Method for evaluating semiconductor substrate

Abstract

Method for evaluating electrical properties of a semiconductor substrate (1), wherein the method comprises the following steps: Formation of a pn junction on a surface of the semiconductor substrate (1); Attaching the semiconductor substrate (1) to a wafer holder (8) which is equipped with a device (6) for performing light irradiation (4) on the surface of the semiconductor substrate (1) and a device (7) for measuring the amount of light for the irradiation; Applying light irradiation (4) to the surface of the semiconductor substrate (1) for a predetermined time; and Measuring a quantity of carriers generated after irradiation (4) of the pn junction, at least after switching off the irradiation (4), wherein a quantity of carriers generated during the light irradiation (4) of the pn junction is measured in the step of performing the light irradiation (4), the quantity of carriers produced during light irradiation (4) and the quantity of carriers produced after light irradiation (4) are measured with a carrier measuring probe (5) which is provided separately from the equipment (6) for carrying out the light irradiation (4) and the equipment (7) for measuring the quantity of light for the irradiation, and the carrier measuring probe (5) is a non-contact Kelvin probe or a mercury probe.

Description

TECHNICAL AREA

[0001] The present invention relates to a method for evaluating a semiconductor substrate. STATE OF THE ART

[0002] As the miniaturization and increased performance of semiconductor devices such as solid-state image sensors, including memory and CCDs (charge-coupled devices), progress, silicon wafers as a material must also exhibit higher quality to improve the yield of such products. Accordingly, various silicon wafers are being developed to meet this requirement. It is assumed that the crystallinity at a specific section of the wafer's surface layer, in particular, has a direct and significant influence on the product's properties.To improve crystallinity, the following were developed: 1) an annealed wafer that was heat-treated at a high temperature in an atmosphere containing an inert gas or hydrogen; 2) a defect-free polished wafer with reduced embedded defects by improving the drawing conditions; 3) an epitaxial wafer obtained by epitaxial growth; and the like.

[0003] A conventional method for evaluating the electrical properties of a silicon wafer's surface quality is the evaluation of the dielectric strength of an oxide film (GOI). This is performed as follows: a gate oxide film is formed on the surface of a silicon wafer by thermal oxidation; electrical stress is applied to the silicon oxide film, which is an insulator, by forming an electrode on the oxide film; and the surface quality of the silicon wafer is evaluated by the degree of insulation. That is, if defects and metal impurities are present on the surface of the original silicon wafer, these will be incorporated into the silicon oxide film by thermal oxidation, or an oxide film will form that conforms to the surface shape, resulting in an uneven insulator, and so on.Therefore, if defects or impurities are present, the insulating properties are impaired. Conventional methods use such characteristics to evaluate the surface quality of a silicon wafer.

[0004] The actual component in question here is the reliability evaluation of a gate oxide film of a MOSFET (metal-oxide-semiconductor field-effect transistor), and various wafers have been developed to improve it. However, component yields can be affected even if no problems arise during the GOI evaluation. Such phenomena have increased, particularly in recent years, with the higher integration of components. For example, in solid-state image sensors, when considering the improvement in sensitivity through reduced dark current, reducing the wafer-induced leakage current leads to a decrease in dark current and ultimately contributes to improved component performance.

[0005] In recent years, even trace amounts of metallic impurities have been shown to influence the performance of devices, particularly when the cause is identified. Various metals are now being detected in chemical analyses in an effort to achieve greater sensitivity. However, it is currently extremely difficult to determine which of the detected metallic elements has the greatest impact on an actual component or junction leakage. Furthermore, the analysis of metallic impurities is a technique that involves, for example, etching the surface of a wafer and analyzing the etching solution. Therefore, the information from the wafer surface is analyzed vicariously, and it is generally not possible to obtain information about the distribution across the plane.On the other hand, when evaluating the leakage current, many pn junctions are formed on the surface of a silicon substrate, and the reverse current of each pn junction is determined. In this way, the leakage distribution at the substrate level can be determined.

[0006] Solid-state image sensors, such as CCD and CMOS (Complementary MOS) image sensors, employ different methods for extracting the electrical charge generated by electron-hole pairs that arise upon incident light. However, the underlying principle of converting light into electrical charge (photoelectric conversion) is the same: forming a pn junction and having a depletion layer as its structure. The phenomenon of electron-hole pairs forming in a depletion layer due to the presence of defects or impurities, thus generating an electrical charge even when no light is incident, is known as white noise or dark current.The properties of the reverse current of a wafer with a pn junction formed in this way can be evaluated by a dark current in a solid-state image sensor and can be used as an indicator for improvements in the suspicion of a cause or the development of materials.

[0007] In addition to dark current, residual image properties are also known as material properties that influence a solid-state image sensor. It is known that residual image properties are closely related to the materials, especially the substrate (Non-Patent 1 and Non-Patent 2). For example, Non-Patent 1 and Non-Patent 2 state that light elements in a silicon substrate influence the residual image properties, and that the defects responsible for this effect are composites of boron and oxygen. Since the influence of the substrate on the residual image properties has become clear as described, in addition to evaluating the reverse current properties, which corresponds to dark current evaluation, a method for evaluating a substrate that addresses residual image properties has become necessary in order to evaluate the substrate properties.

[0008] A method described in patent specification 1 is such a method, but to carry out this method it is necessary to form a photodiode, which is a light-receiving part of a solid-state image sensor. For this purpose, in addition to the diffusion of a dopant, the formation of an element isolation structure is required. To form this element isolation structure, process equipment for carrying out photolithography, a subsequent etching process, etc., is required. Furthermore, time and large-scale facilities are required to carry out an evaluation. As described, the fabrication of an actual device was essential for evaluating the residual image properties, and the problem is that evaluation in the wafer state is difficult.

[0009] Patent specification 2 relates to a method for measuring the recombination lifetime of a minority carrier in a semiconductor. The method comprises a step for applying a voltage in an inversion voltage region to a semiconductor to generate an excessive minority carrier within the semiconductor. Furthermore, the method comprises a process for determining the recombination lifetime of the minority charge carriers in the semiconductor based on the time of the capacitance change. QUOTE LIST PATENT LITERATURE Patent specification 1: JP 2019 - 9 212 A Patent specification 2: JP H11 - 186 350 A NON-PATENT LITERATURE Non-Patent Document 1: The 77th Japan Society of Applied Physics Autumn Meeting, Meeting Proceedings, 14p-P6-10, Tasuku Kaneda, Akira Ohtani, “Analysis of a residual image in CMOS image sensor 1”. Non-Patent Document 2: The 77th Japan Society of Applied Physics Autumn Meeting, Meeting Proceedings, 14p-P6-11, Akira Ohtani, Tasuku Kaneda, “Analysis of a residual image in CMOS image sensor 2” Non-patent specification 3: S. Rein “Lifetime spectroscopy” p. 398, Springer, 2005 SUMMARY OF THE INVENTIONAL PROBLEM

[0010] The present invention was made with regard to the problem described above. One object of the present invention is to enable the same evaluation in a wafer state as if an actual solid-state image sensor had been formed, without producing a device, by using process equipment when evaluating features corresponding to the residual image features of a wafer intended for use in products requiring high yields, such as CCD and CMOS image sensors. Another object is to contribute to achieving higher quality semiconductor substrates by enabling simple measurement at the wafer level. SOLUTION TO THE PROBLEM

[0011] To achieve this goal, the present invention provides a method for evaluating the electrical properties of a semiconductor substrate, the method comprising the following steps: Forming a pn junction on a surface of the semiconductor substrate; Mounting the semiconductor substrate onto a wafer holder, which is equipped with a device for irradiating the surface of the semiconductor substrate with light and a device for measuring the amount of light used for irradiation; Applying light irradiation to the surface of the semiconductor substrate for a predetermined time; and Measuring the amount of carriers generated after irradiation of the pn junction, at least after switching off the irradiation.

[0012] With such a method, poor residual image properties, which are due to the semiconductor substrate and are considered problematic in CCD and CMOS image sensors, etc., can be evaluated simply and with high precision at a substrate level, and a high-quality semiconductor substrate can be provided.

[0013] According to the invention, a quantity of the carriers generated during the light irradiation of the pn junction is measured in the light irradiation step.

[0014] By measuring the amount of carriers produced during light irradiation, it can be more reliably avoided that differences in the amount of carriers produced impair the residual image properties.

[0015] Furthermore, according to the invention, the quantity of carriers produced during light irradiation and the quantity of carriers produced after light irradiation are measured with a carrier measuring probe, which is provided separately from the equipment for carrying out the light irradiation and the equipment for measuring the amount of light for the irradiation.

[0016] This method makes the measurement easier.

[0017] According to a first alternative of the invention, the carrier measuring probe is a contactless Kelvin probe.

[0018] A non-contact Kelvin probe is suitable as a carrier measuring probe.

[0019] According to a second alternative of the invention, the carrier measuring probe is a mercury probe.

[0020] A mercury probe is also suitable as the carrier measuring probe.

[0021] Furthermore, the semiconductor substrate is preferably subjected to heat treatment after the pn junction formation step and before the semiconductor substrate application step.

[0022] By subjecting the semiconductor substrate to heat treatment beforehand, it is possible to create and enlarge defects in the semiconductor substrate, thus making the poor residual image properties more apparent. It is also possible to reproduce the behavior of heat treatment in a device manufacturing process.

[0023] Furthermore, a semiconductor substrate for a solid-state image sensor is preferably used as the semiconductor substrate, and the residual image properties of the solid-state image sensor are preferably evaluated from the ratio of the amount of carriers produced after light irradiation to the amount of carriers produced during light irradiation.

[0024] The generation of carriers during light irradiation varies depending on the type of semiconductor substrate. However, standardization is possible using such a method by determining the ratio between the amount of carriers generated after light irradiation and the amount of carriers generated during light irradiation, thus enabling evaluation independent of the type of semiconductor substrate. ADVANTAGEOUS EFFECTS OF THE INVENTION

[0025] With the inventive method for evaluating a semiconductor substrate, poor residual image properties due to the semiconductor substrate, which are considered problematic in CCD, CMOS image sensors, etc., can be easily and with high precision evaluated at the substrate level, and a high-quality semiconductor substrate can be provided. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 is a schematic view showing an example of an embodiment of the present invention. Fig. Figure 2 is a diagram showing an example of a transition structure formed in the inventive method for evaluating a semiconductor substrate. Fig. Figure 3 is a diagram showing an example of a measurement sequence of the inventive method for evaluating a semiconductor substrate. Fig.Figure 4 is a graph showing the relationship between: the ratio between the amount of carriers produced after light irradiation and the amount of carriers produced during light irradiation in Example 1 and Example 2; and the oxygen concentration in the substrate. DESCRIPTION OF THE EXECUTION FORMS

[0026] As described above, a method for evaluating a semiconductor substrate was needed that would allow properties to be measured at the substrate level that correspond to the residual image properties of a semiconductor substrate intended for use in products requiring high yield, such as CCD and CMOS image sensors.

[0027] To achieve this goal, the present inventor has conducted serious research and discovered that the following method can solve the problem described above. That is, a method for evaluating the electrical properties of a semiconductor substrate, the method comprising the following steps: Forming a pn junction on a surface of the semiconductor substrate; Mounting the semiconductor substrate onto a wafer holder, which is equipped with a device for irradiating the surface of the semiconductor substrate with light and a device for measuring the amount of light used for irradiation; Applying light irradiation to the surface of the semiconductor substrate for a predetermined time; and Measuring the amount of carriers generated after irradiation of the pn junction, at least after switching off the irradiation.

[0028] The present invention is described below with reference to the drawings, but the present invention is not limited thereto.

[0029] First, a pn junction structure is formed in a semiconductor substrate (the step of forming a pn junction on a surface of the semiconductor substrate). The pn junction structure is not particularly limited, and any pn junction structure is possible, but preferably the leakage current (surface component) caused by the pn junction structure is reduced as much as possible.

[0030] As an example of such a pn transition structure, Fig.Figure 2 shows an example of a transition structure formed in the process according to the invention for evaluating a semiconductor substrate. Such a transition structure can be produced, for example, as follows: boron or the like is diffused onto a semiconductor substrate 1 doped with phosphorus or the like to form a highly concentrated diffusion layer 2 whose conductivity type is opposite to that of the semiconductor substrate 1; and a pn junction is formed by bringing the semiconductor substrate 1 and the highly concentrated diffusion layer 2, whose conductivity types are opposite to each other, into contact. Here, electrons and holes combine near the pn junction and form a depletion layer 3 in which no supports are present. As mentioned above, Fig.2 is only one example for the description of a pn junction that can be formed within the scope of the present invention, and the conductivity type of the semiconductor substrate, the pn junction structure and so on are not particularly limited.

[0031] To describe the inventive method for evaluating a semiconductor substrate, shows Fig. Figure 1 is an example of an embodiment of the present invention. The step of attaching the semiconductor substrate 1 with the pn junction structure produced in the manner above to a wafer holder 8, which is equipped with a light source 6 for performing light irradiation and a light meter 7 for measuring the amount of light as shown in Figure 1, is described in Figure 1. Fig.The following steps are performed: 1. Then, after performing light irradiation 4 on the surface of the semiconductor substrate 1 at a predetermined illuminance for a predetermined time (the step of performing light irradiation), the step of measuring the quantity of carriers produced after the light irradiation 4 is performed after the light irradiation 4 is switched off.

[0032] Here, it can be assumed that a light source such as an LED emitting white light is preferred when using an actual device. However, if the device is specialized for infrared light, for example, a light source with a wavelength adapted to the purpose can also be selected. Furthermore, it is desirable that the amount of light (illuminance) does not vary from measurement to measurement. Accordingly, the light source preferably has both a mechanism for measuring the illuminance and a mechanism for adjusting the illuminance.

[0033] Furthermore, with regard to the amount of light at the time of measurement in an actual device, residual image characteristics occur when a shutter is temporarily closed to obtain an image after the incidence of strong light, and then the shutter is opened to obtain the next image. The carriers produced by the previous light are not sufficiently eliminated, and their influence persists. It must therefore be assumed that a comparatively high amount of light is required. In an actual experiment, it is necessary to determine the optimal illuminance beforehand, in some cases by adjusting the illuminance. However, it is usually sufficient to set the illuminance of a commercially available light source to its maximum. Specifically, the illuminance is preferably around 500 lux.

[0034] Furthermore, the duration of light exposure is preferably 1 to 10 seconds, more preferably 3 to 7 seconds. If the duration of light exposure is 1 second or more, it is possible to ensure that the amount of light from the illumination stabilizes after the light exposure is switched on, and the illuminance can be kept constant with greater reliability. If the duration is 10 seconds or less, the measurement time can be shortened.

[0035] In this way, the amount of carriers generated in the formed pn junction is measured. A conceptual diagram of the specific time sequences of light irradiation and measurement is shown in Fig. 3 shown. Fig. Figure 3 is a diagram showing an example of a measurement sequence of the inventive method for evaluating a semiconductor substrate.

[0036] The present inventor assumes that the amount of carriers produced by light irradiation 4 is influenced by the type of semiconductor substrate 1 or a light element, in particular carbon, contained in the semiconductor substrate 1. In order to prevent the initial difference in the amount of carriers produced by light irradiation 4 from influencing the residual image properties, the amount of carriers produced at that time (the amount of carriers produced during light irradiation) is measured during the light irradiation process, as shown in Fig. Figure 3 shows that the semiconductor substrate can be evaluated in this way, taking into account the initial difference in the amount of carriers produced.

[0037] It is also planned to switch off the light irradiation and measure the amount of carriers produced (the amount of carriers produced after light irradiation) again, after measuring the amount of carriers produced during light irradiation, in order to determine the ratio between the amount of carriers produced after switching off the light irradiation and the amount of carriers produced during light irradiation. The amount of carriers produced during light irradiation varies depending on the type of semiconductor substrate. However, by determining the ratio between the amount of carriers produced after light irradiation and the amount of carriers produced during light irradiation, standardization is possible, so that an evaluation is possible regardless of the type of semiconductor substrate.

[0038] Furthermore, the time required to measure the quantity of carriers produced after light irradiation, once the light is switched off, also depends on the performance of the measuring equipment. Therefore, it is advisable to perform a check beforehand. The measurement time can, for example, be 1 second and accumulated.

[0039] In the present invention, the time between switching off the light irradiation and starting the measurement of the quantity of carriers generated after light irradiation is not specifically limited. The measurement can be started simultaneously with switching off the light irradiation, or it can be started after a predetermined time has elapsed. The measurement of the quantity of carriers generated after light irradiation is intended to capture the phenomenon of carriers generated by the light irradiation being trapped and then released again, and the released carriers are measured. The time at which the carriers are released again depends on the type of carrier traps and the measurement environment, so it is desirable to perform a check beforehand.For example, it is possible to start the measurement at the same time as switching off the light irradiation and to carry out the measurement with a measurement time of 1 second, and so on.

[0040] Furthermore, in Fig. 3. The reason for pausing the measurement once before measuring the quantity of carriers generated after switching off the light irradiation is that noise is more reliably avoided when the light irradiation is switched off. Therefore, the interruption is not strictly necessary, and depending on the performance of the measuring equipment and the condition of the object being measured, this step is not always required.

[0041] The quantity of carriers produced during and after light irradiation is measured with a carrier measuring probe 5, which is provided separately from the equipment for carrying out the light irradiation and the equipment for measuring the quantity of light for the irradiation, as shown in Fig. Figure 1 shows that by using a non-contact Kelvin probe or a mercury probe as the carrier probe 5, the measurement can be carried out more easily and without device insulation or similar.

[0042] The residual image properties can then be evaluated based on the ratio of the current value of the carrier probe when the light irradiation is on and off. If the current value is high after the light irradiation is switched off, this means that many carriers are trapped, and it can be assumed that the residual image properties are poor.

[0043] In current solid-state image sensors, electron-hole pairs, generated by incoming light when the shutter is open, create an electrical charge, and this charge is then captured to construct an image. After the shutter closes, it is crucial that the electron-hole pairs discharge quickly; if this happens too slowly, it will result in residual image artifacts in the next image.

[0044] Furthermore, sometimes a significant difference cannot be achieved solely through the formation of a pn junction in the semiconductor substrate. In this case, the semiconductor substrate is preferably subjected to heat treatment after the pn junction formation step. For example, it is effective to add heat treatment to create defects in the semiconductor substrate. Such heat treatment allows defects in the semiconductor substrate to form and grow, thus making poor residual image properties more apparent. Moreover, the residual image properties, as described in Non-Patent 1 and Non-Patent 2, are referred to as boron-oxygen composites in a substrate, and it is also effective to perform the measurement after the heat treatment in cases where these defects reproduce their behavior during heat treatment in a device manufacturing process.It is known that defects forming from clusters of light elements such as boron and oxygen arise at relatively low temperatures and become unstable at high temperatures. Therefore, the temperature of the additional heat treatment is preferably a relatively low temperature of about 100 to 500 °C, as described in non-patent 3.

[0045] According to such a method, poor residual image properties, which are due to the semiconductor substrate and are considered problematic in CCD and CMOS image sensors, etc., can be easily and with high precision evaluated at the substrate level, and a high-quality semiconductor substrate can be provided. EXAMPLE

[0046] The present invention is described in more detail below with reference to examples. However, the present invention is not limited to the following examples. [Example 1]

[0047] Using phosphorus-doped CZ silicon wafers with a resistivity of 10 Ω·cm and a diameter of 200 mm, three samples were prepared, each with an oxygen concentration (Oi) in the substrate of 3.38, 3.58, and 3.71 ppma (JEITA). Boron was deposited onto these silicon wafers at 10 keV and a dose of 6.0 × 10 13 atoms / cm² 2 Ion implantation was performed. Then the silicon wafers were annealed at 1000 °C under a nitrogen atmosphere to form a pn junction structure.

[0048] Next, in the Fig.In the measurement sequence shown in Figure 3, the amount of carriers generated (current value) was measured with a non-contact Kelvin probe during a 1-second light irradiation. The light irradiation was then switched off, and the current value was determined in the same way over a 1-second period. The residual image properties were evaluated based on the ratio of the current value (also called the current ratio) before and after the light irradiation. The results are shown in Table 1 and Fig. 4 shown.

[0049] As a result, it can be observed that the higher the oxygen concentration in the substrate, the higher the current ratio and the worse the residual image properties. Boron-oxygen composites are cited as one cause of residual image quality in solid-state image sensors. It can be assumed that the present results demonstrate the following: Phosphorus-doped substrates were used, and the boron concentration was controlled by ion implantation. Therefore, the difference in oxygen concentration in the semiconductor substrates determines the concentration of the boron-oxygen composites. The higher the oxygen concentration, the higher the concentration of the boron-oxygen composites, which affects the residual image properties. [Table 1] Oi (ppma) Current ratio before and after light irradiation (au) 3,38 1,32 3,58 1,41 3,71 1,43 [Example 2]

[0050] Next, pn junction structures were formed in three samples fabricated with CZ silicon wafers similar to those in Example 1, i.e., exhibiting the same oxygen concentrations, using the same procedure as in Example 1. Then, in accordance with non-patent 1, annealing at 450°C under a nitrogen atmosphere was performed for 70 hours. The same measurement as in Example 1 was then carried out.

[0051] The relationship between the oxygen concentration of the substrate and the current ratio after annealing at 450°C is shown in Table 2 and Fig.Figure 4 illustrates this. The higher the oxygen concentration of the substrate, the higher the current ratio, and it can be observed that the residual image properties deteriorate. Furthermore, the difference between the samples is greater when annealing at 450 °C is performed than when annealing is not. Composites of boron and oxygen are said to cause residual images in solid-state image sensors, and it has been shown that these defects increase after annealing at 450 °C. It has been shown that poor residual image properties can become more pronounced after this heat treatment. [Table 2] Oi (ppma) Current ratio before and after light irradiation (au) 3,38 2,70 3,58 3,25 3,71 3,63

[0052] It has been shown that by applying the method according to the invention in this way it is possible to evaluate residual image properties at the substrate level without the use of photolithography or etching equipment by a technique that is simpler and faster than in the past, and that the method according to the invention is effective as a method for evaluating a semiconductor substrate for a solid-state image sensor.

[0053] It should be noted that the present invention is not limited to the embodiments described above. These embodiments are merely examples.

Claims

[1] Method for evaluating electrical properties of a semiconductor substrate (1) wherein the method comprises the following steps: Formation of a pn junction on a surface of the semiconductor substrate (1); Attaching the semiconductor substrate (1) to a wafer holder (8) which is equipped with a device (6) for performing light irradiation (4) on the surface of the semiconductor substrate (1) and a device (7) for measuring the amount of light for the irradiation; Applying light irradiation (4) to the surface of the semiconductor substrate (1) for a predetermined time; and Measuring a quantity of carriers generated after irradiation (4) of the pn junction, at least after switching off the irradiation (4), wherein a quantity of carriers generated during the light irradiation (4) of the pn junction is measured in the step of performing the light irradiation (4), the quantity of carriers produced during light irradiation (4) and the quantity of carriers produced after light irradiation (4) are measured with a carrier measuring probe (5) which is provided separately from the equipment (6) for carrying out the light irradiation (4) and the equipment (7) for measuring the quantity of light for the irradiation, and the carrier measuring probe (5) is a non-contact Kelvin probe or a mercury probe. [2] Method for evaluating electrical properties of a semiconductor substrate (1) according to claim 1, wherein the semiconductor substrate (1) is subjected to heat treatment prior to the step of forming the pn junction and prior to the step of applying the semiconductor substrate (1). [3] Method for evaluating electrical properties of a semiconductor substrate (1) according to claim 1, wherein the semiconductor substrate (1) is a semiconductor substrate (1) for a solid-state image sensor and residual image properties of the solid-state image sensor are evaluated from a ratio of the amount of carriers produced after light irradiation (4) to the amount of carriers produced during light irradiation (4).

Images