TUNNELING RESISTANCE OF A ONE DIMENSIONAL RANDOM LATTICE ISABEL CRISTINA DOS SANTOS^CARVALHO .Sc., Universidade Federal Do Rio De Janeiro, B r a s i l , 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT . OF THE REQUIREMENTS FOR THE DEGREE OF THE FACULTY OF GRADUATE STUDIES (Department of Physics) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA ®Isabel C r i s t i n a dos Santos Carvalho, 1984 • By MASTER OF SCIENCE i n September, 1984 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t 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 study. I f u r t h e r agree 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 copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of 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 allowed without my w r i t t e n p e r m i s s i o n . Department of ? \\ V S i c % The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date (DcJLsJU^ 13 ABSTRACT The r e s i s t i v i t y of a one-dimensional l a t t i c e consisting of randomly d i s t r i b u t e d conductivity and i n s u l a t i n g s i t e s i s considered. Tunneling resistance of the form R D i e^* i s assumed for a c l u s t e r of i adjacent i n s u l a t i n g s i t e s . Three d i f f e r e n t ensembles are considered and compared. In the f i r s t ensemble the number of i n s u l a t i n g "atoms" i s fixed and d i s t r i b u t e d i n a l i n e a r chain; i n the second one there exists a fixed p r o b a b i l i t y p of having an in s u l a t o r "atom" occupying a s i t e i n a l i n e a r chain, and f i n a l l y the third one consists of a l i n e bent into a c i r c l e and the p r o b a b i l i t y p i s considered. I t i s observed that i n the thermodynamic l i m i t , the average ensemble resistance per s i t e diverges at the c r i t i c a l f i l l i n g f r a c t i o n p c = e~^, while the variance of the resistance 2 diverges at the lower f i l l i n g f r a c t i o n p = p . Computer simulations of c i c large but f i n i t e systems, however, ex h i b i t a much weaker divergence of the resistance per s i t e at p c and no divergence of the variance at P C l « - i i i -TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i i i LIST OF FIGURES . v DEDICATION v i i ACKNOWLEDGEMENTS v i i i CHAPTER ONE: INTRODUCTION 1 1.1 Motivation for the Work 1 1.2 Description of the Model 2 CHAPTER TWO: STATISTICS OF THE MODEL 6 2.1 Fixed Number of INSULATOR "ATOMS" Dist r i b u t e d i n the Chain 6 2.2 P r o b a b i l i t y p of having an Insulator Occupying a S i t e 10 2.3 Ensemble with Fixed P r o b a b i l i t y p and Line Bent in t o C i r c l e 17 2.4 Comparison Between Ensembles 23 CHAPTER THREE: COMPUTER SIMULATIONS 28 CHAPTER FOUR: CONCLUSION 35 REFERENCES 37 APPENDIX A: Ca l c u l a t i o n of < K± > for Fixed Number of Insulator "Atoms" Distributed i n the Chain...................... 38 APPENDIX B: Calcu l a t i o n of < K± Kj > for Fixed Number of Insulator "Atoms" Distrbuted i n the Chain 41 - i v -Page 2 APPENDIX C: Calcu l a t i o n of < R > for Fixed Number of Insulator "Atoms" Distributed i n the Chain 46 APPENDIX D: Calcu l a t i o n of < K ± > for Fixed P r o b a b i l i t y p of having an Insulator Occupying a Site i n a Line. 48 APPENDIX E: Ca l c u l a t i o n of < K ± K, > for Fixed P r o b a b i l i t y p of having an Insulator Occupying a Site i n a Line 50 APPENDIX F: Ca l c u l a t i o n of < R 2 > for Fixed P r o b a b i l i t y p of having an insula t o r occupying a s i t e i n a l i n e . . 55 APPENDIX G: Ca l c u l a t i o n of < K ± > for Fixed P r o b a b i l i t y p of having an Insulator Occupying a Site i n a Line Bent into a C i r c l e 58 APPENDIX H: Ca l c u l a t i o n of < YL± > for Fixed P r o b a b i l i t y p of having an Insulator Occupying a Site i n a Line Bent i n t o a C i r c l e 60 APPENDIX I: C a l c u l a t i o n of < R2 > for Fixed P r o b a b i l i t y p of having an Insulator Occupying a S i t e i n a Line Bent into a C i r c l e 65 - v -LIST OF FIGURES Figure Page 1 Diagram of a trapezoidal b a r r i e r i n an i n s u l a t i n g f i l m between two si m i l a r electrodes 3 2 Diagram of a l i n e a r chain of metal-insulator-metal junctions 4 3 a Dependence of the ensemble average resistance per s i t e on the f r a c t i o n nj/N for N = 100. Ensembles with fixed n^ and fixed p r o b a b i l i t y p i n a l i n e 8 b Dependence of the ensemble average resistance per s i t e on the f r a c t i o n n^N for N = 200. Ensembles with fixed ^ and fixed p r o b a b i l i t y p i n a l i n e 8 4 Dependence of the variance « R 2 > - < R > 2)/N 2R 2 on the f r a c t i o n n^/N. Ensemble with fixed n^ 11 2 2 2 2 5 Dependence of the variance (<R > - < R > ) / N R on the p r o b a b i l i t y p. Ensemble with fixed p r o b a b i l i t y p i n a l i n e 18 6 Dependence of the variance (< R 2 > - < R > 2)/N 2R 2 on the array s i z e . Ensemble with fixed p r o b a b i l i t y p i n a l i n e 19 7 Dependence of the ensemble average resistance per s i t e on the p r o b a b i l i t y p. Ensemble fixed p r o b a b i l i t y p and l i n e bent into a c i r c l e 21 8 Dependence of the variance « R 2 > - < R > 2)/N 2R 2 on the p r o b a b i l i t y p. Ensemble fixed p r o b a b i l i t y p and l i n e bent into a c i r c l e 24 - v i -Dependence of value of < RI >/NRQ, the contribution to < R > from a l l islands of size I, on the isl a n d s i z e . Ensembles fixed n\ and fi x e d p r o b a b i l i t y p i n a l i n e (a) P = 0.45, ni = 45 N = 100 (b) P = 0.48, ni = 48 N = 100 (c) P = 0.50, n l = 50 N = 100 (d) P = 0.52, n l = 52 N = 100 Sample mean resistance per s i t e for M runs versus the p r o b a b i l i t y p. Ensemble with fixed p r o b a b i l i t y p i n a l i n e Computer simulation geometric mean over M runs versus £n(N). Ensemble with fixed p r o b a b i l i t y p i n a l i n e . . . Computer simulation cumulative average resistance per s i t e versus the number of runs. Ensemble with f i x e d p r o b a b i l i t y p i n a l i n e Histogram of the logarithm of the sample mean resistance per s i t e for M runs. Ensemble with f i x e d p r o b a b i l i t y p i n a l i n e - v i i -DEDICATION To My Son - v i i i -ACKNOWLEDGEMENTS I would l i k e to thank my supervisor, Professor R. Barrie, for h i s support and guidance throughout this work. I would also l i k e to thank Professor B. Bergersen and Dr. P. Palffy-Muhoray for their help with the computer. F i n a l l y , I wish to thank the funding through CAPES, B r a s i l , and NSERC. - 1 -CHAPTER ONE INTRODUCTION 1.1 Motivation f o r the Work The properties of composite materials of metals and insulators have been the subject of many studies (ABE75). These materials have three d i f f e r e n t e l e c t r o n i c conduction regimes: m e t a l l i c regime, d i e l e c t r i c regime and t r a n s i t i o n regime. The o r i g i n a l motivation f o r . t h i s work was a study of the mechanism of conduction i n these granular metals i n the d i e l e c t r i c regime. In this regime metal p a r t i c l e s are dispersed i n a d i e l e c t r i c continuum. Abeles (ABE75), i n his work, calculates the lo w - f i e l d conductivity, o^, i n the d i e l e c t r i c regime of granular metals taking into consideration the tunneling through an i n s u l a t o r , from one i s o l a t e d m e t a l l i c grain to the next one. He takes into account the charging energy, E C, that i s necessary to remove an electron from one neutral metal grain and place i t on another neutral metal grain. Therefore, the expression for the t o t a l l o w - f i e l d conductivity, OL» i s the sum of products of mobility, charge and number density of charge c a r r i e r s over a l l possible percolation paths: where B(s) i s the density of percolation paths associated with the value s and s i s the grain separation. The exponent e""^:s i s the tunneling p r o b a b i l i t y wi th x = (2mcT/'n2) m the electron mass, <|> the bar r i e r - 2 --E°/2KT height and h Planck's constant. The exponent e i s the Boltzmann f a c t o r , which i s proportional to the number density of a l l charge c a r r i e r s with charging energy E°. Aft e r making some approximations, Abeles evaluates the expression and obtains: - 2 ( C / K T ) 1 / 2 oT = a e L o where O q i s a constant independent of temperature and C = 3(sE°. In order to i s o l a t e the role of tunneling i n this work, one can look at zero temperature and lo w - f i e l d conductivity, removing the idea of a charging energy f a c t o r . Thus one can consider the conduction to be through a one dimensional series of metal-insulator-metal junctions. 1.2 Description of the Model The tunneling e f f e c t i n junctions of metal-insulator-metal has been widely discussed (DUK69). In the present work, the regime of low voltage and the tunneling between two metals of the same kind at zero temperature w i l l be considered. Since our i n t e r e s t i s i n the low voltage ohmic regime, the p o t e n t i a l b a r r i e r w i l l be considered as rectangular. The expression for the tunneling current, for the above case, was obtained by Sommerfeld and Bethe (SOM33) and i s reproduced i n the paper by Holm (H0L51). Figure (1) shows the diagram of a trapezoidal b a r r i e r i n an i n s u l a t i n g f i l m between two s i m i l a r metal electrodes. In the ohmic regime the b a r r i e r can be considered as a rectangular one. - 3 -WORK FUNCTION FERMI ENERGY \ \ TUNNEL CURRENT OCCUPIEO LEVEL APPLIED VOLTAGE As-* METAL METAL INSULATOR Figure 1: Diagram of a trapezoidal b a r r i e r i n an i n s u l a t i n g f i l m between two s i m i l a r electrodes. The equation for the tunneling resistance per unit cross section area where h = Planck's constant q = electron charge As = thickness of the junction m = electron mass cj> = work function For our purposes we w i l l write the b a r r i e r thickness as As = i£ Q» where £ Q i s the unit of "atomic" spacing and i corresponds to the number, of s i t e s with unit length £ Q . The resistance R of a single i n s u l a t i n g c l u s t e r of i s i t e s due to tunneling w i l l then be of the form: i s (H0L51): R = ( h 2 As/q 2(2m4>) 1 / 2) exp [ (4T r As/h) (2m(j>) 1 / 2]; (1) R = R i e 1 o (2) where R q = h 2 Hja} (Imfy)1^ and b = (4 TT £ Q/h) (2m<)))1/'2. To obtain equation ( 1 ) , the tunneling p r o b a b i l i t y for a rectangular b a r r i e r was calculated and this was done by the evaluation of the transmission c o e f f i c i e n t D ( E X ) using the WKB approximation. As a model of the one dimensional granular metal, consider a l i n e a r chain of length N £ Q of metal and i n s u l a t o r "atoms" d i s t r i b u t e d randomly. Each s i t e of length iQ can be occupied by a metal or an i n s u l a t o r "atom".A clu s t e r of i consecutive i n s u l a t i n g u n i t s , each of unit length lQy w i l l form an i s l a n d of size i , Figure (2) i = 2 — M I I M M I M o N£ 0 Figure 2: Diagram of a l i n e a r chain of metal-insulator-metal junctions. In the future we w i l l talk about an array' of N s i t e s . The resistances of the metal islands are going to be considered equal to zero and the resistance of an i n s u l a t i n g i s l a n d w i l l be given by equation (2). In other words, the l i n e a r array w i l l be composed of a series of metal-insulator-metal junctions. A t y p i c a l junction of metal-insulator-metal, 1/2 with unit length (£ ) equal to 5A, would have b = 5.125<|> . Considering, - 5 -f o r instance, the case of SiU2(ABE75), b would be of the order of 10. In order to i l l u s t r a t e the study of the s t a t i s t i c s of the problem, the value of b w i l l be taken to be Zn2. This choice w i l l become clear l a t e r (See P. 9). - 6 -CHAPTER TWO STATISTICS OF THE MODEL The present model w i l l be concerned with the ohmic regime and zero temperature. Consider a l i n e a r chain of s i t e s that i s randomly occupied, either by an i n s u l a t o r "atom" or a metal "atom". The end two s i t e s of the chain are f i l l e d with metal. Each arrangement of metal-insulator-raetal w i l l form a ju n c t i o n . We consider the l i n e a r chain to be formed of a set of independent junctions. Take n^ as the number of d i e l e c t r i c atoms, (n2 - 2) the number of metal atoms such that N = n^ + n2 i s the number of s i t e s i n the l i n e a r chain. The c a l c u l a t i o n of the resistance and variance for d i f f e r e n t ensembles, i n this model, i s described below. 2.1 Fixed Number of Insulator "Atoms" Distributed In the Chain In the case of a fixed number of insulator "atoms" d i s t r i b u t e d i n a chain with metal ends, the t o t a l ensemble average of the resistance w i l l be given by: n l r b i < R > = R l i e < K, >, o . , i ' i=l where < > i s the average number of d i e l e c t r i c c l u s t e r s of s i z e i ( i = 1, 2 , . . . , n j ) . - 7 -In analogy with the d i s t r i b u t i o n of runs (WIL62), the p r o b a b i l i t y of having d i e l e c t r i c clusters of size i ( i = 1, 2, n^) i s : n 2 - 1 K! \ K PCK.f^}) - K I v ! .. .K ! /N - 2 1 2 n. 1 1 n x n l n l where fc. } = {K, , K~ , ... K }, I K. = K, £ i K, = n, and K i s the 1 i=l i=l 1 1 t o t a l number of metal-insulator-metal junctions. The value of K ranges from 0 to ni and the value of n\ from 0 to N-2. The expression for < > can then be evaluated < K^> = Jo (A ztn T^W K* I t i s shown i n Appendix A that: t n l The t o t a l resistance w i l l then be given by: < R > = R Q i e TN-TTy 1^ n x - i ' ( 5 ) We have not been able to express this sum i n any s i g n i f i c a n t l y simpler fashion. Figures (3a) and (3b) show the graph of the resistance per s i t e versus the f r a c t i o n n^/N for values of N = 100 and N = 200. - 8 -Figure 3; Dependence of the ensemble average resistance per site on the fraction n^/N. Ensembles with fixed n^ and fixed probability p in a l ine , (a) N = 100 (b) N = 200 - 9 -In the l i m i t of N, n\ and n2 tending to i n f i n i t y , such that n l n2 — = p and -JJ— = (1 - p), the expression for the resistance per s i t e i s : T T * « ' - » 2 7 - ^ 2 « > o (1 - pe ) • Therefore, the average resistance per s i t e diverges for an i n f i n i t e chain, when p = P c = e ^. The value of b i s then chosen to be equal to Zri2 such that the f i l l i n g factor p c = 1/2 and i s at the centre of the i n t e r v a l from p = 0 to p = 1. In order to determine < >, the moment < Kj > has to be calculated by using the following expression: v2"1 < K. K. > = I I — /.T K ' K. K., (7) with I K± = K and I K± i = ^ . i i From Appendix B, the expression for < Kj > i s : (n - D ( n 2 - 2) / 4 , < K. K. > = - 2 j ( 4 - 4 J + 6 i j < K i > ' <*) i t i s to be noted that the f i r s t term i s zero unless i + j < n^ This leads to: < R2 > = R2 11 i j e b ( i + J } < K K. >. (9) o . . J i j The average of the square of the resistance i s calculated i n - 10 -Appendix C to be: < R2 > - R2 r ( n 2 " 1 ) ( n 2 " 2 ) y 1 (4 m3 - m) bm /N - m - 4\ < R > _ R ° 1 ' N n " 2 j Jo 1 2 6 V n 2 " 4 ' x ( n 2 " 1 } V 1 ,2 2bi /N - i - 3\ i Again, we have not been able to express these sums i n any simpler form. In the l i m i t of N, n^, and n 2 tending to i n f i n i t y , such that n^/N = p and n 2/N = (1 - p), the expression for the variance i s < r Z > - < R > 2 • m - P) 2 p e 2 " a + ffi R (1 - pe ) Therefore, the variance diverges for an i n f i n i t e chain when -2b p = p = e 1 In order to evaluate the variance, the < R > and < R^ > are calculated numerically. The behavior of the variance with the r a t i o n^/N i s shown i n Figure (4). 2.2 Probability p of Having an Insulator Occupying a Site In this ensemble, there i s a fi x e d p r o b a b i l i t y p of having an i n t e r i o r s i t e occupied by an in s u l a t o r , and p r o b a b i l i t y (1-p) for such a s i t e to be occupied by a metal. The two end exterior s i t e s are occupied by metal. The p r o b a b i l i t y (M0041) of having xx\ d i e l e c t r i c "atoms" d i s t r i b u t e d i n Ki clu s t e r s of size i ( i = 1, 2,...,n]_), i s : r - 11 -- 12 -p < v K ; { K I } ) = P n i (i - p ) v 2 ^ K i , ^ ^ , y n2 " *\ (11) n 1 n x w i t n I K. - K, I i K. = n, and n, taking values from zero to (N - 2), i=l i=l 1 The evaluation of < > is performed on Appendix D, and i t gives: < K ± > = Y I 1 1 , P(n ; K; {K }) K p" 1 (1 - p)" 2 2 (12) n 1 =0 K=l {K 1} 1 1 1 with I K = K and I K, i = n,. i i 1 1 Therefore, j P ^ l " P){(1 " p) (N - i - 3) + 2}, i < N - 2 ^ > = I i = N - 2 The total resistance is then evaluated N-3 , < R > = R Q { I i p x ( l - p) {2 + (1 - p)(N - 3 - i) } e b l i=l + (N - 2) (pe b) N" 2} (14) After performing the sum over the island sizes: < R > / R Q = (1 - p) [2 +(1 - p)(N - 3)] [1 -(N - 2)(pe b) N" 3 + (N - 3) (pe b) N"" 2] x p e b b 2 - (1 - P ) 2 [l + p e b - ( N - 2) 2 ( p e b ) N ~ 3 + (2N 2 - ION + 11) (1 - pe ) x ( p eb)N-2 _ ( N _ 3 ) 2 (^bjN-lj Pe b + ( N _ 2 ) ( p eb ) N-2 ( 1 5 ) (1 - p e V - 13 -The graph of the ensemble average resistance versus the f i l l i n g f a c tor p i s plotted i n Figures (3a) and (3b). In the l i m i t of N •*• °°, the geometric series converges only for P < Pc = e _ b * At the p r o b a b i l i t y p c = e - b , the expression for the average resistance i s : N-3 N-3 9 9 < R > = R { I (1 - p)[2 + (1 - p)(N - 3)] i - I il (1 - P r + (N - 2)} C i=l i=l (16) Therefore, > < R C > = R q { ( 1 - p)[2 + (1 - p)(N - 3)] (N - 3)(N " 2) ( 1 - p ) 2 ( N - 3 ) ( N - 2 ) ( 2 N - 5) + ( N _ 2 ) } 6 At the thermodynamic l i m i t , of N + » and at the c r i t i c a l p r o b a b i l i t y p c , the average resistance per s i t e i s : < R c > 1 2 2 The c a l c u l a t i o n of the moment < Kj >, i s performed i n Appendix E, by performing the sums: N-2 n l /n - 1\ n „ -2 K ' 1 2 I K ± K. p \ l - p) 2 K l K i > = ^ £ , * K J n 1 =0 K=l {K } K ! K2J....K ! \ with I K, = K and I K. i = n,. (19) i i - 14 -Therefore, 1 < KA K. > = { (N - i - j - 4) (N - i - j - 5) (1 - p ) 2 + 6(N - i - j - 4) (1 - p) + 6} p i + j (1 - p ) 2 .8(1 + j , N - 3) + 2 ( l - p ) pN"3 6 ( i + j > N _ 3 ) + ^ [ ( N - 1 - 3) (1 - p) + 2] (1 - p) p 1 6(1, N - 2) + P N" 2 6 l j 6i,N-2 <20> | 1 i f m < n where 8(m, n) = I (_0 otherwise Using the above expression, the average of the square of the resistance can then be evaluated. This is done in Appendix F, and gives: ,2 N _ D2 rv V b ( i + j ) i j Therefore, 2 2 < R2 > - ( 1 1 2 P ) [(4A - C) S 3 + BS2 - 4BS4 + 4CS5 - AS^ ^o + I ( p e b ) N _ 3 (1 - p ) (N - 2) (N - 3) (N - 4) + (1- p ) {[2 + (N - 3) (1 - p ) ] T 2 - (1 - p ) T3} + (N - 2) 2 ( p e 2 b ) N ~ 2 (21) w h e r e B = 6(1 - p ) + (2N - 9) (1 - p ) 2 H l o o d a p p e a r s t o h a v e a n e r r o r i n < Kj >. - 15 -A = 6 + 6(N - 4) (1 - p) + (N - 4) (N - 5) (1 - p)' ,2 s v = I fflk ( p e b ) m ; C - (1 - p) N-4 m=0 T, = . I i k ( p e 2 1 3 ) 1 k i=0 S l " pe {1 - (N - 3) (p e b ) N ~ 4 + ( N - 4 ) (peV"""} (1 - P e b ) 2 bxN-3 (22) S2 = {pe b - ( p e b ) N - 3 (N - 3 ) 2 + ( p e V " 2 [2(N - 4) / , K 3 (1 - pe ) + 2(N - 4) - 1] - ( p e b ) N - 1 (N - 4 ) z + ( p e V } K 2 (23) S3 " pe {l - ( p e b ) N 4(N.- 3 ) J + ( p e 0 ) " J [3(N - 4 ) J + 6(N - 4 ) Z - 4] (1 - pe ) KN-3 + ( p e b ) N 2 [ -3(N - 4 ) 3 - 3(N - 4 ) 2 + 3(N - 4) - 1 ] + ( p e b ) N _ 1 (N - 4 ) J + 4(peD) + ( p e V } bx2- (24) S4 " pe 4 , KN-4 (1 - pe ) {l - (N - 3 T (peD) + (pe b ) N _ 3 [ 4(N - 4 ) 4 + 12(N - 4 ) 3 + 6(N - 4 ) 2 - 12(N - 4) - 11] - (pe b ) N _ 2 [ 6(N - 4 ) 4 + 12(N - 4 ) 3 - 6(N - 4 ) 2 - 12(N - 4) + 11] + ( p e b ) N _ 1 [4(N - 4 ) 4 + 4(N - 4 ) 3 - 6(N - 4 ) 2 + 4(N - 4) - 1] - ( p e b ) N (N - 4) H + l l ( p e D ) + l l ( p e V + ( p e V } bN2 bx3 (25) - 16 -(1 - pe ) + (pe b) N - 3[5(N - 4) 5+ 20(N - 4) 4+ 20(N - 4 ) 3 - 50(N - 4) -26 -20(N -4) 2] + (pe b) N~ 2[10(N - 4 ) 5 + 30(N - 4 ) 4 - 60(N - 4 ) 2 + 66] + (pe b) N - 1[10(N - 4 ) 5 - 20(N - 4) 3+ 20(N - 4 ) 4 - 20(N - 4) 2+ 50(N - 4)-26] + (pe b) N[5(N - 4 ) 5 + 5(N - 4 ) 4 - 10(N - 4) 3+ 10(N - 4 ) 2 - 5(N - 4) + 1] + ( p e b ) N + 1 ( N - 4 ) 5 + 26(peb) + 66(pe b) 2 + 26(pe b) 3 + (pe b) 4 } (26) „ 1 r 2b , 2KN-2 ... o N2 T 2 = 2 b 3 t p e " ( p e } ( N " 2 ) (1 - pe ) + ( p e 2 b ) N _ 1 [ 2 ( N - 3 ) 2 + 2(N - 3) - 1] - ( p e 2 b ) N (N - 3 ) 2 + (pe 2 b) 2} (27) (1 - pe ) + ( p e 2 b ) N ~ 2 [3(N - 3 ) 3 + 6(N - 3 ) 2 - 4] + ( p e 2 b ) N _ 1 [-3(N - 3) 3 - 3(N - 3 ) 2 + 3(N - 3) - 1] + ( p e 2 b ) N (N - 3) 3 + 4 (pe 2 b) + (pe 2 b) 2} (28) In the limit of N •* °°, the geometric series a l l converge only for p < e~2b. Consequently, the variance w i l l have a divergence at the probability p c^ equal to e - ^ . Considering the leading terms in the variance, at the thermodynamic 2 —2b limit and at p c = p c = e , the variance diverges as: < R2 > - < R >2 •*• N4 (1 - p ) 2 IZ Cj^ (29) - 17 -The behavior of the variance with the p r o b a b i l i t y p i s shown i n Figure (5) and the pl o t of the variance versus the array size i s i n Figure(6). 2.3 Ensemble with Fixed Probability p and Line Bent into a Circle This ensemble consists i n taking the l i n e a r array, with fixed p r o b a b i l i t y p, and bending i t into a c i r c l e . The case of n^ = 0 ( n 2 = N) and n^ = N (n^ = 0) w i l l be excluded. A s i m i l a r ensemble i s described by David and Barton (DAV62). Therefore, the p r o b a b i l i t y of having n^ d i e l e c t r i c s d i s t r i b u t e d i n {K.£ } c l u s t e r s of size i i s given by: »(n i ; K , { K . } ) = p" 1 (1 - p ) " 2 | ^ n 2 - IV K , K - 1 / K , ! K ' . . . K ! * / 1 2 n x With the above expression, the moment < KJ; >.can be calculated. This i s done i n Appendix G. The f i n a l expression for < K-^ > i s : < K > = i ' N p 1 (1 - p ) 2 for i N< N ' 2 UP*" 1 ( i -P) F O R I = N ~ U (31) The t o t a l resistance i s then evaluated - N(l - P) 2 Y K p e V + N(N - 1)(1 - p)(pe b ) N 1 R o i=l Therefore: - 18 -- 1 9 -Figure 6 : Dependence of the variance (< R2 > - < R >2)/N2R^ on the array size for three different values of p. Ensemble with fixed probability p in a l ine . - 20 -< j p = N ( 1 _ p )2 Pe b {1 - (N - 1) (peV" 2 + (N - 2) (peV" 1} o (1 - pe ) + (1 - p) N(N - 1) ( p e b ) N - 1 . (32) Again, as N •*• 0 0 the average resistance diverges at a c r i t i c a l probability Pc = e" b. Analysing the average resistance at p c one obtains: N-2 < R > = N(l - p) 2 I i + N(N - 1) (1 - p) R A 1 o i=l Therefore, <-|2- N ( i - p ) 2 <N - 2><N - " + N (N - 1) (1 - P) (33) o Figure (7) shows the graph of the ensemble average resistance versus the probability p. At the thermodynamic limit of large N, the average resistance per site goes as: %|^4N 2 (1 - P) 2. (34) The calculation of the moment < Ki Kj > is done in Appendix H. The expression for < Kj > is then given by: < K± K > = N p i + j (1 - p ) 3 {(1 - p) (N - i - j - 3) + 2} 0(i + j , N - 2) + N p i + J (1 - p) 2 6 ( ± + ^ N _ 2 ) + N p i ( l - p) 2 9(1, N-l) 6 ± J + N (1 - p) p 1 « ( ± f N _ ^ 6 ± J (35) - 21 -Figure 7 Dependence of the ensemble average resistance per site on the probability p. Ensemble with fixed probability p and line bent into a circ l e (N = 100). - 22 -The average of the square of the resistance can then be evaluated. And the f i n a l expression which i s i n Appendix I i s : 2 < R2 > = f j ( 1 " P ) 3 {4[2 + (N - 3) ( 1 - p)] S3' R0 - [2 + (N - 3) ( 1 - p)] Sj} + ( 1 - p ) 3 { s ^ ( l - p) - 4 ( 1 - p) S^} + N ( p e b ) N " 2 ( 1 - p ) 2 (N - D ( N - 2)(N - 3) 6 + N(l - p ) 2 T 2 + (N - l ) 2 ( p e 2 b ) N - 1 N(l - p) (36) N-3 where = I m" (pe ) m ; / m=l Tl = I i 2 ( P e 2 b ) X i=l The value of each sum being equal to: S1 " ^ h ? t l - ^ - 2 ) ( P e b ) N - 3 + (N - 3 ) ( P e b ) N " 2 } 1 ( 1 - P e b r S* = ^ b 4 { 1 - ( P e b ) N " 3 (N - 2 ) 3 + ( 1 - pe ) + ( p e b ) N ~ 2 [3 (N - 3 ) 3 + 6(N - 3 ) 2 - 4] + ( p e b ) N _ 1 [-3 (N - 3 ) 3 - 3(N - 3 ) 2 + 3(N - 3) - 1 ] + ( p e b ) N (N - 3 ) 3 + 4pe b + ( p e b ) 2 } - 23 -S4' = P e t ? b 5 {1 - (N - 2 ) 4 ( p e b ) N - 3 (1 - pe ) + ( p e b ) N ~ 2 [4(N - 3 ) 4 + 12(N - 3 ) 3 + 6(N - 3 ) 2 - 12(N - 3) - 11] - ( p e b ) N _ 1 [6(N - 3 ) 4 + 12(N - 3 ) 3 - 6(N - 3 ) 2 - 12(N - 3) + 11] + ( p e b ) N [4(N - 3 ) 4 + 4(N - 3 ) 3 - 6(N - 3 ) 2 + 4(N - 3) - 1] - ( p e b ) N + 1 (N - 3 ) 4 + l l ( p e b ) 2 + l l ( p e b ) + ( p e b ) 3 } „, 1 r 2b , 2b NN-l ,„ n 2 x T 2 = ~ 2bTT t p e " ( p e > (N - 1) + (1 - pe ) + ( p e 2 b ) N [2(N - 2 ) 2 + 2(N - 2) - 1] - ( p e 2 b ) N + 1 (N - 2 ) 2 + ( p e 2 b ) 2 } Once more, as N + » the variance w i l l diverge at the p r o b a b i l i t y ~~2 b 21} p = e . The variance at p = e and at the l i m i t of large N, w i l l C l C l tend to the l i m i t i n g value: < R 2 > - < R > 2 1 ( .2 ..2 2~1 ' V 3 ( 1 " P c ) N * N R 1 Using the expressions for < R^ > and < R >, the variance was calculated numerically. The r e s u l t of this c a l c u l a t i o n i s shown i n Figure ( 8 ) . 2.4 Comparison Between Ensembles Figures (3a) and (3b) shows the t o t a l resistance per s i t e plotted against the p r o b a b i l i t y p for the approaches i n sections 2.1 and 2.2. The two ensembles are related by the condition that (n^/N) = p. For small p r o b a b i l i t y p, the two ensembles agree. / - 24 -10", {0 - f 1 1 r 1 1— 0.2O 0.22 0.24 0.26 0.28 0.30 PROBABILITY P Figure 8 : Dependence of the variance (< R2 > - < R > 2)/N 2R 2, on the probability p. Ensemble with fixed probability p and line bent into a c i rc le (N = 100). - 25 -The numerical value of the contribution to the resistance per s i t e due to each is l a n d size was calculated for the two approaches. The r e s u l t of i t i s shown i n Figures (9a), (9b) (9c), (9d), for an array of size N = 100. D i f f e r e n t values of p were considered. For f i n i t e N and large p the two approaches disagree. This discrepancy arises because the ensemble with fixed p r o b a b i l i t y p allows d i e l e c t r i c c l usters larger than the corresponding n^, of the ensemble with fixed n ^ The contribution of these large islands i s dominant i n the t o t a l resistance. Comparing the variances of approaches i n Sections 2.1 and 2.2, i t was observed that the variance for the ensemble with fixed p r o b a b i l i t y p has a higher value than the one with f i x e d n\. Following we quote the numerical values of the variance for the two ensembles for N = 100. Variance Ensemble f i x p Ensemble f i x n 1 nj_/N = 0.27 1.1 x 10 5 4.5 nj/N = 0.25 4.6 x 10 2 2.3 n x/N = 0.24 5.9 x 10 1.7 This behaviour i s again a r e s u l t of the contribution of the large i s l a n d s . The exact way that the resistance diverges, for the ensemble with fixed n i , was not obtained. The reason i s that i t was not possible to perform the sum over the i s l a n d sizes i n the expression for the resistance. Therefore, the thermodynamic l i m i t was taken before the sum was done. The ensembles of approaches 2.2 and 2.3 allow the comparison of two ensembles with f i x e d p but d i f f e r e n t boundary conditions. Comparing equations (13) and (31), i t can be seen that the way of putting an i s l a n d - 26 -• » «o *c te ae c n .0 H ISLAND SI7C B L A N D SJ7C ( C ) o . ae z 40 t0 B L A N D size o ac z .e u B L A N D S17C Figure 9: Dependence of the value of (< RI >/NRQ) on the island s i z e . < RI > i s the contribution to < R > from a l l the islands of size I. Ensemble with fixed n i (dashed l i n e ) and fixed p r o b a b i l i t y p ( s o l i d l i n e ) i n a (a) p = 0.45, nj (b) P = 0.48, nj (c) p = 0.50, nj (d) p = 0.52, rxi l i n e (N = 100). = 45 N = 100 = 48 N = 100 = 50 N = 100 = 52 N = 100 - 27 -of s i z e i i n a l i n e i s d i f f e r e n t from the one i n a c i r c l e . Equations (13) and (31) d i f f e r because the embedding of an island of size i i n a l i n e of length N i s ( N - i + 1 ) / N whereas i n a c i r c l e i t i s 1 (embedding of an is l a n d of size i i s the number of ways i t can be put i n divided by the t o t a l number of s i t e s ) . In the l i m i t of N * 0 0 these two embeddings are both unity. However, i n order to obtain the average value of the r e s i s t -ance, a summation over i has s t i l l to be performed. If this summation i s done before the l i m i t N •*• 0 0 i s taken, then these two ensembles produce d i f f e r e n t r e s u l t s . In p a r t i c u l a r , we have noted that for large N the l i n e problem gave the r e s u l t (18) « R > / N R « (1/6)(1 - p ) 2 N 2 ) whereas o c the c i r c l e problem gave the r e s u l t (34) ( < R > / N R q •> (1/2)(1 - p £ ) 2 N 2 ) . A s i m i l a r difference occurs for the variance. I t i s to be remembered that equation (13) i s for f i n i t e i . Thus by considering a l l three ensembles we have demonstrated, for this p a r t i c u l a r problem, the differences that occur for fixed n^ and fixed p and for d i f f e r e n t boundary conditions. - 28 -CHAPTER THREE COMPUTER SIMULATIONS I t transpires that for the computer simulations i t makes l i t t l e d i f f e r e n c e which boundary conditions, metal ends or c y c l i c boundary conditions, are used nor whether the randomness i s generated according to the ensemble of paragraph 2.1 or 2.2 of the previous chapter. We quote here the r e s u l t s for metal ends and f i l l i n g according to a fixed p. Figure (10) shows < R >^/NR0, the sample mean resistance per s i t e for M computer simulations. Computer simulations for d i f f e r e n t N were done; they herald a Jtn 2^ divergence of < R >M at p = p c as N becomes larger (PAL84). The value of < R >M at p c i s orders of magnitude smaller than the corresponding ensemble value obtained by approaches 2.1 or 2.2 of the previous sections. The question then arises as to the nature of the r e l a t i o n between the computer simulation of this chapter and the ensemble theory of the previous one. A previous work (PAL84) describes the problem and what should be the representative ensemble for the computer simulations. A summary of the r e l a t i o n between the computer simulation and the ensemble theory i s described i n the following paragraph. What the computer program does i s generate repeatedly a set {K^ } N v 2 b i and then calculates R = l i e K.. I t then averages this R over M i=l 1 computer runs. I t i s found that for a given M the computer each time produces zero for K i for i larger than some i m a x « In the computer - 29 -60 - i 50 40 A A T. AS v 30 H 20 10 0.1 0.2 0.3 0.4 PROBABILITY P 0.5 0.6 Figure 10: Sample mean resistance per s i t e < R >j^/NR0 for M runs versus the p r o b a b i l i t y p i n a l i n e (M - 1000, N = 1000). For each p the 1000 runs were done 3 times. - 30 -averaging over M runs, islands of size greater than i m a x w i l l not provide t h e i r very large contribution to the average resistance. An acceptable an a l y t i c d e s c r i p t i o n of the computer simulations i s to cut off the sum over i i n the ensemble approach, at some imax* I n this way, the e f f e c t of rare events, islands bigger than i m a x » a r e n o t taken i n t o account. The an a l y t i c description approaches the ensemble one when the number of runs i s approximately equal to M = eN-£nN (pAL84). In systems l i k e t h i s , where rare events dominate the ensemble arithmetic mean, we do not expect the sample mean to agree with the ensemble arithmetic mean; since as we found the sample mean does not pick up the rare events. The question then arises as to how i n the ensemble approach of Chapter 2 we are to predict t h e o r e t i c a l l y what experiment (ei t h e r computer simulation or laboratory measurements) should expect. I t has been suggested that the ensemble geometric mean, since i t i s not so dominated by rare events might be more comparable to the experiment geometric sample mean. Unfortunately, by approach of Chapter 2 we were not able to calculate the ensemble geometric mean. However, as a check on the relat e d suggestion that the computer simulation geometric means w i l l be less divergent than the arithmetic means, we i n fac t obtained also the computer simulation geometric mean. Figure (11) shows the geometric mean over M runs, of the resistance per s i t e against the logarithm of the array s i z e . I t appears that the geometric mean grows at p = p c as j£n(N), which i s a slower growth than the one of the arithmetic mean at p = p c which grows as In2 (NM) (PAL84). By comparing three groups of 1000 runs the geometric mean shows smaller f l u c t u a t i o n s i n comparison with the arithmetic mean. The pl o t of the cumulative average resistance per s i t e against the - 31 -25 - i 20 -15 -i 10 -i A A 6 A A A A A A A —1— 7.5 4.5 5.5 6 6.5 InN 8 5 Figure 11 : Computer simulation geometric mean over M runs versus l n ( N ) . Ensemble with fixed p r o b a b i l i t y p i n a l i n e . p = 0.5 and M = 1000, for each point such that the f l u c t u a t i o n can be observed. - 32 -number of runs for a set of 1000 runs i s shown i n Figure (12). From the p l o t of the cumulative average resistance per s i t e against the number of runs, the behavior of the average value of the resistance per s i t e can be analysed. I t can be seen that the mean of the resistance per s i t e exhibits sudden jumps followed by decreases. These jumps are caused by the appearance of large islands i n a c e r t a i n computer simulation. Then as the number of runs increases the importance i n the average of those large islands decreases. These decreases w i l l continue u n t i l another jump occurs. A s i m i l a r behavior has already been observed by Montroll and Shlesinger (MON83). In discussing income d i s t r i b u t i o n s , Montroll and Shlesinger (MON83) introduce the process of a m p l i f i c a t i o n , a m p l i f i c a t i o n of a m p l i f i c a t i o n , etc. S t a r t i n g with a basic d i s t r i b u t i o n , they can then construct a new d i s t r i b u t i o n which allows for the p o s s i b i l i t y of continuing le v e l s of a m p l i f i c a t i o n . Each amplifier class has the same basic d i s t r i b u t i o n function but the mean value of the quantity measured i s amplified N times with a small p r o b a b i l i t y A. Analogously, i n the cumulative average resistance per s i t e the mean of the resistance per s i t e suffers amplifications or jumps equal to e*5 and p r o b a b i l i t y X = p. This behavior i s also r e f l e c t e d i n the histogram of &n(R/NR0), Figure (13), where for large values of the resistance per s i t e the histogram has spikes at constantly spaced i n t e r v a l s . A histogram was done with a d i f f e r e n t value of b and i t was checked that the distances between the spikes were equal to the new value of b. In other words, each considerable increase i n the resistance i s mainly me^, where m i s an integer. - 33 -p=0.50 N=1000 NUMBER OF RUNS Figure 1 2 Computer simulation cumulative average resistance per s i t e for M runs. Ensemble with fixed probabilty p i n a l i n e . - 34 -HISTOGRAM OF In OF RESISTANCES PER SITE 8 0 n In O F R E S I S T A N C E S PER SITE Figure 13: Histogram of the logarithm of the sample mean resistance per s i t e for M runs. Ensemble with fixed p r o b a b i l i t y p i n a l i n e (M = 1000, N = 1000, p = 0.5). In order to be able to exhibit the whole histogram on one diagram, the horizontal scale i s in fa c t x with x - U n « R >M/NR0) - (-0.5)]/0.1 + 1. - 35 -CHAPTER FOUR CONCLUSION The s t a t i s t i c s of the problem of a random series of metal-insulator-metal junctions was studied considering three d i f f e r e n t ensembles. I t was found that the e f f e c t of large insu l a t o r c l u s t e r s was dominant i n the behavior of the ensemble average resistance and v a r i ance. Therefore, the main difference between the ensembles with fixed p r o b a b i l i t y p and fixed nj_ > d i s t r i b u t e d i n a l i n e a r arrangement, was due to the l i m i t a t i o n on the second ensemble to pick up larger island s i z e s . When comparing two ensembles with fixed p r o b a b i l i t y p, of having an i n s u l a t o r occupying a s i t e , but one i n a l i n e a r arrangement and the other one i n an arrangement of a l i n e bent into a c i r c l e , a discrepancy i s observed i n the way the ensemble average resistance diverges for N -»• 0 0 at p = p c. This f a c t i s due to the d i f f e r e n t embeddings i n the two ensembles. If the l i m i t N * 0 0 i s taken before the summation of i s l a n d contributions for d i f f e r e n t size i s l a n d s , then the two approaches agree. The divergence of the ensemble average resistance and variance at the l i m i t N •*• 0 0 was observed for the three ensembles at p = p c = e - b and p = p c^ = e - 2 ^ , r e s p e c t i v e l y . This behavior i s also t y p i c a l of the s o - c a l l e d G r i f f i t h s s i n g u l a r i t i e s (GRI69). In these, as i n our case, i n the l i m i t N ->• 0 0 the coherence length ( i n our case, the average d i e l e c t r i c i s l a n d size) does not go to i n f i n i t y u n t i l p = 1, i n one-dimensional systems. Before that, there occurs a divergence i n the resistance at p c« - 36 -As a suggestion for future work, the temperature dependence In the tunneling r e s i s t i v i t y i n a randomly-filled chain problem, should be considered. I t may be that at non-zero temperatures the behavior of the divergence i s s i g n i f i c a n t l y modified. Another i n t e r e s t i n g problem i s the extension of the present problem to a two-dimensional one, which would be a better approximation to the simulation of granular metals. In conclusion we would l i k e to point out that this project has made a s i g n i f i c a n t contribution to the study of a one-dimensional random system which displays a s e n s i t i v i t y to s t a t i s t i c a l l y rare events. - 37 -REFERENCES ABE75 Abeles, B., P. Sheng, M.D. Coutts, and A. A r i e . Advances i n Physics, 24_, 407 (1975). DAV62 David, F.N. and D.E. Barton. Combinatorial Chance, Charles G r i f f i n , London (1962). DUK69 Duke, C.B. Tunneling i n Solids, S o l i d State Physics, Supplement 10, Academic Press. New York, London (1969). GRA Gradshteyn, S./I.M. Ryzhlk. Table of Integrals Series and Products, p. 4, equation KR64 (71.1). GRI69 G r i f f i t h s , R.B. Phys. Rev. L e t t . 23, 17 (1979). GUN66 Gundlach, K.H. So l i d State Electron. j)» 9 4 9 (1 966). H0L51 Holm, R. J . Appl. Phys. 22, 569 (1951). MON83 E l l i o t t , W. Montroll and Michael F. Shlesinger. Journal of S t a t i s t i c a l Physics, V o l . 32, No. 2, 1983. M0041 Mood, A.M. Annals of Mathematical S t a t i s t i c s , V o l . 11-12, 1940-1941. PAL84 P. Palffy-Muhoray, R. Barrie, B. Bergersen, I. Carvalho and Freeman, M. Journal of S t a t i s t i c a l Physics, V o l . 35, Nos. 1/2, 1984. SOM33 Sommerfeld, A. and H. Bethe. Handbuch der Physik von Geiger and Scheel ( J u l i u s Springer-Verlag, B e r l i n , 1933), V o l . 24/2, p. 450. WIL62 Wilks, S.S. Mathematical S t a t i s t i c s , p. 144, Princeton, 1962. - 38 -APPENDIX A CALCULATION OF THE MOMENT < K± > FOR FIXED NUMBER OF INSULATOR "ATOMS" DISTRIBUTED IN THE CHAIN ru, - 1 < K i > i ) | i l K 1 , K ^ . . . S l | M - S ^ 1 . with 0 < n x ^ N - 2, O ^ K ^ r i j , O ^ i ^ n ^ {K± } = {K ±, K 2 > .. .K } I i K. = n, and I K. = K. i 1 1 i 1 But: I H P 1 K ' K. ! K„ ! .. .K ! V K J \ {K.} V * 2 l — \ -K T (K - 1)! ( n2 " 1] ' • K,!...(K,-!)!...K n ! \ K / 1 { K- } ^ i . . . ^ . -with I i K, = n, and I K. = K. i 1 i 1 Making the change of va r i a b l e s : K' = K - 1, K'. = J K! / S " *\ K_ = K ± - 1 i f j = i .K. otherwise K,! K0!...K ! I K j i Tl, \ K V KM / n2 " 1 r ,f, K\!...K!1...K* ! V K ' 1 1 l n, v {K!} V 1 with {K|} = {K{, K»,...,K^}, I K' = K - 1 and £ i K' = ^ - i . - 39 -Making use of the procedure given i n Wilks (Wil62), the above sum can be written as: Y K'! _ ( N 1 - 1 - 1 ) ! r£,i K!! K!!...K' ! (K - 2)! (n. - i - K + 1)! ' ( A 2 ) with I KI = K - 1 and I i K! = n. - i . i 1 i 1 Therefore I t i s worth n o t i c i n g that: - The range of K i s from K = 1 to the smaller - 1 and n^ - i + 1; - If n^ =0, then K = 0 which corresponds to < > =0; - For a given n^ ^ 1, the term K = 1 i s zero unless i = n^ .. Consequently, f o r a given nj and i < n^, the f i r s t moment i s given by: Settin g K" = K - 2 and not i c i n g that K" ^ n 2 - 3; the sum over K then reduces to: - 40 -i /N - a.\ K i Q U " 3 - K 7 V K " ~ "2)s i n c eq I 0 ( : ( p - % = n + p > ( G R A 6 5 ) ; where m is a natural number. It follows that ( n2 ~ X ) /n - -i - ^ < K l > - 7 H i T i r - N ( n . - i ) - l « * l < » - 3 In the case of i = n^, the smaller of (n 2 - 1) and (n^ - i + 1) is (n^ - i + 1), unless n 2 = 2 when they are equal. Therefore, < K > = _ 1 f n2 " M /-A ( n2 " 1 } /N - 2\ V 1 / \ - U " /N - 2 J * I n , The two cases, i < n^ and i = n^ can be combined as: < K. > = 2\ V - ' i ^ w i t h 1 < N - 3 <A4> Notice that i f nj = N - 2, < K± > = 0 unless i = N - 2 when < Ki > = 1. - 41 -APPENDIX B CALCULATION OF THE MOMENT < K ± R\. > FOR FIXED NUMBER OF INSULATOR "ATOMS" DISTRIBUTED IN THE CHAIN n 2 - 1 with I K = K, ^ i K = n p 0 < KV< n and 0 v< i s< n,. i i 1 1 Separating out the diagonal and off-diagonal terms i n (BI), the sum over {K^} can then be written as: £ K! £ K! 2 { K ± } K X ! K 2 ! . . . K N I ! K i K j = 6 i j { K ±} K l ! ! . . .Kn ! K i + ( 1 " V K ] [ ! V . . . K N ! K i K r <B2> with I K = K and £ i K , = n.. i i 1 1 Consider the diagonal term i n equation (B2) I K ! -> - K ! K -{Kj} K X ! ^ . . . ( K , - ! ) ! . . . ^ ! with £ K = K and £ i K = n.. i i 1 1 - 42 -Making the change of variables: K1 = K - 1 f K - 1 i f j = i K'. = j J J L K, o t h e r w i s e ' J e q u a t i o n (B3) i s e q u a l to: K' ! (K1. + 1) KM K' v y i t7 y 1 t.'llK!!...K!I...R' ! r , i KM .. .K! !.. .K' ! t K i l 1 1 " x 1 i n v y KM r M x rv,S K.M ...KM ...K' ! with I KJ = K - 1 and T i K'. = n, - i . i 1 f 1 1 Perform once more, the change of variables in the f i r s t equation of the right-hand side (B4) such as: K" = K' - 1, K" = j f K'. - 1 i f j = i J . KI otherwise. Equation (B3) is then written as: v(v - n V K"! y KM V ; K;,!...K"!...K" ! r f, KM ...KM ...K' ! » t ^ l 1 i ^ {K{1 1 i n x wi th I K^ = K - 1; £ i K| = ^ - i , i i £ KV = K - 2 and 5! 1 K £ = ,n1 - 2 i . From Appendix A, the value of the sum over {K^ } can then be obtained. And equation (B3) reduces to: - 43 -y K 2 K! fn - 21 - l \ /n - i - 1\ {K±} V V T T T " < K ( K " l ) I K - 3 J + K l K - 2 ) ( B 5 ) Analogously, the off-diagonal term i n (B2) i s : ( 1 - 6 , . ) I - K. K. 2 {K.} K.! K.I...K ! J 1 i J 1 2 n^ /n - i - j - l \ = K ( K - 1 ) ( K _ 3 (B6) With (B5) and (B6), the expression for (Bl) i s then: < K K > = {I K(K - 1) I " 1 K \ n, - 21 - 1\ / n 2 - 1' K - 3 / V K n i ,'n. - i - j - 1\ /n„ - l \ (1 - 6. .) • i ) (B7) with 0 ^ K < n j . Consider, f i r s t , the sum over K for the off-diagonal term i n (B7): n - 44 ->9 1) (B8) Sett i n g K" = K - 2 and rearranging terms, then equation (B8) i s given by: " I - 1 " ! /n - i - j - A /„ - 3 (°2 " '><»2 " 2> I ( „ - i - j - r ( V , K"=0 X V with the condition 0 ^ K" ^ n^ - i - j . By r e l a t i o n (A3), the f i n a l expression for (B8) i s /n, - i - j - 4\ = ( n 2 - l ) ( n 2 - 2) f j (B9) Consider now, the sums over K for the diagonal terms i n (B7). In analogy with equation (B9), the f i r s t term i s given by: n l In - 21 - l \ / n - 1 \ J o K ( K - 1 } v K _ 3 ; v * ; /N - 21-4 = ( n 2 - l ) ( n 2 - 2 ) ^ ^ j (BIO) - 45 -The second sum over K, i n equation (B7), can be s i m p l i f i e d as: ^ o ^ ' H * 1 * - 2 ' 1 ) 0 ^ - 0 ( b u ) Setting K' = K - 1 i n ( B l l ) and using r e l a t i o n (A3), equation ( B l l ) i s given by: Therefore, X k ( ? * - "2') C 2 * " ' ) = < n 2 • n f V - " 1 ' ) ( B u > With the re l a t i o n s (B9), (BIO) and (B12), the expression for < K ± Kj > i s : < h «j > - «° 2 - » ( N - 4 : T 4 ) - ( N ^ - V ) ! 013) - 46 -APPENDIX C CALCULATION OF < R 2 > FOR FIXED NUMBER OF INSULATOR "ATOMS" DISTRIBUTED IN THE CHAIN The average of the square of the resistance i s given by: < R 2 > - R20 I I U e b ( i + j ) < K ±K >, (Cl) with 0 < i + j < n j . From equation (B13), the < R2 > i s given by: < R 2 > _ ( n 2 1 ) ( n 2 " 2 ) n b(i+j) / N - i - j - 4 ' 2 / N - 2 \ L L ±3 e n - 4 R I I n • 1 j V 2 . ( n 2 " l ) v .2 2bi /N - i - 3\ ._ 0. + /N - 2^ I 1 6 [ n 2 - 3 1 ( C 2 ) n l Rearranging terms and making a change of variables i n (C2) such that: 9 i + j = m and i - j = n, the expression for < R > can be written as: . Jl ^ (n„ - l ) ( n 0 - 2) , 2 2 N , /„ ,\ < R > _ 2 2 _ y y (m - n ) bm / N - m - 4 D 2 " | N - 2 \ 1 1 4 e l n - 4 RQ ^ ^ 1 \ 2 (n 2 1) „ 2bi I N - i - 3 "l wi th - m ^ n ^ m and 0 ^ m ^ n^ 47 -Performing the summation over n, one obtains: < R > = ( n 2 ~ 1 ) ( n 2 " 2 ) r (4 m3 - m) bm ( N - m - 4 R o ( n x J \ 1 / . ( n 2 " 1 } v .2 2 b i / N - i - 3\ + i n - 2^ I 1 e I n - i ) ( C 3 ) n l w i t h m = i + j , O^m^n^ and 0 ^ i ^ n±, - 48 -APPENDIX D CALCULATION OF < > FOR FIX PROBABILITY p OF HAVING AN INSULATOR OCCUPYING A SITE IN A LINE The moment < > i s given by: < h>- I I j ^ r r r f b T \ p 1 a - p> 2 . nj K {Kt} 1 V K2 n l with [ K. = K, I i K ± = n L, 0 < K < n x , i i 0 ^ i ( nj and 0 ( ii j s< N - 2 (Dl) The sum over {%} and the subsequent sum over K i n equation (Dl) has already been performed i n Appendix A, equations (A2) and (A4). Therefore: < K. > = ^ ( n 2 - 1) IN -n*-*l p" 1 (1 - p ) " 2 ' \ ( D 2 ) with 0 < nl ^ N - 2 and i ^ N - 3. Equation (D2) can be rewritten as: n, N-n,-2 / . . \ < K i > = I p ( l - p ) (N - i - 3) ( ~ V i nl \• 1 1 / n, N-n,-2 / „ . ~ \ + 2 I p 1 (1 - p) 1 / N l^SiJl (D3) - 49 -with i ^ N - 3 and 0 ^ nx ^ N-2. Making the change of variables in (D3), nj = - i and rearranging / n _ i _ A the limits in the sum over n! such that . 1 * 0 and 1 V ni I 'N - 1 - 3^ , 1 * 0 , one obtains nj 1 ' i m • / N-i-4 n, / „ . , < K. > = (1 - p) 2 p l ( l - p) 1*" 1- 4 I ( r E - ) 1 N " ^ 4 (N - i - 3) nj=0 + 2(1 - p) p ^ l - p ) ^ 1 " 3 I ( r ^ - ) N-i-3 _ n{ / N _ ± _ 3 nj=0 with i < N - 3. (D4) By the Binomial series definition the sums in (D4) can be calculated. Then: < K ± > = p 1 (1 - p) {(1 - p) (N - i - 3) + 2}, for i < N - 2. (D5) For the case of i = N-2, < K^ > is equal to p N - 2 . APPENDIX E CALCULATION OF < K. K. > FOR FIX PROBABILITY p OF HAVING AN INSULATOR OCCUPYING A SITE IN A LINE r r r K' f °2 " 1 \ H l n 2 ~ 2 < K K > = I I J K 1 K I V K K i K i P ( 1 " P ) ; KM...K ! \ / J 1 L l J 1 wi th 0 ^ n. « N - 2, 0 ^ K < n , I K = K, i 1 i K ± = n l f 0 ^ i < n 1 and {K± } = {Kj, K 2 K N }. ( E l ) The expression for the sums over {K^ } and K can be obtained from Appendix B. Equation (El) can then be written as: < K 4 K. > - | • « „ 2 - 1) („ 2 - 2, I « " L-J- ") • + o. . (n„ - 1) _ i j 2 V 3 \ n, n 0-2 J } P 1 (1 " P) 2 , with 0 < n x ^ N - 2. (E2) Taking i n t o consideration that: "1=0 i f eit h e r r > n or r < 0; equation (E2) i s rewritten as: - 51 -< K. K. > = {I ( n 2 - 1) ( n 2 - 2) ^ " ^'J' ^ p" 1 (1 - p)" 2"'} x 9 ( i + j , N - 3 ) + 2 ( l - p ) p N " 3 6 ( 1 + j ) N _ 3 ) + + I ( n 2 - 1) (^N l^.'/j P1"1 (1 " P ) " 2 2 Q(i> N - 2) + 6 i j 6i,N-2 P N" 2' <E3> 1 i f m < n w i t h O s < n i < N - 2 and 9 ( m , n) - | Q o t h e r w i s e . Each sum over n^ , i n equation (E3), w i l l be evaluated separately i n the next paragraphs. The f i r s t sum over n j , i n equation (E3) can be rewritten as: X P (1 " P) (*2 " 1) ( n 2 - 2) ^ ^ J 4 j = n i N - n i ~ 2 / m • _ • I P (1 - P) {(N - i - j - 4)[(N - i - j - 5) ( I _\ n l \ 2 with 0 ,< ni O - 2 -Consider the f i r s t term i n the right-hand side of equation (E4). After making the change of variables n'^ = n^ - i - j and rearranging the l i m i t s of the sum, such that the binomial c o e f f i c i e n t i s not equal to zero one obtains: - 52 -(N - 1 - j - 4) (N - i - j - 5) (1 - p , « " ? (» - . ^ . V . - J 6 ' (N - i - J - 4) <N - 1 - ] - 5) ( 1 - p ) N " 2 f r ^ - p - ) 1 4 = (N - i - j - 4) (N - i - j - 5) p i + j (1 - p ) 4 (E5) Using the same procedure used to evaluate expression (E5), the others sums i n the right-hand side of equation (E4) are given by: 6(N - i - j - 4) (1 - p ) 3 p i + J ; (E6) and Considering equations (E5), (E6) and (E7), equation (E4) can then be evaluated, being equal to: I p 1 (1 - P) 1 <n2 - 1) ( „ 2 - 2 ) t" - ' 'J' 4 n^ \ 2 {(N - i - j - 4) [(N - i - j - 5) (1 - p ) 2 + 6(1 - p)] + 6} (1 - p ) 2 p i + j (E8) - 53 -Now, the second sum i n the expression for < Kj >, equation (E3), w i l l be calculated. Therefore v , . /N - i - 3\ n l "2 " 2 I ( n2 - 1 } ( n 2 - 3 ) P ( 1 " P ) V1 P j \ "1 i ' + 2 2 with n* = n - i . ( 1 - P) N 3 1 1 Therefore v . /N - i - 3\ n l n ^ J ( N 2 - 1 } ( n 2 - 3 J P ( 1 " P ) [(N - i - 3) (1 - p) + 2] (1 - p) p 1 , (E9) with O ^ n ^ ^ N - 2 . With equations (E8) and (E9), equation (E3) can then be evaluated. I t i s given by: - 54 -< K ± K > = {(N - i - j - 4) [(N - i - j - 5) (1 - p ) 2 + 6(1 - p)] + 6} (1 - p ) 2 p i + j 6(1 + j , N - 3) + 2 ( l - p ) p N - 3 « ( 1 + J > H . 3 ) + [(N - i - 3) (1 - p) + 2] (1 - p) p 1 6 6(1, N - 2) + 6 l j 6i,N-2 P N-2 (E10) APPENDIX F CALCULATION OF < R 2 > FOR FIX PROBABILITY p OF HAVING AN INSULATOR OCCUPYING A SITE IN A LINE < R2 > - R2D {I I l j e b ( i + J } < K K >}, i j with O ^ i ^ N - 2 and <*L> - Hi i e b ( i + J > {p i +J (1 - p ) 2 [6 + 6<N - ± - j - 4) (1 - p) + R o 1 J (N - i - j - 4) (N - i - j - 5) (1 - p) 2} 9 ( i + j , N - 3) + 2 p N " 3 (1 - p) 6 ( . + j j N _ 3 ) + p 1 (1 - p){2 + (N - i - 3) (1 - p)} 6 ±. 9(1, N - 2 ) + 6^ p N ~ 2 6 i ) N_ 2} ( F l ) In the next paragraphs each term i n equation (Fl) w i l l be considered separately. a) I I i j ( p e b ) i + j (1 - p ) 2 {6 + 6(N - i - j - 4) (1 - p) i j + (N - i - j - 4) (N - i - j - 5) (1 - p ) 2 } , with i + j < N - 3. (F2) Performing a change of variables such that m = i + j ; n = i - j and rearranging the terms and the l i m i t s of the sums; equation (F2) i s then written as: - 56 -I ( p e b ) m (1 - p ) 2 {A - 6m (1 - p) - (2N - 9) mC + m2c} £ ^ " ^ = m n ( 11 2 P ) 2 I ( p e b ) m {(4A - C) m3 - 4Bm4 + Bm2 + 4Cm5 - Am} = m ( 11 2 ~ P ) {(4A - C) S 3 + BS 2 - 4BS 4 + 4CS 5 - ASj }, (F3) where: -m^n^m, O ^ m ^ N - 4 ; A = 6 + 6(N - 4) (1 - p) + (N - 4) (N - 5) ( 1 - p ) 2 ; B = 6(1 - p) + (2N - 9) (1 - p ) 2 ; C = (1 - p ) 2 N-4 b m K s v = I (pe ) m . K m=6 b) I I i j ( p e V " 3 2 (1 - p ) 6 i + j ) N _ 3 = 2 ( p e b ) N " 3 (1 - p) I H (N - 3 ) 2 - *f [(N - 3) - 2 j ] 2 } = j=0 j=0 i ( p e b ) N ~ 3 (1 - p) (N - 2) (N - 3) (N - 4) (F4) c) I I i j p 1 e b ( 1 + j ) (1 - p){2 + (N - i - 3)(1 - p) } 6 x 9(1, N - 2) = i j N-3 I i 2 ( p e 2 1 5 ) 1 (1 - p) {2 + (N - i - 3) (1 - p)} = 1-0 (1 - p) [2 + (N - 3) (1 - p)] T 2 - (1 - p ) 2 T 3, (F5) - 57 -where T = £ i ( p e 2 " ) 1 . K i=0 d) X I i j e b< i +J> P ( N " 2 ) 6 6. - (N - 2 ) 2 ( p e 2 b ) N - 2 (F6) i j ' With equati ons (F3), (F4), (F5) and (F6) the f i n a l expression for < R 2 > i s 2 2 < R2 > = ( 1 1 2 P ) {(4A - C) S 3 + BS 2 - 4BS 4 + 4CS 5 - ASj} o + | ( p e b ) N 3 (1 - p) (N - 2) (N - 3) (N - 4) + (1 - p) [2 + (N - 3) ( 1 - p)] T 2 - (1 - p ) 2 T 3 + (N - 2 ) 2 ( p e 2 b ) N _ 2 . (F7) and with s = I ( p e b ) m K m=0 N-3 „ 0, . _ V - K / 2b.l K i=0 - 5 8 -APPENDIX G CALCULATION OF < K± > FOR FIX PROBABILITY p OF HAVING AN INSULATOR OCCUPYING A SITE IN A LINE BENT INTO A CIRCLE r v v K! I n2 " M N n l n2 n x K {K } K l * h. n 1 with 1 < iij < N - 1, I 4 K 4. nlf {Ki} = |Kj, K 2,...K n }, I K ± = K, 1 i I i K. = n, and 1 < i < n, . (Gl) i The sum over {K-j_} can be performed i n the same way as i n Appendix A leading to: with 1 ^ N - 2 and 1 ^ K ( i i p (G2) Performing the change of variables i n equation (G2) such that K' = K - 1 and rearranging the l i m i t s of the sums i n order that the binomial c o e f f i c i e n t i s not zero; equation (G2) then reduces to: < K i > - J |, N ( \ ^ ( n ! - i - K') ( 1 " *" 2> With 1 ^ n x ^ N - 1 and 1 ^ K' < n x - i . (G3) From equation (A3), expression (G3) i s given by: n2 1 - P) , (G4) - 59 -with 1 ^ rij ^ N - 1 and i ^ N - 2. Making the change of variables such that nj = n^ - i and rearranging the l i m i t s of the sum, equation (G4) i s then equal to: By the multinomial formula i t i s straightforward that: < K ± > = N p 1 (1 - p ) 2 for i ^ N - 2. A more general way of writing < > i s : N p 1 (1 - p ) 2 i f i ^ N - 2 < K i > = 1 ( N - n <G6> N p (1 — p) i f i = N - 1 and n 2 = 1. - 60 -APPENDIX H CALCULATION OF < K, K. > i J FOR FIX PROBABILITY p OF HAVING AN INSULATOR OCCUPYING A SITE IN A LINE BENT INTO CIRCLE K! „ „ n l , n2 N n2 ~ 1 with 1 ^ K ^ n p I ^ = K, I i K. = tij and 1 ^ i ^ n 1 (HI) Equation (HI) can be written as: < K K > - I I J K ! , K K p" 1 (1 - p ) " 2 1 [ A , (H2) with the same r e s t r i c t i o n s as i n equation (HI). Referring to Appendix B, the r e s u l t for the sum over {K^ } can be obtained leading to: < s K j > -11 p n i d - P)"2 i . (K(K - o - * : \ - i ) • wi th 1 ^ n\ 4 N ~ 1 a n d 1 < K < . In order to perform f i r s t the sum over K, consider now the f i r s t term at the right-hand side of equation (H3) and rearrange the terms. - 61 -Therefore: n 9 w / n 9 \ /n, - i - j - l \ K - 3 / K 2 N | ( n2 " 1 } ( K - 2 ) ( 1 K - 3 1 ) ' W l t h 1 < K < V ( H 4 ) Making the change of variables i n equation (H4), K" = K - 2 and using r e l a t i o n (A3); the f i n a l expression for equation (H4) i s then given by: K"=0 \ / \ K" - 1 / Now take i n t o consideration the second term i n the right-hand side i n equation (H3) and rearrange the terms. T n l „ , n 2 N F / M l n l i ' ( 1 " p ) ^ K U H r n n \ /n, - i - 1> K - 2 , n, ^ /a, - 1\ /tij - i - 1\ K - 1 j V K - 2 I I P X ( l - P) N K 2 - 1 ( V ) C 1 n„ / n„ - 1\ / n, - i - 1 K' - 1 > with 1 < K < n\ and K' = K - 1. Again using r e l a t i o n (A3), one gets: - 62 -n„ „ ITLA /n^ - i - l\ r n l n , n2 N „ ( r I p (1 - p) - K y ^ K _ 2 N p 1 (1 - p) 2 / N n J > r e s t r i c t e d to 1 < K ^ n ^ Considering equations (H5), (H6) and the l i m i t s at which the binomial c o e f f i c i e n t s are not equal to zero; the expression for < K i Kj >, equation (H3), reduces to: N-3 n n / < K. K. > = I p \l - p) N(n - 1) I 2 n =1 \ 1 , „ f _ . y _V_y) e(i + j , N - 2) + N p i + J ( 1 - p ) 2 6 ( ± + j ) N _ 2) + N f ^ p"1 (1 - P ) " 2 N ^N n ^ - ' i ^ 6 ± j ^ N " D + N(l - p) p 1 6 ± j 6 i > N_ 1 (H7) In order to evaluate the sum over n 1 at the f i r s t term i n equation (H7), perform the change of v a r i a b l e s , = - i - j • Therefore: I P (1 " p) N(N - n - 1) J_ n =1 \ n l 2 j n,' N-i-j-3 (H8) nj=0 With the r e l a t i o n n ! [ N i 7 J ~ 3 | = ( N - i - i - 3 ) ( N ~ i ~ J ~ 4 and using the d e f i n i t i o n of the binomial c o e f f i c i e n t s , equation (H8) reduces to: N( 1 - n ) N / P V+.J f(N - 1 - J - 1) _ K1-*) ( I - P ) ^ " 3 N-i-j-4 / \ n,"+l /„ . . . \ where n^ = nj - 1. Again using the d e f i n i t i o n of the binomial c o e f f i c i e n t s , one gets: N-3 n N-n / \ I P\l-P) 1 N(N- n - 1) N : 1 V ~ i 3 ) = n -1 \ n l 1 2 j N p i + j (1 - p ) 3 {(1 - p) (N - i - j - 3) + 2}. (H9) Consider now the second sum i n the right-hand side of equation (H7), and perform a change of var i a b l e s : nj = n^ - i . Therefore: N " f p" 1 (1 " P ) " 2 N ( N " 1 " 2 ) 6.. 9(1, N - 1) = Na-p)N i;rHnl+i ( b " 4 " 2 ) = - i ( i - p ) 2 Therefore, with (H9) and (H10), the expression for the moment < >, equation (H7), i s given by: - 64 -< K ± K. > = N p i + J (1 - p ) 3 {(1 - p) ( N - i - j - 3) + 2} 9(i + j , N - 2) + N-2 + N P — ( l - p ) ' 6 ( i + j > N _ 2) ( H l l ) + N p i ( l - p ) 2 6 ± j 0 ( i , N - 1) + N(l - p) p N 1 6 ± j 6 ± N - 1 - 65 -APPENDIX I CALCULATION OF < R 2 > FOR FIX PROBABILITY p OF HAVING AN INSULATOR OCCUPYING A SITE IN A LINE BENT INTO A CIRCLE < R2 > = R T I i j e b ( i + j ) < K. K. > (II) o I j 1 J i J Substitute equation ( H l l ) into equation ( I I ) . R 2 o I I i j ( p e b ) i + j N (1 - p ) 3 [2 +(N - 3)(1 - p)] 6(i + j , N - 2) -i j I I i j ( p e b ) i + J N (1 - p ) 4 ( i + j ) 9(1 + j , N - 2) + i j I I i j ( p e b ) i + j N (1 - p ) 2 6(1 + j , N - 2) + (12) i j H i j N p 1 e b ( ± + j ) (1 - p ) 2 6(1, N - 1) 6 ± + i j I I i j N(l - p) e b ( i + j ) p 1 6(1, N - 1) 6 i j The sums i n equation (12) are going to be evaluated separately i n the subsequent paragraphs. - 66 -a) I I i j ( p e b ) i + j 9(i + j , N - 2) = i j N-3 m 2 2 , N-3 , 3 I I ( P e b ) m = I ( p e b ) m = m=0 n=-m m=0 {4S^ - S'^ }; where i + j = m , i - j = n N-3 and S' = I ( p e V m\ (13) m=0 b) Analougously to the deduction of equation (13), i t is obtained that: I I i j ( p e b ) i + j ( i + j) 8(1 + j , N - 2) = L [4 S' - S£] (14) J N-2 ( p e b ) N 2 I i(N - 2 - i ) = I (N - 1) (N - 2) (N - 3) (pe b ) N ~ 2 (15) 1=0 d ) I I U P 1 e b ( i + j ) 6 9(1, N - 1) = N " f i 2 (pe 2 1 3 ) 1 = T' , (16) i j 3 1=0 1 N-2 where T' = \ ( p e 2 1 5 ) 1 i K . K i=0 - 67 -Consequently, with (13), (14), (15) and (16), equation (12) i s given by: 2 < R2 > = 72 ( ^ - S|) N(l - p) 3[2 +(N - 3)(1 - p)] -R o \j (4S> - S^) N(l - p ) 4 + N(l - p ) 2 i - (N - 1) (N - 2) (N - 3) ( p e b ) N ~ 2 + N(l - p ) 2 V2 + N(N - l ) 2 ( p e 2 b ) N _ 1 (1 - p) (17) N-3 . where S' = I ( p e b ) m mK; K m=0 T« - Y (pe 2V i K . K 1-1
- Library Home /
- Search Collections /
- Open Collections /
- Browse Collections /
- UBC Theses and Dissertations /
- Tunneling resistance of a one dimensional random lattice
Open Collections
UBC Theses and Dissertations
Featured Collection
UBC Theses and Dissertations
Tunneling resistance of a one dimensional random lattice Carvalho, Isabel Cristina Dos Santos 1984
pdf
Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
Page Metadata
Item Metadata
Title | Tunneling resistance of a one dimensional random lattice |
Creator |
Carvalho, Isabel Cristina Dos Santos |
Publisher | University of British Columbia |
Date Issued | 1984 |
Description | The resistivity of a one-dimensional lattice consisting of randomly distributed conductivity and insulating sites is considered. Tunneling resistance of the form R[sub o] i e[sup bi] is assumed for a cluster of i adjacent insulating sites. Three different ensembles are considered and compared. In the first ensemble the number of insulating "atoms" is fixed and distributed in a linear chain; in the second one there exists a fixed probability p of having an insulator "atom" occupying a site in a linear chain, and finally the third one consists of a line bent into a circle and the probability p is considered. It is observed that in the thermodynamic limit, the average ensemble resistance per site diverges at the critical filling fraction p[sub c] = e[sup –b], while the variance of the resistance diverges at the lower filling fraction p[sub c1] = p[sup 2/c] . Computer simulations of large but finite systems, however, exhibit a much weaker divergence of the resistance per site at p[sub c] and no divergence of the variance at P[sub c1]. |
Genre |
Thesis/Dissertation |
Type |
Text |
Language | eng |
Date Available | 2010-05-10 |
Provider | Vancouver : University of British Columbia Library |
Rights | For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use. |
DOI | 10.14288/1.0096045 |
URI | http://hdl.handle.net/2429/24557 |
Degree |
Master of Science - MSc |
Program |
Physics |
Affiliation |
Science, Faculty of Physics and Astronomy, Department of |
Degree Grantor | University of British Columbia |
Campus |
UBCV |
Scholarly Level | Graduate |
AggregatedSourceRepository | DSpace |
Download
- Media
- 831-UBC_1984_A6_7 C37.pdf [ 2.56MB ]
- Metadata
- JSON: 831-1.0096045.json
- JSON-LD: 831-1.0096045-ld.json
- RDF/XML (Pretty): 831-1.0096045-rdf.xml
- RDF/JSON: 831-1.0096045-rdf.json
- Turtle: 831-1.0096045-turtle.txt
- N-Triples: 831-1.0096045-rdf-ntriples.txt
- Original Record: 831-1.0096045-source.json
- Full Text
- 831-1.0096045-fulltext.txt
- Citation
- 831-1.0096045.ris
Full Text
Cite
Citation Scheme:
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}]}"
data-media="{[{embed.selectedMedia}]}"
async >
</script>
</div>
Our image viewer uses the IIIF 2.0 standard.
To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096045/manifest