Page 1
1
Pershan, Weizmann,
Jan.’06
Nano-Liquids, Nano-Particles, Nano-Wetting: X-ray Scattering
StudiesPhysics of Confined Liquids with/without Nanoparticles:
Confinement Phase transitions are suppressed and/or shifted. When do Liquids fill nano-pores?
(i.e. wetting and capillary filling). Contact Angles vary with surface structure. (i.e. roughness & wetting) Attraction/repulsion between surfaces. (i.e. dispersions or aggregation) Important for formation of Nanoparticle arrays:
(i.e. electronic/optical properties, potential use for sensors, catalysis, nanowires)
How will these affect nano-scale liquid devices?How will these affect processes that are essential for
nano-scale liquid technology?
Page 2
2
Pershan, Weizmann,
Jan.’06
Applications of Nano-Liquids/Nano-Particles
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
A. Terray, J. Oakey, and D. W. M. Marr, Science 296, 1841 (2002).
•Particle rotation by optical traps Pump
•3m Silica in 6m Channel
Sawitowski, T., Y. Miquel, et al. (2001). "Optical properties of quasi one-dimensional chains of gold nanoparticles." Advanced Functional Materials 11(6): 435-440.
Nano Particle Structures
Page 3
3
Pershan, Weizmann,
Jan.’06
Co Workers
Harvard Students and Post DocsK Alvine Graduate Student PhD Expected Jan/Feb 06 D. Pontoni Post Doc.O. Gang Former Post Doc. Current: Brookhaven National Lab.O. Shpykro Former Grad. Student & Post Doc. Current: Argonne National LabM. Fukuto Former Grad. Student & Post Doc. Current: Brookhaven National LabY. Yano Former Guest. Current: Gakushuin Univ., Japan
OthersB. Ocko Brookhaven National Lab.D. Cookson Argonne National Lab.A. Checco Brookhaven National Lab.F. Stellacci MITK. Shin U. Mass. AmherstT. Russell U. Mass. AmherstC. Black I.B.M.
Page 4
4
Pershan, Weizmann,
Jan.’06
Liquid SurfacesTraditional Tools and/or
TechniquesContact AngleEllipsometry Non-Linear Optics
P(2ω) ~χsurf (2ω,ω)Ein(ω)Quartz
Microbalance
Macroscopic •Length Scale: m
•Interpretation: TheoryAbsorption
AFM Imaging
Page 5
5
Pershan, Weizmann,
Jan.’06
Noncontact AFM imaging of liquids A. Checco, O. Gang and B. Ocko (Brookhaven National Laboratory)
lock-in Deflectionsensor
sine-wave generator
AFMpiezo-scanner
dither piezoA
A
A
R
A
R = 270 kHzQ=500
Aset<10nmvan der Waals forces
Powerful: surface topology
Adsorbed Liquid
Chemical Pattern
Page 6
6
Pershan, Weizmann,
Jan.’06
AFM Visualization of Condensation of ethanol onto
COOH nanostripes 1 8 . 0 0 n m
0 . 0 0 n m
COOH
200 400 600 800 1000 12000
3
6
9
12
15
nm
nm200 400 600 800 1000 12000
3
6
9
12
15
nm
nm
AFM topography across the
stripes
300nm
T>>0
300nm
T~0
300nm
T<0
200 400 600 800 1000 12000
3
6
9
12
15
nm
nm
1 2 3
1
2
3
Limited by size of probes.
Page 7
7
Pershan, Weizmann,
Jan.’06
Wetting & Nano Thin Films
Macroscopic Liquid/Solid:
Contact Angle
Non-wetting Wetting
Macroscopic Meniscus
δ ~D−3
Vapor Pressure Thickness
δδP
Van der Waals
Nano Thin Films
Page 8
8
Pershan, Weizmann,
Jan.’06
Control of
Bulk liquid reservoir: at T = Trsv.
Wetting film on Si(100) at T = Trsv + T.z
Outer cell: 0.03C
Saturated vapor
• Chemical potential was controlled by offset T between substrate and liquid reservoir.
• Dominant contribution to is from latent heats of pure materials:
Inner cell: 0.001C
[n(s°v – s°l)] T.
Page 9
9
Pershan, Weizmann,
Jan.’06
System I: Structure normal to the surface : X-Ray Reflectivity
qz
log R
qc
Reflectivity
el
Density Profile
qz
log Rel
z
z
Qz = 4π ( )sinα
€
Φ(Qz )2
~ A2 + B2 + 2AB cos QzD[ ]
R(Qz )=RF (Qz) Φ(Qz)
2exp −Qz
2σeff2
( )
Page 10
10
Pershan, Weizmann,
Jan.’06
Example of 1/3 Power Law
T [K]
[J/cm3]
Thi
ckne
ss L
[Å
] L (2Weff /)1/3 (T)1/3
Methyl cyclohexane (MC) on Si at 46 °C
• Via temperature offset
Comparisons
• Via gravity
For h < 100 mm,
< 105 J/cm3
L > ~500 Å
small , large L
• Via pressure under-saturation
For P/Psat > 1%,
> 0.2
J/cm3
L < 20 Å
large , small L
Page 11
11
Pershan, Weizmann,
Jan.’06
System II: Capillary Filling of Nano-Pores (Alumina)
Energy Cost of Liquid
or TCapillary Filling:
Transition
2πγ R −D⎡⎣ ⎤⎦Surface
Min: DR0
π R
2 − R −D( )2⎡
⎣⎢⎤⎦⎥
Volume
Min: D0
Page 12
12
Pershan, Weizmann,
Jan.’06
Anodized Alumina (UMA)
Fig. 1: AFM image (courtesy UMA) of anodized alumina sample. The ~15nm pores are arranged in a hcp array with inter-pore distance ~66nm
Fig 2: SEM (courtesy of UMA) showing hcp ordering of pores and cross-section showing large aspect ratio and very parallel pores.
~90 microns thick
Top
Side~ 15nm
Page 13
13
Pershan, Weizmann,
Jan.’06
SAXS Data
Pore fills with liquid Contrast Decreases
<10>
<11> <20>
Short Range Hexagonal Packing
∆T decreasing
Thin films
Condensation
Page 14
14
Pershan, Weizmann,
Jan.’06
Capillary filling—film thickness
Wal
l film
thi
ckne
ss [
nm]
~ 2γ/D
Transition
Page 15
15
Pershan, Weizmann,
Jan.’06
Adsorpton vs Shape: Phase Diagram
1/γ
System III: Sculpted Surfaces
Theory: Rascon & Parry, Nature (2000)
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Variety of Shapes (γ
Long Channels
Planar CrossoverGeometry to Planar
GeometryDominated
Height =L
xL
⎛
⎝⎜⎞
⎠⎟
γ Adsorbed Liquid ∞
Page 16
16
Pershan, Weizmann,
Jan.’06
Parabolic Pits: Tom Russell (UMA)
Diblock Copolymer in
Solvent
Self Alignment on Si
PMMA removal by UV
degradation & Chemical RinseReactive Ion
EtchingC. Black (IBM)
~40 nm Spacing
~20 nm Depth/Diameter
Height ~ r2
γ ≈2
Page 17
17
Pershan, Weizmann,
Jan.’06
X-ray Grazing Incidence Diffraction (GID) In-plane surface structure
Diffraction Pattern of Dry PitsHexagonal Packing
Thickness D~3Cross over to other filling!
Liquid Fills Pore: Scattering Decreases:
Page 18
18
Pershan, Weizmann,
Jan.’06
X-ray Measurement of Filling
GID
Electron Density vs T
Filling
Reflectivity
Filling
Page 19
19
Pershan, Weizmann,
Jan.’06
Results for Sculpted Surface
Γc ~ T( )
−βc
R-P Predictionβc~3.4
βc3
Observedβc
Sculpted Crossover to
Flat
Flat Sample
Sculpted is Thinner than Flat
Page 20
20
Pershan, Weizmann,
Jan.’06
Gold Nanoparticles & Controlled Solvation
Conventional Approach:Dry Bulk Solution Imaging of Dry Sample
Controlled Wetting:Dry Monolayer Adsorption
LangmuirIsotherms
Formation
Liftoff AreaOf Monolayer
Page 21
21
Pershan, Weizmann,
Jan.’06
Au Particles: Coating Stellacci et al OT: MPA (2:1)OT=CH3(CH2)7SHMPA=HOOC(CH2)2SHTEM
bi-modal distribution
Size Segregation
Page 22
22
Pershan, Weizmann,
Jan.’06
GID: X-ray vs Liquid Adsorption
(small particles)
GIDAdsorpt
ion
Return to Dry
Qz
QxyQxy Qxy
Page 23
23
Pershan, Weizmann,
Jan.’06
Three FeaturesThat Can Be Understood!
Solid lines are just guides for the eye!
Temperature Dependence of Reflectivity:
1-Minimum at low qz
2-Principal Peak Reduces and Shifts
3-2nd Minima Moves to Lower qz
Page 24
24
Pershan, Weizmann,
Jan.’06
Construction of Model: Dry Sample
Core size distribution
Vertical electrondensity profile
Model Fit: Based on Particle Size Distribution
Page 25
25
Pershan, Weizmann,
Jan.’06
Fits of Physical Model
1-Minimum at low qz
2-Principal Peak Reduces and Shifts
3-Second Minima Moves to Lower qz
Page 26
26
Pershan, Weizmann,
Jan.’06
Evolution of Model with Adsorption
Thin wetting film regime
Beginning of bilayer transition
Thick wetting film regime
Page 27
27
Pershan, Weizmann,
Jan.’06
DRY
toluene T~3K
tolueneT~.5K
tolueneT~5mK
tolueneT~.5K
tolueneT~3K
Bimodal/polydisperse Au nanocrystals in equilibrium with undersaturated vaporGood Solvent Poor vs Good Solvent
Rev
ersi
ble
Aggregation in Poor Solvent
Dissolution in GoodSolvent
Self Assembly
(1) dry
(2) ethanol T~K
(3)ethanolT~5mK
(4)dryagain(etOHextracted)
(5)tolueneT=5K
()tolueneT~5mK
()tolueneT~3K
Summary of Nano-particle experiments
QuickTime™ and aTIFF (LZW) decompressor
are needed to see this picture.
Page 28
28
Pershan, Weizmann,
Jan.’06
NanoParticle Assembly in Nanopores: Tubes
Empty
SEM of empty pores, diameter~30nm
50 nm
Fill with Particles ~2nm dia.
FilledTEM of nanoparticles in pores.
Page 29
29
Pershan, Weizmann,
Jan.’06
SAXS Experimental Setup
Brief experiment overview:
•Study in-situ SAXS/WAXS of particle self assembly as function of added solvent.
•Solvent added/removed in controlled way via thermal offset as in flat case.
Scattered x-rays
T
Incident x-ray's
Toluene
Alumina membraneWith nano-particles
Small Qx: Pore-Pore Distances
Large Qx, Qy.Qz: Particle-Particle Distances
z
x
Q
Qz
Qx
Top
Page 30
30
Pershan, Weizmann,
Jan.’06
Heating/Cooling, w/ nanoparticles
Hex. Packing
Small Q peaks pore filling hysteresis
<01>
<11>
<02>
With nanoparticles
• Decrease/Increase in contrast indicates pores filling/emptying.
Below: w/o nanoparticles
•Capillary transition shifts from ~2K for pores w/o nanoparticles to about ~8K w/ nanoparticles
•Strong hysteresis T~ /R
Note: Shift in Capillary Condensation
Page 31
31
Pershan, Weizmann,
Jan.’06
Larger Q Data / WAXS (Particle-Particle Scattering)
SlicesImages
Inte
ns
ity
q radial (spherical coord.)
Inte
ns
ity
q radial In
ten
sit
y
q radial
Thin film
Filled pore
1
2
3
Page 32
32
Pershan, Weizmann,
Jan.’06
Modeling WAXS with Shell/Tile Model
1) Break shell up into ~flat tiles no correlation between tiles.
2) Powder average over all
tiles of a given orientation.
3) Scattering from 2) is same from flat monolayer
• S(q) is 2D lorentzian ring
• F(q) is form factor for distribution of polydisperse spheres (Shulz)
4) Add up scattering from all tile orientations
Page 33
33
Pershan, Weizmann,
Jan.’06
Shell model fits for thin films:
•fit slices simultaneously with 3 global parameters plus backgnd.
•Nanoparticle radius, polydispersivity from bulk meas.
•Fits in good agreement with data.
Fitted Data
High T
Page 34
34
Pershan, Weizmann,
Jan.’06
Summary of Au-Au Scattering(Drying)
Real space modelSlices
q radial
Inte
ns
ity
q radial
Inte
ns
ity
q radial
Images
Inte
ns
ity
Cylind.Shell
Shell + Isotropicclusters
Shell + Isotropicsolution
Heatin
g
Page 35
35
Pershan, Weizmann,
Jan.’06
Summary nanoparticle self-assembly
• Strong dependence upon solvent:– Subtle confinement effect for aggregation in “poor” solvent
– Most systems reversible upon adding/removing solvent• Able to probe different geometries:
– Flat sheets– Pores tubes– Some similarity, interesting differences
• Thermal offset method gives us precise control of self-assembly process while doing in-situ measurements.
Page 36
36
Pershan, Weizmann,
Jan.’06
Critical Casimir Effect in Nano-Thick LiquidsBinary Liquid
47.7 °C
46.2 °C
45.6 °C
[Heady & Cahn, J. Chem. Phys. 58, 896 (1973)]
Tc = 46.13 0.01 °C, xc = 0.361 0.002
x (PFMC mole fraction)
Tem
pera
ture
[C
]PFMC rich
MC rich
Methylcyclohexane (MC)
Perfluoro-methylcyclohexane
(PFMC)
Page 37
37
Pershan, Weizmann,
Jan.’06
Thermodynamic Casimir effect in critical fluid filmsFisher & de Gennes (1978): Confinement of critical fluctuations in a fluid
produces “force” between bounding interfaces
Bulk MC + PFMC reservoir:(x ~ xc = 0.36) at T = Trsv.
wetting film on Si(100)
T = Trsv + T.
Outer cell: 0.03C
Inner cell: 0.001C
Same Experiment: Thickness of Absorbed Film
T=(T-Tc)/Tc
Film-TRes2 Phase
Coexistence
Vapor Phase
Liquid Phase
Critical Point
ExperimentalPaths
ExperimentalPaths
Page 38
38
Pershan, Weizmann,
Jan.’06
X-ray reflectivity Film thickness L
Tfilm [°C]
Film
thic
knes
s L
[Å
]
0.50 K
0.10 K
0.020 K
x = 0.36 ~ xc
Tc =
46.
2 °C
T
qz [Å1]
R/R
F
Paths
Page 39
39
Pershan, Weizmann,
Jan.’06
Theory
( ) ( )22 L
LTk
L
WLLF cBeff ξθ
μ ++Δ≈ΔExcess free energy/area of a wetting film:
Casimir term
( ) 3/12⎥⎦
⎤⎢⎣
⎡Δ
Θ+≈⇒
μ
ξLTkWL cBeff “Force” or “pressure” balance: 0=
∂∂−LF
y = (L/)1/ = t (L/0+)1/ y = (L/)1/ = t (L/0
+)1/
+
,(y
) (+,)
+
,(y
) /
+,(
0)
(+,)
(+, +)(+, +)
d = 4 Ising (mean field)[M. Krech, PRE 1997]
d = 2 Ising (exact)[R. Evans & J. Stecki, PRB 1994]
Page 40
40
Pershan, Weizmann,
Jan.’06
Experiment vs Theory
y = (L/)1/ = t (L/0+)1/
T 0.020 K0.10 Kd = 2 (exact)
d = 4 (MFT)
+,(y)
+,(0)
Theory for d=3 does not exist!
There is prediction for for 3D.
Page 41
41
Pershan, Weizmann,
Jan.’06
Universal “Casimir amplitudes”
• At bulk Tc (t = 0), scaling functions reduce to:
For d = 3 Ising systems
Renormalization Group (RG)Monte Carlo [M. Krech, PRE 1997]
-0.326-0.345
2.392.450
“Local free energy functional” theory (LFEF)[Z. Borjan & P. J. Upton, PRL 1998]
-0.42 3.1
Our Result N/A 3 ± 1
For recent experiments with superfluid He (XY systems), see: R. Garcia & M.H.W. Chan, PRL 1999, 2002; T. Ueno et al., PRL 2003
(0) = (0)/(d – 1)
Page 42
42
Pershan, Weizmann,
Jan.’06
Summary
Flat Surfaces: van der Waals1/3 power law
Porous Alumina: Capillary filling
Sculpted Surfaces: Cross over behavior
Nano Particles: Flat Surface Self Assembly & Solvent Effects. Size Segregation.
Nano Particles: Porous Alumina- Reversible self assembly, dissolution within the pore. Capillary filling changed be presence of the particles
Casimir Effect.
Monodisperse Particle Vary force/solvent effects (Casimir
effects) Variation in Self Assembly
Test Casimir effect for symmetric bc.
Delicate Control of Thickness of Thin Liquid Layers T)
Future