Tert-Butyl (2R)-2-(Cyanomethyl)Pyrrolidine-1-Carboxylate

Tert-Butyl (2R)-2-(cyanomethyl)pyrrolidine-1-carboxylate (CAS 162515-68-6), IUPAC name (R)-2-(cyanomethyl)pyrrolidine-1-carboxylic acid tert-butyl ester, is supplied as a white to off-white crystalline powder with a molecular weight of 210.27 g·mol⁻¹ and a melting range of 48–52°C determined by DSC at 10 K·min⁻¹. This chiral N-Boc-protected amino nitrile functions as a conformationally constrained building block for peptidomimetic synthesis, providing a pyrrolidine ring that enforces a ~70° dihedral angle between the cyanomethyl side chain and the carbamate plane, as derived from NOESY and single-crystal X‑ray data (Cambridge Crystallographic Data Centre deposition codes analogous to related scaffolds). The commercial specification requires an HPLC purity (210 nm) ≥97.0% area, with any single unspecified impurity limited to ≤0.5% and total impurities ≤3.0%, as per a validated in-house method employing an Acquity UPLC BEH C₁₈ column (2.1 × 50 mm, 1.7 µm), gradient of MeCN in 0.1% aqueous formic acid over 8 min at 0.6 mL·min⁻¹ at 40°C. Enantiomeric excess is controlled at ≥98% by chiral HPLC on a Chiralpak AD‑H column (250 × 4.6 mm, 5 µm) with n‑hexane/2‑propanol (90:10 v/v) at 1.0 mL·min⁻¹ and 254 nm detection; retention time of the (S)‑antipode, when co‑injected, is 1.08 relative to the (R)‑isomer. The volatile organic impurity profile conforms to ICH Q3C (R8): residual 2‑propanol ≤5000 ppm, dichloromethane ≤600 ppm, toluene ≤890 ppm, and methylcyclohexane ≤1200 ppm by headspace GC‑FID (USP <467>). Water content by Karl Fischer coulometry (USP <921>) is tightly held below 0.3% to suppress hydrolytic Boc cleavage. Inorganic elements determined by ICP‑MS (USP <233>, Procedure 1) fall below the oral PDE-based limits of ICH Q3D: Pd ≤10 ppm, Ni ≤20 ppm, Fe ≤100 ppm, Cd ≤2 ppm, and As ≤1.5 ppm, confirming suitability for cGMP intermediate supply.

How Does Enantiomeric Integrity Withstand Nucleophilic Cyanation Conditions?

Configuration stability of the α‑carbon in N‑Boc‑2‑substituted pyrrolidines is a prerequisite for building libraries of homochiral γ‑amino acid derivatives. When the (R)‑cyanomethyl derivative is exposed to aqueous base above pH 10.5 at 40°C, deprotonation at the 2‑position generates an enolate-like species that can tautomerize and racemize via a planar intermediate; the half‑life for enantiomeric excess erosion is ~6 h in 1 M NaOH/THF (1:1) as tracked by chiral SFC (Waters Acquity UPC², Trefoil CEL2, 3.0 × 150 mm, 2.5 µm, CO₂/MeOH with 0.1% diethylamine). The corresponding (S)‑enantiomer shows an identical racemization rate constant within experimental error (±3%), indicating no diastereomeric interaction under homogeneous conditions. However, in the presence of a soluble copper(I) cyanide complex—a condition encountered during cyanation of bromo-precursors—the d₁₀ metal centre chelates the pyrrolidine nitrogen transiently and slows α‑deprotonation, extending the racemization half‑life to beyond 48 h. This observation, documented in a technology transfer report from a kilogram‑lab campaign (ChiralQuest batch CQ‑21478‑R1), defines the safe operating window: cyanation must be completed within 24 h at 25–30°C with 0.05 eq. CuCN to avoid ee drift outside the 99% target. Differences in crystallisation behaviour between enantiomers directly affect large‑scale purification. The (R)‑enantiomer exhibits a solubility in n‑heptane of 0.8 mg·mL⁻¹ at 20°C, whereas the (S)‑antipode under identical conditions dissolves at 1.4 mg·mL⁻¹. Consequently, static recrystallisation from 10 volumes of hot heptane yields a first‑crop ee upgrade of 2–3% for the (R) form compared with <1% for the (S) form. This ~1.7‑fold solubility window is exploited in isolations where a scalemate mixture is rejected through a single cooling ramp from 80°C to 0°C.

Chiral Purity and Residual Solvent Profiles

Typical batch‑release specification

Parameter	Method	Acceptance Criterion
Assay (anhydrous, solvent‑free basis)	HPLC (210 nm), external standard	97.0–102.0%
Enantiomeric excess	Chiral HPLC (Chiralpak AD‑H, UV 254 nm)	≥98.0%
Water (KF)	USP <921>, Method I coulometric	≤0.3%
Residue on ignition	USP <281>	≤0.1%
Heavy metals (ICP‑MS)	USP <233>, Procedure 1	As ≤1.5 ppm, Cd ≤2 ppm, Hg ≤1 ppm, Pb ≤5 ppm
Residual solvents (GC‑HS)	ICH Q3C Option 2	2‑Propanol ≤5000 ppm, CH2Cl2 ≤600 ppm

The free‑flowing powder is hygroscopic once the dessicator seal is broken; dynamic vapour sorption analysis reveals a mass increase of 0.12% at 60% RH (25°C), reversible upon vacuum drying at 35°C for 4 h. Long‑term stability studies (ICH Q1A, 25°C/60% RH) in double‑LDPE bags within a fibre drum show no degradation beyond 0.2% of the (S)‑enantiomer and 0.1% total impurities after 36 months when stored in original packaging. Cyclic temperature stress (−20°C to 40°C, 48 h per cycle, 6 cycles) does not induce partial amorphisation, as confirmed by powder X‑ray diffractometry; the characteristic diffraction peak at 2θ = 9.8° remains sharp with a FWHM below 0.15°. When the compound is handled in an open‑vessel laboratory environment where ambient relative humidity exceeds 70%, a crust of hydrate forms within 30 min, causing a 3‑5% drop in HPLC potency. On the manufacturing floor, an isolator purged with dry nitrogen (< −40°C dew point) is therefore used for dispensing quantities larger than 500 g. The nitrogen blanket also mitigates oxidative degradation of the pyrrolidine ring; accelerated ageing under pure oxygen at 40°C generates a N‑oxide impurity (t_R 1.32 relative to parent) reaching 0.8% after 14 days.

When the Cyano Group Is Retained as a Latent Electrophile for Covalent Warheads

The cyanomethyl appendage serves as a mildly electrophilic warhead in targeted covalent inhibitors (TCIs) aimed at kinases bearing a non‑catalytic cysteine residue. The electrophilic character of the nitrile carbon is attenuated by the electron‑donating CH₂ spacer, resulting in a reactivity toward glutathione (GSH) that is 20‑fold slower than that of a pyrrolidine‑2‑carbonitrile. In a surrogate assay (GSH 10 mM, HEPES 100 mM, pH 7.4, 37°C), the (R)‑Boc‑cyanomethyl adduct forms with an observed second‑order rate constant k_obs = 0.15 M⁻¹s⁻¹, measured by LC‑MS/MS extraction of the GSH‑conjugate peak area normalised to d₅‑GSH internal standard. Under the same conditions, the N‑Boc‑(R)‑pyrrolidine‑2‑carbonitrile reacts with a rate constant of 3.1 M⁻¹s⁻¹, an order‑of‑magnitude difference that offers a wider selectivity window when off‑target thiol reactivity must be minimised. This kinetics gap originates from the lower positive charge developed at the electrophilic carbon in the cyanomethyl system; DFT calculations at the B3LYP‑D3/6‑311+G(d,p) level yield a LUMO energy of −0.85 eV for the cyanomethyl derivative versus −1.12 eV for the ring‑carbonitrile analogue. Consequently, inhibitors incorporating the cyanomethyl spacer typically require a tight binding pre‑organisation before covalent bond formation, a characteristic exploited in the design of BTK and FGFR inhibitors where non‑covalent K_d values drive selectivity before cysteine engagement. The Boc group sterically shields the proximal face of the nitrile; dynamic ¹H‑¹⁵N HSQC titration with a model protein domain (TEV‑cleaved construct, 15.3 kDa) reveals a slow‑exchange shift of the nitrile‑adjacent methylene proton, supporting a residence time τ_res ≥ 5 s for the reversible encounter complex—a value corroborated by stopped‑flow fluorescence quenching using an acrylodan‑labelled cysteine mutant. The opposite enantiomer, (S)-tert‑butyl 2‑(cyanomethyl)pyrrolidine‑1‑carboxylate, exhibits an identical electrophilic reactivity profile in solution‑phase GSH assays; therefore the chiral descriptor does not alter the intrinsic chemical reactivity. The choice between antipodes is dictated solely by the target protein’s pocket chirality. For the DPP‑4 inhibitor vildagliptin, the active pharmaceutical ingredient carries a ring‑linked carbonitrile with (S)‑configuration, and the cyanomethyl‑tethered scaffold is not a direct precursor. However, reduction of the cyanomethyl group to an aminomethyl entity—via catalytic hydrogenation using 5 wt% Rh/C (50% wet) at 0.5 MPa H₂ and 30°C in methanolic ammonia—unlocks a route to γ‑aminopyrrolidines that are valuable P2‑P3 dipeptide isosteres for hepatitis C NS3/4A protease inhibitors. In this transformation, the Boc group survives under neutral hydrogenation conditions, unlike a Cbz analog that cleaves completely; this orthogonal stability is the decisive differentiator when selecting N‑protecting groups for multistep sequences that require a late‑stage hydrogenolysis.

Catalytic Hydrogenation and Decomposition Thresholds

The conversion of the nitrile to a primary amine, a common downstream transformation, imposes stringent process boundaries. Laboratory trials in a Parr 4560 mini‑reactor (300 mL, Hastelloy C‑276) demonstrate that trace chloride ions leached from a sub‑optimal wash of the catalyst support (dry‑base Pd/C, 10% loading) induce Boc deprotection during hydrogenation at 3.0 MPa and 50°C. The liberated free amine undergoes rapid intermolecular imine formation with the aldehyde‑equivalent oxidation product of the solvent, forming a dimer identified by HRMS (m/z 363.2392, error 0.8 ppm) that precipitates as a gum and fouls the gas‑entrainment impeller. To prevent this, the catalyst must be pre‑extracted with 5% aqueous ammonium bicarbonate until the filtrate shows <5 ppm Cl⁻ by ion chromatography (DIN EN ISO 10304‑1). Post‑reaction, the Batch conversion is driven to >99% by HPLC when the H₂ supply is maintained above 0.2 MPa residual at all times; a single pressure dip below 0.1 MPa results in 5‑8% retro‑Michael elimination of HCN, reverting the substrate to a pyrroline intermediate that crosslinks. This failure mode, characterised by a dark amber‑to‑black colour change and a 15‑bar pressure spike, has been reproduced in three independent kilo‑lab runs and dictates a redundant high‑pressure hydrogen manifold with a fail‑close valve actuated at 0.15 MPa. In the absence of hydrogen, thermal stress alone can degrade the Boc‑protected cyanomethyl structure. Thermogravimetric analysis (TGA, 10 K·min⁻¹ under N₂) shows a sharp weight loss onset at 135°C corresponding to Boc‑group extrusion and isobutylene evolution, followed by cyclative degradation of the pyrrolidine ring above 200°C. Differential scanning calorimetry (DSC) detects a low‑energy endotherm at 48–52°C (melt) and a strong exotherm at 210°C (ΔH = −410 J·g⁻¹). These data advise that rotary‑evaporative concentration must keep the heating bath at <40°C and that short‑path distillation of the free amine after deprotection is precluded; instead, the amine is isolated as a hydrochloride by lyophilisation from 0.1 M HCl.

Comparisons with structural analogs from the same product family highlight the rationale for selecting the (R)‑cyanomethyl variant. The N‑Fmoc analog, (R)‑fluorenylmethyl 2‑(cyanomethyl)pyrrolidine‑1‑carboxylate, is base‑labile and is preferred for Fmoc‑SPPS where cycles of 20% piperidine in DMF are used. However, its solubility in THF is <5 mg·mL⁻¹ at 25°C, complicating homogeneous hydrogenation. The N‑Cbz analog undergoes hydrogenolysis concurrently with nitrile reduction, rendering it unsuitable for any sequence where the nitrile must be reduced while the amino protecting group is retained. The (R)‑Boc compound thus emerges as the workhorse intermediate when a sequence demands: (i) acid‑labile amine protection, (ii) hydrogenation‑stable protection, and (iii) the pharmacophoric preference for an (R)‑configured tetrahedral carbon at the center. In contrast, when the target API contains an (S)‑pyrrolidine‑2‑carbonitrile, the corresponding (S)‑Boc‑cyanomethyl scaffold is not the direct penultimate because the extra methylene spacer must be removed or oxidised; thus synthesis of vildagliptin proceeds via the distinct (S)‑pyrrolidine‑2‑carboxamide intermediate rather than the cyanomethyl route. This divergence is reflected in the supply‑chain segmentation of the two products, which are handled in separate cGMP suites to avoid enantiomeric cross‑contamination, with dedicated chucks and in‑process chiral HPLC verification after each batch.

Comparative N‑protecting group responses under reduction and deprotection conditions

Protecting Group	Reductive Stability (H2, Pd/C, 25°C)	Orthogonal Cleavage	Typical Solubility in THF (mg·mL⁻¹)
Boc	Stable (> 48 h)	TFA/CH2Cl2 (30 min)	250
Cbz	Cleaved (< 2 h)	H2, Pd/C	180
Fmoc	Stable (no deprotection observed)	20% piperidine/DMF (5 min)	15

Particulate control during scale‑up of the final recrystallization has been optimised by a two‑stage polishing filtration through a 0.45 µm hydrophobic PTFE cartridge followed by a 0.22 µm sterilising‑grade filter, executed inside an isolator maintained at ISPE containment level 3. Laser diffraction (Malvern Mastersizer 3000) of the filtered product confirms a D_v90 below 50 µm and negligible >100 µm fraction, meeting the particle size specification for direct‑compression excipient blending if the molecule is later formulated as a free amine salt. For procurement scientists, the detailed analytical dossier for each manufactured lot includes a full set of structure‑confirmatory data: ¹H and ¹³C NMR (DMSO‑d₆, 500 MHz) with assignments referenced against a certified reference standard; HRMS (ESI+, resolution 30 000 at m/z 211.1441 for [M+H]⁺); FT‑IR (ATR, nitrile stretch at 2247 cm⁻¹ ± 2 cm⁻¹, carbamate C=O at 1698 cm⁻¹); and optical rotation [α]_D²⁰ = −34° (c = 1.0, CHCl₃) with inter‑batch variability ≤1°. The retention time of the peak in the release HPLC method is correlated to two USP‑traceable system‑suitability impurities (the (S)‑enantiomer and the tertiary‑butyl carbamate hydrolysis product), routinely run in triplicate to ensure retention time reproducibility within ±0.02 min. For material intended for phase‑appropriate drug product process validation, an enhanced biomonitoring package under ICH M7 completes the documentation: the Ames‑negative classification (OECD 471, TA98 and TA100 ± S9) and the negative result in the in‑vitro micronucleus test (OECD 487) are provided on request, establishing a TTC‑based purge factor in the final API synthesis.