permanent-magnet helicon sources for etching, coating, and thrust

Permanent-magnethelicon sources for

etching, coating, and thrust

Francis F. Chen, UCLA

Low Temperature Plasma Teleseminar, June 14, 2013; originally prepared for (but not given at):2013 Workshop on Radiofrequency Discharges, La Badine, La Presqu’ile de Giens, France, 29-31 May 2013

Helicons are RF plasmas in a magnetic field

UCLA

Prf % 200 16% 400 21% 600 35% 800 38%

1000 42%

Density increase over ICP

Helicon sources usually use heavy magnets

UCLA

The MØRI source of Plasma Materials Technologies, Inc.

UCLA

Size of the MØRI source

UCLA

The source has been simplified to this

This uses a commercially available 2 x 4 x 0.5 inch

NeFeB magnet

The discharge is 5 cmhigh and 5 cm in diameter

UCLA

How did we get here?

Lc

a b

h

Loop antennaHelical antenna

B0

D. Arnush, Phys. Plasmas 7, 3042 (2000).

First, the HELIC code of Arnush

Examples of HELIC results: Pr , Pz

UCLA

0

1000

2000

3000

4000

5000

0.000 0.005 0.010 0.015 0.020 0.025r (m)

P(r

) (a

rb.)

2.5E+114.0E+116.3E+111.0E+12

n (cm-3)

0

1

2

3

4

5

6

0 5 10 15 20 25 30z (cm)

P(z

)

1.6E+122.5E+124.0E+126.3E+121.0E+13

13 MHz, 80G

Most important, RnB: power deposition

UCLA

RnB results organized in matrices

UCLA

100W

300W

500W

D = 7.4" D = 11" D = 13"

Folder Folder Folder

1 mTorr, ei = -1D = 0

Folder

Folder Folder Folder

FolderD13W100 D0W100

Folder

D7W100Folder

D11W100Folder

D0W500

D7W300 D11W300 D13W300 D0W300

D7W500 D11W500 D13W500Folder

conductor

insulator

1 mTorr 10 mTorr

Folder FolderP1cond P10cond

D = 0, 300W

Folder FolderP1ins P10ins

Matrix A: B-field Matrix C: pressure

Similar matrices for tube size determined the tube design.

After that, there are matices for the experimental results.

Optimized discharge tube: 5 x 5 cm

UCLA

1. Diameter: 2 inches

2. Height: 2 inches

3. Aluminum top

4. Material: quartz

5. “Skirt” to prevent eddy currents canceling the antenna current

5.1 cm

10 cm

5 cmANTENNA

GAS INLET (optional)

The antenna is a simple loop, 3 turns for 13 MHz, 1 turn for 27 MHz. The antenna must be as close to the exit aperture as possible.

Antenna: 1/8” diam tube, water-cooled

Design of matching circuit (1)

UCLA

R, L

R, L

R, L

R, L

PS

N loads

Z2 - short cables

Distributor

Z1Z2

Z1 - long cable

Matching ckt.(standard) 50W

C1C2

Matching ckt.(alternate)

C1C2

Design of 50W matching circuit (2)

UCLA

Let the load be RL + jXL :

R and X have first to be transformed by cable length

k = R0C, R0 = 50W

Forbidden regions of L, R, and Z (N = 8)

UCLA

0

500

1000

1500

2000

0.5 1.5 2.5 3.5 4.5R (ohms)

C (p

F)

C1(S)C2(S)

0

200

400

600

800

1000

1200

1400

1600

0 50 100 150 200Z2 (cm)

C (p

F)

C1(S)C2(S)

0

200

400

600

800

0 0.5 1L (mH)

C2

(pF)

86421

N

0

500

1000

1500

2000

0 0.5 1 1.5 2 2.5L (mH)

C1

(pF)

86421

N

Instead, we can use annular PMs

UCLA

-16

-14

-12

-10

-8

-6

-4

-2

0

2

4

6

8

-10 -8 -6 -4 -2 0 2 4 6 8 10

NdFeB ring magnet3” ID, 5” OD, 1” high

Stagnation point

Put plasma here

or here, to adjust B-field

The far-field is fairly uniform

PM Permanent magnet

Experimental chamber

UCLAGate Valve

To Turbo Pump

PUMP BAFFLE AND GROUND PLANE

WALL MAGNETS

Port 1

Port 3

Port 2

38

46

6.8

16.9

27.2

19 cm

Langmuir probes at three ports

The magnet height is set for optimum B-field

UCLA

Optimization of the B-field

Peak density in Port 2 vs. B and Prf @ 27 MHz

30-60 G is best

UCLA

Typical scan of “Low-field peak”

UCLA

Density of ejected plasma

0

1

2

3

4

5

-25 -20 -15 -10 -5 0 5 10 15 20 25r (cm)

n (1

011

/cm

3 )

n11, 65Gn11, 280G

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

-20 -10 0 10 20r (cm)

n (1

011

cm

-3)

6.816.927.2

62 G, 400Wcm below source

Gate Valve

To Turbo Pump

PUMP BAFFLE AND GROUND PLANE

WALL MAGNETS

Port 1

Port 3

Port 2

38

46

6.8

16.9

27.2

19 cm

Vertical probe for inside plasma

UCLA

UCLA

0

2

4

6

8

10

12

14

16

0 2 4 6 8 10 12z (cm)

n 11

Jan. 4 w. Mk2Jan. 5 w. ESPion

Bottom of tube

400W, 62G

Horiz. Probe point

0

4

8

12

16

0 2 4 6 8 10 12z (cm)

n 11

400W, 13.56 MHz, 15 mTorr

Axial density profile inside the tube

Density inside the tube is low (<1013 cm-3) because plasma is efficiently ejected.

UCLA

Downstream density: 6 x 1011 cm-3

0

1

2

3

4

5

6

7

0 200 400 600 800 1000 1200Prf (W)

n (1

011 c

m-3

)

HeliconICP

15 mTorrPort 2

Prf % 200 16% 400 21% 600 35% 800 38% 1000 42%

Density increaseover ICP

UCLA

Port 2 density is higher at 13 MHz

0

1

2

3

4

5

6

7

8

9

0 200 400 600 800 1000 1200Prf (W)

n (1

011 c

m-3

) 13 MHz

27 MHz

Port 2

This was a surprise and is contrary to theory.

Cause and location of the “double layer”

1/20½, /s sn n e

1/2 2( / ) ½ ½is s e is is ev c KT M W Mv KT

20 0 0/ / ( / )B B n n r r

F.F. Chen, Phys. Plasmas 13, 034502 (2006)

n

xs x

ne = ni = n

PLASMA

SHEATH

ni

ne

+

ns

PRESHEATH

v = cs

0 , where e /e en n e V KT Maxwellian electrons

Bohm sheath criterion

1/40/ 1.28r r e

A sheath must form here

Single layer forms where r has increased 28%

UCLA

Where a diffuse “double layer” would occur

0

50

100

150

200

250

5 10 15 20 25 30z (cm)

B (G

)

Approx. location of "double layer"

Ion energy distribution by Impedans SEMion

UCLA

IEDFs vs. Prf in Port 1

UCLA

0E+00

5E-07

1E-06

2E-06

2E-06

0 5 10 15 20 25 30V

IED

F

600W500W400W400W repeat300W200W100W

Vs

Conditions at Port 1 (r = 0) at 400W

UCLA

0

1

2

3

4

5

6

-20 -10 0 10 20r (cm)

n (1

011 c

m-3

)

Port 1400W

Te (eV)

n11

0

1

2

3

4

5

6

7

-100 -80 -60 -40 -20 0 20V

I2 (mA

)2

Ii squaredIi squared (OML)

0.01

0.1

1

10

100

-5 0 5 10 15 20V

I e (m

A)

IeIe(fit)Ie (0)

Te eVs iVs n11

2.14 15.9 27 4.21

Plasma potential

Mach probe (in Port 2)

UCLA

0.50

1 ln 0.75ln

( 1.34)

flow up up

s down down

v I Ic K I I

K

Hutchinson:

Result:

0.14f

s

vc

0.14f

s

vc

This seems very low. It should be much higher in Port 1

UCLA

helicon ARRAY sources

SUBSTRATE

TO PUMP

DISTRIBUTOR

MEDUSA MEDUSA 1

The Medusa 2 large-area array

UCLA

An array source for roll-to-roll processing

UCLA

165 cm

30 cm

15 cm

Probe ports

Aluminum sheetHeight can be adjusted electrically if desired

The source requires only 6” of vertical space above the process chamber. Two of 8 tubes are shown.

Z1Z2

65"

21"

7"

7"

7"

7"29"

3.5"

12"

7"3/4” aluminum

Endplates: gas feed and probe port at each end.

Tubes set in deeply (½ inch)

1/2" aluminum

Top view of Medusa 2

Possible positions shown for 8 tubes.

Substrate motion

Two arrangements of the array

Staggered array

Covers large area uniformly for substrates moving in the y-direction

165 cm

53.3 cm

17.8

17.8

17.8

35.6 cm

x

y

165 cm

53.3 cm

17.8

17.8

17.8

x

yTop view 17.8 Compact array

Gives higher density, but uniformity suffers from

end effects.

Operation with cables and wooden magnet tray

It’s best to have at least 3200W (400W per tube) to get all tubes lit equally.

Details of distributor and discharge tube

UCLA

The top gas feed did not improve operation.

A rectangular 50W transmission line

50-W line with ¼” diam Cu pipe for cooled center conductor

Operation with rectangular transmission line

Radial profile at Z2 across rows

0

0.5

1

1.5

2

2.5

3

3.5

-25 -20 -15 -10 -5 0 5 10 15 20 25r (cm)

n (1

011 c

m-3

) nKTe

Density profiles with staggered array

Staggered configuration, 2kW

Bottom probe array

0

1

2

3

4

5

-8 -6 -4 -2 0 2 4 6 8 10 12 14 16x (in.)

n (1

011 c

m-3

)

-3.503.5

Staggered, 2kW, D=7", 20mTorr

y (in.)

Argon

UCLA

Density profiles with compact array

Compact configuration, 3kW

Bottom probe array

0

2

4

6

8

10

-8 -6 -4 -2 0 2 4 6 8 10 12 14 16x (in.)

n (1

011 c

m-3

)

3.5-03.5

Compact, 3kW, D=7", 20mTorr

y (in)

Data by Humberto Torreblanca, Ph.D. thesis, UCLA, 2008.

Plans for an 8-tube array for round substrates

UCLANo center tube is necessary!