Synthesis of terminal alkynes based on (1 S ,3 R ,4 R )- and (1 S ,3 S ,4 R )-2-azabicyclo[2.2.1]heptane

Two approaches to terminal alkynes based on the enantiomerically pure epimers (1 S ,3 R ,4 R )- and (1 S ,3 S ,4 R )- 2-azabicyclo[2.2.1]heptane aldehydes were established: i) a non-Wittig route through a dichloroalkene intermediate; and, ii) a Corey-Fuchs approach via dibromoalkene. Among various organometallic reagents tested, the use of n -butyllithium was efficient. The resulting alkynes were fully characterized, and one epimer was used in a click chemistry triazole synthesis. For one of the products containing a bulky N -Boc-proline substituent, the existence of atropisomers was observed. The absolute stereochemistry was determined by electronic circular dichroism spectroscopy (ECD) and optical rotation supported by quantum chemical simulations.

Herein, we focus on the preparation of terminal alkynes based on the 2-azanorbornane scaffold using two approaches, non-Wittig and Wittig-type Corey-Fuchs syntheses.The application of the prepared enantiopure compound in a CuAAC click reaction will also be presented.

Results and Discussion
To further explore 2-azanorbornanes in click reactions, we decided to investigate two possible methods of introducing a terminal alkyne function into this chiral scaffold.Epimeric enantiopure aldehydes, (1S,3R,4R)-4 and (1S,3S,4R)-5, were chosen as precursors in these preparations.The two tested strategies rely on the formation of dihaloalkene intermediates which were supposed to undergo elimination with strong bases, such as organolithium compounds.
Firstly, we attempted a non-Wittig approach (Scheme 2), using trichloroacetic acid and sodium trichloroacetate as sources of halogen atoms.The advantage of this route lies in the fact that the reaction is carried out without phosphorous derivatives offering lower toxicity of the reaction mixture and less impurities as compared to Wittig-type reactions. 11,12The reaction of aldehyde 4 gave the dichloroalkene (1S,3R,4R)-6 in a moderate yield (32%) (Scheme 2).In the next step, the dichloroalkene 6 was subjected to elimination conditions.Several organolithium bases and conditions were tried, including the recommended methyllithium or n-butyllithium and temperatures in the range between -78°C and room temperature, as well as different reaction times.However, all attempts failed, the desired alkyne was not formed and the substrate could be recovered.
Attempts to grow crystals suitable for the X-ray measurement of compound 13 were partially successful.From the platelets received, only a model of structure was obtained with a R parameter of 24%.However, the model supported our configuration assignments (Supporting Information, Figure S19).To further confirm the absolute stereochemistry of compound 13 we used chiroptical methods.The absolute configuration was determined by carrying out the experimental study (electronic circular dichroism spectroscopy and optical rotation) supported by quantum chemical simulations.][21][22][23][24][25] In Figure 1 the measured ECD/UV spectra of compound 13 in hexane solution are presented (black line).As can be seen, in the measurement range there are two ECD bands: a minimum at 215 nm, maximum with well-developed fine structure centered at 262 nm, and one shoulder at ca. 227 nm.TD-DFT simulations of ECD spectra were preceded by a thorough conformational analysis at the molecular mechanic level and further re-optimization of the subsequent structures at the DFT level (see experimental part for details).Finally, 6 conformers within the range of 1.5 kcal/mol were identified.The 3 most abundant ones cover more than ca.83% of all conformers in solution equilibrium (Figure 2).They show variations both in conformation of the pyrrolidine ring (Conf.13a vs. Conf.13b) and in arrangement of the triazole moiety (Conf.13a/13b vs. Conf.13c).
As can be seen in Figure 1, the calculated ECD/UV spectra are in good agreement with the experimental ones.However, some incompatibility is visible in the range of 245-275 nm.This is owed to the nature of 1 Lb aromatic band which makes TD-DFT calculations problematic without considering vibronic effects. 26To independently confirm a conformational equilibrium, TD-DFT calculations of optical rotation ([α]D) at the wavelength of 589 nm were carried out for the same set of conformers.Such a complementary approach is considered as a good practice in assigning the absolute configuration, in particular for floppy molecules. 24,27,28As can be seen in Table 1, the simulated [α]D shows the same sign as the experimental data, regardless of the functionals and basis sets used (B3LYP vs. CAM-B3LYP or aug-cc-pVDZ vs. def2-TZVP).
Table 1.Experimental [α]D value for compound 13 recorded in CH2Cl2 (c = 0.23 g/100 cm 3 ) at 298 K and simulated ones Furthermore, we noticed that for conformers 13a and 13b the sign of [α]D is positive (very close to the experimental value and does not depend on the level of theory used), while for 13c its value is always negative (Supplementary Information, Table S1).
Calculations were also performed for the epimer of compound 13 (epi-13) differing by the configuration at C-3 (Supplementary Information, Figure S21, Tables S2-S4).The obtained results showed, however, that the calculated sign of [α]D is opposite to the experimental one (negative), and calculated ECD spectrum exhibits lower factor of enantiomeric similarity index (ESI) determined by SpecDis (=0.845), in respect to compound 13 for which this factor equals 0.875. 29o sum up, such analysis supports an aforementioned configurational assignment, regardless the amplitude of the [α]D.Another important point to note is that in the equilibrium one can distinguish two main atropisomers divided by different arrangement of the substituted triazole group (Figure 2), namely conformers 13a/13b and conformer 13c.The calculated energy barrier at the B3LYP/6-311+G(d,p) level of theory for the transformation from one into another, i.e. by rotating around C3-triazole bond amounts 6.4 kcal/mol.

Conclusions
A route towards terminal alkynes based on the intrinsically chiral 2-azanorbornane scaffold was developed.From the two proposed routes, the one proceeding via dibromoalkenes yielded two epimeric enantiomerically pure terminal alkynes.Alkyne (1S,3R,4R)-6 was successfully applied in two CuAAC click-chemistry reactions.This result, together with the possible use of 2-azabicyloalkane-derived azides (e.g.12) opens the route to a variety of 1,2,3-triazoles based on an intrinsically chiral bicyclic backbone.The results of combined experimental and theoretical analysis of chiroptical data of compound 13 allowed the effective assignment of its three-dimensional structure.

Experimental Section
General.Preparative methods.For the chromatographic separation silica gel 60 (70-230 mesh) was used, and thin layer chromatography was carried out on silica gel 60 precoated plates.Analytical measurements.The NMR spectra were recorded on Jeol 400yh and Bruker Avance II 600 instruments.The measurements were conducted at 298 K.The spectra were calibrated using residual solvent signals.Infrared spectra in the range of 500 -4000 cm -1 were recorded using a Perkin Elmer 2000 FTIR spectrophotometer.High resolution mass spectra were collected using a Waters LCT Premier XE TOF instrument with electrospray ionization.Melting points were determined on the Schmelzpunkt Bestimmer Apotec apparatus using the standard open capillary.The optical rotations were measured using an Optical Activity Ltd.Model AA-5 automatic polarimeter; [α]D values are given in 10 -1 deg cm 2 g -1 .ECD spectra measurement was carried out using a Jasco J-815 spectrometer (Tokyo, Japan) at room temperature using spectroscopic grade hexane in quartz cells (0.0033 mol/dm 3 ) with a path length of 0.1 and 0.02 cm.ECD spectra were measured using a scanning speed of 100 nm/min, a step size of 0.2 nm, a band-width of 1 nm, a response time of 0.5 seconds, and an accumulation of 5 scans.The spectra were background-corrected using solvent recorded under the same conditions.CCDC-2020530 contains the supplementary crystallographic data for this paper.These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, by e-mailing data_request@ccdc.cam.ac.uk, or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44(0)1223-336033.Aldehydes 4 and 5 were synthesized using the formerly described method. 10Azide 12 was prepared as described previously. 16

Figure 1 .
Figure 1.Experimental ECD (left) and UV (right) spectra of compound 13 recorded in hexane at room temperature (black lines) confronted with TD-DFT simulations performed at the CAM-B3LYP/def2-TZVP level of theory (red lines).Enantiomeric Similarity Index (ESI) for calculated curve equals 0.875.

Figure 2 .
Figure 2. Structures of most abundant conformers of compound 13 (conf.13a-13c) calculated at the B97X-D/6-311+G(d,p) level of theory with their relative energy (given in kcal/mol) and populations in equlibrium calculated at 298 K. Hydrogen atoms are omitted for clarity.
Computational Details.Conformational analysis and TD-DFT simulations.