Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

ISOLET Kerf Structure and Math Model for Improved Sensitivity to Bipolar Leakage Components

IP.com Disclosure Number: IPCOM000047580D
Original Publication Date: 1983-Dec-01
Included in the Prior Art Database: 2005-Feb-07
Document File: 3 page(s) / 76K

Publishing Venue

IBM

Related People

Chakravarti, SN: AUTHOR [+2]

Abstract

In the construction of bipolar silicon integrated circuits of small device dimensions, it is well known that cumulative stresses, induced by dopant diffusions, silicon dioxide isolation structures, silicon nitride layers used for masking or selective oxidation barriers, or passivation, etc., can result in microscopic substrate defects, such as dislocations, which can produce emitter-to-collector "pipes", and other forms of defect leakage. Process quality and enhancements can be studied via finished-product functionality, via test wafers, via test sites at limited positions of product wafers, and via kerf structures.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 44% of the total text.

Page 1 of 3

ISOLET Kerf Structure and Math Model for Improved Sensitivity to Bipolar Leakage Components

In the construction of bipolar silicon integrated circuits of small device dimensions, it is well known that cumulative stresses, induced by dopant diffusions, silicon dioxide isolation structures, silicon nitride layers used for masking or selective oxidation barriers, or passivation, etc., can result in microscopic substrate defects, such as dislocations, which can produce emitter- to-collector "pipes", and other forms of defect leakage. Process quality and enhancements can be studied via finished-product functionality, via test wafers, via test sites at limited positions of product wafers, and via kerf structures. For the fundamental studies of process interactions with silicon wafers of different origins and/or histories, different degrees of lattice perfection, different bulk concentrations and/or precipitation characteristics of impurities (such as oxygen or metallic contaminants), it is very helpful to use kerf structures (rather than the other named alternatives) which can be tested before product metallization, and which offer a large sample size, well distributed over a given wafer. One drawback of leakage characterization via bipolar kerf structures is the difficulty in making these "macroscopic" devices sensitive to the proximity and stress effects operating in the microscopic product devices. Another difficulty is that of separating the various contributors to defect leakage, as perimeter, area, isolation scheme, photolithographic control, as well as the silicon bulk phenomena. Prior art in bipolar kerf structures has tended toward either (a) simple geometry devices with total critical area comparable to that of all devices in the product chip, or (b) complex geometry structures which are inseparably sensitive to process and materials interactions, photo-alignment problems, photo defect-induced product defects, etc. Either of these limits of kerf leakage characterization suffer further loss of discrimination, when used with simple "pass/fail" test analysis. Because of the listed shortcomings of the prior art, and because of the continuing enhancements and innovations in LSI bipolar isolation technologies, a compromise structure is described which offers the following advantages: (a) Relaxes state-of-the-art photolithographic groundrules (dimensions, alignments, run-out, small photo defects, etc.); (b) Separates isolation and non-isolation perimeter effects on similar emitter structures; (c) Usable in kerf (as well as test site), with inherent freedoms in on-wafer sample size and locations (i.e., compared to test sites, or specially metallized product chips); (d) Suitable for automatic testing, before metallization; (e) Intended for computer analysis, with more discrimination than "pass/fail-tracking" to produce (e.g., by scaling of critical dimensions); (f) Merges devices and simplifies probing to save kerf area; and (g) Can be...