effects of catalytic site size on activity of electrodes ... · effects of catalytic site size on...
TRANSCRIPT
Effects of Catalytic Site Size on Activity of Model Electrodes Prepared By Mass‐Selected Cluster Deposition
Alexander von Weber, Eric Baxter, Henry S. White and Scott Anderson
Electrochemistry:
Related gas‐surface chemistry:
(Early work by Sebastian Proch, Mark Wirth, Michael Rosenfelder)
In situ electrochemistry in a UHV system is hard – Why bother?
Electrochemical cleaning is fine for:• Bulk/single crystals electrodes• Large supported nanoparticles Not for small, size‐selected clusters!
Potential cycling may change the size by etching or potential‐induced diffusion
Ethanol oxidationPt4/ITO in 0.1 M HClO4in situ
Avoid complications of adventitious adsorbates
HER
OER
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
-25
-20
-15
-10
-5
0
5
10
15
20
2. cycle 55. cycle
Cur
rent
den
sity
(A
cm
-2 geo)
Potential (V vs. NHE)
CV #2CV #55
Identical conditions with air exposure
HER
OER
Size‐Selected Electrochemistry Without Air Exposure
Electrochemistry Antechamber• Separately pumped
UHV chamber• Can be vented
without affecting main chamber UHV
Triple differential seal
Electrochemistry Setup
Fritted glassElectrolyte OutflowElectrolyte Injection
Working Electrode
Ag/AgCl ReferenceElectrode
Pt mesh CounterElectrode
Viton O‐rings
Working compartment
Protocol• Clean/characterize electrode substrate in UHV• Deposit 1.5 x 1014 Pt atoms/cm2 as Ptn+
– Equivalent to 0.1 ML Pt• XPS characterization• Lower into antechamber, pressurize with argon• Seal to electrochemical cell• Pump in electrolytes, do series of cyclic voltammograms (CVs)
Problems• In situ requirement → sta c experiments only• Need for a dry, bakeable reference electrode
– Ag/AgCl, reproducible to ~15 meV from day to day
• Limited working time due to Cl‐ diffusion/poisoning.• Cell alignment to cluster spot
Fritted glassElectrolyte OutflowElectrolyte Injection
Cluster WorkingElectrode
Pt mesh CounterElectrode
Viton O‐rings
Working compartment
Oxygen Reduction Reaction (ORR)• Limiting factor in fuel cell efficiency
– Need better catalysts to reduce overpotential– Reduce amount of Pt required
• O2 + 4 e‐ + 4 H+ → 2 H2O • O2 + 2 e‐ + 2 H+ → H2O2
Marković, Schmidt, Stamenković, Ross, Fuel Cells 1, 105 (2001)
Start with Ptn/glassy carbonElectrolyte = 0.1 M HClO4
O2 saturatedN2 sparged
Size‐selected ORR chemistryPtnano/glassy carbon
Size‐selected ORR chemistry
Area Inside
cell
Pt7
Area of cluster spot
AreaOutsideCell
J. Am. Chem. Soc. 2013, 135, 3073−3086
0.0 0.2 0.4 0.6 0.8 1.0 1.2
-80
-60
-40
-20
0
20
40
60
curr
ent d
ensi
ty [
A c
m-2]
potential vs. Ag/AgCl [V]
N2-saturated O2-saturated
Pt7/glassy carbon Ptnano/glassy carbon
-0,1 0,0 0,1 0,2 0,3 0,40
20
40
60
80
100
curr
ent d
ensi
ty [m
A c
m-2]
potential vs. Ag/AgCl [V]
N2-saturated O2-saturated
Pt4
Many small Ptn catalyze the reactionC + 2 H2O → CO2 + 2 H2
With essentially no overpotential and high efficiency: J. Am. Chem. Soc. 2013, 135, 3073−3086Interesting result, but it prevents study of ORR or other reactions
Unexpected Chemistry
• Huge oxidation currents• Lots of bubbles• What is oxidizing what?
Anti‐correlations between Pt 4f binding energy and ORR catalysis
Glassy carbon oxidation by water, catalyzed by Ptn
High Pt 4f BE is associated with no carbon oxidation“normal ORR”
2 4 6 8 10
71,3
71,4
71,5
71,6
71,7
71,8
71,9pe
ak p
ositi
on o
f Pt 4
f 7/2 [
eV]
cluster size [atoms]
Bulk Pt = 71.2 eV
Only “normal”behavior “near‐
Normal”
Switch to an electrode support than can’t be oxidized:Indium Tin Oxide (ITO).
Polycrystalline with 15 – 20 nm domain size• AFM Ra = 1.1 nm• Roughness may be good
• Help stabilize clusters?1200 1000 800 600 400 200 0
0.0
50.0k
100.0k
150.0k
200.0k
C 1s
In 4s
In 3p
Sn 3d
In 3d
In 4p
O 1s
CP
S
Binding energy (eV)
Pristine UHV cleaned
In 4d
Adventitious carbonaceous adsorbates• Cannot be eliminated by heating in
UHV or in O2• Do affect chemistrySputter/anneal in O2
Avoid carbon oxidation:ITO electrode substrates
Most important point: UHV‐cleaned ITO is quite inert in the potential range of interest
‐0.1 to 1.5 V vs. NHE
-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6-20
-15
-10
-5
0
5
Cur
rent
den
sity
(A
cm
-2 geo)
Potential (V vs. NHE)
Potential Range impt forORR and EOR chemistry
Double layer charging/Capacitive current
Slow onset of hydrogen evolution (HER)
Polycrystalline with 15 – 20 nm domain size• AFM Ra = 1.1 nm• Roughness may be good
• Help stabilize clusters?
-80
-60
-40
-20
0
Pt1 - N2
Pt1 - O2-15
-10
-5
0
5
Pt2 - N2
Pt2 - O2
-80
-60
-40
-20
0
Pt5 - N2
Pt5 - O2Cur
rent
den
sity
(A
cm-2 ge
o)
-60
-40
-20
0
Pt7 - N2
Pt7 - O2
0.0 0.2 0.4 0.6-140
-120
-100
-80
-60
-40
-20
0
Pt10 - N2
Pt10 - O2
Potential (V vs. NHE)0.0 0.2 0.4 0.6 0.8
-40
-20
0
ITO - N2
ITO - O2
Pt14 - N2
Pt14 - O2
Pt1/ITO Pt2/ITO
Pt5/ITO Pt7/ITO
Pt10/ITO Pt14/ITO
ORR is clearly seen even for Pt1
NOTE alignment problemCurrents are lower limits
Mass activity is at least ~10 times higher than for Ptnano/ITO under same conditions.
ORR over Ptn/ITO
electrodesStart with narrow range CVs to avoid Pt redox chem
-100
-80
-60
-40
-20
0
20
40
60
Lin. fit Pt1 - N2
Pt1 - O2 -40
-20
0
Pt2 - N2
Pt2 - O2
-60
-40
-20
0
20
Pt5 - N2
Pt5 - O2Cur
rent
den
sity
(A
cm
-2 geo)
-60
-40
-20
0
20
Pt7 - N2
Pt7 - O2
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4-140-120-100-80-60-40-20
02040
Pt10 - N2
Pt10 - O2
Potential (V vs. NHE)0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
-60
-40
-20
0
20
Pt14 - N2
Pt14 - O2
Pt1/ITO Pt2/ITO
Pt5/ITO Pt7/ITO
Pt10/ITO Pt14/ITO
Cluster –containing electrodesFull range CVsAll sizes active for HER and ORR
• ~1.2 V Oxidation peak cannot be Pt oxidation‐ too large
• Only seen after sweeping through the ORR wave ‐ oxidation of
ORR products‐ H2O2 oxidation
(HPOR)
Hydrogen Peroxide Oxidation Rxn (HPOR)
0.8 0.9 1.0 1.1 1.2 1.3 1.4-5
0
5
10
15
20
25
Cur
rent
den
sity
(A
cm
-2 geo)
Potential (V vs. NHE)
Lower potential limit (V vs. NHE): 0.55 0.15 0.45 0.05 0.35 - 0.05 0.25
Pt14/ITO
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
-60
-40
-20
0
20
-60
-40
-20
0
20
Pt14 - N2
Pt14 - O2Cur
rent
den
sity
(A
cm
-2 geo)
Potential (V vs. NHE)
We can vary the amount of H2O2produced by changing the lower potential limit. HPOR peak should increase as we go further into the ORR wave.
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4100
-80
-60
-40
-20
0
20
40
60
-100
-80
-60
-40
-20
0
20
40
60
Cur
rent
den
sity
(A
cm
-2 geo)
Lin. fit Pt1 - N2
Pt1 - O2
ORR
HPOR
H2O2 is created during ORR at low potentials
O2 + 4 e‐ + 4 H+ → 2 H2OO2 + 2 e‐ + 2 H+ → H2O2
React catalytically on the oxidized Pt surface at high potentials
2 Pt(OH) + H2O2 → 2 Pt(H2O) + O2,2 Pt(H2O) → 2 Pt(OH) + 2 H+ + 2 e‐
H2O2 → 2 H+ + O2 + 2 e‐
HPOR = hydrogen peroxide oxidation reaction
Pt oxidized
Katsounaros, I.; Schneider, W. B.; Meier, J. C.; Benedikt, U.; Biedermann, P. U.; Auer, A. A.; Mayrhofer, K. J. J. Phys. Chem. Chem. Phys. 2012, 14, 7384−7391.
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4100
-80
-60
-40
-20
0
20
40
60
-100
-80
-60
-40
-20
0
20
40
60
Cur
rent
den
sity
(A
cm
-2 geo)
Lin. fit Pt1 - N2
Pt1 - O2
ORR
HPOR
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4-150
-100
-50
0
Potential (V vs. NHE)
Ptnano/ITO - N2
Ptnano/ITO - O2
C.Pt5/ITO Ptnano/ITO
ORR
HPOR
MUCH more HPOR for small clusters than for Ptnano or Ptpoly
Cluster size dependence of HPOR branchingand ORR onset potential
Much more H2O2formed for smallclusters. Why?
1 2 3 4 5 6 7 8 9 10 11 12 13 14 160.60
0.65
0.70
0.75
0.80
0.85
0.90
ORR onset Ampl. ratio
Platinum cluster (atoms)
OR
R o
nset
(V v
s. N
HE
)
Ptnano
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Am
plitu
de ra
tio H
PO
R/O
RR
High H2O2 branching for small clusters is not attributable to electronic effects (see below)Small clusters lack sites for O2dissociation – leads to H2O2rather than H2O
Ethanol Oxidation• Direct Ethanol Fuel Cells
– Would like to produce CO2 but it is hard to break the C‐C bond
– Mostly produce acetaldehyde and acetic acid• Very complex chemistry
– DEMS– Vibrational spectroscopy– And all kinds of electrochemical tricks– Mechanism is still not completely understood
• We will focus on effects of Pt site size
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
-50
0
50
100
150
200
250
300
Cur
rent
den
sity
(µA c
m-2 ge
o)
Potential (V vs. NHE)
Ethanol oxidation on Ptnano/ITOPtnano 1 vol% EtOH in 0.1 M HClO4
H2OOxid.
1stOxidation
2ndOxidation
PtOxreduction
ReactivationPeak
HER
Surface Oxidized
CVs for EOR on 0.1 ML Ptn/ITO in 0.1 M HClO4 + 1 % EtOH; scan rate: 0.1 V/s.Strong oscillations of activity with size.
Ptnano/ITO
Size dependence of EOR activity
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
-10
0
10
20
-10
0
10
20
30
-10
0
10
20
30
40
Potential (V vs. NHE)
Pt 8 Pt 10 Pt 9 Pt 14
Cur
rent
den
sity
(A
cm
-2 geo) Pt 4 Pt 7
Pt 5 Pt 8 Pt 6
A.
C.
Pt 1 Pt 2 Pt 4
B.
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
-50
0
50
100
150
200
250
300
Cur
rent
den
sity
(µA c
m-2 ge
o)
Potential (V vs. NHE)
Activity vs. Size in CV #13 on each sample
1. Relative intensity of the 3 EOR peaks varies with cycling2. Size dependence is preserved for at least 30 CVs, implying that the
clusters remain intact.
Efficiency Amps/gram PtPtnano Ptn Max Ptn Min
1st Oxidation 46 130 ~02nd Oxidation 72.5 650 300Reactivation 75 450 30
1 2 3 4 5 6 7 8 9 10 11 12 13 140
25
50
75
100
125
150
1. O
x. p
eak
(A/g
Pt)
1. Ox. 2. Ox. Reac. Pt 4d3/2
Platinum cluster (atoms)
200
300
400
500
600
7002.
Ox.
pea
k (A
/gP
t)
0
100
200
300
400
500
Rea
c. p
eak
(A/g
Pt)
334.0
333.5
333.0
332.5
Bin
ding
ene
rgy
Pt 4
d 3/2 (
eV)
Activity vs. Size in CV #13 on each sample
1. Strong fluctuations show that the clusters remain intact.2. Anti‐correlation with Pt 4d core level binding energy 3. No sign of covergence toward bulk‐like properties
331.8 eV for bulk Pt
Anti‐correlations between Pt 4f binding energy and ORR catalysis
Glassy carbon oxidation by water, catalyzed by Ptn
High Pt 4f BE is associated with no carbon oxidation“normal ORR”
2 4 6 8 10
71,3
71,4
71,5
71,6
71,7
71,8
71,9pe
ak p
ositi
on o
f Pt 4
f 7/2 [
eV]
cluster size [atoms]
Bulk Pt = 71.2 eV
Only “normal”behavior “near‐
Normal”
Other examples of anticorrelationsbetween XPS BEs and Oxidation catalysis
Cluster Size (Number of Atoms)0 5 10 15 20 25 D
evia
tion
from
N-S
Cha
rge
Sca
ling
(eV
)
-0.4
-0.2
0.0
0.2
Act
ivity
(CO
2 pe
r TP
R x
109 )
0
2
4
6
8
10 ActivityPd 3d XPS Shift
Single Layer Islands
Growth of 2nd Layer
CO oxidation over Pdn/TiO2(110) under gas‐surface conditions
Other examples of anticorrelationsbetween XPS BEs and Oxidation catalysis
CO oxidation under gas‐surface conditionsAnticorrelationwith Valence band onset and with a particular type of binding site.
Ptn/alumina/Re(0001)
What causes the Anticorrelations?Two perspectives on core level shifts:Initial State Final StateHigh BE means that Pt or Pdis in an electron‐poor (positively charged) state which stabilizes all the electronic orbitals.
Low BE means that Pt or Pdis electron rich, which increases electronic energy due to e‐‐e‐ repulsion
High BEs result from poor screening of the final state core hole. Means that Pt or Pd are electron poor.
Low BE means that core hole is well screened and delocalized. Pt or Pd must have electron rich and polarizable valence shells
Conclusions• Small, monodisperse clusters can have strongly size‐dependent
chemistry– Activities (currents) are strongly size dependent– Peak or onset potentials are weakly size dependent
• Oxidation catalysis activity is dependent on electron rich clusters – rate limiting step is bond activation in water, EtOH
• H2O2/H2O branching in ORR may be controlled by the size of the available sites for O2 binding/dissociation
• Can see ORR, OER, HER, EOR, HPOR even on very small clusters –one to two atoms!
• For small clusters – need to avoid adventitous adsorbates to see inherent size dependence
• Mass activities are substantially higher than for Ptnano