Theoretical investigation of reaction

mechanisms – from methodological

development to applications

Summary of PhD dissertation

János Daru

Supervisors:

András StirlingInstitute of Organic Chemistry,

Research Centre for Natural Sciences

and

Gergely TóthDepartment of Physical Chemistry,

Eötvös Loránd University

Eötvös Loránd University

Doctoral School of Chemistry

Head of School: György Inzelt

Theoretical and Physical Chemistry,

Structural Chemistry PhD program

Head of Program: Péter Surján

2015

1 Introduction

Numerical simulations and theoretical modelling of chemical reac-tions have become an emerging field of chemistry. The atomisticlevel insight provided by theoretical chemistry is an important assetin the development of new, more efficient chemical procedures. Inparticular the theoretical exploration of reaction mechanisms canassist the deeper understanding of synthetic reactions. My thesissummarizes our results in this field.

The first chapter of the thesis is an introduction to the theoreticalframework of our studies. It summarizes the applied methods andtheories; and contrasts them with alternative approaches. Thesecond chapter focuses on the experimentally motivated researchon organic and organometallic reactions. We have explored threecomplex reaction mechanisms in cooperation with the group ofZoltán Novák and thoroughly investigated the H2 activation bythree linked Frustrated Lewis-Pairs. Finally the last part of thethesis is based on our methodological developments in the fields ofrate constant calculations and optimization of reaction coordinates.

3

2 Results

2.1 Palladium catalysed C–H activation [1]

We have studied the palladium catalysed C–H activation andcoupling reaction of acetanilide and benzaldehyde in the presence oftButyl hydroperoxide (TBHP).

Figure 1. oxidative coupling of acetanilide and benzaldehyde

The main conclusions from our mechanistic study are thefollowings:

• the rate determining step is the carbopalladation

• the catalytic effect of acids can be attributed to the stabi-lization of the active palladium mono-acetate complex by theprotonation of the dissociated acetate ligands

• the C–C coupling proceeds through a bimetallic palladiumcomplex

• the reductive elimination features a Pd(III)→ Pd(I) process

Our mechanistic picture is in good agreement with recent exper-imental evidences concerning the rate determining step [7] andis supported by the presence of bimetallic complexes observed atsimilar reaction conditions [8].

4

2.2 Silver-mediated furan formation [2]

Lei and co-workers have reported a selective synthesis of substitutedfurans by the oxidative coupling of alkynes and β-keto esters[9]. We have investigated the mechanism of this interestingtransformation. To this end we have examined the reaction betweenethyl acetoacetate and phenylacetylene.

Figure 2. Silver-mediated oxidative C-H/C-H functionalization

The study provided us with the following important results:

• the C–C coupling reaction features the radical intermediate ofethyl acetoacetate and silver phenylacetylide

• the mechanism of the cyclisation reaction is ionic for terminalalkynes

• internal alkynes require higher activation for the C–C couplingreaction

• the cyclisation reaction of intermediates from internal alkynesproceeds through a radical pathway

• silver ions have double role in the reaction: radical generationand the catalysis of the ring formation

• the silver-furanyl organometallic intermediate decomposes be-fore the acidic work-up of the reaction

Our mechanism is supported by vibrational spectroscopic observationof silver phenylacetylide and isotope labelling experiments. Thepresence of radical intermediates has been verified by experimentswith scavengers.

5

2.3 Trifluoroethylation of indoles via C–H

functionalization [3]

Our experimental partners have developed an efficient and selectivesynthetic procedure for C3 trifluoroethylation of indoles. Thereaction required the presence of 2,6-di-tert-butylpyridine (DTBPy)in order to achieve maximum yield.

Figure 3. Trifluoroethylation of indoles

Our mechanistic picture of the reaction can be summarizedas follows:

• the C–C bond formation reaction is a direct electrophilic attackby the iodonium complex

• the role of the DTBPy is the deprotonation of the σ complexfrom the electrophilic attack

• in the absence of DTBPy, dimerization side reaction takesplace

• bases with modest steric congestion undergoN-trifuoroethylation under the reaction conditions

Our scheme of competing N- and C-alkylations is able to predict thethe efficiency of a given substrate-base combination by computingactivation barriers for these two elementary steps.

6

2.5 Divided Saddle Theory – new method for cal-

culating rate constant [5]

We have derived a new theory and an efficient method to calculaterate constants from molecular dynamics simulations. Our methoddevelopment can be summarized as follows:

• the rate constant is factorised to a dynamical kSD and a staticquantity αSD

RS

• αSD

RScan be calculated from the free energy profile along a

suitable reaction coordinate

• kSD can be calculated from short dynamical trajectories initi-ated at the saddle region

• Bennett-Chandler-like methods [10] can be analytically de-rived from our method

• the method has been validated through numerical simulations

• in terms of the number of trajectories our method proved tobe more efficient then the effective positive flux method

Figure 5. Divided sadle chair – ilustration for the divided states in our

theory

8

3 Papers Forming the Basis of the

Dissertation

[1] F. Szabó, J. Daru, D. Simkó, T. Zs. Nagy, A. Stirling, Z.Novák, “Mild Palladium-Catalyzed Oxidative Direct ortho-C–HAcylation of Anilides under Aqueous Conditions”, Adv. Synth.

Catal. 2013, 355, 685.

[2] J. Daru, Zs. Benda, Á. Póti, Z. Novák, A. Stirling, “Mecha-nistic Study of Silver-Mediated Furan Formation by OxidativeCoupling”, Chem. Eur. J. 2014, 20, 15395.

[3] G. L. Tolnai, A. Székely, Z. Makó, T. Gáti, J. Daru, T. Bihari,A. Stirling, Z. Novák, “Efficient Direct 2,2,2-Trifluoroethylationof Indoles via C-H Functionalization”, Chem. Commun. 2015

DOI: 10.1039/c5cc00519a

[4] J. Daru, I. Bakó, A. Stirling, I. Pápai, “AIMD modelling offrustrated Lewis pairs”, manuscript in preparation

[5] J. Daru, A. Stirling, “Divided Saddle Theory: A New Ideafor Rate Constant Calculation”, J. Chem. Theory Comput.

2014, 10, 1121.

[6] J. Daru, G. Tóth, “Committor map collective variable”,manuscript in preparation

4 Literature References

[7] W. Y. Chan, Z. Zhou, W. Yu, Adv. Synth. Catal. 2011, 353,2999.

[8] X. Zhao, C. S. Yeung, V. M. Dong, J. Am. Chem. Soc. 2010,132, 5837.

[9] C. He, S. Guo, J. Ke, J. Hao, H. Xu, H. Chen, A. Lei, J.

Am. Chem. Soc. 2012, 134, 5766.

[10] P. G. Bolhuis and C. Dellago, in Rev. Comp. Chem., edited byK. B. Lipkowitz (Wiley, New York, 2010), Vol. 27, pp. 111–202

10