JP5369588B2 - Bonding evaluation method, bonding evaluation apparatus, substrate bonding apparatus, evaluation gauge, and stacked semiconductor device - Google Patents

Description

  The present invention relates to an evaluation gauge, and more particularly, to a bonding evaluation method, a stacked semiconductor device, a bonding evaluation apparatus, and a substrate bonding apparatus using the evaluation gauge.

Patent Document 1 describes that a stacked semiconductor device is manufactured by stacking wafers on which elements are formed.
JP 2007-103225 A

  In the manufacture of stacked semiconductor devices, wafers on which elements are formed are aligned and bonded with high accuracy. For this reason, it is required to evaluate the alignment accuracy of the bonding apparatus. Effective evaluation of alignment accuracy involves observation of a number of alignment indices. In the conventional evaluation, for example, a technique of measuring a deviation amount between the upper and lower wafers using a vernier mark is used. In this case, for example, it takes a lot of time to observe all 500 or more chips in a 300 mm wafer, and it has been difficult to evaluate the alignment accuracy inspection on the production line.

  Therefore, in order to solve the above problem, as a first aspect of the present invention, there is provided a bonding evaluation gauge for evaluating the alignment accuracy of a pair of substrates that are aligned and bonded, at least one of which is for inspection light. A pair of transmissive substrates, a first partial pattern formed on one of the pair of substrates as part of the repeated pattern, and a pattern obtained by removing the first partial pattern from the repeated pattern, A second partial pattern formed on the other side of the substrate, and when the pair of substrates are aligned and bonded, the inspection light irradiated on the repetitive pattern formed by the first partial pattern and the second partial pattern A junction evaluation gauge that produces diffracted light is provided.

  According to a second aspect of the present invention, there is provided a stacked semiconductor device including a pair of substrates, at least one of which is transparent to inspection light and bonded in alignment with each other. A first partial pattern formed on one scribe line of the pair of substrates and a pattern obtained by removing the partial pattern from the repetitive pattern, and formed on the other scribe line of the pair of substrates Provided is a stacked semiconductor device that includes a second partial pattern, and in which, when a pair of substrates are bonded, the inspection light applied to the repeated pattern formed by the first partial pattern and the second partial pattern generates diffracted light The

  Furthermore, as a third aspect of the present invention, there is provided a bonding evaluation method for evaluating the alignment accuracy of a pair of substrates that are aligned and bonded, wherein a part of a repetitive pattern is formed on one of the pair of substrates. A first partial pattern forming step for forming one partial pattern, and a second partial pattern forming step for forming a second partial pattern in which the partial pattern is removed from the repeated pattern and the pair of substrates are aligned with the other of the pair of substrates. A joining step for joining, a repetitive pattern formed by joining, a test light irradiating step for irradiating inspection light passing through one of the pair of substrates, and a measuring step for measuring diffracted light generated in the test light by the repetitive pattern A bonding evaluation method is provided.

  Still further, as a fourth aspect of the present invention, one substrate having a first partial pattern that forms part of a repetitive pattern, and the other substrate having a second partial pattern obtained by removing the first partial pattern from the repetitive pattern; A holding unit that holds a bonded substrate that is bonded by aligning a pair of substrates, and a repeating pattern of the bonded substrate held by the holding unit that passes through one of the pair of substrates and generates diffracted light by the repeating pattern There is provided a bonding evaluation apparatus including a light source for irradiating light and a measurement unit for measuring the intensity of diffracted light generated by the repeated pattern of inspection light.

  Furthermore, as a fifth embodiment of the present invention, the above-described bonding evaluation apparatus, an alignment apparatus that aligns and bonds one of the pair of substrates to the other, and a pair of substrates are bonded by the alignment apparatus. There is provided a substrate bonding apparatus including a pressure device that pressurizes a bonding substrate and bonds the pair of substrates.

  The above summary of the present invention is not an exhaustive list of all features of the invention. Further, a sub-combination of these feature groups can be an invention.

  Hereinafter, the present invention will be described through embodiments of the invention. The embodiments described below do not limit the invention according to the claims. In addition, not all combinations of features described in the embodiments are essential for the solution of the invention.

  FIG. 1 is a plan view schematically showing the overall structure of the bonding apparatus 100. The bonding apparatus 100 includes a normal temperature part 102 and a high temperature part 202 formed inside a common casing 101.

  The normal temperature unit 102 has a plurality of substrate cassettes 111, 112, 113 and a control panel 120 facing the outside of the housing 101. The control panel 120 includes a control unit that controls the operation of the entire bonding apparatus 100. In addition, the control panel 120 has an operation unit that is operated by a user from the outside when the joining apparatus 100 is turned on, various settings, and the like. Further, the control panel 120 may include a connection unit that connects the other devices provided to the bonding apparatus 100.

  The substrate cassettes 111, 112, and 113 accommodate the substrates 180 bonded in the bonding apparatus 100 or the substrates 180 bonded in the bonding apparatus 100. The substrate cassettes 111, 112, and 113 are detachably attached to the housing 101. As a result, a plurality of substrates 180 can be loaded into the bonding apparatus 100 at once. Further, the substrates 180 bonded in the bonding apparatus 100 can be collected at a time.

  The substrate 180 loaded in the bonding apparatus 100 may be a single silicon wafer, compound semiconductor wafer, glass substrate, or the like, in which elements, circuits, terminals, and the like are formed. In addition, the loaded substrate 180 may be a laminated substrate that is already formed by laminating a plurality of wafers.

  The room temperature unit 102 includes a pre-aligner 130, an aligner 140, a holder stocker 150, an evaluation unit 160, and a pair of robot arms 171 and 172, which are arranged inside the housing 101. When the substrate is loaded on the aligner 140, the pre-aligner 130 provisionally aligns the positions of the individual substrates 180 so that each substrate 180 is loaded in a narrow adjustment range of the aligner 140 because of high accuracy. Thereby, the positioning of the substrate 180 in the aligner 140 is completed quickly and reliably.

  The aligner 140 includes an upper stage 141, a lower stage 142, and an interferometer 144. The upper stage 141 and the lower stage 142 carry the substrate 180 alone or the substrate holder 190 holding the substrate 180.

  In addition, a heat insulating wall 145 and a shutter 146 are disposed around the aligner 140. The space surrounded by the heat insulating wall 145 and the shutter 146 communicates with an air conditioner or the like and is precisely temperature-controlled, so that the alignment accuracy in the aligner 140 is maintained. The holder stocker 150 accommodates a plurality of substrate holders 190 and makes them stand by.

  Of the pair of robot arms 171 and 172, the robot arm 171 disposed on the side closer to the substrate cassettes 111, 112, and 113 transports the substrate 180 between the substrate cassettes 111, 112, 113, the pre-aligner 130, and the aligner 140. . The robot arm 172 disposed on the side far from the substrate cassettes 111, 112, 113 conveys the substrate 180 and the substrate holder 190 between the aligner 140, the evaluation unit 160 and the air lock 220. The robot arm 172 is also responsible for loading and unloading the substrate holder 190 with respect to the holder stocker 150.

  Further, the robot arm 172 has a function of turning over one of the substrates 180 to be joined. Note that the robot arms 171, 172, 230 adsorb and hold the substrate holder 190 by vacuum adsorption, electrostatic adsorption, or the like. Further, the substrate holder 190 attracts and holds the substrate 180 by, for example, electrostatic attraction. The structure and operation of the evaluation unit 160 will be described later separately.

  The high temperature unit 202 includes a heat insulating wall 210, an air lock 220, a robot arm 230, and a plurality of heating and pressing units 240. The heat insulating wall 210 surrounds the high temperature part 202 to maintain a high internal temperature of the high temperature part 202 and blocks heat radiation from the high temperature part 202 to the outside. Thereby, the influence which the heat of the high temperature part 202 has on the normal temperature part 102 can be suppressed.

  The robot arm 230 conveys the substrate 180 and the substrate holder 190 between any one of the heating and pressing units 240 and the air lock 220. The air lock 220 includes shutters 222 and 224 that open and close alternately on the normal temperature part 102 side and the high temperature part 202 side.

  When the substrate 180 and the substrate holder 190 are carried into the high temperature unit 202 from the normal temperature unit 102, first, the shutter 222 on the normal temperature unit 102 side is opened, and the robot arm 172 carries the substrate 180 and the substrate holder 190 into the air lock 220. . Next, the shutter 222 on the normal temperature part 102 side is closed, and the shutter 224 on the high temperature part 202 side is opened.

  Subsequently, the robot arm 230 unloads the substrate 180 and the substrate holder 190 from the air lock 220 and inserts them into one of the heating and pressing units 240. The heating / pressurizing unit 240 pressurizes the substrate 180 carried into the heating / pressurizing unit 240 while being sandwiched between the substrate holders 190. Thereby, the substrate 180 is permanently bonded.

  When the substrate 180 and the substrate holder 190 are carried out from the high temperature unit 202 to the normal temperature unit 102, the above series of operations are executed in reverse order. Through a series of these operations, the substrate 180 and the substrate holder 190 can be carried into or out of the high temperature part 202 without leaking the internal atmosphere of the high temperature part 202 to the normal temperature part 102 side.

  FIG. 2A to FIG. 2E are diagrams showing the transition of the state of the substrate 180 and the substrate holder 190 in the bonding apparatus 100 for bonding two substrates. Hereinafter, the operation of the bonding apparatus 100 will be described with reference to FIGS. 2a to 2e.

  As shown in FIG. 2a, at the beginning of the operation of the bonding apparatus 100, each of the substrates 180 to be bonded is individually accommodated. The substrate holder 190 is also individually accommodated in, for example, a holder stocker.

  When the bonding apparatus 100 starts operation, the substrates 180 are held one by one in the substrate holder 190 as shown in FIG. 2b.

  When the substrate holder 190 is the second one, as shown in FIG. 2 c, the stage is moved while monitoring the position with an interferometer, for example, so as to face the first substrate 180. Align. The aligned substrate 180 is temporarily bonded.

  As shown in FIG. 2D, the substrate 180 and the substrate holder 190 positioned in the aligner are connected by a plurality of stoppers 192 fitted in a groove 191 formed on the side surface, for example, to maintain the positioned state. The pair of connected substrates 180 and substrate holder 190 are conveyed together and inserted into the heating and pressing unit 240 as required.

  In this embodiment, the alignment accuracy in the aligner 140 is evaluated by the evaluation unit. In such a case, the substrate 180 that has been aligned and temporarily bonded in the aligner 140 is taken out of the substrate holder 190 before being carried into the heating and pressing unit 240 and carried into the evaluation unit. At this time, the substrate 180 and the substrate holder 190 are apparently in the form shown in FIG. 2e.

  FIG. 3 is a cross-sectional view schematically showing the structure of the evaluation gauge 18 used for evaluating the alignment accuracy in the aligner. The evaluation gauge 18 includes a transparent substrate 181, a semiconductor substrate 183, evaluation patterns 182 and 184, and alignment patterns 186 and 187.

  For the semiconductor substrate 183, a member equivalent to the substrate 180 to be bonded in the bonding apparatus 100 is used. Specifically, a Si wafer or the like can be arbitrarily selected. On the other hand, the transparent substrate 181 is formed of a material having a high transmittance with respect to inspection light described later. When using inspection light in the visible band, for example, a quartz substrate can be used. In addition, when using inspection light in the infrared band, a Si wafer can be used as the transparent substrate 181.

  In FIG. 3, the set of evaluation patterns 182 and 184 and the alignment patterns 186 and 187 are enlarged, but a plurality of sets of evaluation patterns 182, 184 and Alignment patterns 186 and 187 are formed.

  For example, the set of evaluation patterns 182 and 184 may have the same size as a single die of the stacked semiconductor device and may be arranged in the same manner as on the substrate 180. In addition, small evaluation patterns 182 and 184 may be formed in regions not used in the elements formed on the substrate 180.

  FIG. 4 is a plan view showing one set of evaluation patterns 182 and 184 and alignment patterns 186 and 187 extracted. Each of the evaluation patterns 182 and 184 includes a plurality of linear patterns formed at the same constant pitch 2p. The linear pattern has a line width such that, for example, the duty is 25% when viewed in the repeating direction of the evaluation patterns 182 and 184.

The alignment patterns 186 and 187 have shapes complementary to each other and are arranged in the vicinity of the evaluation patterns 182 and 184. Distance _{D 1} for the evaluation pattern 182 of one of the alignment pattern 186, with respect to the spacing _{D 2} of the evaluation pattern 184 of the other alignment pattern 187 has a difference in p corresponding to half the pitch 2p evaluation pattern 182 .

  FIG. 5 is a cross-sectional view of the evaluation patterns 182 and 184 when the transparent substrate 181 and the semiconductor substrate 183 are bonded. In the aligner 140, the transparent substrate 181 and the semiconductor substrate 183 are aligned so that the alignment patterns 186 and 187 coincide with each other, and are further pressed and temporarily joined.

When the transparent substrate 181 and the semiconductor substrate 183 are temporarily bonded, the alignment patterns 186 and 187 are integrated with each other. On the other hand, since the evaluation patterns 182 and 184 have different distances D ₁ and D ₂ with respect to the alignment patterns 186 and 187, linear patterns are alternately arranged.

  FIG. 6 is a plan view showing a state in which the synthesized evaluation patterns 182 and 184 are viewed from the transparent substrate 181 side. As shown, the alignment patterns 186 and 187 are integrated. On the other hand, the evaluation patterns 182 and 184 are nested with each other to form a composite pattern 188 that repeats a linear pattern at a pitch p. The transparent substrate 181 and the semiconductor substrate 183 including the composite pattern 188 formed in this way are used as the evaluation gauge 18 for evaluating the alignment accuracy of the aligner 140.

  FIG. 7 is a diagram schematically showing the structure of the evaluation unit 160 that evaluates the alignment accuracy of the aligner 140 using the evaluation gauge 18. An evaluation stage 168 on which the evaluation gauge 18 is mounted, an X drive unit 163, a Y drive unit 165, and a θ drive unit 167 that move or swing the evaluation stage 168, and a light source 162 that generates inspection light to be applied to the evaluation gauge 18. And an imaging unit 164 for observing the diffracted light generated in the inspection light by the evaluation gauge 18.

  The imaging unit 164 includes a curved reflecting mirror 169 that causes the inspection light generated by the light source 162 to be parallel light and incident on the evaluation gauge 18. Thereby, the inspection light is incident on each part of the evaluation gauge 18 at the same incident angle. Therefore, diffracted light is generated on the entire surface of the evaluation gauge 18 under the same conditions. Note that, as indicated by a dotted line in the figure, a plurality of imaging units 164 may be provided according to the order of the diffracted light to be observed.

  The evaluation gauge 18 is mounted on the evaluation stage 168 with the transparent substrate 181 side as an upper surface. The evaluation stage 168 also adsorbs and fixes the evaluation gauge 18 by electrostatic adsorption, vacuum adsorption or the like, similarly to the substrate holder 190 or the like. The light source 162 emits an inspection of a specific band or a specific wavelength designated in advance.

  The evaluation stage 168 is supported by the X driving unit 163, the Y driving unit 165, and the θ driving unit 167 stacked on each other. The X drive unit 163 moves the evaluation stage 168 along the guide unit 161 in the direction indicated by the arrow X in the drawing. The Y drive unit 165 moves the evaluation stage 168 on the X drive unit 163 in the direction indicated by the arrow Y in the drawing.

  Further, the θ drive unit 167 swings the evaluation stage 168 supported by the spherical seat on the Y drive unit 165. By the operations of the X driving unit 163, the Y driving unit 165, and the θ driving unit 167, the inspection light emitted from the light source 162 can be applied to the evaluation gauge 18 on the evaluation stage 168 at an arbitrary incident angle. Therefore, arbitrary diffracted light can be generated in the evaluation gauge 18 by changing optical conditions such as the irradiation angle of the inspection light.

  The light source 162, the curved reflecting mirror 169, and the imaging unit 164 may form a telecentric optical system. Thereby, the incident angles of the inspection light with respect to the evaluation gauge 18 and the imaging unit 164 are the same over the entire surface of the evaluation gauge 18, and the diffracted light can be observed at a time under uniform conditions. If the telecentric optical system is not used, the entire evaluation gauge 18 cannot be observed at a time under uniform conditions, but can be observed separately for each area of the evaluation gauge 18. However, the work efficiency of evaluation is reduced.

  FIG. 8 is a partial enlarged cross-sectional view showing a phenomenon that occurs in the evaluation gauge 18 irradiated with the inspection light. 8 is an enlarged view of the area surrounded by the dotted line A in FIG.

  As already described, each of the individual evaluation patterns 182, 184 and the composite pattern 188 forms a repetitive pattern having a constant pitch. Here, the evaluation patterns 182 and 184 each have a pitch 2p. When the transparent substrate 181 and the semiconductor substrate 183 of the evaluation gauge 18 are ideally aligned, the composite pattern 188 formed by synthesizing the evaluation patterns 182 and 184 is a repetitive pattern with a pitch p and a duty of 50%. Form.

Here, when the inspection light having the wavelength λ is incident on the repetitive pattern with the pitch p, the incident angle of the inspection light is represented by θ _i , the diffraction light angle is represented by ▲ ▼, and the diffraction order is represented by n, the following formula 1 is satisfied. Reflected diffracted light is generated under certain conditions.

Under the same conditions, the condition that the repetitive pattern with the pitch 2p generates diffracted light can be expressed as the following Expression 2.

On the other hand, the condition that the diffracted light of the inspection light incident on the repeating pattern with the pitch p cancels each other and does not generate diffracted light can be expressed by the following Equation 3.

  In the above formulas 2 and 3, for example, when the order n is n = 2m + 1, the two conditions are the same, and therefore, as shown in FIG. No order diffracted light is generated.

  FIG. 10 is a partially enlarged cross-sectional view showing a modification of the evaluation patterns 182 and 184. As shown in FIG. As shown in the figure, when the positions of the transparent substrate 181 and the semiconductor substrate 183 are shifted for some reason, the pitch of the composite pattern 188 formed by overlapping the evaluation patterns 182 and 184 each having the pitch 2p changes. That is, the pitch (p ± Δp) is obtained by increasing or decreasing the positional deviation amount Δp with respect to the ideal pitch p.

  FIG. 11 is a diagram showing the diffracted light intensity with respect to the pitch 2p generated by the evaluation patterns 182 and 184. As shown in FIG. As described above, when the pitch of the repetitive pattern of the pitch p is disturbed, the condition for canceling the first-order diffracted light is not satisfied, and the first-order diffracted light that has never existed appears. Therefore, by observing the change in the intensity of the first-order diffracted light, the presence or absence of misalignment can be determined, and the alignment accuracy in the aligner 140 can be evaluated. That is, the first-order diffracted light intensity with respect to the pitch 2p increases and the second-order diffracted light decreases.

  Further, in the evaluation unit 160, the θ driving unit 167 is set at an angular position where the first-order diffracted light with respect to the pitch 2p can be received, so that the distribution of the local misregistration amount can be grasped collectively in the wafer. When there is no position shift, zero-order first-order diffracted light appears, and it is most effective to observe changes in the first-order diffracted light. However, since the second-order diffraction decreases as the first-order diffracted light increases, the second-order diffracted light may be observed. Further, the decrease in the first-order diffracted light with respect to the pitch p corresponding to the increase in the displacement of the positions of the transparent substrate 181 and the semiconductor substrate 183 may be measured. Indistinguishable from Origami).

  FIG. 12 is a graph showing the results of calculating the relationship between the shift amount of the substrate 180 and the first-order diffracted light intensity with respect to the pitch 2p by the RCWA (Rigorous Coupled Wave Analysis) method. In the RCWA method, incident light as an electromagnetic wave is developed into a plane wave for each wavelength, and the diffraction efficiency is obtained analytically.

  Monochromatic light is preferably used as the inspection light, and E-ray (wavelength 546 nm (TE polarized light)) is used here. Further, a quartz wafer is used as the transparent substrate 181, and a Si wafer is used as the semiconductor substrate 183.

  The refractive index of the quartz wafer was 1.46 and the thickness was 725 μm. The evaluation patterns 182 and 184 formed on the transparent substrate 181 and the semiconductor substrate 183 are formed of a resist having a refractive index of 1.52 and a thickness of 200 nm. Further, the Si wafer had a refractive index of 4.09-0.03i and a thickness of 725 μm.

  As shown in the figure, attention is focused on the first-order diffracted light when the 2p period is 0.3 μm. If it is ideally aligned and there is no displacement, no first-order diffracted light is generated.

  However, in the region where the positional deviation amount exceeds 20 nm, the pitch of the composite pattern 188 is deviated from p due to the positional deviation, so that the first-order diffracted light appears clearly. Furthermore, the intensity of the first-order diffracted light increases as the amount of misalignment increases. Accordingly, by preparing the evaluation patterns 182 and 184 having a 2p period of 0.3 μm and observing the increase in the intensity of the first-order diffracted light in the evaluation unit 160, the amount of positional deviation can be grasped instantaneously.

  As described above, the first-order diffracted light intensity increases when the amount of positional deviation increases. For this reason, when the ideal alignment is performed, the first-order diffracted light does not appear at all. However, in order to effectively detect the first-order diffracted light, the emission intensity of the light source 162 or the sensitivity of the imaging unit 164 must be set appropriately.

  Therefore, when the positional deviation is evaluated with reference to the intensity of the first-order diffracted light, the reference pattern with the intentional pitch shifted is used as the evaluation pattern 182, so that it can be used when the light source 162 or the imaging unit 164 is calibrated. It is preferable to provide it together with 184. By providing such a reference pattern, it is possible to distinguish between a case where the imaging unit 164 is not sensitive for some reason and a case where the alignment is ideal.

  FIG. 13 is a graph showing the result of calculating the relationship between the shift amount of the substrate 180 and the intensity of the second-order diffracted light by the RCWA method. Note that the materials and specifications of the transparent substrate 181, the semiconductor substrate 183, and the evaluation patterns 182 and 184 used are the same as those in the calculation shown in FIG.

  Here, attention is focused on the second-order diffracted light when the 2p period is 0.7 μm. In the case of ideal alignment and no displacement, the composite pattern 188 formed by the evaluation patterns 182 and 184 forms a repetitive pattern with a pitch p. Corresponding to the composite pattern 188, strong second-order diffracted light is generated in the graph.

  However, in the region where the amount of misalignment exceeds 80 nm, the pitch of the composite pattern 188 deviates from p due to misalignment, so the second-order diffracted light is clearly reduced. Furthermore, the intensity of the second-order diffracted light decreases for a while as the positional deviation amount increases.

  Therefore, by preparing evaluation patterns 182 and 184 having a 2p period of 0.7 μm and observing a decrease in the intensity of the second-order diffracted light in the evaluation unit 160, the amount of positional deviation can be grasped instantaneously. Note that the evaluation unit 160 may be provided with a plurality of imaging units 164 to observe the first-order diffracted light intensity and the second-order diffracted light intensity simultaneously.

  FIG. 14 is a diagram schematically showing a specific method for evaluating the joining accuracy. That is, when attention is paid to specific diffracted light, the presence or absence of the positional deviation amount or the amount of the positional deviation amount is observed as a difference in brightness. Accordingly, as shown in FIG. 7, the entire substrate 180 to be evaluated is irradiated with inspection light, and the substrate 180 is collectively observed by the imaging unit 164, whereby the displacement amount distribution of the entire substrate 180 is distributed. Can be grasped instantly.

  In the example shown here, the amount of displacement is large in the element region 189 that appears bright. This makes it clear at a glance that the distribution of misalignment has a certain tendency.

  That is, the positional deviation amount to be evaluated can be quantified by previously theoretically and experimentally identifying the relationship between the diffraction intensity and the positional deviation amount with respect to the intended condition. Moreover, when using for a production line test | inspection, the acceptance criteria can be provided based on the relationship grasped | ascertained beforehand.

  FIG. 15 is a plan view showing another form of the evaluation patterns 182 and 184. As shown in the figure, in this evaluation gauge 18, two types of composite patterns in which the repeated directions of the composite patterns 188 are orthogonal to each other are mixed in one evaluation gauge. Thereby, the distribution of alignment accuracy in the aligner 140 can be grasped two-dimensionally.

  Note that the diffracted light is generated in the repeating direction of the composite pattern 188. Therefore, when simultaneously measuring the composite pattern 188 having different repeating directions, it is desirable to provide the evaluation unit 160 with a plurality of light sources 162 and a plurality of imaging units 164.

  Further, by irradiating the composite pattern 188 with inspection light vertically from a single light source 162, diffracted light can be simultaneously generated in the composite pattern 188 having different repeating directions. In this case, it is sufficient to provide a plurality of imaging units 164.

  Note that the types of repetition directions are not limited to two types orthogonal to each other. Therefore, a variety of evaluation patterns 182 and 184 having a repetitive direction corresponding to a stress having a high probability of occurring on the substrate 180 in the bonding apparatus 100 or the aligner 140 may be formed.

  FIG. 16 is a plan view showing another form of the evaluation patterns 182 and 184. In this embodiment, the composite pattern 188 formed by the evaluation patterns 182 and 184 is arranged on the scribe line generated between the element regions 189 where the elements are formed.

  With such an arrangement, the alignment accuracy can be evaluated using the composite pattern 188 in the manufacturing process of the stacked semiconductor device in which elements are formed in the element region 189. Further, since the scribe line disappears when the element region 189 is cut by dicing, the evaluation can be performed without reducing the utilization efficiency of the substrate 180.

  The arrangement of the composite pattern 188 is not limited to the scribe line. The composite pattern can be arranged not only on the outside of the element region 189 in the vicinity of the edge of the substrate 180 but also on an unused area inside the element region 189. Furthermore, the composite pattern 188 can be formed over the elements formed in the element region 189.

  In this way, by using the evaluation gauge 18 including a repeated pattern, the alignment accuracy distribution in the aligner 140 can be comprehensively and instantaneously grasped. Therefore, in the maintenance and inspection of the aligner 140, the state of the aligner can be grasped easily and quickly.

  Moreover, since the evaluation method using the evaluation patterns 182 and 184 requires a short time for the evaluation work, the alignment accuracy of the aligner 140 can be evaluated in the production process by mounting the evaluation unit 160 on the joining device 100. Therefore, it is possible to take measures such as loading the substrate 180 having undergone a remarkable positional shift into the aligner 140 again without sending it to the next step. Thereby, the yield of the material can be improved and the productivity of the stacked semiconductor device can be improved.

  Furthermore, the evaluation method using the evaluation patterns 182 and 184 can also be applied to inspection of a stacked semiconductor device that has become a product. That is, it is generated in the heating and pressurizing unit 240 by evaluating the stacked semiconductor device that is heated and pressed by the heating and pressurizing unit 240 in the bonding apparatus 100 and using the evaluation patterns 182 and 184. Can also be evaluated. In other words, the operation accuracy of the heating and pressing unit 240 can also be evaluated by the evaluation unit 160 and the evaluation method using the evaluation gauge 18.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

  The order of execution of each process such as operations, procedures, steps, and stages in the apparatus, system, program, and method shown in the claims, the description, and the drawings is particularly “before” or “prior to”. It should be noted that it can be realized in any order unless it is explicitly indicated as “,” etc., and the output of the previous process is not used in the subsequent process. Regarding the operational flow of the claims, the description, and the drawings, even if it is described using “first”, “next”, etc. for the sake of convenience, it means that it is essential to carry out in this order. is not.

1 is a plan view schematically showing the entire structure of a bonding apparatus 100. FIG. It is a figure which shows typically the transition of the state of the board | substrate 180. FIG. It is a figure which shows typically the transition of the state of the board | substrate 180. FIG. It is a figure which shows typically the transition of the state of the board | substrate 180. FIG. It is a figure which shows typically the transition of the state of the board | substrate 180. FIG. It is a figure which shows typically the transition of the state of the board | substrate 180. FIG. It is sectional drawing of the evaluation patterns 182 and 184. It is a top view of evaluation patterns 182 and 184. It is sectional drawing of the synthesis | combination evaluation patterns 182 and 184. FIG. It is a top view of synthesized evaluation patterns 182 and 184. It is a figure which shows the structure of the evaluation part 160 typically. It is a partial expanded sectional view which shows the phenomenon in the evaluation patterns 182 and 184. It is a figure which shows the diffracted light intensity produced by the evaluation patterns 182 and 184. It is a partial expanded sectional view showing modification of evaluation patterns 182 and 184. It is a figure which shows the diffracted light intensity produced by the evaluation patterns 182 and 184. It is a graph which shows the relationship between the deviation | shift amount of the board | substrate 180, and 1st-order diffracted light intensity. It is a graph which shows the relationship between the deviation | shift amount of the board | substrate 180, and a 2nd-order diffracted light intensity. It is a figure which shows typically the evaluation method of joining precision. It is a top view which shows the other form of the evaluation patterns 182 and 184. It is a top view which shows the other form of the evaluation patterns 182 and 184.

Explanation of symbols

18 Evaluation gauge, 100 Bonding device, 101 Case, 102 Room temperature part, 111, 112, 113 Substrate cassette, 120 Control panel, 130 Pre-aligner, 140 Aligner, 141 Upper stage, 142 Lower stage, 144 Interferometer, 145, 210 Thermal insulation Wall, 146, 222, 224 Shutter, 150 Holder stocker, 160 Evaluation unit, 161 Guide unit, 163 X drive unit, 162 Light source, 164 Imaging unit, 165 Y drive unit, 167 θ drive unit, 168 Evaluation stage, 169 Curved surface reflection Mirror, 171, 172, 230 Robot arm, 180 substrate, 181 Transparent substrate, 182, 184 Evaluation pattern, 183 Semiconductor substrate, 186, 187 Alignment pattern, 188 Compound pattern, 189 Element area, 190 Substrate holder, 191 Groove, 92 stop, 202 high temperature section, 220 airlock 240 pressurizing and heating unit

Claims (18)

A first partial pattern forming step of forming a first partial pattern forming a part of a repetitive pattern on one of a pair of substrates;
A second partial pattern forming step of forming a second partial pattern on the other of the pair of substrates by removing the first partial pattern from the repetitive pattern;
A bonding step of bonding the pair of substrates;
An inspection light irradiation step of irradiating inspection light transmitted through one of the pair of substrates to the repetitive pattern formed by the bonding in the bonding step;
A measurement stage for measuring the intensity of diffracted light generated in the repetitive pattern;
A bonding evaluation method including an evaluation step of evaluating a positional deviation between the pair of substrates based on the state of the light measured in the measurement step.   The bonding evaluation method according to claim 1, further comprising a step of detecting whether or not the pair of substrates is misaligned based on the presence or absence of the diffracted light to be measured.   The bonding evaluation according to claim 1, further comprising a step of detecting a positional deviation amount of the pair of substrates based on the measured intensity of the diffracted light with reference to a relationship between the diffracted light intensity and the positional deviation amount obtained in advance. Method.   The bonding evaluation method according to claim 1, wherein the inspection light is irradiated under conditions that generate diffracted light by the repetitive pattern.   5. The bonding evaluation method according to claim 4, wherein the measuring step measures an increase in first-order diffracted light intensity with respect to a pitch twice the pitch of the repetitive pattern, which is generated by the repetitive pattern.   5. The bonding evaluation method according to claim 4, wherein the measuring step measures a decrease in second-order diffracted light intensity with respect to a pitch twice the pitch of the repetitive pattern caused by the repetitive pattern.   The bonding evaluation method according to claim 4, wherein in the measuring step, a decrease in the first-order diffracted light with respect to the pitch of the repetitive pattern caused by the repetitive pattern is measured.   The joint evaluation method according to any one of claims 1 to 7, wherein the step of illuminating the inspection light and the step of measuring the diffracted light use a telecentric optical system.   The joining evaluation method according to claim 1, wherein the measuring step is executed after the joining step.   The bonding evaluation method according to any one of claims 1 to 8, wherein the measuring step is performed on the pair of substrates heated and pressurized after the bonding step. A pair of substrates composed of one substrate having a first partial pattern forming a part of a repeated pattern and another substrate having a second partial pattern obtained by removing the first partial pattern from the repeated pattern are bonded to each other. A substrate holding unit for holding the bonded substrate;
An illumination unit that irradiates the repeating pattern of the bonding substrate held by the holding unit with light that passes through one of the pair of substrates;
A joint evaluation device comprising: a measurement evaluation unit that measures the intensity of diffracted light generated in the repetitive pattern and observes a change in the intensity of the diffracted light to evaluate alignment accuracy . The bonding evaluation apparatus according to claim 11, wherein the illumination unit irradiates the repetitive pattern with diffracted light, and the measurement evaluation unit measures the intensity of the diffracted light generated in the repetitive pattern. The measurement evaluation unit
An odd-order diffracted light measuring unit for measuring odd-order diffracted light generated by the repetitive pattern;
The bonding evaluation apparatus according to claim 12, further comprising: an even-order diffracted light measurement unit that measures even-order diffracted light generated by the repetitive pattern. The illumination unit irradiates the light along a normal line of the bonding substrate,
The bonding evaluation apparatus according to claim 11, wherein the measurement evaluation unit measures the diffracted light from different directions intersecting each other. The bonding evaluation apparatus according to any one of claims 11 to 14,
An alignment device for aligning one of the pair of substrates with respect to the other;
A pressure device that pressurizes and bonds the pair of substrates aligned by the alignment device;
A substrate bonding apparatus comprising:   The alignment device is adjusted so that alignment accuracy of the pair of substrates satisfies a given evaluation standard when the bonding evaluation device evaluates the pair of substrates aligned by the alignment device. The substrate bonding apparatus according to claim 15.   When the pair of substrates aligned by the alignment device is evaluated by the bonding evaluation device, the pressurizing device is concerned when the alignment accuracy of the pair of substrates satisfies a given evaluation criterion. The board | substrate bonding apparatus of Claim 15 or Claim 16 which bonds a pair of board | substrate.   The board | substrate bonding apparatus of any one of Claim 15 to 17 in which the said evaluation apparatus evaluates the joining precision of the said pair of board | substrate bonded by the said pressurization apparatus.

Images