Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Partitioning of Water Between Voids and the Polymer Matrix in a Molding Compound by Proton NMR



David L. VanderHart, Michael A. Schen, G T. Davis


A commercial electronic packaging material (EPM), based on a silica-filled cresol-novolac epoxy, was fabricated into a 0.4 mm sheet, using methods mimicking commercial fabrication of EPM's used as encapsulants. Solid-state proton nuclear magnetic resonance methods were employed to quantify the volume fraction of voids, i.e., empty regions larger than 4 nm in radius, and to study their possible role in producing the undesirable phenomena of popcorning' and delamination--the latter associated with solder reflow for electronics packages that have been exposed to humid environments. Simple, single-pulse spectra, when taken at ambient temperature in the presence of magic-angle sample spinning, allow one to distinguish signals arising from the rigid matrix protons and the water protons. Spectral difference techniques further allow one to isolate the water signal from that of other, more mobile protons, most of which are probably associated with coupling agents on the surfaces of the silica filler particles. Samples, immersed in water in the absence of other gases, were equilibrated, thereby filling the voids with water. Water present in such voids can be distinguished from water dissolved in the polymer matrix on the basis of linewidth differences. The volume fraction of voids was determined to be 0.0020 0.0003. These immersion-equilibrated samples were subjected to increasing times of exposure to vacuum and the loss of matrix-water and void-water was monitored. Voids do not show low-impedance paths to the surfaces; hence their role in accelerating moisture transport during solder reflow is not obvious. Also, fast heating of high-humidity-equilibrated samples to solder reflow temperatures does not produce new low impedance paths. Equilibration at high humidity (89% RH) does not begin to fill these voids; hence, they are not expected to function as avoidable reservoirs for water prior to solder reflow. Based on the lack of condensation of water in these voids at 89% RH, a minimum radius of curvature (4.5 nm) characterizing these voids is estimated. The possibility that these voids arise from unwetted surfaces, most likely facilitated by the entrainment of a small amount of air, is raised. Accelerated aging under 2 atmospheres of water-vapor pressure (121 C) was also conducted. Since samples had to be decompressed and brought back to ambient temperature, little can be said about the distribution of water under test conditions. However, after immersion of such samples in water, the fraction of void water was reduced by 20%, the fraction of matrix-water increased by 45% and the longitudinal relaxation time for the void water became significantly shorter. The explanations for these changes are not clear but can include possibilities like chemical degradation of the matrix, water going into interface regions, and chemical leaching of impurities from the (naturally-occurring) silica.
International Journal of Microcircuits and Electronic Packaging


aging, delamination, electronic packaging material, heating, NMR partitioning, polymer, popcorning, proton, voids


VanderHart, D. , Schen, M. and Davis, G. (2008), Partitioning of Water Between Voids and the Polymer Matrix in a Molding Compound by Proton NMR, International Journal of Microcircuits and Electronic Packaging (Accessed June 10, 2023)
Created October 16, 2008