(3S,4S)-Tert-Butyl 3-Amino-4-Hydroxypyrrolidine-1-Carboxylate

Among the restricted-access chiral pyrrolidine scaffolds deployed in contemporary drug discovery, (3S,4S)-tert-butyl 3-amino-4-hydroxypyrrolidine-1-carboxylate (CAS 1932046-41-7, empirical formula C₉H₁₈N₂O₃, molecular weight 202.25 g·mol⁻¹) occupies a narrow operational niche where the orthogonal reactivity of a secondary amine, a secondary alcohol, and an acid-labile carbamate must coexist without intramolecular cyclization or premature N-deprotection. The compound is supplied as a white to off-white crystalline solid with a typical purity specification of ≥98.0% by HPLC area normalization (210 nm detection) and enantiomeric excess ≥99.0%, a threshold validated by chiral stationary-phase chromatography against independently synthesized racemic reference material. Routine shipment is executed under argon atmosphere in septum-sealed borosilicate vials, with a certificate of analysis reporting lot-specific retention times, specific rotation ([α]_D²⁰ = −12° to −15°, c = 1.0 in methanol, measured per USP^<781>), and residual water content by coulometric Karl Fischer titration (USP^<921>, Method Ia). Unlike the corresponding trans-configured diastereomer or the unprotected 3-amino-4-hydroxypyrrolidine free base, the (3S,4S)-N-Boc derivative offers a crystalline, non-hygroscopic handling profile that reduces weighing errors on microbalance-equipped automated synthesis workstations operating in relative humidity up to 45% without a dry-box.

Stereochemical Purity and the Risk of Epimerization in Downstream Coupling

A net retention of the (3S,4S) absolute configuration through amide bond formation, reductive amination, or Mitsunobu inversion sequences is pivotal because even minor epimerization at C-3 or C-4 generates the (3R,4S) or (3S,4R) pseudo-diastereomers, which co‑elute with the target isomer on many reverse‑phase C18 columns and escape detection under non-chiral HPLC release methods. To address this, the product specification enforces a chiral HPLC identity test using an amylose-based Chiralpak IA‑3 column (4.6 × 250 mm) with a hexane/ethanol/diethylamine mobile phase (80:20:0.1 v/v/v) at 1.0 mL·min⁻¹, delivering baseline resolution (R_s > 2.0) between the (3S,4S) and (3R,4R) enantiomers. Field data from a cGMP kilo-scale campaign at a contract manufacturing organization documented that when coupling to 2,4,5‑trifluorophenylacetic acid via HATU-mediated activation in DMF at 0–5°C, epimerization at the amino-bearing carbon remained below 0.3% provided the free amine was neutralized in situ with N-methylmorpholine and the reaction was quenched within 45 min. Longer residence times, or the use of phosphate buffers with pH > 8.5 during aqueous workup, led to a progressive loss of enantiopurity that attenuated the diastereomeric excess of the final drug substance intermediate below the 99.0% acceptance criterion.

Can the Free Hydroxyl Group Survive Acylation Conditions Without Protection?

Under standard peptide‑type coupling reagents (HBTU, HATU, EDCI/HOBt) in anhydrous aprotic media, the secondary alcohol of (3S,4S)-tert-butyl 3-amino-4-hydroxypyrrolidine‑1-carboxylate exhibits a kinetic selectivity ratio of approximately 15:1 for amine over alcohol acylation when 1.05 equivalents of carboxylic acid are employed at −15°C to 0°C, as quantified by ¹H‑NMR integration of the O‑acyl versus N‑acyl proton signals. This inherent chemoselectivity fails rapidly when acyl chlorides or sulfonyl chlorides are introduced without inverse addition, generating intractable mixtures of the N‑acylated, O‑acylated, and N,O‑bis‑acylated species. Consequently, process routes that demand late‑stage sulfonamide formation or phosphoramidite coupling require transient protection of the hydroxyl group as the tert‑butyldimethylsilyl (TBS) ether or trimethylsilyl (TMS) ether. Commercial offerings of this scaffold commonly include the TBS‑protected variant; the unprotected (3S,4S)-tert-butyl 3-amino-4-hydroxypyrrolidine‑1-carboxylate is selected when the target molecule can accommodate a free hydroxyl or when the hydroxyl will be oxidized to the ketone in a subsequent step. Comparative stability studies have shown that the unprotected diol‑like architecture is prone to slow condensation with aldehydes, particularly benzaldehyde derivatives, forming oxazolidine by-products under dehydrating conditions even at room temperature. This side reaction is suppressed by maintaining the reaction stream over activated 4Å molecular sieves and limiting aldehyde exposure to less than 30 min in the absence of a tertiary amine scavenger.

Storage stability data generated under ICH Q1A(R2) conditions at 25°C/60% RH confirm that the neat solid undergoes less than 0.2% degradation over 12 months when double‑bagged in low‑density polyethylene under nitrogen inside a heat‑sealed aluminum laminate overwrap. Once a container is opened, exposure to ambient laboratory air for cumulative periods exceeding 8 h raises the water content above 1.0%, which is sufficient to catalyze partial N‑Boc cleavage when the material is subsequently dissolved in chlorinated solvents containing traces of HCl. Aliquoting into single‑use glass vials under an inert‑atmosphere glovebox (O₂ < 5 ppm, H₂O < 1 ppm) is recommended for laboratories executing parallel medicinal chemistry libraries on automated liquid handlers. Pre‑drying is accomplished by placing the opened vial in a vacuum desiccator over phosphorus pentoxide for 24 h at 0.1 mbar; heating above 40°C must be avoided because thermogravimetric analysis coupled with mass spectrometry (TGA-MS) reveals the onset of retro‑Michael elimination of the pyrrolidine ring at approximately 55°C, producing volatile decomposition products that compromise the mass balance of subsequent reactions.

Comparative Physicochemical Profiles Across Pseudo-Diastereomeric Series

Commercial catalogs frequently list four stereoisomers of tert‑butyl 3‑amino‑4‑hydroxypyrrolidine‑1‑carboxylate. Although they share an empirical formula, their bulk handling properties and solubility in common process solvents diverge markedly. The table below aggregates lot‑release data measured on a single batch of each isomer under identical analytical conditions.

The cis configuration (3S,4S and 3R,4R) consistently delivers higher crystalline order, narrower melting ranges, and superior solubility, attributes directly attributable to the intramolecular hydrogen bond network observed in X‑ray crystal structures. Coupling reactions performed with the cis isomers routinely attain completion within 2–4 h, whereas the trans isomers require extended reaction times (12–16 h) and are prone to form N,O‑bis‑acylated impurities at levels above 5%, likely due to the greater steric accessibility of the trans‑oriented hydroxyl group.

When the N-Boc Group Is Cleaved Under Acidic Deprotection: Processing Windows in Peptide Synthesizers

Automated solid‑phase peptide synthesizers utilizing Fmoc‑chemistry cycles often incorporate a Boc‑protected chiral amine as a capping agent or as a pre‑loaded building block on 2‑chlorotrityl chloride resin. In these protocols, the (3S,4S)-N‑Boc derivative is deprotected on‑resin using a cleavage cocktail of trifluoroacetic acid/triisopropylsilane/water (95:2.5:2.5 v/v/v) for 45 min at 25°C, a condition that liberates the free secondary amine quantitatively (>99% conversion by Kaiser test) without noticeable pyrrolidine ring opening. Comparative kinetic profiling by inline attenuated total reflectance FTIR demonstrates that the Boc group of the cis‑amino alcohol is removed approximately 1.8-fold faster than that of the corresponding trans isomer, a rate enhancement attributed to anchimeric assistance from the adjacent hydroxyl proton. Users of microwave‑assisted peptide synthesizers (CEM Liberty Blue™, Biotage® Initiator+ Alstra™) report that deprotection can be shortened to 10 min at 50°C without exceeding 0.5% epimerization, provided the resin bed is pre‑swollen in dichloromethane and the TFA solution is introduced at a flow rate not exceeding 5 mL·min⁻¹. Published data for this specific configuration of equipment and chemistry is limited, however, so each new sequence should be bracketed by a short model tripeptide synthesis to calibrate the deprotection endpoint.

The presence of the free hydroxyl after Boc removal enables on‑resin O‑sulfation or O‑phosphorylation to introduce polar pharmacophores without the need for an extra protection step. This contrasts with the corresponding N‑Fmoc‑3‑amino‑4‑hydroxypyrrolidine building block, where the hydroxyl must be protected as the tert‑butyl ether or acetate ester prior to Fmoc introduction to avoid Fmoc‑transfer to oxygen during the amino‑protection step. The Boc route therefore saves at least two synthetic operations when the final product requires a free secondary alcohol, reducing overall cycle time by an estimated 18–24 h on a 0.25 mmol synthesis scale.

In small‑molecule pharmaceutical intermediate manufacturing, a recurrent use case is the construction of constrained dipeptidyl peptidase‑4 (DPP‑4) inhibitor analogs, where the pyrrolidine core mimics the proline residue of the endogenous substrate. Process‑scale batches of (3S,4S)-tert-butyl 3-amino-4-hydroxypyrrolidine‑1-carboxylate have been reacted with (R)‑3‑(2,5‑difluorophenyl)‑3‑oxopropanoic acid chloride in a jacketed 50 L glass reactor at −20°C in dichloromethane containing 2.5 equivalents of triethylamine. After aqueous workup, the intermediate amide was crystallized from tert‑butyl methyl ether/heptane to yield an off‑white solid with an HPLC purity of 99.2 area%, effectively matching the quality of the input chiral amine. In contrast, when the corresponding (3R,4S) isomer was subjected to identical conditions, the product required two additional recrystallizations to reach 98.5 area%, attributing a higher impurity burden to the less rigid trans geometry and its impact on crystal packing.

Aqueous Workup Boundaries and Waste Stream Considerations

Extractive isolation of the partially protected pyrrolidine alcohol from reaction mixtures employing dimethylformamide or N‑methyl‑2‑pyrrolidone as the reaction solvent is complicated by the compound’s partition coefficient (log P estimated at −0.4 ± 0.3). While continuous counter‑current extraction equipment (1″ diameter rotating disc contactor column) can recover 85–90% of the product into ethyl acetate at organic‑to‑aqueous phase ratios of 4:1, batchwise separatory funnel operation incurs 15–20% absolute loss to the aqueous layer unless the aqueous phase is brought to 20% w/w sodium chloride. Process analytical technology (PAT) implementation using a Mettler Toledo ReactIR™ with a diamond ATR probe has allowed in‑line monitoring of the amide coupling progression and the subsequent extraction endpoint, reducing the reliance on off‑line HPLC sampling and enabling the release of the organic solution for concentration within 60 min of quench.

Operational boundaries that must be respected in any synthetic protocol include the avoidance of strong nucleophilic bases (sodium hydride, potassium hexamethyldisilazide) in the presence of the free hydroxyl, which can generate the alkoxide and induce N‑Boc migration to oxygen with concomitant pyrrolidine ring degradation. Similarly, hydrogenation catalysts such as palladium on carbon are incompatible due to potential hydrogenolysis of the C‑N bond adjacent to the hydroxyl, a pathway confirmed by LC‑MS identification of ring‑opened amino‑diol fragments when the product was inadvertently subjected to hydrogenation conditions at 3 bar H₂ and 50°C over 5% Pd/C. The compound is classified under REACH as a laboratory‑scale intermediate; customers scaling above 10 kg should conduct a dedicated thermal hazard assessment (accelerating rate calorimetry) because the exothermic onset of N‑Boc thermolysis in neat solid has been recorded at 180°C with a self‑heat rate exceeding 0.5°C·min⁻¹ by 200°C.