TW201909246A - Wafer manufacturing method and wafer - Google Patents

Abstract

Disclosed is a wafer manufacturing method comprising preparing a substrate, the substrate including a dopant doped at a low concentration equal to or less than 1E17 atom/cm ³ , having a frequency equal to or greater than 0.2 Ω·cm Resistivity, and no void defects, a rapid thermal annealing is performed on the prepared substrate; a nitride layer is removed from the substrate subjected to the rapid thermal annealing, the substrate is annealed, and a protrusion is formed on the substrate subjected to annealing Crystal layer.

Description

Wafer manufacturing method and wafer

It involves wafer fabrication methods and wafers.

In the case where the defect-free tantalum layer is fabricated in the active region of the epitaxial wafer, deterioration of device characteristics and yield reduction due to crystal defects and metal contamination can be prevented in advance. Here, such a defect may be a hole or a telluride defect caused by metal contamination or the like.

For the above reasons, epitaxial wafers are used as substrates for various semiconductor devices, such as microprocessors, flash memories, image sensors, and power devices.

In particular, as the demand for wafer microscales increases with higher integration and higher performance of devices, the technical preference for the perfection and strength of epitaxial wafers is increasing.

Conventionally, a substrate is used to fabricate an epitaxial wafer doped with a high concentration of boron of 1E18 atom/cm ³ or higher. In this case, the metal contaminant absorption capacity and strength of the epitaxial wafer can be increased by increasing the effect of oxygen precipitation and boron dislocation pinning.

In the case of manufacturing an epitaxial wafer using a substrate doped with boron at a high concentration, various technical problems such as self-doping may be caused by heat treatment in the process of device fabrication. Therefore, there is a tendency to fabricate epitaxial wafers using a substrate doped with boron at a low concentration rather than a high concentration.

However, epitaxial wafers fabricated using substrates doped with boron at low concentrations rather than high concentrations can cause metal contaminant absorption and mechanical properties compared to epitaxial wafers fabricated using substrates doped with high concentrations of boron. The strength is reduced.

Japanese Patent Publication No. 2011-54763

Japanese Patent Publication No. 2017-50490

The examples provide methods of fabricating wafers having uniform and high oxygen precipitation (bulk micro defect (BMD) density and gettering ability and high strength, and wafers fabricated therefrom.

Embodiments may provide a wafer manufacturing method including the steps of: preparing a substrate including a dopant doped at a low concentration equal to or less than 1E17 atom/cm ³ , having a frequency equal to or greater than 0.2 Ω·cm resistivity, and no void defect; rapid thermal annealing on the prepared substrate; removal of a nitride layer from the substrate subjected to the rapid thermal annealing, the substrate is annealed; An epitaxial layer is formed on the annealed substrate.

For example, the prepared substrate has an initial oxygen concentration of 5E17 atoms/cm ³ to 7E17 atoms/cm ³ .

For example, the prepared substrate may contain a constituent element and a heterogeneous element of a different type from the constituent element.

For example, the constituent element may contain cerium, oxygen, and boron, wherein the heterogeneous element may contain nitrogen, and the concentration of the nitrogen is 1E12 atom/cm ³ to 1E14 atom/cm ³ .

For example, the rapid thermal annealing may be performed in a temperature range of 1100 ° C to 1300 ° C for 1 second to several tens of seconds.

For example, the rapid thermal annealing may be performed using a mixed gas of argon and a nitrogen-based compound.

For example, the rapid thermal annealing can be performed by a plurality of heat treatment procedures, and the heat treatment procedures can be performed at different temperatures.

For example, the heat treatment processes may include a first rapid thermal annealing treatment of the rapid thermal annealing on the substrate at a first temperature during a first period; and after the first rapid thermal annealing treatment, During a second period, a second rapid thermal annealing treatment of the rapid thermal annealing is performed on the substrate at a second temperature different from the first temperature.

For example, the first temperature may range from 1100 ° C to 1150 ° C, the first period may range from 1 second to several tens of seconds, and the second temperature may range from 1150 ° C to 1200 ° C, and the second The period can range from 1 second to 10 seconds.

For example, a temperature rise rate from the first temperature to the second temperature may range from 20 ° C/s to 100 ° C/s.

For example, in the removing step, at least one of a chemical treatment or a machining may be performed on the substrate subjected to the rapid thermal annealing.

For example, the chemical treatment may use a cleaning liquid containing hydrofluoric acid.

For example, the machining can include polishing the substrate subjected to the rapid thermal annealing.

For example, the annealing step may be carried out at a temperature ranging from 600 ° C to 950 ° C for 10 minutes to several hours.

Another embodiment provides a wafer comprising: a substrate; and an epitaxial layer disposed on the substrate, wherein the substrate comprises a dopant having a low concentration equal to or less than 1E17 atom/cm ³ It has a resistivity equal to or greater than 0.2 Ω·cm and has no void defects and has an oxygen precipitation density which varies in the radial direction within 20% of the uniformity.

Yet another embodiment provides a wafer comprising: a substrate; and an epitaxial layer disposed on the substrate, wherein the substrate comprises a dopant having a low concentration equal to or less than 1E17 atom/cm ³ It has a resistivity equal to or greater than 0.2 Ω·cm, and has no void defects, and has dislocation propagation equal to or less than 40 μm.

For example, the wafer may contain a constituent element and a heterogeneous element of a different type from the constituent element.

For example, the dopant may be a p-type conductive dopant, the initial oxygen concentration of the wafer may be 5E17 atom/cm ³ to 7E17 atom/cm ³ , and the constituent element may include germanium, oxygen, and boron. The heterogeneous element may comprise nitrogen at a concentration of 1E12 atom/cm ³ to 1E14 atom/cm ³ .

For example, the oxygen precipitation density may range from 7E9 atom/cm ³ to 1.5E10 atom/cm ³ .

For example, the wafer can have a gettering capability between 80% and 90%.

In the wafer manufacturing method and wafer according to the embodiment, even if the wafer is fabricated using a substrate doped with a low concentration dopant, although there is no crystal originated particles (COP), the wafer is realized in the radial direction. Uniform BMD density distribution, and also achieve high BMD density, high levels of gettering capacity and sufficient strength to resist deformation or slippage during processing.

Hereinafter, exemplary embodiments will be described in detail with reference to the drawings. While the exemplary embodiments are susceptible to various modifications and alternative forms, the specific embodiments are illustrated by way of example, and are described in detail herein, however, it should be understood that The present invention is to be construed as being limited to the modifications and equivalents of the embodiments.

In the description of the embodiments, it is understood that when an element is referred to as "on" or "under" another element, it can The intermediate members are formed indirectly.

In addition, it should be understood that the components "upper" or "lower" can be described with respect to the drawings.

In addition, relative terms such as "first", "second", "upper/over/over" and "lower/lower/lower", the above terms used in the following description may be used for any substance or A component is distinguished from another substance or component and does not require or contain any physical or logical relationship or sequence between such materials or components.

Hereinafter, the term "epi-wafer" may refer to a wafer in which an epitaxial layer (for example, 220 of FIG. 2B) is formed on a substrate (for example, 210A of FIG. 2B), and "epitaxial bare" A "Bare wafer" (or "bulk wafer") may refer to a wafer in which the epitaxial layer 220 is not formed on the substrate 210A. For example, the bulk wafer may correspond to substrate 210A that has been subjected to steps 120 through 150 shown in FIG.

In addition, hereinafter, a wafer not specifically referred to as an "epitaxial wafer", an "epitaxial die" or a "bulk wafer" may be an epitaxial wafer or a bulk wafer.

Hereinafter, a wafer fabrication method according to an embodiment will be described with reference to the drawings.

FIG. 1 is a flow chart for explaining a wafer fabrication method 100 in accordance with an embodiment.

2A and 2B are exemplary process cross-sectional views showing wafers fabricated by the wafer fabrication method 100 shown in FIG. 1.

Referring to Figures 1 and 2A, a substrate 210 is prepared that includes dopants doped at a low concentration (step 120). Here, the dopant may be a p-type dopant such as boron (B), or an n-type dopant such as phosphorus (P), but the embodiment is not limited to the kind of dopant.

In addition, the maximum value of the dopant concentration doped at a low concentration may be 1E17 atom/cm ³ , but the embodiment is not limited thereto. In this case, the substrate 210 may have a resistivity of 0.2 Ω·cm or higher.

In addition, the minimum value of the dopant concentration doped at a low concentration may be 1E13 atom/cm ³ , but the embodiment is not limited thereto. In this case, the substrate 210 may have a resistivity of 100 Ω·cm or less.

In summary, the substrate 120 can be prepared in step 120, and the substrate 120 includes a dopant having a low concentration ranging from 1E13 atom/cm ³ to 1E17 atom/cm ³ . That is, the substrate 210 having a resistivity in the range of 0.2 Ω·cm to 100 Ω·cm can be prepared.

In addition, the substrate 210 prepared in step 120 may be free of void defects, ie, crystal originated particles (COPs). In the presence of COP, as will be described below, when heterogeneous element 212 is present in substrate 210, the COP may be changed to have a needle shape. This may cause defects on the surface of the epitaxial wafer on which the epitaxial layer 220 is formed on the substrate 210A, and thus may cause deterioration in device yield. Therefore, the substrate 210 having no COP can be used to improve the crystallinity of the epitaxial layer 220.

For example, an oxygen atom contained in a single crystal ingot grown by the Czochralski method is an impurity that affects oxygen precipitation and dislocation pinning. Higher concentrations of oxygen atoms (hereinafter referred to as "initial oxygen concentration") ensure higher wafer strength.

FIG. 3 is a graph showing dislocation propagation depending on initial oxygen concentration Oi and nitrogen concentration. The horizontal axis represents the initial oxygen concentration in ppma, and the vertical axis represents misalignment propagation in units of mm.

The result of measuring the length of misalignment propagation (or slip propagation) by heat treatment after the Vickers indentation test while changing the initial oxygen concentration Oi, as shown in FIG. The length of the shift propagation is continuously decreased as the initial oxygen concentration Oi increases. Therefore, when the initial oxygen concentration is high, sufficient wafer strength can be ensured.

A method of evaluating wafer strength will be briefly described below.

In order to evaluate the strength of the wafer, after the wafer surface is pressed with a force of 100 gf using a Vickers hardness tester, for example, in order to generate and propagate misalignment under pressure damage, diffusion heat treatment is performed at a temperature of 900 ° C. 1 hour. Thereafter, selective etching is performed to measure the length of the misalignment, and the strength of the wafer is evaluated by measuring the length of the misalignment generated at the pressing point. Here, in the selective etching, a Wright etchant or a Secco etchant may be used.

There are strength evaluation methods, for example, an indentation rosettes method, which evaluates the slip resistance to mechanical stress and thermal stress. The indentation strain flowering method is well known to those skilled in the art, and thus a detailed description thereof will be omitted.

When the initial oxygen concentration Oi is greater than 7E17 atom/cm ³ , oxygen in the substrate 210 may diffuse into the epitaxial layer 220 during processing of the device, thereby causing defects such as oxygen precipitation in the epitaxial layer 220, and therefore, using Lei A wafer fault may be caused in a device of a wafer. In addition, when the initial oxygen concentration of oxygen contained in the substrate 210 is less than 5E17 atom/cm ³ , oxygen precipitation may not occur.

Therefore, in order to ensure the mechanical strength of the wafer while avoiding the formation of defects, the substrate 210 prepared in step 120 may have an initial oxygen concentration ranging from 5E17 atom/cm ³ to 7E17 atom/cm ³ . That is, based on the new American Society for Testing Materials (ASTM) standard, the substrate 210 prepared in step 120 may have an initial oxygen concentration of 10 ppma to 15 ppma, such as 10 ppma to 14 ppma or 11.5 ppma to 15 ppma. Alternatively, based on the old ASTM standard, the substrate 210 prepared in step 120 may have an initial oxygen concentration of 20 ppma to 30 ppma, such as 20 ppma to 28 ppma or 23 ppma to 30 ppma.

Referring again to FIG. 1, the substrate 210 prepared in step 120 may include a heterogeneous element 212 in addition to its constituent elements. The heterogeneous element refers to an element different from the type of constituent elements of the substrate 210.

The constituent elements and heterogeneous elements of the substrate 210 may vary depending on the characteristics of the device using the epitaxial wafer. Here, the kind of the hetero element and the concentration of the hetero element may be independent of the concentration of the dopant doped into the substrate 210. For example, constituent elements of the substrate 210 may include bismuth (Si), oxygen (O), and boron (B), and the heterogeneous element may include nitrogen (N), but the embodiment is not limited thereto.

In the case where the substrate 210 includes the heterogeneous element 212 having a concentration or higher concentration, a stable growth-in precipitate nuclei is formed at a high temperature, thereby increasing bulk micro defect (BMD). The density of (or oxygen precipitation), which increases the gettering abiblity of the wafer. Here, the growth of the precipitated nuclei may refer to a precipitate nuclei formed during the growth of the single crystal ingot. Here, oxygen precipitation or BMD in the wafer is an index indicating the oxygen precipitation ability of the wafer. After prolonged heat treatment (for example, heat treatment at 800 ° C for 4 hours and heat treatment at 1000 ° C for 16 hours), precipitation easily occurs, and oxygen precipitates or BMD formed in the wafer may be subjected to a laser-scattering method or a single-sided etching method. measuring.

4 is a graph illustrating the BMD density of an epitaxial wafer depending on the concentration ((a) to (d)) of nitrogen (N) as the heterojunction 212.

In FIG. 4, (a) shows the BMD density when nitrogen (N) is not included in the substrate 210, and (b) shows the BMD density when the substrate 210 contains nitrogen having a concentration of 1E12 atom/cm ³ , (c) The BMD density when the substrate 210 contains nitrogen having a concentration of 1E13 atom/cm ³ and (d) indicates the BMD density when the substrate 210 contains nitrogen having a concentration of 1E14 atom/cm ³ .

It can be understood that, as compared with the case where the heterogeneous element 212 is not included in the substrate 210, as shown in (a) of FIG. 4, when the heterogeneous element 212 is included in the substrate 210, the BMD density is increased, as shown in FIG. (b) to (d).

In addition, it is known that various atomic composite structures made of nitrogen (N) in the body of the ruthenium substrate 210, which are heterogeneous elements 212, contribute to misalignment pinning. Therefore, when the heterogeneous element 212 is included in the substrate 210, the strength of the wafer can be enhanced. In FIG. 3, it can be understood that the case of including nitrogen (N) as the heterogeneous element 212 in the substrate 210 (320, 330, and 340) is compared with the case where the heterogeneous element 212 is not included in the substrate 210 (310). In the middle, the length of the slip propagation is more reduced. That is, when the heterogeneous element 212 is included in the substrate 210, dislocation pinning can be enhanced.

In Fig. 3, the symbol "320" indicates a nitrogen concentration ranging from 1E12 atom/cm ³ to 1E13 atom/cm ³ , and the symbol "330" indicates a nitrogen concentration ranging from 1E13 atom/cm ³ to 1E14 atom/cm ^{3 .} In the case of the symbol "340", the concentration of nitrogen is in the range of 1E14 atom/cm ³ to 1E15 atom/cm ³ , and the label "350" corresponds to the third comparative example in which the dopant is doped at a high concentration, which will Described below.

In particular, referring to FIG. 3, it can be understood that when the concentration of nitrogen ranges from 1E12 atom/cm ³ to 1E13 atom/cm ³ (320), when the concentration of nitrogen is equal to or greater than 1E13 atom/cm ³ (330, At 340), misalignment is clearly suppressed more.

However, when the concentration of nitrogen as the heterogeneous element 212 is greater than 1E14 atom/cm ³ , the size of the grown oxygen precipitate excessively increases, resulting in defects on the surface of the epitaxial wafer, which may lower the epitaxial wafer quality.

FIG. 5 is a graph illustrating nitrogen-induced large defects (NILD) occurring or not occurring in an epitaxial wafer according to an initial oxygen concentration Oi and a nitrogen concentration [N].

Referring to FIG. 5, it can be understood that when the concentration [N] of nitrogen is equal to or less than 1E14 atom/cm ³ (NILD FREE), no NILD is found in the epitaxial wafer. When the nitrogen concentration [N] is greater than 1E14 atom/cm ³ (NILD FOUND), NILD is found in the epitaxial wafer. In particular, it is understood that when the concentration [N] of nitrogen is equal to or less than 1E14 atom/cm ³ , the presence or absence of NILD is not related to the initial oxygen concentration Oi.

6A and 6B are views showing the presence or absence of a pattern on the surface of the epitaxial layer 220 in the epitaxial wafer.

As shown in FIG. 5, when the NILD is found (410), a defect pattern may appear in the center of the surface of the crystal layer 220 as shown in FIG. 6A. On the other hand, as shown in FIG. 5, when the NILD is not found (420), no pattern appears in the center of the surface of the epitaxial layer 220, as shown in FIG. 6B.

In addition, when the nitrogen concentration of the heterogeneous element 212 is less than 1E12 atom/cm ³ , as shown in FIG. 3, the misalignment propagation suppression is small, which may cause the wafer strength to decrease, and the BMD density to decrease, as shown in FIG.

In summary, the nitrogen density [N] of the BMD density is high, the wafer strength is significantly increased, and NILD does not occur, and the nitrogen concentration [N] can be from 1E12 atom/cm ³ to 1E14 atom/cm ³ , for example, from 1E12 atom/cm ^{3 .} Up to 6.8E13 atom/cm ³ , but the embodiment is not limited thereto.

In summary, as described above, the substrate 210 prepared in the step 120 may have an initial oxygen concentration ranging from 5E17 atom/cm ³ to 7E17 atom/cm ³ , may have no void defects, and may include 1E10 atom/cm ³ to 1E17 atom/cm. ^A range of low concentration doped dopants may have a resistivity ranging from 0.2 Ω·cm to 100 Ω·cm, and may include heterogeneous elements doped at a concentration of 1E12 atom/cm ³ to 1E14 atom/cm ³ 212. The substrate 210 that satisfies these various conditions can be manufactured by various methods.

Generally, a floating zone (FZ) method or a Czochralski (CZ) method is often used as a method of manufacturing the substrate 210. Since it is difficult to manufacture the large-diameter substrate 210 and the process cost of growing the single crystal ingot by the FZ method is very high, the CZ method is usually used to grow the single crystal ingot.

In the case where the constituent element of the substrate 210 is tantalum, after the polycrystalline germanium is loaded into the quartz crucible and melted by heating the graphite heater by the CZ method, the seed crystal is immersed in the obtained crucible melt so that it occurs at the melt interface. Crystallization, and then the seed crystal is lifted while rotating, thereby growing a single crystal germanium ingot. Thereafter, the grown single crystal germanium ingot is sliced, etched, and polished to fabricate the substrate 210 in the form of a wafer.

At this time, in order to manufacture the substrate 210 satisfying the above various conditions, at least one of various factors such as the amount introduced, the time point of introduction, and the introduction of the heterogeneous element 212 and the dopant into the ruthenium melt may be complicatedly adjusted. Speed, speed, number, position of the melt, and the heating temperature of the heater that heats the helium melt, the position and strength of the magnetic field in the helium melt, and the rate and speed of the seed crystal.

Referring again to FIG. 1, the substrate 210 prepared in step 120 is subjected to rapid thermal annealing (RTA) (or rapid thermal processing (RTP)) (step 130). Step 130 may be performed to form a vacuum-supersaturated area in the body of the substrate 210. In addition, by doping the matrix 210 with heterogeneous elements such as nitrogen and oxygen, various complexes can be induced to form, during the RTA, by the interaction of vacancies, oxygen and nitrogen, which can advantageously act on the formation of oxygen precipitates in subsequent processing. nuclear. The complex formed at this time may generally be a vacuum-oxygen complex (VOx) and a nitrogen-vacancy complex (N-V), but the embodiment is not limited thereto.

Meanwhile, when rapid thermal annealing (RTA) is performed at a temperature lower than 1,100 ° C, the supersaturation of the void and the formation of various complexes may not sufficiently occur. In addition, when rapid thermal annealing (RTA) is performed at a temperature higher than 1300 ° C, slippage may occur in the epitaxial wafer, resulting in deterioration of wafer strength. Therefore, rapid thermal annealing (RTA) can be performed in a temperature range of 1100 ° C to 1300 ° C, but the embodiment is not limited thereto.

In addition, when rapid thermal annealing (RTA) is performed for more than several tens of seconds, slippage may occur in the epitaxial wafer, resulting in a decrease in wafer strength. Therefore, rapid thermal annealing (RTA) can be performed in the range of 1 second to several tens of seconds, but the embodiment is not limited thereto.

In addition, rapid thermal annealing (RTA) can be carried out using a mixed gas in which argon (Ar) and a nitrogen-based compound are mixed.

In addition, when a void is introduced into the body of the substrate 210 at a high temperature by rapid thermal annealing (RTA), an environment in which oxygen precipitation (or BMD) is generated may be generated. Thus, in order to sufficiently introduce the void into the body, rapid thermal annealing (RTA) can be performed by a plurality of heat treatment steps, and each heat treatment step can be performed at different temperatures.

Hereinafter, an example in which rapid thermal annealing (RTA) is performed by two heat treatment steps will be described with reference to the drawings. Here, in the two heat treatment steps, the heat treatment step performed first is referred to as "first RTA processing step", and the heat treatment step performed later is referred to as "second RTA processing step".

FIG. 7 is an exemplary diagram for explaining an RTA process according to an embodiment.

Referring to FIG. 7, the first RTA processing step may be performed during the first period P1 at the first temperature T1, after which the second RTA processing step may be performed during the second period P2 at the second temperature T2. Here, the first period P1 is a period from the second time t2 to the third time t3, and the second period P2 is a period from the fourth time t4 to the fifth time t5.

Here, the first temperature T1 and the second temperature T2 may be different, and the second temperature T2 may be higher than the first temperature T1. In addition, the second period P2 may be equal to or longer than the first period P1, and each of the first temperature T1 and the second temperature T2 may be in a range of 1100 ° C to 1300 ° C, in the first period P1 and the second period P2 Each of them can range from 1 second to several tens of seconds.

Next, referring to FIG. 7, while a mixed gas in which a nitrogen-based gas and an argon (Ar) gas are mixed is introduced into the chamber, the temperature in the chamber (or the temperature of the substrate 210) is increased in the first temperature rising period #1. The first temperature rising period #1 is a period from the first time t1 to the second time t2, during which the rate of temperature rise may be from 20 ° C / s to 100 ° C / s.

Thereafter, the first RTA processing step may be performed during the first period P1 at the first temperature T1. For example, the range of the first period P1 may be 1 second to 10 seconds, and the range of the first temperature T1 may be 1100 ° C to 1300 ° C. At this time, a nitrogen-based gas may be introduced into the chamber at a flow rate of 10% or more of the total mixed gas flow rate. For example, when the total mixed gas is introduced into the chamber at a flow rate of 400 sccm, the nitrogen-based gas can be introduced at a flow rate of 41 sccm. During the first period P1, when the first RTA processing step is performed, the reason for introducing the nitrogen-based gas into the chamber is to control the vacancy concentration in the substrate 210. However, when the second RTA processing step is performed, the introduction of the nitrogen-based gas is stopped.

After the first RTA processing step is performed, the temperature inside the chamber (or the temperature of the substrate 210) is increased (second temperature rising period #2). The second temperature rising period #2 is a period from time t3 to time t4, and the rate of temperature rise during this period may be in the range of 20 ° C / s to 100 ° C / s. Thereafter, a second RTA processing step can be performed at the second temperature T2 for the second period P2. For example, the second period P2 may range from 1 second to 10 seconds, and the second temperature T2 may range from 1150 °C to 1200 °C.

In FIG. 7, the flow rate of the argon gas supplied from the first time t1 to the sixth time t6 may be in the range of 10 slm to 30 slm, but the embodiment is not limited thereto.

After step 130, the nitride layer can be removed from the substrate 210 that has been rapidly thermally annealed (step 140). During the RTA, there may be a thermal nitride layer (hereinafter referred to as a "nitridation layer") generated during the heat treatment and metal contamination caused in the furnace. Since the perfection of the final epitaxial wafer may not be ensured when the epitaxial layer 220 is formed without controlling the nitride layer and metal contamination, step 140 may be performed to remove the source of contamination.

Alternatively, when an additional heat treatment is performed in a state where such a thermal nitride layer remains on the surface of the substrate 210, since a nitride layer exists on the surface of the substrate 210, nitrogen may diffuse into the body of the substrate 210, resulting in A nitrogen-oxygen precipitate is formed in excess. In addition, when the epitaxial layer 220 is grown in a state in which the thermal nitride layer remains, the reaction between the germanium substrate 210 and the germanium-based gas is suppressed to inhibit the growth of the epitaxial layer 220, and thus the epitaxial layer 220 may not be secured. Perfection. Therefore, in order to prevent the possibility of occurrence of undesired defects on the surface during the subsequent heat treatment, step 140 may be performed to remove the nitride layer as a source of contamination.

According to an embodiment, in step 140, at least one of chemical treatment or machining may be performed on the substrate that has been subjected to rapid thermal annealing to remove the nitride layer.

For example, when chemical treatment is performed to remove the nitride layer, RCA or some other cleaning method similar to RCA may be used, but embodiments are not limited to any particular cleaning method. Further, when chemical treatment is performed to remove the nitride layer, a cleaning liquid containing hydrofluoric acid (HF) may be used, but the embodiment is not limited thereto.

Alternatively, when mechanical processing is performed to remove the nitride layer, the substrate that has been subjected to rapid thermal annealing may be polished.

After step 140, the substrate 210 from which the nitride layer has been removed is annealed (step 150). Based on the vacancy-supersaturated regions formed by the blocks in step 130 and the various complexes formed therefrom, step 150 is performed to induce enhanced oxygen precipitate nucleation and to obtain additional thermal stability. As the vacancy supersaturation increases, the oxygen precipitate nucleation rate can increase, and as the amount of complex that can act as a nucleation site increases, the oxygen precipitate nucleation rate can also be increased. For example, by using the vacancy-oxygen complex and the nitrogen-vacancy complex formed in the RTA process according to step 130 of the embodiment as a nucleation site, an enhanced oxygen precipitation nucleation rate can be obtained in the annealing process. Therefore, since a larger amount of oxygen precipitated nuclei is formed after the method according to the embodiment, a sufficient amount of oxygen precipitated nuclei may exist even after the step 160 of forming the epitaxial layer 220, and may be grown during subsequent processes. It forms a precipitate and may contribute to gettering, but the embodiment is not limited thereto.

For example, when the annealing temperature is higher than 950 ° C, thermal slippage may occur, which may deteriorate the crystallinity of the epitaxial layer 220 in the epitaxial wafer. In addition, when the annealing temperature is lower than 600 ° C, oxygen precipitation nucleation may not occur sufficiently. Therefore, the annealing treatment can be performed in a temperature range of 600 ° C to 950 ° C, but the embodiment is not limited thereto.

Further, the annealing treatment can be carried out in the range of 10 minutes to several hours in consideration of the culture time.

FIG. 8 is an exemplary chart for explaining an annealing process according to an embodiment.

Referring to FIG. 8, when argon gas is introduced into the chamber, the temperature in the chamber (or the temperature of the substrate 210) increases in the third temperature rising period #3. The third temperature rising period #3 is a period from the seventh time t7 to the eighth time t8, and the rate of temperature rise during this period may be in the range of 2 ° C / min to 10 ° C / min.

Thereafter, annealing may be performed at the third temperature T3 for the third period P3. For example, the third stage P3 may be in the range of 0.1 hour to 2 hours, and the third stage T3 may be in the range of 600 °C to 950 °C. Thereafter, the temperature in the chamber (or the temperature of the substrate 210) is lowered at the first deceleration time #1. The first deceleration period #1 is a period from the ninth time t9 to the tenth time t10, and the rate of temperature drop during this period may be in the range of 2 ° C/min to 10 ° C/min, but the embodiment is not limited thereto. .

In FIG. 8, the flow rate of the argon gas supplied from the seventh time t7 to the tenth time t10 may be in the range of 10 slm to 30 slm, but the embodiment is not limited thereto.

After step 150, as shown in FIG. 2B, an epitaxial layer 220 is formed on the annealed substrate 210A, thereby completing the fabrication of the epitaxial wafer (step 160).

Although not shown in FIG. 1, the pre-processing procedure may be further performed after performing step 150 and before performing step 160. The reason for performing the pretreatment procedure is to remove organic contaminants and native oxides present on the substrate 210A on which the step 150 has been performed.

FIG. 9 is an exemplary diagram for explaining a pre-processing procedure and step 160 in accordance with an embodiment.

Referring to FIG. 9, in the case of the pre-processing procedure, after the substrate 210A in which the step 150 has been performed is loaded into the chamber, the temperature inside the chamber (or the temperature of the substrate 210A) is increased in the fourth temperature rising period #4. The fourth temperature rising period #4 is a period from the eleventh time t11 to the twelfth time t12, and the rate of temperature rise during this period may be in the range of 1 ° C / s to 20 ° C / s.

Subsequently, the pre-processing procedure may be performed at a fourth temperature T4 via hydrogen (H ₂ ) baking and hydrogen chloride (HCl) etching in the fourth period P4. Here, the fourth period P4 may be a period from the twelfth time t12 to the thirteenth time t13, and may be in the range of 20 seconds to 40 seconds, and the fourth temperature T4 may be in the range of 1110 ° C to 1160 ° C . Here, hydrogen is used to remove the native oxide formed on the substrate 210 on which the step 150 has been performed, and the substrate 210 on which the step 150 has been performed is cleaned. In addition, hydrogen chloride is used to remove surface defects from the substrate 210A on which the step 150 has been performed.

Subsequently, the temperature in the chamber (or the temperature of the substrate 210A) is lowered at the second deceleration time #2. The second deceleration period #2 is a period from the thirteenth time t13 to the fourteenth time t14, and the rate of temperature drop during this period may be in the range of 1 ° C / s to 20 ° C / s.

Subsequently, an epitaxial layer 220 is formed on the substrate 210A while the reactant is supplied at the fifth temperature T5 at the fifth period P5. Here, the fifth period P5 may be a period from the fourteenth time t14 to the fifteenth time t15, and may be in the range of 10 seconds to 30 seconds, and the fifth temperature T5 may be in the range of 1090 ° C to 1140 ° C . When a reactant such as SiH ₄ , Si ₂ H ₆ , dichlorodecane (DCS) or tetrachlorosilane (TCS), for example, a TCS gas containing ruthenium and a doping gas such as B ₂ H ₆ (PH ₃ ) are The fifth period P5 is introduced into the chamber, and a chemical deposition reaction may be formed on the surface of the substrate 210A to form the epitaxial layer 220, but the embodiment is not limited thereto.

Subsequently, the temperature in the chamber (or the temperature of the substrate 210A) is lowered at the third deceleration time #3. The third deceleration period #3 is a period from the fifteenth time t15 to the sixteenth time t16, and the rate of temperature drop during the period may be in the range of 1 ° C / s to 20 ° C / s.

In FIG. 9, the flow rate of the hydrogen gas (H2) supplied from the eleventh time t11 to the sixteenth time t16 may be in the range of 50 slm to 100 sl, and the reactant supplied from the fourteenth time t14 to the fifteenth time t15. The flow rate may be in the range of 0.01 slm to 0.1 slm, but the embodiment is not limited thereto.

For example, the thickness of the epitaxial layer 220 formed on the substrate 210A may be in the range of 1 mm to 10 mm, but the embodiment is not limited thereto.

Hereinafter, embodiments and comparative examples will be compared and described with reference to FIGS. 10 to 13. Further, when the steps 130, 140, 150, and 160 shown in FIG. 1 are performed as shown in FIGS. 7 to 9, the above-described FIGS. 4 to 6 and FIGS. 10 to 13 which will be described below may correspond to the obtained results.

The first comparative example is a case where an epitaxial wafer is manufactured without performing RTA processing and annealing treatment.

The wafer manufacturing method according to the second comparative example (hereinafter referred to as "second comparative example") is a case where step 150 is omitted from the wafer manufacturing method 100 shown in FIG. 1. Further, in the case of the second comparative example, the step 140 shown in FIG. 1 may be omitted.

The wafer manufacturing method according to the third comparative example (hereinafter referred to as "third comparative example") is a case where steps 130 and 140 are omitted from the wafer manufacturing method 100 shown in FIG. 1.

According to the wafer manufacturing method of the fourth comparative example (hereinafter referred to as "fourth comparative example"), the epitaxial wafer is manufactured using the substrate 210, and the substrate 210 includes a high concentration dopant such as a concentration of 1E18 atom/cm ³ The case of boron or higher is different from the embodiment.

In order to increase the thermal stability of the oxygen precipitate, the second comparative example may include an RTA method, and the third comparative example may include an annealing treatment.

In the case where the RTA treatment was not performed and only the annealing treatment was performed as in the third comparative example, the oxygen precipitated nuclei present in the grown state grew, and only slight additional nucleation was induced. Therefore, when the process of forming the epitaxial layer 220 is performed at an ultrahigh temperature, for example, at 1100 ° C in step 160, the density of the nuclei remaining undissipated may be insufficient.

In addition, the substrate 210 without COP generally has a non-uniform distribution of growth point defects, that is, vacancy defects in the initial state and/or self-interstitial dot defects. Therefore, the uniformity of BMD density is poor in the radial direction of the wafer. When the RTA treatment was not performed and only the annealing treatment was performed as in the third comparative example, the unevenness of the BMD density could not be overcome.

In the case where only the RTA process is performed as in the second comparative example and the annealing process is not performed, when the substrate 210 is rapidly heated to a high temperature, a vacancy-supersaturation region is formed in the body of the substrate 210, and then the substrate 210 is rapidly cooled again to room temperature. It can suppress the recombination of point defects and save the vacancy-supersaturation area. Even if only the RTA process is performed, as long as the uniformity of the heat treatment is ensured, the vacancy-supersaturated region can be equally formed over the entire substrate 210 to compensate for the unevenness of the BMD density. However, since the vacancy-supersaturated region formed in the RTA process is not in a stable state during the high-temperature heat treatment process of the epitaxial process in the step 160 described later, the process of forming the epitaxial layer 220 in most of the vacancy-supersaturated regions is performed. The middle will dissipate. Therefore, after the step 160 of forming the epitaxial layer 220, the effect of the RTA process may disappear and the BMD density of the epitaxial wafer may deteriorate.

In summary, when a wafer according to the first comparative example, the second comparative example, or the third comparative example is fabricated using the substrate 210 including the low concentration dopant, the BMD density may be low and uneven.

On the other hand, in the case of the wafer manufacturing method according to the embodiment, the RTA process is performed in step 130 to form a void-supersaturated region and various composites in the body of the substrate 210, after which the annealing process is performed in step 150. Therefore, the number of oxygen-precipitating nuclei present increases even after the high-temperature heat treatment at 1100 ° C or higher, compared to the process of forming the epitaxial layer 220. Therefore, when the epitaxial layer 220 is formed in step 160, metal contaminant absorption can be ensured by using an oxygen precipitate grown from the precipitated nuclei present in the device processing using the epitaxial wafer.

FIG. 10 is a graph comparing the BMD densities of the epitaxial wafers and the examples manufactured according to the first comparative example to the fourth comparative example. In FIG. 10, (a) corresponds to the first comparative example, (b) corresponds to the second comparative example, (c) corresponds to the third comparative example, (d) corresponds to the embodiment, and (e) corresponds to the first Four comparison examples.

It can be understood that the BMD density of the epitaxial wafer fabricated by the embodiment shown in FIG. 10(d) is higher than that shown in (a) to (c) of FIG. 10 by the first comparative example, respectively. The BMD density of the epitaxial wafers produced in the second comparative example and the third comparative example. It can be understood that the BMD density shown in (d) of FIG. 10 is substantially equal to the BMD density of the epitaxial wafer shown in (e) of FIG. 10 manufactured by the fourth comparative example. The BMD density of the wafer according to the embodiment may range from 7E9 atom/cm ³ to 1.5E10 atom/cm ³ .

11A to 11E are graphs comparing the uniformity of the BMD density of the epitaxial wafers and the examples manufactured according to the first comparative example to the fourth comparative example. In each of the figures, the horizontal axis represents the position of the wafer in the radial direction, "0" represents the center of the wafer, and the vertical axis represents the BMD density.

The uniformity of the BMD density of the epitaxial wafers fabricated by various methods is compared with reference to FIGS. 11A through 11E. The uniformity of the BMD density can be calculated by the following formula 1.

Equation 1: D = (MAX-MIN)/2AVG

Here, "D" is the uniformity of BMD density, "MAX" is the maximum value of BMD density, "MIN" is the minimum value of BMD density, and "AVG" is the average value of BMD density.

As shown in FIG. 11A, the uniformity of the BMD density in the epitaxial wafer fabricated by the first comparative example was 141.4%.

As shown in FIG. 11B, the uniformity of the BMD density in the epitaxial wafer fabricated by the second comparative example was 51.8%.

As shown in FIG. 11C, the uniformity of the BMD density in the epitaxial wafer manufactured by the third comparative example was 96.6%.

As shown in Fig. 11D, the uniformity of the BMD density in the epitaxial wafer fabricated by the example was 10.4%.

As shown in FIG. 11E, the uniformity of the BMD density in the epitaxial wafer manufactured by the fourth comparative example was 1.8%.

In summary, the uniformity of the BMD density in the wafer according to the embodiment may be within 20%, preferably, may be about 10%, for example, 10.4%, but the embodiment is not limited thereto.

12A to 12E show photographs of epitaxial wafers manufactured according to the first comparative example to the fourth comparative example and the embodiment. In each figure, black dots represent BMD.

12A to 12E are photographs of an etched epitaxial wafer obtained by an optical microscope.

As shown in FIG. 12A, the epitaxial wafer fabricated by the first comparative example has a very small amount of BMD. The epitaxial wafer fabricated by the second comparative example was processed using an RTA without annealing treatment, as shown in Fig. 12B, and had a very small number of BMD as in Fig. 12A. On the other hand, the epitaxial wafer fabricated by the third comparative example used an annealing treatment without RTA treatment, as shown in FIG. 12C, and had a larger number of BMD than in FIG. 12A or FIG. 12B.

The epitaxial wafer fabricated by the embodiment is simultaneously subjected to RTA processing and annealing treatment as shown in FIG. 12D, and it can be understood that the epitaxial crystal of the embodiment is compared to those of the epitaxial wafers of FIGS. 12A to 12C. The circle has a significantly larger number of BMDs.

Thus, when the epitaxial wafer is fabricated according to the embodiment, the number of BMDs is increased to correspond to the number of BMDs in the epitaxial wafer fabricated by the fourth comparative example shown in FIG. 12E.

Fig. 13 is a graph showing the gettering ability of the epitaxial wafers manufactured according to the first comparative example to the fourth comparative example and the examples. Figure 13 shows the gettering ability of contaminated metals, particularly contaminated metal nickel (Ni).

In the epitaxial wafer manufactured by the first comparative example, as shown in (a) of FIG. 13, the gettering ability is very low, in the range of 50% to 60%. The epitaxial wafer manufactured by the third comparative example was annealed without RTA treatment, and as shown in FIG. 13(b), the gettering ability for nickel was increased to 70% as compared with FIG. 13(a).

At this time, the epitaxial wafers processed using the RTA and the annealed according to the embodiment are as shown in FIG. 13C, and it can be understood that the epitaxial wafers of (a) and (b) of FIG. 13 are sucked. Compared with the heterogeneous ability, the gettering ability of nickel is further increased to 80% or more. Thus, when the epitaxial wafer is manufactured according to the embodiment, its gettering ability is enhanced to correspond to the gettering ability of the epitaxial wafer fabricated by the fourth comparative example shown in (d) of FIG.

In summary, it can be understood that when the epitaxial wafer is fabricated by using RTA processing and annealing treatment as in the embodiment, the BMD density of the epitaxial wafer is increased, the uniformity of BMD density is improved, and the gettering ability of the wafer is improved. .

Hereinafter, a wafer (for example, an epitaxial wafer or a bulk wafer) according to an embodiment will be described with reference to the drawings.

The bulk wafer (or epitaxial die) according to the embodiment may be fabricated by performing steps 120 to 150 of the above-described wafer fabrication method 100 shown in FIG. 1, but the embodiment is not limited thereto. That is, a bulk wafer (or epitaxial die) according to another embodiment may be fabricated by a method other than the wafer fabrication method 100 shown in FIG.

In addition, the epitaxial wafer of the embodiment may be fabricated by performing steps 120 to 160 of the above-described wafer fabrication method 100 illustrated in FIG. 1, but the embodiment is not limited thereto. That is, the epitaxial wafer according to another embodiment can be fabricated by a method other than the wafer fabrication method 100 shown in FIG.

The wafer according to the embodiment may include a dopant doped at a low concentration of 1E17 atom/cm ³ or less.

In addition, the minimum value of the concentration of the dopant included in the wafer may be 1E13 atom/cm ³ , but the embodiment is not limited thereto. Here, the conductive dopant may be a p-type dopant or an n-type dopant. For example, boron can be included in the wafer as a p-type dopant. That is, the wafer according to the embodiment may have a resistivity of 0.2 Ω·cm or more. In addition, the wafer according to the embodiment may have a resistivity of 100 Ω·cm or less.

Further, the uniformity of the BMD (oxygen precipitation) density in the wafer according to the embodiment may be within 20%, preferably, about 10%, for example, 10.4% in the radial direction. Here, the uniformity of the BMD density can be calculated using Equation 1 above. In particular, the wafer according to an embodiment may have no hole defects, ie COP. When the wafer has no COP, the wafer may be perfect.

Generally, when the wafer has no COP, its BMD density is not uniform. On the other hand, although the wafer according to the embodiment has no COP, for example, as shown in FIG. 11D, its BMD density is uniform because the wafer is subjected to RTA processing and annealing treatment as shown in FIG.

In addition, the initial oxygen concentration in the wafer according to the embodiment may be in the range of 5E17 atom/cm ³ to 7E17 atom/cm ³ . Thus, since the initial oxygen concentration of the wafer has a high value of 7E17 atom/cm ³ , as shown in FIG. 3, the mechanical strength of the wafer can be improved. In addition, when the wafer has an initial oxygen concentration within the above range, no defects are generated in the apparatus using the wafer, and therefore, no leakage is caused and oxygen precipitation can be stably generated.

In addition, the wafer according to the embodiment may include constituent elements and heterogeneous elements. Heterogeneous elements and constituent elements are of different types. For example, the constituent elements may include ruthenium, and the heterogeneous elements may include nitrogen. Since the wafer according to the embodiment includes a heterogeneous element, the strength of the wafer can be enhanced, as shown in FIG. 3, and the BMD density is increased, so that the gettering ability can be enhanced, as shown in FIG.

In addition, the BMD density in the wafer according to the embodiment may be in the range of 7E9 atom/cm ³ to 1.5E10 atom/cm ³ , and the metal gettering ability of the wafer may be in the range of 80% to 90%.

For example, the concentration of the hetero element (for example, nitrogen) may be in the range of 1E12 atom/cm ³ to 1E14 atom/cm ³ , but the embodiment is not limited thereto. When the concentration of the heterogeneous element is within this range, as shown in FIG. 4, the BMD density is high, as shown in FIG. 3, the wafer strength is remarkably enhanced, and as shown in FIGS. 5 and 6B, NILD does not occur. Referring to FIG. 3, it can be seen that the wafer according to the embodiment has misalignment propagation of 40 mm or less, thereby having sufficient strength to resist deformation or sliding during processing.

Although the embodiments have been described above, the embodiments are presented by way of example only and are not intended to limit the invention. It will be apparent to those skilled in the art that various alternatives, modifications and variations can be made in the spirit and scope of the embodiments. For example, various components specifically described in the embodiments may be modified in various ways. In addition, the differences associated with these modifications and applications are to be construed as being included within the scope of the invention as defined by the appended claims.

100‧‧‧ Wafer manufacturing method

210, 210A, 310, 320, 330, 340, 350‧‧‧ substrates

212‧‧‧Heterogeneous elements

220‧‧‧ epitaxial layer

FIG. 1 is a flow chart for explaining a wafer manufacturing method according to an embodiment. 2A and 2B are exemplary process cross-sectional views showing wafers fabricated by the wafer fabrication method shown in Fig. 1. FIG. 3 is a view showing misalignment propagation depending on initial oxygen concentration and nitrogen concentration. 4 is a graph illustrating that the BMD density of an epitaxial wafer depends on the concentration of nitrogen ((a) to (d)) as a heterogeneous element. Figure 5 is a graph illustrating the occurrence or non-occurrence of NILDs in epitaxial wafers depending on initial oxygen concentration and nitrogen concentration. 6A and 6B are views illustrating the appearance or absence of a pattern on the surface of an epitaxial layer in an epitaxial wafer. FIG. 7 is an exemplary diagram for explaining an RTA process according to an embodiment. FIG. 8 is an exemplary chart for explaining an annealing process according to an embodiment. FIG. 9 is an exemplary diagram for explaining a pre-processing procedure and step 160 in accordance with an embodiment. FIG. 10 is a graph comparing the BMD densities of the epitaxial wafers and the examples manufactured according to the first comparative example to the fourth comparative example. 11A to 11E are graphs comparing the uniformity of the BMD density of the epitaxial wafers and the examples manufactured according to the first comparative example to the fourth comparative example. 12A to 12E show photographs of epitaxial wafers manufactured according to the first comparative example to the fourth comparative example and the embodiment. Fig. 13 is a graph showing the gettering ability of the epitaxial wafers manufactured according to the first comparative example to the fourth comparative example and the examples.

Claims (20)

A wafer manufacturing method comprising: preparing a substrate, wherein the substrate comprises a low concentration of a dopant equal to or less than 1E17 atom/cm ³ , and one of the dopants has a resistivity equal to or greater than 0.2 Ω·cm, And the dopant has no void defects; performing a rapid thermal annealing on the prepared substrate; removing a nitride layer from the substrate subjected to the rapid thermal annealing; annealing the substrate; and on the substrate subjected to annealing An epitaxial layer is formed. The method of claim 1, wherein the substrate is prepared to have an initial oxygen concentration of 5E17 atoms/cm ³ to 7E17 atoms/cm ³ . The method of claim 1, wherein the substrate is prepared comprising: a constituent element; and a heterogeneous element different from the type of the constituent element. The method of claim 3, wherein the constituent element comprises ruthenium, oxygen and boron, wherein the heterogeneous element comprises nitrogen at a concentration of from 1E12 atom/cm ³ to 1E14 atom/cm ³ . The method of claim 1, wherein the rapid thermal annealing is performed in a temperature range of 1100 ° C to 1300 ° C for 1 second to several tens of seconds. The method of claim 1, wherein the rapid thermal annealing is performed using a mixed gas of argon and a nitrogen-based compound. The method of claim 1, wherein the rapid thermal annealing is performed by a plurality of heat treatment procedures, and wherein the heat treatment procedures are performed at different temperatures. The method of claim 7, wherein the heat treatment process comprises: performing a first rapid thermal annealing treatment of the rapid thermal annealing on the substrate at a first temperature during a first period; and in the first rapid heat After the annealing treatment, a second rapid thermal annealing treatment of the rapid thermal annealing is performed on the substrate at a second temperature different from the first temperature during a second period. The method of claim 8, wherein the first temperature ranges from 1100 ° C to 1150 ° C, the first period ranges from 1 second to several tens of seconds, and wherein the second temperature ranges from 1150 ° C to 1200 °C, the second period ranges from 1 second to 10 seconds. The method of claim 9, wherein a temperature rise rate from the first temperature to the second temperature is from 20 ° C/s to 100 ° C/s. The method of claim 1, wherein in the step of removing the nitride layer from the substrate subjected to the rapid thermal annealing, at least one of a chemical treatment or a mechanical processing is performed on the substrate subjected to the rapid thermal annealing. One. The method of claim 11, wherein the chemical treatment uses a cleaning solution containing hydrofluoric acid. The method of claim 11, wherein the machining comprises polishing the substrate subjected to the rapid thermal annealing. The method of claim 1, wherein the annealing step is performed in a temperature range of 600 ° C to 950 ° C for 10 minutes to several hours. A wafer comprising: a substrate; and an epitaxial layer disposed on the substrate, wherein the substrate comprises a dopant having a low concentration equal to or less than 1E17 atom/cm ^{3 , and} one of the dopants has a resistance The rate is equal to or greater than 0.2 Ω·cm, the dopant has no void defects, and the dopant has an oxygen precipitation density which varies in a radial direction within a range of 20% of uniformity. A wafer comprising: a substrate; and an epitaxial layer disposed on the substrate, wherein the substrate comprises a dopant having a low concentration equal to or less than 1E17 atom/cm ^{3 , and} one of the dopants has a resistance The rate is equal to or greater than 0.2 Ω·cm, and the dopant has no void defects and has misalignment propagation of 40 μm or less. The wafer of any one of claims 15 or 16, wherein the wafer comprises: a constituent element; and a heterogeneous element different from the type of the constituent element. The wafer of claim 17, wherein the dopant is a p-type conductive dopant, wherein the wafer has an initial oxygen concentration of 5E17 atoms/cm ³ to 7E17 atoms/cm ³ , and wherein the composition The element comprises cerium, oxygen and boron, wherein the heterogeneous element comprises nitrogen at a concentration of from 1E12 atom/cm ³ to 1E14 atom/cm ³ . The wafer of claim 15 wherein the oxygen precipitation density ranges from 7E9 atoms/cm ³ to 1.5E10 atoms/cm ³ . The wafer of any one of claims 15 or 16, wherein the wafer has a gettering capability of between 80% and 90%.