*2.3. Determination of the Absolute Configuration of Nitroalkanes (***4***)*

The absolute configuration of the reduction product, the nitroalkane (**4**), was experimentally established by converting **4a** into a known compound. To achieve this goal, two approaches were attempted (Scheme 4).

**Scheme 4.** Strategies for the experimental determination of the absolute configuration of nitroalkane **4a**.

The first strategy (Scheme 4, method A) explored the transformation of **4a** into the known compound **7**, by a three-step procedure involving the alkylation of nitroalkane **4a**, followed by the reduction of the nitro group and hydrolysis of the ester moiety as depicted in Scheme 4 [22–24]. However, despite several conditions being attempted, the alkylation did not lead to the desired product, but mixtures of unreacted starting material and Oalkylated products were obtained (In an additional experimental effort, we have observed that nitroalkane **4a** reacted with "softer" electrophiles, such as methyl vinyl ketone in a Michael reaction, additional studies are underway to further explore these transformations).

Thus, our efforts were focused on an alternative approach (Scheme 4, Method B). The decarboxylation of the ester moiety of **4a** (as 50:50 mixture of *syn/anti* isomers, enantioenriched) to afford the corresponding trisubstituted nitroalkane **8** was then explored [25]. This simple approach was found to be effective and the desired decarboxylated nitroalkane **8** was finally synthetized. The experimental optical rotation was measured using a polarimeter and by comparison with the literature data, the compound **8**, derived from the major enantiomer of product **4a** (*4a syn A* and *4a anti A*)**,** was established to have the *R*-configuration [26].

DFT computational studies on the enantioselective reduction of tetrasubstituted nitroalkanes were preliminarily performed to rationalize the stereochemical outcome of the reaction, using the Gaussian g16 package using Catalyst **A** as the model catalyst. All

geometries of reactants and products (ground states and transition states) were located at a B3LYP/6-31G (d,p) level of theory and finer electronic energies were successively obtained, increasing the basis set up to 6/311 + (2df,2pd) with B3LYP functional [27]. In Figure 3 four possible complexes of nitroalkene **3a** with catalyst **A** and the geometries of the TS leading to the formation of the four stereoisomers of nitroalkane **4a** are represented.

```
( &220H
```
*%/<3 JGISG %/<3 JGS OHYHO RI WKHRU\*

**Figure 3.** The four possible TS geometries for nitroalkene **3a** complexes with catalyst **A**.

Transition states responsible for the hydride transfer were located assuming the coordination of the nitro group of the nitroalkane **3a** to the thiourea moiety and of the Hantzsch ester NH group with the catalyst carboxyamide group, according to the so-called Takemoto model. The energy profile is depicted in Figure 4.

*%/<3JGISG%/<3 JGSOHYHORIWKHRU\*

**Figure 4.** DFT calculations performed for the enantioselective reduction of tetrasubstituted nitroal-kene **3a**.

In Figure 5 the geometries of the transition states originated by the *E*-isomer of compound **3a** are illustrated.

**Figure 5.** Transition states formed by the *E*- isomer of nitroacrylate **3a**.

The blue spheres represent the nitrogen atom of the nitro group of nitroalkane **3a** and of the thiourea catalyst **A**, the yellow one represents the sulfur atom of the thiourea moiety, and the red spheres represent the oxygen atoms of nitro and ester groups of compound **3a,** and of the carboxyamide group of the catalyst. The pink sphere represents the *tert*-butyl groups of the Hantzsch ester, carbon atoms are grey and hydrogen atoms are white. The broken lines showed the H-bonding interaction between the nitro group of **3a** and the thiourea moiety of the catalyst, and the transfer hydride between Hantzsch ester and carbon C3 of the nitroolefin.

According to the calculations, among the two transition states originating from *Z*-olefin, TS-*Z*-(*3R*) is the lowest in energy and would lead to the formation of the final product with a *R-*configuration at the C3 carbon, in agreement with the experimental data. However, the (*Z*) isomer was found to be very poorly reactive, while, as established in NMR analysis, the more reactive isomer is the (*E*)-nitroacrylate, that, according to the calculations should preferably afford the (S) enantiomer at C3 carbon of compound **4a**. These findings are in contrast with the experimentally established absolute configuration (*R*) for the major enantiomer derived from the reaction of the E isomer of **3a**. Therefore, we can conclude that, at the moment, the proposed TS according to the Takemoto model, is not able to explain why the *E* isomer should be more reactive than the *Z* isomer, and, furthermore, cannot predict the correct configuration at the C3 of the nitroalkane. Those results are probably an indication that other coordination modes are active in the TS of the reactions, and other models need to be taken into consideration to rationalize the stereochemical outcome of the reaction.
