JPH0335531A - Manufacture of mos transistor and integrated circuit - Google Patents

Abstract

PURPOSE:To accurately and finely control threshold voltages of individual MOS transistors without reducing the operating speed of the transistor by a simple process technique by radiating a gate region of the transistor with an electron beam for controlling the flow of electrons from a source to a drain. CONSTITUTION:After an oxide film and a polysilicon film are formed on an N-type Si substrate, a gate region is formed with a first mask. Then, source, drain regions are formed with second mask, contact holes for electrodes are formed with third mask, and Al is deposited. Eventually, source, gate, drain electrodes are formed with a fourth mask. After a threshold voltage is previously measured, it is set to a threshold voltage controller, computer-controlled by a radiating position controller 5 to irradiate it with an electron beam by a predetermined dose. The electron beam dose is set to a desired value of a threshold voltage of MOS transistors for forming a MOS-IC according to the current amount of the beam and the scanning speed of the beam.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a method for manufacturing a MOS transistor and an integrated circuit in which MOS transistors manufactured by this method are integrated.

(Prior art) Conventionally, analog circuits using MOS transistors have been slow to develop compared to digital circuits, and their applications have been limited.
The reason for this is due to the characteristics of the MOS transistor, one of which is the problem of threshold voltage.

The threshold voltage of a MOS transistor is determined by the N content of the gate oxide film and the impurity concentration in the semiconductor substrate of the gate region. Therefore, as a method of controlling the threshold voltage of a conventional MOS transistor, the following is explained.

There are two known methods: (a) controlling the thickness of the gate oxide film, and (b) controlling the impurity concentration in the semiconductor substrate in the gate region.

Method (a) simultaneously controls the oxide film thickness of the entire wafer of the MOS transistor, so it is difficult to individually control the threshold voltage of each transistor.

The first method (b) is to control the impurity concentration by injecting impurities into the gate using an ion implanter, and by selecting the ion implantation position using a mask, the threshold value of each transistor can be adjusted. The voltage can be controlled separately. However, in this method, if too many ions are implanted, the implanted impurity will be
! Affects flow characteristics and mobility, '1! Since this reduces the flow characteristics and mobility, the threshold value can only be controlled up to about 2V at most without adversely affecting others.

Furthermore, in actual Si gate MO3LSI, it is extremely difficult to control the IJfJ voltage to the full voltage due to the redistribution of implanted ions due to the high temperature treatment accompanying the Si gate process.

In addition, in this method (b), a mask material such as photoresist is used to select the ion implantation location, so if transistors with different threshold values are to be manufactured on one wafer L, prepare as many masks as necessary. However, since it is necessary to perform ion implantation for the number of masks, as the number of transistors whose threshold voltages are to be controlled increases, the number of masks also increases, and the number of additional steps increases. Furthermore, a high temperature treatment (annealing) step is also required to activate the ions after ion implantation. As a result, the number of steps increases, the process becomes more complex,
In addition to the resulting decrease in yield, the amount of change in the threshold value does not change linearly with the amount of ion implantation due to high-temperature processing, and tends to decrease somewhat, but this decrease is absorbed by the development process. Accurate threshold control is not possible because it is only possible to grasp the target.

Another problem caused by high-temperature processing is that the implanted ions spread out due to re-spun, making it difficult to concentrate the implanted ions in a narrow enough area relative to the gate length. Therefore, there is a problem that a desired threshold value cannot be obtained.

On the other hand, in method (b), it is necessary to implant the necessary amount of impurities to change the impurity concentration of the substrate at the initial stage of the process, that is, before the first step of forming the gate film on the substrate. , in addition to the above-mentioned problem that the threshold value changes due to subsequent processing, the final threshold value is changed to [!
I can't think of anything to do.

For the reasons mentioned above, MOS transistors whose threshold values are controlled using conventional methods (a) or (b) have uneven source-drain voltage-current characteristics, making it difficult to create integrated circuits that operate in an analog manner. The problem is that it is difficult to configure.

(Objects and Structure of the Invention) The present invention has been made in view of the above points, and uses a simple process technology to accurately calculate the VIJ value voltage of each MOS transistor without reducing the operating speed of the MOS transistor. In order to achieve this goal, the gate region of the MOS transistor is irradiated with a predetermined amount of electron beam to control the flow of electrons from the source to the drain.

(Example) The present invention will be explained below based on the drawings.

FIG. 1 schematically shows the configuration of a threshold voltage control device used to control the threshold voltage of a MOS transistor according to the present invention, in which an electron beam EB generated by a filament 1 is focused by a condenser lens 2, and The amount of beam is adjusted to an appropriate amount by the aperture 3, and a predetermined amount is irradiated onto an appropriate position on the wafer 6 by the deflection coil 4, which is controlled by the irradiation position control device 5 having a built-in computer. The irradiated electron beam may be a Gaussian beam with a Gaussian current density distribution, or may be a shaped beam with a rectangular shape, for example.

The method of irradiating the electron beam is as shown in FIG. As shown in Figure (b), the gate region between the source S and drain D may be partially irradiated linearly as shown by the broken line (partial irradiation).

Figure 3 shows the source-drain gap vI) and the source-drain open current (2) when the entire gate region of an actual p-MOS transistor is uniformly irradiated with an electron beam. V represents the relationship with. - This is the T end characteristic. In the figure, the electron beam irradiation amount is A is 0, B, C,...
The threshold voltage of the MOS transistor V Th I * increases as the number of irradiated electron beams increases.
It can be seen that V Th2 r... becomes smaller. Furthermore, what is noteworthy about this characteristic is that although the threshold value changes depending on the amount of electron beam irradiation, the slope thereof, that is, the resistance value between the source and drain hardly changes, and a substantially parallel characteristic is obtained. Such characteristics are advantageous in creating integrated circuits that perform analog circuit operations.

On the other hand, Figure 4 is a plot of the change in threshold voltage when the entire gate region is irradiated, and the horizontal axis is l.
It is the number of irradiated electrons per cm". From this figure, it can be seen that the amount of change in the threshold voltage is proportional to the index of the irradiation amount of the electron beam. From this characteristic, 1lII depending on the irradiation amount of the electron beam.
It can be seen that the value voltage can be controlled.

FIG. 5 shows the change in mobility when the entire gate region is irradiated, and the irradiation amount is the same as in FIG. 3. When irradiated with electron beam (B, C, D, E, F)
The change in mobility shifts to the negative side by the threshold voltage compared to the case (A) without electron beam irradiation, but the shape is almost the same, and the rate of decrease in mobility is about 20% of that in case A. It is kept within.

Figure 6 shows the change in source-drain current 1 in the case of partial irradiation in which a part of the gate region of a MOS transistor is irradiated with an electron beam as shown in Figure 2 (b), but the threshold voltage changes. Since only a portion of the gate region of the MOS transistor is exposed, the V or ID characteristics are as shown in the figure, and once the channel is opened, the MOS transistor
The current flowing through the S transistor is approximately equal to the current flowing through a MOS transistor that is not irradiated with the electron beam. In this way, even if the threshold voltages are different, the source-drain voltage-current characteristics are the same, which is advantageous when applied to an integrated circuit that operates in an analog manner.

In the above, the threshold voltage control in the present invention has been explained from the low point of full irradiation and partial irradiation to the gate region. However, as an electron beam irradiation method, a focused electron beam is used as shown in FIGS. In addition to the method of sweeping linearly and irradiating as shown in b), it is also possible to irradiate the entire gate area with an unfocused electron beam at once, or to irradiate with a focused electron beam by sweeping nonlinearly. Well, in short.

The gate region may be irradiated with an electron beam so as to control the flow of electrons from the source to the drain of the MOS transistor.

Incidentally, in the present invention, the gate region of the MOS transistor is irradiated with an electron beam after the process is completed.
This was after the OS transistor was completely sold out.

At this stage, the threshold voltage of the transistor is measured, and while checking the measured value, the amount of electron beam irradiation is controlled so that the desired threshold value is achieved. In this way, the VR value voltage can be adjusted to the designed value in order to perform threshold value control in the final process, and an integrated circuit with desirable characteristics can be realized.

The basics of threshold voltage control of a MOS transistor according to the present invention have been explained above. First, a method for manufacturing an integrated circuit using the threshold voltage control method according to the present invention will be explained.

A normal MOS-IC manufacturing process can be used as is. In other words, p-MOS-
To give an example of the IC process, first, oxidation is performed on an nNjSi substrate! After forming I (S i Og) and a polysilicon film, a first mask is applied to form a gate region. Next, a second mask is applied to form source and drain regions. Next, a third mask is applied, contact holes for electrodes are formed, and pattern A is deposited. Finally, a fourth mask is applied to each of the source, gate, and drain 'Iv,
form a pole.

Each MOS constituting the MOS-I C manufactured in this way
After measuring the threshold voltage of the transistor in advance, it is set in the threshold voltage control device shown in FIG. 1, and a predetermined amount of electron beam is irradiated to the gate region of each transistor of the MOS-IC under computer control by the irradiation position control device 5. do. The amount of electron beam irradiation is controlled for each transistor by the amount of current of the electron beam and the scanning speed of the beam. In this way, the threshold voltage of each MO3-transistor constituting the MOS-IC is set to a desired value.

Next, the MOS-I whose threshold voltage is individually controlled in this way
An example of an integrated circuit constructed using C will be described.

The first integrated circuit example is the A/D converter shown in Figure 7, which converts an analog signal in the range of 0 to -1.5V into a 4-bit digital signal and outputs it. be.

The A/D converter is composed of a variable threshold voltage circuit lO, an exclusive OR circuit 20, and a hexadecimal-binary encoder 30, and here, in all logic circuits,
-5V is "o" and ov is "l". The variable threshold voltage circuit 10 has a threshold voltage controlled by the method according to the invention.
Five p-MOS transistors Tr, -Tr, ,
The drain of each transistor is connected to a power supply voltage VD of -5V, and the source is grounded. When an analog signal V in is input from input terminal A, output terminals B+, B2 . From Bz, .B+s, either -5V or Ov is output as an output signal.The threshold voltage of each transistor is set to a different value by the method of the present invention, and the threshold voltage of transistor Tr is 10, IV, the threshold voltage of the transistor Tr2 is -0.2V, . . . The threshold voltage is increased by 0.1 in sequence, and the threshold voltage of the transistor Tr2 is -1.5V.

The exclusive OR circuit 20 is also composed of 15 exclusive OR circuits E.
xOR+, ExORz, ExOR* ...ExOR,
5, and is configured to input output signals from two adjacent transistors of the variable threshold type rE circuit/circuit 10. for example.

The output signals of transistors T r and T r are input to ExOR, and the output signals of transistors T
The output signals of r2 and Tr= are input. The last ExOR+s has a transistor T r ,,
The output signals of and T r , , are input. Note that the output signal of the transistor Tr and the ground potential are input to the first E x ORl.

The hexadecimal-binary encoder 30 has 15 exclusive OR circuits ExOR+ and ExORt that constitute the exclusive OR circuit 20.
, ExOR=, -ExOR+s and the output signals from the transistors T r , , constituting the variable threshold voltage circuit 10, a total of 16 output signals are inputted and output as 4-bit signals to the output terminals D + , D. 2. D:l,
A known circuit configuration output to D4 is shown.

Next, the circuit operation of this A/D converter will be explained.

If we now input an analog input signal V of 1, OV to the input terminal A of the variable threshold voltage circuit 10, the threshold voltage will be -
Transistor T r 1 from 0,IV to -1,Ov
Since S-T r to is turned on, its output terminal 81~
1'' (Ov) is output to all B IQ, and the remaining transistors Trl+ to Tr and 5 are OFF, so all output terminals Bll to B4 are output ``O'' (-5V). Table 1 below shows the outputs at the output terminals B and Bls when an analog input signal (2) from Ov to -1.5V is input.

When such an output signal from the Ti (variable IsJ value voltage circuit lO) is input to the exclusive OR circuit 20, its output terminal C0
~C8 outputs a digital signal as shown in Table 2 below. When the output signal from this exclusive OR circuit 20 is passed through a hexadecimal-binary encoder 3o, output terminals D + ,
A 4-bit binary digital signal as shown in Table 3 below can be obtained from D2, D:l, and D4.

For example, the input signal V to the variable dark value voltage circuit 10 described above
,. When is -1, OV, the output signal of the hexadecimal-binary encoder 30 becomes "1010", which means that A/D conversion has been performed.

As described above, by realizing a MOS transistor in which the threshold voltage is finely changed in steps of O and IV according to the present invention, a highly accurate A/D converter can be obtained.

A second example of an integrated circuit is the operational amplifier shown in FIGS. 8 and 10. In both cases, -24V is used as the power supply voltage VD.

The two operational amplifiers shown in FIG. 8 and FIG. 1O have the same basic circuit configuration, but have different amplification characteristics.

This is because it is known that the resistance when the MOS transistor is turned on is proportional to the ratio of the channel width W to the channel length, that is, the reciprocal of W/L, so changing the value of W/L can improve the amplification characteristics. It is a changed version of .

The operational amplifier in Figure 8 has the following threshold voltage vT and W/
10 CQp-MO3)-IF transistors Tr, ~Tr, and another MOS transistor Tr, which is always turned on by a separate power supply regardless of the input signal, are illustrated. It is connected and configured as follows.

(Left below) Transistor threshold voltage V, (V) W/LT
r + O, l l/176T
r 2 0, 2 1/146.7
Try O, 31/120Tr,
-0,4I/96Tr 5-0,5 1/74.7Tr
r, 0, 6 1156Tr
70.7 +/40 T". -0,81/26.7 T". -0,91/16 Tr,, -1,01/8 In this operational amplifier, the signal V input to input terminal A
For example, when In is V, and n>-0, IV, all transistors Tr, . ut is Ov, but when the input signal V + n is 10, IV≧V, n>-0,2V"?, only the transistor Tr is turned on, so the output terminal B has the transistor Tr turned on. Yesterday's resistor and transistor T', ON
The output signal V is divided by the resistor at the time. It outputs 2v of ut. When the input signal V in is, for example, 10.45V, the transistors T r , ~T r , are ONL.
, , other transistors Tr, , ~Tr, o are OFF
Therefore, the output signal V o u L is -8V. Output signal V a u for other input signal V An
The change in L, that is, the amplification characteristic, is as shown in FIG.

Looking at FIG. 9, the amplification characteristics are step-like.

This can be made closer to a straight line by controlling the threshold voltage Va more finely.

Is the operational amplifier in Figure 1O the same as the operational amplifier in Figure 8 in its basic circuit configuration, or is there no transistor equivalent to transistor Tr, and the threshold voltage of each transistor is the same in O, IV increments? /L is a completely different value as follows.

(Left below) Transistor threshold voltage V, (V) W/LTr
+ o, 1 1/: 168T
r: l O, 3V337T r 4
0, 4 +/: 108Tr,
-0,51/196 T r s 0,6 1/115
Trt -0,71/82 T r O,81/:19 T r 9
-0,9 1/16.7Tr,,
-1,0! /3.2 The circuit operation of this operational amplifier is exactly the same as the amplifier shown in Fig. 8, but since the amplification characteristics are square-law characteristics as shown in Fig. 11,
There is no transistor Tr2 with a threshold voltage vT of -0.2V, and as the threshold voltage approaches 10 and IV, W/
The rate of change in L is increasing.

By changing the threshold voltage vT and W/L of each transistor in combination in this way, in addition to linearity as shown in FIG. 9 and square law characteristics as shown in FIG. 11, exponential characteristics,
It can have arbitrary characteristics such as polygonal line characteristics.

The characteristics of these operational amplifiers are as follows.

(Since the lea configuration is simple, the operating time is basically determined by the ON/OFF of the MOS transistor, so it is not affected by the slew rate.The slew rate is the change that can be made per unit time (usually 1g5) of the output voltage of the operational amplifier. When the level of the input signal is large, the output signal cannot change at a rate of change greater than the slew rate, and the output signal waveform is distorted, so conventional operational amplifiers are unable to follow changes in the input signal. There was a problem.

(2) All circuit elements are composed of MOS transistors, and it is easy to combine with other functional ICs to create a single chip with the same function.Furthermore, the threshold voltage can be set arbitrarily, so when combined with W/L, input The amplification characteristics of the 0-wire operational amplifier, which allows you to freely set the corresponding range of voltage and output voltage at the design stage, are step-like, but by increasing the resolution of the threshold voltage, W/
By setting L finely, the characteristics can be brought closer to an ideal curve as shown by the broken line in FIG.

Although two types of integrated circuit examples manufactured according to the present invention have been described above, it goes without saying that the present invention can be similarly applied to other integrated circuits.

(Effects of the Invention) As explained above, in the present invention, the threshold voltage is controlled by irradiating the gate region of the MOS transistor with a predetermined amount of electron beam that controls the flow of electrons from the source to the drain. Therefore, the threshold voltage of each transistor can be set individually and precisely without the need for masks or multiple processes. Therefore, analog circuit design becomes possible, and analog integrated circuits can be realized for various purposes.

In particular, even if the source-drain voltage-current characteristics or threshold values are different, the characteristics are parallel or coincident, which widens the range of analog design. Furthermore, since ions are not implanted into the semiconductor substrate, there is no reduction in the mobility of holes or electrons, and therefore, there is no reduction in the operating speed of the transistor.In this sense, there are various methods using MOS transistors. Functional integrated circuits can be manufactured, and each integrated circuit has improved accuracy and characteristics compared to integrated circuits manufactured using conventional MOS transistors.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram showing an example of the threshold voltage control device according to the present invention, and the fJIJz diagram shows the irradiation mode of the electron beam according to the present invention, where (a) shows full irradiation, (b) shows partial irradiation, and FIG. is a graph showing the relationship between the source-drain voltage and source-drain current of a transistor when the electron beam irradiation amount is changed in the case of full-surface irradiation, and Figure 4 shows the change in threshold voltage with respect to the electron beam irradiation amount. Figure, 5th
Figure 6 shows the relationship between the source-drain voltage and mobility of a transistor when the electron beam irradiation amount is changed.
The figure is a graph showing the relationship between the source-drain voltage and the source-drain current of a transistor when the electron beam irradiation amount is changed in the case of partial irradiation. Figure 8 is a circuit diagram of an operational amplifier as an example of an integrated circuit manufactured by controlling the threshold voltage according to the present invention. Amplification characteristics of the operational amplifier. FIG. 1O shows a circuit example of an operational amplifier as another example of an integrated circuit manufactured by controlling the threshold voltage according to the present invention. FIG. 11 shows the amplification characteristics of the operational amplifier shown in FIG. l...Filament, 2...Condenser lens, 3
... Squeeze, 4... Deflection coil, 5... Irradiance control device, 6... Wafer

Claims (4)

[Claims] (1) A method for manufacturing a MOS transistor, which comprises irradiating a gate region of the MOS transistor with a predetermined amount of electron beam so as to control the flow of electrons from the source to the drain. (2) The method for manufacturing a MOS transistor according to claim 1, wherein the electron beam is reciprocated over the gate region multiple times to sweep the irradiation. (3) An integrated circuit comprising a plurality of MOS transistors manufactured by the method according to claim 1. (4) The integrated circuit according to claim 3, wherein W/L of each MOS transistor is set to a desired value.