FLUORESCENCE YIELD OF DOUBLE K VACANCIES IN KRYPTON

HAL Id: jpa-00227212

https://hal.archives-ouvertes.fr/jpa-00227212

Submitted on 1 Jan 1987

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

FLUORESCENCE YIELD OF DOUBLE K VACANCIES IN KRYPTON

M. Hribar, A. Kodre, D. Glavić

To cite this version:

M. Hribar, A. Kodre, D. Glavić. FLUORESCENCE YIELD OF DOUBLE K VACANCIES IN KRYP- TON. Journal de Physique Colloques, 1987, 48 (C9), pp.C9-625-C9-628. �10.1051/jphyscol:19879105�.

�jpa-00227212�

FLUORESCENCE YIELD OF DOUBLE K VACANCIES IN KRYPTON

M. HRIBAR, A. KODRE and D.

GLAVIE

VrO Fizika FNT and Institute J. Stefan, E. Kardelj University of Ljubljana, PO Box 199/IV, YU-61001 Ljubljana, Yugoslavia

The formation of double K-shell vacancy in absorption of single photon is studied in krypton. A multiwire wall-less proportional counter filled with a mixture of krypton and propane is used to detect the double vacancies separately according to their mode of decay. The analysis of spectra of pulses demonstrated the independent particle model of decay to be inappropriate. Using the collective model the probabi- lities for radiative, radiative-nonradiative and nonradiativy modes in tb+e decay of double vacancy state were preliminary determined as < 0.1

-

^0.1, 0.35

-

^{0.05 and}

0.54 2 0.05, respectively. The double K-shell ionization cross section at the edge was determined to be less than 8.2(E-3) of the total photoelectric cross section.

introduction

The multiple photoionization has extensively been studied in noble gases where cross section for the process is relatively large (1). However, asymmetric excitations with one vacancy in a lower shell and the other(s) in higher shells have only been investigated, being a part of theaccompanyingstructure of single absorp- tion edges.

Symmetric excitations, in particular the double K photoeffect, exhibit most strongly the effects of interaction of the two vacancies, in creation as well as in decay. One of the rare instances of a direct observation of the double K photoeffect is a recent experiment of Salem and Kumar (2) on copper. However, some properties of double vacancies have been known from the study of satellite and hypersatellite lines of X-ray spectra in ion bombardment.

The interaction of the vacancies, resulting in the hypersatellite energy shift, can also affect the rates of radiative and nonradiative modes of decay. The number of fluorescent photons per vacancy, i.e. the fluorescence yield can conveniently be studied by the use of a proportional counter (3). Provided that the gas mixture in the counter is sufficiently thin the fluorescent photons largely escape from the sensitive volume. The energy detected by the device in the absorption of a photon by an atom of the gas mixture depends on the mode of decay of the vacancy. The events ending with emission of a fluorescent photon are registered with an energy loss.

IJe looked for the double K photoeffect in krypton using a multiwire proportio- nal counter built for fluorescence yield experiments ( 4 ) . In a separate experiment

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19879105

JOURNAL DE PHYSIQUE

the existence of K-- state was confirmed by a scan of the absorption coefficient in the region above twice K-edge energy of krypton (28.646 keV).

Experiment

Results

The multiwire proportional counter is filled with a mixture of 45 mbar of krypton and of 520 mbar of propane. The gas is irradiated along the diameter of the detector by a thin beam of X-rays provided by a convent;ional tungsten anode X-ray tube and a Bragg monochromator with a LiF (220) crystal. For the purpose of the experiment the energy scale of the monochromator is locally calibrated with scans of the K shell absorption edges of indium (27.928 keV) and tin (29.190 keV). The reso- lution of the device is estimated at 110 eV.

A spectrum of pulses from the detector at the photon energy 29.5 keV is shown in Fig.1. Although the energy is above the threshold for double K photoeffect, the features of the spectrum are governed by the single vacancy process: the prominent peaks at 29.5 and 16.9 keV correspond to the Auger and the fluorescent decay of the K vacancy, respectively. The double photoeffect events are hidden within the same

peaks: the resolution of the detector is too low to show separately the escape of hypersatellite fluorescent photons.

We tried to interpret the experimental data by using the independent vacancy model. In this model one of the two vacancies decays with hypersatellite energy shift and with the shifted flu~rescence yield W ' while the other behaves as a single vacancy with well known fluorescence yield & = 0.66 (3). The nonradiati- ve processes appear with the probability 1- d j or 1-w respectively. The number of counts in the full energy peak acquired during the pasge of N photons through the counter containing n atoms in the bean volume is then

counts 150M) -

low -

5000

-

To extract the double photo- effect contribution a series of

escape peak spectra is recorded with the inci-

dent photon energy increasing from below to above the tentative double K edge. An argon filled counter is used as a beam monitor to integrate

full energy peak the dose delivered per spectrum.

' 3. The numbers of counts in the peaks

. .

. .

were corrected for propane absorp-

. .

tion and for the diminishing and mixing associated with the anti- coincident connection of the main

. . . . . . . . . ' .

and ring counter. The corrections are within 4-6 tal number of counts in both peaks %. In Fig. 2 the to-

. . _. . .

normalized to monitor counts vs in-

*: ⁸

I

..

cident photon energy is plotted in

. 1

.L+ a double log graph. In the same

50 Im 150 Im3

$m$tf

graph the relative contribution of the escape peak to the joined num- Fig. I ber.is displayed. The steady slope of the joined number plot follows the detector efficiency while the jump between the points at 28.9 keV and 29.4 keV shows the onset of the double K photoeffect. From the plot the relative jump is de- termined by extrapolating below and above edge trends to the point in the middle of the interval where the edge is supposed to appear. The value A (Ne+Nf )/(Ne-1-Nf) =

= (7.5 2 0.5) (E-3) is obtained. Similarly the relative jump values werz determined for full energy as yell as escape peak, respectively: hNf/Nf = (10.5

-

0.5) (E-3) and 4Ne/Ne = 5 . 5

-

0.05)(E-3). The jump in the escape peak ratio is indicative of the change in the fluorescence yield ; it was determined as

-

2.4 (E-3).

N A 0.3

d l section and double vacancy production

/' cross section, respectively and PK=dK!da

*/'

/' the relative K-shell photo cross section.

.

^/ Similarly we find for the escape peak Ne=nN(PKda&2 + $K(~'(l-~)+cJ(~-~')) In the analysis we did not consider the proposed double escape peak containing N, = ~ N ~ ~ ~ w w ~ ~ u ~ s ~ s as it was not detec- ted due to structured low energy back- ground of the proportional counter. From the two inde~endent relative differences of the numbek of counts at the K-- edge

0.2 aNf./Nf=(dKK/da) (1-@I ( I-dl)/( 1 - u ~ ~ ) and -1 0 E d N,/N,= (dKK/dap@) (&I( l-d)+ o(

I-u')

)

I n

-

2 E ~ one can obtain the desired values for

- 0.55

-0-.-

dKK

and U)' as compared to

da

and 3

.

.-.=;/.

/. ^/*-.r

However, using the experimental data and the tabulated values& 0.66 and PK=0.86 we obtained W' to be negative in clear contradiction to the definition of the

-

^0,6 coefficient. The result shows the in-

N I adequacy of the proposal

,

so the collec-

Ln

-

N f + N, tive model of deexcitations was intro-

duced.

Fig. 2 We define wrr as the probability for the double vacancy state to decay via pure radiative decay, w~~ as the probability for mixed radiative-nonradiative decay and WAA as the probability for nonradiative decay. From the definition it follows that WAA

+

WAr +

"

^{rr = I}

The full energy peak in the spectra can then be attributed to the non- radiative decays of both single and double vacancy states

Nf = ~n(d,(l-UP,) + dKKwAA)

The escape peak comprises the pulses from radiative decays of single vacancy states as well as the pulses from mixed decays of double vacancy states

Ne = ~ n ( d ~ w ~ ~ + dKKwAr)

The second escape peak should appear composed of pulses from pure radiative decay of double vacancy states, mounting to

Ne = Nnd w KK rr

Below the double K-edge only the contributions of single vacancy states have to be considered.

From the two independent jump values obtained at the K-- edge:

4 N,/N, = (dKK/<) (wAr/&PK) and A Nf/NE = (SK/da) wAA/( 1- WPd together with the jump in the total number of counts

ANt/Nt =

(dKK/da)

( l-wrr)

we determined the following values for the relations b~tween unknown coefficients

6

KK

/d

a = 7.5 (E-3)/(1-wrr), wAA/(l-wrr) = 0.61 f 0.05

,

wAr/(l-wrr) = 0.&$? 0.05 To check the consistency the relative jump in the ratio N,/(N,+N~) Was CalCula- ted using the determined coefficients and cornparedto experimental observations. The value

-

2.35(E-3) was obtained as compared Co the expekhental

-

^2.4(E-3).

JOURNAL DE _PHYSIQUE

From statistical considerations the detection limit for the second escape peak was found to be less than (E-3) of the total number of counts, limiting

wrr to be less than 0.1

'

^0.1.

Conclusions

The proportional counter with internal fluorescer proved to be ideal tool for the study of deexcitation processes of single as well as of double vacancy states.

The coefficients for collective transitions have to be introduced in deciphering the spectra which indicates very strong coupiirig of the two vacancies. More experimental as well as theoretical work will be required to resolve the indicated effect.

References

( 1 ) Deslattes R.D. et al., Phys. Rev. A

7

(1983) 923

( 2 ) Salem S.I. and Kumar A., J. Phys. B: At. Mol. Phys.

19

(1986) 73

(3) Eanbynek W. et al., Rev. Med. Phys. 44 (1972) 710

( 4 ) Pahor J . et al., Z. Phys.

227

(1969)490

(5) Storm E . and Israel H.I., Nucl. Data Tab. (1970) 565