The case for soft Xrays: Improved compositional contrast for structure and morphology determination

1
The case for soft X-rays Improved
compositional contrast for structure and
morphology determination using real and
reciprocal space methodsAppalachian State
UniversitySept 18, 2009H. AdeDepartment of
PhysicsNorth Carolina State Universityhttp//www
.physics.ncsu.edu/stxm/
2
Objectives of presentation

• Soft X-rays The relative latecomer with unique
  attributes
• Soft X-rays require synchrotrons
• no laboratory source user community that could
  feed the synchrotron community
• Provide an overview, give a flavor of what is
  possible
• Not always the latest examples
• Sometimes work in progress
• Provide Tutorial on X-ray Physics and Resonant
  Soft X-ray Scattering (RSoXS)
• Complement to conventional x-rays and neutrons
• Combination of best aspects of both
• Potential of index-matched GI-SoXS
• This could be a big deal
• Initial steps
• Most examples are related to organic electronic
  devices
• Talk is really about a tool though

3
Soft X-raysUnique and specific interactions
with matter yield unique contrastandunique
characterization capabilities(without contrast,
we see nothing!)
4
Scattering factors and optical constants of C,N,
and O
Assumed density of 1 g/cm3
Complex index of refraction n1-diß
Scattering contrast
Assumed density of 1 g/cm3

• Quantitative absorption microscopy
• Beers Law II0e-µ?t
• 20-200 nm thick samples

5
Near Edge X-ray Absorption Fine Structure
(NEXAFS) Spectroscopy Resonant effects are more
than just elemental
Unsaturation CO
Unoccupied Molecular Orbital
e-
Unsaturation CC
C 1s edge 290 eV N 1s edge 405 eV O 1s edge
540 eV
hn
6
Resonant Scattering/Reflectivity (contrast is
almost as good as selective deuteration)
Absorption (NEXAFS)
Scattering factors f and f (optical const. d
and ß, respectively) show strong energy dependence
Dispersion
R or I ? (?d2?ß2)

• Bond specific scattering!
• Substantial potential as complementary tool!

7
Structure/morphology Determination in Real
Space Absorption X-ray Microscopy
Best for NEXAFS Relatively slow Low damage
8
MicrogelsInvestigated in-situ in solvated
state with real-space microscopy
9
Direct Imaging and Spectroscopic Characterization
of Stimuli-Responsive MicrogelsT. Araki, H. Ade
(NCSU), S. Fujii, S.P. Armes (U. Sheffield, UK)

• Microgel particles (100 nm to 1 µm)
• Swell, deswell by 30X in volume
• pH, temperature, addition of electrolyte etc
• Application
• drug delivery, chemical separations, sensors,
  catalysis, dynamically tunable microlenses,
  templates for synthesis of inorganic
  nanoparticles, water purification, viscosity
  modifiers, and as smart particulate emulsifiers
  
• Important parameters
• Size, shape, dispersion
• Stabilizer, crosslinker, silica distribution
• Water content hydration
• Charge distribution, protonation
• In situ, real space observation difficult

SEM image of dried lightly cross-linked
poly(4-vinylpyridine)-silica (P4VP-SiO2)
nanocomposite microgel particles.
10
Two types of microgels were investigated
P2VP
P4VP
20 nm silica sol
DVB cross-linker
n
cross-linker
Polymeric stabilizer
N
Poly(ethylene glycol) monomethacrylate PEGMA
Samples S.P. Armes group (Sheffield)
Ethylene glycol dimethacrylate (EGDMA)
11
Lightly cross-linked poly(4-vinylpyridine)-silica
nanocomposite microgel particlesReal space
correlation of size to charge state has been
achieved
wet cell in STXM
S. Fujii, S.P. Armes, T. Araki, H. Ade. J. Am.
Chem. Soc. (Communication) 127, 16808 (2005).
12
Sterically stabilized P2VP microgels show a size
dependenceon pH and salt concentration
S. Fujii D. Dupin, T. Araki, S. P. Armes, and H.
Ade, "First Direct Imaging of Electrolyte-Induced
Deswelling Behavior of pH-Responsive Microgels in
Aqueous Media Using Scanning Transmission X-ray
Microscopy" . Langmuir 25, 2588 (2009)
13
Sterically stabilized P2VP microgels show a size
dependenceon pH and salt concentrationDirect
spectroscopic characterization
S. Fujii D. Dupin, T. Araki, S. P. Armes, and H.
Ade, "First Direct Imaging of Electrolyte-Induced
Deswelling Behavior of pH-Responsive Microgels in
Aqueous Media Using Scanning Transmission X-ray
Microscopy" . Langmuir 25, 2588 (2009)
14
Resonant Soft X-ray Scattering
15
NC STATE University
Ade Research Group (Polymer Physics/X-ray
Characterization Techniques)
PMMA/P(BA-co-S)
P(MA-b-MMA) / PS
J.M. Stubbs, D.C. Sundberg Polymer 46 (2005) 1125
Different process and composition

• Fuzzy TEM
• modified core/shell structure?
• Phase less separated?
• Where really is the PS?

Idealized, consensus particle
PS
PS
PS-shell
200 nm
150 nm
Scattering Intensity a.u
100 nm

• PS about same size than PMMA/acrylate!
• Scattering indicates PS is slightly more in
  center of nanoparticle relative to
  PMMA/P(BA-co-S)

q (1/nm)

• PS effective radius larger than PMMA radius

T. Araki et al. Appl. Phys. Lett. 89, 124106
(2006)
16
Organic DevicesSolar Cells and Light Emitting
DiodesInterfaces and nano-structure are
critically important Part I Nanostructures
17
Organic Electronic Devices- Cheap,Efficient
Solar Cells- Cheap efficient LEDsMotivationCha
racterization ChallengesReport explicitly
statesTo increase the operational
understanding of these (organic) solar cells, new
experimental approaches will be needed to
correlate the chemical and physical properties of
the active components and layers with their
performance in operating PV devices.
Setting up the challenge!
http//www.sc.doe.gov/bes/reports/files/SEU_rpt.pd
f.
18
Bulk heterojunction organic PVAn emerging, third
generation technology

• Photo-excitation creates exciton (EB0.2 eV)
• not free carriers as in p-n junctions
• Excitons are neutral
• 10 nm random-walk mean free path

• Acceptor/Donor components

E. Moons, J. Physics-Condensed Matter 14, 12235
(2002).

• Organics can be cast from solution
• Potentially cheap process
• Non-equilibrium structures at multiple length
  scales are formed
• Target Solar paint on flexible substrate

Absorption ? diffusion ? dissociation ? charge
transport
19
Fabrication of organic solar cell
Solvent casting
Phase separation
T or vapor annealing

• Ideal structure needs to be achieved
• Exciton diffusion length 10-20 nm
• Photon absorption length 100-200 nm
• Bi-continuous interpenetration 10 nm wide
  network of 150 nm thick film (BHJ)
• Process control and optimization
• Heuristic principles, Edisonian approach works,
  but is usually tedious
• Advanced characterization might provide insight
  that lead to shortcut

20
MDMO-PPVPCBM blends solar cells

• Continuous domain of 45 (wt) MDMO-PPV / 55 (wt)
  PCBM.
• PCBM rich clusters are 100 pure within a few
  weight percent.
• Exciton harvesting suppressed

C.R. McNeill, B. Watts, L. Thomsen, W.J. Belcher,
N.C. Greenham and P.C. Dastoor, Small 2, 1432
(2006)
21
poly(9,9-dioctylfluorene-co-bis-N,N-(4,butylphen
yl)-bis-N,N-phenyl-1,4-phenylene-diamine) (PFB)
and poly(9,9?-dioctylfluorene-co-benzothiadiazole)
(F8BT) solar cells
Cast from p-xylene
Nanoscale phase separation has to be evoked to
explain remaining discrepancies
C. R. McNeill et al. Macromolecues 40, 3263 (2007)
22
85 nm thick 11 PFBF8BT blend all polymer solar
cell cast from chloroform has much finer domain
structure

• Have insufficient spatial resolution
• Need 3D resolution, i.e. tomography

Tg 160C
22
C. R. McNeill, B. Watts, L. Thomsen, W. Belcher,
S. Swaraj, H. Ade, and P. C. Dastoor,
Nanotechnology, in press (2008).
23
RSoXS provides more details about domain
evolution in 11 PFBF8BT Good enough contrast
to measure these thin films in transmission
RSoXS
23
C. R. McNeill, B. Watts, L. Thomsen, W. Belcher,
S. Swaraj, H. Ade, and P. C. Dastoor,
Nanotechnology, in press (2008).
24
RSoXS provides more details about domain
evolutionGood enough contrast to measure these
thin films in transmission
Correlation function
24
C. R. McNeill, B. Watts, L. Thomsen, W. Belcher,
S. Swaraj, H. Ade, and P. C. Dastoor,
Nanotechnology, in press (2008).
25
Same other recent NEXAFS microscopy work on
organic devices

• C. R. McNeill, B. Watts, L. Thomsen, H. Ade, N.
  C. Greenham, and P. C. Dastoor, "X-ray Microscopy
  of Photovoltaic Polyfluorene Blends Relating the
  Nanomorphology to Device Performance",
  Macromolecules 40, 3263 (2007).
• C. R. McNeill, B. Watts, L. Thomsen, W. Belcher,
  S. Swaraj, H. Ade, and P. C. Dastoor, "Evolution
  of the nanomorphology of photovoltaic
  polyfluorene blends Sub-100 nm resolution with
  X-ray spectromicroscopy", Nanotechnology 19,
  424015 (2008).
• C. R. McNeill, B. Watts, L. Thomsen, W. J.
  Belcher, N. C. Greenham, P. C. Dastoor, and H.
  Ade, "Evolution of Laterally Phase-Separated
  Polyfluorene Blend Morphology Studied by X-ray
  Spectromicroscopy", Macromolecules 42, 3347
  (2009).
• K. B. Burke, W. J. Belcher, L. Thomsen, B. Watts,
  C. R. McNeill, H. Ade, and P. C. Dastoor, "Role
  of Solvent Trapping Effects in Determining the
  Structure and Morphology of Ternary Blend Organic
  Devices", Macromolecules 42, 3098 (2009).
• B. Watts, W. J. Belcher , L. Thomsen, H. Ade and
  P. C. Dastoor, A Quantitative Study of PCBM
  Diffusion During Annealing of P3HTPCBM Blend
  Films, Macromolecules, (2009).

26
Status of the field

• Efficiency advances seem to be leveling off
  (still 6)
• Field would benefit from improved (detailed)
  understanding of fundamental operation, limits
  and identification of bottlenecks
• Band-alignment
• Morphology
• Composition
• We need to measure in order to understand
• High spatial resolution compositional mapping
  (real space or reciprocal space)
• High spatial resolution of charge generation and
  carrier properties
• Interfacial properties (width - roughness,
  interdiffusion)

27
Soft X-ray Beam Induced Current
(SoXBIC)Motivation In-situ local efficiency
of PFBF8BT PV devices

• Segregation can be controlled by choice of
  solvent
• Segregated devices are not as bad as expected.
• Snaith et. al correlated efficiency with
  interfacial area between domains.
• Snaith, H. J. Friend, R. H. Thin Solid Films
  2004, 451-452, 567-571.
• McNeill used near-field photocurrent mapping to
  show that domain centers are responsible for the
  majority of the generated photocurrent.
• Need to quantify measurements to determine what
  is (or is not) determining device efficiency.

McNeill, C. R. Frohne, H. Holdsworth, J. L.
Dastoor, P. C. Nano Lett. 2004, 4, 2503-2507.
28
(No Transcript)
29
Where is the collected current being generated?
Exponential absorption within active layer
30
The collected current has a strong contribution
from the boundary layer. This implies space
charge limitations due to limited p-mobility in
PFB
F8BT composition map
- - PFB
F8BT
31
Organic DevicesSolar Cells and Light Emitting
DiodesInterfaces and nano-structure are
critically importantPart II Interfaces
32
PFB/F8BT bilayer photodiode devicesDark-current
subtractedDevices made by floating and
transfer, then annealing
S. Swaraj, C. Wang, H. Yang, N C. Greenham, H.
Ade. C.R. McNeill
33
The Role of Interface in Photovoltaic Devices

• After exciton is created, it needs to find an
  interface with a potential difference
• Then it needs to be dissociated
• Two steps
• Roughness ? increased area
• Sharp ? increased field E?V/d
• Use resonant reflectivity to investigate
  interfaces in model bilayer systems
• Or deuterate one component and use neutron
  reflectivity.
• This used be method of choice, as conventional
  x-rays have low contrast to polymer/polymer
  interfaces

34
Reflectivity Tuturial-1
270 eV
35
Reflectivity Effect of roughness (simulation)
Reflectivity from rough interface
Surface Roughness
100 nm PMMA single layer
36
Reflectivity Tutorial-3 (simulations)
PS 50nm/PMMA 200nm
37
Soft X-ray Resonant Reflectivity First
DataPS/PMMA bilayer
NC STATE University
Ade Research Group (Polymer Physics/X-ray
Characterization Techniques)
Complementary Tool to Neutrons and hard X-rays
C. Wang, T. Araki, H. Ade Appl. Phys. Lett. 87,
214109 (2005)

• Observed strong photon energy dependence

Potential Diffuse scattering from interfaces
Data Araki, BL6.3.2. ALS, Berkeley
38
Interpretation of Reflectivity Data
39
PFB/F8BT bilayer photodiode devicesDark-current
subtracted
-6V, it seems that having a rougher interface can
increase device performance (increased exciton
dissociation?) Yet, charge separation easiest in
unannealed bilayer, which has sharpest interface
S. Swaraj, C. Wang, H. Yang, N C. Greenham, H.
Ade. C.R. McNeill
40
PFB/F8BT bilayer devicesWork in progress. Data
is 1.5 week old
Annealing increases interface and surface widths
substantially! Roughness or interdiffusion????
41
PFB/F8BT bilayer devicesWork in progress. First
fits. Analysis update from 9/4/2009
Annealing increases interface and surface widths
substantially! Roughness or interdiffusion???? N
eed to analyze more spots/samples (we have six)
to get better statistics
42
Comparison to device performance is now possible

• Interfaces with smallest width are best
• Non-equilibrium processing might have to be
  explored
• But, how much roughness/diffusion do we actually
  have?

43
Organic light emitting diodesEfficient use of
electricity
Differential casting of materials often employed
Physics of OLED is the reverse of the PV systems
44
PLED interfaces Model bilayerspoly2-methoxy-5-(
2'-ethylhexyloxy)-p-phenylene vinylene (MEH-PPV)
and poly9,9-bis(6-N,N,N,-trimethylammoniumhexyl)
fluorene-co-alt-1,4 phenylene bromide (PFNBr)
PFNBr
MEH-PPV
With A. Garcia, T.-Q. Nguyen, G. C. Bazan, K.E.
Sohn, and E.J. Kramer (UCSB), and C. Wang, A.
Hexemer (ALS) ,

• MEH-PPV is a neutral conjugated polymer spun-cast
  from non-polar solvent Toluene
• PFNBr is a charged conjugated polymer (conjugated
  polyelectrolyte) spun-cast from polar solvent
  Methanol

Direct casting
Bilayer
45
Interface between donor and acceptor materials is
critical, but little understood

• How does device efficiency/performance depend on
  roughness and interdiffusion
• How do we distinguish and quantify roughness and
  diffusion at buried interfaces?
• What is the band alignment and charge transfer at
  the interface?
• Is that also dependent on roughness/interdiffusion
  ?
• ? Use resonant reflectivity to investigate
  interfaces in model bilayer systems.

46
Hard X-ray reflectivity and RSoXR Experimental
comparison
47
Quantitative Analysis
PFNBr/MEH-PPV Bilayers
PPV single layer
(1) PFNBIm4-/MEH-PPV, (2) PFNBr-/MEH-PPV, (3)
PFNBr-/annealed MEH-PPV, and (4) annealed
PFNBr-/MEH-PPV bilayer
48
Comparison to device performance is now possible

• Devices with smallest width are best (again)
• Non-equilibrium processing seems to be required!
• But, how much roughness/diffusion do we actually
  have?

• What contribution does chemical interdiffusion
  and physical roughness make to the width?

? opportunity for index matched GI-SoXS (diffuse
scattering)
49
Unique opportunity GISAXS from buried
polymer/polymer interface
50
Unique opportunity GISAXS from buried
polymer/polymer interface
(Figure from http//www.gisaxs.de/theory.html)
51
Incoherent vs. Coherent X-Ray Scattering
Figure courtesy J. Stohr
Resonant scattering and resonant speckle should
be excellent tools for organic materials !!!!
52
Outlook for Soft X-rays

• Lots of great science possible (Its also fun!)
• Software development required to capture
  information at all energies
• Dedicated instrument development
  required/preferred
• Turning surface reflections off
• Index-matched GIXS from interfaces
• Beware of
• Radiation damage
• Parasitic scattering
• Higher order spectral contamination
• Carbon contamination in beamline (on optics,
  filters, etc.)
• Need thin samples (this is good and bad)

53
Thank you for your attention!!!!Thanks to
members of my groupB. Watts (now SLS), S.
Swaratj, T. Araki (now Toyota), H. Yanand Bill
Slotter, Jan Luning (Stanford). C. McNeill
(Cambridge), C. Wang, A. Hexemer (ALS), A.
Garcia, T.-Q. Nguyen, G. C. Bazan, K.E. Sohn, and
E.J. Kramer (UCSB),D. Kilcoyne and T. Tylisczek,
and others as indicated previouslyI hope some
of you will end up using soft x-rays for your
applications
54
Spectroscopy selective contrast
Dhez, Ade, and Urquhart J. Electron Spectrosc.
128, 85 (2003)
vectra?
PC
PAR
Kevlar
NEXAFS microscopy/Resonant scattering is really
only game in town for quantitative compositional
analysis of soft matter at high spatial resolution
Data Stony Brook STXM at NSLS
PBT
PNI
PET
Fingerpt.ppt 8 May 1998