1. Introduction:
Several hundred industrial gamma irradiation facilities (especially 60Co source) are in operation worldwide, as listed in the IAEA Database[1].
Tunisian legislation, which has been promulgated in April and May 2002, concerned, respectively, sterilization of medical devices with single use (syringe, surgical gowns, suture...) and the radiation processing of fresh-foods (fruits, vegetables, salads and meat products, potato, garlic and onions...) and well as dry-foods (dehydrated or dried spices, dried fruits and dried vegetables...).
Over the years, the decrease of the source activity and thus the dose rate affects the irradiation time. Consequently, it is necessary to compensate the dose rate decrease by incrementally increasing the irradiation time, which has a significant impact on the cost and scheduling of the radiation processing. This causes doubt on the accuracy and the homogeneity of the dose distribution in the irradiated product.
The safe management and the dosimetry quality control of the source during the first half life period (commissioning at 1999) was studied by different ways[2-5].
Since the Tunisian 60Co source is near the end of life, the present study is focused on sterilization of medical devices with single use. The objective was to evaluate the efficiency of sterilization of medical devices by gamma irradiation at the end of source life.
The procedures established at the 60Co source process control for the irradiation of medical devices had been consistent with the relevant Good Radiation Processing (GRP) requirements. In addition, IAEA Guidance[6] and ISO 11137 standards under parts: 1[7], 2[8], and 3[9] related to radiation sterilization for healthcare products are provided.
At the third half life of the 60Co source, and due to the very long irradiation time necessary to deposit the desired dose in the irradiated product, it becomes necessary to control the quality of the processing in terms of absorbed dose rate and dose rate uniformity to verify the dose distribution as well as the dose uniformity with the actual activity equal to 384 TBq (November 2016).
The experimental measurements are expensive and need a lot of time. Thus, it was useful to resort to GEANT4 in dose rate calculations. As already demonstrated by Kadri et al.,[10], the excellent agreement seen between the dose rates calculated by GEANT4 and the measured values allows us to apply such code in order to determine the 60Co irradiation facility efficiency after 16 years of use.
In order to better understand the spatial variation of the dose and dose rate distributions during a radiation sterilization process, the dose distribution around the 60Co source was estimated. For this purpose, Monte Carlo simulations were performed to model and simulate the irradiation room, including the 60Co source with its container. An extensive dose rate mapping study was carried out, then, the absorbed dose rate uniformity was deduced. Minimum and maximum doses measured experimentally in the medical product over the years were plotted.
To our knowledge, the measurement of the dose distribution delivered by the 60Co industrial irradiator and the efficiency of medical devices irradiation at the end of source life had never been done before.
2. Materials and Methods:
2.1 Description of the 60Co irradiator:
The Pilot-scale gamma irradiation facility is utilizing 60Co pencils. The maximum amount of activity which can be loaded into the facility was 3.626 PBq on 09/04/1999 (date of commissioning). The C-188 MDS Nordion 60Co gamma irradiator is a 'category II' facility. Source rack is a panoramic dry source storage for continuous operations (see Fig.1). It contains eight 60Co source pencils of 7.2 mm and 410 mm of diameter and height, respectively. The sources, doubly encapsulated in a welded stainless steel cage, are cylindrically arranged on two pitch-circles of 102 mm and 56 mm diameters, respectively.
The design of cylinders holding the source pencils with 20 housing allows future-loading. The size of the irradiation chamber is 6 m x 6m x 3 m. It is surrounded by walls made of 1.7 m high density concrete. When sources are not in use, they are stored in a lead cylindrical shield container with a radius of 0.332 m and height of 1.166 m. The irradiation process can be monitored from the control room adjacent to the irradiation chamber. More description of the facility is shown in reference[2].
2.2 Dosimeters:
Given its sensitivity to the absorbed dose range (5–50 kGy) linear response, Red Perspex dosimeter type 4034 (Harwell-UK), is suitable for radiation processing applications, especially in sterilization of soft materials such as medical devices with single use[11, 12].
The active substance of dosimeters is composed of a dye dispersed in the Poly-MethylMethAcrylate (PMMA) and sealed in polyethylene-aluminum sachets, with sheets cut down into plaques of 30 mm x 11 mm of size and 3 mm of thickness (±0.55). Gamma radiation leads to the ionization of PMMA and the added dye. The radicals of ionized polymer react with the dyed molecules to produce an optical absorption in the visible spectrum.
The perusal of dosimeters was realized 24 hours after irradiation using the Aerial Optical Dosimetry Equipment at 640 nm wavelength[13].
The irradiation has been performed in air at the 60Co gamma-irradiation facility at the dose rate: 100 Gy/min[2]. The dose rate was established with the alanine/EPR dosimetry system in term of absorbed dose traceable to the National Physical Laboratory, UK[12]. Before the experiment, the dose rate was verified by a standard Fricke dosimeter.
Uncertainties of measurements at 1σ (95% confidence interval) are evaluated following ASTM Standards[12] and was found to be ± 4.5% for the calibrated dosimeter.
2.3 Simulated setup in the Geant4 Monte Carlo code:
GEANT4 (GEometry ANd Tracking, version 4)[14-15] is used in te present study.
For the dose rate calculation, the geometry of the 60Co facility is constructed using the DetectorConstruction class of GEANT4. For this purpose, the standard geometry bodies included in GEANT4 were used while respecting the dimensions and the real compositions.
In this work, we are interested in the weak decay of the 60Co source and electromagnetic interactions of generated gammas (γ1 and γ2 of the 60Co) with matter (air and dosimeters). Particles, processes, and the default Cut Value (1 mm) are implemented using the G4Physicslist class. The generation of gamma photons with their characteristics is modeled using the G4PrimaryGenerationAction class.
PMMA dosimeters were assumed to be spheres in order to obtain the same surface exposed to the radiation. Based on statistical studies, the most appropriate radius of these spheres is 4 cm [16]. This radius leads to a low statistical error (less than 5%).
The mass of the dosimeter is defined as md=d×V=1.19×(4π r3/3)
With: d: Material density of the dosimeter (g/cm3)
V: Dosimeter volume (cm3)
r: Sphere radius (cm)
After defining the parameters of particles generation, the absorbed dose rate were calculated by transforming the deposited energy in the dosimeter during a series of events, N, as follows:
With:
Ed: Total energy deposited in a dosimeter (MeV),
C: Conversion factor (J/MeV), equal to 1.602 x 10-10,
A: Activity in (Bq),
N: Number of generated events, equal to 107,
2: Factor to take into account two emitted photons by the 60Co source,
t: The irradiation time, equal to 60 min.
3. Results and discussion:
3.1 Experimental results:
In this section, we are interested to follow the absorbed dose in a routine processing of single use medical products from 2009 to 2015.
3.1.1 Process control in the product over the years:
Besides the transversal variation of the dose, there is also the dose variation in the lateral direction. Both types of the dose variation contribute to the non-uniformity of the dose delivered to the product. However, one must keep in mind that dose variation in the irradiated product is unavoidable. The concept of Dose Uniformity Ratio (DUR), defined as the ratio of the maximum to the minimum of dose in the irradiated product[16], is used in the present work.
The recorded maximum and minimum absorbed doses during the third source half life allowed the deduction of the DUR.
Dosimeters were placed at the minimum and maximum dose positions during routine irradiation of a process load (Fig. 2). The maximum and the minimum doses are determined after a cartography of the irradiated product.
In this part of study, the Red Perpex dosimeters described in the section (2.2) were used. The dosimeters associated with the device are analyzed after irradiation of the product is complete, to confirm that the required dose has been delivered.
The dose distribution measurements were carried out with the same single use medical product with a density of 0.114 g/cm3. The size and the mass of the product boxes are 58 x 39 x 31 cm and 7 kg, respectively. Sixteen boxes with dimensions of 80 cm x 58 cm x 62 cm were used for the processing batch. Four product boxes were used for irradiation (see Fig. 1).
In order to reach its desired dose and to improve dose distribution, the irradiated product has to receive dose increments. Similar conditions of irradiation were respected, except the irradiation time and the dose rate, which were adjusted to compensate the source decay.
The regions of minimum and maximum absorbed doses within the product were determined and the DUR was deduced.
The results of irradiation of several batches of product, done during the third source half life period (2009-2015), are shown in Fig. 3. Consistency of dose received by irradiated products per irradiation is demonstrated.
3.1.2 Dose rate and processing irradiation time compensation:
Currently, the irradiation time of the product is much longer in comparison with the first years. It took a few days in the beginning of source installation and now takes weeks for the sterilization of single use medical products.
Fig. 4 shows evidence of satisfactory processing. Even with the decrease of source activity, compensation with the increase of the irradiation time and the decrease of dose rate is a proper manner of processing.
Process interruption is sometimes encountered. Some interruptions may be planned, for example double side irradiations may be used to improve the dose distribution. Other interruptions may be the result of unplanned irradiator shutdowns[17].
3.1.3 Dose Uniformity Ratio:
The dose variation in the lateral direction of the product is presented in the DUR value, which is a method of describing the non-uniformity of the dose delivered to the product. Fig. 5 recapitulates the deduced DUR as the ratio of the measured maximum and minimum doses delivered to the product over the third source half life period.
Keep in mind that the dose limit ratio is between 1.5 and 3 for many applications[18] and sometimes even larger depending on the product and the process. We assume that dose variation in the irradiated product is small and the dose delivered to the product is uniform.
3.2 Monte Carlo calculations:
· In the first part, validation of the GEANT4 setup was carried out by comparing our results of absorbed dose rates to those obtained previously by reference[11]. A comparison between dose rates at the first source half life (2004) and at the third source half life (2015) was also studied.
· The second part is dedicated to the absorbed dose rate calculations resulting from the gamma energy deposited in PMMA dosimeters (simulated dosimeters put equidistantly around the source). This study concerns the years between 1999 and 2018: commissioning and planned pencils reloading date.
· The third part is focused on the study of the dose distribution uniformity during the operation years.
3.2.1 Validation of the GEANT4 setup:
Previously, Kadri[16] measured the dose rates experimentally and compared them with GEANT4 calculations in seven critical points. PMMA's dosimeters were placed in the irradiation room at 157 cm of height. Seven points are sorted from the most distant to the closest one to the source.
In the present work, dose rate calculations were carried out using GEANT4. The hardware platform used to run the GEANT4 version 9.2 code is a Linux (Scientific Linux CERN 5) Personal Workstation with 8 GB RAM and a 3.4 GHz CPU. Similar results (less than a 3% difference) were obtained compared with those determined by reference[16], which used the version 6.1 to calculate dose rates at different dosimeters' locations.
As illustrated in Fig. 6, the absorbed dose rate distribution is well preserved. The closest (Position 7) and the farthest (Position 1) points from the source, have respectively the highest and the lowest absorbed dose rates, as expected. Points 2, 4, and 6 are equidistant with regard to the source, which explain their approximately equal dose rates. Points 3 and 5 are symmetric with respect to the point 4. However point 3 is closer to the irradiation room corner, which explains its slight elevation compared to point 5.
In the next part, we calculate the dose rates at the seven points in the year 2015, which corresponds to the third source half life.
The total activity of the source in 2015 is about 430 ± 0.18 TBq. It is lower than the activity of a single pencil in the year of the installation.
At the third half life (2015) of the 60Co source, dose rates are around one third to one fourth compared to the values obtained during the first half life (2004), as shown in Fig. 7, which is explained by the source decay.
3.2.2 Study of the absorbed dose rate distribution over the years:
The present study is focused on the dose rate distribution uniformity inside the irradiation room from the year of commissioning (April, 1999) until the expected year of pencils reloading (April, 2018). This study is very important for the transversal isotropy verification of gamma ray emission.
To verify that the angular distribution isotropy is ensured in all directions of the irradiation room, a spherical PMMA filled scoring dosimeter geometry is implemented for dose rate calculations. G4PVPlacement is used to place the 120 dosimeters around the source on 10 circular beam levels with radii extending from 40 to 200 cm at the same height of 157 cm (midplane of irradiation chamber)[2]. In each level, the angle between two consecutive dosimeters is 30 ° (Fig. 8).
A comparison between the mapping of the calculated absorbed dose for years 1999 and 2015 is illustrated in Fig. 9. The dose rate distribution keeps the circular behavior, and the gamma isotropy is insured at the conveyer level.
The excellent transversal isotropy of the gamma ray emission in the year 1999 remains similar after 16 years of the source use. For both years (1999 and 2015), the deposited energy at the level of conveyer keeps approximately the same value around the 60Co source, which explains the good dose distribution inside the irradiation room. The contribution of photon scattering by the wall around the source was demonstrated lower than statistical errors[3]. At the third source half life, the anisotropy of the gamma ray emission is preserved far away from the conveyer level.
3.2.3 The dose distribution uniformity over the years at the level of the conveyer:
Since the year of the installation (1999) until the year of pencils reloading (2018), the dose rate distribution showed a clear decrease. For this purpose, the value of the level dose rate uniformity (at the conveyer level) is deduced.
The level dose rate uniformity is defined as the ratio between the maximum and the minimum dose rates taken at the same distance from the source (Dmax/Dmin)conveyer level.
As shown in Fig. 10, the level dose rate uniformity is close to the unity till the end of the source life. Replenishment of source pencils at 2018 to compensate the lost activity could preserve the level uniformity close to the unity.
4. Conclusion:
The application of GEANT4 for dose mapping in the 60Co irradiator has been found very promising because of its accuracy, fast simulation, and because it is a cheaper way to validate measurements and optimize dosimetry quality control.
Good agreement was obtained over the period starting from the irradiator commissioning in 1999 until the third source half life, which confirmed the stability of the dosimetry measurements and the Monte Carlo modeling and reconfirmed the half-life of 60Co.
Although, the decrease of the source activity can be compensated by the increase of the irradiation time. In practice, this is not possible for certain cases, e.g., processing to preserve fresh food in terms of microbiological safety becomes not feasible, and the range of irradiated products is limited to only soft materials.
To add perspective, the study of the adequate arrangement and pencils activities are underway to compensate the decrease of the source activity and maintain an excellent isodose distribution inside the irradiation room.
IAEA-DGPF/CD, Directory of Gamma Processing Facilities in Member States. A technical document issued by the International Atomic Energy Agency (2004)
. http://www-pub.iaea.org/MTCD/publications/PDF/dgpf-cd/PDF/Contents.pdf. AccessedDose measurements for characterization of semi-industrial 60Co gamma irradiation facility
. Radiat. Meas. 41, 201-208 (2006). doi: 10.1016/j.radmeas.2005.03.003Validation of GEANT4 code of CERN a predictive tool of dose rate measurement in the Tunisian gamma irradiation facility
. Radiat. Phys. Chem. 74, 102-110 (2005). doi: 10.1016/j.radphyschem.2005.02.006Monte Carlo improvement of dose uniformity in gamma irradiation processing using the GEANT4 code
. Nucl. Instrum. Meth. B. 239, 391-398 (2005). doi: 10.1016/j.nimb.2005.05.052Monte Carlo optimization of some parameters in gamma irradiation processing
. Nucl. Instrum. Meth. B. 245, 459-463 (2006). doi: 10.1016/j.nimb.2005.11.142IAEA-TECDOC-539, Guidelines for industrial radiation sterilization of disposable medical product. A technical document issued by the international Atomic Energy Agency (1990)
. http://www-pub.iaea.org/MTCD/publications/PDF/te_539_web.pdf. AccessedApplication of GEANT4 code in gamma irradiation processing
. Int. J. Theor. Phys. 14, 1-15 (2009).Dosimetry systems for radiation processing
. Radiat. Phys. Chem. 46, 1163-1174 (1995).A dedicated tool for the quality assurance in the field of radiation processing
.GEANT4—a simulation toolkit
. Nucl. Instrum. Meth. A. 506 (3), 250-303 (2003). doi: 10.1016/S0168-9002(03)01368-8GEANT4 Developments and Applications
. IEEE. T. Nucl. Sc. 53 (1), 270-278 (2006). doi: 10.1109/TNS.2006.869826Monte Carlo studies of the Tunisian gamma irradiation facility using GEANT4 code, Appl
. Radiat. Isotopes. 64, 170-177 (2006). doi: 10.1016/j.apradiso.2005.07.009Process control and dosimetry in a multipurpose irradiation facility
. Radiat. Phys. Che. 55, 781-784 (1999).An Empirical Model for Predicting Harwell Dose Variation During Gamma Process Interruption Using Experimental Design
. MAPAN-Journal of Metrology Society of India. 30 (2), 85-90 (2015). doi: 10.1007/s12647-014-0129-6