Process window testing method and automatic process optimization method in wire bonding system

Abstract

To provide a process window testing method for process parameters in a wire bonding system, particularly concerning wire bonding operation.SOLUTION: The method includes a step 1000 for identifying process parameters used for wire bonding operation in a wire bonding system, a step 1002 for testing at least one response of the process parameters with multiple values of the process parameters in the wire bonding system, and a step 1006 for automatically recording at least one response of the process parameters with multiple values of the process parameters.SELECTED DRAWING: Figure 10

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Application No. 63 / 560,190, filed March 1, 2024, the contents of which are incorporated herein by reference.

[0002] The present invention relates to wire bonding operations, and more particularly to a method for testing the process window of process parameters in a wire bonding system. [Background technology]

[0003] In semiconductor device fabrication and packaging, wire bonding remains the primary method for providing electrical interconnection between two locations within a package (e.g., between a die pad on a semiconductor die and a lead on a lead frame). More specifically, a wire bonder (also known as a wire bonding machine) is used to form a wire loop between the locations to be electrically interconnected. The primary methods for forming the wire loop are ball bonding and web bonding. Various types of bonding energy can be used to form the bond between (a) the end of the wire loop and (b) each bonding location (e.g., die pad, lead, etc.), including ultrasonic energy, thermosonic energy, and thermocompression energy. Wire bonding machines (e.g., stud bump machines) are also used to form conductive bumps from portions of the wire.

[0004] During a ball bonding operation, the tail of the wire extending from the tip of the bonding tool (e.g., capillary, etc.) is melted into a free-air ball using a spark from an electric flame-off (EFO) device. A first bond (e.g., a ball bond, etc.) of the wire loop is then formed at a first bonding location using the free-air ball. The wire is then extended from the ball bond to a second bonding location, where a second bond (e.g., a stitch bond, etc.) of the wire loop is formed by bonding (joining) a portion of the wire to the second bonding location using the bonding tool. For example, the first bonding location may be a bonding pad on a semiconductor die, and the second bonding location may be a lead on a lead frame.

[0005] Several difficult situations can arise in connection with such wire bonding operations. For example, during bonding of the first bond of a wire loop, a free air ball may not properly bond to the bonding location. This situation is often referred to as an NSOP condition (i.e., no stick on pad condition). In another example, the second bond of the wire loop may not properly bond to the bonding location. This situation is often referred to as an NSOL condition (i.e., no stick on lead condition). An exemplary process for dealing with such "no stick" conditions is disclosed in U.S. Pat. No. 8,899,469, entitled "AUTOMATIC REWORK PROCESSES FOR NON-STICK CONDITIONS IN WIRE BONDING OPERATIONS."

[0006] Another challenge in wire bonding operations relates to the so-called "short tail condition." For example, after a stitch bond is formed at the second bond location of a wire loop, the bonding tool may be elevated to a short tail detection height, where the wire may be tested (e.g., electrical continuity tested) to ensure continued continuity with the stitch bond at the second bond location. If a short tail is not detected, the bond head (i.e., carrying the bonding tool and wire clamp, with the clamp now closed) is elevated to sever the wire at the stitch bond. The remaining wire tail length can be used to form another free air ball for another wire loop. However, a short tail may also be detected. A short tail condition can lead to various problems during wire bonding, such as inconsistencies in the size and shape of the free air ball. Exemplary techniques for addressing such "short tail" conditions are disclosed in U.S. Patent No. 9,165,842, entitled "SHORT TAIL RECOVERY TECHNIQUES IN WIRE BONDING OPERATIONS." The contents of each of U.S. Patent Nos. 8,899,469 and 9,165,842 are incorporated herein by reference in their entirety.

[0007] Certain process parameters used in connection with wire bonding operations may be varied to address certain undesirable conditions (e.g., NSOP, NSOL, etc.) Therefore, it is desirable to provide a method for testing the process window of such process parameters in a wire bonding system. Summary of the Invention [Means for solving the problem]

[0008] According to an exemplary embodiment of the present invention, there is provided a method for testing a process window of a process parameter in a wire bonding system, the method including: (a) identifying the process parameter to be used in a wire bonding operation in the wire bonding system, (b) testing a response of at least one of the process parameters at multiple values ​​of the process parameter in the wire bonding system, and (c) automatically recording the response of the at least one process parameter at the multiple values ​​of the process parameter.

[0009] According to another embodiment of the present invention, the method described in the immediately preceding paragraph has the following features: further comprising a step (d) of identifying an acceptable process window for the process parameters based on a result of step (b); further comprising a step of identifying optimal values ​​for the process parameters based on a result of step (b); and further comprising a step of automatically recovering from a process error during step (b) without operator intervention; the step of automatically recovering from a process error includes at least one of recovering from a floating ball condition, recovering from a floating stitch condition, recovering from a short tail condition, bonding the floating ball to another location and continuing step (b), forming a wire tail after bonding the floating ball using predetermined parameters, bonding the floating stitch to another location and continuing step (b), and / or forming a wire tail after bonding the floating stitch using predetermined parameters. the process parameter is related to at least one of ultrasonic energy applied during formation of a wire bond (e.g., a ball bond on a wire loop, a stitch bond on a wire loop, a first wire bond on a wire loop, a second wire bond on a wire loop, an intermediate wire bond on a wire loop, a conductive bump bond, a wire bond on a vertical wire structure, etc.), bonding force applied during formation of a wire bond, and table scrub energy applied during formation of a wire bond; the at least one response is related to at least one of ball lift of a ball bond, stitch lift of a stitch bond, short tail condition, wire bond tensile value, wire bond shape, and / or wire bond dimension; each of steps (b) and (c) is repeated a predetermined number of iterations at each of the plurality of values ​​of the process parameter;Each of steps (b) and (c) may be repeated while varying another process parameter to test the response of the at least one process parameter in the wire bonding system to the multiple values ​​of the process parameter and the other process parameter; further comprising step (d) after repeating steps (b) and (c), identifying an acceptable process window for the process parameter and a desired value for the other process parameter; and / or further comprising step (d) after repeating steps (b) and (c), identifying an optimal value for the process parameter and a desired value for the other process parameter;

[0010] According to another exemplary embodiment of the present invention, there is provided a method for testing a process window of a process parameter in a wire bonding system, the method including: (a) identifying the process parameter to be used in a wire bonding operation in the wire bonding system; (b) testing a response of at least one of the process parameters at multiple values ​​of the process parameter in the wire bonding system; and (c) identifying an acceptable process window for the process parameter based on the results of step (b).

[0011] According to another embodiment of the present invention, the method described in the immediately preceding paragraph further comprises the step of automatically recording the at least one response of the process parameter at the plurality of values ​​of the process parameter; the at least one response related to at least one of ball lift of a ball bond, stitch lift of a stitch bond, short tail condition, wire bond pull value, wire bond shape, and / or wire bond dimension; the process parameter related to at least one of ultrasonic energy applied during formation of a wire bond (e.g., first wire bond of a wire loop, second wire bond of a wire loop, middle wire bond of a wire loop, conductive bump bond, wire bond of a vertical wire structure, ball bond of a wire loop, stitch bond of a wire loop, etc.), bonding force applied during formation of a wire bond, and table scrub energy applied during formation of a wire bond; and further comprising the step of automatically recording the at least one response of the process parameter at the plurality of values ​​of the process parameter based on a result of step (b). identifying optimal values ​​of process parameters; further including the step of automatically recovering from a process error during step (b) without operator intervention; the step of automatically recovering from a process error includes at least one of recovering from a floating ball condition, recovering from a floating stitch condition, and recovering from a short tail condition, bonding the floating ball to another location and continuing step (b), bonding the floating stitch to another location and continuing step (b), and / or forming a wire tail after bonding the floating ball using predetermined parameters; step (b) is repeated a predetermined number of times at each of the multiple values ​​of the process parameters before step (c); step (b) is repeated while varying other process parameters, and before step (c), the wire bonding system tests the response of the at least one process parameter to the multiple values ​​of the process parameter and the other process parameters;The method may further include one or more of: after repeating step (b), a step (d) of identifying desired values ​​of the other process parameters; and / or after repeating step (b), a step of identifying optimal values ​​of the process parameters and desired values ​​of the other process parameters;

[0012] According to another exemplary embodiment of the present invention, there is provided a method for testing a process window of a process parameter in a wire bonding system, the method including: (a) identifying the process parameter to be used in a wire bonding operation in the wire bonding system; (b) testing a response of at least one of the process parameters at multiple values ​​of the process parameter in the wire bonding system; and (c) identifying an optimal value for the process parameter based on the results of step (b).

[0013] According to another embodiment of the present invention, the method described in the immediately preceding paragraph further comprises the following features: (d) identifying an acceptable process window for the process parameter based on results of (b); and (d) automatically recording the at least one response of the process parameter at the plurality of values ​​of the process parameter; the at least one response being related to at least one of a ball lift of the ball bond, a stitch lift of the stitch bond, a short tail condition, a tensile value of the wire bond, a shape of the wire bond, and / or a dimension of the wire bond; the process parameter being selected from ultrasonic energy applied during formation of a wire bond (e.g., a first wire bond of a wire loop, a second wire bond of a wire loop, a middle wire bond of a wire loop, a conductive bump bond, a wire bond of a vertical wire structure, a ball bond of a wire loop, a stitch bond of a wire loop, etc.), a wire bond, ... the process error relates to at least one of a bonding force applied during formation of the bond and a table scrub energy applied during formation of the wire bond; further comprising automatically recovering from a process error during step (b) without operator intervention; the automatically recovering from the process error includes at least one of recovering from a floating ball condition, recovering from a floating stitch condition, and recovering from a short tail condition, bonding the floating ball to another location and continuing step (b), forming a wire tail after bonding the floating ball using predetermined parameters, bonding the floating stitch to another location and continuing step (b), and / or forming a wire tail after bonding the floating stitch using predetermined parameters; step (b) is repeated a predetermined number of times at each of the plurality of values ​​of the process parameters before step (c);Step (b) may be repeated while varying other process parameters, and before step (c), testing the response of the at least one process parameter in the wire bonding system to the multiple values ​​of the process parameter and the other process parameters; further comprising step (d) after the repeated step (b), identifying an acceptable process window for the process parameter and a desired value for the other process parameter; and / or further comprising step (d) after the repeated step (b), identifying a desired value for the other process parameter;

[0014] The methods of the present invention may also be implemented as an apparatus (e.g., as part of the intelligence of a wire bonding apparatus / system) or as computer program instructions on a computer-readable medium (e.g., a computer-readable medium containing a wire bonding program for use in connection with a wire bonding apparatus / system). [Brief explanation of the drawings]

[0015] The invention is best understood from the following detailed description when read in conjunction with the accompanying drawings. According to common practice, the various elements of the drawings are not drawn to scale. Rather, the dimensions of the various elements have been arbitrarily increased or reduced for clarity. The drawings include the following figures: [Figure 1A] 1A-1B are block diagram side views of a wire bonding system illustrating the testing of process parameters in certain exemplary embodiments of the present invention. [Figure 1B] 1A-1B are block diagram side views of a wire bonding system illustrating the testing of process parameters in certain exemplary embodiments of the present invention. [Figure 2A] 2A-2B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 2B] 2A-2B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 3A] 3A-3B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 3B] 3A-3B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 4A] 4A-4B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 4B] 4A-4B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 5A] 5A-5B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 5B] 5A-5B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 6A] 6A-6B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 6B] 6A-6B are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 7A]7A-7C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 7B] 7A-7C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 7C] 7A-7C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 8A] 8A-8C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 8B] 8A-8C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 8C] 8A-8C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 9A] 9A-9C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 9B] 9A-9C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 9C] 9A-9C are block diagram side views of the wire bonding system of FIGS. 1A-1B illustrating further testing of process parameters in certain exemplary embodiments of the present invention. [Figure 10]FIG. 10 is a flow diagram illustrating a method for automatically testing the process window of process parameters in a wire bonding system, according to various exemplary embodiments of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] As used herein, the term "semiconductor device" is intended to refer to any structure that includes (or is configured to include at a later stage) a semiconductor chip or die. Exemplary semiconductor devices include, among others, bare semiconductor die, semiconductor die on a substrate (e.g., a lead frame, PCB, carrier, etc.), packaged semiconductor devices, flip-chip semiconductor devices, die embedded in a substrate, and stacks of semiconductor dies. Additionally, semiconductor devices can include devices configured to be bonded to or otherwise included in a semiconductor package (e.g., spacers, substrates, etc., bonded in a stacked die structure). In the context of the present invention, a semiconductor device is an example of a workpiece. Other examples of workpieces include semiconductor devices mounted on a substrate (e.g., semiconductor die mounted on a lead frame). Yet other examples of workpieces include multiple semiconductor devices.

[0017] As used herein, the term "process parameter" is intended to be broadly construed to include any parameter or setting used in a wire bonding process (e.g., USG, bonding force, scrub amplitude, scrub frequency, time, bonding duration, bonding system settings, bonding system mode, etc.) The value of a process parameter can be any quantitative and / or qualitative value (e.g., numeric value, percentage of reference / set point, on / off condition, mode active / inactive, etc.).

[0018] As will be appreciated by those skilled in the art, the term "wire portion" is intended to be broadly construed and is not limited to a particular length.

[0019] Various error conditions for the wire bonding process are described herein, which may include, among others, an NSOP condition, an NSOL condition, a short tail condition, and a long tail condition.

[0020] According to certain exemplary embodiments of the present invention, a wire bonding system is provided with software functionality that enables automated process window testing on the wire bonding system. Exemplary process windows that are tested include a USG window, a bond force process window, and a scrub amplitude process window. Such process windows may be tested for various wire bonds, such as a first wire bond (e.g., a ball bond) of a wire loop, a second bond (e.g., a stitch bond) of a wire loop, etc. Exemplary process responses used in connection with process window testing include a ball lift response, a stitch lift response, a short tail response, a pull test response (e.g., using a pull tester integrated into the wire bonding system), and a shear test response (e.g., using a shear tester integrated into the wire bonding system).

[0021] For example, when testing a USG process window, multiple values ​​(e.g., input setting levels, percentages relative to a reference set point, etc.) are tested and the process response (e.g., ball lift) is automatically determined and recorded in the wire bonding system. According to certain exemplary embodiments of the present invention, the method includes an automatic recovery mechanism for addressing process error conditions (e.g., ball lift conditions).

[0022] It should be understood that like reference numerals used throughout this specification (including the drawings) are intended to refer to like elements unless otherwise stated.

[0023] Referring now to the drawings, FIGS. 1A-1B illustrate various elements of a wire bonding system 100 associated with a wire bonding process (e.g., a test wire bonding process). The wire bonding system 100 includes a support structure 102 and a bond head assembly 104. The bond head assembly 104 includes a transducer 108 (e.g., an ultrasonic transducer). The transducer 108 is configured to carry a wire bonding tool 110 (e.g., a capillary, etc.). The wire bonding tool 110 is configured to create conductive structures, such as wire loops, vertical wire structures, conductive bumps, etc., using a wire 112. For example, in the case of a wire loop, such a wire loop can provide an electrical interconnection between (i) a bonding location on a semiconductor device 106 (e.g., a bond pad on the semiconductor device 106) and (ii) a bonding location on a substrate 114 (e.g., a lead on a lead frame). A semiconductor element 106 (e.g., a semiconductor die) is shown proximate to a substrate 114 and supported by a support structure 102. The wire bonding system 100 is shown including a computer 116 and a detection system 118. The computer 116 can be programmed to execute a wire bonding process (e.g., a wire bonding program, etc.). For example, the wire bonding system 100 can be programmed to execute a wire bonding process using multiple process parameters (e.g., ultrasonic energy applied during wire bond formation, bonding force applied during wire bond formation, and table scrub energy applied during wire bond formation). Generally, a robust wire bonding process is desired in which an acceptable process window for each of the multiple process parameters (and / or optimal values ​​for each of the multiple process parameters) is known. According to an aspect of the present invention, multiple values ​​of one or more process parameters are tested by detecting a response to the wire bonding process using multiple values ​​of the process parameter.

[0024] Various figures provided herein show example values ​​of process parameters that may be tested for a particular response (or responses), which may be recorded (e.g., automatically by wire bonding system 100, including computer 116).

[0025] 1A , the parameters tested may include, for example, ultrasonic energy (i.e., “USG”) applied during the formation of the wire bond, bonding force (i.e., “BF”) applied during the formation of the wire bond, and table scrub energy (i.e., “SCRUB”) applied during the formation of the wire bond. In FIG. 1A , a free air ball 112 a is shown contacting a bonding location on a semiconductor device 106 during a wire bonding operation (e.g., during a test operation). During bonding of the free air ball 112 a to the bonding location (or while such bonding is attempted), values ​​of one or more process parameters may be tested.

[0026] A parameter status indicator 120 (e.g., a screen, a graphical user interface, a virtual representation of a parameter status, a digital array of data, etc.) of the computer 116 is shown in FIG. 1A. For example, the parameter status indicator 120 indicates that the USG value being tested is 40% of a reference (e.g., a predetermined set point). FIG. 1A also indicates that the BF value being tested is 40% of a reference and / or that the SCRUB value being tested is 40% of a reference. It should be understood that one process parameter may be tested at a time. Thus, while FIG. 1A (and other figures herein) shows various parameters on the parameter status indicator 120, it should be understood that it may be desirable to test single process parameters (e.g., USG, BF, SCRUB, etc.) one at a time. In this manner, Figures 1A-1B, 2A-2B, 3A-3B, and 4A-4B can be thought of as testing a single process parameter (e.g., USG) by detecting the response to that process parameter at various values ​​(e.g., 40% in Figures 1A-1B, 60% in Figures 2A-2B, 70% in Figures 3A-3B, and 100% in Figures 4A-4B).

[0027] In certain iterations (e.g., when testing certain values ​​of process parameters), the free air ball 112a may not properly bond to the bonding location on the semiconductor device 106. For example, this condition may be referred to as a "no stick on pad" (i.e., NSOP) condition. Wire bonding systems often include a detection system (e.g., detection system 118) that detects whether a portion of the wire is properly bonded to the bonding location. For example, wire bonding systems sold by Click & Soffer Industries, Inc. often use a "BITS" process (i.e., bond integrity test system) to verify whether a proper wire bond has (or has not) been formed. International Patent Application Publication No. WO 2009 / 002345, incorporated herein by reference in its entirety, provides exemplary details of such a process and associated systems.

[0028] 1B , as the wire bonding tool 110 moves away from the semiconductor device 106, the detection system 118 detects that the free air ball 112a′ (i.e., the deformed free air ball) was not properly bonded to the bonding location. Thus, using the detection system 118, the wire bonding system 100 can determine that a floating ball condition (e.g., an NSOP condition) occurs when the USG process parameter is 40% of the reference. Alternatively, the wire bonding system 100 can determine that a floating ball condition (e.g., an NSOP condition) occurs when the bonding force process parameter is 40% and / or the scrub amplitude process parameter is 40%.

[0029] The wire bonding process (e.g., test process) of FIGS. 1A-1B can be repeated multiple times with different values ​​of the process parameters (see, e.g., FIGS. 2A-2B, 3A-3B, and / or 4A-4B). FIGS. 2A-2B show the response of another lifted ball condition (e.g., an NSOP condition) when the process parameters (e.g., USG, BF, and / or SCRUB) are tested at 60% of the standard. FIGS. 3A-3B show the response of an acceptable (e.g., not an NSOP condition) wire bond 112b when the process parameters (e.g., USG, BF, and / or SCRUB) are tested at 70% of the standard. FIGS. 4A-4B also show the response of an acceptable (e.g., not an NSOP condition) wire bond 112b when the process parameters (e.g., USG, BF, and / or SCRUB) are tested at 100% of the standard. The response to each test (eg, ball lift status, acceptable wire bond status, etc.) can be (automatically) recorded.

[0030] The exemplary processes shown in connection with Figures 1A-1B, 2A-2B, 3A-3B, and 4A-4B can be used to test the process window of one or more process parameters in wire bonding system 100. Table 1 shows various test points (i.e., values) for testing process parameters (e.g., USG, BF, SCRUB, etc.), including those shown in Figures 1A-1B, 2A-2B, 3A-3B, and 4A-4B. [Table 1]

[0031] Table 1 (row 1) identifies the process parameters to be tested (USG, BF, SCRUB, etc.). The response at each value of the process parameter recorded is related to the ball lift condition (or its absence).

[0032] The results and / or responses of the tests (e.g., tests using the methods shown in Figures 1A-1B, 2A-2B, 3A-3B, and 4A-4B) are shown in the second row of Table 1. More specifically, a "ball float" condition exists at 40%, 50%, and / or 60% of the reference for the process parameter (e.g., USG, etc.). A favorable response (e.g., no ball float) is indicated as "OK" at each test point (e.g., process parameter value) of 70% or greater. Therefore, the process window for a process parameter (e.g., USG, etc.) is defined as any process parameter value that is 70% or greater of the reference USG setting. Additionally, an optimal value may be determined using specific criteria (e.g., algorithms, tolerance considerations, time / energy considerations, etc.). For example, from the results shown in Table 1, the optimal value may be identified as 80% using such criteria.

[0033] Those skilled in the art will appreciate that the testing process (the results of which are summarized in the second row of Table 1) can be performed with many other process parameters. For example, if USG is the process parameter being tested, the test may be performed with other parameters (e.g., bonding force, table scrub, etc.) at some other level. According to certain embodiments of the present invention, each testing and recording step may be repeated while varying other process parameters, such that the response of at least one of the process parameters can be tested at multiple values ​​of the process parameter and the other process parameters.

[0034] Table 2 shows an example of such a process. The process parameters tested may be, for example, USG, BF, SCRUB, etc., and the detected response is still the presence (or absence) of a ball lift condition. In a particular example, the process parameter tested is USG. While testing the response at multiple values ​​of USG (e.g., 40%, 50%, ...100%, etc.), it may be desirable to also vary other process parameters (e.g., variable parameter values) and test the response. Lines 2-4 of Table 2 represent varying the other process parameters while testing the response to multiple values ​​of USG. In an example where the other process parameter is BF (i.e., bonding force), variable parameter value #1 may be 90% of the reference bonding force, variable parameter value #2 may be 100% of the reference bonding force, and variable parameter value #3 may be 110% of the reference bonding force. Thus, by testing the process parameter response at multiple values ​​of the process parameter (i.e., 40%-100%) and other process parameters, it is found that the largest process window occurs at variable parameter value #3 (e.g., a reference bonding force of 110%). In this case, the USG process window is any USG value that is 50% or greater than the reference USG setting. An optimal value may also be determined using specific criteria (e.g., algorithms, tolerance considerations, time / energy considerations, etc.). For example, for the results shown in Table 2 (using such criteria and a BF of 110%), the optimal USG value may be identified as 60%. [Table 2]

[0035] Each of the variable parameter values ​​may represent an adjustable parameter in the wire bonding operation (e.g., USG, bonding force, scrub amplitude, scrub frequency, wire bonding tool speed, wire bonding tool acceleration, etc.) and / or a setting of the wire bonding system 100 (e.g., constant speed mode, search mode, calibration mode, etc.).

[0036] 1A-1B, 2A-2B, 3A-3B, and 4A-4B, and Tables 1 and 2 relate to testing process parameters whose response is the presence (or absence) of a lifted ball condition. It should be understood that the response detected during testing of the process parameters may be any desired response. For example, exemplary responses include lifted ball in a ball bond, lifted stitch in a stitch bond, a short tail condition, a wire bond pull value, a wire bond shape, and a wire bond dimension. 5A-5B, 6A-6B, 7A-7C, 8A-8C, and / or 9A-9C relate to testing process parameters whose response is the presence (or absence) of a lifted stitch condition and / or a short tail condition.

[0037] Additionally, FIGS. 1A-1B, 2A-2B, 3A-3B, and 4A-4B, as well as Tables 1 and 2, are described in connection with identifying an acceptable process window, where the process window has a single constraint (e.g., greater than 70% of the reference USG, greater than 60% of the reference USG, etc.) and / or a minimum threshold. It should be understood that the acceptable process window may also be identified by constraints on both ends of the process window (e.g., minimum and maximum thresholds). Each of FIGS. 5A-5B, 6A-6B, 7A-7C, 8A-8C, and / or 9A-9C is described in connection with such a process window.

[0038] Another wire bonding process (e.g., a test process) is shown in Figures 5A-5B. According to an embodiment of the present invention, multiple values ​​of one or more process parameters are tested by detecting a response to the wire bonding process using multiple values ​​of the process parameters in Figures 5A-5B.

[0039] 5A , the parameters tested may be, for example, ultrasonic energy applied during the formation of the wire bond ("USG"), bonding force applied during the formation of the wire bond (i.e., "BF"), and / or table scrub energy applied during the formation of the wire bond (i.e., "SCRUB"). FIG. 5A shows a wire bond 112b (e.g., a bonded free air ball) formed at a bonding location on a semiconductor device 106 during a wire bonding operation. A length of the wire 112 extends from the wire bond 112b to a bonding location on a substrate 114. FIG. 5A shows a wire bonding tool 110 attempting to bond (attach) a portion 112c of the wire (e.g., a stitch bond) to the bonding location on the substrate 114. Ultrasonic energy is applied to attach the portion 112c of the wire to a bonding location (e.g., a lead) on the substrate 114 (e.g., a lead frame). During bonding of wire portion 112c to the bonding location (or during an attempted such bonding), the values ​​of one or more process parameters may be tested. As shown in Figure 5A, parameter status indicator 120 indicates that the USG value being tested is 40% of nominal (e.g., a predetermined set point), the BF value being tested is 40% of nominal, and / or the SCRUB value being tested is 40% of nominal.

[0040] 5B, as wire bonding tool 110 moves away from substrate 114, detection system 118 detects that portion 112c of wire was not properly bonded to the bonding location. Thus, using detection system 118, wire bonding system 100 can determine that process parameters (e.g., USG, BF, and / or SCRUB) are 40% below the standard to produce a no-stick-on-lead (NSOL) condition.

[0041] The wire bonding process (e.g., test process) of FIGS. 5A-5B can be repeated multiple times with different values ​​of the process parameters (see, e.g., FIGS. 6A-6B, 7A-7C, 8A-8C, and / or 9A-9C). FIGS. 6A-6B show the response of an NSOL condition when the process parameters (e.g., USG, BF, and / or SCRUB) are tested at 60% of the standard. FIGS. 7A-7C show the response of wire bond 112c′ that is acceptable (i.e., an acceptable stitch bond) (e.g., not an NSOL condition) when the process parameters (e.g., USG, BF, and / or SCRUB) are tested at 70% of the standard. FIGS. 8A-8C also show the response of wire bond 112c′ that is acceptable (e.g., not an NSOL condition) when the process parameters (e.g., USG, BF, and / or SCRUB) are tested at 100% of the standard.

[0042] 9A-9C also show that the response of wire bond 112c' is acceptable (e.g., not an NSOL condition) when the process parameters (e.g., USG, BF, and / or SCRUB) are tested at 110% of the nominal. However, as shown in FIG. 9C, when wire bonding tool 110 is moved upward, a short wire tail 112d' is formed. Thus, using detection system 118, wire bonding system 100 can determine that the process parameters (e.g., USG, BF, SCRUB, etc.) tested at 110% of the nominal result in an unacceptable condition (e.g., a short tail condition). The response to each test (e.g., an NSOL condition, an acceptable wire bond condition, or a short tail condition, etc.) can be (automatically) recorded.

[0043] The exemplary processes illustrated in connection with Figures 5A-5B, 6A-6B, 7A-7C, 8A-8C, and 9A-9C can be used to test the process window of one or more process parameters in wire bonding system 100. Table 3 illustrates various test points (i.e., values) for testing process parameters (e.g., USG, BF, SCRUB, etc.), including those illustrated in Figures 5A-5B, 6A-6B, 7A-7C, 8A-8C, and / or 9A-9C. [Table 3]

[0044] For example, in Table 3, the process parameter is scrub amplitude, and the response is the presence (or absence) of a no-stick ("NSOL") condition and / or a short-tail ("SHTL") condition. The results of the test process are shown in the second row. More specifically, the NSOL condition occurs at 40%, 50%, and / or 60% of the nominal scrub amplitude. A desirable response (e.g., a non-NSOL condition) is indicated as "OK" at 70% or above at each test point, with another undesirable response indicated at 110%. As shown, a short-tail condition occurs at 110% and / or 120% of the nominal scrub amplitude. Therefore, the process window for scrub amplitude is defined as any scrub amplitude value between 70% and 100% of the nominal scrub amplitude value or setting. The optimal value may also be determined using specific criteria (e.g., algorithms, tolerance considerations, time / energy considerations, etc.). In the example in the table below, it may be desirable to allow some tolerance for the parameters (e.g., + / - 10%), in which case the optimum value may be 80% or 90% of the nominal scrub amplitude. In other examples, the optimum value may be the midpoint of the process window (e.g., 85% of the nominal scrub amplitude).

[0045] Those skilled in the art will appreciate that the test process shown in row 2 can be performed with many other process parameters. For example, if SCRUB is the process parameter being tested, the test can be performed with other parameters (e.g., USG, BF, etc.) at some other level. According to certain embodiments of the present invention, each testing and recording step can be repeated while varying other process parameters, such that the response of at least one of the process parameters can be tested at multiple values ​​of the process parameter and the other process parameters.

[0046] Table 4 shows an example of such a process, where the process parameters tested are, for example, USG, BF, SCRUB, etc., and the response detected is still the presence (or absence) of an NSOL condition and / or a short-tail condition. In a particular example, the process parameter tested is SCRUB. While testing the response at multiple values ​​of SCRUB (e.g., 40%, 50%, ...120%, etc.), it may be desirable to also vary other process parameters (e.g., variable parameter values) and test the response. Rows 2-4 of Table 4 represent varying other process parameters while testing the response to multiple values ​​of SCRUB. In an example where the other process parameter is BF (i.e., bonding force), variable parameter value #1 may be 90% of the reference bonding force, variable parameter value #2 may be 100% of the reference bonding force, and variable parameter value #3 may be 110% of the reference bonding force. Thus, by testing the response of a process parameter (e.g., SCRUB) at multiple values ​​of the process parameter (i.e., 40%-120%) and with the other process parameters described above, it is found that the largest process window occurs at variable parameter value #3 (e.g., 110% bonding force). In this case, the SCRUB process window is any value of SCRUB greater than or equal to 50% of the reference SCRUB setting and less than or equal to 110% of the reference SCRUB setting. Additionally, the optimal value may be determined using specific criteria (e.g., algorithm, tolerance considerations, time / energy considerations, etc.). For example, for the results shown in Table 4, using such criteria, for a BF of 110%, the optimal value of SCRUB may be identified as 60%, 100%, or a value in between (e.g., 80%). [Table 4]

[0047] Each variable parameter value may represent an adjustable parameter in the wire bonding operation (e.g., USG, bonding force, scrub amplitude, scrub frequency, wire bonding tool speed, wire bonding tool acceleration, etc.) and / or a setting of the wire bonding system 100 (e.g., constant speed mode, search mode, calibration mode, etc.).

[0048] 10 is a flow diagram of a method for testing the process window of a process parameter in a wire bonding system, such as wire bonding system 100. As will be understood by those skilled in the art, certain steps included in the flow diagram may be omitted, certain additional steps may be added, and the order of steps may be changed from the order shown, all within the scope of the present invention.

[0049] In step 1000, process parameters are identified for use in a wire bonding operation in a wire bonding system (see, e.g., the process parameters identified and / or illustrated by the parameter status indicators 120 of the wire bonding system 100 in FIGS. 1A-1B, 2A-2B, 3A-3B, 4A-4B, 5A-5B, 6A-6B, 7A-7C, 8A-8C, and / or 9A-9C). In step 1002, a response of at least one of the process parameters is tested in the wire bonding system at multiple values ​​of the process parameter (see, e.g., the test process in FIGS. 1A-1B, 2A-2B, 3A-3B, 4A-4B, 5A-5B, 6A-6B, 7A-7C, 8A-8C, and / or 9A-9C). In certain embodiments, the process parameters relate to at least one of ultrasonic energy applied during formation of the wire bond, bonding force applied during formation of the wire bond, and table scrub energy applied during formation of the wire bond (see, e.g., USG, BF, and SCRUB references in Figures 1A, 2A, 3A, 4A, 5A, 6A, 7A, 8A, and / or 9A). In certain exemplary embodiments, the wire bond may be a first wire bond of a wire loop, a second wire bond of a wire loop, an intermediate wire bond of a wire loop (e.g., a wire bond between the first and last wire bond in a wire loop), a conductive bump bond, a ball bond of a wire loop, a stitch bond of a wire loop, and / or a wire bond of a vertical wire structure (e.g., a vertical wire structure having only one end bonded to the workpiece). In certain exemplary embodiments, the at least one response relates to a ball lift of a ball bond, a stitch lift of a stitch bond, a short tail condition, a pull value of a wire bond, a shape of a wire bond, and / or a dimension of a wire bond. In certain embodiments, step 1002 is performed automatically (e.g., without stopping the machine, without operator intervention, etc.).

[0050] In certain exemplary embodiments, at optional step 1004, the wire bonding system automatically recovers from the process error during step 1002 without operator intervention. By automatically recovering from the process error, the process window testing method described above can continue without operator intervention.

[0051] In certain exemplary embodiments, automatically recovering from a process error includes recovering from a floating ball condition. For example, recovering from a floating ball condition may include (a1) bonding the floating ball to another location and (b1) continuing to test the process window after bonding the floating ball to the other location. Further, after step (a1), a wire tail may be formed after bonding the floating ball in step (a1) using predetermined parameters.

[0052] In certain exemplary embodiments, automatically recovering from a process error includes recovering from a floating stitch condition. For example, recovering from a floating stitch condition may include (a2) bonding the floating stitch to another location and (b2) continuing to test the process window after bonding the floating stitch to the other location. Further, after step (a2), a wire tail may be formed after bonding the floating stitch in step (a2) using predetermined parameters.

[0053] In certain exemplary embodiments, automatically recovering from a process error includes recovering from a short tail condition. For example, predetermined parameters may be used to form a wire tail in connection with recovering from a short tail condition.

[0054] Exemplary techniques for recovering from certain error conditions and for forming wire tails in connection with such recovery are set forth in U.S. Patent Nos. 8,899,469 and 9,165,842, the contents of which are incorporated herein by reference in their entireties.

[0055] In optional step 1006, the response of at least one of the process parameters at multiple values ​​of the process parameter is recorded automatically (e.g., without shutting down the equipment, without operator intervention, etc.). In decision block DB1, it is determined whether certain aspects of the method (e.g., each of steps 1002, 1004, and / or 1006) are to be repeated a predetermined number of iterations. If the answer in decision block DB1 is "yes," certain aspects of the test method are repeated, as described in the following two paragraphs.

[0056] For example, it may be desirable to repeat step 1002 (and / or steps 1004 and 1006, as appropriate) a predetermined number of iterations. By repeating step 1002, a more accurate process window can be determined for a process parameter of interest, or a more accurate optimum value for that process parameter can be determined.

[0057] In other examples, it may be desirable to repeat step 1002 (and / or steps 1004 and 1006, as appropriate) a predetermined number of iterations while varying other process parameters to test the response of at least one of the process parameters in the wire bonding system at multiple values ​​of the process parameter and other process parameters. For example, see Tables 2 and 4 above, where the process parameter of interest is tested while the other process parameters are varied. By repeating step 1002 a predetermined number of iterations while varying the other process parameters, a more accurate process window for the process parameter of interest can be determined, a more accurate optimum value for the process parameter can be determined, and / or more accurate values ​​for the other process parameters can be determined.

[0058] If the answer at decision block DB1 is no, the process proceeds to step 1008. In optional step 1008, an acceptable process window for the process parameters is identified based on the results of step 1002 (e.g., including any iterative executions of step 1002). Desired values ​​for other process parameters may also be identified (e.g., after iterations of steps 1002, 1004, and / or 1006).

[0059] In optional step 1010, optimal values ​​for the process parameters are identified based on the results of step 1002 (e.g., including any iterative executions of step 1002). Desired values ​​for other process parameters may also be identified (e.g., after iterations of step 1002, step 1004, and / or step 1006).

[0060] Although the present invention has been described primarily with respect to testing the process window of particular process parameters (e.g., USG, BF, SCRUB) in a wire bonding system, the invention is not so limited. Different and / or additional process parameters (and different and / or additional responses) are contemplated within the scope of the present invention.

[0061] Although the invention has been shown and described herein with reference to specific embodiments, the invention is not intended to be limited to the details shown. Rather, various modifications can be made in the details within the scope of the claims and their equivalents without departing from the invention.

Claims

1. 1. A method for testing a process window of a process parameter in a wire bonding system, comprising: (a) identifying the process parameters to be used in a wire bonding operation in the wire bonding system; (b) testing a response of at least one of the process parameters at a plurality of values ​​of the process parameter in the wire bonding system; (c) automatically recording the at least one response of the process parameter at the plurality of values ​​of the process parameter; A method having the following.

2. 10. The method of claim 1, further comprising the step (d) of identifying an acceptable process window for the process parameters based on the results of step (b).

3. 2. The method of claim 1, further comprising the step of identifying optimal values ​​of the process parameters based on the results of step (b).

4. 10. The method of claim 1, further comprising the step of automatically recovering from a process error during step (b) without operator intervention.

5. 5. The method of claim 4, wherein the step of automatically recovering from the process error includes at least one of recovering from a lifted ball condition, recovering from a lifted stitch condition, and recovering from a short tail condition.

6. 10. The method of claim 1, wherein the process parameters relate to at least one of ultrasonic energy applied during formation of a wire bond, bonding force applied during formation of a wire bond, and table scrub energy applied during formation of a wire bond.

7. 7. The method of claim 6, wherein the wire bond is selected from the group consisting of a first wire bond of a wire loop, a second wire bond of a wire loop, an intermediate wire bond of a wire loop, a conductive bump bond, and a wire bond of a vertical wire structure.

8. 7. The method of claim 6, wherein the wire bond is selected from the group consisting of a wire loop ball bond and a wire loop stitch bond.

9. 2. The method of claim 1, wherein the at least one response is related to at least one of ball lift of the ball bond, stitch lift of the stitch bond, short tail condition, pull value of the wire bond, shape of the wire bond, and dimension of the wire bond.

10. 10. The method of claim 1, wherein the process parameter relates to ultrasonic energy applied during formation of a ball bond.

11. The method of claim 10 , wherein the at least one response is related to ball lift of the ball bond.

12. 12. The method of claim 11, further comprising the step of automatically recovering from a process error during step (b) without operator intervention.

13. 13. The method of claim 12, wherein the step of automatically recovering from the process error includes bonding the floating ball to another location and continuing with step (b).

14. 14. The method of claim 13, wherein automatically recovering from the process error comprises forming a wire tail after bonding the floating ball using predetermined parameters.

15. 10. The method of claim 1, wherein the process parameter relates to ultrasonic energy applied during formation of a stitch bond.

16. 16. The method of claim 15, wherein the at least one response is related to a stitch lift of the stitch bond.

17. 17. The method of claim 16, further comprising the step of automatically recovering from a process error during step (b) without operator intervention.

18. 20. The method of claim 17, wherein automatically recovering from the process error comprises bonding the floating stitch to another location and continuing with step (b).

19. 20. The method of claim 18, wherein automatically recovering from the process error comprises forming a wire tail after bonding the floating stitch using predetermined parameters.

20. 2. The method of claim 1, wherein each of steps (b) and (c) is repeated a predetermined number of iterations at each of the plurality of values ​​of the process parameter.

21. 10. The method of claim 1, wherein each of steps (b) and (c) is repeated while varying another process parameter, and testing the response of the at least one process parameter in the wire bonding system at the multiple values ​​of the process parameter and the other process parameter.

22. 22. The method of claim 21, further comprising the step (d) of identifying an acceptable process window for the process parameter and a desired value for the other process parameter after repeating steps (b) and (c).

23. 22. The method of claim 21, further comprising the step of identifying an optimum value of the process parameter and a desired value of the other process parameter after repeating steps (b) and (c).

24. 1. A method for testing a process window of a process parameter in a wire bonding system, comprising: (a) identifying the process parameters to be used in a wire bonding operation in the wire bonding system; (b) testing a response of at least one of the process parameters at a plurality of values ​​of the process parameter in the wire bonding system; (c) identifying an acceptable process window for the process parameters based on the results of step (b); A method having the following.

25. 25. The method of claim 24, further comprising automatically recording the response of the at least one process parameter at the plurality of values ​​of the process parameter.

26. 25. The method of claim 24, further comprising identifying optimal values ​​of the process parameters based on the results of step (b).

27. 25. The method of claim 24, further comprising automatically recovering from a process error during step (b) without operator intervention.

28. 28. The method of claim 27, wherein automatically recovering from the process error includes at least one of recovering from a lifted ball condition, recovering from a lifted stitch condition, and recovering from a short tail condition.

29. 25. The method of claim 24, wherein the process parameters relate to at least one of ultrasonic energy applied during formation of a wire bond, bonding force applied during formation of a wire bond, and table scrub energy applied during formation of a wire bond.

30. 30. The method of claim 29, wherein the wire bond is selected from the group consisting of a first wire bond of a wire loop, a second wire bond of a wire loop, an intermediate wire bond of a wire loop, a conductive bump bond, and a wire bond of a vertical wire structure.

31. 30. The method of claim 29, wherein the wire bond is selected from the group consisting of a wire loop ball bond and a wire loop stitch bond.

32. 25. The method of claim 24, wherein the at least one response is related to at least one of ball lift of the ball bond, stitch lift of a stitch bond, a short tail condition, a pull value of a wire bond, a shape of a wire bond, and a dimension of a wire bond.

33. 25. The method of claim 24, wherein the process parameter relates to ultrasonic energy applied during formation of a ball bond.

34. 34. The method of claim 33, wherein the at least one response is related to ball lift of the ball bond.

35. 35. The method of claim 34, further comprising automatically recovering from a process error during step (b) without operator intervention.

36. 36. The method of claim 35, wherein automatically recovering from the process error comprises bonding the floating ball to another location and continuing with step (b).

37. 37. The method of claim 36, wherein automatically recovering from the process error comprises forming a wire tail after bonding the floating ball using predetermined parameters.

38. 25. The method of claim 24, wherein the process parameter relates to ultrasonic energy applied during formation of a stitch bond.

39. 39. The method of claim 38, wherein the at least one response is related to a stitch lift in the stitch bond.

40. 40. The method of claim 39, further comprising automatically recovering from a process error during step (b) without operator intervention.

41. 41. The method of claim 40, wherein automatically recovering from the process error comprises bonding the floating stitch to another location and continuing with step (b).

42. 42. The method of claim 41, wherein automatically recovering from the process error comprises forming a wire tail after bonding the floating stitch using predetermined parameters.

43. 25. The method of claim 24, wherein step (b) is repeated a predetermined number of times at each of the plurality of values ​​of the process parameter before step (c).

44. 25. The method of claim 24, wherein step (b) is repeated while varying other process parameters, and prior to step (c), testing the response of the at least one process parameter in the wire bonding system to the multiple values ​​of the process parameter and the other process parameter.

45. 45. The method of claim 44, further comprising the step (d) of identifying desired values ​​of the other process parameters after repeating step (b).

46. 45. The method of claim 44, further comprising the step of identifying an optimum value of the process parameter and a desired value of the other process parameter after the repeated step (b).

47. 1. A method for testing a process window of a process parameter in a wire bonding system, comprising: (a) identifying the process parameters to be used in a wire bonding operation in a wire bonding system; (b) testing a response of at least one of the process parameters at a plurality of values ​​of the process parameter in the wire bonding system; (c) identifying optimal values ​​of the process parameters based on the results of step (b); A method having the following.

48. 48. The method of claim 47, further comprising the step (d) of identifying an acceptable process window for the process parameters based on the results of step (b).

49. 48. The method of claim 47, further comprising automatically recording the response of the at least one process parameter at the plurality of values ​​of the process parameter.

50. 48. The method of claim 47, further comprising automatically recovering from a process error during step (b) without operator intervention.

51. 51. The method of claim 50, wherein automatically recovering from the process error includes at least one of recovering from a lifted ball condition, recovering from a lifted stitch condition, and recovering from a short tail condition.

52. 48. The method of claim 47, wherein the process parameters relate to at least one of ultrasonic energy applied during formation of a wire bond, bonding force applied during formation of a wire bond, and table scrub energy applied during formation of a wire bond.

53. 53. The method of claim 52, wherein the wire bond is selected from the group consisting of a first wire bond of a wire loop, a second wire bond of a wire loop, an intermediate wire bond of a wire loop, a conductive bump bond, and a wire bond of a vertical wire structure.

54. 53. The method of claim 52, wherein the wire bond is selected from the group consisting of a wire loop ball bond and a wire loop stitch bond.

55. 48. The method of claim 47, wherein the at least one response is related to at least one of ball lift of the ball bond, stitch lift of a stitch bond, a short tail condition, a pull value of a wire bond, a shape of a wire bond, and a dimension of a wire bond.

56. 48. The method of claim 47, wherein the process parameter relates to ultrasonic energy applied during formation of a ball bond.

57. 57. The method of claim 56, wherein the at least one response is related to ball lift of a ball bond.

58. 58. The method of claim 57, further comprising automatically recovering from a process error during step (b) without operator intervention.

59. 60. The method of claim 58, wherein automatically recovering from the process error comprises bonding the floating ball to another location and continuing with step (b).

60. 60. The method of claim 59, wherein automatically recovering from the process error comprises forming a wire tail after bonding the floating ball using predetermined parameters.

61. 48. The method of claim 47, wherein the process parameter relates to ultrasonic energy applied during formation of a stitch bond.

62. 62. The method of claim 61, wherein the at least one response is related to a stitch lift in the stitch bond.

63. 63. The method of claim 62, further comprising automatically recovering from a process error during step (b) without operator intervention.

64. 64. The method of claim 63, wherein automatically recovering from the process error comprises bonding the floating stitch to another location and continuing with step (b).

65. 65. The method of claim 64, wherein automatically recovering from the process error comprises forming a wire tail after bonding the floating stitch using predetermined parameters.

66. 48. The method of claim 47, wherein step (b) is repeated a predetermined number of times at each of the plurality of values ​​of the process parameter before step (c).

67. 48. The method of claim 47, wherein step (b) is repeated while varying other process parameters, and prior to step (c), testing the response of the at least one process parameter in the wire bonding system at the multiple values ​​of the process parameter and the other process parameter.

68. 68. The method of claim 67, further comprising the step of: (d) after the repeated step (b), identifying an acceptable process window for the process parameter and desired values ​​for the other process parameters.

69. 68. The method of claim 67, further comprising the step of identifying desired values ​​of the other process parameters after repeating step (b).

Images