Open-Source Chromatographic Data Analysis for Reaction Optimization and Screening

Abstract

Automation and digitalization solutions in the field of small molecule synthesis face new challenges for chemical reaction analysis, especially in the field of high-performance liquid chromatography (HPLC). Chromatographic data remains locked in vendors' hardware and software components, limiting their potential in automated workflows and data science applications. In this work, we present an open-source Python project called MOCCA for the analysis of HPLC-DAD (photodiode array detector) raw data. MOCCA provides a comprehensive set of data analysis features, including an automated peak deconvolution routine of known signals, even if overlapped with signals of unexpected impurities or side products. We highlight the broad applicability of MOCCA in four studies: (i) a simulation study to validate MOCCA's data analysis features; (ii) a reaction kinetics study on a Knoevenagel condensation reaction demonstrating MOCCA's peak deconvolution feature; (iii) a closed-loop optimization study for the alkylation of 2-pyridone without human control during data analysis; (iv) a well plate screening of categorical reaction parameters for a novel palladium-catalyzed cyanation of aryl halides employing O-protected cyanohydrins. By publishing MOCCA as a Python package with this work, we envision an open-source community project for chromatographic data analysis with the potential of further advancing its scope and capabilities.

Figure 1

Proposed analytical workflow starting with HPLC systems controlled by vendor-specific software. HPLC–DAD raw data are exported in nonproprietary and open data formats, preferably, a metadata-enriched standardized format (Allotrope) implementing FAIR data principles. After parsing in Python, HPLC–DAD data sets are analyzed in context to each other by MOCCA. From the analysis results, structured data sets are generated for data-based decision making.

Figure 2

Summary of the data analysis features implemented in MOCCA.

Scheme 1. Knoevenagel Condensation Reactions of Benzaldehyde (1a), 4-Methoxybenzaldehyde (1b) and 4-(Dimethylamino)benzaldehyde (1c) with Malononitrile (2) in Methanol (MeOH) to Yield Benzylidenemalononitriles 3a–c

Figure 3

(a) Results of the competition experiment with two benzaldehydes (1a and 1b). (b) Results of the competition experiment with three benzaldehydes (1a–c). Top: Chromatographic signals of the benzaldehydes using different gradient lengths. MOCCA indicates results of purity checks (green passed, red failed) and centers of retention profiles modeled by the deconvolution algorithm (vertical black dashed lines). Bottom: Deconvolution results of the overlapping signal recorded with a gradient length of 0.5 min. The modeled retention profiles (left, colored lines) described the retention profile of the impure peak (black dashed line). The modeled UV–Vis traces (right, colored lines) correspond to the UV–Vis spectra of the benzaldehydes as exemplified for 1a (black dashed line).

Figure 4

Second-order reaction kinetics plots of benzaldehyde (1a) in the Knoevenagel condensation recorded with five different HPLC methods employing varying gradient lengths. (a) Competition experiment with two benzaldehydes (1a and 1b). (b) Competition experiment with with three benzaldehydes (1a–c).

Figure 5

Closed-loop optimization cycle employed in this work. Blue: Experimental Design via Bayesian Optimization (EDBO) Python package from the Doyle group and translation of the suggested parameters to a LabVIEW experimental protocol. Yellow: Experimental execution by a microfluidic reactor platform employing an oscillatory droplet reactor design. 0.02 μL HPLC samples are taken out of the droplet after diltution with acetonitrile. Green: HPLC system with a photodiode array detector (DAD) and an automated HPLC–DAD raw data export routine. Red: Data analysis by the MOCCA tool and a project-specific script for the extraction of objective values and process control values.

Scheme 2. Optimization Campaign on the Alkylation of 2-Pyridone (4) with 1-Iodobutane (5) Yielding 1-Butylpyridone (6)

The domain space of the optimization campaign spans over two continuous variables, reaction time (low boundary: 10 min, high boundary: 60 min), and temperature (low boundary: 35 °C, high boundary: 100 °C), as well as two categorical variables, identity of base (DBU, TMG, DIPEA) and solvent (DMF, toluene, n-butanol). The objective value of the optimization is the yield of 6. Two side products were identified with 2-butoxypyridine (7) and butylated DBU (8).

Figure 6

Results of the closed-loop optimization on the alkylation of 2-pyridone (4). (a) Objective values as a function of optimizer choices in each round. Top: Objective value (yield of 6) with marker shape indicating the chosen solvent and marker color indicating the chosen base; middle: Chosen reaction temperature; bottom: Chosen reaction time. (b) Chromatogram of the reaction under optimal conditions with an impure product peak (∼1.7 min). (c) Modeled retention profiles (dashed line: impure peak) and UV–Vis spectra (dashed line: reference UV–Vis spectrum of (6) of the product 6 (yellow) and the unexpected impurity 8 (blue).

Scheme 3. (a) Palladium-Catalyzed Cyanation of Aryl Chlorides Developed by Guimond et al. Based on the Slow Addition of Acetone Cyanohydrin via Syringe Pump. (b) Newly Developed Cyanation Method Using Protected Cyanohydrins (PG: protecting group) for in Situ Release of Cyanide

Figure 7

(a) Reaction conditions for the well plate screening of the cyanation of 2-chlorotoluene (9) yielding o-tolunitrile (11) using palladium(π-cinnamyl) chloride dimer as the catalyst precursor. (b) Screened O-protected cyanohydrins. (c) Screened bases. (d) Screened ligands. (e) Yield of o-tolunitrile (11) in dependency on the employed protected cyanide-releasing agent 10 as well as the chosen ligand and base. CPME: cyclopentyl methyl ether; Bz: benzoyl group; Ac: acetyl group; TMS: trimethylsilyl group.