Thermodynamics of Materials

(1)

Thermodynamics of Materials

24th Lecture 2007. 6. 4 (Monday)

1. N.M. Hwang and D.Y. Yoon

"Driving Force for Deposition in the Chemical Vapor Deposition Process,"

J. Mater. Sci. Lett. vol. 13 (1994) 1437-1439.

2. N.M. Hwang

“Thermodynamic Analysis of the Chemical Vapor Deposition of Diamond in the C-H, C-O and C-H-O Systems,"

J. Crystal Growth , vol. 135 (1994) 165-171.

3. N.M. Hwang and D.Y. Yoon

"Thermodynamic Approach to the Chemical Vapor Deposition Process,"

J. Crystal Growth , vol. 143 (1994) 103-109.

Thermodynamics of CVD

(2)

CVD Diamond

(3)

Gas Gas

Diamond

Diamond GraphiteGraphite

Thermodynamic and Kinetic Analysis of Diamond Deposition in CVD

Diamond

Graphite Gas

Chem. Potential of Carbon

Hwang et al.

Diamond Relat. Mater.

1 (1992) 9

Since graphite is more stable than diamond at low pressure used in CVD, the driving force for precipitation of graphite from the gas phase is larger than that

of diamond.

However, the larger driving force does not guarantee the

dominant formation, which is determined by kinetic barrier.

If the kinetic barrier for diamond

formation from the gas phase

is lower than that for graphite,

diamond can be formed dominantly

over graphite.

(4)

Capillary Pressure in the Nucleation Capillary Pressure in the Nucleation

Stage of Diamond or Graphite Stage of Diamond or Graphite

r

P

₂

P

₁

P r P

P 2 σ

1

2

− =

= Δ

Cluster of 1 nm in radius

2 9

2 2 3.7 / 10

J M

r M

σ

−

= × = 7400 MPa

0 500 1000 1500 2000

-10 -5 0 5 10

graphite diamond

Gibbs free energy, 10

-20

KJ

Number of carbon atoms Size Dependence of Clusters on

Stability between Diamond and Graphite

Hwang et al.

1 (1992) 9

No Gas Activation

→ Graphite Gas Activation

→ Diamond

(5)

Surface Energy is the only one that can be varied by the processing condition

n* = 36 π ( σ

dia

(Ω

dia

)

³

_

2

_ σ

gra

(Ω

gra

)

³

2

)

³

f f

gra dia

Role of the Gas Activation is to

modify the surface energy in the way favoring the stability of diamond over that of graphite

essential to the dominant formation of diamond over graphite

0 500 1000 1500 2000

-10 -5 0 5 10

20%

10% graphite diamond

Gibbs free energy, 10

-20

KJ

Number of carbon atoms Size Dependence of Clusters on

Stability between Diamond and Graphite

Hwang et al.

J. Crystal Growth 172 (1997) 416

(6)

Winter et al. J. Computer- Aided Mater. Des. 5 (1998) 279 Barnard et al. Diamond Relat. Mater. (2003)

Fe Si

Si substrate Fe substrate

Hwang et al : J. Crystal Growth 162 (1996) 55

(7)

Preferential formation of Soot at the Edge

These results imply that the growth source might be electrically charged.

Gerhardt and Homann Combst. Flame 81 (1990) 289

Formation Mechanism of Soot Particles Formation Mechanism of Soot Particles

→ → Neutralization of Charged Nanoparticles Neutralization of Charged Nanoparticles

(8)

Current Measurements in CVD Diamond Process

Current Measurements (d= 1 cm, 20 Torr)

-26000 -21000 -16000 -11000 -6000 -1000

1000 1200 1400 1600 1800 2000

Temperature (^oC) Current (nA/cm2)

~ 20 μA/cm²

→ 2 ⅹ10¹⁴ ions /s

CH4+ H2

Current Meter Gas

CH4(1 sccm) H2 (99 sccm)

Temperature 1000 ℃ 1100 ℃ 1200 ℃ . . . . 2000 ℃

Pressure 20 Torr

Distance (d) 1 cm

Pump

CH4+ H2

Current Meter Gas

CH4(1 sccm) H2 (99 sccm)

Temperature 1000 ℃ 1100 ℃ 1200 ℃ . . . . 2000 ℃

Pressure 20 Torr

Distance (d) 1 cm

Pump

1 A = ~10

1 A = ~10¹⁹¹⁹ions /sions /s

Ion-Induced Nucleation

3

4

2

3 4

G π r f π σ r

Δ = Δ +

radius

Gibbs Free Energy

3 2

4

2

3 4

r e

G f r

r

π π σ

Δ = Δ + +

radius

Gibbs Free Energy

(9)

- Designed to test the nucleation behavior

Wilson Cloud Chamber

- Ion induced nucleation could not be avoided!

- Utilized to detect the track of the high energy particle (Bubble Chamber Experiment)

Ion-induced nucleation

in cloud and bubble chambers

Discovery of the Positron (the anti-electron)

(10)

C.T.R. Wilson in 1927

For his invention

of Wilson cloud chamber

Nobel Prize Works related to Nobel Prize Works related to Ion Ion -Induced Nucleation - Induced Nucleation

C.D. Anderson in 1936

For his discovery of positron

P.M.S. Blackett in 1948

For his works on

of Wilson cloud chamber method and cosmic rays

Wilson or Bubble Chamber Experiments - Small amount of ions

- Large amount of a medium to precipitate

Thin Film, Nanotube, Nanowire CVD Reactor - Large amount of ions

- Small amount of a medium to precipitate

Why don’t we worry about Ion-Induced Nucleation in Thin Film, Nanotube, Nanowire Processes?

→ Charged nuclei grow instantly into visible size.

→Charged nuclei maintain invisible nanosize,

suspended in the gas phase like nano-colloids.

(11)

Slow Flocculation Self-Assembly

Fast Flocculation Random Coagulation

Fe Si

(a) Si substrate (b) Fe substrate

Sedimentation Behavior of Micron

Sedimentation Behavior of Micron- -Size Colloids Size Colloids

Electrical double layer stabilizes dielectric diamond cluster.

Pd Pt Rh Ir Ni Fe Au W Nb Mo Cu Ta Al Ti

SUBSTRATE

SOOT

DIAMOND

Low Charge Transfer Rate High

(12)

T_s,1= s,2=1263 K

f,1=2473 K, _f,2= 2323K Deposition Time = 2 hr (* s : substrate, f : filament)

T T

(a) Fe substrate (b) Fe substrate on quartz T

Park, Park, Hwang, Ihm, Tejima, Nakamura, Phys. Rev. B 69 (2004) 195411

(13)

References References

1.1. N.M. Hwang, J.H. Hahn, and D.Y. Yoon, J. Crystal Growth 162, 55 N.M. Hwang, J.H. Hahn, and D.Y. Yoon, J. Crystal Growth 162, 55 (1996)(1996) 2.2. I.D. I.D. JeonJeon, C.J. Park, D.Y. Kim and N.M. Hwang, J. Crystal Growth 213, 79 , C.J. Park, D.Y. Kim and N.M. Hwang, J. Crystal Growth 213, 79 (2000)(2000) 3.

3. N.M. Hwang and D.Y. Kim, J. Crystal Growth 218, 40 (2000)N.M. Hwang and D.Y. Kim, J. Crystal Growth 218, 40 (2000)

Diamond

Simultaneous Diamond Deposition and Graphite Etching

Yabrough, J. Am. Ceram. Soc., 75, (1992) 3179 : Thermodynamic Paradox Directions of atomic flux violate the 2^ndLaw of Thermodynamics.

gra dia

c c

μ < μ

^μ ^μ

μ μ μ

μ

gas gra

dia gr

c c

gas c

a

dia c c

c

graphite etching :

diamond deposition : < ⎫⎪ <

> ⎬⎪⎭

Atomic Hydrogen Hypothesis

Atomic hydrogen etches graphite much faster than diamond.

Therefore, less stable diamond can deposit and stable graphite etches simultaneously.

↔ The second law of thermodynamics requires that if stable graphite etches, the less stable diamond should etch also.

(14)

Diamond Deposition on the Etch Pit of Graphite Substrate

A.R. Badzian et al., Mat. Res. Bull., 23 (1988) 531

Weight Change of Graphite Ring

M. C. Salvadori et al, J. Electrochem. Soc., 139 (1992) 558 0.5%CH₄-H₂, 80 Torr. Tsub= 870^oC

Deposition Time (h)

(15)

before experiment after experiment 37.06 mg

43.21 mg (6.15 mg)

T

_F

: 2100

^o

C, T

_s

: 1050

^o

C, 1% CH

₄

- H

₂

, 2h Simultaneous Diamond Deposition and Graphite Etching

Diamond

Simultaneous Diamond Deposition and Graphite Etching

Hwang et al.

1 (1992) 9

Thermodynamic Paradox in CVD Diamond Growth

(16)

CVD phase diagram of C-H system at 2700 Pa

Gas + Solid

Gas

(17)

Paradox-free direction of carbon flux

Hwang et al. J. Crystal Growth, 160 (1996) 98 : Diamond growth by clusters The thermodynamic paradox can be avoided if diamond deposits by clusters.

atoms clusters

Paradox or Diamond Deposition by Clusters ?

Chemical potential of Carbon

Diamond

Chemical potential of Carbon

Before Gas Phase Nucleation

Diamond

Graphite

Gas

After Gas Phase Nucleation Gas Phase Nuclei or Clusters

10^-5torr 6 torr

5

6 4 Repellor

Detector Ground

E = ½ mv

²

: energy of cluster

Velocity is determined by Wien filter.

Energy is determined by energy analyzer.

Mass determination ! Wien Filter for Experimental Verification for

Existence of Charged Clusters

(18)

0 5 10 15 20 25 30 35 40 0

-100 -200 -300 -400 -500 -600

1.5% CH₄, 2100^oC, 6.0 torr

dI/dm (nA/amu)

Mass, 10

³

amu

J. Crystal Growth 213 (2000) 79

Negatively-Charged Clusters of 200 ~ 300 Carbon Atoms Exist in the Gas Phase in Abundance!

0 5 10 15 20 25 30 35 40

0 -10 -20 -30 -40 -50 -60

1.0%

1.5%

3.0%

5.0%

dI/dm (μA/amu)

Mass (X1000amu)

J. Crystal Growth, 223 (2001) 6

Small Clusters

Large Clusters Cluster size Increases with

increasing CH

₄

concentration.

(19)

For TEM observation, individual clusters were captured on a TEM grid with Mo mesh and a silica membrane during oxy-acetylene synthesis of diamond.

(a) TEM observation of clusters of 1 ~ 1.5 nm

(b) Some clusters show the lattice fringe of diamond with (111) spacing of 2.06 Å.

J. Crystal Growth 234 (2002) 399

Cauliflower Structure in Diamond and Silicon CVD Processes

(a) Diamond (b) Silicon

(20)

What would happen in Si CVD?

SiH

₄

: HCl : H

₂

= 1:1:98, 1333 Pa, 1123K

J. Crystal Growth, 218 (2000) 27 (b) Fe substrate (a) Si substrate

Fe Si

Si substrate

Si substrate Fe substrateFe substrate

Time Evolution of Microstructure on Ni substrate in Si CVD

(a) 3 min (b) 30 min

SiH

₄

: HCl : H

₂

= 1:1:98, 1333 Pa, 1123K

J. Crystal Growth, 218 (2000) 27

(21)

(c) after 6 min on Si

Selective Deposition and Nanowire Growth by Charged NPs

Electrostatic Interaction between CNPs and charged rods

→ Nanowire Growth

J. Crystal Growth, 218 (2000) 33

(a) after 3 min on Mo

Conductor (Mo)

(a) after 24 min on SiO₂ (b) after 3 min on Si

Insulator (SiO₂) CNPs have difficulty in landing on insulators.

→ Selective Deposition

J. Crystal Growth, 206 (1999) 177

SiH₄: HCl : H₂= 3 : 1 : 96, 950^oC, 10 Torr

Rev. Mod. Phys., Vol. 77, No. 2, April 2005

Colloquium: Reactive plasmas as a versatile nanofabrication tool K. Ostrikov*

School of Physics, The University of Sydney, Sydney NSW 2006, Australia

(22)

KIST

(23)

People tend not to believe the existence of invisible things.

Conventional Paradigm of CVD

New Paradigm (Films or Nanowires by CVD) Concept similar to Polymorphous Film Growth Ludwig Boltzmann

(1844-1906) Atomic Theory

S = k log W

atoms → invisible

Nanoparticles → invisible

> 40 papers Inter. Mater. Rev.

49 (2004) 171

CVD chamber

DMA

MFC

Faraday Cup Electrometer

Vacuum Pump

N₂ O₂

P

filter pressure gauge

P

MFC

filter

Purge Quartz tube

Quartz tube

In-situ Measurements of CNPs by nano-DMA

Differential Mobility Analyzer

1. ZnO Nanowires 2. Carbon Nanotubes 3. Si CVD

Atmospheric CVD

(24)

HW-CVD System & Charge Effect

Silicon HWCVD 66.7 Pa

20%-SiH

₄

-80%-H

₂

Thin Solid Film, in press.

SEM and AFM microstructures of silicon deposited at 450

^o

C

μm

μm μm

μm

Mobility = 175 cm

²

/Vsec (measured in LG-Phillips)

Pure & Appl. Chem. 78 (2006) 1715

Low Temp. Depo. of Polycrystalline Silicon

by Hot Wire CVD

(25)

HW-CVD System & Charge Effect

Lower Part

Upper Part

Effect of Charge Removal on Film Growth

20% SiH4(30 sccm) Inlet gas

0.5 torr Pressure

450 ℃ Substrate Temp.

6.5 cm 1650 ℃

Distance Wire Temp.

6 min 5 min

4 min 3 min

2 min

Î As Si film grows, charge mobility changes (Glass vs Si)

(26)

Selected Papers on Theory of Charged Clusters (TCC)

• N.M. Hwang and J.H. Hahn and D.Y. Yoon,

• "Chemical Potential of Carbon in the Low Pressure Synthesis of Diamond,"

• J. Crystal Growth, vol. 160, (1996) 87-97.

•

• N.M. Hwang and D.Y. Yoon,

• "Thermodynamic Approach to Paradox of Diamond Formation with Simultaneous Graphite Etching in the Low Pressure Synthesis of Diamond,"

• J. Crystal Growth, vol. 160 (1996) 98-103.

•

• N.M. Hwang, J.H. Hahn and D.Y. Yoon

• "Charged Cluster Model in the Low Pressure Synthesis of Diamond,"

Selected Papers on TCC

• N.M. Hwang

• "Deposition and Simultaneous Etching of Si in the CVD Process: Approach by the Charged Cluster Model,"

• J. Crystal Growth, vol. 205 (1999) 59-63.

• • N.M. Hwang, W.S. Cheong and D.Y. Yoon

• "Deposition Behaviors of Si on Insulating and Conducting

• Substrates in the CVD Process: Approach by Charged Cluster Model"

• J. Crystal Growth, vol. 206 (1999) 177-186.

(27)

Selected Papers on TCC

• I.D. Jeon, C.J. Park, D.Y. Kim, and N.M. Hwang

• “Experimental Confirmation of Charged Carbon Clusters in the Hot Filament Diamond Reactor"

•

• N.M. Hwang, W.S. Cheong, D.Y. Yoon and D.Y. Kim

• "Growth of Silicon Nanowires by Chemical Vapor Deposition :

• Approach by Charged Cluster Model,"

•

• W.S. Cheong, D.Y. Yoon, D.-Y. Kim and N.M. Hwang

• "Effect of Substrates on Morphological Evolution of a Film in the

• Silicon CVD Process: Approach by Charged Cluster Model,"

Selected Papers on TCC

• I.D. Jeon, C.J. Park, D.Y. Kim and N.M. Hwang

• “Effect of Methane Concentration on Size of Charged Clusters in the Hot Filament Diamond CVD Process,"

• J. Crystal Growth, 223 (2001) 6-14.

• H.S. Ahn, H.M. Park, D.Y. Kim and N.M. Hwang, "Observation of

• Carbon Clusters of a few Nanomters in the Oxyacetylene Diamond

• CVD Process"

• J. Crystal Growth, 234 (2002) 399-403.

•

• S.C. Lee, B.D. Yu, D.Y. Kim and N.M. Hwang,

• "Effects of Cluster Size and Substrate Temperature on the

• Homoepitaxial Deposition of Au Clusters,"

• J. Crystal Growth, 242, (2002) 463-470.

(28)

Selected Papers on TCC

• N.M. Hwang, D.Y. Kim, "Charged Clusters in Thin Film Growth,"

• Int. Mater. Rev. 49 (3-4) (2004) 171-190

• J.Y. Kim, D.Y. Kim and N.M. Hwang,

• “Spontaneous Generation of Negatively Charged Clusters and Their Deposition as Crystalline Films during Hot-Wire Silicon Chemical Vapor Deposition,”

• Pure Appl. Chem., 78 [9] (2006) 1715-1722.

•

• J.-H. Song, W. Jo and N.M. Hwang,

• “Non-Uniform Deposition in the Early Stage of Hot-Wire Chemical

• Vapor Deposition of Silicon: The Charge Effect Approach,”

• Thin Solid Films, in press (2007)