Cobalt-catalyzed C–H cyanations: Insights into the reaction mechanism and the role of London dispersion

Article abstract of DOI:10.3762/bjoc.14.130

Carboxylate-assisted cobalt(III)-catalyzed C–H cyanations are highly efficient processes for the synthesis of (hetero)aromatic nitriles. We have now analyzed the cyanation of differently substituted 2-phenylpyridines in detail computationally by density f

Full text of DOI:10.3762/bjoc.14.130

Cobalt-catalyzed C–H cyanations: Insights into the reaction
mechanism and the role of London dispersion
Eric Detmar1, Valentin Müller2, Daniel Zell2, Lutz Ackermann*2 and Martin Breugst*1
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2018, 14, 1537–1545.
1Department für Chemie, Universität zu Köln, Greinstraße 4, 50939
Köln, Germany and 2Institut für Organische und Biomolekulare
Chemie, Georg-August-Universität Göttingen, Tammannstraße 2,
37077 Göttingen, Germany
doi:10.3762/bjoc.14.130
Received: 27 February 2018
Accepted: 02 June 2018
Published: 25 June 2018
Email:
Lutz Ackermann* - Lutz.Ackermann@chemie.uni-goettingen.de;
Martin Breugst* - mbreugst@uni-koeln.de.
This article is part of the Thematic Series "Dispersion interactions".
Guest Editor: P. Schreiner
* Corresponding author
© 2018 Detmar et al.; licensee Beilstein-Institut.
License and terms: see end of document.
Keywords:
catalysis; C–H activation; density functional theory; London
dispersion; reaction mechanisms
Abstract
Carboxylate-assisted cobalt(III)-catalyzed C–H cyanations are highly efficient processes for the synthesis of (hetero)aromatic
nitriles. We have now analyzed the cyanation of differently substituted 2-phenylpyridines in detail computationally by density func-
tional theory and also experimentally. Based on our investigations, we propose a plausible reaction mechanism for this transformat-
ion that is in line with the experimental observations. Additional calculations, including NCIPLOT, dispersion interaction densities,
and local energy decomposition analysis, for the model cyanation of 2-phenylpyridine furthermore highlight that London disper-
sion is an important factor that enables this challenging C–H transformation. Nonbonding interactions between the Cp* ligand and
aromatic and C–H-rich fragments of other ligands at the cobalt center significantly contribute to a stabilization of cobalt intermedi-
ates and transition states.
Introduction
For a long time, large and bulky substituents have intuitively tance for organic transformations has only been fully realized
been considered to act through unfavorable steric interactions, within the last decades [3]. Among others, these interactions
although London dispersion – the attractive part of the van-der- explain the hexaarylethane riddle [4] and are responsible for the
Waals interaction – is known for more than 100 years [1,2]. The high stability of singly bonded diamondoid dimers resulting in
stabilizing nature of C–H···H–C interactions and their impor- very long C–C bonds [5,6], or very short H···H contacts in
1537

Beilstein J. Org. Chem. 2018, 14, 1537–1545.
Results and Discussion
tris(3,5-di-tert-butylphenyl)methane [7]. Besides a remarkable
effect on organic structures, dispersion can also affect the Analysis of the reaction mechanism
outcome of chemical transformations. Presumably due to attrac- To unravel the importance of London dispersion on the cobalt-
tive dispersive interactions between two adamantyl groups in catalyzed C–H cyanation of 2-phenylpyridine (1a), the under-
the transition state of a [4 + 2] cycloaddition of benzynes lying catalyst’s mode of action has to be fully understood. The
(Scheme 1), the seemingly sterically more hindered product is available experimental data indicated a reversible C–H metala-
formed preferentially [8].
tion, which led to the suggested catalytic cycle of Scheme 2
[30]. As computational investigations also allow the study of
Similar to other noncovalent interactions [9-11], London disper- intermediates that are too unstable to be observed under the ex-
sion can also play a crucial role in different transition-metal-cat- perimental conditions, we have analyzed the underlying reac-
alyzed reactions [12-17]. The C–H-rich di-1-adamantylphos- tion mechanism in more detail employing density functional
phine oxide – a typical dispersion element – was experimental- theory. A complete free energy profile on the B3LYP-D3BJ/
ly found to be an excellent preligand for ruthenium- and palla- def2-QZVP/COSMO//B3LYP-D3BJ/def2-TZVP potential
dium-catalyzed C–H functionalizations [18-23]. Similarly, energy surface is depicted in Figure 1 (black line), while the
computational studies revealed the importance of dispersion free-energy profile on the M06-L surface is summarized in Sup-
effects in palladium-catalyzed cross-coupling reactions [24-27]. porting Information File 1. Selected intermediates and transi-
For example, the contribution of London dispersion (up to tion states are shown in Figure 2.
37 kcal mol−1) has a huge influence on the ligand dissociation
process within the Pd(PPh3)4 system [25]. Furthermore, only The computational analysis starts with the catalytically active
the results obtained from dispersion-corrected density func- cobalt(III) acetate complex 4 which is generated in situ from the
tional theory [28,29] were in agreement with the experimental precatalyst [Cp*CoI2(CO)], AgSbF6, and KOAc. While the
observations and dispersion reduces the activation free energies iodine ions are captured by Ag+, carbon monoxide dissociates
by up to 30 kcal mol–1 [27].
and leaves the reaction mixture as a gas. Although the SbF6–
counter ion to the cationic cobalt complexes is considered to be
Currently, the strategic application of London dispersion in ca- weakly coordinating [37], specific interactions cannot be com-
talysis is still very difficult to achieve and, as a consequence, pletely ruled out. We have assumed that all of the positively
detailed insights in how dispersion influences organic reactions charged cobalt complexes on the reaction path are similarly
continue to be in high demand. Therefore, we have computa- affected by ion pairing and therefore, we base the following in-
tionally analyzed the recently developed cobalt-catalyzed C–H vestigation mainly on the reactions of the cobalt complexes and
cyanation of arenes (Scheme 2) [30-34]. Dispersion effects can do not include ion pairing in our analysis. Coordination of
be envisioned to be highly important in this system, as the rela- 2-phenylpyridine (1a) to this 16-electron species leads to the
tively C–H-rich ligand Cp* can interact with both substrates intermediate 5a (Figure 2) in a highly exergonic reaction step
within the cobalt complexes. In 2015, Li and Ackermann have (ΔG = −18.8 kcal mol−1), which also is the resting state of the
proposed the catalytic cycle (C–H cobaltation, ligand coordina- catalytic cycle. This intermediate could therefore be amenable
tion, insertion) shown in Scheme 2 which served as the starting to spectroscopic characterization. Based on our computational
point of this investigation [30]. We now report on our computa- analysis, the subsequent C–H cobaltation (5a → 7a) is ender-
tional findings supported by novel kinetic investigations to gonic (ΔΔG = +12 kcal mol−1) and proceeds in a step-wise
establish the reaction mechanism of this synthetically useful fashion. A similar mechanism has previously been described by
C–H activation and to elucidate the role of London dispersion in McMullin, Williams, and Frost [38], as well as by Ackermann
these transformations.
[39,40] for ruthenium-catalyzed C–H alkenylations. In the first
Scheme 1: Cycloaddition reaction of in situ generated benzynes resulting in the sterically more hindered adduct (Ad = 1-adamantyl) [8].
1538

Beilstein J. Org. Chem. 2018, 14, 1537–1545.
Scheme 2: Recently developed cobalt-catalyzed C–H cyanation [30].
transition state (TS1a, Figure 2), the κ2-coordination of the the cobalt–carbon bond takes place through TS3a. Within the
acetate ligand changes to a κ1-coordination. The resulting inter- four-membered transition state (Figure 2), the C–C bond to be
mediate 6a is stabilized by an agostic interaction between the formed is still rather long (C–C distance 1.92 Å), while the C–N
C–H bond and the metal atom as well as by an additional weak distance is already significantly elongated (1.15 Å in 9a, 1.22 Å
hydrogen bond between the C–H bond and the acetate oxygen in TS3a, 1.26 Å in 10a). Furthermore, a significant reorganiza-
(O…H distance 2.26 Å). A natural population analysis of struc- tion has to take place during this step: the former almost linear
ture 6a further confirms the stabilizing nature of these interac- N–C–N fragment (179.2°) changes to 137.9° in TS3a and
tions. In the second transition state TS2a (Figure 2), the C–H 124.4° in 10a, which results in a high barrier for this step.
bond is broken and the proton is transferred to the acetate which Subsequent coordination of acetic acid leads to intermediate
results in the formation of the cobaltacycle 7a.
11a. No transition states could be obtained for the following
β-elimination and proto-demetalation resulting in product 3a,
Acetic acid dissociates, and N-cyano-N-phenyl-p-toluenesulfon- the cobalt(III) acetate complex 4, and N-phenyl-p-toluene-
amide (2a) coordinates to the 16-electron intermediate 8a sulfonamide. All attempts starting from different potential
yielding 9a. Next, the insertion of the cyanating agent 2a into transition state structures resulted in barrierless reactions
1539

Beilstein J. Org. Chem. 2018, 14, 1537–1545.
Figure 1: Calculated free-energy profile for the cobalt-catalyzed C–H cyanation of 2-phenylpyridine (1a) [in kcal mol−1, [Co] = Cp*Co, black lines indi-
cate that dispersion (D3 correction with Becke–Johnson damping) [35,36] was included in the calculations while red lines indicate that dispersion was
not included].
when a proton approaches the amidine substructure (→ 12a). In contrast to previous computational studies on manganese(I)-
As the cyanated 2-phenylpyridine 3a is less Lewis-basic com- catalyzed fluoro-allylation reactions where β-fluoride and HF
pared to the starting material 1a, 12a could also react with 1a eliminations played an important role [41], similar reactions in-
in a thermodynamically favorable ligand exchange reaction volving amine eliminations seem to be not relevant in this reac-
(ΔG = −2.2 kcal mol−1) to yield complex 5a.
tion. Furthermore, a comparison with previous computational
1540

Beilstein J. Org. Chem. 2018, 14, 1537–1545.
Figure 2: Calculated structures, selected bond lengths (in Å), and
imaginary frequencies for representative intermediates and transition
states for the cobalt-catalyzed C–H cyanation of 2-phenylpyridine (1a).
investigations on copper-catalyzed ortho C–H cyanations of
vinylarenes revealed that those reactions take place via a com-
pletely different mechanism involving two distinct catalytic
cycles (copper-catalyzed electrophilic cyanative dearomatiza-
tion and base-catalyzed hydrogen transposition) [42,43].
Scheme 3: Kinetic profile of the cobalt-catalyzed C–H cyanation with
differently substituted cyanating agents 2.
computational analysis depicted in Table 1, the turnover-
limiting step for all substrates 1 is represented by the insertion
Inspired by this computational analysis, we experimentally of the cyanating agent 2a into the cobalt–carbon bond, which
probed the effect of differently substituted cyanation agents 2 can also be concluded based on the kinetic data of Scheme 3.
on the kinetics of the cobalt(III)-catalyzed C–H cyanation
(Scheme 3). Thus, we observed that electron-withdrawing Based on the computational analysis of Figure 1 and the experi-
groups significantly facilitated the desired transformation. As mental data depicted in Scheme 2, the turnover-limiting step for
the calculated rate-limiting transition state TS3 benefits from a this transformation is the insertion of 2a with an overall barrier
stabilization of the developing negative charge on the sulfon- of 25.5 kcal mol−1. The initial C–H cobaltation occurs with a
amide, the relative rates of Scheme 3 provide further support for smaller activation free energy of 15.5 kcal mol−1. These values
the migratory insertion representing the rate determining step are also in good qualitative agreement with the experimental
[44].
findings: The calculated high barriers match the prolonged reac-
tion times and high temperature required in the experimental
As differently substituted 2-phenylpyridines 1 have been em- studies and the reversible C–H metalation [30].
ployed experimentally, we included five representative sub-
strates (R = H, CH3, F, C(O)CH3, CN) into the computational Influence of London dispersion
analysis as well. For these calculations, only one functional In recent years, London dispersion, the attractive part of the
(B3LYP-D3BJ) and a smaller basis set (def2-SVP for non- van-der-Waals force, has been repeatedly identified as key to
metals and def2-TZVP for Co) were employed during the opti- stabilizing organic structures and facilitating novel reactivities
mization to reduce the computational cost. These results are [3]. As the Cp* ligand is a C–H-rich molecule, we envisioned
summarized in Table 1.
that dispersive interactions should be important for this transfor-
mation as well. As a consequence, we have analyzed this reac-
For the unsubstituted 2-phenylpyridine (1a), both computa- tion additionally with B3LYP without dispersion correction and
tional methods (Figure 1 and Table 1) and the optimized struc- the dispersion-corrected M06-L functional under otherwise
tures are generally rather similar to one another. Based on the identical conditions as a first starting point. Independent of the
1541

Beilstein J. Org. Chem. 2018, 14, 1537–1545.
Table 1: Calculated free energies for the reaction mechanism involving differently substituted 2-phenylpyridines 1a–e [B3LYP-D3BJ/def2-QZVP/
COSMO//B3LYP-D3BJ/def2-SVP, def2-TZVP for Co].
12
4
5
TS1
6
TS2
7
8
9
TS3
10
11
3
R = H (1a)
0.0
0.0
0.0
0.0
0.0
−20.1 −12.4 −14.1
−21.0 −14.2 −15.7
−19.8 −12.0 −12.4
−5.4
−7.1
−4.8
−3.4
−3.1
−8.4
−9.6
−9.7
−7.4
−8.9
−4.2
−4.5
−4.4
−2.1
−3.5
−15.5
−18.5
−17.5
−17.7
−16.8
+3.4
+0.6
+4.5
+3.4
+5.4
−18.3 −16.0 −34.5 −18.0
−18.8 −16.1 −35.5 −18.9
−17.3 −14.9 −32.3 −16.8
−17.5 −13.2 −30.9 −16.4
−14.9 −12.4 −29.4 −15.5
R = CH3 (1b)
R = F (1c)
R = C(O)CH3 (1d)
R = CN (1e)
−18.1
−18.0
−9.6
−9.9
−11.2
−9.9
computational method, the overall reaction free energy for the stabilizations of individual complexes have been calculated
transformation of Scheme 2 is almost identical [−15.9 (B3LYP- here.
D3BJ), −17.0 (B3LYP), and −15.9 (M06-L) kcal mol−1] indi-
cating that dispersion is less important for the overall thermody- A closer qualitative analysis of the intramolecular interactions
namics of this reaction. In contrast, a strong effect of the func- in these complexes employing the NCIPLOT program [48,49]
tional was observed for the complete energy profile. While the furthermore confirms these noncovalent interactions. While all
dispersion-corrected functional M06-L (see the Supporting plots are shown in Supporting Information File 1, Figure 3
Information File 1 for details) resulted in a comparable profile summarizes those for selected intermediates and transition
to that obtained with B3LYP-D3BJ (black lines in Figure 1), a states. For all structures, significant interactions can be found
significant deviation was observed when the latter was used between the Cp* ligand and the various phenyl groups of the
without any dispersion correction (red lines in Figure 1). All reagents. In addition, the presence of additional stabilizing
cobalt complexes are substantially stabilized by dispersive interactions such as further hydrogen bonds can also be con-
interactions resulting in a significant net reduction of the activa- firmed by this analysis (e.g., in TS3, see also the Supporting
tion free energy by 11 kcal mol−1. Comparable contributions of Information File 1).
London dispersion have also been calculated with other func-
tionals (TPSS [45] and PBE [46,47]). As expected, complexes To further probe the dispersive interaction of the Cp* ligand
with more nonbonding contacts (e.g., 10a) are better stabilized and the other ligands, we have additionally calculated the
than complexes where the Cp* ligand is located farther away dispersion interaction densities (DID) [50] for all intermediates
from other ligands (e.g., 8a). In comparison to computational and transition states at the SCS-LMP2/def2-TZVPP level of
investigations of Pd-catalyzed reactions [27], similar dispersive theory. The DID plots of Figure 4 reveal that medium to strong
Figure 3: Noncovalent interaction (NCI) analysis for selected intermediates and transition states. The gradient isosurfaces (s = 0.5 au) are colored ac-
cording to the sign of (λ2)ρ over the range of −0.05 (blue) to +0.05 (red).
1542

Beilstein J. Org. Chem. 2018, 14, 1537–1545.
Figure 4: Projected dispersion interaction density (DID) plots for selected intermediates and transition states. The molecular density isosurfaces
(0.1 e/Bohr3) are colored from zero interaction energy (blue) to the strongest dispersion interaction (red).
dispersive interactions can be found between the Cp* ligand erate the calculations [56,57]. For the analysis of the substitu-
and the aromatic and C–H-rich fragments in its proximity. In ent effect, the B3LYP functional with Grimme’s dispersion
line with the analyses presented in Figure 3 and Figure 4, a correction D3 (Becke–Johnson damping) was employed
local energy decomposition (LED) analysis [51] using DLPNO- together with the def2-SVP basis set for all non-metals and the
CCSD(T)/cc-pVDZ also confirmed medium to strong disper- def2-TZVP basis set for Co. Vibrational analysis verified that
sive interactions up to 12 kcal mol−1 between the Cp* ligand each structure was a minimum or transition state (iω < 30 cm−1
and the other ligands. Based on the computational analysis, were tolerated). Thermal corrections were calculated from
London dispersion is not only highly beneficial for the syntheti- unscaled harmonic vibrational frequencies at the same levels of
cally important cobalt-catalyzed C−H cyanation reaction, but it theory and refer to a standard state of 298.15 K and 1 mol L−1.
also emphasises that the Cp* ligand does not exclusively act as Entropic contributions to the reported free energies were ob-
a sterically demanding ligand in transition-metal-catalyzed reac- tained from partition functions evaluated with Truhlar’s quasi-
tions.
harmonic approximation [58]. This method uses the same
approximations as the usual harmonic oscillator approximation
except that all vibrational frequencies lower than 100 cm−1 are
Conclusion
We have analyzed the cobalt(III)-catalyzed C–H cyanation of set equal to 100 cm−1. Energies were subsequently derived from
differently substituted 2-phenylpyridines with N-cyano-N-aryl- single-point calculations employing the functionals described
p-toluenesulfonamide using density functional theory. On the above, the quadruple-ζ basis set def2-QZVP [52] and the
basis of our computational and experimental data, we can COSMO solvation model [59] for dichloroethane (ε = 10.125).
propose a reaction mechanism for this transformation. After an The dispersion interaction densities (DID) [50] were calculated
initial and reversible C–H cobaltation, the subsequent insertion at the SCS-LMP2/def2-TZVPP level of theory using MOLPRO
of the cyanating agents is the rate-limiting step. In addition, our 2015 [60,61]. The local energy decomposition analysis [51] was
calculations unravel that all the cobalt intermediates are consid- performed employing Neese’s domain-based local pair-natural
erably affected by London dispersion, which also results in a orbital (DLPNO) approach to the CCSD(T) method [DLPNO-
significant stabilization of the rate-limiting transition state.
CCSD(T)] [62-64] with tightPNO settings and the double-ζ
cc-pVDZ basis set as implemented in ORCA 4 [65]. All DFT
calculations were performed with Turbomole 7.1 [66,67] and
Computational Details
For all structures, geometry optimizations were performed with the NCIPLOT code was employed for the visualization non-
three different functionals using the def2-TZVP (def2-TZVPP covalent interactions [48,49].
for M06-L) basis set [52] and the m4 numerical quadrature grid
in the gas phase. The hybrid functional B3LYP [53,54] with and Experimental Details
without Grimme’s dispersion correction D3 (Becke–Johnson General remarks: Catalytic reactions were carried out in
damping) [35,36] as well as Truhlar’s dispersion-corrected Schlenk flasks under nitrogen atmosphere using predried glass-
M06-L [55] functional were employed in this investigation. For ware. 1,2-Dichlorethane (DCE) was dried and distilled over
the latter, the density fitting RI-J approach was used to accel- CaH2 under N2. N-Cyano-N-phenyl-p-toluenesulfonamide (2a)
1543

Beilstein J. Org. Chem. 2018, 14, 1537–1545.
3. Wagner, J. P.; Schreiner, P. R. Angew. Chem., Int. Ed. 2015, 54,
12274–12296. doi:10.1002/anie.201503476
[68] and Cp*Co(CO)I2 [69] were synthesized according to pre-
viously described methods. Other chemicals were obtained from
commercial sources and were used without further purification.
4. Grimme, S.; Schreiner, P. R. Angew. Chem., Int. Ed. 2011, 50,
12639–12642. doi:10.1002/anie.201103615
5. Schreiner, P. R.; Chernish, L. V.; Gunchenko, P. A.; Tikhonchuk, E. Y.;
Hausmann, H.; Serafin, M.; Schlecht, S.; Dahl, J. E. P.;
Carlson, R. M. K.; Fokin, A. A. Nature 2011, 477, 308–311.
doi:10.1038/nature10367
Kinetic experiments of the cobalt(III)-catalyzed C–H cyana-
tion: A suspension of 1 (78 mg, 0.50 mmol), 2 (0.75 mmol),
[Cp*Co(CO)I2] (6.0 mg, 2.5 mol %), AgSbF6 (8.6 mg,
5.0 mol %) and KOAc (2.5 mg, 5.0 mol %) in DCE (2.0 mL)
was heated at 120 °C. Aliquots up to ca 15% conversion
(25 µL; 10, 20, 30, 40, 50 min) were periodically removed by a
syringe and directly analyzed by GC using n-dodecane (30 µL)
as internal standard.
6. Fokin, A. A.; Chernish, L. V.; Gunchenko, P. A.; Tikhonchuk, E. Yu.;
Hausmann, H.; Serafin, M.; Dahl, J. E. P.; Carlson, R. M. K.;
Schreiner, P. R. J. Am. Chem. Soc. 2012, 134, 13641–13650.
doi:10.1021/ja302258q
7. Rösel, S.; Quanz, H.; Logemann, C.; Becker, J.; Mossou, E.;
Cañadillas-Delgado, L.; Caldeweyher, E.; Grimme, S.; Schreiner, P. R.
J. Am. Chem. Soc. 2017, 139, 7428–7431. doi:10.1021/jacs.7b01879
8. Aikawa, H.; Takahira, Y.; Yamaguchi, M. Chem. Commun. 2011, 47,
1479–1481. doi:10.1039/C0CC03025B
2-(Pyridin-2-yl)benzonitrile (3a): 1H NMR (CDCl3,
400 MHz) δ 8.73–8.70 (ddd, J = 4.7, 1.8, 0.9 Hz, 1H),
7.82–7.71 (m, 4H), 7.64 (dd, J = 7.6, 1.4 Hz, 1H), 7.49 (dd, J =
7.6, 1.2 Hz, 1H), 7.31 (ddd, J = 7.4, 4.7, 1.2 Hz, 1H); 13C NMR
(CDCl3, 125 MHz) δ 155.1 (Cq), 149.8 (CH), 143.4 (Cq), 136.7
(CH), 134.0 (CH), 132.7 (CH), 129.9 (CH), 128.6 (CH), 123.2
(CH), 123.1 (CH), 118.6 (Cq), 111.0 (Cq); IR (ATR): 3350,
2224, 1560, 1464, 758, 509 cm−1. EIMS m/z (relative intensity):
180 (100) [M+], 154 (5), 140 (5), 126 (5), 102 (5), 75 (5);
HRMS (EI) m/z: [M+] calcd. for C12H8N2, 180.0687; found,
180.0684. The analytical data are in accordance with those re-
ported in literature [30].
9. Hobza, P.; Müller-Dethlefs, K. Non-covalent Interactions; The Royal
Society of Chemistry: Cambridge, 2009. doi:10.1039/9781847559906
10.Wheeler, S. E.; Seguin, T. J.; Guan, Y.; Doney, A. C. Acc. Chem. Res.
2016, 49, 1061–1069. doi:10.1021/acs.accounts.6b00096
11.Walden, D. M.; Ogba, O. M.; Johnston, R. C.; Cheong, P. H.-Y.
Acc. Chem. Res. 2016, 49, 1279–1291.
doi:10.1021/acs.accounts.6b00204
12.Daugulis, O.; Roane, J.; Tran, L. D. Acc. Chem. Res. 2015, 48,
1053–1064. doi:10.1021/ar5004626
13.Wencel-Delord, J.; Glorius, F. Nat. Chem. 2013, 5, 369–375.
doi:10.1038/nchem.1607
14.Yeung, C. S.; Dong, V. M. Chem. Rev. 2011, 111, 1215–1292.
doi:10.1021/cr100280d
15.Ackermann, L. Chem. Rev. 2011, 111, 1315–1345.
doi:10.1021/cr100412j
Supporting Information
16.Chen, X.; Engle, K. M.; Wang, D.-H.; Yu, J.-Q. Angew. Chem., Int. Ed.
2009, 48, 5094–5115. doi:10.1002/anie.200806273
17.Bergman, R. G. Nature 2007, 446, 391–393. doi:10.1038/446391a
18.Ghorai, D.; Müller, V.; Keil, H.; Stalke, D.; Zanoni, G.;
Tkachenko, B. A.; Schreiner, P.; Ackermann, L. Adv. Synth. Catal.
2017, 359, 3137–3141. doi:10.1002/adsc.201700663
19.Graux, L. V.; Giorgi, M.; Buono, G.; Clavier, H. Dalton Trans. 2016, 45,
6491–6502. doi:10.1039/C5DT04683A
Supporting Information File 1
Cartesian coordinates, energies of all calculated structures,
and details of computational methods.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-14-130-S1.pdf]
20.Zell, D.; Warratz, S.; Gelman, D.; Garden, S. J.; Ackermann, L.
Chem. – Eur. J. 2016, 22, 1248–1252. doi:10.1002/chem.201504851
21.Ackermann, L.; Barfüsser, S.; Kornhaass, C.; Kapdi, A. R. Org. Lett.
2011, 13, 3082–3085. doi:10.1021/ol200986x
Acknowledgements
Financial support from the Fonds der chemischen Industrie
(Liebig scholarship to M.B.), the DFG (SPP 1807), and from
the University of Cologne within the excellence initiative is
gratefully acknowledged. We are very grateful to Professor
Ricardo Mata and Axel Wuttke for help with the calculation of
dispersion interaction densities. We thank the Regional
Computing Center of the University of Cologne (RRZK) for
providing computing time on the DFG-funded High Perfor-
mance Computing (HPC) system CHEOPS as well as for their
support.
22.Ackermann, L.; Althammer, A.; Born, R. Angew. Chem., Int. Ed. 2006,
45, 2619–2622. doi:10.1002/anie.200504450
23.Ackermann, L. Org. Lett. 2005, 7, 3123–3125. doi:10.1021/ol051216e
24.Sperger, T.; Sanhueza, I. A.; Schoenebeck, F. Acc. Chem. Res. 2016,
49, 1311–1319. doi:10.1021/acs.accounts.6b00068
25.Ahlquist, M. S. G.; Norrby, P.-O. Angew. Chem., Int. Ed. 2011, 50,
11794–11797. doi:10.1002/anie.201105928
26.Hansen, A.; Bannwarth, C.; Grimme, S.; Petrović, P.; Werlé, C.;
Djukic, J.-P. ChemistryOpen 2014, 3, 177–189.
doi:10.1002/open.201402017
27.Lyngvi, E.; Sanhueza, I. A.; Schoenebeck, F. Organometallics 2015,
34, 805–812. doi:10.1021/om501199t
References
1. London, F. Z. Phys. 1930, 63, 245–279. doi:10.1007/BF01421741
2. London, F. Trans. Faraday Soc. 1937, 33, 8b–26.
doi:10.1039/tf937330008b
28.Ehrlich, S.; Moellmann, J.; Grimme, S. Acc. Chem. Res. 2013, 46,
916–926. doi:10.1021/ar3000844
29.Grimme, S. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2011, 1,
211–228. doi:10.1002/wcms.30
1544

Beilstein J. Org. Chem. 2018, 14, 1537–1545.
30.Li, J.; Ackermann, L. Angew. Chem., Int. Ed. 2015, 54, 3635–3638.
doi:10.1002/anie.201409247
56.Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor. Chem. Acc.
1997, 97, 119–124. doi:10.1007/s002140050244
31.Yu, D.-G.; Gensch, T.; de Azambuja, F.; Vásquez-Céspedes, S.;
Glorius, F. J. Am. Chem. Soc. 2014, 136, 17722–17725.
doi:10.1021/ja511011m
57.Deglmann, P.; May, K.; Furche, F.; Ahlrichs, R. Chem. Phys. Lett.
2004, 384, 103–107. doi:10.1016/j.cplett.2003.11.080
58.Ribeiro, R. F.; Marenich, A. V.; Cramer, C. J.; Truhlar, D. G.
J. Phys. Chem. B 2011, 115, 14556–14562. doi:10.1021/jp205508z
59.Klamt, A.; Schüürmann, G. J. Chem. Soc., Perkin Trans. 2 1993,
799–805. doi:10.1039/P29930000799
32.Pawar, A. B.; Chang, S. Org. Lett. 2015, 17, 660–663.
doi:10.1021/ol503680d
33.Liu, W.; Ackermann, L. Chem. Commun. 2014, 50, 1878–1881.
doi:10.1039/c3cc49502g
60.Werner, H.-J.; Knowles, P. J.; Knizia, G.; Manby, F. R.; Schütz, M.
Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 242–253.
doi:10.1002/wcms.82
34.Cui, J.; Song, J.; Liu, Q.; Liu, H.; Dong, Y. Chem. – Asian J. 2018, 13,
482–495. doi:10.1002/asia.201701611
35.Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys. 2010,
132, 154104. doi:10.1063/1.3382344
61.MOLPRO, version 2015.1; a package of ab initio programs,
http://www.molpro.net.
36.Grimme, S.; Ehrlich, S.; Goerigk, L. J. Comput. Chem. 2011, 32,
1456–1465. doi:10.1002/jcc.21759
62.Riplinger, C.; Sandhoefer, B.; Hansen, A.; Neese, F. J. Chem. Phys.
2013, 139, 134101. doi:10.1063/1.4821834
37.Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004, 43, 2066–2090.
doi:10.1002/anie.200300620
63.Riplinger, C.; Neese, F. J. Chem. Phys. 2013, 138, 034106.
doi:10.1063/1.4773581
38.Leitch, J. A.; Wilson, P. B.; McMullin, C. L.; Mahon, M. F.; Bhonoah, Y.;
Williams, I. H.; Frost, C. G. ACS Catal. 2016, 6, 5520–5529.
doi:10.1021/acscatal.6b01370
64.Liakos, D. G.; Neese, F. J. Chem. Theory Comput. 2015, 11,
4054–4063. doi:10.1021/acs.jctc.5b00359
65.Neese, F. Wiley Interdiscip. Rev. Comput. Mol. Sci. 2012, 2, 73–78.
doi:10.1002/wcms.81
39.Kumar, N. Y. P.; Rogge, T.; Yetra, S. R.; Bechtoldt, A.; Clot, E.;
Ackermann, L. Chem. – Eur. J. 2017, 23, 17449–17453.
doi:10.1002/chem.201703680
66.TURBOMOLE, V7.1, 2015; TURBOMOLE GmbH, since 2007,
http://www.turbomole.com.
40.Bu, Q.; Rogge, T.; Kotek, V.; Ackermann, L. Angew. Chem., Int. Ed.
2018, 57, 765–768. doi:10.1002/anie.201711108
41.Zell, D.; Dhawa, U.; Müller, V.; Bursch, M.; Grimme, S.; Ackermann, L.
ACS Catal. 2017, 7, 4209–4213. doi:10.1021/acscatal.7b01208
42.Yang, Y.; Buchwald, S. L. Angew. Chem., Int. Ed. 2014, 53,
8677–8681. doi:10.1002/anie.201402449
a development of University of Karlsruhe and Forschungszentrum
Karlsruhe GmbH, 1989–2007.
67.Furche, F.; Ahlrichs, R.; Hättig, C.; Klopper, W.; Sierka, M.;
Weigend, F. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2014, 4,
91–100. doi:10.1002/wcms.1162
68.Liu, W.; Richter, S. C.; Mei, R.; Feldt, M.; Ackermann, L.
Chem. – Eur. J. 2016, 22, 17958–17961. doi:10.1002/chem.201604621
69.Sun, B.; Yoshino, T.; Matsunaga, S.; Kanai, M. Adv. Synth. Catal.
2014, 356, 1491–1495. doi:10.1002/adsc.201301110
43.Yang, Y.; Liu, P. ACS Catal. 2015, 5, 2944–2951.
doi:10.1021/acscatal.5b00443
44.Orientating calculations using 2b and 2c as the cyanating agent
parallel the trend observed in the kinetic data (ΔG‡(2b) < ΔG‡(2b) <
ΔG‡(2c)).
45.Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E.
Phys. Rev. Lett. 2003, 91, 146401.
License and Terms
doi:10.1103/PhysRevLett.91.146401
46.Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,
3865–3868. doi:10.1103/PhysRevLett.77.3865
47.Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1997, 78,
1396. doi:10.1103/PhysRevLett.78.1396
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
48.Johnson, E. R.; Keinan, S.; Mori-Sánchez, P.; Contreras-García, J.;
Cohen, A. J.; Yang, W. J. Am. Chem. Soc. 2010, 132, 6498–6506.
doi:10.1021/ja100936w
49.Contreras-García, J.; Johnson, E. R.; Keinan, S.; Chaudret, R.;
Piquemal, J.-P.; Beratan, D. N.; Yang, W. J. Chem. Theory Comput.
2011, 7, 625–632. doi:10.1021/ct100641a
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(https://www.beilstein-journals.org/bjoc)
50.Wuttke, A.; Mata, R. A. J. Comput. Chem. 2017, 38, 15–23.
doi:10.1002/jcc.24508
51.Schneider, W. B.; Bistoni, G.; Sparta, M.; Saitow, M.; Riplinger, C.;
Auer, A. A.; Neese, F. J. Chem. Theory Comput. 2016, 12, 4778–4792.
doi:10.1021/acs.jctc.6b00523
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.14.130
52.Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297–3305. doi:10.1039/b508541a
53.Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
doi:10.1063/1.464913
54.Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789.
doi:10.1103/PhysRevB.37.785
55.Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125, 194101.
doi:10.1063/1.2370993
1545