# US 2015/0106780 A1 · Semiconductor device reliability model and methodologies for use thereof

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2015/0106780 A1
- Document ID: US20150106780A1
- Status: Published application
- Kind: A1
- Application number: 14/551,693
- Filed: 2014-11-24
- Published: 2015-04-16
- Earliest claimed: 2013-06-20
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Jeanne P. BICKFORD; Nazmul HABIB; Baozhen LI; Pascal A. NSAME
- Web page: https://patentgrep.com/patent/US20150106780A1

## CPC

- G06F 30/398: Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM] (optical proximity correction [OPC] design processes G03F1/36)
- G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30)
- G06F 2111/08: Probabilistic or stochastic CAD
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Systems and methods for semiconductor device reliability qualification during semiconductor device design. A method is provided that includes defining performance process window bins for a performance window. The method further includes determining at least one failure mechanism for each bin assignment. The method further includes generating different reliability models when the at least one failure mechanism is a function of the process window, and generating common reliability models when the at least one failure mechanism is not the function of the process window. The method further includes identifying at least one risk factor for each bin assignment, and generating aggregate models using a manufacturing line distribution. The method further includes determining a fail rate by bin and optimizing a line center to minimize product fail rate. The method further includes determining a fail rate by bin and scrapping production as a function of a manufacturing line excursion event.

## Claims

### Claim 1

A method comprising:
defining performance process window bins for a performance window;
assigning a percentage of manufacturing line distribution to each of the performance process window bins;
determining at least one failure mechanism for each bin assignment;
determining whether the at least one failure mechanism determined for each bin assignment is a function of the process window;
when the at least one failure mechanism determined for each bin assignment is the function of the process window, generating different reliability models for the at least one failure mechanism for each bin assignment;
when the at least one failure mechanism determined for each bin assignment is not the function of the process window, generating common reliability models for the at least one failure mechanism for each bin assignment;
identifying at least one risk factor for each bin assignment; and
generating aggregate models using the manufacturing line distribution,
wherein at least the step of generating the aggregate models is performed using a processor.

### Claim 2 (depends on 1)

The method of claim 1, further comprising obtaining input data comprising at least one of product data, manufacturing data, and physics of failure models, wherein the input data is used to determine the at least one failure mechanism for each bin assignment.

### Claim 3 (depends on 1)

The method of claim 1, wherein the performance process window bins are assigned based on performance characteristics of a semiconductor product, and the percentage of the manufacturing line distribution assigned to each of the performance process window bins comprise a subset of the semiconductor product.

### Claim 4 (depends on 3)

The method of claim 3, wherein the determining the at least one failure mechanism for each bin assignment comprises examining electrical characteristics of the semiconductor product, and the at least one failure mechanism is selected from at least one of time-dependent dielectric breakdown (TDDB), hot carrier injection (HCl), biased temperature instability (BTI), soft error rate (SER), retention disturbance, electromigration (EM), stress migration (SM), and TDDB between metal lines.

### Claim 5 (depends on 4)

The method of claim 4, wherein the determining whether the at least one failure mechanism is the function of the process window comprises plotting each failure mechanism as the function of the process window bin.

### Claim 6 (depends on 5)

The method of claim 5, wherein the at least one risk factor is a factor that will cause a first semiconductor device of the subset of semiconductor product assigned to the corresponding performance process window bin to fail first.

### Claim 7 (depends on 1)

The method of claim 1, wherein the generating the aggregate models comprise aggregating fails for each of the performance process window bins using the formula
F
=
1
-
∏
i


(
1
-
F
i
)
,
where i=the at least one failure mechanism, and Fi=is a function of the at least one failure mechanism.

### Claim 8 (depends on 1)

The method of claim 1, wherein the performance process window bins are defined based on operating or performance characteristics of semiconductor devices, the assigning comprises assigning a percentage of the semiconductor devices representative of a subset of the manufacturing line distribution to each of the performance process window bins.

### Claim 9 (depends on 8)

The method of claim 8, wherein the at least one risk factor is a factor that will cause a first semiconductor device of the subset of the manufacturing line distribution assigned to a performance process window bin to fail first.

### Claim 10 (depends on 1)

The method of claim 1, further comprising:
selecting a line center within a process window of at least one aggregate model of the aggregate models;
assigning a percentage of the manufacturing line distribution to each performance process window bin within the process window;
calculating a product fail rate for each of the performance process window bins within the process window; and
optimizing the line center to minimize the product fail rate for each of the performance process window bins within the process window.

### Claim 11 (depends on 10)

The method of claim 10, further comprising calculating a product reliability fail rate for a semiconductor product manufactured at the selected line center.

### Claim 12 (depends on 11)

The method of claim 11, wherein the optimizing the line center comprises:
comparing the product reliability fail rate to a manufacturing line centering fail rate;
determining whether the product reliability fail rate meets product reliability requirements;
when the product reliability fail rate does not meet product reliability requirements, reevaluating the at least one failure mechanism and selecting at least one new failure mechanism;
identifying a process point within the process window that has a lower fail rate for the new reliability failure mechanism; and
repeating the steps of selecting the line center, assigning the percentage of the manufacturing line distribution, and calculating the product fail rate using a newly selected line center and the at least one new failure mechanism.
