Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mobile ion contamination of MOS structures investigated by computer-controlled laser scanning internal… Tsoi, Hak-Yam 1980

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

Item Metadata

Download

Media
831-UBC_1980_A1 T86.pdf [ 7.1MB ]
Metadata
JSON: 831-1.0095217.json
JSON-LD: 831-1.0095217-ld.json
RDF/XML (Pretty): 831-1.0095217-rdf.xml
RDF/JSON: 831-1.0095217-rdf.json
Turtle: 831-1.0095217-turtle.txt
N-Triples: 831-1.0095217-rdf-ntriples.txt
Original Record: 831-1.0095217-source.json
Full Text
831-1.0095217-fulltext.txt
Citation
831-1.0095217.ris

Full Text

MOBILE ION CONTAMINATION OF MOS STRUCTURES INVESTIGATED BY COMPUTER-CONTROLLED LASER SCANNING INTERNAL PHOTOEMISSION AND SELF-HEALING BREAKDOWN TECHNIQUES by HAK-YAM jTSOI B . S c , The Chinese Un i v e r s i t y of Hong Kong, 1969 M.Sc, The Chinese Un i v e r s i t y of Hong Kong, 1974 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of E l e c t r i c a l Engineering We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA JANUARY, 1980 (c) Hak-Yam Ts o i In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f< an advanced degree at the U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e tha the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department o f The U n i v e r s i t y o f B r i t i s h Co lumbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date U f l J t c L , L f & P > < $ i ABSTRACT With the present trend of s i l i c o n IC's towards very large scale i n t e g r a t i o n , premature device breakdown i s inc r e a s i n g l y important. Contam-i n a t i o n of MOS devices by mobile sodium ions i s of p a r t i c u l a r i n t e r e s t as a source of device f a i l u r e . The e f f e c t s of contamination of MOS structures by mobile sodium ions were investigated by using a combination of two techniques: computer-co n t r o l l e d laser scanning i n t e r n a l photoemission and s e l f - h e a l i n g breakdown measurements. The i n t e r n a l photoemission measurements involved focussing UV r a d i a t i o n from a He-Cd la s e r on to a spot a few microns i n diameter. The spot was then scanned across the sample surface and a map of the i n t e r n a l photoemission current was obtained. The scanning i n t e r n a l photoemission measurements were made using a PDP-8E based minicomputer-controlled set up. A s p e c i a l technique for focussing the UV l a s e r beam and for measuring the l i g h t spot s i z e with the sample i n s i t u was developed. R e f l e c t i v i t y imaging was used to p o s i t i o n the sample and to i d e n t i f y c e r t a i n types of defects on the sample surface. The s e l f - h e a l i n g breakdown tests were made using a s p e c i a l l y b u i l t e l e c t r o n i c device. This allowed the operator to breakdown the sample a preset number of times i n order to avoid the occurrence of both weak spot and i n t r i n s i c breakdowns on the same sample. The photocurrent i s low i n uncontaminated samples and also i n contaminated samples before a p p l i c a t i o n of p o s i t i v e gate voltage. However, i f a p o s i t i v e bias-temperature stress treatment i s applied to the contaminated samples, very pronounced photocurrent peaks were observed. The posi t i o n s i i on the surface at which breakdowns occurred were mostly found to coincide with strong peaks i n the i n t e r n a l photoemission current. The i n t e r n a l photoemission process i n MOS structures was modelled. The dependence of the quantum y i e l d on the photon energy, on the semiconductor doping, on the semiconductor surface band bending and on the applied f i e l d were calculated. The model predicted that a large downward band bending at the surface of a heavily doped p-type semiconductor would s h i f t the photoemis-sion threshold to a lower value. The photoemission current imaging and the r e f l e c t i v i t y imaging of MOS structures and defects s i t e s which are presented show that the two techniques complement each other i n revealing defect s i t e s . The transfer of Na + from aqueous solutions to the oxide before and a f t e r aluminum gate deposition was examined. I t was found that the Na + which was transferred a f t e r the aluminum gate deposition always d i s t r i b u t e d i n a non-uniform fashion. For Si02 films prepared i n the presence of a small amount of HC£, mobile ion n e u t r a l i z a t i o n at the Si-SiO^ i n t e r f a c e was observed. For samples which were exposed to room ambient for a prolonged period, photocurrent peaks were observed along the periphery of the aluminum gate a f t e r the a p p l i c a t i o n of a p o s i t i v e bias-temperature stress treatment. This indicated that mobile ions could d i f f u s e into the sample along the periphery of the aluminum gate. Nucleation of breakdowns was again observed corresponding to the photocurrent peaks. This indicates that any storage of MOS devices before encapsulation i s best done i n an environment which does not cause sodium ion contamination, since the presence of the gates does not provide t o t a l protection. i i i TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i v LIST OF ILLUSTRATIONS v i ACKNOWLEDGEMENT x i 1. INTRODUCTION 1 2. REVIEW OF PREVIOUS WORK 3 2.1 The E f f e c t s of A l k a l i Metal Ions on MOS Structures . . . . 3 2.2 Internal Photoemission i n MOS Structures 6 2.3 D i e l e c t r i c Breakdown i n MOS Structures 9 3. THEORY OF INTERNAL PHOTOEMISSION IN MOS STRUCTURES 13 3.1 Interaction of Radiation with the MOS Structure 13 3.2 Formulation of the Problem 16 3.3 Remarks 33 4. DESIGN OF THE COMPUTER CONTROLLED LASER SCANNER 37 4.1 P r i n c i p l e of Operation 37 4.2 Optics of the Laser Scanner 39 4.2.1 C h a r a c t e r i s t i c s of the Focussed Light Spot 39 4.2.2 Focussing Optics 44 4.2.3 Measurement of the Light Spot Size 49 4.3 Hardware of the Laser Scanner 51 4.4 Software and Operation of the Laser Scanner 56 5. DESIGN OF THE SELF-HEALING BREAKDOWN TESTER 59 6. SAMPLE DESIGN AND FABRICATION 65 6.1 Design Considerations 65 6.2 Sample Fab r i c a t i o n 67 6.2.1 Reinforcement Oxide Preparation 67 6.2.2 Gate Oxide Preparation 70 6.2.3 M e t a l l i z a t i o n 71 i v Page 7. EXPERIMENTAL RESULTS 74 7.1 Imaging with the Laser Scanner 74 7.2 The E f f e c t s of Na + on Scanning Internal Photoemission . . . 86 7.2.1 Na + Contamination a f t e r Aluminum Gate Deposition . . 86 7.2.2 Na Contamination before Aluminum Gate Deposition . . 98 7.3 R e v e r s i b i l i t y of Photocurrent Peaks under Bias-Temperature Stress Treatment 105 7.4 C o r r e l a t i o n between Mobile Ion Induced Photocurrent Peaks and Self-Healing Breakdown i n MOS Structures 107 7.5 Scanning Internal Photoemission and Self-Healing Breakdown Studies of HC£-Grown S i l i c o n Dioxide Films 116 8. CONCLUSIONS 125 8.1 Suggestions for Further Research 127 REFERENCES 128 APPENDIX 1 THE ELECTROSTATIC FIELD AND THE BAND BENDING AT THE SEMICONDUCTOR SURFACE 130 APPENDIX 2 SOFTWARE OF THE LASER SCANNER 134 APPENDIX 3 CIRCUIT DIAGRAMS OF THE SELF-HEALING BREAKDOWN TESTER . . 156 APPENDIX 4 PRE-FURNACE CLEANING OF SILICON WAFERS 163 v LIST OF ILLUSTRATIONS Page F i g . 2.1 T o t a l mobile ion density at the Si-Si02 i n t e r f a c e as a function of the bias-temperature stress time and temperature 4 F i g . 2.2(a) Current vs. time curve for contaminated sample under constant bias 11 F i g . 2.2(b) Calculated v a r i a t i o n of tun n e l l i n g distance with the p o s i t i o n of a layer of p o s i t i v e charge i n Si02 . . . 11 F i g . 3.1(a) The three-step photoemission process 17 F i g . 3.1(b) Energy band diagram of MOS structure under p o s i t i v e bias 17 F i g . 3.2 O p t i c a l absorption c o e f f i c i e n t of S i (a) vs. wave-length ( A ) 18 F i g . 3.3 Schottky lowering of b a r r i e r height and b a r r i e r p o s i t i o n measured from the Si-Si02 i n t e r f a c e as a function of applied f i e l d 20 F i g . 3.4 S i l i c o n surface p o t e n t i a l vs. oxide f i e l d at the Si-SK> 2 i n t e r f a c e 22 F i g . 3.5 E f f e c t i v e Debye length vs. doping i n s i l i c o n . . . . 24 F i g . 3.6 Calculated d i f f e r e n t i a l y i e l d (dY/dX) versus d i s -tance from the Si-SiO^ i n t e r f a c e (X) with the escape depth (L^) as the parameter 28 F i g . 3.7 Calculated d i f f e r e n t i a l y i e l d (dY/dX) versus d i s -tance from the S i - S K ^ i n t e r f a c e (X) with incident photon energy (hv) as the parameter 29 F i g . 3.8 Calculated d i f f e r e n t i a l y i e l d (dY'/dX) versus d i s -tance from the Si-SiO^ i n t e r f a c e (X) with surface p o t e n t i a l (V ) as the parameter 31 F i g . 3.9 Calculated d i f f e r e n t i a l y i e l d (dY'/dX) versus d i s -tance from the Si-Si02 i n t e r f a c e (X) with s i l i c o n doping (P^) as the parameter 32 F i g . 3.10 E f f e c t of s i l i c o n doping and surface p o t e n t i a l on the quantum y i e l d versus photon energy curve . . . . 34 F i g . 3.11 Calculated quantum y i e l d (Y) versus applied f i e l d (E) 35 v i Page F i g . 4.1(a) P r i n c i p l e of operation of the scanner 38 F i g . 4.1(b) Scan pattern 38 F i g . 4.2 Intensity d i s t r i b u t i o n of the l i g h t spot as a function of the incoming beam width 40 F i g . '4.3 To t a l power and the s i z e of the focussed l i g h t spot versus incoming beam width 43 F i g . 4.4 Focussing optics of the l a s e r scanner 46 F i g . 4.5 Image of the l i g h t spot produced by the focussing lens 1 and lens 2 47 F i g . 4.6(a) Light spot si z e and R2 as a function of lens 1 displacement from the focus 48 Fig . 4.6(b) Phase s e n s i t i v e detector output as a function of the p o s i t i o n of the focussing lens 1 48 F i g . 4.7(a) Scan a l i g h t spot over two regions of d i f f e r e n t r e f l e c t a n c e 50 F i g . 4.7(b) Edge p r o f i l e as calculated by equation 4.12 50 F i g . 4.8 Set up of the la s e r scanner 52 F i g . 4.9(a) Measurement of the photocurrent 53 F i g . 4.9(b) Temperature cycle recorded by a s t r i p chart recorder. 53 F i g . 4.10 Sample l o c a t i o n by using r e f l e c t i v i t y scan 54 F i g . 4.11 Flow chart of the operation of the scanner 57 F i g . 5.1 Measurement of s e l f - h e a l i n g breakdown i n MOS structures 60 F i g . 5.2 Block diagram of the automatic s e l f - h e a l i n g breakdown tester 62 F i g . 6.1 Transmittance of aluminum vs. f i l m thickness . . . . 66 F i g . 6.2 MOS capacitor with two layer m e t a l l i z a t i o n 66 F i g . 6.3 Pattern of a MOS capacitor with a duplex structure . 68 F i g . 6.4 Sample f a b r i c a t i o n 69 F i g . 6.5 Photograph of a f i n i s h e d sample 73 v i i Page F i g . 7.1 R e f l e c t i v i t y map of Sample 001-4311 75 F i g . 7.2(a) Photograph of Sample 001-4311 76 F i g . 7.2(b) Photograph of Sample 005-3332 76 F i g . 7.3 R e f l e c t i v i t y map of Sample 005-3332 77 F i g . 7.4 R e f l e c t i v i t y map of the larger scratch on Sample 005-3332 79 F i g . 7.5 Photocurrent map of the larger scratch on Sample 005-3332 80 F i g . 7.6 Magnified p i c t u r e of the s t a i n on Sample 005-3332 . . . 81 F i g . 7.7 R e f l e c t i v i t y map of the s t a i n on Sample 005-3332 . . . 82 F i g . 7.8 Photocurrent map of the s t a i n on Sample 005-3332 . . . 83 F i g . 7.9 Photocurrent map of Sample 005-3332 84 F i g . 7.10 Photocurrent map of Sample 005-3332 with the background noise removed by the computer software 85 F i g . 7.11 The e f f e c t of negative bias-temperature stress t r e a t -ment on the photocurrent map 88 F i g . 7.12 The e f f e c t of p o s i t i v e bias-temperature stress t r e a t -ment on the photocurrent map 89 F i g . 7.13(a) Photocurrent map of a p o r t i o n of Sample 005-3311 which was contaminated by 10 N NaCA so l u t i o n 90 F i g . 7.13(b) Photocurrent map of a p o r t i o n of Sample 005-3311 which was contaminated by 10 N NaC& so l u t i o n 90 F i g . 7.14 Photocurrent map of Sample 005-3311 which was contamin-ated by 10~ N NaC£ sol u t i o n 92 F i g . 7.15 Shadow map of Sample 005-3311 93 F i g . 7.16 Photocurrent map of Sample 005-3311 measured with a bias voltage of +19.2V across the sample 95 F i g . 7.17 Photocurrent map of Sample 005-3311 measured with a bias voltage of +28.9V across the sample 96 F i g . 7.18 Photocurrent map of Sample 005-3311 measured with a bias voltage of +38.4V across the sample 97 v i i i Page F i g . 7.19 Photocurrent map of Sample 001-2331 obtained before a p o s i t i v e bias-temperature stress treatment was applied to the sample 99 F i g . 7.20 Photocurrent map of Sample 001-2331 obtained a f t e r s t r e s s i n g the sample at 118°C, +9.0 V for 30 min . . . 100 Fig. 7.21 Photocurrent map of Sample 001-2331 obtained a f t e r s t r e s s i n g the sample at 118°C, +9.0 V for 1 hr . . . . 101 Fi g . 7.22 Photocurrent map of Sample 001-2331 obtained a f t e r s t r e s s i n g the sample at 118 C, +9.0 V for 3 hr 40 min . 102 Fi g . 7.23 Photocurrent map of Sample 001-2331 obtained a f t e r s t r e s s i n g the sample at 118°C, +9-0 V for 15 hr 10 min. 103 Fi g . 7.24 Photocurrent map of Sample 001-2331 obtained a f t e r s t r e s s i n g the sample at 118°C, +9:0'V for 27 hr 10 min. 104 Fi g . 7.25 R e v e r s i b i l i t y of photocurrent peaks studied under b i a s -temperature (B-T) stress treatment 106 Fi g . 7.26 Photocurrent map of Sample 004-1218 obtained before Na d r i f t 108 F i g . 7.27 R e f l e c t i v i t y map of Sample 004-1218 obtained before Na d r i f t 109 Fi g . 7.28 Photocurrent map of Sample 004-1218 obtained a f t e r s t r e s s i n g the sample at 118 C, +19.2V for 1 hr . . . . I l l F i g . 7.29 R e f l e c t i v i t y map of Sample 004-1218 obtained a f t e r s t r e s s i n g the sample at 118°C, +19.2V for 1 hr . . . . 112 Fi g . 7.30 Photograph of Sample 004-1218 obtained a f t e r s e l f -healing breakdown test 113 Fi g . 7.31 R e f l e c t i v i t y map of Sample 004-1218 obtained a f t e r s e l f - h e a l i n g breakdown te s t 114 Fi g . 7.32(a) Photocurrent map of Sample 004-1218 a f t e r Na + d r i f t : (Fig. 7.28) i s presented as a shadow map 115 Fi g . 7.32(b) R e f l e c t i v i t y map of Sample 004-1218 a f t e r breakdown (Fig. 7.31) i s presented as a shadow map 115 Fi g . 7.33 Photocurrent map of Sample 004-1218 obtained a f t e r s e l f - h e a l i n g breakdown 117 i x Page F i g . 7.34 119 F i g . 7.35 Photocurrent mag of Sample 007-4116 obtained 4.25 X 10 cm mobile ions were d r i f t e d to af t e r the 120 F i g . 7.36 Photocurrent man of Sample 007-4116 obtained 1.33 X 10 cm mobile ions were d r i f t e d to a f t e r the 121 F i g . 7.37 Photograph of Sample 007-4116 obtained a f t e r a s e l f - 122 F i g . 7.38 R e f l e c t i v i t y map of Sample 007-4116 obtained a f t e r 123 F i g . A l Energy band diagram of a MIS structure . . 131 F i g . A3.1 157 F i g . A3. 2 Preamplifier, threshold detector and pulse shaper 158 F i g . A3.3 159 F i g . A3.4 160 F i g . A3.5 Overvoltage cowbar and ramp generator c i r c u i t . . . . 161 F i g . A3.6 . 162 ACKNOWLEDGEMENT I would l i k e to thank my supervisor, Dr. L. Young for h i s patient help and guidance during the course of t h i s research. I wish to express my appreciation to Mr. J . Stuber for h i s assistance i n making the scanning stage of the l a s e r scanner, to Mr. A. Yan for many h e l p f u l discussions during the preparation of the numerical compu-t a t i o n programs and for proof-reading the manuscript and to Mrs. K, Brindamour for typing a portion of the t h e s i s . The U n i v e r s i t y of B r i t i s h Columbia i s g r a t e f u l l y acknowledged for the fellowship which supported t h i s work. x i To Susan x i i 1 Chapter 1 INTRODUCTION Because of t h e i r importance i n integrated c i r c u i t s , MOS devices have been studied extensively during the l a s t two decades. The MOSFET approach i s one of the most promising i n current work to achieve VLSI. The drive to increase c i r c u i t complexity has imposed increasingly high require-ments on f a b r i c a t i o n technology to achieve both acceptably high product y i e l d and adequate device r e l i a b i l i t y . Defects and contamination introduced during processing steps are important i n reducing product y i e l d and c i r c u i t r e l i a b i l i t y . They include: (i) Defects introduced by the photolithography process. For example pinholes, scratches or p a r t i c l e s on the photomask; incomplete removal of photoresist on the s i l i c o n wafer; pinholes on the photoresist. ( i i ) Dust p a r t i c l e s on the wafer which may r e s u l t i n pinholes or thin spots i n the d i e l e c t r i c f i l m , ( i i i ) Defects i n m e t a l l i z a t i o n . For example, short c i r c u i t s r e s u l t i n g from metal bridging and open c i r c u i t s r e s u l t i n g from discontinu-i t i e s i n the metal f i l m at the base of steep steps. (iv) Contamination of the gate oxide by mobile ions which may r e s u l t i n lower breakdown voltages and i n s t a b i l i t y due to i o n i c motion. This thesis i s devoted to the study of item ( i v ) , that i s the e f f e c t s of i o n i c contamination. A combination of computer-controlled l a s e r scanning i n t e r n a l photoemission and s e l f - h e a l i n g breakdown measurements were used. 2 In Chapter 2, a review of the previous work i s given. In Chapter 3, the i n t e r n a l photoemission process i s modelled. The photoemission of electrons from S i into SiO^ i s considered as a three step process,namely, the e x c i t a t i o n of electrons i n the s i l i c o n , the transport of photo-excited electrons to the energy b a r r i e r , and f i n a l l y the escape of the photo-excited electrons over the b a r r i e r . The e f f e c t s of s i l i c o n band bending, of impact i o n i z a t i o n i n s i l i c o n , of s c a t t e r i n g i n S i C ^ a n d of Schottky lowering of the energy b a r r i e r on photoemission are calc u l a t e d . In Chapter 4, the design of a computer c o n t r o l l e d laser scanner i s described. In Chapter 5, the design of an automatic s e l f - h e a l i n g breakdown tester i s discussed. In Chapter 6, the design of a s p e c i a l MOS sample s u i t a b l e f o r both scanning i n t e r n a l photoemission and s e l f - h e a l i n g breakdown measurements i s discussed. The experimental r e s u l t s are summarized i n Chapter 7. These include the imaging of MOS samples with the l a s e r scanner; the study of sodium trans-f e r process i n MOS structures; the c o r r e l a t i o n between sodium contamination and breakdown and the study of mobile ion n e u t r a l i z a t i o n i n HC£-grown Si02 .films. Chapter 8 contains conclusions and suggestions for further work. 3 Chapter 2 REVIEW OF PREVIOUS WORK In t h i s chapter, a b r i e f review i s given of previous work on mobile ion migration i n thermal Si02 f i l m s , on i n t e r n a l photoemission i n MOS structures and on breakdown i n MOS structures. 2.1 The e f f e c t s of a l k a l i metal ions on MOS structures The ion transport phenomenon i n thermally grown Si02 films was f i r s t studied by Snow et a l . (1965). MOS samples were d e l i b e r a t e l y contaminated with e i t h e r sodium or l i t h i u m by r i n s i n g them i n a d i l u t e s o l u t i o n of NaC£ or LiC£'just p r i o r to m e t a l l i z a t i o n . A f t e r the a p p l i c a t i o n of a prolonged p o s i t i v e bias to the metal electrode at an elevated temperature, the f l a t -band voltages of the contaminated samples were observed to s h i f t to more negative values. The flat-band voltages were determined from room temperature capacitance-voltage (C-V) c h a r a c t e r i s t i c s measured before and a f t e r the str e s s . The s h i f t i n flat-band voltage was interpreted as a result'.of the migration of sodium or l i t h i u m ions from the metal-oxide i n t e r f a c e to the s i l i c o n - o x i d e i n t e r f a c e under the p o s i t i v e bias-temperature stress treatment. Snow et a l . studied the accumulation of ions at the oxide-semiconductor i n t e r -face as a function of time, temperature and applied voltage. They found that at a given applied voltage the number of mobile ions per unit area at the oxide-semiconductor i n t e r f a c e increased l i n e a r l y with the square root of time during the i n i t i a l period of d r i f t as shown i n F i g . 2.1 and then approached a saturation value. Saturation was approached more r a p i d l y at higher temperatures but the f i n a l value of N. remained the same for a l l * This known as p o s i t i v e bias-temperature stress treatment. Fig.2.1 Tot a l mobile ion density at the Si-SiC> 2 i n t e r f a c e as a function of the bias-temperature stress time and temperature. (After Snow et al.,1965) 5 temperatures. The rate of increase i n ion density N. at the oxide-ion semiconductor i n t e r f a c e was also observed to increase with the applied f i e l d . At room temperature and at lower applied f i e l d s , the migration of a l k a l i metal ions was slow. I t resu l t e d i n a slow d r i f t or i n s t a b i l i t y of MOS device c h a r a c t e r i s t i c s . Besides i n s t a b i l i t y , c o n t a m i n a t i o n by the p o s i t i v e l y charged ions was also reported to lower the breakdown f i e l d of MOS structures. A review of the e f f e c t s of mobile ions on MOS breakdown i s given i n Section 2.3. Both i n s t a b i l i t y and premature breakdown can s e r i o u s l y degrade the r e l i a b i l i t y of MOS devices. There are three established methods for measuring mobile ion den-s i t y i n MOS structures, namely the capacitance-voltage (C-V) method (Snow et al.,1965), the integr a t i n g electrometer (Q-t) method (Snow et al.,1965), and the t r i a n g u l a r voltage sweep (TVS) method (Chou, 1971) . The C-V method measures the t o t a l ions which are e l e c t r i c a l l y a c t i v e at the Si-Si02 i n t e r -face. The Q-t and TVS measure the t o t a l ions d r i f t e d through the Si02« How-ever, none of these methods can measure the d i s t r i b u t i o n of mobile ions on the surface of MOS structures. For integrated c i r c u i t s , a s i n g l e element f a i l u r e may lead to malfunction of the ent i r e c i r c u i t . Detection of l o c a l i z e d contamination by mobile ions i s important. Thus an average mobile ion den-9 2 s i t y of 10 ion/cm i s usually regarded as a low contamination l e v e l . If 2 these mobile ions are located e n t i r e l y i n an area of say 10 x 10 um i n a 2 13 2 sample of 1 x 1 mm , the l o c a l density of ion w i l l be 10 ions/cm , which i s an unacceptably high value. A review of the scanning i n t e r n a l photo-emission method which can be used to detect small patches of ions at the oxide-semiconductor i n t e r f a c e i n MOS structures w i l l be given i n Section 2.2. There are two methods for reducing the e f f e c t of mobile ions on MOS c h a r a c t e r i s t i c s . The f i r s t i s to stop the migration of mobile ions from 6 the metal-SiC^ i n t e r f a c e to the Si-SiG^ i n t e r f a c e by f a b r i c a t i n g a layer of phosphosilicate glass on SiO^ (Eldridge and Kerr, 1971). The second i s to n e u t r a l i z e the mobile ions at the S i - S i C ^ i n t e r f a c e a f t e r d r i f t by incorpor-ating Cl i n the SiO^ (Kriegler,1972). The p o l a r i z a t i o n of phosphosilicate glass makes the f i r s t method undesirable for device a p p l i c a t i o n . The second method i s now commonly used to passivate MOS devices against i o n i c contamination. 2.2 Internal photoemission i n MOS structures Photoemission of electrons from s i l i c o n to s i l i c o n dioxide i n MOS structures was f i r s t studied by Williams (1965). On i l l u m i n a t i n g a MOS capaci-tor with r a d i a t i o n of wavelength shorter than 2900A", a photocurrent through the S102 layer was observed. The photocurrent was bias p o l a r i t y dependent which suggested i t was not due to the photoconductivity i n the oxide. For p o s i t i v e gate bias the photocurrent was found to be independent of the gate material. This suggested that electron emission from S i to Si02 was the dominating process. The photoemission threshold, which corresponded to the b a r r i e r height measured from the top of the S i valence band to the bottom of the Si02 conduction band, was found to be 4.25 eV. The b a r r i e r energies of the metal-Si02 b a r r i e r and the S i - S i O ? bar-r i e r were studied by Snow, Deal and Mead (1966) using photoemission and C-V measurement techniques. The two techniques gave r e s u l t s i n good agreement with each other. Goodman (1966) studied the v a r i a t i o n of the photoemission threshold with applied e l e c t r i c f i e l d . The r e s u l t was found to be consistent with b a r r i e r height lowering by the Schottky e f f e c t . The b a r r i e r height at the Si-Si02 i n t e r f a c e f o r zero f i e l d was found to be 4.2 eV. The contributions of electrons emitted by the cathode and of holes emitted by the anode to the photocurrent were studied by Powell (1969). 7 Au-SiO^-Si devices were made with thicknesses of oxide such that o p t i c a l interference made the l i g h t i n t e n s i t y a maximum at ei t h e r the Au-SiC^ or Si-SiO^ i n t e r f a c e . Photoemission of c a r r i e r s from Au into SiO^ and from S i into SiO^ could be studied separately. Powell observed that for a given bias d i r e c t i o n , hole photocurrent was small when compared with electron photocurrent from the opposite electrode. In p a r t i c u l a r , the quantum y i e l d for photoemission of holes from Au into SiC^ was at le a s t two orders-of-magnitude lower than that of electron emission from S i into S i O ^ The e f f e c t s of ion migration on photoemission were studied by Williams (1966). A f t e r the ap p l i c a t i o n of a p o s i t i v e bias-temperature stress treatment to the sample, a new low-energy branch i n the photoemission spec-trum with a threshold of 3.05eV was observed. Two explanations were suggest-ed. The f i r s t was due to the formation of a degenerate n-type surface. Con-t r i b u t i o n of the photoelectrons emitted from the conduction band resulted i n a lower threshold. The second, a l t e r n a t i v e explanation involved the lowering of the b a r r i e r height by the Schottky e f f e c t . A lowering of the photoemission threshold a f t e r the ap p l i c a t i o n of a p o s i t i v e bias-temperature stress treatment was also observed by Viswanathan and Ogura (1968/9), but these authors favoured a d i f f e r e n t explanation. The ef f e c t was inter p r e t e d as a r e s u l t of the large band bending at the semi-conductor surface which raised the Eermi l e v e l and reduced the photoemission threshold. Scanning i n t e r n a l photoemission measurements were f i r s t made by DiStefano (1971). A monochromatic l i g h t beam of 5 - 50 um diameter was used to scan samples contaminated by sodium. Maps of photocurrent were r e -corded. I t was found that a f t e r p o s i t i v e bias-temperature stress treatment, 8 the photocurrent was not uniform. Very pronounced photocurrent peaks were observed. At the positi o n s of the photocurrent peak, lowering of photoemis-sion threshold were observed. The lowering i n photoemission threshold was believed to be caused by the sodium dipole la y e r , s i m i l a r to the e f f e c t of cesium on photoemitters. A He-Cd l a s e r of wavelength 3250$ (3.81eV) was f i r s t employed by Williams and Woods (1972) f o r scanning i n t e r n a l photoemission studies. The photon energy of the las e r i s not s u f f i c i e n t to excite electrons from the s i l i c o n valence band into the SiO^ conduction band. Aft e r the a p p l i c a t i o n of a p o s i t i v e bias-temperature stress treatment to the MOS samples, very pro-nounced photocurrent peaks were observed. This was explained as a r e s u l t of the l o c a l i z e d b a r r i e r height lowering which made photoemission from the va-lence band poss i b l e . The positi o n s of the photocurrent peak were observed to coincide with the spots at which s e l f - h e a l i n g breakdown occurred. DiStefano (1973) also observed a correspondence between the positio n s of photocurrent peak and the points at which d i e l e c t r i c breakdown occurred. A r e l a t i o n s h i p between b a r r i e r height reduction and breakdown was thought to e x i s t . The d i s t r i b u t i o n of sodium ions i n the Si-Si02 structure was studied t h e o r e t i c a l l y by Williams and Woods (1973) and Williams (1974). It was argued that a uniform d i s t r i b u t i o n of ions was inherently unstable and tended to condense into patches of high concentration. DiStefano (1974) however, observed that the photocurrent peaks were d i s t r i b u t e d i n large and i r r e g u l a r spacing. The l o c a l i z e d high density of sodium was thought to be associated with defects i n the sample rather than an e f f e c t of sodium ion segregation. 9 2.3 D i e l e c t r i c breakdown In MOS structures The c l a s s i f i c a t i o n of breakdown i n s i l i c o n dioxide i n t o weak spot and i n t r i n s i c breakdown was f i r s t made by K l e i n et a l . ( K l e i n & Gafni, .1966; K l e i n , 1966) and Frit z s c h e ( F r i t z s c h e , 1967). K l e i n and h i s co-workers applied the non-shorting or s e l f - h e a l i n g technique to investigate breakdown i n MOS structures. This technique o u t i l i z e s a thi n metal electrode (<1000A). A slowly increasing voltage i s applied to the sample u n t i l breakdown occurs. Localized breakdown of the oxide provides a. high conduction filament. The energy stored i n the capaci-tor discharges through t h i s conducting filament and results i n high l o c a l joule heating. The heat evaporates the oxide and the t h i n metal electrode surrounding the filament. The breakdown spot i s then disconnected from the rest of the metal electrode so that the MOS capacitor i s s t i l l operational. The breakdown process can be observed as a spark discharge and the formation of a macroscopic hole. K l e i n observed that a succession of breakdown events could be i n i t i a t e d at low f i e l d s (<2 MV-cm "*") . The breakdown strength of a MOS capacitor increased as the weak spots i n the oxide were eliminated. By continued t e s t i n g , an ultimate breakdown strength i n the range of 5 - 10 MV-cm ^ could be reached. The breakdowns at lower f i e l d s (<5 MV-cm "*") were believed to be caused by weak spots i n the MOS structure. The breakdowns at higher f i e l d s were believed to be a property of the defect free MOS str u c -tures. The breakdowns at higher f i e l d s are c a l l e d i n t r i n s i c to di s t i n g u i s h them from the breakdowns at weak spots . In a d i f f e r e n t approach, Fritzsche studied breakdown i n MOS structures by using a sing l e breakdown per capacitor method. The breakdown f i e l d at which the f i r s t breakdown event occurred i n a capacitor was record-10 ed. The s t a t i s t i c a l d i s t r i b u t i o n of breakdown strength values of a large number of MOS capacitor was examined. The d i s t r i b u t i o n showed three maxima. According to F r i t z s c h e , these three peaks were interpreted as: the " t e r t i a r y " peak i n the range of 0 - 0.6 MV-cm ^ which was at t r i b u t e d to the e f f e c t of mic-ron si z e defects such as pinholes; the "secondary" peak i n the range of 1 - 3 MV-cm ^ which was a t t r i b u t e d to the presence of small randomly d i s t r i b u t e d , c r y s t a l l i z e d regions; and the "primary" peak i n the range of 6 - 9 MV-cm ^ represents the i n t r i n s i c breakdown strength of the oxide. The primary break-down strength obtained by Frit z s c h e corresponds to the ultimate or i n t r i n s i c d i e l e c t r i c strength values reported by K l e i n . Time dependent breakdown was f i r s t reported by Worthing (1968). On st r e s s i n g the sample with a constant voltage V, a Peek's law dependence of the time before breakdown (t)on the voltage (V)was observed; i . e . , V^t This r e l a t i o n s h i p was only found at room temperature with the s i l i c o n biased negatively. No time dependent breakdown was observed when the s i l i c o n was biased p o s i t i v e l y or at l i q u i d nitrogen temperature. Similar breakdown c h a r a c t e r i s t i c s were observed by Raider (1973) and Osburn and Raider (1973). The time dependent breakdown was explained as a re s u l t of sodium contamination. On st r e s s i n g sodium-contaminated samples with a constant voltage, metal electrode p o s i t i v e , the specimen current was observed to increase slowly with time as shown i n F i g . 2.2(a). The enhancement i n current was explained as an e f f e c t of the Na + motion i n the Si02« As the sodium ions d r i f t e d through the SiO^ f i l m , the space charge d i s t r i b u t i o n was continually modified ( F i g . 2.2(b)). The tr i a n g u l a r poten-t i a l b a r r i e r width was a function of both the applied f i e l d and the changing space charge f i e l d . The e f f e c t of the sodium ion motion caused the t r i -angular b a r r i e r width f i r s t to decrease, then to pass through a minimum, and 11 e o 2 0 0 400 E = 4.4MV/cm + 13 2 Na = 1 . 9 X 1 0 / c m 800 1000 1200 600 t ( SEC ) F i g . 2 . 2 ( a ) C u r r e n t v s . t i m e c u r v e f o r c o n t a m i n a t e d s a m p l e u n d e r c o n s t a n t b i a s . C u r r e n t s p i k e s r e p r e s e n t s e l f - h e a l i n g b r e a k d o w n s , ( a f t e r R a i d e r , 1 9 7 3 ) 800 850 S i O r 9 0 0 (X) 950 1000 F i g . 2 . 2 ( b ) C a l c u l a t e d v a r i a t i o n o f t u n n e l l i n g d i s t a n c e w i t h t h e p o s i t i o n o f a l a y e r o f p o s i t i v e c h a r g e i n S i 0 2 < T h e p o s i t i o n s o f t h e s h e e t c h a r g e a n d t h e t u n n e l l i n g d i s t a n c e s a r e : ( a ) 0 A * ( a t A 1 ) , 6 4 X ; ( b ) 850X , 35X ; ( c ) 967X , 33X (m in ) ; ( d ) 995X , 58X . ( A f t e r R a i d e r , 1 9 7 3 ) 12 f i n a l l y to increase again. The e l e c t r o n i c tunnelling current therefore increased i n the beginning, passed through a maximum and f i n a l l y decreased again. The f i r s t type of time dependent breakdown was observed before the current peak was reached. I t was believed to be caused by the b a r r i e r width reduction. For times beyond the current maximum, the b a r r i e r width had passed through i t s minimum and started to increase again. At t h i s time the sodium ions were very closed to the Si-SiO^ i n t e r f a c e , the high f i e l d region was too narrow to sustain an e l e c t r o n i c avalanche. The second type of time dependent breakdown was a t t r i b u t e d to the e f f e c t of b a r r i e r height lowering by the sodium dipole layer. Osburn and Chou (1973) and Osburn and Ormond (1974) studied the e f f e c t s of bias at elevated temperatures and of low temperature annealing i n nitrogen on MOS structures. The breakdown-related defect density was increased s i g n i f i c a n t l y a f t e r annealing or p o s i t i v e bias-temperature stress treatment. A f t e r p o s i t i v e bias-temperature stress treatment, a current enhancement which could lead to breakdown was observed. The current enhance-ment could be quenched by moisture or l i g h t but not by a negative b i a s -temperature stress treatment. This suggested that sodium ions were not the cause. The current enhancement was thought to be caused by some u n i d e n t i f i e d electrochemical process. 13 Chapter 3 THEORY OF INTERNAL PHOTOEMISSION IN MOS STRUCTURES 3.1 Interaction of r a d i a t i o n with the MOS structure When a MOS structure i s i r r a d i a t e d with photons of s u f f i c i e n t l y high energy, a photocurrent can be observed to flow through the i n s u l a t o r . The photocurrent may due to one or more of the following : 1) Photodepopulation of f i l l e d electron traps, 2) I n t r i n s i c photoconductivity of the i n s u l a t o r , 3) Photo-injected c a r r i e r s from the metal electrode, 4) Photo-injected c a r r i e r s from the semiconducting electrode. Photodepopulation of f i l l e d electron traps usually r e s u l t s i n a transient photocurrent as opposed to the continuing photocurrent produced by the other three mechanisms. For the metal-SK^-Si systems, there i s every reason to believe that the d-c steady state photocurrent i s due to the photo-i n j e c t e d c a r r i e r s from the electrodes rather than to the i n t r i n s i c bulk photoconductivity of the i n s u l a t o r . Experiments of Powell and Morad (1978) and of DiStefano and Eastman (1971) showed that the energy gap of SIO^ i s about 8 - 9 eV. The i n t e r f a c i a l b a r r i e r heights of several metal-Si02 i n t e r f a c e s and also of the Si-Si02 i n t e r f a c e are shown i n Table 3.1. For these b a r r i e r heights, i n t e r n a l photoemission can be stimulated with photons of energy l e s s than 6 eV, a value well below the i n t r i n s i c photoconductivity threshold of the s i l i c o n dioxide. Besides, the strong p o l a r i t y dependence of the photocurrent (Williams, 1965) also indicates that i t i s a photoinjection process rather than a photoconduction process. Internal photoemission experiments are usually done with a voltage 14 Metal or s i l i c o n B a r r i e r energy (eV) S i l i c o n 4.35 Aluminum 3.2 Gold 4.1 Nic k e l 3.7 S i l v e r 4.2 Copper 3.8 Table 3.1 B a r r i e r energy values for the s i l i c o n - s i l i c o n dioxide i n t e r f a c e and f o r various m e t a l - s i l i c o n dioxide i n t e r f a c e s . (After Deal et a l . , 1966.) The Si-SiO^ b a r r i e r i s measured from the top of the s i l i c o n valence band to the bottom of the SiC^ conduction band. The metal-SiG^ b a r r i e r i s measured from the metal Fermi l e v e l to the bottom of the SiO^ con-duction band. 15 bias applied across the MOS sample. Electrons emitted by the cathode and holes emitted by the anode can be swept by the f i e l d to the opposite elec-trode. For the case that the metal i s biased p o s i t i v e l y , photocurrent can be produced by photoemission of electrons from the s i l i c o n into the s i l i c o n dioxide or photoemission of holes from the metal into the s i l i c o n dioxide. The experiment of Williams (1965) showed that f or metal biased p o s i t i v e l y , the photocurrent was independent of the metal electrode material. This indicates that the photoemission of holes from the metal into the Si02 i s small compared to the photoemission of electrons from the s i l i c o n into the Si02- The con-t r i b u t i o n s of hole emission and electron emission i s better demonstrated by the o p t i c a l interference experiment of Powell (1969). Because of i n t e r n a l multiple r e f l e c t i o n , by sui t a b l e choice of oxide thickness, the l i g h t inten-s i t y can be made a maximum either at the metal-Si02 or at the Si-Si02 i n t e r -face. Contributions from anode hole emission or cathode electron emission can be studied separately. Powell confirmed that electron emission from the s i l i c o n into the Si02 i s the main constituent to the photocurrent when the metal gate i s biased p o s i t i v e l y . In the following section, we w i l l assume that f o r a p o s i t i v e l y biased metal gate, the photocurrent i n a MOS structure i s mainly contributed by electron emission from the S i into the SK^. The corresponding quantum y i e l d w i l l be calculated by taking into account the e f f e c t s of band bending at the semiconductor surface, of Schottky lowering of the b a r r i e r height, of impact i o n i z a t i o n i n the semiconductor and of sca t t e r i n g i n the in s u l a t o r . * More p r e c i s e l y i t i s the photoemission of electrons from Si02 to the anode thus generating holes. 3.2 Formulation of the problem A simple approximation to the volume photoemission process i s the three step model (Spicer, 1976) as shown i n F i g . 3.1(a). The three successive steps are: 1) Photoexcitation of electrons to a higher energy s t a t e , 2) Transport of electrons to the b a r r i e r , 3) Escape over the b a r r i e r . During the transportation step, photo-excited electrons may s u f f e r s c a t t e r i n g . In the escape step, only those electrons that s a t i s f y a certain energy and momentum condition can escape. F i g . 3.1(b) shows the band diagram of a MOS structure under p o s i -t i v e gate b i a s . The diagram i s shown for a p-type substrate with a depleted surface. However, the discussion which follows applies to n-type substrates and to surface conditions other than depletion. When a monochromatic r a d i a -ti o n reaches the s i l i c o n substrate, i t s i n t e n s i t y w i l l decay as i t tra v e l s into the s i l i c o n : T f -L. \ T -a(hv)x ,o T\ I(x,hv) = I e (3.1) o where I i s the l i g h t i n t e n s i t y at the Si-Si0„ i n t e r f a c e , a = a(hv) = 1/L i s o 2 a the absorption c o e f f i c i e n t i s the l i g h t attenuation length. The v a r i a t i o n of the absorption c o e f f i c i e n t a with photon energy hu o i s shown i n F i g . 3.2 (Dash and Newman, 1955). At a wavelength of 3250 A, the o l i g h t attenuation length i s about 300 A. I f the l i g h t attenuation length i s comparable to the semiconductor surface space charge layer width, photoexcitation of electrons occurs c h i e f l y i n the space charge l a y e r . The e f f e c t i v e b a r r i e r heights for photoelectron SiCL Si K P H O T O N P R O P A G A T I O N (3) E L E C T R O N E S C A P E , SCATTERING IN SiO, / (1) ELECTRON EXCITATION ^ (2) T R A N S P O R T I M P A C T I O N I Z A T I O N L I N S I L I C O N BARRIER J * Si-Si0 2 INTERFACE 3.1(a) The three-step photoemission process. M E T A L SiO, SILICON 3.1(b) Energy band diagram of M O S structure under positive bias (Illustrated with a p-type semiconductor) 18 1 . 2 "X.(pm) Fig. 3 . 2 Optical absorption coefficient of Si (oC) vs. wavelength ( > 0 • (After Dash and Newman,1955.) 19 o r i g i n a t i n g from d i f f e r e n t depths i n the space charge layer are not the same (Fig. 3.1(b)) unless the semiconductor i s under flat-band conditions. For p o s i t i v e gate bias, the e f f e c t i v e b a r r i e r height i s lower than the zero f i e l d b a r r i e r height $ f o r two reasons. F i r s t , i n the presence of an e l e c t r i c f i e l d E i n the oxide, the b a r r i e r i s lowered by the image force as described by the Schottky e f f e c t , i . e . : 2 / ire K o with the p o s i t i o n of the b a r r i e r maximum at x = Apz ^5" (3.3) o / I D T T E KE o where K i s the d i e l e c t r i c constant of the i n s u l a t o r . Second, the downward bending of the valence band w i l l cause the b a r r i e r to reduce by an amount Ai|)(x) = * s - <Kx) (3.4) where ^ g i s the surface p o t e n t i a l and ^(x) i s the amount of band bending measured from the bulk. The e f f e c t i v e b a r r i e r height seen by an electron at a depth x i n space charge layer i s therefore <j)(x) = <t>B-A<j>(E) - Ai|)(x) (3.5) The e f f e c t of image force on i n t e r n a l photoemission i n MOS struc-tures has been investigated by Goodman (1966) and Berglund and Powell (1971). F i g . 3.3 shows the value of the Schottky b a r r i e r lowering A<j> and the p o s i t i o n of the b a r r i e r maximum x as a function of the e l e c t r i c f i e l d i n SiO„. A o z high frequency d i e l e c t r i c constant of 2.15 (Goodman, 1966) i s assumed for these c a l c u l a t i o n s . For high e l e c t r i c f i e l d s , the b a r r i e r maximum i s very close to Si-SiO„ i n t e r f a c e . When the distance x i s the atomic dimensions, 2 o the c l a s s i c a l continuum d e s c r i p t i o n of b a r r i e r lowering may not be adequate. The smallest distance for which the Schottky model i s v a l i d i s not well 20 Fig. 3.3 Schottky lowering of barrier height ana" barrier position measured from the Si-SiO interface as a function of applied f i e l d . 21 established. Experiments by DiStefano (1976) in d i c a t e d that the Schottky 6 model i s applicable for distances from the i n t e r f a c e down to about 4 A. The e f f e c t of band bending on photoemission was f i r s t observed experimentally by Scheer (1960) on the cesium-covered s i l i c o n surface. p-type samples were found to give a higher quantum y i e l d than the n-type samples. A s i m i l a r e f f e c t was observed by Gobeli andAllen (1962) on cleaved s i l i c o n surfaces. A t h e o r e t i c a l model for vacuum photoemission which took in t o account the surface band bending was proposed by Gobeli andAllen. The model showed good agreement with the experimental data. I t should be noted that both the l i g h t attenuation length and the e l e c t r o n escape depth increase with decreasing incident photon energy. The lower photoemission threshold for Si-SiO^ (~4.2 eV) as compared to S i - vacuum (-5.15 eV) implies that photo-electrons can be excited from deeper i n the space charge l a y e r i n the case of Si-SiO^ i n t e r f a c e . We would expect a higher influence of the band bending on the quantum y i e l d f o r the Si-SiO^ system. The t o t a l amount of band bending (surface potential) at the s i l i c o n surface of a MOS structure i s r e l a t e d to the surface f i e l d . The v a r i a t i o n of the band bending with the oxide f i e l d E at the S i - S i 0 9 i n t e r f a c e i s shown i n F i g . 3.4. i\i i s given by the equation: (Appendix 1) e cosh(U +V ) ! E= B-\4r [ S -V tanhU, - l ] } " 2 ^ ± (3.6) £„...,. T2 coshU, s b e SxO^ L b where e„ = p e r m i t t i v i t y of s i l i c o n e = p e r m i t t i v i t y of Si0„ b i u _ £• 2 kT U, = bulk p o t e n t i a l of s i l i c o n (normalized to — unit) b e = -In P b/n 1 (p-type) = in n ^ / n ^ (n-type) I l I I I I I I I I I 1 M i l l IO5 IO6 E ( V/cm) Fig. 3.4 Silicon surface potential vs. oxide f i e l d at the S i - S i 0 2 interface. 23 kT V = surface p o t e n t i a l (normalized to — unit) s e . .kT = \p /— s e L = e f f e c t i v e Debye length E kT s e ( p b + nb) p^, n^ are the hole and electron d e n s i t i e s i n the semiconductor bulk res p e c t i v e l y . The e f f e c t i v e Debye length which characterizes the space charge layer width, i s shown as a function of doping i n F i g . 3.5. Electrons, photo-excited' at a depth x i n the semiconductor space charge layer, have to t r a v e l a t o t a l distance x + X q before they can reach the b a r r i e r maximum. During the transport process, some of the electrons w i l l be scattered and lose energy. The best i d e n t i f i e d energy loss process i n semiconductors with a photoemission threshold higher than the bandgap i s the impact i o n i z a t i o n process (Spicer, 1960). In such a process a s u f f i -c i e n t l y energetic electron excites a valence band electron into the conduc-t i o n band and produces a electron-hole p a i r . The threshold energy f o r the pair production process i s about twice the bandgap i n s i l i c o n (Spicer, 1960). The mean free path i s small, somewhere between 15 - 100 X (Spicer, 1960). I f a constant mean free path i s assumed as a f i r s t approximation, the f r a c t i o n of electrons which have not suffered a c o l l i s i o n a f t e r t r a v e l l i n g a distance x i s given by -x/L S (x) = e 6 (3.8) e where L^ = mean free path of the impact i o n i z a t i o n process. Those electrons that escape through the semiconductor surface may suf f e r a d d i t i o n a l s c a t t e r i n g as they t r a v e l from the Si-SiO^ i n t e r f a c e to the T N_ or N ( cm ) Fig. 3.5 Effective Debye length vs. doping in silicon. 25 b a r r i e r maximum. The mean free path f o r s c a t t e r i n g i n SiC^ was observed to be about 34 A at hu = 5 eV by Berglund and Powell (1971). Experiments by Peisner et a l . (1976), however, indicated the mean free path i s i n the range of 40 - 70 A f o r photon energy of 5 - 6 eV. I f a constant mean free path i s again assumed, the sc a t t e r i n g factor i s given by -x /L S (x ) = e ° p (3.9) P o where i s the mean free path of the s c a t t e r i n g process i n Si02« The distance x i s a function of the e l e c t r i c f i e l d i n the oxide, o I f we neglect the p r o b a b i l i t y of escape a f t e r one or more s c a t t e r i n g pro-cesses, the product of S and S would determine the f r a c t i o n of electrons e p which can escape the b a r r i e r . This i s a good approximation when the photo-emission i s studied near the threshold. In this case, the escape cone i s small. Electrons s u f f e r i n g one or more s c a t t e r i n g events are u n l i k e l y to have a normal component of momentum l y i n g within the escape cone. The photoemission process for a f l a t band semiconductor has been discussed t h e o r e t i c a l l y by Kane (1962) and experimentally v e r i f i e d by p Gobeli and A l l e n (1962). The quantum y i e l d has the form (hv-h v ) , where hv^ i s the photoemission threshold and P has values of 1,3/2,2 or 5/2. The actual value of P depends on the e x c i t a t i o n process and the s c a t t e r i n g mechanism assumed. For an unscattered volume photoemission process, P i s equal to 1 for d i r e c t o p t i c a l e x c i t a t i o n and 5/2 for i n d i r e c t o p t i c a l e x c i t a t i o n . Experimentally, a number of investigators have found that the best f i t to the power law usually gives a value of P from 2 to 3 (Deal et a l . , 1966 ; Goodman, 1966 ; Distefano, 1976), but the same data can often be f i t t e d within experi-mental er r o r by e i t h e r a square root or a cube root dependence. The basic difference i s i n the estimated value of the photoemission threshold. The 26 threshold f i t t e d by P = 3 can be about 0.15 eV lower than the threshold f i t t e d with P = 2. To calculate the quantum y i e l d , l e t us consider the photoelectrons o r i g i n a t i n g from a slab dx at a depth x i n the semiconductor. The d i f f e r e n t i a l y i e l d dY can be written as dY = c E^-[hv-(<|) B-(^ s(E)-^(x))-AME))] P s e ( x ) s p ( x o ) dx (3.10) o where I(x) i s the l i g h t i n t e n s i t y at x. I f we define an escape depth as LA L a L e (3.11) Then -x/L p -x /L dY = C e A[hv-<}>13+(^ -<Kx))+A<j>r e ° F dx (3.12) a s - , T 7 k T • • t, (Appendix 1) The r e l a t i o n s h i p between x and I|J = V — i s given by i J i r V _ d V (3.13) x = + ^ / v s ^ V„ F(U,,,V) where cosh(U+V) v F < V V ) = [ coshV - V t a n h V 1 ] 2 ( 3 - 1 4 ) b The negative sign i n Equation (3.13) i s used for a downward band bending while the p o s i t i v e sign i s used for an upward band bending. I t should be noted that A<j>, X q and tj> are dependent on the applied f i e l d . These quan-t i t i e s depend on the f i e l d as described by Equations 3.2, 3.3 and 3.6 re-s p e c t i v e l y . The quantum y i e l d Y w i l l be an in t e g r a t i o n of dY over those values of i> which give [hu-4>B+(ij;s-ijj)+A<t>] >0 (3.15) The d i f f e r e n t i a l y i e l d of Equation (3.12) can be rewritten as 27 1 fV L dV L A V s / 2 F ( U b ' V ) " X " ( E ) / L HY k T P o p d x = C e [hv-<PB+(^s(E)-V^-)+A<f)(E)] e (3.16) The quantum y i e l d i s 1_ rV L dV T J T L A V s /2 F ( U b ' V ) Y = C e [hv-* +(* (E)-V—)+A<f)(E)]^ V B s e o -x (E)/L T where V Q = ~ (hv-<J)g+^s(E)+A(t,(E)) Both dY / dx and Y of Equations (3.16) and (3.17) have been computed by a computer numerical method. The r e s u l t w i l l be discussed here. The physics of the photoemission process can best be described by a set of d i f f e r e n t i a l y i e l d curves dY/dx versus the distance x i n the space charge layer. F i g . 3.6 shows the e f f e c t of the escape depth on the d i f f e r e n t i a l y i e l d dY/dx. A l a r g e r escape depth L allows electrons excited deeper i n the space charge l a y e r , to escape the b a r r i e r . The area under the curves i s the quantum y i e l d Y. A larger escape depth L implies a higher quantum y i e l d . F i g . 3.7 shows the e f f e c t of change i n photon energy near the threshold. For lower photon energies eg. hv=3.7 eV and hv=3.85 eV, electrons close to the in t e r f a c e cannot be excited. The e f f e c t i v e b a r r i e r height (<j>g - (ip - i|>(x)) - A(f>(E)) seen by an electron at p o s i t i o n x decreases with increasing distance x. For photon energies hv. l e s s than (J) -A<j>, only those electrons deep i n the space charge l a y e r , where the e f f e c t i v e b a r r i e r height i s l e s s than the photon energy h v, can escape the b a r r i e r . This i s 28 Fig. 3.6 Calculated differential yield j(dY/dX) > versus distance from, the Si-SiO„ interfaae (X)- with the escape depth .(ti:) as-the parameter. 29 x ( X ) F i g . 3.7 Calculated differential yield (dY/dX) versus distance from the Si - S i 0 2 interface (X) with incident photon energy (h)>) as the parameter. 30 diemonstrated by the low d i f f e r e n t i a l y i e l d at small distances from the Si-Si02 i n t e r f a c e where the photon energy hv i s below the e f f e c t i v e b a r r i e r height. F i g . 3.8 shows the e f f e c t of band bending on the d i f f e r e n t i a l y i e l d dY/dx. Since the oxide f i e l d s are d i f f e r e n t for d i f f e r e n t band bendings, the curves are corrected for the Schottky e f f e c t f o r comparison. This can be done by introducing the new variables Y' = Y/exp (-X Q/L ) and hv 1 = hv+A(f> i n the c a l c u l a t i o n of dY'/dx. An increase i n band bending can be seen as a promotion of the photoemission closer to the i n t e r f a c e . F i g . 3.9 shows the e f f e c t of substrate doping on the d i f f e r e n t i a l y i e l d dY/dx. The curves are again corrected f or the Schottky e f f e c t . A photon energy hv<(cj>B-Ac}>) was assumed for these c a l c u l a t i o n s . It should be noted that the semiconductor space charge layer width increases with decreasing doping. A larger space charge layer width implies the p o s i t i o n s , f o r which hv><f> (x) = (tj> -(ip -Tp (x) )-A<)>), are deeper i n the space charge layer. I f these B s posit i o n s are large compared with the electron escape depth L ^ , the d i f f e r -e n t i a l y i e l d w i l l be small. This i s demonstrated by a s h i f t of the dY/dx curve to a larger x value and a shrinkage of the area under the curve as the doping i s decreased. I t should be noted that the scale f or dY/dx i s large for F i g . 3.9. The maximum d i f f e r e n t i a l y i e l d f o r p^=10^ i s twenty orders lower than that for p^=10^^. The d i f f e r e n t i a l y i e l d f o r p^=10^ i s too small to be p l o t t e d on the diagram. The low y i e l d of l i g h t l y doped semiconductors as shown i n F i g . 3.9 corresponds to the case of sub-threshold i l l u m i n a t i o n . I t i s useful to consider a flat-band approximation to the quantum y i e l d c a l c u l a t i o n s . The semiconductor i s e f f e c t i v e l y i n f l a t band condition i f : 1) The band bending i s small. For example a heavily doped n-type semi-conductor i n accumulation condition. 31 1 1 1 r V =+35.0 0 50 100 150 200 250 300 X <X) F i g . 3.8 Calculated d i f f e r e n t i a l y i e l d (dYVdX) versus distance from the Si-SiO i n t e r f a c e (X) with surface p o t e n t i a l (V ) as the parameter. 32 1 I 1 | I I I 1 1 1 1 1 1 I I I F i g . 3.9 Calculated d i f f e r e n t i a l y i e l d (dY'/dX) versus distance from the Si- S i C ^ i n t e r f a c e (X) with s i l i c o n doping (P^) as the parameter. C a l c u l a t i o n parameters: hx> '=h)J+A4>=4.00eV, L =23A\ L =347A, p=3, , A p V = +28.0(kT/e u n i t s ) , <b =4.15eV, Y = Y/ -X /L . The curves are s T B / e o p corrected f o r the Schottky e f f e c t . 33 2) The space charge l a y e r width i s large so that the amount of band bending within the electron^ escape depth L, i s small. This holds f o r the case of A high r e s i s t i v i t y semiconductors. For the f l a t band approximation, the quantum y i e l d described by equation 3.17 can be expressed i n a closed form -x (E)/L P Y = C e ° P[hv-<j> B+A<KE)] L A (3.18) Equation 3.18 i s i n the same form as obtained by Berglund and Powell (1971). F i g . 3.10 shows the quantum y i e l d c a l c ulated by using the flat-band approximation as compared to that calculated for other non-flatband conditions. The quantum y i e l d s of n-type semiconductors with an accumulation surface and the quantum yields of l i g h t l y doped p-type semiconductors are very close to the f l a t band approximation. F i g . 3.11 shows the quantum y i e l d as a function of the applied f i e l d for hv = 3.81 eV (He-cd l a s e r radiation) calculated by the f l a t band approxi-mation. The quantum y i e l d curve shows that the photocurrent turns on at a f i e l d of about 2 MV/cm. 3.3. Remarks Some remarks should be made on the model described i n this section. 1) In our c a l c u l a t i o n , photoelectrons are assumed to originate i n the valence band. Photoemission from the conduction band i s neglected. This assump-ti o n i s based on the experimental evidence of Goodman (1966). Goodman showed that the quantum y i e l d of photoemission from the conduction band i s several orders lower than photoemission from valence band even for 20 3 strongly n-type degenerate semiconductor (10 /cm ). 34 3.75 F i g . 3.10 1/3 NOTE: The Y vs. hv curves f o r the following cases f a l l between (c) and (d) and they are not shown in the diagram, n-type: N = 1 0 w, V =3.0, 1 fi NR=10 , V =5.0, 15 Nfi=10 , Vg=7.0. p-type: V 1 0 ' V28-0' P B=10 1 6, Vg=30.0. 3.95 4.55 4.75 4.15. . -4.35 hv> (eV) E f f e c t of s i l i c o n doping and surface p o t e n t i a l on the quantum y i e l d versus photon energy curve-. 4.95 35 T 1 r 1 2 3 4 5 6 E (MV/cm) Fig. 3.11 Calculated quantum yield (Y) versus applied f i e l d (E). Flat-band approximation was used for the calculation. 36 The e f f e c t of impact i o n i z a t i o n w i l l create hole-electron p a i r s i n the semiconductor surface space charge la y e r . These excess charges w i l l modify the band-bending i n the semiconductor. Since we have no knowledge about the density of these excess charges and t h e i r d i s t r i b u t i o n i n the space charge region, t h e i r e f f e c t on the surface p o t e n t i a l p r o f i l e i s excluded i n our c a l c u l a t i o n . Further experimentation i s required to test t h e i r influence. The i n t e r f a c e between Si-SiO^ i s assumed to be an abrupt t r a n s i t i o n . However, experiments by Chang (1978), Spicer (1976) and others indi c a t e d that there i s a gradual t r a n s i t i o n through the Si-SiO^ i n t e r f a c e . For the impact i o n i z a t i o n and the s c a t t e r i n g processes, constant mean free paths are assumed. This i s used only as a f i r s t approximation. Also escape a f t e r one or more sca t t e r i n g s i s neglected. These assumptions are not yet tested. 37 Chapter 4 DESIGN OF"THE COMPUTER CONTROLLED LASER SCANNER 4.1 P r i n c i p l e of operation The basic p r i n c i p l e of operation of the scanner, s p e c i a l l y designed for MOS studies, i s i l l u s t r a t e d i n F i g . 4.1(a). The beam from a h o r i z o n t a l l y o mounted RCA ID 2186 3mW He-Cd la s e r producing UV r a d i a t i o n of wavelength 3250A i s d i r e c t e d on to a front surface aluminum mirror mounted at 45° from the h o r i z o n t a l . The mirror r e f l e c t s the beam v e r t i c a l l y downwards in t o a focus-sing lens. The MOS sample to be examined i s positioned on top of a stepping motor driven X-Y t r a n s l a t i o n a l stage. The distance between the lens and the top surface of the sample i s adjusted to one f o c a l length f o r proper focussing, Scanning i s done with the l a s e r beam held stationary and the specimen moving i n a plane perpendicular to the o p t i c a l axis of the focussing l e n s . The f i x e d beam, moving sample approach has ce r t a i n advantages over the r o t a t i n g mirror method (Williams and Woods, 1972; DiStefano and Viggiantf, 1974). The focus of the l a s e r of a r o t a t i n g mirror scanner l i e s i n a curved surface. When the l a s e r beam scans over a f l a t surface, the angle of rotation of the mirror must be kept small to minimize focussing error and s p a t i a l non-l i n e a r i t y . This l i m i t s the maximum f i e l d area that can be examined. In the technique used here, the sample moves i n the f o c a l plane of the lens, provided the stage i s properly adjusted. A PDP8-E computer was used for c o n t r o l l i n g the sample movement, data reading, data storage, data manipulation, data p r i n t i n g and curve p l o t t i n g . Scanning i s done by stepping the sample under the l a s e r beam i n a pattern as shown i n F i g . 4.1(b). The computer software controls the scan according to 38 SHIELDING BOX He-Cd LASER MIRROR COMPUTER INTERFACE STEPPING MOTOR CONTROL Fig. 4 . 1(a) Principle of operation of the scanner. FIELD /COVERAGE Fig. 4 . 1(b) Scan pattern. 39 the sampling space (step size) , f i e l d coverage (number of X and Y steps) and scan speed as s p e c i f i e d by the operator. In each scan, a two dimensional array of data points i s c o l l e c t e d and i s stored on DECtape f or future use. The r e s u l t s can be printed out as a 3-D map or i n other appropriate forms. The d e t a i l s of the scanner operation w i l l be described i n Section 4.5. 4.2 Optics of the l a s e r scanner 4.2.1 C h a r a c t e r i s t i c s of the focussed l i g h t spot The f i n e s t geometry of the sample that the scanner can examine i s determined by two fa c t o r s , namely the incremental step si z e of the X-Y trans-l a t i o n a l stage and the spot s i z e of the focussed l a s e r beam. The step s i z e of the X-Y t r a n s l a t i o n a l stage i s l i m i t e d by the mechanical p r e c i s i o n of the gears, backlash, etc. The step s i z e i s two microns. The spot si z e depends on how the focussing optics i s set up. Small spot s i z e can only be obtained at the expense of reduced beam power. The factors a f f e c t i n g the spot s i z e and beam power w i l l be discussed i n t h i s section. Consider the case that the l a s e r i s operating i n the TEM mode oo with a Gaussian beam i n t e n s i t y p r o f i l e and the focussing lens i s aberration-free. The i n t e n s i t y d i s t r i b u t i o n and the spot s i z e of the focussed l i g h t spot depends on the width of the incoming beam r e l a t i v e to the lens diameters as i l l u s t r a t e d i n F i g . 4.2. As the incoming beam width i s reduced, two s i g n i f i -cant things happen. The power concentration i n the beam center increases and the s i z e of the focussed spot increases. For the case that the incoming beam width i s much la r g e r than the lens diameter, the lens can be considered uni-formly illuminated. The l a s e r beam i s d i f f r a c t e d by the lens aperture into 40 BEAM LENS OF GIVEN SPOT SIZE DIAMETER NUMERICAL AND POWER PROFILE REDUCES APPERTURE INCREASES F i g . 4 . 2 I n t e n s i t y d i s t r i b u t i o n o f t h e l i g h t s p o t a s a f u n c t i o n o f t h e i n c o m i n g b e a m w i d t h . an Airy pattern of bright and dark rings. The s i z e of the focussed l i g h t spot i s conventionally defined as the diameter of the c e n t r a l bright Airy d i s c . The d i f f r a c t i o n - l i m i t e d spot s i z e i s small. Since only part of the beam can pass through the lens aperture and l i g h t i s d i f f r a c t e d into side lobes, the c e n t r a l A i r y d i s c r e t a i n s only a f r a c t i o n of the incoming beam power. The spot diameter d i s given by d A = 2.44 Af/D £ (4.1) with an i n t e n s i t y d i s t r i b u t i o n f v 2 J (p) 4 ^ = t — — ] (4-2) where p = TT D^r/Af = lens diameter f = lens f o c a l length = Bessel function of order 1 r = radius measured from the o p t i c a l axis For the case that the l a s e r beam width i s much smaller than the lens diameter, the l a s e r beam i s v i r t u a l l y u n d i f f r a c t e d and the focussed l i g h t spot has a Gaussian i n t e n s i t y d i s t r i b u t i o n . The radius of the focussed l i g h t spot i s conventionally defined as the distance measured from the beam axis to 2 the point where the l i g h t i n t e n s i t y drops to 1/e of i t s maximum a x i a l value. For a given lens, the s i z e of a Gaussian spot i s usually l a r g e r than the s i z e of the A i r y d i s c . Since most of the la s e r beam can pass through the lens aperture, the t o t a l power i n the Gaussian spot i s higher. The spot diameter d i s given by g d = ^ L i - (4.3) g TT D b with an i n t e n s i t y d i s t r i b u t i o n 42 8ri ~ d 2 ^ = e 8 (4.4) o where D, = l a s e r beam diameter, b The power than can pass through the lens i n both cases i s given by D * 2 -2(—) D P = P [1-e ] • T (4.5) o where P q i s the t o t a l laser beam power, T i s the lens transmittance. For the case that the numerical aperture of the lens i s f i x e d , the power and the l i g h t spot s i z e as a function of the incoming beam width i s shown i n F i g . 4.3. It should be noted that the actual value of the spot s i z e depends on the lens numerical aperture. A lens with a larger numerical aperture gives a smaller spot s i z e . Furthermore, F i g . 4.3 shows i f a small spot s i z e i s desirable for high r e s o l u t i o n scanning a p p l i c a t i o n s , the l a s e r beam should be expanded. The lens aperture should be o v e r f i l l e d f o r d i f f r a c t i o n - l i m i t e d operation. However, t h i s would be accompanied by some loss i n beam power. The primary concern of our experiment i s to obtain a high beam power with a reasonable r e s o l u t i o n . The l a s e r beam i s used without expansion. There i s also another factor that a f f e c t s the l i g h t spot s i z e , namely the l a s e r beam divergence. The spot s i z e contributed by beam di v e r -gence 6 i s given by D = f 8 (4.6) If a small spot si z e i s desirable, the l a s e r beam should be s p a t i a l l y f i l t e r e d i n addition to expansion to reduce the beam divergence. A3 T Fig. A.3 Total power and the size of the focussed light spot versus incoming beam width. 44 4.2.2 Focussing optics To obtain a minimum spot s i z e f o r a given o p t i c a l set up, the d i s -tance between the focussing lens and the specimen should be adjusted to one f o c a l length. The experimental method to achieve t h i s condition i s , however, not t r i v i a l . Several points should be noted i n designing a focussing set up. i ) Although the l a s e r operates i n the u l t r a - v i o l e t , a crude v i s u a l inspec-t i o n of the l i g h t spot s i z e i s possible by examining the fluorescence ; caused by the UV s t r i k i n g a sui t a b l e target. The UV i t s e l f i s , of course, a p o t e n t i a l hazard. i i ) Various photographic techniques can be used to determine the l i g h t spot s i z e . Photographic methods, however, require a series of t r i a l and er r o r measurement to obtain the smallest spot si z e so that they are time consuming and inconvenient. Since the thicknesses of the photographic emulsion and of the sample are usually d i f f e r e n t , i f the lens i s properly adjusted f o r the photographic plate thickness i t w i l l not n e c e s s a r i l y be i n focus when the sample i s substituted, i i i ) The l i g h t spot i s usually a few microns i n diameter so that i t i s too small to be analysed by a photo-detector array. An i d e a l focussing method should achieve the following objectives, i ) V i s u a l inspection (which involves exposure to scattered UV) w i l l not be required. i i ) The focussing condition can be accessed with the sample on s i t e . i i i ) E f f e c t s of sample thickness w i l l be eliminated. i i i ) The size o f the focussed spot can be determined with the sample on s i t e , i v ) For the best use of the computer con t r o l method, the focussing set up should provide a p o s s i b i l i t y of automatic focussing. 45 A focussing set up which can achieve these goals i s shown i n F i g . 4.4. A d i r e c t study of the l i g h t spot i s not attempted. In t h i s set up the magnified image of the l i g h t spot i s instead studied. The l i g h t spot on the p a r t i a l l y r e f l e c t i v e sample surface i s considered as a secondary l i g h t source. The image of t h i s secondary l i g h t source i s formed by lenses 1 and 2 at a p i n -hole. Here, the p o s i t i o n of lens 2 i s f i x e d at a distance equal to i t s f o c a l length f 2 from the pinhole. The focus i s set by varying the distance of the focussing lens 1 from the sample surface. The actual o p t i c a l path i s shown i n F i g . 4.5. In the f i g u r e , lens 1 i s shown to be off-focused by an amount Af. For the s i m p l i c i t y of dicussion, the l i g h t spot i s assumed to be a disc of radius R^. I t s magnified image i s a disc of radius R^- The r e l a t i o n s h i p between R^ and R 2 i s given i n f i r s t approximation by the equations (Buchl,1970) R1 = R O + ^ M (4.7) R 2 = ^ ( R 1 + - ^ ) (4.8) where R q = spot s i z e when the beam i s properly focussed. D, = l a s e r beam width, b f f 2 = f o c a l length of lens 1 and lens 2 r e s p e c t i v e l y . Af = displacement of the focussing lens 1 from i t s focus. F i g . 4.6(a) shows the v a r i a t i o n of R^ and R 2 with the displacement Af/f as calculated by Equations 4.7 and 4.8. R^ and R 2 are at t h e i r minima when lens 1 i s properly focussed. F i g . 4.6(b) shows the measured photodetector output as a function of the r e l a t i v e p o s i t i o n of lens 1. The s i g n a l shows a maximum as predicted. The photodetector output would provide a means for automatic focussing with the sample on s i t e . BEAM SPLITTER MIRROR PHOTO-DETECTOR Fig. 4.4 Focussing optics of the laser scanner. When the laser beam focussed, the photodetector output i s at a maximun. F i g . 4.5 Image of the l i g h t spot produced by the focussing lens 1 and lens 2. 48 •^orbr ^OTOT A f / f o.oi 0 . 0 2 F i g . 4.6(a) Light spot s i z e and as a function of lens 1 displacement from the focus. 7.2 7.4 7.6 7.8 MICROMETER POSITION (mm) F i g . 4.6(b) Phase s e n s i t i v e detector output as a function of the p o s i t i o n of the focussing lens 1. 49 4.2.3 Measurement of the l i g h t spot s i z e A method which can determine the l i g h t spot s i z e with the sample i n s i t u i s discussed i n this section. F i g . 4.7(a) shows a surface with two regions of d i f f e r e n t r e f l e c t i v i t y separated by a s t r a i g h t edge. Such a structure i s often found i n planar semiconductor devices, f or example, an aluminum layer on top of SiO^ with an edge defined by photolithography. I f a Gaussian beam i s scanned across t h i s edge, the t o t a l power r e f l e c t e d by the surface can be calculated by convolving the i n t e n s i t y d i s t r i b u t i o n of the l i g h t spot I(x,y) with the reflectance of the surface R(x,y). The reflectance of the surface can be defined as R(x) = < R x<0 R 2 x>0 (4.9) The i n t e n s i t y d i s t r i b u t i o n of the Gaussian beam i s : 8 .2 2, 5- (x + y ) 8P D I(x,y) = ^ 2 e (4.10) where D = diameter of the Gaussian spot as defined i n Section 4.2.1 P = t o t a l power c a r r i e d by the Gaussian spot Assuming the l i g h t spot scans along the x-axis, the l i g h t power r e f l e c t e d from the surface as a function of the p o s i t i o n of the l i g h t spot i s O  00 P R(x) = _/ _/ R(n) K n-x . O d n d S (4.11) The convolution i n t e g r a l y i e l d s a rather simple r e s u l t R R~ P R(x) = P [ - y (1-2 e r f (^X) ) + ~- (1+2 e r f ( £ x ) ) ] (4.12) where the error function i s defined as - t 2 / 2 erf(x) = — / X e dt (4.13) /27 ° F i g . 4.7(b) Edge p r o f i l e as calculated by equation 4.12. 51 F i g . 4.7(b) shows the v a r i a t i o n of the normalized r e f l e c t e d l i g h t power as the l i g h t beam i s scanned across the edge. The edge i s "rounded by the Gaussian beam as expected. The curves always pass through an invariant point with P /PR0 = 0.5(1+K) regardless of the value of the l i g h t spot diameter. F i g . 4.7(b) shows that the edge p r o f i l e i s a unique function of t h e ' l i g h t spot diameter D. Hence the l i g h t spot diameter can be deduced from the edge p r o f i l e . 4.3 Hardware of the laser scanner F i g . 4.8 shows the set up of the laser scanner. The beam of the He-Cd la s e r was directed onto a fused quartz f l a t which was used as a beam s p l i t t e r . A small percentage of the beam power (about 6%) was r e f l e c t e d into a photodetector for laser power l e v e l monitoring. The beam that passed through the quartz was r e f l e c t e d downwards and focussed on the sample. The photoemission current induced by the l a s e r beam was measured by a e l e c t r o -meter with a b i a s i n g c i r c u i t as shown i n F i g . 4.9(a). Since the photo-—10 —8 current was i n the 10 to 10 ampere range, care had to be taken .-to min-imize noise pick up during measurement. Noise l e v e l was kept under 20 mV (2% of f u l l scale) at the output of the electrometer. The electrometer amplified s i g n a l was connected to an analog channel at the computer i n t e r -face. This enabled the computer to read i n the photocurrent. A specimen heater was b u i l t on the scanning stage to provide heat treatment to the sample. This allowed scanning and bias-temperature stress treatment to be done i n s i t u without removing the sample from the scanner. In s i t u heating was p a r t i c u l a r l y u seful when the same area was to be scanned before and a f t e r bias-temperature stress treatment. Removal of the sample PHOTO-DETECTOR He-Cd LASER /\/\ z 4>- RECORDER (HP 7100BM) (RCA LD 2186) ^ > QUARTZ ( LIGHT CHOPPER BROOKDBAL < > l ^ V MIRROR 9479 LENS 1 M A W LENS 2 — P I N HOLE DETECTOR TEMP. CONTROL (YSI 72) ELECTROMETER + BIAS SAMPLE-HEATER (KEITHLEY 602b-c> A/Df SCAN STAGE MOTOR CONTROL PDP8-E INTER-FACE PHASE SENSITIVE DETECTOR (PAR HR-fc) A/lt Fig. 4.8 Set up of the laser scanner. 5 3 TO ANALOG CHANNEL ELECTRO-METER LO A METAL GATE SPECIMEN BATTERY SUPPLY Fi g . 4.9(a) Measurement of the photocurrent. TIME COOLING HEATING 40 30 . 20 TIME ( MINUTE ) 10 F i g . 4.9(b) Temperature cy c l e recorded by a s t r i p chart recorded. 54 (b) F i g . 4.10 Sample l o c a t i o n by using r e f l e c t i v i t y scan, (a) Photocurrent map of sample 005-32fl. The edge of the aluminum gate i s not v i s i b l e on the map. (b) R e f l e c t i v i t y map of the same sample. The c e n t r a l plateau on the map i s the aluminum gate. 55 from the scanning stage often resulted i n alignment error . The specimen holder had a large thermal mass. To provide rapid temperature c y c l i n g , a temperature c o n t r o l l e r was used to speed up the r i s e time of a heating cycle while forced a i r cooling was used to reduce the f a l l time. A t y p i c a l heating and cooling cycle i s shown i n F i g . 4.9(b). The l i g h t r e f l e c t e d from the sample surface would retrace i t s i n -cident path and be r e f l e c t e d by the r i g h t hand surface of the quartz f l a t down to lens 2. I t was focussed onto a pinhole and then detected by a photo-detector. The i n t e n s i t y of the r e f l e c t e d l i g h t was usually low. A phase s e n s i t i v e detector was used to improve the s i g n a l to noise r a t i o . The assembly, co n s i s t i n g of lens 2, pinhole and photodetector, served two purposes. F i r s t , i t was an a i d for l a s e r beam focussing as discussed i n Section 4.2.2. Second, i t was used to analyse the r e f l e c t i v i t y of the sample surface. The reason for a r e f l e c t i v i t y measurement i s apparent i f we consider the photocurrent map as shown i n F i g . 4.10(a). This photocurrent map has a very low photo-current i n the background. The edge of the aluminum gate i s not v i s i b l e on t h i s map. In other words, the sample cannot be located. To f i n d the p o s i t i o n of the sample, a simultaneous scan of the photocurrent and the r e f l e c t i v i t y was used. A r e f l e c t i v i t y map (Fig. 4.10(b)) i n addition to the photocurrent map was printed to i d e n t i f y the l o c a t i o n of the sample. In the r e f l e c t i v i t y mode, the scanner operated as a scanning o p t i c a l microscope. As a precaution, to protect the operator and the people working i n the neighbourhood from exposure to the d i r e c t or r e f l e c t e d i n v i s i b l e UV l a s e r beam, the scanner was contained i n a black box. A safe-lock was incorporated i n the unit so that the power to the laser would be cut o f f i f the l i d of the box was opened a c c i d e n t a l l y . The black box had forced a i r fan cooling. This 56 was used to improve the v e n t i l a t i o n and to provide cooling of the He-Cd las e r for better long term s t a b i l i t y . 4.4 Software and operation of the laser-scanner A PDP8-E based 0S8 operating system was used for the control of the l a s e r scanner. The system consisted of a PDP8-E ce n t r a l processing u n i t , i n i t i a l l y with a 16K words of core memory (now 32K), a dual drive DEC tape for computer program and user's measurement data storage, a VT52 DEC scope as the conversational terminal, a teletype for hard copy output and a DP10 incremental p l o t t e r for graphic output. Programs were written i n an OS8-compatible mixture of Fortran II and SABR assembly language. One routine was written i n 0S8 Fortran IV. The programs and t h e i r function are l i s t e d i n Appendix 2. The software of the scanner were prepared i n a way to make the operation of the scanner s e l f -i n s t r u c t i o n a l . The user only required to give Y/N answers to a ser i e s of questions and to provide c e r t a i n parameters to the computer. A f i l e MEMO was a v a i l a b l e f or the explanation of the scanner operation. There were two pro-grams for operating the scanner. The program SIPM was used for normal opera-t i o n while the program SIP8R was used for preliminary scanning. In the normal mode of operation, the program SIPM was used i n con-junction with the map p r i n t i n g routines DMAP and LABEL. The flow chart for the operation of the routine SIPM i s shown i n F i g . 4 . 1 1 . The program began by s h i f t i n g the sample to a desired area for scanning. The computer then requested the operator to specify the incremental step s i z e , the scan speed and the area to be scanned. It should be noted that both the electrometer c i r c u i t and the phase s e n s i t i v e detector had f i n i t e response times. The X-Y t r a n s l a t i o n a l stage could only be stepped at a rate s i g n i f i c a n t l y slower than 57 MOVE SAMPLE FROM KEYBOARD V  READ IN SCAN PARAMETERS (1) STEP SIZE,(2)SCAN AREA, (3) TIME CONSTANT, (4) FILE NAME SCAN TIME ESTIMATION SCAN THE SAMPLE WITH THE STEP AND READ PROCEDURE OUTPUT DATA TAPE DATA PRESENTATION ( 3 OPTIONS ) READ DATA FROM TAPE \ f \ \ PRINT A MATRIX ON THE VT-52 (FOR SIP8R ONLY) SHADOW MAP ON DP-10 3-DIMENSIONAL MAP ON DP-10 F i g . 4.11 Flow chart of the operation of the scanner. 58 the c i r c u i t response time. This was done by i n s e r t i n g a waiting loop i n the subroutine "step". The t r a n s l a t i o n a l stage moved the sample one step and stopped f o r a time T = K T , where x was the maximum of the electrometer and the phase s e n s i t i v e detector time constant. K was a user constant usually between 3 to 10. The computer also requested the operator to specify a data f i l e name i f the r e s u l t was to be stored on tape. The t o t a l time f o r a scan was estimated by the program and output to the VT-52 f or the information of the user. Af t e r a l l the i n s t r u c t i o n s were received, the program started the scan. When the scan was f i n i s h e d , a message "scan completed" was output to the VT-52. Since the data were saved on a tape, the operator had the option of p r i n t i n g the r e s u l t immediately or at a l a t e r time. The program DMAP was used to p r i n t e i t h e r a 3-dimensional map or a shadow map. The routine LABEL was used to l a b e l the map axes and to give a t i t l e to the map. The program SIP8R was a fa s t version of the SIPM. I t was pr i m a r i l y used f o r preliminary scanning. To increase the speed of the operation, data were not stored on tape. The maximum data sampling point was l i m i t e d to 18 X 18. The 3-dimensional graph p l o t t i n g was also not used. Data was printed on the VT-52 CRT Terminal as a matrix. The time for a SIP8R scan was usually a f r a c t i o n of a minute. The time f o r a SIPM scan va r i e d from a few minutes to a few tens of minutes, depending on the f i e l d width and the scanning speed used. 59 Chapter 5 DESIGN OF THE SELF-HEALING BREAKDOWN TESTER One of the objects of t h i s work was to study the c o r r e l a t i o n between enhanced photoemission and sodium induced breakdown. The s e l f - h e a l i n g break-down technique was used for producing breakdowns on MOS samples. The advant-ages to t h i s procedure are: (1) Since the sample i s by d e f i n i t i o n not shorted a f t e r s e l f - h e a l i n g breakdown, further experimentation on the same sample i s possible. (2) The p o s i t i o n s of the s e l f - h e a l i n g breakdown s i t e can be located and correlated with other information concerning the sample, for example scanning i n t e r n a l photoemission map. To produce s e l f - h e a l i n g breakdown on a MOS sample, two types of voltage t e s t can be used. The f i r s t i s the ramp voltage test as used by K l e i n et a l . (Klein and Gafni, 1966; K l e i n , 1966). The second i s the step voltage te s t as used by Worthing (1968) and Raider (1973). The step voltage test i s easier to set up. However, the rapid decrease of breakdown time with the applied f i e l d and sample-to-sample v a r i a t i o n s make the breakdown time d i f f i c u l t to p r e d i c t . The choice of an appropriate stress voltage has to be arr i v e d by t r i a l and error. For t h i s reason, the ramp voltage test was used. F i g . 5.1 shows the basic set up for a ramp voltage breakdown te s t . In t h i s set up, the output of a ramp generator is connected to the specimen i n s e r i e s with a current l i m i t i n g r e s i s t o r (>10Kfi). The specimen voltage i s measured as a function of time with a s t r i p chart recorder. When the oxide of a MOS structure breaks down, a conducting filament i s formed at the break-down spot. The charge stored i n the c a p a c i t o r - l i k e MOS structure discharges through t h i s conducting filament and r e s u l t s i n a drop i n the specimen voltage. 60 Resistor •A/W-RAMP GENERATOR Specimen i c RECORDER t=0 Measurement of self-healing breakdown in MOS structures. 61 Af t e r the oxide and the metal gate surrounding the filament are evaporated by the l o c a l i z e d heating, the specimen voltage r i s e s again to the supply voltage. A spike w i l l be seen on the recorded ramp voltage curve as shown i n F i g . 5.1, provided that the s t r i p chart recorder i s 1 f a s t enough to r e g i s t e r the voltage spike. The spike on the ramp voltage curve marks the time and the voltage at which breakdown occurs. I t should be noted that during breakdown, the specimen voltage collapses to a minimum value i n a very short time, from few nanoseconds to few tenths of a microsecond (Klein,1966). The specimen voltage then r i s e s to the ramp voltage with a time constant determined by the capacitance of the MOS structure and the resistance of the current l i m i t i n g r e s i s t o r . This c i r c u i t time constant i n our experiment was of the order of microsecond. A voltage spike of few microsecond duration was too f a s t to be recorded by a mechanically driven s t r i p chart recorder. Hence, spikes did not occur, on the recorded ramp voltage curve i n our experiment. To mark a breakdown event i n our case, the technique used by Carnes and Duffy (1971) was employed. In t h i s technique, the ramp voltage was reset to zero by an e l e c t r o n i c c i r c u i t when an abrupt increase i n specimen current was sensed during breakdown. The time and the voltage at which s e l f - h e a l i n g breakdown occurred were defined by the peak of the ramp voltage. The block diagram of the breakdown tester which was designed to achieve the above objectives i s shown i n F i g . 5.2 with the d e t a i l s of the e l e c t r o n i c c i r c u i t r y shown i n Appendix 3. The ramp generator i n the breakdown tester was b a s i c a l l y a large value capacitor charged by a constant current source. The use of a large r e s e r v o i r capacitor (20 yF) reduced the loading e f f e c t of the specimen on the ramp generator. The ramp rate was continuously HIGH VOLTAGE SUPPLY MANUAL START/ STOP CONTROL OVERVOLTAGE COWBAR START/ STOP RAMP GENERATOR RESET AUTO STOP. O/P CURRENT LIMITING RESISTOR SCR RESET CIRCUIT COUNTER STOP CONTROL COUNTER 200 mS SCR RESET CIRCUIT J 1 THRESHOLD DETECTOR PULSE SHAPER R>10K 1000M PREAMPLIFIER ] SPECIMEN I I M I N I M U M P U L S E D E T E C T A B L E =10.0jiA / 0.7;JS F i g . 5.2 Block diagram of the automatic s e l f - h e a l i n g breakdown t e s t e r . (S3 63 adjustable from 5 V/sec to 100 V/sec. The maximum output voltage was 400 v o l t s which gave a f i e l d strength of 10 MV/cm or higher f o r oxide thickness up to 4000 X. The voltage across the sample was measured by a HP 7100BM s t r i p chart recorder of 1 Mfi input resistance i n series with a 1000 r e s i s t o r . The r e s i s t o r was used to reduce the loading e f f e c t of the recorder on the ramp generator and also to provide a su i t a b l e attenuation of the ramp voltage. The specimen current was monitored by a current sensing c i r c u i t c o nsisting of a preamplifier and a threshold detector. The current threshold was adjustable from 10' pA to 10 mA. The smallest current pulse width that could be detected was 0.7 ysec. The voltage drop across the current sensing c i r c u i t was 100 mV when the threshold current was reached. This voltage was small compared to the voltage across the specimen. The current sensing c i r c u i t was protected against sample short c i r c u i t s since i n the event of non-self-healing breakdown, a large f r a c t i o n of the ramp voltage could appear across the input of the current sensing c i r c u i t . . The counter stop control i s a s p e c i a l feature of the breakdown tester. I t was used to terminate the experiment when a desired number of breakdown was reached. I t could be disabled when continuous ramp breakdown tests were desirable, but i n p r a c t i c e i t was more useful to employ the counter stop con-t r o l so that a desired number of breakdown events could be examined at a time. The operation of the s e l f - h e a l i n g breakdown tester was as follows, (1) The breakdown test was started manually by throwing a switch. This s t a r t s the ramp voltage with a pre-selected ramp rate. (2) When the breakdown voltage was reached as defined by the specimen cur-rent exceeding a preset threshold, the current sensing c i r c u i t produced a TTL compatible voltage pulse. The width of t h i s voltage pulse was expanded to 200 msec. The output of the ramp generator was reset to zero by the r i s i n g edge of the expanded pulse and kept at the zero voltage l e v e l for 200 msec. This reset period was s u f f i c i e n t l y long f o r the recorder pen to return to zero. At the same time, the counter r e g i s t e r e d a breakdown event. A f t e r the reset period, the ramp generator automatically started again to repeat the breakdown t e s t . When the desired number of breakdowns was registered by the counter, a s i g n a l was automatically put out by the counter stop control to terminate the breakdown t e s t . The ramp generator could be re-started again only by a manual s t a r t c o n t r o l . The breakdown test could also be stopped at anytime by a over-riding manual stop c o n t r o l . 65 Chapter 6 SAMPLE DESIGN AND FABRICATION 6.1 Design considerations It was required to design a sample, b a s i c a l l y a metal-SiO^-Si capacitor, which could be used both for laser scanning i n t e r n a l photo-emission and for s e l f - h e a l i n g breakdown measurements. For laser scanning i n t e r n a l photoemission measurement, the metal gate i s also the o p t i c a l window for the laser beam. This requires the metal gate to be s u f f i c i e n t l y t h i n so that i t i s semitransparent to the laser wavelength concerned. The m e t a l l i z a t i o n s that were r e a d i l y preparable i n this laboratory were gold and aluminum. Gold has a higher o p t i c a l trans-mittance than aluminum but i t shows poor adherence to S i 0 o unless a f i r s t layer of chromium i s added. This second property makes gold undesirable to be the gate material. Aluminum was therefore chosen to be the material for the gate electrode. F i g . 6.1 shows the transmittance of aluminum as a function of f i l m thickness. The curve shows the aluminum transmittance decreases as i t s thickness increases. To achieve an o p t i c a l transmittance of higher than about 10%, the thickness of the aluminum should be less than 200 X . Films of such thickness are not mechanically strong enough for making e l e c t r i c a l connections using a pressed contact. A m e t a l l i z a t i o n layer of several thousand angstrom thickness i s usually required. A two layer m e t a l l i z a t i o n process as shown i n F i g . 6.2 was used to provide both an e l e c t r i c a l contact and a semitransparent o p t i c a l window for the sample. E l e c t r i c a l connection was made by means of a spring loaded, spherical-ended press contact probe. 66 W H H M 70 _ 60 L 50L AO 30 20 10 40 7^=4000$ 80 120 160 200 FILM THICKNESS ( 8 ) 240 Fig.6.1 Transmittance of aluminum vs. f i l m thickness. THICK AL (CONTACT PAD) (^5000^) THIN AL (OPTICAL WINDOW) (<160S) S±Or S i AL F i g . 6.2 MOS capacitor with two layer m e t a l l i z a t i o n . 67 This prevent the p o s s i b i l i t y of probe penetration into the oxide and subsequent breakdown underneath the probe. The sample designed so far would not be s u i t a b l e for s e l f - h e a l i n g breakdown measurement since the f i e l d strength i n the SiO^ i s the same underneath the t h i n and the thick aluminum. Breakdown would occur equally probably i n the two regions and breakdown i n the thick aluminum region would not be s e l f - h e a l i n g . To prevent shorting breakdown underneath the thick aluminum contact pad, the structure i n F i g . 6.3 was used. A thick SiC^ was used to reduce the f i e l d strength underneath the thick aluminum and hence the p r o b a b i l i t y of shorting breakdown. With t h i s structure, o p t i c a l and breakdown measurements were done i n the t h i n aluminum region while the thick aluminum served purely as an e l e c t r i c a l connection. 6.2 Sample f a b r i c a t i o n The sample were fabricated by a three mask process. The f a b r i -cation steps are shown i n F i g . 6.4. The substrates were two inches diameter s i l i c o n wafers supplied by Monsanto. Both n-type (100) phosphorus doped wafers of 1-2 Q-cm r e s i s t i v i t y and p-type (111) boron doped wafers of 4-16 Q - cm r e s i s t i v i t y were used. Before oxidation, the wafers were cleaned by a RCA cleaning process using ^2^2 + ^ 4 ^ ' d i l u t e d HF and ^2^2 + s ° l u t i - o n s # The d e t a i l s of the process are outlined i n Appendix 4. 6.2.1 Reinforcement oxide preparation The preparation of the reinforcement oxide i s shown i n steps 2 and 3 of Fig. 6.4. The thick reinforcement oxide was grown i n a burnt hydrogen oxidation furnace at a temperature of 1100°C. The 5+5+50+10+20 oxidation cycle was as follows: THICK AL ( « 0 0 0 8 ) THICK SiCL (i^.5400S) / / J ^ THIN AL (120-160/2) THIN SiO ? (1000A) t ^300_um SILICON I I I I I I I Au OR A l (3000-5000A) / / / / / / / / / TOP VIEW THICK SiO THIN SiO r 0. 7mm F i g . 6.3 Pattern of a MOS capacitor with a duplex structure. 69 (1) S i Pre-furnace cleaning SiO, (2) (3) / / / / / i l l I I I I I I I S i > i i i i i i i i i i i i r-r S i 0 o SiO 7777771 S i 0 2  1/77771 S i 1st oxidation (reinforcement oxide) Reinforcement oxide etching (4) T7TT71 IT77777 Pre-furnace cleaning (5) (6) (8) (9) jmhrrrmJlV Thick A l ( 7 ) 7W>W^77777 Au or A l (10) (ii) -TrTTL^Jrrhv Thin A l I I l l I I I I I I I I i l-TZLTL 2nd oxidation (gate oxide) Thick A l evaporation Backside oxide removal Ohmic contact evaporation Ohmic contact s i n t e r i n g (470°C,30 minutes,N ambient) Pattern d e f i n i t i o n of the thick A l Thin A l evaporation (12) 11 i i i i i n i i i i i m Pattern d e f i n i t i o n of the t h i n A l F i g . 6.4 Sample f a b r i c a t i o n . 70 Time (minute) Ambient Description. 5 °2 Heat up 5 0 2 + 5% HC£ Dry oxidation 50 H 2 + °2 + 5 % H C £ Wet oxidation 10 °2 HC£ purge 20 N2 Annealing The gas flow rates for 0^, I^, HC£ and N 2 were 1 l i t r e / m i n , 1.6 l i t r e / m i n , 50 cc/min and 1 l i t r e / m i n r e s p e c t i v e l y . The oxide obtained was 5500$ thick with an interference color of yellow-green. The pattern of the reinforcement oxide was then defined by using a standard photolithographic technique. 6.2.2 Gate oxide preparation The preparation of the gate oxide i s shown i n steps 4 and 5 of F i g . 6.4. Before oxidation, the wafers were cleaned again by using the RCA cleaning process. The oxidation temperature was 1100°C. Two types of gate oxide were prepared as follows: HC£ oxide Time (minute) Ambient Description 2 °2 Heat up 29 HC£ + 0 2 HC£ oxidation 5 °2 HC£ purge 20 Annealing 71 Dry oxide Time (minute) Ambient Description 36 °2 Oxidation 20 N2 Annealing The gas flow rates were the same as i n the reinforcement oxide preparation. The oxide obtained was 1000>! thick with an interference color of r e d - v i o l e t . 6.2.3 M e t a l l i z a t i o n The preparation of an ohmic contact on the back of the wafer and two aluminum gate layers on the front of the wafer i s shown i n steps 6 - 1 2 of F i g . 6.4. The m e t a l l i z a t i o n s were prepared by using an E-gun evaporator. The thick aluminum contact pad was f i r s t prepared by evaporating an aluminum layer of about 5000A1 on the front side of the wafer. The a l u -minum layer was then covered by photoresist and the SiO^ on the back of the wafer was removed by using a buffered HF etchant. A f t e r s t r i p p i n g o f f the photoresist, an aluminum layer of about 1 um was evaporated on to the back side of the wafer to form the ohmic contact. The wafer was then annealed at 470°C i n a N 2 ambient for 30 minutes. This i s to s i n t e r the ohmic contact and also to anneal out the E-gun induced r a d i a t i o n damage at the oxide-semiconductor i n t e r f a c e . A photolithographic process was then applied to the aluminum on the front side of the wafer to define the pattern of the gate contact pad. A t h i n aluminum layer of about 160X was evaporated on to the front side of the wafer and a photolithographic process was applied to t h i s t h i n aluminum layer to define the pattern of the semitransparent gate. 72 The p i c t u r e of a f i n i s h e d sample i s shown i n F i g . 6.5. The s i z e 2 of a unit c e l l , which consists of 22 MOS capacitors, measures 1 X 1 cm . 73 F i g . 6.5 Photograph of a f i n i s h e d sample. 74 Chapter 7 EXPERIMENTAL RESULTS 7.1 Imaging with the l a s e r scanner The imaging of MOS samples and defect s i t e s with the l a s e r scanner i s demonstrated by a s e r i e s of pic t u r e s and maps i n t h i s s e c t i o n . F i g . 7.1 shows a r e f l e c t i v i t y map of Sample 001-4311 which may be compared with the photograph of the same sample i n F i g . 7.2(a). On the photograph, stains are observed on the upper r i g h t corner and i n the middle of the r i g h t edge on the aluminum gate of Sample 001-4311. A pinhole i s also v i s i b l e as a dark spot on the upper area. The dark brown area i n the background i s the t h i n Si02« The green area i s the thick Si02« The s p e c t r a l response of the photographic f i l m r e s u l t s i n a brownish color i n the p i c t u r e . The actual color of both the t h i n and thick aluminum are m e t a l l i c . The color of the t h i n S1O2 i s r e d - v i o l e t . The color of the thick SiO 2 i s green. On the r e f l e c t i v i t y map of Sample 001-4311 of F i g . 7.1, the regions of the t h i n oxide, of the thick oxide and of the aluminum gate are c l e a r l y distinguished. The stains and the pinhole are observed as a drop i n r e f l e c t i v i t y at the corres-ponding p o s i t i o n s . Sample imaging by r e f l e c t i v i t y map and also by photocurrent map i s demonstrated by a second example. F i g 7.2(b) shows the photograph of Sample 005-3332. There are two scratches on i t s surface. A l i g h t but curved scratch i s on the l e f t and a wider but almost s t r a i g h t scratch i s on the r i g h t . A s t a i n which i s bearly v i s i b l e i s also observed on the sample. The r e f l e c t i v i t y map of the same sample i s shown i n F i g . 7.3. The two scratches are well imaged i n r e f l e c t i v i t y scanning. The wider F i g . 7.1 R e f l e c t i v i t y map of Sample 001-4311. 7 6 i 1 200um F i g . 7.2(a) Sample 001-4311 ' '200um F i g . 7.2(b) Sample 005-3332 200 100 £ 0 0 8 0 0 M I C R O N S F i g . 7.3 R e f l e c t i v i t y map of Sample 005-3332. 1 0 0 2 0 0 R E F L E C T I V I T Y MAP R U N N Q « 0 0 S 3 3 3 2 A 3 78 scratch was then scanned at a higher magnification ( i . e . smaller step s i z e ) . The corresponding r e f l e c t i v i t y map and photocurrent map are shown i n F i g . 7.4 and F i g . 7.5. The scratch and the edge of the aluminum gate, which are j u s t barely v i s i b l e on the photocurrent map of F i g . 7.5, are shown c l e a r l y on the r e f l e c t i v i t y map of F i g . 7.4. However under c e r t a i n circumstance the photocurrent imaging has better revealing a b i l i t y than the r e f l e c t i v i t y imaging. F i g . 7.6 shows a magnified o p t i c a l microscope p i c t u r e of the s t a i n near the l i g h t scratch. F i g . 7.7 shows the r e f l e c t i v i t y scan of the same area. The scratch i s c l e a r l y v i s i b l e . The s t a i n however i s d i f f i c u l t to image due to the small v a r i a t i o n i n contrast. Only the c e n t r a l dark spot of the s t a i n i s observed as a drop i n r e f l e c t i v i t y i n the r e f l e c t i v i t y map of F i g . 7.7. The sample was then subjected to a p o s i t i v e bias-temperature stress treatment at +19.2 V and 118°C for 15 minutes. The photocurrent map of the same area was then taken and shown i n F i g . 7.8. The photocurrent was not uniform. A higher photocurrent was observed at the p o s i t i o n of the s t a i n . The shape and the si z e of the photocurrent enhancement region c l e a r l y resembled that of the s t a i n . As already noted the photocurrent was studied i n the range of —8 —10 10 to 10 ampere. Although extreme care had been taken i n s h i e l d i n g to minimize noise pick up and i n wiring to prevent the formation of ground loops, the e f f e c t s of noise were s t i l l v i s i b l e . F i g . 7.9 shows a photo-current map with background noise. This was the t y p i c a l background noise -9 obtained when higher current s e n s i t i v i t y (<L0 ) was used. A more e f f e c t i v e way of eliminating the background noise i s by means of software. F i g . 7.10 shows a software processed map of F i g . 7.9 with the background noise re-moved. 5 6 0 M I C R O N S F i g . 7.5 Photocurrent map of the large scratch on Sample 005-3332. oo o 25um 1 1 7.6 Magnified picture of the s t a i n on Sample 005-3332. MICRONS F i g . 7.7 R e f l e c t i v i t y map of the s t a i n on Sample 005-3332. co R M P S X 10 EXP C-3 D 3-a _ l . s J 200 120 MICRONS i i 1 1G0 200 S I P M A P RUN NO:00S 3332 B3 F i g . 7 . 8 P h o t o c u r r e n t map o f t h e s t a i n o n S a m p l e 0 0 5 - 3 3 3 2 . T h e map wa s t a k e n a f t e r s t r e s s i n g t h e s a m p l e a t 1 1 8 ° C , + 1 9 . 2 V f o r 1 h r . oo LO AMPS X10EXPC-9 D 1 . 0 0. s 8 0 0 R P 5 3332 A3 MICRONS Fig. 7.9 Photocurrent map of Sample 005-3332, oo F i g . 7.10 Photocurrent map of Sample 005-3332 with the background noise removed by the computer software. oo 86 7.2 The e f f e c t s of Na' on scanning i n t e r n a l photoemission 7.2.1 Na + contamination a f t e r aluminum gate deposition In F i g . 7.6, the s t a i n on the sample was produced by an accidental contamination. The a p p l i c a t i o n of a p o s i t i v e bias-temperature stress t r e a t -ment resulted i n an increase i n photocurrent at the p o s i t i o n of the s t a i n . The same area was also examined by applying a negative bias-temperature stress treatment to the sample for 1 hour at 118°C, -19.2V. The photocurrent was reduced to an immeasurably small value. This suggested that the enhancement i n photocurrent was due to i o n i c motion. We would postulate that mobile a l k a l i metal ions were introduced by an accidental contamination during or a f t e r metal gate deposition. A p o s i t i v e bias-temperature stress treatment drove the p o s i t i v e l y charged ions from the Al-SiO^ i n t e r f a c e to the S i - S i C ^ i n t e r f a c e and caused a photocurrent enhancement. The subsequent a p p l i c a t i o n of a negative bias-temperature stress treatment removed the mobile ions from the Si-SiO^ i n t e r f a c e . The photocurrent was then returned to a low value. To confirm that the above observed e f f e c t was caused by the mobile a l k a l i metal ions. Sample 005-3332 was i n t e n t i o n a l l y contaminated. This -3 was done by applying a few drops of 10 N NaCJl so l u t i o n on top of the sample. The amount of the s o l u t i o n was j u s t enough to cover Sample 005-3332 so that the rest of wafer would not be contaminated. The NaC& solu t i o n was l e f t on the sample for 5 minutes and then blown o f f with N 2 gas. A p o s i t i v e bias-temperature stress treatment of 118°C, +19.2V was then applied to the sample for 1 hour. New photocurrent peaks were found a f t e r t h i s contamination procedure. Sample 005-3332 was studied by applying 3 cycles of p o s i t i v e bias-temperature stress treatment followed by a 87 negative bias-temperature stress treatment. I t was found that a negative bias-temperature stress treatment always removed a l l the photocurrent peaks while a p o s i t i v e bias-temperature stress treatment brought them back at exactly the same l o c a t i o n . F i g . 7.11 and F i g . 7.12 show the t y p i c a l photocurrent maps a f t e r a negative and a p o s i t i v e bias-temperature stress treatment respectively. Photocurrent peaks A, B and C were observed before the sample was contaminated by NaC£ s o l u t i o n . They had res u l t e d from an accidental contamination. The other peaks were caused by an i n t e n t i o n a l contamination. Peak A corresponded to" the s t a i n of F i g . 7.6. O p t i c a l microscope examination also showed that peak B coincided with a s t a i n . However no s t a i n was observed at the p o s i t i o n of peak C with a magnification up to X650. The peaks produced by an acciden-t a l contamination and by an i n t e n t i o n a l contamination could both be removed by a negative bias-temperature stress treatment. Na + transfer a f t e r aluminum gate deposition i s further demonstrated with Sample 005-3311. Sample 005-3311 was f i r s t contaminated by placing a -4 few drops of 10 N NaC£ s o l u t i o n on the sample. The quantity of s o l u t i o n was j u s t enough to cover the aluminum gate. The s o l u t i o n was l e f t on the sample for 10 minutes and then blown o f f with gas. A one hour p o s i t i v e bias-temperature stress treatment of 118°C, +19.2V was then applied to the sample to drive the Na + to the Si-Si02 i n t e r f a c e . A scanning photocurrent map of the sample was taken and i s shown i n F i g . 7.13(a). The map shows a -9 photocurrent peak of about 0.3 X 10 ampere on a background current of -9 0.1 X 10 ampere. -3 The sample was then contaminated again by using enough 10 N NaC£ s o l u t i o n to cover the gate area. The so l u t i o n was l e f t on the sample 200 100 600 800 1000 M I C R O N S F i g . - -7.1.1 The e f f e c t of negative bias-temperature stress treatment on the photocurrent map. The stress conitions were 118°C, -19.2V, Ihr. The sample was 005-3332. oo 00 AMPS X 10 EXP C-3 D 1.0 0 . 5 .800 S I P M R P RUN NQ:005 3332 R3 100 600 MICRONS F i g . 7.12 The ef f e c t of p o s i t i v e bias-temperature stress treatment on the photocurrent map. The stress conditions were 118°C, +19.2V, 1 hr. The sample was 005-3332. oo AMPS X10EXPC-3 3 1.0 _ 0 . 5 2 0 0 2 0 0 1 0 0 £ 0 0 MICRONS 8 0 0 1 0 0 0 S I P M A P RUN NO:00S 3311 02 -4, F i g . 7.13(a) Photocurrent map of a portion of Sample 005-3311 which was contaminated by 10 N NaC£ solution. The map was obtained a f t e r stressing the sample at 118 C, +19.2V for 1 hr. • • i i i i i i i i i i J 120 2 1 0 3 8 0 1 8 0 £ 0 0 7 2 0 8 1 0 3 8 0 1 0 8 0 1 2 0 0 MICRONS -3 F i g . 7.13(b) Photocurrent map of a portion of Sample 005-3311 which was contaminated by 10 N NaC£ soltuion. The map was obtained a f t e r stressing the sample at 118 C, +19.2V for 1 hr. 91 for 5 minutes and then blown o f f . A one hour bias-temperature stress t r e a t -ment of +19.2 V, 118°C was again applied to the sample. A scanning i n t e r n a l photocurrent map of the sample was taken ( F i g . 7.13(b)). The photocurrent was observed to increase non-uniformly. In some areas, the photocurrent was increased by more than an order of magnitude. I t should be noted that the current scale of Fig. 7.13(b) i s three times that of F i g . 7.13(a). O p t i c a l microscope examination of the sample did not reveal any difference between the high photocurrent area and the low photocurrent areas. The same s i t u a t i o n was also observed on Sample 005-3322 of F i g . 7.2(b) and on other samples. The picture i n F i g . 7.2(b) was taken a f t e r NaCl sol u t i o n contamination. The region of high photocurrent (Fig. 7.10) shows no diffe r e n c e from the neigh-bouring area. It should be noted that i t i s not always easy to compare the positions of p h y s i c a l defects recorded on.a two dimensional microscope picture with the positions of high photocurrent regions indicated on a three dimensional map. This i s due to the d i f f i c u l t y i n fi n d i n g the h o r i -zontal coordinates of some of the high photocurrent regions. For instance, Fig. 7.14 shows the photocurrent map of Sample 005-3311. The h o r i z o n t a l coordinate of the summit of peak AA cannot be read d i r e c t l y from the map but has to be estimated. _Also the high photocurrent i n the foreground blocks the portion of the photocurrent map behind i t . This r e s u l t s i n a portion of the photocurrent map which cannot be read. The p o s i t i o n of a high photocurrent region can be better i d e n t i f i e d ; by marking t h e i r projection on the h o r i z o n t a l plane. The shadow map i s used for t h i s purpose. In the shadow map, the regions of photocurrent higher than a c e r t a i n l e v e l are projected onto the h o r i z o n t a l plane and drawn as a shaded area. F i g . 7.15 shows a shadow map corresponding to the photocurrent map of Fig. 7.14. F i g . 7.14 Photocurrent map of Sample 005-3311 which was contaminated by a 10 N NaC£ s o l u t i o n . The map was obtained a f t e r stressing the sample at 118°C. +19.2V for 1 hr. Shadow map of Sample 005-3311. The shadow map i s redrawn from the photocurrent map of f i g . 7.14. The shaded area corresponds to the region in which the photocurrent is higher than 1.2 X 10 amp. OJ 94 The shaded area corresponds to the region of photocurrent larger than 1.2 X -9 10 ampere (40% of f u l l scale reading). The threshold 40% i s selected a r b i t a r i l y . The shape of high photocurrent region and the p o s i t i o n of the photocurrent peak are better demonstrated on a shadow map. The usefulness of the shadow map w i l l be further demonstrated i n Section 7.4. Internal photoemission current was measured with a bias voltage applied across the sample. The p o s i t i v e side of the voltage was connected to the gate. The e f f e c t of t h i s bias voltage on the photocurrent map i s shown i n F i g . 7.16 to F i g . 7.18. Note that the current scales on the three maps are d i f f e r e n t . Generally speaking the photocurrent increases with i n -creasing bias voltage. There were some changes i n the contrast of the photo-current map but the basic pattern remained the same. The use of a higher bias voltage to achieve a higher photocurrent i s l i m i t e d by sample breakdown and current i n s t a b i l i t y at high f i e l d s . The spikes X i n F i g . 7.17 and Y i n Fi g . 7.18 were due to s e l f - h e a l i n g breakdown i n the sample. In most of our experiments, a bias f i e l d strength of about 2 MV/cm ^ was used. This i s also the turn-on f i e l d strength as shown i n F i g . 3.11 of Chapter 3. The re s u l t of th i s section can be summarized as followings: ( i ) The 160X aluminum i s not an e f f e c t i v e b a r r i e r to sodium contam-_3 i n a t i o n . Contamination with NaC& so l u t i o n of about 10 N resulted i n very pronounced increase i n photocurrent. ( i i ) For some unknown reasons, the sodium which i s transferred i s always d i s t r i b u t e d i n a non-uniform fashion, ( i i i ) O p t i c a l microscope examination of the i n t e n t i o n a l l y contaminated samples did not reveal any difference between the regions of high photocurrent and the res t of the sample, (iv) A c c i d e n t i a l contamination during wafer processing would produce RMPS X10EXPC-8 D 1. B 0 . 5 720 S I P M A P RUN NQ:005 3311 05 MICRONS F i g . 7.16 Photocurrent map of Sample 005-3311 measured with a bias voltage of +19.2V across the sample. Ln SELF-HEALING BREAKDOWN AMPS X 10 EXP C-8 3 3 . 0 _ 1.5 M R P 005 3311 0G M I C R O N S F i g . 7.17 Photocurrent map of Sample 005-3311 measured with a bias voltage of +28.9V' across the sample. SELF-HEALING BREAKDOWN 210 180 720 380 1200 M I C R O N S F i g . 7.18 Photocurrent map of Sample 005-3311 measured with a bias voltage of +38.4V across the sample. VO 98 l o c a l i z e d photocurrent enhancement, s i m i l a r to those produced by 10~ 3 N NaC£ so l u t i o n , (v) Accidental contamination may r e s u l t i n stains on the sample, (vi) Photocurrent enhancement a f t e r a p o s i t i v e bias-temperature stress treatment can be removed by a negative bias-temperature stress treatment. This applies to both i n t e n t i o n a l l y and, a c c i d e n t a l l y contaminated samples. 7.2.2 Na + contamination before aluminum gate deposition A wafer without the thi n aluminum gate was fabricated. This included the f a b r i c a t i o n steps up to 10 as shown i n F i g . 6.4 of Chapter -3 6. Contamination was done by immersing the wafer i n 140 ml of 10 N NaC£ < solut i o n at 28°C. The wafer was placed at the bottom of a 80 mm diameter X 40 mm high pyrex dish with the SiO^ facing upward. No ul t r a s o n i c a g i t a t i o n was used. The t o t a l immersion time was 5 minutes. The sample was blown dry with N 2 gas and the t h i n aluminum gate then deposited. A sequence of p o s i t i v e bias-temperature stress treatments was again applied to the samples to study the Na + transfer process and the t y p i c a l r e s u l t i s presented i n t h i s section. For Sample 001-2331, eleven scanning i n t e r n a l photocurrent maps were taken with a t o t a l accumulated stress time of 27 hours. The p o s i t i v e bias-temperature stress were done at 118°C with a f i e l d of 0.9 MV/cm. Six of these maps are shown i n Figures 7.19 to 7.24. F i g . 7.19 shows the SIP map before p o s i t i v e bias-temperature stress treatment. This i s the., t y p i c a l ' SIP map for a. sample before d r i f t . The photocurrent was small over the entire sample. The photocurrent was observed to increase with the stress time under p o s i t i v e bias-temperature 800 A M P S X 1 0 E X P C - 9 D B.S J M R P 00 1 2331 0 1 200 100 G00 M I C R O N S 800 1000 F i g . 7.19 Photocurrent map of Sample 001-2331 obtained before a p o s i t i v e bias-temperature stress treatment was applied to the sample. AMPS X10EXPC-9 D 1.0 ^ 0 . S 8 0 0 S I P M R P RUN NO:001 2331 02 1000 MICRONS F i g . 7.20 Photocurrent map of Sample 001-2331 obtained a f t e r s t r e s s i n g the sample at 118 C, +9.0 V for 30 min. o o F i g . 7.21 Photocurrent map of Sample 001-2331 obtained a f t e r s t r e s s i n g the sample at 118 C, +9.0 V for 1 hr. F i g . .7.22 Photocurrent map of Sample 001-2331 obtained a f t e r stressing, the sample at 118°C, +9.0 V for 3 hr. 40 min. i—• o F i g . 7.23 Photocurrent map of Sample 001-2332 obtained a f t e r s t r e s s i n g the sample at 118°C, +9.0 V for 15 hr. 10 min. o OJ F i g . 7.24 Photocurrent map of Sample 001-2331 obtained a f t e r s t r e s s i n g the sample at 118°C, +9.0 V for 27 hr. 10 min. o 4> 105 treatment. F i g . 7.19 to F i g . 7.24 show that both the peaks A, B, C and the background photocurrent were increased with the stress time. Capacitance-voltage measurement also revealed a continuous s h i f t of the flat-band voltage to more negative values as stress time was increased. The major difference between photocurrent maps of the samples contaminated before aluminum de-p o s i t i o n and the samples contaminated a f t e r aluminum deposition was i n the background photocurrent. Samples contaminated before aluminum gate deposition usually showed a f a i r l y high and uniform background photocurrent a f t e r p o s i t i v e bias-temperature stress treatment. Under the conditions of our experiment, the background photocurrent did not develop into patches of high l e v e l s even a f t e r prolonged p o s i t i v e bias-temperature stress treatment. Segregation of Na + ions into small patches of high density has not been observed. The photocurrent peaks on the maps were observed at the very beginning stage of the Na + d r i f t process and also t h e i r p o s i t i o n s coincided with stains on the sample. I t i s more l i k e l y that these photocurrent peaks were caused by l o c a l i z e d a c c i d e n t a l contamination rather than a r e s u l t of Na + segregation. 7.3 R e v e r s i b i l i t y of photocurrent peaks under bias-temperature stress  treatment If photocurrent enhancement a f t e r a p o s i t i v e bias-temperature stress treatment i s produced by a process other than mobile ion migration, f o r example c r y s t a l l i z a t i o n as suggested by Osburn (1974), the photocurrent enhancement should not be removable by a negative bias-temperature stress treatment. Among the samples fabricated i n t h i s laboratory, i r r e v e r s i b l e photocurrent peak was observed only on one sample. F i g . 7.25(a) shows the photocurrent map of Sample 0221 which was obtained a f t e r s t r e s s i n g the sample at 118°C, -29.2V for % hour. Photocurrent peaks were observed on the map. 3 1 100 200 300 (c) F i g . 7.25 R e v e r s i b i l i t y of photocurrent peaks studied under bias-temperature (B-T) stress treatment. The photocurrent maps were obtained(a) after 30 min negative B-T stress treatment of 118 C, -29.2V, (b) aft e r 35 min p o s i t i v e B-T stress treatment of 118°C, +29.2V, (c) a f t e r 30 min negative B-T stress treatment of 118 C, -29.2V. o ON 107 A p o s i t i v e bias-temperature stress treatment of 118 C, +29.2V was then applied to the sample for 35 minutes. The photocurrent map a f t e r the stress treatment i s shown i n F i g . 7.25(b). A photocurrent enhancement region i n a d d i t i o n to the photocurrent peaks was observed on the map. A negative bias-temperature stress treatment was then applied to the sample for % hour at 118°C, -29.2V. The photocurrent map a f t e r the stress treatment i s shown i n F i g . 7.25(c). The negative bias-temperature stress treatment removed the photocurrent enhancement region but not the photocurrent peaks. The cause of the photocurrent peaks was not i d e n t i f i e d . I t may be damage s i t e s at the Si-Si02 i n t e r f a c e . 7.4 C o r r e l a t i o n between mobile ion induced photocurrent peaks and s e l f - healing breakdown i n MOS structures It was observed repeatedly i n our experiments that the positions of the photocurrent peak coincided with the spots at which s e l f - h e a l i n g breakdown occurred. A t y p i c a l example i s given i n t h i s section. F i g . 7.26 and F i g . 7.27 show the photocurrent and the r e f l e c t i v i t y maps of Sample 004-1218 before bias-temperature stress treatment r e s p e c t i v e l y . The photocurrent throughout the sample i s low. The r e f l e c t i v i t y map shows three defect s i t e s A, B and C on the sample. Defect s i t e A corresponds to a scratch while defect s i t e s B and C correspond to two pinholes. Sample 004-1218 was contaminated a c c i d e n t a l l y during wafer processing. A comparatively large number of stains s i m i l a r to the one shown i n F i g . 7.6 were observed on the sample. Again, these stains were not revealed by the r e f l e c t i v i t y map because of t h e i r low contrast. The sample was then subjected to a p o s i t i v e bias-temperature stress o VO 110 treatment of one hour at 118°C, +19.2V. Capacitance-voltage (C-V) curves were recorded before and a f t e r the p o s i t i v e bias-temperature stress treatment. The s h i f t i n flat-band voltage of the C-V curves indicated that the mobile 12 2 ion density at the Si - S i C ^ i n t e r f a c e was approximately 1.6 X 10 /cm . The sample was then scanned again. The photocurrent and r e f l e c t i v i t y maps are shown i n F i g . 7.28 and F i g . 7.29 re s p e c t i v e l y . Very d i s t i n c t photo-current peaks were observed on the photocurrent map of F i g . 7.28. The r e f l e c t i v i t y map of F i g . 7.29 shows l i t t l e change a f t e r the bias-temperature stress treatment. The sample was removed from the scanner for microscope examination a f t e r each scan. This resulted i n sample displacement as shown i n F i g . 7.27 and F i g . 7.29. The sample was then subjected to a series of s e l f - h e a l i n g break-down tests u n t i l a breakdown f i e l d strength of about 5 MV/cm was reached. The t o t a l number of breakdowns accumulated was 140. The f i e l d strength of 5 MV/cm was so chosen that i t was well below the i n t r i n s i c breakdown f i e l d strength of SiC^ but high enough to produce breakdown at the weak spots. A picture of the sample a f t e r s e l f - h e a l i n g breakdown test i s shown i n F i g . 7.30. The sample was then scanned again. The re f l e c t i v i t y " m a p a f t e r breakdown test i s shown i n F i g . 7.31. The depressions on the r e f l e c t i v i t y map correspond to the s e l f - h e a l i n g breakdown spots. Since i t i s d i f f i c u l t to compare the positi o n s of the photocurrent peaks of Fig.7.28 with the positi o n s of the s e l f - h e a l i n g breakdown spots of Fig.7.31 i n three dimensional mappings, shadow maps are given to f a c i l i t a t e the comparison. F i g . 7.32(a) and F i g . 7.32(b) show the shadow maps of F i g . 7.28 and F i g . 7.31 re s p e c t i v e l y . The shaded areas of F i g . 7.32(a) correspond to the positi o n s of the photocurrent peaks. The shaded areas of F i g . 7.32(b) correspond to the s e l f - h e a l i n g break-MICRONS Fig. 7.28 Photocurrent map of Sample 004-1218 obtained after stressing the sample at 118°C, +19.2V for 1 hr. I I I I I I I I I I I l | I I I I I I I I l | I I I I I I I I I) I I I I I I I I I I I I Ij 11 l I I I i n | . . . . . . . . . . . . , 200 400 600 800 1000 MICRONS F i g . 7.29 R e f l e c t i v i t y map of sample 004-1218 obtained a f t e r s t r e s s i n g the sample at 118°C, +19.2V for 1 hr. F i g . 7.31 R e f l e c t i v i t y map of Sample 004-1218 obtained a f t e r s e l f - h e a l i n g breakdown t e s t . 4^  F i g . 7.32(a) Photocurrent map of Sample 004-1218 a f t e r Na d r i f t (Fig.. 7.28) i s redrawn as a shadow map. The photocurrent i n the shaded areas were higher than 15% of the f u l l scale current. • i i ' ••• 1 'i • l i i i i i ^ n j i u M M M i M i M M M i ; ; ; : ; ; ; : : ; ; : ; ; : ; ; ; : ; ; : i F i g . 7.32(b) R e f l e c t i v i t y map of Sample 004-1218 a f t e r breakdown (Fig. 7.31) \ i s redrawn as a shadow map. The shaded areas correspond to the Si0„ and the positions of s e l f - h e a l i n g breakdown . 116 down spots and the areas not covered by aluminum. A good c o r r e l a t i o n i s observed between the positions of the high photocurrent peaks and the posi t i o n s of the s e l f - h e a l i n g breakdown spots. F i g . 7.33 shows the photocurrent map of the sample a f t e r breakdown tes t . A number of photocurrent peaks disappeared as compared to the photo-current map before the breakdown test ( F i g . 7.28). This also substantiated that s e l f - h e a l i n g breakdowns had occurred at the posi t i o n s of the photocurrent peak. The removal of the gate metal during breakdown e f f e c t i v e l y destroyed the portion of the sample at these p o s i t i o n s . There were some photocurrent peaks remaining on the sample as shown i n F i g . 7.33. These areas correspond to the portion of the sample which required a f i e l d strength higher than 5 MV/cm to breakdown. 7.5 Scanning i n t e r n a l photoemission and s e l f - h e a l i n g breakdown studies  of HC£-grown s i l i c o n dioxide films I t was observed i n MOS samples with HC£-grown oxide that the p o s i t i v e mobile ion density accumulated at the Si-Si02 i n t e r f a c e a f t e r a p o s i t i v e b i a s -temperature stress treatment,as derived from the flat-band voltage s h i f t of the capacitance-voltage (C-V) curves, was less than the mobile ion density that d r i f t e d across the SiO 2 layer as measured by the i n t e g r a t i n g electrometer (Q-t) method (Kriegler et a l . , 1972; van der Meulen et a l . , 1975; Singh and Balk, 1978; Rohatgi et a l . , 1977/9a,b). Some of the mobile ions which d r i f t e d towards the S i - S K ^ i n t e r f a c e during the p o s i t i v e bias-temperature stress treatment were e l e c t r i c a l l y n e u t r a l i z e d . MOS specimens were prepared with the gate oxide grown at 1100°C i n the presence of 5 v o l . % ECU. This oxidation process was often used i n our laboratory and i s described i n Section 6.2.2. The oxide was evaluated by 200 400 600 800 1000 MICRONS F i g . 7.33 Photocurrent map of Sample 004-1218 obtained a f t e r s e l f - h e a l i n g breakdown t e s t . 118 using scanning i n t e r n a l photoemission and s e l f - h e a l i n g breakdown techniques. F i g . 7.34 shows the scanning i n t e r n a l photoemission map of Sample 007-4116. The photocurrent throughout the map was low. A p o s i t i v e b i a s -temperature stress treatment of 118°C, +19.2V was then applied to the sample 12 -2 for 10 minutes so that 4.25 X 10 cm mobile ions were d r i f t e d towards the Si-SiO^ i n t e r f a c e as measured by the int e g r a t i n g electrometer method. Capacitance-voltage curves were also taken before and a f t e r the p o s i t i v e bias-temperature stress treatment. The s h i f t i n flat-band voltage of the C-V curves indicated that the charged mobile ion density at the Si - S i C ^ i n t e r f a c e 11 -2 was approximately 2.4 X 10 cm . This indicated that a portion of the mobile ions which d r i f t e d towards the Si-Si02 i n t e r f a c e was e l e c t r i c a l l y n e u t ralized. The photocurrent map which was obtained a f t e r the p o s i t i v e bias-temperature stress treatment i s shown i n F i g . 7.35. The photocurrent throughout the map was low. The amount of mobile ions which remained e l e c t r i c a l l y unneutralized at the Si-Si02 i n t e r f a c e did not produce photo-current peaks on the map. The sample was then subjected to a s e l f - h e a l i n g breakdown test and was broken down for four times. The breakdown f i e l d strengths of the sample were 6 MV/cm for the f i r s t event breakdown and increased to 8.7 MV/cm i n the fourth breakdown. The rapid increase of breakdown strength with the number of breakdown event indicated that the number of weak spots on the specimen was small. An a d d i t i o n a l p o s i t i v e bias-temperature stress treatment of 118°C, +19.2V was then applied to the sample for 50 minutes. The t o t a l charge accumulated at the Si-Si02 i n t e r f a c e a f t e r the two bias-temperature stress 12 -2 treatments was about 1.25 X 10 cm (from C-V measurement) while the t o t a l 13 -2 mobile ions d r i f t e d towards the S i - S i 0 o i n t e r f a c e were 1.33 X 10 cm AMPS X 1 0 E X P C - 3 D 1.0 _ • - s 210 180 720 . 3G0 MICRONS F i g . 7.34 Photocurrent map of Sample 007-4116. g=/ SIP MAP ==/ RUN NO: 007 «rlie 0 1 1200 M R P 0 0 7 0 2 210 180 720 M I C R O N S 360 1200 12 -2 F i g . 7.35 Photocurrent map of Sample 007-4116 obtained a f t e r 4.25 X 10 cm mobile ions were d r i f t e d to the Si-Si02 i n t e r f a c e . o 13 -2 F i g . 7.36 Photocurrent map of Sample 007-4116 obtained a f t e r 1.33 X 10 cm mobile ions were d r i f t e d to the Si-SiO^ i n t e r f a c e . ho '720 A R B I T R A R Y U N I T S 1.0 _ M A P R U N N Q : 0 0 7 4 1 1 G 0 4 240 480 720 M I C R O N S 360 F i g . 7.38 R e f l e c t i v i t y map of Sample 007-4116 obtained a f t e r a s e l f - h e a l i n g breakdown t e s t . 124 (from Q-t measurement). The photocurrent map which was obtained a f t e r the a d d i t i o n a l 50 minutes p o s i t i v e bias-temperature stress treatment i s shown i n F i g . 7.36. Photocurrent peaks were observed along the periphery of the t h i n aluminum gate. Sample 007-4116 was then subjected to a ser i e s of s e l f - h e a l i n g breakdown te s t s . The photograph and the r e f l e c t i v i t y map of the specimen a f t e r the breakdown test are shown i n F i g . 7.37 and F i g . 7.38 r e s p e c t i v e l y . S e l f - h e a l i n g breakdowns occurred again at the p o s i t i o n of the photocurrent peaks. I t should be noted that f i v e other samples on the same wafer were studied and photocurrent peaks were observed along the periphery of the t h i n aluminum gate of a l l these samples. I t was believed that the samples were contaminated e i t h e r during the aluminum etching step or from prolonged exposure of the sample surface to room atmosphere. Mobile ions, which were i n i t i a l l y on the sample surface, d i f f u s e d along the periphery of the aluminum gate into the sample and resulted i n photocurrent peaks. The scanning i n t e r n a l photoemission map gave d i r e c t evidence of mobile ion transfer along the metal gate periphery. The existence of photocurrent peaks and weak spot breakdowns indicated that unneutralized mobile ions would degrade the performance of MOS structure. More chlorine should be incorporated into the HC£ oxide for mobile ion passivation. 125 Chapter 8 CONCLUSIONS The purpose of th i s work was to set up a computer-controlled laser scanning i n t e r n a l photoemission experiment and a s e l f - h e a l i n g breakdown experiment and to apply these two experimental techniques to investigate the ef f e c t s of mobile ion contamination on MOS structures. In p a r t i c u l a r , the imaging of defect s i t e s on MOS structures by using the las e r scanner, the contamination of MOS structures by using an aqueous NaCSJ, s o l u t i o n , the cor-r e l a t i o n between mobile ion contamination and breakdown, the n e u t r a l i z a t i o n of mobile ions i n HC£-grown oxide and the transfer of mobile ions along the metal gate periphery were examined. In addition,the i n t e r n a l photoemission process i n MOS structures was modelled t h e o r e t i c a l l y . The scanning i n t e r n a l photoemission method was proved to be a powerful non-destructive technique i n detecting l o c a l i z e d mobile ion d i s t r i b u t i o n . I t was also a useful t o o l for semiconductor process evaluation. The contributions which were made to the subject may be summarized as follows: (1) A t h e o r e t i c a l model of the i n t e r n a l photoemission process i n MOS structures was developed. The model predicted q u a n t i t a t i v e l y the dependence of the quantum y i e l d on the semiconductor doping, on the incident photon energy, on the band bending at the semiconductor surface and on the applied f i e l d . I t was shown that a large downward band bending i n a heavily doped p-type semiconductor may s h i f t the photoemis-sion threshold to a lower value. When the amount of band bending within the electron escape depth was small, the flat-band approximation can be used. The f i e l d dependence of the quantum y i e l d at hv = 3.81 eV 126 (He-Cd laser radiation) as calculated by the flat-band model indicated the existence of a turn on f i e l d strength of about 2 MV/cm. A computer controlled laser scanner was designed. The problem of the UV laser focussing was solved by analysing the radiation reflected from the specimen surface. The special focussing set up which was designed provided a possible means of automatic focussing. A method was also devised to measure the light spot size with the sample in s i t u . An automatic self-healing breakdown tester was designed. The counter control feature of the tester allowed the operator to breakdown a MOS sample for a desired number of times. This gave the operator more control of the self-healing breakdown test, and the danger of having S both weak spot and intrinsic breakdown occurred on the same sample was eliminated. The defect imaging capability of the laser scanner was demonstrated. It was shown that the r e f l e c t i v i t y imaging and the photocurrent imaging are complementary techniques in revealing defect sites on the MOS structures. Mobile ion transfer i n MOS structures was investigated. It was observed that contamination by NaC£ solution after metal gate deposition always resulted in very localized Na + distribution. The reason for this was not known. The possibility of contamination after metal gate fabrication implies that mobile ions are potential hazards to thin metal gate MOS devices. A correlation between photocurrent peaks induced by a localized dis-tribution of mobile ions and weak spot breakdowns was observed. Transfer of the mobile ions along the metal gate periphery was observed. Scanning internal photoemission was proved to be useful in locating the 127 positions where the mobile ions entered the sample. 8.1 Suggestions for further research A t h e o r e t i c a l model for i n t e r n a l photoemission i n MOS structures was developed i n t h i s work. However, some of the predictions of the theory, for example, the dependence of the quantum y i e l d on the photon energy, on the semiconductor doping, on the band bending and on the applied f i e l d have not been tested by experiment at the present. A c o r r e l a t i o n between mobile ion induced photocurrent peaks and the spots of s e l f - h e a l i n g breakdown was observed. However, the mechanisms of the e f f e c t of mobile ions on photocurrent and breakdown are not well understood. Further experiments are required on t h i s . Peaks i n the photocurrent were observed but evidence was obtained that they were not due to mobile ion segregation as suggested by Williams and Woods (1972). However, i t may be that segregation of mobile ions only occurs beyond a c e r t a i n threshold condition. A wider set of experimental conditions should be investigated. I t was shown that the HC& oxide prepared according to the procedure of Section 6.2.2 did not contain enough chlorine f or mobile ion passivation. More work i s needed on the HC£ oxidation process to determine the threshold percentage of HC£ needed f o r mobile ion passivation under d i f f e r e n t oxidation conditions. 128 REFERENCES Buchl, K. (1970), Appl. Optics j), 113. Carnes, J.E. and Duffy, M.T. (1971), J. Appl. Phys. 42. 4350. Chang, CC. (1978), Extended Abstracts number 219, The Electrochemical Society Spring Meeting, Seattle, Washington, May 21-26, 1978. Chou, N.J. (1971), J. Electrochem. Soc. 118, 601. Dash, W.C. and Newman, R. (1955), Phys. Rev. 99, 1151. Deal, B.E., Snow, E.H. and Mead, C.A. (1966), J. Phys. Chem. Solids 27, 1873. Berglund, C.N. and Powell, R.J. (1971), J. Appl. Phys. 42, 573. DiStefano, T.H. and Eastman, D.E. (1971), Solid State Comm. 9_, 2259. DiStefano, T.H. (1971), Appl. Phys. Lett. 19, 280. DiStefano, T.H. (1973), J. Appl. Phys. 44, 527. DiStefano, T.H.and Viggiano, J.M. (1974), IBM J. Res. Develop.,18 , 94. DiStefano, T.H. (1976), J. Vac. Sci. Technol. _1_3, 856. Eldridge, J.M. and Kern, D.R. (1971), J. Electrochem. Soc. JJL8, 986. Fritzsche, C. (1967), Z. Angew. Phys. 24, 43. Gobeli, G.W. and Allen, F.G. (1962), Phys. Rev. _127, 141. Goodman. A.M. (1966), Phys. Rev. 144, 588. Goodman, A.M. (1966), Phys. Rev. 152, 785. Johannessen, J.S., Spicer, W.E. and Strausser, Y.E. (1976), J. Appl. Phys. 47, 3028. Kane, E.O. (1962), Phys. Rev. 127_, 131. Kern, W. and Puotinen, D.A. (1970), RCA Review, 31, 186. Klein, N. and Gafni, H. (1966), IEEE Trans. Electron Devices ED-13, 281. Klein, N. (1966), IEEE Trans. Electron Devices ED-13, 788. Kriegler, R.J., Chang, Y.C. and Colton, D.R. (1972), J. Electrochem. Soc. 119, 388. Madland, G.R. et a l . (1972), Integrated Circuit Technology, pp4-14, Integrated Circuit Engineering Corporation, Phoenix, Ariz.. Osburn, CM. and Raider, S.I. (1973), J . Electrochem. Soc. 120, 1369. Osburn, CM. and Chou, N.J. (1973), J . Electrochem. Soc. 120, 1377. Osburn, CM. and Ormond, D.W. (1974), L. Electrochem. Soc. 121, 1195. Peisner, J . , Aszodi, G.,Nemeth-Sallay, M. and Forgacs, G. (1976), Thin S o l i d Films 36, 251. Powell, R.J. (1969), J . Appl. Phys. 40, 5093. Powell, R.J. and Morad, M. (1978), J . Appl. Phys. 49, 2499. Raider, S.I. (1973), Appl. Phys. L e t t . 23, 34. Rohatgi, A., Butler, S.R., F e i g l , F.J., Kraner, H.W. and Jones, K.W. (1977), Appl. Phys. L e t t . 30, 104. Rohatgi, A., Butler, S.R., F e i g l , F.J., Kraner, H.W. and Jones, K.W. (1979a), J . Electrochem. Soc. 126, 143. Rohatgi, A., Butler, S.R. and F e i g l , F.J. (1979b), J . Electrochem. Soc. 126, 149. Scheer, J . J . (1960), P h i l i p s Res. Repts. 15, 584. Singh, B.R. and Balk, P. (1978), J . Electrochem. Soc. ^ 25, 453. Snow, E.H., Grove, A.S., Deal, B.E. and Sah, C T . (1965), J . Appl. Phys. 36, 1644. Spicer, W.E. (1960), J . Appl. Phys. 3J,, 2077. Spicer, W.E. (1976), C r i t . Rev. S o l i d State S c i . , 6, 317. van der Meulen, Y.J., Osburn, CM. and Z i e g l e r , J.F. (1975), J . Electrochem. Soc. 122, 284. Viswanathan, C.R. and Ogura, S. (1968), Appl. Phys. L e t t . JL2, 220. Viswanathan, C.R. and Ogura, S. (1969), Proc. IEEE 57, 1552. Williams, R. (1965), Phys. Rev. J^40, A569. Willaims, R. (1966), J . Appl. Phys. 37, 1491. Williams, R. and Woods, M. (1972), J . Appl. Phys. 43, 4142. Williams, R. and Woods, M. (1973), Appl. Phys. L e t t . 22_, 458. ' Williams, R. (1974), J . Appl. Phys. 45, 1239. Worthing, F.L. (1968), J . Electrochem. Soc. 115, 88. 130 Appendix 1 THE ELECTROSTATIC FIELD AND THE BAND BENDING AT THE SEMICONDUCTOR SURFACE Figure A . l shows the energy diagram of a MIS structure. Although, a n-type semiconductor with a depleted surface i s used for the i l l u s t r a t i o n , the discussion which follows applies also to p-type semi-conductors and surface conditions other than depletion. The dimensionless p o t e n t i a l U i s defined as U = (E_ - E.)/ kT (A.l) F 1 where E i s the Fermi l e v e l i n the semiconductor^ r E. i s the i n t r i n s i c Fermi l e v e l i n the semiconductor,, l k i s Boltzmann 's constant, T i s the absolute temperature. In the bulk of the semiconductor, U i s c a l l e d the bulk p o t e n t i a l U, . b The band bending V i s defined as V = U - U b (A. 2) At the semiconductor surface, V i s c a l l e d the surface p o t e n t i a l V g. V i s p o s i t i v e for a downward band bending and negative for an upward band bending. In a non-degenerated semiconductor, the electron density n and hole 'density p are given at every point by p = p ue , p K =n.e P (A.3) Jb ' F b i :U, n = n be , n^ = n..e b (A. 4) where n. i s the i n t r i n s i c c a r r i e r density, x n, and n, are the electron and hole densities i n the bulk of the b r b semiconductor r e s p e c t i v e l y . METAL INSULATOR SEMICONDUCTOR X=0 Al Energy band diagram of a MIS structure. (Illustrated with n-type substrate and a depleted surface.) 132 For the case of constant impurity concentration and complete i o n i z a t i o n , Poisson's equation for the surface space charge region can be written as 2 .2 = - — :( N - N + p - n ) (A.5) dX Z e-kT D A S-J. . where e g i s the p e r m i t t i v i t y of the semiconductor, e i s the e l e c t r o n i c charge. and N^ are the donor and acceptor impurity d e n s i t i e s respectively. Charge n e u t r a l i t y i n the bulk requires ND " N A - n b " P b ( A - 6 ) Therefore, 2 2 — 9 = ( n, - P K + P - n ) dX 2 e k T ^ b s - e 2 / •• . -V -V, ( n, - p, + p e - n, e ) s l sinh(U +V) 72 ( cosh II - t a n h U b > (A-7) •L b where L i s the e f f e c t i v e Debye length e kT L = /-j (A. 8) e ( n b + p b ) Integration of the Poisson's equation by using the boundary condition dV/dX = 0 at V = 0 gives /2- F(V,UJ dX dV = - -^'W (A.9) L and x = + -±_ ^  -^M-j (A. 10) s b cosh(U +V) t where F(V,U, ) = ( — - V tanh U - 1 )^ (A. 11) b cosh U, b 133 The negative sign r e f e r s to V > 0 and the p o s i t i v e r e f e r s to V < 0. The e l e c t r o s t a t i c f i e l d i n s i d e the semiconductor j u s t next to the semiconductor i n s u l a t o r i n t e r f a c e i s given by E = kT dV S e dX X = 0 kT /2 , cosh (IL + V ) „ . , T T , sh ,k ,„x = + — — ( b s' - V g tanh Ufe - 1 ) 2 (A. 12) cosh U, D E i s p o s i t i v e when directed from the semiconductor surface outward, s The e l e c t r o s t a t i c f i e l d E i n the i n s u l a t o r j u s t next to the semi-conductor-insulator i n t e r f a c e i s given by - £ s kT 72 C O S h ( U b + V s ) V R A I O N E = + _ £ i S l ^ . ( - V tanh IL - 1 )'5 ( A ' 1 3 ) e + e L cosh U, s b i b where i s the p e r m i t t i v i t y of the i n s u l a t o r . 134 Appendix 2 SOFTWARE OF THE LASER SCANNER There are four main programs used to control the l a s e r scanner, to c o l l e c t data and to present the data i n an appropriate form. A l l programs and sub-programs except LABEL were written i n a mixture of 0S8 Fortran II and SABR assembly language. The program LABEL was written i n 0S8 Fortran IV. A program MEMO was also written as a help f i l e for the convenience of user. The function of each program and sub-program i s described i n t h i s appendix. 135 MAIN PROGRAM F i l e Program Subroutines Function Name Name Galled SIP8R.SV SIP8R MVORG, STEP, A fast routine f o r pre-RANAL, DAPRTD liminary scanning. Data p r i n t out as a matrix on the VT-52 CRT terminal. SIPM.SV SIP4 MVORG, STEP, RANAL. Scanning i n t e r n a l photo-emission main program. This i s to be used together, with DMAP to obtain a 3-D map or a shadow map on the DPlcJ> incremental p l o t t e r . Data are saved on tape. DMAP.SV TMAP D3MAP, D3AXES, MSCALE, TPMAP, DAXIS, PLOTD. A routine f o r drawing a 3-D map or a shadow map on the DPl(j> p l o t t e r . LABEL.SV LABEL For l a b e l l i n g and print-ing t i t l e s on a map. MEMO.SV MEMO Memo for SIPM & SIP8R. SUBROUTINES F i l e Name Subroutine Name Function MVORG MVORG Move the sample to the desired p o s i t i o n (for s e t t i n g up o r i g i n ) . STEP STEP Step the scanning stage fo r - a desired number of steps and then hold the stage for a desired period of time. RANAL RANAL Read the voltage from a desired analog channel. DAPRTD DAPRTD Pr i n t a matrix on VT52, D3MAP D3MAP Plot a 3-dimensional map on the DP1<J) incremental p l o t t e r . D3AXES D3AXES Draw the axes for a 3-D map. MSCALE MSCALE Scale the map to f i t a p l o t t i n g paper. TPMAP TPMAP Pr i n t out a shadow map on the DPltt, DAXIS DAXIS Draw an axis. PLOTD PLOTD Perform 3 functions: (1) move p l o t t e r pen from one p o s i t i o n to another. (2) draw a l i n e between two points. (3) move the p l o t t e r pen from A to B and pl o t a point at B. 137 PROGRAM NAME: SIP4 C SCANNING INTERNAL PHOTO-EMISSION PROGRAM FOR SCANNING THE C SAMPLE AND OUTPUTTING DATA TO THE DECTAPE.THIS IS TO BE USED C TOGETHER WITH THE DMAP ROUTINE FOR PLOTTING A D3MAP OR A C TPMAP ON THE DP 10 PLOTTER. . COMMON IP, I B DIMENSION IPC209),IRC200) 5 WRITEC1,6) 6 FORMAT<• '> WRITEC1,8) 8 FORMAT?'NEW ORIGIN LOCATION') READC1,9) IOX,IOY,ID 9 FORMATC' OX='I3,' OY='I3,' OIDXY='I3) ' C LOCATING A NEW ORIGIN IOX=IOX*ID IOY=IOY*ID CALL MVORGCIOX,IOY) -WRITEC1,10) 10 FORMATC'SCAN DIMENSION') ' READC1,11) NX,NY,IDXY 11 FORMATC' NX='I3,' NY='I3,' IDXY='I3) .-WRITEC1,60) 60 FORMATC'CIRCUIT TIME CONSTANT CALCULATIONI-'> READC1,61) FV,FI,C 61 FORMATC' FSV='F5.2,' CURRENT RANGECPA)=•F6.2,' C C P F ) = ' F 5 » 1 ) CTK=FV*C/FI WRITEC1,62) CTK ' 62 FORMATC * TI ME CONSTANTCMSEC)='F6•2> READC1,13) ITK 13 FORMATC' ITKCMSEC) = 'I 3) DXY=FLOATC1DXY) -X=FLOATCNX) Y= FLOAT C NY > TK=FLOATCITK) . ' " . SCANTK=CDXY*12.5/1000.0+TK/1000.0)*X*Y/60.0 WRITEC1,14) SCANTK 14 FORMATC'SCANNING TIMECMINUTES)='F6•2) READC1,16) F I L E 16 FORMATC'DATA F I L E NAME='A6) WRITEC1,15) 15 FORMATC'START') C * * * * * I N I T I A L I Z A T I O N OF THE SCAN CALL O O P E N C ' D T A 1 « , F I L E ) MC0DE=48 J J 2 = 0 C *****SCAM THE SAMPLE DO 20 1=1,NY J,J1 = 0 C *****SCAN ONE LINE DO 2 1 J-- 1, NX JJ1=JJ1+1 ICHAN=13 CALL RANALCI PC,I CHAN) IPCJ J 1 ) = I P C ICHAN=12 CALL RANALCIRL,I CHAN) IRCJJ1)=IRL C *****LAST SCAN POINT OF THE LINE' ? IFCJ-NX) 22,23,23 • * • • 22 CALL STEPCIDXY,ITKjMCODE) GO TO 21 ' 2 3 MC0DE=192 CALL STEPCIDXY,ITK,MCODE) 2 1 CONTINUE JJ2=JJ2+1 C ODD OR EVEN TEST FOR LINE JUST 3CANNEDCJJ2=1 ODD,JJ2=2 EVEN) ****ODD LINEJ TRANSFER DATA IN THE SEQUENCE 1 TO NX DO 26 J = l i N X WRITE<4>27> IP(J> . ( WRITE<4>27> IR<J> -FORMAT<13) MC0DE=32 GO TO 20 ****EVEN LINE:TRANSFER DATA IN THE SEQUENCE NX TO DO 88 J=1>NX K=NX-J+1 • V7RITE(4,29> IPCK) WRITE<4.,29> IRCK) FORMATC13) MC0DE=48 JJ2=0 ' CONTINUE CALL OCLOSE *****RETURN TO ORIGIN NOTE! IXTOT=0 IF JJ2=0,IXTOT=NX-1 I F JJ2= 1 IYTOT=-NY*IDXY IF ( J J 2 - 1 ) 30,31»3t IXTOT=0 GO TO 32 IXTOT=-(NX - l)*IDXY CALL MVORGCIXTOT*IYTOT) WRITE<1.,33> FORMAT('SCAN COMPLETED*) VRITEC l.,49> FORMATC'PROGRAM COMPLETED , READY FOR NEXT RUN*) GO TO 5 END 139 P R O G R A M N A M E : S I P 8 R C P R O G R A M F O R S C A N N I N G I N T E R N A L P H O T O - E M I S S I O N W I T H C A M A T R I X P R I N T O U T O N V T 5 2 C O M M O N I P , I R D I M E N S I O N I P C S 0 0 ) » I R C 5 0 0 ) 5 W R I T E C 1 , 6 ) 6 F O R M A T C ' • ) W R I T E C 1 , 8 ) 8 F O R M A T C * N E W O R I G I N L O C A T I O N ' ) R E A D C 1 , 9 ) I O X , l O V , I D 9 F O R M A T C ' 0 X = ' I 3 , ' O Y = ' I 3 , ' O I D X Y = » I 3 ) C L O C A T I N G A N E W O R I G I N I O X = I O X * I D I C ) Y = I O Y * I D C A L L M V O R G C I O X , I O Y ) W R I T E C 1 , 1 0 ) 1 0 F O R M A T C ' S C A N D I M E N S I O N ' ) R E A D C 1 , 1 1 ) N X , N Y , I D X Y 1 1 F O R M A T C * N X = ' I 3 , ' N Y = ' I 3 , ' I D X Y = ' I 3 ) W R I T E C 1 , 6 0 ) 6 0 F O R M A T C " C I R C U I T T I M E C O N S T A N T C A L C U L A T I O N : - ' ) R E A D C 1 , 6 1 ) F V , F I , C 6 1 F O R M A T C ' F S V = ' F 5 . 2 , ' C U R R E N T R A N G E C P A ) = • F 6 . 2 , ' C C P F ) = ' F 5 * 1 ) C T K = F V * C / F I W R I T E C 1 , 6 2 ) C T K 6 2 F O R M A T C ' T I M E C O N S T A N T C M S E C ) = ' F 6 « 2 ) R E A D C 1 , 1 3 ) I T K , . 1 3 F O R M A T C ' I T K C M S E C ) = ' 1 3 ) -D X Y = F L O A T C I D X Y ) X = F L O A T C N X ) Y = F L O A T C N Y > T K = F L O A T C I T K ) . S C A N T K = C D X Y * 1 2 . 5 / 1 0 0 0 . 0 + T K / 1 0 0 0 . 0 ) * X * Y / 6 0 . 0 W R I T E C 1 , 1 4 ) S C A N T K 1 4 F O R M A T C ' S C A N N I N G T I M E C M I N U T E S ) = * F 6 . 2 ) W R I T E C 1 , 1 2 ) 1 2 F O R M A T C ' S T A R T ' ) C * * * * * I N I T I A L I Z A T I O N O F T H E S C A N ' J J 1 = 1 J J 2 = 0 N---1 M C 0 D E = 4 8 C * * + * * S C A N T H E S A M P L E D O 2 0 1 = 1 , N Y C + * : ] . * * S C A N O N E L I N E D O 2 1 J = 1 , N X I C H A N = 1 3 C A L L R A N A L C I P C , I C H A N ) I P C J J 1 ) = I P C I C H A N = 1 2 C A L L R A N A L C 1 R L , I C H A N ) I R C J J 1 ) = I R L C * * * * + L A S T S C A N P O I N T O F T H E L I N E ? I F C J - N X ) 2 2 , 2 3 , 2 3 2 2 C A L L S T E P C I D X Y , I T K , M C O D E ) J J 1 = J J 1 + 1 * N G O T O 2 1 • 2 3 • M C 0 D E = 1 9 2 C A L L S T E P C I D X Y , I T K , M C O D E ) J J 1 = J J 1 + N X 2 1 ' C O N T I N U E J J 2 = J J 2 + 1 C * * * * P A F I T Y T E S T F O R L I N E J U S T S C A N N E D C * * * * J J 2 = 1 O D D , J J 2 = 2 E V E N . 140 IFC J J 2 - 2 ) 2 4 , 2 5 , 2 5 24 N=-l MC0DE=32 , GO TO 20 ( 2 5 N=l MC0DE=48 JJ2=0 20 CONTINUE C *****RETURN TO ORIGIN C NOTE: IXTOT=0 IF JJ2=0,IXTOT=NX-1 I F JJ2=1 IYTOT=-NY*IDXY • IF C J J 2 - 1 ) 30,31*31 30 IXTOT=0 GO TO 32 31 IXTOT=-CNX-l)*IDXY 32 CALL MVORGCIXTOT,IYTOT) C ****PRINT A MATRIX ON VT52 READC1*40) IANS1 4 0 FORMAT<*REFLECTANCE MAP ?*A1> IF C I A N S l - 1 6 3 2 ) 42,41,42 4 1 CALL DAPRTDCNX,NY,IR) 42 READC1,43) IAN52 4 3 FORMATC'SIP MAP ?*A1) IFCIANS2-1632) 45,44,45 44 CALL DAPRTDCNX,NY,IP) 45 READC1,46) I AN S 3 46 FORMATC'REFLECTANCE MAP AGAIN ?*A1) IFCIANS3-1632) 48,41,48 4 8 WRITEC 1,49) 49 FORMATC'SCAN COMPLETED*) GO TO 5 . END 141 PROGRAM NAMEt MVORGCSUBROUTINE) C LOCATING A NEW ORIGIN BEFORE SCANNING C COORDINATE OF THE NEW ORIGIN W.R.T. THE OLD ONE IS CIOX,IOY) SUBROUTINE MVORGCIOX,IOY) IFCI OX) 1,2,3 1 IOX=-IOX DO 10 J=1,IOX S TAD CK40 S JMS MOVE ' . , 10 CONTINUE " ' . -x GO TO 2 3 DO 2 J=1,IOX S TAD CK60 S JMS MOVE 2 CONTINUE ' IFCIOY) 4,5,6 4 IOY=-IOY DO 1 2 J= 1, I OY 5 TAD CK200 5 JMS MOVE 1 2 CONTINUE GO TO 5 6 DO 5 J=1,IOY . S TAD CK300 S JMS MOVE 5 CONTINUE . . RETURN S MOVE,0 S 6332 S A,6331 S JMP A S 6336 S CLA S TAD WAIT . ' ' S B.IAC S S?.A S JMP B S JMP I MOVE . S WAIT,1777 END PROGRAM NAME! STEPCSUBROUTINE) C MOTOR STEPPING SUBROUTINE WITH A WAITING LOOP C IDXY IN UNITS OF 2 MICRON,ITK IN UNITS OF 1MSEC SUBROUTINE STEP<IDXY,ITK,MCODE) DO 1 J= 1, I DXY S CLA S TAD I \MCODE S 6332 S A,6331 S JMP A S 6336 S CLA C *** A 12.5MSEC WAITING LOOP IS USED TO LIMIT THE MAXIMUM C *** MOTOR STEPPING RATE TO 80 SPS S TAD WAITA S B,IAC S SZA S JMP B 1 CONTINUE IFCITK-J) 4,3,3 C 1 MSEC WAITING LOOP 3 DO 4 1=1,ITK S CLA ; S TAD WAITB S C I A C S SZA , -S JMP C 4 CONTINUE RETURN 5 WAITA,1712 S UAITB,7405 END 143 P R O G R A M N A M E t R A N A L ( S U B R O U T I N E ) C S U B R O U T I N E F O R R E A D I N G A N A L O G C H A N N E L V O L T A G E S U B R O U T I N E R A N A L ( I L V L , I C H A N > S C L A S T A D I W C H A N , S 6 3 2 3 S A * 6 3 2 1 S J M P A S 6 3 2 4 S C M A . S D C A \ L V L X L V L = F L O A T C L V L ) X L V L = X L V L / I o 0 2 4 I L V L = > I F I X C X L V L > R E T U R N END P R O G R A M N A M E ! D A P R T D C S U B R O U T I N E ) C D A T A P R I N T S U B R O U T I N E ? F O R 1 8 * 1 8 M A T R I X ) S U B R O U T I N E D A P R T D C N X * N Y , I X ) D I M E N S I O N I X C 5 0 0 ) D I M E N S I O N N J C 2 0 ) D O 1 J = > I . N X I N J C J ) = » J K = N X * C N Y - 1 ) D O 5 I M . N Y I N D Y « = N Y - I + 1 W R I T E C 1 , 6 ) I N D Y , C I X C J * K ) , J = 1 , N X ) 6 F O R M A T C 1 2 , 1 8 1 4 ) K o K - N X 5 C O N T I N U E W R I T E C 1 , 4 ) C N J C J ) , J a l , N X ) 4 F O R M A T C 2 X » 1 8 1 4 ) 7 R E T U R N E N D . 144 PROGRAM NAME: TMAP C PROGRAM FOR PLOTTING A 3-D MAP OR A SHADOW MAP ON THE DP-10 C PLOTTER 1 READC1,2) NX,NY,FILE 2 FORMATC• NX='13,' NY='I3,* FILE=*A6) READC1,3) MODE 3 FORMATC' D3MAP OR TPMAP?'Al) IFCMODE-288) 4,5,4 -5 CALL D3MAPCNX,NY,FILE) ' GO TO 6 . ' 4 CALL TPMAPCNX, NY, F I L E ) 6 GO TO 1 END P R O G R A M N A M E ! D 3 M A P C S U B R O U T I N E ) C S U B R O U T I N E F O R D R A W I N G A 3 - D I M E N S I O N A L M A P S U B R O U T I N E D 3 M A P ( N X » N Y , F I L E ) C O M M O N I P * I R D I M E N S I O N I P C 2 0 0 ) , I R C 2 0 0 ) , I R E F C 6 0 0 ) , I X C B 0 0 ) C * * * * P L O T T E R I N I T I A L I Z A T I O N 1 R E A D < 1 * 2 ) M O D E 2 F O R M A T ( * R E F L E C T A N C E O H P H O T O - C U R R E N T M A P ? E N T E R R F O R RMAP 1 F O R P M A P ' A l ) C * * * * * M A P S C A L I N G R E A D C 1 , 7 ) I M A Z 7 F O R M A T C ' Z - A X I S S C A L E F A C T O R < N * 1 0 ) ) » • 1 4 ) C A L L M S C A L E < N X , N Y , I M A Z , I D X 2 ) R E A D C 1 , 3 ) N O U S E 3 F O R M A T C ' P L O T T E R R E A D Y ? H I T R E T U R N K E Y I F Y E S ' A l ) R E A D C 1 , 4 ) N O I S E 4 F O R M A T C ' N O I S E T H R E S H O L D L E V E L » * 1 3 ) C * * * * D R A W A X I S C A L L D 3 A X I S < N X , N Y , I M A Z , I D X 2 ) C * * * * A F T E R D R A W I N G T H E A X I S , T H E P E N S H O U L D H A V E R E T U R N E D T O C * * * * T H E O R I G I N C * * * * P L O T M A P I N I T I A L I Z A T I O N I D X ° I D X 2 * 2 I P E N = 0 . . . C * * S E T A L L I R E F T O 0 I S I Z = N X * 2 * N Y . D O 1 0 I = J 1 , I S I Z 1 0 I R E F C I ) » 0 C P O S I T I O N P E N T O C O - O R D I N A T E C 1 , 1 ) I I D X = C 2 + 1 > * I D X 2 I I D Y = 2 * I D X e I S E G * 0 C A L L P L O T D C I I D X , 1 1 D Y , I P E N , I S E G ) C * * * * S T A R T T H E P L O T T I N G CALL IOPENC'DTAI'PFILE) D O 1 0 0 J o l , N Y . C * * O B T A I N A N D S C A L E T H E D A T A I S F = I M A Z / 1 0 0 D O 1 9 K « » 1 , N X R E A D C 4 , 1 4 ) I P C K ) R E A D C 4 , 1 4 ) I R ( K ) 1 4 FORMATC13) C * * D A T A F O R R M A P O R P M A P ? I F C M O D E - 1 0 5 6 ) 1 6 , 1 7 * 1 6 1 6 I X C K > = < I R C K ) - N O I S E > / 1 0 G O T O 1 5 1 7 I X ( X ) = C I P ( K ) - N O I S E ) / 1 0 1 5 I X C K ) = » I X C K ) * I S F I F C I X C K ) ) 1 8 , 1 8 , 1 9 1 8 I X C K ) a 0 1 9 C O N T I N U E C # * * * P L O T O N E L I N E ( B Y U S I N G DO L O O P D O W N T O S T A T E M E N T 1 0 1 ) D O 1 0 1 1 = 1 , N X I F C I - l ) 2 0 , 2 0 , 2 1 2 0 C O N T I N U E C * * P L O T T H E F I R S T P O I N T < I o l ) I I Z = I X < I > I S E G o 0 I D X o 0 C A L L P L O T D C I D X , I I Z , I S E G , I P E N ) 2 2 G O T O 1 0 1 2 1 C O N T I N U E C * * P L O T T H E R E S T N X - 1 D A T A C * * P L O T T H E H O R I Z O N T A L S E C T I O N I D X = I D X 2 I D Y = 0 I N D E X = S * < I - 1 ) * J D O 3 5 K * l , 2 I F < I Z I - I R E F < I N D E X ) ) 3 0 , 3 0 , 3 1 3 0 I F U Z l - I R E F C I N D E X + l ) ) 3 8 , 3 2 , 3 3 3 1 I S E G * » 1 I R E F ( I N D E X ) = I Z 1 G O T O 3 4 3 3 I S E G » 1 . G O T O 3 4 3 2 I S E G = 0 3 4 C A L L P L O T D C I D X , I D Y , I P E N , I S E G > 3 5 I N D E X - I N D E X + l C * * P L O T T H E V E R T I C A L S E C T I O N I D X » 0 I N D E X a 2 * I + J I Z 1 = I X < I - l ) + J * 2 * I D X 2 I Z 2 = I X < I ) + J * 2 * I D X 2 I F C I Z 1 - I R E F ( I N D E X > > 4 0 , 4 0 , 4 1 4 0 I F C I Z 2 - I R E F C I N D E X ) ) 4 2 , 4 2 , 4 3 4 1 I F U Z 2 - I R E F U N D E X ) ) 4 4 , 4 4 , 4 5 C * * * I Z 1 , I Z 2 « I R E F 4 2 I S E G = 0 I I D Z = I Z 2 - I Z 1 C A L L P L O T D < I D X , 1 1 D Z , I P E N , I S E G > G O T O 4 8 C * * * * I Z 1 < I R £ F < I Z 2 . 4 3 I S E G = > 0 I I D Z = I R £ F < I N D E X ) - I Z 1 C A L L P L 0 T D ( I D X , I I D Z » I P E N , 1 S E G ) I S E G - 1 •••••• 1 1 D Z = I Z 2 ~ I R E F < I N D E X ) C A L L P L 0 T D ( I D X , I I D Z , I P E N , 1 S E G > I R E F C I N D E X ) = I Z 2 , G O T O 4 8 C * * * * I Z 1 > I R E K > I Z 2 4 4 I S E G = 1 I I D Z < = I R E F < I N D E X ) - I Z 1 C A L L P L O T D < I D X , 1 1 D Z , I P E N , I S E G ) I S E G = » 0 I I D Z = I Z 2 - I R E F < I N D E X ) C A L L P L O T D ( I D X , I I D Z , I P E N , I S E G ) I R E F ( I N D E X ) = I Z 1 I R E F = I Z 1 G O T O 4 8 C * * * * I R E F < I Z 1 , I Z 2 4 5 I S E G = 1 I I D Z = I Z 2 - I Z 1 C A L L P L O T D f I D X , 1 1 D Z * I P E N , I S E G ) I I D Z « = I Z 2 - I Z 1 I F ( I I D Z ) 4 6 , 4 6 , 4 7 4 6 I R E F < I N D E X ) = I Z 1 G O T O 4 8 4 7 I R E F < I N D E X ) I Z 2 4 8 C O N T I N U E 1 0 1 C O N T I N U E C * * A F T E R P L O T T I N G A L I N E , P O S I T I O N T H E P E N T O ( 1 , J + 1 ) I I D Z » - ( 1 X « N X ) - 1 D X 2 * 2 ) I D X > » - < < N X - 1 ) * I D X 2 * 2 - I D X 2 ) I S E G = 0 147 C A L L PLOTD<IDX>IIDZ*IPEN*ISEG) 100 CONTINUE C * * * * P L O T T I N G COMPLETED*RETURN TO ORIGIN FROM POSI T I O N d >NY+1 > I D X o - ( ( N Y + 1 ) * I D X 2 + I D X a * 2 ) I I D Z o - < N Y * 1 ) * I D X 2 * 2 ISEG»0 . . . C A L L PLOTDCIDX>IIDZ#IPEN#ISEG> C * * * * T E S T I F ANOTHER PLOT IS NEEDED READ<1,102) IANS 102 F O R M A T ( ' P L O T ANOTHER M A P ? ' A l ) I F U A N S - 1 6 3 2 ) 103*1*103 103 RETURN END 148 P R O G R A M N A M E ! D 3 A X E S ? S U B R O U T I N E ) C D R A W A X E S W I T H S C A L E F O R D 3 M A P S U B R O U T I N E D 3 A X I S ? N X » N Y , I M A Z , I D X 2 ) I D X = I D X 2 * 2 C * * * > * * D R A t f X - A X I S N X 1 = N X + 1 I D l t > 1 0 I D 2 - 3 4 I D 3 » 1 7 I A N G o 0 • C A L L D A X I S ? I D X , N X 1 , I D 1 , I D 2 , I D 3 , I A N G ) C * * * * * D R A V Y - A X I S ? I D Y = I D X * 1 . 1 2 » I D X 2 « I D Y 8 > I D l o l 0 I D 2 = 1 0 I D 3 = 5 I A N G - 2 C A L L D A X I S ? I D X 2 , N Y , I D 1 » I D 2 , I D 3 , I A N 6 ) C * * * * * R E T U R N T O O R I G I N ? I X T O T = - ? N X 1 * 2 + N Y ) * I D X 2 ) I X T O T " - N X U I D X 8 I Y T 0 T * > - N Y - * 2 * I D X 2 I P E N n 0 I S E G = 0 C A L L P L O T D ? I X T O T , I Y T O T , I P E N , I S E G ) I X T 0 T = - N Y * 1 D X 8 I Y T O T - 0 C A L L P L O T D ? I X T O T , I Y T O T * I P E N , I S E G ) I X T 0 T = - N X l * I D X 2 I Y T O T = 0 C A L L P L O T S ? I X T O T , I Y T O T , I P E N , I S E G ) C * * * * * D R A W Z - A X I S ? I M A Z S H O U L D B E I N M U L T I P L E O F 1 0 ) N Z = I 0 I D Z o I M A Z / 1 0 I D 1 - 1 8 I D 2 » 6 I D 3 = 9 I A N G o 0 C A L L D A X I S ? I D Z , N Z , I D 1 , I D 2 , I D 3 , I A N G ) C * * * * * R E T U R N T O O R I G I N I Z T O T = - I M A Z I P E N = > 0 I S E G o 0 I X T O T = » 0 C A L L P L O T D ? I X T O T , I Z T O T , I P E N , I S E G ) R E T U R N E N D PROGRAM NAME: MSCALECSUBROUTINE) C *****MAP SCALING SUBROUTINE MSCALEC NX.NY,IMAZ,IDXS) ZMAZ=FLOATCIMAZ) HMAX=2600.0 VMAX=1600.0 5 7 YMAX=VMAX-ZMAZ XMAX=HMAX-YMAX/2.0 Y=FLOAT( NY) * , X=FLOATCNX) YD2=YMAX/2.0/Y XD2=XMAX/2.0/X IXDS=IFIXCXD2) IYD2=IF1XCYD2) IFCIYD2-IXD2) 50,50,51 50 IDX2=IYD2 GO TO 52 5 1 I D X 2 = 1 X D 2 5 2 DX2=FLOATCIDX2) HMAX=DX2*X*2.0+DX2*Y VMAX=DX2*Y*2.0+ZMAZ WRITEC1,53) HMAX,UMAX 53 FORMATC'SIZE OF THE MAP: HMAXC2600•0)='F6•1,• VMAXC1600.0) 1=*'F6.1) READ C1,54) I ANSI 54 FORMATC"SIZE OK? Y OR N*A1), IFCIANS1-1632) 55,56,55 5 5 READ C l , 5 8 ) HMAX,UMAX 58 FORMATC'ENTER NEW MAP SIZE,HMAXC2600•0)='F6•1, ' VMAXC1600.8 1='F6.1) GO TO 57 , . 5 6 CONTINUE RETURN END ' . 150 PROGRAM NAME: TPMAPCSUBROUTINE) C SUBROUTINE FOR DRAWING A TOPOGRAGHIC MAP ON THE INCREMENTAL C , PLOTTER C0MPL0T DP-I 0 SUBROUTINE TPMAPCNX,NY,FILE) COMMON IP,IR DIMENSION IPC200), IRC200) ' C ****PLOTTER INITIALIZATION C ****REFLECTANCE OR PHOTO-CURRENT MAP? READC1,3) MODE 3 FORMATC'REFLECTANCE OR PHOTO-CURRENT MAP?ENTER R FOR RMAP,P 1 FOR PMAP'Al) READC1,2) NOUSE 2 FORMATC'PLOTTER READY? PRESS RETURN KEY IF Y E S ' A l ) C ****DRAW THE FRAME C1400MAX*1400MAX) IFCNX-NY) 4,4,5 4 NXY=NY GO TO 6 5 NXY=NX 6 IMAX=1400 IDX=IMAX/NXY IANG=0 NX1=NX+1 NY1=NY+1 C ****DIRE INDEX:ID1=DIRE OF AXIS,ID2=DIRE OF SCALE, C ****ID3=DIRE OPPOSITE TO ID2 ID1-10 , ID2=34 ID3=17 CALL DAXISCIDX,NX1,ID1,ID2,ID3,IANG) IDl=18 ID2=10 ID3=5 CALL DAXISCIDX,NY1,IDl,ID2,ID3,IANG) C ****RETUPN TO ORIGIN v IIDX = -CMX+D*IDX I IDY = -CNY+1 )* I DX ISEG=0 IPEN=0 CALL PLOTDC11DX,11DY,I PEN,ISEG) I D1 = 18 ID2=6 ID3=.9 CALL DAXISCIDX,NY1,IDl,ID2,ID3, IANG) ID1=10 ID2 =18 ID3=33 CALL DAXISCIDX,NX1,IDl, ID2, ID3,IANG) C ****AFTER DRAWING THE FRAME, THE PEN SHOULD RETURN C TO THE ORIGIN I IDX = -CNX+1)*IDX I IDY = -CNY+1)*IDX ISEG=0 IPEN=0 CALL PLOTDC11DX,11DY, I PEN, I SEG) 100 READC1,9) LVLL,LVLH 9 FORMATC'DISCRIMINATION LEVELSCI 3),LVLL=•I 3,•LVLH =•I 3> C DRAW THE TOPOGRAGHIC MAP . CALL I O P E N C ' D T A l « , F I L E ) INDX0=0 INDY0=0 DO 20 J=1,NY - -JJ=0 DO 20 1=1,NX JJ=JJ+1 READC4,10) I P C J J ) 151 READC4*10) I R C J J ) 10 FORMATCI 3 > IFCMODE-1056) 11.12*11 11 IZ=IRCJJ) GO TO 13 12 I Z = I P C J J ) 13 I F CIZ-LVLL) 20*14*14 14 I F CIZ-LVLH) 15*15*20' 15 INDX=I INDY=J IIDX=CINDX-INDX0)*IDX ., IIDY=CINDY-IN DY"0 ) * I DX IDOT=l ISEG=0 CALL PLOTDC11DX*11DY*I DOT*ISEG) INDX0=INDX INDY0=INDY 2 0 CONTINUE C ****RETURN TO ORIGIN IIDX=-INDX0*IDX I IDY = -INDY0*IDX IDOT=0 ISEG=0 CALL PLOTDC11DX*IIDY*I DOT*ISEG) READC1*21) IANS 2 1 FORMATC'PLOT ANOTHER L E V E L ? ' A l ) IF CI ANS-1632) 101* 100* 101 101 RETURN END P R O G R A M N A M E t D A X I S < S U 8 R 0 U T I N E > C D R A W A N A X I S W I T H S C A L E S U B R O U T I N E D A X I S < I D X . N X , I D 1 * I D 2 * I D 3 * I A N G ) C I D 1 = D I R E C O D E O F T H E A X I S * I D 2 = > D I R E C O D E O F T H E S C A L E S * C I D 3 ° D I R E O P P O S I T E T O 1 D 2 * I A N G = A N G L E O F T H E A X I S M A D E C W I T H T H E H O R I Z O N T A L . S 6 5 0 0 S C L A S , 6 5 0 6 J J = . 0 D O I K = 1 * N X J J « » J J + 1 C O 2 I = 1 * I D X S T A D I N I D I S J M S X L A T E C D R A W A N A X I S W I T H A N I N C L I N A T I O N C I A N G = 0 F O R 0 O R 9 0 D E G R E E D E P E N D I N G O N I D l * I A N G n l F O R 4 5 * C I A N G = 2 F O R 6 3 . 4 3 I F U A N G - 1 > 2 * 1 1 * 1 1 1 1 D O 2 J o l * I A N G S T A D < K 2 2 S J M S X L A T E 2 C O N T I N U E I F ( J J - 1 0 > 3 * 4 * 4 3 I S X = 1 0 G O T O 5 4 I S X - » 2 0 J J = 0 5 D O 6 J » 1 , I S X S T A D I \ I D 2 5 J M S X L A T E 6 C O N T I N U E D O 1 J«= 1 » I S X S T A D I \ I D 3 S J M S X L A T E 1 C O N T I N U E R E T U R N S X L A T E * 0 S A * 6 5 0 1 S J M P A S 6 5 0 6 S C L A C L L S J M P I X L A T E E N D PROGRAM NAMEt PLOTDCSUBROUTINE) C SUBROUTINE FOR MOVING THE PEN FROM A TO B OR FOR DRAWING A LINE C FROM A TO B OR FOR MOVING THE PEN FROM A TO B AND PRINT A DOT C AT B. C FROM A TO B» ISEG=1 FOR PLOT SEGMENT*ISEG=0 FOR MOVE PEN C AT Bt IPEN«0 FOR PEN UP*I PEN*t FOR PEN DOWN SUBROUTINE PLOTD<IDX*IDY*I PEN*ISEG) S 6500 S CLA CLL . S 6506 IDIRE=9*ISEG IDIRS=33+ISEG .IDIRW=5+ISEG IDIRMM7+ISEG IFUDX) 1*2*3 1 IDX=-IDX DO 10 J» 1 * I DX S TAD \IDIRW S JMS XLATE 10 CONTINUE GO TO 2 3 DO 2 J=1*IDX S TAD \IDIRE S JMS XLATE 2 CONTINUE IFCIDY) 4*5*6 4 IDY=-IDY DO 12 J-1*IDY • 5 TAD MD1RS 5 JMS XLATE 12 CONTINUE GO TO S 6 DO 5 J=1*IDY S TAD WDIRN S JMS XLATE . 5 CONTINUE IF(I PEN- 1> 14*13*14 13 CONTINUE S JMS WAIT S 6505. S JMS WAIT S 6503 • 14 RETURN S XLATE,0 S XA*6501 ' S JMP XA S 6506 S CLA CLL S JMP I XLATE • S WAIT*0 S XB*6501 S JMP XB S 6502 S JMP I WAIT END P R O G R A M N A M E ! L A B E L C P R O G R A M F O R L A B E L L I N G A X E S A N D P R I N T I N G T I T L E O N A D 3 M A P C T H I S P R O G R A M I S W R I T T E N I N 0 S 8 F O R T R A N 4 W R I T E C 4 * i > 1 F O R M A T ? ' E N T E R T H E S I X P A R A M E T E R S F O R S C A N S I Z E & M A P S I Z E J • ) W R I T E C 4 # 2 > R E A D C 4 * 3 > N X 2 F O R M A T C • N X = » * > $ ) 3 F O R M A T C 1 5 ) . W R I T E C 4 , 4 > 4 F O R M A T C • N Y ^ ' ^ S ) R E A D C 4 > 3 ) N Y -W R I T E C 4 > 5 > 5 F O R M A T C * I D X Y = » ' > S ) R E A D C 4 » 3 > I D X Y V R I T E C 4 » 6 ) 6 F O R M A T C • R E A D C 4 , 3 > I Z W R I T E C 4 * 7 ) 7 F O R M A T C * I H S I Z = ' * S ) R E A D C 4 , 3 > I H S I Z W R I T E C 4 , 8 > 8 F O R M A T C • I V S I Z ? « , $ ) R E A D C 4 . 3 ) I V S I Z W R I T E C 4 , 9 ) 9 F O R M A T C * E N T E R M A R K I N T E R V A L S ! * > W R I T E C 4 , 1 0 ) 1 0 F O R M A T C ' I T X Y = ' * $ ) R E A D C 4 > 3 ) I T X Y W R I T E C 4 , U ) 1 1 F O R M A T C • I T Z = S $ ) R E A D C 4 , 3 > I T Z C A L L P L O T S C . 0 0 5 , 0 ) 1 0 0 W R I T E C 4 . 4 0 ) 4 0 F O R M A T C * E N T E R P F O R P M A P > R F O R R M A P J ' , 8 ) R E A D C 4 * 4 1 ) M O D E 4 1 F O R M A T C A l ) D A T A P M A P / ' P V I F C M O D E . N E . P M A P ) G O T O 4 2 U R I T E C 4 . . 1 2 ) 1 2 F O R M A T C ' E N T E R C U R R E N T S C A L E t * ) W R I T E C 4 * 1 3 ) 1 3 F O R M A T C • 1 0 = ' * $ ) R E A D C 4 » 3 ) 1 0 W R I T E C 4 * 1 4 ) 1 4 F O R M A T C • I E X P n ' , 5 ) R E A D C 4 , 3 ) I E X P 4 2 W R I T E C 4 , 1 5 ) 1 5 F O R M A T C ' E N T E R R U N N O % •> W R I T E C 4 , 1 6 ) 1 6 F O R M A T C * R U N N O = > ' , $ ) R E A D C 4 W 7 ) R U N . N O 1 7 F O R M A T C 8 A 6 ) W R I T E C 4 . 6 0 ) 6 0 F O R M A T C ' P R I N T T I T L E ? Y O R N ' » S ) R E A D C 4 » 6 1 ) P T I T L E 6 1 F O R M A T C A 1 ) C P L O T T E R I N I T I A L I Z A T I O N C A L L X Y P L O T C 0 > 0 » - 3 ) C H A R S ° 0 . 1 4 I D E L 2 = I H S I Z / C N X * 2 * N Y ) ' T X Y = C I D E L 2 * 2 * I T X Y ) $ . 0 0 5 C M A R K X - A X I S N N X = > N X / I T X Y DO 19 J = 1 * N N X I X U N I T = I D X Y * 2 * I T X Y * J C A L L N U M B E R C T X Y * J - C H A R S * - * C H A R S * 0 . 1 5 ) , C H A R S * I X U N I T * 0 * - 1 > C A L L S Y M B O L * I D E L 2 * N X * 0 . 0 0 5 - 3 * 0 . 2 1 * - . 6 5 * 0 . 2 1 # ' M I C R O N S ' » 0 * 7 ) P R I N T T I T L E A N D R E M A R K D A T A Y E S / ' Y * / I F < P T I T L E . N E . Y E S ) GO TO 6 8 I F C M O D E . N E . P M A P ) GO TO 4 4 C A L L S Y M B O L * I D E L 2 * 2 * * N X + 1 ) * . 0 0 5 * 1 . 4 * . 8 * . 2 8 * ' S I P M A P ' * 0 * 7 ) 6 0 T O 4 6 . X A X I S - - * I D E L 2 * g * ( N X + l ) * . 0 0 5 C A L L S Y M B O L C X A X I S + 1 . 0 * . 8 * . 2 4 5 * ' R E F L E C T I V I T Y M A P ' * 0 * 1 6 ) C O N T I N U E C A L L S Y M B O L * 1 D E L 2 * 2 * ( N X + 1 ) * . 0 0 5 + 1 . 0 * 0 * . 2 1 * ' R U N N O j ' * 0 » 7 ) C A L L S Y M B 0 L * I D E L 2 * 2 * ( N X + 1 ) * . 0 0 5 + 1 . 0 + 7 * . 2 1 * 0 * . 2 1 * R U N * 0 » 6 ) C A L L S Y M B O L < I D E L 2 * 2 * ( N X + 1 ) * . 0 0 5 + 1 . 0 + 1 3 * . 2 1 * 0 * . 2 l * N 0 , 0 * 6 ) C O N T I N U E MARK Y - A X I S N N Y o N Y / I T X Y ; DO 2 0 J = 1 * N N Y I Y U N I T = » I D X Y * 2 * I T X Y * J C A L L N U M S E R * I D E L 2 * ( 2 * ( N X + 1 ) + I T X Y * J ) * . 0 0 5 * 0 . 1 5 * I D E L 2 * 2 * I T X Y * U * . 0 0 5 * C H A R S * I Y U N I T * 0 * - I ) C O N T I N U E M A R K Z - A X I S I F C M O D E . E Q . P M A P ) GO TO 4 8 Z U i ^ I T = 1 . 0 C A L L N U M B E B * - * C H A R S * 3 + . 1 5 ) * I Z * . 0 0 5 * C H A R S * Z U N I T * 0 * 1 ) C A L L S Y M B O L < - 1 . 7 * I Z * . 0 0 5 + . 5 » . 2 1 * ' U N I T S ' » 0 , 5 > C A L L S Y M B O L * - 1 . 7 * I Z * . 0 0 5 + . 9 * . 2 1 * * A R B I T R A R Y ' * 0 * 9 ) GO T O 5 0 C O N T I N U E T Z = I Z / 1 0 * I T Z * 0 . 0 0 5 * ' ' N N Z = 1 0 / I T Z DO 3 0 J o l , N N Z Z U N I T = I 0 * J T Z / 1 0 . 0 * J C A L L M U M B E R * - C C H A R 5 * 3 + . 1 5 ) * T Z * J * C H A R S * Z U N I T * 0 » 1 ) C A L L S Y M B O L C - 1 . 7 * I Z * . 0 0 5 + . 5 * . 1 7 5 * ' X 1 0 E X P C » * 0 » 7 ) C A L L N U M B E R C - 1 . 7 + 7 * . 1 7 5 . I Z * . 0 0 5 + . 5 * . 1 7 5 * I E X P * 0 * - 1 ) C A L L S Y M B O L * - 1 . 7 + 1 0 * . 1 7 5 * I Z * . 0 0 5 + . 5 * 0 . 1 7 5 * ' ) a * 0 * 1 ) C A L L S Y M B O L ( - 1 * I Z * . 0 0 5 + . 9 * . 2 1 * * A M P S ' * 0 * 4 ) C A L L X Y P L O T < 0 * 0 * - 3 ) GO T O 1 0 0 E N D Appendix 3 CIRCUIT DIAGRAMS OF THE SELF-HEALING BREAKDOWN TESTER. I D / THRESHOLD CURRENT SELECTOR THRESHOLD CURRENT SELECTOR PREAMPLIFIER INPUT Fig. A3.1 Threshold current selector circuit 1 5 8 3.9K' -6V I f ) 1 2 u 0+5V to 2A Fig. A 3 . 2 Preamplifier .threshold detector and pulse shaper cir c u i t +5V 200mS PULSE 2A from IA ^~ +5V O- ( ~ -i—^"O-400 15 150 ,ED (RESET INDICATOR) 2N3053 I K 2L ^  > to 4B J O . O l u ^ ^ A A i SN7493 MANUAL START SWITCH V Y '390 +5V ~ ^ to 3A ^ > to 3D to 3B 2 2 to 3C Fig. A3.3 SCR reset control and counter circuit 160 MANUAL STOP SWITCH COUNTER STATE INDICATOR 1 -O, 560 15 I S 390 1150 ?LED (STOP INDICATOR) -> X oo 2N4441 > Y 1K$ i . O l j i ->to 4A 5V LED (START INDICATOR) 150 390 - ~ ~ * 2N3053 F i g . A3.4 Counter c o n t r o l l e d stop c i r c u i t (+) o HIGH VOLTAGE INPUT (-) O-INPUT OVERVOLTAGE INDICATOR Q+20V 5/5W \7LED 1N4005 >560 \ 390 - 4 0 -IM 2N4124 20K 2 N 4 4 4 4 2N3053 TT IK| - ro.oiu 10K 10K O.Ol/i 100K (TEN TURN) IOV | ZENER ^ 2N5214 250K 20/u 550WV 30/5W 2N4444 _10K>_ 4-RAMP OUTPUT 4A COUNTER RESET CONTROL 4B RESET CONTROL Fig.. A3.5 Overvoltage crowbar and ramp generator c i r c u i t 162 H A M M O N D 167J28 15VX2 O+20V 0+12V -0+5V JT lOOOu _ 25WV -r- 0.1*1 - G R O U N D 220a 25WV ^ 6V 450 AAAr 300 -AAAr -O -6V j i 2 V -O -12V F i g . A3.6 Power supply of the tester Appendix 4 PRE-FURNACE CLEANING OF SILICON WAFERS This process was f i r s t developed by Werner Kern and David A. Puotinen of the RCA laboratories i n 1970. I t i s often r e f e r r e d to the RCA cleaning process and i s now used extensively i n the industry. The process co n s i s t s of the following steps. (1) D.I. H 20 + NH^OH + H 20 2 i n a volume r a t i o of 5:1:1 use for 10 minutes at 75 - 85°C. (2) D.I. H 20, r i n s e for 10 minutes. (3) 10% HF, etching f o r 30 seconds. (4) D.I. H 20, r i n s e for 10 minutes. (5) D.I. H 20 + HCJi + ^2®2 ^ n a v o-'- u m e r a t i ° °f 6:1:1, use for 10 minutes at 75 - 85°C. (6) D.I. H 20, r i n s e f o r 10 minutes. (7) Blown dry by using N 9 gas. 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0095217/manifest

Comment

Related Items