- Email: [email protected]
Neutron flux mapping inside a cubic and a head PMMA phantom using indirect neutron radiography
ARTICLE IN PRESS Applied Radiation and Isotopes 67 (2009) S190–S194
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Neutron flux mapping inside a cubic and a head PMMA phantom using indirect neutron radiography Pi-En Tsai a, Yuan-Hao Liu a,, Chun-Kai Huang a, Hong-Ming Liu b, Shiang-Huei Jiang a a b
Engineering and System Science Department, National Tsing Hua University, Taiwan Nuclear Science and Technology Development Center, National Tsing Hua University, Taiwan
a r t i c l e in f o
Keywords: BNCT Indirect neutron radiography Imaging plate Neutron flux mapping
a b s t r a c t This study aims to measure the two-dimensional (2D) neutron spatial distribution inside a cubic and a head PMMA phantom for the purpose of further comparison with the treatment planning. The measurements were made by using the indirect neutron radiography (INR), which utilized a thin copper foil and the imaging plate. The developed image provides satisfactory spatial resolution and very low statistical error (o 1%). As to the time cost, the whole procedure normally takes less than 3 h. The result shows that the indirect neutron radiography can be a quick and reliable method to provide a 2D neutron spatial distribution inside a phantom. & 2009 Elsevier Ltd. All rights reserved.
1. Introduction The thermal neutron flux distribution inside a tumor is proportional to the boron reaction rate distribution concerned in the boron neutron capture therapy (BNCT). The boron dose delivered to a tumor, which is an important indication of the effectiveness of BNCT, is an integrated value of the boron spatial distribution multiplied by the neutron spatial distribution over the tumor volume. The more precise the neutron flux and the boron compound distribution, the more precise the boron dose. This paper will focus on the measurement of neutron flux distribution. Generally, the flux mapping is performed by only limited number of activation detectors due to the time cost and the counting ability. Hence, the quality of the flux mapping is not satisfactory. Recently, it is demonstrated that the indirect neutron radiography (INR) can quickly and precisely picture the twodimensional (2D) neutron spatial distribution of an epithermal neutron beam with a very high resolution (Liu et al., in press). In this study, the INR is applied to obtain the 2D neutron spatial distribution inside a cubic and a head PMMA phantom for the purpose of further comparison with the treatment planning.
2. Materials and methods
center of the beam exit were measured to be 1.34 108 and 1.07 109 neutron cm2 s1 for thermal (o0.5 eV) and epithermal neutrons (0.5 eVoEepio10 keV), respectively at a reactor power of 1.2 MW (Liu et al., 2009).
2.2. Indirect neutron radiography The INR utilizes the idea of neutron activation and later development. In this study, it is applied to measure the 2D neutron spatial distribution inside a phantom. The activity of the irradiated target is proportional to the product of the target crosssection and the average-neutron flux. For an epithermal neutron beam incident into a PMMA phantom, it is quickly moderated into thermal neutrons. Hence, the activity distribution over the target is the relative distribution of the concerned thermal neutron flux. As a result, the developed image can be processed into the neutron flux map after proper translation as expressed in the following equation: I / a0 / fn
(1)
where I is the number of signals collected by the image reader; a0 the activity of the activated target at the end of irradiation; and fn the thermal neutron flux cross the target.
2.1. THOR BNCT beam
2.3. Experiment setup
The BNCT beam of the THOR is an epithermal neutron beam with an aperture of 14 cm in diameter. The neutron fluxes at the
Concerning the availability, cost, neutron capture crosssection, and the activated radionuclide properties, a 15 20 0.0125 cm3 copper foil with purity 99.9% is chosen to be an activation converter for the neutron flux mapping. The uniformity of the copper foil is 70.5%. The dominant radionuclide in the
Corresponding author. Tel.: +886 3 5715131x35803.
E-mail address: [email protected] (Y.-H. Liu). 0969-8043/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2009.03.047
ARTICLE IN PRESS P.-E. Tsai et al. / Applied Radiation and Isotopes 67 (2009) S190–S194
copper foil is 64Cu formed from the 63Cu(n,g)64Cu reaction, which is sensitive to thermal neutrons. Measurements were performed inside a solid cubic phantom and a head phantom. The cubic phantom is composed of PMMA slices with an area 20 20 cm2 and with different thicknesses from 1 mm to 5 cm such that the copper can be put at the position of interest. All the slices are put inside a PMMA container, whose dimensions are 21 21 21 cm3. The head phantom is composed of 12 PMMA slices of thickness 2.54 cm, from the shoulder to the top of head (see Fig. 6(a)). There are four cross-hair lasers installed in the BNCT irradiation room; all the lasers crosses through a reference point. The front laser pointing to the beam center emits a cross passing through the horizontal and vertical planes normal to the beam exit surface. The cubic phantom with a copper foil inside is positioned against the beam exit surface and aligned according to the front laser to ensure that the beam direction is vertical to the PMMA surface. This position method is similar to the head phantom. For the BNCT beam, the intensity distribution of epithermal neutrons at the beam exit is also quite important. Therefore, a 0.5 mm thick cadmium plate was put in front of the copper foil to filter the thermal neutrons whose energies are below the cadmium cut-off energy (0.55 eV).
2.4.
63
Cu capture reaction rate vs. thermal neutron flux
Note that, the 63Cu(n,g)64Cu reaction also has response to epithermal neutrons due to its main resonance peak around 580 eV. Nevertheless, the incident epithermal neutrons are quickly moderated into thermal neutrons in a hydrogen-rich phantom within a quite short length. For the sake of demonstration, a comparison between the 63Cu(n,g)64Cu reaction rate distribution and the thermal neutron flux distribution inside the cubic PMMA phantom was performed using MCNP(Briesmeister, 2000). The self-shielding effect was also investigated by corresponding calculations. Calculated distributions can be found in Results section.
S191
2.5. Image development The radiation area detector applied in INR is an image plate (IP) which was first introduced by Fuji Film Co. in 1983 (Sonoda et al., 1983). It is comprised of phosphors whose material is BaFBr:Eu2+ that can trap and store the radiation energy. The IP applied in this work is type BAS-IIIs or the so-called white IP manufactured by Fuji, and the IP reader is Fuji FLA-3000 with the reading density selected to be 20 pixels/mm. The dimensions of BAS-IIIs IP are 20 cm wide and 40 cm long. As a 2D integral-type detector, IP was chosen in this work due to its superior linearity range exceeding five orders and its high sensitivity which is several ten times higher than the traditional X-ray film (Thoms, 1997). Moreover, it is reusable. Such advantages make the arrangement of INR much flexible. After the irradiation and cooling, the activated foil was placed into the IP cassette where the copper foil is attached to the surface of IP for exposure. Hence the response of IP to the radioactivity is contributed dominantly from the beta-plus and beta-minus particles accompanying the 64Cu beta decay than the decay gamma rays. Such an arrangement improves the quality of image. For minimizing the influence caused by fading phenomenon and non-uniform exposure, all the exposure are performed in a well temperature-controlled dark room (2071 1C). After the exposure, the IP was moved into the IP reader for development. All the following image process is done by Fujifilm Image Gauge software.
2.6. Linearity calibration When incident radiations deposit energy to IP’s phosphor layer, the excited electrons are then trapped to form fluoride-centers. The trapped electrons are stimulated using 633-nm HeNe laser and released as 415-nm blue lights called photostimulated luminescence (PSL). Theoretically, the PSL value is directly proportional to the deposited energy caused by radiation. In order to ensure that the response of IP is directly proportional to the activity of radiation source, the linearity between the PSL value
Fig. 1. The calculated 63Cu capture reaction rate distributions (with and without foil in the model) and calculated thermal neutron flux distribution inside the cubic PMMA phantom.
ARTICLE IN PRESS S192
P.-E. Tsai et al. / Applied Radiation and Isotopes 67 (2009) S190–S194
and the number of incident radiation was checked by a 64Cu source which is a 2.5 2.5 cm2, 0.1-mm thick, pure copper foil.
3. Results and discussions 3.1. Calculated reaction rate and thermal neutron flux distributions According to calculated results as shown in Fig. 1, the calculated reaction rate (with foil) distribution along the central axis deviates from the calculated thermal neutron flux distribution only at a very shallow region. For clarity, all the calculated distributions were normalized to their maximum values separately. For depth larger than 0.5 cm, the maximum deviation between the two calculated distributions is less than 2.5%. Furthermore, calculations with and without the copper foil in the model deliver identical distributions. Thus, the selfshielding effect does not affect the reaction rate distribution. 3.2. Linearity of IP Fig. 2 shows the result of linearity calibration. The copper foil described in Section 2.6 was first irradiated and activated in the epithermal neutron beam for 2 h to accumulate sufficient activity. Its activity was then determined by a well-calibrated high purity germanium detector. The exposure time was controlled to be 30 min; the time period after exposure, prior to the readout process was 10 min. From Fig. 2, it is clear that the activity of 64Cu is directly proportional to the PSL value. 3.3. Neutron flux mapping in a cubic phantom The copper foils were irradiated from 15 to 30 min, depending on the irradiation position, with a reactor power of 1.2 MW. After 1-h cooling, the activated copper foil was exposed to the IP for
30–60 min according to the degree of activity. All the interesting measured data are with a standard deviation below 1%. The measurements performed inside the cubic phantom were at the depth of 1, 2, and 5 cm. The 64Cu activity profiles, which are proportional to the thermal neutron distributions, along the horizontal axis of the beam aperture are shown in Fig. 3. The measured free-in-air profile (with a cadmium cover) at the beam exit surface is also given in the figure. It can be seen from the free-in-air measurement that the epithermal neutron distribution is quite uniform within a 12-cm diameter at the beam center. When the epithermal neutron beam is incident into the cubic phantom, owing to the scattering effect and its angular distribution, the relative intensity sharply decreases on the edge. As the depth increases in the phantom, the neutron flux becomes more divergent. Hence, the 90% relative intensity of the profiles is located at 4, 3.75, and 3.25 cm away from the beam center for the depths of 1, 2, and 5 cm, respectively. The measured profiles in the cubic phantom are quite symmetric, which is consistent with the free-in-air measurement. Due to the build-up effect of the epithermal neutrons in the hydrogen-rich phantom, the peak of the thermal neutron flux along the central axis will be a small distance away from the phantom surface. The measured profile along the center line is plotted in Fig. 4. The profile calculated by MCNP is also given in the figure. All the calculated values have a statistical uncertainty below 1% within 95% confidence interval. Apparently, the calculated curve is slightly different from the measured one. This is because of the imperfect energy spectrum and angular distribution of the neutron source description applied in the calculation. The measured 2D profile of the horizontal plane cross the central axis is shown in Fig. 5. The depths before 5 cm are much more concerned in the treatment, and therefore the resolution is higher than the rest of depths. The element size of the processed image before 5 cm is 0.2 (depth) 0.6 (wide) cm2; for the depths after 5 cm, it is 0.4 (depth) 0.6 (wide) cm2/element. The hot spot
Fig. 2. The linear response curve of BAS-IIIs to a 0.1-mm thick
64
Cu source.
ARTICLE IN PRESS P.-E. Tsai et al. / Applied Radiation and Isotopes 67 (2009) S190–S194
Fig. 3. The comparison between the
64
S193
Cu activity profiles along the horizontal axis at different depths in the cubic phantom, and free-in-air at beam exit surface.
Fig. 4. Measured and calculated depth profiles of the
emerges at the depth of 2.3 cm along the central axis. The upper 50% intensity along the center line locates between 0.4 and 6.2 cm. Note that all the concerned elements have a statistical error less than 1%.
64
Cu activity along the central axis of the beam.
spatial distribution. This result is quite delicate and can be used for the comparison with the treatment planning.
4. Conclusions 3.4. Neutron flux mapping in a head phantom The INR can also be performed in a human-like, sophisticated phantom. To simulate the clinical treatment of brain tumor, the copper foil was positioned at the vertical plane cross the beam central axis and implanted between layers 4 and 5 (about the height of eyes) as the head phantom was lying with its face side up as shown in Fig. 6(a). Fig. 6(b) shows the measured neutron
This study successfully measured the 2D thermal neutron flux distribution according to the measured 63Cu(n,g)64Cu reaction rate distribution, in the cubic and head phantom using the indirect neutron radiography. The procedure only takes less than 3 h. The developed image has satisfactory resolution and very low statistical error. The hot spot emerges at the depth of 2.3 cm in the cubic phantom. The measured image inside the head phantom is ready to be used for the comparison with the treatment planning and neutron
ARTICLE IN PRESS S194
P.-E. Tsai et al. / Applied Radiation and Isotopes 67 (2009) S190–S194
Fig. 5. The horizontal mapping image of the 64Cu relative distribution inside a cubic PMMA phantom with different range of depths. The depth in: (a) is from 0 to 5 cm and the depth in (b) is from 0 to 15 cm.
Fig. 6. The indirect neutron radiography was utilized in a head phantom: (a) is the experimental setup and (b) is the processed image representing the distribution in the phantom.
source adjustment. Still, caution should be paid at the shallow region, where epithermal neutrons still contribute partial reaction rates. As such, for the verification of the treatment planning, it is suggested to compare the calculated and measured 63Cu(n,g)64Cu reaction rate distributions, which has less uncertainty.
Acknowledgments This study was supported by the National Science Council of the Republic of China under contract number NSC95-2221-E-007135 and -136. We will like to show our sincere appreciation to Mr. Ang-Yu Chen for his assistance of the MCNP calculations.
64
Cu 2D relative
References Briesmeister, J., MCNP—a general Monte Carlo N-particle transport code. Version 4C, Los Alamos National Laboratory, 2000. Liu, Y.H., et al., 2009. Coarse-scaling adjustment of fine-group neutron spectra for epithermal neutron beams in BNCT using multiple activation detectors. Nucl. Instum. Methods A 598, 764–773. Liu, Y.-H., et al., BNCT epithermal neutron beam mapping by using indirect neutron radiography. Nucl. Technol., in press. Sonoda, M., et al., 1983. Computed radiography utilizing scanning laser stimulated luminescence. Radiology 148, 833–838. Thoms, M., 1997. The dynamic range of X-ray imaging with image plates. Nucl. Instrum. Methods A 264, 437–440.
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Neutron flux mapping inside a cubic and a head PMMA phantom using indirect neutron radiography Pi-En Tsai a, Yuan-Hao Liu a,, Chun-Kai Huang a, Hong-Ming Liu b, Shiang-Huei Jiang a a b
Engineering and System Science Department, National Tsing Hua University, Taiwan Nuclear Science and Technology Development Center, National Tsing Hua University, Taiwan
a r t i c l e in f o
Keywords: BNCT Indirect neutron radiography Imaging plate Neutron flux mapping
a b s t r a c t This study aims to measure the two-dimensional (2D) neutron spatial distribution inside a cubic and a head PMMA phantom for the purpose of further comparison with the treatment planning. The measurements were made by using the indirect neutron radiography (INR), which utilized a thin copper foil and the imaging plate. The developed image provides satisfactory spatial resolution and very low statistical error (o 1%). As to the time cost, the whole procedure normally takes less than 3 h. The result shows that the indirect neutron radiography can be a quick and reliable method to provide a 2D neutron spatial distribution inside a phantom. & 2009 Elsevier Ltd. All rights reserved.
1. Introduction The thermal neutron flux distribution inside a tumor is proportional to the boron reaction rate distribution concerned in the boron neutron capture therapy (BNCT). The boron dose delivered to a tumor, which is an important indication of the effectiveness of BNCT, is an integrated value of the boron spatial distribution multiplied by the neutron spatial distribution over the tumor volume. The more precise the neutron flux and the boron compound distribution, the more precise the boron dose. This paper will focus on the measurement of neutron flux distribution. Generally, the flux mapping is performed by only limited number of activation detectors due to the time cost and the counting ability. Hence, the quality of the flux mapping is not satisfactory. Recently, it is demonstrated that the indirect neutron radiography (INR) can quickly and precisely picture the twodimensional (2D) neutron spatial distribution of an epithermal neutron beam with a very high resolution (Liu et al., in press). In this study, the INR is applied to obtain the 2D neutron spatial distribution inside a cubic and a head PMMA phantom for the purpose of further comparison with the treatment planning.
2. Materials and methods
center of the beam exit were measured to be 1.34 108 and 1.07 109 neutron cm2 s1 for thermal (o0.5 eV) and epithermal neutrons (0.5 eVoEepio10 keV), respectively at a reactor power of 1.2 MW (Liu et al., 2009).
2.2. Indirect neutron radiography The INR utilizes the idea of neutron activation and later development. In this study, it is applied to measure the 2D neutron spatial distribution inside a phantom. The activity of the irradiated target is proportional to the product of the target crosssection and the average-neutron flux. For an epithermal neutron beam incident into a PMMA phantom, it is quickly moderated into thermal neutrons. Hence, the activity distribution over the target is the relative distribution of the concerned thermal neutron flux. As a result, the developed image can be processed into the neutron flux map after proper translation as expressed in the following equation: I / a0 / fn
(1)
where I is the number of signals collected by the image reader; a0 the activity of the activated target at the end of irradiation; and fn the thermal neutron flux cross the target.
2.1. THOR BNCT beam
2.3. Experiment setup
The BNCT beam of the THOR is an epithermal neutron beam with an aperture of 14 cm in diameter. The neutron fluxes at the
Concerning the availability, cost, neutron capture crosssection, and the activated radionuclide properties, a 15 20 0.0125 cm3 copper foil with purity 99.9% is chosen to be an activation converter for the neutron flux mapping. The uniformity of the copper foil is 70.5%. The dominant radionuclide in the
Corresponding author. Tel.: +886 3 5715131x35803.
E-mail address: [email protected] (Y.-H. Liu). 0969-8043/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2009.03.047
ARTICLE IN PRESS P.-E. Tsai et al. / Applied Radiation and Isotopes 67 (2009) S190–S194
copper foil is 64Cu formed from the 63Cu(n,g)64Cu reaction, which is sensitive to thermal neutrons. Measurements were performed inside a solid cubic phantom and a head phantom. The cubic phantom is composed of PMMA slices with an area 20 20 cm2 and with different thicknesses from 1 mm to 5 cm such that the copper can be put at the position of interest. All the slices are put inside a PMMA container, whose dimensions are 21 21 21 cm3. The head phantom is composed of 12 PMMA slices of thickness 2.54 cm, from the shoulder to the top of head (see Fig. 6(a)). There are four cross-hair lasers installed in the BNCT irradiation room; all the lasers crosses through a reference point. The front laser pointing to the beam center emits a cross passing through the horizontal and vertical planes normal to the beam exit surface. The cubic phantom with a copper foil inside is positioned against the beam exit surface and aligned according to the front laser to ensure that the beam direction is vertical to the PMMA surface. This position method is similar to the head phantom. For the BNCT beam, the intensity distribution of epithermal neutrons at the beam exit is also quite important. Therefore, a 0.5 mm thick cadmium plate was put in front of the copper foil to filter the thermal neutrons whose energies are below the cadmium cut-off energy (0.55 eV).
2.4.
63
Cu capture reaction rate vs. thermal neutron flux
Note that, the 63Cu(n,g)64Cu reaction also has response to epithermal neutrons due to its main resonance peak around 580 eV. Nevertheless, the incident epithermal neutrons are quickly moderated into thermal neutrons in a hydrogen-rich phantom within a quite short length. For the sake of demonstration, a comparison between the 63Cu(n,g)64Cu reaction rate distribution and the thermal neutron flux distribution inside the cubic PMMA phantom was performed using MCNP(Briesmeister, 2000). The self-shielding effect was also investigated by corresponding calculations. Calculated distributions can be found in Results section.
S191
2.5. Image development The radiation area detector applied in INR is an image plate (IP) which was first introduced by Fuji Film Co. in 1983 (Sonoda et al., 1983). It is comprised of phosphors whose material is BaFBr:Eu2+ that can trap and store the radiation energy. The IP applied in this work is type BAS-IIIs or the so-called white IP manufactured by Fuji, and the IP reader is Fuji FLA-3000 with the reading density selected to be 20 pixels/mm. The dimensions of BAS-IIIs IP are 20 cm wide and 40 cm long. As a 2D integral-type detector, IP was chosen in this work due to its superior linearity range exceeding five orders and its high sensitivity which is several ten times higher than the traditional X-ray film (Thoms, 1997). Moreover, it is reusable. Such advantages make the arrangement of INR much flexible. After the irradiation and cooling, the activated foil was placed into the IP cassette where the copper foil is attached to the surface of IP for exposure. Hence the response of IP to the radioactivity is contributed dominantly from the beta-plus and beta-minus particles accompanying the 64Cu beta decay than the decay gamma rays. Such an arrangement improves the quality of image. For minimizing the influence caused by fading phenomenon and non-uniform exposure, all the exposure are performed in a well temperature-controlled dark room (2071 1C). After the exposure, the IP was moved into the IP reader for development. All the following image process is done by Fujifilm Image Gauge software.
2.6. Linearity calibration When incident radiations deposit energy to IP’s phosphor layer, the excited electrons are then trapped to form fluoride-centers. The trapped electrons are stimulated using 633-nm HeNe laser and released as 415-nm blue lights called photostimulated luminescence (PSL). Theoretically, the PSL value is directly proportional to the deposited energy caused by radiation. In order to ensure that the response of IP is directly proportional to the activity of radiation source, the linearity between the PSL value
Fig. 1. The calculated 63Cu capture reaction rate distributions (with and without foil in the model) and calculated thermal neutron flux distribution inside the cubic PMMA phantom.
ARTICLE IN PRESS S192
P.-E. Tsai et al. / Applied Radiation and Isotopes 67 (2009) S190–S194
and the number of incident radiation was checked by a 64Cu source which is a 2.5 2.5 cm2, 0.1-mm thick, pure copper foil.
3. Results and discussions 3.1. Calculated reaction rate and thermal neutron flux distributions According to calculated results as shown in Fig. 1, the calculated reaction rate (with foil) distribution along the central axis deviates from the calculated thermal neutron flux distribution only at a very shallow region. For clarity, all the calculated distributions were normalized to their maximum values separately. For depth larger than 0.5 cm, the maximum deviation between the two calculated distributions is less than 2.5%. Furthermore, calculations with and without the copper foil in the model deliver identical distributions. Thus, the selfshielding effect does not affect the reaction rate distribution. 3.2. Linearity of IP Fig. 2 shows the result of linearity calibration. The copper foil described in Section 2.6 was first irradiated and activated in the epithermal neutron beam for 2 h to accumulate sufficient activity. Its activity was then determined by a well-calibrated high purity germanium detector. The exposure time was controlled to be 30 min; the time period after exposure, prior to the readout process was 10 min. From Fig. 2, it is clear that the activity of 64Cu is directly proportional to the PSL value. 3.3. Neutron flux mapping in a cubic phantom The copper foils were irradiated from 15 to 30 min, depending on the irradiation position, with a reactor power of 1.2 MW. After 1-h cooling, the activated copper foil was exposed to the IP for
30–60 min according to the degree of activity. All the interesting measured data are with a standard deviation below 1%. The measurements performed inside the cubic phantom were at the depth of 1, 2, and 5 cm. The 64Cu activity profiles, which are proportional to the thermal neutron distributions, along the horizontal axis of the beam aperture are shown in Fig. 3. The measured free-in-air profile (with a cadmium cover) at the beam exit surface is also given in the figure. It can be seen from the free-in-air measurement that the epithermal neutron distribution is quite uniform within a 12-cm diameter at the beam center. When the epithermal neutron beam is incident into the cubic phantom, owing to the scattering effect and its angular distribution, the relative intensity sharply decreases on the edge. As the depth increases in the phantom, the neutron flux becomes more divergent. Hence, the 90% relative intensity of the profiles is located at 4, 3.75, and 3.25 cm away from the beam center for the depths of 1, 2, and 5 cm, respectively. The measured profiles in the cubic phantom are quite symmetric, which is consistent with the free-in-air measurement. Due to the build-up effect of the epithermal neutrons in the hydrogen-rich phantom, the peak of the thermal neutron flux along the central axis will be a small distance away from the phantom surface. The measured profile along the center line is plotted in Fig. 4. The profile calculated by MCNP is also given in the figure. All the calculated values have a statistical uncertainty below 1% within 95% confidence interval. Apparently, the calculated curve is slightly different from the measured one. This is because of the imperfect energy spectrum and angular distribution of the neutron source description applied in the calculation. The measured 2D profile of the horizontal plane cross the central axis is shown in Fig. 5. The depths before 5 cm are much more concerned in the treatment, and therefore the resolution is higher than the rest of depths. The element size of the processed image before 5 cm is 0.2 (depth) 0.6 (wide) cm2; for the depths after 5 cm, it is 0.4 (depth) 0.6 (wide) cm2/element. The hot spot
Fig. 2. The linear response curve of BAS-IIIs to a 0.1-mm thick
64
Cu source.
ARTICLE IN PRESS P.-E. Tsai et al. / Applied Radiation and Isotopes 67 (2009) S190–S194
Fig. 3. The comparison between the
64
S193
Cu activity profiles along the horizontal axis at different depths in the cubic phantom, and free-in-air at beam exit surface.
Fig. 4. Measured and calculated depth profiles of the
emerges at the depth of 2.3 cm along the central axis. The upper 50% intensity along the center line locates between 0.4 and 6.2 cm. Note that all the concerned elements have a statistical error less than 1%.
64
Cu activity along the central axis of the beam.
spatial distribution. This result is quite delicate and can be used for the comparison with the treatment planning.
4. Conclusions 3.4. Neutron flux mapping in a head phantom The INR can also be performed in a human-like, sophisticated phantom. To simulate the clinical treatment of brain tumor, the copper foil was positioned at the vertical plane cross the beam central axis and implanted between layers 4 and 5 (about the height of eyes) as the head phantom was lying with its face side up as shown in Fig. 6(a). Fig. 6(b) shows the measured neutron
This study successfully measured the 2D thermal neutron flux distribution according to the measured 63Cu(n,g)64Cu reaction rate distribution, in the cubic and head phantom using the indirect neutron radiography. The procedure only takes less than 3 h. The developed image has satisfactory resolution and very low statistical error. The hot spot emerges at the depth of 2.3 cm in the cubic phantom. The measured image inside the head phantom is ready to be used for the comparison with the treatment planning and neutron
ARTICLE IN PRESS S194
P.-E. Tsai et al. / Applied Radiation and Isotopes 67 (2009) S190–S194
Fig. 5. The horizontal mapping image of the 64Cu relative distribution inside a cubic PMMA phantom with different range of depths. The depth in: (a) is from 0 to 5 cm and the depth in (b) is from 0 to 15 cm.
Fig. 6. The indirect neutron radiography was utilized in a head phantom: (a) is the experimental setup and (b) is the processed image representing the distribution in the phantom.
source adjustment. Still, caution should be paid at the shallow region, where epithermal neutrons still contribute partial reaction rates. As such, for the verification of the treatment planning, it is suggested to compare the calculated and measured 63Cu(n,g)64Cu reaction rate distributions, which has less uncertainty.
Acknowledgments This study was supported by the National Science Council of the Republic of China under contract number NSC95-2221-E-007135 and -136. We will like to show our sincere appreciation to Mr. Ang-Yu Chen for his assistance of the MCNP calculations.
64
Cu 2D relative
References Briesmeister, J., MCNP—a general Monte Carlo N-particle transport code. Version 4C, Los Alamos National Laboratory, 2000. Liu, Y.H., et al., 2009. Coarse-scaling adjustment of fine-group neutron spectra for epithermal neutron beams in BNCT using multiple activation detectors. Nucl. Instum. Methods A 598, 764–773. Liu, Y.-H., et al., BNCT epithermal neutron beam mapping by using indirect neutron radiography. Nucl. Technol., in press. Sonoda, M., et al., 1983. Computed radiography utilizing scanning laser stimulated luminescence. Radiology 148, 833–838. Thoms, M., 1997. The dynamic range of X-ray imaging with image plates. Nucl. Instrum. Methods A 264, 437–440.