Alpha Particle Spark Counter (Part 2)

Send to Kindle

By Timothy Raney…Bald Engineer Guy with Glasses

Part 2 Introduction
This section of the paper describes the concluding experiments of the multiple-wire anode spark counter designed to detect alpha (α) particles. Read Part 1 here.

The results indicated the best operating potential ranged from 7kV to 10kV. The ~10kV was the upper limit for proper functioning. The detector was not connected to the oscilloscope at this time. Therefore, visual observation was the only means to verify that the detector was responding to the alpha source. The detector orientation was “grid up” (normal orientation) as shown in figure 2. An experiment was then performed to see if orientation affected detector functioning. In this case, the detector was inverted and fixed above the 95Am241 source. The intent was to verify its function for later experiments using loose uraninite granules. The inverted orientation would preclude loose granules from falling into the detector. In practice, the detector was inverted and held above the source with a burette clamp attached to a lab stand. The potential was adjusted to 9 and 10kV. Ambient temperature was 22oC. At both values, consistent discharges occurred at ~30mm. As expected, this experiment confirmed the detector would operate properly when inverted. There was no observable difference in its response. The alpha particle-induced discharges were confirmed again via interposing a 1.6mm thick aluminum alloy sheet between the source and detector.

In the last series of experiments, the oscilloscope was connected to the detector’s output via a 690pF capacitor to the positive lead through a 5MΩ resistor. The other side of the capacitor (anode side) was connected through a 10MΩ resistor to the negative oscilloscope lead. Circuit components for this attenuator circuit were selected to produce an input pulse that did not exceed the oscilloscope’s scale. A commercial attenuator probe intended for this oscilloscope was not on-hand. This circuit worked very well with the ~1μCi 95Am241 source[1]. The corresponding oscilloscope pulse heights were in the 10V range. When the spark counter was used with uraninite source, it did not elicit a consistent response, even though its activity was ~5 x 104 counts per minute (CPM). However, it was important to recognize the ~5 x 104 CPM was only equivalent to ~0.022 μCi (or 22 nanocuries).

Thus, it is understandable the uraninite did not elicit a response similar to the 1mCi 95Am241 source. Moreover, uncontrolled detector discharge occurred from 6 to 7kV with the oscilloscope in the circuit. These continuous discharges occurred without a source being present. When the potential was lowered to 5kV, the counter responded fairly well to the uraninite, i.e., sparks indicated alpha particles were being emitted. However, the response did not coincide with its activity as measured independently by a Geiger-Müller counter with the alpha particle probe. In other words, the spark counter could detect alpha particles from uraninite. Additionally, it would yield a rough order of magnitude result when comparing different sources. However, the total recorded counts might not agree very well when compared to better instrumentation. One might also conclude the relative alpha particle energies of 92U238 and 95Am241, i.e., ~4.15MeV (million electron-volts) and 5.44MeV, respectively[2] is another factor. Though these results showed the spark detector clearly worked much better with a source in the mCi activity range. A comparison between equal alpha emitters with different particle energies would eliminate the former as a variable.

This paper described experiments with a Chang-Rosenblum type multiple-wire anode spark counter designed to detect alpha particles. These experiments were the first step in attempting to characterize this spark counter prototype’s performance. This particular prototype worked very well when used with an alpha particle source of at least one microcurie. This spark counter can also detect the alpha emissions from uranium-bearing minerals, but the results are less consistent given activities in the nanocurie range. Though more work to refine the system is indicated, the spark counter is an interesting and relatively simple means of detecting alpha particles. Once again, in the immortal words of Professor Julius Sumner Miller, “No Experiment is a failure!”

Cited References.

[1]    C.C.H. Washtell, An Introduction to Radiation Counters & Detectors, Philosophical Library, Inc., New York, 1960, pp. 63-65.

[2]    O.C. Allkofer, Spark Chambers, Verlag Karl Thiemig KG, Munich, Germany, 1968, pp. 37-38 and 45-47.

[3]    W. Y. Chang and S. Rosenblum, A Simple Counting System for Alpha-Ray Spectra and the Energy Distribution of Polonium Alpha-Particles, Palmer Physical Laboratory, Princeton University, 29 December 1944.

[4]    F.M. Mims III, Semiconductor Projects (vol. 1), Radio Shack – A Tandy Corporation Company, Fort Worth, TX, 1975, pg. 46.

[5]    B.B. Rossi and H.H. Staub, Ionization Chambers and Counters: Experimental Techniques (5th Ed.), McGraw-Hill Book Company, Inc., New York, 1949, pp. 125 and 132.

[6]    One mCi equals 3.7 x 104 disintegrations per second (becquerel). See almost any modern physics text.

[7]    R.A. Dunlop, Experimental Physics: Modern Methods, Oxford University Press, Inc., New York, 1988, pp. 290 (table 11-8).

End Notes:

[Abstract] W. Y. Chang and S. Rosenblum, Palmer Physical Laboratory, Princeton University, Received 29 December 1944. An alpha-ray counting system, consisting of parallel wire electrical counters and amplifying circuits, etc., is described. The conditions of reliable operation have been carefully examined and well defined. The mechanism of the counters as to why they answer only to alpha-particles but not to even strong beta- or gamma-rays is generally discussed. The distribution form of the Po-alpha particles determined by these counters in conjunction with an alpha-ray magnetic spectrograph agrees well with that obtained from a photographic line. The resolving power for the alpha-ray lines is very high, owing to the small active region of the counters. ©1945 The American Physical Society.

[1] One mCi equals 3.7 x 104 disintegrations per second (becquerel).

[2] R.A. Dunlop, Experimental Physics: Modern Methods, Oxford University Press, Inc., New York, 1988, pp. 290 (table 11-8).

This entry was posted in Experimentation, Instrumentation, Measurement, Projects, Uncategorized. Bookmark the permalink.

One Response to Alpha Particle Spark Counter (Part 2)

  1. Dave says:

    It may be interesting to characterize the distributed capacitance of the spark gap along with the equivalent series resistor. Such a circuit composes a RC network, such that a discharge will drop the voltage across the gap, where it will exponentially return to the steady state voltage. However, while the voltage is below the steady state value, it may not produce a spark for a subsequent alpha particle. This may result in a smaller number of counts than expected for a high activity source.

    Additionally, the gas (air?) will require a time to deionize following a spark event. GM tubes use a gas mixture which has a quenching effect. This, too, may affect the linearity of the count with respect to the activity of the source.

    One should also consider the photoelectric effect, which may produce electrons from the electrodes, which can cascade as they transit the gap, possibly producing false counts.


Leave a Reply