Browse Prior Art Database

Contact Opening in Shallow Junction Transistors

IP.com Disclosure Number: IPCOM000089961D
Original Publication Date: 1968-Dec-01
Included in the Prior Art Database: 2005-Mar-05
Document File: 2 page(s) / 59K

Publishing Venue

IBM

Related People

Anantha, NG: AUTHOR

Abstract

This process is for forming openings in a passivating layer without undercutting. In high-speed junction transistors, particularly those with very shallow emitter junctions, it is difficult to open contact windows without undercutting the oxide over the junction, thus exposing the junction. The reason for the undercutting is that, during the phosphorus emitter diffusion, a layer of P(2)O(5) glass forms on the mask sidewalls of the emitter window. This glass etches very fast in buffered HF, normally used to reopen the opening for an emitter contact. The problem becomes very significant in small geometry devices where the diffusion window and the contact opening are the same dimensions.

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 91% of the total text.

Page 1 of 2

Contact Opening in Shallow Junction Transistors

This process is for forming openings in a passivating layer without undercutting. In high-speed junction transistors, particularly those with very shallow emitter junctions, it is difficult to open contact windows without undercutting the oxide over the junction, thus exposing the junction. The reason for the undercutting is that, during the phosphorus emitter diffusion, a layer of P(2)O(5) glass forms on the mask sidewalls of the emitter window. This glass etches very fast in buffered HF, normally used to reopen the opening for an emitter contact. The problem becomes very significant in small geometry devices where the diffusion window and the contact opening are the same dimensions.

In this method, emitter diffusion window 10 is formed in thermal oxide layer 12 over base region 14 as in A. A layer of silicon nitride 16 approximately 300 to 500 angstroms thick is deposited over the wafer either by pyrolytic decomposition or by sputtering. Layer 18 of SiO(2) is then deposited over layer 16 as in B. Emitter diffusion window 20 is then formed in layer 16 and 18 by etching layers 12 and 14 using HF and H(3)PO(4) acid as in C. The emitter diffusion is then carried out as indicated in D and the contact window opened using buffered HF.

During emitter diffusion, layer 16 protects the sidewall of the opening since it serves as a barrier to phosphorous and is not easily etched in HF. Thus the undercutting problem which could...