Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A performance analysis of a direct-conversion digital X-ray imager Salhotra, Amith Kumar 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_november_Salhotra_Amith.pdf [ 1.97MB ]
Metadata
JSON: 24-1.0074381.json
JSON-LD: 24-1.0074381-ld.json
RDF/XML (Pretty): 24-1.0074381-rdf.xml
RDF/JSON: 24-1.0074381-rdf.json
Turtle: 24-1.0074381-turtle.txt
N-Triples: 24-1.0074381-rdf-ntriples.txt
Original Record: 24-1.0074381-source.json
Full Text
24-1.0074381-fulltext.txt
Citation
24-1.0074381.ris

Full Text

A Performance Analysis of aDirect-Conversion Digital X-rayImagerbyAmith Kumar SalhotraA THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF APPLIED SCIENCEinTHE COLLEGE OF GRADUATE STUDIES(Electrical Engineering)THE UNIVERSITY OF BRITISH COLUMBIA(Okanagan)September 2014c© Amith Kumar Salhotra, 2014AbstractIn this thesis, the performance of direct-conversion digital X-ray imagersis evaluated using an elementary model that draws upon the material prop-erties and the dimensions of the X-ray photoconductor employed within suchimagers. Five possible X-ray photoconductors are considered in this anal-ysis, namely amorphous selenium, cadmium zinc telluride, mercury iodide,lead iodide, and thallium bromide. The collected charge per unit area is theperformance metric considered in this analysis. The collected charge perunit area related to the motion of the electrons and holes individually, anddue the motion of both types of charge carriers, is evaluated. The fractionalcontributions to the collected charge per unit area is also evaluated. Theapplication of both positive and negative biases to the radiation receivingterminals is considered. It is found that the collected charge per unit area,for the case of both positive and negative biases, is higher for the case ofcadmium zinc telluride when compared with the other X-ray photoconduc-tors considered in this analysis. This suggests that cadmium zinc telluridemay be a better material to employ as the X-ray photoconductor withindirect-conversion digital X-ray imagers.iiTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . xviDedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xviiChapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . 1Chapter 2: Background . . . . . . . . . . . . . . . . . . . . . . . 112.1 Digital X-ray imagers . . . . . . . . . . . . . . . . . . . . . . 112.2 General principles underlying the digital X-ray imager . . . . 122.3 The direct-conversion approach . . . . . . . . . . . . . . . . . 132.4 The indirect-conversion approach . . . . . . . . . . . . . . . . 19Chapter 3: Photoconductors . . . . . . . . . . . . . . . . . . . . 213.1 Scope of Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 21iiiTABLE OF CONTENTS3.2 Properties of the ideal X-ray photoconductor . . . . . . . . . 223.3 Comparison of materials properties . . . . . . . . . . . . . . . 243.3.1 Amorphous Selenium (a-Se) . . . . . . . . . . . . . . . 293.3.2 Cadmium Zinc Telluride (CdZnTe) . . . . . . . . . . . 293.3.3 Mercury Iodide (HgI2) . . . . . . . . . . . . . . . . . 293.3.4 Lead Iodide (PbI2) . . . . . . . . . . . . . . . . . . . 343.3.5 Thallium Bromide (TlBr) . . . . . . . . . . . . . . . . 343.4 The induced external photocurrent and the resultant collectedcharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353.5 Trapping and its role in shaping the device performance . . . 373.6 Performance model for a direct-conversion X-ray imager . . . 39Chapter 4: Modeling Results . . . . . . . . . . . . . . . . . . . . 424.1 Comparative analysis . . . . . . . . . . . . . . . . . . . . . . . 424.2 Performance analysis . . . . . . . . . . . . . . . . . . . . . . . 444.3 Performance for the different materials considered . . . . . . 444.3.1 Imager results using a-Se . . . . . . . . . . . . . . . . 444.3.2 Imager results using CdZnTe . . . . . . . . . . . . . . 554.3.3 Imager results using HgI2 . . . . . . . . . . . . . . . . 684.3.4 Imager results using PbI2 . . . . . . . . . . . . . . . . 784.3.5 Imager results using TlBr . . . . . . . . . . . . . . . . 904.4 Comparative analysis . . . . . . . . . . . . . . . . . . . . . . . 101Chapter 5: Conclusions . . . . . . . . . . . . . . . . . . . . . . . 108References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110ivList of TablesTable 3.1 Basic physical properties of the candidate X-ray pho-toconductors considered in this analysis [9]. . . . . . . 27Table 3.2 Properties of the X-ray photoconductor materials con-sidered for this analysis. ‘a’ is at F = 10 V/µm and‘b’ is at F = 20 V/µm [8, 10–15]. . . . . . . . . . . . . 28Table 3.3 The applications of X-ray image detectors [10]. . . . . 28vList of FiguresFigure 1.1 The electromagnetic spectrum . . . . . . . . . . . . . 2Figure 1.2 The conventional X-ray imager . . . . . . . . . . . . . 5Figure 1.3 Digital X-ray imager . . . . . . . . . . . . . . . . . . 8Figure 2.1 Digital X-ray imager . . . . . . . . . . . . . . . . . . 14Figure 2.2 An illustration of the active matrix array . . . . . . . 15Figure 2.3 Direct-conversion approach . . . . . . . . . . . . . . . 16Figure 2.4 The cross-section of an individual pixel within a direct-conversion digital X-ray imager. . . . . . . . . . . . . 18Figure 2.5 The in-direct conversion approach . . . . . . . . . . . 20Figure 3.1 The mass density of the various X-ray photoconduc-tors considered in this analysis. . . . . . . . . . . . . . 25Figure 3.2 The energy gaps associated with various X-ray pho-toconductors considered in this analysis. . . . . . . . 26Figure 3.3 The linear attenuation coefficient, α, for the variousX-ray photoconductors considered in this analysis atthe X-ray photon energy of 20 KeV is considered inthis analysis. . . . . . . . . . . . . . . . . . . . . . . . 30viLIST OF FIGURESFigure 3.4 The product of the electron mobility, µe, and the elec-tron trapping-time,τe, for the various X-ray photo-conductors are considered in this analysis. . . . . . . 31Figure 3.5 The product of the hole mobility, µh, and the holetrapping-time, τh, for the various X-ray photocon-ductors considered in this analysis. . . . . . . . . . . . 32Figure 3.6 The electron-hole pair creation energy (W±), for thevarious X-ray photoconductors considered in this anal-ysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33Figure 3.7 The representation of an electron drifting across theX-ray photoconductor. . . . . . . . . . . . . . . . . . 36Figure 4.1 The collected charge per unit area as a function ofthe applied electric field strength for the case of ana-Se based X-ray imager for the imager thickness setto 200 µm for the case of negative bias. . . . . . . . . 45Figure 4.2 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for an a-Se based X-ray imager for the im-ager thickness set to 200 µm for the case of negativebias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47Figure 4.3 The collected charge per unit area as a function ofthe applied electric field strength for the case of ana-Se based X-ray imager for the imager thickness setto 200 µm for the case of positive bias. . . . . . . . . 48viiLIST OF FIGURESFigure 4.4 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for an a-Se based X-ray imager for the im-ager thickness set to 200 µm for the case of a positivebias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50Figure 4.5 The collected charge per unit area as a function ofthe thickness of the X-ray imager for the case of ana-Se based X-ray imager for the electric field strengthset to 100 kV/cm for the case of negative bias. . . . . 51Figure 4.6 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for an a-Se based X-ray imager for the elec-tric field strength set to 100 kV/cm for the case ofnegative bias. . . . . . . . . . . . . . . . . . . . . . . . 53Figure 4.7 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of an a-Sebased X-ray imager, for the electric field strength setto 100 kV/cm for the case of positive bias. . . . . . . 54Figure 4.8 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for an a-Se based X-ray imager for the elec-tric field strength set to 100 kV/cm for the case ofpositive bias. . . . . . . . . . . . . . . . . . . . . . . . 56viiiLIST OF FIGURESFigure 4.9 The collected charge per unit area as a function ofthe applied electric field strength for the case of aCdZnTe based X-ray imager for the imager thicknessset to 2000 µm for the case of negative bias. . . . . . 57Figure 4.10 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a CdZnTe based X-ray imager for theimager thickness set to 2000 µm for the case of neg-ative bias. . . . . . . . . . . . . . . . . . . . . . . . . 59Figure 4.11 The collected charge per unit area as a function ofthe applied electric field strength for the case of aCdZnTe based X-ray imager for the imager thicknessset to 2000 µm for the case of positive bias. . . . . . . 60Figure 4.12 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a CdZnTe based X-ray imager for theimager thickness set to 2000 µm for the case of apositive bias. . . . . . . . . . . . . . . . . . . . . . . . 62Figure 4.13 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of a CdZnTebased X-ray imager for the electric field strength setto 100 kV/cm for the case of negative bias. . . . . . . 63ixLIST OF FIGURESFigure 4.14 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a CdZnTe based X-ray imager for theelectric field strength set to 100 kV/cm for the caseof negative bias. . . . . . . . . . . . . . . . . . . . . . 65Figure 4.15 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of a CdZnTebased X-ray imager, for the electric field strength setto 100 kV/cm for the case of positive bias. . . . . . . 66Figure 4.16 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a CdZnTe based X-ray imager for theelectric field strength set to 100 kV/cm for the caseof positive bias. . . . . . . . . . . . . . . . . . . . . . 67Figure 4.17 The collected charge per unit area as a function of theapplied electric field strength for the case of a HgI2based X-ray imager for the imager thickness set to250 µm for the case of negative bias. . . . . . . . . . . 69Figure 4.18 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a HgI2 based X-ray imager for the im-ager thickness set to 250 µm for the case of negativebias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70xLIST OF FIGURESFigure 4.19 The collected charge per unit area as a function of theapplied electric field strength for the case of a HgI2based X-ray imager for the imager thickness set to250 µm for the case of positive bias. . . . . . . . . . . 71Figure 4.20 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a HgI2 based X-ray imager for the im-ager thickness set to 250 µm for the case of a positivebias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73Figure 4.21 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of a HgI2based X-ray imager for the electric field strength setto 100 kV/cm for the case of negative bias. . . . . . . 74Figure 4.22 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a HgI2 based X-ray imager for the elec-tric field strength set to 100 kV/cm for the case ofnegative bias. . . . . . . . . . . . . . . . . . . . . . . . 76Figure 4.23 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of a HgI2based X-ray imager, for the electric field strength setto 100 kV/cm for the case of positive bias. . . . . . . 77xiLIST OF FIGURESFigure 4.24 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a HgI2 based X-ray imager for the elec-tric field strength set to 100 kV/cm for the case ofpositive bias. . . . . . . . . . . . . . . . . . . . . . . . 79Figure 4.25 The collected charge per unit area as a function of theapplied electric field strength for the case of a PbI2based X-ray imager for the imager thickness set to83 µm for the case of negative bias. . . . . . . . . . . 80Figure 4.26 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a PbI2 based X-ray imager for the im-ager thickness set to 83 µm for the case of negativebias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82Figure 4.27 The collected charge unit area as a function of theapplied electric field strength for the case of a PbI2based X-ray imager for the imager thickness set to83 µm for the case of positive bias. . . . . . . . . . . 83Figure 4.28 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a PbI2 based X-ray imager for the im-ager thickness set to 83 µm for the case of a positivebias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85xiiLIST OF FIGURESFigure 4.29 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of a PbI2based X-ray imager for the electric field strength setto 100 kV/cm for the case of negative bias. . . . . . . 86Figure 4.30 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a PbI2 based X-ray imager for the elec-tric field strength set to 100 kV/cm for the case ofnegative bias. . . . . . . . . . . . . . . . . . . . . . . . 88Figure 4.31 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of a PbI2based X-ray imager, for the electric field strength setto 100 kV/cm for the case of positive bias. . . . . . . 89Figure 4.32 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a PbI2 based X-ray imager for the elec-tric field strength set to 100 kV/cm for the case ofpositive bias. . . . . . . . . . . . . . . . . . . . . . . . 91Figure 4.33 The collected charge per unit area as a function of theapplied electric field strength for the case of a TlBrbased X-ray imager for the imager thickness set to500 µm for the case of negative bias. . . . . . . . . . . 92xiiiLIST OF FIGURESFigure 4.34 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a TlBr based X-ray imager for the im-ager thickness set to 500 µm for the case of negativebias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94Figure 4.35 The collected charge per unit area as a function of theapplied electric field strength for the case of a TlBrbased X-ray imager for the imager thickness set to500 µm for the case of positive bias. . . . . . . . . . . 95Figure 4.36 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a TlBr based X-ray imager for the im-ager thickness set to 500 µm for the case of a positivebias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96Figure 4.37 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of a TlBrbased X-ray imager for the electric field strength setto 100 kV/cm for the case of negative bias. . . . . . . 97Figure 4.38 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a TlBr based X-ray imager for the elec-tric field strength set to 100 kV/cm for the case ofnegative bias. . . . . . . . . . . . . . . . . . . . . . . . 99xivLIST OF FIGURESFigure 4.39 The collected charge per unit area as a function of thethickness of the X-ray imager for the case of a TlBrbased X-ray imager, for the electric field strength setto 100 kV/cm for the case of positive bias. . . . . . . 100Figure 4.40 The fractional contributions to the collected chargeper unit area related to the motion of the electronsand holes for a TlBr based X-ray imager for the elec-tric field strength set to 100 kV/cm for the case ofpositive bias. . . . . . . . . . . . . . . . . . . . . . . . 102Figure 4.41 The collected charge per unit area plotted as a func-tion of the electric field for the case of five X-ray pho-toconductors considered in this analysis for the caseof negative bias. . . . . . . . . . . . . . . . . . . . . . 103Figure 4.42 The collected charge per unit area plotted as a func-tion of the electric field for the five X-ray photocon-ductors considered in this analysis for the case of pos-itive bias. . . . . . . . . . . . . . . . . . . . . . . . . 105Figure 4.43 The collected charge per unit area plotted as a func-tion of the thickness of the X-ray photoconductor forthe case of five X-ray photoconductors considered inthis analysis for the case of negative bias. . . . . . . . 106Figure 4.44 The collected charge per unit area plotted as a func-tion of the thickness of the X-ray photoconductor forthe case of five X-ray photoconductors considered inthis analysis for the case of negative bias. . . . . . . . 107xvAcknowledgementsI would like to take this opportunity to thank Dr. Stephen K. O’ Leary,my supervisor, for his continuing support and encouragement throughoutthe course of my Master’s studies. Without him, this dream and ambitionof my life, of completing a Master’s program, could not have been fulfilled. Iwould also like to thank my Committee Members Dr. Wilson Eberle and Dr.Jahangir Hossain, and Dr. Yang Cao for their encouragement and support.Sincere gratitude and thanks are also extended to my parents, Mr. RajKumar Salhotra and Late. Mrs.Madhurani Salhotra, my brother Mr. NithinKumar Salhotra, and all my family members, with whose support, I was ableto come to Canada and further my education.I would also like to thank Reddi P. Cheekoori, Vishal Pittalwar, PramodNakod, Hannah Anita, Kallol Mondal, Aparajitha Mondal, Erfan Baghani,Anant Parghi, Atul Polwar, Rajveer Dhillon, Jaspreet Shergill, all the friends,and my near and dear ones, for their help and support. I would also like tothank Shanon Hohle, Amanda Brobble, and all of the staff of the College ofGraduate Studies, the School of Engineering, and the Centre for ScholarlyCommunication for their support.xviDedicationI would like to dedicate this thesis to my parents, Mr. Raj KumarSalhotra and Late. Mrs. Madhurani Salhotra. My mother always had adream to see me completing higher education. Unfortunately, you are notaround me to see this dream coming true, but I always believe that you arewith me and happy to see me fulfilling this dream. Cheers to you, MOM.xviiChapter 1IntroductionX-rays have played a major role in the field of medical diagnostics sincethey were first discovered in 1895. Wilhelm Rontgen, a German scientistworking in his laboratory at the University of Wurzburg [1], is credited asbeing the first researcher to observe X-rays. He observed that phosphors,located at other points in his laboratory, glowed at the same time as highvoltages were being applied across evacuated glass tubes in a darkened room.He dubbed the new kind of radiation as “X-radiation”,“X” being the un-known variable in a mathematical equation. His discovery was reported in ascientific paper, entitled “On a new kind of rays”, which was published in thejournal Science in 1896 [2]. Despite all of the work that has been performedon understanding and characterizing these X-rays since the pioneering con-tribution of Rontgen, the term X-ray captured the popular imagination, andhas remained in use up to the present day.Subsequent analysis has shown that X-rays are merely a form of elec-tromagnetic radiation, as is visible light. In Figure 1.1, an overview of theelectromagnetic spectrum is provided [3]. In this figure, the various bands inthe electromagnetic spectrum are depicted, the corresponding wavelengthsand photon energies also being shown. While visible light has a wavelengththat runs the gamut from 4000 to 7100 A˚, where 1 A˚ = 1 × 10−10 m, X-rays1Chapter 1. Introduction Figure 1.1: The electromagnetic spectrum. The wavelength and photonenergy scales are depicted at the top and bottom of the figure, respectively.This image is after O’Leary [3].2Chapter 1. Introductionpossess wavelengths that are smaller than 100 A˚. While most solid physi-cal objects are opaque to visible light, X-rays have the ability to penetratethrough solid objects. This allows one to image the internal characteristicsof such objects. It is this penetrating property of X-rays that make themuseful for medical imaging purposes. Indeed, shortly after Rontgen’s dis-covery of X-rays, X-rays were starting to be deployed in medical settings.Essentially, the discovery of X-rays has allowed for the genesis of medicalimaging and radiology, fields of medicine that have greatly improved lifeexpectancy and the quality of life in the developed world.Conventional X-ray imagers have been in use since the late 19th Century.Typically, these machines are comprised of an X-ray tube, a phosphor screen,and a film cassette. The X-ray tube emits a uniform flux of X-rays, whichpass through the human subject. Since dense objects will absorb moreof the X-ray flux than less dense objects, bones will absorb more of theX-ray flux than the flesh. Thus, the X-ray flux that emerges from thehuman subject will no longer be uniform in intensity, the regions underbones having a less intense X-ray flux than those under the flesh, i.e., ashadowed X-ray flux emerges. The resultant X-ray flux then interacts withthe underlying phosphor screen, which emits light with an intensity thatis ideally proportional to the X-ray flux intensity received by the screenat that particular point. The light, which the phosphor screen emits, isthen captured by the cassette containing the film. A negative of the X-rayimage is thus captured. The image becomes permanent following processingthrough chemical means. The resultant image allows the medical specialistworking with the machine to acquire insights into the internal workings of3Chapter 1. Introductionthat particular patient. A simplified diagram, depicting the operationalprinciples of the conventional X-ray imager, is depicted in Figure 1.2.There are a number of shortcomings associated with the conventionalX-ray imager [4]. The patients scanned through this process are exposedto a substantial X-ray flux, and this is known to pose a potential healthrisk. Conventional X-ray imagers are large in size, and therefore, occupy aconsiderable amount of space. The images that result are of limited resolu-tion, and no further image processing may be performed following exposureand processing, i.e., no post-exposure image processing and feature enhance-ment may be performed. The chemical means required in order to processthe films is hazardous, the materials used for such processing being difficultto handle and potentially harmful to the environment. Finally, the imagethat results must be stored somewhere else for subsequent retrieval, cross-referencing, and further examination. This has become a major problemfor the cash-strapped health-care sector, as a visit to the medical recordssection of any modern hospital will attest to.The conventional X-ray imager, of the form depicted in Figure 1.2, hasbeen in use in medicine for more than a century now. With all of its afore-mentioned limitations, researchers have been exploring means of moderniz-ing and improving the architecture of the conventional X-ray imager. Thisquest has been guided by a critical thought. Surely, with the abundance oftechnology available in the 21st Century, a device conceived of in the 19thCentury, when technological options were much more limited, can be im-proved. As a result of these explorations, the digital X-ray imager has beenconceived of and manufactured. For the last five years, the digital X-ray4Chapter 1. Introduction Figure 1.2: The conventional X-ray imager. This image is after O’Leary [3].5Chapter 1. Introductionimager has started to be deployed in medical and dental clinics throughoutthe world, and it appears likely that most conventional X-ray imagers willbe displaced by their digital counterparts within the next decade, in thedeveloped world at least.The digital X-ray imager performs all of the functionalities of the conven-tional X-ray imager, but with a substantially reduced amount of radiationdose for the same level of image quality, thereby benefiting the patient [5].The core technology underlying the digital X-ray imager is the active ma-trix array, which is essentially an array of thin-film transistors upon whichthe rest of the imager resides. This active matrix array provides the elec-tronic framework within which a digital X-ray image may be captured andarchived. Each element of the array contains a transistor and a capacitorcorresponding to an individual pixel of the X-ray image. Fundamentally, thedigital X-ray imager works much like the conventional X-ray imager, i.e., auniform X-ray flux passes through the human subject, and the emergingshadowed X-ray flux is captured by the imager. In the conventional X-rayimager, the X-ray image is captured through the use of the phosphor screenand the film cassette. In the digital X-ray imager, however, the image area ispartitioned into individual pixels, and the amount of charge collected corre-sponding to each pixel provides a measure of the intensity of the X-ray fluxcorresponding to that particular pixel. Essentially, while the conventionalX-ray imager captures the intensity of the light emitted off of the underly-ing phosphor screen, the digital X-ray imager converts the X-ray flux into apixelated array of charges, the charge associated with each pixel being pro-portional to the amount of X-ray flux received by it. These charges are then6Chapter 1. Introductionstored by the capacitors associated with the different pixels. The resultantimage is then read-off through the use of the peripheral electronic circuitryassociated with the active matrix array. A simplified diagram of the digitalX-ray imager detector is depicted in Figure 1.3 [3].Digital X-ray imagers actually come in two distinct types: (1) direct-conversion, and (2) indirect-conversion. In the direct-conversion case, X-raysare absorbed by a photoconductor, leading to the generation of electron andhole concentrations in proportion to the incident X-ray flux intensity. Thatis, the absorbed X-ray flux is directly converted into charge. These chargesare then collected through the application of an electric field, i.e., a voltage,on a pixel-by-pixel basis. In contrast, for the indirect-conversion case, theX-ray flux is first converted into light through the use of a scintillator, i.e.,a phosphorus screen. The resultant light, emerging from the scintillator,is then detected by a pixelated array of photodiodes which converts thelight intensity into electrical charges, the collected charge correspondingto each pixel ideally being proportional to the intensity of the X-ray fluxreceived by it. In this case, the X-ray flux is indirectly converted into charge.The collected charges are then read-off through the use of the peripheralelectronics, the resultant X-ray image thus being captured.The performance of a direct-conversion digital X-ray imager is deter-mined, in large measure, by the choice of X-ray photoconductor used withinthe imager. Crystalline materials, such as silicon, cannot be depositedinexpensively and uniformly over large areas, and thus, the X-ray photo-conductors used within direct-conversion digital X-ray imagers are eitheramorphous or polycrystalline in nature. A number of materials have been7Chapter 1. Introduction Figure 1.3: The digital X-ray imager. This image is after O’Leary [3].8Chapter 1. Introductionconsidered as candidates for such photoconductive applications. Amorphousselenium (a-Se), for example, a material used in the past for xerographicalpurposes [6], has been identified as an excellent potential photoconductorfor X-rays, i.e., a large number of electrons and holes are produced foreach X-ray photon absorbed within it. At present, a-Se is being used fordirect-conversion digital X-ray imagers, the resultant images being noted bymedical practitioners for their exceptional quality. A number of other ma-terials have also been considered for such applications, including cadmiumzinc telluride (CdZnTe), mercury iodide (HgI2), lead iodide (PbI2), and thal-lium bromide (TlBr). To date, however, only direct-conversion digital X-rayimagers based on a-Se have actually been fabricated.Polycrystalline and amorphous materials possess defect states which po-tentially can act as traps as the electrons and holes drift under the action ofan applied electric field within the photoconductor. The trapping of chargewithin such a material leads to a reduction in the amount of collected chargereceived by the capacitors associated with the active matrix array followingX-ray exposure. This obviously will detract from the performance of a digi-tal X-ray imager, and understanding how the presence of such traps plays arole in shaping the performance of such detectors has become a major sub-ject of research within this field. Through the quantitative understandingof trapping, and its role in shaping the amount of collected charge received,the performance of such a digital X-ray imager may be better understandand potentially optimized.It is the aim of this thesis to understand how the performance of a digitalX-ray imager is shaped by the particular selection of X-ray photoconductor9Chapter 1. Introductionused and then to use this understanding in order to critically evaluate theperformance obtained for a number of different possible X-ray photocon-ductor candidates. This analysis will be cast within the framework of anelementary model for the collected charge associated with a direct-conversiondigital X-ray imager. Trapping will be considered. The performance of fivedifferent X-ray photoconductors will be considered in this analysis, i.e., a-Se,CdZnTe, HgI2, PbI2, and TlBr. As the image intensity at a given pixel isproportional to the collected charge received by this pixel, for the purposesof this analysis, the focus will be on determining the collected charge per unitarea. The collected charge attributable to the motion of the electrons andholes individually within the X-ray photoconductor, and due to the motionof both types of charges, will be considered. The sensitivity of the results tovariations in the polarity of the applied voltage, and to the thickness of thephotoconductor, will also be considered. Ultimately, recommendations willbe made based on a critical comparison of the predicted performance for thedifferent types of materials considered for the X-ray photoconductor.This thesis is organized in the following manner. The background re-lated to this work is provided in Chapter 2. Then, in Chapter 3, the prop-erties of the different candidate X-ray photoconductors considered in thisanalysis are discussed, an elementary model for the performance of a direct-conversion digital X-ray imager being provided. In Chapter 4, results for theperformance of the different materials considered are presented, along withcomparisons between all of the materials. Finally, in Chapter 5, conclusionsare drawn based on the results presented in Chapter 4.10Chapter 2Background2.1 Digital X-ray imagersDigital X-ray imagers offer improved performance at reduced radiationdoses, thereby benefiting the patient. They allow for all of the functionalitiesof the conventional X-ray imager, with a number of additional benefits.With economies of scale, their use is bound to provide cost savings for acash-strapped sector of the economy. At present, digital X-ray imagers havebeen fabricated, and they are starting to be deployed in medical settingsaround the globe. Practising radiologists in the field are noting that theimages acquired through the use of digital X-ray imagers are superior inquality to those obtained through the use of the conventional X-ray imager.With current rates of adoption, it is expected that over the next decade thedigital X-ray imager will displace most conventional X-ray imagers, withinthe developed world at least.In this chapter, the background related to this work is presented. First,general principles, underlying the operation of a digital X-ray imager, arepresented in greater detail than that presented in Chapter 1. Then, theoperational characteristics of the direct-conversion digital X-ray imager arediscussed. Finally, for the sake of completeness, the operational charac-112.2. General principles underlying the digital X-ray imagerteristics of the indirect-conversion digital X-ray imager are featured. Thediscussion within this chapter will remain at a reasonably high-level. Dis-cussions related to the subsequent technical results, and to means wherebythe performance of a direct-conversion digital X-ray imager may be quanti-tatively evaluated, will be presented in the subsequent chapter.This chapter is organized in the following manner. In Section 2.2, thegeneral principles, underlying the digital X-ray imager, are discussed. Then,the operational characteristics of the direct-conversion digital X-ray imagerare laid out in Section 2.3. Finally, the operational characteristics of theindirect-conversion digital X-ray imager are featured in Section 2.4.2.2 General principles underlying the digitalX-ray imagerThe image captured by a digital X-ray imager is partitioned into an ar-ray of pixels, each pixel being associated with an individual element of theunderlying active matrix array. The intensity of each pixel in the resul-tant digital X-ray image is proportional to the X-ray intensity received bythat particular pixel, which in turn, should be proportional to amount ofcharge collected by the given pixel following X-ray exposure. These collectedcharges are then stored on a capacitor associated with the pixel. FollowingX-ray exposure, the image is then read-off, pixel-by-pixel, through the use ofthe peripheral electronics. The flow of charge off the capacitors during theread-off is externally orchestrated through the activation of the underlyingarray of thin film transistors, this array of transistors often being referred122.3. The direct-conversion approachto as the active matrix array.In the conventional X-ray imager, a phosphor screen is used in order togenerate light in response to X-ray exposure. This light is then capturedthrough the use of the accompanying film cassette. In contrast, in the digitalX-ray imager, shown in Figure 2.1, the phosphor screen and film cassetteof the conventional X-ray imager are reduced into a single unit which iscapable of producing an X-ray image in response to a given incident X-rayflux. The associated active matrix array architecture employed for a digitalX-ray imager, with individual pixels depicted, is presented in Figure 2.2.As was mentioned in Chapter 1, the conversion of X-rays into chargesmay be accomplished through two distinct means: (1) direct-conversion,and (2) indirect-conversion. In direct-conversion, an X-ray photoconductordirectly converts the absorbed X-ray photons into charges, i.e., electron-holepairs. These electron-hole pairs are then collected through the application ofan applied bias at the radiation receiving terminal. These collected chargesreside on the capacitors associated with each pixel until read-off [7]. Theamount of charge collected by each pixel is ideally proportional to the X-rayflux received by it.2.3 The direct-conversion approachThe operating principles underlying the direct-conversion digital X-rayimager are illustrated in Figure 2.3. In this approach, an X-ray photocon-ductor is deposited onto an underlying active matrix array, which includesthin-film transistors, electrical leads to the peripheral electronics, the bot-132.3. The direct-conversion approach Figure 2.1: A representative digital X-ray imager. This image is afterO’Leary [3].142.3. The direct-conversion approach Figure 2.2: An illustration of the underlying active matrix array within adigital X-ray imager. This image is after O’Leary [3].152.3. The direct-conversion approach Charged surface (gives rise to electric field ) X-ray + _   +  +  +  +  +  +  +  +  +  +  +  +  +        Conductive surface at Ground potential Figure 2.3: The direct-conversion approach. This image is after O’Leary [3].162.3. The direct-conversion approachtom electrode associated with each pixel, the transparent top electrode (thiselectrode is transparent so that very little of the X-ray flux is lost passingthrough it), and the associated pixel capacitor. The absorption of the X-rayphotons by the photoconductor will lead to the creation of a large numberof electron-hole pairs within the X-ray photoconductor, i.e., charges. As theenergy associated with a given X-ray photon is orders of magnitude greaterthan the energy gap associated with these materials, a single absorbed X-ray photon can lead to the creation of a large number of electron-hole pairs.Through the application of a voltage to the radiation receiving top electrode,the electron-hole pairs generated within the X-ray photoconductor inducean external photocurrent, and it is this external photocurrent that generatescollected charge associated with each pixel [8]. Figure 2.4 shows the processof charge collection within a single pixel. These charges are then storedby the capacitor associated with the pixel. Ultimately, the charge associ-ated with a given pixel will be read-off through the activation of the gateassociated with the thin film transistor.At the present moment, direct-conversion digital X-ray imagers are fabri-cated using a-Se as the X-ray photoconductor. Familiarity with this materialis the primary reason for this, i.e., a-Se has been used for a number of im-portant applications in the past. There are, however, other types of X-rayphotoconductors, including CdZnTe, HgI2, PbI2, and TlBr, which are alsobeing critically evaluated for possible future use in such imagers. The aim ofthis thesis is to critically evaluate the performance of direct-conversion dig-ital X-ray imagers using a variety of different X-ray photoconductors, withthe hope of identifying the most promising candidate X-ray photoconductors172.3. The direct-conversion approach Figure 2.4: The cross-section of an individual pixel within a direct-conversion digital X-ray imager. This image is after O’Leary [3].182.4. The indirect-conversion approachfor such an application.2.4 The indirect-conversion approachThe operating principles underlying the indirect-conversion digital X-ray imager are illustrated in Figure 2.5. In this approach to X-ray imaging,incident X-ray photons interact with a phosphorus screen, i.e., a scintillator,thereby producing light. The light produced by the screen is then convertedinto charge through an array of photodiodes, which are positioned over anunderlying active matrix array of thin film transistors. The light incident tothe photodiode produces charges, which are then stored on the capacitorsassociated with the individual pixels on the underlying active matrix array.As with the direct-conversion digital X-ray imager, these stored charges areultimately read-off through the use of the peripheral electronics.Unfortunately, the indirect-conversion digital X-ray imager has a funda-mental limitation. In particular, the phosphor grains within the scintillatorlead to light scattering. This will lead to image blurring, and ultimatelylimit the effectiveness of this imaging technique [4]. Thus, while indirect-conversion digital X-ray imagers are at present the dominant digital X-raytechnology, the fundamental advantage offered by the direct-conversion dig-ital X-ray imager will most likely lead to its widespread adoption in thecoming years.192.4. The indirect-conversion approach 3hotodiode surface (gives rise to electric field )  +  _    +  +  +  +  +  +  +  +  +  +  +  +  +       Conductive surface at Ground potential X-ray /ight scatter 3hosphor grain        /ight Figure 2.5: The indirect-conversion approach. This image is afterO’Leary [3].20Chapter 3Photoconductors3.1 Scope of AnalysisDirect-conversion digital X-ray imagers offer a number of advantageswhen contrasted with their indirect-conversion counterparts. Within suchan imager, electrons and holes are created in response to the absorption ofX-ray photons. These charge are then collected on a capacitor associatedwith a given pixel through the application of a voltage on the radiation re-ceiving terminal. In order to evaluate the performance of such an imager,one should have knowledge of the material properties of the X-ray photocon-ductors that are used within such imagers and how these material propertiesimpact upon the corresponding device performance. This will require one todevelop a relationship between the drifting charge carriers within the X-rayphotoconductor and the corresponding collected charge.It is the aim of this chapter to serve this goal. First, a detailed exposi-tion of the ideal properties required by an X-ray photoconductor within suchan imager is provided. Then each of the potential X-ray photoconductorsconsidered, i.e., a-Se, CdZnTe, HgI2, PbI2, and TlBr is discussed, a tabu-lation of the related material properties being provided. Finally, how thedrifting charges within an X-ray photoconductor relate to the corresponding213.2. Properties of the ideal X-ray photoconductorcollected charge is discussed, an elementary model for the charge collectedfrom a direct-conversion digital X-ray imager being presented.This chapter is organized in the following manner. In Section 3.2, theideal properties of an X-ray photoconductor, for use within a direct-conversiondigital X-ray imager, are presented. Then, the material properties, corre-sponding to the X-ray photoconductors considered in this analysis, i.e., a-Se, CdZnTe, HgI2, PbI2 and TlBr, are tabulated in Section 3.3. Finally,in Section 3.4, an elementary model for the performance of such a direct-conversion X-ray imager, i.e., the charge collected per unit area, is presented,this model relating the material properties and dimensions of the underlyingX-ray photoconductors to the device performance. This model provides theframework for the subsequent performance analysis.3.2 Properties of the ideal X-ray photoconductorIn order to choose which X-ray photoconductor is best suited for appli-cations within a direct-conversion digital X-ray imager setting, a summaryof the ideal X-ray photoconductor attributes provides a useful benchmark.Following the review of Kasap and Rowlands [9], a good X-ray photocon-ductor for such applications should possess the following properties:• Most of the incident X-ray flux should be absorbed by the X-ray pho-toconductor.• The X-ray photoconductor should produce a very large number ofelectron-hole pairs in response to the absorption of a single X-ray pho-ton.223.2. Properties of the ideal X-ray photoconductor• There should be minimal bulk recombination within the X-ray photo-conductor.• There should be limited trapping of the electron-hole pairs within theX-ray photoconductor, i.e., the electrons and the holes should be tran-siting sufficiently fast, and the trapping should be sufficiently slow, sothat very little such trapping occurs. This means that µτF >> L, forboth the electrons and holes, where µ represents the mobility, τ rep-resents the trapping time, F represents electric field, and L representsthe thickness of the photoconductor.• The dark current, i.e., the current that occurs without X-ray exposure,should be insignificant. That is, the contrast with the current thatoccurs following X-ray exposure must be significant.• The duration of any charge carrier transit-time must be less than thepixel access time.• The material properties should not degrade when subjected to re-peated exposure to X-rays.• The X-ray photoconductor must be easily deposited over the activematrix array.This tabulation of ideal X-ray photoconductor properties provides asense of the constellation of issues which the designers of a direct-conversiondigital X-ray imager must deal with. In the next section, a tabulation ofmaterial properties corresponding to the X-ray photoconductors under con-233.3. Comparison of materials propertiessideration in this analysis, i.e., a-Se, CdZnTe, HgI2, PbI2, and TlBr is pro-vided.3.3 Comparison of materials propertiesThe coating of an X-ray photoconductor over a large area (in a micro-electronics sense, surfaces of the order of 20 cm × 20 cm in dimensions areconsidered large areas) must be deposited using thin-film technologies. Asingle crystal, such as crystalline silicon, the workhorse of conventional mi-croelectronics, can not be uniformly and inexpensively deposited over suchlarge areas. Thus, alternate materials must be used instead for such appli-cations. As high temperatures cannot be used for such depositions, i.e., soas not to damage the underlying active matrix array, instead polycrystallineor amorphous materials are used for such purposes. For this analysis, anumber of X-ray photoconductor materials are considered, including a-Se,CdZnTe, HgI2, PbI2, and TlBr. Some of the basic physical properties ofthese materials, such as their atomic numbers, mass densities, and energygaps, are listed in Table 3.1, and pictorially represented in Figures 3.1 and3.2.The performance of the digital X-ray imager critically depends on theamount of the collected charge per unit area, which itself is directly re-lated to some of the material properties, such as the mobility-trapping-timeproduct, i.e., µτ . The greater the mobility-trapping-time product, the lesscharge trapping that occurs, and thus, the better performance of the direct-conversion digital X-ray imager. Mobility-trapping-time products for the243.3. Comparison of materials propertiesa - S e C d Z n T e H g I 2 P b I 2 T l B r01234567891 0P b I 2Density (g/cm3 )H g I 2Figure 3.1: The mass density of the various X-ray photoconductors consid-ered in this analysis. X-rays absorption is proportional to the mass density.253.3. Comparison of materials propertiesa - S e C d Z n T e H g I 2 P b I 2 T l B r01234P b I 2Energy Gap (eV)H g I 2Figure 3.2: The energy gaps associated with various X-ray photoconductorsconsidered in this analysis. Lower energy gaps favour electron-hole pairgeneration, i.e., there is less of a barrier.263.3. Comparison of materials propertiesTable 3.1: Basic physical properties of the candidate X-ray photoconductorsconsidered in this analysis [9].Photoconductor Atomic numbers (Z) Mass density (g/cm3) Energy gap (eV)a-Se Se-34 4.3 2.22CdZnTe Cd-48, Te-52, Zn-30 5.8 1.7HgI2 Hg-80, I-53 6.3 2.1PbI2 Pb-82, I-53 6 2.3TlBr Tl-81, Br-35 7.56 2.68electrons can be represented as µeτe, while for holes it is represented as µhτh.These properties of the material, along with typical X-ray photoconductorthicknesses, the absorption linear attenuation coefficient, α, for the case of a20 keV X-ray photon energy, and the electron-hole pair creation energy, W±,are represented in Table 3.2. A comparison of the linear attenuation coeffi-cients, α, for the X-ray photoconductors considered in this analysis, is shownin Figure 3.3. Similarly, a comparison between the mobility-trapping-timeproduct for electrons and holes, and the electron-hole pair creation ener-gies, W±, are shown in Figures 3.4, 3.5, and 3.6, respectively. The digitalX-ray imager is used for a variety of different medical applications, suchas mammography, chest radiology, and fluoroscopy. Guidelines for theseapplications are provided in Table 3.3.273.3. Comparison of materials propertiesTable 3.2: Properties of the X-ray photoconductor materials considered forthis analysis. ‘a’ is at F = 10 V/µm and ‘b’ is at F = 20 V/µm [8, 10–15].X-ray photoconductor α at 20 kev (cm−1) µeτe(cm2/V) µhτh (cm2/V) W± (eV)a-Se 208.33 7.2×10−7 7.0×10−6 45a, 20bCdZnTe 125 2.0×10−4 3.0×10−6 5HgI2 312.5 6.4×10−6 7.0×10−8 5PbI2 357.1 7.0×10−8 2.0×10−6 5TlBr 555.5 1.7×10−4 6.4×10−5 6.5Table 3.3: The applications of X-ray image detectors [10].Clinical task Detector size X-ray spectrum Mean exposure (X)Chest radiology 32 cm×43 cm 120 kVp 300 µRMammography 18 cm×24 cm 30 kVp 12 mRFluroscopy 25 cm×25 cm 70 kVp 1 µR283.3. Comparison of materials properties3.3.1 Amorphous Selenium (a-Se)It is well known that a-Se may be inexpensively and uniformly depositedover large areas through the use of vacuum deposition technique [16]. Theonly drawback of using a-Se within direct-conversion digital X-ray imagersis its high electron-hole creation energy, W±; recall Figure 3.6. For thisreason, while a-Se is the only material currently in use in direct-conversiondigital X-ray imager applications, other materials are being considered foruse in such imagers [17].3.3.2 Cadmium Zinc Telluride (CdZnTe)Polycrystalline CdZnTe is fabricated using the high-pressure Bridgemanmethod [18]. CdZnTe is one of the most attractive materials for the direct-conversion digital X-ray imager applications owing to its high detection ef-ficiency and energy resolution [19]. Because of its high atomic number andhigh mass density, more of the X-ray photons will be absorbed by this ma-terial. Unfortunately, its high leakage current and the high concentration ofgrain boundaries reduce the effectiveness of this material.3.3.3 Mercury Iodide (HgI2)Through the use of the screen printing and physical vapor deposition,polycrystalline HgI2 photoconductors can be easily fabricated [10]. Unfortu-nately, as with the case of CdZnTe, HgI2 is characterized with high leakagecurrents. The particle-in-binder method can also be used to deposit a layerof HgI2 onto the underlying active matrix array [20]. The main advantage ofHgI2 over the a-Se is its small electron-hole creation energy, W±, and there-293.3. Comparison of materials propertiesa - S e C d Z n T e H g I 2 P b I 2 T l B r01 0 02 0 03 0 04 0 05 0 06 0 0P b I 2Linear Attenuation Coefficient α (cm-1 )H g I 2Figure 3.3: The linear attenuation coefficient, α, for the various X-ray pho-toconductors considered in this analysis at the X-ray photon energy of 20KeV.303.3. Comparison of materials propertiesa - S e C d Z n T e H g I 2 P b I 2 T l B r1 E - 1 01 E - 91 E - 81 E - 71 E - 61 E - 51 E - 41 E - 3 µ eτ e (cm2 /V)P b I 2H g I 2Figure 3.4: The product of the electron mobility, µe, and the electrontrapping-time, τe, for the various X-ray photoconductors considered in thisanalysis.313.3. Comparison of materials propertiesa - S e C d Z n T e H g I 2 P b I 2 T l B r1 E - 1 01 E - 91 E - 81 E - 71 E - 61 E - 51 E - 41 E - 3 µ hτ h (cm2 /V)P b I 2H g I 2Figure 3.5: The product of the hole mobility, µh, and the hole trapping-time,τh, for various X-ray photoconductors considered in this analysis.323.3. Comparison of materials propertiesa - S e C d Z n T e H g I 2 P b I 2 T l B r01 02 03 04 05 0  W ± (eV)P b I 2H g I 2Figure 3.6: The electron-hole pair creation energy, W±, for the various X-rayphotoconductors considered in this analysis.333.3. Comparison of materials propertiesfore greater sensitivity; recall Figure 3.6. The disadvantage of HgI2 materialis its grain boundary effects [21]. This results in non-uniform surfaces andincreased leakage currents.3.3.4 Lead Iodide (PbI2)Polycrystalline PbI2 goes through a process of purification before it canbe used as an X-ray photoconductor. This material is typically depositedusing vacuum evaporation [22]. The grain sizes of this material are smallerthan those found within HgI2. It has the additional advantage of allowing foruniform deposition. The disoriented polycrystalline structure of the materialleads to high image lag, which restricts its use from certain applications inmedical diagnostics, such as fluoroscopy [23].3.3.5 Thallium Bromide (TlBr)As with the case of PbI2, intrinsic TlBr possesses high concentrationsof impurities. Accordingly, it must be processed in order to be fit to serveas an X-ray photoconductor for applications within a direct-conversion dig-ital X-ray imager. This processing occurs through the use of either themultipass zone refining technique or the Bridgeman-Stockbarger method, inorder to limit the impact of the impurities [14]. Then the material is de-posited, using technique such as thermal evaporation or the spray coatingmethod. The advantage of this material is that it has a very large atomicnumber and it shows a good spectrometric performance at steady roomtemperature. Another feature of this material is that the electron and holemobility-trapping-time product is quite similar, which results in uniform343.4. The induced external photocurrent and the resultant collected chargecharge collection through the surface and also yields higher energy perfor-mance. Unfortunately, owing to grain boundary effects within the material,which results in high leakage currents, TlBr is not considered an ideal ma-terial. Because of the high leakage current, a shot noise is induced in thematerial, leading to a lowering of the performance of this material.3.4 The induced external photocurrent and theresultant collected chargeThe performance of a direct-conversion digital X-ray imager is deter-mined, in large area measure, by how the drifting charges within the X-rayphotoconductor induce a photocurrent in the corresponding external circuit.Accordingly, it is instructive to determine the photocurrent related to sucha drifting charge within the X-ray photoconductor, under the action of anapplied electric field, and the corresponding collected charge. In Figure 3.7,a representation of an electron drifting across the X-ray photoconductor,under the action of an applied electric field, and the corresponding pho-tocurrent in the external circuit, are depicted. In this case, the radiationreceiving terminal is positively biased and the electron is moving towards it.Ramo’s theorem assert’s that the induced photocurrent associated with themotion of this particular electron may be expressed as;ie = q(veL), (3.1)where ve denotes the electron drift velocity, q represents the electron charge,and L is the separation distance between the terminals within which the X-ray photoconductor is present, i.e., L is the X-ray photoconductor thickness.353.4. The induced external photocurrent and the resultant collected charge 7op (lectrode 9 X-rays Charge Amplifier F Electron   Photoconductor Figure 3.7: The representation of an electron drifting across the X-ray pho-toconductor under the action of an applied electric field. The correspondinginduced photocurrent, Iph, is depicted [24].363.5. Trapping and its role in shaping the device performanceNoting that the electron will drift a distance x before reaching the radiationreceiving terminal, the electron will drift for a time xve . Thus, the collectedcharge that results from the movement of the electron from its point oforigin, x, to reaching the positively biased radiation-receiving terminal,qe = q(xL), (3.2)Similarly, for a hole, initially at x, the collected charge that results from themovement of the hole may be shown to beqh = q[1−(xL)]. (3.3)3.5 Trapping and its role in shaping the deviceperformanceThe electrons and holes that drift across the X-ray photoconductor underthe action of an applied electric field have the potential to be trapped by thetraps that are present within these materials; as these are non-crystallinematerials, the potential for trapping is considerable. Typically, the potentialfor trapping is characterized in terms of a mean trapping time, τ . Trappingclearly takes away from the possibility of an entire charge emerging from thecreation of an electron and hole at a given point. The impact that trappinghas on the collected charge may be characterized in terms of the Hechtrelationship, i.e., the fraction of charge that is collected by the externalcurrent may be expressed asη( xµτF)=µτFx[1− exp(−xµτF)](3.4)373.5. Trapping and its role in shaping the device performancewhere µ, τ , and F represent the drift mobility, the mean trapping-time, andthe electric field, respectively, and x denotes the point where the charge wasinitially generated, η denotes the fraction of the charge that is collected.The charge carrier drift velocity is the product of the drift mobility andthe applied electric field strength, i.e., µF , and thus, µτF correspond to theexpected distance that a charge carrier, be it an electron or a hole, will driftbefore trapping occurs. Note that as1− z ≤ exp(−z), (3.5)for all z, it follows that1− exp(−z) ≤ z, (3.6)and therefore, for z > 0,1− exp(−z)z≤ 1. (3.7)Letting z = xµτF , it can be shown that η in Eq. (3.4) is less than or equalto unity, i.e., η ≤ 1. It is observed that in the limiting case of xµτF << 1,i.e., when the distance from the collection terminal is less than the averagetrapping displacement, µτF , that the fraction of charge that is collected, η,as expressed in Eq. (3.4), may be reduced into a power series, i.e.,η =(µτFx)(1−[1−(xµτF)+12!(xµτF)2− · · · · · ·])(3.8)where the ‘· · · ’ represents higher order terms. Thus, in the limit that( xµτF ) → 0, η(xµτF ) → 1. Alternatively, for (xµτF ) → ∞, it is seen thatη( xµτF )→µτFx .383.6. Performance model for a direct-conversion X-ray imagerUsing this Hecht relationship, Egs. (3.4) and (3.7) may be re-expressedasqe(x) = q(xL)η(xµeτeF), (3.9)andqh(x) = q[1−(xL) ]η(L− xµhτhF), (3.10)where the total collected chargeQ(x) = qe(x) + qh(x). (3.11)3.6 Performance model for a direct-conversionX-ray imagerThere are number of measures available whereby the performance ofa digital X-ray imager can be evaluated. The collected charge acquiredper unit area following X-ray exposure is one such measure. An elementarymodel for the charge collected per unit area for such a digital X-ray imager isprovided in the analysis of Kabir and Kasap [11]. The following assumptionsunderlie the analysis of Kabir and Kasap [11]• The electric field is assumed to be uniform and constant.• The diffusion of charge carriers is neglected.• Each type of charge carrier is ascribed a mobility, µ, and a trapping-time, τ .• Bulk recombination is neglected.393.6. Performance model for a direct-conversion X-ray imager• The X-ray excitation pulse is treated as being instantaneous.• Pixel differences are neglected.The collected charge obtained per unit area may be determined throughthe solution of the charge continuity equation. Following the analysis ofKabir and Kasap [11], the total collected charge related to the motion of theelectrons, for the case of applying a negative bias to the radiation receivingterminal, maybe shown to be:Qe = Qo(µeτeFL)((1− e−αL) +11µeτeF− 1(e(−LµeτeF) − e−αL)), (3.12)where µe denotes the electron mobility, τe represents the electron trapping-time, F is the magnitude of the applied electric field, i.e., F = V/L, whereV is the applied voltage, L is the thickness of the X-ray photoconductor,and α is the linear attenuation coefficientQo =5.45× 1013 × eAX(αairρair )W±(αenα), (3.13)where αen is the energy absorption coefficient, A is the detector area, αairis the energy absorption coefficient of air, ρair is the density of air, and Xis the X-ray exposure.A similar analysis indicates that the charge related to the motion of holesfor the case of applying a negative bias to the radiation receiving terminalmay by shown to be expressed asQh = Qo(µhτhFL)((1− e−αL) +11µhτhF− 1(1− e−αL−( LµhτhF))), (3.14)where µh denotes the hole mobility and τh represents the hole trapping-time,all other terms being define earlier. The total collected charge, Q, may be403.6. Performance model for a direct-conversion X-ray imagerexpressed as the sum of Qe and Qh, i.e.,Q = Qe +Qh (3.15)Similarly, the total collected charge related to the motion of the electrons,for the case of applying positive bias to the radiation receiving terminalmaybe shown to be:Qe = Qo(µeτeFL)((1− e−αL)+11µeτeF− 1(1− e−αL−(LµeτeF))), (3.16)where all the terms are defined earlier. The charge related to the motionof holes for the case of applying a positive bias to the radiation receivingterminal may by shown to be expressed asQh = Qo(µhτhFL)((1− e−αL)+11µhτhF− 1(e( −LµhτhF)− e−αL)), (3.17)where all of the terms are defined earlier.This elementary model for the performance of a direct-conversion digitalX-ray imager, which allows for the evaluation of the collected charge perunit area in terms of a number of basic material properties, i.e., Eqs. (3.12),(3.14), and (3.15) for the case of negative bias and Eqs. (3.15), (3.16), and(3.17) for the case of positive bias, will be used in the subsequent analysis.41Chapter 4Modeling Results4.1 Comparative analysisDigital X-ray imagers, capable of providing all of the functionalities ofthe conventional X-ray imager, but with a substantially reduced amount ofradiation, are starting to be deployed in the medical sector. The direct-conversion approach to digital X-ray imaging is viewed as one of the mosteffective means of implementing digital X-ray imaging. In order to assesshow the performance of such an imager is shaped by the X-ray photocon-ductor employed, the determination of the collected charge per unit areais an effective performance metric. The quality of the image depends onthe amount of charge collected following X-ray exposure. A number of ma-terials have been considered as possible materials for serving as the X-rayphotoconductor within direct-conversion digital X-ray imagers, including a-Se, CdZnTe, HgI2, PbI2, and TlBr. A critical comparative analysis, in whichthe performance of these different materials, within the context of a digitalX-ray imager application, is considered, is the aim of this chapter.In this chapter, the elementary model for the performance of the direct-conversion digital X-ray imagers, introduced in the previous chapter, is usedin order to form the basis of this critical comparative analysis. Five differ-424.1. Comparative analysisent X-ray photoconductors are considered in this analysis, namely a-Se,CdZnTe, HgI2, PbI2, and TlBr. For the purposes of this analysis, the col-lected charges, attributable to the motion of the electrons and holes individ-ually, and due to the motion of both types of charge carriers, are considered.The voltage is applied to the radiation receiving terminal, and both positiveand negative biases are considered, i.e., for the case of the positive bias,the electrons will drift towards the radiation receiving terminal and theholes will drift in the opposite direction, while for the case of the negativebias, the electrons will drift away from the radiation receiving terminal andholes will drift towards it. The fractional contributions, corresponding tothe different types of charges, will also be considered. Finally, performancecomparisons, corresponding to the different X-ray photoconductors consid-ered in this analysis, will be offered. For all cases, the X-ray exposure, X,is set to 12 mR, this being the exposure corresponding to mammography;recall Table 3.3.This chapter is organized in the following manner. Section 4.2 specifiesthe imager performance corresponding to the different materials which canbe used as a photoconductor. A comparison of the basic material propertiesof the material considered is then presented in Section 4.3. Finally, resultsare examined and a comparison of the charge collection per unit area forthe different photoconductors considered in this analysis are presented inSection 4.4.434.2. Performance analysis4.2 Performance analysisThe collected charge per unit area is the performance metric that is em-ployed for the purposes of this analysis, the performance of such an imagerbeing tied to the amount of collected charge per unit area. The perfor-mance analysis that is performed for the different materials considered inthis analysis builds upon Eqs. (3.11) and (3.12) for the case of a negativebias being applied to the radiation receiving terminal, and upon Eqs. (3.15)and (3.16) for the case of a positive bias being applied to the radiation re-ceiving terminal. The materials considered for this analysis include a-Se,CdZnTe, HgI2, PbI2, and TlBr, the material parameters corresponding tothese materials being drawn from Tables 3.1, 3.2 and 3.3. For each mate-rial, the contributions to the collected charge per unit area related to themotion of the electrons and holes are determined individually, as is the totalcollected charge due to the motion of both types of charge carriers. Thedependence of the collected charge per unit area on the applied electric fieldand the photoconductor thickness is evaluated for the two different types ofbiasing conditions.4.3 Performance for the different materialsconsidered4.3.1 Imager results using a-SeIn Figure 4.1, the collected charge per unit area associated with an a-Se based X-ray imager is plotted as a function of the applied electric field444.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10000.511.522.533.544.5holeelectrontotala−Senegative bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 ) Figure 4.1: The collected charge per unit area as a function of the appliedelectric field strength for the case of an a-Se based X-ray imager for theimager thickness set to 200 µm for the case of negative bias. The X-rayflux, X, is set to 12 mR. The collected charge per unit area related to themotion of electrons and holes are depicted, as is the total collected charge,which of course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version is in color. The online version of the figure isin color.454.3. Performance for the different materials consideredstrength for the case of the imager thickness being set to 200 µm, this resultcorresponding to the application of a negative bias. The collected chargeper unit area related to the motion of the electrons and holes are depictedindividually, as is the total collected charge per unit area. It is noted that thecharge related to the motion of electrons exceeds that related to the motionof the holes by a factor of about 3.42. This might have been expectedas electrons, on average, will drift longer before they reach the terminalelectrode while holes drift a shorter distance; recall Ramo’s theorem. It isalso noted that the collected charge monotonically increases with the appliedelectric field strength, i.e., there are less opportunities for trapping if thecharge carriers are moving faster. By taking the no-trapping limits, bothassociated with the motion of the electrons and the holes, upper boundson the charge collection per unit area may be obtained, both associatedwith the motion of the electrons and holes, and the corresponding totalcollected charge. The fractional contributions to the collected charge perunit area, related to the motion of the electron and holes, corresponding tothe functional dependencies depicted in Figure 4.1, are presented in Figure4.2. As expected, the motion of electrons is the dominant contribution tothe collected charge per unit area for this particular case.In Figure 4.3, the collected charge per unit area associated with an a-Se based X-ray imager is plotted as a function of the applied electric fieldstrength for the case of the imager thickness being set to 200 µm, this resultcorresponding to the application of a positive bias. The collected chargeper unit area related to the motion of the electrons and holes are depictedindividually, as is the total collected charge per unit area. It is noted that the464.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10000.10.20.30.40.50.60.70.80.91holeelectrona−Senegative bias12 mRElectric field (kV/cm)Fractional contribution Figure 4.2: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for an a-Se based X-ray imager for the imager thickness set to 200 µm for the case of negativebias. The X-ray flux, X, is set to 12 mR. These results are plotted as afunction of the applied electric field strength. These results correspond tothe results presented in Figure 4.1. Eqs. (3.12), (3.14), and (3.15) are usedto generate this plot. The online version is in color.474.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10000.511.522.533.544.5electron holetotala−Sepositive Bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 ) Figure 4.3: The collected charge per unit area as a function of the appliedelectric field strength for the case of an a-Se based X-ray imager for theimager thickness set to 200 µm for the case of positive bias. The X-ray flux,X, is set to 12 mR. The collected charge per unit area related to the motionof electrons and holes are depicted, as is the total collected charge per unitarea, which of course is equal to the sum of that due to the motion of theelectrons and holes. The corresponding no trapping limits, depicted withthe dashed lines, are also shown. Eqs. (3.12), (3.14), and (3.15) are used togenerate this plot. The online version is in color. The online version of thefigure is in color.484.3. Performance for the different materials consideredcharge related to the motion of holes exceeds that related to the motion of theelectrons by a factor of about 3.479. This might have been expected as holes,on average, will drift longer before they reach the terminal electrode whileelectrons drift a shorter distance; recall Ramos theorem. It is also noted thatthe collected charge per unit area monotonically increases with the appliedelectric field strength, i.e., there are less opportunities for trapping if thecharge carriers are moving faster. By taking the no-trapping limits, bothassociated with the motion of the electrons and the holes, upper bounds onthe charge collection per unit area may be obtained, both associated withthe motion of the electrons and holes, and the corresponding total collectedcharge per unit area. The fractional contributions to the collected chargeper unit area, related to the motion of the electrons and holes, correspondingto the functional depenedies depicted in Figure 4.3, are presented in Figure4.4. As expected, the motion of holes is the dominant contribution to thecollected charge per unit area for this particular case.In Figure 4.5, the collected charge per unit area associated with an a-Sebased X-ray imager is plotted as a function of the imager thickness for thecase of electric field strength being set to 100 kV/cm, this result correspond-ing to the application of a negative bias. The collected charge per unit arearelated to the motion of the electrons and holes are depicted individually, asis the total collected charge. It is noted that the charge per unit area relatedto the motion of electrons exceeds that related to the motion of the holesby a factor of about 2.773. This might have been expected as electrons,on average, will drift longer before they reach the terminal electrode whileholes drift a shorter distance; recall Ramos theorem. It is also noted that494.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10000.10.20.30.40.50.60.70.80.91electronholea−Sepositive bias12 mRElectric field (kV/cm)Fractional contributionFigure 4.4: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for an a-Se based X-ray imager for the imager thickness set to 200 µm for the case of a positivebias. The X-ray flux, X, is set to 12 mR. These results are plotted as afunction of the applied electric field strength. These results correspond tothe results presented in Figure 4.3. Eqs. (3.15), (3.16), and (3.17) are usedto generate this plot. The online version is in color. The online version is incolor.504.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.511.522.533.544.5holeelectron totala−Senegative bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.5: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of an a-Se based X-ray imager for the electricfield strength set to 100 kV/cm for the case of negative bias. The X-ray flux,X, is set to 12 mR. The collected charge per unit area related to the motionof the electrons and the holes are depicted, as is the total collected charge,which of course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version is in color. The online version of the figure isin color.514.3. Performance for the different materials consideredthe amount of collected charge initially monotonically increases with theimager thickness, achieves a maximum, and then monotonically decreasesin response to further increase in the X-ray imager thickness, i.e., there isan opportunity to collect a maximum amount of charge per unit area at aparticular thickness of the imager. By taking the no-trapping limits, bothassociated with the motion of the electrons and holes, upper bounds on thecharge collection per unit area may be obtained associated with the motionof the electrons and the holes, and the corresponding total collected chargeper unit area. The fractional contributions to the collected charge per unitarea, related to the motion of the electrons and holes, corresponding to thefunctional dependencies depicted in Figure 4.5, are presented in Figure 4.6.As expected, the motion of the electrons is the dominant contribution to thecollected charge per unit area for this particular case.In Figure 4.7, the collected charge per unit area associated with an a-Sebased X-ray imager is plotted as a function of the X-ray imager thickness forthe case of the electric field strength being set to 100 kV/cm for the case ofpositive bias. The collected charge per unit area related to the motion of theelectrons and holes are depicted individually, as is the total collected charge.It is noted that the charge per unit area related to the motion of holes exceedsto that related to the motion of the electrons by a factor of about 5.563.This might have been expected as holes, on average, will drift longer beforethey reach the terminal electrode while electrons drift a shorter distance;recall Ramos theorem. It is also noted that the amount of collected chargemonotonically increases with imager thickness and achieves a maximum, andthen monotonically decreases in response to further increase in the thickness524.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91holeelectrona−Senegative bias12 mRThickness L (micron)Fractional contributionFigure 4.6: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for an a-Se based X-ray imager for the electric field strength set to 100 kV/cm for the case ofnegative bias. The X-ray flux, X, is set to 12 mR. These results are plottedas a function of the imager thickness. These results correspond to the resultspresented in Figure 4.5. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version of the figure is in color.534.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.511.522.533.544.5electronhole totala−Sepositive bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.7: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of an a-Se based X-ray imager, for theelectric field strength set to 100 kV/cm for the case of positive bias. TheX-ray flux, X, is set to 12 mR. The collected charge per unit area relatedto the motion of electrons and holes are depicted, as is the total collectedcharge per unit area, which of course is equal to the sum of that due to themotion of the electrons and holes. The corresponding no trapping limits,depicted with the dashed lines, are also shown. Eqs. (3.15), (3.16), and(3.17) are used to generate this plot. The online version of the figure is incolor.544.3. Performance for the different materials consideredof the X-ray imager, i.e., there is an opportunity to collect a maximumamount of charge per unit area at a particular thickness of the X-ray imager.By taking the no-trapping limits, both associated with the motion of theelectrons and holes, upper bounds on the charge collection per unit areamay be obtained associated with the motion of the electrons and the holes,and the corresponding total collected charge. The fractional contributionsto the collected charge per unit area related to the motion of the electronsand holes, corresponding to the functional dependence depicted in Figure4.7, are presented in Figure 4.8. As expected, the motion of the holes isthe dominant contribution to the collected charge per unit area for thisparticular case.4.3.2 Imager results using CdZnTeIn Figure 4.9, the collected charge per unit area associated with a CdZnTebased X-ray imager is plotted as a function of the applied electric fieldstrength for the case of the imager thickness being set to 2000 µm, this re-sult corresponding to the application of a negative bias. The collected chargeper unit area related to the motion of the electrons and holes are depicted in-dividually, as is the total collected charge per unit area. It is noted that thecharge related to the motion of electrons exceeds to that related to the mo-tion of the holes by a factor of about 24.05. This might have been expectedas electrons, on average, will drift longer before they reach the terminalelectrode while holes drift a shorter distance; recall Ramo’s theorem. It isalso noted that the collected charge per unit area monotonically increaseswith the applied electric field strength, i.e., there are less opportunities for554.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91electronholea−Sepositive bias12 mRThickness L (micron)Fractional contributionFigure 4.8: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for an a-Se based X-ray imager for the electric field strength set to 100 kV/cm for the case ofpositive bias. The X-ray flux, X, is set to 12 mR. These results are plottedas a function of the X-ray imager thickness. These results correspond to theresults presented in Figure 4.7. Eqs. (3.15), (3.16), and (3.17) are used togenerate this plot. The online version is in color.564.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 100051015202530354045holeelectron totalCdZnTenegative bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 ) Figure 4.9: The collected charge per unit area as a function of the appliedfield strength for the case of a CdZnTe based X-ray imager, for the imagerthickness set to 2000 µm for the case of negative bias. The X-ray flux, X, isset to 12 mR. The collected charge related to the motion of the electron andhole individually shown, as is the total charge, which of course is equal to thesum of that due to the motion of the electrons and holes. The correspondingno trapping limits, depicted with the dashed lines, are also shown. Eqs.(3.12), (3.14), and (3.15) are used to generate this plot. The online versionof the figure is in color.574.3. Performance for the different materials consideredtrapping if the charge carriers are moving faster. By taking the no-trappinglimits, both associated with the motion of the electrons and the holes, upperbounds on the charge collection per unit area may be obtained, both asso-ciated with the motion of electrons and holes, and the corresponding totalcollected charge. The fractional contributions to the collected charge perunit area, related to the motion of the electrons and holes, corresponding tothe functional dependencies depicted in Figure 4.9, are presented in Figure4.10. As expected, the motion of electrons is the dominant contribution tothe collected charge per unit area for this particular case.In Figure 4.11, the collected charge per unit area associated with aCdZnTe based X-ray imager is plotted as a function of the applied elec-tric field strength for the case of the imager thickness being set to 2000 µm,this result corresponding to the application of a positive bias. The collectedcharge per unit area related to the motion of the electrons and holes aredepicted individually, as is the total collected charge. It is noted that thecharge related to the motion of holes exceeds to that related to the motion ofthe electrons by a factor of about 23.98. This might have been expected asholes, on average, will drift longer before they reach the terminal electrodewhile electrons drift a shorter distance; recall Ramos theorem. It is alsonoted that the collected charge monotonically increases with the appliedelectric field strength, i.e., there are less opportunities for trapping if thecharge carriers are moving faster. By taking the no-trapping limits, bothassociated with the motion of the electrons and the holes, upper bounds onthe charge collection per unit area may be obtained, both associated withthe motion of the electrons and holes, and the corresponding total collected584.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10000.10.20.30.40.50.60.70.80.91holeelectronCdZnTenegative bias12 mRElectric field (kV/cm)Fractional contribution Figure 4.10: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a CdZnTe based X-ray imager for the imager thickness set to 2000 µm for the case of negativebias. The X-ray flux, X, is set to 12 mR. These results are plotted as afunction of the applied electric field strength. These results correspond tothe results presented in Figure 4.9. Eqs. (3.12), (3.14), and (3.15) are usedto generate this plot. The online version is in color.594.3. Performance for the different materials considered0 50 100 150 200 250 300 350 400 450 500051015202530354045electronholetotalCdZnTepositive bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2Figure 4.11: The collected charge as a function of the applied electric fieldstrength for the case of a CdZnTe based X-ray imager for the imager thick-ness set to 2000 µm for the case of positive bias. The X-ray flux, X, is set to12 mR. The collected charge per unit area related to the motion of electronsand holes are depicted, as is the total collected charge per unit area, whichof course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.604.3. Performance for the different materials consideredcharge per unit area. The fractional contributions to the collected chargeper unit area, related to the motion of the electrons and holes, correspond-ing to the functional dependencies depicted in Figure 4.11, are presented inFigure 4.12. As expected, the motion of holes is the dominant contributionto the collected charge per unit area for this particular case.In Figure 4.13, the collected charge per unit area associated with aCdZnTe based X-ray imager is plotted as a function of the imager thick-ness for the case of electric field strength being set to 100 kV/cm, this resultcorresponding to the application of a negative bias. The collected charge perunit area related to the motion of the electrons and holes are depicted indi-vidually, as is the total collected charge. It is noted that the charge relatedto the motion of electrons exceeds that related to the motion of the holesby a factor of about 10.503. This might have been expected as electrons,on average, will drift longer before they reach the terminal electrode whileholes drift a shorter distance; recall Ramos theorem. It is also noted thatthe amount of collected charge initially monotonically increases with the im-ager thickness, achieves a maximum, and then monotonically decreases inresponse to further increase in the X-ray imager thickness, i.e., there is a op-portunity to collect a maximum number of charge at a particular thicknessof the imager. By taking the no-trapping limits, both associated with themotion of the electrons and holes, upper bounds on the charge collection perunit area may be obtained associated with the motion of the electrons andthe holes, and the corresponding total collected charge per unit area. Thefractional contributions to the collected charge per unit area, related to themotion of the electrons and holes, corresponding to the functional depen-614.3. Performance for the different materials considered0 50 100 150 200 250 300 350 400 450 50000.10.20.30.40.50.60.70.80.91electronholeCdZnTepositive bias12 mRElectric field (kV/cm)Fractional contributionFigure 4.12: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a CdZnTe based X-ray imager for the imager thickness set to 2000 µm for the case of a positivebias. The X-ray flux, X, is set to 12 mR. These results are plotted as afunction of the applied electric field strength. These results correspond tothe results presented in Figure 4.11. Eqs. (3.15), (3.16), and (3.17) are usedto generate this plot. The online version is in color.624.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 2000051015202530354045holeelectrontotalCdZnTenegative bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.13: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of a CdZnTe based X-ray imager for theelectric field strength set to 100 kV/cm for the case of negative bias. TheX-ray flux, X, is set to 12 mR. The collected charge related to the motionof the electrons and the holes are depicted, as is the total collected charge,which of course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version of the figure is in color.634.3. Performance for the different materials considereddence depicted in Figure 4.13, are presented in Figure 4.14. As expected,the motion of the electrons is the dominant contribution to the collectedcharge per unit area for this particular case.In Figure 4.15, the collected charge per unit area associated with aCdZnTe based X-ray imager is plotted as a function of the X-ray imagerthickness for the case of the electric field strength being set to 100 kV/cmfor the case of positive bias. The collected charge per unit area related tothe motion of the electrons and holes are depicted individually, as is thetotal collected charge. It is noted that the charge related to the motionof holes exceeds to that related to the motion of the electrons by a factorof about 3.559. This might have been expected as holes, on average, willdrift longer before they reach the terminal electrode while electrons drift ashorter distance; recall Ramos theorem. It is also noted that the amount ofcollected charge per unit area monotonically increases with imager thicknessand achieves a maximum, and then monotonically decreases in response tofurther increase in the thickness of the X-ray imager, i.e., there is an oppor-tunity to collect a maximum amount of charge per unit area at a particularthickness of the X-ray imager. By taking the no-trapping limits, both as-sociated with the motion of the electrons and holes, upper bounds on thecharge collection per unit area may be obtained associated with the motionof the electrons and holes, and the corresponding total collected charge. Thefractional contributions to the collected charge per unit area related to themotion of the electrons and holes, corresponding to the functional depen-dence depicted in Figure 4.15, are presented in Figure 4.16. As expected,the motion of the holes is the dominant contribution to the collected charge644.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91holeelectronCdZnTenegative bias12 mRThickness L (micron)Fractional contributionFigure 4.14: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a CdZnTe basedX-ray imager for the electric field strength set to 100 kV/cm for the caseof negative bias. The X-ray flux, X, is set to 12 mR. These results areplotted as a function of the imager thickness. These results correspond tothe results presented in Figure 4.13. Eqs. (3.12), (3.14), and (3.15) are usedto generate this plot. The online version of the figure is in color.654.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 2000051015202530354045electronholetotalCdZnTepostive bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.15: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of a CdZnTe based X-ray imager, for theelectric field strength set to 100 kV/cm for the case of positive bias. TheX-ray flux, X, is set to 12 mR. The collected charge per unit area relatedto the motion of electrons and holes are depicted, as is the total collectedcharge, which of-course is equal to the sum of that due to the motion of theelectrons and holes. The corresponding no trapping limits, depicted withthe dashed lines, are also shown. Eqs. (3.15), (3.16), and (3.17) are used togenerate this plot. The online version of the figure is in color.664.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91electronholeCdZnTepositive bias12 mRThickness L (micron)Fractional contributionFigure 4.16: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a CdZnTe basedX-ray imager for the electric field strength set to 100 kV/cm for the case ofpositive bias. The X-ray flux, X, is set to 12 mR. These results are plottedas a function of the imager thickness. These results correspond to the resultspresented in Figure 4.15. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.674.3. Performance for the different materials consideredper unit area for this particular case.4.3.3 Imager results using HgI2In Figure 4.17, the collected charge per unit area associated with a HgI2based X-ray imager is plotted as a function of the applied electric fieldstrength for the case of the imager thickness being set to 250 µm, this resultcorresponding to the application of a negative bias. The collected charge perunit area related to the motion of the electrons and holes are depicted indi-vidually, as is the total collected charge. It is noted that the charge relatedto the motion of electrons exceeds to that related to the motion of the holesby a factor of about 6.8668. This might have been expected as electrons,on average, will drift longer before they reach the terminal electrode whileholes drift a shorter distance; recall Ramo’s theorem. It is also noted thatthe collected charge monotonically increases with the applied electric fieldstrength, i.e., there are less opportunities for trapping if the charge carri-ers are moving faster. By taking the no-trapping limits, both associatedwith the motion of the electrons and the holes, upper bounds on the chargecollection per unit area may be obtained, both associated with the motionof electrons and holes, and the corresponding total collected charge. Thefractional contributions to the collected charge per unit area, related to themotion of the electrons and holes, corresponding to the functional depen-dencies depicted in Figure 4.17, are presented in Figure 4.18. As expected,the motion of electrons is the dominant contribution to the collected chargeper unit area for this particular case.In Figure 4.19, the collected charge per unit area associated with a HgI2684.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 100051015202530354045holeelectrontotalHgI2negative bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 )Figure 4.17: The collected charge per unit area as a function of the appliedelectric field strength for the case of a HgI2 based X-ray imager for the imagerthickness set to 250 µm for the case of negative bias. The X-ray flux, X,is set to 12 mR. The collected charge related to the motion of electronsand holes are depicted, as is the total collected charge, which of course isequal to the sum of that due to the motion of the electrons and holes. Thecorresponding no trapping limits, depicted with the dashed lines, are alsoshown. Eqs. (3.12), (3.14), and (3.15) are used to generate this plot. Theonline version of the figure is in color.694.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10000.10.20.30.40.50.60.70.80.91holeelectronHgI2negative bias12 mRElectric field (kV/cm)Fractional contributionFigure 4.18: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a HgI2 based X-rayimager for the imager thickness set to 250 µm for the case of negative bias.The X-ray flux, X, is set to 12 mR. These results are plotted as a functionof the applied electric field strength. These results correspond to the resultspresented in Figure 4.17. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version is in color.704.3. Performance for the different materials considered0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000051015202530354045electronholetotal chargeHgI2positive bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 )Figure 4.19: The collected charge per unit area as a function of the appliedelectric field strength for the case of a HgI2 based X-ray imager for theimager thickness set to 250 µm for the case of positive bias. The X-ray flux,X, is set to 12 mR. The collected charge per unit area related to the motionof electrons and holes are depicted, as is the total collected charge, whichof course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.714.3. Performance for the different materials consideredbased X-ray imager is plotted as a function of the applied electric fieldstrength for the case of the imager thickness being set to 250 µm, this resultcorresponding to the application of a positive bias. The collected chargerelated to the motion of the electrons and holes are depicted individually, asis the total collected charge. It is noted that the charge related to the motionof holes exceeds to that related to the motion of the electrons by a factor ofabout 6.8265. This might have been expected as holes, on average, will driftlonger before they reach the terminal electrode while electrons drift a shorterdistance; recall Ramos theorem. It is also noted that the collected chargemonotonically increases with the applied electric field strength, i.e., thereare less opportunities for trapping if the charge carriers are moving faster.By taking the no-trapping limits, both associated with the motion of theelectrons and the holes, upper bounds on the charge collection per unit areamay be obtained, both associated with the motion of the electrons and holes,and the corresponding total collected charge per unit area. The fractionalcontributions to the collected charge per unit area, related to the motionof the electrons and holes, corresponding to the functional dependenciesdepicted in Figure 4.19, are presented in Figure 4.20. As expected, themotion of holes is the dominant contribution to the collected charge perunit area for this particular case.In Figure 4.21, the collected charge per unit area associated with a HgI2based X-ray imager is plotted as a function of the imager thickness for thecase of electric field strength being set to 100 kV/cm, this result correspond-ing to the application of a negative bias. The collected charge per unit arearelated to the motion of the electrons and holes are depicted individually,724.3. Performance for the different materials considered0 1000 2000 3000 4000 5000 6000 7000 8000 9000 1000000.10.20.30.40.50.60.70.80.91electronholeHgI2positive bias12 mRElectric field (kV/cm)Fractional contributionFigure 4.20: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a HgI2 based X-rayimager for the imager thickness set to 250 µm for the case of a positive bias.The X-ray flux, X, is set to 12 mR. These results are plotted as a functionof the applied electric field strength. These results correspond to the resultspresented in Figure 4.19. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version is in color.734.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200005101520253035holeelectrontotalHgI2negative bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.21: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of a HgI2 based X-ray imager for the electricfield strength set to 100 kV/cm for the case of negative bias. The X-rayflux, X, is set to 12 mR. The collected charge related to the motion of theelectrons and the holes are depicted, as is the total collected charge, whichof course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version of the figure is in color.744.3. Performance for the different materials consideredas is the total collected charge. It is noted that the charge related to themotion of electrons exceeds that related to the motion of the holes by a fac-tor of about 14.98. This might have been expected as electrons, on average,will drift longer before they reach the terminal electrode while holes drift ashorter distance; recall Ramos theorem. It is also noted that the amountof collected charge initially monotonically increases with the imager thick-ness, achieves a maximum, and then monotonically decreases in response tofurther increase in the X-ray imager thickness, i.e., there is a opportunityto collect a maximum number of charge at a particular thickness of the im-ager. By taking the no-trapping limits, both associated with the motion ofthe electrons and holes, upper bounds on the charge collection per unit areamay be obtained associated with the motion of the electrons and the holes,and the corresponding total collected charge. The fractional contributionsto the collected charge per unit area, related to the motion of the electronsand holes, corresponding to the functional dependencies depicted in Figure4.21, are presented in Figure 4.22 . As expected, the motion of the electronsis the dominant contribution to the collected charge for this particular case.In Figure 4.23, the collected charge per unit area associated with a HgI2based X-ray imager is plotted as a function of the X-ray imager thicknessfor the case of the electric field strength being set to 100 kV/cm for thecase of positive bias. The collected charge related to the motion of the elec-trons and holes are depicted individually, as is the total collected charge.It is noted that the charge related to the motion of holes exceeds to thatrelated to the motion of the electrons by a factor of about 1.416. This might754.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91holeelectronHgI2negative bias12 mRThickness L (micron)Fractional contributionFigure 4.22: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a HgI2 based X-ray imager for the electric field strength set to 100 kV/cm for the case ofnegative bias. The X-ray flux, X, is set to 12 mR. These results are plottedas a function of the imager thickness. These results correspond to the resultspresented in Figure 4.21. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version of the figure is in color.764.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200005101520253035electron hole totalHgI2positive bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.23: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of a HgI2 based X-ray imager, for the electricfield strength set to 100 kV/cm for the case of positive bias. The X-ray flux,X, is set to 12 mR. The collected charge per unit area related to the motionof electrons and holes are depicted, as is the total collected charge, whichof course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.774.3. Performance for the different materials consideredhave been expected as holes, on average, will drift longer before they reachthe terminal electrode while electrons drift a shorter distance; recall Ramostheorem. It is also noted that the amount of collected charge monotoni-cally increases with imager thickness and achieves a maximum, and thenmonotonically decreases in response to further increase in the thickness ofthe X-ray imager, i.e., there is a opportunity to collect maximum numberof charge at a particular thickness of the X-ray imager. By taking the no-trapping limits, both associated with the motion of the electrons and holes,upper bounds on the charge collection per unit area may be obtained associ-ated with the motion of the electrons and holes, and the corresponding totalcollected charge. The fractional contributions to the collected charge perunit area related to the motion of the electrons and holes, corresponding tothe functional dependence depicted in Figure 4.23, are presented in Figure4.24. As expected, the motion of the holes is the dominant contribution tothe collected charge per unit area for this particular case.4.3.4 Imager results using PbI2In Figure 4.25, the collected charge per unit area associated with a PbI2based X-ray imager is plotted as a function of the applied electric fieldstrength for the case of the imager thickness being set to 83 µm, this resultcorresponding to the application of a negative bias. The collected charge perunit area related to the motion of the electrons and holes are depicted indi-vidually, as is the total collected charge. It is noted that the charge relatedto the motion of electrons exceeds to that related to the motion of the holesby a factor of about 2.5217. This might have been expected as electrons,784.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91electronhole HgI2positive bias12 mRThickness L (micron)Fractional contributionFigure 4.24: The fractional contributions to the collected charge related tothe motion of the electrons and holes for a HgI2 based X-ray imager for theelectric field strength set to 100 kV/cm for the case of positive bias. TheX-ray flux, X, is set to 12 mR. These results are plotted as a function ofthe imager thickness. These results correspond to the results presented inFigure 4.23. Eqs. (3.15), (3.16), and (3.17) are used to generate this plot.The online version of the figure is in color.794.3. Performance for the different materials considered0 100 200 300 400 500 600 700 800 900 1000051015202530354045holeelectrontotalPbI2negative bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 ) Figure 4.25: The collected charge per unit area as a function of the appliedfield strength for the case of a PbI2 based X-ray imager, for the imagerthickness set to 83 µm for the case of negative bias. The X-ray flux, X, isset to 12 mR. The collected charge related to the motion of the electron andhole individually shown, as is the total charge, which of course is equal to thesum of that due to the motion of the electrons and holes. The correspondingno trapping limits, depicted with the dashed lines, are also shown. Eqs.(3.12), (3.14), and (3.15) are used to generate this plot. The online versionof the figure is in color.804.3. Performance for the different materials consideredon average, will drift longer before they reach the terminal electrode whileholes drift a shorter distance; recall Ramo’s theorem. It is also noted thatthe collected charge monotonically increases with the applied electric fieldstrength, i.e., there are less opportunities for trapping if the charge carri-ers are moving faster. By taking the no-trapping limits, both associatedwith the motion of the electrons and the holes, upper bounds on the chargecollection per unit area may be obtained, both associated with the motionof electrons and holes, and the corresponding total collected charge. Thefractional contributions to the collected charge per unit area, related to themotion of the electrons and holes, corresponding to the functional depen-dencies depicted in Figure 4.25, are presented in Figure 4.26. As expected,the motion of electrons is the dominant contribution to the collected chargeper unit area for this particular case.In Figure 4.27, the collected charge per unit area associated with a PbI2based X-ray imager is plotted as a function of the applied electric fieldstrength for the case of the imager thickness being set to 83 µm, this resultcorresponding to the application of a positive bias. The collected chargeper unit area related to the motion of the electrons and holes are depictedindividually, as is the total collected charge. It is noted that the chargerelated to the motion of holes exceeds that related to the motion of theelectrons by a factor of about 2.54. This might have been expected as holes,on average, will drift longer before they reach the terminal electrode whileelectrons drift a shorter distance; recall Ramos theorem. It is also noted thatthe collected charge per unit area monotonically increases with the appliedelectric field strength, i.e., there are less opportunities for trapping if the814.3. Performance for the different materials considered0 100 200 300 400 500 600 700 800 900 100000.10.20.30.40.50.60.70.80.91holeelectronPbI2negative bias12 mRElectric field (kV/cm)Fractional contributionFigure 4.26: The fractional contributions to the collected charge related tothe motion of the electrons and holes for a PbI2 based X-ray imager for theimager thickness set to 83 µm for the case of negative bias. The X-ray flux,X, is set to 12 mR. These results are plotted as a function of the appliedelectric field strength. These results correspond to the results presented inFigure 4.25. Eqs. (3.12), (3.14), and (3.15) are used to generate this plot.The online version is in color.824.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 1000510152025303540holeelectrontotalPbI2positive bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 ) Figure 4.27: The collected charge per unit area as a function of the appliedelectric field strength for the case of a PbI2 based X-ray imager for theimager thickness set to 83 µm for the case of positive bias. The X-ray flux,X, is set to 12 mR. The collected charge per unit area related to the motionof electrons and holes are depicted, as is the total collected charge per unitarea, which of course is equal to the sum of that due to the motion of theelectrons and holes. The corresponding no trapping limits, depicted withthe dashed lines, are also shown. Eqs. (3.15), (3.16), and (3.17) are used togenerate this plot. The online version of the figure is in color.834.3. Performance for the different materials consideredcharge carriers are moving faster. By taking the no-trapping limits, bothassociated with the motion of the electrons and the holes, upper boundson the charge collection per unit area may be obtained, both associatedwith the motion of the electrons and holes, and the corresponding totalcollected charge. The fractional contributions to the collected charge perunit area, related to the motion of the electrons and holes, corresponding tothe functional dependencies depicted in Figure 4.27, are presented in Figure4.28. As expected, the motion of holes is the dominant contribution to thecollected charge per unit area for this particular case.In Figure 4.29, the collected charge per unit area associated with a PbI2based X-ray imager is plotted as a function of the imager thickness for thecase of electric field strength being set to 100 kV/cm, this result correspond-ing to the application of a negative bias. The collected charge per unit arearelated to the motion of the electrons and holes are depicted individually, asis the total collected charge per unit area. It is noted that the charge relatedto the motion of electrons exceeds that related to the motion of the holes bya factor of about 1.49. This might have been expected as electrons, on av-erage, will drift longer before they reach the terminal electrode while holesdrift a shorter distance; recall Ramos theorem. It is also noted that theamount of collected charge per unit area initially monotonically increaseswith the imager thickness, achieves a maximum, and then monotonicallydecreases in response to further increase in the X-ray imager thickness, i.e.,there is an opportunity to collect a maximum amount of charge per unit areaat a particular thickness of the imager. By taking the no-trapping limits,both associated with the motion of the electrons and holes, upper bounds844.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10000.10.20.30.40.50.60.70.80.91holeelectronPbI2positive bias12 mRElectric field (kV/cm)Fractional contributionFigure 4.28: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a PbI2 based X-rayimager for the imager thickness set to 83 µm for the case of a positive bias.The X-ray flux, X, is set to 12 mR. These results are plotted as a functionof the applied electric field strength. These results correspond to the resultspresented in Figure 4.27. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version is in color.854.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200005101520253035holeelectron totalPbI2negative bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.29: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of a PbI2 based X-ray imager for the electricfield strength set to 100 kV/cm for the case of negative bias. The X-ray flux,X, is set to 12 mR. The collected charge per unit area related to the motionof the electrons and the holes are depicted, as is the total collected charge,which of course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version of the figure is in color.864.3. Performance for the different materials consideredon the charge collection per unit area may be obtained associated with themotion of the electrons and the holes, and the corresponding total collectedcharge per unit area. The fractional contributions to the collected chargeper unit area, related to the motion of the electrons and holes, correspond-ing to the functional dependencies depicted in Figure 4.29, are presentedin Figure 4.30. As expected, the motion of the electrons is the dominantcontribution to the collected charge per unit area for this particular case.In Figure 4.31, the collected charge per unit area associated with a PbI2based X-ray imager is plotted as a function of the X-ray imager thicknessfor the case of the electric field strength being set to 100 kV/cm for the caseof positive bias. The collected charge per unit area related to the motionof the electrons and holes are depicted individually, as is the total collectedcharge. It is noted that the charge related to the motion of holes exceedsto that related to the motion of the electrons by a factor of about 7.79.This might have been expected as holes, on average, will drift longer beforethey reach the terminal electrode while electrons drift a shorter distance;recall Ramos theorem. It is also noted that the amount of collected chargemonotonically increases with imager thickness and achieves a maximum, andthen monotonically decreases in response to further increase in the thicknessof the X-ray imager, i.e., there is a opportunity to collect maximum numberof charge at a particular thickness of the X-ray imager. By taking theno-trapping limits, both associated with the motion of the electrons andholes, upper bounds on the charge collection may be obtained associatedwith the motion of the electrons and holes, and the corresponding totalcollected charge. The fractional contributions to the collected charge per874.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91holeelectronPbI2negative bias12 mRThickness L (micron)Fractional contributionFigure 4.30: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a PbI2 based X-ray imager for the electric field strength set to 100 kV/cm for the case ofnegative bias. The X-ray flux, X, is set to 12 mR. These results are plottedas a function of the imager thickness. These results correspond to the resultspresented in Figure 4.29. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version of the figure is in color.884.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 20000510152025303540totalelectronholePbI2positive bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.31: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of a PbI2 based X-ray imager, for the electricfield strength set to 100 kV/cm for the case of positive bias. The X-ray flux,X, is set to 12 mR. The collected charge per unit area related to the motionof electrons and holes are depicted, as is the total collected charge, whichof course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.894.3. Performance for the different materials consideredunit area related to the motion of the electrons and holes, corresponding tothe functional dependencies depicted in Figure 4.31, are presented in Figure4.32. As expected, the motion of the holes is the dominant contribution tothe collected charge per unit area for this particular case.4.3.5 Imager results using TlBrIn Figure 4.33, the collected charge per unit area associated with a TlBrbased X-ray imager is plotted as a function of the applied electric fieldstrength for the case of the imager thickness being set to 500 µm, thisresult corresponding to the application of a negative bias. The collectedcharge per unit area related to the motion of the electrons and holes aredepicted individually, as is the total collected charge. It is noted that thecharge related to the motion of electrons exceeds that related to the motionof the holes by a factor of about 44.21. This might have been expectedas electrons, on average, will drift longer before they reach the terminalelectrode while holes drift a shorter distance; recall Ramo’s theorem. It isalso noted that the collected charge monotonically increases with the appliedelectric field strength, i.e., there are less opportunities for trapping if thecharge carriers are moving faster. By taking the no-trapping limits, bothassociated with the motion of the electrons and the holes, upper boundson the charge collection per unit area may be obtained, both associatedwith the motion of the electrons and holes, and the corresponding totalcollected charge. The fractional contributions to the collected charge perunit area, related to the motion of the electron and holes, corresponding tothe functional dependencies depicted in Figure 4.33, are presented in Figure904.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91electronholePbI2positive bias12 mRThickness L (micron)Fractional contributionFigure 4.32: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a PbI2 based X-ray imager for the electric field strength set to 100 kV/cm for the case ofpositive bias. The X-ray flux, X, is set to 12 mR. These results are plottedas a function of the imager thickness. These results correspond to the resultspresented in Figure 4.31. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.914.3. Performance for the different materials considered0 5 10 15 20 25 30 35 40 45 5005101520253035hole electrontotalTlBrnegative bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 ) Figure 4.33: The collected charge per unit area as a function of the appliedfield strength for the case of a TlBr based X-ray imager, for the imagerthickness set to 500 µm for the case of negative bias. The X-ray flux, X, isset to 12 mR. The collected charge per unit area related to the motion of theelectron and hole individually shown, as is the total charge, which of courseis equal to the sum of that due to the motion of the electrons and holes.The corresponding no trapping limits, depicted with the dashed lines, arealso shown. Eqs. (3.12), (3.14), and (3.15) are used to generate this plot.The online version of the figure is in color.924.3. Performance for the different materials considered4.34. As expected, the motion of electrons is the dominant contribution tothe collected charge per unit area for this particular case.In Figure 4.35, the collected charge per unit area associated with a TlBrbased X-ray imager is plotted as a function of the applied electric fieldstrength for the case of the imager thickness being set to 500 µm, this resultcorresponding to the application of a positive bias. The collected chargerelated to the motion of the electrons and holes are depicted individually,as is the total collected charge. It is noted that the charge related to themotion of holes exceeds that related to the motion of the electrons by afactor of about 44.43. This might have been expected as holes, on average,drift longer before they reach the terminal electrode while electrons drift ashorter distance; recall Ramos theorem. It is also noted that the collectedcharge monotonically increases with the applied electric field strength, i.e.,there are less opportunities for trapping if the charge carriers are movingfaster. By taking the no-trapping limits, both associated with the motion ofthe electrons and holes, upper bounds on the charge collection per unit areamay be obtained, both associated with the motion of electrons and holes,and the corresponding total collected charge. The fractional contributionsto the collected charge per unit area, related to the motion of the electronsand holes, corresponding to the functional dependencies depicted in Figure4.35, are presented in Figure 4.36. As expected, the motion of holes isthe dominant contribution to the collected charge per unit area for thisparticular case.In Figure 4.37, the collected charge per unit area associated with a TlBrbased X-ray imager is plotted as a function of the imager thickness for the934.3. Performance for the different materials considered0 5 10 15 20 25 30 35 40 45 5000.10.20.30.40.50.60.70.80.91holeelectronTlBrnegative bias12 mRElectric field (kV/cm)Fractional contributionFigure 4.34: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a TlBr based X-rayimager for the imager thickness set to 500 µm for the case of negative bias.The X-ray flux, X, is set to 12 mR. These results are plotted as a functionof the applied electric field strength. These results correspond to the resultspresented in Figure 4.33. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version is in color.944.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10005101520253035holeelectrontotalTlBrpositive bias12 mRElectric field (kV/cm)Collected charge Q (nC/cm2 ) Figure 4.35: The collected charge per unit area as a function of the appliedelectric field strength for the case of a TlBr based X-ray imager for theimager thickness set to 500 µm for the case of positive bias. The X-ray flux,X, is set to 12 mR. The collected charge related to the motion of electronsand holes are depicted, as is the total collected charge, which of-course isequal to the sum of that due to the motion of the electrons and holes. Thecorresponding no trapping limits, depicted with the dashed lines, are alsoshown. Eqs. (3.15), (3.16), and (3.17) are used to generate this plot. Theonline version of the figure is in color.954.3. Performance for the different materials considered0 10 20 30 40 50 60 70 80 90 10000.10.20.30.40.50.60.70.80.91holeelectronTlBrpositive bias12 mRElectric field (kV/cm)Fractional contributionFigure 4.36: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a TlBr based X-rayimager for the imager thickness set to 500 µm for the case of a positive bias.The X-ray flux, X, is set to 12 mR. These results are plotted as a functionof the applied electric field strength. These results correspond to the resultspresented in Figure 4.35. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version is in color.964.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200005101520253035holeelectron totalTlBrnegative bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.37: The collected charge as a function of the thickness of the X-ray imager for the case of a TlBr based X-ray imager for the electric fieldstrength set to 100 kV/cm for the case of negative bias. The X-ray flux, X,is set to 12 mR. The collected charge related to the motion of the electronsand the holes are depicted, as is the total collected charge, which of courseis equal to the sum of that due to the motion of the electrons and holes.The corresponding no trapping limits, depicted with the dashed lines, arealso shown. Eqs. (3.12), (3.14), and (3.15) are used to generate this plot.The online version of the figure is in color.974.3. Performance for the different materials consideredcase of the electric field strength being set to 100 kV/cm, this result cor-responding to the application of a negative bias. The collected charge perunit area related to the motion of the electrons and holes are depicted indi-vidually, as is the total collected charge. It is noted that the charge relatedto the motion of electrons exceeds that related to the motion of the holesby a factor of about 9.455. This might have been expected as electrons,on average, will drift longer before they reach the terminal electrode whileholes drift a shorter distance; recall Ramos theorem. It is also noted thatthe collected charge monotonically increases with the imager thickness, andachieves a maximum, and then monotonically decreases in response to fur-ther increase in the imager thickness, i.e., there is an opportunity to collecta maximum number of charge at a particular thickness of the X-ray imager.By taking the no-trapping limits, both associated with the motion of theelectrons and holes, upper bounds on the charge collection per unit areamay be obtained, associated with the motion of the electrons and holes,and the corresponding total collected charge. The fractional contributionsto the collected charge per unit area, related to the motion of the electronsand holes, corresponding to the functional dependencies depicted in Figure4.37, are presented in Figure 4.38. As expected, the motion of the electronsis the dominant contribution to the collected charge per unit area for thisparticular case.In Figure 4.39, the collected charge per unit area associated with a TlBrbased X-ray imager is plotted as a function of the X-ray imager thicknessfor the case of the electric field strength being set to 100 kV/cm for the caseof positive bias. The collected charge per unit area related to the motion984.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91holeelectronTlBrnegative bias12 mRThickness L (micron)Fractional contributionFigure 4.38: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a TlBr based X-ray imager for the electric field strength set to 100 kV/cm for the case ofnegative bias. The X-ray flux, X, is set to 12 mR. These results are plottedas a function of the imager thickness. These results correspond to the resultspresented in Figure 4.37. Eqs. (3.12), (3.14), and (3.15) are used to generatethis plot. The online version of the figure is in color.994.3. Performance for the different materials considered0 200 400 600 800 1000 1200 1400 1600 1800 200005101520253035electronholetotalTlBrpositive bias12 mRThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.39: The collected charge per unit area as a function of the thicknessof the X-ray imager for the case of a TlBr based X-ray imager, for the electricfield strength set to 100 kV/cm for the case of positive bias. The X-ray flux,X, is set to 12 mR. The collected charge per unit area related to the motionof electrons and holes are depicted, as is the total collected charge, whichof course is equal to the sum of that due to the motion of the electronsand holes. The corresponding no trapping limits, depicted with the dashedlines, are also shown. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.1004.4. Comparative analysisof the electrons and holes are depicted individually, as is the total collectedcharge. It is noted that the charge related to the motion of holes exceeds tothat related to the motion of the electrons by a factor of about 8.09. Thismight have been expected as holes, on average, will drift longer before theyreach the terminal electrode while electrons drift a shorter distance; recallRamos theorem. It is also noted that the amount of collected charge mono-tonically increases with imager thickness and achieves a maximum, and thenmonotonically decreases in response to further increase in the thickness ofthe X-ray imager, i.e., there is an opportunity to collect maximum numberof charge at a particular thickness of the X-ray imager. By taking the no-trapping limits, both associated with the motion of the electrons and holes,upper bounds on the charge collection per unit area may be obtained associ-ated with the motion of the electrons and holes, and the corresponding totalcollected charge. The fractional contributions to the collected charge perunit area related to the motion of the electrons and holes, corresponding tothe functional dependencies depicted in Figure 4.39, are presented in Figure4.40. As expected, the motion of the holes is the dominant contribution tothe collected charge per unit area for this particular case.4.4 Comparative analysisA critical comparison between the results is now presented. In Fig-ure 4.41, the collected charge per unit area is plotted as a function of theelectric field for the case of the five X-ray photoconductors considered in thisanalysis, i.e., a-Se, CdZnTe, HgI2, PbI2, and TlBr, for the case of negative1014.4. Comparative analysis0 200 400 600 800 1000 1200 1400 1600 1800 200000.10.20.30.40.50.60.70.80.91electronholeTlBrpositive bias12 mRThickness L (micron)Fractional contributionFigure 4.40: The fractional contributions to the collected charge per unitarea related to the motion of the electrons and holes for a TlBr based X-ray imager for the electric field strength set to 100 kV/cm for the case ofpositive bias. The X-ray flux, X, is set to 12 mR. These results are plottedas a function of the imager thickness. These results correspond to the resultspresented in Figure 4.39. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.1024.4. Comparative analysis0 10 20 30 40 50 60 70 80 90 100051015202530354045PbI2(83µm)a−Se(200 µm)CdZnTe(2000µm)HgI2(250 µm)TlBr(500µm)negative biasElectric field (kV/cm)Collected charge Q (nC/cm2 )Figure 4.41: The collected charge per unit area plotted as a function of theelectric field for the case of five X-ray photoconductors considered in thisanalysis for the case of negative bias. The X-ray flux, X, is set to 12 mR.Eqs. (3.12), (3.14), and (3.15) are used to generate this plot. The onlineversion of the figure is in color.1034.4. Comparative analysisbias. It is noted that, for all cases, CdZnTe allows for the greatest amountof collected charge per unit area. This is primarily on account of its smallelectron-hole creation energy, W±. An analogous result is depicted in Figure4.42 for the case of positive bias.In Figure 4.43, the collected charge per unit area is plotted as a functionof the thickness of the X-ray photoconductor as a function of the detectorthickness for the case of the five X-ray photoconductors considered in thisanalysi, i.e., a-Se, CdZnTe, HgI2, PbI2, and TlBr, for the case of negativebias. For all cases, this collected charge corresponds to the motion of bothtypes of charge carriers. It is noted that CdZnTe has a greater amount ofcollected charge per unit area than any other X-ray photoconductor consid-ered in this analysis. An analogous result is depicted in Figure 4.44 for thecase of positive bias.1044.4. Comparative analysis0 10 20 30 40 50 60 70 80 90 100051015202530354045PbI2 (83 µm)a−Se (200 µm)CdZnTe (2000 µm)HgI2 (250 µm)TlBr (500 µm)positive biasElectric field (kV/cm)Collected charge Q (nC/cm2 )Figure 4.42: The collected charge per unit area plotted as a function of theapplied electric field for the case of the five X-ray photoconductors consid-ered in this analysis for the case of positive bias. The X-ray flux, X, is setto 12 mR for all cases. Eqs. (3.15), (3.16), and (3.17) are used to generatethis plot. The online version of the figure is in color.1054.4. Comparative analysis0 200 400 600 800 1000 1200 1400 1600 1800 2000051015202530354045PbI2a−SeCdZnTeHgI2TlBrnegative biasThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.43: The collected charge per unit area plotted as a function of thethickness of the X-ray photoconductor for the five X-ray photoconductorsconsidered in this analysis for the case of negative bias. The electric field isbeing set to 100 kV/cm. The X-ray flux, X, is set to 12 mR for all cases.Eqs. (3.12), (3.14), and (3.15) are used to generate this plot. The onlineversion is in color.1064.4. Comparative analysis0 200 400 600 800 1000 1200 1400 1600 1800 2000051015202530354045HgI2a−SePbI2CdZnTeTlBrpositive biasThickness L (micron)Collected charge Q (nC/cm2 )Figure 4.44: The collected charge per unit area plotted as a function of thethickness of the X-ray photoconductor for the case of five X-ray photocon-ductors considered in this analysis for the case of positive bias. The electricfield is being set to 100 kV/cm. The X-ray flux, X, is set to 12 mR for allcases. Eqs. (3.15), (3.16), and (3.17) are used to generate this plot. Theonline version is in color.107Chapter 5ConclusionsIn this thesis, the performance of direct-conversion digital X-ray im-agers was evaluated using an elementary model that draws upon the mate-rial properties and dimensions of the X-ray photoconductor employed. Fivepossible X-ray photoconductors were considered in this analysis, namely a-Se, CdZnTe, HgI2, PbI2, and TlBr. The collected charge per unit area wasthe performance metric considered in this analysis. The collected chargeper unit area related to the motion of the electrons and holes individually,and that due to the motion of both types of charge carriers, was evaluated.The fractional contributions were also evaluated. The application of bothpositive and negative biases to the radiation receiving terminals were con-sidered. It was found that the collected charge per unit area for the caseof both positive and negative bias, is higher in CdZnTe, when compared toother materials considered in this analysis. This suggests that CdZnTe isthe better material in case of amount charge collected per unit area.This thesis presents a number of original contributions that add ontothe understanding of the performance of direct-conversion digital X-ray im-agers. While the results presented are based upon the analytical expressionsprovided by Kabir and Kasap [11], there are number of novel aspects to thisanalysis which distinguish it from that of Kabir and Kasap [11]. The eval-108Chapter 5. Conclusionsuation of the performance of such an imager with respect to the differentmaterials considered in this analysis was not performed by Kabir and Kasap[11], and represents a useful contribution to the field. This is particularlytrue with regards to the performance comparison results that were presentedin Section 4.4. The identification of the individual contributions to the col-lected charge attributable to the electrons and holes is another novel aspectof this analysis.There area a variety of topics that could be considered for possible futurework. Recombination effects, between the electrons and holes, somethingnot considered in this analysis could play an important role in shaping theresultant device performance. The finiteness of the trapping that can occur,something not considered in this analysis, also has the potential to shapedevice performance. Finally, a comparison with the results of the experi-mental work would be a useful contribution to the field. These topics willhave to be dealt with in the future.109References[1] O. W. Linton, “Medical applications of X-rays,” Beam Line, vol. 25,no. 2, pp. 25-34, 1995.[2] W. C. Ro¨ntgen, “On a new kind of rays,” Science, vol. 3, no. 59, pp.227-231, 1896.[3] S. K. O’Leary, “Digital Flat Panel X-Ray Image Detectors using Amor-phous Selenium,” presented in Boise State University, Boise, Idaho,U.S.A, July 2002.[4] M. Yaffe, and J. A. Rowlands, “X-ray detectors for digital radiogra-phy,” Physics in Medicine and Biology, vol. 42, no. 1, pp. 1, 1997.[5] G. Bansal, “Digital radiography. A comparison with modern conven-tional imaging,” Postgraduate Medical Journal, vol. 82, no. 969, pp.425-428, 2006.[6] J. Mort, Anatomy of Xerography: Its Invention and Evolution, Mc-Farland, Jefferson, 1989.[7] Y. Izumi and O. Teranuma, T. Sato, K. Uehara, H. Okada, S. Tokuda,and T. Sato, “Development of Flat Panel X-ray Image Sensors,” SharpTechnical Journal, pp. 25-30, 2001.110Chapter 5. References[8] S. O. Kasap, M. Z. Kabir, and J. A. Rowlands, “Recent advances inX-ray photoconductors for direct conversion X-ray image detectors,”Current Applied Physics, vol. 6, no. 3, pp. 288-292, 2006.[9] S. O. Kasap, and J. A. Rowlands, “Direct-conversion flat-panel X-rayimage sensors for digital radiograph,” Proceedings of the IEEE, vol.90, no. 4, pp. 591-604, 2002.[10] R. A. Street, M. Mulato, M. M. Schieber, H. Hermon, K. S. Shah, P.R. Bennett, Y. Dmitryev, J. Ho, R. Lau, E. Meerson, S. E. Ready,B. Reisman, Y. Sado, K. V. Schuylenbergh, A. I. Vilensky, and A.Zuck, “Comparative study of Pbl2 and Hgl2 as direct detector mate-rials for high-resolution X-ray image sensors,” Medical Imaging 2001.International Society for Optics and Photonics, pp. 1-12, 2001.[11] M. Z. Kabir and S. O. Kasap, “Charge collection and absorption-limited sensitivity of X-ray photoconductors: Applications to a-Se andHgI2,” Applied Physics Letters, vol. 80, no. 9, pp. 1664-1666, 2002.[12] K. Hitomi, M. Matsumoto, O. Muroi, T. Shoji, and Y. Hiratate, “Thal-lium bromide optical and radiation detectors for X-ray and gamma-rayspectroscopy,” Nuclear Science, IEEE Transactions, vol. 49, no. 5, pp.2526-2529, 2002.[13] S. O. Kasap, M. Z. Kabir, J. A. Rowlands, O. Tousignant, J. Leboeuf,L. Laperriere, and Y. Demers, “Dependence of the detective quantumefficiency of photoconductive X-ray image detectors on charge trans-111Chapter 5. Referencesport parameters and exposure: Application to a-Se,” Applied PhysicsLetters, vol. 81, no. 18, pp. 3482-3484, 2002.[14] K. Shah, J. Lund, F. Olschner, L. Moy, and M. Squillante, “Thalliumbromide radiation detectors,” Nuclear Science, IEEE Transactions,vol. 36, no. 1, pp. 199-202, 1989.[15] R. Redus, “Charge Trapping in XR-100T-CdTe and CZT Detectors,”Application Note (ANCT-2 Rev 2) Revised Jan, vol. 9, 2003.[16] G. Belev, and S. O. Kasap, “Amorphous selenium as an X-ray photo-conductor,” Journal of Non-Crystalline Solids, vol. 345, pp. 484-488,2004.[17] S. O. Kasap, and J. A. Rowlands, “Review X-ray photoconductorsand stabilized a-Se for direct conversion digital flat-panel X-ray image-detectors,” Journal of Materials Science: Materials in Electronics, vol.11(3), pp. 179-198, 2000.[18] Y. Nemirovsky, A. Ruzin, G. Asa, and J. Gorelik, “Study of the chargecollection efficiency of CdZnTe radiation detectors,” Journal of Elec-tronic Materials, vol. 25, no. 8, pp.1221-1231, 1996.[19] C. Szeles, “CdZnTe and CdTe materials for X-ray and gamma rayradiation detector applications,” Physica Status Solidi (b), vol. 241,no. 3, pp.783-790, 2004.[20] S.-H. Cho, S.-U. Heo, C.-W. Choi, S.-S. Kang, S.-H. Nam, J.-K. Park,S.-K. Park, “Properties of Polycrystalline HgI2 Films Fabricated by112Chapter 5. ReferencesUsing a Particle in Binder Method,” Journal of Korean Physical So-ciety, vol. 52, no. 1123, 2008.[21] R. A. Street, S. E. Ready, L. Melekhov, J. Ho, A. Zuck, and B. N.Breen, “Approaching the theoretical X-ray sensitivity with Hgl2 directdetection image sensors,” Medical Imaging 2002. International Societyfor Optics and Photonics, pp. 414-422, 2002.[22] R. A. Street, J. T. Rahn, S. E. Ready, K. S. Shah, P. R. Bennett, Y. N.Dmitriyev, P. Mei, J.-P. Lu, R. B. Apte, J. Ho, et al., “X-ray imagingusing lead iodide as a semiconductor detector,” Medical Imaging 1999.International Society for Optics and Photonics 1999, pp. 36-47, 1999.[23] G. Zentai, L. D. Partain, R. Pavlyuchkova, C. Proano, G. F. Virshup,L. Melekhov, A. Zuck, B. N. Breen, O. Dagan, A. Vilensky, et al.,“Mercuric iodide and lead iodide x-ray detectors for radiographic andfluoroscopic medical imaging,” Medical Imaging 2003. InternationalSociety for Optics and Photonics 2003, pp. 77-91, 2003.[24] M. Z. Kabir, “Modeling of x-ray photoconductors for x-ray im-age detectors,” Ph.D. Dissertation, Dept. Elect. Eng., Univ. ofSaskatchewan, Saskatoon, Canada, 2005.113

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0074381/manifest

Comment

Related Items