The effect of Helium on vacuum photodetectors

R M Brown Nov 1999

The effect of Helium on vacuum photodetectors
According to data gleaned from the catalogues of manufacturers, He ingress through the glass envelope of a phototube causes:
- an increase in noise associated with after-pulsing produced by positive ions

- electrical breakdown when the He partial pressure reaches 10^-2 Torr (1.3 Pa).

Since He is chemically inert, poisoning of the photocathode (and dynode) surfaces is presumably not a problem.
In the Philips catalogue it is claimed that the permeability of fused silica (‘quartz’) is 10x greater than that of borosilicate glass which in turn is 100x greater than that of soda lime glass. However, while reference 1 (thanks to PSF for uncovering this), confirms the ratio of 100 between borosilicate and soda lime, it indicates that the ratio between fused silica and borosilicate is also 100 (not 10). For borosilicate glass at 20^O C, the value of the permeation velocity¹ is:
D = 5.7x10^-16 (cm³ at NTP)/s/cm²/(Pa/mm thickness)
The rise in pressure within a phototube is proportional to the surface area of the envelope and inversely proportional to the volume enclosed. Therefore an aspect ratio may be defined:
A = S/V
For a sphere: A = 3/r. For our VPTs: A ≈ 2 cm^-1.
After a time, t, the He pressure , P_I, inside a tube would be:
P_I = P_E D A T/273 10⁵ t/d
where P_E is the external partial pressure of He, T is the absolute temperature, and d is the thickness of the envelope in mm.
The Philips catalogue states that in the presence of He at the natural partial pressure in the atmosphere (0.7 Pa), one may expect ‘an average life of some ten years’. The partial pressure of He within a tube with an enevelope thickness of 1mm would then be:
P_I = 0.7 5.7x10^-16 2 293/273 10⁵ 3.15x10⁸ = 2.7x10^-2 Pa
Presumably this can be taken as a rough guide to the partial pressure at which after-pulsing becomes significant.
The Philips catalogue also states that a tube immersed in He at atmospheric pressure fails within a few days. After 1 day, the internal pressure would be:
P_I = 10⁵ 5.7x10^-16 2 293/273 10⁵ 8.6x10⁴ = 1.0 Pa
which is completely consistent with the pressure associated with the onset of sparking.
Thus the permeation rate of He in borosilicate glass, obtained from published data, leads to a prediction of phototube breakdown that is compatible with the information given in the phototube catalogues.
Conclusion We should consider coating or wrapping the tubes to reduce the rate of ingress of He. We should take precautions to keep the ambient partial pressure of He in the region of the VPTs at or below the natural pressure in the atmosphere of 7x10^-6 bar.

1 Glass Science, Robert H Doremus, Wiley, P141.