DD292549A5 - ARRANGEMENT FOR DAMPING THE SEQUENCE ELECTRON RECYCLED OUTPUT SIGNAL - Google Patents

Abstract

The invention relates to an arrangement for damping the secondary electron multiplier output signal. The invention is applied in the surface inspection with Streulichtabtastern, z. As in semiconductor manufacturing, since the radiation detector is always loaded with intense light pulses when the scanning laser beam moves over the rounded wafer edge. Because of the highly sensitive particle detection in the sub-mm range, the SEV is always strongly dazzled. Other applications arise everywhere where a process is not only associated with an intense light pulse, but is triggered by this. These include all suggestions with pulsed radiation sources, such. B. by accelerators and pulsed light sources. But also on the reverse radiation situation, the invention is applicable, d. h., When a reaction associated with much radiation is to be triggered by a weak light signal. This is z. As in the control of bubble chambers of the case. The arrangement is characterized in that the cathode of the secondary electron multiplier and one of its dynodes are each connected to the poles of a switch, the on / off control of the switch with a Daempfungssteuerung, the cathode of the secondary electron multiplier with the negative pole of the operating voltage. Fig. 1 {shutter protection; Sekundaerelektronenvervielfacher; SEV; Daempfungssteuerung; Glare due to intense light pulses; Cathode; Wafer edge; Radiation detector; Light pulse; dynodes}

Description

Field of application of the invention

For economic reasons, in the production of HL components in microelectronics, the widest possible utilization of the wafer is demanded to the edge.

The surface inspection of Sl wafers must take account of this with a high accuracy of particle counting over the entire surface up to the edge region of the wafer.

During surface inspection with scattered light scanners, the radiation detector is subjected to intensive light pulses whenever the scanning laser beam moves over the rounded wafer edge. Because of the highly sensitive particle detection in the Sub ^ m range, the SEV is strongly blended each time. This can lead both to Lirstörung the SEV and subsequent electronic components, as well as the measurement of the extremely weak, immediately following particle signals difficult or even impossible.

Other applications arise wherever a process is not only associated with an intense light pulse, but is triggered by this. This includes all excitations with a pulsed radiation source, such as e.g. through accelerators and pulsed light sources. But also on the reverse radiation situation, the invention is applicable, d. h "when a reaction associated with much radiation is to be triggered by a weak light signal. This is e.g. in the control of bubble chambers the case.

Characteristic of the known state of the art

The switching off and on of an SEV (by "switching off" is here also meant a very great reduction of its sensitivity) or in general the control of its amplification or its sensitivity to a greater or lesser degree are realized in the state of the art in a very diverse manner. Since the first beginnings of 1939, essentially three main processes can be distinguished (without control with crossed electric or magnetic fields):

1. Control of the SEV operating voltage Ue

2. Generation of a reverse voltage between a control electrode, which is not identical to a dynode, and the photocathode.

3. Short circuit of dynodes or control of the potential difference between them; Increasing the resistance of voltage divider elements to infinity. ^N

An effective switching off of the SEV with respect to damping and switching time after 1. requires a very strong change of the operating voltage of the SEV (DE 1564097: U = 400 V, 10 x). In another example, the gain of a conventional SEV decreases when its operating voltage is reduced from Ue = -1200V to Ue = -600V only by a factor of 4 x 10 ² . This method fails due to known technical limitations, since the required high-performance, low-impedance SEV power supplies are not able to switch high operating voltage reduction fast enough (Farinelli, Malvano, 1958). In addition, during the time of low operating voltage, the linearity range of the SEV and its upper limit frequency are lowered (EP 0155377). In addition, the strongly fluctuating base temperature of the heat-sensitive SEV due to the strong fluctuations of the voltage divider current.

The method according to 2. can not eliminate all the deficiencies of the prior art. Either additional power sources are necessary for the generation of steeper and occasionally also more intense (up to 110V) countervoltage pulses (Farinelli, Malvano, 1958, US 2951941, EP 0155377) or, for example, interference pulses are generated at the SEV output and / or the reset is activated runs slowly (note in Roose, 1965, Keller, Nefkens, 1964). In 2-grid control systems with a pre-acceleration grid Si (positive with respect to the cathode) and a control grid S ₂ (negative with respect to the cathode), disturbing residual pulses are produced by groups of electrons which were originally blocked and after passing through S ₁ three times during a restart against S ₂ into one Opening time fall (Pohl, 1968). Finally, not all SEVs that are more or less suitable for surface inspection have a control electrode (Hamamatsu, 1988), which is not only due to complicated construction and higher manufacturing costs.

Although the method according to 3 saves (in the case of the short-circuit principle) the additional powerful voltage source, it still has considerable shortcomings:

- small attenuation factor of the SEV output signal (Roose, 1965: <90 x)

- slow reconnection (Roose, 1965: => 3μβ; Keller, Nfifkens, 1964: = <30μβ)

- glitches at the S V output (Roose, 1965, Keller, Nefkens, 1964)

Since in general and also in the surface inspection of the SEV by setting different operating voltages Ub must have different measuring ranges (sensitivity ranges), eliminating all voltage divider arrangements in which the non-shorted, active dynodes e.g. Zener diodes on stabilized, fixed dynode potentials. This has in the prior art (except for DE 2353573, dependent claim 10) further disadvantages:

- Reduction of the linearity range of the SEV

- Variation of the SEV temperature

Object of the invention

It should be guaranteed with little circuit complexity optimal Waferqusnutzung or equivalent good performance characteristics of each device (depending on the application).

Explanation of the essence of the invention

The object was to provide an arrangement for damping the secondary electron multiplier (SEV) output signal, which realizes a high attenuation with short switching times, widest possible linearity range, without glitches, without constructive additives such as control electrodes and the lowest possible power of the power supply. According to the invention the object is achieved in an arrangement for damping the secondary electron multiplier (SEV) output signal in which a SEV is connected to a voltage divider, an anode resistor and an operating voltage, characterized in that the cathode of the SEV and one of its near-cathode dynodes with the poles a switch, the EinVAussteuerung the switch with a damping control the Kr'.hode the SEV is connected to dom negative pole of the operating voltage. It is advantageous to use cathode-near dynodes. In the case of damping, the otherwise open switch closes and the dynode is connected to the cathode potential. This reduces both the number of primary photoelectrodes transferred from the cathode to the first dynode and reduces the secondary electron multiplication. It is advantageous to insert Zener diodes in the voltage divider for the dynodes between the cathode and the dynode connected to the switch. This shortens the switching times (compared to ohmic resistances).

The realization of the arrangement for damping the SEV output signal, in which an SEV is connected to a voltage divider, an anode resistor and an operating voltage, can be further solved by connecting the cathode of the SEV and one of its cathode-near dynodes to the poles of a switch, the cathode of the SEV is connected via a voltage source to the negative pole of the operating voltage, the voltage source to a parallel switch, the EirWAussteuerung both switches with a damping control. Since, according to the first implementation proposal, the current through the voltage divider and thus the voltage difference between the non-shorted anode-near dynodes increases, which has a negative effect on the attenuation, an equivalent of the voltage applied to the open switch, when closing the selbigen between the cathode and negative operating voltage pole switched, z. B. by the otherwise closed second switch opens and releases the shorted components same.

The constancy of the voltage divider current must be ensured even during the switching edges, in order to achieve short switching times and to prevent glitches. This requires a simultaneous course of the cut-in edge of the compensation voltage and the Ausschaltflanke the voltage between the dynode and cathode (short circuit) when switching on the damping and analogue when switching back. With transistors as switches, this is relatively expensive or even impossible. Advantageous is the replacement of the resistors commonly used in the voltage divider and the resistors used by Zener diodes according to this second solution variant, because this steep, linear switching edges are realized without the asymptotically slow turn-on of the switching transistor needs to be shortened with aufwanderhöhen circuit means , The reason for this advantageous effect of the Zener diode is the fact that the maximum SEV voltage Tag'lerstrom is reached suddenly and even at very low resistance of the switching transistor, namely, when the voltage drop across the transistor of the Z voltage of the Zener diode equivalent. In contrast, the maximum voltage divider current in ohmic Spannungssteilerwiderständen is only then and gradually adopted when the resistance of the switching transistor is much greater than the sum of the short-circuited voltage divider resistors, which are generally some lOkOhm to several lOOkOhm. The realization of the arrangement for damping the SEV output signal, in which a SEV is connected to a voltage divider, an anode resistor and an operating voltage, can be further solved by connecting the cathode of the SEV and at least one of its dynodes to the poles of a switch, the cathode of the SEV is connected via a voltage source to the negative pole of the operating voltage, another dynode of the SEV via another switch with the negative pole of the operating voltage and the on / off control of both switches with a damping control. As a result, all dynodes between the cathode and the dynode connected to the negative pole of the operating voltage are at a more negative potential than the cathode, which leads to a further increase in attenuation. The realization of the arrangement for damping the SEV output signal, in which a SEV is connected to a voltage divider, an anode resistor and an operating voltage, can be further solved by connecting the cathode of the SEV and at least one of its dynodes to the poles of a switch, the cathode of the SEV via two series voltage source with the negative pole of the operating voltage, another dynode of the SEV via a second switch with the negative pole of the cathode nearer voltage source, the other voltage source with a third parallel switch and the on / off control of three switches is connected to a 6r damping control. This realization shows a possible combination of the essential features of variants 2 and 3.

embodiments

Fig. 2: SEV-BS without reverse voltage and with current compensation, Fig. 3: SEV-BS with reverse voltage and without current compensation Fig. 4: SEV-BS with reverse voltage and with current compensation

Fig. 5: SEV-BS with reverse voltage and without current compensation with pure ohmic voltage divider, Fig. 6: SEV-BS with reverse voltage and with operating voltage independent current compensation with pure ohmic voltage divider.

Figs. 1-3 show a secondary electron multiplier 1 (SEV) in the conventional design with photocathode K and anode A, between which z. B. 11 dynode stages D ₁ -D ₁ I are arranged. With the aid of a voltage divider 2, the operating voltage is applied to the dynodes of the secondary multiplier 1, that between the cathode and anode gradually increasing voltages are effective. In FIGS. 1 and 2, the necessary voltages are generated at the cathode-near dynode stages D ₁ -D ₃ instead of otherwise conventional ohmic resistances by zener diodes 5, 6, 7. Switching transistor 3 is driven by high-frequency transformer 4 of the damping control. The high-frequency transformer 4 ensures the galvanic decoupling to the high voltage side of the voltage divider 2 and allows the realization of any long turn-off and turn-on of the anti-glare. In normal operation (undamped), the switching transistor 3 is locked, controlled by in the case of damping. In the case of damping, the dynodes 1, 2 and 3 are approximately at cathode potential, which reduces the sensitivity of the SEV. The fact that, however, the voltage increases over the remaining resistors of the voltage divider 2, this attenuation is partially compensated by higher secondary electron multiplication. If one puts on the cathode in Fig. 1 z. B. a negative

High voltage U = -1200V, the anode is at Massepotenial and have the Zener diodes 5,6,7 a voltage U _z = 100V, so when shorted cathode dynode 3, the anode current of a dazzle, which flows through the anode resistance to a damping factor 10 ⁶ x reduced, the switching times of the anti-glare are less than 1 με. Dampening factors of> 10 ⁹ x are necessary for the surface inspection, so that the damped iris impulse does not affect the particle detection at the lower detection limit. Switchover times of <1 με are required for an edge exclusion of 1 mm.

In the event of a short circuit, the current through the voltage divider 2 increases from I = 2mA to I = 2.7mA and thus to the voltage across each of the 9 dynode resistors from U = 100V to U = 133V and the total voltage across all resistors from U = 900V to U = 1200V. This requires an increase in the secondary multiplication of these short-circuit-free dynode stages by a factor of 8x. This factor can be compensated if, according to FIG. 2, the blond protection works in current compensation J. h.

the voltage divider current is unchanged even in the case of a short circuit I = 2 mA. By the current compensation is also achieved that the linearity range dqs unattenuated and attenuated SEV is equal to high anode currents. In Figure 2, the cathode K (in Kurzschlußfa. ¹ , ie "attenuated") is at a higher potential than the negative pole of the operating voltage source, because to the Zener LModen 5,6,7 between the cathode and dynode 3 Zener diodes 8 identical , 9,10 between the cathode and the negative pole of the operating voltage source were used

Switching transistor 11 (via high-frequency transformer 12) takes place inversely to switching transistor 3, that is, in the event of a short-circuit across the zener transistor.

Diodes 5, 6, 7, the bridged voltage is switched (by removing the short circuit via Zener diodes 8, 9, 10) between the cathode and the negative operating voltage. The voltage across the remaining resistors of the voltage divider 2 remains constant. In Figure 3, the Zener diodes 8,9,10 are replaced symbolically by switch 21 by a resistor 14 and switching transistor 3 / high-frequency transformer. Switching transistor 11 and high-frequency transformer 12 omitted. Between negative operating voltage and dynode 1 is a switch 22, at the same time works with switch 21. Thus, in the case of attenuation, dynodes 1 and 2 become more negative than the cathode, which in combination with the short-circuit cathode dynode 3 results in a very high attenuation value. Between the cathode and zener diode 5, a diode 13 is additionally connected in series (avoidance of a short circuit between the cathode and the negative operating voltage when the switch 22 is closed). If an operating voltage of U = -1600 V is used in FIG. 3 and the ohmic resistance is 14 = 4 × 5OKQ and the Z voltage of the Zener diodes 5, 6, 7 is 100 V in each case, the operating voltage between negative pole of the operating voltage is simultaneously short-circuited and dynode 1 by means of switch 22 and between cathode and dynode 1 generates a counter-voltage of U = 220V between the cathode dynode 1. The dynodes 2 and 3 have the same potential as the cathode. The current through the voltage divider 2 increases in the event of a short circuit of I = 2 mA to I = 3.1 mA, which leads to an increase of the secondary multiplication of the short-circuit-free dynode stages by a factor of 25 x.

FIG. 4 shows, analogously to FIG. 2, the compensation of the voltage increase at the voltage divider resistors. In this case, the zener diodes 16-20 are switched inversely by switch 23.

FIGS. 5 and 6 show a variant in which the zener diodes 5, 6, 7 and 16-20 are replaced by ohmic resistors 14, 15 and 24, 25, 26.

If a negative voltage U = -1600V is used in FIG. 5, the anode is at ground potential, the ohmic resistors 24, 25, 26 are each 5OK, and the ohmic resistor 14 is 4 × 5OK, becomes short-circuit between negative pole and dynode 1 by means of switch 22 and between the cathode and dynode 3 by means of switch 21 generates a counter-voltage of U = 96V between the cathode dynode 1 and a reverse voltage of U = 48 V between the cathode dynode 2; Dynode 3 has the same potential as the cathode. The current in the voltage divider increases in the event of a short circuit of I = 2mAlmp = 3.3mA, which leads to an increase in the secondary multiplication of the short-circuit-free dynode stages D ₃ -D _n by a factor of 41 χ. This factor can be compensated if, according to FIG. 6, an otherwise short-circuited ohmic resistance 15 = 321 Ω is released for current compilation. The arrangement according to FIG. 6 has the advantage that the current compensation is maintained independently of the respectively selected operating voltage of the secondary electron multiplier.

Claims (4)

1. An arrangement for attenuating the Sekundärelektronenvervielfacher output signal, in which a Sekundärelektronenvervielfacher is connected to a voltage divider, an anode resistor and an operating voltage, characterized in that - The cathode of the photomultiplier (1) and one of its dynodes each with the Poland a switch, - the on / off control of the switch with a damping control - The cathode of the photomultiplier (1) with the negative pole of operating voltage
connected is. 2. An arrangement for attenuating the Sekundärelektronenvervielfacher output signal, in which a Sekundärelektronenvervielfacher is connected to a voltage divider, an anode resistor and an operating voltage, characterized in that the cathode of the photomultiplier tube (1) and one of its dynodes, each with the poles of a switch, - The cathode of the photomultiplier (1) via a voltage source to the negative pole of the operating voltage - The voltage source with a parallel switch - The input / output control of both switches is connected to a damping control. 3. An arrangement for attenuating the Sekundärelektronenvervielfacher output signal, in which a Sekundärelektronenvervielfacher is beschältet with a voltage divider, an anode resistor and an operating voltage, characterized in that the cathode of the photomultiplier tube (1) and one of its dynodes, each with the poles of a switch, - The cathode of the photomultiplier (1) via a voltage source to the negative pole of the operating voltage - Another dynode of the secondary electron multiplier (1), which lies between the dynode connected to the cathode via the first switch and the cathode, which via another switch with the negative pole of the operating voltage - The input / output control of both switches is connected to a damping control. 4. An arrangement for attenuating the Sekundärelektronenvervielfacher output signal, in which a Sekundärelektronenvervielfacher is connected to a voltage divider, an anode resistor and an operating voltage, characterized in that the cathode of the photomultiplier tube (1) and one of its dynodes, each with the poles of a switch, - The cathode of the photomultiplier (1) via two series voltage sources with the negative pole of the operating voltage, - Another dynode of the photomultiplier (1), which is located between the connected via the first switch to the cathode dynode and the cathode, via a second switch to the negative pole of the cathode nearer voltage source - The other voltage source with a third parallel switch - The EinVAussteuerung the three switches is connected to a damping control. For this 6 pages drawings