984 



THE BELL SYSTEM TECHNICAL JOURNAL, JULY 1957 



tion of the field. If ai is fixed, the field at breakdo\ni can be determined. 



The breakdo\\ii voltage is / Ed.> 



Jo 



Fig. 4 is a plot of breakdown voltage as a function of space-charge p 

 width for PN and PIN diodes. The PIN values are calculated; the PN 

 data is previously unpublished data supplied by K. G. McKay. 



Some interesting observations can be made from Fig. 4: 



1. The plot of breakdown voltage versus barrier width for a PN step \ 

 junction assumes that the space-charge region does not extend through ; 

 the high resistivity side of the junction. For this class of junctions the 

 breakdown voltage is determined by the impurity concentration as shown 

 in Fig. 5. The plot of breakdown versus space-charge width for a PIN 

 diode assumes that the space-charge region extends from the P to N 

 region at very low bias, and that it is limited by the width of the I re- 

 gion. If a constant field is assumed in the I region, the breakdown volt- 

 age is a function of the barrier width. 



2. Although the space-charge region can reach through the I region 

 at low bias, the avalanche breakdo^\^l voltage is a function of the width 

 of the I region. 



3. For the devices considered here with tt regions in the order of 10' ' 

 cm, the maximum breakdown voltage is in the order of 2,000 volts. 



10' 



o 

 > 



< 



o 

 > 



z 

 5 



o 



10^ 



5 102 





10 



4 6 8,^-2 2 



4 6 8, 



10"" " " "10"^ " " " "10""^ " "" " "10"' 



BARRIER WIDTH FOR PN AND PIN JUNCTIONS IN CM 



Fig. 4 — Breakdown vohage as a function of barrier width for PN and PIN 

 junctions. 



