JP5545066B2 - Defect inspection method, process management method, and defect inspection apparatus - Google Patents

Description

  The present invention relates to a defect inspection method, a process management method, and a defect inspection apparatus. For example, the present invention relates to a defect inspection for performing manufacturing process management of electronic devices such as semiconductor devices in a variety of batches.

  A defect inspection using a defect inspection apparatus is performed as a discovery of an abnormality in a semiconductor manufacturing apparatus or a quality check in the course of a semiconductor device manufacturing process. Management of the number of defects output from the defect inspection apparatus is performed by a trend chart or statistical process control (SPC).

  As a defect inspection apparatus in this case, a defect is detected using reflected light of laser light or lamp light. The output information of the defect inspection includes the number of detected defects and coordinate information of each defect location.

Here, a conventional defect inspection process will be described with reference to FIG. FIG. 8 is a flowchart of a conventional defect inspection process. Here, a defect inspection process for a gate electrode will be described as an example.
s ₁ : First, a polycrystalline silicon film to be a gate electrode is formed.
s ₂ : Pattern the resist by lithography.
s ₃ : The polycrystalline silicon film is etched using the resist pattern as a mask to form a gate electrode.
s ₄ : Inspection is performed using a defect inspection apparatus. For example, after the gate electrode is processed, a defect inspection such as an etching residue or a resist pattern defect is performed.
s ₅ : Impurities are ion-implanted using the gate electrode as a mask.

  In addition, after the element isolation region forming step, the sidewall forming step, the first layer via forming step, the first layer wiring forming step, the second layer via forming step, etc. After the process, defect inspection is sequentially performed as necessary.

  FIG. 9 shows an example of defect number data in the output information. The vertical axis represents the number of defects per wafer, the horizontal axis represents the defect inspection date and time, and the number of defects is plotted as a trend chart. A management value related to the number of defects is set in the table and used for manufacturing process management. When the number of defects exceeds the control value, the manufacturing equipment and the like are investigated.

  In many cases, system LSIs are produced in a variety of types and in small quantities, and the circuit pattern is different for each type, so the number of defects varies depending on the type. Therefore, at present, a trend chart of the number of defects for each type is created and managed, but halation is a problem.

  FIG. 10 is an explanatory view of halation, FIG. 10 (a) is a plan view of the main part showing the reflected light distribution of the semiconductor chip of type A, and FIG. 10 (b) is an explanation of the cumulative probability of the reflected light intensity. FIG. As shown in FIG. 10A, each random logic circuit 31 surrounded by a frame in the figure is a region where halation occurs.

  As shown in FIG. 10B, when the reference for determining that halation has occurred is set to reflected light intensity = 230, this product A has halation occurring in an area of 30% of the entire chip. .

  FIG. 11 is a plan view of the main part showing the reflected light distribution of the semiconductor chip of the product type B. Compared with the product type A, the area causing halation is remarkably reduced. Incidentally, the area where halation occurs is 2% when judged under the same conditions and the same standards as those of the kind A.

  As described above, the area of the halation-induced region depends on the circuit pattern, and thus varies depending on the product type. Therefore, it is desirable to set the inspection conditions for each product type each time, but it takes time to set the inspection conditions for each product type, which is difficult in terms of factory operation.

  Therefore, at present, a type having a relatively large amount of production is selected as a representative type, defect inspection is performed, a trend chart is created, and process management is performed. In this case, as will be described later, since defect inspection cannot be performed in a halation-proposed area, the inspection conditions are set so that halation is difficult to occur in the laser power, etc., and standard inspection conditions are set. This standard inspection condition is also applied to.

  FIG. 12 is an explanatory diagram of the defect inspection principle. First, as shown in the upper diagram, for example, an ultraviolet laser beam having a beam spot size of 1000 nm is scanned on a chip manufactured on a wafer by moving the stage. The middle diagram is an image diagram of the scanned position, and shows a case where the center chip has a defect. The lower diagram is a diagram in which reflected light from the chip is imaged by an image sensor and converted into light intensity.

  At the time of inspection, the presence or absence of a defect is determined from the difference by comparing the reflected light intensities at the same location of adjacent chips. In the figure, the center chip is the inspection target, the left and right chips are the reference chips, and when the reflection intensity is compared with the left and right, a difference appears in the reflection intensity in any case, so it is determined that the center chip is defective.

  FIG. 13 is an explanatory diagram of the case where there is halation. When the reflected light is saturated and halation occurs, there is no difference in reflection intensity between the inspection target chip and the left and right reference chips, and the inspection target If a chip is defective, it cannot be certified. Therefore, as described above, defect inspection is performed by setting the laser power or the like as an inspection condition so that halation is less likely to occur.

  FIG. 14 is an example of a trend chart. FIG. 14A is a trend chart for the product type A, and FIG. 14B is a trend chart for the product type B. Here, the average number of defects is 55 for the type A having the same chip size, and the average number of defects is 95 for the type B.

JP 2009-088026 A

  However, as shown in FIG. 14, when a trend chart is created for each product type, since the number of production is small, the production date is not continuous over a long period of time, so the data becomes discrete and periodic process management is performed. There is a problem that becomes difficult.

  In order to solve such a problem, as shown in FIG. 15, when a plurality of types are plotted on the same trend chart, high-precision management is performed due to variations in the number of defects between types due to differences in inspection sensitivity caused by halation. There is a problem that becomes difficult.

  Accordingly, an object of the present invention is to plot defect number data of a plurality of types on the same chart and to perform process management regularly and accurately.

  From one aspect to be disclosed, the step of irradiating the wafer surface on which the electronic device chip is formed with light, measuring the reflected light intensity of the irradiated light in the inspection target region of the wafer, and the measured reflected light intensity. The step of determining the presence or absence of halation compared to a preset threshold value, the step of calculating the area of the non-halation region in the wafer based on the presence or absence of the halation, and the measured reflected light intensity are adjacent to each other. A step of determining the presence or absence of defects by comparing between electronic device chips, a step of counting the number of defects of the entire wafer from the determination result, and dividing the total number of defects by the area of the non-halation area And a step of calculating a defect density.

  Further, from another viewpoint to be disclosed, the above-described defect inspection method performs defect inspection on a plurality of types of wafers on which electronic device chips of different varieties are formed under the same inspection condition, and A defect inspection method characterized by plotting the substantial defect density of a wafer on the same trend chart is provided.

  From another viewpoint to be disclosed, when a management value is preset for the substantial defect density acquired by the defect inspection method described above, and the substantial defect density exceeds the management value, the substantial defect density is There is provided a process management method characterized by investigating a manufacturing apparatus or manufacturing conditions for manufacturing a wafer exceeding the control value and managing the manufacturing process.

Further, from still another aspect to be disclosed, a mounting unit for mounting a wafer on which a plurality of electronic device chips to be measured are mounted, and light irradiation means for irradiating the wafer while scanning a linear light beam And a light receiving means for measuring the reflected light from the wafer, and an area calculating means for calculating the area of the non-halation area in the wafer by comparing the reflected light intensity measured by the light receiving means with a preset threshold value. Defect number inspection means for inspecting the presence or absence of defects by comparing reflected light intensities at the same position between the electronic device chips having the same surface pattern adjacent to each other, and defects in the wafer unit acquired by the defect number inspection means A defect inspection apparatus comprising: a defect defect calculating means for calculating a defect defect density by dividing the number by the area of the halation-free region. There is provided.

  According to the disclosed defect inspection method, process management method, and defect inspection apparatus, it becomes possible to plot defect number data of electronic devices in the middle of manufacturing a plurality of types on the same chart, and to perform periodic management with high accuracy. .

It is a principle flowchart of process management using the defect inspection method of an embodiment of the invention. It is a correlation diagram of the number of defects per unit area and RAM scale. It is a notional block diagram of the defect inspection apparatus used for embodiment of this invention. It is a conceptual diagram explaining the determination of the presence or absence of halation. It is a factory management system figure using the defect inspection method and process management method of Example 1 of this invention. It is explanatory drawing of a defect detection result. Is a trend chart that you created in step S _9. It is a flowchart of the conventional defect inspection process. It is an example of defect number data in output information. It is explanatory drawing of halation. It is a principal part top view which shows the reflected light distribution of the semiconductor chip of the kind B. It is explanatory drawing of a defect inspection principle. It is explanatory drawing in case there exists halation. It is an example of a trend chart. It is a trend chart plotting multiple varieties.

Here, with reference to FIG. 1 thru | or FIG. 4, the defect inspection method of embodiment of this invention is demonstrated. FIG. 1 is a principle flowchart of process management using the defect inspection method according to the embodiment of the present invention. As shown in FIG. 1, it performs process control at step S ₁ to S _8.
S ₁ : Measure reflected light intensity in the region to be inspected.
S ₂ : The presence or absence of halation is determined by comparing the reflected light intensity with a preset threshold value.
S _3: calculating the area of no halation area wafer units.
S _4: determining the presence or absence of a defect.
S ₅ : Count the number of defects per wafer.
S ₆ : The substantial defect density is calculated by dividing the total number of defects by the area of the area without halation. S _7: to plot a substantial defect density in the trend chart.
S _8: to implement the process managed by the trend chart.

  This kind of defect inspection method and process management reflect the results of intensive studies that have found that the difference in the number of defects due to differences between varieties largely depends on the amount of halation, not the chip size. The situation will be described with reference to FIG.

  FIG. 2 is a correlation diagram between the number of defects per unit area and the RAM scale. Conventionally, it was considered that the difference in the number of defects between the types was greatly influenced by the difference in the chip size, but in reality, it was found that the influence of the circuit pattern mounted on the type was large. This is because if the RAM scale is small, the proportion of the area occupied by random logic circuits that are prone to halation increases, and the area that can be inspected decreases. As shown in FIG. 2, the number of defects per unit area tends to decrease. Is seen.

  Therefore, if the number of defects per unit area in a non-halation area capable of substantial defect inspection, that is, the substantial defect density is used as a criterion, different types are plotted on the same trend chart and process control is performed. It becomes possible to do.

  FIG. 3 is a conceptual configuration diagram of the defect inspection apparatus used in the embodiment of the present invention. As shown in the figure, a stage 12 on which a wafer 11 on which an electronic device chip is formed is placed, a light irradiation means 13, a lens 14 for condensing the reflected light from the wafer 11, and an image sensor 15 for receiving the collected reflected light. And a control unit 20.

  The wafer 11 is typically a semiconductor wafer on which a system LSI is formed, but is not limited to a semiconductor wafer. For example, the present invention is also applied to other electronic device devices such as a wafer on which a ferroelectric device is formed and a wafer on which a magnetoresistive effect element such as a GMR effect element is formed.

  The light irradiation means 13 is typically an ultraviolet laser, but a visible light laser may be used, and is not limited to a laser, and lamp light may be used.

  The control unit 20 includes an area calculating unit 21 that calculates the area of the non-halation area in the wafer 11, a defect number inspecting unit 22 that inspects for the presence or absence of defects, and a real defect density calculating unit 23 that calculates a real defect density. I have. These means substantially comprise a program.

  The area calculation means 21 calculates the area of the area without halation in the wafer 11 by comparing with a preset threshold value using the position coordinates where the reflected light intensity measured by the image sensor 15 is recorded. FIG. 4 is a conceptual diagram illustrating the determination of the presence / absence of halation. Both the center inspection target chip and the adjacent comparison target chip are recognized as halation when the reflected light intensity is equal to or higher than a preset threshold value. The position information is recorded in the memory 24 of the control unit 20.

  The defect inspection means 22 inspects the presence or absence of a defect by comparing the reflected light intensities at the same position between adjacent electronic device chips, as in the prior art. Further, the real defect density calculation means 23 obtains the real defect density by dividing the number of defects in the wafer unit acquired by the defect number inspection means 22 by the area of the area without halation.

  In the embodiment of the present invention, since the substantial defect density is used as a criterion, the defect number data of electronic devices in the middle of manufacturing a plurality of types is plotted on the same chart, and the periodic management is performed with high accuracy. Is possible.

  Based on the above, a process management method using the defect inspection method according to the first embodiment of the present invention will now be described with reference to FIGS. FIG. 5 is a factory management system diagram using the defect inspection method and the process management method according to the first embodiment of the present invention, and includes automatic operation using a defect inspection apparatus and process management using a trend chart and the like.

Automatic operation using the defect inspection apparatus comprises the following steps S ₁ to S _8.
S ₁ : Irradiation is performed while scanning a semiconductor wafer on which a system LSI is formed with ultraviolet laser light. Here, for example, 365 nm ultraviolet laser light is used.
S _2: measuring the intensity of the reflected light detected by the image sensor.
S ₃ : The presence or absence of halation is determined by comparing the reflected light intensity with a threshold value.
S _4: recording the halation and determined position coordinates or halation without the determined position coordinates.
S ₅ : Move to the next inspection area, perform steps S _{1 to} S ₄ , and proceed to the next step S ₆ when the entire area is scanned.
S ₆ : For all inspection targets, the area of the region where no halation has occurred is calculated in wafer units as the actual inspection area using the recorded position coordinates.
S ₇ : Defect inspection is performed in the same manner as in the past to determine the number of defects per wafer.
S ₈ : Divide the number of detected defects by the substantial inspection area to calculate the substantial defect density.

FIG. 6 is an explanatory diagram of the defect detection results, and shows the measurement results for two types of system LSIs of type A and type B both having a size of 7 mm × 7 mm. As described above, since the ratio of the area in which the halation of type A does not occur is 70%, the actual inspection area is 34.2 mm ² . Since the number of defects per wafer is 55, the number of defects per actual unit inspection area, that is, the actual defect density is 1.60 / mm ² .

On the other hand, since the ratio of the area where the halation of the type B does not occur is 98%, the actual inspection area is 48.0 mm ² . Since the number of defects per wafer is 95, the number of defects per actual unit inspection area, that is, the actual defect density is 1.98 / mm ² .

  Therefore, when compared with the real number of defects, the type B / type A = 1.73 times, whereas when compared with the real defect density, type B / type A = 1.23 times. It can be seen that it is possible to evaluate the product types B and A on the same soil.

Then, process control using a trend chart, etc. in FIG. 5 is composed of steps S ₉ and S S _10.
S ₉ : Plot the real defect density on the trend chart.
S ₁₀ : A management value is set in the trend chart, and statistical process management is performed.

FIG. 7 is a trend chart that you created in step S _9. As shown in FIG. 7, compared with the trend chart of FIG. 15 described above, even when different varieties are plotted on the same trend chart, the variation among the varieties is reduced.

  As described above, since the effective defect density is used as the management standard, different types can be managed on the same trend chart. In addition, since the difference between types can be ignored, the defect trend chart can be appropriately updated by periodically updating without depending on the requirements of each type. In FIG. 7, two types are plotted, but three or more types may be plotted on the same trend chart.

DESCRIPTION OF SYMBOLS 11 Wafer 12 Stage 13 Light irradiation means 14 Lens 15 Image sensor 20 Control part 21 Area calculation means 22 Defect number inspection means 23 Real defect density calculation means 24 Memory 31 Random logic circuit

Claims (5)

Irradiating the wafer on which the electronic device chip is formed with light, and measuring the reflected light intensity of the irradiated light in the inspection pair region on the wafer surface;
Comparing the measured reflected light intensity with a preset threshold to determine the presence or absence of halation;
Calculating the area of the non-halation area in the wafer according to the presence or absence of the halation;
For the measured reflected light intensity, comparing the electronic device chips adjacent to each other to determine the presence or absence of defects;
A step of counting the number of defects of the entire wafer from the determination result;
And a step of calculating a substantial defect density by dividing the total number of defects by the area of the non-halation region.   The defect inspection method according to claim 1, wherein defect inspection is performed under the same inspection condition on a plurality of types of wafers on which electronic device chips of different varieties are formed, and the substantial defect density of the plurality of types of wafers A defect inspection method characterized by plotting a graph on the same trend chart.   A management value is set in advance for the substantial defect density acquired in claim 1 or claim 2, and when the substantial defect density exceeds the management value, a wafer in which the substantial defect density exceeds the management value is determined. A process management method characterized by investigating a manufactured manufacturing apparatus or manufacturing conditions and managing a manufacturing process.   The process management method according to claim 3, wherein the electronic device chip is a system large-scale integrated circuit chip. A mounting unit for mounting a wafer on which a plurality of electronic device chips to be measured are formed;
Light irradiation means for irradiating the wafer while scanning a linear light beam;
A light receiving means for measuring reflected light from the wafer;
An area calculating means for calculating the area of the halation-free region in the wafer by comparing the reflected light intensity measured by the light receiving means with a preset threshold;
Defect number inspection means for inspecting the presence or absence of defects by comparing the reflected light intensity at the same position between the electronic device chips having the same surface pattern adjacent to each other;
A defect inspection apparatus, comprising: a substantial defect density calculation unit that obtains a substantial defect density by dividing the number of defects in the wafer unit acquired by the defect number inspection unit by an area of the non-halation region.

Images