Transcript Absorption on the surface - physisorption and Chemisorption

Adsorption on the surface
- physisorption and Chemisorption
Physisorption
1. Van der Waals bonding
(dipole)
2. Weak (<0.4 eV)
3. always atomic/molecular
4. Reversible and fast
5. surface symmetry
insensitive
6. may form multilayers
Chemisorption
1. Covalent/metallic/ionic
2. Strong (>0.4 eV)
3. May be dissociative
4. Often irreversible
5. surface symmetry specific
6. limited to monolayer
Typical several eVs
Typical 10 – 100 meV
Typical 3 – 10 Å
Typical 1 – 3 Å
Chemisorption
It is chemical reaction, like the chemical bonding in the molecules.
Schematic drawing show covalent
chemisorption bonding between a
molecule and a transistion metal (d
bands dominate at Fermi surface).
The
rearrangement
of the
electronic
shell during
chemisorption
may lead to
physisorption
dissociative
dissociative adsorption
adsorption
Work Function change induced by adsorption
Different
material
have
different
f.
Moreover,
chemisorption
change
the
charge distribution and cause f
change. For semiconductor
even band bending is changed.
The dipole of the adsorbate can
change f.
Under different condition (T for
example), different adsorption
bonding for the same adsorbate can
be formed on the same surface with
different f.
Phase diagram for 2 D adsorbates
When
without
long
range
order
then is
2D liquid
droplets.
Random, dilute phase of 2D gas
2D crystallites with internal order
The relative strength of the lateral and vertical interactions determines if an
ordered adsorbate layer is with registry with the substrate. The lateral ones
can be: van der Waals attraction, dipole forces (between parallel ones,
repulsive), the repulsive force due to orbital overlap between neighbors,
substrate-mediated interactions (modification of electronic structure, elastic
properties to compensate the change of the lattice.)
Phase diagram for 2 D adsorbates
Describe the
inter-particle
potential
For 3D system, van der Vaals equation ( p  an2 )(1  nb)  nKT
a 2 1
For 2D system:
( 
Splay pressure d
)(  f p )  kT

Area density
Minimum area of particle
Adsorption Kinetics
Adsorption rate:
Coverage then:
dN
p
uS
S
dt
2m kT
dN
   udt   S
dt Or
dt
S  2mkT / p
In case for activated adsorption, S as  function:
S ( )  f ( ) exp(Eact / kT )
Condensation coefficient
Occupation factor
d
dt
Adsorption isotherm
Together with adsorption process, there is desorption process which can
be described by description rate:
v   ( ) f ( ) exp(Edes / kT )
Occupation factor
Desorption efficient
Adsorption isotherm describe the equilibrium adsorbate coverage at
certain temperature between adsorption and desorption: u = v
More

f
(

)
1
f
(

)
complcated
p
2m kTe  EB / kT


f ( ) A f ( )
when
adsorbate
For non-dissociative adorpstion: f()=1-
atoms allow
And f()=. Therefore,
adsorption
Ap
 ( p) 
1  AP
Curves tell the strength of
the adsorption
Surface Diffusion
The adsorbate atoms on the surface although lost some kinetic energies
to the substrate after adsorption, it can vibrate in potential well with
frequency n0 and also hop (surface diffusion) from one site to the
neighbor site with frequency n:
1
n  n 0 exp(  E (diff ) / kT )
z
z is the number of possible neighboring wells.
Ordered monolayers can only form if surface diffusion is high.
Typically, ~ 1013 s-1 and E(diff) 100-300 kJ·mol-1
Film epitaxy
The expression “epitaxy” has been introduced on a geometrical basis. It
characterizes the oriented growth of a material A on top of a crystalline
substrate B. If A and B are the same material, the growth is called
homoepitaxy. Otherwise, it is called heteroepitaxy.
For equilibrium
condition
Bauer’s criteria: Dg=ga+gab-gb
Dg < 0 Layer-by-layer growth (Frank-van der Merwe (FM))
Dg > 0 Island growth (Vollmer-Weber mode)
Dg is function of thickness, initially Dg < 0 then Dg > 0
Transition layer/island growth(Stranski-Krastanov mode)
Growth kinetics
the microscopic kinetic processes taking place at the surface:
• Adsorption from the vapor phase
• Surface diffusion followed by binding or re-desorption
• Nucleation of 2-dimensional or 3-dimensional cluster
• Capture by an existing cluster or surface defect sites (steps)
• Interdiffusion with the substrate
in most cases the growth are done at relatively low temperature and high
deposition rate. Even thermodynamics favors 3-dimensional islands formation,
monolayer-like growth can be approached by reducing atomic mobility and so that
the equilibrium shape cannot form. The growth modes also depends on defects.
Surfactant-to achieve FM growth
The interlayer mass transport is a
prerequisit for layer-by-layer growth,
however, the potential barrier can
prevent such mass transport.
Surfactant is a contaminant
which, when deposited once,
promotes a smoother growth.
Possible mechanisms:
•Surfactant decorate edges and reduce the step-edge barrier
•It induce a potential energy gradient to attract deposited atoms toward the steps.
•It reacts as nucleation centers by attracting adatoms to increase the island density.
•Surfactant atoms are practically immobile on the surface and act repulsively to
deposited diffusing adatoms.
Some examples for surfactant
The comparison between
growth of Cu on Ru(0001)
with/without Oxygen as
surfactant, Cu on Cu(111)
with/without Pb.
Phys. Rev. Lett. 81, 850-853 (1998)
Growth techniques
As the kinetics process is important to growth, the techniques have their
distinct important factors: Deposition temperature, rate, energy of the
particles deposited, pressure, vacuum requirement, substrate morphology,
etc.
Molecular Beam Epitaxy (MBE)
Typical MBE chamber
In MBE, the constituent
elements in the form of
“molecular beams” are
deposited
onto
a
crystalline substrate to
form thin epitaxial layers.
The “molecular beams”
are
typically
from
thermally
evaporated
elemental sources.
Knudsen Effusion Cell
basis of nearly all beam
generation. The cells
contain the condensed
phase and its vapor in
equilibrium. For a cell of
orifice area, A, a distance l
from the substrate, and at
temperature T(K), the flux
of molecules or atoms
striking a unit are of
substrate :
J = (1.118x1022)pA/[l2(MT)1/2] molecules cm-2s-1
where p is the source pressure in the cell in Torr, and M is the Molecular
weight of the source material.Therefore T essentially determines the rate
and need to be controlled.
The distribution across a flat substrate for such a cell can be
expressed as: J = Jo cos4
Knudsen Effusion Cell
Normally the materials in the cell
need to be heated to increase the
vapor pressure till 10-3 Torr to get a
reasonable growth in the high
vacuum (typical rate 1 m/h or 200
Å/min). Therefore, from the
temperature dependence of the
equilibrium vapor pressure of certain
material, one can already determine
a suitable temperature for MBE
growth of this particular material. If
the temperature is too high, the
growth by MBE will be very difficult
or even practically impossible and
other methods need to be
considered.
Normally the cell is heated by
resistive way, but some new types
can use e-beam with high voltage.
cell
Electron Beam (EB) source
For the low-vapor-pressure materials, electron beam (EB) with high
voltage is used to heat the source materials to reach very high
temperature for evaporation.
No crucible is
necessary and
can fast reach
extremely high
temperature.
Magnetron Sputtering (MS)
Sputter sources for film deposition can be categorized in two ways: glow
discharge and ion beam. Magnetron Sputtering belongs to the first category. In
all cases, atoms are removed from the target by momentum transfer from the
incident ions with a large accelerating energy to the surface atoms of target
material, then the sputtered atoms for the film growth.
(c)
The plasma is important for glow discharge sputtering process and is typically
formed by partially ionizing a gas at a pressure well bellow atomspheric
(normally from 0.1 to 1000 mtorr). For the most part, these plasmas are very
weakly ionized, with an ionization ration of 10-5 to 10-1. Plasmas are
generally neutral, in that in the body of the plasma there are roughly
equal numbers of electrons and ions. They are conductive with the
dominant charge carriers: the electrons. There are generally three types
for plasma generation: excited between two powered electrode;
application of electric fields, typically through an insulator; injection of
large currents of electrons to ionize the gas particle.
Dc/rf sputtering
Dc diode sputtering with typical voltages 3-5 kV and a current from 50-250
mA at a pressure of 50-250 mtorr. For deposition purpose, the target material
will be use as cathode and samples are placed on the anode or nearby. It is
slow, and it needs high gas density, high discharge voltage, and
conductive target as the cathode (no insulator for dc sputtering).
Rf sputtering is more easily to be used with low gas density and low
voltage, and works with insulator too. The reason is the quickly change
polarity of the E field with MHz, the electron will quickly react and to
compensate any charging on the sample surface, while the ion with big
mass is still sputtering.
The bias in RF sputtering
Due to the rather bigger mass than electron, the powered electrode will
in the first several cycles collect more electrons than ions, therefore,
there will be a negative dc bias generated, which prevent electrons be
further collected and only ions will be accelerated by the dc bias
toward the electrode (the material needed to be sputtered) and less
influenced by rf field. This bias is just half of the voltage of rf
oscillation.
Magnetron Sputtering (MS)
Charged particles will have drift
motion in the presence of E and B
with drift velocity:
Vdrift = E/B
Electrons circles in a too large
circles compared with chamber
size, but ions will circle and drift
inside the chamber.
Magnetron is to keep ions
circles over the target to be
easily generate plasma.
Pulse Laser Deposition (PLD)
Pulsed laser radiation is used vaporize materials and collect the vapor onto a
substrate. The laser-solid interaction leads to evaporation, ablation, plasma
formation, and exfoliation. The plume generated by the laser beam consists of
mixture of energetic species including atoms, molecules, electrons, ions,
clusters and even micron-sized particulates.
Pulse Laser Deposition (PLD)
Schematic diagram for three fluence levels on the target surface. (a) At low
fluence ripple formation. (b) At higher fluence, surface uniformly melted and
larger scale capillary waves develop. (b) At PLD fluence, melt depth is
greater and vaporization occurs, and capillary waves still form This change
of morphology will strongly influenced the stability of the growth rate.
A series of plots of
deposition rate versus
exposure for different YBCO
target density. They all show
a clear decay.
Summary of the three deposition techniques
Techniques
(Torr)
MBE
MS
PLD
Energy(eV)
0.1
1-1000
1-1000
rate (Å/s)
0.01-1
1-100
1-100
Vacuum Operation pressure
UHV
HV
HV
10-9
10-1 and 10-3
10-9 to 10-1
MBE: good control of film (purity and thickness), too
slow, difficulties in multi-components.
MS: large area evaporation, fast growth, workable
with reactive gas, impurities, the inefficient usage of
the targetmaterial, difficulties in magnetic targets
PLD: keep stoichiometry, almost every material target,
super-high instant growth rate, reactive gas, easy to
change targets, particulates and the difficulty to grow
in large area
Thickness calibration
a. The intensity of MEED (medium energy electron diffraction) and RHEED,
or even other scattering methods like Atom scattering is sensitive to surface
roughness, which can be used to monitor the thickness during growth for
layer-by-layer mode.
b. Auger/XPS/absorption, etc,
have certain free path length,
which can be used to study the
thickness.
c. Quartz crystal deposition monitor uses the piezoelectric sensitivity of a
quartz monitor crystal to added mass, which corresponds to film thickness
when provided known density.
Auger/XPS ratio
1. Think about the mean free path length of the electrons.
2. Think about the coverage (morphology)
Chemical Vapor Deposition (CVD)
Techniques
thin films are deposited on a substrate from vapor phase precursors through chemical
reactions at the surface, CVD involves many coupled processes, including fluid flow,
heat and mass transfer, chemical kinetics, and nucleation and growth of the film on the
substrate
CVD versus physical depostion
•Compared to other deposition techniques the CVD method is
perhaps the most complex.
•Unlike growth by physical deposition such as evaporation or
MBE, this method requires numerous test runs to reach suitable
growth parameters, especially for single-crystal growth.
•The complexity of this method results from the facts that:
(i) it generally includes multicomponent species in the
chemical reactions,
(ii) the chemical reactions generally produce intermediate
products,
(iii) the growth has numerous independent variables, and
(iv) the growth includes more consecutive steps than in
physical methods.
CVD process
– CVD—thermally activated (or pyrolytic) chemical vapour deposition;
– MOCVD—metalorganic CVD;
– PCVD—photo CVD;
– PECVD—plasma-enhanced CVD;
– ALE—atomic layer epitaxy.
Technique
Energy promoting
the reactions
What is specific?
Pressure (Torr)
(Thermal) CVD
Thermal
Inorganic sources
10- 5 –760
MOCVD
Thermal
Organometallic
sources
10–760
PECVD
Plasma
Lower temperatures
10–760
PCVD
Light
Lower temperatures,
selected area
0.01–10
ALE
Thermal/Light
monolayer control
0.01–760