Method for judging abnormal communication of infrared focal plane device flip-chip interconnection

Abstract

The application discloses a method for judging abnormality of flip-chip interconnection of an infrared focal plane device, and comprises the following steps: selecting a same layer lay out layer as a reference surface on four corners of a readout circuit, and the reference surface is close to an antireflection film of an infrared detector focal plane; performing 3D measurement on the four corners of the infrared detector focal plane, and the measurement area needs to contain the antireflection film of the infrared detector focal plane and the same layer lay out layer found on the readout circuit; performing height measurement on the measurement area, and measuring the height from the antireflection film to the readout circuit; calculating the height difference of the four corners, and then judging whether the flip-chip interconnection is abnormal or not. The application can quickly and effectively judge whether the abnormality of the interconnection is caused by the flip-chip interconnection abnormality, and will not cause any damage to the infrared focal plane.

Description

Technical Field

[0001] This invention belongs to the field of infrared detector technology, specifically relating to a method for determining abnormal interconnection of inverted infrared focal plane array devices. Background Technology

[0002] Flip-chip interconnect technology, as one of the important technologies in infrared detector manufacturing, not only possesses excellent electrical properties (due to short interconnects and low parasitic impedance, it has good resistance to electromagnetic interference), but also enables simultaneous electrical and mechanical interconnection. With the continuous development of infrared detector technology, pixel pitch is constantly shrinking. While small pixel devices bring advantages in imaging, they also present challenges to device manufacturing processes.

[0003] Currently, the fabrication process of the focal plane of an infrared detector using flip-chip interconnect technology is as follows: the detector chip and the readout circuit are connected together using flip-chip interconnect. The interconnect reliability is improved by the underfill process, the substrate is removed by the back thinning process, and then an anti-reflection film is deposited by the back anti-reflection process. At this point, the focal plane process is completed and testing can begin.

[0004] For flip-chip interconnects, the requirements for leveling and alignment are even higher. Large leveling errors can lead to uneven pressure during flip-chip interconnect bonding, resulting in excessive or insufficient pressure at the corners, causing signal anomalies at the corners during device imaging. Signal anomalies mainly fall into two categories: one is excessive pressure, causing excessive deformation of the indium pillars during flip-chip interconnection, bridging adjacent indium pillars and leading to short circuits and signal anomalies; the other is insufficient pressure, causing the indium pillars to shrink, failing to establish a connection between the device and the readout circuit, resulting in weak or no signal.

[0005] Currently, the judgment of flip-chip interconnect results is mainly based on test results, including response signal graphs and GPOL-V characteristic curves. From the response signal graph, the response graph corresponding to areas with heavy or light pressure shows small signals or dead cells. From the GPOL-V characteristic curve, insufficient pressure leading to no connection is manifested as an unopened state, while excessive pressure causing adhesion is manifested as a saturated state. However, these electrical performance states and flip-chip interconnect anomalies are not a one-to-one relationship. Simply put, flip-chip interconnect anomalies inevitably lead to the above electrical performance states, but these electrical performance states can be caused by multiple factors occurring individually or together. For example, many steps in the device manufacturing process, such as PN junction openings not being open or excessive hole depth, and abnormal metal electrode contact, can lead to connectivity anomalies. Therefore, test results alone cannot be used to determine connectivity anomalies.

[0006] Another method is to etch the thin film on the infrared focal plane and then confirm it by the shape and height of the indium pillar. However, this method causes irreversible damage to the infrared focal plane and is cumbersome and time-consuming.

[0007] Therefore, there is currently no fast, economical, and simple method to identify flip-chip interconnect anomalies. Summary of the Invention

[0008] The purpose of this invention is to overcome the above-mentioned shortcomings and provide an economical and convenient method for judging the abnormality of flip-chip interconnection in infrared focal plane array devices. This invention can quickly and effectively determine whether the abnormality is caused by the abnormality of flip-chip interconnection without causing any damage to the infrared focal plane array.

[0009] The infrared detector focal plane array, after the back antireflection process is completed, consists of a readout circuit, a mercury cadmium telluride thin film, an antireflection coating, and a filler adhesive.

[0010] The method for determining flip-chip interconnect anomalies according to the present invention has the following steps:

[0011] Step (1): Prepare a microscope with 3D measurement capabilities and an infrared detector focal plane that needs to be tested.

[0012] Step (2): Under the microscope, find the same layout layer at the four corners of the infrared detector readout circuit as the reference plane. This position should be close to the anti-reflection film of the infrared detector focal plane to facilitate subsequent measurements.

[0013] Step (3): Perform 3D measurement on the four regions found in step (2). The measurement region needs to include the reference plane of the readout circuit and the anti-reflection coating region of the focal plane of the infrared detector.

[0014] Step (4): Measure the height of the 3D measurement image obtained in step (3) to measure the height of the antireflection film up to the selected reference surface on the circuit read out in step (2);

[0015] Step (5): Calculate the relative height difference of the four points using the height obtained in step (4), and determine whether the inverted interconnection is abnormal, or whether the welding pressure is too high or too low based on the height difference.

[0016] Furthermore, the antireflection coating thickness is typically 3000–5000 angstroms, with a non-uniformity of less than 10%, so the height difference of the infrared focal plane antireflection coating can be denoted as Δh1. The mercury cadmium telluride film is grown on a substrate, and thickness variations also exist during growth, generally within 0.8 μm in a 25 × 20 mm range; this thickness difference is denoted as Δh2. During the soldering process, leveling cannot be absolutely uniform, resulting in differences in the height of the indium pillars in different areas after soldering; this height difference is denoted as Δh3. When the relative height difference from the antireflection coating to the readout circuit is greater than H = Δh... 1+ Δh 2+ Δh3 can be considered as an abnormality in the inverted interconnect connection.

[0017] Furthermore, if the height difference of a certain measurement point exceeds the threshold ΔH, and at the same time, "black and white dots" appear at the corresponding position of the focal plane signal response map and the signal of some pixels in the GPOL-V characteristic curve is saturated, then it is determined that the welding pressure at that measurement point is too high; if the height difference of a certain measurement point exceeds the threshold ΔH, and at the same time, "black and white dots" appear at the corresponding position of the focal plane signal response map and the signal of some pixels in the GPOL-V characteristic curve is not in a conducting state, then it is determined that the welding pressure at that measurement point is too low.

[0018] The beneficial effects of this invention are:

[0019] This invention provides an economical and simple method for determining the interconnection abnormality of an infrared focal plane array device. It can quickly and effectively determine whether the interconnection abnormality is caused by the flip interconnection abnormality, and will not cause any damage to the infrared focal plane array. Attached Figure Description

[0020] Figure 1 The diagram shows a cross-sectional view of the focal plane array of an infrared detector, where: 1-antireflective coating; 2-mercury cadmium telluride thin film; 3-indium pillar; 4-filler adhesive; 5-readout circuit.

[0021] Figure 2 The image shown is a top-down view of the infrared detector focal plane array, where: 1-the anti-reflection coating on the focal plane of the infrared detector; 2-the readout circuit. I, II, III, and IV on the focal plane in the image are the measurement points for 3D measurement in four regions.

[0022] Figure 3 This is a diagram of the signal response at the focal plane.

[0023] Figure 4 The graph shows the characteristic curves of GPOL-V. In the graph, the signal inside the solid box is in a saturated state, and the signal inside the dashed box is in a non-conducting state. Detailed Implementation

[0024] like Figure 1As shown, the infrared detector focal plane array after the back antireflection process consists of a readout circuit 1, a mercury cadmium telluride thin film 2, an antireflection film 3, and a filler adhesive 4. The thickness of the antireflection film 4 is generally 3000–5000 angstroms (4000 angstroms is taken), with a non-uniformity of less than 10% (0.04 μm). Therefore, the height difference of the infrared focal plane array antireflection film can be denoted as Δh1. The mercury cadmium telluride thin film 2 is grown on the substrate, and there are also differences in thickness during growth. Generally, the thickness difference is within 0.8 μm in the range of 25 × 20 mm, and this thickness difference is denoted as Δh2. During the soldering process, leveling cannot be absolutely uniform, so the height of the indium pillars in different areas varies after soldering. This height difference of the indium pillars is denoted as Δh3. When the relative height difference between the antireflection film 3 and the readout circuit 1 is greater than ΔH = Δh1 + Δh2 + Δh3, it can be considered as an abnormal flip-chip interconnect connection.

[0025] Example 1

[0026] like Figure 3 As shown in the figure, taking the signal response of a 1280×1024 array detector with a center-to-center distance of 15μm as an example, it can be seen from the figure that the corresponding signal at point A in the upper right corner is abnormal, and the image shows "black and white dots". Figure 4 The GPOL-V characteristic curves shown indicate that some pixels exhibit signal saturation (solid box) and others show non-conductivity (dashed box). Based on the principle, this is likely due to signal cascading between adjacent pixels. However, we cannot yet determine if this phenomenon is caused by a leveling issue during flip-chip interconnection, resulting in excessive pressure in the upper right corner.

[0027] The method proposed in this invention was used to determine the height of the infrared detection focal plane antireflection film to the surface of the readout circuit. The measurement results are shown in Table 1.

[0028] Table 1 Measurement results at four measurement points

[0029] Measurement points I II III IV Height (μm) 13.2 10.2 12.9 13.7

[0030] For a 1280×1024 area array detector with a center-to-center distance of 15μm, an indium pillar deformation of approximately 2–3.5μm after pressure bonding is considered effective for interconnection, so Δh3 is 1.5μm. Assuming a thickness difference of 0.8μm for the mercury cadmium telluride film, H can be calculated to be at least 2.34μm. Therefore, according to this invention, if the height difference between the antireflection coating at the focal plane of the infrared detector and the readout circuit is greater than 2.34μm, the inverted interconnection is considered abnormal. Based on this, the measurement results show that the signal anomaly at measurement point "II" is caused by a problem with leveling during the inverted interconnection, resulting in excessive pressure at measurement point "II". This excessive pressure causes some adjacent indium pillars at "II" to stick together, resulting in "black and white dots" on the signal response graph and saturation of the GPOL-V characteristic curve.

[0031] The focal plane H of different types of detectors is different, so it is determined based on the specific focal plane of the detector, and then used to determine whether there is an abnormality in the flip-chip interconnect.

Claims

1. A method for determining abnormal interconnect connectivity in an infrared focal plane array device, characterized in that, Includes the following steps: Step 1: Prepare a microscope with 3D measurement capabilities and an infrared detector focal plane device fabricated using flip-chip interconnect technology to be tested. The infrared detector focal plane device consists of a readout circuit, a mercury cadmium telluride thin film, an anti-reflection coating, and a filler adhesive. Step 2: Under the microscope, locate the same lay-out layer at the four corners of the infrared detector readout circuit as the reference plane. This position should be close to the anti-reflection coating of the infrared detector focal plane. Step 3: Perform 3D measurement on the four corner areas found in Step 2. The measurement area needs to include the reference plane of the readout circuit and the anti-reflection coating area of ​​the focal plane of the infrared detector. Step 4: Measure the height of the 3D measurement image obtained in Step 3, and measure the height of the antireflection film to the selected reference surface on the readout circuit in Step 2. Step 5: Calculate the relative height difference of the four points using the height obtained in Step 4, and determine whether there is an abnormality in the reverse interconnection based on whether the relative height difference of each measurement point exceeds the threshold. The height difference of the antireflection film is denoted as Δh1, the thickness difference of the mercury cadmium telluride film is denoted as Δh2, and the height difference of the indium pillar is denoted as Δh3. When the relative height difference between the antireflection film and the readout circuit exceeds the threshold ΔH = Δh1 + Δh2 + Δh3, it is considered an abnormal flip-chip interconnection.

2. The method for determining abnormal interconnect connectivity of an infrared focal plane array device according to claim 1, characterized in that, Step 5 further includes: After determining that there is an abnormality in the inverted interconnection, further determine whether the welding pressure is too high or too low.

3. The method for determining abnormal interconnection of an infrared focal plane array device according to claim 2, characterized in that: If the height difference at a certain measurement point exceeds the threshold ΔH, and at the same time, "black and white dots" appear at the corresponding position of the focal plane signal response map and the signal of some pixels in the GPOL-V characteristic curve is saturated, then it is determined that the welding pressure at that measurement point is too high.

4. The method for determining abnormal interconnection of an infrared focal plane array device according to claim 2, characterized in that: If the height difference at a certain measurement point exceeds the threshold ΔH, and at the same time, black and white dots appear at the corresponding position of the signal in the focal plane signal response map, and the signal of some pixels in the GPOL-V characteristic curve is in a non-conductive state, then it is determined that the welding pressure at that measurement point is too low.

Images