Subscriber access provided by NJIT | New Jersey Institute of Technology
Article 2
Structural Diversity of Copper-CO Complexes – Infrared Spectra and Structures of [Cu(CO)]- Clusters 2
n
Benjamin J. Knurr, and J. Mathias Weber J. Phys. Chem. A, Just Accepted Manuscript • Publication Date (Web): 15 Oct 2014 Downloaded from http://pubs.acs.org on October 16, 2014
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 free 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 accessible to all readers and 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.
The Journal of Physical Chemistry A 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 30
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
The Journal of Physical Chemistry
Structural Diversity of Copper-CO2 Complexes – Infrared Spectra and Structures of [Cu(CO2)n]- Clusters
Benjamin J. Knurr and J. Mathias Weber*
JILA and Department of Chemistry and Biochemistry, University of Colorado, Boulder, CO 80309, USA Abstract: We present infrared spectra of [Cu(CO2)n]- (n = 2 – 9) clusters in the wavenumber range 1600 cm-1 to 2400 cm-1. The CO stretching modes in this frequency region encode the structural nature of the cluster core and are interpreted with the aid of density functional theory. We find a variety of core species in [Cu(CO2)n]- clusters, but the dominant core structure is a [Cu(CO2)2]core where the two CO2 ligands are bound to the Cu atom in a bidentate fashion. We compare the results of [Cu(CO2)n]- clusters to those of other [M(CO2)n]- clusters (M = Au, Ag, Co, Ni) in order to establish trends of how the metal-CO2 interaction depends on the metal partner.
Keywords: Transition metals, carbon dioxide, metal-organic chemistry, photodissociation spectroscopy, mass spectrometry
*
Corresponding Author: J. M. Weber: phone, ++1-303-492-7841; e-mail: [email protected]. 1 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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 30
Introduction: Transition metals are employed in a wide variety of catalytic applications, but in many cases the mechanistic, molecular level details of how the catalysts function are poorly understood, hindering rational catalyst design. Experiments on catalyst complexes in situ are challenging, in particular because speciation is often complicated, and the solvent environment is difficult to control and characterize and can affect the chemical nature of the complex under study.
Unlike experiments in the condensed phase, molecular beam experiments on mass
selected cluster ions in vacuo allow for better control over the species and the solvation environment under study, making these experiments ideal for the investigation of model catalyst systems. Copper has been used in a number of catalytic systems and fuel cell applications to convert CO2 to CO1-3, methane3-5, alcohols and other small organic species.3,6,7 While there has been some success in the conversion of CO2, efficiency and selectivity of the catalysts used still leave much to be desired. Overall, there has been little effort to characterize the molecular level details of the reactions involved, even though understanding the catalyst-reactant interaction is necessary to accurately model and design better catalysts. We have previously studied the interactions of CO2 with a number of different metals (Au, Ag, Co and Ni) and have been able to characterize the resulting metal-CO2 interactions and the effect of solvation on them.8-11 In the context of previous metal CO2 studies, copper is unique as it is both a member of the 9th column (like Au and Ag) and the top row transition metals (like Co and Ni). The dominant structural motifs in these two groups of metals are very different. While the coinage metals form a [MCO2]- core ion,8,9 the first row transition metals bind to CO2 ligands primarily through [CO2MCO2]- complexes.10,11
2 ACS Paragon Plus Environment
Here we present IR
Page 3 of 30
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
The Journal of Physical Chemistry
photodissociation spectra of [Cu(CO2)n]- clusters.
Aided by density functional theory
calculations, we discuss the structural nature of these clusters and the solvent response of the core species. Finally, we discuss how the interaction of Cu with CO2 compares with other metal CO2 cluster anions.
Methods Experimental Our experiment has been described in detail elsewhere.12 Briefly, Cu vapor is created by laser vaporization of material on the rim of a rotating Cu disc employing the third harmonic of a pulsed Nd:YAG laser. The Cu vapor is subsequently entrained in a pulsed supersonic expansion of CO2 generated from an Even-Lavie pulsed valve operating with a stagnation pressure of 5.5 bar. The anionic clusters formed in the expansion are injected into a Wiley-McLaren time-offlight mass spectrometer. Using a pulsed mass gate, a single cluster size is mass selected and subsequently irradiated by the tunable output of an optical parametric converter operating in the range of 600 cm-1 to 4400 cm-1 with a pulse width of 7 ns and a bandwidth of 2 cm-1. Irradiation occurs in a multipass cell based on a design by Liu and coworkers,13 modified to adjust the number of passes (typically 10-14). Absorption of a photon leads to the loss of a CO2 molecule and the resulting fragment anions are detected using a reflectron as a secondary mass analyzer. Infrared photodissociation spectra are collected by monitoring the fragment ion intensity as a function of infrared wavenumber. The spectra are corrected for laser power, and multiple scans from different days are averaged to increase the signal-to-noise ratio and ensure reproducibility. Multiphoton effects were tested for by varying the laser fluence and determined not to be present in this experiment. The experimental repetition rate is 20 Hz.
3 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Computational Density functional theory calculations were performed for [Cu(CO2)n]- clusters (n = 1 – 6) using the TURBOMOLE v. 6.2 suite of programs.14
The B3-LYP functional15,16 with
dispersion correction17 was used in all cases, and def2-TZVPP basis sets18 were assigned to all atoms. Harmonic frequencies were calculated using the AOFORCE program19,20 and were scaled by 0.9754 to correct for anharmonicity. The scaling factor was obtained by comparing the calculated antisymmetric stretch of free CO2 to that of the same experimentally determined value.21 Analysis of partial charges was carried out using a natural population analysis.
Results and Discussion: The infrared spectra of [Cu(CO2)n]- clusters have two distinct regions (see Figure 1) similar to the spectra of other metal CO2 clusters.8-11 Peaks observed in the range 2300 cm-1 to 2400 cm-1 originate from solvent type CO2 molecules. These CO2 molecules are not part of the charge carrier but rather solvate the core species and exhibit a very weakly perturbed CO2 antisymmetric stretching mode around 2345 cm-1, shifted only by a few cm-1 from its position at 2349 cm-1 in free CO2.21 The second region, from 1600 cm-1 to 2150 cm-1, contains peaks originating from CO stretching vibrations that belong to ligands bound to the metal center and encode the structural nature of the charged core species of the [Cu(CO2)n]- clusters. We refer to these ligands as “core” CO2 units, and they are strongly perturbed by interaction with the metal and the excess charge. If the core CO2 units maintain their O-C-O connectivity, the observed core CO vibrations correspond to antisymmetric CO stretching modes of the core ligands.
4 ACS Paragon Plus Environment
Page 4 of 30
Page 5 of 30
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
The Journal of Physical Chemistry
While the binary [CuCO2]- complexes do not dissociate upon infrared absorption in the spectral region under study, photodissociation spectra are observed for [Cu(CO2)2]- clusters with peaks appearing in both the solvent and core regions. The average predicted binding energy of a core CO2 unit in metal CO2 clusters is on the order of 104 cm-1 while the binding energy of a solvent CO2 molecule is estimated to be ca. 1600 cm-1 for these systems.8-11 It follows that a solvent CO2 molecule must be present for dissociation to be readily observed and that the charge carrying species of [Cu(CO2)2]- giving rise to the observed spectrum is likely of the form [CuCO2]-·CO2, composed of a [CuCO2]- core ion with one solvent CO2 molecule. Previous studies suggest that the most likely structure of a [CuCO2]- core ion is a formate-like species where the hydrogen in HCOO- has been replaced by the metal atom,8,9 referred to as an η1 motif in the metal-organic literature.22 Figure 2 shows the calculated lowest energy structures for [CuCO2]-. Structure A is the formate binding motif and is predicted to be the lowest energy structure for a complex of this composition. The only other stable structure is the insertion structure (B), which is predicted to be much higher in energy than the formate structure at this cluster size. Moreover, there should be a high barrier to formation of the insertion structure, since a CO bond must be broken during formation of this isomer. Exploratory calculations put the estimated barrier for interconversion to the insertion motif at ca. 2 eV. Therefore, we would not expect to observe a substantial population of the insertion isomer in the experiment. All starting geometries for [CuCO2]- converged to one of these two structures. In [Cu(CO2)2]- clusters, a [CuCO2]- core ion can be solvated by the second CO2 unit. Figure 2 shows the previously mentioned core structures A and B in their singly solvated forms. To assign the experimental spectrum for [Cu(CO2)2]-, we only need to consider structures that contain a solvent molecule, since the photon energies used here are insufficient for the release of
5 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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 30
a core CO2 ligand (see above). The formate core can be solvated in a few unique locations (see Figure 2, structures A(1a) and A(1b)), which lead to distinct spectral signatures. Only solvation around the CO moiety of isomer B is predicted to preserve the core structure of that isomer, as all other starting geometries converged to isomer C. Figure 3 shows a comparison of the predicted infrared spectra for different solvation positions around formate and insertion cores to the experimental spectrum for [Cu(CO2)2]-. The peak at 1750 cm-1 reflects structure A(1b), and the shoulder at 1730 cm-1 is due to structure A(1a), while the solvated insertion core structure B does not recover any part of the spectrum. In addition to singly solvated [CuCO2]- core ions, both CO2 units in [Cu(CO2)2]- could be part of the core ion, similar to the cases of [Co(CO2)2]- and [Ni(CO2)2]-.10,11
The new core
structures predicted for [Cu(CO2)2]- (see Figure 2, structures C, D, D* and E) have all been found before in calculations performed in previous metal CO2 studies.10,11 Isomer C is an insertion-type structure where the second CO2 moiety is covalently bound to the lone oxygen atom and the copper atom. Isomer C is predicted to be the global minimum for CO2 complexes with Co and Ni at this cluster size, but was assumed not to be significantly populated due to the high barrier to formation (see analogous discussion of isomer B above).10,11 Isomers D and D* contain two CO2 molecules making both carbon-copper and oxygen-copper bonds and are referred to as butterfly structures. The two unique butterfly structures differ in that the two lone CO groups can point in roughly the same (D) or opposite directions (D*).
Butterfly-type
structures with various angles between the two CO2 moieties have been observed before as the primary structures in [Co(CO2)n]- and [Ni(CO2)n]- clusters.10,11 The final new core structure predicted for this cluster size (see Figure 2, structure E) contains a C2O4 subunit bound to the
6 ACS Paragon Plus Environment
Page 7 of 30
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
The Journal of Physical Chemistry
copper atom by one oxygen atom on each of the CO2 units, resulting in a five-membered cyclic structure. The core species described above involving two strongly bound CO2 units should only become observable when a solvent CO2 is present, i.e., at cluster sizes n ≥ 3. Consequently, the experimental spectrum becomes more complicated when the total number of CO2 units is increased from two to three. Three major peaks observed at 1752 cm-1, 1780 cm-1 and 1812 cm-1 dominate the spectrum at this cluster size. In addition to binary [CuCO2]- core ions solvated by two CO2 molecules, it is now possible that a [Cu(CO2)2]- core ion is solvated by the remaining CO2 molecule. The unique solvation positions around the different [Cu(CO2)2]- core structures can be seen in Figure 2.
Moreover, there is one additional (unsolvated) core ion structure to
consider at this cluster size. Attaching a CO2 molecule in a bidentate fashion to the empty side of the Cu atom in the cyclic structure E results in structure F (see Figure 2). This structure is energetically relevant (421 meV above the global minimum) but should not be experimentally observable until [Cu(CO2)4]-, when it can be solvated. Figure 4 shows a comparison of the calculated spectra for isomers A(2), C(1), D(1a), D(1b) and D*(1a) to the experimental spectrum of [Cu(CO2)3]-. Only the formate (A) and butterfly (D) isomers are good candidate structures to assign the three dominant observed signatures. Neither E(1) nor D*(1b) (both are not shown in Figure 4) are consistent with the experimental spectrum. The unsolvated butterfly structure (isomer D) would give rise to two peaks, which correspond to the in-phase (higher energy) and out-of-phase (lower energy) combinations of the antisymmetric CO stretching modes of the two CO2 ligands. Without solvation, the vibrational amplitudes of the atoms in the two ligands are equal for both of these combinations due to the C2v symmetry of the core ion.
7 ACS Paragon Plus Environment
However, solvation affects the
The Journal of Physical Chemistry
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
vibrational patterns and frequency positions significantly and is essential for the assignment of peaks to structural isomers. Out of the three dominant peaks, we will first assign the signatures at 1810 cm-1 and 1750 cm-1. The only structure that accounts for the higher energy feature is isomer D when there is preferential solvation around one of the two bidentate CO2 moieties of the [CO2CuCO2]- core (see Figure 2 D(1b)), and we refer to this situation as asymmetric solvation. Asymmetric solvation results in a change in the vibrational patterns, because the excess charge will be polarized by the presence of the solvent molecule, and the amount of excess charge in the solvated CO2 ligand will be higher (-0.84 e) than in the unsolvated ligand (-0.74 e) using structure D(1b) as an example. This asymmetric charge distribution breaks the symmetry of the CO2 ligands and leaves the CO bonds in the unsolvated ligand stronger than in the solvated ligand, since the excess charge is localized in antibonding CO orbitals. This results in the inphase superposition of the two antisymmetric CO2 stretching vibrations (the higher energy feature) shifting to the blue and in a localization of the amplitudes of this mode largely on the unsolvated ligand. The last feature among the three dominant peaks that needs assignment is the central peak at 1780 cm-1. This feature is only accounted for when both or neither of the bidentate CO2 ligands of isomer D are solvated (for example see Figure 2, structures D(1a) and D(1c)), a situation we term symmetric solvation. Core isomer D is therefore responsible for all three of the dominant features of the spectrum. The next features to consider are the minor peaks at 1920 cm-1 and 1955 cm-1. These features are not observed until [Cu(CO2)4]- which suggests that the core structure that exhibits these vibrations is of the form [Cu(CO2)3]-. The only structure that can account for these peaks is isomer F (see Figures 2 and 5), which has a [Cu(CO2)3]- core, with different solvent positions
8 ACS Paragon Plus Environment
Page 8 of 30
Page 9 of 30
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
The Journal of Physical Chemistry
occupied. If solvation occurs on the bidentate CO2 moiety, the feature at 1920 cm-1 is recovered, while leaving this site empty and solvating anywhere around the C2O4 moiety recovers the peak at 1955 cm-1 (see Figure 6 for spectral comparison and Figure 2 for the solvation positions). In both solvation conformers of core isomer F, a second vibrational mode is predicted to be in the region of 1750 cm-1. Since isomer F is a minor contributor to the spectrum, this feature is likely obscured by the lowest energy dominant peak assigned to core isomer D. The last spectral signature to discuss is the feature to the low energy side of the dominant three peaks. The formate structure (core isomer A) gives a vibrational signature consistent with this minor peak. Solvation exclusively around the CO2 moiety of the formate core gives rise to a redshift of the antisymmetric stretch of the CO2 unit as more charge is polarized onto the CO2 ligand with each additional solvent CO2. We note that the insertion structure B is also predicted to exhibit a vibration in the vicinity of the lowest energy feature. However, there is also a feature predicted at 2025 cm-1 that is not observed experimentally. The absence of this feature and the high barrier to formation of the insertion structure lead us to conclude that lowest energy feature should not be attributed to the insertion structure but to the formate structure. The assignments are further corroborated by the solvation behavior of the signatures discussed above. As the number of CO2 molecules increases, the three major peaks observed in the [Cu(CO2)3]- spectrum and assigned to core D split, and the experimental spectra become increasingly congested (see Figure 1). In addition, the two peaks centered around 1780 cm-1 and 1810 cm-1 shift to the blue by about 5 cm-1 per additional CO2 while the peak at 1750 cm-1 shifts to the red by approximately 5 cm-1.
The experimental trends are summarized in Figure 6
together with the predicted solvation behavior of the core isomers. As the number of solvent molecules increases, exploratory calculations show that any solvent conformation around isomer
9 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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 10 of 30
D characterized by symmetric solvation will exhibit a vibrational mode around 1780 cm-1 (see Figure 6). Both symmetric and asymmetric solvation are predicted to exhibit a core stretching frequency in the region of the lowest energy dominant peak at 1750 cm-1. As the number of solvent molecules increases, there are small but noticeable shifts to this peak and these shifts and splittings are generally recovered in our exploratory calculations. Increasing solvation of isomer F causes its predicted signatures at 1920 cm-1 and 1955 cm-1 to shift to the blue, consistent with the experimental results (see Figure 6). At [Cu(CO2)8]all of the unique solvation positions around core structure F are filled, and the two peaks collapse to one peak at 1920 cm-1 where the solvation position around the bidentate CO2 moiety is filled. Although the absolute wavenumber position of the modes are less well recovered than those for the other conformers, isomer F has been observed before in [Ni(CO2)n]- clusters where it was assigned to similar features, based on its solvation behavior.11 Lastly, the shoulder at 1717 cm-1 on the low frequency side of the three dominant peaks in [Cu(CO2)3]- shifts to the red with increasing cluster size. Exploratory calculations of larger cluster sizes show good agreement with the observed peak positions and spectral shifts (see Figure 6). This solvation behavior has been observed before in [Ag(CO2)n]- clusters where additional CO2 molecules added exclusively around the CO2 moiety of a formate core structure.9 In contrast, the signature of the insertion structure is not predicted to exhibit a significant solvent induced spectral shift, while the experimental feature progressively shifts to lower wavenumbers, strengthening our conclusion that the insertion structure is not the origin of the observed lowenergy feature. With the experimental spectra assigned, we can now compare and contrast [Cu(CO2)n]clusters with other metal-CO2 cluster anions. The dominant core structure in [Cu(CO)n]- is the
10 ACS Paragon Plus Environment
Page 11 of 30
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
The Journal of Physical Chemistry
butterfly core (structure D). This core type has also been observed as the dominant core species in [Co(CO2)n]- and [Ni(CO2)n]-.10,11 As one moves from left to right across the periodic table, the polarizability of the core is calculated to increase by roughly 10% from Co to Ni and again by 10 % from Ni to Cu. The charge distribution strongly impacts the vibrational frequencies of the core CO2 moieties, and the solvent induced shifts will become stronger as core polarizability increases. The greater variability of the spectroscopic signatures of the Cu core compared to Ni and of both the Cu and Ni core ions compared to the Co complexes are therefore completely in accord with the computational result. Copper is a first-row transition metal, and it is also a coinage metal in the series Cu-AgAu. In [Ag(CO2)n]- and [Au(CO2)n]- clusters, the formate-type core MCOO- is the dominant structural motif.8,9 While this core ion is only a minor contributor to the spectra of [Cu(CO2)n]-, it is nonetheless present, and its spectroscopic signature is strongly influenced by solvation. The larger the electronegativity of the metal, the more the excess charge will localize on the metal resulting in less charge transfer to the CO2 moiety and a less red-shifted spectral feature. The CO2 moiety in a [Au(CO2)]- core exhibits the smallest redshift (1840 cm-1 for [Au(CO2)2]-)
8
while the [Cu(CO2)]- formate core exhibits the largest redshift (1750 cm-1 for [Cu(CO2)2]-). Comparing the electronegativities of Cu, Ag and Au (1.90, 1.93 and 2.4 respectively),23 the observed spectroscopic trend is reflected in the electronegativity values. With its position at the intersection of the first row and the 9th column of the periodic system, it is interesting to observe that Cu exhibits characteristics and fits trends of both series.
Conclusions:
11 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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 12 of 30
[Cu(CO2)n]- clusters exhibit a large structural variability of the copper-CO2 interaction motif. Aided by density functional theory calculations, we assign the observed spectroscopic signatures to a number of different core structures including the butterfly structure (D), the formate structure (A) and the cyclic structure (F). All of these structural motifs exhibit solventinduced vibrational shifts that are recovered by calculations. The splitting of the experimental peaks suggests that the core species (D) is strongly affected by the nature of the solvation environment, consistent with calculated trends in polarizability. The cluster behavior observed here is similar to what has been observed before in the case of other metal-CO2 clusters. The copper complexes in the present work exhibit core ions with similar structures as other first-row transition metals in complexes with CO2 molecules (bidentate interaction motifs). In addition, there are core ions that have the same structural motif as those of coinage metal complexes with CO2 (formate-type structures), highlighting the position of copper in the periodic table at the intersection of both metal families.
Acknowledgements: We gratefully acknowledge the National Science Foundation for funding through Grant CHE0845618 (for graduate student support of BJK) and for instrumentation funding through Grant PHY-1125844. JMW is an Alfred P. Sloan Research Fellow.
12 ACS Paragon Plus Environment
Page 13 of 30
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
The Journal of Physical Chemistry
References: (1)
Hori, Y.; Murata, A.; Takahashi, R. Formation of Hydrocarbons in the
Electrochemical Reduction of Carbon-Dioxide at a Copper Electrode in Aqueous-Solution. Journal of the Chemical Society-Faraday Transactions I 1989, 85, 2309-2326. (2)
Kyriacou, G.; Anagnostopoulos, A. Electrochemical Reduction of CO2 at Cu+Au
Electrodes. J. Electroanal. Chem. 1992, 328, 233-243. (3)
Hara, K.; Tsuneto, A.; Kudo, A.; Sakata, T. Electrochemical Reduction of CO2 on
a Cu Electrode under High-Pressure - Factors that Determine the Product Selectivity. J. Electrochem. Soc. 1994, 141, 2097-2103. (4)
Hori, Y.; Kikuchi, K.; Murata, A.; Suzuki, S. Production of Methane and Ethylene
in Electrochemical Reduction of Carbon-Dioxide at Copper Electrode in Aqueous Hydrogencarbonate Solution. Chem. Lett. 1986, 897-898. (5)
Hori, Y.; Kikuchi, K.; Suzuki, S. Production of CO and CH4 in Electrochemical
Reduction of CO2 at Metal-Electrodes in Aqueous Hydrogencarbonate Solution. Chem. Lett. 1985, 1695-1698. (6)
Cook, R. L.; Macduff, R. C.; Sammells, A. F. Evidence for Formaldehyde,
Formic-Acid, and Acetaldehyde as Possible Intermediates during Electrochemical CarbonDioxide Reduction at Copper. J. Electrochem. Soc. 1989, 136, 1982-1984. (7)
Lim, R. J.; Xie, M. S.; Sk, M. A.; Lee, J. M.; Fisher, A.; Wang, X.; Lim, K. H. A
Review on the Electrochemical Reduction of CO2 in Fuel Cells, Metal Electrodes and Molecular Catalysts. Catal. Today 2014, 233, 169-180. (8)
Knurr, B. J.; Weber, J. M. Solvent-Driven Reductive Activation of Carbon
Dioxide by Gold Anions. Journal of the American Chemical Society 2012, 134, 18804-18808.
13 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
(9)
Knurr, B. J.; Weber, J. M. Solvent-Mediated Reduction of Carbon Dioxide in
Anionic Complexes with Silver Atoms. J. Phys. Chem. A 2013, 117, 10764-10771. (10)
Knurr, B. J.; Weber, J. M. Infrared Spectra and Structures of Anionic Complexes
of Cobalt with Carbon Dioxide Ligands. J. Phys. Chem. A 2014, 118, 4056-4062. (11)
Knurr, B. J. W., J. M. Interaction of Nickel with Carbon Dioxide in [Ni(CO2)n]-
Clusters Studied by Infrared Spectroscopy. J. Phys. Chem. A 2014, 118, 8753-8757. (12)
Weber, J. M. A Pulsed Ion Source for the Preparation of Metal Containing Cluster
Ions Using Supersonic Entrainment of Laser Vaporized Metal. Rev. Sci. Instrum. 2005, 76, 043301. (13)
Riedel, J.; Yan, S. N.; Kawamata, H.; Liu, K. P. A Simple Yet Effective
Multipass Reflector for Vibrational Excitation in Molecular Beams. Rev. Sci. Instrum. 2008, 79. (14)
Ahlrichs, R.; Bär, M.; Häser, M.; Horn, H.; Kölmel, C. Electronic-Structure
Calculations on Workstation Computers - the Program System Turbomole. Chem. Phys. Lett. 1989, 162, 165-169. (15)
Becke, A. D. Density-Functional Exchange-Energy Approximation with Correct
Asymptotic-Behavior. Phys. Rev. A 1988, 38, 3098-3100. (16)
Lee, C. T.; Yang, W. T.; Parr, R. G. Development of the Colle-Salvetti
Correlation-Energy Formula into a Functional of the Electron-Density. Phys. Rev. B 1988, 37, 785-789. (17)
Grimme, S. Semiempirical GGA-type Density Functional Constructed with a
Long-Range Dispersion Correction. J. Comput. Chem. 2006, 27, 1787-1799.
14 ACS Paragon Plus Environment
Page 14 of 30
Page 15 of 30
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
The Journal of Physical Chemistry
(18)
Weigend, F.; Ahlrichs, R. Balanced Basis Sets of Split Valence, Triple Zeta
Valence and Quadruple Zeta Valence Quality for H to Rn: Design and Assessment of Accuracy. Phys. Chem. Chem. Phys. 2005, 7, 3297-3305. (19)
Deglmann, P.; Furche, F.; Ahlrichs, R. An Efficient Implementation of Second
Analytical Derivatives for Density Functional Methods. Chem. Phys. Lett. 2002, 362, 511-518. (20)
Deglmann, P.; Furche, F. Efficient Characterization of Stationary Points on
Potential Energy Surfaces. J. Chem. Phys. 2002, 117, 9535-9538. (21)
Shimanouchi, T.; National Bureau of Standards: 1972, p 1-160.
(22)
Gibson, D. H. The Organometallic Chemistry of Carbon Dioxide. Chem. Rev.
1996, 96, 2063-2095. (23)
Pauling, L. The Nature of the Chemical Bond; 3 ed.; Cornell University Press:
Ithaca, United States, 1960.
15 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure Captions:
Figure 1: Experimental spectra of [Cu(CO2)n]-. Numbers denote the number of CO2 molecules present for a given spectrum. All spectra are individually normalized so the left and right traces are on different scales (see text for discussion).
Figure 2: Selected calculated lowest energy structures for [Cu(CO2)n]- (n = 1 – 4). The capital letter denotes the core structure and the number/letter combination in parentheses indicates the number of solvent molecules and their conformations. Bare core structures are shown on the top row while solvation conformers are shown in the descending columns. Relative energies are given in meV and are with respect to the lowest energy isomer for a given cluster size.
Figure 3: Comparison of calculated spectra of solvation conformers of the formate type core and of the insertion structure to the experimental spectrum of [Cu(CO2)2]-.
Figure 4: Comparison of predicted spectra for calculated structures of [Cu(CO2)3]- (lower traces) to the experimental spectrum for [Cu(CO2)3]- (top trace). Relative energies are given for each calculated trace (see also Figure 2 for structural labels).
Figure 5: Comparison of predicted spectra for calculated structures of [Cu(CO2)6]- (lower traces) to the experimental spectrum for [Cu(CO2)6]- (top trace). See also Figure 2 for structural labels. For structure D, we compare symmetric and asymmetric solvation. For structure F, we compare presence and absence of a solvent molecule at the CO2 moiety (see text).
16 ACS Paragon Plus Environment
Page 16 of 30
Page 17 of 30
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
The Journal of Physical Chemistry
Figure 6: Spectral shifts of the experimental and simulated features for [Cu(CO2)n]- (n = 3 – 6). The top trace show the spectrum for n = 6 for comparison. The lines are meant to guide the eye, the letters indicate the assignments of the core isomers as shown in Figure 2.
17 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 1:
18 ACS Paragon Plus Environment
Page 18 of 30
Page 19 of 30
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
The Journal of Physical Chemistry
Figure 2:
19 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 3:
20 ACS Paragon Plus Environment
Page 20 of 30
Page 21 of 30
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
The Journal of Physical Chemistry
Figure 4:
21 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 5:
22 ACS Paragon Plus Environment
Page 22 of 30
Page 23 of 30
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
The Journal of Physical Chemistry
Figure 6:
23 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 1: Experimental spectra of [Cu(CO2)n]-. Numbers denote the number of CO2 molecules present for a given spectrum. All spectra are individually normalized so the left and right traces are on different scales (see text for discussion). 62x46mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 24 of 30
Page 25 of 30
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
The Journal of Physical Chemistry
Figure 2: Selected calculated lowest energy structures for [Cu(CO2)n]- (n = 1 – 4). The capital letter denotes the core structure and the number/letter combination in parentheses indicates the number of solvent molecules and their conformations. Bare core structures are shown on the top row while solvation conformers are shown in the descending columns. Relative energies are given in meV and are with respect to the lowest energy isomer for a given cluster size. 339x190mm (150 x 150 DPI)
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 3: Comparison of unique formate solvation isomers and the insertion structure to the experimental spectrum of [Cu(CO2)2]-. 83x84mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 26 of 30
Page 27 of 30
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
The Journal of Physical Chemistry
Figure 4: A comparison of calculated structures for [Cu(CO2)3]- to the experimental spectrum for [Cu(CO2)3]-. Relative energies are given for each calculated trace (see also Figure 2 for structural labels). 91x102mm (300 x 300 DPI)
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 5: Comparison of predicted spectra for calculated structures of [Cu(CO2)6]- (lower traces) to the experimental spectrum for [Cu(CO2)6]- (top trace). See also Figure 2 for structural labels. For structure D, we compare symmetric and asymmetric solvation. For structure F, we compare presence and absence of a solvent molecule at the CO2 moiety (see text). 107x140mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 28 of 30
Page 29 of 30
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
The Journal of Physical Chemistry
Figure 6: Spectral shifts of the experimental and simulated features for [Cu(CO2)n]- (n = 3 – 6). The top trace show the spectrum for n = 6 for comparison. The lines are meant to guide the eye, the letters indicate the assignments of the core isomers as shown in Figure 2. 147x121mm (300 x 300 DPI)
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
TOC Figure 338x190mm (96 x 96 DPI)
ACS Paragon Plus Environment
Page 30 of 30
Article 2
Structural Diversity of Copper-CO Complexes – Infrared Spectra and Structures of [Cu(CO)]- Clusters 2
n
Benjamin J. Knurr, and J. Mathias Weber J. Phys. Chem. A, Just Accepted Manuscript • Publication Date (Web): 15 Oct 2014 Downloaded from http://pubs.acs.org on October 16, 2014
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 free 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 accessible to all readers and 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.
The Journal of Physical Chemistry A 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 30
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
The Journal of Physical Chemistry
Structural Diversity of Copper-CO2 Complexes – Infrared Spectra and Structures of [Cu(CO2)n]- Clusters
Benjamin J. Knurr and J. Mathias Weber*
JILA and Department of Chemistry and Biochemistry, University of Colorado, Boulder, CO 80309, USA Abstract: We present infrared spectra of [Cu(CO2)n]- (n = 2 – 9) clusters in the wavenumber range 1600 cm-1 to 2400 cm-1. The CO stretching modes in this frequency region encode the structural nature of the cluster core and are interpreted with the aid of density functional theory. We find a variety of core species in [Cu(CO2)n]- clusters, but the dominant core structure is a [Cu(CO2)2]core where the two CO2 ligands are bound to the Cu atom in a bidentate fashion. We compare the results of [Cu(CO2)n]- clusters to those of other [M(CO2)n]- clusters (M = Au, Ag, Co, Ni) in order to establish trends of how the metal-CO2 interaction depends on the metal partner.
Keywords: Transition metals, carbon dioxide, metal-organic chemistry, photodissociation spectroscopy, mass spectrometry
*
Corresponding Author: J. M. Weber: phone, ++1-303-492-7841; e-mail: [email protected]. 1 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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 30
Introduction: Transition metals are employed in a wide variety of catalytic applications, but in many cases the mechanistic, molecular level details of how the catalysts function are poorly understood, hindering rational catalyst design. Experiments on catalyst complexes in situ are challenging, in particular because speciation is often complicated, and the solvent environment is difficult to control and characterize and can affect the chemical nature of the complex under study.
Unlike experiments in the condensed phase, molecular beam experiments on mass
selected cluster ions in vacuo allow for better control over the species and the solvation environment under study, making these experiments ideal for the investigation of model catalyst systems. Copper has been used in a number of catalytic systems and fuel cell applications to convert CO2 to CO1-3, methane3-5, alcohols and other small organic species.3,6,7 While there has been some success in the conversion of CO2, efficiency and selectivity of the catalysts used still leave much to be desired. Overall, there has been little effort to characterize the molecular level details of the reactions involved, even though understanding the catalyst-reactant interaction is necessary to accurately model and design better catalysts. We have previously studied the interactions of CO2 with a number of different metals (Au, Ag, Co and Ni) and have been able to characterize the resulting metal-CO2 interactions and the effect of solvation on them.8-11 In the context of previous metal CO2 studies, copper is unique as it is both a member of the 9th column (like Au and Ag) and the top row transition metals (like Co and Ni). The dominant structural motifs in these two groups of metals are very different. While the coinage metals form a [MCO2]- core ion,8,9 the first row transition metals bind to CO2 ligands primarily through [CO2MCO2]- complexes.10,11
2 ACS Paragon Plus Environment
Here we present IR
Page 3 of 30
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
The Journal of Physical Chemistry
photodissociation spectra of [Cu(CO2)n]- clusters.
Aided by density functional theory
calculations, we discuss the structural nature of these clusters and the solvent response of the core species. Finally, we discuss how the interaction of Cu with CO2 compares with other metal CO2 cluster anions.
Methods Experimental Our experiment has been described in detail elsewhere.12 Briefly, Cu vapor is created by laser vaporization of material on the rim of a rotating Cu disc employing the third harmonic of a pulsed Nd:YAG laser. The Cu vapor is subsequently entrained in a pulsed supersonic expansion of CO2 generated from an Even-Lavie pulsed valve operating with a stagnation pressure of 5.5 bar. The anionic clusters formed in the expansion are injected into a Wiley-McLaren time-offlight mass spectrometer. Using a pulsed mass gate, a single cluster size is mass selected and subsequently irradiated by the tunable output of an optical parametric converter operating in the range of 600 cm-1 to 4400 cm-1 with a pulse width of 7 ns and a bandwidth of 2 cm-1. Irradiation occurs in a multipass cell based on a design by Liu and coworkers,13 modified to adjust the number of passes (typically 10-14). Absorption of a photon leads to the loss of a CO2 molecule and the resulting fragment anions are detected using a reflectron as a secondary mass analyzer. Infrared photodissociation spectra are collected by monitoring the fragment ion intensity as a function of infrared wavenumber. The spectra are corrected for laser power, and multiple scans from different days are averaged to increase the signal-to-noise ratio and ensure reproducibility. Multiphoton effects were tested for by varying the laser fluence and determined not to be present in this experiment. The experimental repetition rate is 20 Hz.
3 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Computational Density functional theory calculations were performed for [Cu(CO2)n]- clusters (n = 1 – 6) using the TURBOMOLE v. 6.2 suite of programs.14
The B3-LYP functional15,16 with
dispersion correction17 was used in all cases, and def2-TZVPP basis sets18 were assigned to all atoms. Harmonic frequencies were calculated using the AOFORCE program19,20 and were scaled by 0.9754 to correct for anharmonicity. The scaling factor was obtained by comparing the calculated antisymmetric stretch of free CO2 to that of the same experimentally determined value.21 Analysis of partial charges was carried out using a natural population analysis.
Results and Discussion: The infrared spectra of [Cu(CO2)n]- clusters have two distinct regions (see Figure 1) similar to the spectra of other metal CO2 clusters.8-11 Peaks observed in the range 2300 cm-1 to 2400 cm-1 originate from solvent type CO2 molecules. These CO2 molecules are not part of the charge carrier but rather solvate the core species and exhibit a very weakly perturbed CO2 antisymmetric stretching mode around 2345 cm-1, shifted only by a few cm-1 from its position at 2349 cm-1 in free CO2.21 The second region, from 1600 cm-1 to 2150 cm-1, contains peaks originating from CO stretching vibrations that belong to ligands bound to the metal center and encode the structural nature of the charged core species of the [Cu(CO2)n]- clusters. We refer to these ligands as “core” CO2 units, and they are strongly perturbed by interaction with the metal and the excess charge. If the core CO2 units maintain their O-C-O connectivity, the observed core CO vibrations correspond to antisymmetric CO stretching modes of the core ligands.
4 ACS Paragon Plus Environment
Page 4 of 30
Page 5 of 30
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
The Journal of Physical Chemistry
While the binary [CuCO2]- complexes do not dissociate upon infrared absorption in the spectral region under study, photodissociation spectra are observed for [Cu(CO2)2]- clusters with peaks appearing in both the solvent and core regions. The average predicted binding energy of a core CO2 unit in metal CO2 clusters is on the order of 104 cm-1 while the binding energy of a solvent CO2 molecule is estimated to be ca. 1600 cm-1 for these systems.8-11 It follows that a solvent CO2 molecule must be present for dissociation to be readily observed and that the charge carrying species of [Cu(CO2)2]- giving rise to the observed spectrum is likely of the form [CuCO2]-·CO2, composed of a [CuCO2]- core ion with one solvent CO2 molecule. Previous studies suggest that the most likely structure of a [CuCO2]- core ion is a formate-like species where the hydrogen in HCOO- has been replaced by the metal atom,8,9 referred to as an η1 motif in the metal-organic literature.22 Figure 2 shows the calculated lowest energy structures for [CuCO2]-. Structure A is the formate binding motif and is predicted to be the lowest energy structure for a complex of this composition. The only other stable structure is the insertion structure (B), which is predicted to be much higher in energy than the formate structure at this cluster size. Moreover, there should be a high barrier to formation of the insertion structure, since a CO bond must be broken during formation of this isomer. Exploratory calculations put the estimated barrier for interconversion to the insertion motif at ca. 2 eV. Therefore, we would not expect to observe a substantial population of the insertion isomer in the experiment. All starting geometries for [CuCO2]- converged to one of these two structures. In [Cu(CO2)2]- clusters, a [CuCO2]- core ion can be solvated by the second CO2 unit. Figure 2 shows the previously mentioned core structures A and B in their singly solvated forms. To assign the experimental spectrum for [Cu(CO2)2]-, we only need to consider structures that contain a solvent molecule, since the photon energies used here are insufficient for the release of
5 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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 30
a core CO2 ligand (see above). The formate core can be solvated in a few unique locations (see Figure 2, structures A(1a) and A(1b)), which lead to distinct spectral signatures. Only solvation around the CO moiety of isomer B is predicted to preserve the core structure of that isomer, as all other starting geometries converged to isomer C. Figure 3 shows a comparison of the predicted infrared spectra for different solvation positions around formate and insertion cores to the experimental spectrum for [Cu(CO2)2]-. The peak at 1750 cm-1 reflects structure A(1b), and the shoulder at 1730 cm-1 is due to structure A(1a), while the solvated insertion core structure B does not recover any part of the spectrum. In addition to singly solvated [CuCO2]- core ions, both CO2 units in [Cu(CO2)2]- could be part of the core ion, similar to the cases of [Co(CO2)2]- and [Ni(CO2)2]-.10,11
The new core
structures predicted for [Cu(CO2)2]- (see Figure 2, structures C, D, D* and E) have all been found before in calculations performed in previous metal CO2 studies.10,11 Isomer C is an insertion-type structure where the second CO2 moiety is covalently bound to the lone oxygen atom and the copper atom. Isomer C is predicted to be the global minimum for CO2 complexes with Co and Ni at this cluster size, but was assumed not to be significantly populated due to the high barrier to formation (see analogous discussion of isomer B above).10,11 Isomers D and D* contain two CO2 molecules making both carbon-copper and oxygen-copper bonds and are referred to as butterfly structures. The two unique butterfly structures differ in that the two lone CO groups can point in roughly the same (D) or opposite directions (D*).
Butterfly-type
structures with various angles between the two CO2 moieties have been observed before as the primary structures in [Co(CO2)n]- and [Ni(CO2)n]- clusters.10,11 The final new core structure predicted for this cluster size (see Figure 2, structure E) contains a C2O4 subunit bound to the
6 ACS Paragon Plus Environment
Page 7 of 30
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
The Journal of Physical Chemistry
copper atom by one oxygen atom on each of the CO2 units, resulting in a five-membered cyclic structure. The core species described above involving two strongly bound CO2 units should only become observable when a solvent CO2 is present, i.e., at cluster sizes n ≥ 3. Consequently, the experimental spectrum becomes more complicated when the total number of CO2 units is increased from two to three. Three major peaks observed at 1752 cm-1, 1780 cm-1 and 1812 cm-1 dominate the spectrum at this cluster size. In addition to binary [CuCO2]- core ions solvated by two CO2 molecules, it is now possible that a [Cu(CO2)2]- core ion is solvated by the remaining CO2 molecule. The unique solvation positions around the different [Cu(CO2)2]- core structures can be seen in Figure 2.
Moreover, there is one additional (unsolvated) core ion structure to
consider at this cluster size. Attaching a CO2 molecule in a bidentate fashion to the empty side of the Cu atom in the cyclic structure E results in structure F (see Figure 2). This structure is energetically relevant (421 meV above the global minimum) but should not be experimentally observable until [Cu(CO2)4]-, when it can be solvated. Figure 4 shows a comparison of the calculated spectra for isomers A(2), C(1), D(1a), D(1b) and D*(1a) to the experimental spectrum of [Cu(CO2)3]-. Only the formate (A) and butterfly (D) isomers are good candidate structures to assign the three dominant observed signatures. Neither E(1) nor D*(1b) (both are not shown in Figure 4) are consistent with the experimental spectrum. The unsolvated butterfly structure (isomer D) would give rise to two peaks, which correspond to the in-phase (higher energy) and out-of-phase (lower energy) combinations of the antisymmetric CO stretching modes of the two CO2 ligands. Without solvation, the vibrational amplitudes of the atoms in the two ligands are equal for both of these combinations due to the C2v symmetry of the core ion.
7 ACS Paragon Plus Environment
However, solvation affects the
The Journal of Physical Chemistry
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
vibrational patterns and frequency positions significantly and is essential for the assignment of peaks to structural isomers. Out of the three dominant peaks, we will first assign the signatures at 1810 cm-1 and 1750 cm-1. The only structure that accounts for the higher energy feature is isomer D when there is preferential solvation around one of the two bidentate CO2 moieties of the [CO2CuCO2]- core (see Figure 2 D(1b)), and we refer to this situation as asymmetric solvation. Asymmetric solvation results in a change in the vibrational patterns, because the excess charge will be polarized by the presence of the solvent molecule, and the amount of excess charge in the solvated CO2 ligand will be higher (-0.84 e) than in the unsolvated ligand (-0.74 e) using structure D(1b) as an example. This asymmetric charge distribution breaks the symmetry of the CO2 ligands and leaves the CO bonds in the unsolvated ligand stronger than in the solvated ligand, since the excess charge is localized in antibonding CO orbitals. This results in the inphase superposition of the two antisymmetric CO2 stretching vibrations (the higher energy feature) shifting to the blue and in a localization of the amplitudes of this mode largely on the unsolvated ligand. The last feature among the three dominant peaks that needs assignment is the central peak at 1780 cm-1. This feature is only accounted for when both or neither of the bidentate CO2 ligands of isomer D are solvated (for example see Figure 2, structures D(1a) and D(1c)), a situation we term symmetric solvation. Core isomer D is therefore responsible for all three of the dominant features of the spectrum. The next features to consider are the minor peaks at 1920 cm-1 and 1955 cm-1. These features are not observed until [Cu(CO2)4]- which suggests that the core structure that exhibits these vibrations is of the form [Cu(CO2)3]-. The only structure that can account for these peaks is isomer F (see Figures 2 and 5), which has a [Cu(CO2)3]- core, with different solvent positions
8 ACS Paragon Plus Environment
Page 8 of 30
Page 9 of 30
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
The Journal of Physical Chemistry
occupied. If solvation occurs on the bidentate CO2 moiety, the feature at 1920 cm-1 is recovered, while leaving this site empty and solvating anywhere around the C2O4 moiety recovers the peak at 1955 cm-1 (see Figure 6 for spectral comparison and Figure 2 for the solvation positions). In both solvation conformers of core isomer F, a second vibrational mode is predicted to be in the region of 1750 cm-1. Since isomer F is a minor contributor to the spectrum, this feature is likely obscured by the lowest energy dominant peak assigned to core isomer D. The last spectral signature to discuss is the feature to the low energy side of the dominant three peaks. The formate structure (core isomer A) gives a vibrational signature consistent with this minor peak. Solvation exclusively around the CO2 moiety of the formate core gives rise to a redshift of the antisymmetric stretch of the CO2 unit as more charge is polarized onto the CO2 ligand with each additional solvent CO2. We note that the insertion structure B is also predicted to exhibit a vibration in the vicinity of the lowest energy feature. However, there is also a feature predicted at 2025 cm-1 that is not observed experimentally. The absence of this feature and the high barrier to formation of the insertion structure lead us to conclude that lowest energy feature should not be attributed to the insertion structure but to the formate structure. The assignments are further corroborated by the solvation behavior of the signatures discussed above. As the number of CO2 molecules increases, the three major peaks observed in the [Cu(CO2)3]- spectrum and assigned to core D split, and the experimental spectra become increasingly congested (see Figure 1). In addition, the two peaks centered around 1780 cm-1 and 1810 cm-1 shift to the blue by about 5 cm-1 per additional CO2 while the peak at 1750 cm-1 shifts to the red by approximately 5 cm-1.
The experimental trends are summarized in Figure 6
together with the predicted solvation behavior of the core isomers. As the number of solvent molecules increases, exploratory calculations show that any solvent conformation around isomer
9 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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 10 of 30
D characterized by symmetric solvation will exhibit a vibrational mode around 1780 cm-1 (see Figure 6). Both symmetric and asymmetric solvation are predicted to exhibit a core stretching frequency in the region of the lowest energy dominant peak at 1750 cm-1. As the number of solvent molecules increases, there are small but noticeable shifts to this peak and these shifts and splittings are generally recovered in our exploratory calculations. Increasing solvation of isomer F causes its predicted signatures at 1920 cm-1 and 1955 cm-1 to shift to the blue, consistent with the experimental results (see Figure 6). At [Cu(CO2)8]all of the unique solvation positions around core structure F are filled, and the two peaks collapse to one peak at 1920 cm-1 where the solvation position around the bidentate CO2 moiety is filled. Although the absolute wavenumber position of the modes are less well recovered than those for the other conformers, isomer F has been observed before in [Ni(CO2)n]- clusters where it was assigned to similar features, based on its solvation behavior.11 Lastly, the shoulder at 1717 cm-1 on the low frequency side of the three dominant peaks in [Cu(CO2)3]- shifts to the red with increasing cluster size. Exploratory calculations of larger cluster sizes show good agreement with the observed peak positions and spectral shifts (see Figure 6). This solvation behavior has been observed before in [Ag(CO2)n]- clusters where additional CO2 molecules added exclusively around the CO2 moiety of a formate core structure.9 In contrast, the signature of the insertion structure is not predicted to exhibit a significant solvent induced spectral shift, while the experimental feature progressively shifts to lower wavenumbers, strengthening our conclusion that the insertion structure is not the origin of the observed lowenergy feature. With the experimental spectra assigned, we can now compare and contrast [Cu(CO2)n]clusters with other metal-CO2 cluster anions. The dominant core structure in [Cu(CO)n]- is the
10 ACS Paragon Plus Environment
Page 11 of 30
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
The Journal of Physical Chemistry
butterfly core (structure D). This core type has also been observed as the dominant core species in [Co(CO2)n]- and [Ni(CO2)n]-.10,11 As one moves from left to right across the periodic table, the polarizability of the core is calculated to increase by roughly 10% from Co to Ni and again by 10 % from Ni to Cu. The charge distribution strongly impacts the vibrational frequencies of the core CO2 moieties, and the solvent induced shifts will become stronger as core polarizability increases. The greater variability of the spectroscopic signatures of the Cu core compared to Ni and of both the Cu and Ni core ions compared to the Co complexes are therefore completely in accord with the computational result. Copper is a first-row transition metal, and it is also a coinage metal in the series Cu-AgAu. In [Ag(CO2)n]- and [Au(CO2)n]- clusters, the formate-type core MCOO- is the dominant structural motif.8,9 While this core ion is only a minor contributor to the spectra of [Cu(CO2)n]-, it is nonetheless present, and its spectroscopic signature is strongly influenced by solvation. The larger the electronegativity of the metal, the more the excess charge will localize on the metal resulting in less charge transfer to the CO2 moiety and a less red-shifted spectral feature. The CO2 moiety in a [Au(CO2)]- core exhibits the smallest redshift (1840 cm-1 for [Au(CO2)2]-)
8
while the [Cu(CO2)]- formate core exhibits the largest redshift (1750 cm-1 for [Cu(CO2)2]-). Comparing the electronegativities of Cu, Ag and Au (1.90, 1.93 and 2.4 respectively),23 the observed spectroscopic trend is reflected in the electronegativity values. With its position at the intersection of the first row and the 9th column of the periodic system, it is interesting to observe that Cu exhibits characteristics and fits trends of both series.
Conclusions:
11 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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 12 of 30
[Cu(CO2)n]- clusters exhibit a large structural variability of the copper-CO2 interaction motif. Aided by density functional theory calculations, we assign the observed spectroscopic signatures to a number of different core structures including the butterfly structure (D), the formate structure (A) and the cyclic structure (F). All of these structural motifs exhibit solventinduced vibrational shifts that are recovered by calculations. The splitting of the experimental peaks suggests that the core species (D) is strongly affected by the nature of the solvation environment, consistent with calculated trends in polarizability. The cluster behavior observed here is similar to what has been observed before in the case of other metal-CO2 clusters. The copper complexes in the present work exhibit core ions with similar structures as other first-row transition metals in complexes with CO2 molecules (bidentate interaction motifs). In addition, there are core ions that have the same structural motif as those of coinage metal complexes with CO2 (formate-type structures), highlighting the position of copper in the periodic table at the intersection of both metal families.
Acknowledgements: We gratefully acknowledge the National Science Foundation for funding through Grant CHE0845618 (for graduate student support of BJK) and for instrumentation funding through Grant PHY-1125844. JMW is an Alfred P. Sloan Research Fellow.
12 ACS Paragon Plus Environment
Page 13 of 30
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
The Journal of Physical Chemistry
References: (1)
Hori, Y.; Murata, A.; Takahashi, R. Formation of Hydrocarbons in the
Electrochemical Reduction of Carbon-Dioxide at a Copper Electrode in Aqueous-Solution. Journal of the Chemical Society-Faraday Transactions I 1989, 85, 2309-2326. (2)
Kyriacou, G.; Anagnostopoulos, A. Electrochemical Reduction of CO2 at Cu+Au
Electrodes. J. Electroanal. Chem. 1992, 328, 233-243. (3)
Hara, K.; Tsuneto, A.; Kudo, A.; Sakata, T. Electrochemical Reduction of CO2 on
a Cu Electrode under High-Pressure - Factors that Determine the Product Selectivity. J. Electrochem. Soc. 1994, 141, 2097-2103. (4)
Hori, Y.; Kikuchi, K.; Murata, A.; Suzuki, S. Production of Methane and Ethylene
in Electrochemical Reduction of Carbon-Dioxide at Copper Electrode in Aqueous Hydrogencarbonate Solution. Chem. Lett. 1986, 897-898. (5)
Hori, Y.; Kikuchi, K.; Suzuki, S. Production of CO and CH4 in Electrochemical
Reduction of CO2 at Metal-Electrodes in Aqueous Hydrogencarbonate Solution. Chem. Lett. 1985, 1695-1698. (6)
Cook, R. L.; Macduff, R. C.; Sammells, A. F. Evidence for Formaldehyde,
Formic-Acid, and Acetaldehyde as Possible Intermediates during Electrochemical CarbonDioxide Reduction at Copper. J. Electrochem. Soc. 1989, 136, 1982-1984. (7)
Lim, R. J.; Xie, M. S.; Sk, M. A.; Lee, J. M.; Fisher, A.; Wang, X.; Lim, K. H. A
Review on the Electrochemical Reduction of CO2 in Fuel Cells, Metal Electrodes and Molecular Catalysts. Catal. Today 2014, 233, 169-180. (8)
Knurr, B. J.; Weber, J. M. Solvent-Driven Reductive Activation of Carbon
Dioxide by Gold Anions. Journal of the American Chemical Society 2012, 134, 18804-18808.
13 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
(9)
Knurr, B. J.; Weber, J. M. Solvent-Mediated Reduction of Carbon Dioxide in
Anionic Complexes with Silver Atoms. J. Phys. Chem. A 2013, 117, 10764-10771. (10)
Knurr, B. J.; Weber, J. M. Infrared Spectra and Structures of Anionic Complexes
of Cobalt with Carbon Dioxide Ligands. J. Phys. Chem. A 2014, 118, 4056-4062. (11)
Knurr, B. J. W., J. M. Interaction of Nickel with Carbon Dioxide in [Ni(CO2)n]-
Clusters Studied by Infrared Spectroscopy. J. Phys. Chem. A 2014, 118, 8753-8757. (12)
Weber, J. M. A Pulsed Ion Source for the Preparation of Metal Containing Cluster
Ions Using Supersonic Entrainment of Laser Vaporized Metal. Rev. Sci. Instrum. 2005, 76, 043301. (13)
Riedel, J.; Yan, S. N.; Kawamata, H.; Liu, K. P. A Simple Yet Effective
Multipass Reflector for Vibrational Excitation in Molecular Beams. Rev. Sci. Instrum. 2008, 79. (14)
Ahlrichs, R.; Bär, M.; Häser, M.; Horn, H.; Kölmel, C. Electronic-Structure
Calculations on Workstation Computers - the Program System Turbomole. Chem. Phys. Lett. 1989, 162, 165-169. (15)
Becke, A. D. Density-Functional Exchange-Energy Approximation with Correct
Asymptotic-Behavior. Phys. Rev. A 1988, 38, 3098-3100. (16)
Lee, C. T.; Yang, W. T.; Parr, R. G. Development of the Colle-Salvetti
Correlation-Energy Formula into a Functional of the Electron-Density. Phys. Rev. B 1988, 37, 785-789. (17)
Grimme, S. Semiempirical GGA-type Density Functional Constructed with a
Long-Range Dispersion Correction. J. Comput. Chem. 2006, 27, 1787-1799.
14 ACS Paragon Plus Environment
Page 14 of 30
Page 15 of 30
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
The Journal of Physical Chemistry
(18)
Weigend, F.; Ahlrichs, R. Balanced Basis Sets of Split Valence, Triple Zeta
Valence and Quadruple Zeta Valence Quality for H to Rn: Design and Assessment of Accuracy. Phys. Chem. Chem. Phys. 2005, 7, 3297-3305. (19)
Deglmann, P.; Furche, F.; Ahlrichs, R. An Efficient Implementation of Second
Analytical Derivatives for Density Functional Methods. Chem. Phys. Lett. 2002, 362, 511-518. (20)
Deglmann, P.; Furche, F. Efficient Characterization of Stationary Points on
Potential Energy Surfaces. J. Chem. Phys. 2002, 117, 9535-9538. (21)
Shimanouchi, T.; National Bureau of Standards: 1972, p 1-160.
(22)
Gibson, D. H. The Organometallic Chemistry of Carbon Dioxide. Chem. Rev.
1996, 96, 2063-2095. (23)
Pauling, L. The Nature of the Chemical Bond; 3 ed.; Cornell University Press:
Ithaca, United States, 1960.
15 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure Captions:
Figure 1: Experimental spectra of [Cu(CO2)n]-. Numbers denote the number of CO2 molecules present for a given spectrum. All spectra are individually normalized so the left and right traces are on different scales (see text for discussion).
Figure 2: Selected calculated lowest energy structures for [Cu(CO2)n]- (n = 1 – 4). The capital letter denotes the core structure and the number/letter combination in parentheses indicates the number of solvent molecules and their conformations. Bare core structures are shown on the top row while solvation conformers are shown in the descending columns. Relative energies are given in meV and are with respect to the lowest energy isomer for a given cluster size.
Figure 3: Comparison of calculated spectra of solvation conformers of the formate type core and of the insertion structure to the experimental spectrum of [Cu(CO2)2]-.
Figure 4: Comparison of predicted spectra for calculated structures of [Cu(CO2)3]- (lower traces) to the experimental spectrum for [Cu(CO2)3]- (top trace). Relative energies are given for each calculated trace (see also Figure 2 for structural labels).
Figure 5: Comparison of predicted spectra for calculated structures of [Cu(CO2)6]- (lower traces) to the experimental spectrum for [Cu(CO2)6]- (top trace). See also Figure 2 for structural labels. For structure D, we compare symmetric and asymmetric solvation. For structure F, we compare presence and absence of a solvent molecule at the CO2 moiety (see text).
16 ACS Paragon Plus Environment
Page 16 of 30
Page 17 of 30
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
The Journal of Physical Chemistry
Figure 6: Spectral shifts of the experimental and simulated features for [Cu(CO2)n]- (n = 3 – 6). The top trace show the spectrum for n = 6 for comparison. The lines are meant to guide the eye, the letters indicate the assignments of the core isomers as shown in Figure 2.
17 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 1:
18 ACS Paragon Plus Environment
Page 18 of 30
Page 19 of 30
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
The Journal of Physical Chemistry
Figure 2:
19 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 3:
20 ACS Paragon Plus Environment
Page 20 of 30
Page 21 of 30
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
The Journal of Physical Chemistry
Figure 4:
21 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 5:
22 ACS Paragon Plus Environment
Page 22 of 30
Page 23 of 30
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
The Journal of Physical Chemistry
Figure 6:
23 ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 1: Experimental spectra of [Cu(CO2)n]-. Numbers denote the number of CO2 molecules present for a given spectrum. All spectra are individually normalized so the left and right traces are on different scales (see text for discussion). 62x46mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 24 of 30
Page 25 of 30
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
The Journal of Physical Chemistry
Figure 2: Selected calculated lowest energy structures for [Cu(CO2)n]- (n = 1 – 4). The capital letter denotes the core structure and the number/letter combination in parentheses indicates the number of solvent molecules and their conformations. Bare core structures are shown on the top row while solvation conformers are shown in the descending columns. Relative energies are given in meV and are with respect to the lowest energy isomer for a given cluster size. 339x190mm (150 x 150 DPI)
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 3: Comparison of unique formate solvation isomers and the insertion structure to the experimental spectrum of [Cu(CO2)2]-. 83x84mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 26 of 30
Page 27 of 30
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
The Journal of Physical Chemistry
Figure 4: A comparison of calculated structures for [Cu(CO2)3]- to the experimental spectrum for [Cu(CO2)3]-. Relative energies are given for each calculated trace (see also Figure 2 for structural labels). 91x102mm (300 x 300 DPI)
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Figure 5: Comparison of predicted spectra for calculated structures of [Cu(CO2)6]- (lower traces) to the experimental spectrum for [Cu(CO2)6]- (top trace). See also Figure 2 for structural labels. For structure D, we compare symmetric and asymmetric solvation. For structure F, we compare presence and absence of a solvent molecule at the CO2 moiety (see text). 107x140mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 28 of 30
Page 29 of 30
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
The Journal of Physical Chemistry
Figure 6: Spectral shifts of the experimental and simulated features for [Cu(CO2)n]- (n = 3 – 6). The top trace show the spectrum for n = 6 for comparison. The lines are meant to guide the eye, the letters indicate the assignments of the core isomers as shown in Figure 2. 147x121mm (300 x 300 DPI)
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
TOC Figure 338x190mm (96 x 96 DPI)
ACS Paragon Plus Environment
Page 30 of 30
