Specific Process Knowledge/Lithography/CSAR

Simple e-beam pattern in this resist has been tested, the results showed on this page. If you have questions to the process or wish to use this e-beam resist, please contact lithography at DTU Nanolab.

Spin Curves

The thickness is measured on VASE Ellipsometer using a simple Cauchy model for a transparent polymer on Si. The measurements are performed at one incidence angle (70 degrees) only. 9 points on each 4" wafer has been measured; the standard deviation thus representing the homogeinity of the film on the 4" wafers.

Around 2 ml of resist per wafer has been used when fabricating these curves. If you use less than 2 ml, the thickness of the final resist might be smaller than reported here.

AllResist AR-P 6200.09 (> 2ml per 4" wafer) spinning on Spin Coater: Manual LabSpin A-5, TIGRE, 09-04-2014. Softbake 5 min @ 150 degC.
Spin Speed [rpm]	Acceleration [1/s2]	Thickness [nm]
2000	4000	226
3000	4000	194
4000	4000	170
5000	4000	151
6000	4000	142
7000	4000	127

AllResist CSAR 6200.09 1:1 in anisole (< 2ml per 4" wafer), Spin Coater: Manual LabSpin A-5, TIGRE, 16-06-2014. Softbake 2 min @ 150 degC.
Spin Speed [rpm]	Acceleration [1/s2]	Thickness [nm]
2000	4000	84
3000	4000	67
4000	4000	59
5000	4000	53
6000	4000	49

AllResist CSAR 6200.18 (< 2ml per 4" wafer), Spin Coater: Manual Standard Resists, E-5, TIGRE, 15-06-2016. Softbake 2 min @ 180 degC.
Spin Speed [rpm]	Acceleration [1/s2]	Thickness [nm]
2000	2000	1003
3000	2000	809
4000	2000	721
5000	2000	639
6000	2000	586
7000	2000	549

Contrast Curve

CSAR 6200.09

100 nm lines in both ~70 nm and ~188 nm thick CSAR has been developed with AR-600-546 (standard CSAR developer) at room temperature to provide the following contrast curves.

CSAR Contrast Curve, Processed by TIGRE, FEB-MARCH 2016
Resist	Spin Coat	E-beam exposure	Development	Characterisation
CSAR AR-P6200.09 AllResist, CSAR AR-P6200.09 diluted 1:1 in Anisole	08-02-2016, LabSpin E-5, 4000 rpm, 60s, softbaked 60s @ 205 degC	09-02-2016, JBX9500 E-2, 2nA aperture 5, doses 40-600 µC/cm2, 100 nm lines and 300 nm spaces	11-02-2016, Fumehood D-2, AR-600-546, rinsed in IPA 60s.	02-03-2016 AFM Icon, F-2, ScanAsyst in Air

AR-P 6200 contrast curves.

Dose to size

Small features need a comparatively higher dose then big features and hence it can be useful to map out the dose and size dependency. Below is a set of cross sectional images of 100, 50 and 20 nm lines written 500, 250 and 180 nm resist at doses from 200 to 600 µC/cm².

Cross section SEM images of 500 nm AR-P 6200.09 exposed at 200-600 µC/cm2. Top image is 100 nm lines, center image is 50 nm lines, bottom image is 20 nm lines. Au coated for SEM imaging.

Cross section SEM images of 250 nm AR-P 6200.09 exposed at 200-600 µC/cm2. Top image is 100 nm lines, center image is 50 nm lines, bottom image is 20 nm lines. Au coated for SEM imaging.

Cross section SEM images of 180 nm AR-P 6200.09 exposed at 200-600 µC/cm2. Top image is 100 nm lines, center image is 50 nm lines, bottom image is 20 nm lines. Au coated for SEM imaging.

CSAR 6200.18

100 nm lines in ~900 nm thick CSAR has been developed with AR-600-546 (standard CSAR developer) at room temperature.

CSAR Contrast Curve, Processed by TIGRE, JUNE 2016
Resist	Spin Coat	E-beam exposure	Development	Characterisation
CSAR AR-P6200.18 AllResist	15-06-2016, LabSpin E-5, 2000 rpm, 60s, softbaked 60s @ 205 degC	15-06-2016, JBX9500 E-2, 2nA aperture 5, doses 40-600 µC/cm2, 100 nm lines and 300 nm spaces	16-06-2016, Fumehood E-4, AR-600-546, 30s/60s/90s, rinsed in IPA 60s.	JUNE/JULY 2016 SEM Supra 2, 10 keV

Dark Erosion

Dark erosion has been measured on a un-exposed 4" wafer spin coated with CSAR 6200.18 to a thickness of approximately 549 nm. The resist thickness has been measured by VASE Ellipsometer before development, and after 3 minutes, 13 minutes, and 30 minutes of development in AR 600 546.

The graphs shows the measured thicknesses; the errorbars represents the standard deviations from the ellipsometric measurements. The average etch rate of CSAR is ~0.1 nm/min.

Development

Many resists can be developed in different developers, CSAR can be developed in: AR 600-546, AR 600-548, ZED N-50 and mix of MIBK and IPA among others.

CSAR and ZEP520A are in principle the same chemical, however the pretreatment (filtration and temperature control) can differ.

Some users have reported residues and residual layers when using ZED N-50 on CSAR and vice verca, hence we recommend to use AR 600-546 or AR 600-548 (3 times stronger) to develop CSAR and not ZED N-50.

When this is said some users still observe residues when using AR 600-546, the producer "All resist GMBH" have recommended to use 3-5s, dip in pure MIBK to remove residues.

AR 600 546 will dissolve different plastic materials, hence never use it on PS compounds.

Etch Tests

If you have wafers or chips with CSAR you would like to have tested, please send me an [email].

Chlorine versus flourine-based etches

We have experienced problems with removal of CSAR after chlorine-based dry etch, see the file File:DryEtchTestsCSAR.pdf. It seems the chlorine etch forms particles of chlorinated CSAR on the surface, and these particles remains on the surface after resist removal with AR-600-71. The C4F8/SF6 etch also forms particles on the surface, but much smaller than those formed in the chlorine etch. It seems these particles are removed after 3 minutes in AR-600-71.

How to mount chips in dry etch tools

All etch rates presented here are measured on chips (i.e. diced 4" wafers) crystal bonded to a carrier. The carrier is either a blank Si wafer, a Si wafer spin coated with resist or a Si wafer coated with ALD grown Al2O3.

Etch Tests of CSAR, recipe 'nano1.42', DRIE PEGASUS, A-1. CSAR thickness measured on Ellipsometer VASE at 70 degrees

Sample	CSAR Etch rate nm/min
Full 4" Si wafer with non-patterned ~180 nm CSAR	~56.5 (based on 2 runs)
Full 4" Si wafer with non-patterned ~240 nm CSAR, postbaked 60 sec @ 130 degC	~56.5 (based on 2 runs)
1/4 4" Si wafer with non-patterned ~125 nm CSAR, not crystal bonded to Si carrier	~83.3 (based on 3 runs)
1/4 4" Si wafer with non-patterned ~125 CSAR, crystal bonded to 4" Si carrier	~54 (based on 1 run)

Etch rates and profile inspection

Continous Etches

Recipe nano1.42 on Deep Reactive Ion Etch PEGASUS A-1

Recipe	Gasses	C4F8 75 sccm, SF6 38 sccm	Profiles of lines exposed at 300 µC/cm2, etched 2:30 minutes (150s) with recipe 'nano1.42'
	Pressure	4 mTorr, Strike: 3 secs @ 15 mTorr
	Power	800 W Coil Power, 40 W Platen Power
	Platen temperature	- 20°C
Conditions	Conditioning	Pre-clean: 10 min oxygen clean 5 min oxygen clean between runs
Etch rates	Si	500 nm lines: ~200 nm/min 190 nm lines: ~200 nm/min 102 nm lines: ~190 nm/min 61 nm lines: ~170 nm/min
	CSAR	~55 nm/min

Recipe processC on Deep Reactive Ion Etch PEGASUS A-1

Recipe	Gasses	C4F8 70 sccm, SF6 38 sccm	Profiles of lines exposed at 300 µC/cm2, etched 60s with recipe 'ProcessC'
	Pressure	4 mTorr, Strike: secs @ mTorr
	Power	450 W Coil Power, 100 W Platen Power
	Platen temperature	10°C
Conditions	Conditioning	Pre-clean: 10 min oxygen clean 5 min oxygen clean between runs
Etch rates	Si	500 nm lines: ~300 nm/min 102 nm lines: ~250 nm/min
	CSAR	158 nm/min

Bosch Etch

Recipe NBoost01 on Deep Reactive Ion Etch PEGASUS A-1

Recipe	Deposition step	Duration	2.5 s	Profiles of lines exposed at 300 µC/cm2, etched 6:00 minutes with recipe 'NBoost01'
		Gasses	C4F8 50 sccm, SF6 0 sccm
		Pressure	10 mTorr
		Powers	500 W Coil
	Etch step (boost)	Duration	1.5 s
		Gasses	C4F8 0 sccm, SF6 60 sccm
		Pressure	5 mTorr
		Powers	400 W Coil, 50 W Platen
	Etch step (main)	Duration	3.5 s
		Gasses	C4F8 40 sccm, SF6 60 sccm
		Pressure	15 mTorr
		Powers	400 W Coil, 20 W Platen
	Platen temperature	20 °C
Conditions	Conditioning	Pre-clean: 10 min oxygen clean 5 min oxygen clean between runs
Etch rates	Si	200 nm lines: ~700 nm/min 130 nm lines: ~580 nm/min
	CSAR	~18 nm/min