Abstract
This work aimed to determine the linear and mass attenuation coefficients of soy-lignin bonded Rhizophora spp. particleboard intended for use as a phantom material using Monte Carlo GATE simulation. At a desired density of 1.0 g cm-3, particleboard constructed of Rhizophora spp. wood trunk bonded with soy flour and lignin was created. The sample’s elemental composition was identified using energy dispersive X-ray spectroscopy. The GATE software was used to simulate the setup with the histories of 1 × 107, and comparison was made between the experimental and simulation data. The disparities between the linear and mass attenuation coefficients of the samples experimentally measured and calculated using GATE at low energy photons were quite small. The result revealed a good agreement between the experimental and simulation data, and the attenuation coefficients were in close proximity with XCOM of water. The outcome revealed GATE adequacy for validation of attenuation coefficient measurement in bioresources phantom material for medical physics application.
Download PDF
Full Article
Attenuation Coefficients of Soy-Lignin Bonded Rhizophora spp. Particleboard as a Potential Phantom Material Using Monte Carlo GATE
Siti Hajar Zuber,a,* Muhammad Fahmi Rizal Abdul Hadi,b Nurul Ab. Aziz Hashikin,b Mohd Fahmi Mohd Yusof,c and Mohd Zahri Abdul Aziz d
This work aimed to determine the linear and mass attenuation coefficients of soy-lignin bonded Rhizophora spp. particleboard intended for use as a phantom material using Monte Carlo GATE simulation. At a desired density of 1.0 gcm-3, particleboard constructed of Rhizophora spp. wood trunk bonded with soy flour and lignin was created. The sample’s elemental composition was identified using energy dispersive X-ray spectroscopy. The GATE software was used to simulate the setup with the histories of 1 × 107, and comparison was made between the experimental and simulation data. The disparities between the linear and mass attenuation coefficients of the samples experimentally measured and calculated using GATE at low energy photons were quite small. The result revealed a good agreement between the experimental and simulation data, and the attenuation coefficients were in close proximity with XCOM of water. The outcome revealed GATE adequacy for validation of attenuation coefficient measurement in bioresources phantom material for medical physics application.
DOI: 10.15376/biores.19.4.8920-8934
Keywords: Monte Carlo simulation; GATE; Attenuation coefficient; Phantom material; Plant-based
Contact information: a: Centre of Diagnostic, Therapeutic and Investigative Studies, Faculty of Health Sciences, Universiti Kebangsaan Malaysia, 50300, Kuala Lumpur, Malaysia; b: School of Physics, Universiti Sains Malaysia, 11800, Penang, Malaysia; c: Faculty of Health and Life Sciences, Management & Science University, 40150 Shah Alam, Selangor, Malaysia; d: Pusat Perubatan USM, Universiti Sains Malaysia, 13200, Penang, Malaysia; *Corresponding author: hajarzuber@ukm.edu.my
INTRODUCTION
Water is frequently utilized as phantom material because of its mass density, which is similar to that of human tissue. Because water cannot always be used as a phantom due to its shape, many water-equivalent phantoms, such as acrylic, have been developed. Due to the instability of plastic-based phantom materials, the majority of acrylic phantoms still fail to exhibit satisfactory scattering and absorption capabilities. It is also well known that solid materials, such as Perspex®, which makes up the RANDO phantom, are often utilized as phantom materials, but their mass density and attenuation characteristics have been found to be very different from those of water and human soft tissue (Khan and Gibbons 2014). In addition, many water-equivalent substances such as polystyrene, acrylic, and solid water were also created. Despite their many benefits, they do have some drawbacks, such as the difficulty to accurately simulate actual human soft tissue (Yohannes et al. 2012).
Further research has resulted in replacements for the widely used phantom, and more natural materials have been employed to make it. Previous studies have reported Rhizophora’s ability as a potential phantom material in medical physics applications (Bradley et al. 1991; Sudin et al. 1988; Yusof et al. 2017c). Previous studies also found that Rhizophora spp. showed a linear attenuation coefficient of 0.0212 mm-1, which is near the value of 0.0205 mm-1 by water (Samson et al. 2020a, 2020b). The potential of Rhizophora spp. as phantom material is also supported by the elemental composition that mimics human soft tissues (Banjade et al. 2001; Marashdeh et al. 2011).
Previous research on Rhizophora spp. demonstrated the feasibility of this particular mangrove tree as a phantom material (Banjade et al. 2001; Abuarra et al. 2014; Tousi et al. 2014; Taghizadeh Tousi et al. 2015; Ababneh et al. 2016; Yusof et al. 2017a). However, because of its instability in terms of shape and homogeneity, particleboard made from wood is considerably preferable over raw wood for the fabrication of phantom material. To increase the physical and mechanical strength of particleboard made from different types of wood over the years, adhesive is frequently added to the manufacturing process. Although phenol formaldehyde resin (PF) is one of the common bonding substances used in the wood industry, many researchers have resorted to alternatives due to its recognized negative effects on the environment and human health. This phenol-based resin is highly hazardous when ingested, inhaled, or comes into contact with the eyes. It is also a probable cause of cancer (Nelson et al. 1986; Partanen et al. 1990; Gerberich et al. 2000; Beane Freeman et al. 2009).
In an effort to replace phenol-based resins and lessen the negative effects, several researchers concentrated on bio-based adhesives, and a variety of adhesives were developed and studied. Among many others, it has been established that protein- and carbohydrate-based adhesives are acceptable for use in the production of particleboard. Soy protein had been widely used as an alternative bonding agent to replace petroleum-based resins, whereas lignin as adhesive made it more accessible to the fibre during hot pressing as it redeposited during mixing (Boon et al. 2019). Both the adhesives are used for particle bonding and contribute to the particleboard’s increased structural and mechanical durability. Particleboard’s rough and irregular surface has led to increased efforts to create a better and smoother board surface, reducing the air gap between each slab. This is crucial for the dosimetry study because an air gap could result in inaccurate dose measurement. To generate correct dosimetry data when compared to human tissue, phantom material characterization is crucial. Several experiments must be conducted, including those on the physical and mechanical properties, characterization, mass density, and attenuation properties, to accommodate the unique characteristics of human phantom material (Zuber et al. 2020; Binti Zuber et al. 2024). This is critical, particularly in radiation studies where the phantom must have characteristics that are comparable to those of human tissue (Sharma et al. 2023), and the best comparison that can be made is with water, as provided by the Photon Cross Sections Database (XCOM) value (Marashdeh et al. 2011).
The mass attenuation coefficient of any material must be determined for medical physics applications because it reflects the material’s properties and radiation study appropriateness. The attenuation properties of a substance are extensively studied in industrial, environmental, and agricultural studies in addition to medical research (Marashdeh et al. 2015). Ionizing radiation’s ability to attenuate as it travels through a material is assessed, and the coefficient can be calculated using a variety of techniques, including direct measurement and the XCOM computer programme. The attenuation coefficient can be directly measured using proton-induced X-ray emission (PIXE) (Abdullah et al. 2010), high energy gamma photon (Yusof et al. 2016, 2017b), X-ray fluorescent (XRF) techniques (Marashdeh et al. 2012; Alshipli et al. 2018).
X-ray fluorescence (XRF) spectroscopy is a non-destructive analytical technique that can be used to determine the elemental composition of a material. This technique makes use of fluorescent X-ray emissions from the sample that are detected after they are excited by a primary radiation source. By configuring the spectrometer to meet the requirements for the measurement and collection of counts at the detector for attenuation studies, this method can measure a material’s linear attenuation coefficient (Shakhreet et al. 2009; Marashdeh et al. 2012; Hamid et al. 2017; Yusof et al. 2017b). Previous experiments have demonstrated the effectiveness of this design, and the materials employed include wood particles, Perspex®, and other phantom materials in radiation studies (Yusof et al. 2017b; Alshipli et al. 2018).
The employment of software and algorithms for the forecast or validation of any experimental investigation is crucial for researchers in the age of developing computer technology (Sharma et al. 2023). The Monte Carlo (MC) method in medical physics involves using statistical simulations to model and analyze complex systems, such as radiation transport in medical imaging. Since the 1930s, when the MC method was first introduced as a computing tool, it has been widely employed, making use of the simulating system to solve physics and mathematical issues. The paradigm change brought about by developments in statistics and physics sparked the creation of tactics and measurements utilizing MC, particularly to validate against diverse experimental findings.
Geant4 Application for Emission Tomography (GATE) is a well-known open-source MC simulation platform that enables straightforward and user-friendly dosimetry, imaging, and radiation simulation in the same environment (Jan et al. 2011, 2004). Nuclear medicine, radiation therapy, and dosimetry have all benefited from the development of the MC toolset, and the creation of GATE now makes it possible to do more broad simulation in a dosimetry setting (Ljungberg and Strand 1989; Harrison et al. 1993; Agostinelli et al. 2003). GATE application is a part of geant4 toolkit that simulates the interaction between same or different matters and able to provide high-level features to ease the simulation and design. GATE’s unique scripting approach enables the smooth and easy development of geometry, resulting in accurate simulation in realistic setup (Agostinelli et al. 2003). Although GATE has been widely verified and utilized for many studies using single photon emission computed tomography (SPECT) and positron emission tomography (PET) imaging, there are only a limited number of articles that can demonstrate its application and dependability in dosimetry studies.
This work aims to determine the linear and mass attenuation coefficient of the soy-lignin bonded Rhizophora spp. particleboard for low energy photons using the Monte Carlo GATE simulation, and comparison was made with experimental data. The elemental fraction of particleboard used in the simulation was obtained from scanning electron microscopy with energy dispersive X-ray analysis (SEM-EDX).
EXPERIMENTAL
Preparation of Particleboard
Wood logs from Rhizophora species were acquired from a coal company in Kuala Sepetang. To create samples with various particle sizes, the logs underwent a number of processing steps, including washing, drying, debarking, grinding, and sieving. Slabs of particleboards bonded with different percentages of soy flour and lignin were then constructed at a target density of 1.0 gcm-3. Rhizophora spp. wood particles were prepared at 1.39 to 6.47% L moisture content at three different particle sizes (0 to 103 µm, 104 to 210 µm, and 211 to 500 µm). Soybean flour (type I) and lignin (alkali, low sulfonate content) (both in powder form), purchased from Sigma Aldrich, Germany, were prepared at two different percentage mixtures, 6% (4.5% soy flour and 1.5% lignin) and 12% (6% soy flour and 3% lignin). Hot pressing was used for the fabrication of particleboard at 200 °C and pressure of 20 MPa for 20 min. The samples were prepared accordingly at (5.0 × 5.0 × 0.5) cm3 dimension.
Experimental Setup for Attenuation Coefficient Study
The experimental attenuation coefficient study was performed using the XRF configuration, with energy calibration carried out using (low energy Germanium) LEGe detector (Hamid et al. 2017). An annular Americium-241 (241Am) source with a nominal activity of 100 mCi was used in conjunction with four metal plates – Niobium (Nb), Molybdenum (Mo), Palladium (Pd), and Tin (Sn) to determine the mass attenuation coefficient of the soy-lignin bonded Rhizophora particleboards at the energy ranges of 16.61 to 25.27 keV. Figure 1 visualizes the experimental setup for the attenuation study (Hamid et al. 2017). The LEGe detector (CANBERRA) was employed to collect the photons transmitted, and the output pulses were amplified by an amplifier (ORTEC 572). Multichannel analyzer (MCA-3 series) was used to collect the spectrum for 60 s. The experimental results were reported in a previous study (Zuber et al. 2021), and comparison with GATE simulation was made in this work.
Fig. 1. Experimental setup for the attenuation study using 241Am
Elemental Analysis Using Energy Dispersive X-Ray (EDX) Spectroscopy
For this analysis, Rhizophora spp. particleboards were prepared at 1.0 g·cm-3 target density with dimension (1.0 × 1.0 × 0.5) cm3. Then, EDX (Field Emission Scanning Electron Microscope (FESEM) – FEI Nova NanoSEM 450 with EDX (FEI Company, Hillsboro, OR, USA) was performed to determine the elemental composition of the samples, where they were mounted onto specimen holders and examined under vacuum conditions using the scanning electron microscope, with weight percentage of elements recorded accordingly.
Monte Carlo Toolkit: GATE for the Simulation
The model of the computer employed in this study was a Lenovo H30-50 with Linux Mint 19 Tara 64-bit operating system (OS). GATE v8.2 with geant4 v10.05.p01 and Root v6.14/06 platform was used in the simulation. The SPECTHead example was modified for this work, and the input file created by GATE simulates the experimental configuration utilizing macro files with a variety of commands. The geometry setup for the simulation study is shown in Fig. 2.
Figure 3 provides a bird eye view of the setup with photon energy directed towards the sample in GATE simulation.
Fig. 2. Geometry setup for the simulation study
The challenge for Monte Carlo studies is to correctly quantify all the variables so that the internal and external influence of the geometry can be consistently detected. A 20‑item checklist – RECORDS – Reporting of Monte Carlo Radiation Transport Studies – was included in this work, in an effort to improve the quality of MC study as proposed by AAPM Research Committee Task Group 268 (Sechopoulos et al. 2018). Table 1 reports the RECORDS checklist.
Command scripts were developed in various macro files, including actors, beam, geometry, physics, visualization, and main, which contains the entire ordered set of commands required to perform the simulation. The master macro was divided into three sections – data, mac, and output.
Output macro files made use of the ROOT graphical user interface TBrowser to analyse and visualize simulation results interactively. The geometry of the setup was developed based on the experimental setup, with detailed measurements prior to the pilot run to validate the energy peak with a predetermined energy window.
Fig. 3. (a) Bird’s eye view (BEV) of the geometry setup for GATE simulation, and (b) BEV of the photon energy directed toward the sample
Table 1. RECORDS Items Checklist for Monte Carlo Study
XRF = x-ray fluorescence; CPU = central processing unit; GPU = graphics processing unit; EM = electromagnetic; VRT = variance reduction technique; AEIT = approximate efficiency improving technique
In this work, soy-lignin bonded Rhizophora spp. samples were placed at 6.0 cm from the source, and 7.0 cm from the LEGe detector. A 3.8-mm collimator at the detector was used in this work. For the GATE simulation, mono-energy of X-ray was preset at the surface of the plate and directed to the detector based on the photon energy in Table 2, to allow for accurate representation of the metal plates. The setup was simulated via the GATE (version 1.2.3) MC package, with histories of 1 × 107. The simulation data revealed results in the form of entries for the predetermined energy window after the launch of ROOT for output, and the linear and mass attenuation coefficients were calculated.
Table 2. Photon Energy of Each Metal Plates Used in this Work
RESULTS AND DISCUSSION
Elemental Analysis for the Soy-Lignin Bonded Particleboard
Elemental analysis was performed for all soy-lignin bonded Rhizophora spp. wood particles. The fractional weight for each element was recorded in Table 3. The elemental composition of samples was documented to prepare for the command scripts in the GATE simulation.
Table 3. Weight Percentages of Elements for All Particleboards
A = Sample with particle size of 211 to 500 µm; B = Sample with particle size of 104 to 210 µm; C = Sample with particle size of 0 to 103 µm; 0 = 0% soy flour and lignin, 6 = 4.5% soy flour and 1.5% lignin, 12 = 9% soy flour and 3% lignin
Measurement of Linear and Mass Attenuation Coefficient in GATE Simulation
The linear and mass attenuation coefficient of the Rhizophora particleboards and the percentage differences in comparison to simulation data are shown in Tables 4 and 5. Table 6 recorded the percentage standard deviation calculated in GATE.
Fig. 4. Percentage differences between the experimental and simulated data for each Rhizophora sample (Each colored bar in the graph represents the percentage difference between the experimental and simulated data. The length of each color-coded bar indicates the magnitude of this difference)
Table 4. Linear and Mass Attenuation Coefficients of Particleboards in Experiment and GATE Simulation at 16.61 and 17.47 keV
A = 211 to 500 µm, B = 104 to 210 µm, C = 0 to 103 µm particle size ranges; 0 = 0% soy flour and lignin, 6 = 4.5% soy flour and 1.5% lignin, 12 = 9% soy flour and 3% lignin
Table 5. Linear and Mass Attenuation Coefficients of Particleboards in Experiment and GATE Simulation at 21.17 and 25.27 keV