An Investigation of Tin Electroless Deposition

Monday, 10 October 2022: 15:00
Room 301 (The Hilton Atlanta)
E. J. O'Sullivan, C. T. Camagong, R. Paranjape, M. Hopstaken, and C. Lavoie (IBM Research Division, Thomas J. Watson Research Center)
Electroless metal/alloy deposition can be an efficient process in certain areas of microelectronic fabrication. In fact, it is often easier to obtain coatings of uniform thickness and composition using electroless deposition than with electrodeposition, since one does not have the current density uniformity problem of the latter. For example, we were able to develop a Ni(P) process as a replacement for the final aluminum interconnect level, significantly decreasing wafer processing cycle time, by selectively depositing a Ni(P) capping layer on the Cu bitline wiring level. In STTM MRAM, we successfully employed this Ni(P) capping process to enable the evaluation of memory state retention via functional testing in an air atmosphere at elevated temperatures (1).

However, there is a need to explore materials + deposition methods for rapidly developing fields, such as Quantum computing, e.g., materials with superconducting properties. Conventional electroless metal deposition, utilizing a separate reducing agent, can deposit materials with unique and useful properties, such as phosphorus-containing alloys in the case of hypophosphite reducing agents. This talk discusses work we have carried out on electroless tin deposition, including aspects of electroless solution preparation and stability, copper substrate surface preparation and catalyzation, and the mechanisms of electroless deposition and solution decomposition.

Electroless processes can deposit a limited number of materials, especially pure metals. This is in part due to conventional electroless processes requiring catalytically active surfaces both to initiate the deposition reaction and to sustain it, the heterogeneous oxidation of the reducing agent being a kinetically hindered process, often with multiple reaction pathways (2). Though not possessing good catalytic activity due to its Periodic Table related, p-block element status, pure Sn, an environmentally robust, superconducting metal, can be electrolessly deposited through a disproportionation reaction involving stannous ions (3) in an alkaline aqueous medium.

We achieved electroless Sn deposition rates of up to 8 – 9 μm/hr for tartrate-citrate complexed electroless Sn solutions in the temperature range 80 – 85 ⁰C with sodium and potassium hydroxides to adjust alkalinity. We found that either in-house formulated, or commercially available, immersion Sn solutions deposited a uniform Sn catalyst layer (≤ 0.5 μm) to initiate the electroless Sn deposition reaction on copper; however, improperly formulated immersion Sn solutions rapidly developed precipitates due to tin ion hydrolysis.

The biggest technical challenge was minimizing unwanted electroless deposition of tin in bulk solution, i.e., deposition not associated with the catalytically active substrate surface. Tin oxide (SnO) is known to be metastable at ambient conditions and to decompose at temperatures above 300 ⁰C with “noticeable rate” into Sn and SnO2 (4). Thus, removal of filterable hydrolysis products of Sn(II) following solution preparation was important, but not always sufficient, for obtaining solutions that were viable for several days of use. The reasons for, and mechanisms of, electroless Sn solution decomposition do not appear to have been adequately addressed in the literature.

We will show SIMS analysis of both immersion and electroless Sn layers along with synchrotron X-Ray analysis results of immersion Sn catalyst films on Cu to determine the extent of Sn-Cu intermetallic formation following their formation. We will discuss the current understanding of the mechanism of electroless Sn deposition including that of concomitant H2 gas evolution. We will conclude with contrasting the Ni(P) and Sn electroless processes in terms of ease of operation and reliability for routine processing.

Present address: Solvay, 1937 West Main Street, Stamford 06902, CT.

Quantum intern at the IBM TJ Watson Research Center, Summer 2019, 2020 and 2021.

[1]. E. J. O'Sullivan, C. Camagong et al., 2019 Meet. Abstr. MA2019-02 916; https://doi.org/10.1149/MA2019-02/15/916.

[2]. E. J. O'Sullivan, Ch 5, Advances in Electrochemical Science and Engineering, Volume 7, https://doi.org/10.1002/3527600264.ch5.

[3]. A. Molenaar and J. W. G. de Bakker, 1989, J. Electrochem. Soc. 136, 378 and refs therein; H. Koyano, M. Kato, and M. Uchida, 1991, Plating and Surface Finishing, 78, 68-74 and refs therein.

[4]. H. Giefers et al, 2005, Solid State Ionics, 176, 199-207; https://doi.org/10.1016/j.ssi.2004.06.006.

Acknowledgements

The authors gratefully acknowledge the efforts of the staff of the Microelectronics Research Laboratory (MRL) at the IBM T. J. Watson Research Center, where some of the fabrication work described in this talk was carried out.