
ORIGINAL ARTICLE 



Year : 2009  Volume
: 34
 Issue : 3  Page : 176179 

Effective atomic numbers and electron density of dosimetric material
SB Kaginelli^{1}, T Rajeshwari^{1}, Sharanabasappa^{1}, BR Kerur^{1}, Anil S Kumar^{2}
^{1} Department of Physics, Gulbarga University, Gulbarga  585 106, India ^{2} Radiation Safety and System Division, Bhabha Atomic Research Centre, Mumbai  400 085, India
Date of Submission  25Dec2008 
Date of Decision  13Apr2009 
Date of Acceptance  18Apr2009 
Date of Web Publication  3Aug2009 
Correspondence Address: B R Kerur Department of Physics, Gulbarga University, Gulbarga  585 106 India
Source of Support: None, Conflict of Interest: None  Check 
DOI: 10.4103/09716203.54853
Abstract   
A novel method for determination of mass attenuation coefficient of xrays employing NaI (Tl) detector system and radioactive sources is described.in this paper. A rigid geometry arrangement and gating of the spectrometer at FWHM position and selection of absorber foils are all done following detailed investigation, to minimize the effect of small angle scattering and multiple scattering on the mass attenuation coefficient, m/r,_{ } value. Firstly, for standardization purposes the mass attenuation coefficients of elemental foils such as Aluminum, Copper, Molybdenum, Tantalum and Lead are measured and then, this method is utilized for dosimetric interested material (sulfates). The experimental mass attenuation coefficient values are compared with the theoretical values to find good agreement between the theory and experiment within one to two per cent. The effective atomic numbers of the biological substitute material are calculated by sum rule and from the graph. The electron density of dosimetric material is calculated using the effective atomic number. The study has discussed in detail the attenuation coefficient, effective atomic number and electron density of dosimetric material/biological substitutes.
Keywords: Mass attenuation coefficient, effective atomic number, electron density, dosimetric materials
How to cite this article: Kaginelli S B, Rajeshwari T, Sharanabasappa, Kerur B R, Kumar AS. Effective atomic numbers and electron density of dosimetric material. J Med Phys 2009;34:1769 
How to cite this URL: Kaginelli S B, Rajeshwari T, Sharanabasappa, Kerur B R, Kumar AS. Effective atomic numbers and electron density of dosimetric material. J Med Phys [serial online] 2009 [cited 2020 Jan 26];34:1769. Available from: http://www.jmp.org.in/text.asp?2009/34/3/176/54853 
Introduction   
In medical physics it is important to evaluate the amount of radiation, delivered by the ionizing radiation, in composite substances. The energy delivered through the photon interactions in composite substances cannot represent the atomic number uniquely across the entire energy region. This number in composite substances is called the effective atomic number and it varies with energy and is denoted here by Z_{ eff} . On the other hand, the concept of zdependence of photon attenuation coefficient has been utilized in many applications of radiation studies.^{ [1]} For example, precise knowledge of effective atomic numbers is very important in medical radiation dosimetry and medical imaging, where the crosssectional anatomy is generated by computer tomography (CT) scans.^{ [2]} It is a common practice to verify the validity of calculation algorithms by comparing the generated doses with the measured doses in tissue equivalent phantom substances. Similarly, tissueequivalent phantoms are specifically designed to study the image quality and performance of the CT scanners. In both instances, a precise knowledge of the effective atomic number and electron density of the composite substances is necessary in the low energy region and have proved to be a convenient parameter for interpretation of xray attenuation by a complex medium like a biological tissue and particularly in the calculation of dose in radiography and radiation dosimetry etc.^{ [3]}
The importance of this paper from diagnostic or therapeutic point of view is that while calculating the effective atomic number of the compound, especially when the photon energy is close to the binding energy of the electron present in the compound, it gives correct information about corrections to be added while calculating the dose to the patient. In such cases the experimental determined effective atomic number may not be agreeable with the theoretical values i.e. deviates from calculated values using the Jackson's formula.
The xray mass attenuation coefficient, m/r, for any material is usually estimated from Bragg's additivity law or more commonly called mixture rule. Thus m/r for any chemical compound/material is given by
μ/ρ=ε ω_{i }(μ/ρ)_{i} ,
where (μ/ρ)_{ i} is the mass attenuation coefficient of the i^{ th} element and ω_{i} is the fraction by weight of the i^{ th} element. For a compound/material with chemical formula (Z_{ 1} )_{ a1} , (Z_{ 2} )_{ a2} , ……(Z_{ n} )_{ an} the weight factor for the i^{ th} element is given by
ω_{i }={( a_{ i} A_{ i)} /(ε a_{ i} A_{ i)}},
where A_{ i} is the atomic weight of the i^{ th} element. Hence an attempt has been made, in this regard, to determine the μ/ρ of xrays for the dosimetric material (sulfates of Mg, Ca, Mn Fe and Zn elements) and then determine the Z_{ eff} of these material by LSF method from ln(μ/ρ) Vs. lnZ graph. These values are compared with the theoretical values. Using the these values of Z_{ eff} the effective electron density calculated by the expression
N_{} e = N_{} A Z_{} eff / A_{ eff}
Where N_{} A is the Avogdro's number, A_{ eff} = A/n_{ i} is the effective atomic weight is the ratio of the molecular weight of the sample divided by the total number of the atoms of all types present in the compound.
Experimental   
The goodgeometry experimental arrangement used in determination of the mass attenuation coefficient is similar to the one described in detail by us earlier ^{ [4]} and the schematic experimental setup is shown in [Figure 1]. Briefly, photons from a variable energy xray source S passed through a collimator C1 and were incident on the specimen A in the form of a thin foil/pellet kept normal to the photon beam. The transmitted beam passed through another collimator C2 and reached a NaI (Tl) xray detector D. The transmitted photon spectrum was recorded using a PCbased multichannel analyzer.
The collimators C1 and C2 were 40 mm thick lead discs that collimated respectively the incident and transmitted beam to 6 mm dia. The scatter acceptance angle equal to the sum of the incident beam divergence and acceptance angle at the detector is found to be less than or equal to 3° degrees. This thickness of the collimator would reduce the intensity of scattered photons of 300 keV by a factor of 10^{ 7} .
A radioactive isotope ^{ 57} Co of strength 0.37 MBq (10 mCi), was obtained from BRIT, Mumbai, India. ^{ 57} Co decays by electron capture to the ^{ 57} Fe ground state, emitting Fe K_{ 1} , K_{α2} and K_{β} xrays. The NaI(Tl) detector can't resolve K_{α} and K_{β} xrays. In the present paper it is proposed that the mass attenuation coefficients are to be determined for the K_{α } xray only and removing the K_{β } xrays. Hence the K_{β } xray intensity has practically been eliminated by the differential absorption technique. So, the transmitted beam was regarded to have the weighted average energy of K_{α1} and K_{α2} i.e. K_{α } is 6.400 keV. Similarly, for the ^{ 65} Zn radioactive source only 8.907 keV xrays selected for the final measurement of mass attenuation coefficient. The variable energy xray source consisted of 10 mCi (370 MBq) ^{ 241} Am as the primary source of excitation radiation and Rubidium target selected to produce fluorescence xray with characteristic energy of the target ie., 13.339 keV. The inner bremsstrahlung intensity was found to be negligible compared to the xray intensity in the region of interest from the radioactive sources. No noticeable impurities were observed in the source spectra.
The dosimetric compounds viz., MgSO_{ 4} .7H_{ 2} O, CaSO_{ 4} .2H_{ 2} O, MnSO_{ 4} .H_{ 2} O, FeSO_{ 4} .7H_{ 2} O, ZnSO_{ 4} .7H_{ 2} O of 99.5% purity were obtained from SD fine chemicals Mumbai, India.. Metal foil (Al, Cu, Mo and Ta) standards were procured from Good Fellow, England was used for standardization purposes only. The dosimetric samples of required thickness in the range of 10 to 200 mg/cm^{ 2} were prepared in the form of 10 mm dia cylindrical pellets by pressing the weighed quantity of the finely ground powder in a hand operated hydraulic press at a pressure of 10 ton. The area density (mass per unit area) of a foil/pellet sample was determined by weighing it using a single pan electronic balance with an accuracy of 0.01 mg and measuring its dimensions using a traveling microscope with an accuracy of 0.001 cm. Thus, the measured areal density expressed in mg/cm^{ 2} had an uncertainty of less than one per cent.
A bicronmade integrated assembly of 25mm dia x 4 mm thick Na(Tl) scintillation mounted on a photomultiplier tube (PMT) served as the xray detector. Oxford model PCAP plus PC plugin single PCI card had on board high voltage supply, preamplifier, amplifier, and 1k ADC. The components on the card were individually controlled and used as 1k channel MCA using the software package OXWIN MCA.
The error involved in each measurement is taken care of by following the procedure counting time conditions as stated in Rose and Shapiro,^{ [5]} viz., background to signal background to foil thickness and signal to foil thickness, systematic errors due to the detection of forward scattered radiation, beam hardening when higher atomic number absorber is used. The Raysum method has been adopted in the present measurement for calculation for the random errors which arises from all aspects of the measurement, further is has also suggested a method for the calculation called the raysum error. In the present measurements, Raysum method is applied to all observations since the random errors arise from all aspects of measurement, in the exponential law of attenuation. The errors presented in the tables are due to propagation of errors calculated according the formula given by Pearson and Osborne.^{ [6]}
No dead time corrections are found for radioactive ^{ 57} Co and ^{ 65} Zn sources. However, in the present case we have selected the live time of the MCA for sources. With these conditions, the transmitted intensity of xrays for various combinations of specimen thickness is recorded and corrected for background intensity^{ .[7]} A plot of logarithm of transmission as a function of specimen thickness yielded a straight line for the entire transmission region, verifying the validity of the BeerLambert's law. This is confirmed for different material too.
Results and Discussion   
The plots of the logarithm of transmitted intensity versus specimen thickness were linear for all the samples and the m/r is obtained from the plots of linear regression over the 502% transmission range. The m/r obtained for the all dosimetric compounds at three different photon energies are presented in [Table 1]. The theoretical results have been calculated by WinXCom^{ [8] } or its predecessor, XCOM^{ [9]} using the mixture rule and theoretical m/r values of the elements. The theoretical estimated errors are lying between one to two per cent as mentioned in the WinXCom.^{ [8]} The error involved in over all experimental values is lying between 23% for the dosimetric samples. The experimental m/r values are presented in second column and theoretical m/r are presented in the third column of [Table 1]. The percent deviation is the difference between the experimental and theoretical μ/ρ values divided by theoretical value. The determined values of μ/ρ_{ } in this transmission range is agreeing well with the theoretical values of WinXCom^{ [8]} within two to three per cent. [Figure 2] clearly shows that the edge effect on the mass attenuation coefficient values otherwise all the three lines would have shown liner graph with a positive slope. As energy increases, m/r decreases, which is well known fact that m/r strongly depends on the atomic number of the absorber and inversely proportional to the energy.
The extrapolated effective atomic number values of the dosimetric material are presented in [Table 2] and these values found to vary from 10.25 to 18.43 for all the sulfates. The theoretical values of the Z_{ eff} are also calculated using the formula given by Jackson and Hawkes^{ [2]} and these values are discussed in the light of the dosimetry point of view and as discussed in the introduction. The experimental and theoretical Z_{ eff} values are agreeing within 5% except the edge region. It is important to mention that the theoretical/calculated values have not considered the edge effects and since the effective atomic numbers are under/over estimated when any element falls below the absorption edge. In the present work, there is a good indication that even in the low photon energy region say that up to 15 keV the effective atomic number can be determined with greater accuracy but one should take into account of edge effects. The electron density of the dosimetric materials is calculated using the experimental Z_{ eff} values and found to vary 0.478 to 0.676 (10^{ 24 } electrons g^{ 1} ).
Conclusions   
The experimental and theoretical Z_{ eff} values are agreeing within five per cent except at the edge region and the determined Z_{ eff} value is agreeing with the theoretical values within five per cent for three energies mentioned.
Acknowledgment   
One of the authors, BRK expresses sincere thanks for awarding him the BRNS Project (sanction no. 2004/37/23/BRNS)
References   
1.  Shivalinge Gowda, S Krishnaveni, T Yashoda, T K Umesh And Ramakrishna Gowda, Photon mass attenuation coefficients, effective atomic numbers and electron densities of some thermo luminescent dosimetric compounds. PRAMANA 2004:63:52941. 
2.  Jackson DF and Hawkes JD, Xray attenuation coefficients of elements and mixtures. Physics Reports 1981;70 : 169233. 
3.  Shivaramu R, Vijayakumar R, Rajasekaran L and Ramamurthy N. Effective atomic numbers for photon energy absorption of some low z substances of dosimetric interst. Radiation Physics and Chemistry 2001;62:3717. 
4.  Kerur BR, Manjula VT, Lagare MT, Anil Kumar S, Mass attenuation coefficient of saccharides for Xrays in the energy range from 8 keV to 32 keV, Radiat. Measurements, 2009 ; 44: 6367 
5.  Pearson JJ and Osborne RR, Practicle Nucleonics, S S N S T N, Ltd., 22, Henrieth Street, London. 1960. 
6.  Rose EM and Shapiro MM, Statistical error in absorption experiments. Phys Rev 1948;74:1853. 
7.  Kerur BR., Thontadarya SR, Hanumaiah B. Measurement of xray mass attenuation coefficients using mixed radiation emitters. Indian J Phys 1993;67A:163. 
8.  Gerward L, Guilbert N, Jensen KB, Levring H, Xray absorption in matter. Reengineering XCOM. Radiat Phys Chem 2001;60:234. 
9.  Berger, M.J., Hubbell, J.H.,. XCOM: Photon Cross Sections Database, Web Version 1.2, Available at http:// physics.nist.gov/xcom. National Institute of Standards and Technology, Gaithersburg, MD 20899, USA (199). 1987/99. 
[Figure 1], [Figure 2]
[Table 1], [Table 2]
This article has been cited by  1 
Robust calculation of effective atomic numbers: The AutoZeff software 

 M. L. Taylor, R. L. Smith, F. Dossing, R. D. Franich   Medical Physics. 2012; 39(4): 1769   [VIEW]  [DOI]  


