JP2000114198A - Surface treatment method and equipment thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the generation of irregularities in the electrical resistance values of a source/drain region and a gate electrode, by introducing elements, molecules and compounds in a substrate through plasma doping which is capable of controlling does quantity. SOLUTION: A semiconductor substrate is held on a substrate holding table 6 in a chamber 5, and source gas 7 is introduced. The source gas 7 is mixed at an arbitrary ratio with a mixing means and introduced in the chamber 5 by an arbitrary flow rate with a flow rate control means such as a valve. After that, B2H6 in the source gas 7 is made into a plasma with an ECR plasma source 8. The state of plasma in the chamber 5 is observed with a plasma- observation apparatus 10, and the luminous intensity is measured. Dose quantity is understood from the luminous intensity. The value of dose quantity understood from the plasma observation is compared with the value of desired dose quantity, and the state of plasma is properly controlled on the basis of this difference. As a result, desired dose quantity can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a surface treatment method for introducing a substance (atom, molecule, compound, alloy, etc.) into a substrate using plasma doping.

[0002]

2. Description of the Related Art When a semiconductor device is manufactured, a small amount of impurities such as phosphorus and boron are introduced into a semiconductor to form an n-type semiconductor.
A step of forming a p-type semiconductor is required.

[0003] As a method of introducing impurities into a semiconductor, an ion implantation method is widely used. However, as the size of the semiconductor becomes smaller, the junction depth becomes shallower, and therefore, it is necessary to reduce the energy in the ion implantation process. Have a point.

Therefore, various novel impurity introduction methods have been proposed in place of the ion implantation method. Above all, it is a room temperature process, is compatible with the conventional ion implantation method, can maintain a high throughput even in a low energy region, and is more inexpensive than an ion implantation device and occupies a small area of the device. Therefore, research on an impurity introduction method called plasma doping has been actively conducted.

[0005] In the case of using a method called plasma doping, before starting mass production, impurities are introduced on a trial basis, and a dose is obtained by using a method called SIMS (secondary ion mass spectrometry). It is indispensable to measure the sheet resistance after the heat treatment step to confirm the increase or decrease of the dose. Then, after adjusting the doping time based on the result, mass production can finally be started.

[0006]

However, in the above-mentioned prior art, no measures are taken against the fact that the dose varies due to the change in the plasma state immediately before or during the plasma doping process. Therefore, the dose cannot be controlled accurately. Therefore, in the semiconductor device manufactured by the conventional method, a defect that the device driving capability becomes non-uniform due to variation in the electric resistance value of the source / drain region and the gate electrode occurs, and the yield also decreases. .

It is possible to improve the yield to some extent by performing trial introduction of impurities in advance, analyzing the result, and adjusting the doping time and the like based on the analysis result. As a result, time is required, resulting in a decrease in productivity in the manufacture of semiconductor devices.

[0008]

A means for solving the above problems is to introduce elements, molecules and compounds into a substrate by plasma doping capable of controlling a dose. For example, if this means is applied to a method for manufacturing a semiconductor device, impurities will be introduced into the semiconductor substrate by plasma doping capable of controlling the dose.

Specifically, an observation step of observing the state of the plasma and a control step of controlling the state of the plasma based on the observation result are performed. Since the amount of the dose depends on the state of the plasma, by observing and controlling the state of the plasma, the dose can be appropriately controlled as a result.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The surface treatment method proposed by the present inventor can be used when introducing elements, molecules and compounds into a substrate, and can be used not only in the manufacture of semiconductor devices but also in various fields. Can be applied to This is because, by introducing an appropriate element or the like into the base, specific properties can be given to the base or specific properties can be improved. Specific properties include, for example, mechanical properties (abrasion resistance, lubricity, mold release properties, corrosion resistance, etc.), electric and magnetic properties (electric conductivity, electromagnetic wave shielding properties, magnetic properties, etc.), and optical properties ( Light absorption, light reflection, glossiness, coloring, etc.) and thermal properties (heat resistance, thermal conductivity, etc.). Specifically, in order to reduce the friction coefficient of the bearing, a material that reduces the friction coefficient is introduced into the surface of the bearing member.

Here, as an example of an embodiment of the present invention, a case of manufacturing a MOS transistor will be described with reference to FIGS. FIG. 1 is a view for explaining a method of manufacturing the same MOS transistor, and FIG.
It is the schematic of the manufacturing apparatus of S transistor.

First, as shown in FIG. 1A, an N well region 2 and a gate oxide film 3 (such as a thermally grown silicon oxide film) are formed on a P-type silicon substrate 1.
Then, a gate electrode 4 is patterned on the gate oxide film 3. Here, the thickness of the gate electrode 4 is 200 nm,
Gate oxide film 3 has a thickness of 3 nm. The gate length is 150 nm.

Next, after the semiconductor substrate of FIG. 1A is held on a substrate holding table 6 provided as a holding means provided inside the chamber 5, a flow rate of 200
A source gas 7 of sccm is introduced.

The source gas 7 contains B ₂ H _{6 as} an impurity. That is, outside the chamber 5, B
A container enclosing ₂ H ₆ and a container enclosing He for dilution are prepared. These are mixed at an arbitrary ratio by a mixing means such as a valve, and then mixed by an arbitrary flow rate by a flow control means such as a valve. It is introduced into the chamber 5. Although the impurities are introduced into the chamber 5 in a gaseous state here, they may be vaporized in the chamber 5 after being introduced in a liquid state.

Thereafter, B ₂ H ₆ in the source gas 7 introduced into the chamber 5 is converted into a 500 W ECR plasma source 8.
Is converted into plasma. At this time, the degree of vacuum in the chamber 5 is 4 × 10 ⁻⁴ Torr. Here, an ECR plasma source is used as the plasma source, but an ICP type or a parallel plate type may be used.

Subsequently, the state of the plasma in the chamber 5 is observed by the plasma measuring device 10 as the plasma observation means. Specifically, emission analysis of ions or radicals of impurities in the plasma region formed in the chamber 5 is performed.
The emission intensity at a wavelength of 4332 ° corresponding to the transition process of (A ¹ {−X ¹ }) is measured by the plasma measuring device 10 (emission analysis).

The state of the plasma is suitably controlled by control means (not shown) based on the emission analysis results.

That is, since there is a correlation between the light emission intensity and the sheet resistance, and there is also a correlation between the sheet resistance and the dose, by observing the plasma and measuring the light emission intensity, The dose amount can be grasped. Then, by comparing the value of the dose amount obtained from the plasma observation result with the value of the desired dose amount, and appropriately controlling the state of the plasma based on the difference, a desired dose amount can be obtained. Methods for controlling the state of the plasma include ECR power control, RF power control, magnetic field control,
There are pressure control in the chamber, source gas flow rate control, source gas concentration control, and the like, and one or more techniques can be appropriately employed. In this embodiment, an RF power supply 11 is provided, and an RF power of a value (for example, 300 W) suitable for a semiconductor substrate is provided.
By applying power, the state of the plasma is controlled. That is, this RF power causes 70
A self-bias of 0 V is generated, and boron ions are
It is introduced into the semiconductor substrate at an acceleration energy of eV. FIG. 1B shows a state in which boron 9 has been introduced into the semiconductor substrate of FIG.

Although attention has been paid here to the relationship between the light emission intensity and the dose, the relationship between the light emission intensity, the bias, and the dose may be noted.

Conventionally, the plasma potential is relatively small (several eV to several hundred e) as compared with the doping energy.
V) Therefore, the influence of plasma doping on ion energy could be neglected. However, in recent years, with the miniaturization of semiconductor devices, lowering of doping energy (several hundred eV to several thousand eV) has been advanced. Therefore, in the plasma doping in the low energy region, the influence of the plasma potential on the ion energy to be doped becomes relatively large, so that the plasma potential cannot be ignored.
At this time, the energy of the ion to be doped is the sum of the plasma potential and the forcible bias (or acceleration voltage) from an external power supply.

Further, the control of the plasma can be performed before the impurity introduction step, and can be performed in parallel with the impurity introduction step. By appropriately controlling the plasma state so as to obtain a desired dose before introducing the impurity, the trial introduction of the impurity which is conventionally required becomes unnecessary, thereby shortening the manufacturing time of the semiconductor device and improving the yield. Can be realized. When the control is performed while observing the plasma state in parallel with the impurity introduction, even if impurities are introduced into a plurality of semiconductor substrates in the same chamber, there is almost no difference in dose amount between the substrates. As a result, it is possible to reduce the manufacturing time of the semiconductor device and improve the yield.

Here, in order to explain the reason why the dose can be controlled by observing and controlling the plasma, the results of experiments conducted by the inventor are shown in FIGS.
This will be described with reference to FIG. The experimental conditions are as follows. Further, the sample was prepared based on the above process.

<Experimental Conditions> Semiconductor substrate: 6-inch N-type silicon substrate Doping device: plasma doping device (manufactured by Matsushita Electric Industrial Co., Ltd.) Doping condition: doping time 100 sec RF power 100 to 300 W ECR power 500 W Source gas: B ₂ H ₆ 200 sccm Chamber vacuum: 1 × 10 ⁻⁴ Torr to 2 × 10 ⁻³ Torr Activation heat treatment: RTA 1000 ° C., 10 sec Sheet resistance measurement: 4-point needle method SIMS measurement: primary ion species O ²⁺ , secondary ion species Positive Primary ion energy 3 keV Emission analysis: Measure the emission intensity at a wavelength of 4332 ° corresponding to the transition process of (A ¹ Π−X ¹ Σ) of the BH radical. The results of the experiment conducted under the above conditions were focused on RF power. The arrangement is as follows (see FIGS. 3 to 5).

FIG. 3 shows the relationship between the RF power applied to the semiconductor substrate, the emission intensity of BH radicals as a result of emission analysis in the plasma region, and the sheet resistance of the substrate after the activation heat treatment. FIG. 3 shows that when the RF power is increased from 100 W to 300 W, the light emission intensity increases and the sheet resistance decreases. Since a decrease in the sheet resistance means an increase in the dose, FIG. 3 shows that the emission intensity increases and the B dose increases with an increase in the RF power. According to the SIMS measurement results when only the RF power was varied (see FIG. 4), when the RF power was 100 W and 300 W, the B dose amount was 4 ×
It is 10 ¹⁵ cm ^-2 and 7 × 10 ¹⁵ cm ^-2 , confirming that the above-mentioned relationship exists among the RF power, the emission intensity, and the B dose.

FIG. 5 shows RF applied to a semiconductor substrate.
4 shows the relationship between the power, the bias between the semiconductor substrate and the plasma, and the sheet resistance of the substrate after the activation heat treatment. FIG. 5 shows that when the RF power is increased from 100 W to 300 W, the bias increases and the sheet resistance decreases, that is, the B dose increases.

Therefore, according to FIGS. 3 to 5, the magnitude of the dose can be determined based on the magnitude of the luminous intensity or the magnitude of the bias measurement result, and the luminous intensity, bias and dose can be determined by controlling the RF power. Can be controlled. That is, at least one of the plasma emission intensity and the bias measurement result is observed,
By controlling the RF power according to the degree, the dose can be set appropriately.

Next, the results of the above-described experiments are arranged as follows, focusing on the degree of vacuum in the chamber (see FIGS. 6 to 9).

FIG. 6 shows the relationship between the degree of vacuum in the chamber, the emission intensity of BH radicals as a result of emission analysis, and the sheet resistance of the substrate after the activation heat treatment. As shown in FIG. 6, the chamber vacuum was changed from 1.5 × 10 ⁻⁴ Torr to 4
It can be seen that the emission intensity increases and the sheet resistance decreases when increasing to × 10 ⁻⁴ Torr. Further, the degree of vacuum of the chamber is increased from 4 × 10 ⁻⁴ Torr to 8 × 10 ⁻⁴ T.
It can be seen that when the intensity is increased to orr, the light emission intensity increases and the sheet resistance increases. In addition, the result of the SIMS measurement when only the chamber vacuum was changed (FIG. 7,
According to FIG. 8), the degree of vacuum in the chamber is 1.5 ×
10 ^-4 Torr, 4 × 10 ^-4 Torr, 8 × 10 ^-4 Tor
In the case of rr, the B dose amount is 3 × 10 ¹⁵ cm ⁻² , 7 ×
It is 10 ¹⁵ cm ⁻² and 5 × 10 ¹⁵ cm ⁻² , confirming that the above-mentioned relationship exists among the three factors of the chamber vacuum degree, the light emission intensity, and the B dose amount.

FIG. 9 shows the relationship between the degree of vacuum in the chamber and the bias and sheet resistance between the semiconductor substrate and the plasma. FIG. 9 shows that the sheet resistance decreases, that is, the B dose increases in a section where the bias is almost constant (1 × 10 ⁻⁴ Torr to ⁴ × 10 ⁻⁴ Torr).

Therefore, according to FIGS. 6 to 9, while the bias is almost constant, the magnitude of the dose can be determined based on the magnitude of the luminous intensity, and the luminous intensity and the dose can be determined by controlling the degree of vacuum of the chamber. It can be seen that the amount can be controlled. Further, it can be seen that when the bias starts to decrease rapidly by lowering the degree of vacuum in the chamber, the emission intensity increases as the degree of vacuum decreases, and the B dose decreases. That is, by observing the measurement results of the plasma emission intensity and the bias and controlling the degree of vacuum in the chamber according to the degree, the dose can be set appropriately.

Next, the results of an experiment performed under the above conditions are as follows:
The following is a summary when focusing on the source concentration (see FIGS. 10 to 12).

FIG. 10 shows the relationship between the source (B ₂ H ₆ ) concentration, the emission intensity of BH radicals as a result of emission analysis in the plasma region, and the sheet resistance of the substrate after the activation heat treatment. From FIG. 10, when the source concentration is increased,
It can be seen that the emission intensity increases and the sheet resistance decreases. Since a decrease in the sheet resistance means an increase in the dose, FIG. 10 shows that the emission intensity increases and the B dose increases with an increase in the source concentration. In addition,
According to the SIMS measurement results when only the source concentration was varied (see FIG. 11), the source concentration was 5%, 0.5%
%, 0.05%, the B dose amount is 4 × 10 ¹⁵ cm each
⁻² , 2 × 10 ¹⁵ cm ⁻² , and 1 × 10 ¹⁵ cm ⁻² , confirming that the above-mentioned relationship exists between the three of the source concentration, the emission intensity and the B dose.

FIG. 12 shows the relationship between the source concentration, the bias between the semiconductor substrate and the plasma, and the emission intensity. 10 and 12, it can be seen that when the source concentration is increased, the bias is substantially unchanged, and during this time, the sheet resistance decreases, that is, the B dose increases.

Therefore, according to FIGS. 10 to 12,
It can be seen that the magnitude of the dose can be determined based on the magnitude of the light emission intensity, and that the light emission intensity and the dose can be controlled by controlling the source concentration. That is, the emission intensity of the plasma is observed, and according to the degree,
By controlling the source concentration, the dose can be suitably set.

From the above, the state of the plasma was observed, and the RF power, the degree of chamber vacuum,
It is clear that the dose can be suitably controlled and set by controlling the plasma concentration by controlling the source concentration and the like. Here, the plasma state (ie, dose amount) was controlled by individually controlling the RF power, the chamber vacuum degree, and the source concentration. However, various parameters (RF power, chamber vacuum degree) capable of controlling the plasma state were used. ,
Needless to say, magnetic field strength, ECR power, source gas flow rate, source gas concentration, etc.) can be used, and the effects of the present invention can be obtained by appropriately combining them.

[0036]

As described above, according to the surface treatment method of the present invention, the dose can be accurately set before or in parallel with the substance introduction step. . As a result, it is not necessary to take a sample and check the dose before the substance introduction step, so that the manufacturing time can be reduced and the yield can be improved. In addition, even when the surface treatment method of the present invention is applied to a method of manufacturing a semiconductor device, a reduction in manufacturing time and an improvement in yield can be realized, and the completed semiconductor device has a source / drain region and a gate electrode. Becomes a predetermined value, and the device driving capability becomes uniform.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a method for manufacturing a MOS transistor according to an embodiment of the present invention;

FIG. 2 is a schematic view of the manufacturing apparatus.

FIG. 3 is a diagram showing a relationship among RF power, emission intensity of BH radical, and sheet resistance.

FIG. 4 is a diagram showing SIMS measurement results when only RF power is changed.

FIG. 5 is a diagram showing the relationship between RF power, bias, and sheet resistance.

FIG. 6 shows the degree of chamber vacuum, the emission intensity of BH radicals,
Diagram showing the relationship between sheet resistance

FIG. 7: SI when only the degree of chamber vacuum is changed
First figure showing MS measurement results

FIG. 8: SI when only the degree of vacuum in the chamber is changed
Second diagram showing the results of MS measurement

FIG. 9 is a diagram showing the relationship between chamber vacuum, bias, and sheet resistance.

FIG. 10 is a diagram showing a relationship among a source concentration, emission intensity of BH radical, and sheet resistance.

FIG. 11: SIMS when only source concentration is changed
Diagram showing measurement results

FIG. 12 is a diagram showing the relationship between source concentration, bias, and emission intensity.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 P-type silicon substrate 2 N well region 3 Gate oxide film 4 Gate electrode 5 Chamber 6 Substrate holder 7 Source gas 8 ECR plasma source 9 Boron 10 Plasma measuring instrument 11 RF power supply

Claims (14)

[Claims] 1. A surface treatment method for introducing an arbitrary substance into a substrate using plasma doping, wherein the dose can be controlled to an arbitrary value. 2. The method according to claim 1, further comprising a plasma-forming step of converting a substance into a plasma, an observation step of observing a state of plasma, and a control step of controlling a dose based on the observation result. Surface treatment method. 3. The surface treatment method according to claim 1, wherein the dose is controlled before the step of introducing the substance into the substrate. 4. The surface treatment method according to claim 1, wherein the dose is controlled in parallel with the step of introducing the substance into the substrate. 5. The observation step includes at least one of an emission analysis step of performing emission analysis of arbitrary ions or radicals in a region where the substance is turned into plasma, and a bias measurement step of measuring a bias between the substrate and the plasma. The surface treatment method according to any one of claims 2 to 4, comprising: 6. The surface treatment method according to claim 5, wherein in the emission analysis step, the intensity of radical emission is used as an index. 7. The surface treatment method according to claim 5, wherein a self-bias is used as an index in the bias measuring step. 8. The surface treatment method according to claim 5, wherein RF power is used as means for generating a bias on the substrate, and the dose is controlled by controlling the RF power. . 9. The surface treatment method according to claim 5, wherein the dose is controlled by controlling the degree of vacuum in a chamber where the plasma process is performed. 10. The surface treatment method according to claim 5, wherein the dose is controlled by controlling the concentration of the substance. 11. A source supply unit for introducing a substance into a chamber, a plasma source for converting the substance into plasma in the chamber, a holding unit for holding a substrate in the chamber, and a state of plasma in the chamber. A surface treatment apparatus, comprising: an observation unit that observes the temperature; and a control unit that controls a dose amount based on a result of the observation by the observation unit. 12. The observation means is at least one of an emission analysis means for performing emission analysis of arbitrary ions or radicals in a region where the substance is turned into plasma, and a bias measurement means for measuring a bias between the substrate and the plasma. The surface treatment apparatus according to claim 11, comprising: 13. A method for manufacturing a semiconductor device, comprising using the surface treatment method according to claim 1 to introduce an impurity into a semiconductor substrate as a base. 14. A source supply unit for introducing impurities into a chamber, a plasma source for converting the impurities into plasma in the chamber, a holding unit for holding a semiconductor substrate in the chamber, and a plasma source in the chamber. An apparatus for manufacturing a semiconductor device, comprising: observation means for observing a state; and control means for controlling a dose based on an observation result by the observation means.