Search tips
Search criteria 


Logo of jrrAbout the journalSubmissionsSubscribeJRR
J Radiat Res. 2016 June; 57(3): 318–324.
Published online 2016 June 21. doi:  10.1093/jrr/rrw025
PMCID: PMC4915547

Radiological properties of nanocomposite Fricke gel dosimeters for heavy ion beams


The radiological properties of nanocomposite Fricke gel (NC-FG) dosimeters prepared with different concentrations of nano-clay, perchloric acid and ferrous ions in deaerated conditions were investigated under carbon and argon ion beam irradiation covering a linear-energy-transfer (LET) range of 10 to 3000 eV/nm. We found that NC-FG exhibits radiological properties distinct from those of conventional Fricke gel. The radiation sensitivity of NC-FG is independent of the LET and is nearly constant even at very high LET (3000 eV/nm) values in the Bragg peak region of the argon ion beam. In addition, whereas conventional Fricke gel dosimeters only operate under acidic conditions, NC-FG dosimeters function under both acidic and neutral conditions. The radiation sensitivity decreases with decreasing nano-clay concentration in NC-FG, which indicates that the nano-clay plays a vital role in the radiation-induced oxidation of Fe2+.

Keywords: Ion beam, LET, Nanocomposite gel, Gel dosimeter


The aqueous ferrous sulfate Fricke chemical dosimeter is based on the radiation-induced oxidation of ferrous ions (Fe2+) to ferric ions (Fe3+) under acidic conditions [1, 2]. The addition of a gel matrix (typically gelatin or agarose) to the aqueous Fricke solution preserves the spatial distribution of Fe3+ ions, which can then be probed using magnetic resonance imaging (MRI) to record three-dimensional (3D) dose distributions in tissue-equivalent gels [3]. Under aerated and acidic conditions, the Fe3+ radiolytic yield G(Fe3+) for Fricke gel (FG) is typically 38–155 × 10–7 mol J–1 for gamma-ray or X-ray irradiation [49], which is 2.3–10 times higher than that for an aqueous Fricke solution (16.2 × 10–7 mol J–1) because the Fe3+ ions are formed via a chain reaction in the former [7]. On the other hand, FG dosimeters do not work under neutral conditions as well as they do under oxygen-free conditions [7, 8, 10].

3D dosimetry is expected to be a useful tool, particularly for the verification of more complex radiation dose planning in advanced radiation cancer therapy [1118]. However, conventional FG dosimeters have two main drawbacks regarding their application to heavy ion beam irradiation: diffusion of the radiation product [8, 11, 19] and linear-energy-transfer (LET)-dependent radiation sensitivity [20]. LET is defined as the amount of energy transferred per unit track length to the medium and is a representative index of radiation quality. The radiation sensitivity of gel dosimeters decreases with increasing LET due to enhanced radical recombination by localized ionization and excitation. It should be noted that the LET dependence of radiation sensitivity is common to virtually all types of 3D dosimeters and is also not unique to gel dosimeters, e.g. film, scintillation and semiconductor dosimeters are LET-dependent as well [2130].

Recently, we reported a nanocomposite Fricke gel (NC-FG) dosimeter that is free from the above-mentioned drawbacks [31]. The NC-FG was prepared by mixing a Fricke solution and nanocomposite gel under deaerated conditions. MRI images of this NC-FG provided stable radiation sensitivities and virtually no diffusion for 9 days following irradiation by a 290-MeV/nucleon carbon-ion beam (see fig. 5 of Reference [31]). Moreover, the radiation sensitivity data obtained from the NC-FG faithfully reproduced the depth–dose distribution measured by an ionization chamber, which indicates LET independence.

Fig. 5.
Comparison between the relative δR1 distributions (normalized to unity at the entrance surface) obtained for the reference composition (Gel1) and other NC-FG compositions (Gel2–Gel4, Gel8 and Gel10) under carbon ion beam irradiation and ...

As an extension of our previous work, in this study, we investigate the radiological properties of NC-FG dosimeters in detail for various preparation conditions and under carbon and argon beam irradiation over an LET range of 10–3000 eV/nm. The radiation sensitivity was found to be LET-independent even at very high LET values (3000 eV/nm), which corresponds to the Bragg peak region of the argon ion beam. While conventional FG dosimeters only work under acidic conditions, NC-FG dosimeters function under both acidic and neutral conditions. We also found that the radiation sensitivity depends critically on the presence of the nano-clay under the deaerated conditions used for NC-FG dosimeters.


Gel preparation

The various conditions under which NC-FG was prepared are summarized in Table 1. The composition of Gel1 was the same as that used in our previous report [31]: 1% (w/w) nano-clay (synthetic hectorite or Laponite XLG; Rockwood Additives Ltd), 3% (w/w) gelatin (300 g Bloom from porcine skin; Sigma-Aldrich), 1 mM ammonium iron (II) sulfate (Ammonium ferrous sulfate hexahydrate; Fluka) and 50 mM perchloric acid (perchloric acid (60%); Wako Pure Chemicals).

Table 1.
Preparation conditions for NC-FG and types of ionizing radiation. Gel1 is a reference condition reported in [31]. The concentration of HClO4, nano-clay, and Fe2+ is varied in Gel2–Gel5, Gel6–Gel8 and Gel9–Gel10, respectively

The procedure for the preparation of NC-FG was as follows. First, ultra-pure water was exposed to bubbling Ar gas for 30 min to remove dissolved oxygen. Next, gelatin and Laponite XLG were added to the deaerated ultra-pure water and mixed using a rotation/revolution vacuum mixer (V-mini300, EME Corporation) for 3 min to obtain a uniform dispersion. The dispersion was then heated in a microwave oven until dissolution took place. Finally, the 5% (w/w) aqueous Fricke stock solution [2] was added at around 40°C and mixed using a rotation/revolution mixer for 3 min to produce NC-FG. The prepared NC-FG was sealed in color comparison tubes made of Pyrex glass (Iwaki Glass Co., COLOR-TUBE25S: 18 mm outer diameter and 18 cm length) under a pure argon atmosphere in a glove box and left to solidify for a day.


The irradiation experiments were performed at the Biological Irradiation Port of the Heavy Ion Medical Accelerator in Chiba (HIMAC) at the National Institute of Radiological Sciences (NIRS). A 12C6+ beam at 290 MeV/nucleon and an 40Ar18+ beam at 500 MeV/nucleon with an irradiation field of ±5% lateral dose uniformity within a 10-cm diameter were used. The color comparison tubes containing NC-FG were irradiated from the bottom. The energies of both ion beams were attenuated with an energy absorber made of polymethyl methacrylate (PMMA) plates to adjust the range of the ion beams to the center of the color comparison tubes. The water equivalent thickness of the energy absorber for the carbon and argon beam irradiation was 63.26 and 11.04 mmH2O, respectively. At the entrance surface of the color comparison tubes containing NC-FG, the absorbed dose was 150, 300, 450 and 600 Gy for the carbon ion beam, and 200, 400, 600 and 800 Gy for the argon ion beam. These values were calibrated using a Markus ionization chamber at the same position as the samples [32, 33]. The depth–dose distribution of the carbon ion beam in this system has been reported in the literature [34].

MRI measurements

In the presence of paramagnetic Fe3+ ions, proton relaxation rates [s–1] increase significantly. The relaxation rate is proportional to the Fe3+ concentration in a gel or solution. We measured the longitudinal MR relaxation rate (R1 [s–1], the inverse of the longitudinal relaxation time T1) for the NC-FG samples using a 1.5-T MRI scanner (Intera Achieva 1.5 T HP Nova Dual Gradient, Philips Medical Systems, Best, the Netherlands). A turbo mixed sequence [35, 36] was used. The conditions for the T1 measurements were as follows: TR = 2260 ms; TE1 = 19 ms; TE2 = 100 ms; TI = 500 ms; ETL = 6; pixel spacing = 0.8 mm; slice thickness = 5 mm with a resulting voxel size of 0.8 × 0.8 × 5 mm3. The MRI measurements were carried out within 2 weeks of the irradiation.

Evaluation of the Fe3+ radiolytic yield G(Fe3+) using a calibration curve

The calibration curve (plot of R1 vs. [Fe3+]) was evaluated by measuring non-irradiated gel samples with known concentrations (0–0.4 mM ammonium iron (III) sulfate mixed with 1% (w/w) nano-cla 3% (w/w) gelatin and 50 mM perchloric acid under aerated conditions), using the same 1.5 T MRI and the same MRI sequence employed for the irradiated NC-FG samples. The slope of the calibration curve was used as a conversion factor (CF) to convert the change in the proton relaxation rate ΔR1 to [Fe3+]. The Fe3+ radiolytic yield G(Fe3+) was then calculated using the following equations:



Where [Fe3+] (mM) is the concentration of ferric ions produced by irradiation, D (kGy) is the absorbed dose and ρ (kg/l is the density of NC-FG. The CF was used to evaluate G(Fe3+) for the reference composition Gel1 only (see Table 1).


Radiological properties of reference NC-FG

Figure 1(a) and (b) shows the R1 distributions measured for the reference sample Gel1 irradiated with a 290-MeV/nucleon carbon ion beam and 500-MeV/nucleon argon ion beam, respectively. The R1 values for the non-irradiated samples (hereafter, referred to as the baseline) are about 1.3 and 1.0 s−1 for the samples used for carbon and argon beam irradiation, respectively. These differences in the baseline values per batch are due to oxygen contamination that occurs during sample preparation. In addition, there is not much difference between the baseline R1 values for NC-FG and those reported for conventional FG [9, 20].

Fig. 1.
(a, b) R1 distributions and (c, d) entrance surface dose dependence of R1 values at 5 mm penetration depth (open squares) and at the Bragg peak (open circles) in the NC-FG dosimeters after irradiation with (a, c) a 290-MeV/nucleon carbon ion beam and ...

The peak positions were adjusted within a single MRI pixel (pixel size 0.78 mm) in such a way that they coincided with each other. The entrance surface dose dependence of R1 near the entrance surface (5 mm depth) and near the Bragg peak (79 mm for carbon irradiation, 105 mm for argon irradiation), where good linearity was observed, is plotted in Fig 1(c) and 1(d), respectively. It should be noted that the horizontal axis is the entrance surface dose and that the dose especially at the Bragg peak is much higher than it.

In Fig. 2, we compare the radiation sensitivity δR1, defined as the rate of increase in R1 per unit entrance surface dose, with the actual dose distribution obtained from the ionization chamber (right vertical axis). The δR1 distribution for Gel1 faithfully reproduces the dose distribution, even at the Bragg peak for the argon ion beam, indicating that the sensitivity for Gel1 did not change even under very high LET radiation. In addition, the radiation sensitivity near the surface is almost the same (0.6 s–1 kGy–1) for both C and Ar ions.

Fig. 2.
Comparison of the δR1 distribution of Gel1 (solid line: left vertical axis) and the physical dose distribution (open symbols: right vertical axis) measured using the ionization chamber (IC) with different ion beams. δR1 is the radiation ...

Figure 3 shows the calibration curve obtained from measurements of R1 carried out with various Fe3+ concentrations. The slope of the straight line is CF = 7.71 s–1 mM–1, which is consistent with the value of 8.37 s–1 mM–1 reported in [3]. We determined the Fe3+ radiolytic yield for the reference composition Gel1 to be 0.78 × 10–7 mol J–1 using equations (1) and (2). This result indicates that the radiolytic yield for NC-FG is significantly smaller than that for the aqueous Fricke solution for carbon ion irradiation [37] (G(Fe3+) = 11 × 10–7 mol J–1 under aerated conditions and G(Fe3+) = 6 × 10–7 mol J–1 under deaerated conditions).

Fig. 3.
Plot of change in relaxation rate R1 vs. [Fe3+] used as the calibration curve. Open symbols represent measured values on gel sample prepared from 1% (w/w) nano-clay, 3% (w/w) gelatin, 50 mM perchloric acid and 0–0.4 mM ammonium iron (III) sulfate. ...

Effect of preparation conditions

The gelling properties depend strongly on the preparation conditions used in their fabrication. For example, the incorporation of perchloric acid into NC-FGs decreases their resilience and firmness. On the other hand, high concentrations of nano-clay and gelatin cause the gels to lose their mobility immediately after mixing. Outside the ranges used in this study, the gels became either too hard or too soft.

Figure 4 shows the R1 distributions for NC-FGs produced under various preparation conditions. R1 is confirmed to increase monotonically with dose for all preparation conditions, except in the case of Gel5 and Gel6. Relative radiation sensitivities as a function of depth are shown in Fig. 5 for various concentrations of (a) perchloric acid, (b) nano-clay and (c) iron. The peak-to-entrance δR1 ratios for Gel1–Gel4, Gel8 and Gel10 are between 3.8 and 4.6. Similar LET-independent radiological properties of NC-FG were confirmed over a wide range of preparation conditions (HClO4: 0–50 mM, nano-clay: 0.75–1% (w/w), Fe2+: 1–5 mM).

Fig. 4.
R1 distributions for NC-FG for several preparation conditions (see Table 1 for list) after irradiation with a 290-MeV/nucleon carbon beam at five different entrance surface doses (0, 150, 300, 450, 600 Gy).

Figure 6 summarizes how the radiation sensitivity near the entrance surface (4–6 mm depth) changes with the preparation conditions with respect to the reference composition (Gel1). Figure 6(a) shows the effect of the nano-clay concentration; the results indicate that the radiation sensitivity vanishes at 0.1% (w/w) nano-clay and below. Conventional FG dosimeters are also not radiation sensitive under deaerated conditions [7]. While, in our previous report [31], we originally incorporated nano-clay into Fricke gel in order to suppress the diffusion of Fe3+ ions, the results shown in Fig. 6(a) reveal that nano-clay also significantly affects the reaction mechanism under deaerated conditions. Although this may be related to the adsorption of Fe2+ ions onto clay nanoparticles [38, 39], further investigation is required in order to reveal the precise underlying reaction mechanism. Figure 6(b) shows the dependence of the radiation sensitivity on the HClO4 concentration, and thus, on pH. Under neutral conditions, the behavior of NC-FG (Gel2) is similar to that under acidic conditions (Gel1). This is in stark contrast to conventional FG (without nano-clay), which only works under acidic conditions [8]. In addition, the radiation sensitivity of NC-FG is suppressed at 150 mM HClO4 (Gel5), whereas that of conventional FG is nearly constant between 6 and 250 mM sulfuric acid [7]. We speculate that the origin of this difference is the following. Nanoclay adsorbs hydrogen ions as well as Fe2+ and Fe3+ ions [40]. Thus, even if the bulk FG is neutral, Fe2+ and Fe3+ ions may be placed under locally acidic conditions, and NC-FG functions. Since hydrogen ions are more strongly adsorbed than Fe2+ and Fe3+ ions [40], too high hydrogen ion density may prevent the adsorption of the latter, leading to radiation sensitivity suppression.

Fig. 6.
Comparison of the entrance surface radiation sensitivity obtained from the reference composition (Gel1) and the other NC-FG compositions (Gel2–Gel10) under carbon ion irradiation. (a) Clay concentration dependence. (b) HClO4 concentration dependence. ...

Figure 6(c) shows the radiation sensitivity of NC-FG increases with Fe2+ concentration, which again is in contrast to conventional FG, for which the sensitivity does not change between 0.2 and 5 mM [8, 41]. Finally, the effect of the gas used for deaeration (Ar and N2O) on the sensitivity is shown in Fig. 6(d). The data plotted for N2O are from our previous report [31]. It appears that the gas used did not greatly affect the radiation sensitivity, even though the preparation methods were not the same; a rotation/revolution vacuum mixer, a glove box, and a microwave oven were used to stabilize the baseline of R1 in this study.


The radiological properties of NC-FG under various preparation conditions were investigated in detail under carbon and argon beam irradiation covering an LET range of 10–3000 eV/nm. By varying the concentration of nano-clay, Fe2+ and perchloric acid from the reference composition described in our previous report [31], the following features were revealed.

  1. The radiation sensitivity is nearly LET-independent for the compositions 1–4, 8 and 10. Surprisingly, the sensitivity did not change even at very high LET values (3000 eV/nm) in the Bragg peak region of the argon ion beam.
  2. NC-FG works not only in acidic but also in neutral conditions.
  3. The radiation sensitivity correlates with the nano-clay concentration and disappeared at nano-clay concentrations below 0.1% (w/w).

The first and second features are distinct from those of conventional Fricke gel. The third feature suggests that nano-clay not only suppresses the diffusion of iron ions, but also plays an essential role in radiation-induced reactions.

Clarification of the reaction mechanism leading to these features requires further study, especially on the behavior of Fe2+ adsorbed onto nano-clay [40].


Funding to pay the Open Access publication charges was provided by the Journal of Radiation Research as this article was a recipient of the JRR Award at the ICRR 2015 conference.


This research was carried out as a Research Project with Heavy Ions at NIRS-HIMAC. TM gratefully acknowledges support from the RIKEN Special Postdoctoral Researchers Program and the Grant-in-Aid for Young Scientists (B-24760725 and B-26820412).


1. Fricke H, Morse S The chemical action of roentgen rays on dilute ferrosulphate solutions as a measure of dose. Am J Roentgenol Rad Ther 1927;18:430–2.
2. Fricke H, Hart EJ Chemical dosimetry. In: Attix FH, Roesch WC, Tochilin E (eds). Radiation Dosimetry: Instrumentation. New York: Academic Press, 1966, 167–239.
3. Gore JC, Kang YS, Schulz RJ Measurement of radiation dose distributions by nuclear magnetic resonance (NMR) imaging. Phys Med Biol 1984;29:1189–97. [PubMed]
4. Duzenli C, Sloboda R, Robinson D A spin-spin relaxation rate investigation of the gelatin ferrous sulphate NMR dosimeter. Phys Med Biol 1994;39:1577–92. [PubMed]
5. Audet C, Schreiner LJ Multiple-site fast exchange model for spin-lattice relaxation in the Fricke-gelatin dosimeter. Med Phys 1997;24:201–9. [PubMed]
6. Olsson LE, Petersson S, Ahlgren L et al. Ferrous sulphate gels for determination of absorbed dose distributions using MRI technique: basic studies. Phys Med Biol 1989;34:43–52. [PubMed]
7. Appleby A, Leghrouz A, Christman EA Radiation chemical and magnetic resonance studies of aqueous agarose gels containing ferrous ions. Int J Radiat Appl Instrum C Radiat Phys Chem 1988;32:241–4.
8. Schulz RJ, Deguzman AF, Nguyen DB et al. Dose-response curves for Fricke-infused agarose gels as obtained by nuclear magnetic resonance. Phys Med Biol 1990;35:1611–22. [PubMed]
9. Olsson LE, Appleby A, Sommer J A new dosimeter based on ferrous sulphate solution and agarose gel. Int J Radiat Appl Instrum Appl Radiat Isot 1991;42:1081–6.
10. Keller BM. Characterization of the NMR-based Fricke-gelatin dosimeter. MSc thesis, Montreal: McGill University, 1994.
11. Schreiner LJ. Review of Fricke gel dosimeters. J. Phys: Conference Series 2004;3:9–21.
12. Vandecasteele J, De Deene Y On the validity of 3D polymer gel dosimetry: II. Physico-chemical effects. Phys Med Biol 2013;58:43–61. [PubMed]
13. Vandecasteele J, De Deene Y On the validity of 3D polymer gel dosimetry: III. MRI-related error sources. Phys Med Biol 2013;58:63–85. [PubMed]
14. Vandecasteele J, De Deene Y On the validity of 3D polymer gel dosimetry: I. reproducibility study. Phys Med Biol 2013;58:19–42. [PubMed]
15. De Deene Y, De Wagter C, Van Duyse B et al. Validation of MR-based polymer gel dosimetry as a preclinical three-dimensional verification tool in conformal radiotherapy. Magn Reson Med 2000;43:116–25. [PubMed]
16. MacDougall N, Pitchford WG, Smith MA A systematic review of the precision and accuracy of dose measurements in photon radiotherapy using polymer and Fricke MRI gel dosimetry. Phys Med Biol 2002;47:R107–21. [PubMed]
17. Baldock C, De Deene Y, Doran S et al. Polymer gel dosimetry. Phys Med Biol 2010;55:R1–63. [PMC free article] [PubMed]
18. Furuta T, Maeyama T, Ishikawa KL et al. Comparison between Monte Carlo simulation and measurement with a 3D polymer gel dosimeter for dose distributions in biological samples. Phys Med Biol 2015;60:6531–46. [PubMed]
19. Olsson LE, Westrin BA, Fransson A et al. Diffusion of ferric ions in agarose dosimeter gels. Phys Med Biol 1992;37:2243–52.
20. Back SA, Medin J, Magnusson P et al. Ferrous sulphate gel dosimetry and MRI for proton beam dose measurements. Phys Med Biol 1999;44:1983–96. [PubMed]
21. Karger CP, Jakel O, Palmans H et al. Dosimetry for ion beam radiotherapy. Phys Med Biol 2010;55:R193–234. [PubMed]
22. Martišíková M, Jäkel O Dosimetric properties of Gafchromic® EBT films in monoenergetic medical ion beams. Phys Med Biol 2010;55:3741–51. [PubMed]
23. Al-Nowais S, Doran S, Kacperek A et al. A preliminary analysis of LET effects in the dosimetry of proton beams using PRESAGE (TM) and optical CT. Appl Radiat Isotopes 2009;67:415–18. [PubMed]
24. Heufelder J, Stiefel S, Pfaender M et al. Use of BANG (R) polymer gel for dose measurements in a 68 MeV proton beam. Med Phys 2003;30:1235–40. [PubMed]
25. Gustavsson H, Bjack SAJ, Medin J et al. Linear energy transfer dependence of a normoxic polymer gel dosimeter investigated using proton beam absorbed dose measurements. Phys Med Biol 2004;49:3847–55. [PubMed]
26. Ramm U, Moog J, Spielberger B et al. Investigations of dose response of BANG (R) polymer-gels to carbon ion irradiation. Radiother Oncol 2004;73:S99–101. [PubMed]
27. Baker CR, Quine TE, Brunt JNH et al. Monte Carlo simulation and polymer gel dosimetry of 60 MeV clinical proton beams for the treatment of ocular tumours. Appl Radiat Isotopes 2009;67:402–5. [PubMed]
28. Kantemiris I, Petrokokkinos L, Angelopoulos A et al. Carbon beam dosimetry using VIP polymer gel and MRI. J. Phys: Conference Series 2009;164:012055.
29. Ramm U, Weber U, Bock M et al. Three-dimensional BANG (TM) gel dosimetry in conformal carbon ion radiotherapy. Phys Med Biol 2000;45:N95–102. [PubMed]
30. Yoshioka M, Tominaga T, Usui S et al. Examination of fundamental characteristics of a polymer gel detector in a proton beam irradiation. Radiat Meas 2011;46:64–71.
31. Maeyama T, Fukunishi N, Ishikawa KL et al. A diffusion-free and linear-energy-transfer-independent nanocomposite Fricke gel dosimeter. Radiat Phys Chem 2014;96:92–6.
32. Torikoshi M, Minohara S, Kanematsu N et al. Irradiation system for HIMAC. J Radiat Res 2007;48:A15–25. [PubMed]
33. Kanai T, Fukumura A, Kusano Y et al. Cross-calibration of ionization chambers in proton and carbon beams. Phys Med Biol 2004;49:771–81. [PubMed]
34. Kanai T, Endo M, Minohara S et al. Biophysical characteristics of HIMAC clinical irradiation system for heavy-ion radiation therapy. Int J Radiat Oncol Biol Phys 1999;44:201–10. [PubMed]
35. Baldock C, Burford RP, Billingham N et al. Experimental procedure for the manufacture and calibration of polyacrylamide gel (PAG) for magnetic resonance imaging (MRI) radiation dosimetry. Phys Med Biol 1998;43:695–702. [PubMed]
36. Denkleef J, Cuppen JJM RLSQ: T1, T2, and ρ calculations, combining ratios and least squares. Magn Reson Med 1987;5:513–24. [PubMed]
37. LaVerne JA, Schuler RH Track effects in water radiolysis—Yields of the Fricke dosimeter for carbon-ions with energies up to 1700MeV. J Phys Chem 1994;98:4043–9.
38. Feng J, Hu X, Yue PL et al. A novel laponite clay-based Fe nanocomposite and its photo-catalytic activity in photo-assisted degradation of Orange II. Chem Eng Sci 2003;58:679–85.
39. Sen Gupta S, Bhattacharyya KG Adsorption of heavy metals on kaolinite and montmorillonite: a review. Phys. Chem. Chem. Phys. 2012;14:6698–723. [PubMed]
40. Swartzen-Allen SL, Matijevic E Surface and colloid chemistry of clays. Chem Rev 1974;74:385–400.
41. Chu KC, Jordan KJ, Battista JJ et al. Polyvinyl alcohol-Fricke hydrogel and cryogel: two new gel dosimetry systems with low Fe3+ diffusion. Phys Med Biol 2000;45:955–69. [PubMed]

Articles from Journal of Radiation Research are provided here courtesy of Oxford University Press