Radiation Protection Dosimetry Advance Access published April 15, 2015 Radiation Protection Dosimetry (2015), pp. 1–7
doi:10.1093/rpd/ncv137
EXPERIMENTAL MICRODOSIMETRY: HISTORY, APPLICATIONS AND RECENT TECHNICAL ADVANCES L. A. Braby* Department of Nuclear Engineering, Texas A&M University, 3133 TAMU, College Station, TX 77802-3133, USA *Corresponding author: [email protected]
INTRODUCTION In the early 1950s, H. H. Rossi and W. Rosenzweig devised a spherical tissue-equivalent proportional counter (TEPC) for the purpose of evaluating the linear energy transfer, L (LET) in unknown radiation fields. Since the L is defined as an average (expectation) quantity at a point, they proposed to operate the proportional counter at low pressure so that the product of the gas density and the diameter of the detector was on the order of 1024 g cm22, the equivalent of 1 micrometre of unit density material(1). The results of measurements of the energy deposited in that small simulated site, when exposed to neutron irradiation, showed much more variation than the distribution of recoil proton path lengths in the sphere. This was a dramatic demonstration of the stochastic nature of energy deposition by charged particles, a phenomenon that was well known from theoretical considerations but had been observed primarily as the qualitative variation in cloud chamber tracks made by identical ionising particles. Dr. Rossi recognised that the range of energy depositions observed in the simulated micrometre volume was characteristic of the energy deposited in critical biological cells and that it was probably significant for initiating biological damage. The study of the stochastic nature of energy deposition in small volumes became known as microdosimetry. EARLY DETECTORS AND THEIR LIMITATIONS The early TEPCs consisted of spherical shells, moulded of A150 tissue-equivalent (TE) plastic, on the order of
10 cm in diameter with 0.5 cm wall thickness. A fine wire, strung from pole to pole of the sphere, was used as the detector anode. In order to produce the constant electric field along the length of the anode wire, a requirement in order to have constant gas avalanche gain for all electrons liberated in the volume, the anode was centred in a cylindrical grid consisting of a wire helix held at an electric potential intermediate between the anode and cathode, Figure 1. These early detectors(2) were made by skilled craftsmen in Dr. Rossi’s laboratory and made available to people interested in microdosimetry research. They worked very well in the laboratory environment but were delicate and complicated to operate. If the helical grid moved relative to the anode wire, the capacitance between grid and anode changed, and because of the potential difference between grid and anode, electrical charge was forced to move. The magnitude of this ‘microphonic’ noise charge could be comparable with the charge generated by an ionising radiation event in the detector. In fact, the charge from a radiation event could be quite small, even with a gas gain of 1000, since it was evident that events that produced as little as a single ionisation in the gas volume were possible. As a result, the electronic noise produced by the preamplifier, which was connected in the traditional way, with high voltage applied to the anode and a coupling capacitor blocking high voltage at the preamp input, limited detector sensitivity to events producing a minimum of several ion pairs in the detector volume. Furthermore, A150 plastic, like most plastics, is permeable to atmospheric gases, including water vapour. Since the formation of the electron avalanche, which yields the proportional counter gain, is sensitive to
# The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
The invention of tissue-equivalent proportional counters simulating micrometre diameter volumes, intended to measure the linear energy transfer of a radiation field, resulted in a practical demonstration of the stochastic nature of energy deposition in small volumes. Besides contributing to a better understanding of the interactions between ionising radiation and biological systems, these detectors have had a significant impact on applied radiation dosimetry. The initial instruments were elegant but suitable only for laboratory experiments because of their sensitivity to environmental conditions and the complex support systems they required. However, their ability to separate the dose due to neutrons from that delivered by photons motivated detector design modifications that eventually resulted in robust detectors suitable for use as radiation survey instruments. Proportional counters simulating micrometre tissue volumes turned out to be the ideal detectors for monitoring the complex radiation environments, including on the space shuttle and International Space Station, and have served as the primary active dosimeters in space for nearly two decades. The need for more sophisticated measurements has led to further improvements in detector design, and the need for smaller and lighter dosimeters is motivating further developments in both detectors and data processing systems.
L. A. BRABY
the atomic composition of the gas, a continuous flow of counting gas was required. Since the detector operated at a fraction of atmospheric pressure, a vacuum pump and a system to precisely control the gas pressure were required.
With appropriate care helical grid, solid-walled TEPCs worked well and valuable measurements of the energy deposition characteristics of different radiations were made. However, it soon became evident that the distribution of energy deposited in a gas-filled cavity in a unit density medium is not identical to the distribution of energy deposited in the equivalent size unit density volume in a uniform medium. The difference is known as the wall effect (3) and, as illustrated in Figure 2, it results in some events that would occur in separate volumes in a uniform medium being combined into a single event in the gas-filled cavity. The result is that measurements made with a solid-walled detector will have a slightly higher average energy deposition, but a smaller number of events, than the same measurement made in a system with uniform density. The wall effect does not change the total energy deposited per mass (the absorbed dose), but it changes the size of individual energy deposition events. The magnitude of the effect was difficult to predict. An experiment using a cylindrical grid-walled proportional counter in an infinite (relative to the range of 14C beta particles) homogeneous medium of 14C-labelled gas and fitted with a removable plastic wall, also labelled with 14C(4), demonstrated that the wall effect could be significant, Figure 3.
Figure 2. Schematic diagrams of events leading to wall effects: (a) delta ray, (b) scattered particle, (c) branching track and (d) multiple interactions by an indirectly ionising particle. In each case, the sketch on the right illustrates what would happen in a uniform medium, and the sketch on the left shows how the substitution of a large-diameter, low-density, detector would result in combining two small events into one larger event (3).
Page 2 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
Figure 1. Cross-sectional view of an early TEPC showing the central anode wire, the helical grid, insulator structure and gas flow fittings(2).
WALL EFFECT AND WALL-LESS DETECTORS
EXPERIMENTAL MICRODOSIMETRY
Figure 3. Energy deposition distribution produced in a gridwalled cylindrical detector by 14C beta particles when the grid is surrounded by a solid 14C-labelled plastic wall or is in a uniform 14C-labelled gas medium(4).
A different approach to a grid-walled detector was used to measure f( y) for typical photon spectra and the effect of the size of the TE plastic vacuum chamber relative to the detector diameter(7). The detector consisted of a set of wire rings, in planes perpendicular to the anode, outlining a spherical volume. Each ring was held at the electric potential necessary to maintain an approximately constant electric field along the anode, Figure 6. The results of measurements with this detector showed that for typical photon irradiation, a vacuum chamber diameter approximately six times the detector diameter was sufficient to minimise the wall effect. For low-energy photons and lowvelocity, heavy-charged particles, where the range of the primary charged particles and delta rays, respectively, is not very much larger than the diameter of the site of interest, the wall effect is very small and wallless detectors are not needed. However, for high-energy protons and heavy ions, such as galactic cosmic rays, the range of the delta rays can be much more than the diameter of the site of interest (8, 9). In this situation, the frequency mean of y in a uniform medium is significantly less than L. However, in this situation, the wall effect, specifically the entry of multiple delta rays produced in the wall, increases y measured in a solid-walled detector(10, 11). Thus, the measured mean lineal energy for cosmic rays approaches the L of the incident radiation. The large size of the vacuum chamber, needed for a grid-walled detector to be effective, is incompatible with some applications of microdosimetry. In those situations, solid-walled detectors continue to be used, and the discrepancy between measured values and those that would occur in a uniform medium is considered in the application of the results.
Figure 4. The electric field lines between two small spheres, when cut off by a properly placed guard cup, define a spherical ion collection volume. If the sphere radius is small enough, the electric field at the surface will produce gas gain and an essentially wall-less proportional counter results. The energy deposition produced by alpha particles and their delta rays, as measured by such a detector, is displayed at the right (5).
Page 3 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
A wide variety of wall-less and grid-walled detectors were developed and used for specific measurements. For example, one using the electric field lines between two small spheres, with those lines falling outside a spherical volume cut off by a guard electrode, Figure 4, was used to determine the contribution of delta rays to the lineal energy distribution produced by 4.5 MeV alpha particles(5). However, alignment of the guard electrode of such a detector is tedious, and most measurements have been made with some form of grid-walled detector. For example, a detector with an A150 plastic mesh cathode and a helical grid around the anode, housed in a largediameter TE plastic vacuum chamber, Figure 5, was used to measure f ( y) for monoenergetic photons(6).
L. A. BRABY
Figure 6. Plan of the right half of a wire ring-walled detector and the results obtained for a variety of photon spectra as a function of the ratio of vacuum chamber diameter to detector diameter(7).
DETECTORS FOR OPERATIONAL MICRODOSIMETRY Early solid-walled and wall-less detectors were used successfully for a wide range of measurements that could be made under controlled conditions(6, 12). The characteristics of energy deposition by delta rays(9, 10),
f(y) for many different radiations(6, 13, 14), the impact of site size and shape on f(y)(15, 16), f(z) as a function of absorbed dose and f(y) in and near radiation therapy beams(13, 17, 18) were all investigated with various versions of these detectors. However, it was clear that the information provided by a TEPC could be very useful for radiation protection in complex
Page 4 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
Figure 5. Cutaway diagram of the grid-walled proportional counter inside a TE plastic sphere and the energy deposition spectra of monoenergetic photons as measured by that detector(6).
EXPERIMENTAL MICRODOSIMETRY
radiation fields, and it was equally clear that the early detector designs would not perform satisfactorily in operational environments. What was needed for operational purposes was an entire dosimetry system that was small, light, required little electrical power, was simple to operate, stable for several years without the need for gas flow and rugged. A variety of changes needed to be made to satisfy these requirements. Gas gain stability
Electronic and microphonic noise The electronic noise produced by the conventional preamplifier connected to the detector through a high voltage blocking capacitor required that early detectors be operated just below their breakdown voltage in order to achieve enough gas gain to resolve the smaller energy deposition events. Significant improvements in detector reliability were obtained by lowering the potential difference between anode and cathode, but this required lowering the electronic noise. The reduction in noise was achieved by directly
Figure 7. Schematic drawing of a grid-walled cylindrical detector with field tubes(4). The field tubes are held at an electrical potential that maintains constant electric field strength at the surface of the anode and defines the ends of the detector volume, resulting in a detector active length equal to the cathode inner diameter. Solid-walled detectors use the same design by simply replacing the grid wall with a TE plastic cylinder with the same inside diameter.
Page 5 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
It soon became evident that two primary factors resulted in the need for gas flow to maintain high detector gain. One was the leakage of the ambient atmosphere (containing oxygen, water vapour, etc.) into the detector, and the other was release of vapours, including water and plasticising agents, from the materials used in the construction of the detectors. The first step in solving this problem was to mount the TE plastic detector inside a metal vacuum chamber. Stainless steel beakers, manufactured for chemistry lab applications, were an early choice for this purpose. In some cases, the vacuum chamber was welded closed, but that made it difficult to repair the detector. In other cases, the chamber was sealed with an O-ring, or later, a metal-to-metal seal such as an indium gasket. These detectors were provided with a fill tube, closed by a valve or by crimping the tube, which allowed them to be evacuated, filled with counting gas, and used for several days without significant loss of gain or pulse height resolution. As sealing techniques (choice of electrical feedthroughs, gasket materials, etc.) improved, the useful life of a gas fill increased, but it became evident that eliminating all leaks would not extend it to a year. The next step was to eliminate, to the extent possible, the vapours being emitted by detector components, primarily the TE plastic. One option was to bake A150 in a vacuum oven for a few days before it was moulded into detector parts. This was effective, but it turned out to be easier to assemble the detector, and then hold it at vacuum (on the order of 1026 torr) and elevated temperature for 30 d. This procedure is now used by NASA and results in detectors that maintain their gain for 3 y or more after filling with counting gas.
coupling a very low noise charge-sensitive preamplifier to the detector anode. This required operating the anode at essentially ground potential, and applying negative high voltage to the cathode, which was acceptable because these detectors were housed inside a metal vacuum chamber that was at ground potential. Besides reducing the electronic noise by reducing stray capacitance to ground, this configuration eliminated some of the microphonic noise that had apparently been generated by the blocking capacitor and greatly reduced the risk to the preamplifier input when the applied voltage changed suddenly (due, for example, to a high-voltage power supply problem) by reducing the charge that was forced to move when the potential difference changed. The majority of the microphonic noise from detectors with a helical grid was produced by motion of the grid. In order to be of use outside the laboratory, this noise had to be eliminated, either by eliminating the grid or by replacing it with one that had minimal motion due to vibration. In a cylindrical detector, uniform gas gain can be achieved by use of field tubes at the ends of the anode, Figure 7. NASA used solidwalled cylindrical chambers with field tubes for dosimetry on the space shuttle and ISS through 2011. However, the need for a larger vacuum chamber due to the dead volume at each end of the chamber, the relatively complex chord length distribution of a cylinder and the non-uniform response as a function of the angle between the charged particle track and the detector anode eventually outweighed the advantages
L. A. BRABY
of the simplicity of the cylindrical detector design. Benjamine et al.(19) proposed a design for a spherical detector using field shaping electrodes at the ends of the anode wire to approximate uniform gain along the length of the anode. A single-wire proportional counter, using a similar approach, is commercially available. However, the field shaping electrodes result in a dead volume at each pole of the detector. Another approach(20) was recently adopted by NASA. The latest dosimeter for the ISS utilises solid-walled detectors based on the design of the grid-walled detector illustrated in Figure 6. The cathode consists of TE plastic rings, moulded to the contours of the sphere and laminated with polyethylene insulators, Figure 8.
Dosimeter size Very small TEPCs have been built for a variety of purposes. Rollet et al.(18) developed a 1 mm` 1 mm cylindrical detector for proton beam dosimetry, and Moro et al.(21) developed twin 1-mm chambers with and without boron for boron neutron capture therapy dosimetry. These detectors were intended to be used in high-dose-rate environments and with conventional laboratory electronics. To build a small dosimeter, one needs small electronics. To collect pulse height data and calculate the dose and mean lineal energy, one needs a
Page 6 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
Figure 8. Laminated cathode spherical detector with 3.8 cm inside diameter, mounted to circuit board with voltage divider and preamplifier and attached to vacuum chamber base.
multichannel analyser or high-speed digitiser with a computer to detect the pulse peaks, and additional computer power to calculate the absorbed dose and radiation quality. If one accepts the argument that the radiation quality is related to the dose mean lineal energy, there is an alternative approach, referred to as the variance method. If one measures the charge produced by an ion chamber or proportional counter in defined time intervals, the average charge per time is proportional to the absorbed dose rate and the variance in the charge measurements is proportional to the dose mean lineal energy(22). Since the charge-sensitive preamplifier required for pulse height analysis can easily be converted to a charge integrator by replacing the feedback resistor with a switch, the electronics to analyse dose and radiation quality using the variance technique is inherently simpler than the electronics for pulse height analysis. As electronic components become more compact, the electronics for the variance system will remain smaller and require less power than electronics for pulse height-based systems. A prototype instrument utilising a single detector to determine radiation quality in situations where the dose rate changes was recently developed for NASA. For practical radiation protection purposes, it is not productive to make the instrument smaller by simply making the proportional counter smaller. In a given radiation field, the count rate is proportional to the area of the detector. Since the dose, and especially the dose equivalent, is strongly dependent on the relatively rare large energy deposition events, an accurate measure of the dose requires good counting statistics for those rare large events. As the diameter of a detector is decreased, the time required to obtain good statistics in the same radiation field increases as the reciprocal of the square of the diameter. Consequently, as miniaturisation of electronics progresses, the detector becomes a progressively larger fraction of the volume and mass of the dosimeter. However, a large number of small detectors with the same surface area as one large detector will fit in a significantly smaller volume. Thus, the use of arrays of small detectors(2, 23) offers a way to further miniaturise the dosimeter. Unfortunately, it has proven difficult to achieve good resolution and low electronic noise in detector arrays, so their application has typically been limited to high LET radiations. Semiconductor detectors have long been considered as potentially more reliable replacements for gas-filled detectors for microdosimetry(24). They would, of course, function as arrays of small detectors, and one of the challenges in building a practical solid-state dosimeter system is to produce enough small detectors to achieve the total area needed to produce good statistics in most radiation environments. Recent developments in superconducting devices suggest that quantum interference devices (SQUIDs) may also be useful as microdosimeters(25).
EXPERIMENTAL MICRODOSIMETRY
FUTURE DIRECTIONS 11. 12. 13.
14.
15.
16.
17.
REFERENCES
18.
1. Rossi, H. H. and Rosenzweig, W. Measurements of neutron dose as a function of linear energy transfer. Radiat. Res. 2, 417–425 (1955). 2. Rossi, H. H. and Zaider, M. Microdosimetry and Its Applications. (Springer-Verlag) (1996) ISBN3-540-58541-9. 3. ICRU. Microdosimetry (International Commission on Radiation Units and Measurements) (1983). 4. Braby, L. A., Roesch, W. C. and Glass, W. A. Energy deposition spectra of 14C beta radiation in a uniform medium. Radiat. Res. 43, 499–503 (1970). 5. Glass, W. A. and Braby, L. A. A wall-less detector for measuring energy deposition spectra. Radiat. Res. 39, 230–240 (1969). 6. Kliauga, P. and Dvorak, R. Microdosimetric measurements of ionization by monoenergetic photons. Radiat. Res. 73, 1– 20 (1978). 7. Braby, L. A. and Ellett, W. H. Ionization in solid- and grid-walled detectors. Radiat. Res. 51, 569– 580 (1972). 8. Chatterjee, A. and Schaefer, H. J. Microdosimetric structure of heavy ion tracks in tissue. Radiat. Environ. Biophys. 13, 215–227 (1976). 9. Metting, N. F., Rossi, H. H., Braby, L. A., Kliauga, P. J., Howard, J., Zaider, M., Schimmerling, W., Wong, M. and Rapkin, M. Microdosimetry near the trajectory of highenergy heavy ions. Radiat. Res. 116, 183–195 (1988). 10. Taddei, J. P., Zhao, Z. and Borak, T. B. A comparison of the measured responses of a tissue-equivalent proportional counter to high energy heavy (HZE) particles and those
19. 20. 21.
22.
23. 24. 25.
Page 7 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
The need for high-quality radiation measurements in complex radiation fields and difficult physical environments will continue to drive the development of microdosimetry detectors and their associated electronics. It is likely that microdosimetry will continue to provide important information in support of space exploration, industrial applications of ionising radiation and medicine. Technological developments in many different fields may contribute to new and innovative approaches to dosimeter design. Development of organic semiconductors may lead to improved solidstate dosimeters. Progress in semiconductor manufacturing and reduced feature size may make it possible to build solid-state microdosimetry detectors with improved chord length distributions and site sizes suitable for use with a wide range of radiations. Three-dimensional printing techniques, which have already been used to aid the design of new detectors, may progress to the point where detectors can be manufactured by 3D printing using TE plastic and suitable insulating material. Developments that have not yet been imagined may make it possible to solve persistent problems in radiation dosimetry. The investigator just needs to be observant of changes in a wide variety of fields, and not limit the approach to established methods when attempting to solve problems.
simulated using the Geant4 Monte Carlo code. Radiat. Meas. 43, 1498–1505 (2008). Nikjoo, H., Khvostunov, I. K. and Cucinotta, F. A. The response of (TEPC) proportional counters to heavy ions. Radiat. Res. 157, 435– 445 (2002). Ellett, W. H. and Braby, L. A. The microdosimetry of 250 kVP and 65 kVp x-rays, Co gamma-rays, and tritium beta particles. Radiat. Res. 51, 229–243 (1972). Loncol, T., Cosgrove, V., Denis, J. M., Gueulette, J., Mazal, A., Menzel, H. G., Pihet, P. and Sabattier, R. Radiobiological effectiveness of radiation beams with broad LET spectra: Microdosimetric analysis using biological weighting functions. Radiat. Prot. Dosim. 52, 347–352 (1994). Amols, H. I., Bradbiru, J. N., Dicello, J. F., Helland, J. F., Kligerman, M. M., Lane, T. G., Paciotti, M. A., Foeder, D. L. and Schillaci, M. E. Dose outside the treatment volume for irradiation with negative pions. Phys. Med. Biol. 23, 385– 396 (1978). Hogeweg, B. and Barendsen, G. W. Local energy distributions for a-particles of different energies in relation to the evaluation of critical sizes and energy requirements involved in the induction of damage in mammalian cells. In: Third Symposium on Microdosimetry. Ebert, H. G., Ed. (Commission of the European Communities) pp. 857–872 (1972). Srdoc, D., Goodman, L. J., Marino, S. A., Mills, R. E., Zaider, M. and Rossi, H. H. Microdosimetry of monoenergetic neutron radiation. In: Seventh Symposium on Microdosimetry. Booz, J., Ebert, H. and Hartfiel, H., Eds. (Harwood Academic Publishers) pp. 765–774 (1981). Dicello, J. F. et al. Microdosimetric comparison of scanned and conventional proton beams used in radiation therapy. Radiat. Prot. Dosim. 143, 513–518 (2011). Rollet, S., Colautti, P., Grosswendt, B., Herault, J., Wind, M., Gargioni, E., Beck, P., Lattocha, M. and Moro, D. Microdosimetric assessment of the radiation quality of a therapeutic proton beam: comparison between numerical simulation and experimental measurements. Radiat. Prot. Dosim. 143, 445– 449 (2011). Benjamin, P. W., Kemshall, C. D. and Redfearn, J. A high resolution spherical proportional counter. Nucl. Instrum. Methods 59, 77–85 (1968). Perez-Nunez, D. and Braby, L. A. Replacement tissueequivalent proportional counter for the international space station. Radat. Prot. Dosim. 143, 394– 397 (2011). Moro, D., Colautti, P., Gualdrini, G., Masi, M., Conte, V., De Nardo, L. and Tornielli, G. Two miniaturized TEPCs in a single detector for BNCT microdosimetry. Radiat. Prot. Dosim. 122, 396– 400 (2006). Bengtsson, L. G. Assessment of dose equivalent from fluctuations of energy deposition. In: Second Symposium on Microdosimetry. Ebert, H. G., Ed., (Commission of the European Communities) pp. 375–400 (1969). Waker, A. J., Aslam and Lori, J. Design of a multielement TEPC for neutron monitoring. Radiat. Prot. Dosim. 143, 463– 466 (2011). Agosteo, S. and Pola, A. Silicon Microdosimetry. Radiat. Prot. Dosim. 143, 409– 415 (2011). Galer, S., Hao, L., Gallop, J., Palmans, H., Kirkby, K. and Nisbet, A. Design concept for a novel SQUID-based microdosimeters. Radiat. Prot. Dosim. 143, 427– 431 (2011).
doi:10.1093/rpd/ncv137
EXPERIMENTAL MICRODOSIMETRY: HISTORY, APPLICATIONS AND RECENT TECHNICAL ADVANCES L. A. Braby* Department of Nuclear Engineering, Texas A&M University, 3133 TAMU, College Station, TX 77802-3133, USA *Corresponding author: [email protected]
INTRODUCTION In the early 1950s, H. H. Rossi and W. Rosenzweig devised a spherical tissue-equivalent proportional counter (TEPC) for the purpose of evaluating the linear energy transfer, L (LET) in unknown radiation fields. Since the L is defined as an average (expectation) quantity at a point, they proposed to operate the proportional counter at low pressure so that the product of the gas density and the diameter of the detector was on the order of 1024 g cm22, the equivalent of 1 micrometre of unit density material(1). The results of measurements of the energy deposited in that small simulated site, when exposed to neutron irradiation, showed much more variation than the distribution of recoil proton path lengths in the sphere. This was a dramatic demonstration of the stochastic nature of energy deposition by charged particles, a phenomenon that was well known from theoretical considerations but had been observed primarily as the qualitative variation in cloud chamber tracks made by identical ionising particles. Dr. Rossi recognised that the range of energy depositions observed in the simulated micrometre volume was characteristic of the energy deposited in critical biological cells and that it was probably significant for initiating biological damage. The study of the stochastic nature of energy deposition in small volumes became known as microdosimetry. EARLY DETECTORS AND THEIR LIMITATIONS The early TEPCs consisted of spherical shells, moulded of A150 tissue-equivalent (TE) plastic, on the order of
10 cm in diameter with 0.5 cm wall thickness. A fine wire, strung from pole to pole of the sphere, was used as the detector anode. In order to produce the constant electric field along the length of the anode wire, a requirement in order to have constant gas avalanche gain for all electrons liberated in the volume, the anode was centred in a cylindrical grid consisting of a wire helix held at an electric potential intermediate between the anode and cathode, Figure 1. These early detectors(2) were made by skilled craftsmen in Dr. Rossi’s laboratory and made available to people interested in microdosimetry research. They worked very well in the laboratory environment but were delicate and complicated to operate. If the helical grid moved relative to the anode wire, the capacitance between grid and anode changed, and because of the potential difference between grid and anode, electrical charge was forced to move. The magnitude of this ‘microphonic’ noise charge could be comparable with the charge generated by an ionising radiation event in the detector. In fact, the charge from a radiation event could be quite small, even with a gas gain of 1000, since it was evident that events that produced as little as a single ionisation in the gas volume were possible. As a result, the electronic noise produced by the preamplifier, which was connected in the traditional way, with high voltage applied to the anode and a coupling capacitor blocking high voltage at the preamp input, limited detector sensitivity to events producing a minimum of several ion pairs in the detector volume. Furthermore, A150 plastic, like most plastics, is permeable to atmospheric gases, including water vapour. Since the formation of the electron avalanche, which yields the proportional counter gain, is sensitive to
# The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
The invention of tissue-equivalent proportional counters simulating micrometre diameter volumes, intended to measure the linear energy transfer of a radiation field, resulted in a practical demonstration of the stochastic nature of energy deposition in small volumes. Besides contributing to a better understanding of the interactions between ionising radiation and biological systems, these detectors have had a significant impact on applied radiation dosimetry. The initial instruments were elegant but suitable only for laboratory experiments because of their sensitivity to environmental conditions and the complex support systems they required. However, their ability to separate the dose due to neutrons from that delivered by photons motivated detector design modifications that eventually resulted in robust detectors suitable for use as radiation survey instruments. Proportional counters simulating micrometre tissue volumes turned out to be the ideal detectors for monitoring the complex radiation environments, including on the space shuttle and International Space Station, and have served as the primary active dosimeters in space for nearly two decades. The need for more sophisticated measurements has led to further improvements in detector design, and the need for smaller and lighter dosimeters is motivating further developments in both detectors and data processing systems.
L. A. BRABY
the atomic composition of the gas, a continuous flow of counting gas was required. Since the detector operated at a fraction of atmospheric pressure, a vacuum pump and a system to precisely control the gas pressure were required.
With appropriate care helical grid, solid-walled TEPCs worked well and valuable measurements of the energy deposition characteristics of different radiations were made. However, it soon became evident that the distribution of energy deposited in a gas-filled cavity in a unit density medium is not identical to the distribution of energy deposited in the equivalent size unit density volume in a uniform medium. The difference is known as the wall effect (3) and, as illustrated in Figure 2, it results in some events that would occur in separate volumes in a uniform medium being combined into a single event in the gas-filled cavity. The result is that measurements made with a solid-walled detector will have a slightly higher average energy deposition, but a smaller number of events, than the same measurement made in a system with uniform density. The wall effect does not change the total energy deposited per mass (the absorbed dose), but it changes the size of individual energy deposition events. The magnitude of the effect was difficult to predict. An experiment using a cylindrical grid-walled proportional counter in an infinite (relative to the range of 14C beta particles) homogeneous medium of 14C-labelled gas and fitted with a removable plastic wall, also labelled with 14C(4), demonstrated that the wall effect could be significant, Figure 3.
Figure 2. Schematic diagrams of events leading to wall effects: (a) delta ray, (b) scattered particle, (c) branching track and (d) multiple interactions by an indirectly ionising particle. In each case, the sketch on the right illustrates what would happen in a uniform medium, and the sketch on the left shows how the substitution of a large-diameter, low-density, detector would result in combining two small events into one larger event (3).
Page 2 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
Figure 1. Cross-sectional view of an early TEPC showing the central anode wire, the helical grid, insulator structure and gas flow fittings(2).
WALL EFFECT AND WALL-LESS DETECTORS
EXPERIMENTAL MICRODOSIMETRY
Figure 3. Energy deposition distribution produced in a gridwalled cylindrical detector by 14C beta particles when the grid is surrounded by a solid 14C-labelled plastic wall or is in a uniform 14C-labelled gas medium(4).
A different approach to a grid-walled detector was used to measure f( y) for typical photon spectra and the effect of the size of the TE plastic vacuum chamber relative to the detector diameter(7). The detector consisted of a set of wire rings, in planes perpendicular to the anode, outlining a spherical volume. Each ring was held at the electric potential necessary to maintain an approximately constant electric field along the anode, Figure 6. The results of measurements with this detector showed that for typical photon irradiation, a vacuum chamber diameter approximately six times the detector diameter was sufficient to minimise the wall effect. For low-energy photons and lowvelocity, heavy-charged particles, where the range of the primary charged particles and delta rays, respectively, is not very much larger than the diameter of the site of interest, the wall effect is very small and wallless detectors are not needed. However, for high-energy protons and heavy ions, such as galactic cosmic rays, the range of the delta rays can be much more than the diameter of the site of interest (8, 9). In this situation, the frequency mean of y in a uniform medium is significantly less than L. However, in this situation, the wall effect, specifically the entry of multiple delta rays produced in the wall, increases y measured in a solid-walled detector(10, 11). Thus, the measured mean lineal energy for cosmic rays approaches the L of the incident radiation. The large size of the vacuum chamber, needed for a grid-walled detector to be effective, is incompatible with some applications of microdosimetry. In those situations, solid-walled detectors continue to be used, and the discrepancy between measured values and those that would occur in a uniform medium is considered in the application of the results.
Figure 4. The electric field lines between two small spheres, when cut off by a properly placed guard cup, define a spherical ion collection volume. If the sphere radius is small enough, the electric field at the surface will produce gas gain and an essentially wall-less proportional counter results. The energy deposition produced by alpha particles and their delta rays, as measured by such a detector, is displayed at the right (5).
Page 3 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
A wide variety of wall-less and grid-walled detectors were developed and used for specific measurements. For example, one using the electric field lines between two small spheres, with those lines falling outside a spherical volume cut off by a guard electrode, Figure 4, was used to determine the contribution of delta rays to the lineal energy distribution produced by 4.5 MeV alpha particles(5). However, alignment of the guard electrode of such a detector is tedious, and most measurements have been made with some form of grid-walled detector. For example, a detector with an A150 plastic mesh cathode and a helical grid around the anode, housed in a largediameter TE plastic vacuum chamber, Figure 5, was used to measure f ( y) for monoenergetic photons(6).
L. A. BRABY
Figure 6. Plan of the right half of a wire ring-walled detector and the results obtained for a variety of photon spectra as a function of the ratio of vacuum chamber diameter to detector diameter(7).
DETECTORS FOR OPERATIONAL MICRODOSIMETRY Early solid-walled and wall-less detectors were used successfully for a wide range of measurements that could be made under controlled conditions(6, 12). The characteristics of energy deposition by delta rays(9, 10),
f(y) for many different radiations(6, 13, 14), the impact of site size and shape on f(y)(15, 16), f(z) as a function of absorbed dose and f(y) in and near radiation therapy beams(13, 17, 18) were all investigated with various versions of these detectors. However, it was clear that the information provided by a TEPC could be very useful for radiation protection in complex
Page 4 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
Figure 5. Cutaway diagram of the grid-walled proportional counter inside a TE plastic sphere and the energy deposition spectra of monoenergetic photons as measured by that detector(6).
EXPERIMENTAL MICRODOSIMETRY
radiation fields, and it was equally clear that the early detector designs would not perform satisfactorily in operational environments. What was needed for operational purposes was an entire dosimetry system that was small, light, required little electrical power, was simple to operate, stable for several years without the need for gas flow and rugged. A variety of changes needed to be made to satisfy these requirements. Gas gain stability
Electronic and microphonic noise The electronic noise produced by the conventional preamplifier connected to the detector through a high voltage blocking capacitor required that early detectors be operated just below their breakdown voltage in order to achieve enough gas gain to resolve the smaller energy deposition events. Significant improvements in detector reliability were obtained by lowering the potential difference between anode and cathode, but this required lowering the electronic noise. The reduction in noise was achieved by directly
Figure 7. Schematic drawing of a grid-walled cylindrical detector with field tubes(4). The field tubes are held at an electrical potential that maintains constant electric field strength at the surface of the anode and defines the ends of the detector volume, resulting in a detector active length equal to the cathode inner diameter. Solid-walled detectors use the same design by simply replacing the grid wall with a TE plastic cylinder with the same inside diameter.
Page 5 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
It soon became evident that two primary factors resulted in the need for gas flow to maintain high detector gain. One was the leakage of the ambient atmosphere (containing oxygen, water vapour, etc.) into the detector, and the other was release of vapours, including water and plasticising agents, from the materials used in the construction of the detectors. The first step in solving this problem was to mount the TE plastic detector inside a metal vacuum chamber. Stainless steel beakers, manufactured for chemistry lab applications, were an early choice for this purpose. In some cases, the vacuum chamber was welded closed, but that made it difficult to repair the detector. In other cases, the chamber was sealed with an O-ring, or later, a metal-to-metal seal such as an indium gasket. These detectors were provided with a fill tube, closed by a valve or by crimping the tube, which allowed them to be evacuated, filled with counting gas, and used for several days without significant loss of gain or pulse height resolution. As sealing techniques (choice of electrical feedthroughs, gasket materials, etc.) improved, the useful life of a gas fill increased, but it became evident that eliminating all leaks would not extend it to a year. The next step was to eliminate, to the extent possible, the vapours being emitted by detector components, primarily the TE plastic. One option was to bake A150 in a vacuum oven for a few days before it was moulded into detector parts. This was effective, but it turned out to be easier to assemble the detector, and then hold it at vacuum (on the order of 1026 torr) and elevated temperature for 30 d. This procedure is now used by NASA and results in detectors that maintain their gain for 3 y or more after filling with counting gas.
coupling a very low noise charge-sensitive preamplifier to the detector anode. This required operating the anode at essentially ground potential, and applying negative high voltage to the cathode, which was acceptable because these detectors were housed inside a metal vacuum chamber that was at ground potential. Besides reducing the electronic noise by reducing stray capacitance to ground, this configuration eliminated some of the microphonic noise that had apparently been generated by the blocking capacitor and greatly reduced the risk to the preamplifier input when the applied voltage changed suddenly (due, for example, to a high-voltage power supply problem) by reducing the charge that was forced to move when the potential difference changed. The majority of the microphonic noise from detectors with a helical grid was produced by motion of the grid. In order to be of use outside the laboratory, this noise had to be eliminated, either by eliminating the grid or by replacing it with one that had minimal motion due to vibration. In a cylindrical detector, uniform gas gain can be achieved by use of field tubes at the ends of the anode, Figure 7. NASA used solidwalled cylindrical chambers with field tubes for dosimetry on the space shuttle and ISS through 2011. However, the need for a larger vacuum chamber due to the dead volume at each end of the chamber, the relatively complex chord length distribution of a cylinder and the non-uniform response as a function of the angle between the charged particle track and the detector anode eventually outweighed the advantages
L. A. BRABY
of the simplicity of the cylindrical detector design. Benjamine et al.(19) proposed a design for a spherical detector using field shaping electrodes at the ends of the anode wire to approximate uniform gain along the length of the anode. A single-wire proportional counter, using a similar approach, is commercially available. However, the field shaping electrodes result in a dead volume at each pole of the detector. Another approach(20) was recently adopted by NASA. The latest dosimeter for the ISS utilises solid-walled detectors based on the design of the grid-walled detector illustrated in Figure 6. The cathode consists of TE plastic rings, moulded to the contours of the sphere and laminated with polyethylene insulators, Figure 8.
Dosimeter size Very small TEPCs have been built for a variety of purposes. Rollet et al.(18) developed a 1 mm` 1 mm cylindrical detector for proton beam dosimetry, and Moro et al.(21) developed twin 1-mm chambers with and without boron for boron neutron capture therapy dosimetry. These detectors were intended to be used in high-dose-rate environments and with conventional laboratory electronics. To build a small dosimeter, one needs small electronics. To collect pulse height data and calculate the dose and mean lineal energy, one needs a
Page 6 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
Figure 8. Laminated cathode spherical detector with 3.8 cm inside diameter, mounted to circuit board with voltage divider and preamplifier and attached to vacuum chamber base.
multichannel analyser or high-speed digitiser with a computer to detect the pulse peaks, and additional computer power to calculate the absorbed dose and radiation quality. If one accepts the argument that the radiation quality is related to the dose mean lineal energy, there is an alternative approach, referred to as the variance method. If one measures the charge produced by an ion chamber or proportional counter in defined time intervals, the average charge per time is proportional to the absorbed dose rate and the variance in the charge measurements is proportional to the dose mean lineal energy(22). Since the charge-sensitive preamplifier required for pulse height analysis can easily be converted to a charge integrator by replacing the feedback resistor with a switch, the electronics to analyse dose and radiation quality using the variance technique is inherently simpler than the electronics for pulse height analysis. As electronic components become more compact, the electronics for the variance system will remain smaller and require less power than electronics for pulse height-based systems. A prototype instrument utilising a single detector to determine radiation quality in situations where the dose rate changes was recently developed for NASA. For practical radiation protection purposes, it is not productive to make the instrument smaller by simply making the proportional counter smaller. In a given radiation field, the count rate is proportional to the area of the detector. Since the dose, and especially the dose equivalent, is strongly dependent on the relatively rare large energy deposition events, an accurate measure of the dose requires good counting statistics for those rare large events. As the diameter of a detector is decreased, the time required to obtain good statistics in the same radiation field increases as the reciprocal of the square of the diameter. Consequently, as miniaturisation of electronics progresses, the detector becomes a progressively larger fraction of the volume and mass of the dosimeter. However, a large number of small detectors with the same surface area as one large detector will fit in a significantly smaller volume. Thus, the use of arrays of small detectors(2, 23) offers a way to further miniaturise the dosimeter. Unfortunately, it has proven difficult to achieve good resolution and low electronic noise in detector arrays, so their application has typically been limited to high LET radiations. Semiconductor detectors have long been considered as potentially more reliable replacements for gas-filled detectors for microdosimetry(24). They would, of course, function as arrays of small detectors, and one of the challenges in building a practical solid-state dosimeter system is to produce enough small detectors to achieve the total area needed to produce good statistics in most radiation environments. Recent developments in superconducting devices suggest that quantum interference devices (SQUIDs) may also be useful as microdosimeters(25).
EXPERIMENTAL MICRODOSIMETRY
FUTURE DIRECTIONS 11. 12. 13.
14.
15.
16.
17.
REFERENCES
18.
1. Rossi, H. H. and Rosenzweig, W. Measurements of neutron dose as a function of linear energy transfer. Radiat. Res. 2, 417–425 (1955). 2. Rossi, H. H. and Zaider, M. Microdosimetry and Its Applications. (Springer-Verlag) (1996) ISBN3-540-58541-9. 3. ICRU. Microdosimetry (International Commission on Radiation Units and Measurements) (1983). 4. Braby, L. A., Roesch, W. C. and Glass, W. A. Energy deposition spectra of 14C beta radiation in a uniform medium. Radiat. Res. 43, 499–503 (1970). 5. Glass, W. A. and Braby, L. A. A wall-less detector for measuring energy deposition spectra. Radiat. Res. 39, 230–240 (1969). 6. Kliauga, P. and Dvorak, R. Microdosimetric measurements of ionization by monoenergetic photons. Radiat. Res. 73, 1– 20 (1978). 7. Braby, L. A. and Ellett, W. H. Ionization in solid- and grid-walled detectors. Radiat. Res. 51, 569– 580 (1972). 8. Chatterjee, A. and Schaefer, H. J. Microdosimetric structure of heavy ion tracks in tissue. Radiat. Environ. Biophys. 13, 215–227 (1976). 9. Metting, N. F., Rossi, H. H., Braby, L. A., Kliauga, P. J., Howard, J., Zaider, M., Schimmerling, W., Wong, M. and Rapkin, M. Microdosimetry near the trajectory of highenergy heavy ions. Radiat. Res. 116, 183–195 (1988). 10. Taddei, J. P., Zhao, Z. and Borak, T. B. A comparison of the measured responses of a tissue-equivalent proportional counter to high energy heavy (HZE) particles and those
19. 20. 21.
22.
23. 24. 25.
Page 7 of 7
Downloaded from http://rpd.oxfordjournals.org/ at Freie Universitaet Berlin on May 11, 2015
The need for high-quality radiation measurements in complex radiation fields and difficult physical environments will continue to drive the development of microdosimetry detectors and their associated electronics. It is likely that microdosimetry will continue to provide important information in support of space exploration, industrial applications of ionising radiation and medicine. Technological developments in many different fields may contribute to new and innovative approaches to dosimeter design. Development of organic semiconductors may lead to improved solidstate dosimeters. Progress in semiconductor manufacturing and reduced feature size may make it possible to build solid-state microdosimetry detectors with improved chord length distributions and site sizes suitable for use with a wide range of radiations. Three-dimensional printing techniques, which have already been used to aid the design of new detectors, may progress to the point where detectors can be manufactured by 3D printing using TE plastic and suitable insulating material. Developments that have not yet been imagined may make it possible to solve persistent problems in radiation dosimetry. The investigator just needs to be observant of changes in a wide variety of fields, and not limit the approach to established methods when attempting to solve problems.
simulated using the Geant4 Monte Carlo code. Radiat. Meas. 43, 1498–1505 (2008). Nikjoo, H., Khvostunov, I. K. and Cucinotta, F. A. The response of (TEPC) proportional counters to heavy ions. Radiat. Res. 157, 435– 445 (2002). Ellett, W. H. and Braby, L. A. The microdosimetry of 250 kVP and 65 kVp x-rays, Co gamma-rays, and tritium beta particles. Radiat. Res. 51, 229–243 (1972). Loncol, T., Cosgrove, V., Denis, J. M., Gueulette, J., Mazal, A., Menzel, H. G., Pihet, P. and Sabattier, R. Radiobiological effectiveness of radiation beams with broad LET spectra: Microdosimetric analysis using biological weighting functions. Radiat. Prot. Dosim. 52, 347–352 (1994). Amols, H. I., Bradbiru, J. N., Dicello, J. F., Helland, J. F., Kligerman, M. M., Lane, T. G., Paciotti, M. A., Foeder, D. L. and Schillaci, M. E. Dose outside the treatment volume for irradiation with negative pions. Phys. Med. Biol. 23, 385– 396 (1978). Hogeweg, B. and Barendsen, G. W. Local energy distributions for a-particles of different energies in relation to the evaluation of critical sizes and energy requirements involved in the induction of damage in mammalian cells. In: Third Symposium on Microdosimetry. Ebert, H. G., Ed. (Commission of the European Communities) pp. 857–872 (1972). Srdoc, D., Goodman, L. J., Marino, S. A., Mills, R. E., Zaider, M. and Rossi, H. H. Microdosimetry of monoenergetic neutron radiation. In: Seventh Symposium on Microdosimetry. Booz, J., Ebert, H. and Hartfiel, H., Eds. (Harwood Academic Publishers) pp. 765–774 (1981). Dicello, J. F. et al. Microdosimetric comparison of scanned and conventional proton beams used in radiation therapy. Radiat. Prot. Dosim. 143, 513–518 (2011). Rollet, S., Colautti, P., Grosswendt, B., Herault, J., Wind, M., Gargioni, E., Beck, P., Lattocha, M. and Moro, D. Microdosimetric assessment of the radiation quality of a therapeutic proton beam: comparison between numerical simulation and experimental measurements. Radiat. Prot. Dosim. 143, 445– 449 (2011). Benjamin, P. W., Kemshall, C. D. and Redfearn, J. A high resolution spherical proportional counter. Nucl. Instrum. Methods 59, 77–85 (1968). Perez-Nunez, D. and Braby, L. A. Replacement tissueequivalent proportional counter for the international space station. Radat. Prot. Dosim. 143, 394– 397 (2011). Moro, D., Colautti, P., Gualdrini, G., Masi, M., Conte, V., De Nardo, L. and Tornielli, G. Two miniaturized TEPCs in a single detector for BNCT microdosimetry. Radiat. Prot. Dosim. 122, 396– 400 (2006). Bengtsson, L. G. Assessment of dose equivalent from fluctuations of energy deposition. In: Second Symposium on Microdosimetry. Ebert, H. G., Ed., (Commission of the European Communities) pp. 375–400 (1969). Waker, A. J., Aslam and Lori, J. Design of a multielement TEPC for neutron monitoring. Radiat. Prot. Dosim. 143, 463– 466 (2011). Agosteo, S. and Pola, A. Silicon Microdosimetry. Radiat. Prot. Dosim. 143, 409– 415 (2011). Galer, S., Hao, L., Gallop, J., Palmans, H., Kirkby, K. and Nisbet, A. Design concept for a novel SQUID-based microdosimeters. Radiat. Prot. Dosim. 143, 427– 431 (2011).
