Method for fabricating metal-oxide-metal capacitor

Patent metadata

Field	Value
Publication number	US-11527605-B2
Application number	US-202117364935-A
Country	US
Kind code	B2
Filing date	Jul 1, 2021
Priority date	May 13, 2021
Publication date	Dec 13, 2022
Grant date	Dec 13, 2022

Abstract

A method for fabricating a MOMCAP includes steps as follows: An Nth metal layer is formed on a substrate according to an Nth expected capacitance value of the Nth metal layer. An Nth capacitance error value between an Nth actual capacitance value of the Nth metal layer and the Nth expected capacitance value is calculated. An N+1th expected capacitance value of an N+1th metal layer is adjusted to form an N+1th actual capacitance value according to the Nth capacitance error value, and the N+1th metal layer with an N+1th actual capacitance value is formed on the Nth metal layer according to the adjusted N+1th expected capacitance value, to make the sum of the Nth actual capacitance value and the N+1th actual capacitance value equal to the sum of the Nth expected capacitance value and the N+1th expected capacitance value. N is an integer greater than 1.

First claim

What is claimed is:

1. A method for fabricating a metal-oxide-metal capacitor (MOMCAP), comprising:
calculating an Nth expected pitch of an Nth metal layer and an N+1th expected pitch of an N+1th metal layer according to an Nth expected capacitance value of the Nth metal layer;
forming the Nth metal layer on a substrate according to the Nth expected capacitance value and the Nth expected pitch of the Nth metal layer;
calculating an Nth capacitance error value between an Nth actual capacitance value of the Nth metal layer and the Nth expected capacitance value;
calculating an Nth pitch error value between an actual pitch of the Nth metal layer and the Nth expected pitch;
adjusting an N+1th expected capacitance value of an N+1th metal layer to form an adjusted N+1th expected capacitance value according to the Nth capacitance error value;
adjusting the N+1th expected pitch according to the Nth pitch error value; and
forming the N+1th metal layer with an N+1th actual capacitance value on the Nth metal layer according to the adjusted N+1th expected capacitance value, so as to make a sum of the Nth actual capacitance value and the N+1th actual capacitance value substantially equal to a sum of the Nth expected capacitance value and the N+1th expected capacitance value, wherein, N is an integer greater than 1.

2. A method for fabricating a MOMCAP, comprising:
calculating an Nth expected thickness of an Nth metal layer and an N+1th expected thickness of an N+1th metal layer according to an Nth expected capacitance value of the Nth metal layer;
forming the Nth metal layer on the substrate according to the Nth expected thickness;
calculating an Nth thickness error value between an Nth actual thickness of the Nth metal layer and the Nth expected thickness;
adjusting the N+1th expected thickness according to the Nth thickness error value to form an adjusted N+1th expected thickness of the N+1th metal layer; and
forming the N+1th metal layer with an N+1th actual thickness on the Nth metal layer according to the adjusted N+1th expected thickness, so as to make a sum of the Nth actual thickness and the N+1th actual thickness substantially equal to a sum of the Nth expected thickness and the N+1th expected thickness;
wherein the step of forming the N+1th metal layer with the N+1th actual thickness comprises:
forming a dielectric layer on the Nth metal layer;
forming at least one recess in the dielectric layer;
forming a metal material layer on the dielectric layer and filling the at least one recess;
performing a planarization process with a removal thickness that is equal to a target removal thickness of the planarization process subtracting the Nth thickness error value; and
adjusting a thickness of the dielectric layer according to the Nth thickness error value and the target removal thickness.

3. The method according to claim 2 , wherein the N+1th metal layer has a N+1th thickness error tolerance value; when the Nth thickness error value is greater than the N+1th thickness error tolerance value, the N+1th expected thickness is adjusted only based on the N+1th thickness error tolerance value.

4. The method according to claim 3 , further comprising:
adjusting N+2th expected thickness of an N+2th metal layer to form an adjusted N+2th expected thickness according to a summed-up error value which is formed by summing up a N+1th thickness error value of the N+1th metal layer and a difference between the Nth thickness error value and the N+1th thickness error tolerance value; and
forming the N+2th metal layer on the N+1th metal layer according to the adjusted N+2th expected thickness.

5. The method according to claim 2 , wherein the MOMCAP has a plurality of metal layers and an overall expected thickness which is a sum of a plurality of expected thicknesses of the plurality of metal layers; and the overall expected thickness is proportional to an overall capacitance value of the MOMCAP.

6. The method according to claim 5 , wherein a cumulative thickness and a cumulative capacitance value measured between the substrate and the Nth metal layer form a linear relationship function; and a slope of the linear relationship function is identical to a ratio of the overall expected thickness to the overall capacitance value.

7. A method for fabricating a MOMCAP, comprising:
calculating an Nth expected thickness of an Nth metal layer and an N+1th expected thickness of an N+1th metal layer according to an Nth expected capacitance value of the Nth metal layer;
forming the Nth metal layer on the substrate according to the Nth expected thickness;
calculating an Nth thickness error value between an Nth actual thickness of the Nth metal layer and the Nth expected thickness;
adjusting the N+1th expected thickness according to the Nth thickness error value to form an adjusted N+1th expected thickness of the N+1th metal layer; and
forming the N+1th metal layer with an N+1th actual thickness on the Nth metal layer according to the adjusted N+1th expected thickness, so as to make a sum of the Nth actual thickness and the N+1th actual thickness substantially equal to a sum of the Nth expected thickness and the N+1th expected thickness;
wherein the step of forming the N+1th metal layer with the N+1th actual thickness comprises:
forming a dielectric layer on the Nth metal layer;
forming at least one recess in the dielectric layer with a depth that is equal to a target depth of the at least one recess plus the Nth thickness error value;
forming a metal material layer on the dielectric layer and filling the at least one recess; and
performing a planarization process.

8. The method according to claim 7 , further comprising adjusting a thickness of the dielectric layer according to the Nth thickness error value and the target depth.

Assignees

• United Microelectronics Corp

Inventors

• Chen Shu-Ji
• Feng Jing
• JIANG Xiao-hong
• Tey Ching-Hwa

Classifications

• H10P74/207

  Electrical properties, e.g. testing or measuring of resistance, deep levels or capacitance-voltage characteristics · CPC title

• H10P74/203

  Structural properties, e.g. testing or measuring thicknesses, line widths, warpage, bond strengths or physical defects · CPC title

• H10P74/23

  characterised by multiple measurements, corrections, marking or sorting processes · CPC title

• H10P72/0604

  Process monitoring, e.g. flow or thickness monitoring · CPC title

• H10W20/496 Primary

  Capacitor integral with wiring layers · CPC title