Reverse flow mechanism through the transition is relatively simple. Charge carriers that are minor for one of the regions, drifting in the electric field of the volume charge region, fall into the region where they are already the main carriers. Since the concentration of majority carriers usually significantly exceeds the concentration of minority carriers in the neighboring region (n n>> np and p p >> p n), then the appearance in a particular region of the semiconductor of an insignificant additional amount of the main charge carriers practically does not change the equilibrium state of the semiconductor.
A different picture is obtained with the flow of direct current . In this case, the diffusion component of the current dominates, consisting of majority charge carriers that overcome the potential barrier and penetrate into the region of the semiconductor for which they are minority carriers. In this case, the concentration of minority carriers can increase significantly in comparison with the equilibrium concentration. The phenomenon of the introduction of nonequilibrium carriers is called injection.
When direct current flows through p-n- the transition from the electronic region to the hole region will be the injection of electrons, from the hole region to the electronic region, the injection of holes.
For simplicity, we will further consider only the injection of holes from the hole region of the semiconductor into the electron region, then extending all the conclusions made to the counter process of electron injection into the hole region. If applied to p-n- voltage transition in the direction of transmission (Fig. 3.13), then the height of the potential barrier will decrease, and a certain number of holes will be able to penetrate into n- region.
Fig.3.13. Scheme of the flow of direct current through the junction
Before these holes n- the region was electrically neutral, i.e. positive and negative charges in each of the sufficiently small volumes n- the areas summed up to zero.
holes injected from R- areas in n- area, represent some positive space charge. This charge creates an electric field that propagates in the volume of the semiconductor and sets in motion the main charge carriers - electrons. The electric field created by the holes attracts electrons to the holes, the negative space charge of which must compensate for the positive space charge of the holes. However, the concentration of electrons near the space charge of the injected holes will lead to a decrease in their concentration in adjacent volumes, i.e. to the violation of electrical neutrality and the appearance of a space charge in these volumes.
Since no redistribution of free charges inside an electrically neutral semiconductor can compensate for the space charge of holes, then in order to restore the state of electrical neutrality of the semiconductor, an additional number of electrons must enter from the external output, the total charge of which will be equal to the total charge of the injected holes. Since an electron and a hole have charges equal in magnitude and opposite in sign, the number of electrons entering the bulk of the semiconductor from the external terminal must be equal to the number of injected holes.
Thus, at the same time as the appearance of n- areas of a certain number of injected holes - minor nonequilibrium carriers - the same number of electrons appear the main nonequilibrium carriers. Both those and other carriers are non-equilibrium, since they create a concentration that differs from the concentration of thermodynamic equilibrium.
The process of compensation of the space charge of minor nonequilibrium carriers by the space charge of the main nonequilibrium carriers proceeds exceptionally fast. The settling time of this process is determined by the relaxation time
and is for germanium (ε = 16), the resistivity of which is 10 Ohm. cm, about 10-11 sec. The establishment of the process can therefore be considered instantaneous.
Since the carrier concentration is high directly at the junction, the carriers, due to the presence of a concentration gradient, will propagate deep into the bulk of the semiconductor in the direction of lower concentrations. At the same time, the concentration of nonequilibrium carriers will decrease due to recombination, so that the total value of the concentration will tend to the equilibrium value.
Fig.3.14. The distribution curve of the concentration of non-equilibrium minor
carriers (holes) in the electronic region of the pn junction
If the non-equilibrium concentration is small compared to the concentration of equilibrium main carriers (low injection level), then the decrease in the concentration of non-equilibrium carriers in the direction from the transition deep into the semiconductor will occur according to an exponential law (Fig. 3.14):
(3.23)
L characterizes the average distance over which the carriers have time to diffuse during their lifetime.
At a point far enough away from the transition (x →¥ ) the equilibrium concentration of charge carriers will be maintained.
At a low level of injection, the concentration of nonequilibrium carriers in n- the region near the interface will depend exponentially on the magnitude of the voltage applied to the junction:
(3.24)
(at U= 0; increases rapidly with increasing positive values U).
Note that the change in voltage at the junction by Δ u will lead to an increase in the concentration of nonequilibrium holes in n- areas, i.e. to a change in charge. The change in charge caused by a change in voltage can be thought of as the action of some capacitance. This container is called diffusion , since it appears due to a change in the diffusion component of the current through the junction.
It can be concluded that diffusion capacitance will manifest itself at forward currents through the junction or at low reverse voltages, when the diffusion current cannot be neglected in comparison with the conduction current.
We represent the diffusion capacity as a change in charge Δ Q, related to the voltage change Δ that caused it u:
and estimate the effect of the current through the junction on the diffusion capacitance.
The total charge of minority nonequilibrium carriers in n-domain can be obtained by integrating the expression (3.23).
A change in the external voltage dU at the p-n junction leads to a change in the charge dQ accumulated in it. Therefore, the p-n junction behaves like a capacitor, the capacitance of which is C \u003d dQ / dU.
Depending on the physical nature of the changing charge, barrier (charging) and diffusion capacitances are distinguished.
The barrier (charging) capacitance is determined by the change in the uncompensated charge of the ions when the width of the barrier layer changes under the influence of an external reverse voltage. Therefore, an ideal electron-hole junction can be considered as a flat capacitor, the capacitance of which is determined by the relation
where P, d are the area and thickness of the p-n junction, respectively.
Relations (1.41) and (1.31) imply
.
In the general case, the dependence of the charging capacity on the reverse voltage applied to the p-n junction is expressed by the formula
,
where C 0 is the capacitance of the p-n junction at U OBR = 0; g - coefficient depending on the type of p-n transition (for sharp p-n transitions g = 1/2, and for smooth g = 1/3).
The barrier capacitance increases with an increase in N A and N D, as well as with a decrease in the reverse voltage. The character of dependence С BAR = f(U OBR) is shown in fig. 1.13, a.
Consider diffusion capacity. With an increase in the external voltage applied to the p-n junction in the forward direction, the concentration of injected carriers near the junction boundaries increases, which leads to a change in the amount of charge due to minority carriers in the p- and n-regions. This can be seen as a manifestation of some capacity. Since it depends on the change in the diffusion component of the current, it is called diffusion. Diffusion capacitance is the ratio of the increment of the injection charge dQ inzh to the change in voltage dU pr that caused it, i.e. 
. Using equation (1.30), one can determine the charge of injected carriers, such as holes in the n-region:
Figure 1.13 Dependence of the barrier (a) and diffusion (b) capacitances of the p-n junction on voltage.
Then the diffusion capacitance due to the change in the total charge of nonequilibrium holes in the n region is determined by the formula
.
Similarly, for the diffusion capacitance due to the injection of electrons into the p region,
.
Figure 1.13 Equivalent circuit p-n junction.
Total diffusion capacity
The dependence of the capacitance on the forward voltage at the p-n junction is shown in Figure 1.13, b.
The total capacitance of the p-n junction is determined by the sum of the charging and diffusion capacitances:
.
When the p-n junction is turned on in the forward direction, the diffusion capacitance prevails, and when turned on in the opposite direction, the charge capacitance prevails.
On fig. 1.14 shows the equivalent circuit of the pn ac junction. The circuit contains a differential resistance of the p-n junction r D, a diffusion capacitance C DIF, a barrier capacitance C BAR and a volume resistance of p- and n-regions r 1 . Based on equation (1.37), we can write:
.
If with direct inclusion of the p-n junction U pr \u003e j t, then:
At room temperature 
; (1.42)
(in relation (1.42) the current value is substituted in amperes). The leakage resistance r UT takes into account the possibility of current passing over the surface of the crystal due to the imperfection of its structure. With direct inclusion of p-n junction C BAR<< С ДИФ, дифференциальное сопротивление r Д ПР мало и соизмеримо с r 1 , поэтому эквивалентная схема принимает вид, показанный на рис. 1.15, а.
Figure 1.15 Simplified p-n junction equivalent circuits.
With a reverse bias r D OBR >> r 1 , C BAR >> C DIF and the equivalent circuit has the form shown in fig. 1.15, b.
A semiconductor diode is inert with respect to sufficiently fast changes in current or voltage, since a new carrier distribution is not immediately established. As is known, an external voltage changes the width of the junction, and hence the magnitude of the space charges in the junction. In addition, during injection or extraction, the charges in the base region change (the role of charges in the emitter is insignificant). Therefore, the diode has a capacitance that can be considered connected in parallel with the p-n junction. This capacity can be divided into two components: barrier capacity, reflecting the redistribution of charges in the transition, and diffusion capacity, reflecting the redistribution of charges in the base. Such a division is generally conditional, but it is convenient in practice, since the ratio of both capacitances is different for different polarities of the applied voltage. With a forward voltage, the main role is played by excess charges in the base and, accordingly, by the diffusion capacitance. When the voltage is reversed, the excess charges in the base are small and the barrier capacitance plays the main role. We note in advance that both capacitances are not linear: the diffusion capacitance depends on the forward current, and the barrier capacitance depends on the reverse voltage.
Let us determine the value of the barrier capacitance, considering the transition as asymmetric type n + -p. Then the extent of the negative charge in the p-type base can be considered equal to the entire width of the transition: . Let's write the module of this charge:
where N is the impurity concentration in the base; S - transition area. The same (but positive) charge will be in the emitter layer.
Imagine that these charges are located on the plates of an imaginary capacitor, the capacitance of which can be defined as
Taking into account the expression for the width of the transition during reverse switching, and differentiating the charge Q with respect to voltage, we finally obtain:
(7.10)
where and are, respectively, the width and height of the potential barrier in the equilibrium state.
Keeping in mind that the diode has a capacitance, you can draw up its complete equivalent circuit for alternating current (Fig. 3.10a).
The resistance R 0 in this circuit represents the total relatively small resistance of the n- and p- regions and the contacts of these regions with the leads. The non-linear resistance R nl with direct connection is equal to R pr, i.e. is small, and at reverse voltage R nl = R arr, i.e. it is very big. The given equivalent circuit in various frequency cases can be simplified. At low frequencies, the capacitance is very large and the capacitance can be ignored. Then, with a forward bias, only the resistances R 0 and R pr remain in the equivalent circuit (Fig. 7.5b),
Fig.7.5b. Fig. 7.5c.
and with a reverse voltage - only the resistance R arr, since R 0<< R обр (рис.7.5в).
At high frequencies, capacitances have relatively little resistance. Therefore, with a forward voltage, a circuit is obtained according to Fig. 7.5d, (if the frequency is not very high, then C diff has practically no effect),
Fig.7.5d. Fig. 7.5e.
and in the opposite case, R arr and C b remain (Fig. 7.5e).
It should be borne in mind that there is still capacitance C in between the terminals of the diode, which can significantly shunt the diode at very high frequencies. The inductance of the leads can also appear on the microwave.
Classification of diodes.
The classification of diodes is carried out mainly:
1) on technological methods for creating electrical junctions and diode structures
2) according to the function of the diodes.
According to the manufacturing technology, diodes can be point and planar. The main characteristics of point diodes: the p-n-junction area is small, they have a small capacitance (less than 1pF), low currents (no more than 1 or tens of mA). Used at high frequencies up to microwave. Technology: a tungsten filament coated with an acceptor impurity (for germanium - indium, for silicon - aluminum) is welded to a plate of n-type germanium or n-type silicon using a large current pulse.
Planar diodes: manufacturing technology can be either fusing or diffusion. When fusing, a tablet of metal acceptor material, such as aluminum, is placed on the cleaned surface of a semiconductor wafer, usually n-type, if the semiconductor is silicon. When heated to 600 ... 700 0 C, it melts and dissolves the adjacent layer of silicon, the melting point of which is much higher. After cooling near the surface of the plate, a p + -type silicon layer saturated with aluminum (p-type emitter, n-type base). Diffusion: impurity atoms usually come from a gaseous medium into a semiconductor plate through its surface at a high temperature (about 1000 0) and propagate in depth due to diffusion, i.e. thermal movement. The process is carried out in special diffusion furnaces, where the temperature and time of the process are maintained with high accuracy. The longer the time and temperature, the further the impurities penetrate into the depth of the plate. The diffusion pn junction turns out to be flat, and its area is large and equal to the area of the original plate, the operating currents reach tens of amperes.
According to the function performed, rectifier, pulse, converter, switching, detector diodes, zener diodes, varicaps, etc. are distinguished. Separate classes of diodes can be subdivided into subclasses depending on the operating frequency range (low-frequency, high-frequency, microwave diodes, diodes of the optical range). Diodes are also distinguished by semiconductor material: silicon is most widely used, replacing the previously common germanium. Silicon diodes have a higher maximum operating temperature (Si - 125 ... 150 0 C, Ge - 70 ... 80 0 C) and several orders of magnitude lower reverse current. The number of gallium arsenide diodes (in particular, metal-semiconductor ones), which are superior in parameters to silicon diodes, is constantly increasing.
Consider some types of diodes and their main parameters.
1.Rectifier low frequency diodes. They are used in AC power supplies.
The main electrical parameters of the diode are the values of U ex.avg at a given I ex.avg, as well as I arr.avg at a given amplitude (maximum) value of the reverse voltage (U arr.max) values of forward voltage and reverse current for the period). For silicon diodes with a p-n junction, which are most common, U ave.av does not exceed 1..1.5V at T \u003d 20 0 C. With increasing temperature, this value decreases, and TKN depends on the value of the forward current; decreases with increasing current, and at high current it can even become positive. The reverse current of silicon diodes at T = 20 0 C, as a rule, does not exceed tenths of μA, and increases with increasing temperature (the doubling temperature is about 10 0 C). At T=20 0 C, the reverse current can be neglected. The breakdown voltage of silicon diodes is hundreds of volts and increases with increasing temperature.
The forward voltage of silicon diodes with a metal-semiconductor junction is approximately two times less than in diodes with a p-n junction. And the reverse current is somewhat larger and more strongly dependent on temperature, doubling for every 6..8 0 С.
When choosing the type of diode, the maximum allowable rectified current, reverse voltage and temperature are taken into account. Depending on the allowable current, diodes of small (<300мА), средней (<1А) и большой (>10A) power. The limiting reverse voltage is limited by the breakdown of the transition and lies in the range from 50 to 1500V. To increase the allowable reverse voltage, the diodes are connected in series. Several diodes connected in series, manufactured in a single technological cycle and enclosed in a common housing, are called a rectifier pole. The maximum operating temperature of silicon diodes reaches 125..50 0 C and is limited by the growth of the reverse current.
Low-power diodes with a small p-n junction area (less than 1 mm 2) are created by fusing, high-power ones with a large area - by diffusion. Power diodes with a p-n junction can operate up to frequencies usually not more than 1 kHz, and diodes with a metal-semiconductor junction - up to frequencies of hundreds of kHz.
Germanium diodes have a forward voltage approximately 1.5..2 times less than silicon diodes (usually no more than 0.5 V) due to the smaller band gap. It is mainly determined by the voltage drop across the base resistance, in this case, TC U pr >0. The reverse current at T=20 0 C is 2..3 orders of magnitude greater than in silicon diodes, and depends more strongly on temperature. Doubling for every 8 0 C, in connection with this, the maximum operating temperature is much lower (70 ... 80 0 C).
The thermal breakdown mechanism leads to the fact that germanium diodes fail even with short-term impulse overloads. This is a significant disadvantage. The breakdown voltage decreases with increasing temperature.
Due to the small junction area, the maximum allowable direct currents of high-frequency diodes are small (usually less than 100mA), breakdown voltages, as a rule, do not exceed 100V.
3. pulse diodes. Designed for operation in pulsed mode, i.e. in devices for the formation and conversion of pulse signals, key and digital circuits.
The most important parameter of pulsed diodes is the recovery time of the reverse resistance. It characterizes the transitional process of switching the diode from a state with a given forward current I CR to a state with a given reverse voltage U arr. Figure 7.6 shows the timing diagrams of voltage and current through the diode.
The recovery time t is counted from the moment t 1 of the voltage change on the diode from direct to reverse until the moment t 2 when the reverse voltage reaches a value of 0.1 pr. moment t 1), as well as the process of recharging the barrier capacitance. In switching diodes, the recovery time should be as short as possible; it is necessary to reduce the lifetime of minority carriers in the base, for which silicon diodes with a p-n junction are doped with gold. But for silicon diodes, it is not possible to obtain a recovery time of the order of less than 1 ns. In gallium arsenide, the lifetime is much shorter than in silicon, and in diodes with a p-n junction, it is possible to obtain t res of the order of 0.1 ns. The reduction in barrier capacitance is achieved by reducing the transition area. The shortest recovery time (t re<0.1нс) имеют диоды с переходом металл-полупроводник, в которых отсутствует накопление неосновных носителей при протекании прямого тока. В них время восстановления порядка C б r б определяется процессом перезаряда барьерной емкости перехода через сопротивление базы.
For all pulse diodes, the capacitance is specified at a certain reverse voltage and frequency of the AC signal used in the measurement. The minimum capacitance values are 0.1…1 pF.
The specific parameters of pulsed diodes include the maximum pulsed reverse current I rev.i.max and the maximum impulse resistance r pr.i.max, equal to the ratio of the maximum forward voltage in the process of its establishment to the forward current. It is desirable to have the values of these quantities as small as possible.
For pulsed diodes, static parameters are also important, which determine the steady-state values of current and voltage in the circuits. These include forward voltage at a given forward current and reverse current at a given reverse voltage.
4. Zener diodes. A zener diode is a semiconductor diode designed to stabilize voltages in circuits. Zener diodes are used in power supplies, limiters, level clamps, voltage references and other devices. The principle of operation of zener diodes is based on the use of avalanche or tunnel breakdown in the p-n junction. Figure 7.7 shows a typical volt-ampere characteristic of a zener diode with reverse voltage.
In the breakdown section - the working section of the CVC, the voltage depends very weakly on the current. The minimum value of the operating current I st.min corresponds to the beginning of the "vertical" section of the I–V characteristic, where a small differential resistance r diff =ΔU/ΔI is achieved. The maximum current I st.max is determined by the allowable power dissipation. The main parameter is the stabilization voltage U st, which is practically equal to the breakdown voltage, is set at a certain value of the current I st in the working area.
The zener diode switching circuit is shown in Fig. 7.8.
Here R ogr is a limiting resistor; R n - load resistor, the voltage at which U n \u003d U st. The current flowing through the limiting resistor is I \u003d (E-U st) / R ogr, and the current through the zener diode I st \u003d I-I n, where I n \u003d U st / R n, which corresponds to the operating point c in Fig. 3.11. If the power supply voltage deviates by an amount from the nominal value, the current through the zener diode changes by Δ I st = ΔE) / R limit at r diff<<(R огр ││ R н) и рабочая точка перемещается в пределах участка C ’ C”; напряжение на нагрузке изменяется на очень малую величину
(7.11)
If the load current changes and therefore, the load on the value of Δ I n, then the current through the zener diode and Δ U \u003d - r diff ΔI n will change in approximately the same way. The “-” sign means that as the load current increases, the zener diode current decreases. To obtain good stabilization, the differential resistance should be as low as possible.
The breakdown voltage of the p-n junction decreases with an increase in the concentration of base impurities. For devices of various types, U st can be from 3 to 200V.
The effect of temperature is estimated by the temperature coefficient of the stabilization voltage of the TKN, which characterizes the change in voltage U st with a change in temperature by one degree, i.e.
(7.12)
Voltage temperature coefficient can be from 10 -5 to 10 -3 K -1 . The value of U article and the sign of TKN depend on the resistivity of the main semiconductor. Zener diodes for voltages up to 7V are made of silicon with low resistivity, i.e. with a high concentration of impurities. In these zener diodes, the p-n junction has a small thickness, a field with a high intensity acts in it, and the breakdown occurs mainly due to the tunnel effect. In this case, the TKN turns out to be negative. If silicon with a lower concentration of impurities is used, then the p-n junction will be thicker. Its breakdown occurs at higher voltages and is an avalanche. Such zener diodes are characterized by a positive TKN.
The temperature coefficient of stabilization of high-voltage zener diodes can be reduced by 1 ... 2 orders of magnitude, using thermal stabilization. To do this, the back-connected p-n junction of the zener diode is connected in series with one or two p-n junctions connected in the forward direction. It is known that the forward voltage at the p-n junction decreases with increasing temperature, which compensates for the increase in breakdown voltage. Such thermally compensated zener diodes are called precision. They are used as reference voltage sources.
Most often, the zener diode operates in such a mode when the source voltage is unstable, and the load resistance R n is constant. To establish and maintain the correct stabilization mode in this case, the resistance R limit must have a certain value. Usually R ogr is calculated for the midpoint with the characteristics of the zener diode. If the voltage E changes from E min to E max, then R limit can be found using the following formula
(7.13)
where E cf \u003d 0.5 (E min + E max) - the average voltage of the source;
I cf \u003d 0.5 (I min + I max) - the average current of the zener diode;
I n \u003d U st / R n - load current.
If the voltage E begins to change in one direction or another, then the current of the zener diode will change, but the voltage on it, and therefore on the load, will be almost constant. Since all changes in the source voltage must be absorbed by the limiting resistor, the largest change in this voltage, equal to E max - E min , must correspond to the largest possible change in current, at which stabilization is still preserved, i.e. I max - I min. It follows that if the value of E changes by ΔE, then stabilization will be carried out only if the condition
The second possible stabilization mode is used when E=const, and R n varies from R n min to R n max . For such a regime, R limit can be determined from the average values of the currents according to the formula
(7.15)
I n cf \u003d 0.5 (I n min + I n max), and I n min \u003d U st / R n max and I n max \u003d U st / R n min.
To obtain higher stable voltages, a series connection of zener diodes designed for the same currents is used.
5. Varicaps. Diodes are called varicaps, the principle of operation of which is based on the dependence of the barrier capacitance of the p-n junction on the reverse voltage. Thus. Varicaps are capacitors of variable capacity, controlled not mechanically, but electrically, i.e. reverse voltage change. They are used as elements with electrically controlled capacitance in frequency tuning circuits of an oscillatory circuit, frequency division and multiplication, frequency modulation, controlled phase shifters, etc.
The simplest circuit for switching on a varicap to adjust the frequency of the oscillatory circuit is shown in Fig. 7.9.
The control voltage U is applied to the varicap VD through a high-resistance resistor R, which reduces the shunting of the varicap and the oscillatory circuit by the voltage source. To eliminate the direct current through the inductance element, the oscillatory circuit is connected in parallel to the varicap through a high-capacity separating capacitor Cp. By changing the magnitude of the reverse voltage and, consequently, the capacitance of the varicap and the total capacitance of the oscillatory circuit, the resonant frequency of the latter is changed.
The main semiconductor material for the manufacture of the varicap is silicon, gallium arsenide is also used, which provides a lower base resistance.
The electrical parameters of the varicap include capacitance at nominal, maximum and minimum voltages, measured at a given frequency, capacitance overlap coefficient, quality factor, frequency range, temperature coefficients of capacitance and quality factor. In different types of varicaps, the nominal capacitance can range from a few units to several hundred picofarads.
Diffusion capacitance is a virtual capacitance that simulates the effect of the finite time of "resorption" of the non-equilibrium charge of minority carriers in the high-resistance part p-n- transition.
If, as before, we consider the case when the region R is more high-resistance, i.e.
n n >> p n,
then in the area R electrons are minority carriers and their equilibrium concentration is low. When a forward bias is applied, the electrons are the main carriers of the layer n- in large numbers pass into the layer R, creating there a space charge of nonequilibrium minority carriers.
If we abruptly change the applied voltage to the blocking voltage, then the transition of electrons from n-region will stop, but the electrons of the layer n, caught in R- layer (nonequilibrium space charge), will, as minority carriers, return to the layer n, while the space charge of minority carriers in R- region will not decrease to equilibrium. Physically, this means that for some time after the change in voltage from direct to reverse through p-n- the transition will flow a reverse current much greater than the equilibrium value I S(Fig. 3.12, a).
Rice. 3.12. The manifestation of diffusion capacity p-n- transition:
a– at a low signal change rate;
b– at a high rate of signal change
On fig. 3.12, b shows how the diffusion capacitance at a high frequency of voltage changes leads to the loss of the property of one-way conduction pn-transition. It is obvious that the greater the direct current, the greater the nonequilibrium charge, the more time is needed for its absorption (discharge of the diffusion capacitance), the greater the inertia p-n- transition.
3.7. Breakdown pn-transition
An increase in the reverse voltage to a certain critical value causes the phenomenon of an avalanche-like increase in the reverse current, which, if measures are not taken to limit it, will cause destruction p-n- transition. This phenomenon is called breakdown. The physical breakdown mechanism is rather complicated and can be conditionally divided into two types: thermal and electrical.
thermal breakdown
Thermal breakdown can be simplified as follows: when reverse current flows through p-n- transition power is released P=U 0 I 0 , which leads to heating of the bulk of the semiconductor. A positive thermal bond occurs, which, if thermal equilibrium is not ensured (due to efficient heat removal), will lead to thermal destruction p-n- transition. Prevention of thermal runaway is a major engineering challenge and is achieved by limiting the amount of reverse voltage and ensuring good heat dissipation from pn-transition (installation pn-transition to heat-removing plates-radiators, active ventilation).
In an ideal reverse current, even with a relatively small reverse voltage, it does not depend on the value of the latter. However, in real studies, a rather strong increase in the reverse current is observed with an increase in the applied voltage, and in silicon structures, the reverse current is 2–3 orders of magnitude higher than the thermal one. Such discrepancy between the experimental data and the theoretical ones is explained by the hermetic generation of charge carriers directly in the region and the existence of channel currents and leakage currents.
Channel currents are due to the presence of surface energy states that bend the energy bands near the surface and lead to the appearance of inverse layers. These layers are called channels, and the currents flowing through the transition between the inverse layer and the neighboring region are called channel currents.
Capacitances p-n-junction.
Along with electrical conductivity, the -junction also has a certain capacitance. Capacitive properties are due to the presence of electric charges on both sides of the boundary, which are created by impurity ions, as well as mobile charge carriers located near the boundary.
The capacitance is divided into two components: barrier, reflecting the redistribution of charges in , and diffusion, reflecting the redistribution of charges near . In the forward bias of the junction, the diffusion capacitance mainly manifests itself, while in the reverse bias (extraction mode), the charges near (in the base) change little and the barrier capacitance plays the main role.
Since the external voltage affects the width , the value of the space charge and the concentration of the injected charge carriers, the capacitance depends on the applied voltage and its polarity.
The barrier capacitance is due to the presence of donor and acceptor impurity ions in the junction, which form, as it were, two charged capacitor plates. With a change in the blocking voltage, for example, an increase, the width of the junction increases and part of the mobile charge carriers (electrons in the region and holes in the region) is sucked off by the electric field from the layers adjacent to the junction. The movement of these charge carriers induces a current in the circuit
where is the change in the charge of the depleted junction layer. This current becomes equal to zero at the end of the transient process of changing the boundaries of the transition.
The value for a sharp transition can be determined from the approximate expression
where are the area and thickness at .
With an increase in the applied reverse voltage U, the barrier capacitance decreases due to an increase in the thickness of the transition (Fig. 2.10, a).
The dependence is called the capacitance-voltage characteristic.
When a direct voltage is connected to the p-n junction, the barrier capacitance increases due to a decrease in . However, in this case, the increment of charges due to injection plays an important role, and the capacitance of the -junction is determined mainly by the diffusion component of the capacitance.
Diffusion capacity reflects the physical process of changing the concentration of mobile charge carriers accumulated in regions due to changes in the concentration of injected carriers.
The effect of diffusion capacity can be explained by the following example.
Let a forward current flow through due to the injection of holes into the base region. The charge accumulated in the base is created by minority carriers, proportional to this current, and the charge of the majority carriers, which ensures the electroneutrality of the semiconductor. With a rapid change in the polarity of the applied voltage, the injected holes do not have time to recombine and, under the action of the reverse voltage, go back to the emitter region. The main charge carriers move in the opposite direction and leave along the power rail. In this case, the reverse current is greatly increased. Gradually, the additional charge of holes in the base disappears (dissolves) due to their recombination with electrons and return to the -region. The reverse current is reduced to a static value (Fig. 2.10. b).
Rice. 2.10. Capacitance-voltage characteristics (a) and change in current with a change in voltage polarity (o): 1 - smooth transition; 2 - abrupt transition
The junction behaves like a capacitance, and the charge of the diffusion capacitance is proportional to the forward current that previously flowed through the junction.
