Calibration kits for RF passive devices

Patent metadata

Field	Value
Publication number	US-9425112-B2
Application number	US-201213491364-A
Country	US
Kind code	B2
Filing date	Jun 7, 2012
Priority date	Jun 7, 2012
Publication date	Aug 23, 2016
Grant date	Aug 23, 2016

Abstract

A method includes measuring a first calibration kit in a wafer to obtain a first performance data. The wafer includes a substrate, and a plurality of dielectric layers over the substrate. The first calibration kit includes a first passive device over the plurality of dielectric layers, wherein substantially no metal feature is disposed in the plurality of dielectric layers and overlapped by the first passive device. The method further includes measuring a second calibration kit in the wafer to obtain a second performance data. The second calibration kit includes a second passive device identical to the first device and over the plurality of dielectric layers, and dummy patterns in the plurality of dielectric layers and overlapped by the second passive device. The first performance data and the second performance data are de-embedded to determine an effect of metal patterns in the plurality of dielectric layers to overlying passive devices.

First claim

What is claimed is:

1. A method comprising:
measuring a first calibration kit in a chip to obtain a first performance data, wherein the chip comprises a substrate, and a plurality of dielectric layers over the substrate, and wherein the first calibration kit comprises:
a first passive device over the plurality of dielectric layers, wherein substantially no metal feature is disposed in the plurality of dielectric layers and overlapped by the first passive device;

measuring a second calibration kit in the chip to obtain a second performance data, wherein each of the measuring the first calibration kit and the measuring the second calibration kit is performed using a three-step de-embedding method, and wherein the second calibration kit comprises:
a second passive device over the plurality of dielectric layers, wherein the second passive device is identical to the first passive device; and
metal features comprising first dummy patterns in the plurality of dielectric layers and overlapped by the second passive device, and wherein all metal features that are overlapped by the first passive device in combination are different from all metal features that are overlapped by the second passive device in combination; and

de-embedding the first performance data and the second performance data to determine an effect of the metal features on the second passive device.

2. The method of claim 1 , wherein the first and the second passive devices comprise portions in a layer selected from the group consisting of a metal pad layer, a Post-Passivation Interconnect (PPI) layer, an Under-Bump Metallurgy (UBM) layer, and combinations thereof.

3. The method of claim 1 further comprising:
measuring a third calibration kit in a chip to obtain a third performance data, wherein the third calibration kit comprises:
a third passive device over the plurality of dielectric layers, wherein the third passive device is identical to the first passive device; and
second dummy patterns in the plurality of dielectric layers and overlapped by the third passive device, wherein one of the dielectric layers comprises a portion of the first dummy patterns formed therein, and does not comprise any portion of the second dummy patterns therein, and wherein the step of de-embedding comprises de-embedding the first performance data, the second performance data, and the third performance data to determine the effect.

4. The method of claim 1 further comprising:
measuring a third calibration kit in the chip to obtain a third performance data, wherein the third calibration kit comprises:
a third passive device over the plurality of dielectric layers, wherein the third passive device is identical to the first passive device; and
second dummy patterns in the plurality of dielectric layers and overlapped by the third passive device, wherein the first and the second dummy patterns have different pattern densities, and wherein the step of de-embedding comprises de-embedding the first performance data, the second performance data, and the third performance data to determine the effect.

5. The method of claim 1 further comprising:
simulating a third and a fourth performance data of the first and the second passive devices, respectively, from a model;
comparing the third and the fourth performance data with the first and the second performance data, respectively; and
updating the model based on a result obtained from the step of comparing.

6. The method of claim 1 , wherein the first dummy patterns are electrically floating.

7. The method of claim 1 , wherein conductive features directly underlying the first passive device have a layout different from a layout of additional conductive features directly underlying the second passive device.

8. A method comprising:
designing a plurality of calibration kits from a design specification, wherein the design specification comprises:
a first specification of a passive device;
a second specification for routing metal lines in dielectric layers that are under the passive device, wherein the routing metal lines are overlapped by the passive device; and
a third specification of pattern densities of the routing metal lines;

manufacturing a chip comprising the plurality of calibration kits, wherein each of the plurality of calibration kits comprises the passive device, wherein the passive devices in the plurality of calibration kits are identical to each other, and wherein all routing metal lines in combination in each of the plurality of calibration kits and overlapped by the respective one of the plurality of calibration kits are different from all routing metal lines in combination in any other one of the plurality of calibration kits; and
measuring the plurality of calibration kits to generate a first plurality of performance data, wherein the measuring the plurality of calibration kits is performed using a three-step de-embedding method, and the three-step de-embedding method comprises:
measuring a short device having first test pads shorted with each other;
measuring an open device having second test pads disconnected from each other, wherein the second test pads are identical to the first test pads; and
measuring a through-device comprising third test pads and the passive device in a respective one of the plurality of calibration kits connected between the third test pads, wherein the third test pads are identical to the first test pads.

9. The method of claim 8 further comprising simulating a first plurality of performance data of the plurality of calibration kits from a model.

10. The method of claim 9 further comprising:
measuring the plurality of calibration kits to generate a second plurality of performance data;
comparing the first and the second plurality of performance data; and
updating the model based on a result obtained from the step of comparing.

11. The method of claim 10 , wherein the step of generating the second plurality of performance data comprises de-embedding results measured from the plurality of calibration kits.

12. The method of claim 8 comprising:
designing a first one of the plurality of calibration kits having first pattern densities of routing metal lines same as the pattern densities in the third specification; and
designing a second one of the plurality of calibration kits having second pattern densities of routing metal lines different from the pattern densities in the third specification.

13. The method of claim 8 comprising:
designing a first one of the plurality of calibration kits, wherein a metal routing in the first one of the plurality of calibration kits is placed in first dielectric layers same as the dielectric layers that are specified in the second specification; and
designing a second one of the plurality of calibration kits, wherein a metal routing in the second one of the plurality of calibration kits is placed in second dielectric layers different from the dielectric layers that are specified in the second specification.

14. A device comprising:
a chip comprising:
a substrate; and
a plurality of dielectric layers over the substrate;
a first calibration kit in the chip, wherein the first calibration kit comprises:
a first passive device over the plurality of dielectric layers, wherein substantially no metal feature is disposed in the plurality of dielectric layers and overlapped by the first passive device; and
a second calibration kit in the chip, wherein the second calibration kit comprises:
a second passive device over the plurality of dielectric layers, wherein the second passive device is identical to the first passive device, wherein eac

Assignees

• Chen Jie
• Tsai Hao-Yi
• Chen Hsien-Wei
• Kuo Hung-Yi
• Taiwan Semiconductor Mfg Co Ltd

Inventors

• Chen Jie
• Tsai Hao-Yi
• Chen Hsien-Wei
• Kuo Hung-Yi

Classifications

• H10W72/252

  comprising solid metals or solid metalloids, e.g. PbSn, Ag or Cu · CPC title

• G01R1/07328

  for testing printed circuit boards · CPC title

• G06F30/20

  Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) · CPC title

• G01R35/005

  Calibrating; Standards or reference devices, e.g. voltage or resistance standards, "golden" references (G01R33/0035, G01R35/002 take precedence) · CPC title

• H10P74/277

  Circuits for electrically characterising or monitoring manufacturing processes, e.g. circuits in tested chips or circuits in testing wafers · CPC title