6.26 Zener and Avalanche Diodes

Sec­tion 6.24 ex­plained that nor­mally no sig­nif­i­cant cur­rent will pass through a p-n junc­tion in the re­verse di­rec­tion. The ba­sic rea­son can be read­ily ex­plained in terms of the schematic of the p-n junc­tion fig­ure 6.33. A sig­nif­i­cant re­verse cur­rent would re­quire that the ma­jor­ity n-side con­duc­tion elec­trons and p-side holes both move away from the junc­tion. That would re­quire the cre­ation of sig­nif­i­cant amounts of elec­tron-hole pairs at the junc­tion to re­plen­ish those that leave. Nor­mally that will not hap­pen.

But if the re­verse volt­age is in­creased enough, the diode can break down and a sig­nif­i­cant re­verse cur­rent can in­deed start to flow. That can be use­ful for volt­age sta­bi­liza­tion pur­poses.

Con­sider fig­ure 6.33. One thing that can hap­pen is that elec­trons in the va­lence band on the p side end up in the con­duc­tion band on the n side sim­ply be­cause of their quan­tum un­cer­tainty in po­si­tion. That process is called tun­nel­ing. Diodes in which tun­nel­ing hap­pens are called Zener diodes.

The process re­quires that the en­ergy spec­trum at one lo­ca­tion is raised suf­fi­ciently that its va­lence band reaches the level of the con­duc­tion band at an­other lo­ca­tion. And the two lo­ca­tions must be ex­tremely close to­gether, as the quan­tum un­cer­tainty in po­si­tion is very small. Now it is the elec­tro­sta­tic po­ten­tial, shown in green in fig­ure 6.33, that raises the p-side spec­tra rel­a­tive to the n-side ones. To raise a spec­trum sig­nif­i­cantly rel­a­tive to one very nearby re­quires a very steep slope to the elec­tro­sta­tic po­ten­tial. And that in turn re­quires heavy dop­ing and a suf­fi­ciently large re­verse volt­age to boost the built-in po­ten­tial.

Once tun­nel­ing be­comes a mea­sur­able ef­fect, the cur­rent in­creases ex­tremely rapidly with fur­ther volt­age in­creases. That is a con­se­quence of the fact that the strength of tun­nel­ing in­volves an ex­po­nen­tial func­tion, chap­ter 7.13 (7.74). The fast blow-up of cur­rent al­lows Zener diodes to pro­vide a very sta­ble volt­age dif­fer­ence. The diode is put into a cir­cuit that puts a nonzero tun­nel­ing cur­rent through the diode. Even if the volt­age source in the cir­cuit gets per­turbed, the volt­age drop across the Zener will stay vir­tu­ally un­changed. Changes in volt­age drops will re­main re­stricted to other parts of the cir­cuit; a cor­re­spond­ing change over the Zener would need a much larger change in cur­rent.

There is an­other way that diodes can break down un­der a suf­fi­ciently large re­verse volt­age. Re­call that even un­der a re­verse volt­age there is still a tiny cur­rent through the junc­tion. That cur­rent is due to the mi­nor­ity car­ri­ers, holes from the n side and con­duc­tion elec­trons from the p side. How­ever, there are very few holes in the n side and con­duc­tion elec­trons in the p side. So nor­mally this cur­rent can be ig­nored.

But that can change. When the mi­nor­ity car­ri­ers pass through the space charge re­gion at the junc­tion, they get ac­cel­er­ated by the strong elec­tric field that ex­ists there. If the re­verse volt­age is big enough, the space charge re­gion can ac­cel­er­ate the mi­nor­ity car­ri­ers so much that they can knock elec­trons out of the va­lence band. The cre­ated elec­trons and holes will then add to the cur­rent.

Now con­sider the fol­low­ing sce­nario. A mi­nor­ity elec­tron passes through the space charge re­gion. Near the end of it, the elec­tron has picked up enough en­ergy to knock a fel­low elec­tron out of the va­lence band. The two elec­trons con­tinue on into the n side. But the cre­ated hole is swept by the elec­tric field in the op­po­site di­rec­tion. It goes back into the space charge re­gion. Trav­el­ing al­most all the way through it, near the end the hole has picked up enough en­ergy to knock an elec­tron out of the va­lence band. The cre­ated con­duc­tion elec­tron is swept by the elec­tric field in the op­po­site di­rec­tion of the two holes, back into the space charge re­gion... The sin­gle orig­i­nal mi­nor­ity elec­tron has set off an avalanche of new con­duc­tion elec­trons and holes. The cur­rent ex­plodes.

A diode de­signed to sur­vive this is an “avalanche diode.” Avalanche diodes are of­ten loosely called Zener diodes, be­cause the cur­rent ex­plodes in a sim­i­lar way. How­ever, the physics is com­pletely dif­fer­ent.

Key Points

$\begin{picture}(15,5.5)(0,-3) \put(2,0){\makebox(0,0){\scriptsize\bf0}} \put(12... ...\thicklines \put(3,0){\line(1,0){12}}\put(11.5,-2){\line(1,0){3}} \end{picture}$

Un­like the ide­al­ized the­ory sug­gests, un­der suit­able con­di­tions sig­nif­i­cant re­verse cur­rents can be made to pass through p-n junc­tions.

$\begin{picture}(15,5.5)(0,-3) \put(2,0){\makebox(0,0){\scriptsize\bf0}} \put(12... ...\thicklines \put(3,0){\line(1,0){12}}\put(11.5,-2){\line(1,0){3}} \end{picture}$

It al­lows volt­age sta­bi­liza­tion.