Proton pulse radiolysis

CHEhllCAL

Volume 22, number 3

PHYSICS LETTERS

15 October 1973

PROTON PULSE RADIOLYSIS H.C. CHRISTENSEN, ABA

G. NILSSON,

fomenergi, Srudsvik.

K.-A. THUOMAS

N_rktipinn, Sweden

and T. REITBERGER Division of Nuclear Chemistry, Royal Itistinrre of Techriology, Stockholm, Swederr Received 1 August 1973

A 5 hlcV proton accelerator hhs been used for pulse radiolysis of organic gases. The transient spectra obtained from the alkanes methane, ethane, propane. n.butanc and neopentane have tentatively been assigned to alkyl radicals. Some methodological aspects of this new technique are discussed.

1. Introduction One objection

to the use of protons

in connection

with pulse radiolysis has been their comparatively short range unless the energy is made very high. High energy protons require expensive accelerators and there is also a risk that radioactivity is induced in cells and equipment [I]. In certain cases, however, protons with an energy of only a few MeV can be used to advantage. (i) Investigations of radiation chemical reactions in the surface phase of liquids and solids. It is obvious that protons are more suitable than electrons for producing

radicals

close to the surface

in such

In principle the experimental set up can be made quite simple, i.e., one can irradiate the free surface under controlled atmosphere in ai open vessel. (ii) Investigations of samples which have considerable light absorption or light scattering properties, such as polymers and polycrystalline solids. These can be studied in the form of thin films. (iZ> Investigations of gases at atmospheric or subatmospheric pressure where a very efficient use of the radiation energy is possible due to the short range of the protons and the small spread of the proton studies.

beam.

Electron

accelerators

normally

used for gas

studies deposit only about 1% of the energy ir. the gas [2]. In this paper we report - to our knowledge - the fust pulse radiolysis study carried out with protons. Examples of items (ii>and (iii) above are given with emphasis on methodological aspects.

2. Experimental The 5 MeV proton accelerator (van de GraafQ at Studsvik, which normally operates with a mezn current of 8-10 PA and pulses of l-100 nsec length 81 a frequency of 1 MHz, was modified for this experiment to give 12 Dsec single pulses with 0.5 mA peak

current. The proton energy after psssage through the exit window

was 4.6 MeV and the total energy

of rhe

pulse was 1.7 X 1017 eV or 0.03 J. This is much iess thzn generally accepted in normal puke work with electrons. However, since the range of 4.6 MeV protons is only about 42 mg/cm2 (2 = 10) and the beam area is about I cm2 the dose absorbed wilI be quite high, i.e., about 60 krad. The range of4.6 MeV protons in water is about 0.03 cm while the range in a gas is about 103 times longer. Pulse radioI@ experiments on gases are therefore more feasible and were chosen as a starting point. 533

Volume

22;

number

CHEMICAL

3

PHYSICS

Since the proton beam does not spread very much in a gas the proton energy is most efficiently utilized if the proton and light beams are collinear. The experimental set-up was arranged accordingly. If we assume that the proton energy is homogeneously de. posited the optical density D of a radiation-produced species can be calculated from:

D = ! .5 X lo-’

eEItG/nr2 ,

(1)

where E is the extinction coefficient (M-l cm-l), E the proton energy (MeV), I the beam current (mA), t the pulse length @set), G the yield and r the beam radius (cm). The optical path > the proton range. The experimental set-up for gas-phase studies is shown in fig. 1, The gas cell was open towards the proton beam and the gas under investigation was continuously

flushed

through

LETTERS

arrangement

October 1973

is far from ideal since about 40% of the

useful light beam is lost through the hole in the mirror. After passage through the cell, the light reflected by the plane mirror was focused by a front surface aluminized spherical mirror onto the entrance slit of a Bausch and Lomb high-intensity grating monochromator. The slit chosen corresponded to an optical bandwidth of 5 nm. An EMI 9558 Q photomultiplier tube served as the light detector and its amplification was adjustable by changing the dynode current, as described by Keene [3]. The signal from the tube was amplified and displayed on a Tektronix 549 storage oscilloscope after automatic subtraction of the dc level, which in turn was read on a digital voltmeter. An estimate of the detection limit for the optical density of the entire system is F=5 X 10m4.

the cell. The optical alignment

was quite critical since the protons irradiate a narrow volume in the centre of the cell and only the light which passes this volume should be analyzed. In our first attempts we tried to irradiate through a mirror made from a thin aluminized foil placed in the same position as the plane mirror in fig. 1. However, this caused difficulties due to vibrations induced in the

foil by the proton pulse. Instead, we used a surface Guminized glass mirror with a 6 mm diameter central hole for the passage of the proton

beam (fig. 1). The

3. Results A quantitative test of the equipment bvasobtained by irradiating pure gas since the G-value (we have used the high-dose rate value of 12.8) [5] and extinction coefficient (3180 M-1 cm-l) [6] of ozone, which is formed during irradiation, are known. These values are also high, which facilitates the experiment. We have obtained an absorption spectrum with a maxi-

Preumpli kier

Digilol

vollmler

Fig. 1. The experimental

534

15

arrangement.

Volume

15 October 1973

CHEMICAL PHYSICS LETTERS

22. number 3 r

30 -

IL1

"a 3 % = z a% 20131

0

0

5 .v 0"

- 01

_g)

_I11

\--a_

“\,

200

250

Wavelength

lO--

Ill I1 300

Inm)

Fig. 2. Transient absorption spectra obtained by pulseradiolysis of alkanes; .: methane (1); 0: ethanc (2); A: propane (3); o: n-butane (4) and f: neopentane (5). The zero-line is dis-

placed as marked on the ordinate.

mum at 252 nm which agrees well with the reported value of 255 nm [5j. The optical density at the peak is 1.2 X 10e2 and from eq. (1) we would expect D = 3.5 X 1 O-2. The optical density is thus about 30% of the expected value, which is a good result considering the primitive gas cell used in the experiment. We have also irradiated methane, ethane, propane,

170

200

250

Fig. 3. Transient absorption spectra obtained by pulse radiolysis of unsaturated hydrocarbons; o: isobutyiene (I); C: butadiene (2); q : propylene (3) and c: ethylene (4). The zeroline is displaced as marked on the ordinate.

The absorption of the methyl radical is known the position of the peak as given by Herzberg [7] cated at 216 run which agrees well with the peak tion in fig. 2. In flames of ethyl ether and ethane don et al. [B] have observed absorptions at about

and is loposiGay324

Table 1

n-butane, neopentane, ethylene, propylene, butadim and isobutylene and have in the range 200 to 705) nm found transient absorptions below 300 run (figs. 2 and 3). For the alkanes (fig. 2) the transient absorption is located in the range where one would expect the corresponding alkyl radical to absorb. We therefore tentatively assign the spectra in alkanes to alkyl radicals. If we use the ozone system as a dosimeter, EC-values can be estimated for the peak of the transient absorptions. These values and the peak positions are given in table 1.

300

Wovelength Inn)

methane ethane propane n-butane neopentanc

EG -I cm-’ ( 100 07-l (x. IO’)

hm_x nm

hf

217 227 225 232 245

14 22 23 19 22

535

Volume 22, number 3

CHEMICAL PHYSICS LETTERS

run which they provisionally assigned to the ethyl radical. This absorption band is close to our peak posi-

tion in ethane. Recentiy, Stevens et al. [9] have presented

spectra

of the methyl

tained after pulse aqueous solutions.

and ethyl

radiolysis

of methane

Our methyl

radicals

ob-

and ethane

radical spectrum

in

is in

fairly good agreement whereas our ethyl radical spectrum disagrees with their published spectrum. Stevens et al. did not find any maximum

15 October 1973

hydrocarbons are more difficult to interpret since at short wavelengths these compounds also possess intrinsic absorptions. On the basis of the present data we zre not in n position ticular species.

to assign the spectra

to par-

A transient absorption was observed in a sample of commercial polyethylene foil (0.3 mm thick). The irradiation was carried out in an argon atmosphere.

in the wavelength

range 210-270 nm for the ethyl radical spectrum; the rapid rise of the absorption near 3 10 nm may,

Acknowledgement

however, be caused by a charge transfer interaction of a similar type as that described for the hydrogen atom in water [lo], since the electron affinities of these two radicals are about the same [ 111. Fig. 4 shows the result of an irradiation of cyclo-

The authors wish to thank Dr. T. Wiedling and the staff of the VdG accelerator for their kind cooperation and Miss J. Eriksen for technical assistance. The authors are grateful to Professor P.O. Kinell for stirn-

hexane

ulating

vapour

with

argon

as the carrier

gas. The ob-

served transient shows a marked shoulder at about 245 nm where the cyclohexyl radical is reported to absorb in liquid cyclohexane [ 1Z-141. In this case we observed aerosol formation after each pulse and this probably caused some light scattering. The transient absorptions obtained in unsaturated

discussions.

References

Ill MS. Matheson and L.hl. Dorfman, Pulse radiolysis (MIT Press, Cambridge,

1969) p. 7.

PI R.F. Firestone and L.M. Dorfman, Actions Chim. Biol. Radiations

15 (1971) 7.

i31 J.P. Keene. E.D. Black and E. Hayon, Rev. Sci. Instr. 40 (1969) 1199. [41 J.P. Keene, J. Sci. Instr. 41 (1964) 493. 151 C. Willis, A.W. Boyd, h1.J. Young and D.A. Armstrong, Can. J. Chem. 48 (1970) 1505. Jr. and W.A. hlulac. J. Chem. Phys. 56 (1972) 4995. G.T. Herzberg, Proc. Roy. Sot. 2G2A (1961) 291. A.G. Gaydon, G.N. Spokes and J. van Suchte!cn, Proc. Roy. Sot. 256A (1960) 323. C.C. Stevens, R.M. Clarke and EJ. Hat, J. Phys. Chem. 76 (1972) 386?. P. Pagsberg, H. Christensen, J. Rnbani, G. Nilsson, J. Fenger and S.O. Nielsen. J. Phys. Chem. 73 (1969) 1029. R-P. Blnunstein and L.C. Christophorau, Radiation Res. Rev. 3 (1971) 69. M. Ebert, J.P. Keene, E.J. Land and A,J. Swallow, Proc. Roy. Sot. 287A (1965) 1. K.C. Sauer and I. hfani, I. Phys. Chem. 72 (1968) 3856. V.I. hiakarov and S.A. Kabakchi, Khim. Vys. Energ. 5

161 M.C. Sauer [71 [‘31 [91 1101

I

I IllI

100

I

I

I

200

250

300

Wavelength

Fig. 4. Transient absorption vapour.

(nm)

in pulse-irradiated

1

[I21 1131

cyclohcxane

1141

(197 1) 5.