Browse Prior Art Database

Method And Device for High-Speed, High-Resolution Detection Of Thin Organic Films On Metals

IP.com Disclosure Number: IPCOM000102032D
Original Publication Date: 1990-Oct-01
Included in the Prior Art Database: 2005-Mar-17
Document File: 3 page(s) / 109K

Publishing Venue

IBM

Related People

Braun, JB: AUTHOR [+3]

Abstract

This article describes a high-speed, non-destructive technique for detecting the presence of thin organic films (residues of substrate manufacturing processes) on metal interconnecting pads (C-4 connections).

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Method And Device for High-Speed, High-Resolution Detection Of Thin Organic Films On Metals

       This article describes a high-speed, non-destructive
technique for detecting the presence of thin organic films (residues
of substrate manufacturing processes) on metal interconnecting pads
(C-4 connections).

      Thin organic residues (primarily polyimides) that are very
difficult to detect visually cause wettability problems and poor
bonds on C-4 pads when joining chips to substrates. This detection
technique (device and method) is sensitive to film thickness, offers
a spatial resolution significantly better than the minimum feature
size, and lends itself to an automated, high-speed inspection
process.  An electron beam (E-beam) bombards a particular metal
surface and the E-beam primary voltage is set to cause secondary
electron emission from the polymer (film) but not from the
underlaying metal. Thus, the presence of an organic film is detected
by simply scanning the C-4 pads or other metallic surfaces of
interest with the E-beam.  The ratio of the number of secondary
electrons to primary electrons is called the yield w.  The yield from
an atomically clean surface is characteristic of the material
bombarded and varies with the primary beam energy, Ep (KV), as shown
in Fig. 1, which is a plot of w vs Ep. Fig. 2 illustrates
schematically the yield curves for bulk gold 1, bulk polyimide 2 and
a thin film of polyimide on gold 3. Thin film curves must become
identical to the bulk polyimide curve at a thickness corresponding to
the primary beam penetration depth, which varies as Ep 1.5 and is
approximately 500 Angstroms at 1 Kv in polyimide.

      Of particular importance to this test technique is an
understanding of how a material is charged by irradiation with a
primary beam of a given energy.  Referring to Fig. 1, if the S.E.
yield is unity, there...