Document text
Machine-read from the scan, so expect some errors. Lines that couldn’t be read are left out.
Page 2
Unclassified Health ana Biology University of California School of Medicine, San Francisco Department of Radiology for period ending June 33, i956 Atomic Energy Commission Contract No. AT-U-1-GEN-10, Project No. 2
Page 3
IV. Chemical Dosimetry V. Operation of the Synchrotron VI. Activities by Person 11. Influence of Strain on Acute X-Ray Lethality LD,, and Death Rate Studiesin the Moue. 111. Aczte Lethality Studies with the Hamster. LDso, Death Rate, and Recovery Rate Late Effects of Whole-Body X-Irradiation on the Mouse. Correlations Between Changes in Organ Weight and Histological Findings Response in the Dog C UN ICAL SECTION
Page 4
The professional complement of this section is still incomplete. Although most of the present staff are still relatively new t o the synchrotron and related program, they are nevertheless responsible for many of the accomplishments listed below. ter s relating to physics, 16 series of irradiation s were made for the Biology Section and the treatment of patients with cancer has been started,all with the synchrotron. I n addition to them at - As reported earlier, induced activity(in may,esim foils) has been used t o monitor exposures. It was clear that integration of ioniz at isn current from a transmission monitor chamber wcilld be preferable, but development time for a reliable current integrator seemed not t o be available. The Radocon (Victoreen Instrument CompaEy) i s a current integrator (with rate indicatior, also), and one was obtained. for evaluation. Although normally used with a probe, our Radocon was converted for use with the existing monitor chamber. m e input resis - tor was adjusted so that the unit of current in tem ation would be approximately 1 esu/cc atthe peak of the depth-dose curve i n water wher, An ext&ded series ?f irradiation s was then made t o intercompare and evaluate four methods of measuring t h e radiation. methods were essentially indicators of exposure (induced activity and Radocon) while t h e other two were indicators of absorbed dose. These latter were the standard ferrous sulfate dosimetry solution and a Victoreen ratemeter with the rate continuously recorded on a strip-chart; both were located i n a Presdwood phantom (10 cm deep and 2 meters from Two of these Ratios between these indicators were examined f o r constancy. The result then was that none was more reliable than the Radocon, and induced activity has been entirely abandoned. integration unit i s now being standardized daily by exposure of a chemical dosimeter in standard geometry. The size of the Radocon It should be noted that failure t o standardize daily for a part of this report period has ledto an unfortunate position. A change
Page 5
occurred in the Radocon circuit, noticed on one specific day both by indications on meters and by results of irradiation of a biological system, which change prevented the circuit from responding linearly to the pulses of current it was expected to integrate. was corrected at once but the duration of partial non-linearity from this cause can never be known. Before starting the irradiations of biological systems, the calorimeter was used to measure the energy output from the synchrotron for radiation leakage was estimated with a polyethylene chamber in a lead cave which, i n turn, was surrounded with a moderate amount of neutron shielding. Before adding the neutron shielding, the background reading of this chamber was comparable with the scattered radiation measurement being made. As before, the correction The synchrotron output was measured for a bean seasuring 5 cm d i m . at 2 meters from the target, for a beam measuring 3.5 CP diam. at the same distance from the target, and for this same beam but with the target distance increased to 2.85 meters so that the diameter of the beam was 5 m. Supplementary measurements were also made, first, with a large water phantom interposed and, second, with a 10-cm lead cube interposed to imitate the condition obtaining with the calorimeter during the second measurement above. Thirty-two per cent of the incident energy penetrates the 62.4 cm of the water phantom, corresponding to an absorption coefficient of 0.0182 cm-’. This coefficient for water implies a mean photon energy of about 20 MeV, which is quite reasonable for the filtered 70-Mev Bremsstrahlung spectrum. incident energy was transmitted through the lead cube, corresponding to an absorption coefficient of 0.53 cm-’ (mean photon energy - 12 Mev). The ionization current collected in a probe chamber at a large number of points throughout the water phantom has been measured for both the 5-cm and 3.5-cm diameter beams. ments can all be converted i n t o a single arbitrary unit such as per cent of peak ionization. (per cent ionization found in that gram) throughout the whole water phantom gives a mcasure of the energy absorbed. however, must be that part of the incident energy which did not escnpe either in the direct beam or to the sides. was measured with t3e calorimeter and the side leakage, which was small, was estimated by extrapolation of ionization in the phantom, by photographic films around the phantom, and by chamber measurements external to the phantom. For a given beam the measureIntegrating the product (gram of water) times The energy absorbed, The direct beam leakage Because of the density effect, the stopping power of water relative to the air in the ionization chamber depends on the electron
Page 6
side it, then a simple division of the energy absorbed in the whole phantom per monitor unit by the gram x per cent integral would give directly the ergs per gram absorbed in water at the I$ ionization level and per monitor unit. in the beam has been made by numerical methods, with the result that the expected density effect correction to the classical st0ppir.g power amounts to 15%. The estimate made for the spectrum outside the beam is less exact but correction as such is taken to be 5 % . ionization, about lo$ less energy absorption is indicated in the beam relative to outside. This correction has been included in the -am x per cent integral, which inclusion was relatively easy since the inte-. gral was naturally divided into two parts: ionization current per unit volume, the result of the analysis of the data from the 5-cm diem. beam is 0.91 rad per esu/cc at normal tenperature and pressure. If the electron spectrum were the same in the beam and outA reasonably complete estimate of the spectrum Thus, per unit -- the bean and the rest of Using the extrapolation chaQber described below to measure The data for the 7.5-cm dim. be- have not yet Because of uncertainties in interpreting readicgs from condensertype integrating ionization meters due to spurious effects such as insulator conductivity, this section prefers to make all ionization measurements with chambers used as rate devices. Each chamber constructed in this laboratory has an extensive guard electrode placed so that there exists no conduction path from high voltage to collector that is not intercepted by the guard. The monitor chamber, the quadrants associated with the filter-positioning servo system, the extrapolation chamber, probes (even those with 3 m3 volume), and others, all have this feature The extrapolation chamber is taken as the laboratory working standard. Briefly, the construction is as follows: The upper surface is 0.001-in. M y l a r which has been lightly aluminized (100ohms per square). The rear surface (8.8cm diem.) is the face of a 28-rnm thick block of polyethylene. This face was coated with Aquadag and then four concentric grooves (0.010in. wide) were turned so that included areas are about 8.0, 4.8, 2.4, and 0.9 an2. holes enable connections to be made at the bottom of the block to these small areas and the large area outside the largest groove. The supporting structure is of Plexiglas and includes a large nut containing a 1-mm pitch thread for moving the polyethylene block relative to the Mylax sheet. A simple switch containing spring-loaded aluminum contacts enables any of the small areas to be connected to the central electrode of an attached coaxial cable; in each connection, all outside areas form part of the guard electrode and are automatically connected to the braid of that coaxial cable. Another such cable bring6 the high voltage to the aluminized surface of the Mylar sheet, the braid being
Page 7
grounded. Thus, fora given area used for collection, extrapolation t o zero can be achieved by gap adjustments or, f o r a given area, extrapolation can be made by changing the area. One conduction path not seen a t first w a s from the high-voltage surface through the Mylar, then through the air surrounding the whole chamber structure to the switch area a t the bottom. This was sealed w i t h an additional Mylar sheet, the aluminized surface being grounded. To date t h i s chamber has been used only with 70-Mv x-rays and essenti ally at the peak of the depth-dose curve i n Presdwood. Technical difficulties have prevented extensive use i n the water phantom,so other chambers have been u e to relate the peak ionization i n water to that The only known point i n question about use of this chamber i s that the saturation curve does not agree with the predictions of Boag's analysis for parallel - plate chambers ina pulsed radiation beam. There i s remarkable agreement, seemirgly fortuitous, for positive applied voltage (1v olt and higher), but there is a substantial indicated current for zero applied voltage and the current collected for negative high voltage behaves ina way not covered atall by the theory. is not the first chamber i n which this kica of discrepancy has been spite no applied voltage, the cause of which is not clearly understood. However, the saturation value i s the same vith in experimental uncertaint y for either positive o r negative higll voltage. Apparently this discrepancy i s associated with current flow deOne possible correction was investigated briefly : When ionization current is collected, it i s passed through a resistance thus generating a small voltage which is in fact the quantity measured. This voltage, however, appears between the collector and adjacent guard electrode s o thata small rearrangement of field pattern must occur i n such a direction that the actual readihg is a bit less than the calculated geometry should have delivered. Arguing on the basis of superposition of fields, the amount of applicable correction was estimated by applying small voltages between guard and collector with t h e high voltage grounded. The apparent corrections were generally 1% or less, but since they might be more appreciable i f recombination were not present, the practice has now been adopted of setting manually the reference potential of the guard t o be that which i s currently being read on the collector. One comes upon another curious discrepancy by comparing the indication of Victoreen thimbles with, for example, that of the extrapol ation chamber. it might be expected t o read approximately i n esu per cc when placed readings so obtained (corrected t o 760 mm Hg, 22 C)are 10 t o 12% less than extrapolation chamber readings (corrected t o 760 m Hg, 0 Insulator conductivity inthe thimble should reverse the discrepsncy, If a thimble chamber i s used with a Coco calibration,
Page 8
in the extrapolation chamber should prohibit extrapolation to the same zero for the several areas and different gaps. ancy is not understood. A small development effort has been in progress to find chemical systems which would be reliably useful to assess energy absorption from radiation under two sets of circumstances. It is desired to have systems which are more stable than the conventional ferrous sulfate dosimeter and either more sensitive per se or used with a more sensitive measuring system. The two conditions are indicated here by example: (a) An extremely stable solution which can be used over extended periods for routine daily standardization of output from x-ray stable - such that integral dose measurements are precise for a small beam passing through a small container. single solution may not be found to fit both conditions. (b) A high sensitivity solution which is reasonably It is recognized that a Work started with the ferrous sulfate system. Not having a spectrophotometer at the start, a high quality potentiometer was used to capare the single electrode potential of the unirradiated ferrous solution with that of the ferric-ferrous solution after irradiation. Although considerable care is required in the use of this scheme, it is more sensitive than the spectrophotometer. It is not as easy to use as the spectrophotometer, so work is currently concentrated on the latOne method to increase the apparent sensitivity involves the use of a resin ion exchange column to separate ferric ion from a solution containing both ferric and ferrous ions. Successful separations have been made. However, it appeared that the resin in the column promoted some oxidation during elution of the solution. suspended until a routine solution for condition (a) has been found to be satisfactory. Work here has been Our first observations of instability in the ferrous suLfate The potentialdosimeter were made using the potentiometer method. difference, hence ferric ion concentration, increased vith time in the direction of the cell chamber more exposed to air. In order to obtain reproducible results, it was necessary to control conditions rigidly, particularly atmospheric exposure, solution temperature, and method of loading samples. Because it was not clear that all systems proposed for test would be suited to this method and because of relative ease of use, a decision to use the spectrophotometer was made. ber of other metal ions have been exemined but no evidence was found that any would be as useful as ferrous. Addition of other chemicals was then tried, the idea being that a complex ion might be more stable than the ferrous ion alone. by us, but is less stable. To date, a large number of organic acids have been tried with varying degree of success. Thiocyanate has been tried by others and
Page 9
One solution of extreme stability is currently being used for routine output measurements. added, in order, 450 cc conc. H,SO 0.50 p p benzoic acid, and 3.20 gm FeS04. 7HM,0. This solution has a 6'in the neighborhood of 38 when the density cnange is mea w e d at what appears to be a ferric-benzoic acid resonance peak, 2700 %,and is stable over many months. Adding 16 ~JCI oxalic acid to the above recipe increases the G to about 60, but the system undergoes a slow photochemical change which will limit the useful life of a solution to a few months. A working hypothesis has been developed to explain the utility of mainly carboxyl and hydroxyl groups in the complexing agent, and other additives and more data are being accumulated to test this tentative explanation. The result of this investigation may well be a solution which will enable integral dose measurements to be …
The text continues in the original scan (28 pages).