Dynamic Electric Field Complicates Chemical ... - ACS Publications

Subscriber access provided by Bibliothèque de l'Université Paris-Sud

Chemical and Dynamical Processes in Solution; Polymers, Glasses, and Soft Matter

Dynamic Electric Field Complicates Chemical Reactions in Solutions Yaokun Lei, Jun Zhang, Zhen Zhang, and Yiqin Gao J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.9b01038 • Publication Date (Web): 16 May 2019 Downloaded from http://pubs.acs.org on May 16, 2019

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 21 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 Letters

Dynamic Electric Field Complicates Chemical Reactions in Solutions Yaokun Leia,#, Jun Zhanga,#, Zhen Zhangb, Yi Qin Gaoa,* aInstitute

of Theoretical and Computational Chemistry, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China. bDepartment of Physics, Tangshan Normal University, Tangshan 063000, China

Keywords: dynamic effects, electrostatic stabilization, oriented external electric field, heterogeneous reaction paths

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters 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

Abstract: Chemical reactions can be strongly influenced by external electric field (EEF). But as EEF is often time-dependent and in case that it doesn’t adapt quickly enough to reaction progress especially during fast barrier crossing processes, dynamic effects could be important. Here, we find that electrostatic interactions can reduce the height of reaction barrier for a Claissen Rearrangement reaction and accelerate the key motions for bonding. Meanwhile, strong electrostatic interactions can modify the barrier into an effective potential well, confining the system into the barrier until solvents adjust themselves to provide an appropriate EEF for charge redistribution. In this case, the otherwise concerted mechanism becomes a stepwise one. Consequently, the motion of solvents modulates the reaction dynamics and leads to heterogeneous reaction paths even in a seemingly homogenous aqueous solution. In addition, an excessive stabilization of TS retards the barrier crossing process, making the thermodynamically favorable pathway less favored dynamically. TOC Graphic：

All chemical reactions can be viewed as the movement of electrons and/or nuclei;


Page 2 of 21

Page 3 of 21 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


as such, one expects that their kinetics and thermodynamics could be influenced by external fields. Many theoretical studies predict that electrostatic effects should in principle influence the stabilities of chemical species by stabilizing charge-separated resonance contributors1,2. More specifically, if the transition state (TS) is more polar or can be more polarized by external field than the reactant, then the reaction barrier can be largely reduced by a favorably oriented external electric field (OEEF). Indeed, the elegant experiment of an OEEF-catalyzed Diels-Alder reaction confirmed these previous theoretic predictions3. In addition to artificially applied electric field4,5, it’s important to note that nature has devised chemical species with built-in electric fields that catalyze a variety of reactions by stabilizing ionic resonant structures in their TSs. For example, the catalytic power of enzymes has been attributed6,7 to preorganized polar groups that stabilize transition states and impart electrostatic catalysis on their respective reactions. By mutating different residues in the active site of ketosteroide isomerase, Stephen D. Fried and Steven G. Boxer found8,9 by vibrational stark spectroscopy that the increment of activation barrier due to key mutations strongly correlates with the extent of electric field decrease caused by the mutant 10,11. Two additional problems related to OEEF need to be addressed. Firstly, when it comes to catalytic effects of OEEFs, more attention was paid to the decrease of the height of activation barrier than to the shape of potential surface and therefore the force. In fact, previous theoretic calculations have predicted that the mechanism of DielsAlder reaction (concerted or stepwise) can be changed by the external electric field12,13. Secondly, the electric field exerted by solvents is time dependent and may not adjust rapidly enough to charge transport in chemical reactions14 especially during barrier crossing so that not only the free energy but also dynamic effects should be taken into account15,16. As a result of unfavorable EEF, the charge redistribution will be trapped until solvents adapt themselves to this process, which is called dynamic polarization cage17 in Grote-Hynes Theory18–20. In our previous article hereafter referred as Part I21, we found that in aqueous solution there are two classes of trajectories for Claisen rearrangement (Scheme 1 ) with different transition path lengths by machine leaning method36 (Supporting information, Text section Ⅲ), which we name Type A and Type B respectively for convenience. The bond formation occurs in a more concerted way with bond breaking in Type A than in Type B trajectories. Along Type B trajectories the solute is more likely trapped in a potential well created by the external interaction, forming an intermediate state. The continuation of the reaction awaits for the proper adjustment of solvents. Actually, analysis in part I suggests that the hydrogen bonding with the solvent has an important influence on the progression of the reaction. Moreover, the hydrogen bond vibration shows a similar timescale with the reaction coordinate change in the TS region, which implies that this external interaction is “dynamic” and non-equilibrium effects should be considered. In this article, we will show the important role dynamic effects play in barrier crossing and explain why there are distinct reaction paths in the Claisen rearrangement reaction in homogenous solutions. Our results also reveal that an excessive stabilization of TS may retard the barrier cross process, making the thermodynamically favorable pathway not necessarily favored in the dynamic sense.

Scheme 1 Scheme of forward and backward reactions: cis-2-vinylcyclopropanecarboxaldehyde (left)



Page 4 of 21

interconverts to 2,5-dihydrooxepin (right), with the bond forming/breaking sites indexed in red characters.

In part I, the Claisen rearrangement reaction (Scheme 1) is trapped during the barrier crossing along type B trajectories (Figure 3 (d)). According to Grote-Hynes Theory, the barrier crossing cannot be finished until the environment adapts itself to this process in the polarization cage regime, where the barrier is modified into an effective well by strong solvent force. Based on our earlier calculation21, the system falls well into this regime (Supporting Information, Text section Ⅳ). Modified Effective Potential Well. Firstly, we perform a detailed analysis on the effective potential surface for the two types of trajectories. For the convenience of description, we define the bond length of C3-C8 and C12-O15 as 𝑑1 and 𝑑2, respectively. And in all trajectories, the origin of time is realigned, so that the first 2000 fs corresponds to reactant, the last 2000fs to product, and the reaction takes place at around ~2000 fs. By calculating the force constant (𝐾1, 𝐾2) of the carbon-carbon bond (C3-C8) and carbon-oxygen bond (C12-O15), it was found that the interval between the critical points (𝐾𝑖 < 0,𝑖 = 1,2) for bond breakage and formation is longer in type B than type A trajectories (Figure 1 (a) and (b)). For type B trajectories, it’s clear that 𝐾1 decreases to below zero before being trapped (2000-2300fs) but 𝐾2 remains positive till about 2300fs, showing that during this trapping period (2000-2300fs) C3C8 is broken without C12-O15 being formed (Figure 1 (b)). In fact, potential barriers exist for both bond making and bond breaking steps. Therefore, the longer the time the reactant resides in this region, the more likely the potential well is modified by solvents which retards the barrier crossing. ∂2𝐻

∂2𝐻

𝐾1 = ∂𝑑1∂𝑑1;𝐾2 = ∂𝑑2∂𝑑2

(1)

where 𝐻 is the Hamiltonian of entire system. From Figure 1 (a) and (b), it’s obvious that the reactant resides in this region for a longer time in a type B than a type A trajectory, which is consistent with the existence of a potential well for reaction along the former as previously mentioned. The bond forming and breaking steps are more concerted in Type A than in Type B trajectories. It is intriguing to understand the reason behind the lingering of the reactant near the transition state. Therefore, we try to identify the processes that are related to the trapping of the reactant. Since the most predominant change for the chemical reaction in the TS region is the redistribution of electrons, we suspect this trapping could result from the retarded charge redistribution of the reacting species. Figure 1 (c) and (d) show that the negative charges (electrons) mainly flow from C12 to O15, and as the reaction proceed, further to C1. Comparing these two figures, it is clear that the charge redistribution is very rapid for the Type A trajectory, whereas it becomes significantly hindered during the transition state trapping (2000-2300fs) in a type B trajectory. This difference is a common feature for all trajectories examined.


Page 5 of 21 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 (a) Force Constant of C3-C8 and C12-O15 for a type A trajectory. 𝐾1 (Black Line), 𝐾2 (Blue Line). Bond Length of C3-C8: 𝑑1 (Red Line). When force constant is less than zero, bonding or bond breaking has been finished. For the Type A trajectory, bond breaking and bonding is very close in time. (b) The same as (a) but for Type B trajectory. (c) Mulliken Charge for C1 (blue line), C12 (green line) and O15 (red line) for the Type A trajectory. (d) The same as (c) but for a Type B trajectory. Charge redistribution is retarded during 2000-2300fs.

Electrostatic interaction aids reactant in approaching TS. Next, we investigate the contributions of the solvents to electrostatics during the reaction progress. Specifically, we examine how solvents assist the system in approaching TS by activating the reactant (or rather, imposing work on solute) and/or altering the reaction barrier. Firstly, the electric field induced by water can significantly reduce the reaction barrier22,23. To elucidate this effect, we investigate the intra-molecular potential 𝐸𝑖𝑛𝑡𝑟𝑎 and the intermolecular potential 𝐸𝑖𝑛𝑡𝑒𝑟 following the reaction progression (RC) averaging over all trajectories. 𝑅𝐶 = 0.82𝑑1 ― 0.18𝑑2 In the above equation, RC is the reaction coordinate optimized using the Bayesian measure24,25 in Part I. According to Figure 2 (a) and (b), the intra-molecular barrier is about 17kcal/mol and the electrostatic interaction causes the stabilization of TS as well as destabilization of the reactant. This change leads to a reduction of the barrier up to ~10 kcal/mol (Figure 2 (b)). This reduction can be interpreted in term of the change of charge distribution. In the previous section we showed that the electron is initially enriched at O15 as the system approaches TS(the first step of charge redistribution, Figure 2 (d)). Therefore, the TS can be effectively stabilized by electrostatic interaction with nearby water, resulting in a shallow barrier. Such an effect echoes a mechanism called electrostatic catalysis6,26,27. On the other hand, due to a strong electrostatic interaction between the solute and solvent, work can be done on solute to accelerate certain specific motions. In addition, due to the rotational motion of the solute, strong coupling between the rotation and stretch motions of C12-O15 due to large derivative of the moment of inertia with respect to the bond length of C12-O15 will strongly influence the approach of C12 and O15, corresponding to a Coriolis effect. The charge redistribution is prompted by the compression of the C12-O15 distance, which is also promoted by solvents and Coriolis effects. To quantify these effects, we calculate the work (𝑊𝑠𝑜𝑙, 𝑊𝑄𝑀, eq (2)) done on the stretch motion of C12-O15, by solvents as well as the rest part of the solute, and the energy change 𝑊𝑣𝑏 due to the vibration-rotation coupling by applying eq (3) derived from Poisson bracket28,29.



𝑡

𝑡

𝑊𝑠𝑜𝑙(𝑡) = ∫0𝐹𝑠𝑜𝑙𝑣𝑐𝑜𝑑𝑡;𝑊𝑄𝑀(𝑡) = ∫0𝐹𝑄𝑀𝑣𝑐𝑜𝑑𝑡 𝑡

∂𝐼𝑖

𝑊𝑣𝑏(𝑡) = ―0.5 ∗ ∑𝑖 = 𝑥,𝑦,𝑧∫0𝑣𝑐𝑜∂𝑑2𝜔2𝑖 𝑑𝑡

Page 6 of 21

(2) (3)

𝑑𝑑2

𝑣𝑐𝑜 = 𝑑𝑡 . 𝐹𝑠𝑜𝑙 and 𝐹𝑄𝑀 are the forces applied to the stretch motion of C12-O15 by solvents and solute, respectively. 𝐼𝑖 is the moment of inertia and 𝜔𝑖 angular velocity. As Figure 2 (c) shows, the work contribution from solvents and Coriolis effect outweighs that of the solute before 2000fs, showing that solvents and Coriolis effects play an important role in activating the reactant and promoting the reactive modes. Furthermore, Figure 2 (a) and (b) show that the interaction between solutes and solvent can modify the intramolecular potential surface into a well near the TS. More interestingly, this well is deeper in type B than type A trajectories due to stronger electrostatic interaction.

Figure 2 (a) Intramolecular Potential Energy averaging over all reactive trajectories (b) Interaction between solute and solvents (Electrostatic Energy plus van der waals Energy) averaging over all reactive trajectories (Blue Line), only Type A trajectories (Black Line) and only Type B trajectories (Red Line) (c) Work done on stretch motion of C12-O15 by solvents (Red Line) and solute (Black Line). Energy change due to vibration-rotation coupling (Blue Line) (d) O15 charge averaging over all reactive trajectories (Blue Line), only Type A trajectories (Black Line) and only Type B trajectories (Red Line)

Oriented External Electric Field. The above analysis shows that the reactant could be trapped in a transient potential well induced by improper solvent configurations for a significant period of time near the TS, during which electrostatic interactions play a critical role. Therefore, we calculated the external electric field imposed on C12 projected along the C12-O15 vector and that imposed on O15 projected along the C1O15 vector by applying eq (4) and eq (5). 𝑞𝑖 𝒓𝑂15,𝑖 ∙ 𝒓𝑂15 ― C1 (4) 𝐸2 = ∑𝑖𝜖𝑠𝑜𝑙𝑣𝑒𝑛𝑡𝑠 2 ∗ (| 𝒓 | ∗ |𝒓 |) |𝒓𝑂15,𝑖| 𝑞𝑖

𝐸1 = ∑𝑖𝜖𝑠𝑜𝑙𝑣𝑒𝑛𝑡𝑠|𝒓

𝐶12,𝑖|

2

𝑂15 ― C1

𝑂15,𝑖

𝒓𝐶12,𝑖 ∙ 𝒓𝐶12 ― 𝑂15

∗ (|𝒓𝐶12 ― 𝑂15| ∗ |𝒓𝐶12,𝑖|)

(5)

𝒓𝑂15,𝑖 = 𝒓𝑂15 ― 𝒓𝑖, 𝒓𝐶12,𝑖 = 𝒓𝐶12 ― 𝒓𝑖, 𝒓𝑂15 ― 𝐶1 = 𝒓𝑂15 ― 𝒓𝐶1, 𝒓𝐶12 ― 𝑂15 = 𝒓𝐶12 ― 𝒓𝑂15.

𝐸1 and 𝐸2 are sketched in Scheme 2


Page 7 of 21 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


Scheme 2 sketch map for electric field (𝐸2, 𝐸1) and angle θ

Figure 3 (a) The change of 𝐸1 following reaction progression averaging over all reactive trajectories. (b) The change of 𝐸2 following reaction progression averaging over all reactive trajectories, Type A trajectories and Type B trajectories. Time series of External Electric Fields (𝐸1, 𝐸2) and key bond lengths (𝑑1, 𝑑2) for (c) Type A trajectory (d) Type B trajectory (e) Unreactive trajectory.

From Figure 3 (a), one observes that 𝐸1 gradually increases to 50MV/cm as the reaction proceeds (along RC), favoring the charge flux from C1 to O15. To understand in more details the differences among type A, type B and nonreactive trajectories, we plot the specific time series of 𝐸1 for these three cases in Figure 3 (c), (d) and (e). It’s clear from these figures that for reactive trajectory the external electric field pre-adjusts relatively sharply before 𝑑1 starts to increase, and then fluctuates slightly at around 50MV/cm. In contrast, along the nonreactive trajectory we do not observe a significant enhancement of 𝐸1 prior to the reaction. Instead, 𝐸1 becomes more negative during the barrier crossing process (at around 2000fs) (Figure 3 (e)), which is unfavorable for keeping C12 and O15 close to each other. Therefore, in the latter case the system is



likely to re-cross back to the reaction basin which leads to a transition failure even after it has entered the TS. To understand the molecular origin for the change of external electric field, we decompose it into contributions from individual water molecules. Interestingly, we find that the electrostatic interaction to which the charge redistribution is susceptible is very local and the change of EEF results majorly from the motion of water molecules Hbonded with O15 (WAT2,WAT3 in Figure 4), and the water molecule nearest to C12 (WAT1 in Figure 4). These two H-bonded water molecules (WAT2, WAT3) can have attractive interactions with both C12 and O15, confining their separation into small distances (positive contribution to 𝐸1 as shown in Figure 4 (a-c)). Moreover, there is another water molecule WAT1 which behaves differently in reactive and non-reactive trajectories. In reactive trajectories, it plays a similar role as WAT2 and WAT3, but in non-reactive trajectories it functions to pull C12 apart from O15 (negative contribution to 𝐸1 as shown in Figure 4 (c)). In reactive trajectories, the oxygen atom in WAT1 is closer to the negatively charged C12 than in non-reactive ones, as shown in Figure 4 (d-e) , thus restricting C12 from moving away from O15 (repulsive interaction between O15 and Oxygen atom of WAT 1). In contrast, in non-reactive trajectories WAT1 can form a H-bond with C12 as shown in Figure 4 (f), leading to an inappropriately oriented electric field. These results show that the pre-adjustment of 𝐸1 due to solvent motion plays an important role for the successfulness of the reaction.

Figure 4 Decomposition of 𝐸1 by water molecules (a) Type A trajectory (b) Type B trajectory (c) Unreactive trajectory. The conformation in TS for (d) Type A trajectory (e) Type B trajectory (f) Unreactive trajectory

Reaction dynamics modified by motion of solvent. The pre-adjustment of 𝐸1 mentioned above occurs in both type A and type B reactive trajectories. Next, we averaged 𝐸2 over all reactive trajectories as a function of RC which increases from 100MV/cm to -25MV/cm. This change suggests that the solvent’s initial configuration


Page 8 of 21

Page 9 of 21 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


is unfavorable for the charge redistribution (from O15 to C1) but re-adjusts as the reaction progresses (Figure 3 (b)). More interestingly, 𝐸2 is higher in Type A than in Type B trajectories. Such a difference arises because 𝐸2 increases rapidly when RC increases from 1 to 1.6 Å in type A trajectories but stay at around -125 MV/cm in type B trajectories, implying that solvents can adjust more promptly in Type A trajectories. It can be seen from the time series of 𝐸2 that in type A trajectories 𝐸2 sharply increases almost synchronously as 𝑑1 decreases (around 2000fs in Figure 3 (c)), whereas for type B trajectories at about the same reaction progress (around 2000fs in Figure 3 (d)), 𝐸2 stays below -100MV/cm. In the latter case the electron flux from O15 to C1 is strongly hindered. Actually, the stronger hydrogen bond in type B trajectories obstructs the electron flux from O15 to C1. We also decompose 𝐸2 into contributions from individual water molecules. The major contributions to the change of 𝐸2 again comes from H-bonded water molecules (WAT2 and WAT3, Figure 5). Since the most contribution to 𝐸2 stems from water molecules H-bonded to O15, we use the H-bond length and 𝜃 (Scheme 2) to characterize the motion of these water molecules. Since in the current model, the charge of water molecules is fixed, the increase of H-bond length and the decrease of 𝜃 both increase 𝐸2 (Scheme 2). Time evolutions of these two variables reveal that the H-bond length is largely elongated and θ decreases when 𝐸2 increases, as shown in Figure 5. These results together clearly show that although the electronic stabilization plays a significant part in the reaction process, the external electric field is time-dependent rather than static and not necessarily responds instantly to reaction process especially during a fast barrier crossing process.

Figure 5 The decomposition of 𝐸2 by water molecules for (a) Type A trajectory (b) Type B trajectory (c) Unreactive trajectory. H-bond length and θ of WAT2 for (d) Type A trajectory (e) Type B trajectory (f) Unreactive trajectory. H-bond length and θ of WAT3 for (g) Type A trajectory (h) Type B trajectory (i) Unreactive trajectory.

In summary, the main question we want to address in this study is whether and, if yes, why, there are multiple reaction paths in a seemingly homogenous reaction such as the Claisen rearrangement in a simple solvent. The current analysis shows that although strong H-bonding results in a reduction of barrier height by stabilizing the TS (Figure 2 (b), Type B trajectories) with an elevated negative charge on O15 (Figure 2 (d), Type B trajectories), it may, on the other hand, lead to an inappropriately oriented EEF (𝐸2) which blocks the charge redistribution (from O15 to C1). In other words, the stronger



electrostatic interaction can modifies the barrier into a deeper transient potential well (Figure 2 (a) and (b)) so that it takes a longer time for the system to cross the second barrier, transforming the concert mechanism into a stepwise one. Actually, Figure 2 (a) and (b) show that electrostatic interaction can modify the intramolecular potential surface into a well near the barrier and this well is a few (about 5) kcal/mol deeper for type B than type A pathways. As a result, the reaction becomes halted along type B trajectories until an appropriate external electric field is generated by motion of solvents. Moreover, since the adjustment of 𝐸2 majorly results from the elongation of H-bond and the decrease of θ, stronger H-bonds in turn render a slower adjustment of water molecules. We conclude from these observations that 1), the motion of solvents is coupled to the reaction dynamics, leading to two distinctly different reaction pathways even in a seemingly homogenous aqueous solution due to the existence of different transient solvation configurations. Such an effect was recognized as temporal heterogeneousity30,31. 2), the thermodynamically favored pathway is not necessarily favored in the dynamic sense. In fact, these phenomena correlate with non-equilibrium reaction dynamics that appear when there are mismatched timescales for the chemical transition and the motion of solvents, with the former being so rapid that the latter cannot follow instantly to reach an equilibrium state. In such a case the dynamical effects of the reaction environment should be considered. The non-equilibrium reaction dynamics can result in the existence of different transient solvation configurations with distinctly different strengths of interaction between solutes and solvents. Since the dominant interaction between solvents and solute is due to the local H-bond, the transition paths can be separated into two rather distinctive types, one with and another without H-bonding to oxygen atom (O15 in Scheme 2). We speculate that when the interaction between the solute and solvents become less local (as in ionic liquid and enzymatic active site), the media may provide a more continuously distributed electrostatic environment, and less distinctive transition path types. Therefore, we expect that similar effects will be generally present in many other condensed-phase reactions, and might be even more pronounced in heterogeneous reaction media such as enzymes and interfaces. Actually, previous studies32,33 have found that fast motions of enzyme lead to an increase of the electric field by preparing an enzymatic configuration which is electrostatically favorable. And a single enzyme is able to sample different active configurations with different catalytic efficiency34. Therefore, these effects should be considered in designing efficient enzymes and other catalysts. Experimental Methods QM/MM simulations were performed on AMBER 10 MD platform. Simulation details are included in Supporting information, Text section I. For configurational and reactive trajectory sampling, the MD simulation were carried out under an NPT ensemble (300K, 1atm). Transition path shootings were executed under an NVE ensemble. To increase the efficiency of configurational and reactive trajectory sampling, the Multilevel Integrated Tempering Sampling (MITS21, Supporting information, Text section Ⅱ) and enhanced sampling of reactive trajectories (ESoRT35) protocol were adopted. The clustering of reactive trajectories was performed with spectral clustering (Supporting information, Text section Ⅲ)


Page 10 of 21

Page 11 of 21 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


Acknowledgement The authors thank the Beijing Computing Center for providing the computational resources. This research is financially supported by National Natural Science Foundation of China (21573006, 21821004, 21873007) and National Key R&D Program of China (2017YFA0204702).

Supporting Information The supporting information is available free of charge on the ACS Publication website at xxx. Simulation details, Multilevel Integrated Tempering Sampling (MITS), Spectral Clustering of Transitions Paths, Grote-Hynes Theory.

Author Information Corresponding authors: *(Yi Qin Gao): Phone: 86-10-6275-2431. Email: [email protected] ORCID Yi Qin Gao: 0000-0002-4309-9376 Author contributions: #Y-K.L and J.Z. contributed equally to this work. Y-K.L., J.Z., Z.Z. and Y.Q.G. designed research; Y-K.L., J.Z., Z.Z. and Y.Q.G. performed research; Y-K.L., J.Z., Z.Z. and Y.Q.G. analyzed the data; Y-K.L., J.Z. and Y.Q.G. wrote the paper. Notes: The authors declare no conflict of interest.

References: (1) (2)

(3)

(4) (5) (6) (7) (8)

Shaik, S.; Mandal, D.; Ramanan, R. Oriented Electric Fields as Future Smart Reagents in Chemistry. Nat. Chem. 2016, 8 (12), 1091–1098. https://doi.org/10.1038/nchem.2651. Hiberty, P. C.; Megret, C.; Song, L.; Wu, W.; Shaik, S. Barriers of Hydrogen Abstraction vs Halogen Exchange: An Experimental Manifestation of ChargeShift Bonding. J. Am. Chem. Soc. 2006, 128 (9), 2836–2843. https://doi.org/10.1021/ja053130m. Wallace, G. G.; Diez-Perez, I.; Ciampi, S.; Coote, M. L.; Aragonès, A. C.; Darwish, N.; Haworth, N. L.; Bloomfield, N. J. Electrostatic Catalysis of a Diels–Alder Reaction. Nature 2016, 531 (7592), 88–91. https://doi.org/10.1038/nature16989. Alemani, M.; Peters, M. V.; Hecht, S.; Rieder, K. H.; Moresco, F.; Grill, L. Electric Field-Induced Isomerization of Azobenzene by STM. J. Am. Chem. Soc. 2006, 128 (45), 14446–14447. https://doi.org/10.1021/ja065449s. Darwish, N.; Aragonès, A. C.; Darwish, T.; Ciampi, S.; Díez-Pérez, I. MultiResponsive Photo- and Chemo-Electrical Single-Molecule Switches. Nano Lett. 2014, 14 (12), 7064–7070. https://doi.org/10.1021/nl5034599. Warshel, A.; Sharma, P. K.; Kato, M.; Xiang, Y.; Liu, H.; Olsson, M. H. M. Electrostatic Basis for Enzyme Catalysis. Chem. Rev. 2006, 106 (8), 3210– 3235. https://doi.org/10.1021/cr0503106. Warshel, A. Electrostatic Basis of Structure-Function Correlation in Proteins. Acc. Chem. Res. 1981, 14 (9), 284–290. https://doi.org/10.1021/ar00069a004. Fried, S. D.; Bagchi, S.; Boxer, S. G. Extreme Electric Fields Power Catalysis



(9) (10) (11) (12) (13) (14) (15)

(16) (17) (18) (19)

(20) (21)

(22) (23) (24)

in the Active Site of Ketosteroid Isomerase. Science (80-. ). 2014, 346 (6216), 1510 LP-1514. https://doi.org/10.1126/science.1259802. Fried, S. D.; Boxer, S. G. Electric Fields and Enzyme Catalysis. Annu. Rev. Biochem. 2017, 86 (1), 387–415. https://doi.org/10.1146/annurev-biochem061516-044432. Chattopadhyay, A.; Boxer, S. G. Vibrational Stark Effect Spectroscopy. J. Am. Chem. Soc. 1995, 117 (4), 1449–1450. https://doi.org/10.1021/ja00109a038. Andrews, S. S.; Boxer, S. G. Vibrational Stark Effects of Nitriles II. Physical Origins of Stark Effects from Experiment and Perturbation Models. J. Phys. Chem. A 2002, 106 (3), 469–477. https://doi.org/10.1021/jp011724f. Meir, R.; Chen, H.; Lai, W.; Shaik, S. Oriented Electric Fields Accelerate Diels–Alder Reactions and Control the Endo/Exo Selectivity. ChemPhysChem 2010, 11 (1), 301–310. https://doi.org/10.1002/cphc.200900848. Shaik, S. S. What Happens to Molecules as They React? A Valence Bond Approach to Reactivity. J. Am. Chem. Soc. 1981, 103 (13), 3692–3701. https://doi.org/10.1021/ja00403a014. Shaik, S.; Mandal, D.; Ramanan, R. Oriented Electric Fields as Future Smart Reagents in Chemistry. Nat. Chem. 2016, 8 (12), 1091–1098. https://doi.org/10.1038/nchem.2651. Tuñón, I.; Laage, D.; Hynes, J. T. Are There Dynamical Effects in Enzyme Catalysis? Some Thoughts Concerning the Enzymatic Chemical Step. Arch. Biochem. Biophys. 2015, 582, 42–55. https://doi.org/10.1016/j.abb.2015.06.004. Gertner, B. J.; Wilson, K. R.; Hynes, J. T. Nonequilibrium Solvation Effects on Reaction Rates for Model SN2 Reactions in Water. J. Chem. Phys. 1989, 90 (7), 3537–3558. https://doi.org/10.1063/1.455864. Van Der Zwan, G.; Hynes, J. T. Dynamical Polar Solvent Effects on Solution Reactions: A Simple Continuum Model. J. Chem. Phys. 1982, 76 (6), 2993– 3001. https://doi.org/10.1063/1.443392. Northrup, S. H.; Hynes, J. T. The Stable States Picture of Chemical Reactions. I. Formulation for Rate Constants and Initial Condition Effects. J. Chem. Phys. 1980, 73 (6), 2700–2714. https://doi.org/10.1063/1.440484. Ruiz-Pernía, J. J.; Tuñón, I.; Moliner, V.; Hynes, J. T.; Roca, M. Dynamic Effects on Reaction Rates in a Michael Addition Catalyzed by Chalcone Isomerase. Beyond the Frozen Environment Approach. J. Am. Chem. Soc. 2008, 130 (23), 7477–7488. https://doi.org/10.1021/ja801156y. Grote, R. F.; Hynes, J. T. The Stable States Picture of Chemical Reactions. II. Rate Constants for Condensed and Gas Phase Reaction Models. J. Chem. Phys. 1980, 73 (6), 2715–2732. https://doi.org/10.1063/1.440485. Zhang, J.; Zhang, Z.; Yang, Y. I.; Liu, S.; Yang, L.; Gao, Y. Q. Rich Dynamics Underlying Solution Reactions Revealed by Sampling and Data Mining of Reactive Trajectories. ACS Cent. Sci. 2017, 3 (5), 407–414. https://doi.org/10.1021/acscentsci.7b00037. Blake, J. F.; Jorgensen, W. L. Solvent Effects on a Diels-Alder Reaction from Computer Simulations. J. Am. Chem. Soc. 1991, 113 (19), 7430–7432. https://doi.org/10.1021/ja00019a055. Blake, J. F.; Lim, D.; Jorgensen, W. L. Enhanced Hydrogen Bonding of Water to Diels-Alder Transition States. Ab Initio Evidence. J. Org. Chem. 1994, 59 (4), 803–805. https://doi.org/10.1021/jo00083a021. Hummer, G. From Transition Paths to Transition States and Rate Coefficients.


Page 12 of 21

Page 13 of 21 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


(25) (26) (27)

(28) (29) (30) (31) (32) (33)

(34)

(35) (36)

J. Chem. Phys. 2004, 120 (2), 516–523. https://doi.org/10.1063/1.1630572. Best, R. B.; Hummer, G. Reaction Coordinates and Rates from Transition Paths. Proc. Natl. Acad. Sci. 2005, 102 (19), 6732–6737. https://doi.org/10.1073/pnas.0408098102. Lai, W.; Chen, H.; Cho, K. Bin; Shaik, S. External Electric Field Can Control the Catalytic Cycle of Cytochrome P450cam: A QM/MM Study. J. Phys. Chem. Lett. 2010, 1 (14), 2082–2087. https://doi.org/10.1021/jz100695n. Hirao, H.; Chen, H.; Carvajal, M. A.; Wang, Y.; Shaik, S. Effect of External Electric Fields on the C-H Bond Activation Reactivity of Nonheme Iron-Oxo Reagents. J. Am. Chem. Soc. 2008, 130 (11), 3319–3327. https://doi.org/10.1021/ja070903t. Ingrosso, F.; Rey, R.; Elsaesser, T.; Hynes, J. T. Ultrafast Energy Transfer from the Intramolecular Bending Vibration to Librations in Liquid Water. J. Phys. Chem. A 2009, 113 (24), 6657–6665. https://doi.org/10.1021/jp9022713. Rey, R.; Hynes, J. T. Tracking Energy Transfer from Excited to Accepting Modes: Application to Water Bend Vibrational Relaxation. Phys. Chem. Chem. Phys. 2012, 14 (18), 6332–6342. https://doi.org/10.1039/c2cp23555b. Laage, D.; Elsaesser, T.; Hynes, J. T. Water Dynamics in the Hydration Shells of Biomolecules. Chem. Rev. 2017, 117 (16), 10694–10725. https://doi.org/10.1021/acs.chemrev.6b00765. Duboué-Dijon, E.; Fogarty, A. C.; Hynes, J. T.; Laage, D. Dynamical Disorder in the DNA Hydration Shell. J. Am. Chem. Soc. 2016, 138 (24), 7610–7620. https://doi.org/10.1021/jacs.6b02715. Zoi, I.; Antoniou, D.; Schwartz, S. D. Electric Fields and Fast Protein Dynamics in Enzymes. J. Phys. Chem. Lett. 2017, 8 (24), 6165–6170. https://doi.org/10.1021/acs.jpclett.7b02989. Harijan, R. K.; Zoi, I.; Antoniou, D.; Schwartz, S. D.; Schramm, V. L. Inverse Enzyme Isotope Effects in Human Purine Nucleoside Phosphorylase with Heavy Asparagine Labels. Proc. Natl. Acad. Sci. 2018, 115 (27), E6209– E6216. https://doi.org/10.1073/pnas.1805416115. Hong, N. S.; Petrović, D.; Lee, R.; Gryn’ova, G.; Purg, M.; Saunders, J.; Bauer, P.; Carr, P. D.; Lin, C. Y.; Mabbitt, P. D.; et al. The Evolution of Multiple Active Site Configurations in a Designed Enzyme. Nat. Commun. 2018, 9 (1), 3900. https://doi.org/10.1038/s41467-018-06305-y. Fu, X.; Yang, L.; Gao, Y. Q. Selective Sampling of Transition Paths. J. Chem. Phys. 2007, 127 (15), 154106. https://doi.org/10.1063/1.2779325. Wang, F.; Zhang, C. In Spectral clustering for time series, International Conference on Pattern Recognition and Image Analysis, Springer:Berlin Heidelberg, 2005; pp 345-354.



Scheme 1 Scheme of forward and backward reactions: cis-2-vinylcyclopropanecarboxaldehyde(left) interconverts to 2,5-dihydrooxepin (right), with the bond forming/breaking sites indexed in red characters. 84x25mm (300 x 300 DPI)


Page 14 of 21

Page 15 of 21 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


Scheme 2 sketch map for electric field(E_2, E_1) and angle θ 84x111mm (300 x 300 DPI)



Figure 1 (a) Force Constant of C3-C8 and C12-O15 for a type A trajectory. K_1 (Black Line), K_2 (Blue Line). Bond Length of C3-C8: d_1 (Red Line). When force constant is less than zero, bonding or bond breaking has been finished. For the Type A trajectory, bond breaking and bonding is very close in time. (b) The same as (a) but for Type B trajectory. (c) Mulliken Charge for C1 (blue line), C12 (green line) and O15 (red line) for the Type A trajectory. (d) The same as (c) but for a Type B trajectory. Charge redistribution is retarded during 2000-2300fs. 177x83mm (300 x 300 DPI)


Page 16 of 21

Page 17 of 21 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 2 (a) Intramolecular Potential Energy averaging over all reactive trajectories (b) Interaction between solute and solvents (Electrostatic Energy plus van der waals Energy) averaging over all reactive trajectories (Blue Line), only Type A trajectories (Black Line) and only Type B trajectories (Red Line) (c) Work done on stretch motion of C12-O15 by solvents (Red Line) and solute (Black Line). Energy change due to vibrationrotation coupling (Blue Line) (d) O15 charge averaging over all reactive trajectories (Blue Line), only Type A trajectories (Black Line) and only Type B trajectories (Red Line) 177x93mm (300 x 300 DPI)



Figure 3 (a) The change of E_1 following reaction progression averaging over all reactive trajectories. (b) The change of E_2 following reaction progression averaging over all reactive trajectories, Type A trajectories and Type B trajectories. Time series of External Electric Fields (E_1, E_2) and key bond lengths (d_1, d_2) for (c) Type A trajectory (d) Type B trajectory (e) Unreactive trajectory. 177x142mm (300 x 300 DPI)


Page 18 of 21

Page 19 of 21 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 4 Decomposition of E_1 by water molecules (a) Type A trajectory (b) Type B trajectory (c) Unreactive trajectory. The conformation in TS for (d) Type A trajectory (e) Type B trajectory (f) Unreactive trajectory 177x127mm (300 x 300 DPI)



Figure 5 The decomposition of E_2 by water molecules for (a) Type A trajectory (b) Type B trajectory (c) Unreactive trajectory. H-bond length and θ of WAT2 for (d) Type A trajectory (e) Type B trajectory (f) Unreactive trajectory. H-bond length and θ of WAT3 for (g) Type A trajectory (h) Type B trajectory (i) Unreactive trajectory. 177x78mm (300 x 300 DPI)


Page 20 of 21

Page 21 of 21 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 S1 (a) Schematic illustration of the modified effective barrier under a weak solvent force. (b) Schematic illustration of the effective solvent well in the strong solvent force regime. 84x23mm (300 x 300 DPI)


Dynamic Electric Field Complicates Chemical ... - ACS Publications

Recommend Documents