Semiconductor device

COMPLETE SPECIFICATION FOR THE INVENTION ENTITLED:

SEMICONDUCTOR DEVICE HAVING A SAFETY CIRCUIT.

The following statement is a full description of this invention,including the best method of performing it known 'to me:-

<img file="AU506552B2_D0001.tif" />

if

-1' 1’

<img file="AU506552B2_D0002.tif" />

The invention relates to a semiconductor device having a semiconductor body with a surface-adjoining substrate region of a first conductivity type and comprising an integrated circuit j having at least two supply electrodes, an input electrode and an output electrode, in which there is present between a first and a second of the electrodes a protection circui serving as a protection against excess voltage and including a bipolar lateral protection transistor which is formed by emitter and collector zones of said first conductivity type and a surface-adjoining base reg10 ion of a second conductivity type formed in said substrate region, ft · ft ft ft ft ft ft ft · _Λ •1 ft ft ft ··

<img file="AU506552B2_D0003.tif" />

ft ft ft ft · ft ft ft ft ft ft ft ft ft ft ft said emitter and collector zones adjoining a surface of the semiconductor body and being provided in and surrounded by said surface-adjoining base region, which base region comprises a base connection contact so that a resistor formed by a part of the base region is present in the current path from said base connection contact to the active base zone situated between the emitter and collector zones; the emitter zone, the base connection contact, and said first electrode being connected together electrically and the collector zone being connected electrically to ft ft ft ft ft ft ft ft ft ft ft

<img file="AU506552B2_D0004.tif" />

<img file="AU506552B2_D0005.tif" />

said second electrode. ·

A semiconductor device as described above is known from

British Patent Specification 1,337,220. In this known device the base region is formed by a substrate of the second conductivity type, in which the further elements of the integrated circuit ere also provided.’

One of the drawbacks of this known device is that, when the protection transistor is switched on and off, ft ft ft ft ft* ft ft 9 ft ft ft ft.·· ft ' - >

» the- potential variation of the active base zone of said and can transistor are transferred directly to the substrate be transmitted, vi a said substrate , to other circuit elements ,

225 0 5/7' which in many cases is uncle sirabl e .

one is tection

Ano tli ex' not free in the drawback of this known aevi.ee is that choice of the base doping· of the protransistor which, as a matter of fact, is determined by the substrate doping for tie choice of which factors othei· .than tlie properties of the protection transistor are decisive.

low doping th mb the protection transistor becomes effective only at too high, a voltage.

From United States Patent Specification

9,739,238 a semiconductor device is known in which breakdown between the gate electrode and the underlying substrate of an

IGFET, eithex'· during operation ex'· by charging during manipulating, is avoided by the presence of a bipolax' protection transistor the emitter of which is connected to the ' gate dec trod?, while both the collector of the bipolar transistor and the source electrode of the field effect transistor arc connected to the earthed supply electrode, the base of the bipolai' transistor having no connection and lienee being at a • * ¹ » · · ' ft. a · floating potential.

In. this case-, breakdown occurs between the emittex' and the collector with floating base, the differential resistance after breakdown in both directions being very small·

However, a protection transistoi' having a

I.S 1 floating base involves an imporumt drawback. The transition '5

UV ·

<img file="AU506552B2_D0006.tif" />

from the non-conductive to the conductive state in a bipolar transistor having a floating base actually occurs already at a very small current through the transistor, the switching-on current, while in addition, after the transistor has become conductive due to the occurrence of said breakdown, the minimum current which is necessary to keep the transistor in the conductive state, the switching-off current, is very small. As a result of this, on the one hand the protection often becomes operative too easily, while on the other hand after operation of the protection circuit and after termination of the necessity thereof, the transistor sometimes does not revert automatically to the non-conductive state and can be made non-conductive with difficulty only.

One object of th invention is to provide a semiconductor device having an integrated circuit comprising a protection against excess voltages between two electrodes, in which the differential resistance of the protection circuit in the conductive state is very low in both directions.

Another object of the invention is to provide an integrated circuit having a protection circuit which does not become operative too easily, which protection circuit, after the disappearance of the excess voltage, automatically and rapidly returns to the original, non-oconductive state.

For that purpose, a semiconductor device of the kind described in the preamble is characterized according to the invention in that the base region is surrounded substantially entirely within the semiconductor body by the surface-adjoining substrate region of the first conductivity type which forms a £-n__ junction with the base region.

<img file="AU506552B2_D0007.tif" />

<img file="AU506552B2_D0008.tif" />

• · It is to be noted that, within the scope of this application, when there is referred to the emitter zone and the collector zone of the said bipolar transistor, the collector' zone is to be understood to mean that one of the pnid surface-adjoining zones of the first conductivity type whose jc-n junction with the base zone at the normal operating voltages of the integrated circuit is biassed in the reverse direction.

' The semiconductor device according to the invention comprises a protection circuit which in the conductive state shows a very low differential resistance for both directions of the voltage across it. A further important advantage of a device having- a protection circuit according to the invention is that the current at which the protection circuit switches from the non-conductive into the conductive state, as well as the minimum current at which the protection circuit can remain in the conductive state, are not unacceptably small and can be controlled within certain limits by a suitable ch dee of the geometry of the device. Since the £-n junction between the base region and. the substrate region does not trigger the protection circuit but can always be kept at a substantially fixed reverse voltage, the potential of the substrate region and of the further zones provided therein experience a smaller influence from the switching on and off of the protection circuit. Since furthermore the part of the base region between the connection contact and the active base zone forms a shunt resistance across the £-n junction between the emitter zone and the base region, the switching on and off currents will

-.-.30 —•-'•JV’t*·..·» both be larger than when the base of the transistor is at floating potential.

When a voltage in the reverse direction occurs

<img file="AU506552B2_D0009.tif" />

• · · ft .· · • ·

.. . 15 • · · • ·· • · between the connection contact on the base region and the collector zone, the ja-n junction between the base region and the collector zone being biassed in the forward direction, the protection circuit operates as a £~n diode in the forward direction, in series with a resistance formed by the base region between the connection contact and the collector zone. However, said resistance is generally negligibly small as a result of conductivity modulation due to the large amount of minority charge carriers injected into the base region from the collector zone. Thus the protection operates in both directions.

The protection circuit according to the present invention can. advantageously be used foo? the ·· · 20 • · · • ·· protection against excess voltages between a supply electrode and the input of an integrated circuit. However, the device according to the invention can bo used not only as an input protection but also as a protection between, two arbitrary electrodes, preferably as a supply protection between two supply electrodes, in both cases naturally with suitable _# .... , adaptation of the remaining circuit. In general the ··· ·· · protection circuit according to the invention serves as a protection against excess voltages between the two electrodes ₇ between which the protection circuit is connected. In many • cases said excess voltage can cause direct breakdown between the-two *said electrodes and the protection serves directly to avoid said direct break-down. In other cases on. the

<img file="AU506552B2_D0010.tif" />

-V'TTjysj·.

contrary excess voltage between the two said electrodes will not cause 'damage between said electrodes but elsewhere in the circuit; in that case also a protection circuit connected between the two electrodes serves to avoid such damage. The protection circuit according to the invention cannot only be used in circuits having insulated gate field effect transistors but, as will become apparent hereinafter, may to &lt;

• -·« o ·*· to to to •1

<img file="AU506552B2_D0011.tif" />

<img file="AU506552B2_D0012.tif" />

<img file="AU506552B2_D0013.tif" />

<img file="AU506552B2_D0014.tif" />

also be used advantageously in circuits having, for example, junction field effect transistors (junction FET’s) and in bipolar circuits. j

The invention will now be described in greater detail with reference to a few embodiments and the drawing, in which

Fig. 1 is a diagrammatic cross-sectional view of a device according to the invention, ’ Fig. 2 shows the circuit diagram corresponding to the device shown in Fig. 1, ’

Fig. 3 illustrates diagrammatically the currentvoltage characteristic of the protection circuit of the device shown in Figs·. 1 and 2,

Fig. 4 shows diagrammatically partly a crosssectional view and partly a circuit diagram of another embodiment of the device according to the invention,

Fig·. 5 is a plan view of the device shown in

Fig. 4,

Fig. 6 shows diagrammatically the circuit diagram of a further device according to the invention,

Fig. 7 is a diagrammatic cross-sectional view of again another embodiment of a device according to the invention,

Fig. 8 shows the circuit diagram of a device

<img file="AU506552B2_D0015.tif" />

7according’ to the invention having a bipolar integrated /

circuit, vig·. 9 is a diagrammatic cross-sectional view of the structure of the device shown in Fig. 8, and . Fig. 10 is a diagrammatic cross-sectional view of a device according to the invention having a junction field effect transistor.

The fig¹ :-dS are diagrammatic and not draw to scale. In the cross-sectional views semiconductor regions of the same conductivity type are shaded in the same direction.

<img file="AU506552B2_D0016.tif" />

<img file="AU506552B2_D0017.tif" />

Corresponding parts in the various embodiments are as a rule referred to by the same reference numerals.

Fig. 1 is a diagrammatic cross-sectional view of a semiconductor device according to the invention, the circuit diagram of said, semiconductor device being shown in Fig. 2. The semiconductor device according to this embodiment has a semiconductor body 1 of silicon which comprises an integrated inverter circuit having two complementary field effect transistors, «η n-channel transistor and a nchannel transistor Tp with insulated gate electrode, a single-sided input protection being provided to prevent , breakdown between the insulated gate electrodes and the underlying semiconductor surface via the intermediately located insulating layer. Such a breakdown may occur in the operating ' condition (in which a positive voltage with respect to connection terminal 12 is at the connection terminal 13) > but i it may also occur out of operation by static charging of the gate electrodes during handling.

The integrated circuit has a first supply electrode 2, a second supply electrode 3» ^an input electrode 4 u *

<img file="AU506552B2_D0018.tif" />

-8----...........- .,.......

<img file="AU506552B2_D0019.tif" />

<img file="AU506552B2_D0020.tif" />

<img file="AU506552B2_D0021.tif" />

and an output electrode 5, which, electrodes are partly shown diagrammatically in Fig. 2 as lines but actually they all consist of conductive layers. The electrodes 2, 3 and 5 are connected directly to the connection terminals 12, 13 and 15, while on the contrary the input electrode 4 is connected to the input connection l4 via a series resistor 6 for reasons to be given hereinafter. The resistor 6 which is shown diagrammatically may be an external resistor but may also be a resistance lay?r, for example a layer of polycrystalline silicon, which is separated from the semiconductor surface by an insulating layer, or may be a diffused resistor or

0·· • * ’ 9 ·0·0

Ο ··· a resistor formed by ion implantation, respectively. The input electrode 4 comprises the mutually connected gate electrodes 8 and 7 of the complementary field effect

<img file="AU506552B2_D0022.tif" />

• 00

<img file="AU506552B2_D0023.tif" />

transistors and Tp which are separated from the semiconductor surface by insulating layers 40 and 9 of silicon oxide. The transistor Tp is a p-cliannel transistor having . jo-type source and drain, zones 25 and 26, the transistor is an n-channel transistor having n-type source and drain zones 28 and 27 which are provided in a £-type region 29 which is surrounded entirely by the n-type substrate region 11 and forms therewith a £-n junction 31· For the mutual separation «0.000 Ί • 0 of the transistors T^

Tp and a sunken oxide pattern 30 «00 which surrounds the transistors entirely is used in known manner in this example. The presence of such a sunken oxide ' pattern, however, is not necessary. The region 11 is connected to the terminal 13 via a contact layer 37 and a conductor 38.

<img file="AU506552B2_D0024.tif" />

<img file="AU506552B2_D0025.tif" />

<img file="AU506552B2_D0026.tif" />

<img file="AU506552B2_D0027.tif" />

a ·· · ·♦ ft ft

<img file="AU506552B2_D0028.tif" />

1.5

<img file="AU506552B2_D0029.tif" />

<img file="AU506552B2_D0030.tif" />

, A protection circuit .is present between a first and a second of the said electrodes. In this example has been selected as first electrode the supply electrode 2 which is connected to earth '-nd as second electrode electrode ^1. The protection circuit comprises lateral safety transistor

<img file="AU506552B2_D0031.tif" />

the emitter zone the input a bipolar o f will ch i s connected to the supply electrode 2 and the collector zone is connected to the input electrode 4. This transistor is formed by emitter and collector zones ( 16, 17) of a first conductivity type, in this case _n-type, which adjoin a surface of the semiconductor body and are provided in and surrounded by a surface-adjoining base region 18 of the second conductivity type, in this case the p-type. This base region 18 comprises a connection contact 21 so that a resistor (denoted by 22 in broken lines in Figure l)formed by a part of the base region 18 is present in the current path of the said connection contact 21 to the active base zone 20 situated between the emitter and collector zones (l6, 17)· The emitter zone 16, the connection contact 21, and the said first electrode 2 are connected together electrically, while the collector zone 17 is connected electrically to the said second electrode 4.

According to the invention, the base region 18 is surrounded entirely within the semiconductor body by a surf ace -ad j oining substrate region 11 of the first conductivity type, in this case the n-type, which forms a junction 19 with the base region 13.

The operation of the protection in the device described is based on the following. A voltage which may cause danger of breakdown of the gate electrode insulation layer may occur between the connection terminals 14 and

<img file="AU506552B2_D0032.tif" />

ft ft ft ft* ft ft· ft β · ft ft « ft 0 ·ft·ft

<td> •</td><td> 12 either during a peak at the input voltage or by charging during manipulating. When said voltage has such a polarity that as a result of this the ja-n junction 23 between the . collector zone .17 and the base region 18 is biassed in the</td>

<td> 5</td><td> reverse direction, said £-n junction 2.3 will break down by an avalanche effect when a certain voltage is exceeded. The current which as a result of this flows between the collector zone .17 and the connection contact 21 through the base region 18 causes in the said region 18 a voltage drop across the</td>

<td> 10</td><td> resistance 22 so that a voltage in the forward direction is</td>

<td> ft© ft. ft 9 · • ft ft</td><td> • set up across the ja-n junction 24 between the emitter zone</td>

<td> • · ft · ft · · ft#·©</td><td> 16 and the base region 18 and said junction 24 starts</td>

<td> ft ft ft ·</td><td> injecting charge carriers into the base region 18. As a</td>

<td> ft · · • « ft ft * ‘ ft</td><td> result of this the bipolar transistor T^ becomes conductive</td>

<td> .. .15 ft ft ft ft • ft</td><td> with a very low differential resistance as a result of the interaction between the gain of the transistor and the</td>

avalanche multiplication at the ja-n junction 23 which is biassed in the reverse direction.

Fig·. 3 shows diagrammatically the current.. «0 ft ft ft ft ft ft

<img file="AU506552B2_D0033.tif" />

voltage characteristic of the protection transistor T^ in which, the voltage V between the input electrode 4- and earth, is plotted on the horizontal axis and the current I from the input electrode 4 to earth is plotted on the vertical axis. The switching off current (’’hold, current ) and the switching off voltage (hold voltage) at which after breakdown the protection circuit still remains just in the conductive state, and the trigger current and trigger voltage at which, the protection becomes operative are also shown in Fig. 3· Otherwise this figure is only qualitative and serves only to show the general shape of the resulting

-11« d 3”

V—I curve. The differential resistance (the value ~) after breakdown is very small. The switching off current 1^ and

i.

i i

the switching on. current 1^. are not very low as a result of the presence of the resistance 22 formed by a part of the base region 18, so that it is not difficult to bring the protection circuit again in the non-conductive state and at the same time the premature operation of the protection is counteracted.

In this example, a series resistor 6 is provided between.

the input electrode 4 and the external • · 0

Aft ft ft · ft ft • ••ft ft ft ft ft input connection

4- in order to promote the last-mentioned favourable properties of the protection circuit.

When the ft · ft ft ft ft ft ft ft • ft ft · ft ft ft * ft ft ft ft ft ft &lt;

•« ft value of said resistor 6 is equal to R, then in the state in which the transistor is still just in the state, the voltage drop across said resistor is conduc tive equal to

1-^ x R, while the voltage between the input terminal 14- and earth then, is V, + I, x R. At a given value of h h _ °

ν.^₅ R may be.

ft ft ft ft ft ft ft ft ft ft ft ft

4* ft ft ft ft ft ft. ft ft ft ft chosen to and earth is larger terminal 14· at which transistor is suill than the maximum supply voltage present between terminal 13 and earth. In just whi c tha i the removal of the voltage peak at terminal 14·,

,. ndur tive is normally case, after the protection • ft ft ft ft ft ft ft ft ft ft ft ft ft ft ft · ft circuit will automatically return to the non-conductive state.

lilien the voltage peak at electrode 4 has an’ .25 opposite polarity, the protection circuit between the electrodes 2 and 4 operates as.a diode in the forward direction (j_ ' junction. ' t\vh a resistor in series. In accordance .rth the current which flows through the protection circuit, said resistance may become very small due to

-12Γ

Ϊ conductivity modulation as a result of electrons injected .··

IF £?· across the p-n junction. 23 from the collector zone 17 into the base region 18.

In order to obtain a series resistance r

C

1I which, is as low as be adapted, thereto» connection contact possible ,

As shown the geometry will preferably also in Fig. 1 , in this example the is provided on a surface-adjoining j>-type contact zone 32 belonging to tie base region 18 and. having a higher doping than the adjoining part of the region 18.

In order to prevent the formation of an inversion channel at the surface ft · • ·· and 17, as a result of which the • ft ft · ft •

ft··· • ft • ’ ft ' ft··· « ft ft ft ft ft ft ft ft • ft* 'ft ft ft ft > ft ft ft ·· ft ft ft ft ft ft ft ft ft ft· • 0 is adversely influenced, a field operation of transistor electrode 34, for example of polycrystalline silicon, which is separated from region 18 by an insulating layer 33» ion example oxide, is provided, above the active base zone 20 lateral bipolar transistor T^ and is D.C.

the base of of connected silicon the to the emittox- zone 16. In addition, in order to prevent possible variations in the resistor- 22 as a result of a possible ··· inversion layer- between the contact zone and the emitter • ··· zone l6, in this example a field electrode 36, for example ·· · • -i ft ·· of polycrystalline silicon, which is separated from the base region 18 tv an insulating layer 35; ion exarg. le of silicon . - ft ft····· • ft • ft · • ft. ft ft ft· oxide, is provided between the contact zone 32 and the emitter zone 16 and is also D.C. connected to the emitter zone 16. The presence of the field electrodes 34 and. 3*5 is sometimes desired, but is not necessary.

If requir ed, in order to reduce the resistance .22, the contact zone 32 will preferably be provided so as co adjoin the emitter zone 1(5 which is situp.ted between the connection contact 21 and the collector zone 17· This is

-13illustrated in Fig. 4 which shows only the cross-sectional

I

I

I s l

I

I

I

I view of the protection cir’cuit, while the remainder of the integrated circuit is denoted diagrammatically by IC and the reference numerals correspond to those of Fig. 1. Fig. 4 shows furthermore a more symmetric construction of the protection circuit the plan view of which is shown in Fig. 5·

In addition, in the example of ’Figs. 4 and 5 the contact zone 32 surface pattoi-n and the emitter zone 26 are short-circuited at the by is the metal layer 21. Furthermore, no sunken oxide used in the example shown in

Figs . 4 and 5 ft ft to

Φ· ftft the one

In the examples described

5, the emitter and collector with reference to zones l6 and 17 of protection transistor are connected to two electrodes of which is the supply electrode 2 connected to earth • 15 ft ft ft « ft · (or to a different reference potential.), while the other electrode is the input electrode protection is obtained. However, is may also be connected so that

4. In this manner an input the protection transistor the said 'Lecond electrode ft ft ft ft • 20 • ft ft ft · 9 is is the supply electrode 3· in that case a supply protection obtained against excess Voltages between the connections and 13. The circuit diagram then becomes, for example, as shown in Fig. 6. In this case, however no series resistor ft ft ftft ftft ft ft ft • ft ft ft ftft should be used, between the supply connection 13 and the safety transistor since the voltage drop over such a resistor is impermissible in a supply circuit.

In ornai' to nevertheless keep the minimum voltage at which the protection remains in the conductive state lower than the normal supply voltage so that after return to the normal state the protection

<img file="AU506552B2_D0034.tif" />

-14returns again to the non-conductivo state, the gain factor of the protect! on transistor T may be reduced, for example, by increasing the distance between the emitter and collectox· rr_t , so voltages and V^.

that case, however that also the difference between the (Fig. 3) deco-eases considerably. In the lateral transistor T„ after breakdown J of the collector-base junction, cannot pass a high current.

However, the protection is sufficient nevertheless, since it has boon found that in this case the vertical transistor

1u formed by the zones 16, 18 and. 11 passes a large part of ft ft ft · • · • · ft • ••ft ft ·· ft ft ft ft ft ft ft ft ft ft ft · ft • ft ft ft ft ft ft ft ft ft ft ft ft ft ft ft

<td></td><td> • · • · • · · · • · • · • ··· • ·· · f · · • ··</td><td colspan="2"> 20</td>

<td></td><td> · • ft····· . · ·</td><td></td><td></td>

<td> e</td><td> ·· ·</td><td></td><td></td>

<td></td><td> ·. · · · · · ·</td><td></td><td></td>

<td></td><td> &gt;</td><td> 25</td><td></td>

<td></td><td></td><td></td><td></td>

<td></td><td></td><td></td><td> 1</td>

the current occurring after breakdown. This is a furthor important advantage of tho invention as compared with the said British Patent Speedfication 1,337)220. Of course, this depends on the mutual distances of the p-n junctions 19, 23 and 2h and the dopings used, which quantities cun all bo chosoo suitably by those skilled in the art. in accordance with tiro prevailing conditions. In addition to the supply ' proto..lo„, an input protection (₂₂*. _Tj) i. provided in Fig. C at w-s described in the preceding examples.

As is shown diagrannnat ically in the crosssectional view of Fig. 7, the connection contact 21 may also bo provided between the zones 16 and 17. In that case the resistance 22 is formed by the. spreading resistance of the connection contact 21 and the highly doped zone 32 possibly belonging thereto. In general, however, it will be ebsired to provide the emitter zone 16 between the collector zone 17 and the connection contact 21, as in the preceding examples, so as to obtain values of the switching on and off

<img file="AU506552B2_D0035.tif" />

’CSt’fWL

f.

currents a.nd voltages which are not too low but are not too high either. Combinations of these oases are also / >ssible.

The bipolar lateral protection transistor having a reduced gain factor as shown in Fig. 6 may also be used advantageously at the input, preferably between the terminals 13 and 14. Due to the low gain factor the series resistor 6 may then be avoided sixice the switching on voltage V^_ and the switching off voltage (see Fig. 3) differ only little in that case, while nevertheless the • ··. w · 4 » · · ·

<img file="AU506552B2_D0036.tif" />

• ♦

<img file="AU506552B2_D0037.tif" />

differential resistance remains very low. In this case also, a large part of the current can be received by the vertical transistor (l6, 18, 11).

In the above examples has been described the use of a protection circuit according to the invention to prevent breakdown between an insulated gate electrode and the underlying semiconductor· surface. In practice this certainly is one of the most important cases in which the .

invention can be used advantageously. However, the invention is not restricted thereto as will be illustrated with reference to the following examples.

’ Fig. 8 shows the circuit diagram of a differential amplifier having an input electrode 101, supply electrodes 102, 103, 107, output electrodes 104 and 105 and a current source 106. When the voltage at the input terminal

101 becomes very strongly positive, a danger of breakdown of the emitter—base junction of transistor may occur’ s.nce the said junction may then be polarized considerably in the reverse direction. Therefore, a protection circuit as described in the preceding examples and consisting of transistor

<img file="AU506552B2_D0038.tif" />

and resistance has been provided between

-16the input electrode 101 and earth. In the diagrammatic cross9 it is shown how such a circuit can be integrated.

Finally, view of Fig. 10 it is illustrated how the protection circuit as described in the preceding Examples with the lateral bipolar transistor can be used as a protection between the j2-type gate region 201 and the n-type channel region 202

4of a junction field effect transistor having n source region 203 and n⁺ drain region 204.

0·

00 • 00 ·

0000

•. · • 0 0000 «0

0· • 0·

0 0 ·

00 · > 0 X · •

• 0

0 0

The invention is not restricted to the examples described but many variations are possible to those skilled in the art without departing from the scope of this invention.

For example, the conductivity types of all semiconductor regions may simultaneously be replaced by their opposite types, while observing the polarities of the voltages to be applied w^ich are determined by the given definition of the • · • · • ···

0« • 0 • ··· term collector zone.

applied not only to

In the general circuits having the invention can be complementary MOS . 20 • 0 0 • «0

00 transistors, but to any integrated circuit, in particular also to a device comprising a semiconductor body having only one discrete insulated gate field effect transistor.

Furthermore the invention is not restricted to a semiconductor

9

00device haA^ing a semiconductor body of silicon.

but other suitable semiconductor materials, for example, germanium or compounds, for example GaAs , may also be used. Further more, instead of silicon oxide, the insulation layers used may also be different insulating materials, for example

-17„N^, whATe furthermore tlie said gate and field

ΑΙ^,Ο^ or Si electrodes may, for example, also be a metal instead of polycryctalline silicon.

-18>

r.

··· · ·· ·· ·, ··

9. · ·· ·

<img file="AU506552B2_D0039.tif" />

• · · • · · • «4 ··

<img file="AU506552B2_D0040.tif" />

© - H J t