USF Libraries
USF Digital Collections

Optical response of polycrystalline mercuric iodide photoconductive detectors

MISSING IMAGE

Material Information

Title:
Optical response of polycrystalline mercuric iodide photoconductive detectors
Physical Description:
Book
Language:
English
Creator:
Chegoor, Prashant
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Hgi2
Surface recombination
Bulk recombination
Spectral response
I-V curves
Dissertations, Academic -- Electrical Engineering -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Mercuric Iodide in its tetragonal form has received a lot of attention for many years as a prospective room temperature X-ray and y-ray detector. Its basic properties are well suited for this purpose. Its wide band gap of 2.1eV contributes to a high dark resistivity of 1012ohm-cm or higher. A high atomic number of its constituent atoms (Hg-80, I -53) and a density of 6.3g/cm3 result in its efficient interaction with incident X-ray or y-ray radiation. Single crystalline mercuric iodide has been thoroughly studied and successfully utilized in commercial radiation detectors. But with the urgent need for large area ,low cost efficient X-ray detectors, focus has now shifted towards the development and understanding of the properties of thin film Polycrystalline Mercuric iodide detectors. Such detectors also have the advantage of being most suited for direct X-ray detection i.e.a direct conversion of incident X rays into electric signals which are then used to obtain an equivalent image in digital X-ray imaging. They also can be used in applications where a scintillator intermediate is used to generate visible light from incident high energy photons.Therefore it is important to study their optical response in order to understand and evaluate their Optical Properties. The present work focuses on obtaining the Optical response of the thin film Mercuric iodide photoconductive detectors .These films were grown on TEC-15 LOF glass with a Tin Oxide (SnO2) coating on it, which acts as a growth surface for the films and also functions as the front contact of the detector.Palladium which is sputtered on top of this film acts as the back contact. There are a total of seven contacted devices on each film sample and each device has been tested for its optical response in terms of Spectral Response and I-V characteristics in both light and dark conditions.
Thesis:
Thesis (M.S.E.E.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Prashant Chegoor.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 74 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001681088
oclc - 62751919
usfldc doi - E14-SFE0001026
usfldc handle - e14.1026
System ID:
SFS0025347:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001681088
003 fts
005 20060215071225.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 051228s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001026
035
(OCoLC)62751919
SFE0001026
040
FHM
c FHM
049
FHMM
090
TK145 (Online)
1 100
Chegoor, Prashant.
0 245
Optical response of polycrystalline mercuric iodide photoconductive detectors
h [electronic resource] /
by Prashant Chegoor.
260
[Tampa, Fla.] :
b University of South Florida,
2005.
502
Thesis (M.S.E.E.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 74 pages.
520
ABSTRACT: Mercuric Iodide in its tetragonal form has received a lot of attention for many years as a prospective room temperature X-ray and y-ray detector. Its basic properties are well suited for this purpose. Its wide band gap of 2.1eV contributes to a high dark resistivity of 1012ohm-cm or higher. A high atomic number of its constituent atoms (Hg-80, I -53) and a density of 6.3g/cm3 result in its efficient interaction with incident X-ray or y-ray radiation. Single crystalline mercuric iodide has been thoroughly studied and successfully utilized in commercial radiation detectors. But with the urgent need for large area ,low cost efficient X-ray detectors, focus has now shifted towards the development and understanding of the properties of thin film Polycrystalline Mercuric iodide detectors. Such detectors also have the advantage of being most suited for direct X-ray detection i.e.a direct conversion of incident X rays into electric signals which are then used to obtain an equivalent image in digital X-ray imaging. They also can be used in applications where a scintillator intermediate is used to generate visible light from incident high energy photons.Therefore it is important to study their optical response in order to understand and evaluate their Optical Properties. The present work focuses on obtaining the Optical response of the thin film Mercuric iodide photoconductive detectors .These films were grown on TEC-15 LOF glass with a Tin Oxide (SnO2) coating on it, which acts as a growth surface for the films and also functions as the front contact of the detector.Palladium which is sputtered on top of this film acts as the back contact. There are a total of seven contacted devices on each film sample and each device has been tested for its optical response in terms of Spectral Response and I-V characteristics in both light and dark conditions.
590
Adviser: Don Morel.
653
Hgi2.
Surface recombination.
Bulk recombination.
Spectral response.
I-V curves.
690
Dissertations, Academic
z USF
x Electrical Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1026



PAGE 1

Optical Response of Polycrystalline Merc uric Iodide Photoconductive Detectors by Prashant Chegoor A thesis submitted in partial fulfillment of the requirement s for the degree of Master of Science in Electrical Engineering Department of Electrical Engineering College of Engineering University of South Florida Major Professor: Don L. Morel, Ph.D. Christos S. Ferekides, Ph.D. Yun. L. Chiou, Ph.D. Date of Approval: March 24, 2005 Keywords: HgI2, Surface Recombination, Bulk Recombination, Spectral Response, I-V Curves Copyright 2005, Prashant Chegoor

PAGE 2

ACKNOWLEDGMENTS I would like to take this opportunity to express my sincere gratitude to my Major Professor, Dr Don Morel. He constant ly stood as a source of Inspiration throughout this research. His keen Insi ght and valuable suggestions had motivated me through thick and thin in this research. Apart from this I personally have learnt a lot from him due to his hi gh Integrity and compos ed nature. I would like to thank Dr. Christos Ferekides, a seasoned researcher in the field of Thin Film Solar cells, for all his valuable feedbacks during our research group meetings. I would like to thank Dr Y.L.Chiou for agreeing to be a part of my defense committee. A word of appreciation also goes to my colleagues from the Thin Film Compound Semiconductor Lab at USF espec ially Vikram, Prince and Sridevi.It was indeed a wonderful experience workin g with them.My frie nds in USF and in India also deserve credit for their congeni ality. Last but not least, I am always Indebted to my Parents who were the driving force behind all the earnest endeavors undertaken by me till now.

PAGE 3

i TABLE OF CONTENTS LIST OF TABLES iii LIST OF FIGURES iv ABSTRACT v CHAPTER 1. INTRODUCTION 1 CHAPTER 2. SEMICONDU CTOR PROCESSES 6 2.1 Introduction 6 2.2 Carrier Transport 8 2.3 Carrier Recombinat ion and Generation Processes 12 2.3.1 Re combination Processes 14 2.3.1.1 Band-Band Recombination 14 2.3.1.2 RG Center Recombination 14 2.3.1.3 Auger Recombination 17 2.3.2 G eneration Processes 17 2.4 Direct and Indirect Semiconductors 18 2.5 Surface Recombination-Generation 22 CHAPTER 3. PHOTOGENERATION AND PHOTOCONDUCTIVE DETECTORS 25 3. 1 Photogeneration 25 3. 2 Photoconductor 28

PAGE 4

ii 3.2.1 Q uantum Efficiency (QE) 29 3.2.2 Gain 29 3.2.3 P hotocurrent 31 3.2.4 Dark Current 31 3.2. 5 Spectral Response 32 CHAPTER 4. MERCURIC IODIDE 35 4.1 Literature Review 35 4. 2 Crystal Structure 38 CHAPTER 5. RESULTS AND DISCUSSION 40 5.1 Structur e and Layout of the Films 40 5.2 Experimental Procedure 42 5.3 Optical Resp onse of the Films 43 5.3. 1 Sample 11-19-02 46 5.3.2 Sample 12-02-02 49 5.3. 3 Sample 01-13-03 51 5.3.4 Sample 10-20-03 55 5.3. 5 Sample 03-25-04 55 CHAPTER 6. CONCLUSIONS 59 6.1 Future Work 60 REFERENCES 62 APPENDICES 64 Appendix A Spectral Response Data File 65

PAGE 5

iii LIST OF TABLES Table 1.1 Comparison of Some X –ray Sensitive Semiconductor Materials for Direct Detection 4 Table 5.1 Deposition Parameters of the Samples 44 Table 5.2 Peak QE’s of the Samples at -50V Bias to the Illuminated Contact 45 Table 5.3 Absorption Coefficient Data 48 Table A.1 Spectral Response Data File 66

PAGE 6

iv LIST OF FIGURES Figure 2.1 Visualizatio n of Carrier Drift 9 Figure 2.2 Visualization of Carrier Diffusion Process 11 Figure 2.3 Energy Band Vi sualization of Bulk Recombination and Generation Processes 13 Figure 2.4 The Two Different Types of Semiconductors 19 Figure 2.5 E-k Plots for Visualizat ion of Recombination (band to band) in Direct and Indirect Semiconductors 21 Figure 2.6 Abrupt Termination of Semic onductor Lattice at the Surface 22 Figure 2.7 Processes at the Semiconductor Surface 23 Figure 3.1 Exponential Deca y of Photon Intensity Inside a Semiconductor 26 Figure 3.2 Different Experimental Conf igurations of a Photoconductor 33 Figure 4.1 Tetragonal Unit Cell of HgI2 39 Figure 5.1 Two Different Views of the Sample 41 Figure 5.2 Spectral Response and Light I-V Curves for Sample 11-19-02 47 Figure 5.3 Spectral Response and Light I-V Curves for Sample 12-02-02 50 Figure 5.4 Spectral Response and Light I-V Curves for Sample 01-13-03 52 Figure 5.5 Dark I-V Curves for Sample 01-13-03 53 Figure 5.6 Spectral Response and Light I-V Curves for Sample 10-20-03 54 Figure 5.7 Light I-V Curves for Samples 10-21-03 and 09-23-04 56 Figure 5.8 Spectral Response and Light I-V Curves for Sample 03-25-04 58

PAGE 7

v OPTICAL RESPONSE OF POLYCRYSTALLINE MERCURIC IODIDE PHOTOCONDUCTIVE DETECTORS Prashant Chegoor ABSTRACT Mercuric Iodide in its tetragonal form has received a lot of attention for many years as a prospective room temperature X-ray and -ray detector. Its basic properties are well suited for this purpose. Its wide band gap of 2.1eV contributes to a high dark resistivity of 1012ohm-cm or higher. A high atomic number of its constituent atoms (H g-80, I -53) and a density of 6.3g/cm3 result in its efficient interaction with incident X-ray or -ray radiation. Single crystalline mercuric iodide has been thoroughly st udied and successfully utilized in commercial radiation detectors. But with the urgent need for la rge area ,low cost efficient X-ray detectors, focus has now shifted towards the development and understanding of the properties of thin film Polycrystalline Mercuric iodide detectors. Such detectors also have the ad vantage of being most suited for direct X-ray detection i.e. a direct conversion of incident X rays into electric signals which are then used to obtain an equivalent image in digital X-ray imaging. They also can be used in applic ations where a scintillator intermediate is used to generate visible light from in cident high energy photons.

PAGE 8

vi Therefore it is import ant to study their optical response in order to understand and evaluate their Optical Properties. The present work focuses on obtaining the Optical response of the thin film Mercuric iodide photoconductive det ectors .These films were grown on TEC15 LOF glass with a Tin Oxide (SnO2) coating on it, which acts as a growth surface for the films and also functions as the front contact of the detector. Palladium which is sputter ed on top of this film acts as the back contact. There are a total of seven contacted devices on each film sample and each device has been tested for its optical response in terms of Spectral Response and I-V characteristics in both light and dar k conditions. Results obtained have been tabulated, and some import ant samples have been analyzed. The aim in this research was to obtain films with high QE ,low dark current and uniformity of the results for all devices on a sample.Typical values of high QE’s obtained for the samples ranged from 0.3 0. 4 with a bias of -50V appl ied to the front contact. The photocurrents corresponding to these QE’s ranged from 2.83 A – 3.82 A as obtained from the spectral response dat a file. The dark currents that were measurable were typically below 50nA for the best sample.

PAGE 9

1 CHAPTER 1 INTRODUCTION Semiconductors are playi ng a vital role in influencing us by making our daily life more comfortabl e than before. Semiconductors are the backbone of many electronic devices that we use today. With the rapid developments in the Semiconductor Tec hnology over the last fifty years, electronic devices have become smaller, faster and more reliable. Radiation detectors fabricated with Semiconductor materials is one such category of devices which has received a lot of att ention with the rapid developments in the Semiconductor Technology. Semiconductor Radiation Detectors operate on the basic principle that when radiat ion meeting certain requir ements is incident on the device, a charge pulse of electrons and ho les is created in the volume of the device by some interaction process whic h are then separated and made to flow in opposite direction by t he application of an electric field, thus producing a current in the external circuit which c an be detected. The am ount of charge thus collected is a measure of the energy of radiation incident on the detector. Therefore there are many factors that have to be considered during detector operation such as nature of interacti on of radiation with the semiconductor constituting the detector, the efficiency of the excitation process, the efficiency of

PAGE 10

2 the charge collection process, the external circuit detecting the charge pulse and finally the background noise of the detector. The detect or noise as well as the nature and efficiency of interaction bet ween the particular radiation and the detector volume and the charge collecti on process will determine what materials may be employed to fabricate detectors for a specific purpose. For the Fabrication of room temperature X-ray and ray detectors the semiconductor material should satisfy the following requirements [1]. 1) High Atomic number and Dens ity for high photon absorption. 2) Large Band Gap for high resistivity and to minimize leakage currents at room temperature. 3) High Intrinsic Electron and Hole mobility life time produ cts for efficient charge collection. 4) High purity, homogeneous defect fr ee material with acceptable cross-sectional area and thickness. X-ray detectors are further divided in to two groups known as Indirect and Direct detectors. Indirect Detectors are t hose in which the X-ray photons are first converted to visible photons by adoptin g a phosphor based scintillator and then these visible photons are converted into an electrical signal by means of photodiode arrays. For Digital X ray imag ing this signal is transformed into an equivalent image by sensing it with an el ectronic readout mechanism followed by a analog to digital conversion to obtain the digital image [2]. Obviously this two step process suffers from many disadv antages including low energy transfer efficiency and low energy resolution [3]. In contrast, Direct Detectors offer a one

PAGE 11

3 step conversion of X-ray Photons into El ectrical signals by means of a X-ray photoconductive detector which can then be transformed into an image by the same process as mentioned above. Thus this results in higher QE and resolution for this detector. In this context polycrystalline Mercur ic Iodide detectors show promise as Direct X-ray detectors in di gital radiography.Polycrystalline Mercuric Iodide also finds applications in in direct detection wherein a scintillator intermediate is used to generate visibl e photons from incident high energy photons. The present work theref ore deals with polycrystalline HgI2 photoconductive detectors with emphasis onl y on their Optical response. As shall be discussed later, Photoconductive detectors are the simplest subdivision of the Photodetectors consisting of a semic onductor material sandwiched between two metal contacts. Before proceeding further it is worthwhile to compare some of the properties of Polycrystalline HgI2 in comparison to the other semiconductors used for direct detection .Table 1.1 summarizes these properties. From the table it can be seen that the value of Atomic Number (Z) for HgI2 is high enough for the efficient absor ption of X-rays since the predominant photoelectric absorption process for X-rays in materials is proportional to the Atomic number, Z .The X-ray energy requi red to generate an electron – hole pair for HgI2 is also low which results in high signal strength and resolution. The high mobilitylifetime product in HgI2 results in greater di stance traversed by the generated charges and thus leads to better charge collection and high sensitivity. The lower voltage operat ion allows low voltage electronic design.

PAGE 12

4 Table 1.1 Comparisons of Some X-ra y Sensitive Semiconductor Materials for Direct Detection [4] Finally this chapter ends with a brief overview of some applications of HgI2 detectors [1]. 1) In medical diagnostic applicat ions especially for digital mammography. 2) In non destructive evaluation of materials during security checks at ports. 3) In geological and marine explorations of minerals.

PAGE 13

5 4) In environment pollution monitoring. 5) In x-ray analysis of biological samples and astronomical observations. The following chapter discusses some of the Semiconductor processes relevant to study of Photoconductive detectors. An understanding of these pr ocesses is needed to ex amine the operation and performance of Photoconductive detectors.

PAGE 14

6 CHAPTER 2 SEMICONDUCTOR PROCESSES 2.1 Introduction The variability of the Electric al properties of Semiconductors makes them the perfect choice for fabricat ing electronic devices. Their electrical conductivities lie between those of Meta ls and Insulators. Their electrical conductivity can be varied drastically by means of change in temperature, Optical excitation and by the incorporation of impu rities in the material by a process named as Doping. Every Solid be it Me tal, Insulator and Semiconductor has its own characteristic energy band structure. All band structures are characterized by an upper band of allowed states ca lled the Conduction band and a band of allowed states below this band called the Valence band which is mostly filled with electrons. The valence band and the conduction band may or may not be separated by a forbidden ga p of allowed states ca lled the Band gap. The distribution of states in each of these ba nds according to the energy is given by the Density of States function ,denoted by gc(E) and gv(E) for the conduction and valence band respectively.It thus repres ents the number of states that are available at an energy E. The probability that an av ailable state at an energy E is occupied by an electron is given by t he probability distribution function called

PAGE 15

7 Fermi Function, denoted by f(E).Mathem atically the Fermi function can be expressed at the rmal equilibrium as, 1 f (E)= 1 + exp (E-EF)/ kT Where EF = Fermi energy or Fermi Level in eV k = Boltzmann constant in eV/K T = Temperature in K At a temperature of 0K all available energy states below the Fermi energy level are completely fill ed. The probability that a st ate at a energy E is not occupied is given by 1-f(E). The products gc(E).f (E) and gv(E). (1-f (E)) therefore represent the distribution of electrons and holes in the conduction band and valence band respectively. In metals the conduction and the valence band typically overlap or the conduction band is only partially fi lled. This leads to the intermixing of electrons and empty energy st ates which leads to high conductivity of metals in the presence of an electric field.In Insu lators the magnitude of energy gap is large for e.g. in Diamond an insulator it is about 5eV.At 0 K the valence band in diamond is completely filled whereas t he conduction band is empty. Due to non availability of empty states in the valence band ther e can be no charge transport within the valence band and du e to the absence of elec trons in the conduction

PAGE 16

8 band there can be no charge trans port there either. Therefore insulators possess high resistivity.In semiconductors the band gap is typically much lower (0.1eV3.0eV) than in insulators which allows el ectrons to be easily excited from the valence band to the conduction band by m eans of supplying thermal or optical energy. Such excitation is difficult to achieve in Insulators. In this aspect Semiconductors differ from Insulators even though their band structure at 0 K is similar to that of the Insulators with a completely filled Valence band and an empty Conduction band. 2.2 Carrier Transport Electrical conduction in semiconductors is due to the movement of both the electrons and the holes in the conduction and the valence bands respectively. For this to take place the fi rst requirement is that there should be a partially filled band since a completely filled band and a completely empty band will not conduct. The second r equirement is that the carrier motion should have a net direction and the two mechanisms by which such a directional motion is possible are known as Drift and Diffusion[ 5] .Normally under thermal equilibrium mobile electrons in the conduction band and holes in the va lence band are in random thermal motion resulting in zero net current .The randomness in the motion is caused by carrier scattering mechanisms such as phonon (lattice)scattering, ionized impurity sca ttering ,neutral impurity atom scattering, carrier-carrier scattering, crystal defects sca ttering etc. Now when a external field ( ) is applied across the semiconductor as shown in Fig.2.1( a), the resulting

PAGE 17

9 force on the carriers tends to accelerate t he +q charged holes in the direction of electric field and the –q c harged electrons opposite to t he direction of electric field, where q stands for the electronic charge. But due to scattering caused by the above mentioned mechanisms the motion of carriers for eg holes, though in the direction of the field occurs in a disj ointed fashion involving repeated periods of accelerations and subsequent deceleratio ns as shown in Fig.2.1(b) .Thus the microscopic motion analysis looks so mewhat complicated but measurable quantities being macroscopic in nature reflect the average or overall motion of the carriers. When such an average is carried out over all the carriers the resultant motion can be visualized on a macroscopic scale as in Fig 2.1(c) [6].The motion of each carrier now can be described in terms of a constant velocity vd called drift velocity Such a mechanism of charged particle motion in presence of an applied field is known as Drift. Figure 2.1 Visualization of Carrier Drift [6] It is to be noted that above drifting motion is actually superimposed on the always present thermal motion of the ca rriers which being completely random in nature averages out to zero on a ma croscopic scale and thus does not contribute to carrier transport and is t herefore neglected. T he measured drift

PAGE 18

10 velocity (vd) in semiconductors for low to moderat e field values of electric field ( ) is directly proportional to t he applied electric field. Th e constant of proportionality is termed as Mobility ( ).This is an important paramet er which plays a key role in characterizing the performanc e of a device.It can be interpretated as the ease with which a carrier can drift within a semiconductor crystal. It is a function of the amount of carrier scattering, temperature and the doping of the semiconductor. At higher electric fields the drift velocity tends to saturate thereby becoming field independent [7]. The second mechani sm called Diffusion arises when there is a non uniform density of carriers – electrons and holes. Thus in the absence of any other processes such as drift ,the carriers will di ffuse from a region of high density to a region of low density.Also thermal moti on not interparticle repulsion is the enabling action behind the diffusion proce ss. Fig .2.2 (a) [5] shows a 2-D representation of the Di ffusion process on a micro scopic scale.Under thermal equilibrium each carrier has equal probabilit y of moving in either the –x or +x direction.The graph shows the carrier c oncentration vs. distance at various instances of time t such that t1
PAGE 19

11 (a) (b) Figure 2.2 Visualization of the Carrier Diffusion Process [5, 6] Having discussed about the carri er motion processes in the semiconductors it is now necessary to ex plain certain concepts related to carrier generation and recombination processes whic h are important to the subject of the present work.

PAGE 20

12 2.3 Carrier Recombinati on and Generation Processes When a semiconductor is disturbed from the equilibrium state, an excess or deficit in the carrier concentra tions relative to their equi librium values is invariably created inside the semiconductor. In this si tuation there comes into existence the process of carrier recombination and generation which tries to stabilize the carrier excess or deficit if the pertur bation is still maintained and if not the process tries to eliminate such an excess or deficit of carrier s. Since one often encounters non equilibrium conditions dur ing device operation this process therefore plays a vital role in shaping t he characteristics exhibited by a device. Broadly speaking a Recombination proces s is defined as one which destroys or annihilates electrons and holes. Generati on on the other is a mechanism by which electrons and holes are created. Also unlike Dr ift and Diffusion mechanisms the terms Recombination and Generation do not refer to a single process but they are colle ctive names for a group of similar processes which means that carriers can be destroyed an d created within the semiconductor in a number of ways. It is ther efore convenient to denote these processes as R-G processes. The most import ant of these processes are shown in Fig 2.3 [6]. A brief explanation of each proc ess shall follow thereafter.

PAGE 21

13 Figure 2.3 Energy Band Visualization of Bulk Recombination and Generation Processes [6]

PAGE 22

14 2.3.1 Recombinat ion Processes 2.3.1.1 Band-toBand Recombination This is conceptually the si mplest of all recombination processes. As shown in Fig. 2.3(a) it merely involves the direct annihilation of a conduction band electron and a valenc e band hole. When the electron and hole moving in the semiconductor lattice stray into the same spatial vicinity they annihilate each other and the ex cess energy released during this process typically goes into the production of a phot on of energy approximately equal to the band gap ener gy. This recombination is also called direct thermal recombination. 2.3.1.2 R-G Center Recombination This type of process usual ly plays a central role and often dominates other recombination and generat ion processes. This process is pictured in Fig. 2.3(b) in volves a third party or a intermediary and takes place only at special locations within the semi conductors known as R-G centers. RG centers are lattice defects or special im purity atoms such as gold in Si.Even in semiconductors of highest available pur ity, lattice defects and unintentional impurities are always present. The RG center concentration however is normally very low compared to the a cceptor and donor concentration in device quality materials.The most import ant property of these R-G centers is the introduction of allowed energy levels generally near the center of band gap.It can be shown that the most effi cient R-G centers are those that

PAGE 23

15 introduce energy levels at the mid band gap region [8]. Actual semiconductors can have a number of deep level R-G c enters but the process is usually dominated by only one type of R-G center. As shown in Fig.2.3(b) first one type of carrier and then the other type of carrier is a ttracted to the R-G center. Thus the capture of an el ectron and hole at the sa me site leads to the annihilation of the electron–hol e pair. This can also be equivalently visualized as the transition of a carrier first to the R-G center and then an annihilating transition to the opposite carrier band. R-G center recombination also called indirect thermal recombination is char acteristically non-radiative. Thermal energy (heat) is released dur ing the process or equival ently lattice vibrations also called phonons are pr oduced. An important clarif ication has to be made at this juncture regarding the diffe rence between the terms Trap and R-G center.Broadly speaking the chemical impurities and lattice defects introduce localized allowed energy states in the band gap region.A localized energy state can generally be effective in only one way either as a Trap or a R-G center .If a carrier that is attracted to a localized state is re-excited back after a certain time to its corresponding ener gy band before it recombines with an opposite carrier then the localized state is said to be acting as a Trap for that carrier. If the same state however assist s in recombination then it is termed as a R-G center.Generally energy states introduced by impurities and defects which are close to either conduction or valence band edges act as effective electron and hole traps respectively.Char ge carrier trapping in many devices can play a pivotal role in affecting t heir performance.The ti me that a charge

PAGE 24

16 carrier is free so as to contribute to c onductivity is called free life time of the charge. It is denoted by .It is the time t hat an excited electron spends in the conduction band or the time that an exci ted hole spends in the valence band. The time spent by the electron or hole in the trap is not included in the free life time. It is represented by n and p for electron and hole respectively.For a direct band to band recomb ination both of these life times are equal because annihilation of both electr on and hole occurs at the same time but for R-G center recombination this may not be true.The free lifetime of a charge carrier can be [9] : (a) Terminated by recombination, or if the carriers are extracted from the semiconductor without bei ng replenished from the opposite electrode. (b) Interrupted if the carrier is trapped, to be resumed after the carrier is freed from the trap. (c) Undisturbed if the carrier is extrac ted from the semiconductor by the applied electric field at the same time as an identical carrier is injected into the semiconductor from the opposite electrode. The free life time of a carrier is inve rsely proportional to the product of number of R-G centers per cm3 and the capture cross section ( cm-2) of the R-G center.The capture cross section measur es the effectivenesss of an R-G center or a trap to capture a free carrier. Each localized state in the band gap region has two capture cross sections one for t he electron and one for the hole. For a

PAGE 25

17 energy state in the band gap acting as a true R-G center n= p.For a state acting as True Hole trap p>> n and for true electron trap n>> p [5]. 2.3.1.3 Auger Recombination In the Auger process shown in Fig. 2.3(c) a band to band to recombination occurs simultaneously with the collision between two like carriers.The energy released by the reco mbination is transferred during the collision to the surviving carrier.This hi ghly energetic carrier subsequently “thermalizes “loses energy in small steps through heat producing collisions with the semiconductor lattice as shown in the picture. 2.3.2 Generation Processes All of the above recombination processes can be reversed so as to generate free carriers. Fig. 2.3(d) shows the band to band generation where an electron is excited directly from the va lence band to the conduction band. Either thermal energy or Light energy can be us ed to provide the energy required for band to band transition. If thermal energy is used then the process is termed as Direct Thermal generation. If externally introduced light is absorbed then the process is called Photogener ation .The thermally assi sted generation of carriers with R-G centers acting as intermediaries is envisi oned in Fig 2.3(e).The photoemission of carriers from band gap centers also can al so be pictured but it is typically a rather impr obable process. Finally impact ionization, the inverse of Auger recombination is show n in Fig. 2.3(e).In this process an electron –hole pair

PAGE 26

18 is formed due to the energy released when a highly energetic electron collides with the crystal lattice. Such a generatio n of charges usually occurs in high Electric field regions of the device [6]. The recombination and generation processes discussed above occur at all times in the semiconductors -they even occur when under thermal equilibrium.Under thermal equilibrium c onditions however each fundamental process and its inverse must self balance independent of any other process occurring inside the semiconductor[10].Generally one is concerned with only the process which is dominant among all other processes occurring in a semiconductor under certain conditions.N ot all processes occur at the same rate under certain conditions. In order to get a better understanding of the processes which would domin ate in a semiconductor it is essential that one visualize these processes by means of changes in crystal momentum in addition to energy change. This is because in any R-G process both energy and crystal momentum has to be c onserved. Crystal momentum ( k) related aspects of R-G processes are convenient ly discussed by means of Energy – Momentum (E-k) plots. Semiconducto rs can be classified as Direct and Indirect Semiconductors on the basis of these E-k plots. 2.4 Direct and Indi rect Semiconductors The difference between the two types of semiconductors is evident in the E-k plots of Fig.2.4 [6]. In a Direct semi conductor both the minimum energy of the

PAGE 27

19 conduction band and the maximum energy of the valence band occur at k =0.In an Indirect semiconductor the conduction band is displaced to k 0.Here EC and EV refer to conduction band minimum and val ence band maximum respectively. Figure 2.4 The Two Different Types of Semiconductors [6] In order to visualize a R-G proce ss using a E-k plot an understanding of the nature of transitions associated with the absorpt ion and emission of photons (light) and lattice vibrat ion quanta called phonons is necessary.Photons, being massless identities,carry very little mom entum and a photon assisted transition is essentially vertical on the E-k pl ot.On the other hand the thermal energy associated with lattice vibrations (phonons) is very small but the phonon momentum is comparativ ely large.Thus on the Ek plot the phonon assisted transition is essentially hor izontal.Also it has to be me ntioned here that electrons and holes usually occupy states close to EC minimum and EV maximum respectively[6].Now as shown in Fig. 2.5 (a), a band to band recombination in a

PAGE 28

20 direct semiconductor effectively proceeds because the k values of all the electrons and holes are bunc hed at k=0, leading to a low change in momentum for the recombination process. Conservation of both ener gy and momentum is readily met by emission of a photon as shown. A band to band to recombination in an indirect semiconductor as visualized in Fig.2.5 (b) requires a large change in momentum.Consequently t he emission of a photon mu st be accompanied by a emission or absorption of phonons.The rather involved nature of such a band to band recombination in indirect semiconduc tors leads to a vastly diminished rate.In an indirect semiconductor an elec tron in the conduction band minimum at k 0 cannot recombine with a hole at k= 0 in the valence band maximum unless a phonon of the right energy and momentum is available.Both phonon emission and phonon absorption c an assist the downwar d transition.In order for the right phonon collision to occur the dwell time of the electron in the conduction band increases..It therefore implies that the electron will more likely recombine non radiatively with a hole thr ough R-G centers which are always present due to impurities and lattice defects in the crystal .Thus Band to Band recombination is in fact totally negligible compared to RG center recombinat ion in indirect semiconductors. Therefore t he probability of radiative re combination in a Direct semiconductor is very high compared to radiative recombination in indirect semiconductors. The non radiative competi ng processes reduce the probability of radiative recombination in indirect band gap materials.Similarily Band to Band absorption of light by an elec tron in an Indirect Semiconductor also involves a

PAGE 29

21 phonon for momentum conservation and the probability of this happening is thus reduced when compared to such an absorption in a Direct semiconductor. Figure 2.5 E-k plots for Visualizations of Recombination (band to band) in Direct and Indirect Semiconductors [6] The Recombination -Generation mechanisms discussed so far take place in the bulk of the semiconductor as opposed to Surface Recombination – Generation taking place near the vicinity of Semiconductor surface via interaction with interfacial traps. Though the former is important, the latter is as important and many times more import ant than the former.Since t he properties of many photodetectors are affected by generation and recombinat ion at the surface, a brief discussion is in order.

PAGE 30

22 2.5 Surface Recombination-Generation Figure 2.6 Abrupt Termination of th e Semiconductor Lattice Structure at the Surface [11] As shown in the Fig.2.6 there is abrupt termination of the lattice structure at the semiconductor surface. The surface usually consists of dangling bonds or bonds that are satisfied by atoms other than the host atoms. The presence of dangling bonds, surface imperfections, forei gn atoms on the surfac e, etc result in the introduction of a large number (more than in the bulk) of localized R-G centers at the surface regi on called the surface centers or states.Unlike the bulk R-G centers however ,the surface c enters are found to be continuously distributed in energy throughout the semi conductor band gap .This is pictured in

PAGE 31

23 Fig. 2.7 The same fundamental processes that occur in the bulk also occur at the semiconductor surface.Electrons and ho les can be captured at these surface centers and also can be emi tted from theses surface centers. Though seemingly possible from the energy band diagram the additional transitions occurring between surface centers of different ener gies is extremely unlikely because of the spread out or spat ially isolated nature of centers on the surfac e plane. This is shown in Fig. 2.7(c). (c) Figure 2.7 Processes at the Semiconductor Surface The parameter responsible for gaugi ng the surface recombination and surface generation rate is the Surfac e recombination and surface generation

PAGE 32

24 velocity respectively. Both have the units of velocity (cm/sec).The higher the surface recombination (generation) velocity the greater the surface recombination (generation). Both depe nd on the state of the semiconductor surface whether accumulated, depleted or inverted .Surface recombination(generation) velocity is also proportional to the pr oduct of the total surface centers per unit area (NST) and the capture cross section of the surface center [10]. With the introduction of various processes in a semiconductor which influence carrier motion and their numbers it is now time to analyze one fundamental process of carrier generation relevant to the present work in greater detail namely the process of Photogenerat ion in semiconductors and also the operation of Photoconductive detectors. This shall be the subject of discussion in the next chapter.

PAGE 33

25 CHAPTER 3 PHOTOGENERATION AND PHOTOCONDUCTIVE DETECTORS 3.1 Photogeneration The first stage in the operation of any pho todetector is the absorption of light quanta of appropriate energy resulting in the generation of free charge carriers.This is the basic principle in converting any electromagnetic radiation energy into electrical energy. There ar e several processes by which photons (light) are absorbed in matter [1 1] of which the process of interest for the present study is the Photogeneration. Simply stated it is the process of exciting an electron from the valence band to conduc tion band leading to formation of free electron in the conduction ban d and free hole in the valence band by means of absorption of a photon.The required ph oton energy for photoge neration is thus atleast equal to the band gap energy.As pr eviously discussed this is a special case of Band to Band transition assisted by the absorption of light. Such a band to band absorption process is also called the fundamental (absor ption) process. Independent of the process of absorption, the Intensity (Optical power per unit area) of the light decreases with dist ance inside a Semiconductor according to the well known law known as the Beer’s Law as illustrared in Fig.3.1.The plot shows the exponential decay of intensity of photons with respect to the distance (x direction) in the mate rial .Here the intensity of the incident photons is Io ,It is the

PAGE 34

26 intensity of transmitted photons at the distance t ,where t is the thickness of the Semiconductor. Here it is assumed t hat photons do not undergo any kind reflection at the front and back su rfaces of the semiconductor. The Beer’s Law in the form of an equation can be expressed as I(x) = I0 .e x I(x) is the intensity at a dist ance x in the Semiconductor. The quantity of interest in this equation is (cm-1), known as the absorption coefficient of the semiconductor. The degree of absorptio n of light in a semiconductor is quantified by its absorption coefficient. It is a strong function of material and the wavelength of incident photons.It generally varies with different Io It t Distance Intensity I0 .exFigure 3.1 Exponential Decay of P hoton Intensity Inside a Semiconductor 0

PAGE 35

27 absorption processes.For t he fundamental absorption process discussed at the beginning of this section the absorption coefficient increases rapidly with the photon energy above the band gap energy of the semiconductor under consideration,and for Direct Semiconductors the absorption coefficient is found to be greater than in Indirect Semiconducto rs.This suggests stronger absorption of light in Direct Semiconductors when co mpared to Indirect semiconductors the reasons for which have al ready been discussed in the previous chapter.The significance of the absorption coefficient al so lies in the fact that its reciprocal represents the distance over which the Intensity of photons falls to a value of 1/e. This distance given by x=1/ is called the penetration or absorption depth.For complete absorption of li ght of appropriate wavelength the thickness of the semiconductor should be greater than the absorption depth. T he energy of each photon is given by the relation, Ephoton = hc / Where h is the Planck’s constant in eV-s c is t he speed of light in Vacuum in m/s is the wavelength of light in m Thus based upon the above facts lower wavelength light consisting of higher energy photons is absorbed strongly at the front surface of the semiconductor region compared to longer wa velength light .W hen the energy of each photon is below the band gap of the se miconductor then the absorpti on is negligible and than the semiconductor is said to be transpar ent to such light. This is possible

PAGE 36

28 only if the semiconductor is assumed to be free of im purities and defects which would give rise to energy states in the band gap region and consequently absorption of light of ener gy below band gap is also possible.Such an absorption from the impurity levels in the band gap is known as impurity absorption. Of course there are other absorption proce sses giving rise to absorption below the band gap energy which shall not be discussed here. 3.2 Photoconductor The photoconductor is the simplest photodetector consisting of a slab of semiconductor (in bulk or thin-film form) with two metallic contacts or electrodes fixed at the opposite ends and across which a voltage is applied.The voltage applied is a dc voltage in the present case. Its operation is based on Photoconductivity. Photoconductivity is t he property of a Semiconductor by which its bulk conductivity increases due to absorption of light. Generally in a semiconductor under thermal equilibrium there is a balance between thermally generated free carriers and their recombinat ion which determines their number. Additional free carriers can be produced by m eans of absorption of light resulting in increased conductivity which persists as long as the additional carriers recombine or until they are extracted out at the electr ode without being replenished from the opposite electrode .B ased on the light absorption process photoconductors are divided into two types namely Intrinsic and Extrinsic Photoconductors.In Intrinsic photoconduc tors the light is absorbed by the process of photogeneration in volving creation of free carriers by band to band

PAGE 37

29 transition.A photon of energy equal to or greater than the energy gap is typically required.In an extrinsic photoconductor light is absorbed by filled impurity states in the band gap region resulting in produc tion of free carriers. Focus here will be only on the Intrinsic photoconductor. So me terms related to photoconductors can be defined as below. 3.2.1 Qu antum Efficiency (QE) There are two types of quantum efficiency Internal and External .Internal quantum efficiency is the ratio of number of electron –hole pairs which are created per second and detected to the number of phot ons absorbed per second in the photoconductor.External quan tum efficiency on the other hand is the ratio of the number of electron-hole pairs which are created per second and detected to the number of photons per se cond impinging on the semiconductor surface.Thus the external quantum effi ciency does not take into account the photons lost due to reflection at the surface or transmission.Consequently Internal quantum efficiency is higher th an External quantum efficiency. The maximum value of quantum efficiency can be 1. QE is a function of wavelength. 3.2.2 Gain Before recombination occurs the Gain in a Photoconductor is defined as follows: G = (distance traversed by a electron) + (distance traversed by a hole) Dis tance between the electrodes

PAGE 38

30 Normally when a photon is absorbed in a pho toconductor an electron hole pair is created and if both of them reach their respective el ectrodes, then the photon is said to detected. The Gain in this case is 1.This is usually the case when the contacts to the semiconductor are of bl ocking nature which do not allow carriers to enter the photoconductor from the ex ternal circuit .Thus when an excited carrier exits the photoconductor from one elec trode a similar carrier will not enter the photoconductor from the opposite electrode to maintain charge neutrality.But in the case of Ohmic contac ts which allow carriers to fl ow in either direction with ease the charge neutrality condition is satisfi ed and therefore it is possible for an excited carrier usually the electron( as its mobility is higher than that of hole) to reach its electrode sooner t han the hole and therefore due to charge neutrality the external circuit provides another el ectron to the photoconductor which once again moves faster than the hole reaching its electrode before the hole reaches its electrode.This process will continue un til the electron recombines with the hole.Thus it is possible for the elec tron to traverse the distance between the electrodes more than once.Therefore t he gain of the photoc onductor is more than 1 in this case [9].Of course for this to happen the free lifetime of electron must be greater than its transit time(def ined as the time time taken by the electron to travel the di stance between the electrodes).

PAGE 39

31 3.2.3 Photocurrent The current due to flow of e xcess photogenerated carriers in the presence of an applied field is called is called Photocurrent.An expression for photocurrent in a photoconductor can be written as follows : Iph = e.G. F where, e is the electronic charge,G is photoconductor gain and F is the rate at which the electron hole pairs are created in the photoconductor.It can be shown that G is proportional to the su m of the mobilitylifetime products of electron and hole [9].Therefore higher va lues of these products result in greater gain and consequently higher photocurrent. 3.2.4 Dark Current Basically th e dark current refers to the current flowing through the photoconductor when it is not illuminated. This current is mainly due to the flow of thermally generated carriers in the photoconductor.In some cases these carriers are higher than those excite d by light and therefore this affects the performance of the photoconductor by causing rapid fluctuations in the output current which is a sum of bot h the photocurrent and dark current. Therefore. materials with a high band g ap must be chosen to minimize the effect of the dark currents.

PAGE 40

32 3.2.5 Spectral Response A given photoconductor can be generally characterized by studying its Spectral response, which is plot of its External quantum efficiency(a function of wavelength) vers us the wavelength of light.In the present work the Optical response of the Mercuric iodi de Photoconductive detectors is analyzed by its Spectral response. Photoconductivity is one of a few topics where the experimental setup, contact configuration, prepar ation of the sample and sa mple thickness all play a very vital role in the output and t he analysis of the results.Two most commonly used configurations for photoconductor characterizations are shown in Fig.3.2. Fig 3. 2(a) shows a configuratio n in which the device is illuminated along t he direction of electric fiel d through a transparent contact. Also shown here is the negative bias applied to the illuminated contact. Fig. 3.2 (b) shows a configurat ion in which the device is illuminated perpendicular to the direction of electric field in the photoconductor.

PAGE 41

33 (a) (b) Figure 3.2 Different Experiment al Configurations of a P hotoconductor

PAGE 42

34 The configuration of Fig 3.2(a) is used to obtain the Spectral response I-V curves for Thin Film Polycrystalline Merc uric Iodide photoconductive detectors in this work. The above discussion on photoconductivity thus outlines some of its salient features and it is important to mention here that the photoconduction process involves generation, recombinat ion and the transport of carriers to the electrodes. It is thus a complex process in which many vitals issues relating to above mechanisms have to studied. Inspite of this complexity the photoconduction process provides valuable information about physical properties of materials and thus offers application in photodet ection and radiation measurements.Recent advances in thin film technology particularly in excellent quality crystal growth have gi ven a new dimension to th e field of photoconductors and now quantum well and superlattice photodetectors are a reality [13]. The next chapter addresses some of the important properties and issues of Mercuric Iodide material in relation to its use as detectors and investigated in some of the technical papers on this subject.

PAGE 43

35 CHAPTER 4 MERCURIC IODIDE 4.1 Literature Review Mercuric I odide was one of the earliest materials to be investigated for photoconductivity. In 1903 it was shown that HgI2 could be used with gelatin to form a photographic emulsion .Other wo rkers thereafter studied the spectral response and photoconductivity mechanism in HgI2 .The photoelectric properties of HgI2 were extensively studied by Bube [14] both in its red phase ( -HgI2 ) and on phase transformation to yellow ( HgI2) phase. HgI2 crystallizes in the red tetragonal structure ( -HgI2),which undergoes a phase transformation to yellow orthorhombic ( HgI2 ) form at 400oK.Research carried out on the orthorhombic form of HgI2 revealed that its photosensitivity was about 0.1% of that of the red tetragonal HgI2[14,24] .Therefore the tetragonal form of HgI2 is best suited for Xray and gamma ray detector fabrication. The Crystal growth techniques, Electrical and Optica l properties of HgI2 have been extensively studied with respect to their radiation detection applications [1] and HgI2 has proved to be an excellent material for the fabrication of Ro om temperature X-ray and gamma ray detectors. HgI2 has a band gap of 2.1eV at room temperature and it has a small temperature coefficient of approximately 10-4eV per kelvin which results in a small thermal carrier generat ion over a wide range of temperatures. The dark

PAGE 44

36 resistivity of good quality crystalline material is of the order of 1012 ohm-cm or higher.The density of the material (6.3g/cm3) results in a large absorption coefficient.The high values of the atomic numbers of the constituent elements(Hg-80, I-53) result in a very large photoelectric effect since the photoelectric interactions are proportional to Z5,where Z is the atomic number of the interacting material. Single crystalline HgI2 has been effectively studied and used in the fabrication of X ray and Gamma ray detectors[1,15]. M.Schieber et al.[16,17] have reported that polycrystalline HgI2 fabricated by means of physical vapor deposition (PVD) technique hav e higher sensitivity to radiological X rays which is comparable to the result s measured with single crystals of HgI2 and they also found out that the Electrical properties of the PVD polycrystalline HgI2 were near to those of single crystalline HgI2.Thus large area thin film polycrystalline HgI2 which has a much lower production cost is a prospective candidate for direct X -ray radiology A challenging and sometimes more controversial issue in mercuric iodide grow th is the subject of the stoichiometry [1].The range of composition over whic h mercuric iodide can exist without a change of phase is denoted by HgI2-x,where 0<< |x| <<1,x>0 for mercury rich compositions and x<0 for iodine rich compos itions.Stoichiometric mercuric iodide is the term referring to the case of x = 0.The term “near stoich iometric” refers to the range of x over which the single phase tetragonal mercuric iodide is stable.Evidence that deviati ons on both the iodine and me rcury rich sides of the stoichiometric compositions were detri mental to nuclear radiation detector performance has been reported by Tadjine et al. [18] and it was concluded that

PAGE 45

37 the 2.00 I:Hg ratio was best. HgI2 detectors are exclusively biased along the crystallographic c-axis for carrier co llection and transport properties do depend on the crystallographic direction [1]. Since no dopants are have been found to lower the resisitivity in HgI2 ,shallow levels in HgI2 are either high ly compensated or they do not exist at a ll.Therefore deep levels are important in affecting the performance of HgI2 nuclear spectrometers. R.B James et al. [19] studied the nature and origin of d eep level traps in HgI2 material. They show ed that some of the trap types and their concentration wa s a function of the metal over layer employed as a contact material. Anot her important issue facing the HgI2 detectors is the so called polarization effe ct.The polarization effect is defined as the time dependence of the detector per formance that results from the application of the bias field .The effect usually changes the electric field within the device. This thus alte rs the charge collection effi ciency (proportional to the electric field). Possible causes of pol arization include tr apping, detrapping and change of defect structure in the detecto r [1].Light spot scanning measurements conducted by Bube [14] on HgI2 revealed that the photocu rrent is mainly limited to a small region at or near the cathode i.e. as the li ght spot is moved between the cathode and the anode the cu rrent is maximum when t he spot is near the cathode.This suggests that the photocurrent is mainly due to the contribution from the electrons. Consequently for higher photocurrents the illuminated electrode is negatively biased. T he I-V measurements on the HgI2 -based devices appear often not to be re producible as the resu lts strongly depend on the experimental procedure. Not only the voltage increment and time left before

PAGE 46

38 increasing the voltage but also conditions of polariz ation before recording the I-V curves, or the time during which the sample as been left unbiased before the actual measurement also are known to a ffect the I-V characteristics of the HgI2 device [20].Surface recombination whic h also influences the performance of HgI2 was studied by Levi et al [21] and Z.Burshtein et al. [22].They obtained values for both electron and hole surfac e recombination velocities in their samples from the Photoconductivity vers us Voltage curves .Choice of suitable contact materials to the HgI2 has always posed a challenge because of the reaction between the cont act materials and HgI2.Silver,Indium and Gallium were all tried but they all reacted with HgI2 within minutes causing permanent changes to the device[14].When Cu was tried it was found to be devastating to the detector due to Cu diffusion. Palladi um, Indium tin Oxide (Transparent conducting oxide), Carbon and Gold have been successfully used as contacts for HgI2 detectors. With the exception of carbon all of these contacts are deposited either by therma l evaporation or sputtering .Palladium contacts are however nowadays widely used due to their better performance relatively and the high quality of detectors that c an be fabricated wit h them [1]. 4.2 Crystal Structure The primitive unit ce ll of tetragonal HgI2 is shown in Fig.4.1.The unit cell comprises of two Hg atoms and fo ur I atoms.Thus the unit cell has two molecules of HgI2.The nearest neighbour Hg-I spaci ng is 2.78 .The crystal has

PAGE 47

39 an inversion symmetry about the midpoint between the two Hg atoms. The I atoms form a cubic close packing of spheres, and the Hg atoms occupy tetrahedral voids such as to form layers of cornered shared HgI4 tetrahedra where all I-Hg-I and Hg-I-Hg angles are tetrahedral. The lattice constants are a=b=4.361 and c= 12.450 at room temperature. [1, 23].If one chooses the midpoint between the two Hg atoms as th e origin of the coordinate system, the atomic positions of the two Hg atoms are (-a/4,-a/4,-c/4) and (a/4,a/4,c/4) and those of the four I atoms are (-a/4,a/4,0.111c),(a/4,-a/4,0.111c ),(-a/4,a/4,0.389c) and (a/4, -a/4, -0.389c). Figure 4.1 Tetra gonal Unit Cell of HgI2 [24] The Final chapter in this present work discusses t he Results and Analysis pertaining to the Optical response of Th in Film Polycrystalline Mercuric Iodide Devices.

PAGE 48

40 CHAPTER 5 RESULTS AND DISCUSSION 5.1 Structure and Layout of the Films Thin Films of polycrystalline HgI2 which were anal yzed in this work were deposited on a TEC 15 LOF glass substrates coated with a Tin oxide (SnO2) layer which served as a growth surf ace and as a front contact to the devices on the film. The deposition technique employ ed was the Physical Vapor Deposition (PVD).The surface area of deposition had a diam eter of about 27mm.The thickness of the films ranged from 132.61m to 423.63m.The films showed a preferential (001) orientation such that the c-axis is perpendicular to the substrate. This is also one of t he prerequisites to obtaining high quality polycrystalline PVD HgI2 films [25].Pd contacts were sputtered on to the film through a shadow mask ,with the circular contact area having a diameter of 2mm.Pd contacts function as the bac k contacts of the devices on the film.Seperate leads were t hen attached to front and back contacts for the purpose of application of bias.In addition to this a laye r of parylene surrounds the HgI2 film, ensuring the reliabilit y of the films over a long period. It also prevents accidental short circuit between the front and back contacts during measurements.Each Sample is numbered after the date the film on it was fabricated.Each film sample on the substrat e consists of seven devices (spots)

PAGE 49

41 with each device comprising of substrate (glass),SnO2(front contact),HgI2 film and a Pd back contact. Fig 5.1 (a) shows the layout of the top view of the film Sample and (b) shows the cross-sectional view of the sample. (a) (b) Figure 5.1 Two Different Views of the Sample Glass Parylene Pd HgI2 SnO2 TEC 15 LOF GLASS SnO2 HgI2 Pd Pd Pd Pd Lead for Front Contact Lead for back contact

PAGE 50

42 5.2 Experimental Procedure The Mercuric Iodide Samples were tested for their Spectral response and Current-Voltage (I-V) response in both light and dark conditions.For both these measurements the light sour ce used was a TungstenHalogen Lamp.This lamp is a good choice for recording photoconduct ivity data as its spectral output is in the range 280nm-2500nm [13].Wavelength selection was achieved by means of a Lab Viewtm data logging program in conj unction with an Oriel grating monochromator.The currents were recorded by using a Keithly 617 programmable Electrometer. Sign of the bi as on each device is with respect to the bias applied to the transparent illuminated contact, which is the SnO2 contact in this case. This contact is termed as the ‘Front Contact’ and the palladium acts as the ‘Back Contact’.It is generally obser ved that the photoc urrent and thus the spectral response is higher when the illuminated contact is negatively biased .This is attributed to the better trans port properties of the electrons which traverse the length of the device for this configuration.Therefore the SnO2 contact is negatively biased and the experimental configuration resembles that shown in Fig. 3.2(a).However current s with positive bias to the front contact were also recorded for comparison purposes. The light from the exit window of the monochromator was focused through the glass substrate (TEC 15 LOF) on a circular area of SnO2 of 2mm diameter by means of a focusing lens.This circular area was in line with the circular area of the back contact of palladium of diameter 2mm. The Spectral response fo r each device was obtained at a voltage of -50V applied to the illuminated fr ont contact. For I-V measurements the

PAGE 51

43 voltage was increased from 0V to 50V in steps of 10V and recorded at the peak spectral response wavelength.I-V measur ements were carried out in both Light and dark conditions. Because of the possibilit y of breakdown of the thin films the voltage was not increased beyond 50V. 5.3 Optical Response of the Films A total of 15 samples were studied in this work .Table 5.1 summarizes the important deposition feat ures of these samples. The total time for deposition is also shown. The films were mostly deposited by the three step process of PVD.In the three step process the substrate was in itially maintained at a low temperature then raised to an intermediate temperatur e and elevated to a final temperature which was maintained until the desired thickness was obtained. The source temperature was also varied during growth .In a modified three step process the substrate temperature is increased from an initial temperature to a final temperature gradually in st eps of time. During this increase the source temperature was not changed.In anot her deposition process called the temperature oscillation process the s ubstrate was oscillated between a low and a high temperature for a number of times. The temperature of the substrate was then maintained at either the low or hi gh temperature to obtain the desired thickness [25].Two films were also fabr icated by the two st ep process of PVD wherein the substrate temperature was ma intained at an initia l low temperature and finally elevated to a final temperat ure where it stayed till the desired thickness was obtained.

PAGE 52

44 Table 5.1 Deposition Para meters of the Samples Sample TSRC (oC) TSUB (oC) Thickness ( m) Type of Process Total Time (hr) 11-19-02 70, 105 5,35,59 295. 55 Three Step 4.5 11-22-02 70, 95 5,35,57 270.34 Three Step 4.5 12-02-02 70,105 5,35,59 423.63 Three step 4.0 12-17-02 70,80 5,…50 213.70 Modified Three Step 5.0 12-19-02 70,80 5,…..53 199.90 Modified Three Step 5.5 01-13-03 70 20,……45 132.61 Modified Three Step 4.5 01-14-03 70 20,……45 149.87 Modified Three Step 5.5 01-15-03 70 20,…….47 152.55 Modified Three Step 5.5 10-09-03 80 Oscillation 5,50 181.33 Temperature Oscillation 2.85 10-20-03 80 30,50 261.71 Two Step 4.5 10-21-03 80 Oscillation 50,30 224.26 Temperature Oscillation 3.5 03-10-04 85 Oscillation 5,50 304.401 Temperature Oscillation 4.0 03-25-04 80 20,…….45 141.532 Modified Three Step 4.5 06-02-04 70 Oscillation 50, 30 266.433 Temperature Oscillation 4.0 09-23-04 90 5,40 350.596 Two Step 3.0

PAGE 53

45 Table 5.2 Peak QE’s of the Samples at 50V Bias to the Illuminated Contact Sample Spot # 1 Spot # 2 Spot # 3 Spot # 4 S pot # 5 Spot# 6 Spot # 7 11-19-02 0.209 0.243 0.213 0.288 _ 11-22-02 0.213 0.169 0.174 _ 0.194 12-02-02 0.238 0.221 0.247 0.243 0.277 12-17-02 0.122 0.179 0.174 _ 0.173 12-19-02 0.206 0.215 0.193 0.114 0.304 0.2931 01-13-03 0.132 0.352 0.393 0. 315 0.380 0.176 01-14-03 0.276 0.266 0.271 0. 337 0.275 0.234 01-15-03 0.252 0.205 0.131 0.228 0.174 0.213 10-09-03 0.190 0.145 0.139 0.134 0.216 _ 10-20-03 0.171 0.224 0.169 0. 247 0.206 0.154 10-21-03 0.224 0.128 0.107 0.130 0.161 0.104 03-10-04 0.219 0.209 0.209 0. 274 0.228 0.313 0.181 03-25-04 0.202 0.262 0.279 0. 326 0.321 0.345 0.278 06-02-04 0.167 09-23-04 0.170 0.151 0.119 0.144 0.207 0.240 Table 5.2 lists the Spectral response for the devices on each sample.As seen in the table some of the devices on the samp les failed to produce any response.This may be due to a contact failure .Most of the Sample had QE’s in the range from 0.10.4. These QE’s were recorded at peak wavelengths which ranged from 565nm to 585nm. The absor ption coefficient data for the

PAGE 54

46 polycrystalline mercuric iodide is still in the nascent stage.However the data of single crystal mercuric iodide can be us ed for the present work.This absorption coefficient data is listed in table 5.3 Some important samples shall now be analyzed with respect to their optical response. 5.3.1 Sample 11-19-02 Fig.5.2 (a) shows the Spectral re sponse of sample # 11-19-02. Five devices (spots) on this sample had effi ciency above 0.2.Spot # 5 had the highest efficiency of 0.29 as seen at a wavelength of 565nm. Fig. 5.2 (b) shows the Light I-V curves for this Sample.Dark I-V curv es were however very small and could not be recorded .The peak QE of Spot # 5 at 565nm corresponded to a photocurrent of 2.83 A obtained from the data file (see appendix) used to obtain the spectral response for this spot. But t he photocurrent from Fig 5.2 (b) reveals a value of 2.34 A. The apparent discrepancy may be due to the time lag in obtaining the I-V curves after the measur ement of spectral response. This is observed in all the samples. The measurem ent of I-V curves with both positive and negative polarity helps to separately study the current due to holes and electrons respectively. With the positive polarity to the illuminated front contact and negative polarity to the back contact, holes are attracted to the back contact. Since the carriers are mainly generated within a distance of few microns from the front surface (Table 5.3), the holes have to travel the greater length of the device

PAGE 55

47 SPECTRAL RESPONSE OF SAMPLE 11-19-02 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 400450500550600650700 Wavelength (nm)Q.E. Spot # 1 Spot # 2 Spot # 3 Spot # 5 (a) (b) Figure 5.2 Spectral Response and Li ght I-V Curves for Sample 11-19-02 LIGHT I-V CURVES FOR SAMPLE 11-19-020 0.5 1 1.5 2 2.5-60-40-20 0 20 40 60 Voltage (V)Current (microA) Spot # 1 Spot # 2 Spot # 3 Spot # 5

PAGE 56

48 Table 5.3 Absorption Coefficient Data [26] Wavelength (n m) Absorption Coefficient (cm-1) 400 3.2e5 420 2.55e5 440 1.9e5 460 1.5e5 480 1.25e5 500 1.1e5 520 9.0e4 540 7.0e4 560 2e4 580 300 600 11 620 3.5 till they reach the back contact.Therefore t he current is mainly due to holes.The reverse is true for negative bias to the front illuminated contact.The current then is mainly due to electrons.As expected the photocurrent with negative bias is higher than that with positive bias as show n in Fig.5.2 (b).This is attributed to better electronic properties of electr ons when compared to holes.The holes mainly get trapped or may recombine bef ore they reach the back contact.

PAGE 57

49 A significant feature seen in the s pectral response of sample # 11-19-02 and many other samples too is the initial moderate ri se in the response at a wavelength of 400nm ,a gradual decrea se thereafter with the increase in wavelength,then a sudden increase aroun d the wavelength of about 540nm, observance of a peak around a wavelengt h of around 580nm and then a rapid fall beyond a wavelength of 595nm(since the absorpti on coefficient decreases). Attempts to explain the behaviour of the spectral response in the region from 400nm to 550nm and in the region from 550nm to 640nm was carried out by V.Rupavatharam[26],who propo sed a two region model to match the measured result with the simulated one.The peak response wavelength roughly corresponds to the minimum energy of a ph oton (2.1eV) required to excite an eh pair in HgI2. 5.3.2 Sample 12-02-02 This sample also had five devices with QE’s more than 0.2.The spectral Response and Iight I-V curves for this samp le are shown in Fig. 5.3.The spot # 7 (center spot) had the highest QE of 0.277 at a wavelength of 585nm.This result is a bit surprising because the film on the sa mple is expected to be thicker at the center and consequently the electric fi eld is expected to be weaker when compared to other spots,leading to lo wer carrier collection and hence a lower QE.Fig. 5.3(b) shows the photocurrent in the spot # 7 is about 2.184A at the voltage of -50V to the fr ont contact, obtained at the wavelength of 585nm.The photocurrent for this spot from the Spectral response file is 2.72A

PAGE 58

50 SPECTRAL RESPONSE FOR SAMPLE 12-02-02 0.00 0.05 0.10 0.15 0.20 0.25 0.30 400450500550600650700 Wavelength (nm)Q.E Spot # 1 Spot # 3 Spot # 4 Spot # 5 Spot # 7 (a) (b) Figure 5.3 Spectral Response and Li ght I-V Curves for Sample 12-02-02 LIGHT I-V CURVES FOR SAMPLE 12-02-02 0 0.5 1 1.5 2 2.5 -60-40-20 0204060 Voltage (V)Current (microA) Spot # 1 Spot # 3 Spot # 4 Spot # 5 Spot # 7

PAGE 59

51 and it is this photocurrent that corresponds to the peak QE .The Spectral response reveals that the spot # 3 and s pot # 7 had a almost zero response in the region from 400nm to 555nm.Such a beh avior is suggestive of the high surface recombination being experienc ed by the electrons created at the surface.In this sample too the dark cu rrents were very small and could not be recorded.This is actually a good prospect since the dark currents have to be as minimal as possible to achieve good results in these samples. 5.3.3 Sample 01-13-03 This sample had four spots on it whose QE was greater than 0.3.The QE’s of the other two spots were low as seen in the Fig .5.4 (a).Comparatively this sample was better than all other samples studied in this work.Also in this sample dark currents in nano Amps r ange were recorded and are shown in Fig. 5.5.One more feature of this sample is evident in the Spot # 3 of Fig. 5.4 (a).There is a high rise in the spectral response of this sample at the wavelength of 400nm.The photocurrent at this wavelength with a bias of -50V to the front illuminated contact was recorded as 0.2A from the spectral response data file which was greater than t he photocurrent from a spot on a sample which had a almost zero response at 400nm and the same polarity. The current differed by an order of 2.Such a high rise could not be matched even through simulation carried out by V.Rupavatharam [26].Therefore it is to be concluded at this stage

PAGE 60

52 (a) (b) Figure 5.4 Spectral Response and Light I-V Curves for Sample 01-13-03 SPECTRAL RESPONSE FOR SAMPLE 01-13-03 0 0.1 0.2 0.3 0.4 0.5 400 450 500 550 600 650 700 Wavelength (nm)Q.E Spot # 1 Spot # 2 Spot # 3 Spot # 4 Spot # 5 Spot # 6 LIGHT I-V CURVES FOR SAMPLE 01-13-03 0 0.5 1 1.5 2 2.5 3 -60 -40-20 0 20 40 60 Voltage (V)Current (microA) Spot # 1 Spot # 2 Spot # 3 Spot # 4 Spot # 5 Spot # 6

PAGE 61

53 Figure 5.5 Dark IV Curves for Sample 01-13-03 that such a rise is due to complex elec tronic processes which usually take place at the surface of such devices [22]. T he Spot # 3 had the highest efficiency of 0.39.The photocurrent at a bi as of -50V which corresponds to this QE is 3.82 A from the spectral response data file as opposed to the one obtained from the I-V curve of Fig. 5.4 (b) which is 2.8 A. As expected the dark currents with negative bias are lower than with positive bias show n in Fig.5.5.During the recording of these dark currents it was observed that they changed erratically and were sometimes difficult to tabulat e.The shape of these dark currents is indicative of the nature of contacts to each device.If the contacts were symmetric and ohmic ,the dark currents for both polarities would be linearly symmetric on both sides of DARK I-V CURVES FOR SAMPLE 01-13-030 20 40 60 80 100 120 -60-40-200 204060Voltage (V )Current (nA) Spot # 1 Spot # 2 Spot # 3 Spot # 4 Spot # 5 Spot # 6

PAGE 62

54 SPECTRAL RESPONSE FOR SAMPLE 10-20-03 0.00 0.05 0.10 0.15 0.20 0.25 0.30 400450500550600650700 Wavelength (nm)Q.E Spot # 1 Spot # 2 Spot # 3 Spot # 4 Spot # 5 Spot # 6 (a) (b) Figure 5.6 Spectral Response and Li ght I-V Curves for Sample 10-20-03 LIGHT I-V CURVES FOR SAMPLE 10-20-03 0 0.5 1 1.5 2 2.5-60-40-20 0 204060Voltage (V)Current (microA) Spot # 1 Spot # 2 Spot # 3 Spot # 4 Spot # 5 Spot # 6

PAGE 63

55 the current axis. But this is not t he case here. Samples 01-14-03 and 01-15-03 also showed recordable dark currents and their overall response was similar to the one obtained from sample 01-13-03. 5.3.4 Sample 10-20-03 The Spectral response for this samp le is shown in Fig.5.6 (a).The devices which showed response on this sa mple had QE’s rang ing from 0.17 to 0.25.Spot # 4 has the highest efficiency of 0.25, at a wavelength of 575nm and -50V bias to the front contact.The photoc urrent corresponding to this QE for this spot was recorded as 2.61 A from the spectral respons e file as opposed to the photocurrent recorded from I-V curv e of Fig.5.6(b) which is 2.22 A.Also seen from the spectral response for this sample in Fig.5.6(a) is the zero response of all spots except spot # 4 in the region from 400nm to 550nm.The currents in this region were lower by an order of 2 than those obtained for spot # 4.Also from Fig.5.6(b) the I-V curves fo r the positive bias reveals nearly zero hole currents for 4 spots.This suggests that the prev ention of holes from getting collected due to trapping and/or recombination in the bulk. Spots on Sample 10-21-03 and 0923-04 also showed nearly zero hole curr ents during illumination.This is shown in Fig 5.7.No dark currents could be recorded for these samples either. 5.3.5 Sample 03-25-04 The spectral response and I-V curves for this sample are shown in Fig 5.8. Spot # 6 on this sample had the highes t QE of 0.35 obtained at a wavelength

PAGE 64

56 LIGHT I-V CURVES FOR SAMPLE 10-21-03 0 0.5 1 1.5 2 2.5 -60-40-200204060 Voltage (V)Current (microA) Spot # 1 Spot # 2 Spot # 3 Spot # 4 Spot # 5 Spot # 6 (a) LIGHT I-V CURVES FOR SAMPLE 09-23-04 0.00 0.50 1.00 1.50 2.00 2.50 -60-40-200204060 Voltage (V)Current (microA) Spot # 1 Spot # 3 Spot # 4 Spot # 5 Spot # 6 Spot # 7 (b) Figure 5.7 Light I-V Curves fo r Samples 10-21-03 and 09-23-04

PAGE 65

57 of 565nm and a bias of -50V to the fr ont contact.The photocurrent corresponding to this QE was 3.36 A from spectral response data file as opposed to the value from the I-V curve of Fig .5.8(b) which is 2.74 A.One significant feature seen in the positive bias region of the I-V curves of this figure is the near saturation of hole currents.This suggests maximum collection has been achieved for the holes and no further increase in hole current is expected with increase in voltage. Sample 06-02-04 had one spot wit h a large circular palladium contact. Its QE was measured as 0.17 at a wavelength of 570nm and -50V to the front contact. The photocurrent from spectr al response data file was recorded as 1.95 A and that obtained from t he Light I-V curve was 1.24 A.No dark currents could be recorded in this sample too. Many of the results in this work c ould not be faithfully reproduced and the response of these samples thus depended on initial conditions of voltage,the time for which the sample was left unbiased.This behaviour has also been mentioned in reference [20].

PAGE 66

58 SPECTRAL RESPONSE FOR SAMPLE 03-25-04 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 400450500550600650700 Wavelength (nm)Q.E Spot # 2 Spot # 3 Spot # 4 Spot # 5 Spot # 6 Spot # 7 (a) LIGHT I-V CURVES OF SAMPLE 03-25-04 0 0.5 1 1.5 2 2.5 3 -60-40-200204060 Voltage (V)Current (microA) Spot # 2 Spot # 3 Spot # 4 Spot # 5 Spot # 6 Spot # 7 (b) Fig 5.8 Spectral Response and Light I-V Curves for Sample 03-25-04

PAGE 67

59 CHAPTER 6 CONCLUSIONS A number of polycrystalline mercuric i odide films were st udied as a part of this research.The fabrication conditions of these films which were made by Prince J.Simon [25] have been tabulated in this work.The QE’s for the devices on these films with a bias of -50 V to the front illuminated contact have also been presented in this work.The QE’s obtained ranged from 0.1 to 0.4.The spectral responses obtained were basically of thr ee types.The first type showed a zero response in the region from 400nm -550nm. This is probably due to high surface recombination for electrons at the surfac e which prevent them from traversing the device for negative poarity.The second type had a moderate rise in the spectral response at a wavelength of 400nm and then a gradual decrease till a wavelength of 540nm .This was explained on the basis of a two region model simulation carried out in another work on this subject [26].The third type had a high rise at the wavelength of 400nm and a gradual decrease therafter till the wavelength of 540nm.This could not be expl ained even through simulation and therefore is suggestive of complex electronic processes taking place at the surface.The samples showed a peak re sponse in between the wavelengths from 560nm-585nm which is also affected by ab sorption. I-V curves were obtained at peak spectral response wavelength in thes e samples Most of them had very

PAGE 68

60 low dark currents which could not be recorded. But dark currents could be recorded in three samples viz. 01-13-03,01-14-03 and 01-15-03.The maximum value of dark current value was 77nA wit h –ve bias to the front contact ,the maximum dark conductivity corresponding to this current was computed as 7.0E10( cm)-1. Light I-V curves for all sample s showed greater response in the negative bias region.This was as expect ed due to better electronic properties of electrons. But there was apparent small di screpancy in the photocurrents at peak response obtained from spectral response data file and that obtained from I-V curves.This is attributed to the time lag between the spectral response and I-V measurements.The performance of the different films is judged on the basis of consistent high QE’s and low dark current s obtained from all the devices on it.On this basis samples 01-13-03 and 03-2504 proved to be better than the other samples measured in this work.Four devices on sample 01-13-03 had QE’s greater than 0.3 This sample also had low measurable dark currents in the negative bias region.Sa mple 03-25-04 had all its devices giving a response.Three of the devices had QE ’s greater than 0. 3 and three more devices had QE’s close to 0.3.The dark currents were also small but not within recordable range.Also both these samp les had almost similar fabrication conditions which might have contri buted to their better performance. 6.1 Future Work The surface region of the HgI2 has been an area of concern in this work.The interpretation of the optical resp onse of this region is a challenging task

PAGE 69

61 which has to be carefully carried out by st udying the nature of surface states at the contact -HgI2 interface.The possibility of t he presence of excitons at the surface region has to be investigated thr ough simulation in order to bridge the gap between experimental and si mulated results for both the Spectral and I-V responses. Thermally Stimulated Current (TSC) measurements also have to be carried out on these films to find out the nature of deep traps in HgI2. Transient charge techniques such as the Ti me of Flight (TOF) can be used to evaluate the mobilities of both electrons and holes in these films. Finally the response of these films to X-ray radiat ion is to be analyzed and compared with the response to Light.

PAGE 70

62 REFERENCES [1] R.B.James, T.E.Schlesinger, “ Semiconductor and semimetals” Vol.43, pp.120 and pp.86-216, Academic Press, Inc., (1995). [2] Harrell G.Chotas, James T. Dobbins III, Carl E. Ravin, “Principles of Digital Radiography with LargeArea, Electronically Readable Detectors: A Review of the Basics” Radiology, Vol. 210, pp.595-599, (1999). [3] J.S. Iwanczyk et al., “HgI2 Polycrystalline Films fo r Digital X-Ray Imagers” IEEE Transactions on Nuclear Scienc e, Vol.49, No.1, February 2002. [4] G.Zentai et al., “Large Area Mercuric Iodide X-ray Imager” Proceedings of the SPIE, Vol.4682, Medica l Imaging, 592, (2002). [5] Pallab Bhattacharya, “Semiconductor Optoelectronic Devices” Prentice-Hall, Inc.New Jersey, (1994). [6] Robert F.Pierret, “Semiconductor Device Fundamentals” Addison-Wesley Publishing Company, Inc., (1996). [7] Richard S.Muller and Theodre I.Kamins, “Device Electronics for Integrated Circuits” Second Edition, John Wiley and Sons Inc., New York, 1986. [8] S.M.Sze, “Physics of Semiconductor Devices” 2nd Edition, John Wiley and Sons, Inc., New York, (2003). [9] Richard H.Bube, “Photoconductivity in Solids” John Wiley and Sons, Inc., New York, (1960). [10] Robert F.Pierret, “Advanced Semiconductor Fundamentals” Second Edition, Pearson Education Inc., New Jersey,(2003). [11] http://ece-classweb.uscd.edu:16080 /fall04/ece103/lecture5&6&7.pdf [12] Richard H.Bube, “Electrons in Solids”, Second Edition, Acad emic Press, Inc., (1988). [13] N.V.Joshi, “Photoconductivity” Marcel Decker, Inc.,(1990).

PAGE 71

63 [14] Richard H.Bube, “Opto-Electronic Properti es of Mercuric Iodide” Physical Review, Vol 106, No.4, pp. 703-717, May 15, (1957). [15] L.Van den berg et al., “Mercuric Iodide X-Ray and Gamma Ray Detectors for Astronomy ” Proceedings of the SPIE, Vol.4497, pp. 100-105, (2002). [16] M.Schieber et al., “Near single-crystal electrical properties of polycrystalline HgI2 produced by physical vapor deposition” Proceedings of the IEEE, (2003). [17] M.Schieber et al., “Theoretical and experimental sensitivity to X-rays of single and polycrystalline HgI2 compared with differ ent single-crystal detectors” Nuclear Instruments and Methods in Physics Research A,Vol.458,pp. 41-46,(2001). [18] A.Tajdine et al., “Search for Correlations between Electrical Characteristics and Stoichiometry in Mercuric Iodide” Nuclear Instruments and Methods, Vol.213, pp. 77-82, (1983). [19] T.E.Schlesinger et al., “Carrier Traps and Transport in Mercuric Iodide” Nuclear Instruments and Methods in Ph ysics Research A, Vol.322, pp.414, (1992). [20] J.P.Ponpon et al., “Current Instability in Mercuric Iodide Devices” Solid State Electronics, Vol.44, No.1, pp. 29-35, January, (2000). [21] A. Levi and M.M.Scheiber, “Carrier surface recombination in HgI2 photon detectors” Journal of Applied Physics, Vol.54, No.5, May, (1983). [22] Z.Burshtein et al., “Carrier surface generation an d recombination effects in photoconduction HgI2” Journal of Applied Physi cs, Vol.60, No.9, Nov.1, (1986). [23] D.E Turner and B.N.Harmon, “Electronic Structure of Red Mercuric Iodide” Physical Review B, Vol.40, No.15, Nov.15, (1989). [24] S.B.Hyder, “Trapping effects in silver-doped mercuric iodide crystals” Journal of Applied Physics, Vol.48, No1, January, (1977). [25] Prince J.Simon, “Polycrystalline Mercuric Iodi de Thin Films for Digital Radiation Detectors” Master’s Thesis, USF (2004). [26] Vikram Rupavatharam “Modelling of QE ,I-V charac teristics of MSM (Metal –Semiconductor-Metal )mercuric iodide thin films with MEDICITM ” Master’s Thesis, USF (2004).

PAGE 72

64 APPENDICES

PAGE 73

Appendix A Spectral Response Data File The spectral response curve for each device (spot) on a sample was plotted from the QE data obtained from a file similar to the one shown in Table A.1 on page 66.A part of the data file is shown ti ll the wavelength of 595nm. The second column of the table shows the photocurrent response fr om the device (spot) on a sample for each wavelength.The QE of the device was found out from the photoresponse from a Si photodiode which was subjec ted to the same photon flux to which the device was subjected.T he third and fourth column of the table therefore shows photodiode photocurrent and its QE respectively. The fifth column shows the QE of the device whic h can be found out from the relation as follows. Since the detection of each e-h pair(both the carriers reaching their respective electrodes) contributes to flow one electron worth of current in the external circuit and since the photon flux (num ber of photons/sec) on the device is the same as that on the photodiode for each wavelength, QE of the device = Number of e-h pairs created and detected in the device per second Number of photon s impinging on the device per second = Number of electrons of current in ex ternal circuit of the device per second Number of photons impinging on the device per = (QE of photodiode) X photoc urrent from Device Photocurrent from photodiode 65

PAGE 74

Appendix A (Continued) Table A.1 Spectral Response Data File Wavelength (nm) Device Response ( A) Si Photodiode Response (A) Si Photodiode QE QE of Device 400.01199 6.20E-08 3.95E-07 0.5517 8.66E-02 404.98499 6.69E-08 4.93E-07 0.5556 7.54E-02 409.98999 7.30E-08 6.01E-07 0.5625 6.82E-02 414.992 7.73E-08 7.21E-07 0.569 6.10E-02 419.98901 9.24E-08 8.40E-07 0.5781 6.36E-02 425.01801 1.00E-07 9.70E-07 0.5803 5.98E-02 430.00699 1.03E-07 1.10E-06 0.585 5.49E-02 434.991 1.01E-07 1.26E-06 0.5866 4.74E-02 440.008 1.12E-07 1.41E-06 0.588 4.69E-02 444.983 1.05E-07 1.57E-06 0.5913 3.94E-02 449.991 1.13E-07 1.74E-06 0.5943 3.85E-02 454.99399 1.04E-07 1.92E-06 0.5974 3.22E-02 459.992 1.01E-07 2.11E-06 0.6008 2.86E-02 464.98599 1.12E-07 2.30E-06 0.603 2.92E-02 470.01099 1.18E-07 2.51E-06 0.6053 2.85E-02 474.995 1.21E-07 2.71E-06 0.6075 2.71E-02 480.01099 1.24E-07 2.93E-06 0.6097 2.57E-02 484.98599 1.23E-07 3.15E-06 0.612 2.39E-02 489.991 1.27E-07 3.37E-06 0.6143 2.31E-02 494.992 1.21E-07 3.61E-06 0.6167 2.06E-02 499.98801 1.24E-07 3.87E-06 0.6188 1.98E-02 505.01401 1.29E-07 4.13E-06 0.62 1.94E-02 509.99899 1.31E-07 4.40E-06 0.6183 1.84E-02 515.01502 1.34E-07 4.63E-06 0.6196 1.79E-02 519.98999 1.44E-07 4.90E-06 0.6234 1.83E-02 524.995 1.53E-07 5.13E-06 0.6288 1.87E-02 529.995 1.55E-07 5.38E-06 0.6216 1.79E-02 534.98901 1.74E-07 5.58E-06 0.6267 1.95E-02 540.013 2.55E-07 5.79E-06 0.6263 2.76E-02 544.99597 4.26E-07 6.01E-06 0.6269 4.44E-02 550.00897 8.89E-07 6.28E-06 0.6282 8.89E-02 555.01703 1.88E-06 6.55E-06 0.6277 1.80E-01 560.01801 2.87E-06 6.82E-06 0.6384 2.69E-01 565.013 3.25E-06 7.11E-06 0.6315 2.88E-01 570.00201 3.23E-06 7.41E-06 0.6319 2.76E-01 574.98602 2.92E-06 7.66E-06 0.6335 2.41E-01 579.99799 2.43E-06 7.89E-06 0.6327 1.94E-01 585.00501 1.90E-06 8.07E-06 0.6327 1.49E-01 590.00501 1.38E-06 8.22E-06 0.6339 1.06E-01 594.9902 8.80E-07 8.48E-06 0.6367 6.61E-02 66