1850 NOTIZEN A Superconducting Bolometer as a High Sensitivity Detector for Molecular Beams M. CAVALLINI, G. GALLINARO, and G . SCOLES Istituto di Fisica Sperimentale dell'Universitä, 16132 Genova, Italy Gruppo Nazionale di Struttura della Materia del C.N.R. (Z. Naturforsch. 2t a, 1850—1851 [1969] ; received 31 July 1969) The construction, operation and calibration of a supercon- ducting bolometer is reported. Operated as a molecular beam detector the bolometer has, for Argon, a maximum sensitivity of 7  10® molecules sec -1 corresponding to a N.E.P. of 3  10~ 13 Watt Hz-i. The use of a liquid He cooled Ge infrared detector as a high sensitivity bolometer detector for molecular beams has been recently reported 1 and its usefulness in molecular scattering experiments has been shown 2 - 3 . The present note is intended to report on the con- struction, operation and calibration of a superconduct- ing bolometer which, operated as a molecular beam detector, showed an order of magnitude improvement in signal to noise ratio compared with the previously used Ge bolometer. A superconducting bolometer has been reported, al- most ten years ago, by MARTIN and BLOOR 4 which showed a noise equivalent power (N.E.P.) of about 3-10 -12 Watt Hz~i (reflection coefficient a = 0.1) with a time constant of about 50 10 -3 sec. The sensitive element was an evaporated tin film maintained at a fixed temperature, to within 1 0 -5 C K, in its supercon- ducting transition (around 3.7 ~K). In this condition the film has a very large temperature coefficient of resistance that can be used to transduce a chopped power imput to a voltage output which may then be integrated with standard techniques. Up to the present time the superconducting bolo- meter has not been widely adopted in infrared spectro- scopy for two main reasons. The first is the difficulty of thermo-regulating the sensitive element within 10 -5 K, and the second is the extreme delicacy of its con- struction. Taking into account that the N.E.P. of Martin and Bloor's bolometer was reported to be limited by the electronics available at that time, we undertook the development of a superconducting bolometer with the aim of solving the mechanical ruggedness problem by relaxing the requirement that the film, when operating, should be homogeneous in temperature. Indeed the possibility of a superconducting thin film acting as a non isothermal bolometer is qualitatively quite obvious and has also been quantitatively discus- sed theoretically in the literature by FRANZEN 5 . A superconducting bolometer of the non isothermal kind has also been reported by V. A. KONOVODCHENKO et al. at the 11 th Low Temperature Conference (St. Andrews 1968). With respect to this point it should be noted, however, that it is not clear whether or not a non isothermal bolometer can be operated only in the way described by FRANZEN 5 . Indeed the transition curves obtained by us (see Fig. 1) were of the kind expected for a non isothermal element. Two types of operations are then possible: namely in the region a or in the re- gion b. Region b corresponds to the type describec by Franzen. If sufficient thermal stability is available one can operate in region a which, in our opinion, cor- responds to a different non isothermal kind of opera- R / R N A A 1.0 - —A- Fig. 2. Schematic experimental set-up. Fig. 1. Transition curve of a typical bolometer. Reprint requests to Dr. G. SCOLES, Istituto di Fisica Speri- mentale dell'Universitä di Genova, Viale Benedetto XV, 5, 1-16132 Genova, Italy. 1 M. CAVALLINI, G. GALLINARO, and G. SCOLES, Z. Natur- forsch. 22 a. 413 [1967]. 2 P. CANTINI, M. CAVALLINI. M . G . D O N D I . and G. SCOLES, Phys. Letters 27 A, 284 [1968]. 3 M. G . D O N D I , G. SCOLES. F. TORELLO. and H. PAULY, J. Chem. Phys. 51, 392 [1969], 4 D . H . M A R T I N and D. BLOOR. Cryogenics 1. 159 [1961]. 5 W. FRANZEN, J. Opt. Soc. Am. 53, 596 [1963].