,p* ?>.<y 
 
 %*/ / 
 
 MDDC - 811 
 
 UNITED STATES ATOMIC ENEBGY COMMISSION 
 
 NEW DEVELOPMENTS IN INSTRUMENTS FOR COUNTING 
 AND DETECTING NUCLEAR PARTICLES 
 
 by 
 W . H . Jordan 
 
 Oak Bidge National Laboratory 
 
 Date of Manuscript: 
 Date Declassified: 
 
 March 7, 19^1 
 March 31, 19^7 
 
 This document is for official use. 
 
 Its issuance does not constitute authority 
 
 for declassification of classified copies 
 
 of the same or similar content and title 
 
 and by the same author . 
 
 Technical Information Branch, Oak Bidge Tennessee 
 AEC, Oak Bidge, Term., 3 -9-^9- -750-A337 1 * 
 
NEW DEVELOPMENTS IN INSTRUMENTS FOR COUNTING 
 AND DETECTING NUCLEAR PARTICLES* 
 
 By W. H. Jordan 
 
 Instruments for detecting and counting nuclear particles have 
 improved considerably in the past five years. There have been no revo- 
 lutionary discoveries--one still uses geiger counters, ion chambers, 
 linear amplifiers, scalers, and scopes. The tools are still the same 
 but the design and capabilities of them have improved. This is due to 
 several factors. The tubes are better, circuit techniques are better, 
 and our knowledge of the detailed operation has considerably increased, 
 thus enabling us to use the tubes more effectively. Components such as 
 condensers and resistors are more reliable and have longer life so that 
 one can build more complex circuits without having to worry about whether 
 the instruments will function long enough to get the data. 
 
 One of the first things that any nuclear physics laboratory needs 
 is a good oscilloscope. Chiefly as the result of radar, scopes are 
 vastly improved. Cathode ray tubes now give small, bright spots so that 
 high sweeps are possible. Sweep circuits have developed so that one can 
 see the details of pulses 0.1 jii sec long. (One scope has been described 
 recently^ on which sine waves having a frequency of 10,000 Mc/ Sec can be 
 seen as individual waves. That's 10"10 sec j am talking about now.) 
 Video amplifiers with a bandwidth of 20 Mc are useful adjuncts to these 
 scopes. And for nuclear work the development of good delay lines has 
 been a great boon. One can thereby investigate the pulses that come from 
 geiger counters; for example, by first letting the pulse start the sweep, 
 then delaying it for a fraction of a miscrosecond before applying it to 
 the deflecting plates of the scope. Those of you interested in the de- 
 tails of these developments should procure appropriate volumes of the 
 Radiation Laboratory Technical Series^ t soon to be published. 
 
 One does not detect individual nuclear particles directly but 
 rather by the ionization they produce. Thus we have ionization chambers, 
 geiger counters, proportional counters, cloud chambers, secondary elec- 
 tron multipliers, crystal counters, fluorescent screens. In many cases 
 it is necessary to follow these instruments with amplifiers, pulse height 
 selectors, and scalers. These are the primary tools of the nuclear phys- 
 icist and will be the subject of my talk today. 
 
 First consider the ionization chamber. High velocity electrons, 
 protons, as alpha particles ionize the gas. The ions are collected on 
 one plate, thus producing a current of electricity. Gamma rays are de- 
 tected with poor efficiency by the secondary electrons they produce. 
 
 ♦Presented at the April 19^1 meeting of the Southeastern Section of 
 the American Physical Society, as an illustrated speech. The slides are 
 not being included in this article. 
 
 MDDC - 811 1 
 
2 MDDC - 811 
 
 Neutrons are detected by several reactions . In hydrogenous materials 
 knock-on protons are produced . Alpha particles are produced in boron by 
 the B 10 (n,a) Li7 reaction so that the chamber may be filled with BF3 or 
 lined with boron. A significant development is the improvement in efficiency 
 of detection by the use of the separated isotope BlO which can now be 
 produced in significant quantity. Another very useful reaction for the 
 detection of neutrons is the (n,a) reaction,, By coating the chamber with 
 fissionable material such as Tj235 one obtains very heavily ionizing fission 
 particles o Such fission chambers can be made very small and compact and 
 yet use a large area of plate by making the plates of thin strips wrapped 
 in a spiral. 3 
 
 Let us next consider how one observes the current in the ion chamber. 
 When the current is large, a galvanometer is adequate. Very small currents 
 are usually detected by measuring the voltage across a high resistance, 
 perhaps 10 12 or 10^5 ohms. The use of electrcOBter tubes such as the FP-5U 
 is well known, and recent improvements in its construction have been made. 
 Other small electrometer tubes have recently been produced by Victoreen. 
 Also, well known are the zero drifts that one encounters with DC amplifiers 
 so the vibrating condenser electrometer technique is fast coming into use.^° 
 Here one converts the small DC voltage into an AC voltage by varying the 
 capacity of a condenser connected to the ion chamber. The amplification 
 of this voltage by an AC amplifier and its subsequent rectification is a 
 simple matter . The use of feedback makes a very stable system. 
 
 If one is interested in detecting individual particles the situation 
 is quite different and the use of electron collection techniques ie one of 
 the most interesting developments in the past few years „ Both the ion 
 chamber and the linear pulse amplifier following are involved and it will 
 be necessary to consider each in considerable detail. 
 
 First let us consider what goes on in the ion chamber following the 
 passage of a nuclear particle .7 Electron- ion pairs are created along the 
 path, about 30 ev of energy being substracted from the particle for each 
 pair created. The electron and positive ion move in opposite direction 
 under the influence of the collecting field. So long as the electron 
 does not become attached to a gas molecule, its velocity will be much 
 higher than that of the positive ion. The probability ft of attachment and 
 the velocity of the electron can be seen from Table 1„° (Slide I) 
 
 If a gas with a small value for the probability of electron attach- 
 ment is used, the electrons are nearly all collected before they form 
 heavy negative ions. By using a mixture of gases, such as Argon and CO2 
 the time of collection can be made very small, of the order of O.lju. sec. 
 When this is compared with the several hundred miscrossconds required to 
 collect* heavy ions, one begins to see some of the advantage of electron 
 collection. 
 
 Let us now consider what happens to the potential of the collecting 
 electrode following an ionizing event. As the electron drifts toward the 
 positive electrode (assumed to be the collecting electrode) its potential 
 drops rapidly until the electrons are stopped. Then the potential changes 
 very slowly as the positive ions move out to the other plate. If one now 
 uses an amplifier that is insensitive to slow changes in potential, i.e., 
 discards low frequencies by means of a short time constant coupling circuit, 
 
MDDC - 811 3 
 
 one gets out of it a sharp, high pulse. By making the differentiating 
 time constant only somewhat longer than the electron collection time one 
 can count ionizing events that are separated by only 1 p,sec. This per- 
 mits high counting rates (up to 100,000 per second) and accurate coin- 
 cidence measurements. It permits counting of alpha particles in the 
 presence of an extremely high/3- or^- ray background, since there is 
 little time for "pile-up ," Troubles with microphonics and hum are prac- 
 tically eliminated since the low frequencies have been discarded „ These 
 items represent very real improvements in counting techniques „ 
 
 There is one disadvantage with the scheme of using electron col- 
 lection, namely, the pulse height depends on where the electrons started 
 in the chamber „ This can often be circumvented by insuring that all 
 ionizing particles go through the same part of the chamber and in the 
 same direction . Where this is not feasible, the use of an extra grid 
 near the collecting electrode may be required „ 
 
 Linear amplifiers for fast counting have been considerably improved 
 in the past few years „ 1 would like now to spend a few minutes describing 
 an amplifier which has recently been developed at Oak Ridge National Labo- 
 ratory and is typical of what- can be done with present techniques," 
 (Slides 2 and 3)° A part of the address presented at this point was read 
 directly from the paper, of "A General Purpose Linear Amplifier," by Jordan 
 and Bell o 9 
 
 Following the amplifier and preceding the mechanical register 
 comes the scaler „ A modification of the Eccles-Iordan circuit due to 
 Higginbotham represents a significant improvement in that the scaling is 
 much more reliable . Scalers (Slide k)^ using this circuit are available 
 commercially with resolving timer of 5 j^'sec. Lawson has recently de- 
 veloped a decade scaling circuit which is being marketed as the TY2-1 
 scaler with a 0.1|i sec resolving time. 
 
 Our insight into the operation of geiger counters has been extended 
 in the past few years and described in publication by Montgomery, Korff , 
 Stevens, Bose, and Present .12 The use of quenching gaaes is now generally 
 adopted and the nature of the quenching mechanism fairly well understood,, 
 Bandom delays in the time of firing are troublesome when very accurate 
 coincidence measurements are attempted,, The delays amount to about 0,1 
 miscros®cond as has been shown recently by Sherwin.12 Two possibilities 
 for reducing these delays have been considered recently . One is the use 
 of crystal counters^ and the other the use of electron multiplier tubes 
 
 Ik 
 
 Whether the crystal counter is any better in this respect is now 
 being investigated. They are, however, interesting from another stand- 
 point in that the size of the pulse produced is proportional to the 
 energy of th§ j3 - particle. Crystal counters are much like ionization 
 chambers except that the gas is replaced by a solid, in this case a 
 single silver chloride crystal at low temperature. Ions produced by the 
 passage of a beta particle are collected on one electrode, thus producing 
 a voltage pulse. Since the crystal is dense, a high energy beta particle 
 will be stopped in it, thus producing some 100,000 ion pairs, quite ade- 
 quate for detection when coupled to a linear amplifier. The higher the 
 energy of the beta particle, the more ion pairs produced. One serious 
 difficulty is that beta particles coming into the crystal may be scat- 
 tered back out before they have been stopped. 
 
4 MDDC - 811 
 
 The secondary electron multiplier has proven valuable for counting 
 individual nuclear particles such as protons and alpha particles . It 
 needs considerable improvement to get it into a form convenient for every- 
 day use. But it does have a small delay, and a fast pulse rise time,. 
 The fact that it operates in a vacuum is often a distinct advantage over 
 proportional counters and geiger counters , 
 
 Proportional counters have been used very extensively „ When boron 
 coated or BFo filled, they make excellent neutron detectors . Their char- 
 acteristics have been investigated recently by Rossi and Divan- ■"5 They 
 showed that if care is taken, both in construction and filling, the volt- 
 age pulse obtained is proportional to the ionization produced- The leads 
 supporting the counter wire should be shaped to reduce the effects of 
 fringing electric fields- Also, the gas must have a low electron attach- 
 ment coefficient- Argon, hydrogen, and nitrogen work very well with gas 
 multiplications up to 1000- Multiplication up to 10,000 is possible with 
 the addition of a quenching gas such as methane - 
 
 REFERENCES 
 
 1- Proc- I-R-E. 19^7- 
 
 2- Radiation Laboratory Technical Series, McGraw-Hill (In Press)- 
 
 3- W„ C- Bright, Spiral Fission Chambers, MDDC-91. 
 
 4„ Lafferty and Kingdon, Jour- App- Phys-, Nov- (1945). 
 
 5- Palevsky, H-, R- K- Swank, and R- Grenchick, Design of Dynamic Con- 
 
 denser Electrometer- MDDC -38. 
 
 6- Scherbatskov and Feron, Phys- Rev. 70s96, (1946). 
 
 7. Snyder, T. M., Theory of Ion Chambers, MDDC -475- 
 
 8. Allen, J. S-, and B- Rossi, Time of Collection of Electrons in 
 
 Ionization Chambers, MDDC-448. 
 9- Jordan, W- H-, P. R. Bell, A General Purpose Linear Amplifier. 
 (Has been declassified and submitted to RSI for publication) - 
 
 10- Higgenbotham, W-, Scaler Unit Circuit, MDDC-114. 
 
 11- Van Nostrand, D-, Electron and Nuclear Counters, Korff. 
 
 12- Sherwin, Paper at Jan. 29, (194-7) Meeting of American Physical 
 
 Society. 
 
 13- Van Heerden, P- J-, The Crystal Counter, A New Instrument On Nuclear 
 
 Physics- Published by N. V. Noord - Hollandsche Uitgevers Maat- 
 schappij Amsterdam, 1945- 
 
 14- Allen, J- S-, RSI, 12, 484, (l94l)- Bay, Z„, RSI, 12, 127, (l94l) 
 
 15- Diven, B. C-, Operation of Proportional Counters at Pressures above 
 
 Atmospheric, MDDC -458. 
 
 END OF DOCUMENT 
 
UNIVERSITY OF FLORIDA 
 
 3 1262 08910 5653 
 
 t