Transition Metal Hydride Catalysts for Sustainable Interconversion of

Mar 19, 2018 - Transition Metal Hydride Catalysts for Sustainable Interconversion of CO2 and Formate: Thermodynamic and Mechanistic Considerations ...
1 downloads 3 Views 796KB Size
Subscriber access provided by UniSA Library

Transition Metal Hydride Catalysts for Sustainable Interconversion of CO and Formate: Thermodynamic and Mechanistic Considerations 2

Kate M. Waldie, Felix M. Brunner, and Clifford P. Kubiak ACS Sustainable Chem. Eng., Just Accepted Manuscript • DOI: 10.1021/ acssuschemeng.8b00628 • Publication Date (Web): 19 Mar 2018 Downloaded from http://pubs.acs.org on March 20, 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 26 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 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 60

ACS Sustainable Chemistry & Engineering

Transition Metal Hydride Catalysts for Sustainable Interconversion of CO2 and Formate: Thermodynamic and Mechanistic Considerations

Kate M. Waldie, Felix M. Brunner, and Clifford P. Kubiak*

Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Drive, Mail Code 0358, La Jolla, California 92093, USA * [email protected]

ABSTRACT Hydride transfer between transition metal hydride complexes and carbon dioxide is a known reaction, where the thermodynamically favored direction of hydride transfer determines whether CO2 reduction or formate oxidation occurs. Analysis of a growing database of thermodynamic parameters for transition metal hydride complexes now provides clear demarcation between metal hydrides which will function as oxidases and which as reductases. The turning point is set at the hydricity of formate (44 kcal/mol in acetonitrile). Here we utilize hydricity as a framework to reevaluate the catalytic activity and proposed mechanisms for formate oxidation and CO2 reduction with several Ni and Rh P2N2 (P2N2 = 1,5-diaza-3,7diphosphacyclooctane) complexes, respectively. The series of Ni P2N2 complexes have

ACS Paragon Plus Environment

1

ACS Sustainable Chemistry & Engineering 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 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 60

Page 2 of 26

hydricities between 55 – 64 kcal/mol and are active catalysts for the electrochemical oxidation of formate. A surprising correlation of increased rate of electrochemical oxidation with decreased overpotential, ߟ, is observed. The Rh P2N2 complexes have hydricities between 28 – 34 kcal/mol and function as hydrogenation catalysts for the reduction of CO2 to formate. Learning from the reactivity of these catalysts, design principles for future metal hydride complexes are presented that focus on the ultimate goal of catalyst optimization for improved energy efficiency (overpotential) with high selectivity (Faradaic efficiency) for both formate oxidation and CO2 reduction to formate.

KEYWORDS Hydricity, molecular catalysis, carbon dioxide, formic acid, nickel, rhodium, phosphine ligands

INTRODUCTION The utilization of renewable energy resources such as solar or wind to meet global energy demands will require the storage of energy from these resources in the form of chemical fuels. Liquid carbon fuels such as formic acid are particularly attractive energy carriers due to their high volumetric energy densities and ease of handling.1 The production of formic acid is currently performed by carbonylation of methanol to methyl formate followed by hydrolysis2; however, there is great interest in deriving formate from reduction of CO2, an inexpensive and readily available C1 feedstock. This process may involve either electrochemical reduction of CO2 to formate, or hydrogenation of CO2 to formic acid where the hydrogen is derived from electrochemical water splitting using renewable energy. Conversely, while combustion remains a primary means of energy extraction from fuels, electrochemical oxidation of fuels in a fuel cell

ACS Paragon Plus Environment

2

Page 3 of 26 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 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 60

ACS Sustainable Chemistry & Engineering

offers higher theoretical efficiencies. The development of efficient catalysts for the interconversion of CO2 and formate is thus a critical target for our sustainable energy future, creating a cycle in which electricity from renewable energy drives the reduction of CO2 to formate, and then usable energy is harvested on-demand using a direct formic acid fuel cell (Figure 1).3

Electrochemical Reduction or Hydrogenation

Renewable Energy H2

or

2e−, H+

H+ O CO2

Electrochemical Oxidation

H

O−

2e−, H+

Figure 1. Sustainable energy cycle based on CO2 and formate.

Our group4-6 and others7-17 are interested in the application of transition-metal hydride catalysts to mediate both halves of the energy cycle in Figure 1. The thermal hydrogenation11 or electrocatalytic reduction18 of CO2 to formate typically involves hydride transfer from a metalhydride intermediate to CO2 as a key mechanistic step. The reverse reaction of electrocatalytic formate oxidation may occur via the reverse pathway by heterolytic cleavage of the C-H bond in formate with concomitant hydride transfer to the metal. The free energy associated with both of these hydride transfer reactions can be predicted based on the relative hydride donating ability of the metal hydride versus formate. Hydride donor ability, or hydricity ∆GH−, of a large number (over 100) of transition-metal hydride complexes have been experimentally determined in acetonitrile, in particular for bis(diphosphine) complexes of the general form [HM(P-P)2]n+.7 The

ACS Paragon Plus Environment

3

ACS Sustainable Chemistry & Engineering 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 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 60

Page 4 of 26

hydricity of formate is also known in acetonitrile, thus enabling the thermodynamically favored direction for hydride transfer to be easily assessed.19 The large dataset of known hydricity values for transition-metal hydride complexes has allowed for comparisons of hydricities with other thermodynamic parameters. We recently demonstrated that the hydricities of metal hydrides [M-H](n-1)+ in acetonitrile exhibit a remarkably linear relationship with the first one-electron reduction potential of the parent metal complex [M]n+.19 Similar behavior over smaller data sets of metal hydride classes has been noted by other authors.7, 20-22 This trend is well described by equation (5), which is derived from the thermochemical cycle shown in Scheme 1 relating hydricity and the homolytic bond dissociation free energy of the metal-hydride bond (BDFE) to E1/2(Mn+/(n-1)+).7,

23

Since this reduction

potential is often readily obtained by standard electrochemical methods, the correlation of hydricity with E1/2(Mn+/(n-1)+) may be used to qualitatively predict hydricities simply based on the reduction potential of the parent complex.

Scheme 1. Thermochemical cycle for hydricity and E1/2(Mn+/(n-1)+). [M−H](n-1)+ M(n-1)+ H•

+

e−

[M−H](n-1)+

M(n-1)+ + H•

BDFE

(1)

Mn+

nFE1/2(Mn+/(n-1)+)

(2)

H−

∆G°H•/H−

(3)

Mn+ + H−

∆G°H−

(4)

+

e−

∆G°H− = BDFE + nFE1/2(Mn+/(n-1)+) + ∆G°H•/H-

(5)

The hydricity of a metal-hydride catalyst also provides the opportunity to make generalized statements about the reaction conditions required for catalysis and the limitations thereof for obtaining the correct balance of reaction energies to achieve selective reactivity. In particular, the hydricity of a given metal hydride determines the pKa of exogenous base required for heterolytic cleavage of H2 for CO2 hydrogenation, or the pKa of exogenous Brønsted acid

ACS Paragon Plus Environment

4

Page 5 of 26 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 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 60

ACS Sustainable Chemistry & Engineering

required to avoid competitive hydrogen evolution during electrocatalytic CO2 reduction.19 Efforts have also been directed towards the correlation of hydricity with the rates of catalysis.4-5, 16, 24

Herein, we discuss hydricity in terms of these relationships and reaction condition

limitations for two case studies: electrocatalytic oxidation of formate with [Ni(P2N2)2(CH3CN)]2+ 1 – 8, and thermal hydrogenation of CO2 to formate with [Rh(P2N2)2]+ 10 – 14, where P2N2 = 1,5-diaza-3,7-diphosphacyclooctane (Figure 2). We further comment on future directions with these catalysts and propose new metal-hydride structures based on nickel towards the design of systems for the chemical and electrochemical hydrogenation of CO2 using a first-row metal.

R′

N

R′ N

R

CH3 C N R

P

Ni P P R

P R

2+

N

R′

N

R P

R′ R′

N R′

R

N

R′

P

Rh

N

N R′

P R

P R

[Ni(PR2NR′2)2(CH3CN)]2+

[Rh(PR2NR′2)2]+

1-8

10 - 14

Et2 P

Et2 P

2+

Et2 P

Et2 P

+

Rh

Ni P Et2

P Et2

+

P Et2

P Et2

[Ni(depe)2]2+

[Rh(depe)2]+

9

15

Figure 2. Structures of the parent complexes [M]n+ for the Ni and Rh hydrides of interest.

HYDRICITY CONSIDERATIONS The hydride donor ability of formate ∆GH−(HCO2−) in acetonitrile is 44 kcal/mol.7, 25 This value was obtained by equilibration of formate with [Pt(depe)2]2+ (depe = 1,2bis(diethylphosphino)ethane), which was found to be essentially ergoneutral. The hydricity of this Pt-hydride was in turn calculated from the two-electron reduction potential of [Pt(depe)2]2+ and the pKa of the metal hydride in acetonitrile.26-27

ACS Paragon Plus Environment

5

ACS Sustainable Chemistry & Engineering 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 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 60

Page 6 of 26

Square planar complexes containing two bidentate phosphine ligands are especially well suited for hydricity measurements,7 and the steric and electronic properties of the phosphine ligand are easily modified to enable access to a wide range of environments at the metal center. Additionally, the parent metal complexes typically exhibit reversible electrochemical features due to metal-based redox activity. This behavior is nicely exemplified by complexes 1 – 15. The hydricities of nickel complexes 2 – 6 and 8 – 9 were measured by equilibration of the Ni-hydride and parent complex under an H2 atmosphere in acetonitrile in the presence of an aniline base.4 The same method was applied to determine the hydricity of the rhodium systems 10 – 15 using Verkade’s base in benzonitrile,5 which is not expected to have a significant effect on the thermodynamic properties compared to acetonitrile. For complexes 1 and 7, the hydricities have not been experimentally determined, but are rather estimated by computational methods28 or by linear free energy relationships with closely related complexes.4 The hydricities of the relevant nickel and rhodium systems are tabulated in Table 1, along with the first and second one-electron reduction potentials of the parent complexes and the metal hydride pKa. In Figure 3, each hydricity value is plotted against E1/2(Mn+/(n-1)+) for the respective parent complex [M]n+. These data points are overlaid with the linear scaling relationship recently noted by our group19 for the correlation between hydricity and E1/2(Mn+/(n-1)+) according to equation (5). This set of metal hydrides spans a large range of hydricity values, from 28.3 to 63.7 kcal/mol. The effect of different ligand substituents on the hydricity of metal hydrides bearing P2N2 ligands has been discussed elsewhere.4, 6, 29 However, Figure 3 clearly highlights that the hydricities of the Ni and Rh systems fall into two distinct regions depending on the identity of the metal center: all of the nickel hydrides fall above the horizontal line defined by the hydricity

ACS Paragon Plus Environment

6

Page 7 of 26 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 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 60

ACS Sustainable Chemistry & Engineering

of formate (∆GH−(Ni-H) > 44 kcal/mol), while all of the rhodium hydrides are very strong hydride donors and fall below the formate line (∆GH−(Rh-H) < 44 kcal/mol). Table 1. Thermodynamic data for Ni and Rh hydride complexes. a,b Phosphine Ligand E1/2(d8/d9) E1/2(d9/d10) ∆G°H− Ni-Hydride pKa +/0 c (V vs. Fc ) (V vs. Fc+/0) c (kcal/mol) -0.98 -1.14 55.6 d 18.5 PR2NR'2; R = Ph, R' = Me 1 R R' -0.94 -1.19 57.2 19.6 P 2N 2; R = Ph, R' = Bz 2 R R' P 2N 2; R = Ph, R' = C6H4-OMe -0.87 -1.07 58.6 17.4 3 PR2NR'2; R = Ph, R' = Ph -0.84 -1.02 59.1 16.4 4 PR2NR'2, PR'2NR'2; R = Cy, R' = Ph -0.76 -1.05 60.5 16.6 5 R R’ P 2N 2; R = Cy, R’ = Bz -0.80 -1.28 60.9 21.5 6 PR2NR'2; R = Ph, R' = C6H4-CF3 -0.74 -0.89 61.4 e 13.8 e 7 -0.62 -1.09 63.7 17.3 PR2NR'2; R = Cy, R' = Ph 8 f depe -1.16 -1.29 55.3 23.8 9 Rh-Hydride PR2NR'2; R = Ph, R' = Bz -2.43 -2.43 28.4 44.6 10 R R' P 2N 2; R = Cy, R' = C6H4-OMe -2.45 -2.45 30.2 46.6 11 PR2NR'2; R = Ph, R' = C6H4-OMe -2.27 -2.27 31.3 41.3 12 R R' P 2N 2; R = Cy, R' = Ph -2.39 -2.39 31.9 45.8 13 R R' P 2N 2; R = Ph, R' = Ph -2.21 -2.21 14 depe 28.3 15 a b 45 Values in acetonitrile unless otherwise marked. Data from Kubiak and co-workers unless otherwise indicated. c Reduction potentials refer to the parent metal complex. d DFT-calculated value.28 e Estimated value. See Kubiak and co-workers.4 f Measured in benzonitrile.

Figure 3. Hydricities of Ni (red) and Rh (green) hydride catalysts as a function of E1/2(Mn+/(n-1)+) for the parent metal complex. Hydricity of formate given by the black solid trace. Correlation between ∆GH− and E1/2(Mn+/(n-1)+) for metal hydrides from Kubiak and co-workers19 given by the black dashed trace. Ni-hydrides are in the red region of the

ACS Paragon Plus Environment

7

ACS Sustainable Chemistry & Engineering 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 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 60

Page 8 of 26

graph with ∆GH−(Ni-H) > ∆GH−(HCO2−). Rh-hydrides are in the green region of the graph with ∆GH−(Rh-H) < ∆GH−(HCO2−).

FORMATE OXIDATION WITH NICKEL P2N2 CATALYSTS We have previously shown that nickel complexes 1 – 8 are active for the electrocatalytic oxidation of formate in benzonitrile using a 1:1 buffer of tetra-n-butylammonium formate/formic acid.4 This study was motivated by the fact that the hydricities of [HNi(P2N2)]+ were known to be greater than that of formate in acetonitrile from previous investigations into their thermodynamic properties and activity for electrocatalytic hydrogen evolution.30-31 In the absence of formate, complexes 1 – 9 exhibit two reversible one-electron reductions at ca. -0.8 and -1.0 V versus ferrocenium/ferrocene (Fc+/0), which correspond to the Ni(II/I) and Ni(I/0) couples, respectively. The addition of formate leads to electrocatalytic current enhancement near the potential of the Ni(II/I) couple, reaching saturation near 0.08 M.4 As seen in Table 2, the maximum rate of electrocatalysis, given by the turnover frequency (TOF), depends greatly on the P2N2 substituents. Notably, removal of the pendant base completely shuts down catalysis for [HNi(depe)2]+ 9. The standard reduction potential Eº for the CO2/HCO2− couple in acetonitrile is estimated to be Eº = [-0.77 V – 0.030×pKa] versus Fc+/0 for a 1:1 solution of acid and conjugate base.19 Assuming that thermodynamic values determined in benzonitrile are a reasonable approximation of those in acetonitrile, the overpotential ߟ for formate oxidation can be calculated for each nickel complex. All electrocatalytic studies were performed using a 1:1 buffer of formate/formic acid, which has an estimated pKa of 20.9 in acetonitrile.32 Thus, Eº(CO2/HCO2−) = -1.40 V versus Fc+/0 under these conditions. The overpotential for each catalyst is summarized in Table 2. We note that the potential at half the maximum current33 Ecat/2 is approximated as the potential of the Ni(II/I) couple. This method likely leads to a small overestimation of ߟ since the Ni(II/I)

ACS Paragon Plus Environment

8

Page 9 of 26 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 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 60

ACS Sustainable Chemistry & Engineering

oxidation peak shifts to more negative potentials with increasing formate concentration.4 The estimated overpotentials range from 0.42 – 0.78 V, with 8 (where PR2NR'2; R = cyclohexyl, R' = phenyl) exhibiting the largest overpotential and 1 (where PR2NR'2; R = phenyl, R' = methyl) exhibiting the smallest overpotential.

Table 2. Turnover frequencies (TOF) and overpotentials (ߟ) for electrocatalytic formate oxidation. Ni-Hydride TOF (s-1) a ߟ (V) b 15.8 0.42 1 12.5 0.46 2 8.7 0.53 3 7.4 0.56 4 7.9 0.64 5 9.6 0.60 6 3.4 0.66 7