Controlling Photoisomerization Reactivity Through Single Functional

Subscriber access provided by STEPHEN F AUSTIN STATE UNIV

Communication

Controlling Photoisomerization Reactivity Through Single Functional Group Substitutions in Ruthenium Phosphine Sul-foxide Complexes Gilbert K. Kosgei, Douglas Breen, Robert W. Lamb, Maksim Y. Livshits, Laura A. Crandall, Christopher J. Ziegler, Charles Edwin Webster, and Jeffrey J. Rack J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b05957 • Publication Date (Web): 26 Jul 2018 Downloaded from http://pubs.acs.org on July 26, 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 5 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

Journal of the American Chemical Society

Controlling Photoisomerization Reactivity Through Single Functional Group Substitutions in Ruthenium Phosphine Sulfoxide Complexes Gilbert K. Kosgei,a Douglas Breen,a Robert W. Lamb,b Maksim Y. Livshits,a Laura A. Crandall,c Christopher J. Ziegler,c* Charles Edwin Webster,b* and Jeffrey J. Racka* a. Department of Chemistry and Chemical Biology, 1 University of New Mexico, Albuquerque, NM 87131 b. Department of Chemistry, Box 9573, Mississippi State University, Mississippi State, MS 39762-9573 c. Department of Chemistry, The University of Akron, Akron, OH 44325

Supporting Information Placeholder ABSTRACT: We report the crystallography, emission

spectra, femtosecond pump-probe spectroscopy, and DFT computations for a series of ruthenium complexes that comprise a new class of chelating triphenylphosphine based ligands with an appended sulfoxide moiety. These ligands differ only in the presence of the para-substitutent (e.g., H, OCH3, CF3). The results show a dramatic range in photoisomerization reactivity that is ascribed to differences in the electron density of the phosphine ligand donated to the ruthenium and the nature of the excited state.

ruthenium and osmium polypyridine based complexes whose photochromic action is based on an ES sulfoxide isomerization reaction.1-4 In this case, the GS isomer is S-bonded and the metastable isomer is O-bonded. This reaction can be efficient, with quantum yields nearing unity, and rapid, with electronic absorption changes occurring on the picosecond timescale. We report here a ligand design that permits simple group changes on the periphery of the molecule that leads to remarkable control of the ES isomerization reaction. In the series of three complexes reported here, we have found that the quantum yield of isomerization (ΦS O) may vary from 0.8 to 0.0. Phosphine ligands are well known to modulate the electron density of the metal through a combination of σand π-bonding. Indeed, depending upon the R groups in PR3 ligands, the ligand may be classified as a σ-only donor (e.g., PMe3) or as a strong π-acid or acceptor (e.g., PPh3 or PF3). While commonly employed in organometallic chemistry, this ligand class is not widely utilized in the design of photoactive chromophores.5-17 We questioned whether a family of chelating phosphine sulfoxide ligands would impart tunable properties in the photochromic behavior of these compounds. The salient connection is that different groups on the phosphine ligand alter the energy of the Ligand Field (LF) or MetalCentered (MC) states relative to the charge transfer (CT) states during ES evolution, thus modifying reactivity. →

An important goal in photoscience is the ability to introduce single atom or simple group changes within a molecule in order to alter the fate of an excited state (ES) reaction. In principle, such advances lead to improvements in photocatalysis schemes, molecular photovoltaic devices, and artificial photosynthetic materials. In practice, most such chemical synthetic changes offer only modest or subtle modification of ES processes, because the relaxation processes do not involve vibrations or other atomic motions responsive to these particular synthetic alterations. Moreover, for transition metal complexes, the situation is further confounded by the relatively large number of available electronic states (relative to organic systems), which can further mitigate a substituent effect. Photochromic compounds feature large changes in both electronic and molecular structure following visible light exposure. Thus, these compounds exhibit dramatic changes in the electronic absorption spectra upon irradiation of the ground state (GS) isomer to form a new metastable structure on the GS potential energy surface. We have created and studied a family of photochromic

Scheme 1. Molecular structures determined from the X-ray analysis of the S-bonded isomers of RuL1OH (left), RuL2OOCH3 (center), and RuL3OCF3 (right). Hydrogen atoms are omitted for clarity. ACS Paragon Plus Environment

Journal of the American Chemical Society

typically an ES absorption (reduced bpy, π→π* ligand centered transition) near 380 nm, its detection is obscured by the coincidence of the MLCT bleach. Evidence of this ES absorption can be seen at λ < 350 nm. -1

Wavenumber (cm ) 30000

ε (L mol-1cm-1)

Shown in Scheme 1 are the complexes described in this report. Each molecular structure (determined by single crystal X-ray diffraction) features the identical [Ru(bpy)2]2+ core and a different chelating phosphine sulfoxide ligand that differs only in the substituent in the 4-position on the terminal phenyl rings of the triphenylphosphine moiety. For RuL1OH, this substituent is H, for RuL2OOCH3 this substituent is OCH3, and for RuL3OCF3 this substituent is CF3. This simple substituent modification results in rather modest GS structural differences, but generates rather dramatic and remarkable distinctions in ES reactivity. The electronic absorption spectra of these three complexes are quite similar revealing absorption maxima of 351, 353, and 348 nm, respectively for S[(bpy)2Ru(L1OH)]2+, S-[(bpy)2Ru(L2OOCH3)]2+, and S[(bpy)2Ru(L3OCF3)]2+, respectively (Figure 1A). These data demonstrate the effect of the structurally distant substituent group (H vs. OCH3 vs CF3) on the Ru dπ to bpy π* charge transfer (CT) transition, which is confirmed by TD-DFT (vide infra) computations. Irradiation of RuL3OCF3 at 355 nm results in striking changes in the electronic absorption spectrum (Figure 1B), with the emergence of a new absorption maximum at 422 nm, and isosbestic points at 332 and 376 nm. Based on literature precedence, we ascribe the newly formed spectrum to the O-bonded isomer of the ruthenium phosphine sulfoxide complex. For example, Wolf has reported the absorption maximum of [(bpy)2Ru(PO)]2+ is 412 nm, where PO is 2-diphenylphosphino-(anisole). This complex features a P and O donor on [Ru(bpy)2]2+.18-19 The bold red trace displayed in Figure 1B represents pure O[(bpy)2Ru(L3OCF3)]2+ formed from irradiation (ΦS→O = 0.8±0.2) at 355 nm or 405 nm. Excitation of RuL1OH at either 355 or 405 nm produces a similar change in the absorption spectrum (ΦS→O = 0.2±0.1). In extraordinary contrast, irradiation of RuL2OOCH3 at any wavelength that overlaps with the GS absorption spectrum does not result in any spectral changes, thus providing no indication of isomerization. These observations are supported by computational results and confirmed by time resolved spectroscopy (vide infra). It is remarkable that such a small structural replacement of either H or CF3 with OCH3 alters the photochemical reactivity so dramatically. Ultrafast visible pump probe spectroscopy was employed to monitor the spectral changes in these complexes. For [(bpy)2Ru(L2OOCH3)]2+, the early time spectra are reminiscent of those obtained for many ruthenium polypyridine complexes (Figure 1D). That is, visible light excitation produces an MLCT with expected features that relax to re-form the GS with a lifetime of 790 ± 60 ps. The ES absorption in the red portion of the spectrum is attributed primarily to unreduced bpy → RuIII LMCT transitions, whereas the GS bleach is observed in the blue portion of the spectrum. While there is

27500

25000

22500

20000

17500

A

4000 0

B

4000 0 0.004

ΔA (OD)

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 5

0.000

C

-0.004

D

0.004 0.000

350

400

450

500

550

600

Wavelength (nm)

Figure 1. (A) Absorption spectra for S-[(bpy)2Ru(L1OH)]2+ (black), S-[(bpy)2Ru(L2OOCH3)]2+ (red), and S[(bpy)2Ru(L3OCF3)]2+ (blue) (B) Spectra obtained from bulk photolysis (λexc 355 nm) of [(bpy)2Ru(L3OCF3)]2+ in dichloroethane solution. Spectra obtained at 80 second intervals. Black dashed trace is difference spectrum (O isomer – S isomer) extracted from bulk photolysis data. (C) Time resolved spectra of [(bpy)2Ru(L3OCF3)]2+ obtained at different pump-probe delays of 2.0 ps (red), 20.1 ps (orange), 202 ps (olive green), 2000 ps (green), and 4990 ps (violet). The new absorption maximum centered at 422 is evidence of a photoproduct, consistent with the bulk photolysis data. (D) Time resolved spectra of [(bpy)2Ru(L2OOCH3)]2+ obtained at different pump-probe delays of 0.41 ps (red), 2.51 ps (orange), 199 ps (green), 1000 ps (blue), and 4660 ps (violet). These traces provide no evidence for the formation of a photoproduct (isomerization), consistent with the bulk photolysis data.

Representative pump probe data for RuL3OCF3 document the absorption changes observed during bulk photolysis (Figure 1C). The early time spectral features are emblematic of an MLCT state, as described above. In contrast to RuL2OOCH3, the spectra indicate the formation of a GS photoproduct (isomerization), by the emergence of a new absorption maximum at 422 nm, coincident with loss of absorption in the red portion of the spectrum. The loss of absorption at long wavelengths indicates a transition from the 3MLCT surface to the GS surface. The absorption maximum observed in the bulk photolysis data is nicely reproduced in the pump-probe data, demonstrating formation of the same complex. This is readily observed through visual comparison of the transient spectrum obtained at long pump-probe delays (~5000 ps, vilet trace Figure 1C) with the difference in GS spectra (O-bonded – S-bonded, black dashed trace, Figure 1B) obtained from bulk photolysis. Global fitting analysis and single wavelength kinetics reveal a

ACS Paragon Plus Environment

Page 3 of 5 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

Journal of the American Chemical Society

time constant for formation of the O-bonded isomer of 630±77 ps. Multiplying by ΦS→O (0.80) yields a time constant for isomerization (τS→O) of 790 ps. The absorption changes in RuL1OH and RuL3OCF3 associated with the picosecond isomerization are completely reversible. Reversion rate constants of (300 s)–1 and (770 s)–1 are found for RuL1OH and RuL3OCF3, respectively (Figure S7). Future studies will reveal the bonding details that are responsible for this unusual reactivity. We performed DFT and TDDFT computations to determine if there were differences in electronic structure that might explain the observed differences in photochemistry for this family of complexes. Shown in Figure 2 are the NTOs (natural transition orbitals; see the SI for full computational details) of the hole (left) and particle (electron, right) that describe the CT excited states for RuL2OOCH3 (top) and RuL3OCF3 (bottom). While these images show similar bipyridine contributions to the particle orbital, the Ru and sulfoxide contribution to the hole orbital for RuL3OCF3 is considerable in comparison to the hole orbital for RuL2OOCH3, which exhibits a large contribution from the methoxyphenyl moiety of the phosphine ligand.

Figure 2. Natural Transition Orbitals (NTOs) for the hole (left) and particle (right) for S-[(bpy)2Ru(L2OOCH3)]2+ (Top) and S-[(bpy)2Ru(L3OCF3)]2+ (Bottom). The NTOs for S-[(bpy)2Ru(L1OH)]2+ are qualitatively similar to S[(bpy)2Ru(L3OCF3)]2+.

The relative energies from triplet single-point computations on time-dependent DFT singlet optimized geometries (3TD/BS2//1TD/BS1) demonstrate that the O-bound isomer in the ES (relative to S-bound in the ES) is disfavored for RuL2OOCH3 (ΔGrxn = 2.2 kcal mol–1), competitive for RuL1OH (ΔGrxn = 0.2 kcal mol–1), and favored for RuL3OCF3 (ΔGrxn = –3.5 kcal mol–1). This trend corroborates the experimental observations of isomerization with RuL1OH and RuL3OCF3 and no isomerization with RuL2OOCH3. Furthermore, these data show that the isomerization behavior for this class of complexes can be predictably controlled based on the logic of electron density within the aromatic system of the ligands and, by extension, the degree of electron donation to the metal center. Thus, the computational results provide two insights into the change in reactivity between RuL3O and

RuL2O. First, the NTOs identify differences in Ru and sulfoxide character in the hole orbital; second, while calculations do not provide information on the isomerization pathway, at least they indicate that the O-bonded RuL2O state is thermodynamically uphill. Structural data obtained from single crystal X-ray diffractometry reveal only subtle differences between the three complexes. The Ru–S bond distances vary only over a small range: 2.2185(6) Å in RuL1OH, 2.218(1) Å in RuL2OOCH3, and 2.233(2) and 2.235(2) Å in RuL3OCF3 (there are two unique molecules in this unit cell). The S–O bond distances display a smaller variation: 1.479(9) Å in RuL1OH, 1.475(4) Å in RuL2OOCH3, and 1.479(6) and 1.463(5) Å in RuL3OCF3. Interestingly, the Ru–S and S–O bond distances are equivalent in RuL1OH and RuL2OOCH3, two complexes that display dramatically different photochemical reactivity. The metrical parameters do not reveal any significant differences in the Ru–P distances or in any pertinent angles, thus suggesting that any differences in the photophysical behavior or photochemical reactivity in these complexes is due to an ES phenomenon and not to GS geometric differences. Emission spectra provide compelling evidence for isomerization at 77K. Shown in Figure 3 is the 77 K steady state emission spectrum for S-RuL3OCF3 in a 4:1 mixture of tetrahydrofuran and propylene carbonate. The emission spectrum reveals two maxima at 525 and 650 nm (340 nm excitation). The lifetimes at these two wavelengths are 33 and 50 µs, respectively indicative of a CT ES. Indeed, these lifetimes are much longer than those observed in [Ru(bpy)3]2+ (τ = ~5 µs at 77K; we obtain 7.7 µs on our instrument).20 The two different lifetimes indicate two non-interacting emissive populations. Excitation scans at 525 nm and 650 nm reveal peaks at 340 nm and 440 nm, respectively corresponding to the absorption maxima of S- and O-RuL3OCF3 isomers. The 650 and 525 nm emission bands are assigned to the O- and S-bonded isomers, respectively. We note that the differences in absorption (0.62 V) and emission maxima (0.57 V) are in accord with the difference in reduction potentials (0.66 V) for the S- and O-bonded isomers. While we cannot be certain that there is no Obonded isomer in the original solution, we interpret these results as formation of an O-bonded emissive CT state at 77 K, produced from excitation of an S-bonded GS, which relaxes to both S- and O-bonded GS isomers. The computational data are consistent with these observations. There are weak transitions in these regions for both isomers.

ACS Paragon Plus Environment

Journal of the American Chemical Society Funding Sources

Wavenumbers (cm-1) 24000

21000

18000

15000

12000

1.0

JJR acknowledges the National Science Foundation (Grant CHE 1602240) and the University of New Mexico for financial support. This work was supported by the Mississippi State University Office of Research and Economic Development and the National Science Foundation (OIA1539035). Computations were performed at the Mississippi State University High Performance Computing Collaboratory (HPC2) and the Mississippi Center for Supercomputing Research.

0.6

0.4

0.2

0.0 400

The authors declare no competing financial interest.

ACKNOWLEDGMENT

0.8

Intensity (Normalized)

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

450

500

550

600

650

700

750

800

850

Wavelength (nm)

Figure 3. Corrected steady state emission spectrum of [(bpy)2Ru(L3OCF3)]2+ at 77 K. The 77 K emission data suggest that isomerization occurs adiabatically along the CT surface, whereas the room temperature pump-probe data do not provide evidence for the formation of an O-bonded excited state. We propose that isomerization occurs through nonadiabatic coupling between adiabatic surfaces, which also serve as the basis of the above calculations. Taken together, the temperature dependent data hint at an intricate balance of couplings between surfaces, both excited state and ground state, as well as S-bonded and Obonded. In summary, these results indicate that isomerization occurs efficiently in certain of these ruthenium phosphine sulfoxide complexes, and that the phosphine moiety provides remarkable control of the photoisomerization. Moreover, we report the first data that shows evidence of isomerization at 77 K from an unequivocal CT state. Future studies will focus on understanding the unusual reactivity exhibited by these complexes. ASSOCIATED CONTENT Supporting Information. Detailed synthetic and experimental procedures, and computational details for [(bpy)2Ru(L1OH)]2+, [(bpy)2Ru(L2OOCH3)]2+, and 2+ [(bpy)2Ru(L3OCF3)] . CCDC 1843478-1843480 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures/. The Supporting Information is available free of charge on the ACS Publications website.

AUTHOR INFORMATION Corresponding Author * J.J.R. E-mail: [email protected] * C.E.W. E-mail: [email protected] * C.J.Z. E-mail: [email protected]

REFERENCES 1. Rack, J. J., Coordination Chemistry Reviews 2009, 253 (12), 78-85. 2. King, A. W.; Wang, L.; Rack, J. J., Accounts Chem. Res. 2015, 48 (4), 1115-1122. 3. Garg, K.; King, A. W.; Rack, J. J., J. Am. Chem. Soc. 2014, 136 (5), 1856-1863. 4. Garg, K.; Engle, J. T.; Ziegler, C. J.; Rack, J. J., Chem.Eur. J. 2013, 19 (35), 11686-11695. 5. Angell, S. E.; Rogers, C. W.; Zhang, Y.; Wolf, M. O.; Jones, W. E., Coordination Chemistry Reviews 2006, 250 (13-14), 1829-1841. 6. Dixon, I. M.; Lebon, E.; Sutra, P.; Igau, A., Chem. Soc. Rev. 2009, 38 (6), 1621-1634. 7. Gutmann, R.; Czermak, G.; Dumfort, A.; van der Veer, W. E.; Hong, B.; Kopacka, H.; Ongania, K. H.; Bechtold, T.; Bruggeller, P., Inorg. Chem. Commun. 2005, 8 (3), 319-322. 8. Lebon, E.; Bastin, S.; Sutra, P.; Vendier, L.; Piau, R. E.; Dixon, I. M.; Boggio-Pasqua, M.; Alary, F.; Heully, J. L.; Igau, A.; Juris, A., Chem. Commun. 2012, 48 (5), 741-743. 9. Lebon, E.; Sylvain, R.; Piau, R. E.; Lanthony, C.; Pilme, J.; Sutra, P.; Boggio-Pasqua, M.; Heully, J. L.; Alary, F.; Juris, A.; Igau, A., Inorganic Chemistry 2014, 53 (4), 1946-1948. 10. Litke, S. V.; Ershov, A. Y.; Meyer, T. J., J. Phys. Chem. A 2011, 115 (50), 14235-14242. 11. Rogers, C. W.; Patrick, B. O.; Rettig, S. J.; Wolf, M. O., J. Chem. Soc.-Dalton Trans. 2001, (8), 1278-1283. 12. Sutra, P.; Igau, A., Coordination Chemistry Reviews 2016, 308, 97-116. 13. Dixon, I. M.; Lebon, E.; Loustau, G.; Sutra, P.; Vendier, L.; Igau, A.; Juris, A., Dalton Trans. 2008, (41), 5627-5635. 14. Suzuki, T.; Kuchiyama, T.; Kishi, S.; Kaizaki, S.; Takagi, H. D.; Kato, M., Inorganic Chemistry 2003, 42 (3), 785-795. 15. Carlson, B.; Phelan, G. D.; Kaminsky, W.; Dalton, L.; Jiang, X. Z.; Liu, S.; Jen, A. K. Y., J. Am. Chem. Soc. 2002, 124 (47), 14162-14172. 16. Klassen, D. M.; DelPup, R. V., Inorganic Chemistry 2002, 41 (12), 3155-3160. 17. La Pensee, A. A.; Bickley, J.; Higgins, S. J.; Marcaccio, M.; Paolucci, F.; Roffia, S.; Charnock, J. M., J. Chem. Soc.-Dalton Trans. 2002, (22), 4095-4104. 18. Angell, S. E.; Zhang, Y.; Rogers, C. W.; Wolf, M. O.; Jones, W. E., Inorganic Chemistry 2005, 44 (21), 7377-7384. 19. Rogers, C. W.; Wolf, M. O., Chem. Commun. 1999, (22), 2297-2298. 20. Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; Vonzelewsky, A., Coordination Chemistry Reviews 1988, 84, 85-277.

ACS Paragon Plus Environment

Page 4 of 5

Page 5 of 5 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

Journal of the American Chemical Society

O (bpy)2(R3P)RuII S Me R

Φ

(S → O)

= 0.8

hν

Φ

(S → O)

(bpy)2(R3P)RuII

= 0.2

Φ

ACS Paragon Plus Environment

(S → O)

O S Me R

= 0.0