Patent Application
20230174423
The present disclosure generally relates to a glass substrate, such as a glass wafer, having through glass vias (TGVs). Specifically, the present disclosure is directed to glass wafers having through glass vias with a desired morphology and dimensional tolerances.
Dimensions of the TGVs may vary from via to via as a result of variations in the laser process, the etch process, and material used. Specific nominal hole dimensions and shapes must be in a predetermined range. If the range is exceeded, this may limit the overall product performance and also reduce production yield. Furthermore, when the range is exceeded corrective actions may include additional measurements and process controls, thereby resulting in lost time, capacity and increased cost.
The vias may be metallized to provide interconnections between a first surface and a second surface. Smaller diameter vias are beneficial for reducing spacing requirements and for the metallization process to hermetically seal the vias. Vias having too rough of an interior wall surface may cause electrically conductive material to not adhere to the interior wall surfaces during the metallization process or may delaminate in subsequent service, which may compromise various aspects of function and reliability.
According to one embodiment of the present disclosure, a wafer including a glass substrate is provided. The glass substrate includes a first surface defining a plane and including an average surface roughness Ra of approximately 0.3 nm in an outer via region, wherein the average surface roughness Ra of the plane in the outer via region is an average of at least five measurements, and a second surface. The glass substrate defines a plurality of vias extending from the first surface. The plurality of vias each include an entrance defined by the first surface and including an entrance diameter and an interior sidewall proximate the entrance. A ratio of a depression depth to the entrance diameter of the plurality of vias is not greater than 0.0006. The depression depth is measured from the plane to a transition point from a depressed region to the interior sidewall. The outer via region is at least 250 μm from any one of the plurality of vias.
According to another embodiment of the present disclosure, a wafer including a glass substrate is provided. The glass substrate includes a first surface defining a first plane including an average surface roughness Ra of approximately 0.3 nm in an outer via region, wherein the average surface roughness Ra of the first plane in the outer via region is an average of at least five measurements, and a second surface defining a second plane. The glass substrate defines a plurality of vias extending from the first surface to the second surface. The plurality of vias each include a first opening defined by the first surface and including a first diameter and a second opening defined by the second surface and including a second diameter. The second opening is fluidly coupled to the first opening. An interior sidewall is disposed between the first opening and the second opening. A depressed region surrounds the first opening and includes a surface roughness Ra of less than 0.6 nm, wherein the average surface roughness Ra of the depressed region is an average of at least five measurements. The outer via region is at least 250 μm from any one of the plurality of vias. A ratio of a depression depth to the first diameter of the plurality of vias is not greater than 0.0006. The depression depth is measured from the first plane to a transition point from the depressed region to the interior sidewall.
According to yet another embodiment of the present disclosure, a method of forming a glass wafer is provided. The method includes providing a glass substrate including a surface defining a plane and an average surface roughness Ra of approximately 0.15 nm, wherein the average surface roughness Ra of the plane is an average of at least five measurements. The method further includes applying pulsed laser beams to the glass substrate to form a plurality of laser damage lines within the glass substrate. The glass substrate is etched in an etching solution to enlarge the plurality of laser damage lines to form a plurality of vias within the glass substrate. A ratio of a depression depth to an entrance diameter of the plurality of vias is not greater than 0.0006. The depression depth is measured from the plane defined by the surface to a transition point from a depressed region to an interior sidewall of the plurality of vias.
Additional features and advantages will be set forth in the detailed description which follows, and in part will be readily apparent to those skilled in the art from that description or recognized by practicing the embodiments as described herein, including the detailed description which follows, the claims, as well as the appended drawings.
It is to be understood that both the foregoing general description and the following detailed description are merely exemplary, and are intended to provide an overview or framework to understanding the nature and character of the claims. The accompanying drawings are included to provide a further understanding, and are incorporated in and constitute a part of this specification. The drawings illustrate one or more embodiments, and together with the description serve to explain principles and operation of the various embodiments.
Reference will now be made in detail to the present preferred embodiments, examples of which are illustrated in the accompanying drawings. Whenever possible, the same reference numerals will be used throughout the drawings to refer to the same or like parts.
Referring to
Referring now to
The glass wafer 10 may be made of a variety of materials selected to tailor thermal and elastic properties, including, but not limited to, fused silica, high purity fused silica (Corning) HPFS®, Eagle XG® fusion drawn glass, alkali-free silicate glasses, borosilicate glass, ultra-low expansion glass (e.g. SiO2—TiO2), alkali silicate glasses (e.g. Corning® Gorilla® glass, soda lime glass), and the like. The glass wafer 10 can be made by any suitable process. In some examples, the glass wafer 10 may be made by forming a large boule and coring the boule to include a desired shape (e.g. circular). Following a coring process, the boule may be wire-sawn into slices. Each slice may then be polished or etched to have a desired surface polish/finish and edge finishing for the glass wafer 10. In other examples, the glass wafer 10 may be made in a fusion drawing process forming glass sheets having a desired thickness. The glass sheets may be cut into a desired shape (e.g. circular) for the glass wafer 10.
Referring now to
Referring now to
As illustrated in
As illustrated in
The A-side (e.g. the first surface 14) via opening diameter (e.g. the first opening diameter 44) may be a key parameter for monitoring and controlling processing. The ratio of the A-side via opening diameter to B-side via opening diameter (e.g. the second opening diameter 46) is preferably 1:1, or approximately 1:1. Further, a relatively small waist diameter 50 is beneficial for sealing the vias 20 with copper plating (or other material). The waist diameter 50 is measured independently of A-side via opening diameter and B-side via opening diameter, but it is directly related to the A-side diameter via opening diameter.
Referring now to
While described as including method steps 102-106, the method 100 may include any suitable steps, or procedures, for forming the glass wafer 10 having the plurality of vias 20. For example, the method 100 may include a polishing step prior to step 104 (i.e. an ultra-polishing step performed on a starting material). The polishing step may include a process for ultra-polishing the glass substrate 12. In this way, the ultra-polishing step may result in the glass substrate 12 including a surface roughness Ra of approximately 0.15 nm.
In some examples, the method 100 may result in the depressed regions 32 surrounding the entrances 22 having an average surface roughness Ra of less than 0.6 nm in a range from approximately 10-80 μm from the transition point 30. Average surface roughness Ra in the depressed regions 32 may include an overall average of multiple measurements, which may include at least five measurements of varying locations from front and rear areas of depressed regions 32 of at least two of the plurality of vias 20. Further, an average entrance diameter 24 of the plurality of vias 20 may be in a nominal range of 45-55 μm. Moreover, the method 100 may result in the nominal range of entrance diameters 24 for a sample of the plurality of vias 20 to exhibit a relatively narrow standard deviation. For example, in a sample of 1000 of the plurality of vias 20, the average entrance diameter 24 of 997 of the plurality of vias 20 may be within the nominal range of 45-55 μm. In addition, said 997 of the plurality of vias 20 may be in a range of 6 μm. By providing a plurality of vias 20 having tighter dimensional tolerances (e.g. exhibiting a relatively narrow standard deviation), nominal dimensions may be reduced, including average entrance diameter 24 and average waist diameter 50. Consequently, the glass wafers 10 may provide superior miniaturization and packaging efficiency. Furthermore, with a reduction to the central range (99.7%) of the via dimensions (e.g. average entrance diameter 24), downstream metallization and planarization processes are easier to control. For example, if the nominal waist diameter 50 is reduced, plating time (and consumption of plating chemistry) may be reduced. As plating process time extends until the largest diameter via is completed, a lower nominal waist diameter 50 and lower range are desirable for both product performance and for process efficiency.
Table 1, below shows results of one investigation of various parameters that are monitored to determine conformance of glass wafers. Table 1 shows that an improved A-side (e.g. the first surface 14) diameter central range of 99.7% is achieved using the glass wafers 10 (i.e. wafers formed from starting wafers having undergone the polishing step). Without wishing to be bound by theory, it is believed that the reduced standard deviation of the average entrance diameter 24 of the vias 20 is a direct result of reducing sub-surface damage (SSD) on the glass substrate 12 prior to step 104 (i.e. a starting wafer) as the presence of SSD may interact with both the laser and etch process. This may be explained by a reduction in pitting on the first and second surfaces 14, 18 (i.e. A-side and B-side) which results after step 106 (i.e. etching the vias) as SSD is not readily measurable. The reduced standard deviation of the average entrance diameter 24 of the vias 20 improves overall production yield. Improving overall production yield allows for reduced metrology sampling of roughness on depression region 32 surfaces that is typically associated with a laser damage and etch process.
Additionally, the method 100 may result in a surface roughness Ra in the depressed regions 32 of the vias 20 being smoother and shallower than typical depressed regions 32, or dimples. In some aspects, the surface roughness Ra in the depressed regions 32 of the vias 20 is different from the surface roughness Ra in the outer via regions 16. In one investigation, surface roughness Ra in two depressed regions 32 was measured with atomic force microscopy (AFM) using a 2.0 μm field of view (FOV) in five locations on the first surface 14 and the second surface 18 of one of the glass wafers 10. Surface roughness Ra in two depressed regions of a control glass wafer was also measured in five locations on each side of the glass control wafer. The AFM tip used was a ScanAsyst Air using a scan rate of 0.5 Hz and scan lines of 256×256.
Specifically, the locations measured ranged from approximately 10-80 μm from the transition point 30 as shown in Table 2, below. The difference between the control wafer and the glass wafer 10 is that the control wafer did not include a starting surface roughness Ra of approximately 0.15 nm prior to a laser damage and etch process (i.e. the starting control wafer did not undergo an ultra-polishing step). Table 2 shows that an improved final surface finish is achieved using starting wafers having undergone the polishing step. The average surface roughness Ra in the depressed regions 32 of the final glass wafer 10 according to aspects described herein was found to be less than 0.6 nm in a range from approximately 10-80 μm from the transition point 30. To the contrary, average surface roughness Ra in the depressed regions of the final control glass wafer was found to be approximately 0.8 nm in a range from approximately 10-80 μm from the transition point.
Table 3, below shows inspection data for another investigation including a sample of 256 of the glass wafers 10. The data shows that the glass wafers 10 do have very low average surface roughness Ra—on an order of 0.15 nm (measured with AFM using a 2 um FOV) on the starting material. Reduced sub-surface damage is indicated by the reduced count of pits (and in some examples, particles) after step 106 (i.e. the etching operation). One, or both, of the low average surface roughness Ra and reduced sub-surface damage characteristics appears to result in a laser damage track having less variability, thereby resulting in less entrance diameter 24 variation (e.g. 3σ value).
The glass wafers 10 described herein can be used to improve efficiency during production of glass wafers with through glass vias 20. The glass wafers 10 according to various aspects described here may result in more than a 2× improvement in via dimension tolerance. The improvement of via dimension tolerance is of commercial value due to significant reduction in cost of production, thereby increasing customer value. In one example, the shape of the via entrance and exit (e.g. the first and second openings 40, 42) is an important factor for plating processes and product reliability. Minimizing the depth and roughness of the region around the via facilitates subsequent processing, including metallization. A key factor during TGV manufacture is to precisely control the via dimension (e.g. entrance diameter 24, waist diameter 50, depression depth 28) size range. Further, when manufacturing capability is higher, there is less disruption of the production line and an increase in predictable output. Improved process capability enables more efficient statistical monitoring techniques, which is advantageous from all aspects of lean production.
It will be apparent to those skilled in the art that various modifications and variations can be made without departing from the spirit or scope of the claims.
This application claims the benefit of priority under 35 U. S.C. § 119 of U.S. Provisional Application Ser. No. 63/287,285 filed on Dec. 8, 2021, the content of which is relied upon and incorporated herein by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 63287285 | Dec 2021 | US |
