5±10 81.6±5 8929-2 40.0±5 103.1±10 81.6±5

Table 2: Measurement Setups

Energy Range Grating Incident angle Double Mirror System

80-150eV 300 l/mm 86.9 Ni 14 deg 150-450 eV 300 l/mm 86.9 Ni 5.5 deg or 6 deg* 450-700 eV 800 l/mm 86.9 Ni 4.0 deg 800-1800 eV 800 l/mm 88.1 no

* 5.5deg for June 2002, 6deg for Dec. 2002.

3. MEASUREMENT

3.1. Soft X-ray Transmission

The measurement was performed at Beam line 11A of the Photon Factory in the Institute of Material Structure Science,

High Energy Accelerator Research Organization (KEK-PF). In this beam line, a grazing incident monochrometer is

installed with varied-line-spacing plane gratings[14,15,16]. By changing the grating, this beam line can provide soft

X-rays from ~80 eV to 1900 eV covering the C-K, N-K, O-K and Al-K edges. At the first experiment performed in June

2002, three setups were used covering the energy range from 200eV to 1800eV. At the second experiment done in

Dec. 2002, as well as the three same setups, additional setup was used in order to cover down to 80eV range. The

setup conditions are summarized in table 2. Except for the highest energy range, we used a Ni double-mirror system

to eliminate the higher order light.

The measurement configuration is schematically shown in figure 2. The beam intensity emerging from the

monochrometer (after the Ni double-mirror system) was monitored by measuring the current of the photo-electrons

from an Au-coated W mesh, which was installed in front of the samples. The beam was restricted by a slit, resulting in

a beam size of 2 mm x 5 mm. The OBFs were installed on rotational axles and the OBFs could be put in or out of the

beam. A window-less photodiode (AXW-100 TS) was used as a detector and its current was measured. A

window-less Si(Li) detector cooled by LN2 was also used to check the energy spectrum of the beam.

The transmission of an OBF was calculated as a ratio of currents measured with a photodiode with the OBF and one without it after normalizing by the current of the photo-electrons from the Au-coated W mesh. The calibration of the absolute photon energy was conducted using a K-edge of a 4 µm thick pure Al for the 800-1800 eV setup. For the



other three setups, C-K edge energy of a 7 µm kapton film was used. These were adjusted to 1559 eV and 284 eV, respectively, with an accuracy of ~1 eV. Derived X-ray transmission from Dec. 2002 experiment is shown in figure 3, where the transmissions of six filters between 80eV and 1800eV are compiled. The X-ray Absorption Fine Structures

(XAFSs) around the K-edges of Al, O, N and C were clearly measured. The X-ray transmissions of the six filters are almost same within ±1%. For the comparison of the transmissions derived from June 2002 experiment and this experiment, we plotted the averaged transmission of six OBFs of the two experiments in figure 4. Again the difference between them is within ±1%, except for the region form 780 ev to 950 eV, where the data in Dec. 2002 are

slightly higher than that in June 2002. In the Dec. 2002 experiment, the contamination of the second order light in this region is not negligible. Thus the June 2002 data must be correct.

Figure 2: Schematic view of the experimental configuration. The top view is drawn.

Figure 3: Soft X-ray transmission of six OBFs derived from Dec. 2002 experiment. The transmissions of the six OBFs are the same

within ±1%.



Figure 4. The comparison between the data obtained in June 2002 and Dec. 2002. The average transmission of six OBFs obtained

in the two experiments are plotted.

3.2. Optical Transmission The expected optical transmission is less than 10-6 in the visible range. This large dynamic range makes measurement difficult. The setup of the optical transmission measurement is shown in figure 5. Optical light from a halogen lamp is introduced to a grating monochrometer. The measured wave length range is from 400nm to 900nm. Two kinds of filters were used to eliminate the higher order light, changing according to the measurement wave length. Three samples can be installed in a sample chamber. Usually, we installed two OBFs and one piece of a aluminized mylar.

The direct beam can be measured between the samples. Finally the transmitted (or direct) lights was measured by a photo multiplier.

First we measured the transmission of the aluminized mylar. The transmission was ranging from 10-4 to 10-3. Next we measured the ratio of the transmission between the aluminized mylar and the OBFs. In this case we made the

halogen lamp bright enough to be able to detect the significant signal. Then we calculated the absolute transmission of the OBFs. The resultant transmissions of the six OBFs are shown in figure 6. Two groups can be seen. One group with lower transmission is consisting of the OBFs made from R/N8913-2 film, and the other group with relatively high transmission contains those made from R/N8292-1 and R/N8292-2 films. This difference corresponds to the difference of the Al thickness of roughly 10 nm.

Figure 5: The setup for the optical transmission measurement.



Figure 6: Optical transmissions of six OBFs. Four OBFs were measured two times.

4. RESULTS AND DISCUSSION

4.1. Overall Structure of the soft X-ray transmission To analyze the overall structure of the soft X-ray transmission, we first fitted the transmission, masking the regions of

the complex XAFSs, by a two materials model composed of polyimide and Al using the available absorption coefficients[17]. In the energy region from 750eV to 950eV, the higher order light was contaminating and we removed this region. The thickness of the polyimide and the Al are determined for both the June and Dec. 2002 experiments and listed in table 3, where we assume the densities of Al and polyimide are 2.699 g cm-3 and 1.43 g cm-3. The

thickness of the Al is consistent with the Luxel values but the thickness of the polyimide is significantly thicker. Since the transmissions of the six OBFs are very similar, we calculated the average and the standard deviation for each experiment, respectively. The maximum of standard deviations is 0.56% and most of them are distributed between 0.2% and 0.5%. Thus, we assumed the error of 0.5% uniformly, and fitted the average transmission by the two materials model. The top panel of figure 7 shows the average transmission and the best fit model for the Dec. 2002

experiment. The residuals (data-model) are plotted in the middle panel of figure 7. The best fit thickness is listed in table 3.

The residuals, shown in the middle panel of figure 7, indicate the edge depth of O of the data is deeper than that of the model, whereas the depths of the C and N are shallower. This means the existence of extra oxygen in addition to

the contents of the polyimide. This may be due to the oxidation of Al, which has been pointed out in previous works [7,8, 13]. Therefore, we fitted the data by the three materials model composed of polyimide, Al and Al2O3, where the density of Al2O3 is assumed to be 3.97 g cm-3. The residuals are shown in the bottom panel in figure 7. Due to the extended X-ray absorption fine structures, the residuals are still wavy but the overall edge depths become much refined. The best fit thickness is listed in table 3. Although the thickness of Al is consistent with the Luxel value, the obtained

thickness of the polyimide is still thicker than the Luxel values.



Figure 7: Average soft X-ray transmission obtained in Dec, 2002 experiment. The top panel shows the data and the best fit curve of

the two materials model composed of polyimide and Al. The middle panel shows the residuals (data-model) of the data from the two

materials model. The bottom panel shows the residuals from the three materials model composed of polyimide, Al and Al2O3.

Table 3: Best Fit Model Parameters

Polyimide (nm) Al (nm) Al2O3 (nm)

OBF ID June 2002 Dec. 2002 June 2002 Dec. 2002 June 2002 Dec. 2002 8913-2 137.1 143.6 125.1 124.0

8913-2-2 136.7 142.4 125.7 125.5 8929-1 142.1 144.8 123.1 121.4 8929-1-2 136.7 141.8 122.2 119.9 8929-2 135.5 141.4 125.5 120.9

8929-2-2 135.0 141.2 124.9 122.9 Average 136.1±0.4* 142.5±0.5 124.7±0.2 122.5±0.3 128.4±0.5 134.2±0.5 116.2±0.4 113.1±0.4 10.9±0.4 11.8±0.4 * Errors are only from statistics.



If we compare the results of two experiments, the Al thickness becomes slightly thinner, whereas the polyimide thickness become thicker with roughly 6~7 nm. The difference of the Al2O3 thickness is 0.9 nm and is less than three sigma. The differences of the Al and polyimide thickness might be due to the systematic errors, because the

difference of the actual transmission below C-K edge is less than 1 %. Otherwise, for example due to oxidation or contamination, the low energy part must be seriously affected. Thus the systematic error of our experiment is roughly 6~7 nm.

4.2. XAFS XAFS of C, N, O and Al are measured and shown in figure 8 for both the June and Dec. experiments. The structures of six filters are very similar. The wavy transmission below 250 eV, as seen in figure 3, is also due to the extended X-ray absorption fine structure (EXAFS) of the Al L-edge. The detail of XAFS analysis is beyond of this work. Since the structures are the same within ±1 %, and considering the energy resolution of the CCD of ~50 eV at the O-K

edge [6,11], the average of the six OBFs provides a sufficient transmission model of the OBFs, including the complex XAFSs. The difference of the XAFS between the June 2002 and Dec, 2002 is not significant. A small difference in the near edge structure of Al K edge can be seen. Thus might be due to the difference of the resolution of the monochrometer of the beam line.



Figure 8: XAFS of (a) Al, (b) O, (c) N and (d) C. Both the results from June and Dec. 2002 experiments were plotted.

4.3 Optical transmission The difference of the optical transmission of the two kinds of films are explained by the ~10nm difference of the Al

thickness. The X-ray transmission also indicates that Al of the 8913 film is thicker than that of 8929 film, although the value is ~5 nm difference. This discrepancy may be also due to systematic error.

5. CONCLUSION

We measured the soft X-ray transmission of the optical blocking filter for the CCD cameras onboard Astro-E 2. Six filters were almost the same transmission within ±1% in the energy range from 80 eV to 1800 eV. Thus, the average data provided a good model transmission including XAFS. Small gaps of the measurement could be interpolated and the energy region above 1800 eV could be extrapolated using the best fit model. We found the

existence of the extra O by the comparison of the depth of the absorption edges, comparing the two materials model composed of polyimide and Al. This discrepancy can be reconciled by introducing the Al2O3. The change of the thickness of the Al2O3 was not significant and less than 3 sigma during the half year. The optical transmission of the



six OBFs were measured. All the OBFs have the transmission less than 4×10-6 in the energy range from 400nm to 900 nm. The optical transmission is depending on the film lot. The difference of the optical transmission corresponds to the difference of the Al thickness of about 10 nm, whereas the results of the soft X-ray transmission indicate 5 nm

difference.

6. ACKNOWLEDGMENTS Three of the authors (S.K. K.H. and T.K.) gratefully acknowledge the financial support of the Grant-in-Aid for Scientific Research (Grant No. 14654039 and No. 15037208) and that of the Japan Society for the Promotion of

Science for Young Scientists. The present work has been performed under the approval of the Photon Factory Program Advisory Committee (PF-PAC No. 2002G038).

REFERENCES 1. B. Burke, J. Gregory, M. Bautz, G. Prigozhin, S. Kissel, B. Kosicki, A. Loomis and D. Young, IEEE Trans.

Electron. Devices, 44, 10 (1997). 2. M. W. Bautz, M. J. Pivovaroff, S. E. Kissel, G. Y. Progozhin, T. Isobe, S. E. Jones, R. Thornagel, S. Kraft, F. Scholze and G. Ulm, Proc. SPIE, 4012, 53 (2000). 3. L. Strueder, et al. A¥&A, 365, L18 (2001). 4. M. J. L. Turner, et al. A¥&A, 365, L27 (2001).

5. K. Mori, M. Shouho, H. Katayama, S. Kitamoto, H. Tsunemi, K. Hayashida, E. Miyata, M, Ohta, T. Kohmura, K. Koyama, M. W. Bautz, R. Foster and S. Kissel, NIM A, A459, 191 (2001). 6. M. W. Bautz and J. A. Nousek, 'Science Instruments Calibration Report for the AXAF CCD Imaging Spectrometer(ACIS)' (1999).

7. C. M. Castelli, D. J. Watson, A. Wells, B. J. Kent, M. Barbera, A. Collira and M. Bavdaz, Proc. SPIE 3114, 384 (1997). 8. T. Kohmura, K. Katayama, R. Asakura, S. Kitamoto, H. Tsunemi, K. Hayashida, E. Miyata, K. Hashimotodani, H. Katayama, M. Shouho, K. Koyama, S. Kissel, G. Ricker, M. Bautz, R. Foster and XIS-Team, Adv. Space Res. 25, 877 (2000).

9. K. Katayama, T. Kohmura, H. Katayama, M. Shouho, H. Tsunemi, S. Kitamoto, K. Hayashida, E. Miyata, K. Hashimotodani, K. Koyama, G. Ricker, M. Bautz, R. Foster and Astro-E Team, Adv. Space Res. 25, 881 (2000). 10. M. Shouho, K. Katayama, H. Katayama, T. Kohmura, H. Tsunemi, S. Kitamoto, K. Hayashida, E. Miyata, K. Hashimotodani, K. Yoshita, K. Koyama, G. Ricker, M. W. Bautz, R. Foster and S. Kissel, NIM A, 436, 85 (1999).

11. K. Hayashida, S. Kitamoto, E. Miyata, H. Tsunemi, K. Hashimotodani, K. Katayama, T. Kohmura, R. Asakura, K. Yoshita, H. Katayama, M. Shouho, K. Koyama, T. G. Tsuru, H. Awaki, T. Dotani, M. Ozaki, G. Ricker, J. Doty, M. W. Bautz, S. Kissel, and R. Foster, Proc. SPIE, 3445, 278 (1998). 12. H. Katayama, M. Shouho, T. Kohmura, K. Katayama, K. Yoshita, H. Tsunemi, S. Kitamoto, K. Hayashida, E. Miyata, K. Hashimotodani, K. Koyama, G. Ricker, M. W. Bautz, R. Foster and S. Kissel, NIM A, 436, 74 (1999).

13. S.Kitamoto et al. 2003 NIMA, in press 14. K. Amemiya, Y. Kitajima, T. Ohta and K. Ito, J. Synchrotron Radiat. 3, 282 (1996). 15. K. Amemiya, Y. Kitajima, Y. Yonamoto, T. Ohta, K. Ito, K. Sano, T. Nagano, M. Koeda, H. Sasai, and Y. Harada, Proc. SPIE, 3150, 171 (1997).

16. Y. Kitajima, K. Amemiya, Y. Yonamoto, T. Ohta, T. Kikuchi, T. Kosuge, A. Toyoshima and K. Ito, J. Synchrotron Radiat. 5, 729 (1998). 17. B. L. Henke, E. M. Gullikson and J.C. Davis, Atomic Data and Nuclear Data Tables, 54, 181-342 (1993).



Soft x-ray transmission of optical blocking filters for...

Documents

Transcript of Soft x-ray transmission of optical blocking filters for...