- Email: [email protected]
Dose mapping of complex-shaped foods using electron-beam accelerators
Food Control 18 (2007) 1223–1234 www.elsevier.com/locate/foodcont
Dose mapping of complex-shaped foods using electron-beam accelerators R. Rivadeneira, R. Moreira *, J. Kim, M.E. Castell-Perez Department of Biological and Agricultural Engineering, Texas A&M University, College Station, TX 77843-2117, United States Received 13 February 2006; received in revised form 24 July 2006; accepted 31 July 2006
Abstract Obtaining a uniform dose in inherently irregularly shaped foods, such as an apple, is very difficult, and development of accurate dose calculation methodologies is needed. The objective was to determine the most uniform irradiation treatment on apple surrogates and GAFCHROMIC HD-810 films using electron beams from a 2 MeV Van de Graaff (VDG) accelerator, a l0 MeV Linear Accelerator (LINAC), and X-rays from a 5 MeV LINAC. We manufactured a solid 3-D apple-surrogate from a mixture of paraffin wax, chloroform, and methyl yellow biological indicator. The surrogate showed electron interaction characteristics very similar to an actual apple. Evaluation of food positioning strategies in front of the e-beam source of the VDG demonstrated that tilting axial rotation ensure uniform dose distribution of the entire surface of the surrogate, even reaching the critical regions of the stem and calyx ends. Irradiation with 10 MeV dual e-beam showed uniform penetration (top to bottom) in the HD-810 film and surrogates. X-ray irradiation completely penetrated the apple surrogate (top to bottom) showing excellent lateral uniformity at different penetration depths. The measured dose distributions in the surrogate showed good agreement with the calculated doses. These results support the validity of using chemical surrogate techniques for accurate planning of food irradiation treatments. 2006 Elsevier Ltd. All rights reserved. Keywords: Surrogate; Dosimeter; Irradiation; Food product
1. Introduction For any irradiation treatment, it is critical to control the absorbed dose, i.e., the energy absorbed per unit mass of irradiated material, to provide uniform radiation treatment. In fruits and vegetables, for example, it is important to deliver radiation uniformly throughout the product to achieve bacterial reduction, quarantine control, shelf-life extension, or any other specific purpose (Bresica, Moreira, Braby, & Castell-Perez, 2003). Radiochromic film is used typically in food irradiation to map dose distributions at specific locations in products with predefined packaging and uniform geometries. However, complex-shaped and non-homogenous foods with
*
Corresponding author. Tel.: +1 979 847 8794; fax: +1 979 8453932. E-mail address: [email protected] (R. Moreira).
0956-7135/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodcont.2006.07.023
respect to their 3-D mass distribution make it difficult to produce complete dose map distributions using radiochromic film (Borsa, Chu, Sun, Linton, & Hunter, 2002). Hence, there is a need to develop a dosimetry system capable of easily producing a complete and accurate 3-D dose map distribution to establish adequate radiation treatment (Kim, Rivadeneira, Castell-Perez, & Moreira, 2006). The development of a solid chemical dosimeter will enable visual detection of ionizing radiation in a 3-D object and thus establish quantitative measurements based on relative absorbed dose distributions within the surrogate. The objective of this study was to produce a solid-state chemical mixture that works not only as a radiation indicator, but would also provide the possibility of relatively measuring ionizing radiation in food products. An applesurrogate, a 3-D representation of an apple made of an equivalent biological material, was used as the sample product geometry for electron beam irradiation.
1224
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
2. Materials and methods 2.1. Apple surrogate development Surrogate chemical dosimeters were created using a mold obtained by casting one Red Delicious apple (Fig. 1). The casting material used was a synthetic rubber (Reprorubber No. 16131 catalyst and base, FlexBar Machine Corp, NY). The amount of paraffin wax required for each casting was 325.0 ml/mold after measurement. Table 1 shows the composition required to cast an apple surrogate dosimeter specifically at 20% chloroform and 4 · 104 m methyl yellow concentration. The dosimeter chemical composition provides identical electron density and Z-values (atomic number) to that of an actual apple (Table 2).
Fig. 1. Apple surrogate sample and its cast.
Table 1 Apple-surrogate chemical composition (by wt., at 20% chloroform, 4 · 104 m methyl yellow) Component
Mass (Kg)
Paraffin wax Chloroform Methyl yellow Microcrystalline wax
0.221 0.056 2.5 · 105 2.8 · 103
Total mass
0.280
The surrogate contains 70% of carbon (paraffin’s hydrocarbon structure –C25H52) and 18% of chlorine (from chloroform). No oxygen is present in the surrogate. However, the carbon contained in tissue material substitutes the missing oxygen (ICRU, 1989); thus, in average, the surrogate has identical electron density and Z-value (atomic number) to that of an actual apple. For the surrogate and the actual apple to absorb and scatter electrons to the same extent, both total linear stopping powers (TSP) must be identical over the operating energy range used in the radiation treatment. The stopping power is defined as the rate of energy loss suffered by a charged particle in traversing a unit path length of a medium and it is related to the charge and velocity of the incident particle, and physical properties of the medium. The total linear stopping power, S, for electrons includes the total energy loss, dE, by collision and bremsstrahlung production for a path length dx in the medium as, dE dE dE S¼ ¼ þ q ð1Þ dx q dx col q dx rad where q is the density of the medium, (dE/q dx)col is the mass collision stopping power which includes all energy loses in particle collisions that directly produces secondary electrons (delta rays) and atomic excitations and, (dE/q dx)rad is the mass radiative stopping power which includes all energy losses of the primary electron which lead to bremsstrahlung production (ICRU, 1984). Fig. 2 shows that the total stopping power values for both the surrogate (paraffin apple) and the actual apple overlap throughout the entire range of energy. Therefore, the developed surrogate can be used for practical studies of e-beam irradiation of an actual apple for an energy source range of 0.01–10 MeV. 2.2. GAFCHROMIC HD-810 film contours A 210 · 296 mm paper piece (A4 standard white) with the same dimensions of a HD-810 radiochromic film sheet was used as a grid to fit 6 apple contours per sheet of film. A 210 · 296 mm sheet of film was placed on top of the grid and individual contour slices were cut.
Table 2 The elemental composition and density of an actual apple and the surrogate Material
Surrogate Actual apple (Red delicious)
Element composition (% weight) H
C
N
O
Others
12.99 10.28
70.27 6.07
0.0168 0.04
– 83.47
17.72 Cl 0.01 Mg, 0.01 Ca, 0.01P, 0.11K
Densityb (kg/m3)
Zeffa
1008 1042
7.43 6.58
P ðW i =Ai ÞZ 2 Effective atomic number: calculated as (Tsoulfanidis, 1995): Z eff ¼ Pi ðW =A ÞZi where Ai is the atomic mass, Zi the atomic number, and Wi the weight i i i i fraction. b At room temperature Mohsenin (1986). a
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234 2
Linear Stopping Power [MeV/cm]
10
Actual Apple Paraffin Apple
1
10
0
10
-2
10
-1
10
0
10
1
10
Electron's Kinetic Energy [MeV] Fig. 2. Total stopping power (TSP) of an actual apple and the surrogate (paraffin apple) with corresponding electron energy.
Each individual film contour was held together between two identical paper slices forming a protective envelope for the apple film slice. The apple surrogate was cut in half (along the stem), the radiochromic film envelops placed in between the two halves, and then the two halves were held together by vacuum packaging the whole apple surrogate. 2.3. E-beam irradiation experimental design The apple-surrogates were irradiated with 1.35 MeV electrons using a Van de Graaff electron accelerator Type AK model S (High Voltage Engineering Corporation, Burlington, Massachusetts) located at the Food Safety Engineering Laboratory in the Biological and Agricultural Engineering Department at Texas A&M University.
1225
Each apple-surrogate was rotated at a constant speed around a fixed vertical plane defined as the conveyor’s vertical axis passing through the surrogate center point. In addition, each apple-surrogate was positioned at an angle, a, defined as the horizontal or latitudinal plane passing through the surrogate’s left to right sides (180–0). The e-beam source was fixed at an angle of 22.5 pointing downwards with respect to the apple-surrogate horizontal or latitudinal plane. The dose distributions were obtained for a values of 90, 45, and 22.5 using target doses of 0.05 and 0.1 kGy per angle. The experimental dose rate was fixed at 4.5 · 103 kGy/min and at 7.5 · 103 kGy/ min, respectively. Fig. 3 defines the apple-surrogate dimensions centered at the irradiation point P at 28.36 cm away from the exit. Each apple-surrogate was initially irradiated at the top with half the total target dose at each particular a angle. Next, the apple-surrogate was turned upside down to deliver the remaining dose at the bottom region at the same particular a angle. 2.4. E-beam and X-ray irradiation LINAC The electron beam linear accelerator located in the National Center for Electron-Beam Food Research Facility at Texas A&M University provides two operation modes, i.e., electrons and X-rays. For electron irradiations, two vertically mounted and opposing 10 MeV, 14.4 kW LINAC are used to accelerate electrons to near the speed of light using microwaves so they can be targeted into a food product (Fig. 4a). The electron LINAC can operate in single or dual beam mode so food products receive radiant energy from one or both the upper and lower accelerators so that product flipping is not required. For X-ray irradiation, a single horizontally mounted 5 MeV, 15 kW X-ray linear accelerator produces electrons traveling near the speed of light hitting a dense metal target so X-rays are produced. 2.5. E-beam irradiation experimental design using the 10 MeV LINAC Irradiation treatment of apple-surrogates was performed with a dual beam of 10 MeV electrons. A target dose of 1 kGy was delivered to the samples based on (1) the thicknesses of Plexiglas attenuator of 3 cm; (2) the rolling conveyor system traveling at a speed of 0.3 m/s (60 ft/ min). The attenuating material was utilized so that the apple-surrogates would receive the target dose of 1 kGy when irradiated with a 10 MeV LINAC. 2.6. Polystyrene surrogate holder with Lucite for the 10 MeV E-beam irradiation
Fig. 3. Apple-surrogate dimensions centered at irradiation point P.
Individual polystyrene apple holders were constructed for each apple-surrogate. Each polystyrene holder was 10 cm in length, 10 cm in wide, and 14 cm high with a wall
1226
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
Fig. 4. (a) Apple-surrogate packaging layout for 10 MeV e-beam irradiation; (b) electrons operating ranging and maximum dose in the designed surrogate holder (a).
thickness of 1.5 cm. Each holder contained two removable top and bottom Lucite custom-made absorbers. The Lucite absorber was cut to thicknesses of 3.0 cm to obtain the desired target doses of 1 kGy. Fig. 4a shows the layout of an individual apple-surrogate in its packaging for irradiation with the 10 MeV electrons. Fig. 4b shows the electron operating range passing the different absorber thicknesses. The 10 MeV electrons
penetrate through the Lucite absorber before they reach the apple-surrogate positioned inside the package. Points E and E 0 (Fig. 4b) show the operating range for the electrons after a penetration depth, t (cm), where the electrons have attenuated their average dose from 10 to 1.35 MeV to achieve the desired target doses. Prior to irradiation, the vacuum-packed apple-surrogate was introduced in the holder (Fig. 5a). Finally, the apple
Fig. 5. (a) Polystyrene holder with 4 cm thick Lucite absorber; (b) apple holders for 4 cm thick Lucite absorber before irradiation treatment.
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
1227
holder was taped to a cardboard box (L = 60.9 cm, W = 50.8 cm, and H = 10.2 cm) (Fig. 5b). The apple surrogates were produced in triplicate and irradiated. The irradiations were performed in batches ranked accordingly to its Lucite thickness value. The analyses were carried out 24 h after irradiation treatment to permit the apple surrogate’s chemical solution to stabilize as well as the HD-810 film contours. 2.7. X-ray irradiation experimental design using the 5 MeV LINAC The X-ray irradiation experiment with the 5 MeV LINAC also required use of a secondary holder for the apple-surrogates. Each apple-surrogate was separated by a distance of 16.5 cm center to prevent scattering effects. The height from the bottom of the holder to the center of the apple was 31 cm where the X-ray beam was located. The surrogates’ top surface (stem) was facing the X-ray beam to expose first the top region of the apple. Each apple-surrogate contained its respective radiochromic film slice. Fig. 6a shows the arrangement of three apple surrogates used for X-ray irradiation in a cardboard box holder prior to irradiation. Fig. 6b shows the apple surrogates ready to be irradiated using the 5-MeV X-rays. The X-ray beam exit window was located perpendicular to the conveyor’s direction of motion and separated a distance of 30.48 cm from the custom-made holder. The conveyor was set to travel at a speed of 60.96 cm/min to provide the 0.6 kGy target dose. Each apple-surrogate was positioned inside the holder gap resting on its side so that the apple center of symmetry (where the film was located) was perpendicular to the beam (Fig. 6b). 2.8. Apple-surrogate post irradiation handling After the irradiation experiments were completed, all surrogate samples and HD-810 film were stored in a dark room for a period of 24 h. Next, each apple-surrogate
Fig. 7. (a) GAFCHROMIC HD-810 film calibration curve (Green channel) at irradiation point P; (b) Green channel methyl yellow calibration curve using 3.175 mm thick slices and 1.35 MeV electrons.
sliced into 13.175 mm thick apple contours for a total of 16 slices per surrogate.
Fig. 6. (a) Apple surrogate holder for the X-ray irradiation experiment; (b) configuration layout for the apple-surrogate irradiation at a conveyor speed of 0.61 m/min.
1228
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
2.9. GAFCHROMIC HD-810 film and apple-surrogate slices scanning A Microtek ScanMaker 8700 Pro Series (Microtek USA, Carson, California) was used to perform transmission scans on both radiochromic film and surrogate slices. The scanner was set to capture an image at 300 pixels per inch (ppi). As a result, a 2.64 mega bytes image of 960 pixels · 960 pixels was scanned each time and saved using a Tagged Image File Format (TIFF). 2.10. GAFCHROMIC HD-810 film calibration The film apple 1 cm · 1 cm contours were irradiated using a cylindrical Lucite holder. The holder was designed and constructed to fit the round Farmer ionization chamber with the film in front of the ionization window volume.
A calibration curve for the films was produced by placing individual film slices and irradiating at specific doses. Each individual film was then scanned in transmission mode using the full dynamic range scale. The film intensity data was analyzed for the Green channel and a calibration equation was found between 0.05 and 1.50 kGy. Fig. 7a shows the calibration results. 2.11. Chemical dosimeter calibration The methyl yellow, chloroform, and paraffin wax solution was calibrated using Petri dishes with dimensions equal to the mold (2.54 cm in diameter and 7.6 cm height). The samples were irradiated with 1.35 MeV electrons using target doses from 0 to 0.5 kGy. After the irradiations, each sample was cut and scanned along the beam direction. Fig. 8b shows the calibration results for the green channel.
Fig. 8. (a) HD-810 film dose distribution and (b) HD-810 film dose-depth distribution in the apple-surrogate irradiated at 0.05 kGy with the VDG and a = 90 towards and against a 1.35 MeV plane source; (c) HD-810 film dose distribution; (d) HD-810 film dose-depth distribution in apple-surrogate irradiated at 0.05 kGy with the VDG and a = 45 towards and against a 1.35 MeV plane source; (e) HD-810 film dose distribution; and (f) HD-810 film dose-depth distribution in the apple-surrogate irradiated at 0.05 kGy with the VDG and a = 22.5 towards and against a 1.35 MeV plane source.
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
2.12. Image processing of GAFCHROMIC film and applesurrogate contours A computer code was written using Matlab (The Mathworks Inc., Natick, Massachusetts) for the HD-810 film contours and the apple-surrogate slices to extract and analyze the intensity data corresponding to the green channel of each image file. A 960 pixel · 960 pixel square matrix contained all the intensity data measurements for the green channel only. Each data point was first transformed into optical density by OD = log(I0/I), where I0 is the light intensity of an unexposed film, and I is the light intensity of an exposed film. Then, the dose values where calculated using the calibration equations shown in Fig. 7a and b for the film and apple-surrogate, respectively.
1229
The dose deposition profiles for the film contours and apple-surrogate slices were computed using the Matlab Image Processing Toolbox contouring algorithms, and later transformed to grayscale. 2.13. Energy deposition and dose depth distributions The dose data obtained for this experiment was calculated at the plane that traverses the apple-surrogate from top to bottom, and divides the surrogate in two symmetrical pieces so the radiochromic film response could be compared to that of the apple-surrogate. The dose distributions for both film and apple-surrogate depend on several common factors including the electron entrance angle, the distance from the electron exit window
Fig. 9. (a) HD-810 film dose distribution; (b) HD-810 film dose-depth distribution in the apple-surrogate irradiated at 0.1 kGy with the VDG and a = 45 towards and against a 1.35 MeV plane source; (c) HD-810 film dose distribution and (d) HD-810 film dose-depth distribution in the apple-surrogate irradiated at 0.1 kGy with the VDG and a = 22.5 towards and against a 1.35 MeV plane source.
1230
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
to the surrogate, the path electrons follow because of scattering and the maximum penetration range within the apple-surrogate. Dose deposition data and dose-depth contours were created to provide a visual reference for the different treatments studied. 3. Results 3.1. E-beam irradiation using 1.35-MeV VDG 3.1.1. GAFCHROMIC HD-810 film contours The energy deposition profiles and depth dose distributions around the apple-surrogate periphery (0–360), in its vertical plane of symmetry, (i.e., the apple is divided in half from top to bottom) are shown in Figs. 8 and 9 for the target doses of 0.05 and 0.1 kGy, respectively. Fig. 8a shows that the energy deposition profile at an orientation of 90 away from the apple’s longitudinal axis results in uniform irradiation on the left and right sides of the surrogate. However, the regions near the top and bottom region located between 70 and 90 and 240 and 300 did not receive any irradiation (Fig. 8b). The maximum penetration depth was approximately 4.20 mm from the surface at 180. Fig. 8b shows a high dose region between 210 and 240 below a depth of 0.76 mm. These high doses resulted from the continuous irradiation of such particular points due to the geometrical orientation during the irradiation process. The high doses present on the left and right bottom sides of the surrogate resulted from a higher number of electrons being directed towards these ends due to a shift of the electron beam distribution (Rivadeneira, 2004). The left side of the energy deposition contour in Fig. 8a shows greatest penetration depths than of the right side. This behavior was consistent throughout all samples (Rivadeneira, 2004) and was due to the swiveling motion of the apple-surrogate as it hung from the turning hook and rotated around its vertical axis. Only a very small fraction of the normal surface vectors will be parallel with the incident electrons resulting in a direct hit. Such points were expected to receive higher doses and greater penetration depths due to the absence of electron scattering events. Nonetheless, it was found that due to the beam distribution, higher number of electrons struck the apple-surrogate at the left and right bottom regions resulting in greater penetration depths despite of the higher occurrence of scattering events compared to the surrogate’s top. Fig. 8c shows that the dose deposition data was more uniform at a = 45 compared to a = 90 orientation angle away from the apple’s y-axis. The orientation angle of 45 produces a uniform dose deposition in areas located at the top and bottom of the surrogate. The bottom region (Fig. 8d) of the surrogate (270), however, did not show significant dose deposition, but wider regions (between 210 to 270 and 300 to 330) were irradiated compared to those in Fig. 8b.
The penetration depth (Fig. 8d) showed more uniformity than that irradiated at a = 90 (Fig. 8b). Both orientations showed a maximum penetration depth of approximately 4.57 mm from the surface at different regions. High dose regions at 45 orientation were minimized compared to the a = 90 orientation (Fig. 8a) due to a more uniform irradiation. The apple-surrogate was oriented closer to the e-beam incident angle resulting in more exposed regions (Fig. 8c). Even though the results in Fig. 8d show high dose distributions at 250 and 290, the energy deposition gradient appeared more uniform than at a = 90 and the penetration depths were similar at both orientations. The apple-surrogate orientation parallel to the electron beam at a = 22.5 (Fig. 8e) showed very similar results to that obtained at a = 45. It was expected that the applesurrogate exposure at an angle equal to that of the electron
Fig. 10. (a) Methyl yellow apple contour dose distribution; (b) methyl yellow apple contour dose depth distribution in the apple-surrogate irradiated at 0.05 kGy with the VDG and a = 45 towards and against a 1.35 MeV plane source.
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
beam would result in the best irradiation orientation. In particular, the closer the apple-surrogate is aligned to the beam direction the broader the areas of the apple that are exposed. However, problem areas were still located at the top and bottom regions, where several points were not exposed to irradiation. Fig. 8f shows that for a = 22.5, the surrogate absorbed the highest doses at the top left and right positions located at 60 and 120, respectively. At these locations, it is expected that the electron entrance angle be perpendicular to the surrogate surface resulting in high dose regions. The largest penetration depths at these positions demonstrate that the electrons penetrated nearly perpendicular to the surrogate surface. The effect of a higher dose in the energy deposition and penetration depth profiles, around the surrogate periphery, was studied using the previous apple-surrogate orientation
1231
with a 0.1 kGy dose. Fig. 9 shows the results for the 45 and 22.5 of orientations. As the top of the apple-surrogate was positioned closer to the electron beam plane at 0.1 kGy, the energy deposition data improved as shown in Fig. 9a and c irradiated at a = 45 and a = 22.5, respectively. The dose deposition data shown in Fig. 9a demonstrate that at a = 45 there was complete radiation treatment throughout the surrogate periphery (Fig. 9b). The unexposed regions at the lower energy (0.05 kGy) were completely covered using higher doses. At the top of the surrogate, Fig. 9b shows that between 80 and 120 the maximum penetration depth was at least 1.6 mm from the surface. At the bottom, between 250 and 300, the minimum penetration depth was 3.05 mm from the surface compared to less than 0.76 mm at 0.05 kGy (Fig. 9b). The apple-surrogate received high dose values at the top and bottom right and left corners located at approximately at
Fig. 11. HD-810 film: (a) dose distribution in kGy and (b) depth-dose distribution in kGy using Lucite thickness of 3 cm and a dual beam 10 MeV LINAC accelerator with a conveyor speed of 0.3 m/s (60 ft/min).
1232
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
45, 135, 225, and 315 regions, respectively. At a penetration depth of 2.30 mm from the surface, the dose distribution became uniform throughout the apple periphery. The surrogate irradiated at a = 22.5 and 0.1 kGy showed complete treatment of the apple periphery (Fig. 9c). Fig. 9d shows deeper penetration depth (5.33 mm from the surface) compared to the other two treatments. Nonetheless, the apple-surrogate absorbed higher doses at more spots within its surface compared to any other treatment. The dark areas shown below the penetration depth of 1.52 mm from the surface in Fig. 9d, demonstrate higher doses were absorbed with this orientation. The maximum penetration depths at 100 and 270 regions were at least above 3 mm from the surface ensuring uniform irradiation treatment. The parallel orientation to the electron beam provided uniform irradiation throughout the sample, but it appeared to be overexposed to that at 45.
The evaluation of food positioning strategies in front of the e-beam source demonstrated that tilting the apple-surrogate towards the e-beam ensures uniform dose distributions of the entire surface of the surrogate, even reaching the critical regions of the apple stem and calyx ends. This is important so the risk associated with bacterial contamination, linked to Escherichia coli O157:H7 that may be present in raw apples (Rangel, Sparling, Crowe, Griffin, & Swerdlow, 2005), may be reduced on the apple’s surface by electron beam irradiation. 3.1.2. Apple-surrogate contours The energy deposition profiles and depth dose distributions around the apple-surrogate periphery showed similar results to those obtained with the HD-810 film. Fig. 10a and b show the results for a = 45. Because of chloroform evaporation during the casting method, it was not possible to trace absorbed dose values
Fig. 12. HD-810 film: (a) dose distribution and (b) depth-dose distribution in kGy using a custom-made polystyrene holder and a 5 MeV LINAC accelerator with a conveyor speed of 0.01 m/s (2 ft/min).
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
in the top regions of the methyl yellow apple contours. Improvement in the casting technique should yield better results. Because the pouring orifice in the mold was exactly at the top, during the pouring process the top parts of the mold received less chloroform. The solidification process of the methyl yellow–chloroform–paraffin wax solution inside the mold was very slow, so less chloroform was present in the solution at the end of the pouring. For this reason, no irradiation treatment was detected near the top regions of the apple for both energies and all surrogate orientations towards the e-beam. In general, the methyl yellow surrogate 2-D slices have shown to detect radiation between 0 and 0.5 kGy. The apple-surrogate’s main advantage over the HD-810 film consists in its ability to visually show the absorbed dose in 3-D (Rivadeneira, 2004). 3.2. E-beam Irradiation using 10-MeV LINAC Fig. 11a shows the dose deposition data for the radiochromic film with a Lucite thickness of 3 cm (target dose 1 kGy). The electrons penetrated quite uniformly from the bottom and the top until losing all their energy at the center of the apple-surrogate. The dose depth profile of the electrons measured from top to bottom at specific locations in the x-axis (Fig. 11b). The top entrance dose at the left and central planes were very close showing the electron beam uniformity (Fig. 11a). The right side (x = 64 mm) showed less irradiation; however, this lower dose may have been a result of the positioning of the apple-surrogate inside the box. At the bottom of the apple-surrogate (Fig. 11a), the film dose distribution agrees with the target dose (1 kGy). Fig. 11b shows high dose regions at the top left and right positions of the apple-surrogate. These differences may be due to the positioning of the apple-surrogate in the box and surrogate complex shape. The top left side of the apple-surrogate that shows the highest absorbed dose is closer to the Lucite absorber compared to its top right side. As a result, the electrons traveled less before reaching the top-left part of the sample hence depositing more energy. 3.3. X-ray irradiation using 5 MeV LINAC Fig. 12a shows the energy deposition contour for the radiochromic film. The surrogate received a maximum dose of approximately 0.6 kGy at the top. Fig. 12b shows several curves at penetration depths of 9, 16, 34, 50, and 68 mm from the surface. Notice the dose decrease as the X-rays traveled through the apple at the different depths. The entrance dose was approximately 0.6 below 9 mm from the surface and the exit dose at the bottom of the surrogate was approximately 0.4 kGy at a depth of 68 mm (Fig. 12b). The dose at each particular penetration depth remained uniform across the x-axis.
1233
4. Conclusions The use of a surrogate (to represent a complex-shaped food) composed of paraffin wax, chloroform, and methyl yellow was successful to detect absorbed radiation using 1.35 MeV electrons. A solid-state three dimensional ‘‘apple’’ surrogate helped to visually determine absorbed dose in the surrogate vertical axis of symmetry throughout its periphery. The chemical solution response to absorbed dose was found to be non-linear for 1.35 MeV electrons for doses ranging from 0.01 to 0.5 kGy. The optimal operating range for the chemical solution was between 0.1 and 0.3 kGy. A transmission flatbed scanner proved to be successful and economical tool to obtain dose deposition and depth dose profiles of apple-surrogate slices 3.175 mm thick using Matlab. Image Processing was performed in Red Green Blue (RGB) color space where it was found that the Green channel offered the best response to absorbed doses between 0.01 and 0.5 kGy using 1.35 MeV electrons. In general, tilting the apple-surrogate towards the ebeam source resulted in more uniform irradiations. The surrogates could be over-exposed when oriented exactly parallel to the e-beam at a = 22.5. The 45 orientation ensured uniformity throughout the surrogate periphery for wider penetration depths without overexposing the product. The 90 orientation was found to be the least effective at the top and bottom regions. The methyl yellow apple-surrogate proved to be a good visual of radiation dose in any direction where the radiation treatment was performed. For a Lucite thickness of 3 cm (target dose of 1.0 kGy), the HD-810 film confirmed a very uniform irradiation treatment at the top and bottom regions with a maximum dose of 1.2 kGy located at the top left region of the apple. As expected, the dose decreased in a linear tendency from the surface to the highest penetration depth until the electrons lost all their energy. The X-ray irradiation completely penetrated the entire apple-surrogate from top to bottom. The highest doses observed with the film were located at the top region of the surrogate where the X-ray beam hit first the apple. The entrance dose throughout most of the top region was approximately 0.6 kGy while the dose at the bottom region was approximately 0.4 kGy. X-ray irradiation was uniform across the apple. The dose decreased approximately 25% from the top to the bottom in the radiochromic film.
Acknowledgements This research has been funded by the USDA CSREES Project # 2002-03935 and the USDA NRI Food Safety program Project # 2002-02411. The authors would like to thank Dr. Leslie Braby (Nuclear Engineering) for providing expertise and technical assistance with the irradiation experiments.
1234
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
References Borsa, J., Chu, R., Sun, J. S., Linton, N., & Hunter, C. (2002). Use of CT scans and treatment planning software for validation of the dose component of food irradiation protocols. Radiation Physics and Chemistry, 63(3–6), 271–275. Bresica, G., Moreira, R., Braby, L., & Castell-Perez, E. (2003). Monte Carlo simulation and dose distribution of low energy electron irradiation of an apple. Journal of Food Engineering, 60, 31–39. ICRU (1984). Radiation Dosimetry: Electron Beams with Energies Between 1 and 50 MeV. International Commission on Radiation Units and Measurements, Report No. 35. ICRU (1989). Tissue Substitutes in Radiation Dosimetry and Measurement. International Commission on Radiation Units and Measurements, Report No. 44.
Kim, J., Rivadeneira, R. G., Castell-Perez, M. E., & Moreira, R. G. (2006). Development and validation of a methodology for dose calculation in electron beam irradiation of complex-shaped foods. Journal of Food Engineering, 34(3), 359–369. Mohsenin, N. N. (1986). Physical properties of Plant and Animal Materials (second ed.). New York: Gordon and Breach Science Publishers. Rangel, J. M., Sparling, P. H., Crowe, C., Griffin, P. M., & Swerdlow, D. L. (2005). Epidemiology of Escherichia coli O157:H7 Outbreaks, United States, 1982-2002. Emerging Infectious Diseases, CDC, 11(4), 606–609. Rivadeneira, R. (2004). Development of a chemical dosimeter for electron beam food irradiation. Master Thesis. Department of Biological and Agricultural Engineering. Texas A&M University. College Station, TX. Tsoulfanidis, N. (1995). Measurement and Detection of Radiation. New York: Taylor & Francis.
Dose mapping of complex-shaped foods using electron-beam accelerators R. Rivadeneira, R. Moreira *, J. Kim, M.E. Castell-Perez Department of Biological and Agricultural Engineering, Texas A&M University, College Station, TX 77843-2117, United States Received 13 February 2006; received in revised form 24 July 2006; accepted 31 July 2006
Abstract Obtaining a uniform dose in inherently irregularly shaped foods, such as an apple, is very difficult, and development of accurate dose calculation methodologies is needed. The objective was to determine the most uniform irradiation treatment on apple surrogates and GAFCHROMIC HD-810 films using electron beams from a 2 MeV Van de Graaff (VDG) accelerator, a l0 MeV Linear Accelerator (LINAC), and X-rays from a 5 MeV LINAC. We manufactured a solid 3-D apple-surrogate from a mixture of paraffin wax, chloroform, and methyl yellow biological indicator. The surrogate showed electron interaction characteristics very similar to an actual apple. Evaluation of food positioning strategies in front of the e-beam source of the VDG demonstrated that tilting axial rotation ensure uniform dose distribution of the entire surface of the surrogate, even reaching the critical regions of the stem and calyx ends. Irradiation with 10 MeV dual e-beam showed uniform penetration (top to bottom) in the HD-810 film and surrogates. X-ray irradiation completely penetrated the apple surrogate (top to bottom) showing excellent lateral uniformity at different penetration depths. The measured dose distributions in the surrogate showed good agreement with the calculated doses. These results support the validity of using chemical surrogate techniques for accurate planning of food irradiation treatments. 2006 Elsevier Ltd. All rights reserved. Keywords: Surrogate; Dosimeter; Irradiation; Food product
1. Introduction For any irradiation treatment, it is critical to control the absorbed dose, i.e., the energy absorbed per unit mass of irradiated material, to provide uniform radiation treatment. In fruits and vegetables, for example, it is important to deliver radiation uniformly throughout the product to achieve bacterial reduction, quarantine control, shelf-life extension, or any other specific purpose (Bresica, Moreira, Braby, & Castell-Perez, 2003). Radiochromic film is used typically in food irradiation to map dose distributions at specific locations in products with predefined packaging and uniform geometries. However, complex-shaped and non-homogenous foods with
*
Corresponding author. Tel.: +1 979 847 8794; fax: +1 979 8453932. E-mail address: [email protected] (R. Moreira).
0956-7135/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodcont.2006.07.023
respect to their 3-D mass distribution make it difficult to produce complete dose map distributions using radiochromic film (Borsa, Chu, Sun, Linton, & Hunter, 2002). Hence, there is a need to develop a dosimetry system capable of easily producing a complete and accurate 3-D dose map distribution to establish adequate radiation treatment (Kim, Rivadeneira, Castell-Perez, & Moreira, 2006). The development of a solid chemical dosimeter will enable visual detection of ionizing radiation in a 3-D object and thus establish quantitative measurements based on relative absorbed dose distributions within the surrogate. The objective of this study was to produce a solid-state chemical mixture that works not only as a radiation indicator, but would also provide the possibility of relatively measuring ionizing radiation in food products. An applesurrogate, a 3-D representation of an apple made of an equivalent biological material, was used as the sample product geometry for electron beam irradiation.
1224
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
2. Materials and methods 2.1. Apple surrogate development Surrogate chemical dosimeters were created using a mold obtained by casting one Red Delicious apple (Fig. 1). The casting material used was a synthetic rubber (Reprorubber No. 16131 catalyst and base, FlexBar Machine Corp, NY). The amount of paraffin wax required for each casting was 325.0 ml/mold after measurement. Table 1 shows the composition required to cast an apple surrogate dosimeter specifically at 20% chloroform and 4 · 104 m methyl yellow concentration. The dosimeter chemical composition provides identical electron density and Z-values (atomic number) to that of an actual apple (Table 2).
Fig. 1. Apple surrogate sample and its cast.
Table 1 Apple-surrogate chemical composition (by wt., at 20% chloroform, 4 · 104 m methyl yellow) Component
Mass (Kg)
Paraffin wax Chloroform Methyl yellow Microcrystalline wax
0.221 0.056 2.5 · 105 2.8 · 103
Total mass
0.280
The surrogate contains 70% of carbon (paraffin’s hydrocarbon structure –C25H52) and 18% of chlorine (from chloroform). No oxygen is present in the surrogate. However, the carbon contained in tissue material substitutes the missing oxygen (ICRU, 1989); thus, in average, the surrogate has identical electron density and Z-value (atomic number) to that of an actual apple. For the surrogate and the actual apple to absorb and scatter electrons to the same extent, both total linear stopping powers (TSP) must be identical over the operating energy range used in the radiation treatment. The stopping power is defined as the rate of energy loss suffered by a charged particle in traversing a unit path length of a medium and it is related to the charge and velocity of the incident particle, and physical properties of the medium. The total linear stopping power, S, for electrons includes the total energy loss, dE, by collision and bremsstrahlung production for a path length dx in the medium as, dE dE dE S¼ ¼ þ q ð1Þ dx q dx col q dx rad where q is the density of the medium, (dE/q dx)col is the mass collision stopping power which includes all energy loses in particle collisions that directly produces secondary electrons (delta rays) and atomic excitations and, (dE/q dx)rad is the mass radiative stopping power which includes all energy losses of the primary electron which lead to bremsstrahlung production (ICRU, 1984). Fig. 2 shows that the total stopping power values for both the surrogate (paraffin apple) and the actual apple overlap throughout the entire range of energy. Therefore, the developed surrogate can be used for practical studies of e-beam irradiation of an actual apple for an energy source range of 0.01–10 MeV. 2.2. GAFCHROMIC HD-810 film contours A 210 · 296 mm paper piece (A4 standard white) with the same dimensions of a HD-810 radiochromic film sheet was used as a grid to fit 6 apple contours per sheet of film. A 210 · 296 mm sheet of film was placed on top of the grid and individual contour slices were cut.
Table 2 The elemental composition and density of an actual apple and the surrogate Material
Surrogate Actual apple (Red delicious)
Element composition (% weight) H
C
N
O
Others
12.99 10.28
70.27 6.07
0.0168 0.04
– 83.47
17.72 Cl 0.01 Mg, 0.01 Ca, 0.01P, 0.11K
Densityb (kg/m3)
Zeffa
1008 1042
7.43 6.58
P ðW i =Ai ÞZ 2 Effective atomic number: calculated as (Tsoulfanidis, 1995): Z eff ¼ Pi ðW =A ÞZi where Ai is the atomic mass, Zi the atomic number, and Wi the weight i i i i fraction. b At room temperature Mohsenin (1986). a
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234 2
Linear Stopping Power [MeV/cm]
10
Actual Apple Paraffin Apple
1
10
0
10
-2
10
-1
10
0
10
1
10
Electron's Kinetic Energy [MeV] Fig. 2. Total stopping power (TSP) of an actual apple and the surrogate (paraffin apple) with corresponding electron energy.
Each individual film contour was held together between two identical paper slices forming a protective envelope for the apple film slice. The apple surrogate was cut in half (along the stem), the radiochromic film envelops placed in between the two halves, and then the two halves were held together by vacuum packaging the whole apple surrogate. 2.3. E-beam irradiation experimental design The apple-surrogates were irradiated with 1.35 MeV electrons using a Van de Graaff electron accelerator Type AK model S (High Voltage Engineering Corporation, Burlington, Massachusetts) located at the Food Safety Engineering Laboratory in the Biological and Agricultural Engineering Department at Texas A&M University.
1225
Each apple-surrogate was rotated at a constant speed around a fixed vertical plane defined as the conveyor’s vertical axis passing through the surrogate center point. In addition, each apple-surrogate was positioned at an angle, a, defined as the horizontal or latitudinal plane passing through the surrogate’s left to right sides (180–0). The e-beam source was fixed at an angle of 22.5 pointing downwards with respect to the apple-surrogate horizontal or latitudinal plane. The dose distributions were obtained for a values of 90, 45, and 22.5 using target doses of 0.05 and 0.1 kGy per angle. The experimental dose rate was fixed at 4.5 · 103 kGy/min and at 7.5 · 103 kGy/ min, respectively. Fig. 3 defines the apple-surrogate dimensions centered at the irradiation point P at 28.36 cm away from the exit. Each apple-surrogate was initially irradiated at the top with half the total target dose at each particular a angle. Next, the apple-surrogate was turned upside down to deliver the remaining dose at the bottom region at the same particular a angle. 2.4. E-beam and X-ray irradiation LINAC The electron beam linear accelerator located in the National Center for Electron-Beam Food Research Facility at Texas A&M University provides two operation modes, i.e., electrons and X-rays. For electron irradiations, two vertically mounted and opposing 10 MeV, 14.4 kW LINAC are used to accelerate electrons to near the speed of light using microwaves so they can be targeted into a food product (Fig. 4a). The electron LINAC can operate in single or dual beam mode so food products receive radiant energy from one or both the upper and lower accelerators so that product flipping is not required. For X-ray irradiation, a single horizontally mounted 5 MeV, 15 kW X-ray linear accelerator produces electrons traveling near the speed of light hitting a dense metal target so X-rays are produced. 2.5. E-beam irradiation experimental design using the 10 MeV LINAC Irradiation treatment of apple-surrogates was performed with a dual beam of 10 MeV electrons. A target dose of 1 kGy was delivered to the samples based on (1) the thicknesses of Plexiglas attenuator of 3 cm; (2) the rolling conveyor system traveling at a speed of 0.3 m/s (60 ft/ min). The attenuating material was utilized so that the apple-surrogates would receive the target dose of 1 kGy when irradiated with a 10 MeV LINAC. 2.6. Polystyrene surrogate holder with Lucite for the 10 MeV E-beam irradiation
Fig. 3. Apple-surrogate dimensions centered at irradiation point P.
Individual polystyrene apple holders were constructed for each apple-surrogate. Each polystyrene holder was 10 cm in length, 10 cm in wide, and 14 cm high with a wall
1226
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
Fig. 4. (a) Apple-surrogate packaging layout for 10 MeV e-beam irradiation; (b) electrons operating ranging and maximum dose in the designed surrogate holder (a).
thickness of 1.5 cm. Each holder contained two removable top and bottom Lucite custom-made absorbers. The Lucite absorber was cut to thicknesses of 3.0 cm to obtain the desired target doses of 1 kGy. Fig. 4a shows the layout of an individual apple-surrogate in its packaging for irradiation with the 10 MeV electrons. Fig. 4b shows the electron operating range passing the different absorber thicknesses. The 10 MeV electrons
penetrate through the Lucite absorber before they reach the apple-surrogate positioned inside the package. Points E and E 0 (Fig. 4b) show the operating range for the electrons after a penetration depth, t (cm), where the electrons have attenuated their average dose from 10 to 1.35 MeV to achieve the desired target doses. Prior to irradiation, the vacuum-packed apple-surrogate was introduced in the holder (Fig. 5a). Finally, the apple
Fig. 5. (a) Polystyrene holder with 4 cm thick Lucite absorber; (b) apple holders for 4 cm thick Lucite absorber before irradiation treatment.
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
1227
holder was taped to a cardboard box (L = 60.9 cm, W = 50.8 cm, and H = 10.2 cm) (Fig. 5b). The apple surrogates were produced in triplicate and irradiated. The irradiations were performed in batches ranked accordingly to its Lucite thickness value. The analyses were carried out 24 h after irradiation treatment to permit the apple surrogate’s chemical solution to stabilize as well as the HD-810 film contours. 2.7. X-ray irradiation experimental design using the 5 MeV LINAC The X-ray irradiation experiment with the 5 MeV LINAC also required use of a secondary holder for the apple-surrogates. Each apple-surrogate was separated by a distance of 16.5 cm center to prevent scattering effects. The height from the bottom of the holder to the center of the apple was 31 cm where the X-ray beam was located. The surrogates’ top surface (stem) was facing the X-ray beam to expose first the top region of the apple. Each apple-surrogate contained its respective radiochromic film slice. Fig. 6a shows the arrangement of three apple surrogates used for X-ray irradiation in a cardboard box holder prior to irradiation. Fig. 6b shows the apple surrogates ready to be irradiated using the 5-MeV X-rays. The X-ray beam exit window was located perpendicular to the conveyor’s direction of motion and separated a distance of 30.48 cm from the custom-made holder. The conveyor was set to travel at a speed of 60.96 cm/min to provide the 0.6 kGy target dose. Each apple-surrogate was positioned inside the holder gap resting on its side so that the apple center of symmetry (where the film was located) was perpendicular to the beam (Fig. 6b). 2.8. Apple-surrogate post irradiation handling After the irradiation experiments were completed, all surrogate samples and HD-810 film were stored in a dark room for a period of 24 h. Next, each apple-surrogate
Fig. 7. (a) GAFCHROMIC HD-810 film calibration curve (Green channel) at irradiation point P; (b) Green channel methyl yellow calibration curve using 3.175 mm thick slices and 1.35 MeV electrons.
sliced into 13.175 mm thick apple contours for a total of 16 slices per surrogate.
Fig. 6. (a) Apple surrogate holder for the X-ray irradiation experiment; (b) configuration layout for the apple-surrogate irradiation at a conveyor speed of 0.61 m/min.
1228
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
2.9. GAFCHROMIC HD-810 film and apple-surrogate slices scanning A Microtek ScanMaker 8700 Pro Series (Microtek USA, Carson, California) was used to perform transmission scans on both radiochromic film and surrogate slices. The scanner was set to capture an image at 300 pixels per inch (ppi). As a result, a 2.64 mega bytes image of 960 pixels · 960 pixels was scanned each time and saved using a Tagged Image File Format (TIFF). 2.10. GAFCHROMIC HD-810 film calibration The film apple 1 cm · 1 cm contours were irradiated using a cylindrical Lucite holder. The holder was designed and constructed to fit the round Farmer ionization chamber with the film in front of the ionization window volume.
A calibration curve for the films was produced by placing individual film slices and irradiating at specific doses. Each individual film was then scanned in transmission mode using the full dynamic range scale. The film intensity data was analyzed for the Green channel and a calibration equation was found between 0.05 and 1.50 kGy. Fig. 7a shows the calibration results. 2.11. Chemical dosimeter calibration The methyl yellow, chloroform, and paraffin wax solution was calibrated using Petri dishes with dimensions equal to the mold (2.54 cm in diameter and 7.6 cm height). The samples were irradiated with 1.35 MeV electrons using target doses from 0 to 0.5 kGy. After the irradiations, each sample was cut and scanned along the beam direction. Fig. 8b shows the calibration results for the green channel.
Fig. 8. (a) HD-810 film dose distribution and (b) HD-810 film dose-depth distribution in the apple-surrogate irradiated at 0.05 kGy with the VDG and a = 90 towards and against a 1.35 MeV plane source; (c) HD-810 film dose distribution; (d) HD-810 film dose-depth distribution in apple-surrogate irradiated at 0.05 kGy with the VDG and a = 45 towards and against a 1.35 MeV plane source; (e) HD-810 film dose distribution; and (f) HD-810 film dose-depth distribution in the apple-surrogate irradiated at 0.05 kGy with the VDG and a = 22.5 towards and against a 1.35 MeV plane source.
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
2.12. Image processing of GAFCHROMIC film and applesurrogate contours A computer code was written using Matlab (The Mathworks Inc., Natick, Massachusetts) for the HD-810 film contours and the apple-surrogate slices to extract and analyze the intensity data corresponding to the green channel of each image file. A 960 pixel · 960 pixel square matrix contained all the intensity data measurements for the green channel only. Each data point was first transformed into optical density by OD = log(I0/I), where I0 is the light intensity of an unexposed film, and I is the light intensity of an exposed film. Then, the dose values where calculated using the calibration equations shown in Fig. 7a and b for the film and apple-surrogate, respectively.
1229
The dose deposition profiles for the film contours and apple-surrogate slices were computed using the Matlab Image Processing Toolbox contouring algorithms, and later transformed to grayscale. 2.13. Energy deposition and dose depth distributions The dose data obtained for this experiment was calculated at the plane that traverses the apple-surrogate from top to bottom, and divides the surrogate in two symmetrical pieces so the radiochromic film response could be compared to that of the apple-surrogate. The dose distributions for both film and apple-surrogate depend on several common factors including the electron entrance angle, the distance from the electron exit window
Fig. 9. (a) HD-810 film dose distribution; (b) HD-810 film dose-depth distribution in the apple-surrogate irradiated at 0.1 kGy with the VDG and a = 45 towards and against a 1.35 MeV plane source; (c) HD-810 film dose distribution and (d) HD-810 film dose-depth distribution in the apple-surrogate irradiated at 0.1 kGy with the VDG and a = 22.5 towards and against a 1.35 MeV plane source.
1230
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
to the surrogate, the path electrons follow because of scattering and the maximum penetration range within the apple-surrogate. Dose deposition data and dose-depth contours were created to provide a visual reference for the different treatments studied. 3. Results 3.1. E-beam irradiation using 1.35-MeV VDG 3.1.1. GAFCHROMIC HD-810 film contours The energy deposition profiles and depth dose distributions around the apple-surrogate periphery (0–360), in its vertical plane of symmetry, (i.e., the apple is divided in half from top to bottom) are shown in Figs. 8 and 9 for the target doses of 0.05 and 0.1 kGy, respectively. Fig. 8a shows that the energy deposition profile at an orientation of 90 away from the apple’s longitudinal axis results in uniform irradiation on the left and right sides of the surrogate. However, the regions near the top and bottom region located between 70 and 90 and 240 and 300 did not receive any irradiation (Fig. 8b). The maximum penetration depth was approximately 4.20 mm from the surface at 180. Fig. 8b shows a high dose region between 210 and 240 below a depth of 0.76 mm. These high doses resulted from the continuous irradiation of such particular points due to the geometrical orientation during the irradiation process. The high doses present on the left and right bottom sides of the surrogate resulted from a higher number of electrons being directed towards these ends due to a shift of the electron beam distribution (Rivadeneira, 2004). The left side of the energy deposition contour in Fig. 8a shows greatest penetration depths than of the right side. This behavior was consistent throughout all samples (Rivadeneira, 2004) and was due to the swiveling motion of the apple-surrogate as it hung from the turning hook and rotated around its vertical axis. Only a very small fraction of the normal surface vectors will be parallel with the incident electrons resulting in a direct hit. Such points were expected to receive higher doses and greater penetration depths due to the absence of electron scattering events. Nonetheless, it was found that due to the beam distribution, higher number of electrons struck the apple-surrogate at the left and right bottom regions resulting in greater penetration depths despite of the higher occurrence of scattering events compared to the surrogate’s top. Fig. 8c shows that the dose deposition data was more uniform at a = 45 compared to a = 90 orientation angle away from the apple’s y-axis. The orientation angle of 45 produces a uniform dose deposition in areas located at the top and bottom of the surrogate. The bottom region (Fig. 8d) of the surrogate (270), however, did not show significant dose deposition, but wider regions (between 210 to 270 and 300 to 330) were irradiated compared to those in Fig. 8b.
The penetration depth (Fig. 8d) showed more uniformity than that irradiated at a = 90 (Fig. 8b). Both orientations showed a maximum penetration depth of approximately 4.57 mm from the surface at different regions. High dose regions at 45 orientation were minimized compared to the a = 90 orientation (Fig. 8a) due to a more uniform irradiation. The apple-surrogate was oriented closer to the e-beam incident angle resulting in more exposed regions (Fig. 8c). Even though the results in Fig. 8d show high dose distributions at 250 and 290, the energy deposition gradient appeared more uniform than at a = 90 and the penetration depths were similar at both orientations. The apple-surrogate orientation parallel to the electron beam at a = 22.5 (Fig. 8e) showed very similar results to that obtained at a = 45. It was expected that the applesurrogate exposure at an angle equal to that of the electron
Fig. 10. (a) Methyl yellow apple contour dose distribution; (b) methyl yellow apple contour dose depth distribution in the apple-surrogate irradiated at 0.05 kGy with the VDG and a = 45 towards and against a 1.35 MeV plane source.
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
beam would result in the best irradiation orientation. In particular, the closer the apple-surrogate is aligned to the beam direction the broader the areas of the apple that are exposed. However, problem areas were still located at the top and bottom regions, where several points were not exposed to irradiation. Fig. 8f shows that for a = 22.5, the surrogate absorbed the highest doses at the top left and right positions located at 60 and 120, respectively. At these locations, it is expected that the electron entrance angle be perpendicular to the surrogate surface resulting in high dose regions. The largest penetration depths at these positions demonstrate that the electrons penetrated nearly perpendicular to the surrogate surface. The effect of a higher dose in the energy deposition and penetration depth profiles, around the surrogate periphery, was studied using the previous apple-surrogate orientation
1231
with a 0.1 kGy dose. Fig. 9 shows the results for the 45 and 22.5 of orientations. As the top of the apple-surrogate was positioned closer to the electron beam plane at 0.1 kGy, the energy deposition data improved as shown in Fig. 9a and c irradiated at a = 45 and a = 22.5, respectively. The dose deposition data shown in Fig. 9a demonstrate that at a = 45 there was complete radiation treatment throughout the surrogate periphery (Fig. 9b). The unexposed regions at the lower energy (0.05 kGy) were completely covered using higher doses. At the top of the surrogate, Fig. 9b shows that between 80 and 120 the maximum penetration depth was at least 1.6 mm from the surface. At the bottom, between 250 and 300, the minimum penetration depth was 3.05 mm from the surface compared to less than 0.76 mm at 0.05 kGy (Fig. 9b). The apple-surrogate received high dose values at the top and bottom right and left corners located at approximately at
Fig. 11. HD-810 film: (a) dose distribution in kGy and (b) depth-dose distribution in kGy using Lucite thickness of 3 cm and a dual beam 10 MeV LINAC accelerator with a conveyor speed of 0.3 m/s (60 ft/min).
1232
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
45, 135, 225, and 315 regions, respectively. At a penetration depth of 2.30 mm from the surface, the dose distribution became uniform throughout the apple periphery. The surrogate irradiated at a = 22.5 and 0.1 kGy showed complete treatment of the apple periphery (Fig. 9c). Fig. 9d shows deeper penetration depth (5.33 mm from the surface) compared to the other two treatments. Nonetheless, the apple-surrogate absorbed higher doses at more spots within its surface compared to any other treatment. The dark areas shown below the penetration depth of 1.52 mm from the surface in Fig. 9d, demonstrate higher doses were absorbed with this orientation. The maximum penetration depths at 100 and 270 regions were at least above 3 mm from the surface ensuring uniform irradiation treatment. The parallel orientation to the electron beam provided uniform irradiation throughout the sample, but it appeared to be overexposed to that at 45.
The evaluation of food positioning strategies in front of the e-beam source demonstrated that tilting the apple-surrogate towards the e-beam ensures uniform dose distributions of the entire surface of the surrogate, even reaching the critical regions of the apple stem and calyx ends. This is important so the risk associated with bacterial contamination, linked to Escherichia coli O157:H7 that may be present in raw apples (Rangel, Sparling, Crowe, Griffin, & Swerdlow, 2005), may be reduced on the apple’s surface by electron beam irradiation. 3.1.2. Apple-surrogate contours The energy deposition profiles and depth dose distributions around the apple-surrogate periphery showed similar results to those obtained with the HD-810 film. Fig. 10a and b show the results for a = 45. Because of chloroform evaporation during the casting method, it was not possible to trace absorbed dose values
Fig. 12. HD-810 film: (a) dose distribution and (b) depth-dose distribution in kGy using a custom-made polystyrene holder and a 5 MeV LINAC accelerator with a conveyor speed of 0.01 m/s (2 ft/min).
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
in the top regions of the methyl yellow apple contours. Improvement in the casting technique should yield better results. Because the pouring orifice in the mold was exactly at the top, during the pouring process the top parts of the mold received less chloroform. The solidification process of the methyl yellow–chloroform–paraffin wax solution inside the mold was very slow, so less chloroform was present in the solution at the end of the pouring. For this reason, no irradiation treatment was detected near the top regions of the apple for both energies and all surrogate orientations towards the e-beam. In general, the methyl yellow surrogate 2-D slices have shown to detect radiation between 0 and 0.5 kGy. The apple-surrogate’s main advantage over the HD-810 film consists in its ability to visually show the absorbed dose in 3-D (Rivadeneira, 2004). 3.2. E-beam Irradiation using 10-MeV LINAC Fig. 11a shows the dose deposition data for the radiochromic film with a Lucite thickness of 3 cm (target dose 1 kGy). The electrons penetrated quite uniformly from the bottom and the top until losing all their energy at the center of the apple-surrogate. The dose depth profile of the electrons measured from top to bottom at specific locations in the x-axis (Fig. 11b). The top entrance dose at the left and central planes were very close showing the electron beam uniformity (Fig. 11a). The right side (x = 64 mm) showed less irradiation; however, this lower dose may have been a result of the positioning of the apple-surrogate inside the box. At the bottom of the apple-surrogate (Fig. 11a), the film dose distribution agrees with the target dose (1 kGy). Fig. 11b shows high dose regions at the top left and right positions of the apple-surrogate. These differences may be due to the positioning of the apple-surrogate in the box and surrogate complex shape. The top left side of the apple-surrogate that shows the highest absorbed dose is closer to the Lucite absorber compared to its top right side. As a result, the electrons traveled less before reaching the top-left part of the sample hence depositing more energy. 3.3. X-ray irradiation using 5 MeV LINAC Fig. 12a shows the energy deposition contour for the radiochromic film. The surrogate received a maximum dose of approximately 0.6 kGy at the top. Fig. 12b shows several curves at penetration depths of 9, 16, 34, 50, and 68 mm from the surface. Notice the dose decrease as the X-rays traveled through the apple at the different depths. The entrance dose was approximately 0.6 below 9 mm from the surface and the exit dose at the bottom of the surrogate was approximately 0.4 kGy at a depth of 68 mm (Fig. 12b). The dose at each particular penetration depth remained uniform across the x-axis.
1233
4. Conclusions The use of a surrogate (to represent a complex-shaped food) composed of paraffin wax, chloroform, and methyl yellow was successful to detect absorbed radiation using 1.35 MeV electrons. A solid-state three dimensional ‘‘apple’’ surrogate helped to visually determine absorbed dose in the surrogate vertical axis of symmetry throughout its periphery. The chemical solution response to absorbed dose was found to be non-linear for 1.35 MeV electrons for doses ranging from 0.01 to 0.5 kGy. The optimal operating range for the chemical solution was between 0.1 and 0.3 kGy. A transmission flatbed scanner proved to be successful and economical tool to obtain dose deposition and depth dose profiles of apple-surrogate slices 3.175 mm thick using Matlab. Image Processing was performed in Red Green Blue (RGB) color space where it was found that the Green channel offered the best response to absorbed doses between 0.01 and 0.5 kGy using 1.35 MeV electrons. In general, tilting the apple-surrogate towards the ebeam source resulted in more uniform irradiations. The surrogates could be over-exposed when oriented exactly parallel to the e-beam at a = 22.5. The 45 orientation ensured uniformity throughout the surrogate periphery for wider penetration depths without overexposing the product. The 90 orientation was found to be the least effective at the top and bottom regions. The methyl yellow apple-surrogate proved to be a good visual of radiation dose in any direction where the radiation treatment was performed. For a Lucite thickness of 3 cm (target dose of 1.0 kGy), the HD-810 film confirmed a very uniform irradiation treatment at the top and bottom regions with a maximum dose of 1.2 kGy located at the top left region of the apple. As expected, the dose decreased in a linear tendency from the surface to the highest penetration depth until the electrons lost all their energy. The X-ray irradiation completely penetrated the entire apple-surrogate from top to bottom. The highest doses observed with the film were located at the top region of the surrogate where the X-ray beam hit first the apple. The entrance dose throughout most of the top region was approximately 0.6 kGy while the dose at the bottom region was approximately 0.4 kGy. X-ray irradiation was uniform across the apple. The dose decreased approximately 25% from the top to the bottom in the radiochromic film.
Acknowledgements This research has been funded by the USDA CSREES Project # 2002-03935 and the USDA NRI Food Safety program Project # 2002-02411. The authors would like to thank Dr. Leslie Braby (Nuclear Engineering) for providing expertise and technical assistance with the irradiation experiments.
1234
R. Rivadeneira et al. / Food Control 18 (2007) 1223–1234
References Borsa, J., Chu, R., Sun, J. S., Linton, N., & Hunter, C. (2002). Use of CT scans and treatment planning software for validation of the dose component of food irradiation protocols. Radiation Physics and Chemistry, 63(3–6), 271–275. Bresica, G., Moreira, R., Braby, L., & Castell-Perez, E. (2003). Monte Carlo simulation and dose distribution of low energy electron irradiation of an apple. Journal of Food Engineering, 60, 31–39. ICRU (1984). Radiation Dosimetry: Electron Beams with Energies Between 1 and 50 MeV. International Commission on Radiation Units and Measurements, Report No. 35. ICRU (1989). Tissue Substitutes in Radiation Dosimetry and Measurement. International Commission on Radiation Units and Measurements, Report No. 44.
Kim, J., Rivadeneira, R. G., Castell-Perez, M. E., & Moreira, R. G. (2006). Development and validation of a methodology for dose calculation in electron beam irradiation of complex-shaped foods. Journal of Food Engineering, 34(3), 359–369. Mohsenin, N. N. (1986). Physical properties of Plant and Animal Materials (second ed.). New York: Gordon and Breach Science Publishers. Rangel, J. M., Sparling, P. H., Crowe, C., Griffin, P. M., & Swerdlow, D. L. (2005). Epidemiology of Escherichia coli O157:H7 Outbreaks, United States, 1982-2002. Emerging Infectious Diseases, CDC, 11(4), 606–609. Rivadeneira, R. (2004). Development of a chemical dosimeter for electron beam food irradiation. Master Thesis. Department of Biological and Agricultural Engineering. Texas A&M University. College Station, TX. Tsoulfanidis, N. (1995). Measurement and Detection of Radiation. New York: Taylor & Francis.