Radiological Laboratory University of California School of Medicine, San Francisco, Semiannual Report for June 30, 1955, Atomic Energy Commission Contract No. at-11-1-Gen-10, Project No. 2

Page 1

Unclassified Health and Biology University of California School of Medicine, San Francisco Department of Radiology for period ending June 30, 1955 Atomic Energy Commission Contract No. AT-11-1-GEN-10, Project Bo. 2

Page 2

TABLE OF CONTENTS I. Synchrotron Magnet and Components C. Injectors and Donuts 11. Experiments Involving Radiation Preliminary Use of the Calorimeter 111. Activities by Person I. LD,, and Slope Function of Sunrival Curve for Whole-Body On the Direct and Indirect Effects of X-Rays on the Testis of the Rat IV. Effect of Incubation Time on Macrocolony Formation After V. Combined Effects of Endogenous Respiration and Cell Concentration on the Killing of Microorganisms by X-Rays Current Status of Patients Treated for Hyperthyroidism It should be noted that the semi mud progress reparts of the Radiological Laboratory do not attempt to describe progress In all Of the research programs, but only those in vhich some significant results have been achieved but which a r e not sufficiently CWlete to be mitten as f i u reports. -

Page 3

Three years ago, preliminary biological irradiations using 70 MY & x-rays from the synchrotron were stopped i n order t o improve the reliability of doses delivered from this source. During this report period, development of the last component i n the effort t o achieve reliability appears t o have been coqleted. Afew other developments we still i n progress but these, when complete, vill serve generally t o improve senrlcing ar output rather before returning t o the irradiation of biological specimens has also been A portion of the physical calibration caneidered necessary I. Symhrotron Magnet and Cmponents A . Field Compensation. I n order t o approximate uniform dosage throughout reasonable volumes of tissue-like material and t o assist in the measurement of such doses, the naturally peaked x-ray beam from the synchmtron is compensated by the introduction of a suitably shaped absorbing filter. The last major problem t o be solved was that of providing an electro-mechanical system t o maintain proper alignment of this filter vith respect t o the x-ray The compensating f i l t e r which is used is approximately conical in shape, is 1.25 inches i n diameter at the base, and has an altitude of 1.68 inches. In the center of the beem the absorption is about a factor of five. This large central absorption and the highly peaked beam make the precise alignment of this filter with the beam rather critical. the support can be stated as foUows: positioned coaxially vith the x-ray beam t o produce the flatness desired. Then a tram la tion of the filter (a) perpendicular t o the beam dirtction produces a change i n the relative height of the shoulders of the previously flat beam shape (see Fig. 1). with translatLon (&F/d) i s about 32 per cent per mFllimeter. owing t o the mass of the lead blanket across the face of the synchrotron, no structures were found which were himobile and t o which t h e filter could be but some of the structures show r e l a t i v e motions of at least 1.5 mm between horizontal and vertical positions of the x-ray beem. The problem concerning Assume a filter has been &aped end The r a t e of change of heights of these shoulders Intensity variations of a fev per cent would not be objectionable, Despite the additional complication, a system of two independent and almost identical servomchaaiems has been developed and inatidled. operation of this device may be explained BB follows, referring t o Fig. 2: A quadrant transmission ionization chamber I s used t o Berme any variations from a previously determined condition of flatness. nected i n pairs through gain-stable d.c. amplifiers t o Brown amplifiers. The quadrants are con-

Page 4

C 0 MPE NSAT ED FIG. I. EFFECT OF FILTER MISALIGNMENT ON BEAM SHAPE.

Page 6

Brown amplifiers are chopper amplifiers which have an 8.c. output the phase of which is determined by the sigu of the d.c. input signal. such a.c. output is fed i n t o s two-phase motor so that the direction of rotation of the motor shaft depends on the sign of the d.c. eignsl. tuated by the motor rotation is a piston the motion of which displaces oil ina closed hydraulic system. A t the other end of the oil llnee ie a similar piston which is moved by the oil displaced by motor rotation. It is connected t o the support for the compensating filter. This suppart, is mounted i n independent horizontal sad vertical ways. OverdJ, then, an unsymmetrical change in the shape of the compeneated beam causes the filterto be moved in such a direction ae to ml rh ize thi6 change. The unit re6ponda t o differences of about 1 per cent at the level used and appears t o function reliably over substantial periods of time. The quadrant ionization chamber is temporsrily located in i m n t of the collimator. In order t o make t h e operation of this servmechaniem insensitive t o size and shape a? the field in w e for any particular exposure and t o mahe the front space available f o r other pulpo~ea, the sensing detector must be rebuilt far positioning between filter and colXmator. h &tempt ulll be made tom DU t it O n the BBme SUppOrt 88 the filter. A paper which describes in detail the COMtrUCtiOn and operation o? this semmechanism hss been prepared f o r publication in the open literaB. Vacuum System. Although very l i t t l e the has been devoted recently t o the replacement vacuum 8yStem for the synchrotron, the u d t continues t o operate. small improvements have been made and, aside from miaor annoyances, there now appears no reason not t o ln6taU the unit before starting clFnical therapy Irradiations. Injectors and Donuts. .A new source of electron injectors for theC. synchrotron has been found at Iowa State College, Ames, I-. this Laboratory has offered t o assist Iowa State College in the 11. Experiments Involving Radiation in photographic dens it n lack e ning) with exposure. attractive for utilization in differential dosimetry, particulsrly where time integral information i s useful. A great many exposures have been made and analyzed in this Laboratory primarily in connection with the beam shape and filter position problems. Such materid is very Density fluctuations have been found frm causes other than the p m - eters under investigation. Inadequate at ation during developwnt end Fluctuations remaining appesr t o be associated with th erm lla ion. r position in the developer tank were identified and atw w e taken.

Page 7

Simultaneous exposure of three similar films in the same cassette v i t h subsequent developnent singly i s the method used t o assees this defect. After processing, the densities are measured and, in the UBUal case of a beam shape, are norrnalized t o 100 per cent at the center f o r graphical comparison. Figure 3 shows two such comparisons: 3 depicts the best agreexen3 found at this Laboratory and 3B, the kind of disagreement w h i c h occurs with a freqcency which depends on the type of emulsion used. have been examined. m e MB ( B betatron selected) appears t o have been the best choice but i s no longer produced. ior t o be essentially uselsss f o r precision work. The large fluctuatloPe shown i n Fig. 3B occur much too often in the other types tested. B. Preliminary Use of the Calorimeter. The calorimeter is basicslly of convefitiooal design. The absorbers are twin 10-cm cubes of lead w h i c h are suspended on nylon thread in an evacuated braas container. these blocks are the,-mistors 88 sensing elements and heaters f o r calibration. The radiation absorbers were cast i n the form of cubes so that either horizontal or vertical x - ray beams could be used. The inside of the container and the exterior surface of the cubes are silverplated t o minimize radiation transfer of energy. AdditionaUy, a polished plate of aluminum has been suspended between the cubes. A small amount of Teflon for elect r i c a l insulation and the neoprene vacuum gasket are the only non-metallic or non-vitreous materials exposed tothe evacuated volume. ecules. The few electrical connections t o the cubes are of sufficiently thin wire t o recder negligible the amount of energy transfer by conduction. !he detection of the temperature change resulting fram the absorption of energy by one of the cubee use8 four thermistors in B Wheatstone bridge with a high sensitivity galvanometer used 88 a null detector. thermistors of opposing amm of .the bridge (Fig. 4 ) are potted i n Wood’s metal in one cube and the other pair i n the other cube. The thermistors used have a roam temperature resistance of about 105 This coefficient varies appreciably emong various inohms and a negative coefficient of resistance of about 4.7 per cent per dividual thermistors and i s a slow function of temperature. bridge balance has been made which includes, i n addition tothe normal resistance condition fora Wheatstone bridge, a correctinn term representing unbalance derived from the 6ame temperature change i n both blocks. Proper choice of the-mistors the coefficient of temperature change can be chosen analytically on the basis of measurements of red stance versus temperature of many thermistors and the s e t s were teste& experimentally for mhhm unbalance against temperature changes. v i o u s operation vithout such selection, the present detection ayatem is substantially devoid of ~ f t . temperature of the oil bath BUr rOund irt g the evacuated box t o f0.01 O C, An analysis Of t o vanish at least in first order. Sets of four thermi6tora were By ccmrparison with prePrevioUely it ya8 necessary t o control the

Page 10

whereas a 1 C change in o i l bath temperature is now scarcely detectable. Use of the calorimeter involves irradiation of one block for a time necessary to change the temperature of that block by 2 o r 3 x lo-' This unbalmces the prevlous temperature equilibrium so the next heating is done by electrical power dissipated i n the heaters of the other block. Preceding the first irradiation and between each temperature change interval, the n u l l balance is watched carefully to follow the slow drift which tne bridge balance resistor and a recording potentiometer so as to give a permanent record of the balance condition. After several iterations of the above procedure, the calorlmeter is shifted so that the second block may be irradiated and the first block calibrated with its electrical heater. As shown in Fig. 4, a mechanical linkage has been arranged between For analysis of the thermistor resistance as a function of time, Newton's law of cooling is assumed to hold. Thermistor resistances (Ti) are considered to be sufficient indicators of temperature f o r amall changes and the following analysis is made in those terms. Figure 5 illustrates the behavlor with time of the resistance of a thermistor before, during, and after a typical heating cycle. sistance and corresponding time for each point. Te represents the equilibrium resistance of that thermistor for the existing temperature of the surroundings and is assumed constant during the run. Newton's law is mitten so that, starting from any (Ti, ti), For the case of uniform $eating during t2 S t I t,, there appears a constant additive term, T,,

Page 11

FIG. 5 . CHARACTERISTIC HEATING RESPONSE OBTANED WITH THE CALORIMETER BY MAlMAlNlNG

Page 12

Integrating over t i n e between t, and t, gives the whole change The first two terms of equation 6 are identical t o those i n Laughlin‘s equation 20 (1)and the t h i r d term is a correction which is ordinarily quite negligible and arises out of deviation from linearity of the drift r a t e between heating runs. Thw, considering only the first two term, the graphical construction required i s extension of the pre- and postheating drift rates to the mid-time of the heating period. between these extended lines at that time represents mX. The distance The coefficient & can be evaluated directly from the graph 88 in Apparent values of & from several runs vary from 0.2 to 1.0 x 10-f sec‘’. This is interpreted t o indicate a mean heat transfer time constant of about three days. A computation of & f r a rates of energy tranefer for the conditions i n use leads t o a value of 0.15 x lo-’ sec-1. Fluctuating pressure i n the evacuated space and scopevhat tarnished silver surfaces axe thought t o be responsible for the discrepancy. pIating with a non-tarnishing metal such as gold would be superior t o silver in preserving a surface of Ugh reflectivity. It may be that

Page 13

Preliminary Eeasurements of energy flux density from the synchrotron have been made using 70-Mv x-rays. The beam is monitored i n intensity by a transmission ionizatio9 chamber and in total exposure by total activity induced i n magnesium foils through which the beam also passes. activation is then useful to translate the beam energy measured t o any other secondary effect such as ionization i n a cavity chamber immersed in a water tenaixations obtained with the two blocks. Additional runs will be made and will employ field sizes other than that of 5 cm diameter which has been used A t the monent there i s a discrepancy of 2 per cent between the deCorrections fo the apparent energy resulting from absorption and observed temperature change must be made t o allow for energy escaping i n the form of photons, neutrons, and electrons. The electrons a r e secondary t o photons which did not quite escape and, because of their relatively short range, carry only a few per cent of t h e energy t o be found in the leakage tron-electron annihilation, but in either case have an energy of the order Tnese photons are derived mainly from Compton scattering and posiA guarded ionization chamber was built in a polyethylene block and used with at l e a s t 1 inch of polyethylene on all sides, except 0.5 inches behind. This chamber was placed ina lead cave and the unit was directed towsrds a 10-cm lead cube which was substituted for the calorimeter blocks. Leakage radiation was measured by the ionization produced i n the chamber which was due t o the presence of this cube inthe x-ray beam. and aspects were tested, on the basis of which the leakage intensity was integrated over all dlrections in terms of the arbitrary units of response of the ionization current measuring system. These units were then calibrated by exposure on a 1000 kvp …