Preparation of Mock tile MCPs 01042011

Preparation of Mock tile MCPs

01042011

Anil Mane, Qing Peng, Jeffrey ElamArgonne National Laboratory

Objective:

2

•To prepare 8 workable MCPs for mock tile assembly•Collect within batch MCPs resistance data for resistive coating ALD process •Collect batch-to-batch MCPs resistance data for resistive coating ALD process

Experimental:

• Used 2 batch of 5 MCPs with NiCr electrode• Passivation (53A)• Resistive coating chemistry-2 (~800A)• SEE coating (53A)

3

NiCr deposition at Fermi lab:

NiCr electrode on 1st batch of 5 MCPs

MCP# 125 126 127 128 129

•One of the MCP holder has electrode exposure dimension issue

NiCr electrode on 2nd batch of 5 MCPs

MCP# 130 131 132 133 134

4

MCPs placement prior to ALD in substrate loading tray

MCP with NiCr electrode Before ALD

After ALD

MCP# 125 126 127 128 129

•Uniform deposition on monitors (quartz and Si(100)) as well as on all MCPs

5

ALD coating thickness across the reactor

•Thickness uniformity on monitor Si(100) <2%• The resistive layer thickness ~800A• Similar thickness trend observed on second batch of 5 MCPs Excellent batch-to-batch reproducibility

6

I-V(-10V to +10V) Response in air for 10 MCPs

•Linear I-V response for all MCPs•MCP 131 resistance is out of targeted value (Outlier)•Little scatter in I-V plot

Removed MCP 131 data

7

I-V(-100V to +100V) Response in vacuum (4e-3mbar) for 10 MCPs

•Linear I-V response for all MCPs•MCP 131 resistance is out of targeted value•Little scatter in I-V plot

Removed MCP 131 data

8

I-V Comparison (air vs. vacuum) 9 MCPs

•Linear I-V response for all MCPs•Very little change in I-V values•Little scatter in I-V plot

•Electrical contact cause by electrode underneath?•Related to end spoiling ?•ALD chemistry composition across the reactor? •Electrode area ?

I-V in Air I-V in Vacuum (4e-3mbar)

-10V to +10V -100V to +100V

9

Resistance Comparison for 9 MCPs (air vs. vacuum)

•Very little change in average resistance ( air 111 M vs. vacuum 115 M) •Average resistance in vacuum = 115 ±12 M ~10% resistance variation

10

Resistance for 10 MCPs

What's wrong with MCP# 131?

11

Resistance for 10 MCPs

Cause for outlier (MCP131): •

12

Resistance for 10 MCPs

Cause for outlier (MCP131): •

•Gap in triple points can cause electrode penetration and cause localize low resistance regimes

Will affect greatly on 8”x8” MCPNeed minimum(?) defects on MCPs

Summary

13

Prepared 10 MCPs with ALD resistive layer chemistry-2 and SEE (Al2O3) layer

•Excellent resistive layer uniformity across the ALD reactor

•Within batch good ALD layer reproducibility

•Very good batch-to-batch reproducible of ALD process

All MCPs shows linear I-V response

•Very little change in resistance of MCPs (Air vs. Vacuum)

•Average resistance for MCP = 115M

Big gap @ triple point are responsible for outliers

NiCr electrode deposition need same dimension MCP holders

•Will vary the total # of active pores (resistances)

Next plan for Mock tile:

14

Selected 8 MCPs Gain test in APS test set-up

George and grid (A, B, & C) spacer resistance tuning

Resistance test on stack of [MCPs, George and & spacer (A, B, & C ) on Mock tile

Gain test at UCB (Prof. Ossy’s Lab)