Polycrystalline silicon rod and method for producing polycrystalline silicon rod

Abstract

To provide polycrystalline silicon suitable as a raw material for production of single-crystalline silicon. A D/L value is set within the range of less than 0.40 when multiple pairs of silicon cores are placed in a reaction furnace in production of a polycrystalline silicon rod having a diameter of 150 mm or more by deposition according to a chemical vapor deposition process and it is assumed that the average value of the final diameter of the polycrystalline silicon rod is defined as D (mm) and the mutual interval between the multiple pairs of silicon cores is defined as L (mm).

Claims

1. A method for producing a polycrystalline silicon rod having a diameter of 150 mm or more by a deposition according to a chemical vapor deposition process, wherein a D/L value is set within a range of less than 0.40 when at least two pairs of silicon cores are placed in a reaction furnace and wherein an average value of a final diameter of the polycrystalline silicon rod is defined as D (mm) and a mutual interval between the at least two pairs of silicon cores is defined as L (mm).

2. The method for producing a polycrystalline silicon rod according to claim 1, wherein a reaction pressure in the deposition according to the chemical vapor deposition process of the polycrystalline silicon is set to 0.2 MPa or more.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1A illustrates a view for description of a collection example of a plate-like sample for X-ray diffraction profile measurement, from a polycrystallinee silicon rod deposited and grown according to a chemical vapor deposition process;

(2) FIG. 1B illustrates a view for description of a collection example of a plate-like sample for X-ray diffraction profile measurement, from a polycrystalline silicon rod deposited and grown according to a chemical vapor deposition process;

(3) FIG. 2 illustrates a view for description of an outline of a measurement system example in determination of an X-ray diffraction profile from a plate-like sample according to a φ scanning method;

(4) FIG. 3A represents an X-ray diffraction chart obtained from a <220> plane of a plate-like sample collected from a polycrystalline silicon rod grown under a condition of a D/L value of 0.69;

(5) FIG. 3B represents an X-ray diffraction chart obtained from a <220> plane of a plate-like sample collected from a polycrystalline silicon rod grown under a condition of a D/L value of 0.36;

(6) FIG. 4A represents the result of optical microscope observation of the surface of the plate-like sample represented in FIG. 3A, etched by a mixed liquid of hydrofluoric acid and nitric acid; and

(7) FIG. 4B represents the result of optical microscope observation of the surface of the plate-like sample represented in FIG. 3B, etched by a mixed liquid of hydrofluoric acid and nitric acid.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(8) Hereinafter, embodiments of the present invention are described with reference to the drawings.

(9) The present inventors have proposed, in Japanese Patent Laid-Open No. 2015-3844, an invention relating to a method for selecting a polycrystalline silicon rod for use as a raw material for production of single-crystalline silicon according to an X-ray diffraction method in view of the current state where there is demanded an advanced technique for selecting polycrystalline silicon suitable as a raw material for production of single-crystalline silicon at high quantitativity and reproducibility for the purpose of stable production of single-crystalline silicon at a high yield, and the invention is registered as Japanese Patent No. 5947248.

(10) The method is to select a polycrystalline silicon rod as a raw material for production of single-crystalline silicon in the following case: a polycrystalline silicon rod grown by deposition according to a chemical vapor deposition process is used to collect a plate-like sample in which the cross section perpendicular to the radial direction corresponds to a principal plane, the plate-like sample is placed at a position where the Bragg reflection from a first Miller index plane <111> is detected, in-plane rotation is made at a rotation angle φ with the center of the plate-like sample as the center of rotation so that φ scanning of the principal plane of the plate-like sample is made with an X-ray irradiation region defined by a slit, a chart is determined which represents the dependence of the Bragg reflection intensity from the Miller index plane on the rotation angle (φ) of the plate-like sample, the diffraction intensity value (I.sub.B.sup.<111>) of a baseline is determined from the chart, furthermore the diffraction intensity value (I.sub.B.sup.<220>) of a baseline is determined from a φ scanning chart obtained from a second Miller index plane <220> in the same manner, and a magnitude relationship between the I.sub.B.sup.<111> value and the I.sub.B.sup.<220> value simultaneously satisfies the following two conditions.

(11) Such two conditions are as follows: condition 1: “the I.sub.B.sup.<111> and I.sub.B.sup.<220> obtained with respect to the plate-like sample collected at a position located in the range of R/3 or less from the center in the radial direction of the polycrystalline silicon rod having a radius R satisfy I.sub.B.sup.<111>>I.sub.B.sup.<220>”; and condition 2: “the I.sub.B.sup.<111> value and I.sub.B.sup.<220> value obtained with respect, to the plate-like sample collected at a position located in the range of 2R/3 or more and 3R/3 or less from the center in the radial direction of the polycrystallinee silicon rod having a radius R satisfy I.sub.B.sup.<111><I.sub.B.sup.<220>”.

(12) As a locally heterogeneous crystal having a Miller index plane <111> as a principal plane is more included, a diffraction peak due to a Miller index plane <111>, higher in the intensity than the intensity of a baseline, is more observed in the above φ scanning chart. Similarly, as a needle crystal and a locally heterogeneous crystal having a Miller index plane <220> as a principal plane are more included, a diffraction peak due to a Miller index plane <220>, higher in the intensity than the intensity of a baseline, is more observed in the above φ scanning chart. The presence of such diffraction peaks can be then evaluated as the variation in diffraction intensity of each of Miller index planes <111> and <220>, which can be evaluated from the φ scanning chart. Accordingly, such a variation in diffraction intensity, if can be quantitatively evaluated, can be used as an index of the degree of incorporation of each of a needle crystal and a locally heterogeneous crystal having Miller index planes <111> and <220> as principal planes.

(13) According to studies made by the present inventors, it has been found that, the variation in diffraction intensity of a Miller index plane <111> is easily generated at a site where the load due to current heating in deposition of polycrystalline silicon is increased (mainly the vicinity of a silicon core) and the variation in diffraction intensity of a Miller index plane <220> is easily generated at a site where radiation heat from an adjacent polycrystalline silicon rod is received (mainly the outside of a polycrystalline silicon rod).

(14) It has been already reported by the present inventors (Japanese Patent Laid-Open No. 2016-150885) that a locally heterogeneous crystal having a principal plane <111> is easily generated at the center portion of a polycrystalline silicon rod, and the variation in diffraction intensity of a Miller index plane <111> can be suppressed by proper control of the temperature setting of the center portion of a polycrystalline silicon rod.

(15) On the contrary, a locally heterogeneous crystal having a principal plane <220> is easily generated at a site where radiation heat from an adjacent polycrystalline silicon rod is received (mainly the outside of a polycrystalline silicon rod), and therefore radiation heat from an adjacent polycrystalline silicon rod is demanded to be considered.

(16) The present inventors have made studies about the problem of radiation heat, and thus have found that a needle crystal and a locally heterogeneous crystal having a principal plane <220> can be effectively inhibited from being generated by a D/L value which is set within the range of less than 0.40 when multiple pairs of silicon cores are placed in a reaction furnace in production of a polycrystalline silicon rod having a diameter of 150 mm or more according to a chemical vapor deposition process and it is assumed that the average value of the final diameter of a polycrystalline silicon rod is defined as D (mm) and the mutual interval between the multiple pairs of silicon cores is defined as L (mm).

(17) Specifically, the above method can provide a polycrystalline silicon rod grown by deposition according to a chemical vapor deposition process, wherein the polycrystalline silicon rod has a diameter (2R) of 150 mm or more, and, when an X-ray diffraction chart obtained by in-plane rotation with the center of a plate-like sample collected from each of the center region, the R/2 region and the outer region of the polycrystalline silicon rod, as the center of rotation, at an angle φ of 180 degrees is determined, the degree of variation in diffraction intensity from a <220> plane satisfies 0.15 or less in the center region, 0.30 or less in the R/2 region and 0.58 or less in the outer region in evaluation as a 6σ.sub.n-l/average value. Such evaluation here is made with the 6σ.sub.n-1/average value as the degree of variation in diffraction intensity of a Miller index plane <220> (the degree of variation in diffraction intensity from a <220> plane), and σ.sub.n-1 represents the standard deviation.

(18) In the present invention, the degree of crystal homogeneity is thus evaluated in terms of the degree of variation in diffraction intensity of a Miller index plane <202>. Hereinafter, the evaluation procedure is described.

(19) [Collection of Evaluation Sample]

(20) FIG. 1A and FIG. 1B each illustrate a view for description of a collection example of a plate-like sample 20 for X-ray diffraction profile measurement, from a polycrystalline silicon rod 10 deposited and grown according to a chemical vapor deposition process such as a Siemens method. In the drawings, symbol 1 represents a silicon core for use in deposition of polycrystallinee silicon on the surface and thus production of a silicon rod. In the example, the plate-like sample 20 is collected from each of three sites (CTR: a site closer to the silicon core 1, EDG: a site closer to the outer side surface of the polycrystalline silicon rod 10, R/2: site between CTR and EGD) in order to evaluate crystal homogeneity. In the example illustrated in the drawings, the plate-like sample 20 is collected by hollowing out in a direction vertical to the long axis direction of the polycrystalline silicon rod 10.

(21) The diameter of the polycrystalline silicon rod 10 illustrated in FIG. 1A is 150 mm or more, and a rod 11 having a substantial diameter of 20 mm and a substantial length of 70 mm is obtained by hollowing out from the outer side surface of the polycrystalline silicon rod 10 in a direction vertical to the longitudinal direction of the silicon core 1.

(22) As illustrated in FIG. 1B, plate-like samples (20.sub.CTR, 20.sub.EDG, 20.sub.R/2) each having a cross section perpendicular to the radial direction of the polycrystalline silicon rod 10 as a principal plane and having a substantial thickness of 2 mm are collected from the site (CTR) closer to the silicon core 1 of the rod 11, the site (EDG) closer to the side surface of the polycrystalline silicon rod 10 and the site (R/2) between CTR and EGD, respectively.

(23) The site, the length and the number of samples, with respect to collection of the rod 11, may be appropriately determined depending on the diameter of the silicon rod 10 and the diameter of the rod 11 obtained by hollowing out, and the plate-like sample 20 may also be collected from any site of the rod 11 obtained by hollowing out, but such a site preferably corresponds to one which enables properties of the entire silicon rod 10 to be reasonably presumed.

(24) A case where the diameter of the plate-like sample 20 is substantially 20 mm is merely an example, and the diameter may be properly defined as long as X-ray diffraction measurement is not impaired. In order to perform crystalline texture observation with an optical microscope, the surface of the plate-like sample 20 may be subjected to lap polishing and then etching by a mixed liquid of hydrofluoric acid and nitric acid.

(25) [X-Ray Diffraction Chart]

(26) FIG. 2 illustrates a view for schematically describing a measurement system example in determination of the X-ray diffraction profile from the plate-like sample 20 according to a φ scanning method, and, in the example illustrated in the drawing, an elongated rectangular region defined by a slit in a region over both peripheral ends of the plate-like sample 20 is irradiated with X-ray, and in-YZ-plane rotation (φ=0° to 180°) is made with the center of the plate-like sample 20 as the center of rotation so that scanning of the entire surface of the plate-like sample 20 is made with the region irradiated with X-ray. The plate-like sample 20 is set at an angle which allows the diffraction intensity of a <220> plane to be obtained. X-Ray beam 40 (Cu-Kα ray: wavelength: 1.54 Å) ejected through a slit 30 and collimated enters the plate-like sample 20, and the intensity of the X-ray beam diffracted with respect to each sample rotation angle (θ) is detected by a detector (not illustrated) with rotation of the plate-like sample 20 in a YZ plane (φ scanning measurement), to provide an X-ray diffraction chart (φ scanning chart).

(27) FIG. 3A and FIG. 3B each illustrate an example of the X-ray diffraction chart obtained according to the above procedure, and FIG. 3A illustrates an X-ray diffraction chart, obtained from a <220> plane of a plate-like sample collected from a polycrystalline silicon rod grown under a condition of a D/L value of 0.69, and FIG. 3B illustrates an X-ray diffraction chart obtained from a <220> plane of a plate-like sample collected from a polycrystalline silicon rod grown under a condition of a D/L value of 0.36, when multiple pairs of silicon cores are placed in a reaction furnace and it is assumed that the average value of the final diameter of the polycrystalline silicon rod is defined as D (mm) and the mutual interval between the multiple pairs of silicon cores is defined as L (mm).

(28) In the X-ray diffraction chart in FIG. 3A, there are observed a large number of diffraction peaks from a Miller index plane <220>, higher in the intensity than the intensity of a baseline. On the contrary, in the X-ray diffraction chart in FIG. 3B, there are not observed any diffraction peaks from a Miller index plane <220>, higher in the intensity than the intensity of a baseline. The results indicate that, while the plate-like sample collected from a polycrystalline silicon rod grown under a condition of a D/L value of 0.69 includes a needle crystal and a locally heterogeneous crystal having a Miller index plane <220> as a principal plane in large numbers, the plate-like sample collected from a polycrystalline silicon rod grown under a condition of a D/L value of 0.36 hardly includes a needle crystal and a locally heterogeneous crystal having a Miller index plane <220> as a principal plane.

(29) FIG. 4A and FIG. 4B represents the results of optical microscope observation of the surfaces of such plate-like samples, etched by a mixed liquid of hydrofluoric acid and nitric acid. While a needle crystal and a locally heterogeneous crystal are observed in large numbers in the plate-like sample collected from a polycrystalline silicon rod grown under a condition of a D/L value of 0.69 (FIG. 4A), such any heterogeneous portion is not observed in the plate-like sample collected from a polycrystalline silicon rod grown under a condition, of a B/L value of 0.36 (FIG. 4B).

EXAMPLES

(30) Multiple pairs of silicon cores were placed in a reaction furnace, and a polycrystalline silicon rod was grown according to a Siemens method. The average value D of the diameter after the deposition step of the polycrystalline silicon rod was set within the range from 150 to 300 mm. The mutual interval L (mm) between the multiple pairs of silicon cores was set by changing the mutual distance between center points each connecting two electrodes in which both the lower ends of the silicon cores were received. The reaction furnace had an inner diameter of 1.8 m and a height of 3 m, the concentration of a trichlorosilane gas as a raw material of polycrystalline silicon was 30% by volume, and the flow rate of a hydrogen gas for dilution was 100 Nm.sup.3/hour.

(31) The evaluation results of the polycrystallinee silicon rod in each of Examples 1 to 9 and Comparative Examples 1 to 5 were summarized in Table 1 and Table 2. The description “Locally heterogeneous crystal” in the Tables refers to one confirmed as a locally heterogeneous portion having a grain size of 10 μm or more on an etched surface obtained in etching of the surface of a plate-like sample collected so that the direction vertical to the long axis direction of the polycrystalline silicon rod corresponded to the principal plane direction, by a mixed liquid of hydrofluoric acid and nitric acid.

(32) TABLE-US-00001 TABLE 1 Comparative Example Example 1 2 3 1 2 3 4 D/L 0.90 0.69 0.42 0.40 0.36 0.25 0.11 Reaction pressure (MPa) 0.45 Needle crystal Position Outside Observed Observed Observed Not Not Not Not observed observed observed observed R/2 Observed Observed Observed Not Not Not Not observed observed observed observed Center Not Not Not Not Not Not Not observed observed observed observed observed observed observed Locally Position Outside Observed Observed Observed Not Not Not Not heterogeneous observed observed observed observed crystal R/2 Observed Observed Observed Not Not Not Not observed observed observed observed Center Observed Observed Not Not Not Not Not observed observed observed observed observed Variation in Position Outside 1.08 1.10 0.58 0.54 0.40 0.20 0.10 diffraction intensity R/2 1.00 0.60 0.30 0.30 0.20 0.15 0.10 of <220> (6σ.sub.n−1/ Center 0.40 0.20 0.15 0.12 0.10 0.09 0.08 Ave.) Surface Position Outside 1091 1088 1085 1084 1082 1082 1081 temperature (° C.) R/2 1055 1053 1051 1050 1048 1048 1048 Center 1030 1028 1025 1023 1021 1020 1020 FZ crystal line disappearance Observed or Observed Observed Observed Not Not Not Not not observed observed observed observed observed

(33) TABLE-US-00002 TABLE 2 Comparative Example Example 4 5 5 6 7 8 9 D/L 0.36 Reaction pressure (MPa) 0.05 0.1 0.2 0.3 0.45 0.6 0.9 Locally Outside Observed Observed Not Not Not Not Not heterogeneous observed observed observed observed observed crystal R/2 Observed Observed Not Not Not Not Not observed observed observed observed observed Center Observed Observed Not Not Not Not Not observed observed observed observed observed

(34) The surface temperature shown in Table 1 is a value measured with a radiation thermometer at the center portion in the height direction, and is merely a reference value.

(35) It was found from the results shown in Table 1 that the polycrystalline silicon rod in each of Examples 1 to 4 allowed the degree of variation in diffraction intensity from a <220> plane, in evaluation as the 6σ.sub.n-1/average value, to satisfy 0.15 or less in the center region, 0.30 or less in the R/2 region and 0.58 or less in the outer region, exhibited crystal homogeneity, and was not observed to cause any crystal line to disappear-even when used as a raw material for single-crystallization according to an FZ method.

(36) In addition, as the D/L value was smaller, the degree of variation in diffraction intensity from a <220> plane was also lower. Specifically, the polycrystalline silicon rod in Example 2 allowed the degree of variation in diffraction intensity from a <220> plane to satisfy 0.12 or less in the center region, 0.30 or less in the R/2 region and 0.54 or less in the outer region, the polycrystalline silicon rod in Example 3 allowed the degree of variation in diffraction intensity from a <220> plane to satisfy 0.09 or less in the center region, 0.15 or less in the R/2 region and 0.20 or less in the outer region, and the polycrystalline silicon rod in Example 4 allowed the degree of variation in diffraction intensity from a <220> plane to satisfy 0.08 or less in the center region, 0.10 or less in the R/2 region and 0.10 or less in the outer region.

(37) On the contrary, the polycrystalline silicon rod in each of Comparative Examples did not allow the degree of variation in diffraction intensity from a <220> plane, in evaluation as the 6σ.sub.n-1/average value, to satisfy the condition “0.15 or less in the center region, 0.30 or less in the R/2 region and 0.58 or less in the outer region”, and was observed to cause any crystal line to disappear when used as a raw material for single-crystallization according to an FZ method.

(38) It was found from the results shown in Table 2 that high crystal homogeneity was achieved when the reaction pressure in the deposition step of polycrystalline silicon was set to 0.2 MPa or more.

INDUSTRIAL APPLICABILITY

(39) The present invention provides polycrystalline silicon suitable as a raw material for production of single-crystalline silicon, thereby resulting in contribution to stable production of single-crystalline silicon.