The regioselectivity in the gold-catalyzed cycloisomerization of alcohols: A full DFT study on carbene versus non-carbene mechanisms

doi:10.21203/rs.3.rs-5279805/v1

The mechanistic study of in the gold-catalyzed cycloisomerization of alcohols or amine tethered-vinylidenecyclopropanes has been performed using high-level DFT calculations to find the reasons of the observed selectivity of the products. Two possible pathways were observed, which in the first product, during the cyclization process, the vinylidenecyclopropane moiety converts to methylidenecyclopropane-containing product (P1) via an ionic (non-carbene) intermediate and in the second product, converts to cyclobutene-containing product (P2) via a carbene intermediates. In all six examined derivatives (containing different alkyl group at C1 allenic carbon) P2 was the major product by both thermodynamic (the more stability of P1 than P2) and kinetic (the smaller barrier for producing P2 versus P1) criteria in both gas phase and solvent (THF, using PCM model) media. However, in the derivatives with low-strain allenic group (R), I1 (which leads to P1) is more stable than I3 intermediates (which leads to P2) by 0.6–4.2 kcal/mol in the gas phase and 1.0-3.1 kcal/mol in the solvent and the derivatives with high-strain allenic group, I3 is more stable than I1 by 2.1–2.5 kcal/mol in the gas phase and 3.5–4.2 kcal/mol in the solvent. Therefore, both gas phase and solvent data show that because of the different hinderance between alkyl group and gold cation complex, the selectivity of the reaction is yielded by the favorability of I1 or I3 intermediates. Interestingly, another evidence was provided by atomic charges in the reactants using NBO calculations. These calculations showed that when the atomic charge of C1 allenic carbon was higher than C3, the gold cation prefers to produce I1, leading to P1, and when the atomic charge of C3 allenic carbon was higher than C1, the gold cation prefers to produce I2, leading to P2.

Carbene

Mechanism

DFT

Gold

Isomerization

The mechanistic details of transition metal-catalyzed reactions have been followed by chemists since the first reactions in this field were reported [1–3]. The understanding reaction mechanism helps chemists to make better choices in their future studies and have smart design on the synthesis of new compounds. Despite the extensive developments of such reactions, the experimental methods cannot exactly express the details of these reactions; and instead, theoretical studies have opened new horizons in these areas, especially after the great development of the computational facilities. In this regard, a large number of mechanistic problems have been solved by using various computational methods [4–6].

During the previous study [7], when we were examining the Gold(I)-catalyzed cycloisomerization of alcohol or amine, starting from vinylidenecyclopropanes and producing various morpholine, piperazine or oxazepane derivatives, the chemoselectivity was observed by using different starting materials. Moreover, the mechanism of this reaction is unknown for us, while two possible ways, carbene or non-carbene, are fundamentally possible for this reaction. The reaction is catalyzed by a gold-based complex, and all of these factors made us to make a deeper insight into the mechanistic details of this reaction to provide the appropriate evidences to answer the questions about the reaction mechanism and observed chemoselectivity.

Gold, as a valuable metal, has been attracted many interests related to the catalyzing various reactions [8–12]. Despite its popularity as a noble metal, it showed high abilities in the catalyzing various organic reactions such as synthesis of heterocycles [13], oxidation processes [14], cyclization reactions [15], coupling reactions [16], and electrophilic halogenation [17]. In this line, several studies have been conducted on the mechanistic details of gold-catalyzed reactions [18–21]. However, the possibility of carbene-based, gold intermediate is an interesting point of this study, in addition to the observed chemoselectivity during the isomerization reaction. This type of reaction is valuable for chemists because of its high atom-economy. Via the isomerization reactions, various chemical transformations such as alkane branching [22], imine-enamine conversion [23], olefine transformations [24], and cycloisomerization [25] (similar to the subject of this study) can be achieved.

Another important aspect of the organic reactions is selectivity, including stereoselectivity [27], regioselectivity [28], and chemoselectivity. Since many years ago, this topic has retained its importance because it has important effect on the costs of the chemical processes and the price of the product. Chemists have been always curious about the reasons of the observed selectivities and various experimental and theoretical scientists have been focused on the details of these observations, mostly by studying the reaction mechanisms. During this way, several selective reactions including reductions [29,30], oxidations [31,32], asymmetric process [33], and Diels-Alder reaction [34] have been studied.

Therefore, based on our previous efforts on the searching of the mechanistic details of the chemical transformations [35–38], it seems interesting to take a deep insight into the mechanistic details of the title gold-catalyzed isomerization process, which showed remarkable selectivity, in addition to its notable mechanism, including carbene intermediates. In this line, the energy details of the possible routes and the structures and energies of the major intermediates for different derivatives could be studied, as they performed and presented in the next sections.

The structures of all involved species, including reactants, products, intermediates and transition states, were drawn, processed and analyzed using GaussView 6.1 programs [39]. Then, the optimization of all structures, frequency and NBO calculations and considering solvent effects were performed using Gaussian 09 program package [40]. For all calculations, DFT method was considered because of its high accuracy versus the computation time. In this line, M06-2X functional and def2-TZVP basis set were employed for all calculations since they were proved to be among the highest accuracies functional and basis sets for the study of the reaction mechanisms [41]. The M06-2X functional is a highly parameterized, range-separated hybrid functional designed for a broad spectrum of applications, including main group and transition metal chemistry, as well as noncovalent interactions [42]. The def2-TZVP basis set provides a comprehensive description of the valence electrons, which is crucial for accurate electrochemical predictions [43]. All structures were optimized without any symmetry restriction and the zero imaginary frequency was checked to confirm the structures were true minima. The structures of transition states were obtained from QST3 calculations and confirmed by frequency calculations to have only one imaginary frequency, exactly in the direction of the reaction coordinate. The solvent effects were considered using Tomasi’s continuum model (PCM) employing SCRF keyword. Only THF has been considered as a solvent since it was the solvent of the performed reactions in the experimental work [7].

3.1 The studied mechanisms of the reaction

Based on the results of the experimental work, a reasonable mechanism was proposed for the reaction. Since, two products were obtained in the reaction, two different routes were designed, which each one led to the defined product. Moreover, for faster preformation of the calculations, Me₃P ligand was considered for gold cation complex, instead of the employed ligand in the experimental work. The studied mechanism is completely shown in Scheme 1 with the details and the list of the obtained derivatives in the experimental work with the product ratios was listed in Table 1.

According to this mechanism and via route 1, the allene moiety coordinates to the gold complex via its external C=C double bond to form I1 (intermediate 1). Then, the hydroxy group attacks to the end carbon of the allene, and the π bond transfers between the central atom of allene and gold cation to make I2. In I2, we have ionic-type bonding between the organic structure and gold cation, which is the key difference between these two mechanisms. At the final step of route 1, the proton transfer from the oxonium ion to the adjacent alkene and replaced with the gold cation complex to form the first product (P1) and free gold complex (catalyst).

In route 2, allene moiety coordinates to the gold cation via its central C=C double bond to form I3 (the first difference between route 1 and route 2). Then, an intermolecular rearrangement in the vinylidenecyclopropane moiety is observed and led to I4, which consists of a carbenoid between 4-membered ring and gold cation complex. The nucleophilic attack of hydroxy group to the ending carbon of the remained C=C double bond, followed by the migration of the double bong into the 4-membered ring and conversion of the carbenoid intermediate to ionic intermediate I5. The last step of route 2 is similar to the previous route, which includes the proton transfer from the oxonium ion to the adjacent alkene (in 4-membered ring) and replaced with the gold cation complex to form the second product (P2) and free gold complex (catalyst). Based on these proposals, these mechanisms were studied for the reaction to be confirmed by DFT calculations and more importantly, to investigate about the reasons of the observed selectivity in the experimental work.

3.2 The energy profiles of the mechanistic routes in the Gas and solvent

The mechanism of the reaction for derivative a (the simplest and first case) will be discussed here, as it drawn in Scheme 1, by considering R=Me, which is corresponded to derivative a in the experimental work. All of the structures, including two reactants, two products, 5 intermediates and 5 transition states, were optimized first, which their optimized structures were shown in Figure 1. The optimized structures were obtained according to their expectable situation, as shown in Scheme 1 and all of structures were confirmed by frequency calculations. According to these structures, gold cation is connected to the involved intermediates to effectively catalyze the reaction process. Considering route 1 and in I1, the gold was placed in the top-mid of the external C=C double bond that show the coordination of the π-bond into the empty orbital of gold cation. In I2, there is ionic nature between the carbon and gold cation and gold is connected to only one carbon. In TS1-2 (the transition state between I1 and I2) gold cation will be far away from one carbon and gets closer to another carbon and simultaneously, oxygen atom gets closer to the mentioned carbon atom. In TS2-P1, the hydrogen atom was placed between the oxygen atom and olefinic carbon and gold cation gets further away from that carbon to yield the product. In route 2, the similar configuration with I1 is observed in I3 between gold cation and the internal C=C double bond. TS3-4 is the most complex transition state and the major difference between route 1 and route 2, which consists of ring expansion from 3-membered to 4-membered and forming carbene complex. The remained structures of route 2 are similar to those one in route 1 and no surprise can be observed in their structures. Moreover, since all transition states should have one imaginary frequency (to confirm it is a real saddle point), the value of the imaginary frequency of each transition state was pasted below its image in parentheses. For TS2-P1 and TS5-P2, which include hydrogen transfer, the values of imaginary frequencies (1272 and 1398 cm^-1, respectively) are naturally very larger than the other transition states (between 221 and 268 cm^-1), which includes easier transformations.

After optimization, the total electronic energies, enthalpies, and Gibbs free energies (all in kcal/mol) of all structures were extracted from the outputs of the frequency calculations (after correction by applying scaling factor) and used to obtain the energy profile of the studied reaction (R=Me). All of these energy values were listed in Table 2, which all energies are relative to the energy of the reactants. In addition, the energy profiles for both routes of this reaction were depicted in Figure 2. By considering total electronic energies (E values), in route 1, I1 is 39.1 kcal/mol stable than the reactants, because of the strong interactions between gold cation and organic structure. This highly negative energy is appropriate driving force for the reaction. Then, then I1 can undergo a cyclization to generate an intermediate I2 via transition state TS2 with 10.2 kcal/mol barrier energy in an endothermic process (ΔE = 8.0 kcal/mol). The intermediate I2 is less stable than I1, but still has sufficient negative energy (-31.1 kcal/mol) versus the reactants. At the last step of this route, the intermediate I2 converts to P1 via TS2-P1 with 13.3 kcal/mol barrier energy. This transformation suffers +7.6 kcal/mol energy to the reaction and at the final point, P1 is 23.5 kcal/mol more stable than the reactants. Therefore, this route is exothermic by -23.5 kcal/mol and its highest barrier is -17.8 kcal/mol stable than the reactants and 21.3 kcal/mol less stable than I1, which is the most stable structure of this route. In route 2, I3 is 38.5 kcal/mol more stable than reactants, but 0.6 kcal/mol less stable than I1, its parallel intermediate in route 1. Then, it converts to I4 with 8.6 kcal/mol barrier and ΔE=+2.9 kcal/mol. Transformations of I4 to I5 and I5 to P2 consist of respectively 7.9 kcal/mol and 8.1 kcal/mol barriers with ΔE= 0.0 kcal/mol for the first one and +6.9 kcal/mol for the second one. Resulting, the reaction is exothermic by -28.7 kcal/mol (the relative energy of P2 versus the reactants). By comparing these two routes, it will be concluded that route 2 is more favorable than route 1 since its product (P2) is 5.2 kcal/mol more stable than the product of route 1 (P1) and its highest barrier (belong to TS5-P2 with relative energy=-27.5 kcal/mol) is less than the highest barrier of route 1 (TS2-P1 with -17.8 kcal/mol). This thermodynamic favorability of P2 versus P1 is completely expectable, since based on the reports [44], it consists cyclobutene ring with 31-34 kcal/mol strain, while P1 consisted of methylidenecyclopropane ring with 41 kcal/mol strain (P1 has a higher strain energy than P2). Noticeably, the difference between the strain energies of these two cycles (7-10 kcal/mol) is comparable with the obtained energy differences between two products (5.0-9.4 kcal/mol, next section). However, in the experimental work, P1 is the major product, so thermodynamic preferability of the products is not the reason for this selectivity. Based on the reported energies, it seems that the more stability of I1 versus I3 push the reaction toward producing P1, instead of P2. To examine the correctness of this hypothesis, the same calculations have been performed with five more derivatives and the results will be discussed in the next section.

In addition to the calculations in gas phase, the reaction mechanisms were also investigated in the solvent (THF, the same as the solvent of the experimental work) using PCM model. The relative energy values of all involved species in the solvent were compared with the gas phase values in Table 3. According to these values, the solvent only increased the relative energy values versus the gas phase for intermediates and transition states, which means the reaction performs slower in the solvent. However, both products showed less energy values in the solvent versus the gas phase that show the enhancement in the thermodynamic favorability of the reaction in the solvent. Moreover, like the results in gas phase, route 2 is more favorable than route 1 by considering both thermodynamic (the stability of the product) and kinetic (the overall barrier energy) criteria.

3.3 The investigation about the chemoselectivity of the reactions in the gas phase and solvent

According to all the energy data, the product P2 (from route 2, carbene route) was more stable than P1 (from route 1, non-carbene route) and its overall barrier was also more stable than the overall barrier of P1 in both solvent (THF) and the gas phase. It seems that the reaction is not under common thermodynamic or kinetic control and we made a deeper insight to search about the reason of this selectivity. However, it was decided to calculate the energies of both products in the other derivatives. Therefore, five more derivatives were selected based on the experimental work, in addition to derivative a (R=Me, based on Scheme 1). These derivatives, which were defined with the bold font in Table 1, are a, h, p, q, r, and w. three first derivatives (a, h, and p) gave P1 as the major product and the last three derivatives (q, r, and w) ) gave P2 as the major product. The structures of these derivatives (as reactant, R1) and both possible products (P1 and P2) for each derivative were optimized to examine the thermodynamic favorability of the products. Moreover, since in derivative a, it was shown that the more stability of I1 versus I3 was maybe the reason of the observed product selectivity (pushing the reaction toward producing P1), the structures of I1 and I3 for these derivatives were optimized. The optimized structures of the mentioned intermediates and possible products of these derivatives were shown in Figure 3. From these optimizations, the energies of the optimized structures were extracted and the relative energies (versus the related reactant) were listed and compared in Table 4. In this table, the relative energy (in kcal/mol) of I1, I3, P1 and P2 for all six examined derivatives were listed and at the final two columns, the energy differences between P1 and P2 (ΔE(P2-P1)) and between I1 and I3 (ΔE(I3-I1)) were calculated to make better comparison. According to the calculated values, in all cases, P2 is more stable than P1 by 5.0 – 9.4 kcal/mol. The reason of this stability was discussed in the previous section (less ring strain of cyclobutene ring versus methylidenecyclopropane ring) and the values are comparable with small differences. Therefore, surely the more stability of the product is not the reason of the observed selectivity. Moreover, it was proved in the previous section that the reaction cannot be under the kinetic control because the overall barrier of the P2 was also less than that in P1. Therefore, the energies of I1 (which led to P1) and I3 (which led to P2) intermediates of all derivatives were compared at the last column. Interestingly, it was observed that in the three first derivatives (which produced P1 as the major product), I1 is more stable than I3 by 0.6-4.2 kcal/mol and in three last derivatives (which produced P2 as the major product), I3 is more stable than I1 by 2.1-2.5 kcal/mol. It could be deduced that in the derivatives with low-strain allenic group (R), the gold complex is chelated from C1=C2 allenic bond, which lead to I1 and finally, to P1. In contrast, in the derivatives with high-strain allenic group, the gold complex is chelated from C2=C3 allenic bond to increase its distance with the hindered atom C1, which lead to I2 and finally, to P2. This argument is in accordance with the obtained experimental data and completely justify the observed selectivity.

To examine this reason in the solvent (THF), like previous sections, all calculations were repeated in THF using PCM models and the results (with similar pattern to the gas phase data, Table 4) were shown in Table 5. Noticeably, the solvent results are similar to the gas phase results. This means that in all derivatives, P2 is more stable than P1 by 4.6 – 9.1 kcal/mol. Moreover, in the three first derivatives (which produced P1 as the major product), I1 is more stable than I3 by 1.0-3.1 kcal/mol and in three last derivatives (which produced P2 as the major product), I3 is more stable than I1 by 3.5-4.2 kcal/mol. In fact, in addition to the accordance with the gas phase values, the energy differences between I1 and I3 derivatives in the solvent are more distinct (with higher values) than those in the gas phase. Therefore, the employed models can deduce the reason of the observed selectivity in the products.

At the last step of this study, to provide another reason for the observed selectivity, NBO calculations were performed for the selected derivatives. From the outputs of these calculations, NBO atomic charges for three allenic carbons were extracted and listed in Table 6, in addition to the image of the numbering scheme. Based on the atomic charges of C1 and C3 allenic carbons, the C1 negative charge in three first derivatives (a, h, and p; -0.97 to -0.101) is larger than those in the three last derivatives (q, r, and w; -0.081 to -0.091), which means that the gold cation has higher tendency to be closer to C1, instead of C3, which led to the intermediate I1 and product P1. In contrast, C3 negative charge in three last derivatives (-0.111 to -0.113) is larger than those in the three first derivatives (-0.105 to -0.108), which means that the gold cation has higher tendency to be closer to C3, instead of C1, which led to the intermediate I3 and product P2. This is an interesting aspect that considering a very simple criteria such as atomic charge can be useful in the anticipation of the reaction selectivity.

In conclusion, by reviewing the energy data and in all derivatives (a, h, p, q, r, and w), P2 was the major product, thermodynamically (because of the more stability of P1 than P2) and kinetically (because of the smaller barrier for producing P2 versus P1) in both gas phase and THF (as a solvent). More insight into the energies of the intermediates showed that in the derivatives with low-strain allenic group (R = a, h, and p), I1 was more stable than I3 intermediates by 0.6–4.2 kcal/mol in the gas phase and 1.0-3.1 kcal/mol in the solvent and the derivatives with high-strain allenic group (R = q, r, and w), I3 is more stable than I1 by 2.1–2.5 kcal/mol in the gas phase and 3.5–4.2 kcal/mol in the solvent. Since I1 converts to P1 and I3 converts to P2, both gas phase and solvent data show that the selectivity of the reaction is yielded by the favorability of the intermediates I1 or I3. Observing atomic charges in the reactants using NBO calculations provide another evidence for this selectivity. The NBO atomic charges showed that the higher negative charges of C1 leads to producing I1 and P1 and the higher negative charges of C3 leads to producing I3 and P2.

Author Contribution

H.T. performed calculation, data extraction and paper writingY.W. corrected the manuscript and analyzed the dataM.S. supervised the work and reviewed the manuscript

M. Salciccioli, M. Stamatakis, S. Caratzoulas, D.G. Vlachos, Chem. Eng. Sci., 66, 4319 (2011)
Y. Park, Y. Kim, S. Chang. Chem. Rev. 117, 9247 (2017)
R. Waterman, Chem. Soc. Rev., 42, 5629 (2013)
S. Liao, X. Cheng, L. Zhu, J. Xu, J. Jiang, S. Wang, J. Anal. Appl. Pyrol. 176, 106245 (2023)
J. Niu, Y. Wang, Y. Qi, A.H. Dam, H. Wang, Y.A. Zhu, D. Chen, Fuel, 266, 117143 (2020)
L. Fomina, B. Vazquez, E. Tkatchouk, S. Fomine, A DFT study. Tetrahedron, 58, 6741 (2002).
J.S. Wei, S. Yang, Y. Wei, S. Shamsaddinimotlagh, H. Tavakol, M. Shi, Org. Chem. Front. 10, 1738 (2023)
A.S.K. Hashmi, Chem. Rev., 107, 3180 (2007)
M. Alfonsi, A. Arcadi, M. Aschi, G. Bianchi, F. Marinelli, J. Org. Chem., 70, 2265 (2005)
D. Campeau, D.F. León Rayo, A. Mansour, K. Muratov, F. Gagosz, Chem. Rev., 121, 8756 (2020)
Z. Li, C. Brouwer, C. He, Chem. Rev., 108, 3239 (2008)
A. Corma, A. Leyva-Pérez, M.J. Sabater, Chem. Rev., 111, 1657 (2011)
N. Mehiaoui, Z. Kibou, A. Berrichi, R. Bachir, N. Choukchou-Braham, Res. Chem. Intermed., 46, 5263 (2020)
K.J. Kim, H.S. Seo, H.G. Ahn, Res. Chem. Intermed., 37, 1165 (2011)
B. Alcaide, P. Almendros, Acc. Chem. Res., 47, 939 (2014)
G. Zhang, Y. Peng, L. Cui, L. Zhang, Angew. Chem., 121, 3158 (2009)
A.S.K. Hashmi, T.D. Ramamurthi, M.H. Todd, A.S.K. Tsang, K. Graf, Austr. J. Chem., 63, 1619 (2010)
E. Jimenez-Nunez, A.M. Echavarren, Chem. Rev., 108, 3326 (2008)
L.P. Liu, B. Xu, M.S. Mashuta, G.B. Hammond, J. Am. Chem. Soc., 130, 17642 (2008)
B.N. Nguyen, L.A. Adrio, E.M. Barreiro, J.B. Brazier, P. Haycock, K.K. Hii, J. Szlachetko, Organometallics, 31, 2395 (2012)
A. Nijamudheen, A. Datta, Chem. Eur. J., 26, 1442 (2020)
H. Song, N. Wang, H. Song, F. Li, Res. Chem. Intermed., 42, 951 (2016)
M. Petrisko, J. Krupka, Res. Chem. Intermed., 31, 769 (2005)
H. Wakamatsu, M. Nishida, N. Adachi, M. Mori, The J. Org. Chem., 65, 3966 (2000)
M. Marín-Luna, O. Nieto Faza, C. Silva López, Front. Chem., 7, 296 (2019)
U.M. Lindström, Chem. Rev., 102, 2751 (2002)
M. Stratakis, M. Orfanopoulos, Tetrahedron, 56, 1595 (2000)
C.L. Raston, J.L. Scott, Green Chem., 2, 49 (2000)
R.Q. Long, R.T. Yang, J. Cat., 207, 224 (2002)
K.I. Shimizu, J. Shibata, H. Yoshida, A. Satsuma, T. Hattori, Appl. Cat. B: Environ., 30, 151 (2001)
T.V. Andrushkevich, V.V. Kaichev, Y.A. Chesalov, A.A. Saraev, V.I. Buktiyarov, Cat. Today, 279, 95 (2017)
A.E. de Nooy, A.C. Besemer, H. van Bekkum, Tetrahedron, 51, 8023 (1995)
J.M. Brown, Chem. Soc. Rev., 22, 25 (1993)
H. Wang, Y. Wang, K.L. Han, X.J. Peng, J. Org. Chem., 70, 4910 (2005)
H. Tavakol, Mol. Simul., 36, 391 (2010)
H. Tavakol, M.A. Ranjbari, A. Mahmoudi, Comp. Theor. Chem., 1154, 37 (2019)
M.T. Jafari-Chermahini, H. Tavakol, Res. Chem. Intermed., 49, 4065 (2023)
H. Tavakol, J. Mol. Struct., 956, 97 (2010)
GaussView, Version 6.1, Roy Dennington, A. Todd Keith, and M. John Millam, Semichem Inc. Shawnee Mission, KS, 2016
Gaussian 09, Revision A.02, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, G.A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. Marenich, J. Bloino, B.G. Janesko, R. Gomperts, B. Mennucci, H.P. Hratchian, J.V. Ortiz, A.F. Izmaylov, J.L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V.G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J.A. Montgomery, Jr.J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, J.M. Millam, M. Klene, C. Adamo, R. Cammi, J.W. Ochterski, R.L. Martin, K. Morokuma, O. Farkas, J.B. Foresman, and D.J. Fox, Gaussian, Inc. Wallingford CT, 2016
M. Bursch, J.M. Mewes, A. Hansen, S. Grimme, Angew. Chem. Int. Ed., 61, e202205735 (2022)
Y. Wang, P. Verma, L. Zhang, Y. Li, Z. Liu, D.G. Truhlar, X. He, Proc. Natl. Acad. Sci. U.S.A, 117, 2294 (2020)
U. Ndagi, M.M. Lawal, M.E. Soliman, Russ. J. Phys. Chem. A, 93, 1543 (2019)
E. V. Anslyn, D. A. Dougherty, Modern physical Organic Chemistry, University science books, USA. pp 110-112 (2006)

Table 1. The list of the obtained derivatives in the experimental work with the product ratios, according to scheme 1^a

Der.	R	X	Y	P1/P2	Der.	R	X	Y	P1/P2
a	Me	NTs	O	25	q	i-Pr	NTs	O	0.10
b	Et	NTs	O	10	r	sec-Bu	NTs	O	0.50
c	Pr	NTs	O	6.3	s	cyclopentyl	NTs	O	0.20
d	Hexyl	NTs	O	5	t	cyclohexyl	NTs	O	0.12
e	iPr(OCH₂CH₂O)	NTs	O	5.9	u	cycloheptyl	NTs	O	0.09
f	CH2-Allyl	NTs	O	11.1	v	cyclobutyl	NTs	NBn	0.01
g	Bn	NTs	O	3.2	w	Me	CH₂NTs	O	0.01
h	Cyclopropyl	NTs	O	100	x	i-Pr	CH₂NTs	O	0.01
i	Me	NSO₂Ph	O	27	y	cyclopentyl	CH₂NTs	O	0.01
j	Me	NBs	O	33.3
k	Me	O	O	10
l	Me	CH2	O	100
m	Me	NTs	O(Me_α)	100
n	Me	NTs	NBn	100
o	CH₂-Allyl	NTs	NBn	100
p	Cyclobutyl	NTs	O	1.6

^aThe bold cases are the selected ones in the next steps. R is exactly as defined in Scheme 1, X is the possible other substituents instead of NTs, and Y is the possible substituents instead of ending oxygen atom

Table 2. The relative energies, enthalpies, and Gibbs free energies of all involved structures in both routes of the reaction mechanism of derivative a (R=Me) in the gas phase. All energy values are reported in kcal/mol

Route 1	Reactants	I1	TS1-2	I2	TS2-P1	P1
E (HF+ZPE)	0.0	-39.1	-28.9	-31.1	-17.8	-23.5
H (Enthalpy)	0.0	-38.3	-29.3	-31.8	-18.5	-24.7
G (Gibbs FE)	0.0	-27.9	-15.6	-17.4	-4.2	-22.6
Route 2	Reactants	I3	TS3-4	I4	TS4-5	I5	TS5-P2	P2
E (HF+ZPE)	0.0	-38.5	-29.9	-35.6	-27.7	-35.6	-27.5	-28.7
H (Enthalpy)	0.0	-37.7	-30.2	-35.7	-29.5	-36.8	-28.1	-30.1
G (Gibbs FE)	0.0	-27.4	-16.7	-22.9	-11.7	-20.7	-14.7	-26.8

Table 3. The relative energies, enthalpies, and Gibbs free energies of all involved structures in both routes of the reaction mechanism of derivative a (R=Me) in the solvent. All energy values are reported in kcal/mol

Rel E	Reactants	I1	TS1-2	I2	TS2-P1	P1
in gas	0.0	-39.1	-28.9	-31.1	-17.8	-23.5
in THF	0.0	-29.8	-14.8	-17.4	2.4	-30.6
Rel E	Reactants	I3	TS3-4	I4	TS4-5	I5	TS5-P2	P2
in gas	0.0	-38.5	-29.9	-35.6	-27.7	-35.6	-27.5	-28.7
in THF	0.0	-28.7	-7.6	-23.7	-13.2	-21.4	-15.3	-35.4

Table 4. The relative energies and their differences for the products and important intermediates in all selected derivatives in the gas phase. All energy values are reported in kcal/mol

Deriv./Rel. E	P1	P2	I1	I3	ΔE(P2-P1)	ΔE(I3-I1)
a	-23.5	-28.7	-39.1	-38.5	-5.2	0.6
h	-19.9	-24.9	-40.7	-36.5	-5.0	4.2
p	-15.9	-25.3	-39.9	-38.0	-9.4	1.9
q	-20.0	-27.3	-35.6	-38.1	-7.3	-2.5
r	-17.5	-24.4	-35.5	-37.6	-6.9	-2.1
w	-19.4	-25.7	-37.9	-40.1	-6.4	-2.2

Table 5. The relative energies and their differences for the products and important intermediates in all selected derivatives in the solvent (THF). All energy values are reported in kcal/mol

Deriv./Rel. E	P1	P2	I1	I3	ΔE(P2-P1)	ΔE(I3-I1)
a	-30.6	-35.4	-29.8	-28.7	-4.8	1.0
h	-18.8	-23.4	-29.6	-26.5	-4.6	3.1
p	-14.4	-23.5	-29.0	-27.4	-9.1	1.6
q	-18.5	-25.4	-24.1	-27.6	-6.9	-3.5
r	-16.0	-22.5	-23.9	-27.9	-6.5	-4.0
w	-18.1	-24.2	-25.5	-29.8	-6.1	-4.2

No competing interests reported.

The regioselectivity in the gold-catalyzed cycloisomerization of alcohols: A full DFT study on carbene versus non-carbene mechanisms

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods

3. Results and discussions

4. Conclusions

Declarations

Author Contribution

References

Tables

Additional Declarations

Status:

Version 1