Jump to content

Specific Process Knowledge/Lithography/Descum: Difference between revisions

From LabAdviser
Jehem (talk | contribs)
Jehem (talk | contribs)
 
(6 intermediate revisions by the same user not shown)
Line 8: Line 8:
__TOC__
__TOC__


=Plasma Asher 1=
= Descum Comparison Table =
<span style="color:red">Plasma asher 1 was decommissioned 2024-12-02.</span>
 
[[Specific Process Knowledge/Lithography/Descum/PlasmaAsher1|Information about decommissioned tool can be found here.]]
<br clear="all" />
 
=Plasma Asher 2=
<span style="color:red">Plasma asher 2 was decommissioned 2024-12-02.</span>
 
[[Specific Process Knowledge/Lithography/Descum/PlasmaAsher2|Information about decommissioned tool can be found here.]]
<br clear="all" />
 
=Plasma Asher 3: Descum=
Product name: Diener Pico Plasma Asher<br>
Year of purchase: 2014
 
The user manual(s), quality control procedure(s) and results and contact information can be found in [http://labmanager.dtu.dk/function.php?module=Machine&view=view&mach=423 LabManager] - '''requires login'''
 
Plasma Asher 3 is specifically used for controlled descum process after lithography. Please note that you only can process a single 100 mm wafer, or one small sample, at a time.
The plasma asher is equipped with 2 gaslines: oxygen and nitrogen, but all standard processes use only oxygen (as recommended by Diener).
 
'''Process parameters'''<br>
You can manipulate two different descum process development parameters: you can either change power or chamber pressure.
 
==Power testing - AZ MiR 701==
[[image:AZMIR701_power_settings.png|400px|thumb|Descum results for different power settings]]
 
'''Recipe settings:'''
*Resist: AZ MiR 701
*O2 flow: 5 sccm
*N2 flow: 0
*Pressure: 0.2 mbar
*Power: Varied
 
{| class="wikitable" style="text-align: center;"
|+ style="caption-side: top; text-align: left;" | Experiment parameters
|-
!  !! Forward/reverse !! C2/C1 !! Power
|-
| style="text-align: left;" | '''recipe 1''' || 50/0 || 52/31 || 50%
|-
| style="text-align: left;" | '''recipe 2''' || 100/0 || 53/31 || 100%
|-
| style="text-align: left;" | '''recipe 3''' || 20/0 || 51/34 || 20%
|}
 
==Pressure testing - AZ MiR 701==
[[image:AZMIR701_pressure_settings.png|400px|thumb|Descum results for different pressure settings]]
 
'''Recipe settings:'''
*Resist: AZ MiR 701
*O2 flow: varied
*N2 flow: 0
*Pressure: varied
*Power: V100% (100 W)
 
{| class="wikitable" style="text-align: center;"
|+ style="caption-side: top; text-align: left;" | Experiment parameters
|-
!  !! Forward/reverse !! C2/C1 !! Oxygen !! Pressure
|-
| style="text-align: left;" | '''recipe 1''' || 100/0 || 52/31 || 5 || 0.2
|-
| style="text-align: left;" | '''recipe 2''' || 100/0 || 37/38 || 45 || 0.8
|}
 
==Pressure testing - AZ 5214E==
[[image:AZ5214E_pressure_settings.png|400px|thumb|Descum results for different pressure settings]]
 
'''Recipe settings:'''
*Resist: AZ 5214E
*O2 flow: varied
*N2 flow: 0
*Pressure: varied
*Power: V100% (100 W)
 
{| class="wikitable" style="text-align: center;"
|+ style="caption-side: top; text-align: left;" | Experiment parameters
|-
!  !! Forward/reverse !! C2/C1 !! Oxygen !! Pressure
|-
| style="text-align: left;" | '''recipe 1''' || 100/0 || 52/31 || 17 || 0.4
|-
| style="text-align: left;" | '''recipe 2''' || 100/0 || 37/39 || 45 || 0.8
|}
 
<br clear="all" />
 
=Plasma Asher 4=
Product name: PVA Tepla Gigabatch 380M<br>
Year of purchase: 2024
 
Descum of AZ 5214E on 100 mm wafers. The descum process development was done for a single substrate, as well as 3 substrates (for decreased ashing rate and improved ashing uniformity). The substrates were placed vertically in the glass boat.
 
<gallery style="text-align: center;" widths=250 heights=250>
PA_boat_1Wafer_v2.png|Single vertical substrate
PA_boat_3Wafer_v2.png|3 vertical substrates:<br>Dummy - Test - Dummy
</gallery>
 
For the 3 substrates only the center substrate was used for testing, the front and back wafers were used as dummy wafers. The wafers were placed in consecutive slots - when running multiple wafers in this way, the first and last wafers should always be dummy wafers.
 
The user manual, risk assessment, and contact information can be found in [http://labmanager.dtu.dk/function.php?module=Machine&view=view&mach=530 LabManager] - '''requires login'''
 
'''Typical descum parameters'''<br>
Tested with 1.5 µm AZ 5214E on 100 mm silicon substrate.
*O<sub>2</sub>: 100 sccm
*N<sub>2</sub>: 100 sccm
*Pressure (DSC): 1.3 mbar
*Power: 200 W
*Chamber temperature at start (with door closed): 30°C
*Time (single wafer): 5-10 minutes = 35-72 nm ashed
*Time (multiple wafers): 10-15 minutes = 40-80 nm ashed
 
{| class="wikitable"
{| class="wikitable"
|-
|-
! !! Single substrate !! Center of 3 substrates
!
|-
! [[Specific_Process_Knowledge/Lithography/Descum#Plasma Asher 3: Descum|Plasma Asher 3: Descum]]
! scope=row style="text-align: left;" | Test results: ashing rate
! [[Specific_Process_Knowledge/Lithography/Descum#Plasma_Asher 4|Plasma Asher 4 (Clean)]]
| 5.7 ±2.1 nm/min|| 3.8 ±1.6 nm/min
! [[Specific_Process_Knowledge/Lithography/Descum#Plasma Asher 5|Plasma Asher 5 (Dirty)]]
|-
|-
! scope=row style="text-align: left;" | Test results: non-uniformity
! scope=row style="text-align: left;" | Purpose
| 0.6 ±0.4% || 0.4 ±0.2%
| Resist descum
|
*Resist stripping
*Resist descum
|
*Resist stripping
*Resist descum
|-
|-
! scope=row style="text-align: left;" | Wafers
! scope=row style="text-align: left;" | Method
| 1 || 3
| Plasma ashing
| Plasma ashing
| Plasma ashing
|-
|-
! scope=row style="text-align: left;" | Wafer size
! scope=row style="text-align: left;" | Process gasses
| 100 mm || 100 mm
| O<sub>2</sub> (50 sccm)
|
*O<sub>2</sub> (0-500 sccm)
*N<sub>2</sub> (0-500 sccm)
|
*O<sub>2</sub> (0-500 sccm)
*N<sub>2</sub> (0-500 sccm)
*CF<sub>4</sub> (0-200 sccm)
|-
|-
! scope=row style="text-align: left;" | Boat position
! scope=row style="text-align: left;" | Process power
| Center of chamber || Center of chamber
| 10-100 W (10-100%)
| 150-1000 W
| 150-1000 W
|-
|-
! scope=row style="text-align: left;" | Test wafer position
! scope=row style="text-align: left;" | Substrate batch
| Center of boat || Center of boat
|
*Chips: several
*50 mm wafer: several
*100 mm wafer: 1
|
*Chips: several
*50 mm wafer: several
*100 mm wafer: 1-25
*150 mm wafer: 1-25
*200 mm wafer: 1-25
|
*Chips: several
*50 mm wafer: several
*100 mm wafer: 1-25
*150 mm wafer: 1-25
*200 mm wafer: 1-25
|-
|-
! scope=row style="text-align: left;" | Total gas flow rate
! scope=row style="text-align: left;" | Substrate materials
| 200 sccm || 200 sccm
|
|-
*<span style="color:red">'''No polymer substrates'''</span><br>
! scope=row style="text-align: left;" | Gas mix ratio
*Silicon substrates
| 50% N<sub>2</sub> || 50% N<sub>2</sub>
*III-V substrates
|-
*Glass substrates
! scope=row style="text-align: left;" | Chamber pressure
*Films, or patterned films, of any material except type IV (Pb, Te)
| 1.3 mbar || 1.3 mbar
|
|-
*<span style="color:red">'''No metals'''</span><br>
! scope=row style="text-align: left;" | Power
*<span style="color:red">'''No metal oxides'''</span><br>
| 200 W || 200 W
*<span style="color:red">'''No III-V materials'''</span><br>
|-
*Silicon substrates
! scope=row style="text-align: left;" | Test processing time
*Glass substrates
| Tested parameter || Tested parameter
*Polymer substrates
|-
*Films, or patterned films, of resists/polymers
! scope=row style="text-align: left;" | Test average temperature
|
| 33°C || 33°C
*Silicon substrates
*III-V substrates
*Glass substrates
*Polymer substrates
*Films, or patterned films, of any material except type IV (Pb, Te)
|}
|}
<br clear="all" />
<br clear="all" />


==Single wafer descum ashing rate and uniformity for plasma asher 4 & 5==
{{:Specific Process Knowledge/Lithography/Descum/plasmaAsher03}}
[[File:PA_descum_single_v3.png|400px|thumb|Ashing amount and ashing rate when processing a single 100 mm wafer.|right]]


{| class="wikitable" style="text-align: center;"
{{:Specific Process Knowledge/Lithography/Descum/plasmaAsher04}}
|+ style="caption-side: top; text-align: left;" | Ashing amount and rate
|-
! style="text-align: left;" | Ashing time [min]: !! 1 !! 2 !! 5 !! 10 !! 15
|-
| style="text-align: left;" | Ashing amount [nm]: || 5.2 || 6.2 || 35.1 || 72.3 || 87.1
|-
| style="text-align: left;" | Ashing rate [nm/min]: || 5.2 || 3.1 || 7.0 || 7.2 || 5.8
|}


{| class="wikitable" style="text-align: center;"
{{:Specific Process Knowledge/Lithography/Descum/plasmaAsher05}}
|+ style="caption-side: top; text-align: left;" | Ashing non-uniformity
|-
! style="text-align: left;" | Ashing time [min]: !! 1 !! 2 !! 5 !! 10 !! 15
|-
| style="text-align: left;" | Pre-descum film thickness range [nm]: || 11 || 12 || 10 || 11 || 9
|-
| style="text-align: left;" | Pre-descum non-uniformity [%]: || 0.37 || 0.40 || 0.33 || 0.37 || 0.30
|-
| style="text-align: left;" | Post-descum film thickness range [nm]: || 10 || 10 || 12 || 19 || 33
|-
| style="text-align: left;" | Post-descum non-uniformity [%]: || 0.33 || 0.33 || 0.41 || 0.66 || 1.18
|}


<br clear="all" />
=Decommisioned tools=
<span style="color:red">Plasma asher 1 was decommissioned 2024-12-02.</span>


==Multi wafer descum ashing rate and uniformity for plasma asher 4 & 5==
[[Specific Process Knowledge/Lithography/Descum/PlasmaAsher1|Information about decommissioned tool can be found here.]]
[[File:PA_descum_multi_v3.png|400px|thumb|Ashing amount and ashing rate when processing multiple 100 mm wafers. Test measured on the center of 3 wafers.|right]]


{| class="wikitable" style="text-align: center;"
|+ style="caption-side: top; text-align: left;" | Ashing amount and rate
|-
! style="text-align: left;" | Ashing time [min]: !! 1 !! 2 !! 5 !! 10 !! 15
|-
| style="text-align: left;" | Ashing amount [nm]: || 4.3 || 6.7 || 10.1 || 39.5 || 78.8
|-
| style="text-align: left;" | Ashing rate [nm/min]: || 4.3 || 3.4 || 2.0 || 4.0 || 5.3
|}


{| class="wikitable"  style="text-align: center;"
<span style="color:red">Plasma asher 2 was decommissioned 2024-12-02.</span>
|+ style="caption-side: top; text-align: left;" | Ashing non-uniformity
|-
! style="text-align: left;" | Ashing time [min]: !! 1 !! 2 !! 5 !! 10 !! 15
|-
| style="text-align: left;" | Pre-descum film thickness range [nm]: || 11 || 13 || 11 || 12 || 14
|-
| style="text-align: left;" | Pre-descum non-uniformity [%]: || 0.37 || 0.43 || 0.37 || 0.40 || 0.46
|-
| style="text-align: left;" | Post-descum film thickness range [nm]: || 11 || 9 || 10 || 12 || 21
|-
| style="text-align: left;" | Post-descum non-uniformity [%]: || 0.37 || 0.30 || 0.33 || 0.41 || 0.74
|}
 
<br clear="all" />
 
==Comparison between single substrate processing and multi substrate processing for plasma asher 4 & 5==
Processing a single substrate using the standard descum settings will provide users with a relatively controllable and uniform process. Adding dummy substrates in close proximity with the process substrate will reduce the ashing rate and improve the uniformity:
<gallery mode="packed-hover" heights="150">
PA_descum_compareAmount_v1.png|Ashing amount
PA_descum_compareRate_v1.png|Ashing rate
PA_descum_compareUniformity_v1.png|Non-uniformity
</gallery>


[[Specific Process Knowledge/Lithography/Descum/PlasmaAsher2|Information about decommissioned tool can be found here.]]
<br clear="all" />
<br clear="all" />
==Comparison of ashing rate between substrate sizes==
[[Specific_Process_Knowledge/Lithography/Strip#Comparison_of_ashing_rate_between_substrate_sizes_for_plasma_asher_4_&_5|Comparison]]
=Plasma Asher 5=
Product name: PVA Tepla Gigabatch 380M<br>
Year of purchase: 2024
Plasma asher 5 is identical to plasma asher 4, see descum processing for plasma asher 4 [[Specific_Process_Knowledge/Lithography/Descum#Plasma_Asher_4|here]].

Latest revision as of 12:31, 13 January 2026

The content on this page, including all images and pictures, was created by DTU Nanolab staff, unless otherwise stated.

Feedback to this page: click here

Descum Comparison Table

Plasma Asher 3: Descum Plasma Asher 4 (Clean) Plasma Asher 5 (Dirty)
Purpose Resist descum
  • Resist stripping
  • Resist descum
  • Resist stripping
  • Resist descum
Method Plasma ashing Plasma ashing Plasma ashing
Process gasses O2 (50 sccm)
  • O2 (0-500 sccm)
  • N2 (0-500 sccm)
  • O2 (0-500 sccm)
  • N2 (0-500 sccm)
  • CF4 (0-200 sccm)
Process power 10-100 W (10-100%) 150-1000 W 150-1000 W
Substrate batch
  • Chips: several
  • 50 mm wafer: several
  • 100 mm wafer: 1
  • Chips: several
  • 50 mm wafer: several
  • 100 mm wafer: 1-25
  • 150 mm wafer: 1-25
  • 200 mm wafer: 1-25
  • Chips: several
  • 50 mm wafer: several
  • 100 mm wafer: 1-25
  • 150 mm wafer: 1-25
  • 200 mm wafer: 1-25
Substrate materials
  • No polymer substrates
  • Silicon substrates
  • III-V substrates
  • Glass substrates
  • Films, or patterned films, of any material except type IV (Pb, Te)
  • No metals
  • No metal oxides
  • No III-V materials
  • Silicon substrates
  • Glass substrates
  • Polymer substrates
  • Films, or patterned films, of resists/polymers
  • Silicon substrates
  • III-V substrates
  • Glass substrates
  • Polymer substrates
  • Films, or patterned films, of any material except type IV (Pb, Te)


Plasma Asher 3: Descum

Product name: Diener Pico Plasma Asher
Year of purchase: 2014

The user manual(s), quality control procedure(s) and results and contact information can be found in LabManager - requires login

Plasma Asher 3 is specifically used for controlled descum process after lithography. Please note that you only can process a single 100 mm wafer, or one small sample, at a time. The plasma asher is equipped with 2 gaslines: oxygen and nitrogen, but all standard processes use only oxygen (as recommended by Diener).

Process parameters
You can manipulate two different descum process development parameters: you can either change power or chamber pressure.

Power testing - AZ MiR 701

Descum results for different power settings

Recipe settings:

  • Resist: AZ MiR 701
  • O2 flow: 5 sccm
  • N2 flow: 0
  • Pressure: 0.2 mbar
  • Power: Varied
Experiment parameters
Forward/reverse C2/C1 Power
recipe 1 50/0 52/31 50%
recipe 2 100/0 53/31 100%
recipe 3 20/0 51/34 20%

Pressure testing - AZ MiR 701

Descum results for different pressure settings

Recipe settings:

  • Resist: AZ MiR 701
  • O2 flow: varied
  • N2 flow: 0
  • Pressure: varied
  • Power: V100% (100 W)
Experiment parameters
Forward/reverse C2/C1 Oxygen Pressure
recipe 1 100/0 52/31 5 0.2
recipe 2 100/0 37/38 45 0.8

Pressure testing - AZ 5214E

Descum results for different pressure settings

Recipe settings:

  • Resist: AZ 5214E
  • O2 flow: varied
  • N2 flow: 0
  • Pressure: varied
  • Power: V100% (100 W)
Experiment parameters
Forward/reverse C2/C1 Oxygen Pressure
recipe 1 100/0 52/31 17 0.4
recipe 2 100/0 37/39 45 0.8


Plasma Asher 4

Product name: PVA Tepla Gigabatch 380M
Year of purchase: 2024

Descum of AZ 5214E on 100 mm wafers. The descum process development was done for a single substrate, as well as 3 substrates (for decreased ashing rate and improved ashing uniformity). The substrates were placed vertically in the glass boat.

For the 3 substrates only the center substrate was used for testing, the front and back wafers were used as dummy wafers. The wafers were placed in consecutive slots - when running multiple wafers in this way, the first and last wafers should always be dummy wafers.

The user manual, risk assessment, and contact information can be found in LabManager - requires login

Typical descum parameters
Tested with 1.5 µm AZ 5214E on 100 mm silicon substrate.

  • O2: 100 sccm
  • N2: 100 sccm
  • Pressure (DSC): 1.3 mbar
  • Power: 200 W
  • Chamber temperature at start (with door closed): 30°C
  • Time (single wafer): 5-10 minutes = 35-72 nm ashed
  • Time (multiple wafers): 10-15 minutes = 40-80 nm ashed
Single substrate Center of 3 substrates
Test results: ashing rate 5.7 ±2.1 nm/min 3.8 ±1.6 nm/min
Test results: non-uniformity 0.6 ±0.4% 0.4 ±0.2%
Wafers 1 3
Wafer size 100 mm 100 mm
Boat position Center of chamber Center of chamber
Test wafer position Center of boat Center of boat
Total gas flow rate 200 sccm 200 sccm
Gas mix ratio 50% N2 50% N2
Chamber pressure 1.3 mbar 1.3 mbar
Power 200 W 200 W
Test processing time Tested parameter Tested parameter
Test average temperature 33°C 33°C


Single wafer descum ashing rate and uniformity for plasma asher 4 & 5

Ashing amount and ashing rate when processing a single 100 mm wafer.
Ashing amount and rate
Ashing time [min]: 1 2 5 10 15
Ashing amount [nm]: 5.2 6.2 35.1 72.3 87.1
Ashing rate [nm/min]: 5.2 3.1 7.0 7.2 5.8
Ashing non-uniformity
Ashing time [min]: 1 2 5 10 15
Pre-descum film thickness range [nm]: 11 12 10 11 9
Pre-descum non-uniformity [%]: 0.37 0.40 0.33 0.37 0.30
Post-descum film thickness range [nm]: 10 10 12 19 33
Post-descum non-uniformity [%]: 0.33 0.33 0.41 0.66 1.18


Multi wafer descum ashing rate and uniformity for plasma asher 4 & 5

Ashing amount and ashing rate when processing multiple 100 mm wafers. Test measured on the center of 3 wafers.
Ashing amount and rate
Ashing time [min]: 1 2 5 10 15
Ashing amount [nm]: 4.3 6.7 10.1 39.5 78.8
Ashing rate [nm/min]: 4.3 3.4 2.0 4.0 5.3
Ashing non-uniformity
Ashing time [min]: 1 2 5 10 15
Pre-descum film thickness range [nm]: 11 13 11 12 14
Pre-descum non-uniformity [%]: 0.37 0.43 0.37 0.40 0.46
Post-descum film thickness range [nm]: 11 9 10 12 21
Post-descum non-uniformity [%]: 0.37 0.30 0.33 0.41 0.74


Comparison between single substrate processing and multi substrate processing for plasma asher 4 & 5

Processing a single substrate using the standard descum settings will provide users with a relatively controllable and uniform process. Adding dummy substrates in close proximity with the process substrate will reduce the ashing rate and improve the uniformity:


Comparison of ashing rate between substrate sizes

It is assumed that the low power descum rates, for different substrate sizes, follows the same pattern as the high power ashing rates, which can be seen here.

Plasma Asher 5

Product name: PVA Tepla Gigabatch 380M
Year of purchase: 2024

Plasma asher 5 is identical to plasma asher 4, see descum processing for plasma asher 4 here.

Decommisioned tools

Plasma asher 1 was decommissioned 2024-12-02.

Information about decommissioned tool can be found here.


Plasma asher 2 was decommissioned 2024-12-02.

Information about decommissioned tool can be found here.