Date/Time: 04-23-2019 - Tuesday - 05:00 PM - 07:00 PM
Saaketh Narayan1 2 3 Jack Day1 2 3 Nikhil Suresh1 2 4 Nicole Herbots1 2 3 Robert Culbertson1 2

1, Arizona State University, Tempe, Arizona, United States
2, SiO2 Innovates, LLC, Tempe, Arizona, United States
3, AccuAngle Analytics, LLC, Tempe, Arizona, United States
4, MicroDrop Diagnostics, LLC, Phoenix, Arizona, United States

Surface energy and hydro-affinity play a significant role in device manufacturing and semiconductor processing. Surface reactivity, bonding and passivation is as critical as structure and composition for gate oxides, ohmic contact formation, heterostructures, opto-electronic and piezo-electronic device integration.

In this work, the effect of doping species upon Si(100) total surface energy, γT, hydro-affinity, and reactivity, is investigated for three n-type dopants, phosphorus, arsenic or antimony, and two p-type dopants, boron or gallium, [2] by new quantitative measurements via Three Liquid Contact Angle Analysis (3LCAA) [1] correlated with Ion Beam Analysis (IBA), as a function of concentration.

3LCAA and the van Oss-Chaudhury-Good (vOCG) theory yield γT from three interactions, the Lifshitz-van der Waals interaction energy γLW, the interaction energy with electron donors γ-, and acceptors γ+, measured via the contact angles of three liquids.

A new automated image analysis algorithm, DROP™, [3], yields contact angles within <1° and γT within 3% in minutes. DROP™’s accuracy makes possible quantitative analysis of γT, γLW, γ-, and γ+ as a function of doping species and concentrations.

Ion Beam Analysis (IBA) can detect oxygen coverage and elemental composition. High resolution IBA combes ��(O16, O16)�� 3.039±0.01 MeV nuclear resonance with <111> channeling detects oxygen within 5%, or about 6x1014 at/cm2. Rutherford Backscattering Spectrometry (RBS) combined with <111> axial channeling can detect dopants with a sensitivity that increases quadratically with their atomic number, thus ~4x1020 boron/cm3, or 1 at.%; ~4x1019 phosphorus/cm3, or 0.1 at.%; ~1x1019 gallium/cm3, or 0.02 at.%; ~1x1019 arsenic/cm3, or 0.02 at.%; and ~4x1018 antimony/cm3, or 0.01 at.% [2]. Oxygen coverage measured by IBA, along with doping levels and dopant species, can be correlated with γT, γLW, γ-, and γ+ measured by 3LCAA on both as-implanted and Rapid Thermal Annealed (RTA) Si(100). Si(100) amorphizes with increasing ion dose and defect concentrations, which catalyzes native oxide formation. During RTA, Si(100) recrystallizes and incorporates dopants into substitutional lattice sites. However, native oxide formation and dopant segregation modify final native oxide composition and thickness, as well as total surface energy and the surface interactions with donors and acceptors.

Higher dopant concentrations in both p-type and n-type doped Si(100) result in larger oxygen coverage and thus thicker native oxides, as measured by IBA, and lower surface energies as measured by 3LCAA. As expected, resulting surfaces are more hydrophobic and less reactive than as-implanted Si(100).

3LCAA also finds that as Si resistivity increases, γT also increases, showing that lower doping concentrations result in a more hydrophilic and more reactive surface [3]. Implanted Si is then etched, which removes surface oxides, and recrystallized by RTA to electrically activate dopants, regrowing a new native oxide. Longer RTA processes cause thicker oxides to form, along with higher segregation of n-type dopants at the Si-SiO2 interface. Hence, electrically activated n-type dopants enhance oxidation. The opposite effect is observed for recrystallized p-type doped Si, which shows thinner native oxides than as-implanted Si. 3LCAA, combined with IBA, can accurately measure changes to native oxide surface energies, hydro-affinity and reactivity as a function of dopant species and dopant concentration. These insights from 3LCAA expand understanding of dopants’ influence upon native oxide formation and of surface engineering for heterostructure formation by heteroepitaxy, conventional wafer bonding and NanoBonding™ [3].

[1] US Patents #9,018,077; #9,589,801, Herbots N. et al (2015); (2017)
[2] P. A. Cullen, Ph.D, Massachusetts Institute of Technology PhD Thesis (1991)
[3] US Patents Pending, Nicole Herbots, Saaketh Narayan, Jack Day, et al. (2018)

Meeting Program

Symposium Sessions

5:00 PM–7:00 PM Apr 23, 2019 (US - Arizona)

PCC North, 300 Level, Exhibit Hall C-E