Serra Maia 2018

Subscriber access provided by Kaohsiung Medical University
Energy, Environmental, and Catalysis Applications

Mechanism and Kinetics of Hydrogen Peroxide
Decomposition on Platinum Nanocatalysts
Rui Serra-Maia, Marion Bellier, Stephen Chastka, Kevin Tranhuu, Andrew Subowo,
J. Donald Rimstidt, Pavel M. Usov, Amanda J Morris, and Frederick Marc Michel
ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b02345 • Publication Date (Web): 31 May 2018
Downloaded from http://pubs.acs.org on May 31, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 37 ACS Applied Materials & Interfaces
1
2
3
4 Mechanism and Kinetics of Hydrogen Peroxide
5
6
7 Decomposition on Platinum Nanocatalysts
8
9
10
11
12 Rui Serra-MaiaƗ, Marion BellierƗ, Stephen Chastka Ɨ, Kevin
13
14
15 TranhuuƗ, Andrew SubowoƗ, J. Donald Rimstidt Ɨ, Pavel M. Usov¥,
16
17
18 Amanda J. Morris¥, F. Marc MichelƗ*
19
20
21
Ɨ
22 Department of Geosciences, Virginia Polytechnic Institute and State University, Blacksburg VA
23 24061 USA
24 ¥
Department of Chemistry, Virginia Polytechnic Institute and State University, Blacksburg VA 24061
25
26 USA
27
28
29
30 *
31 Corresponding author: mfrede2@vt.edu
32
33
34
35
36
37 Keywords: hydrogen peroxide, catalysis, H2O2 disproportionation, surface chemistry,
38
39 particle size
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 1
59
60 ACS Paragon Plus Environment
ACS Applied Materials & Interfaces Page 2 of 37
1
2
3
4 Abstract
5
6 The decomposition of H2O2 to H2O and O2 catalyzed by platinum nanocatalysts controls the
7
8 energy yield of several energy conversion technologies, such as hydrogen fuel cells. However,
9
10 the reaction mechanism and rate limiting step of this reaction have been unsolved for more than
11
12 100 years. We determined both the reaction mechanism and rate limiting step by studying the
13
14 effect of different reaction conditions, nanoparticle size and surface composition on the rates of
15
16 H2O2 decomposition by three platinum nanocatalysts with average particle sizes of 3 nm, 11 nm
17
18 and 22 nm.
19
20 Rate models indicate that the reaction pathway of H2O2 decomposition is similar for all three
21
22
nanocatalysts. Larger particle size correlates with lower activation energy and enhanced
23
24
25
catalytic activity, explained by a smaller work function for larger platinum particles, which favors
26
27 chemisorption of oxygen onto platinum to form Pt(O). Our experiments also showed that
28
29 incorporation of oxygen at the nanocatalyst surface results in a faster reaction rate because the
30
31 rate limiting step is skipped in the first cycle of reaction. Taken together these results indicate
32
33 that the reaction proceeds in two cyclic steps and that step 1 is the rate limiting step.
34
35
36 Step 1 + → + ()
37 Step 2
() + →
+ +
38 Overall 2 → + 2
39
40 Establishing relationships between the properties of commercial nanocatalysts and their
41
42 catalytic activity, as we have done here for platinum in the decomposition of H2O2, opens the
43
44
possibility of improving the performance of nanocatalysts used in applications. This study also
45
46
47
demonstrates the advantage of combining detailed characterization and systematic reactivity
48
49 experiments to understand property-behavior relationships.
50
51
52
53
54
55
56
57
58 2
59
1
2
3
4 1 Introduction
5
6 The decomposition of hydrogen peroxide (H2O2) catalyzed by platinum (Pt) through a non-
7
8 electrochemical pathway (without addition of electrons from an anode) is an important reaction
9
10 in many energy conversion applications ranging from miniaturized spacecraft to underwater
11
12 vehicles1-3. This reaction is also important in the oxygen reduction reaction (ORR) in hydrogen
13
14 fuel cells, where it reduces efficiency by acting as a parasitic parallel pathway of the fully
15
16 electrochemical conversion of O2 to H2O4-7 (Figure S1). Despite its importance, the reaction
17
18 pathway of H2O2 decomposition on both supported and colloidal Pt catalysts has been an
19
20 unresolved question for more than 100 years8-14. This is due to a general lack of systematic
21
22
studies that use multivariable analysis to simultaneously understand the effects of both catalyst
23
24
25
properties and reaction conditions. The resulting knowledge gap has prevented scientists from
26
27 explaining the enigmatic variable performance of Pt nanocatalysts in the decomposition of H2O2
28
29 and hindered the design of more efficient catalysts1-2, 15
. Our study aimed to establish how
30
31 surface chemical and structural properties of colloidal forms of two commercial Pt nanocatalysts
32
33 dictate their catalytic activity. For that, we used multivariable analysis to understand the effects
34
35 of five key variables: nanocatalyst size, surface composition, reaction temperature, pH, and
36
37 H2O2 concentration. These results lead to a better understanding of how surface chemical and
38
39 physical properties of Pt and other Pt group element (PGE) catalysts relate to catalytic behavior.
40
41
Several different reaction pathways have been proposed to explain the decomposition of
42
43
H2O2 on PGE surfaces (Table 1)8-14, 16-20.
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 3
59
1
2
3 Table 1. Proposed mechanisms of H2O2 decomposition on platinum group metal
4 surfaces.
5
6 Name/supporting
Mechanism/reaction steps Testable predictions
7 references
Induction period for metallic surfaces11.
8 1) Oxygen-oxide xPt + ½ yO2=PtxOy
Reaction rate does not increase for catalysts
9 mechanism21 PtxOy + yH2O2=xPt + yH2O + yO2
pretreated with H2O2.
10 Surface oxygen concentration of catalysts
11 pretreated with H2O2 depends on relative rates of
2) Peroxide-oxide Pt + H2O2=Pt(O) + H2O step 1 vs step 2 and is independent of 8
12 mechanism4, 5,22 Pt(O) + H2O2=Pt + O2 + H2O (see section 4.3).
13 pH does not have a primary effect in the reaction
14 rate14, 23.
15 Reaction rate is first order in terms of 11.
3) Perhydroxyl H2O2 = O2H- + H+
16 anion-oxide Pt(O) + O2H- = Pt + OH- + O2 Reaction is first order in if all active sites are
17 mechanism11 Pt + O2H- = Pt(O) + OH- covered with oxygen (PtO)11. For lower
coverages reaction order increases11.
18 Lowering catalyst work function increases
19 reaction rate14.
H2O2 + e- = OH- + •OH
20 Hydroxyl radicals are generated10, 24.
H2O2 = O2H- + H+
Open circuit potential matches loss of an electron
21 4) Electrochemical O2H- = •O2H + e-
from platinum25.
22 mechanism10, 14, 23-24 •O2H = O2- + H+
Less conductive surfaces are less catalytically
O2- + H2O2 + OH- + •OH + O2
23 2 •OH = H2O2
active26-27.
Reaction rate does not change for catalysts
24
pretreated with H2O2.
25 H2 + 2Pt = 2Pt-H
26 ½ O2 + Pt = Pt-O
27 Pt+OH- = Pt-(OH)ads + e-
Molecular H2 and O2 are present in solution28.
5) Eley-Rideal Pt-(OH)ads + H2O2 = Pt-(OOH)ads + H2O
28 mechanism19, 28Ɨ Pt-H + H2O2 + e- = Pt + H2O + OH-
Reaction rate is strongly affected by pH when
29 platinum is used as catalystst28-29.
Pt-(OOH)ads + Pt-(H)ads = 2Pt + O2 + H2
30 Pt-O + 2Pt-H = Pt-H2O + 2Pt
Pt-H2O = Pt + H2O
31
32
Ɨ
The study performed in Reference (29) was done under corona discharging conditions.
33
34
35 In most cases, H2O2 decomposition on Pt is first order in terms of H2O2 concentration8-12, 14, 18, 21,
36
37
38
28, 30
although exceptions have been observed for < 10 . Low H2O2 concentrations
39
40 resulted in inconsistent rate measurements and replicability issues, which were attributed to
41
42 insufficient H2O2 to overcome the chemical potential necessary to convert PtOx to Pt8.
43
44 Increasing pH causes an increase in H2O2 decomposition rate to a maximum that was
45
46 determined experimentally to be near pH 10.511, 14, 23. Radical species have been detected when
47
48 the decomposition of H2O2 is catalyzed by certain transition metals13, 23, 31-32, but there currently
49
50 is no evidence for radical species generation on Pt23.
51
52
The catalytic activity of Pt in the decomposition of H2O2 is known to also depend on the
53
54
55
structural and/or surface chemical properties of the catalyst8, 11, 14, 33. Higher concentrations of
56
57
58 4
59
1
2
3 surface oxygen correlate with enhanced H2O2 decomposition rates, but a mechanistic
4
5 understanding for this correlation has not yet been established8, 11, 14, 33
. Higher H2O2
6
7
decomposition rates also match with enhanced cold-working of platinum catalysts. This effect
8
9
10 was initially attributed to an increase in structural dislocations15, 34, but was later ascribed to an
11
12 enhancement of surface oxygen11.
13
14 Particle size is another important factor in Pt reactivity because it influences the ability of
15
16 metal nanoparticles to accept or donate charge35-37. However, this effect has never been studied
17
18 for the decomposition of H2O2. Particle size effects are generally not well understood due to the
19
20 complex interplay of multiple factors7, 37-40. As the size of metal nanoparticles decreases, both
21
22 the work function and the Pauling’s electronegativity increase, which reduces their ability to
23
24 provide charge41. For Pt, the ionization potential varies between 9.0 eV for a single atom and
25
26
5.3 eV for the bulk metal (i.e., the work function)35. With decreasing particle size the ratio of
27
28
29 edges/corners to terraces increases as function of 1/d. These sites are more undercoordinated
30
31 and form stronger bonds in comparison with atoms at terrace sites35. These differences are
32
33 important as they result in variable binding energies of the reactants onto the catalyst surface42.
34
35 The optimum activity is obtained for intermediate binding energy values, which represent a
36
37 compromise between low activation energy for the reaction, and a surface with low reactant
38
39 coverage35, 43.
40
41 This paper reports the rates of H2O2 decomposition as a function of H2O2 concentration, pH
42
43 and temperature by two common Pt nanocatalysts, known as Pt black and Pt nanopowder. Our
44
45
characterization of these samples, which have different average crystal size and shape
46
47
48
characteristics, revealed new information regarding the abundance and speciation of surface
49
50 oxygen [ID am-2018-023446]. The results from this study showed that oxygen at the surfaces is
51
52 primarily chemisorbed with platinum, and that its abundance increases with increasing particle
53
54 size. Rate models obtained for both catalysts were used to infer the reaction pathway of H2O2
55
56 decomposition on Pt nanocatalysts and identify the rate limiting step of reaction, thus providing
57
58 5
59
1
2
3 a molecular-scale understanding of the decomposition mechanism of H2O2 on Pt. The new and
4
5 detailed understanding of the H2O2 decomposition mechanism on Pt nanocatalysts described
6
7
here is crucial for improving current and future catalysts used in energy conversion technologies
8
9
10 that depend on this reaction.
11
12
13
14
15
16 2 Results and Discussion
17
18
19 2.1 Characterization of Pt black and Pt nanopowder
20
21 High-resolution TEM analysis from our previous work showed that Pt black and Pt
22
23 nanopowder consist of nanosized Pt particles [ID am-2018-023446]. Dilute suspensions of the
24
25 samples resulted in small aggregates and isolated crystallites (nanoparticles) when deposited
26
27 on continuous film SiN TEM grids (Figure S2). Average values for crystal size, specific surface
28
29 area, aspect ratio, lattice strain and surface O/Pt ratio of the two nanocatalysts are given in
30
31
Table 2. The elongation observed on Pt black crystallites occurs along random directions and
32
33
34 no particular crystallographic plane is preferentially exposed for either catalyst. When dispersed
35
36 in aqueous solutions the crystallites of both catalyst samples are partly aggregated and the
37
38 extent of aggregation is similar for both [ID am-2018-023446].
39
40
41
42 Table 2. Physical properties of Pt black and Pt
43 nanopowder. Uncertainties shown in parentheses.
44
45 Pt black Pt nanopowder
46
Crystal size (XRD
47 Rietveld), nm
11.31(4) 22.70(4)
48
49 Specific surface area
2 -1 35.5(4) 10.4(9)
50 (BET), m ·g
51 Aspect ratio (TEM) 1.9(8) 1.1(2)
52
53 Lattice strain,
-0.09 % -0.02 %
54 Reference to bulk Pt
55 O/Pt ratio (XPS) 0.14(1) 0.26(5)
56
57
58 6
59
1
2
3
4
2.2 Catalytic activity of Pt black, Pt nanopowder and Pt-Liu-3nm
5 The rate equations reported here are intended to model the rate of H2O2 decomposition as a
6
7
function of different , pH, and temperature (see section 4.2.1 for derivation of rate model).
8
9
10 These variables are known to affect the rate of H2O2 decomposition on Pt8, 10-11, 13-14. Previous
11
12 work by Liu et al. (2014) on colloidal suspensions of 3 nm Pt nanoparticles (Pt-Liu-3nm)13 were
13
14 analyzed along with the results obtained in our study. For Pt nanopowder there are 36 data
15
16 points, for Pt black there are 30 data points and for Pt-Liu-3nm there are 6 data points. Table 3
17
18 gives the notation with units used in this paper. Data from rate measurements performed for Pt
19
20 black, Pt nanopowder and Pt-Liu-3nm as function of , pH, and temperature are given in
21
22
Table S1.
23
24
Table 3 Notation used
25
26
Rate of H2O2 decomposition, !/ ∙ $
27
28 Concentration of H2O2, !/%
29
Time, $&'
30
31 Pre-exponential constant in Arrhenius
(
32 equation, !/ ∙ $
33
)* Activation energy, +/ !
34
35 Gas constant, 8.314 +/ ! ∙ ,
36
- Temperature, ,
37
.
+
38 Activity of H ion
39 Fitted parameters from regression
40 , 0, ', 1
model
41
42 2 Rate constant, !/ ∙ $
43 34 Reaction order for species 5
44
45 Coefficient of determination for

46 regression model
47 1 Crystal size (diameter),
48
49
() Chemisorbed oxygen on Pt
50 Average fraction of surface covered
(67()
51 with Pt(O)
52
(67 Average fraction of metallic surface
53
54 8 (67() /(67
55
56 94 Rate of reaction step 5
57
58 7
59
1
2
3 The catalyzed rate of H2O2 decomposition for each of the three catalysts is shown in Figure
4
5 1.
6
7
8
9
10 d=22.7 nm
-4 R² = 0.91
11
12
13 Measured Log R
d=11.3 nm
14 -5 R² = 0.94
15
16 d=3 nm
R² = 0.97
17
18 -6
19 Pt nanopowder
20 Pt black
21 Pt-Liu-3nm
22 -7
23 -6.75 -5.75 -4.75 -3.75
24 Predicted Log R
25
26
27 Figure 1. Regression model of log measured : vs log
28 predicted : as function of ; , pH and .
<
=
29
30
31
>
32 The rate of H2O2 decomposition as function of , . , and on Pt nanopowder is given
?
33
34
by
35
36 (C±E)>FG >
37 = 10.B> (& H
∙
? )( F.I±F.FJ )(F.FK±F.F>
. )
38
39
On Pt black is given by
40
41
(EE±E)>FG >
42 = 10E.C (& H
∙
? )( F.IJ±F.FC )(F.F±F.F>
. )
43
44
45 And on Pt-Liu-3nm is given by
46
F.F±F.F>
47 = 10E.E (>.FB±F.>

)( . )
48
49 The regression model indicates that the rate of H2O2 decomposition follows first order
50
51 kinetics in terms of H2O2 concentration for all three nanocatalysts. Any differences in the order
52
53
of reaction in terms of H2O2 concentration are not statistically significant (p-value = 0.48). This
54
55
agrees with previous studies8-12, 14, 21, 28, 30
and indicates that in the rate limiting step of the
56
57
58 8
59
1
2
3 reaction one molecule of H2O2 reacts with the active site44. The effect of pH is relatively small
4
5 compared to , but is statistically significant for Pt black, Pt nanopowder, and Pt-Liu-3nm (p-
6
7
8 value = 0.048, p-value < 0.001, and p-value=0.038, respectively) and similar for all three (p-
9
10 value = 0.10). Contrary to these results, some previous studies have suggested that pH has a
11
12 strong effect on the rate of H2O2 decomposition by Pt group metal catalysts8, 10, 14. However, this
13
14 disparity may be simply a consequence of plotting the rate of H2O2 decomposition on a linear
15
16 scale against a log-scale for pH. In any case, the effect of pH on catalytic activity is smaller than
17
18 one log-unit over the range of 6 pH (log-scale) units for all three catalyst samples.
19
20 The small pH effect on the H2O2 decomposition rate is inconsistent with the perhydroxyl
21
22
anion-oxide and the Eley-Rideal mechanisms (Table 1). In the case of the perhydroxyl anion-
23
24
-
25 oxide mechanism, if HO2 were to be the only active species that interacts with the catalyst
26
27 surface, the decomposition of H2O2 should be first order in terms of (directly correlated
28
29
with . ), which is not supported by our results. The Eley-Rideal mechanism, first proposed by
30
31
32
Mededovic et al. (2006), depends on molecular hydrogen and oxygen formed via high-voltage
33
34 aqueous-phase pulsed corona electrical discharge on Pt electrodes. Computational28 and
35
36 experimental29 studies at analogous conditions showed that the H2O2 decomposition rate
37
38 increases several orders of magnitude in the pH range 3.5 to 10.5. Unfortunately, this
39
40 mechanism has been inadvertently used to explain the decomposition of H2O2 on Pt surfaces
41
42 that were not actually subjected to electrical discharge conditions1, 3, 19. Thus, the small pH effect
43
44 on the H2O2 decomposition rate by Pt catalysts reported here indicates that the Eley-Rideal
45
46 mechanism is not suitable to describe the decomposition of H2O2 on Pt surfaces that are not
47
48
subjected to electrical discharges28-29.
49
50
51
The effect of pH alone is not sufficient to reject the remaining three possible mechanisms
52
53 (oxygen-oxide, peroxide-oxide and electrochemical)8-9, 13-14
. In the case of the mechanisms
54
55 involving surface oxides (oxygen-oxide and peroxide-oxide mechanisms), the effect of pH is
56
57
58 9
59
1
2
3 indirect as neither H+ nor OH- ions are directly involved. Bianchi et al. (1961) proposed that the
4
5 lower rates of H2O2 decomposition on Pt surfaces at low pH values are caused by surface
6
7
poisoning effects as H+ ions sorb onto metal surfaces to fulfill their electron deficiency, thus
8
9
10 preventing the formation of Pt(O)8, 18. This idea was corroborated by Ou et al. (2009) who used
11
12 first principle calculations to show that the chemisorption of O2 on Pt (111) is slightly inhibited in
13
14 the presence of hydrogen ions near the surface45. The known inhibition of H2O2 decomposition
15
16 caused by Cl- ions is also consistent with this explanation. Arruda et al. (2008) showed that Cl-
17
18 ions chemisorb to the surface of Pt nanoparticles (1-2 nm diameter), which is in direct
19
20 competition with chemisorbed oxygen46. In the electrochemical mechanism, pH directly affects
21
22 -
23 the concentration of H+ and the balance between HO2 and H2O2, all of which are the active
24
25 species that react with Pt at different stages. Based on the dissociation constants of H2O2 and
26
27 -
HO2 , Weiss (1935) showed that it was possible to predict explicitly that the rate of H2O2
28
29
30
decomposition reaches its maximum at pH~10.5 (see derivation on Mckee et al., 1969)10, 14, 24.
31
32 This is consistent with the fact that experimental values for the pH corresponding to maximum
33
34 rate of H2O2 decomposition are not significantly affected by catalyst composition14, 47
. In this
35
36 mechanism the initial dissociation of the H2O2 molecule is thought to occur due to electron
37
38 donation from the metal. According to Weiss (1935), H2O2 acts as the oxidizing agent to the
39
40 -
metal and HO2 acts as reducing agent10.
41
42
43 In summary, the effect of pH is sufficient to exclude the perhydroxyl anion-oxide and the
44
45 Eley-Rideal mechanisms, but is consistent with the oxygen-oxide, peroxide-oxide, and
46
47 electrochemical mechanisms. As discussed in sections 2.3 and 2.4, we found that Pt
48
49 nanocatalysts with larger particle sizes and/or higher surface chemisorbed oxygen correlate with
50
51 enhanced catalytic activity. Those results are not consistent with the electrochemical or the
52
53 oxygen-oxide mechanisms and support that the H2O2 decomposition on Pt nanocatalysts occurs
54
55 via the peroxide-oxide mechanism.
56
57
58 10
59
1
2
3
4
2.3 Effect of particle (crystallite) size on catalytic activity
5 For the range of conditions studied, the rate of H2O2 decomposition decreases with particle
6
7 size in the order of Pt nanopowder (d=22.7 nm) > Pt black (d=11.3 nm) > Pt-Liu-3nm (d=3 nm).
8
9
The catalytic activity of Pt nanopowder normalized to specific surface area is approximately 10
10
11
12 times greater than Pt black (Figure 1), which is consistent with the 21% lower activation energy
13
14 (see section 4.2.1) of Pt nanopowder compared to Pt black (p-value=0.039). The catalytic
15
16 activity of Pt-Liu-3nm is approximately six times lower than Pt black. The difference in activation
17
18 energy could not be determined since the Liu et al. (2014) experiments were only done at a
19
20 single temperature13. Changes in particle size could lead to differences in surface reactivity if
21
22 certain surface sites become preferentially exposed. However, combined Rietveld and TEM
23
24 analysis suggest that the direction of crystal elongation is random for Pt nanopowder and Pt
25
26 black [ID am-2018-023446]. This means that preferential exposure of specific crystal faces (i.e.,
27
28
active sites) does not explain the 10-fold difference in catalytic activity between Pt nanopowder
29
30
31 and Pt black.
32
33 A correlation between increasing particle size and enhanced catalytic activity has been
34
35 observed previously for Pt nanocatalysts used in hydrocarbon oxidation reactions40, 48 and the
36
37 oxygen reduction reaction37, 49. To evaluate whether catalytic activity is size-dependent for the
38
39 samples in this study, the H2O2 decomposition rates of the three catalysts were regressed
40
41 >
together as function of 1 in addition to , . , and . This regression showed that the
42 ?
43
44 catalytic activity of the three Pt nanocatalysts fall along a unique linear regression curve with R2
45
46 = 0.95 (Figure 2).
47
48
49
50
51
52
53
54
55
56
57
58 11
59
1
2
3
4
5
-4
6 R2=0.95
7
Measured Log R
8
9 -5
10
11
12
13 -6
14 Pt nanopowder
15 Pt black
16 Pt-Liu-3nm
17 -7
18 -6.75 -5.75 -4.75 -3.75
19 Predicted Log R
20
21 Figure 2. Regression model of log measured : vs
22 log predicted : as function of ; , pH and 1/T and
23 1/d.
24
25
26
>
27 The rate of H2O2 decomposition as a function of , . , , and 1 normalized to
?
28
29
30
surface area for the three nanocatalysts regressed together is given by
31
B±×>FG >
32 = 10.F (& H
∙
? )(' >.F>±F.F )(F.F±F.F>
. )(d>.FK±F.F )
33
34
35 The effect of pH and observed for the three catalysts together is similar to the effect
36
37 observed separately for each catalyst. Similarly, the activation energy is intermediate to the
38
39 values for Pt black and Pt nanopowder (see section 4.2.1). Nanoparticle size resulted in first
40
41 order kinetic effects on the specific catalytic activity of the three catalysts. Several different
42
43 explanations for this behavior have been put forward. Early studies pointed out that the average
44
45 coordination number of the surface atoms decreases with decreasing particle size. Calculations
46
47
on perfect or quasi-perfect particle shapes showed that the proportions of low coordination
48
49
50
number (edge and corner) atoms decreased and those of high coordination number atoms (in
51
52 low-index planes) increased as size increased40, 50
. The ratio of edges/corners to terraces
53
54 increases proportionally to the inverse of the diameter40. Because atoms at the edges and
55
56 corners are more undercoordinated than atoms at (planar) facets, the binding energy between
57
58 12
59
1
2
3 reactants and catalyst for these sites shifts away from the peak of the so-called volcano plot7, 37-
4
5 39, 51
, thus causing a reduction in reaction rates42. Alternative explanations attribute size-induced
6
7
changes in behavior to variation in the electronic properties of surface atoms, such as formation
8
9
10 of surface oxides48.
11
12
13
14 More recently, the effect of the position and shape of the valence band on the work function
15
16 of metals has been used to explain enhanced catalytic activity of metal nanoparticles as a
17
18 function of particle size52. The Fermi level increases for larger particles, resulting in an effective
19
20 reduction of the work function35, 52
. For metals, the work function varies linearly with the
21
22 reciprocal of particle diameter in the range of particle sizes studied here53, which is consistent
23
24 with the observed correlation between catalytic activity and size (1). As a result of their lower
25
26
work function, larger Pt nanoparticles have a greater capacity to provide electron density to the
27
28
29 reagent, which happens in the first step of the peroxide-oxide mechanism35, 53
. This is
30
31 corroborated further by the difference between the activation energies of Pt black and Pt
32
33 nanopowder. The activation energy of the spontaneous decomposition of H2O2 in bulk solution
34
35 is about 75 kJ/mol54. We found that the catalytic effect of the samples reduced the activation
36
37 energy to 26±3 kJ/mol for Pt nanopowder and 33±3 kJ/mol for Pt black. A correlation between
38
39 particle size and the Fermi level (which correlates inversely with work function) is known to
40
41 reduce the activation energy for the chemisorption of oxygen on larger Pt particles55-56. This is
42
43 consistent with a rate limiting step that involves oxidation of metallic Pt sites to Pt(O) because
44
45
the lower work function allows the chemisorption of oxygen to occur faster on larger Pt
46
47
48 nanoparticles. Zhou et al. 2012 showed that the difference in the work function of particles with
49
50 sizes similar to Pt nanopowder and Pt black could be up to 0.05 eV for isolated spheres or 0.02
51
52 eV for aggregated nanoparticles, where the smaller nanoparticles have the larger work function
53
54 in both cases53.
55
56
57
58 13
59
1
2
3 Considering the small difference between the activation energies of Pt black and Pt
4
5 nanopowder, it appears unlikely that the effect of size is the only factor causing their 10-fold
6
7
difference in the rate of H2O2 decomposition. This effect is well fitted in the multi-variate linear
8
9
10 regression shown in Figure 2, but only three different particle sizes were considered and the
11
12 effect may be subject to a large uncertainty. This is supported by the results shown in Figure 3
13
14 and Figure 4, which suggest that surface chemisorbed oxygen also has a significant impact in
15
16 the catalytic activity of Pt nanocatalysts in the decomposition of H2O2. Thus, we have also
17
18 considered the effect of surface oxygen on the catalytic activity of Pt black and Pt nanopowder.
19
20
21
22
23 2.4 Effect of chemisorbed oxygen on catalytic activity
24
25 Rate measurements were performed on samples subjected to preheating in order to better
26
27 understand the relationship between surface oxygen content and catalytic activity (Table S2).
28
29 Rietveld analysis showed that heating Pt black and Pt nanopowder at 240 oC for 24h in
30
31
atmospheric conditions did not change their crystal sizes (Table 4 and Table 2). In contrast, the
32
33
O/Pt ratio of Pt black-240 increased approximately 30% (p-value=0.02) while the O/Pt ratio of Pt
34
35
36 nanopowder-240 did not change (p-value=0.17).
37
38
39 Table 4. Rietveld analysis of Pt nanopowder
40 and Pt black heated at 240 oC for 24h.
41
42
43 Pt black-240 Pt nanopowder-240
44 Crystal size (XRD
10.20±0.08 19.89±0.07
45 Rietveld), nm
46
Lattice strain,
47 (XRD Rietveld) -0.04 % -0.02 %
48 reference to bulk Pt
49
50 O/Pt ratio (XPS) 0.18±0.07 0.24±0.08
51
52
53 Rate measurements of the preheated samples show that the H2O2 decomposition reaction is
54
55 faster on Pt nanocatalysts having greater surface oxygen content. The rate of H2O2
56
57
58 14
59
1
2
3 decomposition by Pt black-240 is approximately 2 times faster than Pt black in the as supplied
4
5 state in the range of studied conditions (Figure 3b). In comparison, the rate H2O2 decomposition
6
7
by Pt nanopowder-240 did not change compared to as-received Pt nanopowder (Figure 3a).
8
9
10
11 a
-4
12
13 R² = 0.95
14 -4.5
Measured Log R
15
16
17 -5 R² = 0.91
18
19
20 -5.5
Pt nanopowder
21
22 Pt nanopowder-240
23 -6
24 -6.75 -5.75 -4.75
25 Predicted Log R
26
27 -4.5 b
28
29 R² = 0.99
-5
30
Measured Log R
31
32 -5.5
33 R² = 0.94
34 -6
35
36 Pt black
-6.5
37
38 Pt black-240
39 -7
40 -6.75 -5.75 -4.75
41 Predicted Log R
42
43
44 Figure 3. (a) Comparison of specific catalytic activity
45 of Pt nanopowder-240 with Pt nanopowder as
function of ; , pH and 1/T. (b) Comparison of
46
specific catalytic activity of Pt black-240 with Pt black
47 as function of ; , pH and 1/T.
48
49
50
51
The observed relation between increased surface oxygen content and enhanced catalytic
52
53
54 activity is not consistent with an electrochemical mechanism because oxidized surfaces are
55
56 characterized by larger work function values26-27. This contradicts previous works on Pt that
57
58 15
59
1
2
3 used electrochemical studies to claim that lowering the work function favors the rate of H2O2
4
5 decomposition and used that relationship to support the electrochemical mechanism (Table 1)10,
6
7 14
. The effect observed in those studies is likely explained by the redox activity of H2O2. When
8
9
10 anodic potentials are applied the electrochemical decomposition of H2O2 driven by the external
11
12 supply of electrons outweighs the catalytic effect of Pt. If the electrochemical mechanism truly
13
14 represented this reaction, increasing surface Pt oxidation would reduce the number of active
15
16 sites capable of exchanging electrons with H2O2 and consequently reduce the overall reaction
17
18 rate, which is not observed in our results or in previous studies11, 14
. Furthermore, the
19
20 electrochemical mechanism would require the formation of hydroxyl radical species during the
21
22 decomposition of H2O2. Hydroxyl radical generation has been suggested as a mechanism for
23
24 the decomposition of H2O2 on Pt13, 23, 31-32
, although direct experimental evidence for this is
25
26
lacking in this and in previous studies23 (S.I. Section A).
27
28
29 The correlation between enhanced H2O2 decomposition rate and higher surface chemisorbed
30
31 oxygen content supports that the rate limiting step of the reaction involves the oxidation of
32
33 metallic Pt to Pt(O). The oxidation potential of Pt to Pt(O) observed by cyclic voltammetry for
34
35 both samples at pH ~ 0 and pH ~ 6 is similar (Figure S3). This suggests that this reaction does
36
37 not depend on pH and is consistent with the trivial effect of pH on the rates of H2O2
38
39 decomposition for both catalysts. Thus, the reaction mechanism may be represented by
40
41 reactions 15-17. Reaction 15 (bold) is the rate-limiting step of the reaction.
42
43 Step 1 + → + () (1)
44
45 Step 2
() + →
+ + (2)
46 Overall 2 → + 2 (3)
47
48
49 The presence of initially larger amounts of surface Pt(O) results in a mechanism of H2O2
50
51 decomposition where the oxidation of Pt to Pt(O) is skipped in the first cycle for the oxidized
52
53 sites. Because H2O2 is also decomposed in reaction 16, and our rate measurements were
54
55
performed through the initial rate method (Figure S4), an increase in the initial number of Pt(O)
56
57
58 16
59
1
2
3 sites (i.e., sites that skip the rate limiting step in their first cycle) results in a faster measured rate
4
5 NOPQ(R) S
6 of H2O2 decomposition until the steady state of TOPQ U
is reached at the catalyst surface. This
7
8 explains why Pt nanocatalysts with higher initial surface Pt(O) concentration exhibit enhanced
9
10
catalytic activity in the decomposition of H2O2.
11
12
13 To support this mechanism, the effect of surface oxidation on the catalytic activity of Pt black
14
15 and Pt nanopowder was also evaluated by pretreating Pt black and Pt nanopowder in the as-
16
VWX VWX
17 received state with different H2O2 concentrations ( ) in the range of 9.8 × 10 < <
18
19 9.8 × 10> before rate measurements (Table S3). The catalytic activity of both Pt black and Pt
20
21 VWX
nanopowder subjected to H2O2 pretreatment was similar, regardless of value being tested
22
23
24 (p-value=0.12 and p-value=0.051, respectively). Interestingly, the catalytic activity of Pt
25
26 nanopowder pretreated with concentrated H2O2 was similar to Pt nanopowder in the as supplied
27
28 form (p-value=0.267) (Figure 4a). However, the catalytic activity of Pt black pretreated with
29
30 concentrated H2O2 was approximately 3 times higher than Pt black in the as supplied form (p-
31
32 value=0.03) (Figure 4b).
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 17
59
a
1
2 -4 R² = 0.91
3
Measured Log R
4
5 -4.5
6
7 -5
8
9
10 -5.5
Pt nanopowder
11
12 Pt nanopowder-
-6
13
-6.75 -5.75 -4.75
14
15 Predicted Log R
16 -4.5
17 b
18
19 -5
20
Measured Log R
21
-5.5
22
23 R² = 0.94
24 -6
25
26
27 -6.5 Pt black
28 Pt black-
29
-7
30
-6.75 -5.75 -4.75
31
Predicted Log R
32
33
34 Figure 4. Comparison between the rate of H2O2
35
decomposition of a) Pt nanopowder in the as
36
37 supplied form and pretreated with concentrated
38 H2O2 and b) Pt black in the as supplied form and Pt
39 black pretreated with concentrated H2O2.
40
41
42
43 The oxygen-oxide and the electrochemical mechanisms are both inconsistent with the
44
45 enhanced rates of H2O2 decomposition obtained for Pt black pretreated with H2O2. In the
46
47 oxygen-oxide mechanism, the catalytic activity of H2O2 pretreated Pt surfaces is expected to
48
49 decrease due to an increase in the number of metallic sites generated during step 2 of the
50
51 reaction with H2O2 (Table 1). The addition of this step would result in longer induction periods,
52
53
i.e., slower measured reaction rates. In the electrochemical mechanism, pretreating the catalyst
54
55
56 surface with H2O2 should not have any effect on the catalyst properties and the catalytic activity
57
58 18
59
1
2
3 should be similar. As H2O2 is an oxidizing compound, surface oxidation is plausible. However,
4
5 as discussed above, in the electrochemical mechanism, increased surface oxidation should
6
7
result in a reduction in overall catalytic activity, which is contradicted by our results.
8
9
10 The peroxide-oxide mechanism is consistent with the increase in catalytic activity observed
11
12 for Pt black pretreated with H2O2, as well as the similar catalytic activity obtained for Pt
13
14 nanopowder. This is demonstrated by deriving the ratio of oxidized Pt surface (8) in the
15
16 presence of H2O2 for this mechanism. From reactions 15-17, in steady state
17
18 9>K = 9>C (4)
19
20 2>K T(67 UT U = 2>C N(67() ST U (5)
21
22
NOPQ(R) S [\]
23
TOPQ U
=8= (6)
[\^
24
25
NOPQ(R) S
26 The ratio of at the surface of Pt is only dependent on the ratio of the rate constants of
TOPQ U
27
28
29 reactions (15) and (16), and is totally independent from H2O2 concentration (reaction 20). As
30
31 such, the concentration of H2O2 used in the pretreatment does not affect the surface Pt(O)
32
VWX
33 concentration, which explains the similar effect of different on the catalytic activity of each
34
35 nanocatalyst. This is further supported by previous studies showing that the open circuit
36
37 potential of Pt electrodes immersed in H2O2 solutions is independent from the concentration of
38
39
H2O218. As discussed in section 2.3, the oxidation step of Pt Pt(O) is favored for larger Pt
40
41 [
42 nanoparticles and so one important consequence from equation 20 is that _ \ ` >
[ 67 a*abVbcdXW
43
44
[ NOPQ(R) S
45 _ \` , which results in 867 a*abVbcdXW > 867 fg*h[ . Thus, in the steady state, the TOPQ U
ratio
[ 67 fg*h[
46
47
48 is higher for Pt nanopowder pretreated with concentrated H2O2 than for Pt black. This result is
49
50 consistent with the greater catalytic activity observed for Pt nanopowder compared to Pt black
51
52 when both were pretreated with H2O2 (Figure 4). In support of this hypothesis, we note that the
53
54 total number of surface Pt atoms in an experiment involving Pt black is approximately 2x1019.
55
56
57
58 19
59
1
2
3 For the lowest concentration of H2O2 studied ( = 100 i), the total number of H2O2
4
5
NOPQ(R) S
6 molecules is 6x1019. Bianchi et al. (1962) showed that surface TOPQ U
on bulk Pt ranged between
7
8
9 0.54 – 0.588. Assuming a 8 value in the same range for Pt black, on average each active site
10
11 only has to go through three complete oxidation-reduction cycles (two molecules of H2O2 are
12
13 decomposed per oxidation-reduction cycle) to decompose all of the H2O2. Increasing the initial
14
15 NOPQ(R) S
TOPQ U
ratio would result in a significant increase in the initial rate of H2O2 decomposition
16
17
18 because the rate limiting step of the reaction is skipped in Pt(O) sites. However, as shown in the
19
20 present study, the effect of H2O2 pretreatment on Pt nanopowder was negligible, which
21
22 suggests that the Pt(O) coverage on as received Pt nanopowder is close to the Pt(O) coverage
23
24 that results from pretreating Pt nanopowder with H2O2.
25
26 Although limited in number, theoretical studies of H2O2 decomposition on platinum are in
27
28
general agreement with the peroxide-oxide mechanism57-58. Ab initio calculations indicate that
29
30
31 when hydrogen peroxide comes in contact with surface Pt atoms it becomes unstable and
32
33 dissociates into two hydroxyls or a water molecule and atomic oxygen51, 58. If two hydroxyls are
34
35 initially formed, one hydrogen atom is transferred to the adjacent hydroxyl to form H2O and a
36
37 surface Pt(O)57. To our knowledge, no studies have considered the subsequent interaction of
38
39 Pt(O) with more H2O2. Oxygen chemisorption studies show that increased chemisorbed oxygen
40
41 concentration leads to enhanced electrostatic repulsion, which hinders the chemisorption of
42
43 more oxygen59. This supports the idea that H2O2 molecules are more prone to react with Pt(O)
44
45 sites to regenerate their metallic character, instead of alternative pathways involving the
46
47
generation of adjacent surface oxidized sites from adsorbed O2.
48
49
50
Multiple factors may be important for explaining the observed increase in the amount of
51
52 oxygen at the surface of Pt nanopowder compared to Pt black in the as received condition and
53
54 after pretreatment with H2O2. Pt nanopowder in the as supplied state incorporates more surface
55
56 oxygen than Pt black, which is consistent with an enhanced stability of chemisorbed oxygen on
57
58 20
59
1
2
3 larger catalysts. In addition to size, the differences in synthesis conditions for these two
4
5 catalysts may be important. Synthesis of Pt black involves hydrogenation in aqueous
6
7
suspension at 450 oC. These conditions are known to chemically reduce the surface of metal
8
9
10 nanoparticles to levels below equilibrium with atmospheric oxygen60-61. Residual surface oxygen
11
12 after hydrogenation is likely strongly bonded to Pt, presumably in the form of PtO or PtO2.
13
14 However, these phases are known to be inert species in the decomposition of H2O28. Pt
15
16 nanopowder, in comparison, is synthesized by chemical vapor condensation, which involves
17
18 temperatures in the range of 700 oC – 800 oC. These synthesis conditions favor enhanced
19
20 accumulation of chemisorbed oxygen at the surface39.
21
22 Our results provide compelling evidence that the decomposition of H2O2 on Pt surfaces
23
24 proceeds via peroxide-oxide mechanism. This mechanism, with the rate limiting step involving
25
26
Pt → Pt(O), is also consistent with the fact that some previous studies reported an induction
27
28
29 period for the decomposition of H2O2 on Pt. The catalysts in these studies were subjected to
30
31 reducing conditions in order to obtain nearly perfect metallic surfaces10-11. Under such conditions
32
33 the initial reaction rate is the rate limiting step, as virtually all active sites must go through the
34
35 NOPQ(R) S
oxidation of Pt → Pt(O). In this case the rate constant will increase until surface TOPQ U
reaches
36
37
38 NOPQ(R) S
steady state. Therefore surfaces that originally contain a lower than steady state TOPQ U
exhibit
39
40
41 an induction period after exposure to H2O2. This is corroborated by another study that found that
42
43 the rate constant of a freshly cleaved platinum surface increased asymptotically with time of
44
45 contact with hydrogen peroxide15. Thus, the induction period depends on the surface chemical
46
47 properties of the original catalyst surface. The majority of studies have used different catalysts,
48
49 pretreatments and reaction conditions, all of which play a role in determining surface properties.
50
51
This explains many of the inconsistencies between different studies, which resulted in a lack of
52
53
54 scientific consensus regarding the mechanism of H2O2 decomposition11, 14.
55
56
57
58 21
59
1
2
3
4 3 Conclusions
5
6 The decomposition of hydrogen peroxide on Pt nanocatalysts occurs via the peroxide-oxide
7
8 mechanism. In the first step of this cyclic mechanism H2O2 reacts with the surface of Pt
9
10 nanocatalysts to form Pt(O) releasing one molecule of H2O. In the second step, a second
11
12 molecule of H2O2 reduces Pt(O) to metallic Pt releasing a second molecule of H2O and O2.
13
14 Particles that incorporate more Pt(O) at the surface exhibit a faster rate of H2O2 decomposition
15
16 because the rate limiting step of the reaction is skipped in the first cycle. The catalytic activity is
17
18 enhanced for larger Pt nanoparticles due to their lower work function, which increases the rate
19
20 at which surface Pt atoms react with H2O2 to form Pt(O) (rate limiting step). For the samples in
21
22
this study, the enhanced catalytic activity of Pt nanopowder compared to Pt black results from
23
24
25
the combined effect of higher surface chemisorbed oxygen and lower activation energy due to
26
27 its larger particle size. The reaction mechanism established here rationalizes previous studies
28
29 that reported inconsistent results in terms of the induction periods of Pt nanocatalysts in the
30
31 decomposition of H2O2. The multivariate analysis utilized in this work highlights the integrated
32
33 effect of particle size, surface chemistry and the reaction conditions in the catalytic activity of Pt
34
35 nanocatalysts in the decomposition of H2O2. This approach presents a pathway for unveiling the
36
37 mechanism of other chemical reactions at the atomic level in order to improve the performance
38
39 of platinum or other metal nanoparticles used to catalyze them.
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 22
59
1
2
3
4 4 Materials and Methods
5
6
7 4.1 Platinum black and platinum nanopowder
8
Pt nanopowder is synthesized by combustion chemical vapor condensation and the particle
9
10
11
size reported by the manufacturer is ≤ 50 nm. Pt black is derived from high temperature
12
13 reduction of H2PtCl6 to metallic Pt60-61 and has an average aggregate size reported by the
14
15 manufacturer of ≤ 20 µm. Pt black aggregates are composed of individual nanoparticles,
16
17 despite occurring as micrometer-sized aggregates in the form of dry powder [ID am-2018-
18
19 023446]. Pt black and Pt nanopowder are characterized by different average crystal size,
20
21 shape, and surface oxygen concentration. Average values for crystal size, specific surface area,
22
23 aspect ratio, lattice strain and surface O/Pt ratio of the two materials are given in Table 2.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 23
59
1
2
3
4
4.2 Rate measurements
5 Hydrogen peroxide decomposition rate measurements were performed in well-stirred 1 L
6
7
Erlenmeyer flasks. Data were collected over the concentration range of 100 µM < < 1000
8
9
10 µM, the pH range of 2.7 < pH < 9.3 and the temperature range of 0 oC < T < 55 oC. For
11
12 experiments carried at near 0 oC the reactor was immersed in an ice bath and temperature was
13
14 allowed to equilibrate before initiating the experiments. For experiments carried at temperatures
15
16 above room temperature the reactor was heated directly on a heating-stir plate. Experiments
17
18 were conducted when the temperature variation was smaller than 0.4 oC / min. The temperature
19
20 of the H2O2 solution was recorded at the start and at the end of each run, and the average of
21
22 both measurements was used for data analysis.
23
24
For each run 6 mg of Pt black or Pt nanopowder were dispersed in 5 mL of H2O at pH 11 and
25
26
sonicated for 1 minute to enhance particle dispersion. This dispersion was added to the
27
28
29 Erlenmeyer flask containing 1 L of H2O2 solution previously prepared at the concentration of
30
31 H2O2, pH and temperature conditions specific to each run. The solution was sampled at 10 sec,
32
33 1, 2, 3, 4 and 5 mins. Each sample consists of a 1 mL aliquot that was filtered through a 0.22
34
35 µm pore size syringe filter. The xylenol-orange spectrophotometric method62 was used to
36
37 determine the concentration of H2O2 of the solution samples. For that, 500 µL of filtrate were
38
39 transferred into a spectrophotometer cuvette and immediately mixed with 250 µL of 100 mM of
40
41 xylenol-orange and 250 µL of 250 mM ferrous ammonium sulfate in that order. This protocol
42
43 was designed for a maximum concentration of H2O2 of 100 µM, so when higher concentrations
44
45
of H2O2 were analyzed the filtered solutions were diluted. The cuvettes were then incubated in
46
47
48 the dark for 30 mins and their absorbance was read at 560 nm against a blank containing 500
49
50 µL of water in place of the filtrate solution. The concentration of H2O2 was calculated from the
51
52 absorbance of each aliquot. A new calibration curve was prepared each day.
53
54
55
56
57
58 24
59
1
2
3 The rate of H2O2 decomposition was determined using the initial rate method. The
4
5 concentration of H2O2 was fitted as a function of time with a quadratic equation63 (Figure S4).
6
7
The initial rate of decomposition was calculated from the slope of the curve at t=0.
8
9
10
4.2.1 Rate models
11
12 For each catalyst the specific rate of H2O2 decomposition was regressed against H2O2
13
14 concentration, pH and temperature (see Table 3 for notation).
15
16
17 dkl R a a
no
a a
18 = = 2 ∙ \ ∙ . = (A ∙ & p∙q ) ∙ \ ∙ . (7)
d7
19
20
21 This equation can be linearized by a log transformation.
22
23
24 w >
! r = ! r 2 + 3> ∙ ! r + 3 ∙ ! r . = log ( − o ∙ + 3> ∙ log − 3 ∙ x (8)
25 .EFE∙H ?
26
27 The generalized form of this equation has 4 fitting parameters (a, b, c and d).
28
f
29 ! r = + + ' ∙ ! r − 1 ∙ x (9)
?
30
31
Multi-variate linear regression analysis was applied to each data set, using JMP Pro 12.0.1
32
33 (SAS Institute Inc.), to obtain values for Ea, k, n1 and n2. The regression coefficients obtained
34
35
were used to evaluate the contribution of each independent variable to the overall fit64. The p-
36 value for each coefficient was used to evaluate the statistical significance of the contribution of
37
38 each variable.
39
40
41
42 4.2.2 Concentrated H2O2 pretreatment
43
44 The catalytic activity of Pt black and Pt nanopowder pretreated with H2O2 was evaluated by
45
VWX
46 reacting each nanocatalyst with a range of concentrated solutions of H2O2 ( ) prior to the
47
48 rate measurement. For each pretreatment 6 mg of Pt black or Pt nanopowder were reacted with
49
50
2 mL of H2O2 at a concentration ranging from 0.003% (v/v) to 3% (v/v) of H2O2. The
51
52
53
suspensions were agitated for 20 seconds and then transferred to a falcon tube along with 5 mL
54
55 of water at pH 11. The suspension was sonicated for 1 minute. The sonicated suspension was
56
57
58 25
59
1
2
3 then added into a 1 L Erlenmeyer flask containing H2O2, and the rate of H2O2 decomposition
4
5 was determined following the procedure for rate measurements described earlier. In these
6
7
experiments , pH and temperature were maintained constant, and the only variable studied
8
9 VWX
10 was .
11
12
13 4.2.3 Heat treatment of catalysts
14
15 The rate of H2O2 decomposition by heated samples (in atmospheric conditions) of Pt black
16
17 and Pt nanopowder was compared with the samples analyzed in the as received form. Pt
18
19 nanopowder and Pt black were heated at 240 oC for 24h in atmospheric conditions. These
20
21 samples are designated Pt nanopowder-240 and Pt black-240, respectively. Six rate
22
23 measurements were performed for each heat-treated catalyst:
24
25 = 100 μ, 300 μ and 1000 μ at room (~22oC) and high temperature (~50 o
C)
26
27
conditions. For all runs pH was ~5.9.
28
29
The rate of H2O2 decomposition by heat-treated Pt black and Pt nanopowder was evaluated
30
31
32 in the same way as for samples in the as supplied condition (see section 4.2).
33
34
35
36 4.3 •HO detection
37
38 An attempt was made to detect hydroxyl radicals (•HO) in solution during the decomposition
39
40 of H2O2 on Pt black following an established protocol65. In this experiment, 20 mL of 100 µM or
41
42 1000 µM H2O2 along with 50 mM DMSO (Sigma-Aldrich, St. Louis, MO USA) were added to a
43
44
flask containing 80 mL of 3.7 mg of Pt black. Aliquots of a mixture of 80 mL containing 3.7 mg of
45
46
47 Pt black plus 20 mL of 50 mM DMSO without H2O2 were used as blanks. Reaction of DMSO
48
49 with •HO produces methane sulfinic acid that can be detected by a protocol developed by
50
51 Babbs et al., 198666 with a theoretical sensitivity of 10 nmol·mL-1. A 1 mL aliquot of the aqueous
52
53 mixture of DMSO, Pt particles and H2O2 was filtered through a 0.22 µm filter into a test tube
54
55 containing 1 mL of 2 M sulfuric acid. Then 8 mL of butanol, previously added with 1 mL of 1 M
56
57
58 26
59
1
2
3 sulfuric acid, was added to the sample and mixed for 30 s. The upper butanol phase was
4
5 removed to a second clean tube containing 4 mL of sodium acetate buffer and mixed
6
7
thoroughly. The two phases were allowed to separate for 3 min by gravity. Color development
8
9
10 was initiated by transferring the lower aqueous phase into a third test tube containing 2 mL of 15
11
12 mM Fast Blue BB salt, freshly prepared, to form diazofulfone. The sample was subsequently
13
14 incubated for 10 minutes at room temperature in the dark. Finally, 3.0 mL of 1:1 (v/v)
15
16 toluene/butanol was mixed with the aqueous phase and the lower phase, containing unreacted
17
18 diazonium salt, was discarded. The tube was centrifuged at 500 g for 3 min and the upper
19
20 phase, which contains diazosulfones was transferred into a cuvette. The absorbance of the
21
22 sample was determined at 420 nm against the blank. The calibration curve developed by Babbs
23
24 et al., 198666 was used to determine the concentration of hydroxyl radicals in the samples
25
26
analyzed in this study.
27
28
29
30
31 4.4 X-ray photoelectron spectroscopy (XPS) of heat treated
32
33 catalysts
34
35 XPS analysis was done using a PHI Quantera SXM scanning photoelectron spectrometer
36
37 microprobe. Pt catalysts preheated at 240 oC for 24 h in atmospheric conditions were deposited
38
39 on double coated carbon conductive tape (Ted Pella, USA). High-resolution XPS data were
40
41 collected for Pt4f, O1s, and C1s peaks with 0.1 eV of energy resolution at a pressure of 10-7 Pa.
42
43
The raw data were processed using Phi Multipak software. Charging was referenced to the C1
44
45
46
component of adventitious C1s peak at a binding energy of 284.8 eV for all samples67.
47
48
49 4.5 X-ray diffraction Rietveld analysis
50
Powder X-ray diffraction data were collected for all samples. Rietveld analysis was
51
52
53 performed using Profex 3.20.268. This software uses a fundamental parameters approach (FPA)
54
55 to calculate the instrumental weight function from the device configuration. A silicon NIST640d
56
57
58 27
59
1
2
3 standard for line position and line shape was used to verify the instrumental weight function.
4
5 The data for silicon were fitted with a function (RP2) that had no specimen influence on line
6
7
shapes. No spherical harmonics correction was used. Only the initial unit cell parameter (a),
8
9
10 isotropic displacement (thermal) parameter, and scale factor were allowed to vary freely during
11
12 the refinement. The initial unit cell parameter (a) and isotropic displacement parameters, and
13
14 space group Fm3m (no. 225) for face-centered cubic (FCC) platinum with a Wyckoff value of
15
16 4a were used from crystallographic information file (CIF) number 0011157 obtained from the
17
18
AMCSD69. A pseudo-Voigt function (RP4) to model peak profiles and spherical harmonics
19
20
21
correction (SPHAR4) were used in all fits. The displacement parameter was left fixed at the
22
23 initial value (U=0.001). The fitting strategy used a single Pt phase affected by size-induced
24
25 broadening (B1) and anisotropic microstrain (k2). The unit cell dimension and B1 were first
26
27 allowed to vary freely and then k2 was added to the refinement.
28
29
30
31
32 4.6 Electrochemical analysis of Pt black and Pt nanopowder
33
34 Cyclic voltammetry (CV) was used to compare the oxidation potential of Pt black and Pt
35
36 nanopowder. The experiments were carried out on a BASi EC Epsilon potentiostat using a
37
38 standard three-electrode cell. Glassy carbon disk (diameter = 3 mm) was used as a working
39
40 electrode, while high surface area Pt mesh was used as a counter electrode and Ag/AgCl
41
42 (saturated KClaq.) as a reference electrode. For the measurements, 0.1 M Na2SO4/H2O or 0.5 M
43
44
H2SO4/H2O (prepared from Millipore water, 18.3 MΩ, and analytical grade reagents) were used
45
46
47
as supporting electrolyte, which were bubbled with high purity argon for 15 min. The potential of
48
49 the reference electrode was calibrated in each electrolyte using ferricyanide (E1/2([Fe(CN)6]3-
50
51 /[Fe(CN)6]4-) = 0.164 V vs Ag/AgCl). For the sample preparation, acetonitrile suspensions of Pt
52
53 black and Pt nanopowder were deposited onto the working electrode. Upon evaporation of the
54
55
56
57
58 28
59
1
2
3 solvent, a uniform layer of Pt particles was formed on the electrode surface. The
4
5 voltammograms were collected at room temperature using 100 mV s-1 scan rate.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 29
59
1
2
3
4 Associated Content
5
6
Supporting Information
7 The supporting information is available on the ACS Publications Website at DOI:
8
9 XXXXXXXXX.
10
11
Schematic diagram of proposed mechanisms for the ORR; TEM images of Pt
12 nanopowder and Pt black in the as received condition; tables that compile reaction
13
14 conditions and measured rate for each rate measurement experiment analyzed in this
15
16 study; test results for experimental attempt to detect radical species during the
17
decomposition of H2O2; cyclic voltammetry analysis of Pt black and Pt nanopowder in
18
19 the as received condition; example of initial rate method utilized to determine the rate of
20
21 H2O2 decomposition for each rate measurement experiment.
22
23
24
25
Author Information
26
Email: mfrede2@vt.edu (F.M.M.)
27
28 Email: ruism002@vt.edu (R.S-M.)
29
30
31
32
Acknowledgements
33
34
F. M. M. and R. S. M. gratefully acknowledge financial support from the Virginia Tech
35 Institute for Critical Technology and Applied Science (ICTAS) through a Junior Faculty
36
37 Collaborative Grant (ICTAS-JFC 175884). R. S. M. gratefully acknowledges the Virginia
38
39 Tech College of Science Roundtable Alumni Advisory Group for financial support
40
provided through a “Make-a-Difference” Scholarship, as well as the Department of
41
42 Geosciences for support provided by Sir Aubrey and Madam Eula Orange Scholarship
43
44 funds. The authors also thank Dr. Robert Bodnar and Mr. Charles Farley for support
45
46 with selected experiments.
47
48
49
50
51
52
53
54
55
56
57
58 30
59
1
2
3
4 5 References
5
6 (1) Chen, B.; Garland, N. T.; Geder, J.; Pruessner, M.; Mootz, E.; Cargill, A.; Leners, A.;
7 Vokshi, G.; Davis, J.; Burns, W.; Daniele, M. A.; Kogot, J.; Medintz, I. L.; Claussen, J.
8 C. Platinum Nanoparticle Decorated SiO2 Microfibers as Catalysts for Micro
9 Unmanned Underwater Vehicle Propulsion. ACS Appl. Mater. Interfaces 2016, 8,
10 30941-30947.
11
12 (2) Krejci, D.; Woschnak, A.; Scharlemann, C.; Ponweiser, K. Structural impact of
13
14
honeycomb catalysts on hydrogen peroxide decomposition for micro propulsion.
15 Chem. Eng. Res. Des. 2012, 90, 2302-2315.
16
17 (3) Marr, K. M.; Chen, B.; Mootz, E. J.; Geder, J.; Pruessner, M.; Melde, B. J.; Vanfleet,
18 R. R.; Medintz, I. L.; Iverson, B. D.; Claussen, J. C. High Aspect Ratio Carbon
19 Nanotube Membranes Decorated with Pt Nanoparticle Urchins for Micro Underwater
20 Vehicle Propulsion via H2O2 Decomposition. ACS Nano 2015, 9, 7791–7803.
21
22
(4) Hammerschmidt, A.; Domke, W. D.; Nolscher, C.; Suchy, P., PEM fuel cell. Google
23
24 Patents: 2000.
25
26 (5) Li, X.; Ge, S.; Hui, C.; Hsing, I.-M. Well-dispersed multiwalled carbon nanotubes
27 supported platinum nanocatalysts for oxygen reduction. Electrochem. Solid-State Lett.
28 2004, 7, 286-289.
29
30 (6) Li, Y.; Yang, J.; Song, J. Structure models and nano energy system design for
31 proton exchange membrane fuel cells in electric energy vehicles. Renewable
32
33
Sustainable Energy Rev. 2017, 67, 160-172.
34
35 (7) Nie, Y.; Li, L.; Wei, Z. Recent advancements in Pt and Pt-free catalysts for oxygen
36 reduction reaction. Chem. Soc. Rev. 2015, 44, 2168-2201.
37
38 (8) Bianchi, G.; Mazza, F.; Mussini, T. Catalytic decomposition of acid hydrogen
39 peroxide solutions on platinum, iridium, palladium and gold surfaces. Electrochim.
40 Acta 1962, 7, 457-473.
41
42 (9) Roy, C. B. Catalytic decomposition of hydrogen peroxide on some oxide catalysts. J.
43
44
Catal. 1968, 12, 129-133.
45
46 (10) Weiss, J. The catalytic decomposition of hydrogen peroxide on different metals.
47 Trans. Faraday Soc. 1935, 31, 1547-1557.
48
49 (11) Eley, D. D.; Macmahon, D. M. The decomposition of hydrogen peroxide catalysed
50 by palladium-gold alloy wires. J. Colloid Interface Sci. 1972, 38, 502-510.
51
52 (12) Hall, S. B.; Khudaish, E. A.; Hart, A. L. Electrochemical oxidation of hydrogen
53
peroxide at platinum electrodes. Part II: effect of potential. Electrochim. Acta 1998, 43,
54
55 2015-2024.
56
57
58 31
59
1
2
3 (13) Liu, Y.; Wu, H.; Li, M.; Yin, J.-J.; Nie, Z. pH dependent catalytic activities of
4
5
platinum nanoparticles with respect to the decomposition of hydrogen peroxide and
6 scavenging of superoxide and singlet oxygen. Nanoscale 2014, 6, 11904-11910.
7
8 (14) McKee, D. W. Catalytic decomposition of hydrogen peroxide by metals and alloys
9 of the platinum group. J. Catal. 1969, 14, 355-364.
10
11 (15) Garwig, P. L. Heterogeneous decomposition of hydrogen peroxide by inorganic
12 catalysts. A Literature survey. FMC Corp. Princeton NJ, 1966.
13
14 (16) Choudhary, V. R.; Samanta, C.; Jana, P. Decomposition and/or hydrogenation of
15
16
hydrogen peroxide over Pd/Al2O3 catalyst in aqueous medium: Factors affecting the
17 rate of H2O2 destruction in presence of hydrogen. Appl. Catal., A 2007, 332, 70-78.
18
19 (17) Lousada, C. M.; Yang, M.; Nilsson, K.; Jonsson, M. Catalytic decomposition of
20 hydrogen peroxide on transition metal and lanthanide oxides. J. Mol. Catal. A: Chem.
21 2013, 379, 178-184.
22
23 (18) Bianchi, G.; Mazza, F.; Mussini, T.; UNIV, M. Catathic reduction of oxygen and
24
hydrogen peroxide on platinum, palladium and iridium smooth electrodes, Defense
25
26 Technical Information Center: 1961.
27
28 (19) Hasnat, M. A.; Rahman, M. M.; Borhanuddin, S. M.; Siddiqua, A.; Bahadur, N. M.;
29 Karim, M. R. Efficient hydrogen peroxide decomposition on bimetallic Pt–Pd surfaces.
30 Catal. Commun. 2010, 12, 286-291.
31
32 (20) McKee, D. Catalytic activity and sintering of platinum black. Kinetics of propane
33 cracking. J. Phys. Chem. A 1963, 67, 841-846.
34
35
36
(21) MacInnes, D. A. The mechanism of the catalysis of the decomposition of hydrogen
37 peroxide by colloidal platinum. J. Am. Chem. Soc. 1914, 36, 878-881.
38
39 (22) Hall, S. B.; Khudaish, E. A.; Hart, A. L. Electrochemical oxidation of hydrogen
40 peroxide at platinum electrodes. Part 1. An adsorption-controlled mechanism.
41 Electrochim. Acta 1998, 43, 579-588.
42
43 (23) Ono, Y.; Matsumura, T.; Kitajima, N.; Fukuzumi, S. Formation of superoxide ion
44 during the decomposition of hydrogen peroxide on supported metals. J. Phys. Chem.
45
46
A 1977, 81, 1307-1311.
47
48 (24) Kitajima, N.; Fukuzumi, S.; Ono, Y. Formation of superoxide ion during the
49 decomposition of hydrogen peroxide on supported metal oxides. J. Phys. Chem. A
50 1978, 82, 1505-1509.
51
52 (25) Gerischer, R. Über die katalytische Zersetzung von Wasserstoffsuperoxyd an
53 metallischem Platin. Z. Phys. Chem.. 1956, 6, 178-200.
54
55
56
57
58 32
59
1
2
3 (26) Nolan, P. D.; Wheeler, M. C.; Davis, J. E.; Mullins, C. B. Mechanisms of Initial
4
5
Dissociative Chemisorption of Oxygen on Transition-Metal Surfaces. Acc. Chem. Res.
6 1998, 3, 798-804.
7
8 (27) Schaeffer, J.; Fonseca, L.; Samavedam, S.; Liang, Y.; Tobin, P.; White, B.
9 Contributions to the effective work function of platinum on hafnium dioxide. Appl.
10 Phys. Lett. 2004, 85, 1826-1828.
11
12 (28) Međedović, S.; Locke, B. R. Platinum catalysed decomposition of hydrogen
13 peroxide in aqueous-phase pulsed corona electrical discharge. Appl. Catal., B 2006,
14
15
67, 149-159.
16
17 (29) Kirkpatrick, M. J.; Locke, B. R. Effects of platinum electrode on hydrogen, oxygen,
18 and hydrogen peroxide formation in aqueous phase pulsed corona electrical
19 discharge. Ind. Eng. Chem. Res. 2006, 45, 2138-2142.
20
21 (30) Katsounaros, I.; Schneider, W. B.; Meier, J. C.; Benedikt, U.; Biedermann, P. U.;
22 Auer, A. A.; Mayrhofer, K. J. Hydrogen peroxide electrochemistry on platinum: towards
23
understanding the oxygen reduction reaction mechanism. Phys. Chem. Chem. Phys.
24
25 2012, 14, 7384-7391.
26
27 (31) Chiang, Y. S.; Craddock, J.; Mickewich, D.; Turkevich, J. Study with Fast-Mixing
28 Techniques of the Titanium (III) and Hydrogen Peroxide Reaction. J. Phys. Chem. A
29 1966, 70, 3509-3515.
30
31 (32) Saito, E.; Bielski, B. H. J. The electron paramagnetic resonance spectrum of the
32 HO2 radical in aqueous solution. J. Am. Chem. Soc. 1961, 83, 4467-4468.
33
34
35
(33) Bliznakov, G.; Lazarov, D. Surface heterogeneity of metals of the copper group in
36 the catalytic decomposition of hydrogen peroxide. J. Catal. 1969, 14, 187-192.
37
38 (34) Keating, K. B.; Rozner, A. G.; Youngblood, J. L. The effect of deformation on
39 catalytic activity of platinum in the decomposition of hydrogen peroxide. J. Catal. 1965,
40 4, 608-619.
41
42 (35) Roduner, E. Size matters: why nanomaterials are different. Chem. Soc. Rev. 2006,
43 35, 583-592.
44
45
46
(36) Kinoshita, K. Particle size effects for oxygen reduction on highly dispersed platinum
47 in acid electrolytes. J. Electrochem. Soc. 1990, 137, 845-848.
48
49 (37) Shao, M.; Peles, A.; Shoemaker, K. Electrocatalysis on Platinum Nanoparticles:
50 Particle Size Effect on Oxygen Reduction Reaction Activity. Nano Lett.2011, 11, 3714-
51 3719.
52
53 (38) Burch, R.; Bond, G.; Webb, G. Specialist periodical reports. ACS Catal. 1985, 7,
54
149.
55
56
57
58 33
59
1
2
3 (39) Ono, L. K.; Croy, J. R.; Heinrich, H.; Roldan Cuenya, B. Oxygen Chemisorption,
4
5
Formation, and Thermal Stability of Pt Oxides on Pt Nanoparticles Supported on
6 SiO2/Si(001): Size Effects. J. Phys. Chem. A 2011, 115, 16856-16866.
7
8 (40) Bond, G. C. The origins of particle size effects in heterogeneous catalysis. Surface
9 Science 1985, 156, 966-981.
10
11 (41) Trasatti, S. Electronegativity, work function, and heat of adsorption of hydrogen on
12 metals. J. Chem. Soc., Faraday Trans. 1: Physical Chemistry in Condensed Phases
13 1972, 68, 229-236.
14
15
16
(42) Nørskov, J. K.; Rossmeisl, J.; Logadottir, A.; Lindqvist, L.; Kitchin, J. R.; Bligaard,
17 T.; Jonsson, H. Origin of the overpotential for oxygen reduction at a fuel-cell cathode.
18 J. Phys. Chem. A B 2004, 108, 17886-17892.
19
20 (43) Sabatier, P. La catalyse en chimie organique, Nouveau Monde éditions: 2014.
21
22 (44) Lasaga, A. C. Rate laws of chemical reactions. Rev. Mineral. (United States) 1981,
23 8.
24
25
(45) Ou, L.; Yang, F.; Liu, Y.; Chen, S. First-principle study of the adsorption and
26
27 dissociation of O2 on Pt (111) in acidic media. J. Phys. Chem. A C 2009, 113, 20657-
28 20665.
29
30 (46) Arruda, T. M.; Shyam, B.; Ziegelbauer, J. M.; Mukerjee, S.; Ramaker, D. E.
31 Investigation into the competitive and site-specific nature of anion adsorption on Pt
32 using in situ X-ray absorption spectroscopy. J. Phys. Chem. A C 2008, 112, 18087-
33 18097.
34
35
36
(47) Keating, K.; Rozner, A.; Youngblood, J. The effect of deformation on catalytic
37 activity of platinum in the decomposition of hydrogen peroxide. J. Catal. 1965, 4, 608-
38 619.
39
40 (48) Smith, C. E.; Biberian, J. P.; Somorjai, G. A. The effect of strongly bound oxygen
41 on the dehydrogenation and hydrogenation activity and selectivity of platinum single
42 crystal surfaces. J. Catal. 1979, 57, 426-443.
43
44 (49) Antoine, O.; Bultel, Y.; Durand, R. Oxygen reduction reaction kinetics and
45
46
mechanism on platinum nanoparticles inside Nafion®. Journal of Electroanalytical
47 Chemistry 2001, 499, 85-94.
48
49 (50) van Hardeveld, R.; Hartog, F. Influence of Metal Particle Size in Nickel-on-Aerosil
50 Catalysts on Surface Site Distribution, Catalytic Activity, and Selectivity. Adv. Catal.
51 1972, 22, 75-113.
52
53 (51) Balbuena, P. B.; Calvo, S. R.; Lamas, E. J.; Salazar, P. F.; Seminario, J. M.
54
Adsorption and Dissociation of H2O2 on Pt and Pt− Alloy Clusters and Surfaces. J.
55
56 Phys. Chem. A 2006, 110, 17452-17459.
57
58 34
59
1
2
3 (52) Mason, M. G. Electronic structure of supported small metal clusters. Phys. Rev. B
4
5
1983, 27, 748-762.
6
7 (53) Zhou, L.; Zachariah, M. R. Size resolved particle work function measurement of
8 free nanoparticles: Aggregates vs. spheres. Chemical Physics Letters 2012, 525, 77-
9 81.
10
11 (54) Sweeney, W.; Lee, J.; Abid, N.; DeMeo, S. Efficient Method for the Determination
12 of the Activation Energy of the Iodide-Catalyzed Decomposition of Hydrogen Peroxide.
13 Journal of Chemical Education 2014, 91, 1216-1219.
14
15
16
(55) Han, B. C.; Miranda, C. R.; Ceder, G. Effect of particle size and surface structure
17 on adsorption of O and OH on platinum nanoparticles: A first-principles study. Phys.
18 Rev. B 2008, 77, 075410/1 – 075410/9 .
19
20 (56) Jaeger, N. I.; Jourdan, A. L.; Schulz-Ekloff, G. Effect of the size of platinum
21 particles on the chemisorption of oxygen. J. Chem. Soc., Faraday Trans. 1991, 87,
22 1251-1257.
23
24
(57) Wang, Y.; Balbuena, P. B. Potential Energy Surface Profile of the Oxygen
25
26 Reduction Reaction on a Pt Cluster: Adsorption and Decomposition of OOH and
27 H2O2. J. Chem. Theory Comput. 2005, 1, 935-943.
28
29 (58) Panchenko, A.; Koper, M.; Shubina, T.; Mitchell, S.; Roduner, E. Ab initio
30 calculations of intermediates of oxygen reduction on low-index platinum surfaces. J.
31 Electrochem. Soc. 2004, 151, 2016-2027.
32
33 (59) Gu, Z.; Balbuena, P. B. Absorption of Atomic Oxygen into Subsurfaces of Pt(100)
34
35
and Pt(111): Density Functional Theory Study. J. Phys. Chem. A C 2007, 111, 9877-
36 9883.
37
38 (60) Voorhees, V.; Adams, R. The use of oxides of platinum for the catalytic reduction of
39 organic compounds. J. Am. Chem. Soc. 1922, 44, 1397-1405.
40
41 (61) Mills, A. Final Analysis: Porous Platinum Morphologies: Platinised, Sponge and
42 Black, Platin. Met. Rev. 2007, 51, 52-52.
43
44 (62) Gay, C.; Collins, J.; Gebicki, J. M. Hydroperoxide Assay with the Ferric–Xylenol
45
46
Orange Complex. Anal. Biochem. 1999, 273, 149-155.
47
48 (63) Rimstidt, J. D. Geochemical Rate Models: An Introduction to Geochemical Kinetics,
49 Cambridge University Press: Cambridge, 2013.
50
51 (64) Pollyea, R. M.; Rimstidt, J. D. Rate equations for modeling carbon dioxide
52 sequestration in basalt. Appl. Geochem.2017, 81, 53-62.
53
54 (65) Steiner, M. G.; Babbs, C. F. Quantitation of the hydroxyl radical by reaction with
55
dimethyl sulfoxide. Arch. Biochem. Biophys. 1990, 278, 478-481.
56
57
58 35
59
1
2
3 (66) Babbs, C. F.; Gale, M. J. Colorimetric assay for methanesulfinic acid in biological
4
5
samples. Anal. Biochem. 1987, 163, 67-73.
6
7 (67) Swift, P. Adventitious carbon—the panacea for energy referencing? Surf. Interface
8 Anal. 1982, 4, 47-51.
9
10 (68) Doebelin, N.; Kleeberg, R. Profex: a graphical user interface for the Rietveld
11 refinement program BGMN. J. Appl. Crystallogr. 2015, 48, 1573-1580.
12
13 (69) Downs, R. T.; Hall-Wallace, M. The American Mineralogist crystal structure
14 database. Am. Mineral. 2003, 88, 247-250.
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58 36
59
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26 1273x682mm (96 x 96 DPI)
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59

Serra Maia 2018

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Serra Maia 2018

Uploaded by

Copyright:

Available Formats

Subscriber access provided by Kaohsiung Medical University

Energy, Environmental, and Catalysis Applications

is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

You might also like