Subject: NEW THREAD Name: HSQ Development notes ... WAS Re: cautions on HSQ develop and different dose factors for different patterns
From: "James W. Conway" <>
Date: Wed, 10 Aug 2011 16:50:25 -0700
Wed, 10 Aug 2011 16:50:25 -0700
Zhiping Zhang and Raith Users working with HSQ negative tone resist:

This email follows up on several recent email and office hour discussion threads on developing HSQ for sub-100 nm features and small dense lines in Photonic crystals, dense gratings and tapered coupler patterns.
Please see me during my office hours if you desire more detail information..

There are a lot of different methods reported and I am testing several for high resolution and RIE pattern transfer process capability for publication at EIPBN and SPIE Advance Lithography next year.  Your comments and favorite recipe and dose reports are welcomed.

Several Attachments:

- The entire SNF XR-1541 process sheet is attached.  This is an adaptation from the SNF FoX-12 HSQ material we used last year.  This process is soon to be released to all lab members and is being tested with selected Users working in the Ebeam Lab.

- also attached the HSQ process web page from TU Delft whose process sequence is widely cited in the literature.

- A Paper by Henschel et. al on high resolution development techniques compared different baking and development methods.  This has interesting note that relates to development chemistries that present a theory that the HSQ scums over with Siloxane bonds which then limits the development to about 1 minute time when the development effect crawls to a halt. They then is a dilute 1:4000 HF : water dip to remove that layer and then redevelop in 25% TMAH solution.  Reduces footing and scumming was reported.

- Finally MIT's Joel Yang's and Karl Berggren's classic paper on Salty development using NaOH and NaCl to improve contrast.  This is great work but I have not had the time to pursue it here at SNF -- Yet...
I am in frequent contact with both people, Joel is now in Singapore

In very general terms:
 HSQ is a challenging resist to process in that is not consistent in its batch to batch exposure dose response -- is of poor sensitivity and with normally low to medium contrast. It should be exposed within a week of spinning as it displays significant aging effects over time.  ON the good side it is very high resolution EBL resist material and is 'tough as nails' in normal RIE and liquid etchants.

Would other users like to report their preferred development methods to this list?

Thank you for your excellent questions and support towards further developing the Ebeam Processing Suite in the Ebeam Lab.

James Conway


Development option one:

My process as well as what many people use at other universities worldwide:

  1. Develop in MF319 developer for 60 seconds with very gentle agitation.  If you have applied ESPACER be sure to rinse off the material with water before developing.  I run it through the Semitool SRD.
  1. STOP1: Rinse in 1:9 MF319:H2O mixture for 15 seconds.
    This is a descum processing step that reduces footing or scumming between adjacent lines.
  1. STOP2: Rinse in DI water for 30 seconds.

  2. Dump rinse X 6 cycles in DI water on wbmiscresist.

  3. Spin Rinse Dry in Semitool SRD.

  4. HPV (High Power Visual microscope inspection at 100 and 1000X polarized lighting.  Are area and lines present?  Is there significant scumming between lines? Are there Pinholes present or adhesion issues?

  5. SEM Inspection (optional once you have an established HSQ process and determined your doses for your test pattern CD's).
    Check for Critical Dimensions to match the Pattern CD specifications. Look for pinholes, post, and resist artifacts or scum residues.

On 8/10/2011 12:05 PM, Zhiping Zhang wrote:

Hi James,


As per our discussion just now, when we develop HSQ, to reduce the number of particles, we should use our own clean glassware and dump rinse in the end. Is there any more we should also practice?


Will you have time tomorrow to work with me on getting different dose factors for different cells in my GDS? We can use the PC on the table facing the SEM.





TU Delft - FOx-12 Recipe
Overview full-color pictureTU Delft logo

Kavli Institute of Nanoscience Delft

Content | Menu | Taal/Language | Banners | Portal Menu

U maakt gebruik van een browser die de gebruikte web-standaarden niet of onvolledig ondersteunt.
Hierdoor kunnen afwijkingen in de lay-out ontstaan.

Afhankelijk van uw platform is deze site het beste te bekijken met een recente versie van Microsoft Internet Explorer (5 of hoger), Netscape (6 of hoger), Mozilla (1 of hoger) en Opera (6 of hoger).

Home | Sitemap |

FOx-12 Recipe


Spincoat with HMDS and bake at 200 0C on hotplate for 2 minutes

 special Superior edge roughness
 reference J. Vac. Sci. Technol. B. 16 (1998) 69 - 76

 202 - 128 nm @ 1000 - 6000 rpm

 thinner layers can be obtained at Karl Suss spinner WITH lid closed:
 61 - 44 nm @ 1000 - 2000 rpm or by diluting with MIBK (see spincurve below)


 2 minutes on hotplate @ 150 ░C

 2 minutes on hotplate @ 220 ░C


 about 500 ÁC/cm2 @ 100 kV


 1 min. in MF322

 15 s in MF322 : demi-water = 1 : 9

 15 s in demi-water




Work done in FOx-12

15 nm single pass line in 44nm thick FOx-12, exposed with spot 21nm_740pa_400um_100kv and dose 5460 ÁC/cm2 @ 100kV

20nm dense lines and spaces (single pass lines) in 50nm thick FOx-12, exposed with spot 21nm_740pa_400um_100kv and dose 6970 ÁC/cm2 @ 100kV

Etchrates of FOx-12 resist in different plasmas in comparison with HPR504

GAS 1: SF6/He plasma on Leybold Z401:

 SF6 12.5 sccm
 He 10 sscm
 Pressure 10 Ábar
 P 40 W
 Biaspotential -60 V

GAS 2: CHF3/O2 plasma on Leybold Z401S

 CHF3 50 sccm
 O2 2.5 sccm
 Pressure 7 Ábar
 P 50 W
 Biaspotential -350 V

GAS 3: O2 plasma on Leybold Z401S

 O2 20 sccm
 Pressure 3 Ábar
 P 20 W
 Biaspotential -200 V

Etch rates (nm/min):

  Gas 1 Gas 2 Gas 3
 HPR504 51 16 29
 FOx-12 29 36 0.6


FOx-12/HPR bilayer resist combination

HSQ can also be used in a bilayer combination with hard baked HPR for high aspect ratio nanoscale e-beam lithography.

For this reason a few steps should be added to the processing flowchart:

  • After spincoating and baking HMDS and before following the FOx-12 flowchart HPR504 should be spincoated (5000rpm 937nm) and baked on a hotplate at successive 100, 200 and 250 0C, each temperature for 2 minutes.
  • After development of the FOx-12 resist the HPR is etched in above mentioned O2 plasma (gas 3).

A few examples of work done in a bilayer resist of FOx-12 and HPR

80nm multi pass lines (10 passes) written in 44nm thick FOx-12 on top of 937nm thick HPR, exposed with spot 21nm_740pa_400um_100kv and dose 290 ÁC/cm2 @ 100kV

80nm multi pass lines (10 passes) written in 44nm thick FOx-12 on top of 635nm thick HPR, exposed with spot 21nm_740pa_400um_100kv and dose 990 ÁC/cm2 @ 100kV

26nm semi-dense lines written in 30nm thick FOx-12 on top of 155nm thick HPR, exposed with spot 21nm_740pa_400um_100kv and dose 815 ÁC/cm2 @ 100kV

40nm dense lines and spaces (1:1) written in 30nm thick FOx-12 on top of 129nm thick HPR, exposed with spot 21nm_740pa_400um_100kv and dose 1646 ÁC/cm2 @ 100kV

This Site Tracked by

["application/msword" not shown]
["application/pdf" not shown]
["application/pdf" not shown]