Silicon epitaxial semiconductor wafer and method for its production and method for producing a solid-state image sensor device

Abstract

Silicon epitaxial wafer (200) in which a silicon epitaxial layer (20) is formed on a surface of a silicon wafer (10), wherein the silicon wafer (10) has a modification layer (18) which contains carbon as a solid solution in the surface section, and the half-width of the peak of a carbon concentration profile of the modification layer in the direction of the thickness of the semiconductor wafer is 100 nm or less, and wherein a peak value of a hydrogen concentration profile detected by SIMS in a surface section of the silicon wafer (10) on a side on which the epitaxial layer is formed, 1.0·10 17 atoms / cm² 3 or more and 1.0·10 22 atoms / cm² 3 or less.

Description

TECHNICAL AREA

[0001] This disclosure relates to a silicon epiaxial semiconductor wafer and a method for producing the same, as well as a method for producing a solid-state image sensor device. BACKGROUND

[0002] Silicon epitaxial semiconductor wafers, in which an epitaxial layer is formed on a semiconductor wafer, are used as device substrates for various semiconductor devices such as metal-oxide-semiconductor field-effect transistors (MOSFETs), dynamic random access memories (DRAMs), power transistors, and backlit solid-state image sensor devices.

[0003] In recent years, back-illuminated solid-state image sensor devices have been widely used, for example, in digital video cameras and mobile phones such as smartphones, because, due to the fact that their wiring layer and the like are located in a deeper layer than a sensor unit, they can capture light directly from the outside and thus take sharper pictures or videos even in dark areas.

[0004] With the increasing development of miniaturized and more powerful semiconductor devices in recent years, the silicon epitaxial semiconductor wafers used as device substrates must be of higher quality to improve device properties. To further enhance device properties, techniques are being developed to improve crystal quality using oxygen precipitation heat treatment, getter techniques to prevent heavy metal contamination during epitaxial growth, and similar methods.

[0005] For example, JP 2013 - 197 373 A discloses a method for producing an epitaxial wafer in which, after carrying out an oxygen precipitation heat treatment, an epitaxial layer is formed on a silicon substrate, wherein the conditions for the oxygen precipitation heat treatment are controlled such that the epitaxial layer exhibits a creepage current of 1.5*10 -10 A or less.

[0006] Furthermore, with regard to getter techniques, the applicant of the present application proposes in JP 2010 - 287 855 A a silicon wafer with a surface area of ​​1 · 10⁻⁵ achieved by introducing non-metallic ions at a dosage of 1 · 10⁻⁵ 13 / cm 2 or more and 3 · 10 14 / cm 2or less at a depth of 1 µm or more and 10 µm or less, forming a contamination protection layer. DE 11 2012 002 072 T5 relates to a method for producing an epitaxial semiconductor wafer, an epitaxial semiconductor wafer, and a method for producing a solid-state imaging device. Modifications of surfaces by low-energy cluster ion beams are described in YAMADA, I.: Low-energy cluster ion beam modification of surfaces. In: Nuclear Instruments and Methods in Physics Research B, Vol. 148, 1999, 1-11. SUMMARY (Technical Task)

[0007] As described in JP 2013-197373A and JP 2010-287855A, various attempts have been made to improve the quality of silicon epitaxial semiconductor wafers. In particular, several attempts have been made to improve crystallinity, for example, by reducing the surface depressions in a surface section of an epitaxial layer; however, it was assumed that an inner section of an epitaxial layer possesses sufficiently high crystallinity, so no technique for increasing crystallinity within the epitaxial layer itself has been proposed. If the crystallinity of an inner part of an epitaxial layer can be further increased, an improvement in the device's properties is to be expected. (Solution to the problem)

[0008] With regard to the above task, it could be helpful to provide a silicon epitaxial semiconductor wafer with an epitaxial layer having a higher crystallinity and a method for producing it.

[0009] To solve this problem, the inventors of the present inventions conducted various studies, focusing on ensuring that the peak hydrogen concentration profile in a surface section of a silicon epitaxial semiconductor wafer lies on the side where an epitaxial layer is formed. It is known that hydrogen diffuses here due to the heat treatment during the formation of an epitaxial layer, even when hydrogen, being a light element, is ion-implanted into a semiconductor wafer. Therefore, it was not assumed that hydrogen contributes to improving the device quality of a semiconductor device manufactured using a silicon epitaxial semiconductor wafer.Even if the hydrogen concentration of a silicon epitaxial semiconductor wafer obtained by performing hydrogen ion implantation on a semiconductor wafer under typical conditions, followed by the formation of an epitaxial layer on a surface of the semiconductor wafer, were actually measured, the measured hydrogen concentration was lower than the detection limit of secondary ion mass spectrometry (SIMS), and the effect of hydrogen was unknown. To date, no known literature exists on the concentration peak of hydrogen occurring in an amount exceeding the detection limit of SIMS in a surface region of a semiconductor wafer on the side where an epitaxial layer is formed, nor on the behavior of the hydrogen.However, the results of the experiments carried out by the inventors showed that the crystallinity of an epitaxial layer of the silicon epitaxial semiconductor wafer was clearly improved when the peak value of the hydrogen concentration profile in a surface section of the semiconductor wafer was located on the side where the epitaxial layer was formed. Furthermore, the inventors found that hydrogen in the surface section of the semiconductor wafer contributes to improving the crystallinity of the epitaxial layer. In doing so, they realized the present invention. The inventors also developed a preferred method for producing such a silicon epitaxial semiconductor wafer.

[0010] Specifically, we propose the following features.

[0011] A silicon epitaxial semiconductor wafer according to this disclosure is a silicon epitaxial semiconductor wafer (200) in which a silicon epitaxial layer (20) is formed on a surface of a silicon wafer (10), wherein the silicon wafer (10) has a modification layer (18) which contains carbon as a solid solution in the surface section, and the half-width of the peak of a carbon concentration profile of the modification layer in the direction of the thickness of the semiconductor wafer is 100 nm or less, and wherein a peak value of a hydrogen concentration profile detected by SIMS in a surface section of the silicon wafer (10) on a side on which the epitaxial layer is formed, 1.0·10 17 atoms / cm² 3 or more and 1.0·10 22 atoms / cm² 3 or less.

[0012] The peak value of the hydrogen concentration profile is preferably located at a position in a depth range of 150 nm in the direction of the thickness from the surface of the semiconductor wafer.

[0013] Here, the peak value of the carbon concentration profile is even more preferably located at a position in a depth range of 150 nm in the direction of the thickness from the surface of the silicon wafer.

[0014] Method for producing the silicon-silicon epitaxial semiconductor wafer (200) according to claim 1, comprising: a first step of irradiating a surface of a silicon wafer (10) with cluster ions (16) containing hydrogen and carbon as constituent elements; and a second step of generating a silicon epitaxial layer (20) on the surface of the silicon wafer (10) after the first step, where in the first step the beam current value of the cluster ions is 50 µA or more and 5000 µA or less. Silicon epitaxial semiconductor wafer

[0015] In this case, the beam current value in the first step is preferably 5000 µA or less.

[0016] Furthermore, the cluster ions preferably contain carbon as a constituent element.

[0017] In this case, the semiconductor wafer is preferably a silicon wafer.

[0018] Furthermore, in a method for manufacturing a solid-state image sensor device according to this disclosure, a solid-state image sensor device is produced on the epitaxial layer of one of the aforementioned epitaxial wafers or of the epitaxial wafer produced by one of the aforementioned manufacturing methods. (Beneficial effect)

[0019] A silicon epitaxial semiconductor wafer with a higher crystallinity epitaxial layer can be created. This is achieved because the peak value of the hydrogen concentration profile, as detected by SIMS, lies on the side of the semiconductor wafer where the epitaxial layer is formed. Likewise, a method for fabricating a silicon epitaxial semiconductor wafer with a higher crystallinity epitaxial layer can be provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings show: Fig. 1 a schematic sectional view showing a silicon epiaxial semiconductor wafer 100 according to one of the disclosed embodiments; Fig. 2 a schematic sectional view showing a silicon epiaxial semiconductor wafer 200 according to a preferred embodiment; Fig. 3 a schematic sectional view showing a method for producing a silicon epiaxial semiconductor wafer 200 according to one of the disclosed embodiments; Fig. 4A a schematic view showing the irradiation mechanism for cluster ion irradiation; Fig. 4B a schematic view showing the implantation mechanism for implanting a monomerion; Fig. 5A a graph showing the concentration profiles of carbon and hydrogen in a silicon wafer irradiated with cluster ions according to reference example 1; Fig. 5B a TEM cross-sectional view of a surface section of the silicon wafer according to reference example 1; Fig. 5C a TEM cross-sectional view of a surface section of a silicon wafer according to reference example 2; Fig. 6A a graph showing the concentration profiles of carbon and hydrogen in an epitaxial silicon wafer according to Example 1-1 after the production of an epitaxial layer; Fig. 6B the concentration profile of hydrogen in an epitaxial silicon wafer according to comparison example 1-1; Fig. 7 a graph showing the TO line intensity (transverse optical line intensity) of a silicon epitaxial semiconductor wafer according to Example 1-1 and the conventional Example 1-1; Fig. 8 a graph showing the concentration profiles of carbon and hydrogen in an epitaxial silicon wafer according to Example 2-1; and Fig. 9 a graph showing the TO line intensity of the silicon epitaxial semiconductor wafer according to Example 2-1 and the conventional Example 2-1. DETAILED DESCRIPTION

[0021] Now, the embodiments are described in detail with reference to the drawings. Generally, identical components are designated by the same reference numerals, and the description is not repeated. Fig. 1, Fig. 2 to Fig. Figure 3 shows a semiconductor wafer 10, a modification layer 18 and an epitaxial layer 20, enlarged to simplify the drawings with respect to thickness; therefore, the thickness ratio does not correspond to the actual ratio. (Silicon epitaxial semiconductor wafer)

[0022] With regard to a silicon epitaxial semiconductor wafer 100 according to one of the disclosed embodiments, in the case of a silicon epitaxial semiconductor wafer in which an epitaxial layer 20 is formed on a surface 10A of the semiconductor wafer 10 as in Fig. Figure 1 shows the peak value of the hydrogen concentration profile detected by SIMS in a surface section of the semiconductor wafer 10 on the side where the epitaxial layer 20 is formed. The epitaxial layer 20 is used as a device layer for fabricating a semiconductor device such as a back-illuminated solid-state image sensor device. Its features will now be described in detail.

[0023] The semiconductor wafer 10, for example, is a single-crystal bulk wafer made of silicon or a composite semiconductor (GaAs, GaN, or SiC) in which the surface 10A lacks an epitaxial layer. A single-crystal bulk silicon wafer is typically used in the fabrication of a back-illuminated solid-state imaging sensor device. A silicon wafer can be produced by growing a single-crystal silicon ingot using the Czochralski process (CZ process) or the flow zone melting process (FZ process) and cutting it with a wire saw or the like. It should be noted that a semiconductor wafer 10 to which carbon and / or nitrogen have been added can be used to achieve getter capability.Alternatively, a given dopant can be added in a predetermined concentration, and the resulting semiconductor wafer 10, which is a substrate of the so-called n+ type or p+ type or n- type or p- type, can be used.

[0024] As an example of the epitaxial layer 20, a silicon epitaxial layer can be cited, and the silicon epitaxial layer can be produced under typical conditions. For example, a source gas such as dichlorosilane or trichlorosilane can be introduced into a chamber using hydrogen as the carrier gas, and the source material can be epitaxially grown on the silicon wafer 10 by CVD at a temperature in the range of approximately 1000 °C to 1200 °C, although the growth temperature also depends on the source gas used. The epitaxial layer 20 preferably has a thickness in the range of 1 µm to 15 µm. If the thickness is less than 1 µm, the resistivity of the epitaxial layer 20 changes due to diffusion of dopants from the semiconductor wafer 10, whereas a thickness greater than 15 µm would impair the spectral sensitivity characteristics of the solid-state imaging sensor device.

[0025] One of the unique features of the silicon epitaxial semiconductor wafer 100 is that the peak value of the hydrogen concentration profile detected by SIMS lies in a surface section of the semiconductor wafer 10 on the side where the epitaxial layer 20 is formed. Considering the detection technology using SIMS at this point, the lower limit for detecting the hydrogen concentration by SIMS is 7.0 × 10 16 atoms / cm² 3 The technical significance of using such a feature is described along with its application and effects.

[0026] It has not previously been assumed that ion implantation of hydrogen into a silicon epitaxial semiconductor wafer, such that the hydrogen can be localized in a high concentration within the semiconductor wafer, contributes to improving the semiconductor properties of the device. Under typical conditions for ion implantation of hydrogen into a semiconductor wafer, the hydrogen diffuses outwards after the formation of the epitaxial layer, leaving little hydrogen in the semiconductor wafer, since hydrogen is a light element that diffuses outwards due to the heat generated by the formation of an epitaxial layer.Even when the hydrogen concentration profile of a silicon epitaxial semiconductor wafer, subjected to typical conditions for hydrogen ion implantation, is actually analyzed by SIMS, the hydrogen concentration after the formation of an epitaxial layer is lower than the detection limit. According to the results of experiments carried out by the inventors under specified requirements (the experimental conditions are described in detail below with examples), a region of high hydrogen concentration can be generated in a surface section of a semiconductor wafer on the side where an epitaxial layer is formed, and the inventors focused on the behavior of the hydrogen in this case. This revealed the following facts.

[0027] Although the details are described below in connection with the examples, the inventors observed, using cathode luminescence spectroscopy (CL spectroscopy), the difference between the crystallinity of an epitaxial layer of a silicon epitaxial semiconductor wafer 100 with a peak hydrogen concentration profile and the crystallinity of an epitaxial layer of a conventional silicon epitaxial semiconductor wafer without a peak hydrogen concentration profile. It should be noted that CL spectroscopy is a technique for measuring crystal defects in which a sample is irradiated with electron beams to detect excited light generated due to a transition from the vicinity of the base of the conduction band to the vicinity of the upper end of the valence band. Fig. Figure 7 is a graph showing the transverse optical line (TO) intensities in the direction of thickness of the disclosed silicon epitaxial semiconductor wafer 100 and a conventional silicon epitaxial semiconductor wafer, where a depth of 0 µm corresponds to the surface of the epitaxial layer and a depth of 7.8 µm to the interface between the epitaxial layer and the semiconductor wafer. It is noted that the TO line (the transverse optical line) refers to a spectrum specific to the element Si, corresponding to the band gap of Si observed by crystalline silica (CL) spectroscopy. A higher intensity of the TO line indicates a higher crystallinity of Si.

[0028] As in Fig. As shown in Figure 7, details of which are described below, the disclosed silicon epitaxial semiconductor wafer 100 has the peak TO line intensity in the epitaxial layer 20 on the side near the semiconductor wafer 10. On the other hand, the conventional silicon epitaxial semiconductor wafer has a TO line intensity that tends to decrease gradually from the interface between the semiconductor wafer and the epitaxial layer to the surface of the epitaxial layer. It should be noted that the value at the surface of the epitaxial layer (at a depth of 0 µm) is considered a special case, which is due to the influence of the surface plane on the outermost surface.Assuming that a device is fabricated using the silicon epitaxial semiconductor wafer 100, the inventors next observed the TO line intensity in a case where a heat treatment simulating the fabrication of the device was performed on the silicon epitaxial semiconductor wafer 100. As shown in . Fig. As shown in Figure 9, details of which are described below, it was experimentally demonstrated that the epitaxial layer 20 of the disclosed silicon epitaxial semiconductor wafer 100 maintains the peak value of the TO line intensity and simultaneously exhibits approximately the same level of TO line intensity in regions other than the peak value as the epitaxial layer of the conventional silicon epitaxial semiconductor wafer. This means that it was found that, taking into account all factors involved, the silicon epitaxial semiconductor wafer 100 with the disclosed peak value of the hydrogen concentration profile exhibited an epitaxial layer 20 with a higher crystallinity than the conventional one.

[0029] Although the theoretical background of this phenomenon remains unclear and this revelation is not tied to any theory, the inventors argue as follows. The details are described below; however, it shows Fig. 6 the hydrogen concentration profile of the silicon epitaxial semiconductor wafer 100 immediately after the formation of the epitaxial layer, whereas Fig. Figure 8 is a graph showing the hydrogen concentration profile of the silicon epiaxial semiconductor wafer 100 after performing a heat treatment simulating the fabrication of the device. A comparison between the peak hydrogen concentration values ​​according to Fig. 6 and Fig. Figure 8 shows that the peak hydrogen concentration is reduced by performing the heat treatment simulating the fabrication of the device. Considering the change trends in hydrogen concentration and TO line intensity before and after the simulated heat treatment, it is assumed that point defects in the epitaxial layer 20 are passivated by hydrogen, which is present in high concentration in a surface section of the semiconductor wafer 10 due to the heat treatment simulating the fabrication of the device, thereby increasing the crystallinity of the epitaxial layer 20.

[0030] As described above, the silicon epitaxial semiconductor wafer 100 according to this embodiment has the epitaxial layer 20 with higher crystallinity. The silicon epitaxial semiconductor wafer 100 provided with the epitaxial layer 20 can be used to improve the properties of a semiconductor device for which the wafer is used.

[0031] It is noted that the functionality and mode of operation described above can be achieved if the peak value of the hydrogen concentration profile in the thickness direction from the surface 10A of the semiconductor wafer 10 is located at a depth of 150 nm. Accordingly, the section corresponding to the above position can be defined as the surface section of the disclosed semiconductor wafer. The functionality and mode of operation described above can be better ensured if the peak value of the hydrogen concentration profile in the thickness direction from the surface 10A of the semiconductor wafer 10 is located at a depth of 100 nm.It should be noted that the peak value must be located at a position in a depth range of at least 5 nm or more, as it is physically impossible for the peak value of the hydrogen concentration profile to be located on the outermost surface (at a depth of 0 nm) of the wafer.

[0032] Furthermore, to ensure the aforementioned functionality and mode of operation, the peak concentration of the hydrogen concentration profile is preferably 1.0 · 10 17 atoms / cm² 3 or more, in particular preferably 1.0 · 10 18 atoms / cm² 3 or more. Although there is no intention to limit the invention, the upper limit of the peak hydrogen concentration with regard to the industrial production of the silicon epitaxial semiconductor wafer 100 1.0 · 10 22 atoms / cm² 3 be.

[0033] In this case, a preferred silicon epiaxial semiconductor wafer 200 according to this disclosure has a modification layer 18 which contains carbon as a solid solution in the surface section, as shown in Fig. Figure 2 shows; and the full width at half maximum (FWHM) of the peak of the carbon concentration profile of the modification layer 18 in the direction of the thickness of the semiconductor wafer 10 is preferably 100 nm or less. The modification layer 18 is a region in which carbon is localized as a solid solution at interstitial or substitutional positions in the crystal lattice of the surface section of the semiconductor wafer, the region serving as a strong getter site. Furthermore, with regard to achieving higher getter capability, the FWHM is preferably 85 nm or less, and its lower limit can be set to 10 nm.

[0034] “The carbon concentration profile in the direction of the thickness” here refers to the concentration profile measured by SIMS in the direction of the thickness.

[0035] Furthermore, with regard to achieving a higher getter capability, in addition to the hydrogen and carbon mentioned above, elements other than the main material of the semiconductor wafer (silicon when using a silicon wafer) preferably form the solid solution in the modification layer 18.

[0036] Furthermore, with regard to achieving a higher getter capability, the peak value of the carbon concentration profile of the silicon epitaxial semiconductor wafer 200 is located in the thickness direction preferably at a position at a depth of 150 nm or less from the surface 10A of the semiconductor wafer 10. The peak concentration of the carbon concentration profile is preferably 1 × 10 15 atoms / cm² 3or more, preferably in the range of 1 × 10 17 atoms / cm² 3 up to 1 · 10 22 atoms / cm² 3 , even more preferably in the range of 1 · 10 19 up to 1 · 10 21 atoms / cm² 3 .

[0037] It should be noted that the thickness of modification layer 18 is defined such that the carbon concentration lies within the concentration profile above, but is higher than the background. The thickness can, for example, range from 30 nm to 400 nm. (Method for the production of a silicon epitaxial semiconductor wafer)

[0038] Next, an embodiment of a method for producing the silicon epitaxial semiconductor wafer 200 disclosed above is described. A method for producing the silicon epitaxial semiconductor wafer 200 according to the above embodiment comprises a first step of irradiating the surface 10A of the semiconductor wafer 10 with cluster ions 16 containing hydrogen as a constituent element (step 3A and step 3B in Fig. 3); and a second step to generate an epitaxial layer 20 on the surface 10A of the semiconductor wafer 10 after the first step (step 3C according to Fig. 3), as in Fig. Figure 3 shows. Furthermore, in the first step, the beam current value of the cluster ions is 16-50 µA or more. Step 3C according to Fig. Figure 3 is a schematic sectional view of the silicon epiaxial semiconductor wafer 200 obtained by this manufacturing process. The steps will now be described in detail one after the other.

[0039] First, a semiconductor wafer 10 is produced. Next, as described in steps 3A and 3B, Fig. Figure 3 shows the first step of irradiating the surface 10A of the semiconductor wafer 10 with cluster ions 16 containing hydrogen as a constituent. It is essential that the beam current of the cluster ions 16 in the first step is 50 µA or higher to ensure that the peak value of the hydrogen concentration profile detected by SIMS in the surface section of the semiconductor wafer 10 lies on the side of the epitaxial layer 20. As a result of the irradiation with the hydrogen-containing cluster ions 16 under the aforementioned current conditions, hydrogen contained in the constituents of the cluster ions is localized as a solid solution in the surface section of the semiconductor wafer 10 on the side of surface 10A (i.e., in the irradiated plane) at a concentration exceeding the equilibrium concentration.

[0040] It should be noted that "cluster ions" here refer to clusters formed by the aggregation of several atoms or molecules, which are ionized by becoming positively or negatively charged. A cluster is a mass aggregate comprising several (typically 2 to 2,000) interconnected atoms or molecules.

[0041] The difference in the behavior of the solid solution between the case of irradiation of the semiconductor wafer 10 with cluster ions and the case of implantation of monomer ions is described in detail below. This means that, for example, when monomer ions consisting of certain elements are implanted into a silicon wafer (the semiconductor wafer), the monomer ions sputter silicon atoms into the silicon wafer, which are implanted at a position and a predetermined depth within the silicon wafer, as shown in Fig. Figure 4B shows that the implantation depth depends on the type of constituent elements of the implanted ions and the ion acceleration voltage. Accordingly, the concentration profile of the specific elements is relatively broad in the direction of depth of the silicon wafer, and the region in which the specific implanted elements are present ranges from approximately 0.5 µm to 1 µm from the surface. If implantation is performed simultaneously with several ion species at the same energy, lighter elements are implanted deeper; in other words, elements are implanted at different positions depending on their mass. Consequently, the concentration profile of the implanted elements is broader in such a case. Furthermore, when an epitaxial layer is created after ion implantation, the implanted elements diffuse due to heat, which also contributes to the broader concentration profile.

[0042] Monomer ions are typically implanted at an accelerating voltage of approximately 150 keV to 2000 keV. However, the ions collide with silicon atoms at this energy level, leading to damage to the crystallinity of the surface region of the silicon wafer into which the monomer ions are implanted. Consequently, the crystallinity of an epitaxial layer, which is subsequently to be grown on the wafer surface, tends to deteriorate. Furthermore, the tendency for crystallinity deterioration is greater the higher the accelerating voltage.

[0043] On the other hand, cluster ions 16 are produced when a silicon wafer is irradiated with cluster ions, as in Fig. As shown in Figure 4A, the silicon is instantly brought to a high-temperature state of approximately 1350 °C to 1400 °C by the irradiation energy, causing it to melt. The silicon is then rapidly cooled to create a solid solution of the constituent elements of the cluster ions 16 near the surface of the silicon wafer. The concentration profile of the constituent elements is sharper in the direction of depth of the silicon wafer compared to the case using monomer ions, although it depends on the accelerating voltage and the cluster size of the cluster ions. The region in which the constituent elements used for irradiation are present is approximately 500 nm or less (for example, approximately 50 nm to 400 nm).Furthermore, unlike monomer ions, cluster ions are not transported through the crystal lattice because the ions used for irradiation form clusters, and the thermal diffusion of the constituent elements is suppressed, which also leads to the sharp concentration profile. Accordingly, the constituent elements of cluster ions 16 are deposited in high concentration in a localized area.

[0044] In this process, hydrogen ions diffuse readily, as described above, for example, due to the heat treatment used to create the epitaxial layer 20, since hydrogen is a light element, and tend to remain in the semiconductor wafer only minimally after the epitaxial layer is created. Therefore, it is insufficient to irradiate only the area where hydrogen precipitates locally and heavily with cluster ion beams. To suppress hydrogen diffusion during heat treatment, it is essential to set the beam current of the cluster ions 16 to 50 µA or more so that the surface 10A of the semiconductor wafer 10 is irradiated with hydrogen ions for a relatively short time, thereby increasing the damage to the surface area.A beam current of 50 µA or more can increase damage, causing the peak hydrogen concentration profile detected by SIMS to be located on the epitaxial layer 20 side of the surface section of the semiconductor wafer 10, even after the epitaxial layer 20 has been subsequently formed. If the beam current is lower than 50 µA, the damage to the surface section of the semiconductor wafer 10 is insufficient, and the hydrogen diffuses away due to the heat treatment used to create the epitaxial layer 20. The beam current of the cluster ions 16 can be adjusted, for example, by changing the conditions for the degradation of the source gas in the ion source.

[0045] Following the first step described above, the second step involves the production of the epitaxial layer 20 on the surface 10A of the semiconductor wafer 10. The epitaxial layer 20 described above is produced in the second step.

[0046] This allows the process for manufacturing the silicon epitaxial semiconductor wafer 200 to be created.

[0047] It is noted that the beam current value of the cluster ions 16, even after the formation of the epitaxial layer 20, is preferably 100 µA or more, and more preferably 300 µA or more, to ensure that the peak value of the hydrogen concentration profile detected by SIMS lies in the surface section of the semiconductor wafer 10.

[0048] If the beam current value is excessively high, excessive epitaxial defects will be generated in the epitaxial layer 20. Therefore, the beam current value is preferably 5000 µA or less.

[0049] The conditions for irradiation with the cluster ions 16 are now described. First, the constituents of the cluster ions 16 used for irradiation, other than hydrogen, are not specifically limited; they can include, for example, carbon, boron, phosphorus, arsenic, and / or the like. However, with regard to achieving higher getter capability, the cluster ions 16 preferably include carbon as a constituent. This leads to the formation of the modification layer 18 with a solid solution of carbon. Carbon atoms at a lattice site have a smaller covalent radius than silicon single crystals, and for this reason, a compression site is created in the silicon crystal lattice, resulting in a getter site that attracts impurities in the lattice.

[0050] Furthermore, the elements used for irradiation preferably include elements other than hydrogen and carbon. In particular, the irradiation is preferably carried out using one or more dopants selected from the group consisting of boron, phosphorus, arsenic, and antimony, in addition to hydrogen and carbon. Since the type of metals that can be effectively gettered depends on the types of solid solution elements, solid solutions containing a variety of element species can cover a wider range of metal impurities. For example, carbon can effectively getter nickel (Ni), whereas boron can effectively getter copper (Cu) and iron (Fe).

[0051] A source compound to be ionized is not specifically restricted; however, ethane, methane, or the like can be used as a carbon source compound that can be ionized, whereas diborane, decaborone (B 10 H 14) or the like can be used as a boron source compound that can be ionized. For example, if a mixed gas of dibenzyl and decaborane is used as the material gas, hydrogen compound clusters can be generated in which carbon, boron, and hydrogen are aggregated. Alternatively, when using cyclohexane (C6H₆), 12 Cluster ions formed from carbon and hydrogen are produced as a material gas. In particular, cluster ions are preferably generated from pyrene (C 16 H 10 ), Dibenzyl (C 14 H 14 ) or similar generated C n H m Clusters (3 ≤ n ≤ 16, 3 ≤ m ≤ 10) are used as carbon source compounds. This is because ion beams from small clusters can be easily controlled.

[0052] The cluster size can be adjusted from 2 to 100, preferably 60 or less, and more preferably 50 or less. The cluster size can be adjusted by controlling the pressure of the gas expelled from a nozzle, the pressure of a vacuum chamber, the voltage applied to the wire during ionization, and the like. The cluster size is determined by ascertaining the cluster number distribution by mass spectrometry using the oscillating quadrupole field or by time-of-flight mass spectrometry and determining the average number of clusters.

[0053] Depending on the binding mode, cluster ions can comprise a variety of clusters and can be generated, for example, by known methods described in the following publications. Methods for generating a gas cluster beam are described in JP H09-41138A and JP H04-354865A. Methods for generating an ion beam are described in ISHIKAWA, Zyunzo: Science and technology of charged particle beams. Tokyo: Corona, 2001. INSTITUTE OF ELECTRICAL ENGINEERS OF JAPAN: Electron and ion beam engineering. Tokyo: IEEJ, 1995 and YAMADA, Isao: Cluster ion beam technology: advanced nanofabrication processes. Tokyo: Nikkan Kogyo Shimbun, 2006. In general, a Nielsen ion source or a Kaufman ion source is used to generate positively charged cluster ions, whereas a volume production negative high-current ion source is used to generate negatively charged cluster ions.

[0054] The accelerating voltage of the cluster ions and the cluster size influence the position of the peak value of the concentration profile of the cluster ion constituents in the thickness direction. To ensure that the peak value of the hydrogen concentration profile lies on the epitaxial layer side of the surface section of the semiconductor wafer 10, even after the epitaxial layer has been created, the accelerating voltage of the cluster ions is set higher than 0 keV / cluster and lower than 200 keV / cluster, preferably to 100 keV / cluster or less, and more preferably to 80 keV / cluster or less. Furthermore, two methods are typically used to set the accelerating voltage: (1) electrostatic field acceleration or (2) oscillating field acceleration.Examples of the former method include a technique in which multiple electrodes are arranged at regular intervals and the same voltage is applied between them, thereby generating constant acceleration fields in the direction of the axes. Examples of the latter method include a linear acceleration technique (Linac technique) in which ions are transferred in a straight line and accelerated by radio-frequency waves.

[0055] The dose of cluster ions can be adjusted by controlling the duration of ion irradiation. In this embodiment, the hydrogen dose can be 1 × 10 13 atoms / cm² 2 up to 1 · 10 16 atoms / cm² 2 , preferably 5 · 10 13 atoms / cm² 2 or more. If the hydrogen dose is lower than 1 × 10 13 atoms / cm² 2, the hydrogen would diffuse during the formation of the epitaxial layer, whereas a dose of more than 1 · 10 16 atoms / cm² 2 would cause significant damage to the surface of the epitaxial layer 20.

[0056] Furthermore, the dose of carbon in cluster ions with carbon as a constituent element is preferably 1 · 10 13 atoms / cm² 2 up to 1 · 10 16 atoms / cm² 2 , preferred 5 · 10 13 atoms / cm² 2 or more. If the hydrogen dose is less than 1 × 10 13 atoms / cm² 2 , getter capability is insufficient, whereas a dose of over 1 · 10 16 atoms / cm² 2 would cause significant damage to the surface of the epitaxial layer 20.

[0057] It is noted that, after the first and before the second step, a regeneration heat treatment is preferably also carried out to restore the crystallinity on the semiconductor wafer 10. This regeneration heat treatment can be performed, for example, by holding the semiconductor wafer 10 in an atmosphere of nitrogen gas, argon gas, or the like at a temperature of 900 °C or more and 1100 °C or less for 10 minutes or more up to 60 minutes or less. Alternatively, the regeneration heat treatment can be carried out, for example, outside the epitaxial device using a rapid heating / cooling device for rapid thermal oxidation (RTA), rapid thermal oxidation (RTO), or the like.

[0058] The semiconductor wafer 10 can be a silicon wafer, as described above.

[0059] One embodiment of the method for producing the silicon epitaxial semiconductor wafer 200 has been described in which the peak value of the hydrogen concentration profile detected by SIMS, even after the formation of the epitaxial layer 20, lies on the side of the semiconductor wafer 10 on which the epitaxial layer 20 is formed. However, the disclosed silicon epitaxial semiconductor wafer can of course be produced by other manufacturing methods. (Method for manufacturing a solid-state image sensor device)

[0060] In a method for manufacturing a solid-state image sensor device according to one embodiment, a solid-state image sensor device can be produced on the epitaxial layer 20 arranged in the surface section of the silicon epitaxial semiconductor wafer described above, or on a silicon epitaxial semiconductor wafer produced by the manufacturing process described above, i.e., the silicon epitaxial semiconductor wafer 100, 200. In a solid-state image sensor device obtained by this manufacturing process, white dot defects can be sufficiently suppressed compared to a conventional one.

[0061] This revelation is described in more detail below with the help of examples. However, this revelation is not limited to the following examples. EXAMPLES (Examples of reference experiments)

[0062] The following experiments were first conducted to clarify the difference between the damage to the surface section of each silicon wafer at different beam flux values ​​of cluster ions. (Reference example 1)

[0063] A p-type silicon wafer (diameter: 300 mm, thickness: 775 µm, dopant: boron, resistivity: 20 Ω·cm) derived from a CZ single crystal was fabricated. Subsequently, the surface of the silicon wafer was treated by generating cluster ions from cyclohexane (C6H₆). 12Using a cluster ion generator (CLARIS, manufactured by Nissin Ion Equipment Co., Ltd.), C3H5 cluster ions obtained under irradiation conditions of an accelerating voltage of 80 keV / cluster (accelerating voltage per hydrogen atom: 1.95 keV / atom, accelerating voltage per carbon atom: 23.4 keV / atom, range distance of hydrogen: 40 nm, range distance of carbon: 80 nm) were irradiated to produce a silicon wafer according to Reference Example 1. It should be noted that the dose of cluster ion irradiation was 1.6 × 10 15 atoms / cm² 2 was calculated as the number of hydrogen atoms, and 1.0 · 10 15 atoms / cm² 2 , calculated as the number of carbon atoms. The beam current of the cluster ions was 800 µA. (Reference example 2)

[0064] A silicon wafer according to reference example 2 was produced under the same conditions as in reference example 1, except that the beam current value of the cluster ions was changed to 30 µA. (Concentration profile of the silicon wafer)

[0065] SIMS magnetic field sampling was performed on silicon wafers according to reference examples 1 and 2, which were irradiated with cluster ions to determine the hydrogen and carbon concentration profiles along the wafer thickness. The carbon concentration profile of reference example 1 is shown in Fig. 5A is shown as an illustrative example. The same concentration profile as according to Fig. 5A was also obtained in reference example 2, where only the beam current value was different. This corresponds to... Fig. 5A the surface of the silicon wafer on the side of cluster ion irradiation at a depth of 0 on the horizontal axis. (TEM section view)

[0066] The cross-section of the surface section of the silicon wafer with the cluster-ion irradiated area of ​​each of the silicon wafers according to Reference Examples 1 and 2 was examined using a transmission electron microscope (TEM). The TEM section images of the silicon wafers according to Reference Examples 1 and 2 are shown in the respective Fig. 5B and Fig. 5C shown. The positions in the in Fig. Areas 5B enclosed in bold rectangles, where a black contrast appears, are significantly damaged.

[0067] As in the Fig. As shown in Figures 5A to 5C, in Reference Example 1, where the beam current was 800 µA, significantly damaged areas were produced in the surface section of the silicon wafer, whereas in Reference Example 2, where the beam current was 30 µA, no significantly damaged areas were produced. The concentration profiles of hydrogen and carbon showed similar trends for Reference Examples 1 and 2 due to the identical dose conditions; however, whether or not significantly damaged areas were produced in the surface section of each silicon wafer was attributed to the different beam current values. It should be noted that the Fig. 5A and Fig. Figure 5B shows that significantly damaged areas were created in a range between the peak position of the hydrogen concentration and the peak position of the carbon concentration. (Experimental Examples 1)(Example 1-1)

[0068] A silicon wafer was irradiated with C3H5 cluster ions under the same conditions as in Reference Example 1. The silicon wafer was then placed in an epitaxial growth device (manufactured by Applied Materials, Inc.) for processing single wafers and subjected to hydrogen curing at 1120 °C for 30 s. Subsequently, a silicon epitaxial layer (thickness: 7.8 µm, dopant: boron, resistivity: 10 Ω·cm) was epitaxially grown on one surface of the silicon wafer by CVD at 1150 °C using hydrogen as the carrier gas and trichlorosilane as the source gas, thereby producing an epitaxial silicon wafer according to Example 1-1. (Comparison example 1-1)

[0069] The epitaxial wafer according to comparison example 1-1 was produced under the same conditions as according to example 1-1, except that the beam current value of the cluster ions was changed to 30 µA. (Conventional Example 1-1)

[0070] An epitaxial wafer according to the conventional Example 1-1 was produced under the same conditions as Example 1-1, except that irradiation with cluster ions was not carried out. (Rating 1-1: Evaluation of the concentration profile of the epitaxial wafer using SIMS)

[0071] A SIMS measurement of the magnetic region was performed on the silicon wafers according to Example 1-1 and Comparative Example 1-1, which were irradiated with cluster ions, in order to determine the hydrogen and carbon concentration profiles in the direction of the wafer thickness. The hydrogen and carbon concentration profiles for Example 1-1 are shown in Fig. 6A is shown. Furthermore, the hydrogen concentration profile of comparison example 1-1 is shown in Fig. shown in 6B. Here, the surface of the epitaxial layer corresponds to each of the Fig. 6A and Fig. 6B represents a depth of 0 on the horizontal axis. Depths up to 7.8 µm correspond to the epitaxial layer, whereas depths of 7.8 µm or more correspond to the silicon wafer. When the epitaxial wafers were subjected to SIMS measurement, an unavoidable measurement error of ±0.1 µm was found with respect to the thickness of the epitaxial layer. Accordingly, 7.8 µm in the diagram may not be the exact boundary between the epitaxial layer and the silicon wafer. (Rating of 1-2: Evaluation of TO line intensity by CL spectroscopy)

[0072] Samples processed by chamfering the epitaxial wafers of Example 1-1, Comparative Example 1-1, and Conventional Example 1-1 by polishing were subjected to chromatin-resonance (CL) spectroscopy from the cross-sectional direction, thus obtaining the CL spectrum of each epitaxial layer in the direction of thickness (depth). Irradiation with an electron beam at 20 keV was performed under a measurement condition of 33 K. The measurement results of the CL intensities in the direction of thickness for Example 1-1 and Conventional Example 1-1 are shown in Fig. Figure 7 is shown. It should be noted that the measurement results for comparison example 1-1 were identical to those of the conventional example 1-1.

[0073] As with reference to Fig. As described in 5A, the peak hydrogen concentration after cluster ion irradiation, but before the formation of an epitaxial layer, was located in the surface section of the silicon wafer, regardless of the beam current (see Reference Examples 1 and 2 of the reference experiments). The results of Reference Example 1 and Example 1-1, where the beam current was 800 µA, show that the peak hydrogen concentration before the formation of the epitaxial layer was approximately 7 × 10⁻⁶. 20 atoms / cm² 3 was and that the peak concentration of hydrogen after the formation of the epitaxial layer was approximately 2 · 10 18 atoms / cm² 3 declined ( Fig. 5A and Fig. 6A). On the other hand, if the beam current value was 30 µA, the peak hydrogen concentration occurred before the formation of the epitaxial layer; however, the peak hydrogen concentration occurred after the formation of the epitaxial layer ( Fig. 6B) not. At a beam current of 800 µA, the hydrogen would have remained without completely diffusing through the heat treatment to create the epitaxial layer, as the surface section of the silicon wafer was severely damaged. This can also be considered a phenomenon by which hydrogen in the Fig. The damaged area shown in 5B was captured.

[0074] As in Fig. As shown in Figure 7, in Example 1-1 the peak value of the TO line intensity is located at a depth of approximately 7 µm from the surface of the epitaxial layer. On the other hand, in the epitaxial wafer according to the conventional Example 1-1, the TO line intensity gradually decreases from the boundary of the silicon wafer to the surface of the epitaxial layer. It should be noted that the value of the intensity at the surface of the epitaxial layer (depth: 0 µm), which is a surface, is assumed to be influenced by the surface plane. (Experimental Examples 2)(Example 2-1)

[0075] An epitaxial wafer produced according to Example 1-1 was subjected to a heat treatment simulating the manufacture of the device for 30 minutes at a temperature of 1100 °C. (Conventional Example 2-1)

[0076] As in Example 2-1, an epitaxial wafer manufactured according to conventional Example 1-1 was subjected to a heat treatment simulating the manufacture of the device at a temperature of 1100 °C for 30 minutes. (Rating 2-1: Evaluation of the concentration profile of the epitaxial wafer using SIMS)

[0077] As in assessment 1-1, a SIMS measurement of the magnetic range was performed on the cluster-ion irradiated silicon wafer according to Example 2-1 to determine the hydrogen and carbon concentration profiles along the wafer thickness. The hydrogen and carbon concentration profiles for Example 2-1 are shown in Fig. 8 shown. Here, the surface corresponds to the epitaxial layer, as shown in Figure 8 ... Fig. 6A, a depth of 0 on the horizontal axis. (Grade 2-2: Assessment of TO line intensity using CL spectroscopy)

[0078] As in evaluation 1-2, the CL spectra of the epitaxial wafers were determined according to Example 2-1 and the conventional Example 2-1. The results are presented in Fig. 9 shown.

[0079] Become Fig. 6A and Fig. Compared to example 8, the peak concentration of hydrogen in example 1-1 was approximately 2 · 10 18 atoms / cm² 3 , whereas the peak concentration of hydrogen in Example 2-1 is approximately 3 · 10 17 atoms / cm² 3 decreased. Furthermore, it shows Fig. 9, that in Example 2-1 the same level of TO line intensity was observed in other areas as in the conventional Example 2-1, while the peak value of the TO line intensity was at a position at a depth of approximately 7 µm from the surface of the epitaxial layer (at the same position as the peak value in Fig.7) is held. Accordingly, it can be assumed that an epitaxial wafer that meets the conditions according to this disclosure, taking into account all factors involved, has an epitaxial layer with a higher crystallinity than a conventional one.

[0080] It is assumed that the reason for the change in TO line intensity is that the hydrogen in the epitaxial layer passivated point defects contained in the epitaxial wafer where hydrogen was observed after epitaxial growth. Since, on the other hand, no peak hydrogen concentration was observed in comparison example 1-1, where the beam current was 30 µA, the passivation effect of hydrogen was apparently not achieved in comparison example 1-1. INDUSTRIAL APPLICABILITY

[0081] A silicon epitaxial semiconductor wafer with an epitaxial layer exhibiting higher crystallinity and a method for its fabrication can be created. The created silicon epitaxial semiconductor wafer with such an epitaxial layer can be used to improve the device properties of a semiconductor device fabricated using the wafer. LIST OF REFERENCE MARKS 10 semiconductor wafers 10A Surface of the semiconductor wafer 16 clusterions 18 Modification layer 20 Epitaxial layer 100 silicon epitaxial semiconductor wafers 200 silicon epitaxial semiconductor wafers

Claims

[1] Silicon epitaxial wafer (200) in which a silicon epitaxial layer (20) is formed on a surface of a silicon wafer (10), wherein the silicon wafer (10) has a modification layer (18) which contains carbon as a solid solution in the surface section, and the half-width of the peak of a carbon concentration profile of the modification layer in the direction of the thickness of the semiconductor wafer is 100 nm or less, and wherein a peak value of a hydrogen concentration profile detected by SIMS in a surface section of the silicon wafer (10) on a side on which the epitaxial layer is formed, 1.0·10 17 atoms / cm² 3 or more and 1.0·10 22 atoms / cm² 3 or less. [2] Silicon epitaxial wafer according to claim 1, wherein the peak value of the hydrogen concentration profile in the direction of the thickness is located at a position at a depth of 150 nm from the surface of the silicon wafer. [3] Silicon epitaxial wafer according to claim 1 or 2, wherein the peak value of the carbon concentration profile in the direction of the thickness is located at a position at a depth of 150 nm from the surface of the silicon wafer (10). [4] Method for producing the silicon epitaxial wafer (200) according to claim 1, comprising: a first step of irradiating a surface of a silicon wafer (10) with cluster ions (16) containing hydrogen and carbon as constituent elements; and a second step of generating a silicon epitaxial layer (20) on the surface of the silicon wafer (10) after the first step, where in the first step the beam current value of the cluster ions is 50 µA or more and 5000 µA or less. [5] Method for manufacturing a solid-state image sensor device, wherein a solid-state image sensor device is produced on the silicon epitaxial layer (20) of the silicon epitaxial wafer (200) according to one of claims 1 to 3 or the silicon epitaxial wafer (200) manufactured according to the manufacturing method according to claim 4.

Images