22.101 Applied Nuclear Physics (Fall 2006)

Lecture 21 (11/29/06)

Detection of Nuclear Radiation: P u lse Height Spectra

References :

W. E. Meyerhof, Elements of Nuclear Physics (McGraw-Hill, New York, 1967), Sec.3-6.

We have just concluded the study of radiation interaction with m a tter in which the basic m echanism s of cha r ged particle, neut ron and gamma interactions were discussed separately. A topic which m a kes us e of all this infor m ation is the general problem of detection of nuclear radiation detection. Although this subject properly belongs to a course dealing with experim e ntal aspects of applied nuclear physics, it is nevertheless appropriate to m a ke contact with it at this poi n t in the cours e . There are tw o reasons fo r this. First, radiation detection is a cent ral part of the foundational knowledge for all students in the departm e nt of Nuclear Scie nce and Engineering. Even though we cannot do justice to it due to the lim ited tim e rem a ining in the syllabu s , it is worth w hile to m a ke som e contact with it, however brief. Sec ondly, an analysis of the features observed experim e ntally in pulse-height spectra of gamma radiation is tim ely given what we have just learned about the -interaction processes of Compt on scattering, photoelectric effect

and pair production.

We start by noting that regardless of the t ype of nuclear radiation, the interactions taking p l ace in a m a terial m e dium invariab ly result in ion i zatio n and excitation which then can be detected. Heavy charged pa rticle s a nd elec trons produce ion pairs in ionization cham bers, or light em ission (excitation of atom s) in scintillation counters, or electron-ho le pairs in sem i conductor detector s. Neutrons co llid e with protons which recoil and produce ionization or excitation. In the case of gamm as, all 3 processes we have just discussed give rise to ene r g e tic elec tron s which in turn cause ionization or excitation. Thus the basic m echanism s of nuc lear radiation detection involve m easuring the ion i zatio n or excitation occurr ing in the de tector in a way to allow one to deduce the energy of the incom i ng radiation. A useful su mm ary of the different types of detectors and m e thods of detection is given in the follo wing table [from Meyerhof, p. 107].

1

Common Nuclear Detectors

# T h e i n i t i a l k i n e t i c e n e r g y o f t h e p a r t i c l e i s c a l l e d T 0 .

2

F i g u r e b y M I T O C W .

We focus on detection of radiation. Suppose we are concerned with the problem of detecting tw o rays, at en ergies 1.37 Mev and 2.7 5 Mev, em itted from the

radioactive Na 24 . The m easurem ents are in th e for m of puls e -height spectra, num ber of counts per channel in a multichannel analy zer p l otted ag ains t the puls e height, as sho w n in Fig. 19.1. These are the results obtained by using a detector in the form of Na-I scintillation detector. Th e sp ectra are seen to hav e two sets of features, on e for each incident . By a set we mean a photopeak at the incident energy, a Com p ton edge at an

energy approxim a tely 0.25 Mev (m e c 2 /2) below the incident energy, and two so-called escape peaks denoted as P1 and P2. The escape peaks refer to pair production processes where either one or both anni hilation photons escape detec tion by leaving the counter.

Backscattered radiation

1.368 Mev

C of 1.368-

Mev 

P1 of 2.75-Mev 

0.51 1 Mev

2.75 Mev

P1 of 1.368-Mev 

P2 of 1.368-Mev 

C of 2.75-Mev 

P2 of 2.75-Mev 

10 5

10 4

10 3

10 2

0 400

800

Pulse Height

F i g u r e b y M I T O C W .

1200

Fig. 19.1. Pulse-height spectra of 1.37 Mev and 2.75 Mev obtained using a Na-I detector. (from Meyerhof)

3

Thus, P1 should be 0.511 Mev below the in cident energy and P2 should be 0.511 Mev below P1. The other features that can be seen in F i g. 19.1 are a peak at 0.511 Mev, clearly to b e identified as the ann i hilation pho ton , a backscattered peak as sociated with Com p ton scattering at    which should be positioned at m e c 2 /2, and finally an unidentif ied peak which we can attribute to x -ray s em itted f r om excited a t om s.

One can notice in Fig. 19.1 that the vari ous peaks are quite broad. This is a known characteristic of the scintillation detector, nam e ly, relatively poor energy resolution. In contrast, a sem i conductor dete ctor, such as Li-drifted Ge, would have much higher energy resolution, as can be seen in Fig. 19.2. In addition to the sharper lines, one should notice that th e peaks m easured using the sem i conductor detector have different relative inten s ities com p ared to th e peaks obtained using the scintillation detector. In particular, looking at the rela tive inte nsities of P1 and P2, we see that P1 > P2 in Fig. 19.1, whereas the opposite holds, P2 > P1, in Fig. 19.2.

10 5

10 4

10 3

10 2

10

0 100

200

Pulse Height

F i g u r e b y M I T O C W .

300 400

Fig. 19.2. Sam e as Fig. 19.1 except a sem i conductor detector is used. (from Meyerhof)

4

This difference can be explained by noting that the scin tillation detec t or is large r in physical size than the sem i conductor detector. In this particular case the f o rm er is a cylinder 7.6 cm in dia m e t er and 7.6 cm in lengt h, whereas the latter is 1.9 cm in dia m e t er and 0.5 cm i n height. Th us one can expect th at th e probability that a pho to n will es cap e from the detector can be quite d i fferent in th ese two cases.

To f o llow up on this ide a , let u s def i ne P as the p r obability of escape. In a one- dim e nsional situation P ~ e   x , where is th e linea r atten u ation co ef f i cient and x is the dim e nsion of the detec t o r . Now the probabi lity that one of th e two annihilation gam m a s will es cape is P1 = 2P(1- P ), the f acto r of 2 com i ng f r om the fact th at e ithe r gam m a can escape. For both gamm as to escap e the probability is P2 = P 2 . So we see that whether P1 is larger or sm aller than P2 depends on the m a gn itude of P. If P is s m all, P1 > P2, but if P is close to unity, th en P2 > P1. For the two detectors in question, it is to be expected that P is larg er for the s e m i conductor detector . W ithout putting in actual num b ers we can infer from a n inspection of Figs 19.1 and 19. 2 that P is sm all enough in the case of the

scintillation detector for P1 to be larger than P2, and also P is close enough to unity in the case of the sem i conductor detector f o r P2 to be larger than P 1 .

5