KR20010027001A - Chemical used in wet etching process - Google Patents

Abstract

PURPOSE: Chemicals for wet etch are to provide good repeatability of an etch process by preventing variation of a density caused by volatility of the chemicals, and to improve productivity of the etch process by increasing an etch rate while having good selectivity. CONSTITUTION: Etching liquid is used in a process for wet-etching a silicon nitride layer in a closed chamber by using an oxide layer as a mask, wherein the etching liquid is composed of 85 H3PO4, 0.01-0.1 HF and 1 hydrogen peroxide. In the etching liquid, an etch rate of the silicon nitride layer is from 60 to 74 angstrom/minute and an etch selectivity of a nitride layer and an oxide layer is not lower than 20, at a temperature of 120 deg.C.

Description

Chemical used in wet etching process

The present invention relates to a wet etching chemical used in the wet etching process of the silicon nitride film, and in particular to a wet etching chemical composition that is composed of a new composition and composition ratio to enable a low temperature process and to shorten the process time will be.

In recent highly integrated semiconductor devices, silicon nitride (Si ₃ N ₄ ) is used as a protective film on the surface of a semiconductor or as an oxide or ion implantation mask during a fabrication (FAB) process.

In the etching of the silicon nitride film, a wet etching method is usually performed in which an etching solution composed of 85 phosphoric acid (H ₃ PO ₄ ) and 15 water is heated at a temperature of 163 to 175 ° C.

However, since the photoresist film used as a mask in another etching process cannot withstand the process temperature of the etching solution, a pattern of silicon oxide (SiO ₂ ) is formed as a mask, and then the wet etching process using the phosphoric acid solution described above is used. A method of obtaining a predetermined pattern of a nitride film by removing silicon nitride is used.

The etching rate of the silicon nitride film is greatly influenced by the composition of the etching solution. In particular, the moisture content of the etching solution greatly influences the etching rate of the nitride film and the oxide film.

As the concentration of phosphoric acid increases as the moisture contained in the etchant evaporates, the boiling point temperature rises to the range of 160 to 170 ° C or higher, thereby lowering the etch rate of the silicon nitride film, but in this condition, the silicon oxide film is etched faster. do.

Therefore, in the conventional etching process, pure water is added to an etching bath forming an etching chamber every predetermined time so as to maintain a constant concentration at a predetermined temperature condition.

However, in the conventional wet etching process using 85 phosphoric acid and 15 water etchant, as described above, the process temperature of the etchant is 160-170 ° C., which is much higher than the water evaporation temperature, and thus the concentration of the chemical changes frequently due to water evaporation. Therefore, it is necessary to replenish water frequently, and it is difficult to maintain a constant concentration of the etchant even when replenishing water. In addition, the time required to etch the nitride film having a thickness of 1500 mW also takes about 40 to 70 minutes.

The effect of this drawback is more pronounced in chamber arrangements in which the bath structure is open. Since the open chamber discharges the vaporized water to the outside, it is necessary to replenish water more frequently. Accordingly, the concentration and temperature characteristics of the etchant over time are shown as a severely fluctuating curve.

In addition, the above-described high temperature etching liquid condition generates water vapor even in the hermetically sealed chamber structure, so that a small change in concentration appears due to the circulating water vapor, and the high temperature temperature condition causes thermal damage to the etching chamber to shorten the lifespan.

The present invention devised to solve the above-mentioned problems is an object of the present invention is to provide a wet etching chemical that can be directed to an excellent etching selectivity and etching rate, while being a low temperature process, and having good reproducibility of the process.

1 is a cross-sectional view showing the structure of the experimental wet etching chamber used in the chemical experiment of the present invention,

Figure 2 is a graph showing the degree of silicon removal compared to the concentration of hydrochloric acid confirmed in the experiment of the embodiment of the present invention,

3 is a graph showing the degree of removal of the nitride film compared to the concentration of sulfuric acid confirmed in the experiment of the embodiment of the present invention,

Figure 4a is a graph showing the etching state of the nitride film using the etching solution of 0.01HF concentration according to an embodiment of the present invention,

Figure 4b is a graph showing the etching state of the oxide film using the etching solution of 0.01HF concentration according to an embodiment of the present invention,

Figure 5a is a graph showing the etching state of the nitride film using an etching solution of 0.1HF concentration according to an embodiment of the present invention,

Figure 5b is a graph showing the etching state of the oxide film using the etching solution of 0.1HF concentration according to an embodiment of the present invention.

The wet etching chemical of the present invention for achieving the above object is an etching solution used in the wet etching process of the silicon nitride film using the oxide film as a mask in a closed chamber, wherein the etching solution is phosphoric acid (H ₃ PO ₄ ) 85 , HF is 0.01 to 0.1, and hydrogen peroxide is characterized in that about one.

The etching solution has an etching rate of 60 ns / min to 74 ns / min at a temperature of 120 ° C., and an etching selectivity of the nitride film / oxide is 20 or more.

The wet etching using the etching liquid of this configuration is preferably performed using a closed chamber to minimize the consumption of HF, and the analysis of the etching conditions should include a sensor capable of accurately monitoring the concentration of the etching liquid.

As described above, the present invention enables a low temperature process to significantly reduce the amount of steam generated, thereby keeping the concentration of the chemical constant and reducing the thermal change of the etching chamber, thereby extending the life of the equipment.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, in the experiment of the present invention, as shown in FIG. 1, an experimental chamber of a closed structure having a heater 14, a gas bubbler 16, and a condenser 18 for heating an etchant in the bath 10. The experiment was conducted to demonstrate the present invention by manufacturing.

For reference, an experimental chamber of such a sealed structure has already been disclosed by the applicant of the present applicant.

The present invention is intended to create a wet etching chemical that can be useful in the control of the actual process by accurately grasping various etching mechanisms, such as physical properties on the surface of the nitride film / oxide film by the etching solution using such a laboratory chamber device. .

The present inventors carried out an experiment to confirm the etching reaction of the wafer by injecting chemicals of various compositions and composition ratios into the chamber apparatus described above until the wet etching chemical provided by the present invention is completed.

The analysis of the experimental results is confirmed by measuring the change in the concentration of silicon present in the solution by the ICP-AES by collecting the etching solution in the bath (10) by the time period, and also by collecting the sediment attached to the lower end of the condenser (18) energy It carried out by the method measured by a distributed X-ray spectroscope.

<Base and Background of Embodiment of the Present Invention>

First, reference examples identified in the process up to the present invention will be described.

The present inventors experimented with the possibility of hydrochloric acid as a condition for controlling the etching reaction.

In general, silicon is highly likely to become silicon chloride which reacts with hydrochloric acid during etching to volatilize, and the reaction takes place as follows.

H ₂ SiO ₃ + ₄ HCl → SiCl ₄ + 3H ₂ O

Silicon chloride (SiCl ₄ ) thus formed is known to have a high volatility and a boiling point around 57 ° C. If the process proceeds with such a reaction, it is inferred that the silicon present in HCl can be efficiently removed.However, based on these theories, there are many other variables in the actual process to determine under what conditions the reaction proceeds efficiently. It is not suitable because they work.

Therefore, an experiment was conducted to confirm the etching reaction occurring in the closed chamber by adding hydrochloric acid to phosphoric acid (H ₃ PO ₄ ).

Figure 2 is the result of measuring the concentration change of silicon before and after heating by heating the solution in which HCl 2 is added to phosphoric acid to 125 ℃ and added to it by ICP-AES.

Herein, since the concentration of silicon is 10 times diluted, it is equivalent to 100 ppm in the actual etching container when it is represented by 10 ppm.

Referring to the experimental results shown in the figure, the signal of silicon appearing at low temperature condition before heating was higher in the total concentration range than the signal of silicon appearing at high temperature condition after heating, indicating that the silicon flew to silicon collide. It can be inferred.

The etching of the nitride / oxide wafer with varying the concentration of hydrochloric acid in phosphoric acid from 1 to 10 showed very random etching with little effect on the change of etching rate.

On the other hand, in the same way as the experiment to add the hydrochloric acid described above, when the sulfuric acid was added to the experiment was again tested, the etching rate for 20 minutes at 155 ℃ while changing the sulfuric acid from sulfuric acid from 0 to 20, Figure 3 When sulfuric acid was added as shown, the etching rate was reduced and was confirmed to exhibit an irregular etching rate.

Therefore, it was confirmed that it is not suitable as a wet etching chemical for achieving the object of the present invention.

This time, experiments were performed in which corrosive HF was added to an appropriate amount of phosphoric acid and used as an etching solution.

In general, HF etches the oxide layer more quickly than the nitride and oxide layers, so if the amount of HF is added more than the silicon oxide layer to be etched and melted, the excess HF is more likely to etch the oxide layer. It gets worse.

Accordingly, the present inventor tried to add H ₂ O ₂ into the etchant so that the etching by-products of the nitride film did not become an incomplete oxide film but became a complete oxide film and easily reacted with HF while controlling the amount of HF. This is based on the existing theory that the etch by-products of nitride films are not silicon ions but oxides similar to silicon oxide films.

Prior to the detailed description of the experiment will be described the technical background related to the experiment.

In general, the reaction of HF with the silicon oxide film is known to proceed by the following mechanism.

SiO ₂ + ₆ HF → H ₂ SiF ₆ (aq) + 2H ₂ O

However, it is known that HF has the properties of weak acids except at very low concentrations and does not completely separate H ⁺ and F ⁻ as shown in the following scheme.

HF → H ^{+ +} F ^-

HF + F ^- → HF ^2-

At this time, the generated HF ^2- dissolves silicon dioxide in an aqueous solution by the following reaction.

SiO ₂ + 3 HF ^2- → SiF ₆ ^2- + 2H ₂ O

For reference, other papers that have been reported may predict that the reaction will be as follows.

SiO₂+ 2HF^2-+ 2H₃O⁺→ SiF₄ + 4H₂O

Since SiF _{4, which} is such a reaction product, is so volatile that it exists as a gas even at a temperature of -85 ° C. or lower, it is natural to fly without remaining in the bath 10 as soon as it is produced.

When etching the silicon oxide film and a silicon nitride film to date with HF, the HF solution of 10 Mole or less, whereas the etching rate increases in proportion to the concentration of HF with HF ^2-, F ^- It is known that there are independent changes in the concentration of .

If, based on these results, in the case of the aqueous solution is within a number of HF was present in a SiF ₄ SiF ₆ ^2-, rather than the watching that the temperature of the solution is flying high in SiF ₄ can be determined as a general view of the art Of course, it is natural that the reaction proceeds to the reaction of SiF ₄ at a high temperature during the etching reaction mentioned in the present invention.

When the concentration of HF is very high, it is known that HF exists in the form of H ₂ F ⁺ to give the following reaction.

^{_{SiO 2 + 2H 2 F 2- +}} 2F - → SiF 4 + 2H 2 O

Meanwhile, BHF, which is commonly used in semiconductor processing, has the following reaction.

_{SiO 2 + 4HF + 2NH 4 F} → (NH 4) 2SiF 6 + 2H 2 O

Such a reaction suggests that when the silicon nitride film is etched, salts of anions and ammonium cations of the silicon oxide film are formed.

In addition, there are some documents that introduce a method of etching using a mixed acid containing HF.

In a paper published by ZIGER et al., A study was conducted on the use of an etchant containing HF and HNO ₃ in phosphoric acid. At this time, HNO ₃ is not known to affect the silicon oxide film and the silicon nitride film at all by simply increasing the etch rate of silicon 10 times.

Si + 4HNO ₃ → 3Si (OH) ₄ + 4NO

However, Si (OH) ₄ in the above reaction is expected to be difficult to produce in an acidic environment.

In addition, in the published papers, the concentration of nitric acid used was 3-6, and the concentration of HF was 1-5. In the experiments of the present invention, the same concentration conditions were used, and the above-described closed chamber was As a result of the experiment, reproducibility of the test results could not be obtained.

The reason for this experimental result is that both nitric acid and HF added to the mixed solution are vaporized at a high process temperature, so that it is difficult to exist as a solution. It is assumed that this is because the result is performed at.

In the experimental chamber of the closed structure used to demonstrate the present invention there is no phenomenon that HF and nitric acid are reduced by evaporation during heating, so that their concentration can be kept constant during the experiment.

The experiment was carried out using the above-described hermetically sealed chamber, and the etching solution was etched at 120 ° C. under a condition of 85 HF and 85 phosphoric acid. All oxide films with a thickness of about 1000 mW were completely removed.

As a result, using an etchant consisting of more than 3 HF and 85 phosphate is impossible to precisely control the etching thickness and it is obvious that the etching results are very irregular, so that reproducibility cannot be secured.

In the embodiment of the present invention, based on the various theories and experimental results as described above, experiments using mixed acid as an etchant, but the mixed acid through the experiments of the following examples are phosphoric acid, HF, and hydrogen peroxide (H ₂ O ₂ ), and the composition ratios of 85, 0.01 to 0.1, and 1, respectively, were found to be suitable for the step of wet etching.

First, as a mixed acid to be used as an etchant, phosphoric acid is 85, hydrogen peroxide is 1, and HF is 0.01, and the nitride film and the oxide film were etched using the experimental chamber of the above-described closed structure at a temperature of 120 ° C. The etch rate and the etching thickness of are as shown in the graph shown in Figure 4a and 4b.

In addition, the ratio of HF in the mixed acid was set to 0.1 again, and etching was performed under the same conditions as the above-described environmental conditions at the concentration of 0.01HF. As a result, the etch rate and the etching thickness of each of the nitride film and the oxide film are shown in FIGS. 5A and 5B. As shown in the graph shown.

The etch rate results at the 0.01HF concentration, 0.1HF concentration, and the etching selectivity ratio of the nitride film to the oxide film are shown in Tables 1 and 2 below.

TABLE 1

Classification Average etch rate Etching cost Si ₃ N ₄ 74 23 SiO ₂ 3.3 23

TABLE 2

Classification Average etch rate Etching cost Si ₃ N ₄ 57 33 SiO ₂ 1.7 33

As can be seen from the graphs of FIGS. 4A to 5B and the experimental results of Tables 1 and 2, as the time was changed from 5 minutes to 15 minutes, the etching rate was constant regardless of the HF concentration change in the nitride film. On the other hand, the etching rate of the oxide film is constant when the HF concentration is 0.01, but it was confirmed that the etching rate tends to increase as time passes.

Therefore, at 0.1HF, the etching selectivity of the nitride film to the oxide film is expected to deteriorate with time. As shown in the above two tables, the etching selectivity was about 23 at 0.1, but it was relatively safe for etching at 0.01HF.

The etching rate representing the etching rate is 74 kW / min at 0.1 HF for the nitride film, while 57 kW / min at 0.01 HF, but the etching rate is decreased, but less than that of the oxide film. However, it is expected that the oxide film has a very small etching rate and thus a large experimental error.

These selectivity and etch rate are experimental conditions at the beginning of the experiment, there may be a change in the concentration of reagents and other conditions.

Therefore, in order to confirm the etching results obtained as described above, the results of etching 10 minutes and 15 minutes at 120 ° C. while varying the concentration of HF to 0.01, 0.03, 0.05, 0.08, and 0.10 in a mixture of hydrogen peroxide and phosphoric acid are shown. Shown graphically in 6a and 6b.

Referring to FIG. 1, it can be seen that when etching for 10 minutes, the etching rate increases proportionally as the concentration of HF increases except for 0.03 HF. In the case of 0.03, it is estimated as an error occurrence when measuring the etching thickness.

In the case of etching for 15 minutes, it can be seen that the concentration continuously increases as the concentration of HF increases. The etch rate obtained at this time increased from 40 mW / min to about 70 mW / min as the concentration changed.

These results are shown in Table 3 below to easily identify the result of the comparison between the nitride film and the oxide film.

TABLE 3

HF concentration () Average etch rate (Å / min) Etch selectivity (nitride / oxide) Nitride film Oxide film 0.01 37 0 - 0.03 51 0 - 0.05 53 0 - 0.08 63 0.77 82 0.1 72 3.6 20

As shown in Table 3, the etching rate is the highest when 0.1HF, but there is a possibility that some problems may occur in the etching selection ratio. It is desirable to obtain. In the 0.08 HF concentration condition, the time required for etching the 1500 nm nitride film is about 24 minutes.

These concentration conditions cannot be determined to be optimized, and a close examination of the effects of hydrogen peroxide remains a challenge. For example, by changing the concentration of hydrogen peroxide, it is possible to obtain a good etching selectivity even at a somewhat high HF concentration condition.

Looking at the experimental results described above, it is most preferable to adjust the concentration of HF contained in the etchant between 0.01 to 0.1 and etch in a closed chamber at 120 ° C. temperature, wherein the etching rate is between about 60 μs / min and 75 μs / min. It can be seen that the value of can be obtained.

With this etching rate, the time required for etching the nitride film, which is usually 1500 Å thick, is about 20 to 25 minutes. The present invention is to reduce the time about 2 times compared to the conventional etching the nitride film of 1500 Å 40 minutes or more, and the temperature of the etchant contained in the chamber was 160 ~ 170 ℃ in the past was about 40 ~ 40 ℃ It is possible to lower the 50 ℃, to enable a low temperature process.

As can be seen from the experimental results of this embodiment, when the etching time in the conventional etching conditions as much as the conventional etching time is allowed to further lower the temperature of the etching process is expected to be possible to process below 100 ℃.

In addition, the present invention can prevent the HF contained in the etchant from flying away when the process is performed using the chamber of the closed structure to minimize the consumption of HF.

When the process is performed using the etchant of the present invention and the chamber of the closed structure, it is possible to achieve the required etching rate while maintaining a constant etching selectivity during the etching of the nitride film / oxide film.

In order to apply these experimental results to the actual process, the concentration of the etchant must be precisely controlled.

To this end, an on-line processing system is provided with a sensor that can accurately monitor the concentration of each chemical contained in the etchant used in the process, and a controller that can finely control the inflow of each chemical introduced into the etching chamber according to the signal of the sensor. This on-line treatment system is required, and a chemical concentration control system proposed by the inventor and previously filed by the present applicant may be used.

As described above, when the wet etching chemical provided by the present invention is used in an etching process, a low temperature process is possible, and the effects obtained by such a low temperature process are as follows.

In addition to a low temperature process, a good etching selectivity can be obtained, and a large etching rate can reduce the process time, thereby improving yield and productivity.

Since the low temperature process is possible, the volatilization amount of the etchant is very small. Moreover, when used in a closed chamber, it is possible to further minimize the concentration change of the chemical, to ensure the reproducibility of the process at a constant process temperature, and to obtain a good etching selectivity and etching rate, The consumption is reduced compared to the conventional.

While reducing the labor of a conventional worker who frequently replenishes water during the process, it is possible to reduce the stagnation time required for maintaining the concentration.

Since the capacity of the heater 14 for heating the etchant may be designed to be small, the installation cost and the power consumption may be reduced.

The effect of the present invention is that it can be more doubled in the closed etch chamber than in the open etch chamber.

On the other hand, the present invention is not limited to the specific preferred embodiment, it is a matter of course that a variety of applications are possible by ordinary knowledge in the art within the technical rights described in the claims.

Claims (2)

An etching solution used in a wet etching process of a silicon nitride film using an oxide film as a mask in an airtight chamber, wherein the etching solution is composed of 85 phosphoric acid (H ₃ PO ₄ ), 0.01 to 0.1 HF, and about 1 hydrogen peroxide. Phosphorus wet chemical. The wet etching chemical of claim 1, wherein the etching solution has a silicon nitride film having an etch rate of 60 μs / min to 74 μm / min, and an etching selectivity of the nitride film / oxide being 20 or more.