Yoshikazu Kumamoto, Akihiro Shiragai, Yutaka Noda¡¤Tatsuaki Kanai, Yukio Sato and Takeshi Murakami

Keywords: radiation protection, shielding, neutrons, rem-counter, thermoluminescent dosemeter, LET

¡¡To measure neutron dose equivalent, an ordinary Anderson-Braun rem-counter and a modified rem-counter were used. To compensate for the low sensitivity of the former in the neutron energy range above 20 MeV, C.Birattari et al. modified the counter, by introducing a lead energy moderator. This counter has a response similar to the fluence-to-dose equivalent conversion coefficient for neutrons of up to 1000 MeV. The neutrons were produced by bombarding a copper target with He, C, He and Si ions having energies from 100 to 800 MeV¡¿n. The measurements were carried out behind shielding of 380 cm of concrete plus 50 cm of iron. As shown in fig.1, the difference in dose equivalents between the ordinary and modified rem-counters increases with increasing ion energies, indicating a large component of high energy neutrons. Also the dose difference between carbon and neon bombardment is small when the ion energy per nucleon is the same, although their atomic numbers differ.

¡¡For the absorbed dose measurement of organ doses in patients, the characteristic of the thermoluminescent dosemeter, Mg£²SiO£´:Tb., was studied with the irradiation of 60Co gamma-rays and carbon ions of 13.3 keV¡¿¦Ì and 75 keV¡¿¦Ì. As shown in Fig.2, for all radiations, supralinearity is found; the degree decreases with increasing LET. To investigate the radiation damage of the present dosemeter, 5 irradiations of 5 Gy were made with carbon ions of 13 keV¡¿¦Ì and 194 keV¡¿¦Ì. The sensitivity change was measured with the irradiation of 0.1 Gy of 60Co gamma-rays by repeating the irradiation and annealing. No radiation damages are found.

Fig.1 The neutron dose equivalents per h and per 10£¸ particles per second behind the shielding of 380 cm of concrete and 50 cm of iron. Black symbols show measurements with an ordinary Anderson-Braun rem-counter. Open symbols show those with the modified rem-counter.

Keywords : random rough surface, optical reflection, Monte Carlo simulation

¡¡The proposed light reflection model interconnect the mirror-reflection model and the Lambertian reflection model and is useful for simulation studies of light transport.

[Publications]
Tomitani, T.: IEEE Trans. Nucl. Sci., NS-43(3), 1544-1548, 1996.

Fig.1 Time spectra calculated by Monte Carlo simulation of photon transport inside the parallelepiped scintillator.

 The light source is located at the center of the counter.  

The parameters attached to the curves are explained in the text.

3.Measurements of Metabolic Rate of 11C Auto Activity Induced from 12C Beams in Rabbit Thigh Muscle

Keywords: heavy ions, fragmentation, auto activation, PET, metabolism, biological half-life

¡¡The induced activity may be metabolized inside the human body. To image the 11C distribution, metabolism must be stable during the measurements. The induced activity distributes uniformly irrespective of the organ species. The activity induced in blood vessels will rapidly circulate over the whole body, while that stopped inside the tissues will be metabolized relatively slowly. Prior to clinical measurements, we performed the metabolic rate measurements on animals. To this end, the organ must be uniform and large enough to stop 11C beams within it. Rabbits were chosen since they are easy to anesthetize. Thigh muscle was chosen since its size is large enough for the experiments and the identification of its shape and position can be easily done outside the body, while brain was excluded because of its shape and liver was excluded because of the difficulty in identifying its shape and position externally.

¡¡A rabbit was first anesthetized, fixed to a wooden board and then irradiated with 12C beam collimated with a 6 cm thick brass slab having a hole of 2 cm in diameter. The rabbit was transferred from irradiation vault to the PET within 5 minutes which is sufficient enough to cool down positron emitters of short half-lives except 11C. 11C activity distribution was measured in the dynamic mode for 60 minutes. One example of an induced activity distribution is shown in Fig.1 along with the transmission image of the same section for comparison. After regression analysis of the dynamic study data, the biological half-lives were obtained as 68 min and 90 min for 12C irradiation of 10 Gy and 1.3 Gy, respectively. The activity curve inside the region of interest is shown in Fig.2 as a function of time along with the fitted curve and the curve corresponding to the physical half-life of 11C. These metabolic half-lives were longer than the physical half-life of 11C, 20.39 minutes, and confirms the measurement technique.

[Publications]
1)Yoshikawa, K., Tomitani, T., Kanazawa, M.,
 Yoshida, K., Wada, Y.,  Kanai, T., Imai, 
Y., Suhara, T., Kato, H., Koga, M., Kandatsu, 
S., Yoshioka, H. and Tsujii H., J. Nucl. Med.
Technol.  24, 167-168, 1996.

2)Tomitani, T., Kanazawa, M., Yoshikawa, K.,
 Yoshida, K., Wada, Y., Fukumura, A. and Kanai, T.: NIRS-M-116¡¿HIMAC-013, 169-170, 1996.
Fig.1 Left photograph is the central section image of the 11C auto activity generated from 12C beams inside a rabbit thigh muscle.  Beams were spread out in beam direction with a 2 cm thick ridge filter and collimated by a 6 cm thick brass slab with a hole of 2 cm in diameter. Right photograph is the transmission image of the same section indicating the body contour.

Fig.2 Time-activity curve of 11C in the rabbit thigh muscle.

4.An Experiment on Remote Action against Man in Sense Shielding Condition

Keywords: qi-gong, tohate, sense shielding, suggestion, extrasensory

¡¡Fig.1 illustrates the frequency distribution of time differences between the master's acting time and his pupil's response time of the tohates during 16 trials. Time differences of less than 1 s were observed 6 times (one time in each of 6 trials). The probability of the student stepping back during the 3 tohates in one trial within 1 s is 0.11. The probability that the event described above would accidentally occur in 6 trials or more among the 16 trials is 0.0058. This calculation may imply that all tohates do not depend on the master's suggestion and these is some unknown transmission of the master's acting, since the above result is significant for an approximate synchronous timing between acting and response.

¡¡Fig.2 shows an averaged amplitude topograph of ¦Á wave of the EEG of the qigong master at rest. There exists an area of higher amplitudes at the center of the frontal region. It is known that ¦Á wave amplitudes at the center of the frontal region increase, particularly when a subject is concentrating.

[Publications]
Yamamoto, M., Hirasawa, M., Kawano, K., Yasuda, N. and Furukawa, A.: J. Int. Soc. Life Information Science, 14, 97-101, 1996.

Fig.1 Frequency distribution of time differences between acting time and response time.

Fig.2 Averaged amplitude topograph of ¦Á wave of the EEG of the qigong master at rest.

5.An Experiment on Unknown Subconscious Information Transfer with Auditory Brain Evoked Potentials

Keywords: subconsciousness, extrasensory, information transfer, brain evoked potential, P£² peak latency

The subject listened to a sound pulse of around 630 Hz for 50 ms and his electroencephalogram was recorded for 1 s before and after each pulse at his right frontal region by applying the monopolar method with a reference electrode on the right earlobe. Each trial was composed of 4 pulses at intervals of 3 s, and the subject tried to identify one target pulse which had been randomly determined from the 4 pulses by a computer just before the trial. The choice was made without the subject's knowledge. One hundred trials were repeated with the same subject.

The subject was a healthy 50 year old man. Results of his guessed target are shown in Table 1. "p" indicates the probability of occurrence by chance of guesses that could have no less deviation than the table result has. Fig.1 shows a pair of curves of auditory brain evoked potentials that were obtained from averages of 98 targets or 98 non-targets. The magnitudes of the peaks, P£± to P£² are almost the same between the two curves, but differences in latencies of the peaks, P£± and P£² occur on them. The two-sample-test (t-test) was carried out for two sets of P£² peak latency data, one for the targets, the other for the non-targets. The results are shown in Table 2. The "p" indicates the probability of occurrence by chance of sampling which could give a bigger difference than that calculated from the average value for the latency data of the targets and that of the non-targets as shown in the table.

The result of conscious recognition by means of guessing targets was judged not to be significant at a 5¡ó level of significance (one-tailed), which demonstrated that there is no extrasensory recognition in the subject's consciousness. However, the difference of the latencies of the peaks P£² of the auditory brain evoked potential curves calculated between targets and non-targets was judged significant at a 5¡ó level of significance (one-tailed), which demonstrated that there may be an extrasensory recognition in the subject's subconsciousness.

[Publications]
Hirasawa, M. and Yamamoto, M. : J. Int. Soc. Life Information Science, 14, 32-37, 1996.



Fig.1 Auditory brainevoked two potential curves.
Each was obtained as the average of either 98 targets or 98 non-targets.

6.Ion Recombination Loss and Polarity Effect for Very Small Ionization Chambers

Keywords: ionization chamber,saturation,ion recombination, polarity effect

¡¡The polarity effect is usually caused by the presence of extra cavities around the collecting lead of the chamber. For high energy photon radiations, the polarity effect is also caused by flow in or out of Compton electrons on the collecting lead. This largely depends on the volume of the collector and the insulator. Measurements were carried out using 70 MeV proton beams from the AVF cyclotron to check the effect caused by Compton electrons on the polarity effect. Dose rate was approximately 25Gy¡¿min at the entrance plateau of the beam. Irradiations were made with a 6cm¡ß6cm field in air.

¡¡Measurements were made using 10MV X-rays for saturation characteristics and the polarity effect. Irradiations were carried out with a 10cm¡ß10cm field at the peak depth in a phantom and dose rate was 3Gy¡¿min with the pulse repetition rate of 40pps. Ionization charge was measured by Keithley model 617 electrometer for 10MV X-rays and for 60Co gamma rays, and a vibrating reed electrometer, Cary model 401 was used for 70MeV proton beams.

¡¡The polarity effect as a function of applied voltage was evaluated as the charge ratio for vertical irradiation of the 10MV X-ray beam. All the chambers shown that the effect were less than 5¡ó except for one chamber which was larger than 50¡ó. For the hemispherical chambers, irradiations were made for both vertical and horizontal directions. The value of the charge ratio differed between irradiation directions, however, the shape of the curve was quite similar.

¡¡Using high energy proton beams which are not stopped in the collecting electrode, are one of the best sources to check the polarity effect. For relatively high ion collection efficiency, the polarity effect is small for all chambers used. The polarity effect for 60Co gamma ray beam was measured at one point near the saturation voltage. From the results obtained for 60Co gamma ray and those for the proton beam, the effect did not change very much.

Keywords: dosimetry intercomparison, proton beam, ionization chamber

¡¡2: 19 cm depth in water for the proton beam which has a range of 24 cm in water and the spread-out Bragg Peak of 10 cm width in water;

¡¡3: 9 cm depth in water for the proton beam which has a range of 12 cm in water and the spread-out Bragg Peak of 6 cm width in water. ¡¡According to the European proton dosimetry protocol based on Bragg-Gray cavity theory (S. Vynckier, et al. , Radioth. Oncol., 20, 53, 1991; S. Vynckier, et al. , ibid., 32, 174, 1994), each facility individually evaluated the absorbed dose in water using charge collected during the irradiation and the cobalt calibration factor for its ionization chamber.

¡¡Table 1 summarizes the results of this dosimetry comparison. Each value is the ratio of the absorbed dose evaluated by each facility to the mean value. The JINR data were excluded here because its system could not measure collected charge with good accuracy. Good agreement is seen within about ¡Þ 1¡ó variation for most conditions. This means that it is possible to standardize proton dosimetry internationally when the same protocol is adopted.

Keywords: ion source, Penning, electron cyclotron resonance

¡¡The PIGIS is an indirectly heated-cathode type source. The arc power supply is operated in the pulse mode. Very low-duty pulsed operation realizes a very long source lifetime and high intensities for highly charged ions. The long-term stability of the beam intensity depends strongly on the voltage stability of the arc power supply against the varying cathode condition. This situation can be appreciably improved by adding a stabilizing system of the bombardment power instead of filament current. The arc voltage and the current were about 900 V and 4 A, respectively. The vacuum pressure was about 7¡ß10¡Ý£¶ torr in the vacuum chamber; the pressure at the plasma was estimated to be over 10¡Ý£´ torr. The beam intensity was about 300 e¦ÌA. At present, carbon ions are obtained by sputtering a graphite block with a mixture of N£² and Ne gases. Since the sputtering technique reduces the amount of carbon deposition on the cathode insulators, the technique is very effective for extending the source lifetime. Other ions of solid materials can be produced using the same technique.

¡¡The ECRIS has a single closed ECR zone with 10 GHz microwaves. Plasma confinement is ensure by a simple minimum-B magnetic structure with two mirror coil magnets and a sextupole permanent magnet. The maximum axial mirror fields are 9.3 kG and 7.6 kG. The radial sextupole field is 8.0 kG on the chamber wall. An ECRIS is able to easily maintain stable operation. Although the heat-up of the plasma chamber was the cause of fluctuation, it has almost been surmounted by an improvement in the cooling efficiency of the plasma chamber. The microwave power was about 700 W, and the pulse width was 5 ms. The afterglow peak was appeared after the microwave turn-off, but was not used because of the worse stability. The vacuum pressure in the plasma chamber was estimated to be around 2¡ß10¡Ý£¶ torr. The extraction voltage was 24 kV. The beam intensity was about 340 e¦ÌA. The CH£´ gas is used as ionized gas for the production of C£´¡Ü, whereas CO£² gas is usually used for the C£²¡Ü. It seems that the charge-state distribution depends on the amount of H or O ions which those play the role as a 'support gas'.

¡¡All of the devices of the accelerator components are made controllable through a computer system. All of the parameters and measured values can be saved as a parameter file after beam tuning. Under the well-known condition, since the parameter set up can be obtained automatically, it is only needed for an operator to select the parameter file and to turn on the start-up button. The beam intensity varied during the first one hour, because the vacuum pressure in the plasma chamber changed along with the heat-up of the chamber wall. After one hour, it was seemed that the intensity approached the value set by the given parameters without any tuning. A good reproducibility of the beam was obtained. The typical beam performance is summarized in Table 1. The transmission of the injector included the charge-stripping efficiency located at the output end of the linac. The intensity of the sources was reduced for each requirement by a mesh-type attenuater. In a typical weekly schedule of beam time, on Monday, the whole accelerator facility is inspected and maintenance is carried out during the daytime; beam tuning begins in the evening. Patients are treated during the daytime of other weekdays. Every night and on weekend, beam time is available for experiments. The accelerator facility is in operation from the evening of Monday to the night of Saturday, or the morning of Sunday. The fluctuation and drift of the intensity in the long-term stabiliby of the injector beam throughout a week are less than 10¡ó and less than 2¡ó during about 100 hours without any beam tuning.

[Publications]
A. Kitagawa, et al.:Rev. Sci. Instrum. 67(3), 962 (1996)
S. Shibuya et al.:Rev. Sci. Instrum. 67(3), 1171 (1996)

9.The Very Low Ripple Synchrotron Power Supply for HIMAC

Keywords: synchrotron power supply, ripple, harmonic content, common mode, beam spill

¡¡At HIMAC, a level around 0.1 ppm was achieved. This is realized through a detailed analysis of the source of the ripple and spike. The analysis took into account of the effect of the ground current where it is considered in the normal and common mode. For analytic treatment we established a formulation of the mode separation method. There, the string of magnets is treated as a ladder circuit, which is a circuit of series and parallel resonant circuit. The parallel resonance is treated by assuming stray cpacitances between coils and magnet yokes. We found this resonance is suppressed by bridge resistor which is shown in Figure 1.

¡¡Furthermore the ripple and the spike is made to be confined in the power supply system through a path of the voltage source of the grounded thyristor and the common mode low pass filter. Thus thyristor spikes and hige frequency ripples are removed by the static method, namely the bridge resistor and the low pass mode filter.

¡¡The additional source of 50 Hz and ripple were identified. 50Hz source due to a noise of a current sensor was removed. 100Hz source was suppressed by newly developed active filter. The performance is shown in Figure 2.

[Publications]
M. Kumada, PhD thesis, Graduate Univ. of Advanced Study, 1996, also, submitted to particle Accelerator.
Fig.1 Suppression effect of bridge resistor to resonanct ripple of magnetic field 

Fig.2 Relative current ripple without, with HPF and with HPF¡ÜBPFand with BPF. The ripples are normalized by a rated current.