Computational study of diastereoselective ortho-lithiations of chiral ferrocenes

Abstract

Diastereoselective ortho-lithiations serve for the preparation of many important chiral ferrocenes, however diastereoselections of these lithiations are explained only by simple steric models. We elucidated ortho-lithiations of three important ferrocenes based on DFT calculations. The calculations showed that simple models of transition states involving ferrocenyl substrates and lithium bases can only in some cases account for the experimental results. Transition state models, which include solvent or coordinating additives as distinct entities bound to lithium, can satisfactorily explain diastereoselection of the ortho-lithiations of chiral ferrocenes.

---

Introduction

Ferrocene derivatives with planar chirality serve as successful ligands in transition metal catalysed reactions.^1–3 One of the key steps in their synthesis is usually an introduction of planar chirality by ortho-metalation, such as palladation,^4,5 magnetisation,^6–9 or lithiation. The most versatile and frequently used method is ortho-lithiation. The crucial element of this reaction is the presence of an ortho-directing group, which can coordinate and direct the approach of an alkyllithium or lithium dialkylamide base. If the ortho-directing group is chiral, two diastereoisomers of lithium intermediates are formed (Scheme 1). These Li-intermediates can then be trapped with a variety of electrophiles leading to planar chiral 1,2-disubstituted ferrocene derivatives.

Scheme 1 Formation of two possible lithium intermediates evolved from a chiral ferrocene derivative.

Ugi and co-workers described the first diastereoselective ortho-lithiations of ferrocene derivative – N,N-dimethyl-1-ferrocenylethylamine (Ugi amine).¹⁰ Based on this reaction Hayashi developed chiral amino phosphanes, which later culminated in the development of ferrocenyl diphosphines such as the Josiphos family of chiral ligands.¹¹ Apart from the Ugi amine, other amines can serve as ortho-directing groups.^12–14 Nitrogen, as a stronger nucleophile, serves as a directing atom in oxazolines, which form the basis of another important group of chiral ligands. Three research groups described ortho-lithiations of ferrocenyl oxazolines almost simultaneously.^15–17 Other nitrogen based ortho-directing groups are hydrazones,^18,19 azepines,²⁰ sulfoximines,²¹ and imidazolines.²² Oxygen performs the ortho-directing function in sulfoxides²³ acetals,²⁴ methoxy-derivatives,²⁵ and alcohols.²⁶ The structures of the major products of diastereoselective ortho-lithiations have been well established experimentally, either by X-ray or NMR; however, a detailed understanding of the reasons for the observed diastereoselectivity is lacking. It is usually explained by empirical consideration based on simple models of transition states or even intermediates. DFT calculations, only recently started to help explain the underlying reasons for diastereoselection in ortho-lithiations of ferrocene derivatives.^27,28 We have studied diastereoselective ortho-lithiations of ferrocenophanes²⁹ and have suggested the mechanism of dilithiation reaction leading to the Taniaphos family of ligands.³⁰ A lack of understanding of the lithiation of ferrocenes prompted us to investigate diastereoselective ortho-lithiations of three relevant ferrocene compounds by DFT calculations. Although many factors, such as steric demands or aggregation of alkyllithium reagent, solvent, additive and size of the substituent on a chiral stereogenic centre, seem to influence the diastereoselectivity of ortho-lithiations we argue that even a rather simplistic model including only ferrocene substrate and a lithium base can account, at least on a qualitative level, for the observed diastereoselection. On the basis of these facts, we decided to provide a theoretical explanation for the experimentally observed diastereoselective lithiations on ferrocene derivatives 1–3, which are essential for the synthesis of useful chiral ferrocenyl ligands (Fig. 1).

Fig. 1 Precursors for preparation of relevant planar-chiral ferrocene ligands.

For the compounds 1–3, Table 1 summarizes experimental results obtained by ortho-lithiation.³

Table 1 Experimental results for the ortho-lithiation of compounds 1–3

Compound	Base	Additive	E^+	Solvent	dr	Major diastereomer
(R)-1	n-BuLi	—	PPh2	Et2O	96 : 4	(R,Rp)
(S)-2	s-BuLi	—	Me3Si	THF	8 : 1	(S,Sp)
(S)-2	s-BuLi	TMEDA	Me3Si	Hexane	500 : 1	(S,Sp)
(S)-3	LDA	—	Me3Si	THF	99 : 1	(S,Rp)

---

Results and discussion

The calculations were performed at DFT level using the BP86 functional, because it has been well tested on a variety of compound classes including the main group elements as well as transition metals and their compounds. As a basis set, we used Ahlrich's triple-zeta basis set def2-TZVP. For more details see Experimental. We have calculated complete energy profiles for the lithiation of derivatives 1–3, to gain a better understanding of the processes. Calculations of reaction profiles began from isolated starting materials. Firstly, we located complexes, which were likely to be formed between lithium base and the directing atom of the ferrocene derivative. The complexation of lithium alkyl or amide bases on ortho-directing atoms is thermodynamically favoured. The origin of stabilization can be attributed to a favourable binding energy between the relatively concentrated positive charge of the small lithium atom and negative charge of the electron pair.³¹ This fact is in accordance with the concept of complex-induced proximity effect, described by Snieckus and Beak.³² Already the approach of bases followed by coordination of the free electron pair predicts the formation of transition states. A different situation occurred when the solvent or additive was included in the calculations. The energy of the system did not decrease in the complexation step (the energy difference was just 0–8 kJ mol⁻¹). The major energy gain was observed as a result of Li-coordination with solvent or an additive.

ortho-Lithiation of Ugi amine (1)

Homochiral Ugi amine (1) was the first chiral ferrocenyl amine subjected to diastereoselective lithiations.² Ugi in his seminal work observed the formation of two diastereoisomers in the ratio of 96 : 4. Hayashi later described the same diastereomeric ratio in the preparation of ferrocenyl amino phosphines.³³ Observed diastereoselectivity was explained by strong repulsion of a methyl group with ferrocene moiety in a non-favoured (R,S_p) transition state. This repulsion is not present in the favoured (R,R_p) transition state. Fig. 2 shows energy profiles for the lithiation of amine (R)-1 without inclusion of Et₂O (2a) and with two molecules of Et₂O coordinated to lithium (2b). Table 2 gathers relative energies for the reactants, the transitions states and lithiated intermediates of the lithiation of amine (R)-1 with n-BuLi with and without Et₂O.

Fig. 2 Energy diagrams of the lithiation of amine (R)-1 without Et₂O (2a) and with Et₂O (2b).

Table 2 Relative energies for the lithiation of 1 with n-BuLi

Compound	ΔErel ^a (kJ mol^−1)	Compound	ΔErel ^b (kJ mol^−1)	Compound	ΔErel (kJ mol^−1)
a RI-DFT, BP86, def2-TZVP. b RI-DFT-COSMO (ε = 4.33), BP86, def2-TZVP single point.
(R)-1 + n-BuLi	83.5	(R)-1 + n-BuLi	45.2	(R)-1 + n-BuLi + 2Et2O	95.5
—	—	—	—	(R)-1 + n-BuLi·2Et2O	0.2
(R)-1·n-BuLi (C1a)	0.0	(R)-1·n-BuLi (C1a)	0.0	(R)-1·n-BuLi·2Et2O (C2a)	0.0
(R)-1·n-BuLi (C1b)	2.5	(R)-1·n-BuLi (C1b)	3.6	(R)-1·n-BuLi·2Et2O (C2b)	1.8
(R,Rp)-TS1	59.3	(R,Rp)-TS1	72.7	(R,Rp)-TS2	62.1
(R,Sp)-TS1	68.7	(R,Sp)-TS1	77.1	(R,Sp)-TS2	73.1
(R,Rp)-Int1 + BuH	−25.0	(R,Rp)-Int1 + BuH	−38.0	(R,Rp)-Int2 + BuH	−66.0
(R,Sp)-Int1 + BuH	−21.9	(R,Sp)-Int1 + BuH	−33.9	(R,Sp)-Int2 + BuH	−52.7

---

The complex C1a leads to transition state (R,R_p)-TS1 and C1b to transition state (R,S_p)-TS1. From the two calculated diastereomeric transition states, (R,R_p)-TS1 is energetically more favoured (ΔE = 12 kJ mol⁻¹) than (R,S_p)-TS1. This fact is in accordance with the experimentally observed configurations of products. The fact that transition state TS1 indeed connects starting material with product was confirmed by performing a potential energy scan (see ESI†).

Experimental data show that the course of lithiation often heavily depends on the solvent. The importance of solvation of lithium compounds is well known. Moreover, solvation can influence the formation of transition states or stabilization of the resulting Li-intermediates.³⁴ Taking this fact into account, we proposed to simulate the effect of solvation of Li-derivatives firstly as a dielectric medium and secondly by two molecules of Et₂O, which are included not just as a dielectric medium, but as defined molecules coordinated to lithium.

The calculation performed with COSMO (Et₂O, ε = 4.33) resulted in an energy difference of transition states TS-1 of just 4 kJ mol⁻¹. This result suggests that the solvent in the lithiation did not have influence just as a dielectric medium, but that the solvent molecules needed to be explicitly coordinated on lithium with a direct effect on diastereoselectivity of the reaction.

Interestingly, the inclusion of solvent resulted in approximately the same energy difference between the two diastereomeric transition states. A significant change was observed in the relative energies of lithiated products (R,R_p)-Int2, (R,S_p)-Int2. The energy difference increased to ΔE_rel = 13 kJ mol⁻¹. The inclusion of solvent in the calculation did not affect the formation of transition states but a stabilization of Li-intermediates.

By examination of electron density maps of the calculated transition states TS1, it was possible to see steric repulsions, which were responsible for differences between the transition states. In accordance with Ugi's explanation, there indeed seems to be repulsion of the methyl group with a ferrocene moiety in the non-preferred transition states (R,S_p)-TS1 (Fe–H(CH₃) 3.216 Å) and (R,S_p)-TS2 (Fe–H(CH₃) 3.173 Å) (Fig. 3). The methyl group, in preferred transition states, is located in the plane of the cyclopentadienyl (Cp) ring.

Fig. 3 Electron maps of (R,S_p)-TS1 and (R,S_p)-TS2.

The repulsion of methyl group by the ferrocene skeleton can be visualized by a NCI plot (Fig. 4).³⁵

Fig. 4 The NCI surface of (R,S_p)-TS1.

The alkyllithium bases are present as aggregates in solution. As this can also influence the formation of transition states, we performed a calculation of the lithiation of amine (R)-1 with n-BuLi-dimer. Although in n-BuLi solutions aggregates up to a hexamer can exist, the presence of an amine coordinating atom, such as nitrogen, will likely decrease the aggregation (Fig. 5).

Fig. 5 The reaction profile for the lithiation of amine (R)-1 with n-BuLi dimer.

Therefore we used the n-BuLi-dimer as a model for the aggregated species. The energy difference between the two diastereomeric transition states increased to 18 kJ mol⁻¹ and the correct diastereomer was predicted as being more favourable. The reason for the bigger energy difference between the transition states is probably the increased steric interactions in the presence of n-BuLi-dimer in complexes and transition states. The relative energies of the reactants, the transitions states and the resulting Li-intermediates of the lithiation of amine (R)-1 are summarized in Table 3.

Table 3 The relative energies for the lithiation of amine (R)-1 with n-BuLi dimer

Compound	ΔE (kJ mol^−1)
(R)-1 + (n-BuLi·2Et2O)2	12.4
(R)-1·(n-BuLi)2·3Et2O (C3) + Et2O	0.0
(R,Rp)-TS3 + 3Et2O	71.7
(R,Sp)-TS3 + 3Et2O	89.7
(R,Rp)-Int2 + BuH + (n-BuLi·2Et2O)	−5.9
(R,Sp)-Int2 + BuH + (n-BuLi·2Et2O)	7.4

---

ortho-Lithiation of ferrocenyl oxazoline 2

Diastereoselective ortho-lithiations of ferrocenyl oxazolines leads to several useful chiral ligands.^36,37 Quenching of the respective lithium intermediates with chlorophosphines gives oxazolinyl phosphines.^15,17 The reaction with ketones leads to corresponding hydroxyl oxazolines.³⁸ Other electrophiles such as disulfides and diselenides were also used for the synthesis of chiral ligands. ortho-Lithiation of oxazoline is directed by nitrogen as the stronger nucleophile. Sammakia proved this fact by bridging the oxazoline moiety with the unsubstituted Cp-ring, thus restricting rotation around the Cp–C(N) bond.³⁹ Although Sammakia, Richards, and Uemura performed reactions under different conditions (solvent, additive, base), the major diastereoisomer of product was always the same. The experiments resulted only in different diastereomeric ratios. The first experiment performed with n-BuLi in Et₂O offered product in the diastereomeric ratio of 2.5 : 1. An improvement was reached in the presence of s-BuLi as the base and THF as the solvent. The diastereoselectivity of lithiations changed to 8 : 1. The presence of TMEDA dramatically increased the diastereoselectivity to 100 : 1 (both in diethylether and hexane). The most significant improvement occurred by a change of n-BuLi to s-BuLi where diastereomeric ratio increased to 500 : 1.³⁹ On the basis of the observed results, the authors explained the formation of the major (S,R_p) diastereomer by a simple model, in which the repulsion of the i-Pr group from the oxazoline moiety with lithium base is responsible for the diastereoselectivity (Fig. 6).

Fig. 6 Reaction coordinates of the lithiation of compound 2 (a) without (b) with THF.

The calculations suggest that this simple model may not be sufficient to account for the observed diastereoselection. Diastereomeric transition states TS4 composed of oxazoline 2 and s-BuLi showed only a negligible energy difference ∼1 kJ mol⁻¹. The situation manifests itself already in the calculation of complexes. The energy difference between complexes C4a and C4b was just 0.27 kJ mol⁻¹. As is shown in Table 1, a major influence of THF was experimentally observed. The calculation of transition states, where two molecules of THF are included, confirmed the experimental results. Also, in this case it is not sufficient to include solvent only as the dielectric medium. The energies of the transition states were the same and neither transition state was favoured. However, the experimentally observed diastereoselectivity of the lithiation of compound 2 in THF is 8 : 1.

The calculation of lithiation, where two THF molecules were coordinated on the lithium atom observed a preferred formation of the less hindered transition state (S,R_p)-TS5 by 7.4 kJ mol⁻¹. While in the transition states TS4, we did not observe any repulsion of the i-Pr group with base or ferrocene skeleton, in TS5, THF molecules showed repulsion of the i-Pr group with approaching base. This fact made one transition state more favourable and the diastereoselectivity changed. This theoretical result confirmed the experimental observations of Sammakia and co-workers.³⁹

The inclusion of strongly chelating agent N,N,N′,N′-tetramethylethylenediamine (TMEDA) had a significant impact on the results. As this sterically bulky compound coordinated to lithium, it filled the space around lithium and thus helped increase steric interactions between the base and the substrate.

This effect was manifested both in complexation and was followed by the formation of transition states. It is confirmed by the large energy difference between complexes C6a and C6b. The formation of C6b was less favoured and therefore the formation of (S,S_p)-TS6 was not preferred. The calculated transition state (S,R_p)-TS6 is by 19 kJ mol⁻¹ more stable than (S,S_p)-TS6. These calculated results correlate with the experimentally obtained major diastereomer (Fig. 7). The relative energies of the reactants, the transitions states and the resulting Li-intermediates of the lithiation reactions of oxazoline (S)-2, both with and without TMEDA and THF, are summarized in Tables 4 and 5.

Fig. 7 Reaction coordinates of the lithiation of compound 2 with TMEDA.

Table 4 Relative energies for the lithiation reactions of 2

Compound	ΔE^a (kJ mol^−1)	Compound	ΔE^b (kJ mol^−1)
a RI-DFT, BP86, def2-TZVP. b RI-DFT-COSMO (ε = 7.58), BP86, def2-TZVP single point.
(S)-2 + s-BuLi	86.9	(S)-2 + s-BuLi	37.5
(S)-2·s-BuLi (C4a)	0.0	(S)-2·s-BuLi (C4a)	0.0
(S)-2·s-BuLi (C4b)	0.3	(S)-2·s-BuLi (C4b)	0.1
(S,Rp)-TS4	45.6	(S,Rp)-TS4	57.8
(S,Sp)-TS4	46.4	(S,Sp)-TS4	57.9
(S,Rp)-Int3 + BuH	−65.0	(S,Rp)-Int3 + BuH	−79.1
(S,Sp)-Int3 + BuH	−65.7	(S,Sp)-Int3 + BuH	−80.7

---

Table 5 Relative energies for the lithiation reactions of 2 with THF and TMEDA

Compound	ΔE (kJ mol^−1)
(S)-2 + s-BuLi + 2THF	124.7
(S)-2 + s-BuLi·2THF	6.3
(S)-2·s-BuLi·2THF (C5a)	0.0
(S)-2·s-BuLi·2THF (C5b)	14.6
(S,Rp)-TS5	38.6
(S,Sp)-TS5	46.0
(S,Rp)-Int4 + BuH	−106.6
(S,Sp)-Int4 + BuH	−105.4
(S)-2 + s-BuLi + TMEDA	119.4
(S)-2 + s-BuLi·TMEDA	7.7
(S)-2·s-BuLi·TMEDA (C6a)	0.0
(S)-2·s-BuLi·TMEDA (C6b)	18.1
(S,Rp)-TS6	39.9
(S,Sp)-TS6	59.2
(S,Rp)-Int5 + BuH	−106.3
(S,Sp)-Int5 + BuH	−105.0

---

Inspection of the electron density maps of diastereomeric transition states TS4 shows that the distance between i-Pr group oriented up and alkyllithium base is sufficiently large (2.92 Å) and no repulsion is observed. On the other hand, the presence of TMEDA, as a bulky additive coordinated to the alkyllithium base, leads to larger repulsion with the i-Pr group in (S,S_p)-TS6 and thus the formation of (S,R_p)-TS6 is favoured (Fig. 8).

Fig. 8 The electron density maps of (S,S_p)-TS4 and (S,S_p)-TS6.

ortho-Lithiation of ferrocenyl sulfoxide 3

Diastereoselective ortho-lithiations and electrophilic quenching of ferrocenyl sulfoxide is another useful transformation because the sulfoxide group allows facile modification and thus preparation of chiral ligands. A sulfoxide lithiation strategy led to successful ligands such as the second generation of Taniaphos⁴⁰ and related ligands,⁴¹ diphosphanes with only planar chirality,⁴² or Carretero's Fesulphos family of ligands.⁴³ In the original report, Kagan and co-workers described lithiation of sulfoxide 3 with n-BuLi.²³ They later found a more convenient procedure for the lithiation of ferrocenyl sulfoxide 3 with LDA, because butyl lithium also attacks the C–S bond.⁴⁴ The lithiation proceeds preferentially in the pro-S position. To the best of our knowledge, however, no explanation can be found for the observed diastereoselectivity in the lithiation of sulfoxide 3.

At the beginning of this part of the study, we have realized that LDA conformations play an important role and significantly influence the results. Energy differences of up to ΔE_rel = 5 kJ mol⁻¹ were found (see ESI†). The formation of complex C7a is more favored (3.5 kJ mol⁻¹) than the formation of complex C7b (Fig. 9 and Table 6).

Fig. 9 Reaction coordinates of lithiation of compound 3(a) without and (b) with THF.

Table 6 Relative energies for the lithiation of 3 with and without THF

Compound	ΔE^a (kJ mol^−1)	Compound	ΔE^b (kJ mol^−1)	Compound	ΔE^a (kJ mol^−1)
a RI-DFT, BP86, def2-TZVP. b RI-DFT-COSMO (ε = 7.58), BP86, def2-TZVP, single point.
(S)-3 + LDA	88.5	(S)-3 + LDA	36.4	(S)-3 + LDA + 2THF	114.3
—	—	—	—	(S)-3 + LDA·2THF	5.9
(S)-3·LDA (C7b)	0.0	(S)-3·LDA (C7b)	0.0	(S)-3 + LDA·2THF (C8b)	0.0
(S)-3·LDA (C7a)	3.5	(S)-3·LDA (C7a)	1.6	(S)-3 + LDA·2THF (C8a)	6.5
(S,Rp)-TS7	68.8	(S,Rp)-TS7	74.8	(S,Rp)-TS8	49.5
(S,Sp)-TS7	62.0	(S,Sp)-TS7	61.4	(S,Sp)-TS8	30.7
(S,Rp)-Int6 + (i-Pr)2NH	55.5	(S,Rp)-Int6 + (i-Pr)2NH	26.5	(S,Rp)-Int7 +(i-Pr)2NH	1.1
(S,Sp)-Int6 + (i-Pr)2NH	40.8	(S,Sp)-Int6 + (i-Pr)2NH	12.5	(S,Sp)-Int7 +(i-Pr)2NH	−9.5

---

Diastereoselection seems to be induced by the orientation of a tosyl group. While the tosyl group in C7a and (S,S_p)-TS7 is oriented over the ferrocene skeleton, the position of this group in C7b and (S,R_p)-TS7 is near the ferrocene moiety (Fig. 10).

Fig. 10 The electron density maps of (S,R_p)-TS7 and (S,R_p)-TS8.

The inclusion of the solvent in the calculation increased the energy difference between the complexes and the transition states. The coordination of two molecules of THF on lithium increased the steric demands of the lithium base. The repulsion of LDA with a tosyl group is observed in non-favoured transition state (S,R_p)-TS8. On the other hand, this repulsion is not possible in the transition state (S,S_p)-TS8, because the tosyl group is oriented on the opposite side. The energetically favoured transition states lead to the product (S,R_p)-Int7, which is in accordance with experiment (Fig. 9 and Table 6).

Conclusion

In conclusion, diastereoselective ortho-lithiations of three key ferrocene derivatives were studied by DFT calculations. Calculations can in all examples predict the formation of the experimentally obtained major diastereoisomer. The involvement of solvents or additives influenced energy differences of transition states for compounds 2 and 3. Coordinated solvent or additive increases the steric interactions between the Li-base and ferrocenyl substrate, which helps to differentiate between the diastereomeric transition states.

The electron density maps of transition states help observe interactions in the diastereomeric transition states. Table 7 summarizes the bonding distances, energy differences for complexes, transition states and intermediates of the studied lithiations.

Table 7 The differences of energies of complexes, transition states and intermediates of derivate 1–3 in the theoretical study of lithiations^a

	(R)-1			(S)-2			(S)-3
Int	Int1	Int2	Int2 ^e	Int3	Int4	Int5	Int6	Int7
a Energy of calculated complexes, transition states and intermediates are displayed in kJ mol^−1. b Distances between hydrogen and base in complexes displayed in Å. c Distance between hydrogen and ferrocenyl carbon displayed in Å. d Distance between ferrocenyl carbon and lithium displayed in Å. e Intermediates produced when BuLi-dimer was considered.
C	C1	C2	C3	C4	C5	C6	C7	C8
ΔE C-a	0.0	0.0	0.0	0.0	0.0	0.0	3.5	6.5
d(H–C)^b	3.01	3.40		2.74	2.56	2.61	4.34	2.48
ΔE C-b	2.5	1.8	—	0.3	14.6	18.1	0.0	0.0
d(H–C)^b	2.97	3.48		2.72	2.73	3.85	4.36	5.69
TS	TS1	TS2	TS3	TS4	TS5	TS6	TS7	TS8
ΔE (Rp)-TS	0.0	0.0	0.0	0.0	0.0	0.0	6.8	18.7
d(H–C)^c	1.40	1.44	1.34	1.37	1.41	1.40	1.43	1.45
d(C–Li)^d	2.11	2.40	2.63	2.30	2.75	2.68	2.56	3.11
ΔE (Sp)-TS	9.4	10.9	18.0	0.8	7.4	19.3	0.0	0.0
d(H–C)^c	1.43	1.44	1.35	1.36	1.41	1.39	1.45	1.43
d(C–Li)^d	2.14	2.38	2.60	2.26	2.77	2.64	2.90	3.45
Int	Int1	Int2	Int2^e	Int3	Int4	Int5	Int6	Int7
ΔE (Rp)-Int	0.0	0.0	0.00	0.7	0.0	0.0	14.7	10.6
ΔE (Sp)-Int	3.1	13.3	13.3	0.0	1.2	1.3	0.0	0.0

---

The NCI plot software was use for visualization of non-covalent repulsions in transition states.

Experimental

All structures were first pre-optimized at the semiempirical PM3 level using the Spartan 8 program package.⁴⁵ Transition states were optimized at DFT level using BP86 functional and 6-31G* basis set in Spartan 8. Finally transition states were refined at RI-DFT,^46,47 level with the BP86 functional⁴⁸ using def2-TZVP basis set⁴⁹ as implemented in the Turbomole package.⁵⁰ Minima were also optimized at RI-DFT level (BP86/TZVP). The empirical dispersion correction was evaluated by performing a single point calculation with DTF-D3 as implemented in Turbomole. However, these calculations did not correlate well with the experimental results.⁵¹ Searching for conformers of LDA and (i-Pr)₂NH was performed with HF level and (6-31G*) basis set in Spartan 8. Final conformers of LDA and (i-Pr)₂NH were optimized at DFT level with the BP86 functional using def2-TZVP basis set as implemented in the Turbomole package. All structures were characterized by frequency calculations at RI-DFT level⁵² with the BP86 functional using def2-TZVP. All transition states have only one imaginary vibration corresponding to C–H bond breaking (verified by visualization); minima have only real vibrations.

Acknowledgements

Financial support from Slovak Grant Agency VEGA, grant no. 1/0623/12 is gratefully acknowledged.

Notes and references

1. P. Štepnička, Ferrocene: Ligands, Materials and Biomolecules, Wiley, Chichester, 2008.
2. A. Togni and T. Hayashi, Ferrocenes, VCH, Weinheim, 1995.
3. L.-X. Dai and X.-L. Hou, Chiral Ferrocenes in Asymmetric Catalysis, Wiley-VCH, Weinheim, 2010.
4. V. I. Sokolov, L. L. Troitskaya and O. A. Reutov, J. Organomet. Chem., 1979, 182, 537–546.
5. L. L. Troitskaya, Z. A. Starikova, T. V. Demeshchik, S. T. Ovseenko, E. V. Vorontsov and V. I. Sokolov, J. Organomet. Chem., 2005, 690, 3976–3982.
6. A. H. Stoll, P. Mayer and P. Knochel, Organometallics, 2007, 26, 6694–6697.
7. P. C. Andrikopoulos, D. R. Armstrong, W. Clegg, C. J. Gilfillan, E. Hevia, A. R. Kennedy, R. E. Mulvey, C. T. O'Hara, J. A. Parkinson and D. M. Tooke, J. Am. Chem. Soc., 2004, 126, 11612–11620.
8. K. W. Henderson, A. R. Kennedy, R. E. Mulvey, C. T. O'Hara and R. B. Rowlings, Chem. Commun., 2001, 1678–1679.
9. W. Clegg, K. W. Henderson, A. R. Kennedy, R. E. Mulvey, C. T. O'Hara, R. B. Rowlings and D. M. Tooke, Angew. Chem., Int. Ed., 2001, 40, 3902–3905.
10. D. Marquarding, H. Klusacek, G. Gokel, P. Hoffmann and I. Ugi, J. Am. Chem. Soc., 1970, 92, 5389–5393.
11. A. Togni, C. Breutel, A. Schnyder, F. Spindler, H. Landert and A. Tijani, J. Am. Chem. Soc., 1994, 116, 4062–4066.
12. C. Ganter and T. Wagner, Chem. Ber., 1995, 128, 1157–1161.
13. L. Xiao, R. Kitzler and W. Weissensteiner, J. Org. Chem., 2001, 66, 8912–8919.
14. M. Widhalm, U. Nettekoven and K. Mereiter, Tetrahedron: Asymmetry, 1999, 10, 4369–4391.
15. C. J. Richards, T. Damalidis, D. E. Hibbs and M. B. Hursthouse, Synlett, 1995, 74–76.
16. T. Sammakia, H. A. Latham and D. R. Schaad, J. Org. Chem., 1995, 60, 10–11.
17. Y. Nishibayashi and S. Uemura, Synlett, 1995, 79–81.
18. D. Enders, R. Peters, R. Lochtman and G. Raabe, Angew. Chem., Int. Ed., 1999, 38, 2421–2423.
19. D. Enders, R. Peters, R. Lochtman and J. Runsink, Eur. J. Org. Chem., 2000, 2839–2850.
20. M. Widhalm, K. Mereiter and M. Bourghida, Tetrahedron: Asymmetry, 1998, 9, 2983–2986.
21. C. Bolm, M. Kesselgruber, K. Muñiz and G. Raabe, Organometallics, 2000, 19, 1648–1651.
22. R. Peters and D. F. Fischer, Org. Lett., 2005, 7, 4137–4140.
23. F. Rebière, O. Riant, L. Ricard and H. B. Kagan, Angew. Chem., Int. Ed. Engl., 1993, 32, 568–570.
24. O. Riant, O. Samuel and H. B. Kagan, J. Am. Chem. Soc., 1993, 115, 5835–5836.
25. M. Lotz, T. Ireland, K. Tappe and P. Knochel, Chirality, 2000, 12, 389–395.
26. B. J. Ueberbacher, H. Griengl and H. Weber, Chem. Commun., 2008, 3287–3289.
27. C. Metallinos, J. Zaifman, T. Dudding, L. Van Belle and K. Taban, Adv. Synth. Catal., 2010, 352, 1967–1982.
28. C. Metallinos, J. John, J. Nelson, T. Dudding and L. Belding, Adv. Synth. Catal., 2013, 355, 1211–1219.
29. A. Škvorcová, E. Rakovský, J. Kožíšek and R. Šebesta, J. Organomet. Chem., 2011, 696, 2600–2606.
30. A. Almássy, A. Škvorcová, B. Horváth, F. Bilčík, V. Bariak, E. Rakovský and R. Šebesta, Eur. J. Org. Chem., 2013, 111–116.
31. U. Olsher, R. M. Izatt, J. S. Bradshaw and N. K. Dalley, Chem. Rev., 1991, 91, 137–164.
32. M. C. Whisler, S. MacNeil, V. Snieckus and P. Beak, Angew. Chem., Int. Ed., 2004, 43, 2206–2225.
33. T. Hayashi, Pure Appl. Chem., 1988, 60, 7–12.
34. B. L. Lucht and D. B. Collum, J. Am. Chem. Soc., 1995, 117, 9863–9874.
35. (a) E. R. Johnson, S. Keinan, P. Mori-Sanchez, J. Contreras-Garcia, A. J. Cohen and W. Yang, J. Am. Chem. Soc., 2010, 132, 6498–6506; (b) J. Contreras-Garcia, E. R. Johnson, S. Keinan, R. Chaudret, J.-P. Piquemal, D. N. Beratan and W. Yang, J. Chem. Theor. Comput., 2011, 7, 625–632.
36. O. B. Sutcliffe and M. R. Bryce, Tetrahedron: Asymmetry, 2003, 14, 2297–2325.
37. Y. Miyake, Y. Nishibayashi and S. Uemura, Synlett, 2008, 1747–1758.
38. C. Bolm, K. Muñiz Fernández, A. Seger and G. Raabe, Synlett, 1997, 1052.
39. T. Sammakia and H. A. Latham, J. Org. Chem., 1995, 60, 6002–6003.
40. M. Lotz, K. Polborn and P. Knochel, Angew. Chem., Int. Ed., 2002, 41, 4708–4711.
41. M. N. Cheemala and P. Knochel, Org. Lett., 2007, 9, 3089–3092.
42. G. Argouarch, O. Samuel, O. Riant, J.-C. Daran and H. B. Kagan, Eur. J. Org. Chem., 2000, 2893–2899.
43. S. Cabrera, O. Garcia Mancheno, R. Gomez Arrayas, I. Alonso, P. Mauleon and J. C. Carretero, Pure Appl. Chem., 2006, 78, 257–265.
44. O. Riant, G. Argouarch, D. Guillaneux, O. Samuel and H. B. Kagan, J. Org. Chem., 1998, 63, 3511–3514.
45. Spartan ‘08, Wavefunction, Inc., Irvine, CA.
46. K. Eichkorn, O. Treutler, H. Öhm, M. Häser and R. Ahlrichs, Chem. Phys. Lett., 1995, 242, 652–660.
47. K. Eichkorn, F. Weigend, O. Treutler and R. Ahlrichs, Theor. Chem. Acc., 1997, 97, 119–124.
48. A. D. Becke, Phys. Rev. A, 1988, 38, 3098–3100.
49. F. Weigend, Phys. Chem. Chem. Phys., 2006, 8, 1057–1065.
50. TURBOMOLE V6.4, TURBOMOLE GmbH, Karlsruhe, 2012.
51. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104.
52. P. Deglmann, K. May, F. Furche and R. Ahlrichs, Chem. Phys. Lett., 2004, 384, 103–107.

---

Footnote

† Electronic supplementary information (ESI) available: Energies and Cartesian coordinates of all calculated structures, imaginary frequencies and NCI plots for transition states, table of LDA conformers, and potential energy scan for lithiations of amine (R)-1. See DOI: 10.1039/c3ob41339j

---