Technology Drivers for Plasma Prior to Wire Bonding

The wire bonding process has to be optimized to ensure high device reliability and reduce manufacturing costs, ensuring good yields and bond strengths. Low yields and poor bond strengths are often due to the selection of materials in advanced packaging or upstream contamination sources. Gas plasma technology can be used to clean pads before wire bonding to increase yields and bond strengths.

Gas plasma is a powerful, efficient resource, and when used correctly it can greatly enhance the manufacturability, yield, and reliability of advanced semiconductor packages. Plasma is used to improve the uniformity and pull strength of wire bonds; increase fillet uniformity, fillet height, and underfill adhesion for flip chip devices; and modify surfaces for improved adhesion in mold and encapsulation processes.

A number of factors dictate the effectiveness of a plasma process including the choice of chemistry, power, part placement, time, process parameters, and electrode configuration. For a specific packaging application, electrode configuration and process chemistry are the main factors.

Effective implementation of plasma in packaging these complex devices requires knowledge about the device to be packaged, including its materials of construction, the pre-processing steps and any sensitivity, as well as plasma technology.

Thin Film Metallization

In currently available advanced substrate technologies, low-priced substrates are manufactured using extremely thin gold plating typically on nickel or palladium metallization. The thickness of gold is very thin, normally less than 50 nm.

This gold thickness poses a challenge to the plasma system when looking to use plasma for bond pad contamination removal due to epoxy bleed-out caused during the die attach step. When the epoxy is present, it can lead to poor bonding yields and wire bond pull strengths. The challenge is to effectively eliminate the organic resin bleed with plasma without removing or damaging the thin film gold needed for the wire bonding step.

Two plasma modes can be used to treat the substrate before wire bonding: downstream or direct plasma. The direct plasma mode uses an energy source to ionize and dissociate a source gas producing a gas plasma made up of chemically and physically active components.

Samples to be plasma treated are directly placed in the gas discharge, on or close to the electrode plates of the system with complete exposure to the working species of the plasma (i.e., ions, byproducts, and free radicals). The type of working species that the substrate is exposed to is a function of the source gas selected.

For instance, if argon was used as a source gas, the argon ions generated by plasma would influence the surface of the substrate and eliminate the organic residue through a sputtering mechanism. In the following example, a quad flat no-lead (QFN) package with 25 nm of gold on palladium was checked for wire bond improvement with and without argon direct plasma.

The die was attached with conductive epoxy, oven cured, direct plasma treated, and wire bonded with 25-micron wire. A statistically valid set of samples produced a mean pull strength of 10.00 g with a CpK of 2.07 with plasma, compared to a mean pull strength of 3.89 with a CpK of 0.03 without plasma.

This example reveals that direct plasma can be used to greatly enhance wire bond pull strengths under tightly controlled process conditions.

Oxygen can also be chosen as the source gas. In this case, the active species generated in the plasma include oxygen radicals, oxygen ions, and byproducts such as ozone. The plasma generated oxygen radicals oxidize the organic resin, producing water and gas phase carbon dioxide with slight help from the oxygen ions.

Downstream ion free plasma is an alternative to direction plasma. Ion free plasma (IFP) plasma is a pure chemical plasma, free from photons and ions responsible for the physical component. The IFP process comprises of the generation of active species upstream of the sample processing area, which is followed by diffusion of active species via a gas baffle assembly.

The gas baffle removes the ions, photons, and electrons, enabling the substrate to only be exposed to the radicals, and byproducts produced in the upstream plasma. The downstream plasma mode can be used whenever the die or substrate is sensitive to the exposure of photons or ions generated in the direct plasma.

An example when considering the use of direct plasma versus downstream plasma is when processing substrates using thin metallization. In one plasma cycle, all of the gold on the substrate bond pads can possibly be removed, greatly influencing the wire bond pull strengths.

In the following example, identical QFN packages with wire bond pads comprising of 25 nm of gold on palladium, were die attached with conductive adhesive, oven cured, direct plasma treated using argon source gas under varying power and time conditions, and wire bonded using 25 micron wire.

Table 1. shows the importance of tightly controlling the plasma process to make sure that all of the organic resin bleed is removed without sputtering the thin film gold on the bond pad. The pull strength data (Table 1) was gathered with a constant plasma power, argon source gas, and pressure while the plasma process time was changed.

The “Under Treated” sample exhibited some improvement over the “No Plasma” sample, but when compared to the “Optimized” conditions it was clear that the process did not fully removed the epoxy bleed. An extra set of experiments was performed to demonstrate the significance of tightly controlling the plasma process. The “Over Treated” sample yielded poor pull strengths as the thin film gold bond pad material was removed.

Table 1.

Sample Conditions Mean Pull Strength (grams) CpK
No Plasma 3.72 g 0.07
Under Treated 4.67 g 0.35
Optimized 8.52 g 2.15
Over Treated 4.82 g 0.45

IFP plasma can be applied in cases where the semiconductor device technology or the substrate metallization is sensitive to the direct plasma exposure. A thin film gold QFN package was die attached with conductive adhesive, thermally cured, plasma treated under ion-free and direct plasma conditions, and wire bonded using 25 micron wire.

In ion-free plasma, the QFN package will be exposed only to the chemically active oxygen radicals, restricting the effect of sputtering the gold. Figure 1 shows pull strengths for these QFN packages under direct oxygen plasma, no plasma, and IFP oxygen plasma.

In both plasma cases, the CpK and wire bond pull strength improve considerably when compared to the no-plasma condition. Conversely, the direct plasma condition is a little better, indicating that removal of the organic resin bleed is not the only mechanism for better bond pull strength.

Further studies are being conducted to better understand the above observation.

direct vs ion free plasma

Additive Substrate Technology

There are three key types of metallization techniques in the manufacture of substrates: additive, subtractive, and semi-additive. The conventional techniques use subtractive metallization which involves the application of a blank metal followed by photolithography and metal etch of the metal to form the substrate traces.

In additive plating, the metal traces are directly constructed on the substrate. Additive plating is frequently being used as it offers benefits for small geometries required in high density substrates. There are two standard sources of contamination with additive plating: nickel diffusion from the plating that can influence wire bonding pull strength and yield, and organic contamination from the substrate manufacturing process.

An appropriately configured plasma system can efficiently treat these sources of contamination and enhance wire bond yields.

An additive plated substrate was used to analyze the effectiveness of plasma for boosting wire bond pull strength under conditions of argon-based plasma, oxygen based plasma, and no plasma. The results are illustrated in Table 2, showing that both plasma processes greatly enhance the pull strengths while maintaining high CpK values.

However, the pull strength data does not reveal if the pull strength improvement is because of the reduction of nickel on the bond pad surface or the removal of organic contamination.

Table 2.

Condition Average Pull Strength (grams) CpK
No Plasma 0.86 grams 2.80
Oxygen Based 10.03 grams 3.82
Argon Based 10.93 grams 5.44

To understand more about the plasma-enhanced pull strength improvement, X-ray photoelectron spectroscopy (XPS) was used to assess the performance of the two different plasma processes for removal of the nickel and organic contamination. Relative concentrations of nickel, carbon, and gold were measured on the substrate bond pads. The results are illustrated in Table 3.

The no plasma condition data reveals that the gold bond pad is contaminated with organic contamination as shown by the high nickel and carbon content. While the oxygen-based process is efficient for removing organic contamination through a chemical mechanism, it does not effectively treat the nickel. An argon sputtering process will be highly efficient in eliminating the nickel contamination.

Table 3.

Condition Carbon (%) Nickel (%) Gold (%)
No Plasma 70.9 1.4 27.7
Oxygen Based Plasma 54.1 3.3 42.6
Argon Based Plasma 50.3 Not Detected 49.7

By analyzing the oxygen-based data closely, it can be concluded that a lot of the organic contamination restricting the wire bond pull strength is effectively eliminated and the remaining organic is adventitious carbon as seen both by the increase in pull strengths shown in Table 2, and the relative increase concentration of gold illustrated in Table 3.

In addition, the relative increase in the nickel content for the oxygen-based plasma is probably due to the exposure of the bond pad nickel contamination lying below the organic and the effective reduction of the top layer organic contamination. However, it is observed that the relative small quantity of nickel does not seem to be the major factor in the pull strength improvement when the no plasma condition in Table 2 is compared to both of the plasma conditions.

A sputtering mechanism is used by the argon-based plasma to eliminate both the nickel and the organic. The data in Table 3 shows both the reduction in the nickel and carbon levels. The minor improvement in the pull strengths for the argon process as shown in Table 2 is probably due to the reduction in the nickel content on the bond pad.

When considering the type of plasma chemistry needed for the additive plated substrates, users have to balance throughput needs with the chemistry. Typically, the chemically based processes such as the oxygen plasma will offer shorter cycle times than those driven just by sputtering processes. In either case, the plasma process allows these substrates to be wire bonded.

Conclusion

Advanced packaging technologies continue to promote the progress of material innovations to meet the needs of additional functionality in smaller packages. Using these innovative materials, plasma processing is frequently required in wire bonding applications to allow satisfactory pull strengths and better bonding yields.

Material sensitivity and considerations for the contamination sources have to be taken into consideration while configuring the plasma system. When plasma is optimally configured, it becomes an enabling technology for the improvement of advanced package reliability and yields.

This information has been sourced, reviewed and adapted from materials provided by Nordson MARCH.

For more information on this source, please visit Nordson MARCH.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Nordson MARCH. (2019, September 26). Technology Drivers for Plasma Prior to Wire Bonding. AZoM. Retrieved on February 20, 2020 from https://www.azom.com/article.aspx?ArticleID=13282.

  • MLA

    Nordson MARCH. "Technology Drivers for Plasma Prior to Wire Bonding". AZoM. 20 February 2020. <https://www.azom.com/article.aspx?ArticleID=13282>.

  • Chicago

    Nordson MARCH. "Technology Drivers for Plasma Prior to Wire Bonding". AZoM. https://www.azom.com/article.aspx?ArticleID=13282. (accessed February 20, 2020).

  • Harvard

    Nordson MARCH. 2019. Technology Drivers for Plasma Prior to Wire Bonding. AZoM, viewed 20 February 2020, https://www.azom.com/article.aspx?ArticleID=13282.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback
Submit