Primed for Efficient Motion: Ultrafast Excited State Dynamics and

16 hours ago - All isomers of a four stage rotary molecular motor, dimethyl-tetrahydro-bi(cyclopenta[α]napthal-enylidene), are studied with ultrafast...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/JPCA

Cite This: J. Phys. Chem. A XXXX, XXX, XXX−XXX

Primed for Efficient Motion: Ultrafast Excited State Dynamics and Optical Manipulation of a Four Stage Rotary Molecular Motor Theodore E. Wiley,† Arkaprabha Konar,‡ Nicholas A. Miller,† Kenneth G. Spears,† and Roseanne J. Sension*,†,‡ †

Department of Chemistry, University of Michigan, 930 North University Avenue, Ann Arbor, Michigan 48109-1055, United States Department of Physics, University of Michigan, 450 Church Street, Ann Arbor, Michigan 48109-1040, United States

‡

J. Phys. Chem. A Downloaded from pubs.acs.org by UNIV OF SOUTH DAKOTA on 09/19/18. For personal use only.

S Supporting Information *

ABSTRACT: All isomers of a four stage rotary molecular motor, dimethyltetrahydro-bi(cyclopenta[α]napthal-enylidene), are studied with ultrafast transient absorption spectroscopy. Single and two pulse excitations (pump and delayed repump with a different wavelength) are used to optically probe the excited state dynamics. These measurements demonstrate that this motor is not only designed for unidirectional isomerization, but is also “primed” for efficient rotary motion. The yield for photoisomerization from the stable P-cis isomer to the metastable M-trans isomer is 85% ± 10%, while the yield for the undesired back reaction is ca. 0.08 (+0.02, −0.05). The yield for photoisomerization from stable P-trans to the metastable M-cis isomer is ca. 85% ± 3% and the yield for the back reaction is 15% ± 3%. Excitation of Ptrans in the lowest singlet state results in formation of a dark state on a 3.6 ps time scale and formation of the M-cis isomer on a ca. 12 ps time scale. Excitation of P-cis in the lowest singlet state results in formation of a dark state on ca. 13 ps time scale and formation of the M-trans isomer on a 71 ps time scale. Excitation of either isomer at 269 nm, higher in the excited state manifold, accesses additional excited state pathways, but does not change the ultimate product formation. This result suggests that pulse sequences accessing higher excited states may provide a tool to manipulate the molecular motor. Pulse sequences using a 269 nm pump pulse and a 404 nm repump pulse are able to increase the yield of the P-cis to M-trans reaction but only decrease the yield of the P-trans to M-cis reaction. These pulse sequences are unable to access reaction pathways that bypass the helix inversion step, although other wavelengths and time delays might yet provide optical control of the entire reaction cycle. We propose intermediates and candidate conical intersections between all four isomers.

■

INTRODUCTION Photochemistry provides a potent means for harnessing light energy to manipulate molecular systems through the movement of charge, a change in molecular shape, or the cleavage of a bond. Photochemistry produces action. The potential for optically powered molecular devices covers a wide range of applications.1 However, the practical development of such molecular devices faces a range of challenges. It is not enough that the desired transformation occurs; we require it to occur efficiently and in a controlled fashion. Appropriate compounds must be designed and synthesized. While some of the design requirements can be inferred from the desired action and tested by trial and error, others require understanding the electronic structure and photochemical pathways of the chromophores. A significant effort over the past decade has focused on the design of optically powered molecular motors capable of unidirectional rotary motion.1−7 Prototypical first and second generation molecular motors, classified as stilbene or fluorene motors, are illustrated in Figure 1.6,8 Ultrafast spectroscopies provide a powerful tool to study the fundamental chemistry and physics of the photoisomerization reactions of rotary molecular motors, and have been applied to © XXXX American Chemical Society

Figure 1. Example of a first and second generation rotary molecular motor. The first generation motor pictured is the one studied here.

a few different systems.5,9−12 Meech and co-workers performed fluorescence upconversion measurements on the asymmetric second generation molecular motor in Figure 1, demonstrating fast nonexponential motion out of the Franck−Condon region.12 Subsequent fluorescence and transient absorption measurements of a series of related second generation motors demonstrated similar behavior.5 In this second study, the quantum yield for the light-driven isomerization step was Received: July 6, 2018 Revised: September 4, 2018

A

DOI: 10.1021/acs.jpca.8b06472 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A determined to range from 0.048 to 0.2. The quantum yield for the reverse isomerization reaction was found to be significantly higher than for the forward reaction limiting the efficiency of the motors, especially under broadband excitation and high fluence conditions. In a more recent study, femtosecond stimulated Raman spectroscopy was used to study the structural evolution on the excited state and characterize the modifications of bond order in the central bridge in the excited electronic states.13 Theoretical simulations have been used to further elucidate the photoisomerization reactions of these and other molecular motors.8,14−22 Filatov and co-workers have focused their theoretical efforts on the design of molecular motors optimized for efficiency in the photochemical steps.18,19 Although chemical modification has proven to be a powerful method to optimize the function of molecular motors, optical control with pulse shaping or pulse sequences could also be used to optimize the function of these motors. The modest quantum yields for forward isomerization and the higher quantum yields for back reaction in second generation motors suggest that significant improvement in the light-driven isomerization step might be possible. In addition, optical pulse sequences could make it possible to bypass the thermal helix inversion step allowing optical control of the entire cycle. Potential sequences could use higher excited electronic states to explore distant regions of coordinate space (e.g., longer carbon−carbon bonds along the rotation axis) resulting in direct formation of the stable product or could place vibrational energy into relevant modes of the product resulting in prompt ground state helix inversion from the metastable product to the stable product. In the work reported here broad-band UV−vis pump−probe transient absorption and three pulse pump-repump-probe measurements are used to elucidate and manipulate the excited state dynamics of a symmetric first generation molecular motor, dimethyl-tetrahydro-bi(cyclopenta[α]napthal-enylidene).23 This compound was chosen for study because it is readily synthesized with a photocycle passing through four spectroscopically distinguishable conformations, two stable and two metastable, hereafter referred to as P-cis, P-trans, M-cis, and M-trans, respectively (see Figure 2). Our results demonstrate that this rotary molecular motor is tuned for efficient unidirectional isomerization. In contrast to the second generation motors studied by Meech and coworkers,5,9,12,13 the quantum yields for forward photoisomerization are high, ca. 0.85, while the quantum yields for the reverse reactions are low. The reaction pathway depends on excitation wavelength, with approximately equal branching between internal conversion to the S1 excited state and rapid population of a dark state following excitation at 269 nm. This result suggests that pulse sequences may provide a tool to manipulate the molecular motor. In a first attempt to manipulate the reaction pathway, we find that pulse sequences using 269 nm initial excitation followed by a delayed 404 nm pulse are capable of increasing the yield of the P-cis → M-trans reaction slightly, but only decrease the yield of the P-trans → M-cis reaction. When the second pulse is at 808 nm the isomerization yield is reduced for both P-cis and P-trans. There is no evidence in these initial measurements for reaction pathways bypassing the helix inversion step.

Figure 2. UV−vis absorption (solid lines) and fluorescence (dashed lines) spectra for the stable P-cis and P-trans isomers. The fluorescence spectra are scaled to match the absorption intensities. The excitation wavelengths used in transient absorption measurements are also indicated on the plot. The shoulder of the P-trans spectrum between 390 and 430 nm is assigned to a small population of M-trans as discussed in the text.

■

METHODS UV−vis transient absorption measurements were performed using three different tunable Ti:sapphire based laser systems. In each system ca. 808 nm femtosecond pulses are amplified in a kHz multipass or regenerative amplifier to produce ca. 1 mJ, 150 ps is opposite in sign to the residual bleach/product absorption observed following excitation of Ptrans, supporting assignment of this feature to photoisomerization in the reverse direction (M-cis→P-trans). Summary of Photocycle. Excitation into the lowest allowed electronic absorption band of P-cis populates a series of excited states prior to formation of the metastable M-trans product. The initial optically bright state (A) populated after excitation of P-cis rapidly evolves to a second state or conformation (B), also optically bright, before undergoing conversion to an optically dark state (C) on a ca. 13 ps time scale. Conformational relaxation on the excited state surface, reducing the stimulated emission signal (Figure 5), is similar to the biphasic decay of the bright state observed for a second generation motor.12 The optically dark state C undergoes internal conversion to the ground state on a ca. 70 ps time scale while forming the M-trans product in high yield. M-trans shows no further changes in these measurements, consistent with a measured half-life of ∼18 s at room temperature.23 Excitation of P-trans in the lowest absorption band (368 nm, 352 nm) populates an excited state (A) that relaxes on a 500 fs time scale to an optically bright state or conformation (B). This state converts to a dark state (C) on a 4 ps time scale. The dark state decays to form M-cis on a 13 ps time scale. No further changes to the M-cis product are observed, consistent with the ∼74 min half-life of M-cis at room temperature.23 Excitation of P-cis and P-trans higher in the electronic state manifold at 269 nm introduces additional relaxation pathways. However, the ultimate yield of the M-trans or M-cis photoproduct is similar in magnitude.

Figure 10. Double difference spectrum with 269 nm excitation of Pcis followed by a 404 nm excitation pulse at 300 ps. The color scheme is the same as that used in the upper panel of Figure 12 for excitation of the trans isomers at 404 nm. The long-lived photoproduct formed following excitation of P-cis and the red-edge shoulder in the P-trans spectrum clearly report on the same species.

for a difference of differences (eq 1). The pump-repump-probe data were globally fit with two exponential decay components. The lifetimes obtained from the fit were τ1 ≈ 2.4 ± 0.2 ps and τ2 = 40 ± 5 ps again in agreement with the results obtained following direct excitation of M-trans at 404 nm (Table 1). The population of the M-trans isomer at room temperature may be estimated from the extinction coefficients for the Ptrans and M-trans isomers plotted in Figure 3. The relative magnitude of the red shoulder and the peak absorption provides an estimate of ∼4.5% M-trans. The relative populations may also be estimated from DFT calculations of the energies of these two isomers. Calculations were performed in Spartan 10 (B3LYP functional, 6-31G(D) basis set, PCM solvation). P-cis is the most stable isomer with P-trans calculated to be 11.8 kJ/mol higher. The energy difference between P-trans and M-trans is 8.2 kJ, consistent with relative populations of 3.6% M-trans, 94.4% P-trans at room G

DOI: 10.1021/acs.jpca.8b06472 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A Determination of Quantum Yields. The low-temperature steady state photolysis measurements can be used along with the transient absorption data to place constraints on the quantum yields (ϕ) for the forward and backward reactions in the photocycle. First, the bleach recovery following excitation of the M-trans or M-cis isomers establishes an upper limit for the quantum yields of the M-trans→P-cis and M-cis→P-trans reactions. The bleach recovery following excitation of M-cis (Figure 11) establishes an upper limit of ca. ϕM‑cis→P‑trans ≤ 0.2. The bleach recovery following excitation of M-trans (Figure 9, S16) establishes an upper limit of ϕM‑trans→P‑cis ≤ 0.1. The regions eliminated by the bleach recovery are indicated by horizontal shaded regions in Figures S18 and S20. Second, a lower limit for the P-trans→M-cis and P-cis→Mtrans reactions can be established by comparing the initial bleach at early times with the magnitude of the long-lived product signal. This analysis requires accurate relative extinction coefficients for the P-trans and M-cis or P-cis and M-trans isomers. In principle these can be obtained from the steady state photolysis measurements if the composition of the photostationary state is known. A rigorous lower limit is obtained under the assumption that the photostationary state is entirely product and using these spectra as M-trans and Mcis. The lower limit for P-trans→M-cis is ϕP‑trans→M‑cis ≥ 0.66. The lower limit for P-cis→M-trans is ϕP‑cis→M‑trans ≥ 0.75. The regions eliminated by the product formation are indicated by vertical shaded regions in Figures S18 and S20. Third, these limits can be refined by considering the composition of the photostationary state. The vibronic structure remaining in the P-trans/M-cis product spectrum (Figure 3, left) indicates the presence of P-trans in the photostationary state. Subtraction of the P-trans contribution to eliminate this structure establishes an upper limit of 80:20 for the ratio of M-cis:P-trans in the photostationary state (Figure S17). The requirement that the absorption spectrum of M-cis is everywhere positive establishes a lower limit of 65:35 for this ratio. Using the M-cis spectrum obtained assuming a ratio of 80:20 in the photostationary state, the lower limit for P-trans→M-cis is ϕP‑trans→M‑cis ≥ 0.8. The region eliminated by this limit is indicated by the vertical shaded region labeled “probably eliminated” in Figure S18. The photostationary state for P-cis⥂M-trans shows no clear evidence of P-cis and the lower limit remains at ϕP‑cis→M‑trans ≥ 0.75. Fourth, the composition of the photostationary state and the overlap of the xenon lamp with the reactant and product absorption spectra (see Figures S17 and S19) can be used to further constrain the quantum yields for the forward and backward reactions. The P-trans ⥂ M-cis quantum yields are related by the following equation: σP‐trans ϕP‐trans → M‐cis M‐cis PSS = σM‐cis ϕM‐cis → P‐trans P‐trans PSS

suggests that this is the same species and the same conical intersection is accessed in both reactions. Thus, the quantum yields for the forward and backward reactions should sum to 1 (light blue line in Figure S18). This sets a limit for the quantum yield of P-trans→M-cis of 0.82 to 0.88 with M-cis→ P-trans then 0.12 to 0.18. Using the Strickler−Berg formula,46 the radiative lifetime of the optically allowed S1 excited state of P-cis is estimated at 9 ns and, given the excited state lifetime of up to 16 ps, the corresponding fluorescence quantum yield of the bright state is ≤0.002. Thus, radiative decay can be neglected in this analysis. Analysis of the P-cis ⥂ M-trans photoisomerization is complicated by the fact that direct excitation of M-trans accesses a different dark state minimum (40 ps lifetime) than accessed following P-cis excitation (71 ps lifetime). Thus, the conical intersection for internal conversion to the ground state may also be different and the quantum yields for the forward and backward reactions need not sum to 1. Analysis of the photostationary state suggests that ϕP‑cis→M‑trans = 0.85 ± 0.1 and ϕM‑trans→P‑cis = 0.08 (+0.02, −0.05) (Figure S20). Using the Strickler−Berg formula the radiative lifetime is estimated at 5 ns and the corresponding fluorescence quantum yield of the bright state is ≤0.001. Again, radiative decay can be neglected in the analysis. The quantum yield results are summarized in Figure 12 along with those for the P-trans↔M-cis reactions.

Figure 12. Quantum yields for the forward and backward light-driven reactions of the molecular motor and the room temperature half-life for the dark reactions.23

Filatov and co-workers have used theoretical simulations to explore the factors that influence photoisomerization yields in rotary molecular motors.18,19 They find that introduction of a nitrogen into one of the rings helps to focus the isomerization on axial rotation and increase the quantum yield. Both fast isomerization and relatively high quantum yields (ca. 0.4 to 0.6) are predicted for these systems. The isomerization through an S1/S0 conical intersection is fast (∼100 to 400 fs), and the forward (P-trans→M-cis and P-cis→M-trans) and backward (M-cis→P-trans and M-trans→P-cis) photoreactions are predicted to have comparable yields. The work presented here suggests that alternative approaches may also be fruitful. Access to the conical intersection on an ultrafast time scale is of less importance than a conical intersection primed for efficient unidirectional motion. Theoretical efforts to explore the factors leading to the quantum yields reported here should help in the design of efficient and fast optically powered rotary molecular motors.

(3)

where σP‑trans represents the overlap of the lamp spectrum with P-trans, σM‑cis represents the overlap of the lamp spectrum with M-cis and the right-hand side is the ratio of M-cis to P-trans in the photostationary state. The lines calculated for different PSS ratios are plotted in Figure S18. The similarity in the time constant (ca. 13 ps) and excited state spectrum (see Figures 8 and 11) assigned to the decay of the intermediate dark state C for the forward and backward photoisomerization reactions H

DOI: 10.1021/acs.jpca.8b06472 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A Summary of Intermediates and Pathways. The transient absorption and steady state photolysis measurements reported above reveal two intermediate excited states for Ptrans and P-cis when excited into their lowest optically allowed electronic states. Excitation into the next strong absorption band at 269 nm introduces an additional reaction pathway with branching between direct population of a dark state on an ultrafast time scale and internal conversion to the lowest optically allowed state. The branching ratio in each case is between 60/40 and 40/60. Excitation of P-trans results in formation of a dark state on a 3.6 ps time scale and formation of the M-cis isomer on a ca. 13 ps time scale. The quantum yield for isomerization is high. Excitation of M-cis at 404 nm accesses the same dark state within a few picoseconds followed by internal conversion to the ground state. These results are summarized in the upper panel of Figure 13. States labeled A

invisible to our measurements. The present measurements are not sufficient to determine the quantum yields with sufficient accuracy to distinguish between these options. It should be noted that two distinct conical intersections are also proposed for the second generation molecular motor.50 These results are summarized in the lower panel of Figure 13. The wavelength dependent excited state dynamics uncovered in our transient absorption measurements suggests that pulse sequences may be effective in manipulating the excited state population and modifying the photoisomerization yields. Although the molecule investigated here is already optimized for unidirectional photoisomerization with high quantum yield, there remains some room for improvement. Using Pulse Sequences to Manipulate Product Formation. In addition to pump-repump measurements probing the photochemistry of the metastable M-trans and M-cis photoproducts, shorter time delays can be used to probe the effect of excess energy on the excited state dynamics and photoproduct formation. The repump pulse can be delayed or tuned to target different species during the evolution of the excited states of the molecule.47−49 Repump experiments with 404 and 800 nm were performed for heptane solutions at the time-delays indicated in Figure S21. The data sets are evaluated for pathways and species, including the possibility of the repump pulse driving intermediate states. The probe pulse in each measurement is a broadband continuum spanning the visible region of the spectrum. No evidence is found in these experiments for a change in the photoisomerization reaction. In particular, there is no evidence that excess energy provides a pathway to bypass formation of the metastable ground state conformation and eliminate the ratelimiting thermal helix inversion. Sequential two-pulse excitation of P-trans decreases the overall product yield. A large decrease (∼20%) with a 0.5 ps delay between the 269 and 404 nm pulses is attributed to stimulated emission dumping the excited state population before photoisomerization can occur. At longer time delays for the 404 nm pulse, the transfer of population from states C and D (see Figure 13a) to higher excited states decreases the ultimate yield of M-cis and increases internal conversion to repopulate the P-trans ground state. These results are summarized in Table 2 and Figure S22. Re-excitation of P-

Figure 13. Summary of the excited state progression and quantum yields for isomerization. The top panel (a) is for the P-trans ⥂ M-cis reaction and the lower panel (b) is for the P-cis ⥂ M-trans reaction. All numbers refer to data obtained in heptane solvent. States labeled A and B in both panels are optically coupled to the ground state, characterized by stimulated emission. States labeled C, C′, and D are dark states. Although two distinct conical intersections are suggested for C and D in the upper panel and C and C′ in the lower panel, these could be identical if a single state leading to the conical intersection is short-lived and thus invisible to our measurements.

Table 2. Change in Product Yield after Re-Excitation at the Given Wavelengths and Time Delays

and B are optically coupled to the ground state, characterized by stimulated emission. States labeled C and D are dark states. Although two distinct conical intersections are suggested in the Figure 13 for C and D, these could be identical if a single state leading from these states to the conical intersection has a lifetime ≪13 ps and is thus invisible to our measurements. Excitation of P-cis results in formation of a dark state, C, on ca. 13 ps time scale. Formation of the M-trans isomer follows on a 71 ps time scale. The state C′ populated following excitation of M-trans is shorter-lived (40 ps vs 71 ps) and spectroscopically distinct from the dark state C populated following excitation of P-cis. This suggests that two distinct conical intersections are accessed in these measurements and that the quantum yields for the forward and backward isomerization reactions need not sum to 1. However, the forward and backward isomerization reactions could pass through the same conical intersection if the dark state leading to the conical intersection has a lifetime ≪40 ps making it

P-cis Repump 404 nm 800 nm P-trans Repump 404 nm 800 nm

0.5 ps

−2% 0.5 ps −20% −5%

0.8 ps

10 ps

65 ps

0 ± 2%

6 ± 5%

8.5 ps

7 ± 3% −2% 15 ps

−8%

−4%

−4%

65 ps

trans using an 800 nm pulse 0.5 ps following excitation at 269 nm results depletes the photoproduct yield by ca. 5%, but this depletion is not accompanied by recovery of P-trans ground state (Figure S23), suggesting that the 800 nm pulse populates a higher excited state with additional reactive pathways. However, there is no evidence for formation of the P-cis indicating that the pulse sequence does not bypass the ground state helix inversion. I

DOI: 10.1021/acs.jpca.8b06472 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A Sequential two pulse excitation of P-cis with a 404 nm pulse delayed by 10 or 65 ps results in a modest increase in the formation of the M-trans photoproduct (Figure S24). At 10 ps the initial 269 nm excitation pulse has prepared an excited state population predominantly in the dark state C (Figure 13b). Re-excitation with a 404 nm pulse depopulates this state, producing a higher electronic state with internal conversion to the S1 state surface on a ca. 2 ps time scale followed by return to the ground state as P-cis or M-trans on a 34 (+15, −4) ps time scale (Figure S25). The lifetime of the excited state is similar to that observed following excitation of M-trans (Figure 13b) suggesting that the pulse sequence has populated the dark state C′. This may access a conical intersection with a higher quantum yield for formation of M-trans and account for the increased yield observed in the pump-repump-probe measurements (Figure S25). When the second pulse is at 800 nm, the data suggest a small decrease in photoproduct formation (Figure S26). It is not surprising that the simple pulse sequences used in the current study are unable to bypass the ground state helix inversion in these motors. The steric barrier to helix inversion in the first generation motors is large and the motions involved are complex. Other molecular structures with lower steric barriers could be easier to overcome with optical pulses. In addition, more sophisticated pulse sequences and pulse shapes may provide access to regions of coordinate space (e.g., longer bonds on the rotation axis) where the inversion can be enhanced directly in the photoisomerization process or by placing controlled vibrational excitation in the ground state metastable product.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 734-763-6074. ORCID

Arkaprabha Konar: 0000-0001-6546-4111 Roseanne J. Sension: 0000-0001-6758-0132 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Science Foundation NSF-CHE 1464584. Portions of this work were carried out in the Laboratory for Ultrafast Multidimensional Optical Spectroscopy (LUMOS), supported by NSF-CHE 1428479. The authors also thank Joseph Stackpoole and Joseph Meadows for their assistance in measurement of the low temperature photolysis of P-cis and P-trans, respectively.

■

■

REFERENCES

(1) Feringa, B. L. The Art of Building Small: From Molecular Switches to Motors (Nobel Lecture). Angew. Chem., Int. Ed. 2017, 56, 11059−11078. (2) van Leeuwen, T.; Lubbe, A. S.; Stacko, P.; Wezenberg, S. J.; Feringa, B. L. Dynamic Control of Function by Light-Driven Molecular Motors. Nat. Rev. Chem. 2017, 1, 0096. (3) Kistemaker, J. C. M.; Stacko, P.; Roke, D.; Wolters, A. T.; Heideman, G. H.; Chang, M. C.; van der Meulen, P.; Visser, J.; Otten, E.; Feringa, B. L. Third-Generation Light-Driven Symmetric Molecular Motors. J. Am. Chem. Soc. 2017, 139, 9650−9661. (4) Kassem, S.; van Leeuwen, T.; Lubbe, A. S.; Wilson, M. R.; Feringa, B. L.; Leigh, D. A. Artificial Molecular Motors. Chem. Soc. Rev. 2017, 46, 2592−2621. (5) Conyard, J.; Cnossen, A.; Browne, W. R.; Feringa, B. L.; Meech, S. R. Chemically Optimizing Operational Efficiency of Molecular Rotary Motors. J. Am. Chem. Soc. 2014, 136, 9692−9700. (6) Pollard, M. M.; Meetsma, A.; Feringa, B. L. A Redesign of LightDriven Rotary Molecular Motors. Org. Biomol. Chem. 2008, 6, 507− 512. (7) Wang, Y.; Tian, Y. C.; Chen, Y. Z.; Niu, L. Y.; Wu, L. Z.; Tung, C. H.; Yang, Q. Z.; Boulatovc, R. A Light-Driven Molecular Machine Based on Stiff Stilbene. Chem. Commun. 2018, 54, 7991−7994. (8) Li, Y. Y.; Liu, F. Y.; Wang, B.; Su, Q. Q.; Wang, W. L.; Morokuma, K. Different Conical Intersections Control Nonadiabatic Photochemistry of Fluorene Light-Driven Molecular Rotary Motor: A CASSCF and Spin-Flip DFT Study. J. Chem. Phys. 2016, 145, 244311. (9) Conyard, J.; Stacko, P.; Chen, J. W.; McDonagh, S.; Hall, C. R.; Laptenok, S. P.; Browne, W. R.; Feringa, B. L.; Meech, S. R. Ultrafast Excited State Dynamics in Molecular Motors: Coupling of Motor Length to Medium Viscosity. J. Phys. Chem. A 2017, 121, 2138−2150. (10) Beekmeyer, R.; Parkes, M. A.; Ridgwell, L.; Riley, J. W.; Chen, J. W.; Feringa, B. L.; Kerridge, A.; Fielding, H. H. Unravelling the Electronic Structure and Dynamics of an Isolated Molecular Rotary Motor in the Gas-Phase. Chem. Sci. 2017, 8, 6141−6148. (11) Amirjalayer, S.; Cnossen, A.; Browne, W. R.; Feringa, B. L.; Buma, W. J.; Woutersen, S. Direct Observation of a Dark State in the

CONCLUSIONS A series of transient absorption and low-temperature steady state photolysis measurements have been performed to characterize the photochemical pathways for an efficient firstgeneration four-stage rotary molecule motor. Our results demonstrate that the molecular motor is not only designed for unidirectional isomerization, but it is also “primed” for efficient motion with high quantum yields for the desired photoreactions and low quantum yields for the undesired back reactions. Pulse sequences, exciting the molecule higher into the electronic manifold, have a modest effect on the reaction. The formation of M-trans can be increased using a simple pulse sequence, but the formation of M-cis is only decreased. That only a small increase is observed for the P-cis→M-trans reaction, and no increase for the P-trans→M-cis reaction, is not surprising given the unexpected high quantum yields for the one-photon isomerization reactions. This strategy may be more effective for second generation motors where the intrinsic quantum yield is much lower. In addition, pulse sequences using wavelengths other than 404 and 808 nm and different time delays may access pathways to bypass the ratelimiting thermal helix inversion. Finally, these results should inspire additional theoretical simulations to explore the underlying reasons for the high quantum yields reported here. Such an analysis will help in the design of efficient unidirectional molecular motors.

■

Diagram of the pulse sequence in pump-repump-probe measurements. Additional data plots referred to in the text. A description of the synthesis of the molecular motors. Derivation of the formulas for EADS. Details for the estimate of the fluorescence quantum yields. Additional details of the fitting process. Coordinates obtained for DFT optimized ground state structures of P-tran, P-cis, M-trans, and M-cis conformations (PDF)

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.8b06472. J

DOI: 10.1021/acs.jpca.8b06472 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A Photocycle of a Light-Driven Molecular Motor. J. Phys. Chem. A 2016, 120, 8606−8612. (12) Conyard, J.; Addison, K.; Heisler, I. A.; Cnossen, A.; Browne, W. R.; Feringa, B. L.; Meech, S. R. Ultrafast Dynamics in the Power Stroke of a Molecular Rotary Motor. Nat. Chem. 2012, 4, 547−551. (13) Hall, C. R.; Conyard, J.; Heisler, I. A.; Jones, G.; Frost, J.; Browne, W. R.; Feringa, B. L.; Meech, S. R. Ultrafast Dynamics in Light-Driven Molecular Rotary Motors Probed by Femtosecond Stimulated Raman Spectroscopy. J. Am. Chem. Soc. 2017, 139, 7408− 7414. (14) Pang, X. J.; Cui, X. Y.; Hu, D. P.; Jiang, C. W.; Zhao, D.; Lan, Z. G.; Li, F. L. Watching″ the Dark State in Ultrafast Nonadiabatic Photoisomerization Process of a Light-Driven Molecular Rotary Motor. J. Phys. Chem. A 2017, 121, 1240−1249. (15) Liu, F. Y.; Morokuma, K. Computational Study on the Working Mechanism of a Stilbene Light-Driven Molecular Rotary Motor: Sloped Minimal Energy Path and Unidirectional Nonadiabatic Photoisomerization. J. Am. Chem. Soc. 2012, 134, 4864−4876. (16) Amatatsu, Y. Theoretical Study of Topographical Features around the Conical Intersections of Fluorene-Based Light-Driven Molecular Rotary Motor. J. Phys. Chem. A 2013, 117, 3689−3696. (17) Wang, L. L.; Huan, G.; Momen, R.; Azizi, A.; Xu, T. L.; Kirk, S. R.; Filatov, M.; Jenkins, S. QTAIM and Stress Tensor Characterization of Intramolecular Interactions Along Dynamics Trajectories of a Light-Driven Rotary Molecular Motor. J. Phys. Chem. A 2017, 121, 4778−4792. (18) Nikiforov, A.; Gamez, J. A.; Thiel, W.; Filatov, M. Computational Design of a Family of Light-Driven Rotary Molecular Motors with Improved Quantum Efficiency. J. Phys. Chem. Lett. 2016, 7, 105− 110. (19) Filatov, M.; Olivucci, M. Designing Conical Intersections for Light-Driven Single Molecule Rotary Motors: From Precessional to Axial Motion. J. Org. Chem. 2014, 79, 3587−3600. (20) Filatov, M.; Huix-Rotllant, M. Assessment of Density Functional Theory Based Delta SCF(Self-Consistent Field) and Linear Response Methods for Longest Wavelength Excited States of Extended Pi-Conjugated Molecular Systems. J. Chem. Phys. 2014, 141, 024112. (21) Kazaryan, A.; Kistemaker, J. C. M.; Schafer, L. V.; Browne, W. R.; Feringa, B. L.; Filatov, M. Understanding the Dynamics Behind the Photoisomerization of a Light-Driven Fluorene Molecular Rotary Motor. J. Phys. Chem. A 2010, 114, 5058−5067. (22) Kazaryan, A.; Filatov, M. Density Functional Study of the Ground and Excited State Potential Energy Surfaces of a Light-Driven Rotary Molecular Motor (3R,3 ’ R)-(P,P)-trans-1,1 ’,2,2 ’,3,3 ’,4,4 ’-Octahydro-3,3 ’-dimethyl-4,4 ’-biphenanthrylidene. J. Phys. Chem. A 2009, 113, 11630−11634. (23) ter Wiel, M. K. J.; van Delden, R. A.; Meetsma, A.; Feringa, B. L. Increased Speed of Rotation for the Smallest Light-Driven Molecular Motor. J. Am. Chem. Soc. 2003, 125, 15076−15086. (24) McMurry, J. E. Carbonyl-Coupling Reactions Using LowValent Titanium. Chem. Rev. 1989, 89, 1513−1524. (25) Quick, M.; Berndt, F.; Dobryakov, A. L.; Ioffe, I. N.; Granovsky, A. A.; Knie, C.; Mahrwald, R.; Lenoir, D.; Ernsting, N. P.; Kovalenko, S. A. Photoisomerization Dynamics of Stiff-Stilbene in Solution. J. Phys. Chem. B 2014, 118, 1389−1402. (26) Takeda, A.; Tsubouchi, A. In Science of Synthesis: Houben-Weyl Methods of Molecular Transformations, de Meijere, A., Ed.; Georg Thieme: Stuttgart, Germany, 2009; Vol. 47a, pp 247−325. (27) Gene, G.; Victor, P. Separable Nonlinear Least Squares: The Variable Projection Method and its Applications. Inverse Probl 2003, 19, R1−R26. (28) Golub, G. H.; Pereyra, V. Differentiation of Pseudo-Inverses and Nonlinear Least-Squares Problems Whose Variables Separate. Siam J. Numer. Anal. 1973, 10, 413−432. (29) Kaufman, L. A variable Projection Method for Solving Separable Nonlinear Least Squares Problems. BIT Num. Math. 1975, 15, 49−57.

(30) Mullen, K. M.; van Stokkum, I. H. M. TIMP: An R Package for Modeling Multi-way Spectroscopic Measurements. J. Stat. Soft. 2007, 18, 1 DOI: 10.18637/jss.v018.i03. (31) Nagle, J. F.; Parodi, L. A.; Lozier, R. H. Procedure for Testing Kinetic-Models of the Photocycle of Bacteriorhodopsin. Biophys. J. 1982, 38, 161−174. (32) Snellenburg, J. J.; Laptenok, S. P.; Seger, R.; Mullen, K. M.; van Stokkum, I. H. M. Glotaran: A Java-Based Graphical User Interface for the R Package TIMP. J. Stat. Soft. 2012, 49, 1 DOI: 10.18637/ jss.v049.i03. (33) van Stokkum, I. H. M.; Larsen, D. S.; van Grondelle, R. Global and Target Analysis of Time-Resolved Spectra (vol 1657, pg 82, 2004). Biochim. Biophys. Acta, Bioenerg. 2004, 1658, 262−262. (34) van Stokkum, I. H. M.; Larsen, D. S.; van Grondelle, R. Global and Target Analysis of Time-Resolved Spectra. Biochim. Biophys. Acta, Bioenerg. 2004, 1657, 82−104. (35) Saltiel, J.; Gupta, S. Photochemistry of the Stilbenes in Methanol. Trapping the Common Phantom Singlet State. J. Phys. Chem. A 2018, 122, 6089−6099. (36) Ioffe, I. N.; Quick, M.; Quick, M. T.; Dobryakov, A. L.; Richter, C.; Granovsky, A. A.; Berndt, F.; Mahrwald, R.; Ernsting, N. P.; Kovalenko, S. A. Tuning Stilbene Photochemistry by Fluorination: State Reordering Leads to Sudden Polarization near the FranckCondon Region. J. Am. Chem. Soc. 2017, 139, 15265−15274. (37) Sension, R. J.; Repinec, S. T.; Szarka, A. Z.; Hochstrasser, R. M. Femtosecond Laser Studies of the Cis-Stilbene Photoisomerization Reactions. J. Chem. Phys. 1993, 98, 6291−6315. (38) Kim, S. K.; Courtney, S. H.; Fleming, G. R. Isomerization of Trans-Stilbene in Alcohols. Chem. Phys. Lett. 1989, 159, 543−548. (39) Kim, S. K.; Fleming, G. R. Reorientation and Isomerization of Trans-Stilbene in Alkane Solutions. J. Phys. Chem. 1988, 92, 2168− 2172. (40) Hicks, J. M.; Vandersall, M. T.; Sitzmann, E. V.; Eisenthal, K. B. Polarity-Dependent Barriers and the Photoisomerization Dynamics of Molecules in Solution. Chem. Phys. Lett. 1987, 135, 413−420. (41) Rice, J. K.; Baronavski, A. P. Ultrafast Studies of Solvent Effects in the Isomerization of Cis-Stilbene. J. Phys. Chem. 1992, 96, 3359− 3366. (42) CRC Handbook of Chemistry and Physics, 99th ed.; Rumble, J. R., Ed.; CRC Press: Boca Raton, FL, 2018. (43) Nguyen, T. S.; Koh, J. H.; Lefelhocz, S.; Parkhill, J. Black-Box, Real-Time Simulations of Transient Absorption Spectroscopy. J. Phys. Chem. Lett. 2016, 7, 1590−1595. (44) Shiang, J. J.; Walker, L. A., II; Anderson, N. A.; Cole, A. G.; Sension, R. J. Time-Resolved Spectroscopic Studies of B12 Coenzymes: The Photolysis of Methylcobalamin Is Wavelength Dependent. J. Phys. Chem. B 1999, 103, 10532−10539. (45) Walker, L. A., II; Jarrett, J. T.; Anderson, N. A.; Pullen, S. H.; Matthews, R. G.; Sension, R. J. Time-Resolved Spectroscopic Studies of B12 Coenzymes: The Identification of a Metastable Cob(III)alamin Photoproduct in the Photolysis of Methylcobalamin. J. Am. Chem. Soc. 1998, 120, 3597−3603. (46) Strickler, S. J.; Berg, R. A. Relationship Between Absorption Intensity and Fluorescence Lifetime of Molecules. J. Chem. Phys. 1962, 37, 814−822. (47) Houk, A. L.; Givens, R. S.; Elles, C. G. Two-Photon Activation of p-Hydroxyphenacyl Phototriggers: Toward Spatially Controlled Release of Diethyl Phosphate and ATP. J. Phys. Chem. B 2016, 120, 3178−3186. (48) Ward, C. L.; Elles, C. G. Controlling the Excited-State Reaction Dynamics of a Photochromic Molecular Switch with Sequential TwoPhoton Excitation. J. Phys. Chem. Lett. 2012, 3, 2995−3000. (49) Ward, C. L.; Elles, C. G. Cycloreversion Dynamics of a Photochromic Molecular Switch via One-Photon and Sequential Two-Photon Excitation. J. Phys. Chem. A 2014, 118, 10011−10019. (50) Hall, C. R.; Browne, W. R.; Feringa, B. L.; Meech, S. R. Mapping the Excited-State Potential Energy Surface of a Photomolecular Motor. Angew. Chem., Int. Ed. 2018, 57, 6203−6207.

K

DOI: 10.1021/acs.jpca.8b06472 J. Phys. Chem. A XXXX, XXX, XXX−XXX