Full parametric study of multilayer coating for thermal applications

Paper number

IAC-08.C2.8.3

Author

Dr. Jose Manuel Llorens Montolio, European Space Agency (ESA), The Netherlands

Year

2008

Abstract

The fabrication and design of partially coherent thermal sources has been object of studies in the last years and is of great interest to spacecraft thermal control [ greffet2002cel]. The temporal coherence of a radiation source is tightly related with the spectral width of its emission, and the spatial coherence with the directionality. These properties are exploited in order to tailor to some extent the spectral and spatial distribution of the thermal radiation of a given substrate by controlling its coherence.

The most common technique employed in order to obtain structures which exhibit such properties is writing a grating on the surface of polar or metallic materials [ greffet2002cel, kreiter1999tie, richter1993pct]. Additionally, such behaviour has recently been predicted and observed also in thin film structures arranged in multilayer stacks [ benabdallah2004tab, narayanaswamy2004tec]. In the present work we restrict ourselves to this type of structures given the long tradition of applicability as surface finishes for thermal control in spacecraft engineering. However, their benefits for space applications are beyond thermal control, since they play an important role in partially coherent near-infrared sources and in the field of thermophotovoltaics.

The main objective of this study is to determine which arrangement of thin films of different materials results in a coating exhibiting an emission characterized by a given (i) bandwidth and (ii) angular dependence of the radiation wavelength. We have considered a spectral range from 1 µm to 15 µm and a bandwidth range from 0.01 µm to 0.5 µm. For each combination of wavelength and bandwidth we have studied the angular dispersion.

In order to cope with the design of the multilayer structure we have developed a reverse engineering model. The optical properties of the structure (reflection, transmission and emission) are calculated by means of the transfer matrix formalism [ macleod2001tfo]. The inverse problem is solved by using a genetic algorithm [ holland1992ana].

It is important to stress the extensive numerical calculation required in the optimization process. In order to drastically reduce the required computing time we have integrated the model inside of the platform ACT-DC, a general purpose distributed computing environment (seti@home like) developed internally at ESA’s Advanced Concepts Team [ ACTDC, INF-con-2].

References

[greffet2002cel]

J. Greffet, R. Carminati, K. Joulain, J. Mulet, S. Mainguy, and Y. Chen, Nature 416, 61 (2002).

[kreiter1999tie]

M. Kreiter, J. Oster, R. Sambles, S. Herminghaus, S. Mittler-Neher, and W. Knoll, Optics Communications 168, 117 (1999).

[richter1993pct]

K. Richter, C. Chen, and C. Tien, Optical Engineering 32, 1897 (1993).

[benabdallah2004tab]

P. Ben-Abdallah, Journal of the Optical Society of America A 21, 1368 (2004).

[narayanaswamy2004tec]

A. Narayanaswamy and G. Chen, Physical Review B 70, 125101 (2004).

[macleod2001tfo]

H. Macleod, Thin Film Optical Filters (Institute of Physics Publishing, 2001).

[holland1992ana]

J. Holland, Adaptation in natural and artificial systems (MIT Press Cambridge, MA, USA, 1992).

[ACTDC]

Act-dc homepage. http://www.esa.int/gsp/act/inf/pp/act-dc.htm.

[INF-con-2]

D. Izzo and M. C. Markót (Proceedings of the GO 05 - 4th International Workshop on Global Optimization, 2005), .

Abstract document

IAC-08.C2.8.3.pdf

Manuscript document

(absent)