Erbium etching in Lampoly and P5000

Yiyang Gong yiyangg at
Thu Aug 7 23:20:52 PDT 2008

To whom it may concern,

I would like to etch a Silicon nitride film doped with Erbium (far less 
than 1% by atomic concentration) in the silicon chamber (chamber C) of 
P5000 and Lampoly, or any etchers that has 1:1 selectivty of SiN to 
ebeam resists. The required information from the SNF website is as follows:

1. Name: Yiyang Gong
Phone: 650-723-2279
Coral: yiyang
email: yiyangg at
PI: Jelena Vuckovic

2. Raw erbium MSDS is attached (erbium.htm, from The structure of the 
wafer is 100nm of SiN doped with erbium on top of 200nm of Si, on top of 
thick silicon dioxide, all on a silicon substrate. Another structure is 
a lattice of SiN doped with Er in Si. Because erbium is a dopant in 
Silicon nitride, it's stability should be much more than that of raw erbium.

3. Material is grown at Boston University:
Appl. Phys. Lett. *92*, 181105 (2008)

4. I would like to apply all of the e-beam lithography and know 
fabrication parameters for silicon nitride and Si to this new material 
in order to make photonic crystals or other photonic devices. The 
concentration of erbium in the nitride is very low (less than 1% atomic) 
and etching should proceed as usual. This procedure have been used in 
other plasma etchers, for example in " Hak-Seung Han, S-Y Seo, Jung H. 
Shin, and Kim DS  J. Appl. Phys. 88  2160 (2000)" (attached as 
JAP882160.pdf). In fact, in that paper, they choose to etch SiO2 with a 
standard CF4 etch. Similarly, I believe that a stanford SiN  recipe 
(using HBr) could be used to etch the Er doped SiN without difficulty.

I also believe that Erbium will not sputter and coat wafers. We can 
consider the diffusion coefficient of Aluminum as a guide for Erbium. 
Extrapolating the data from J. Appl. Phys., Vol. 91, No. 9, 1 May 2002, 
5645 (attached as JAP915645.pdf), the diffusion of Aluminum in Silicon 
is 1*10^-19 cm^2/s at 600 degrees celsius. The same number for Erbium at 
the same temperature is quoted as less than 1*10^-17 cm^2/s by J. Appl. 
Phys., 75, 2609 (attached as JAP752609.pdf). Moreover, surface etch 
temperature of Silicon is generally under 200C, so diffusion should not 
be a problem for this process.

I would also like to open up the Si chamber of P5000 to metals. I don't 
know what the fabrication restrictions are for this machine, but it 
seems to be the only dedicated Silicon etcher in SNF used for processing 
ebeam written materials. It would be very helpful to be able to use that 
machine for fabricated small devices such as photonic crystals.

5. Process flow is from Raith to lithography to etcher to possibly 
wetbench GaAs (dedicated labware will be used).

6. I will bring in wafers or pieces of wafers.

7. The wafer is stable, so storage should not be a problem.

8. The wafer is non-hazardous, so minimal disposal requirements are needed.

If I could receive some feedback to this request (i.e. missing 
materials), and about the timeline for processing this request, it would 
be very helpful.

Thank you,
Yiyang Gong
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: JAP882160.pdf
Type: application/pdf
Size: 159459 bytes
Desc: not available
URL: <>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: JAP752609.pdf
Type: application/pdf
Size: 1338914 bytes
Desc: not available
URL: <>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: JAP915645.pdf
Type: application/pdf
Size: 88778 bytes
Desc: not available
URL: <>

More information about the specmat mailing list