WO2018105336A1

WO2018105336A1 - Resistivity measuring method

Info

Publication number: WO2018105336A1
Application number: PCT/JP2017/041005
Authority: WO
Inventors: 史高久米; 光義船木; 久寿樫野
Original assignee: 信越半導体株式会社
Priority date: 2016-12-08
Filing date: 2017-11-15
Publication date: 2018-06-14
Also published as: JP2018098271A; TW201821813A; JP6714867B2; TWI693411B

Abstract

The present invention provides a resistivity measuring method comprising: a mercury electrode forming step of forming a mercury electrode by bringing mercury into contact with a semiconductor single crystalline wafer; a forward bias voltage application step of applying a forward bias voltage to the mercury electrode; a C-V measuring step of applying a reverse bias voltage to the mercury electrode, and measuring a varying capacitance of the semiconductor single crystalline wafer; and a resistivity computing step of computing a resistivity on the basis of a relationship between the reverse bias voltage and the capacitance. In this way, a resistivity measuring method using a C-V technique is provided by which the resistivity of a semiconductor single crystalline wafer can be more reliably measured.

Description

Resistivity measurement method

The present invention relates to a resistivity measuring method, and more particularly to a resistivity measuring method by a CV method.

Conventionally, as one of methods for measuring resistivity, a CV (capacitance-voltage) method using mercury as an electrode is known. In the CV method using mercury as an electrode, a Schottky junction is formed by bringing a mercury electrode into contact with the surface of a semiconductor single crystal wafer such as a silicon epitaxial layer, and the reverse bias voltage is continuously changed from 1 V, for example, to the mercury electrode. When applied, the depletion layer is expanded inside the semiconductor single crystal wafer to change its capacity.

Then, the dopant concentration at the desired depth is calculated from the relationship between the reverse bias voltage and the capacitance, and the dopant concentration is converted into the resistivity using a conversion formula such as ASTM STANDARDS F723. Hereinafter, converting the dopant concentration to resistivity using the conversion formula of ASTM STANDARDS F723 is simply referred to as ASTM conversion.

When measuring the electrical characteristics of a semiconductor single crystal wafer repeatedly using a mercury electrode, this occurs when mercury moves to ground the mercury reservoir or mercury conduit in order to obtain electrical characteristics with improved reproducibility. Patent Document 1 discloses that static electricity is eliminated.

JP-A-2015-99833

For example, in the resistivity measurement method by the CV method using mercury as an electrode, the same surface of the main surface of the semiconductor single crystal wafer is used for daily management of the resistivity measurement device or for re-measurement of the resistivity for product guarantee. If the position is measured several times in succession, the resistivity tends to decrease gradually or increase gradually.

The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a resistivity measurement method by the CV method that can more reliably measure the resistivity of a semiconductor single crystal wafer. .

To achieve the above object, the present invention provides a mercury electrode forming step in which mercury is brought into contact with a semiconductor single crystal wafer to form a mercury electrode, a forward bias voltage applying step of applying a forward bias voltage to the mercury electrode, A CV measurement step of measuring the capacitance of the semiconductor single crystal wafer that changes by applying a reverse bias voltage to the mercury electrode, and a resistivity calculation step of calculating a resistivity from the relationship between the reverse bias voltage and the capacitance. A resistivity measurement method is provided.
In the present invention, the forward bias refers to applying a voltage in a direction in which a current flows between a semiconductor single crystal wafer forming a Schottky junction and a mercury electrode.

As a result, the charge accumulated between the mercury electrode and the semiconductor single crystal wafer is canceled by the reverse charge supplied by the forward bias and disappears, so that the resistivity can be measured more accurately and accurately. It is possible to measure resistivity with good characteristics.

At this time, it is desirable that the applied amount of the forward bias voltage is increased as the resistivity of the semiconductor single crystal wafer is lower.

The lower the resistivity, the higher the impurity concentration and the more likely the electric charge is to accumulate between the semiconductor single crystal wafer and the mercury electrode. Therefore, the resistivity measurement with better reproducibility can be achieved by adjusting the forward bias voltage application amount as described above. Can do.

According to the resistivity measuring method of the present invention, it is possible to reduce the tendency of the resistivity that becomes apparent when the semiconductor single crystal wafer is continuously measured CV multiple times to gradually decrease or gradually increase. The resistivity of the semiconductor single crystal wafer can be measured more reliably.

It is a schematic process drawing which shows an example of the resistivity measuring method of this invention. It is the schematic which shows the mercury probe of an example of the CV method measuring apparatus used for the resistivity measuring method of this invention. It is a CV characteristic measurement system used for the resistivity measurement method of the present invention. This tendency appears when the resistivity of the P-type silicon epitaxial layer is repeatedly obtained by CV measurement by the conventional method. This tendency appears when the resistivity of the N-type silicon epitaxial layer is repeatedly obtained by CV measurement by the conventional method. When determining the resistivity of a P-type silicon epitaxial layer by repeatedly measuring CV, it is an example of resistivity variation when a forward bias voltage is not applied and when it is applied. (A) It is an illustration of the relationship between the resistivity of a P-type silicon epitaxial layer, and a forward bias voltage. (B) It is an illustration of the relationship between the resistivity of an N type silicon epitaxial layer, and a forward bias voltage. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic explanatory diagram when a forward bias is applied to a semiconductor single crystal wafer, where (1) is a state in which a forward bias is applied to a semiconductor single crystal wafer where charge remains, and (2) is the thickness of a depletion layer after forward bias is applied. (3) shows a state in which a depletion layer having a normal level thickness is formed.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic process diagram showing an example of the resistivity measurement method of the present invention. The mercury electrode formation process (process a), the forward bias voltage application process (process b), and the CV measurement process (process c). ) And the resistivity calculation step (step d) indicate that the present invention performs in this order.

FIG. 2 is a schematic view showing a mercury probe 10 as an example of a CV method measuring apparatus used in the resistivity measuring method of the present invention. As the semiconductor single crystal wafer 1, for example,

mercury

31 and 32 stored in the

storage containers

12 and 13 are guided upward from below to the main surface of a silicon epitaxial wafer in which a silicon epitaxial layer is formed on a silicon single crystal substrate. Contact with each other to form mercury electrodes 33 and 34, and CV measurement is performed to obtain the resistivity. The semiconductor single crystal wafer 1 other than silicon can be applied to the resistivity measurement of a compound semiconductor single crystal wafer such as GaP, GaN, SiC.

In the mercury prober 10 shown in FIG. 2, the semiconductor single crystal wafer 1 is held on the stage 11 with its main surface to be measured facing downward, and is placed above the mercury electrodes 33 and 34. When the resistivity is obtained by CV measurement,

mercury

31 and 32 stored in the

storage containers

12 and 13 rise through the

conduits

15 and 16 and come into contact with the main surface of the semiconductor single crystal wafer 1. To form mercury electrodes 33 and 34. A reverse bias voltage V is applied to the mercury electrode 34, while the mercury electrode 33 is connected to the ground 61.

When

mercury

31 and 32 used as electrodes are contaminated by particles adhering to the main surface of the semiconductor single crystal wafer 1 every time they come into contact with the semiconductor single crystal wafer 1, the resistance of the

mercury

31 and 32 gradually increases. Adversely affects the measured values of the physical characteristics. In order to prevent this, the

mercury

31, 32 that has contacted the semiconductor single crystal wafer 1 is temporarily returned to the

storage containers

12, 13, and the

entire mercury

31, 32 is bubbled and cleaned, and then the

mercury

31, 32 is used again for the electrodes. Supply.

When performing this bubbling, the

mercury

31 and 32 rub against the inner walls of the

storage containers

12 and 13 to generate static electricity. When the

storage containers

12 and 13 are made of, for example, polycarbonate, they are charged to the-(minus) electrode. On the other hand,

mercury

31 and 32 are charged to the + (plus) electrode.

Further, when the

mercury

31, 32 rises through the

conduits

15, 16, the

mercury

31, 32 rubs against the inner walls of the

conduits

15, 16 and generates static electricity. When the

conduits

15 and 16 are made of, for example, Teflon (registered trademark), they are charged to the-(minus) electrode. On the other hand,

mercury

31 and 32 are charged to the + (plus) electrode.

Therefore, the

storage containers

12 and 13 are connected to the

ground wires

52 and 54 via the

conductive rubber sheets

41 and 42 covering the

storage containers

12 and 13. The

conduits

15 and 16 are connected to the

ground wires

51 and 53 through

conductive rubber sheets

43 and 44 covering the

conduits

15 and 16. The

conductive rubber sheets

41, 42, 43, 44 are mixed with a conductive material such as carbon or silver and have a low resistivity of about 0.008 Ω · cm. Static electricity generated in the

conduits

15 and 16 can be easily absorbed and escaped to the

ground wires

51, 52, 53 and 54 connected to the

grounds

62 and 63.

A CV characteristic measurement system 100 for measuring resistivity shown in FIG. 3 includes a mercury prober 130 (mercury prober 10 in FIG. 2) that contacts the semiconductor single crystal wafer 1 using

mercury

31 and 32 as electrodes, and the mercury. A high frequency is supplied via

measurement cables

161 and 162 connected to the prober 130, a forward bias voltage is applied, and then a reverse bias voltage is applied to the semiconductor single crystal wafer 1 to form a depletion layer and the depletion layer. LCR meter 140 for measuring the capacitance of the device, and a PC (personal computer) 150 connected to the LCR meter 140 via a GPIB (General Purpose Interface Bus) cable 163, the resistance from the reverse bias voltage and the capacitance of the depletion layer Analysis software for calculating rates

When measuring the resistivity of the semiconductor single crystal wafer 1, a reverse bias voltage V is applied to the mercury electrode 34 formed with a Schottky junction with the semiconductor single crystal wafer 1 placed in the mercury prober 130 using the LCR meter 140. The capacitance C is measured by the LCR meter 140 while changing the capacitance C by expanding the depletion layer inside the semiconductor single crystal wafer 1 while changing the capacitance C continuously. Then, using the analysis software installed in the PC 150, the resistivity is calculated from the relationship between the reverse bias voltage V and the capacitance C of the depletion layer.

When the resistivity of a P-type silicon epitaxial layer formed on, for example, a silicon single crystal substrate is repeatedly measured as a semiconductor single crystal wafer 1 by a conventional method using a mercury probe 10 similar to that of FIG. 2 (CV at the same position). When the measurement is repeated to obtain the resistivity), the resistivity tends to gradually decrease as shown in FIG. Further, when the resistivity of the N-type silicon epitaxial layer is repeatedly measured, the resistivity tends to gradually increase as shown in FIG. This is presumably because charge is accumulated between the mercury electrode 34 of the mercury probe 10 and the semiconductor single crystal wafer 1 every time the resistivity measurement is repeated, that is, every time the reverse bias voltage V is applied.

Therefore, in the present invention, by applying a forward bias voltage, charges accumulated between the mercury electrode 34 and the semiconductor single crystal wafer 1 when the reverse bias voltage V is applied are removed. The timing for applying the forward bias voltage may be before or after the reverse bias voltage V is applied. When the mercury electrode is brought into contact with the semiconductor single crystal wafer and CV measurement is repeatedly performed at the same position, setting the timing for applying the forward bias voltage after applying the reverse bias voltage V applies the forward bias voltage. The timing is set before the application of the next reverse bias voltage V. Therefore, once the reverse bias voltage V is applied and the CV measurement process is performed, the charge accumulated between the mercury electrode and the wafer can be removed by the subsequent forward bias voltage application process. Then, since the next CV measurement is performed, it is possible to prevent the charge from accumulating and the resistivity from gradually increasing / decreasing as in the conventional method.

FIG. 6 illustrates resistivity fluctuations when the forward bias voltage is not applied (conventional method) and when it is applied (the present invention) when the resistivity of the P-type silicon epitaxial layer is repeatedly measured. In the example of FIG. 6, the forward bias voltage is applied from -1V to -18V while increasing the absolute value at 1V intervals, and then the reverse bias voltage V is applied while increasing from + 1V to + 15V at 1V intervals. Measurements were made. When the forward bias voltage is not applied, the resistivity gradually decreases, whereas when the forward bias voltage is applied, the resistivity can be made substantially constant. However, the forward bias voltage application method does not necessarily need to gradually increase the absolute value to the desired voltage. For example, a constant voltage may be applied as the forward bias.

The amount of forward bias voltage to be applied is preferably increased as the resistivity of the semiconductor single crystal wafer 1 is lower. As an example, FIG. 7A shows the relationship between the resistivity of the P-type silicon epitaxial layer and the forward bias voltage, and FIG. 7B shows the relationship between the resistivity of the N-type silicon epitaxial layer and the forward bias voltage. However, the applied amount of the forward bias voltage is repeated at the same position even if it is smaller than the value obtained from the relationship shown in FIGS. 7A and 7B or constant regardless of the resistivity. It is possible to suppress a tendency that the resistivity when the CV measurement is performed gradually increases (or decreases).

More specifically, when the semiconductor single crystal wafer 1 is P-type, a negative (−) voltage, N, is applied to the mercury electrode 34 formed by bringing the semiconductor single crystal wafer 1 and the mercury 32 into contact with each other to form a Schottky junction. In the case of a mold, a positive (+) voltage is applied as a forward bias (FIG. 8 (1)).

Then, the charge remaining between the mercury electrode 34 and the semiconductor single crystal wafer 1 is canceled by the reverse charge supplied by the forward bias and disappears. Therefore, the thickness of the depletion layer 35 formed in the semiconductor single crystal wafer 1 Returns to a normal level (FIG. 8 (2)). When a reverse bias voltage is applied to the mercury electrode 34 in this state, a depletion layer 35 having a normal level thickness is formed (FIG. 8 (3)), so repeated CV measurements were performed at the same position. In this case, the tendency to increase when the resistivity is N type and to decrease when the resistivity is P type is improved.

Since the semiconductor single crystal wafer 1 has a higher impurity concentration as the resistivity is lower, more charges are likely to remain on the mercury electrode 34 in contact with the semiconductor single crystal wafer 1. Therefore, it is desirable to increase the amount of forward bias voltage applied as the resistivity of the semiconductor single crystal wafer 1 is lower.

As described above, in the present invention, first, using a device as shown in FIGS. 2 and 3, mercury is brought into contact with a semiconductor wafer to form a mercury electrode (mercury electrode forming step) (step a). Then, after the forward bias voltage application step (step b) to the semiconductor single crystal wafer 1 is completed, the reverse bias voltage V is applied to the mercury electrode 34 while continuously changing using the LCR meter 140, and the semiconductor single crystal A CV measurement step (step c) is performed in which a depletion layer is expanded inside the wafer 1 to change the capacitance C, and the capacitance C is measured by the LCR meter 140. In the case of repeated measurement at the same position, the forward bias voltage application step (step b) may be performed after the CV measurement step (step c). In this case, as described above, the forward bias voltage application step (step b) is performed after the CV measurement step (step c) before the next CV measurement step (step c) at the same position. A forward bias voltage application step (step b) is performed.

Subsequently, using the analysis software installed in the PC 150, a resistivity calculation step (step d) is performed in which the resistivity is calculated from the relationship between the reverse bias voltage V and the depletion layer capacitance C. At this time, when the relationship between the reverse bias voltage V and the capacitance C is plotted on a graph, a CV characteristic is obtained. Further, by substituting the reverse bias voltage and the capacity into the following formulas (1) and (2), for example, the depth W in the semiconductor single crystal wafer 1 and the dopant concentration N (W) at the depth W are calculated. Therefore, a dopant concentration profile in the depth direction can be obtained.
W = Aε ₀ ε _Si / C (1)
N (W) = 2 / (qε ₀ ε _Si A ² ) × {d (C ⁻² ) / dV} ⁻¹ (2)
Here, A is the electrode area, ε ₀ is the vacuum dielectric constant, ε _Si is the relative dielectric constant of Si, and q is the charge amount of electrons.

Then, when the measurement depth is specified in the dopant concentration profile in the depth direction, the dopant concentration at that depth is obtained. In addition, the dopant concentration can be converted to resistivity by converting the obtained dopant concentration by a conversion formula such as ASTM STANDARDDS F723. According to the present invention as described above, in the conventional method, in particular, in the repeated CV measurement, it is possible to suppress the accumulation of electric charges between the semiconductor wafer and the mercury electrode and the resistivity from gradually increasing / decreasing, It is possible to determine the resistivity with high accuracy and high reproducibility.

Even when the mercury electrode is formed at the same position, the CV measurement, and the resistivity calculation are performed again after repeating the formation of the mercury electrode, the mercury electrode is formed and the C electrode is maintained at the same position while maintaining the mercury electrode. Even when the −V measurement is repeated, the charge accumulates, and the present invention is effective in both cases.
The repeated pattern will be described more specifically.
For example, (Step a) to (Step d) in FIG. 1 can be repeated at the same position. In other words, mercury is brought into contact with the main surface of the wafer to form a Schottky junction, forward bias voltage application, reverse bias voltage application, and resistivity calculation are performed in order, and then the mercury is separated from the wafer main surface for the next measurement. The above procedure can be repeated by bringing mercury into contact with the main surface of the wafer again at the same position (see arrow (1) in FIG. 1).
It is also possible to perform (step a) and repeat (step b) (step c) (step d) at the same position as it is. That is, forward bias voltage application, reverse bias voltage application, and resistivity calculation can be repeated with the Schottky junction still formed (see arrow (2) in FIG. 1).
Furthermore, it is also possible to perform (Step d) after performing (Step a) and repeating (Step b) and (Step c) at the same position. That is, the forward bias voltage application and the reverse bias voltage application are repeatedly performed with the Schottky junction formed, and finally the resistivity calculation can be performed collectively. (See arrow (3) in FIG. 1).
In any case, measurement can be performed while suppressing a gradual increase / decrease in resistivity due to charge accumulation.

Hereinafter, the present invention will be described more specifically with reference to examples and comparative examples of the present invention, but the present invention is not limited to these examples.

[Example 1]
The resistivity measurement method of the present invention was implemented. Specifically, (Step a) to (Step d) shown in FIG. 1 were repeated. When the P-type silicon epitaxial layer 1 having a resistivity of about 0.9 Ωcm is subjected to CV measurement five times using the mercury probe 10 shown in FIG. 2, the absolute value is increased at intervals of 1V from −1V to −18V each time. After applying the forward bias, the resistivity was measured by applying the reverse bias voltage V while increasing it from 1V to + 15V at 1V intervals. As a result, the CV value (coefficient of variation) obtained by dividing the standard deviation σ by the average value of the measured values was 0.03%.

[Comparative Example 1]
The P-type silicon epitaxial layer 1 having a resistivity of about 0.9 Ωcm used in the measurement in Example 1 was subjected to CV measurement five times without applying a forward bias voltage as in the conventional method to obtain the resistivity. . As a result, the CV value was 0.15%.

According to the resistivity measurement method of the present invention, by applying a forward bias voltage, the charge accumulated between the mercury electrode and the semiconductor single crystal wafer can be removed when a reverse bias voltage is applied. The thickness of the depletion layer formed in the single crystal wafer can be maintained at a normal level, and the tendency of the resistivity to gradually decrease or gradually increase can be suppressed. As a result, it is possible to improve the repeatability measurement accuracy of the resistivity.

Note that the present invention is not limited to the above embodiment. The above-described embodiment is an exemplification, and the present invention has any configuration that has substantially the same configuration as the technical idea described in the claims of the present invention and that exhibits the same effects. Are included in the technical scope.

Claims

A mercury electrode forming process in which mercury is brought into contact with a semiconductor single crystal wafer to form a mercury electrode;
Applying a forward bias voltage to the mercury electrode;
A CV measurement step of measuring a capacitance of the semiconductor single crystal wafer that changes by applying a reverse bias voltage to the mercury electrode;
A resistivity measurement method comprising: calculating a resistivity from a relationship between the reverse bias voltage and the capacitance.
The resistivity measurement method according to claim 1, wherein the amount of forward bias voltage applied is increased as the resistivity of the semiconductor single crystal wafer is lower.