Determining bond quality of thermo-compression bonds between dies and substrates based on solder surface tension

Abstract

One or more forces are determined that correspond to surface tension of solder between a die and a substrate during a bonding process while the solder is in a liquid phase. The forces are compared with one or more expected forces that correspond to expected surface tension of the solder between the die and the substrate during the bonding process while the solder is in the liquid phase. Based on the comparison, a bond quality is determined. By quantifying and evaluating the surface tension of the liquid solder, the bond quality can be determined during the bonding process, or at least shortly thereafter, without destructive testing or sampling. Accordingly, bond quality may increase while also minimizing time, materials, and / or cost compared with traditional selective and / or destructive testing.

Description

Docket: SSMP 43930DETERMINING BOND QUALITY OF THERMO-COMPRESSION BONDS BETWEEN DIES AND SUBSTRATES BASED ON SOLDER SURFACE TENSIONFIELD

[0001] This disclosure is directed to determining bond quality of thermo-compression bonds, especially those between dies and substrates.BACKGROUND

[0002] Thermo-compression or thermal-compression bonding is a process by which a die is attached to a substrate via solder to form a package. Prior to the bonding process, solder is placed on pads of the die and / or the substrate. The die and the substrate are placed proximate to one another, and the solder is heated to at least its melting point. The die and substrate are then brought together such that the liquid solder is in contact with the pads of the die and the substrate (e.g., wetting). The solder is then cooled to form a bond between the pads of the die and the pads of the substrate, thereby forming the package.

[0003] Thermo-compression bonding provides a mechanical and electrical connection between the die and substrate. Factors such as improper wetting (e.g., due to oxide layers or residuals on the solder and / or pads) or improper relative movement between the die and the substrate during thermo-compression bonding can affect a quality of the connection (e.g., lowered mechanical strength, electrical performance, and / or long-term reliability). A low-quality connection can lead to the package underperforming or even failing, which can result in decreased yields and / or customer satisfaction.SUMMARY

[0004] A method of determining a bond quality of a thermo-compression bond between a die and a substrate based on solder surface tension is described herein. The method includes determining, based on a force signal received from a force sensor, one or more forces generated by surface tension of solder between the die and the substrate during a bonding process while the solder is in a liquid phase. The method also includes determining one or more expected forces based on expected surface tension of the solder between the die and the substrate during theDocket: SSMP 43930bonding process while the solder is in the liquid phase. The method further includes comparing the one or more forces to the one or more expected forces and determining, based on the comparing, the bond quality.

[0005] A die bonder configured to perform the method discussed above is also described herein. The die bonder includes a force sensor, a displacement sensor, and a processing unit configured to perform the method discussed above.

[0006] A system configured to perform the method discussed above is also described herein. The system includes a processing unit configured to perform the method discussed above. The system may be included within a computing device remote to the die bonder, a cloud computing device, the die bonder, or some combination thereof.

[0007] The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description. In the drawings, like reference numbers indicate identical or functionally similar elements.BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. 1A illustrates a perspective view of an example of a bond head, in accordance with this disclosure, and associated components for bonding.

[0009] FIG. 1B illustrates a side view of the bond head and associated components of FIG. 1A.

[0010] FIG. 2 illustrates the bond head and associated components of FIG. 1 during a bond phase.

[0011] FIG. 3A illustrates example shapes of a solder joint at different displacements during the bond phase, in accordance with this disclosure.

[0012] FIG. 3B illustrates an expected force vs. displacement curve and two example measured force vs. displacement curves, in accordance with this disclosure.

[0013] FIG. 4 illustrates an example process flow of determining bond quality of thermocompression bonds between dies and substrates based on solder surface tension.

[0014] FIG. 5 illustrates details of a portion of FIG. 4.Docket: SSMP 43930

[0015] FIG. 6 illustrates an example system configured for determining bond quality of thermocompression bonds between dies and substrates based on solder surface tension.

[0016] FIG. 7 illustrates an example method of determining a bond quality of a thermocompression bond between a die and a substrate based on solder surface tension.DETAILED DESCRIPTION

[0017] Ensuring quality of thermo-compression bonds (e.g., strength of solder connections) within packages is important for many reasons. For example, poor quality bonds can lead to decreased mechanical or electrical performance of packages or even failure of packages (often referred to as bad or faulty packages). Conventional means of testing bond quality are often only performed on a sample of assembled packages and at some time after the bonding is completed (e.g., on separate equipment, in a downstream process, etc.). For example, conventional tests often require a package to be destroyed (e.g., by pulling the die and the substrate apart or by slicing the package). Accordingly, in a running production flow, because of not testing all of the packages and the time delay between bonding and testing, many further packages may be produced by the time a test identifies a bond problem (e.g., low bond quality). Such further packages may be rejected, which wastes time and money.

[0018] One or more methods, apparatuses, and systems for determining bond quality of thermocompression bonds based on surface tension are described herein. One or more forces are determined that correspond to surface tension of solder between a die and a substrate during a bonding process while the solder is in a liquid phase. The forces are compared with one or more expected forces that correspond to expected surface tension of the solder between the die and the substrate during the bonding process while the solder is in the liquid phase. Based on the comparison, a bond quality is determined.

[0019] By quantifying and evaluating the surface tension of the liquid solder, the bond quality can be determined during the bonding process, or at least shortly thereafter, without destructive testing or sampling. Accordingly, bond quality may increase while also minimizing time, materials, and / or cost compared with traditional selective and / or destructive testing.

[0020] In the following description, numerous specific details are set forth, such as particular structures, components, materials, dimensions, processing steps and techniques, in order toDocket: SSMP 43930provide an understanding of the various embodiments of the present application. However, it will be appreciated by one of ordinary skill in the art that the various embodiments of the present application may be practiced without these specific details. In other instances, well-known structures or processing steps have not been described in detail in order to avoid obscuring the present application.

[0021] FIG. 1A illustrates a perspective view of an example of a bond head 102, in accordance with this disclosure, and associated components for bonding (e.g., a die 108 and substrate 110). FIG. 1B illustrates a side view of the bond head 102 and the associated components. The bond head 102 is part of a die bonder (not shown). FIG. 1A and FIG. 1B show the bond head 102 prior to a bonding phase of the die bonder.

[0022] The bond head 102 is configured to move along an axis 104. For example, the axis 104 may correspond to a z-axis of the die bonder and may be oriented vertically (assuming the bond is in the up / down direction). The die bonder may move the bond head 102 via coarse and fine movements along the axis 104, and the coarse and fine movements may be referred to as different axes (e.g., z-axis and w-axis). The bond head 102 may also be configured to translate along other axes (e.g., orthogonal to the axis 104) and / or rotate around the axis 104 or other axes.

[0023] The bond head 102 includes a bond tool 106 that is configured to attach to, and disengage from, a die 108 (e.g., chip, semiconductor, wafer, package, semiconductor die, die stack, semiconductor package, etc., or any combination thereof). For example, the bond tool 106 may be configured to pull a vacuum therein to couple the bond tool 106 to the die 108. Conversely, the bond tool 106 may be configured to be pressurized to release the die 108 from the bond tool 106.

[0024] The bond head 102 is positioned (e.g., via other axes) such that the die 108 is oriented with a substrate 110 (e.g., wafer, lead frame, organic substrate, glass carrier, printed circuit board, interposer, ceramic, panel, etc., or any combination thereof. Oriented, in the context of this disclosure, means that die pads 109 of the die 108 are aligned with respective substrate pads 112 of the substrate 110.

[0025] Attached to one or more of the die 108 or the substrate 110 are a plurality of solder balls 114. The solder balls 114 may be attached to the die pads 109, to the substrate pads 112, orDocket: SSMP 43930some combination thereof. Each type of die 108 / substrate 110 package (e.g., product, recipe, combination, etc.), has a known configuration of the solder balls 114. The configuration may include numbers and sizes (diameters and / or volumes) of the solder balls 114. The solder balls 114 may have a uniform size or may have different sizes. Furthermore, the solder balls 114 may be disposed in any quantity, configuration, and / or array without departing from the scope of this disclosure. It should be noted that the solder balls 114 are, in most cases, not completely spherical. Solder ball, as used herein, refers to a specific amount of solder in a particular location (e.g., attached to one or more pads) in any shape it assumes before, during, and after the bond process.

[0026] To measure the surface tension generated by the solder balls 114 when they are in a liquid phase and in contact with both the die 108 and the substrate 110, the die bonder has at least one force sensor 116. The force sensor 116 is configured such that forces due to the surface tension of the solder balls 114 may be determined. Multiple force sensors 116 may be used with respective sensing ranges to cover a broad range of configurations of the solder balls 114.

[0027] In the illustrated example, the force sensor 116 is disposed in the bond head 102 above the bond tool 106; however, the force sensor 116 may be anywhere between the bond tool 106 and an end effector of an actuator of the axis 104. In some implementations, the force sensor 116 may not exist in the bond head 102 at all and may instead be comprised within a motor controller that controls the axis 104. For example, the force sensor 116 may correspond to (or be determined from) a current being fed to the actuator of the axis 104.

[0028] FIG. 2 illustrates the bond head 102 and the associated components during a bond phase of the die bonder. The bond phase, for the purpose of this disclosure, refers to a time when the solder balls 114 are in contact with die pads 109 of the die 108 and substrate pads 112 of the substrate 110, and the solder balls 114 may be in a liquid phase or at least partially solidified. Although the term solder ball is used herein, the shape of the solder may take any shape during and after the bond phase.

[0029] As discussed above, the bond head 102 is configured to move along the axis 104. The force sensor 116 is configured to determine forces between the die 108 and the substrate 110 due to surface tension of the solder balls 114 while the solder balls 114 are in the liquid state.Docket: SSMP 43930

[0030] A displacement 200 exists between pads of the die 108 and the substrate pads 112 of the substrate 110 that varies as the bond head 102 moves along the axis 104. For the purposes of this disclosure, the movement of the bond head 102 is measured by the displacement 200. When the solder balls 114 are in the liquid phase and in contact with the die pads 109 and the substrate pads 112, the displacement 200 corresponds to a height of the solder balls 114 in whatever form / shape they are in.

[0031] The solder balls 114, in the illustrated example, are slightly compressed. That is, the displacement 200 is less than the original diameter of the solder balls 114. As will become apparent, the shape of the solder balls 114 may vary with the displacement 200 during the bond phase.

[0032] FIG. 3 A illustrates a plurality of example solder shapes 300 (e.g., solder shapes 300a-300d) between the die pads 109 and the solder pads 112 with corresponding displacements 200 (e.g., displacements 200a-200d). The solder shapes 300 and corresponding displacements 200 may vary without departing from the scope of this disclosure.

[0033] During the bond phase, the bond head 102 is moved such that the solder balls 114 are compressed. Doing so allows for wetting of the solder balls 114 (e.g., adhering of the solder balls 114 to the die pads 109 and the substrate pads 112). After that, the bond head 102 is moved to increase the displacement 200 until a final position where the die 108 is released. Between those points, various solder shapes 300 may exist.

[0034] The solder shape 300a corresponds to a displacement 200a. The displacement 200a may correspond to a wetting position of the bond phase (e.g., a location corresponding to the compression of the solder ball 114). The solder shape 300a at the displacement 200a may have a surface tension that corresponds to a repulsive force exerted on the pads (e.g., pushing the pads apart).

[0035] The solder shape 300b corresponds to a displacement 200b that is greater than the displacement 200a. The displacement 200b may correspond to an initial raising of the bond head 102 from the wetting position during the bond phase. The solder shape 300b at the displacement 200b may correspond to a “natural” shape with very little to no repulsive or attractive forces exerted on the pads.Docket: SSMP 43930

[0036] The solder shape 300c corresponds to a displacement 200c that is greater than the displacement 200b. The displacement 200c may correspond to further raising of the bond head 102 during the bond phase. The solder shape 300c at the displacement 200c may correspond to a high attractive force exerted on the pads (e.g., pulling the pads together).

[0037] The solder shape 300d corresponds to a displacement 200d that is greater than the displacement 200c. The displacement 200d may correspond to a position of the bond head 102 near a release position (e.g., where the die 108 is released by the bond tool 106). The solder shape 300d at the displacement 200d may correspond to a lower attractive force exerted on the pads than at the displacement 200b. This may be due to the mid-section “necking” down to form an hourglass shape. It should be noted, that even if the die 108 is released when the solder balls 114 have an hourglass shape, the surface tension (e.g., attractive force) along with gravity may pull the die 108 and substrate 110 together. Accordingly, the final bond may not take the hourglass shape.

[0038] FIG. 3B illustrates an expected force vs. displacement curve 302 along with two example measured force vs. displacement curves 304 (e.g., force vs. displacement curves 304a and 304b). The displacement axis corresponds to the displacement 200 and the force axis corresponds to forces determined by the force sensor 116. The expected force vs. displacement curve 302 may correspond with theoretical, optimal, and / or expected force values as a function of displacement for a set of solder parameters (e.g., solder ball volume and count). Scales are not shown on the axis in an effort to generalize the curve for any set of solder parameters. It should be apparent that each set of solder parameters may produce a unique curve. As examples, a die size of 1 mm by 1 mm may have a maximum tensile force around 1 g (for typically around 500 pads), while a die size of 50 mm by 50 mm may have a maximum tensile force around 2.5 kg (for typically around 1.5 million pads).

[0039] The displacement axis may be set such that a zero point 306 corresponds to a displacement where there is no force as the bond head 102 is raised after the solder balls 114 have been compressed. In other words, there is a transition between when the force is compressive and when the force is tensile. That point may become the zero point for the displacement axis. For example, the zero point 306 may correspond to a displacement 200 that produces solder shape 300b or a similar solder shape. Solder shape 300a may correspond to aDocket: SSMP 43930portion left of the zero point 306 (e.g., in compression), and solder shapes 300c and 300d may correspond to a portion right of the zero point 306 (e.g., in tension). A force below the zero point 306 indicates an attractive or tensile force generated by the solder balls 114 between the pads and a force above the zero point 306 indicates a repulsive or compression force generated by the solder balls 114 against the pads. A displacement above the zero point 306 indicates an increasing of the displacement 200 from the zero point 306.

[0040] Each of the curves has a minimum (e.g., maximum tensile force) for the span of displacement (e.g., between the zero point 306 and a point in which the die 108 is released). For example, the expected force vs. displacement curve 302 has an expected minimum 308. The measured force vs. displacement curves 304a and 304b have minimums 310a and 310b, respectively.

[0041] A measured force vs. displacement curve 304 may be compared with the expected force vs. displacement curve 302 to determine bond quality. For example, measured force vs. displacement curve 304a may correspond with a poor bond quality because the forces therein are well above, or lower in tension than, those of the expected force vs. displacement curve 302 for most displacements. The measured force vs. displacement curve 304b may correspond with a good bond quality (e.g., the forces are generally within a margin of the expected forces). Any number of points and / or portions of the curves may be compared to determine the bond quality. For example, the curves may be compared holistically, portions of the curves, a plurality of individual points, or single points may be compared.

[0042] FIG. 4 illustrates an example process flow of determining bond quality of thermocompression bonds between dies and substrates based on solder surface tension. The process flow is illustrated as being within a die bonder 401, which may be an example of the die bonder discussed above (e.g., that which houses the bond head 102). However, the process flow, or a portion thereof, may be performed in another device communicatively coupled with the die bonder 401. Furthermore, one or more of the described components, data, signals, or modules may be external to the die bonder 401 without departing from the scope of this disclosure.

[0043] The process flow (including further detail described in regard to FIG. 5) includes various modules that are implemented at least partially in hardware. That is, each of the modulesDocket: SSMP 43930described below may be implemented by one or more processing units (e.g., by executing code corresponding to the respective modules) and / or hardware components within one or more respective systems. For example, a module shown and / or described as being within the die bonder 401 may be implemented by a processing unit of the die bonder 401. If a module is described as being in a different system, then the associated processing unit may be within the different system. If a module may be split amongst multiple systems, then associated processing units may be within both systems.

[0044] The die bonder 401 includes the force sensor 116. As discussed above, the force sensor 116 is configured to determine forces between the die 108 and the substrate 110 that are caused by the solder balls 114 (e.g., due to surface tension thereof). The force sensor 116 is configured to generate a force signal 400 that is a signal corresponding to the force acting on the solder balls 114 (compressive or tensile acting on the solder balls 114 from the die pads 109 and the substrate pads 112). The force sensor 116 may adjust for various weights (e.g., of the bond head 102, the die 108, etc.) and / or apply an offset (e.g., to zero the force signal 400 and / or to move the zero point 306), or a receiving module may compensate for them.

[0045] The die bonder 401 also includes a displacement sensor 402 that is configured to generate a displacement signal 404 (e.g., a signal corresponding to the displacement 200). The displacement sensor 402 may be any device capable of directly determining the displacement 200 or capable to provide information for determining the displacement 200. For example, the displacement sensor 402 may be an encoder or metrology device attached to the bond head 102 or a drive system corresponding to the axis 104. The displacement sensor 402 may be a standalone device used solely for bond quality testing or may be used as part of other systems of the die bonder 401. The displacement signal 404 is a signal corresponding to the displacement 200 between the die pads 109 and the substrate pads 112. The displacement sensor 402 may adjust for various offsets (e.g., to move the zero point 306), or a receiving module may compensate for them.

[0046] The force signal 400 and the displacement signal 404 are received by a bond quality module 406. The bond quality module 406 may be within the die bonder 401 (as illustrated), within another system, or some combination thereof (e.g., one portion within the die bonder 401 and another portion within another system). Thus, the die bonder 401 may include a processingDocket: SSMP 43930unit (not shown) to implement the bond quality module 406 or a portion thereof. Alternatively, or additionally, another system may include a processing unit to implement the bond quality module 406 or a portion thereof.

[0047] The bond quality module 406 contains a bond quality determination module 408 that is configured to determine a bond quality 410 of the solder balls 114 between the die 108 and the substrate 110. The bond quality determination module 408 may generate a measured force vs. displacement curve 304 and determine one or more values of the bond quality 410 based on it (e.g., maximum tensile force, maximum slope, slope gradient, etc.). Furthermore, the bond quality determination module 408 may compare the measured force vs. displacement curve 304 with an expected force vs. displacement curve 302 to determine the bond quality 410. For example, the bond quality 410 may be a percentage of expected force, a curve fit, an integral difference, or some other metric of the comparison. A metric of the bond quality 410 and / or some value of the measured force vs. displacement curve 304 may be compared with a threshold value to determine another metric. For example, a result of the comparison between the expected and measured forces may cause the bond quality determination module 408 to assign a pass or fail to the bond.

[0048] The bond quality 410 may be received by an action module 412. The action module 412 is configured to perform an action 414 based on the bond quality 410. For example, the action module 412 may simply output the bond quality 410 as a characteristic of the product. As another example, the action module 412 may, if the bond quality 410 indicates a low-quality or failing bond, the action 414 may be to cause the die bonder 401 to enter an error state, cause a rebond of the product (e.g., repeat the bonding process to determine a second / updated / new bond quality), change a surface treatment, or any number of other mitigation techniques.

[0049] It should be noted that any number of factors may contribute to the bond quality 410 indicating a low-quality or failing bond. For example, improper wetting of one or more of the solder balls 114, defects in one or more of the solder balls 114 (e.g., inclusion of foreign particles, inclusion of air or gas, incorrectly sized, incorrectly placed), and / or the displacement 200 in any of the process steps being incorrect can lead to low-quality bonds. Thus, when the bond quality 410 indicates a low-quality bond, steps may be taken to evaluate the root cause (if the low-quality is not an outlier) and determine an appropriate mitigation technique.Docket: SSMP 43930

[0050] FIG. 5 illustrates a process flow of the bond quality determination module 408. As discussed above, the bond quality determination module 408 is configured to determine the bond quality 410. Reference is made to FIG. 3B to show examples of how the bond quality determination module 408 may determine the bond quality 410.

[0051] To determine the bond quality 410, a comparison module 500 compares a measured force vs. displacement curve 304 with an expected force vs. displacement curve 302. The expected force vs. displacement curve 302 is based on expected surface tension and corresponds to solder parameters 502. The solder parameters 502 include aspects of the solder balls 114, such as a volume of each of the solder balls and a count of the solder balls 114. Thus, for each configuration of the solder balls 114, there is a corresponding expected force vs. displacement curve 302.

[0052] The expected force vs. displacement curve 302 is based on a model 504 using the solder parameters 502. The model 504 is based on each of the solder balls 114 forming a barrel shape when connected with pads of the die 108 and the substrate 110. The barrel shape includes a height h between the pads (e.g., the displacement 200), a diameter d at the pads, a diameter D at the midsection, a first radius Ri corresponding to a curvature, a second radius R2 corresponding to a distance from a center of to an extent of the midsection, a distance a from an endpoint of Ri to an extent of d, and a distance z that is a difference between Ri and a. It should be noted that R2 equals D / 2, z equals (D-d) / 2, and RI equals a+z.

[0053] The volume V of the solder ball 114 (barrel) may be given according to Equation 1.V = — (2D2+ d2) (1)127

[0054] Solving Equation 1 for D gives Equation 2.D =— — 0.5d2(2) Tth

[0055] Since R2 is equal to D / 2, R2 may be given according to Equation 3.R2z= - l— — 0.5d2(3) 2 y nh

[0056] Ri may be determined via the Pythagorean theorem according to Equation 4 using a and h / 2 as the sides of the triangle with Ri being the hypotenuse.R12= (h / 2)2+ a2(4)Docket: SSMP 43930

[0057] Since z equals (D-d) / 2 and Ri equals a+z, Equation 4 may be solved for Ri with substitutions to give Equation 5.

[0058] The Young-Laplace equation describes the pressure difference across a fluid interface (e.g., exerted on the pads) due to surface tension according to Equation 6.= -r Q- - ^-) (6)r<2' where Ap is the pressure exerted on the pads and y is the surface tension of the solder in the liquid phase.

[0059] Replacing Ri and R2 and multiplying Ap by the area of the pad contact (7td2 / 4) and a count of the solder balls 114 n gives the Force F according to Equation 7, which may be considered a surface tension function.„ -ynnd2I 2z 2 1 F = - - (7)4\ © + ^-0.5d2 / ~0.5d —dwhere z = — -. Thus, the expected forces can be determined for a range of h (e.g., displacements 200) based on values of the volume V, the diameter d of the pad contacts (which may correspond to a maximum diameter afforded by the pads), the surface tension y of the solder, and the count n of solder balls 114.

[0060] Equation 7 assumes that the solder balls 114 have a same size. If the solder balls 114 have different / varying sizes, the expected total forces Ftotai can be determined by using Equation 7 for each solder ball size and summing them according to Equation 8.Ftotai = iFi(Vi,di,ni) (8) where / is the set of solder ball sizes within the package, E is the volume of each solder ball of size i, di is the diameter of the pad contacts for size i. and n, is the count of solder balls of size i,

[0061] The expected force vs. displacement curve 302 may be selected from a database 506 of expected force vs. displacement curves 302 (e.g., expected force vs. displacement curves 302a-302n). Each of the entries in the database 506 may correspond to a set of solder parameters 502. In other words, expected force vs. displacement curve 302a may correspond to solderDocket: SSMP 43930parameters 502a, and expected force vs. displacement curve 302b may correspond to solder parameters 502b. The bond quality determination module 408 may use the solder parameters to select the appropriate expected force vs. displacement curve 302.

[0062] The expected force vs. displacement curves 302 may be determined using Equation 7 or 8 for respective solder parameters 502. The expected force vs. displacement curves 302 may also be determined using a different model (e.g., other than that shown in FIG. 5 and / or other than Equations 7 or 8). Furthermore, the expected force vs. displacement curves 302 may also be determined via experimental testing. For example, many experimental force vs. displacement curves 302 may be determined for a plurality of packages sharing a set of solder parameters 502. The bonds thereof may be evaluated, and correlations may be made between good bonds and good force vs. displacement curves 302. The correlations may be used to determine an expected force vs. displacement curve 302 for the set of solder parameters 502.

[0063] The bond quality determination module 408 may generate the expected force vs. displacement curves 302a-302n, or they may be generated by a separate module and / or device and loaded into the die bonder 401. If the database 506 is not used and / or the expected force vs. displacement curve 302 is not pre-generated or received by the bond quality determination module 408, the bond quality determination module 408 may generate it using the solder parameters 502 and Equation 7.

[0064] Accordingly, with the solder parameters 502 as an input, the bond quality determination module 408 can generate the expected force vs. displacement curve 302 or select the expected force vs. displacement curve 302 from the database 506 of expected force vs. displacement curves 302. Although curve is used herein, the expected force vs. displacement curves 302 and / or the measured force vs. displacement curve 304 may comprise one or more ordered pairs of forces and corresponding displacements.

[0065] Returning to the comparison module 500, the measured force vs. displacement curve 304 is compared with the expected force vs. displacement curve 302 to determine the bond quality 410. Any number of points and / or values may be compared. For example, although curves are used herein, the comparison module 500 may determine single force such as a minimum measured force (e.g., maximum tensile force) during the bond phase and compare that to a minimum expected force for the corresponding solder parameters 502 (e.g., a minimization ofDocket: SSMP 43930Equation 7). In such implementations, the database 506 may include a minimum force value instead of curves and / or a plurality of displaccmcnt / forcc pairs. Furthermore, the displacement 200 may not be used for such a comparison. In any of the implementations, a smoothing algorithm may be used on the measured force(s) and / or curve prior to comparison to expected values.

[0066] In other implementations, the displacements 200 or height h may be factored by the comparison module 500. For example, the comparison module 500 may determine an expected slope from the zero point 306 to the expected minimum 308 and compare that with a measured slope from a point of zero force to the measured minimum measured force 310.

[0067] In other implementations, more complicated analysis may be performed. For example, if curves are used, the comparison module 500 may perform a curve fit or other matching algorithm to determine how well the measured force vs. displacement curve 304 matches the expected force vs. displacement curve 302. Furthermore, the comparison module 500 may integrate the curves to determine areas of difference between them. The comparison module 500 can then determine the bond quality 410 based on the areas between the curves.

[0068] It should be noted that any portion of the curves may be used by the comparison module 500. For example, the comparison module may compare curve portions from the zero point 306 to displacements corresponding to minimum forces. Looking at measured force vs. displacement curve 304b, the forces are below the expected forces (overperforming) up to the minimum values. If, however, the points right of the minimums are used, then the measured force vs. displacement curve 304b has forces that are above the expected forces (underperforming).Measured force vs. displacement curve 304a underperforms the expected force vs. displacement curve 302 throughout most of the displacement range.

[0069] Accordingly, the comparison module 500 may use any number of data points and / or comparison methods to determine the bond quality 410. For example, the comparison module 500 may compare a minimum expected force with a minimum measured force while the solder is in the liquid phase. The comparison module 500 may also determine an expected displacement for the minimum expected force and compare that to a measured displacement at the minimum measured force (e.g., comparing slopes). Furthermore, the comparison module 500 may compare multiple expected displacement / force pairs with measured displacement / force pairsDocket: SSMP 43930(e.g., to determine respective differences between them and determine a metric that describes the respective differences). Finally, the comparison module 500 may compare expected force vs. displacement curves with measured displacement vs. force curves for displacement and / or force ranges within the bond phase while the solder is in the liquid phase.

[0070] As discussed above, the bond quality 410 may comprise any metric. For example, the bond quality 410 may comprise the minimum measured force 310. The bond quality 410 may also comprise one or more measured displacement / force pairs or a curve of measured force vs. displacement. If the comparison module 500 is used, then the bond quality 410 may comprise a pass / fail determination, a percentage of expected determination, or some other quantitative measure of how well the measured forces and displacements meet, match, or exceed the expected forces and displacements.

[0071] FIG. 6 illustrates an example of a system 600 that may be used for determining bond quality of thermo-compression bonds based on solder surface tension. The system 600 includes at least one processing unit 602, at least one computer-readable storage medium 604, and the bond quality module 406. The system 600 (including the processing unit 602) may be contained within the die bonder 401, partially within the die bonder 401, or communicatively coupled with the die bonder 401.

[0072] The processing unit 602 (e.g., one or more of an application processor, central processing unit (CPU), graphics processing unit (GPU), microprocessor, digital- signal processor (DSP), or controller) executes instructions 606 (e.g., code) stored within the computer-readable storage medium 604 (e.g., a non-transitory storage devices such as a hard drive, solid-state drive (SSD), flash memory, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or electrically erasable programmable read-only memory (EEPROM)) to cause the system 600 to perform the techniques described herein. The instructions 606 may be part of an operating system and / or one or more applications of the system 600.

[0073] The instructions 606 cause the processing unit 602 to act upon (e.g., create, receive, modify, delete, transmit, or display) the data 608 (e.g., application data, module data, sensor data (e.g., the force signal 400 and / or the displacement signal 404), or I / O data). Although shown asDocket: SSMP 43930being within the computer-readable storage medium 604, portions of the data 608 may be within a random-access memory (RAM) or a cache of the system 600 (not shown). Furthermore, the instructions 606 and / or the data 608 may be remote to the system 600.

[0074] The bond quality module 406 (or portions thereof) may be comprised by the computer-readable storage medium 604 or be a stand-alone component (e.g., executed in dedicated hardware in communication with the processing unit 602 and computer-readable storage medium 604). For example, the instructions 606 may cause the processing unit 602 to implement or otherwise cause the bond quality module 406 to perform the actions discussed above.

[0075] The system 600 may also contain a communication system (not shown) that may be any wired or wireless communication system configured to communicate data over one or more connections or networks. For example, the communication system may be configured to communicate data between the system 600 and a separate device (e.g., the die bonder 401 if the system 600 is not implemented within the die bonder 401).

[0076] FIG. 7 illustrates a method 700 of determining a bond quality of a thermo-compression bond between a die and a substrate based on solder surface tension. The method 700 may be performed by the die bonder 401, a remote system, or some combination thereof. Steps of the method 700 may be rearranged, split, or combined without departing from the scope of this disclosure.

[0077] At step S702, one or more forces generated by surface tension of solder between a die and a substrate are determined during a bonding process while the solder is in a liquid phase. For example, the bond quality determination module 408 may determine one or more forces from the force signal 400 for respective displacements of the bond head 102 while the solder balls 114 are in a liquid phase. The displacements may be anywhere from a minimum displacement (e.g., a maximum compression of the solder) to a final location of the bond head 102 during the bonding process.

[0078] At step S704, one or more expected forces based on expected surface tension of the solder between the die and the substrate during the bonding process while the solder is in the liquid phase are determined. For example, the bond quality determination module 408 mayDocket: SSMP 43930determine an expected force (e.g., minimum force), one or more expected displacement / force pairs, or the expected force vs. displacement curve 302 based on the solder parameters 502.

[0079] At step S706, the one or more forces are compared to the one or more expected forces. For example, the comparison module 500 may compare the measured force vs. displacement curve 304 with the expected force vs. displacement curve 302. The comparison module 500 may also compare a minimum expected force with a minimum measured force or one or more expected displacement / force pairs with one or more measured displacement / force pairs. In either case, a smoothing algorithm of function may be used to smooth the measured force and / or displacement values.

[0080] At step S708, a bond quality is determined based on the comparing. For example, the comparison module 500 may determine the bond quality 410 based on the result of any of the comparisons of step S706.

[0081] Optionally, at step S710, the bond quality is output. For example, the bond quality determination module 408 may output the bond quality 410 for receipt by the action module 412. The output may be effective to cause the action module 412 to perform the action 414 based on the bond quality 410.

[0082] In some implementations, if the bond quality 410 indicates a poor connection, the action 414 may cause movement of the bond head 102 to be repeated (e.g., to lower the displacement 200 and raise it back up). As such, the measurement and comparison steps may be repeated for the subsequent movement to determine if the subsequent movement created a better bond.

[0083] The above examples generally refer to a single axis force measurement. In some implementations, the force measurement may generate a plurality of forces (e.g., corresponding to different portions of the die 108) and / or one or more moments (e.g., perpendicular to the axis 104). For example, using multiple force sensors or a multi-axis force sensor, the bond quality determination module 408 may determine an unevenness in the force signal 400 or a moment acting on the bond head 102 (e.g., one area of the bond has a different force than another). If an unevenness is detected, the action module 412 may generate an action 414 that causes the bond head 102 to move in one or more dimensions to correct the unevenness.Docket: SSMP 43930

[0084] Example 1: A method of determining a bond quality of a thermo-compression bond between a die and a substrate, the method comprising: determining, based on a force signal received from a force sensor, one or more forces generated by surface tension of solder between the die and the substrate during a bonding process while the solder is in a liquid phase; determining one or more expected forces based on expected surface tension of the solder between the die and the substrate during the bonding process while the solder is in the liquid phase; comparing the one or more forces to the one or more expected forces; and determining, based on the comparing, the bond quality.

[0085] Example 2: The method of example 1, wherein the one or more expected forces correspond to a configuration of the solder.

[0086] Example 3: The method of example 2, wherein: the solder comprises a plurality of solder balls of one or more sizes; and the configuration of the solder comprises, for each size, a volume of each solder ball of the respective size and a count of solder balls of the respective size.

[0087] Example 4: The method of example 3, further comprising generating the one or more expected forces using a surface tension function and the configuration of the solder.

[0088] Example 5: The method of any previous example, wherein the comparing comprises determining respective differences between the one or more forces and the one or more expected forces.

[0089] Example 6: The method of example 5, wherein the determining the bond quality comprises quantifying the differences.

[0090] Example 7: The method of any previous example, further comprising: outputting the bond quality; and responsive to the outputting, causing a repeat of at least a portion of the bonding process.

[0091] Example 8: The method of example 7, further comprising determining a second bond quality for the thermo-compression bond after the portion of the bonding process is completed.

[0092] Example 9: The method of any previous example, further comprising: determining that the bond quality is below a threshold value; and causing a die bonder performing the thermocompression bond to enter an error state.Docket: SSMP 43930

[0093] Example 10: The method of any previous example, wherein: the one or more forces comprise a single force corresponding to a minimum of the force signal; and the one or more expected forces comprise a minimum expected force.

[0094] Example 11: The method of any previous example, wherein: the one or more forces are determined for respective displacements between the die and the substrate; and the one or more expected forces correspond to the displacements.

[0095] Example 12: The method of example 11, wherein the determining the one or more forces comprises generating, based on the force signal and a displacement signal received from a displacement sensor, a measured force vs. displacement curve.

[0096] Example 13: The method of example 12, further comprising applying one or more smoothing algorithms to the measured force vs. displacement curve.

[0097] Example 14: The method of any of example 12 or 13, wherein the determining the one or more expected forces comprises determining an expected force vs. displacement curve.

[0098] Example 15: The method of example 14, wherein the comparing comprises determining a difference between the measured force vs. displacement curve and the expected force vs. displacement curve.

[0099] Example 16: The method of example 15, wherein the determining the bond quality comprises quantifying the difference between the measured force vs. displacement curve and the expected force vs. displacement curve.

[0100] Example 17: A die bonder comprising: a force sensor configured to generate a force signal; a displacement sensor configured to generate a displacement signal; and a processing unit configured to: receive the force signal; determine, based on the force signal, one or more forces generated by surface tension of solder between a die and a substrate for respective displacements between the die and the substrate while the solder is in a liquid phase; determine one or more expected forces; compare the one or more forces to the one or more expected forces; and determine, based on the comparison, a bond quality between the die and the substrate.

[0101] Example 18: The die bonder of example 17, wherein the force sensor is disposed within a bond head of the die bonder.Docket: SSMP 43930

[0102] Example 19: The die bonder of example 17 or 18, wherein the processing unit is further configured to, responsive to determining that the bond quality does not meet a threshold value, cause the die bonder to enter an error state.

[0103] Example 20: A system comprising a processing unit configured to: determine, based on a force signal received from a force sensor, one or more forces generated by surface tension of solder between a die and a substrate for respective displacements between the die and the substrate while the solder is in a liquid phase; determine one or more expected forces; compare the one or more forces to the one or more expected forces; and determine, based on the comparison, a bond quality between the die and the substrate.

[0104] The terminology used herein is for the purpose of describing particular implementations, embodiments, and / or examples only and is not intended to be limiting of this disclosure. As used herein, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “includes,” “comprises,” and / or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Further, the terms up, upper, down, lower, above, below, left, right, forward, rearward, and the like are intended to be understood in the context of the representations described and illustrated above so that a wearable device may have such an orientation in reference to the frame or to various elements as supported by the frame or as illustrated in the drawing figures.

[0105] The corresponding structures, materials, acts, and equivalents of all means or step plus function elements, if any, in the claims below are intended to include any structure, material, or act for performing the function in combination with other claimed elements as specifically claimed. The description of the present disclosure has been presented for purposes of illustration and description but is not intended to be exhaustive or limited to this disclosure in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of this disclosure. The various embodiments were chosen and described in order to best explain the principles of this disclosureDocket: SSMP 43930and the practical application, and to enable others of ordinary skill in the art to understand this disclosure for various embodiments with various modifications as are suited to the particular use contemplated.Docket: SSMP 43930LIST OF REFERENCE NUMERALS USED102 bond head104 axis106 bond tool108 die109 die pads110 substrate112 substrate pads114 solder balls116 force sensor200 displacements300 solder shapes302 expected force vs. displacement curves304 measured force vs. displacement curves306 zero point308 expected minimum force310 minimum measured force400 force signal401 die bonder402 displacement sensor404 displacement signal406 bond quality module408 bond quality determination module410 bond quality412 action module414 action500 comparison module502 solder parameters504 model506 database600 system602 processing unit604 computer - readable storage medium606 instructions608 data700 method

Claims

Docket: SSMP 43930CLAIMSWhat is claimed is:

1. A method of determining a bond quality of a thermo-compression bond between a die and a substrate, the method comprising:determining, based on a force signal received from a force sensor, one or more forces generated by surface tension of solder between the die and the substrate during a bonding process while the solder is in a liquid phase;determining one or more expected forces based on expected surface tension of the solder between the die and the substrate during the bonding process while the solder is in the liquid phase;comparing the one or more forces to the one or more expected forces; and determining, based on the comparing, the bond quality.

2. The method of claim 1, wherein the one or more expected forces correspond to a configuration of the solder.

3. The method of claim 2, wherein:the solder comprises a plurality of solder balls of one or more sizes; andthe configuration of the solder comprises, for each size, a volume of each solder ball of the respective size and a count of solder balls of the respective size.

4. The method of claim 3, further comprising generating the one or more expected forces using a surface tension function and the configuration of the solder.

5. The method of claim 1, wherein the comparing comprises determining respective differences between the one or more forces and the one or more expected forces.Docket: SSMP 439306. The method of claim 5, wherein the determining the bond quality comprises quantifying the differences.

7. The method of claim 1, further comprising:outputting the bond quality; andresponsive to the outputting, causing a repeat of at least a portion of the bonding process.

8. The method of claim 7, further comprising determining a second bond quality for the thermo-compression bond after the portion of the bonding process is completed.

9. The method of claim 1, further comprising:determining that the bond quality is below a threshold value; andcausing a die bonder performing the thermo-compression bond to enter an error state.

10. The method of claim 1, wherein:the one or more forces comprise a single force corresponding to a minimum of the force signal; andthe one or more expected forces comprise a minimum expected force.

11. The method of claim 1, wherein:the one or more forces are determined for respective displacements between the die and the substrate; andthe one or more expected forces correspond to the displacements.

12. The method of claim 11, wherein the determining the one or more forces comprises generating, based on the force signal and a displacement signal received from a displacement sensor, a measured force vs. displacement curve.

13. The method of claim 12, further comprising applying one or more smoothing algorithms to the measured force vs. displacement curve.Docket: SSMP 4393014. The method of claim 12, wherein the determining the one or more expected forces comprises determining an expected force vs. displacement curve.

15. The method of claim 14, wherein the comparing comprises determining a difference between the measured force vs. displacement curve and the expected force vs. displacement curve.

16. The method of claim 15, wherein the determining the bond quality comprises quantifying the difference between the measured force vs. displacement curve and the expected force vs. displacement curve.

17. A die bonder comprising:a force sensor configured to generate a force signal;a displacement sensor configured to generate a displacement signal; anda processing unit configured to:receive the force signal;determine, based on the force signal and the displacement signal, one or more forces generated by surface tension of solder between a die and a substrate for respective displacements between the die and the substrate while the solder is in a liquid phase; determine one or more expected forces based on expected surface tension of the solder between the die and the substrate during the bonding process while the solder is in the liquid phase;compare the one or more forces to the one or more expected forces; and determine, based on the comparison, a bond quality between the die and the substrate.

18. The die bonder of claim 17, wherein the force sensor is disposed within a bond head of the die bonder.Docket: SSMP 4393019. The die bonder of claim 17, wherein the processing unit is further configured to, responsive to determining that the bond quality does not meet a threshold value, cause the die bonder to enter an error state.

20. A system comprising a processing unit configured to:determine, based on a force signal received from a force sensor, one or more forces generated by surface tension of solder between a die and a substrate during a bonding process while the solder is in a liquid phase;determine one or more expected forces based on expected surface tension of the solder between the die and the substrate during the bonding process while the solder is in the liquid phase;compare the one or more forces to the one or more expected forces; and determine, based on the comparison, the bond quality.

Images