ASPECTS OF SOLID-SOLID REACTIONS

1
ASPECTS OF SOLID-SOLID REACTIONS

• Conventional solid state synthesis - heating
  mixtures of two or more solids to form a solid
  phase product.
• Unlike gas phase and solution reactions
• Limiting factor in solid-solid reactions usually
  diffusion, driven thermodynamically by a
  concentration gradient.
• Ficks law J -D(dc/dx)
• J Flux of diffusing species (/cm2s)
• D Diffusion coefficient (cm2/s)
• (dc/dx) Concentration Gradient (/cm4)

2
ASPECTS OF SOLID-SOLID REACTIONS

• The average distance a diffusing species will
  travel ltxgt
• ltxgt (2Dt)1/2 where t is the time.
• To obtain good rates of reaction you typically
  need the diffusion coefficient D to be larger
  than 10-12 cm2/s.
• D Doexp(-Ea/RT) diffusion coefficient increases
  with temperature, rapidly as you approach the
  melting point.
• This concept leads to empirical Tammans Rule
  Extensive reaction will not occur until
  temperature reaches at least 1/3 of the melting
  point of one or more of the reactants.

3
RATES OF REACTIONS IN SOLID STATE SYNTHESIS ARE
CONTROLLED BY THREE MAIN FACTORS

• 1. Contact area surface area of reacting solids
• 2. Rates of diffusion of ions through various
  phases, reactants and products
• 3. Rate of nucleation of product phase
• Let us examine each of the above in turn

4
SURFACE AREA OF PRECURSORS

• Seems trivial - vital consideration in solid
  state synthesis
• Consider MgO, 1 cm3 cubes, density 3.5 gcm-3
• 1 cm cubes SA 6x10-4 m2/g
• 10-3 cm cubes SA 6x10-1 m2/g (109x6x10-6/104)
• 10-6 cm cubes SA 6x102 m2/g
  (1018x6x10-12/104)
• The latter is equal to a 100 meter running
  track!!!
• Clearly reaction rate influenced by SA of
  precursors as contact area depends roughly on SA
  of the particles

5
EXTRA CONSIDERATIONS IN SOLID STATE SYNTHESIS
GETTING PRECURSORS TOGETHER

• High pressure squeezing of reactive powders into
  pellets, for instance using 105 psi to reduce
  inter-grain porosity and enhance contact area
  between precursor grains
• Pressed pellets still 20-40 porous
• Hot pressing improves densification
• Note contact area NOT in planar layer lattice
  diffusion model for thickness change with time,
  dx/dt k/x

6
EXTRA CONSIDERATIONS IN SOLID STATE SYNTHESIS

• x(thickness planar layer) µ 1/A(contact area)
• A(contact area) µ 1/d(particle size)
• Thus particle sizes and surface area connected
• Hence x µ d
• Therefore A and d affect interfacial thickness
  x!!!
• These relations suggest some strategies for rate
  enhancement in direct solid state reactions by
  controlling diffusion lengths!!!

7
MINIMIZING DIFFUSION LENGTHS ltxgt (2Dt)1/2 FOR
RAPID AND COMPLETE DIRECT REACTION BETWEEN SOLID
STATE MATERIALS AT LOWEST T
8
MINIMIZING DIFFUSION LENGTHS ltxgt (2Dt)1/2 FOR
RAPID AND COMPLETE DIRECT REACTION BETWEEN SOLID
STATE MATERIALS AT LOWEST T
9
MINIMIZING DIFFUSION LENGTHS ltxgt (2Dt)1/2 FOR
RAPID AND COMPLETE DIRECT REACTION BETWEEN SOLID
STATE MATERIALS AT LOWEST T

• Johnson superlattice precursor
• Deposition of thin film reactants
• Controlled thickness, composition
• Metals, semiconductors, oxides
• Binary, ternary compounds
• Modulated structures
• Solid solutions (statistical reagent mixing)
• Diffusion length x control
• Thickness control of reaction rate
• Low T solid state reaction
• Designer element precursor layers
• Coherent directed product nucleation
• Oriented product crystal growth
• LT metastable hetero-structures
• HT thermodynamic product

SUPERLATTICE REAGENTS
10
ELEMENTALLY MODULATED SUPERLATTICES -DEPOSITED
AND THERMALLY POST TREATED TO GIVE LAYERED METAL
DICHALCOGENIDES MX2
COMPUTER MODELLING OF SOLID STATE REACTION OF
JOHNSON SUPERLATTICE
11
ELEMENTALLY MODULATED SUPERLATTICES

• Several important synthetic parameters and in
  situ probes.
• Reactants prepared using thin film deposition
  techniques and consist of nm scale layers of the
  elements to be reacted.
• One element easily substituted for another
• Allows rapid surveys over a class of related
  reactions and synthesis of iso-structural
  compounds.

12
ELEMENTALLY MODULATED SUPERLATTICES

• Diffusion distance is determined by the
  multilayer repeat distance which can be
  continuously varied.
• An important advantage, allowing experimental
  demonstration of changes in reaction mechanism as
  a function of inter-diffusion distance and
  temperature.
• Multi-layer repeat distances can be easily
  verified in the prepared reactants and products
  made under different conditions using low angle
  X-ray diffraction.

13
MINIMIZING DIFFUSION LENGTHS ltxgt (2Dt)1/2 FOR
RAPID AND COMPLETE DIRECT REACTION BETWEEN SOLID
STATE MATERIALS AT LOWEST T
Johnson superlattice reagent design (Ti-2Se)6(Nb
-2Se)6n Low T annealing reaction (TiSe2)6(NbS
e2)6n Metastable ternary modulated layered
metal dichalcogenide (hcp Se2- layers, Ti4/Nb4
Oh/D3h interlayer sites) superlattice well
defined PXRD Confirms correlation between
precursor heterostructure sequence and
superlattice ordering of final product
SUPERLATTICE REAGENTS YIELD SUPERLATTICE
ARTIFICIAL CRYSTAL PRODUCT
14
Superlattice precursor sequence
6(Ti-2Se)-6(Nb-2Se) yields ternary modulated
superlattice composition (TiSe 2)6(NbSe 2)6n
with 62 well defined PXRD reflections good
exercise give it a try Confirms correlation
between precursor heterostructure sequence and
superlattice ordering of final product
15
MINIMIZING DIFFUSION LENGTHS ltxgt (2Dt)1/2 FOR
RAPID AND COMPLETE DIRECT REACTION BETWEEN SOLID
STATE MATERIALS AT LOWEST T
John superlattice reagent design (Ti-2Se)6(Nb-2S
e)6n High T annealing reaction (Ti0.5Nb0.5Se2
)n Thermodynamic linear Vegard type solid
solution ternary metal dichalcogenide product
with identical layers Properties of ternary
product is the atomic fraction weighted average
of binary end member components P(TixNb(1-x)Se2)
xP(TiSe2) (1-x)PNbSe2
SUPERLATTICE REAGENTS YIELD HOMOGENEOUS SOLID
SOLUTION PRODUCT
16
NANOCLUSTER PRECURSOR BASED KIRKENDALL SYNTHESIS
OF HOLLOW NANOCLUSTERS

• Synthesis of surfactant-capped cobalt
  nanoclusters
• Co(3)/BH4(-) reduction in oleic acid,
  oleylamine ? ConLm in a high temperature
  surfactant solvent,
• surfactant-sulfur injection, coating of sulfur
  shell on nanocluster
• cobalt sesquisulfide product shell layer formed
  at interface
• counter-diffusion of Co(2)/2e(-) and S(2-)
  across thickening shell
• faster diffusion of Co(2) than S(2-) creates
  VCo in core
• vacancies agglomerate in core
• hollow core created which grows as the product
  shell thickens
• end result a hollow nanosphere made of cobalt
  sesquisulfide Co2S3

17
TURNING NANOSTRUCTURES INSIDE-OUT

• Kirkendall effect a well-known phenomenon
  discovered in 1930s.
• Occurs during reaction of two solid-state
  materials and involves the counter diffusion of
  reactant species, like ions, across product
  interface usually at different rates.
• Special case of movement of fast-diffusing
  component cannot be balanced by movement of slow
  component the net mass flow is accompanied by a
  net flow of atomic vacancies in the opposite
  direction.
• Leads to Kirkendall porosity, formed through
  super-saturation of vacancies into hollow pores
• When starting with perfect building blocks such
  as monodisperse cobalt nanocrystals a reaction
  meeting the Kirkendall criteria can lead to
  super-saturation of vacancies exclusively in the
  center of the nanocrystal.
• General route to hollow nanocrystals of almost
  any given material
• Proof-of-concept - synthesis of Co2S3 nanoshell
  starting from Co nanocluster.

18
Time evolution of a hollow Co2S3 nanocrystal
grown from a Co nanocrystal via the nanoscale
Kirkendall effect Science 2004, 304, 711
19
NANOSCALE PATTERNING OF SHAKE-AND-BAKE
SOLID-STATE CHEMISTRY
MINIMIZING DIFFUSION LENGTHS ltxgt (2Dt)1/2 FOR
RAPID AND COMPLETE DIRECT REACTION BETWEEN SOLID
STATE MATERIALS AT LOWEST T
Younan Xia
20
PDMS MASTER
Whitesides
21
PDMS MASTER
Whitesides

• Schematic illustration of the procedure for
  casting PDMS replicas from a master having relief
  structures on its surface.
• The master is silanized and made hydrophobic by
  exposure to CF3(CF2)6(CH2)2SiCl3 vapor
• SiCl bind to surface OH groups and anchor
  perfluoroalkylsilane to surface of silicon master
  CF3(CF2)6(CH2)2SiO3 for easy removal of PDMS mold
• Each master can be used to fabricate more than 50
  PDMS replicas.
• Representative ranges of values for h, d, and l
  are 0.2 - 20, 0.5 - 200, and 0.5 - 200 mm
  respectively.

22
NANOSCALE PATTERNING OF SHAKE-AND-BAKE
SOLID-STATE CHEMISTRY
Younan Xia
23
NANOSCALE PATTERNING OF SHAKE-AND-BAKE
SOLID-STATE CHEMISTRY
(A) Optical micrograph (dark field) of an ordered
2-D array of nanoparticles of Co(NO3)2 that was
fabricated on a Si/SiO2 substrate by selective
de-wetting from a 0.01 M nitrate solution in
2-propanol. The surface was patterned with an
array of hydrophilic Si-SiO2 grids of 5 x 5 mm2
in area and separated by 5 mm. (B) An SEM image
of the patterned array shown in (A), after the
nitrate had been decomposed into Co3O4 by heating
the sample in air at 600 C for 3 h. These Co3O4
particles have a hemispherical shape (see the
inset for an oblique view). (C) An AFM image
(tapping mode) of the 2-D array shown in (B),
after it had been heated in a flow of hydrogen
gas at 400 C for 2 h. These Co particles were on
average 460 nm in lateral dimensions and 230 nm
in height.
Co(NO3)2
Co3O4
Co
24
NANOSCALE PATTERNING OF SHAKE-AND-BAKE
SOLID-STATE CHEMISTRY
AFM image of an ordered 2-D array of (A) MgFe2O4
and (B) NiFe2O4 that was fabricated on the
surface of a Si/SiO2 substrate by selective
de-wetting from the 2-propanol solution (0.02 M)
that contained a mixture of two nitrates e.g.
12 between Mg(NO3)2 and Fe(NO3)3. The PDMS
stamp contained an array of parallel lines that
were 2 mm in width and separated by 2 mm. Twice
stamped orthogonally. Citric acid HOC(CH2CO2H)3
forms mixed Mg(II)/Fe(III) complex - added to
reduce the reaction temperature between these two
nitrate solids in forming the ferrite. Ferrite
nanoparticles 300 nm in lateral dimensions and
100 nm in height.
MgFe2O4
NiFe2O4
25
NUCLEATION OF PRODUCT PHASE
26
SURFACE STRUCTURE AND REACTIVITY EFFECTS ON
DIRECT REACTION OF SOLIDS
Nucleation depends on surface structure of
reacting phases - crystal faces in contact - MgO
rock salt - different Miller index faces exposed
- ion arrangements in crystal face different -
distinct crystal habits (octahedral,
cubooctahedral, cubic) possible depending on
growth conditions and additives - 100
alternating Mg(2) and O(2-) at corners of square
grid - 111 Mg(2) or O(2-) in hexagonal
arrangement - implies different surface
structures, energies and reactivities
27
DIRECT REACTION OF SOLIDS - SHAKE-AND-BAKE
SOLID STATE SYNTHESIS

• Although this approach may seem to be ad hoc and
  a little irrational at times, the technique has
  served solid state chemistry for well over the
  past 50 years
• It has given birth to the majority of high
  technology devices and products that we take for
  granted every day of our lives
• Thus it behooves us to look critically and
  carefully at the methods used if one is to move
  beyond trial-and-error methods to the new
  chemistry and a rational and systematic approach
  to the synthesis of materials

28
THINKING ABOUT REAGENTS

• Drying reagents MgO/Al2O3 200-800C, maximum SA
• In situ decomposition of precursors at 600-800C
  MgCO3/Al(OH)3 ? MgO/Al2O3
• Intimate mixing of precursor reagents
• Homogenization of reactants using organic
  solvents, grinding, ball milling,
  ultra-sonification

29
THINKING ABOUT CONTAINER MATERIALS

• Chemically inert crucibles, boats
• Noble metals Nb, Ta, Au, Pt, Ni, Rh, Ir
• Refractories, alumina, zirconia, silica, boron
  nitride, graphite
• Reactivity with containers at high temperatures
  needs to be carefully evaluated for each system

30
THINKING ABOUT SOLID STATE SYNTHESIS HEATING
PROGRAM

• Furnaces, RF, microwave, lasers, ion and electron
  beams
• Prior reactions and frequent cooling, grinding
  and regrinding, boost SA of reacting grains
• Overcoming sintering, grain growth, brings up SA,
  fresh surfaces, enhanced contact area
• Pellet and hot press reagents densification and
  porosity reduction, higher surface contact area,
  enhances rate, extent of reaction
• Care with unwanted preferential component
  volatilization if T too high, composition
  dependent
• Need INERT atmosphere for unstable oxidation
  states

31
PRECURSOR SOLID STATE SYNTHESIS METHOD

• Co-precipitation, high degree of homogenization,
  high reaction rate - applicable to nitrates,
  acetates, citrates, carboxylates, oxalates,
  alkoxides, b-diketonates, glycolates
• Concept precursors to magnetic Spinels -
  recording media
• Zn(CO2)2/Fe2(CO2)23/H2O 1 1 solution phase
  mixing
• H2O evaporation, salts co-precipitated solid
  solution mixing on atomic/molecular scale,
  filter, calcine in air
• Zn(CO2)2 Fe2(CO2)23 ? ZnFe2O4 4CO 4CO2
• High degree of homogenization, smaller diffusion
  lengths, fast rate at lower reaction temperature

32
PROBLEMS WITH CO-PRECIPITATION METHOD

• Co-precipitation requires
• Similar salt solubilities
• Similar precipitation rates
• Avoid super-saturation as poor control of
  co-precipitation
• Useful for synthesizing Spinels, Perovskites
• Disadvantage often difficult to prepare high
  purity, accurate stoichiometric phases

33
DOUBLE SALT PRECURSORS

• Known stoichiometry double salts having
  controlled stoichiometry
• Ni3Fe6(CH3CO2)17O3(OH).12Py
• Basic double acetate pyridinate
• Burn off organics at 200-300oC, then calcine at
  1000oC in air for 2-3 days
• Product highly crystalline phase pure NiFe2O4
  spinel

34
Good way to make chromite Spinels, important
tunable magnetic materials - juggling
electronic-magnetic properties of the A Oh and B
Td ions in the Spinel lattice
DOUBLE SALT PRECURSORS

• Chromite Spinel Precursor compound
  Ignition T, oC
• MgCr2O4 (NH4)2Mg(CrO4)2.6H2O 1100-1200
• NiCr2O4 (NH4)2Ni(CrO4)2.6H2O 1100
• MnCr2O4 MnCr2O7.4C5H5N 1100
• CoCr2O4 CoCr2O7.4C5H5N 1200
• CuCr2O4 (NH4)2Cu(CrO4)2.2NH3 700-800
• ZnCr2O4 (NH4)2Zn(CrO4)2. 2NH3 1400
• FeCr2O4 (NH4)2Fe(CrO4)2 1150

35
PEROVSKITE FERROELECTRICS BARIUM TITANATE

• Control of grain size determines ferroelectric
  properties, important for capacitors,
  microelectronics
• Direct heating of solid state precursors is of
  limited value in this respect lack of size and
  morphology control
• BaCO3(s) TiO2(s) ? BaTiO3(s)
• Sol-gel reagents useful to create single source
  barium titanate precursor with correct
  stoichiometry

36
SINGLE SOURCE PRECURSOR SYNTHESIS OF BARIUM
TITANATE - FERROELECTRIC MATERIAL

• Ti(OBu)4(aq) 4H2O ? Ti(OH)4(s) 4BuOH(aq)
• Ti(OH)4(s) C2O42-(aq) ? TiO(C2O4)(aq)
  2OH-(aq) H2O
• Ba2(aq) C2O42-(aq) TiO(C2O4)(aq) ?
  BaTiO(C2O4)2(s)
• Precipitate contains barium and titanium in
  correct ratio and at 920?C decomposes to barium
  titanate according to
• BaTiO(C2O4)2(s) ?? BaTiO3(s) 2CO(g) 2CO2(g)
• Grain size important for control of ferroelectric
  properties
• Used to grow single crystals hydrothermally

37
BASICS FERROELECTRIC BARIUM TITANATE
Cubic perovskite equivalent O-Ti-O bonds in BaTiO3
Tetragonal perovskite long-short axial O-TiO
bonds in BaTiO3
Small grains, tetragonal to cubic surface
gradients, ferroelectricity particle size
dependent
Multidomain ferroelectric dipoles align in E
field and/or below Tc
Multidomain paraelectric above Tc Cooperative
electric dipole interactions within each domain
aligned in domain but random between
Single domain superparaelectric
38
SOL-GEL SINGLE SOURCE PRECURSORS TO LITHIUM
NIOBATE - NLO MATERIAL

• LiOEt EtOH Nb(OEt)5 ? LiNb(OEt)6 ? LiNbO3
• LiNb(OEt)6 H2O ? LiNb(OEt)n(OH)6-n ? ? gel
• LiNb(OEt)n(OH)6-n D O2 ? LiNbO3
• Lithium niobate, ferroelectric Perovskite,
  nonlinear optical NLO material, used as
  electrooptical switch
• Bimetallic alkoxides - single source precursor
• Sol-gel chemistry - hydrolytic polycondensation ?
  gel
• MOH MOH ? MOM H2O
• Yields glassy product
• Sintering product in air - induces crystallization

39
INDIUM TIN OXIDE -ITO

• Indium sesquioxide In2O3 (wide Eg semiconductor)
  electrical conductivity enhanced by p-doping with
  (10) Sn(4)
• ITO is SnnIn2-nO3
• ITO is optically transparent, electrically
  conducting, thin films are vital as electrode
  material for solar cells, electrochromic
  windows/mirrors, LEDs, LC displays, electronic
  ink, photonic crystal ink and so forth
• Precursors - EtOH solution of (2-n)In(OBu)3/nSn(OB
  u)4
• Hydrolytic poly-condensation to form gel, spin
  coat gel onto glass substrate to make thin film
• Dry gel at 50-100?C, heat at 350?C in air to
  produce ITO
• Check electrical conductivity and optical
  transparency

40
SUB -10 NM NANOSCALE DIRECT SOLID STATE REACTION
TiO2 Electron Beam Nanolithography of
Spin-Coated Sol-Gel TiO2 Based Resists
benzoyl acetone
tetrabutoxyorthotitanate
Choosing the right solid state precursor to make
resist
41
SUB -10 NM NANOSCALE DIRECT SOLID STATE REACTION
Electron Beam Nanolithography Using
Spin-Coated TiO2 Resists

• Utilization of spin-coated sol gel based TiO2
  resists by chemically reacting titanium
  n-butoxide with benzoylacetone in methyl alcohol.
  
• They have an electron beam sensitivity of 35 mC
  cm-2 and are gt107 times more sensitive to an
  electron beam than sputtered TiO2 and crystalline
  TiO2 films.

Choosing the right solid state precursor
42
Sub-10 nm Electron Beam Nanolithography Using
Spin-Coated TiO2 Resists

• Fourier transform infrared studies suggest that
  exposure to an electron beam results in the
  gradual removal of organic material from the
  resist.
• This makes the exposed resist insoluble in
  organic solvents such as acetone, thereby
  providing high-resolution negative patterns as
  small as 8 nm wide.
• Such negative patterns can be written with a
  pitch as close as 30 nm.

Choosing the right solid state precursor
43
Nanometer scale precision structures
Nanoscale TiO2 structures offer new
opportunities for developing next generation
solar cells, optical wave-guides, gas sensors,
electrochromic displays, photocatalysts,
photocatalytic mCP, battery materials
44
Nanometer scale tolerances