Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Nov 6;95(46):16801–16809. doi: 10.1021/acs.analchem.3c01613

BINOL as a Chiral Solvating Agent for Sulfiniminoboronic Acids

Robin R Groleau †, Robert S L Chapman †, John P Lowe †, Catherine L Lyall †, Gabriele Kociok-Köhn †, Tony D James †,‡, Steven D Bull †,§,*
PMCID: PMC10666087  PMID: 37931004

Abstract

graphic file with name ac3c01613_0007.jpg

1H NMR spectroscopic studies using BINOL as a chiral solvating agent (CSA) for a scalemic sulfiniminoboronic acid (SIBA) have revealed concentration- and enantiopurity-dependent variations in the chemical shifts of diagnostic imine protons used to determine enantiopurity levels. 11B/15N NMR spectroscopic studies and X-ray structural investigations revealed that unlike other iminoboronate species, BINOL–SIBA assemblies do not contain N–B coordination bonds, with 1H NMR NOESY experiments indicating that intermolecular H-bonding networks between BINOL and the SIBA analyte are responsible for these variations. These effects can lead to diastereomeric signal overlap at certain er values that could potentially lead to enantiopurity/configuration misassignments. Consequently, it is recommended that hydrogen-bonding-CSA-based 1H NMR protocols should be repeated using both CSA enantiomers to ensure that any concentration- and/or er-dependent variations in diagnostic chemical shifts are accounted for when determining the enantiopurity of a scalemic analyte.

We have previously reported versatile three-component chiral derivatization protocols for determining enantiomeric excess (ee) of chiral amines, amino esters, diols, amino alcohols, diamines, hydroxyacids, diacids, and hydroxylamines by 1H/19F NMR spectroscopic analysis.1−12 In a representative protocol, treatment of a scalemic amine (e.g., α-methylbenzylamine 1) with achiral 2-formylphenyl boronic acid (2-FPBA) and an enantiopure diol (e.g., BINOL) produces a mixture of diastereomeric iminoboronate esters (IBEs, e.g., 2a and 2b), whose diastereomeric ratio (dr) can then be determined by integration of baseline-resolved 1H NMR imine resonances. Since no kinetic resolution occurs in this derivatization process, these dr values are an accurate reflection of the enantiomeric ratio (er) of the parent scalemic amine analyte (Figure 1a).1,13,14 This widely used three-component derivatization method is often referred to as the Bull-James assembly,1 with this approach also having been used to determine er using circular dichroism-, fluorescence-, and electrochemical-based approaches.15−17 The excellent yields of IBEs produced in these self-assembly reactions have also been exploited to conduct efficient bioconjugation reactions and functionalize advanced polymeric materials.1,15−18

Figure 1.

Figure 1

Open in a new tab

(a) Three-component complexation of scalemic amine (α-R)-1 (90:10 er), 2-FPBA and enantiopure (R)-BINOL produces diastereomeric IBEs (α-R,R)-2a and (α-S,R)-2b, whose drs are determined by 1H NMR spectroscopic analysis.13 (b) Stepwise three-component complexation of scalemic sulfinamide (RS)-3 (90:10 er), 2-FPBA, and enantiopure pinanediol produces diastereomeric SIBEs (RS,R)-5a and (SS,R)-5b, whose drs can be measured by 1H/19F NMR spectroscopic analysis.19 (c) Two-component reaction of scalemic Ellman’s sulfinamide (SS)-3 (90:10 er) with 2-FPBA produces scalemic SIBA 4 whose enantiomers coordinate to (R)-BINOL to produce noncovalent (R)-BINOL·(SS)-SIBA 4 and (R)-BINOL·(RS)-SIBA 4 complexes.

We recently reported a modified version of this three-component CDA method for determining the enantiopurities of S-chiral sulfinamides using 1H/19F NMR spectroscopy.19 In this case, high-yielding formation of SulfinIminoBoronate Esters (SIBEs) required development of a modified two-step “one-pot” protocol to overcome the lower nucleophilicity of the sulfinamide amino group. This new method involved prior reaction of a scalemic sulfinamide (e.g., Ellman’s sulfinamide 3) with 2-FPBA in CDCl3 to generate a scalemic SulfinIminoBoronic Acid (SIBA, e.g., 4) that was then reacted with enantiopure pinanediol to afford a mixture of SIBE diastereomers (e.g., 5a and 5b) (Figure 1b). These SIBE diastereomers display well-resolved 1H NMR imine resonances, the integral ratios of which could be used to accurately determine dr values. Substituting 3-fluoro-2-FPBA for 2-FPBA in this complexation reaction allows the use of both 1H and 19F NMR spectroscopic analysis to accurately determine SIBE dr values using two independent measurements (Figure 1b).19

Optimization studies aimed at identifying the best chiral reporter diol to derivatize (RS)-SIBA 4 produced pairs of SIBE diastereomers that generally displayed well-resolved imine peaks in their 1H NMR spectra (ΔδH = 0.010–0.085 ppm, Table S1). Pinanediol gave the best result, producing tert-butyl SIBE diastereomers 5a and 5b that displayed sharp 1H NMR imine peaks with a large ΔδH value of 0.085 ppm. This led to pinanediol being chosen as the optimal chiral diol to derivatize other chiral sulfinamide analytes, whose SIBE diastereomers all displayed well-resolved 1H NMR imine peaks. (Table S2).19 However, derivatization of SIBA (RS)-4 with (R)-BINOL gave inferior results, producing 1H NMR spectra exhibiting broader partially overlapped imine resonances. This result was surprising since previous three-component complexation reactions of (rac)-amines, 2-FPBA, and BINOL had produced diastereomeric IBEs displaying well-resolved 1H NMR imine resonances (e.g., ΔδH = 0.17 ppm for 2a/2b).13

These anomalous results prompted us to investigate the SIBA-BINOL complexation reaction further, with this study now reporting that reaction of SIBA 4 with BINOL does not in fact produce SIBE complexes like other diols. Instead, conformationally demanding BINOL (restricted rotation around biaryl bond) acts as a chiral solvating agent (CSA) to generate hydrogen-bonded BINOL–SIBA assemblies that display less well-resolved 1H NMR imine resonances (Figure 1c). Furthermore, investigations into the use of BINOL as a CSA for determining the enantiopurities of scalemic SIBA 4 samples revealed that intermolecular hydrogen-bonding effects produce concentration- and er-dependent variations in the diagnostic 1H NMR imine chemical shift values used to determine er values. These chemical shift variations can result in partially (or fully) overlapped 1H NMR imine resonances at certain analyte concentration and er levels that prevent er from being determined. This signal overlap issue can be resolved by repeating the process the opposite enantiomer of BINOL for derivatization to produce a new 1H NMR spectrum that displays fully resolved imine resonances.

The concentration- and er-dependent variations in diagnostic 1H NMR chemical shifts described in this study could potentially occur in other hydrogen-bonding-CSA-based 1H NMR protocols used to determine er values of other types of analyte. Therefore, it is recommended that duplicate 1H NMR spectra of scalemic analytes complexed to both CSA enantiomers should be acquired to determine whether er-dependent variations in diagnostic 1H NMR chemical shift values need to be accounted for when determining er values.

Experimental Section

General Experimental Details

Reagents and solvents were obtained from commercial suppliers and used without further purification. Reactions were performed without air exclusion or drying, at room temperature, and with magnetic stirring, unless otherwise stated. Anhydrous MgSO4 or Na2SO4 was used as a drying agent for organic solvents. Thin-layer chromatography (TLC) was carried out on Macherey–Nagel aluminum-backed plates that were precoated with silica. Compounds were visualized by either quenching of ultraviolet (UV) fluorescence at 254 nm or by dip-staining (KMnO4, PMA, curcumin,20 I2) followed by gentle heating. Purification by flash column chromatography was performed using high-purity-grade silica gel (60 Å pore size, 40–75 μm particle size). PE refers to petroleum ether 40–60 °C. Capillary melting points were determined using a Stuart digital SMP10 melting point apparatus and are reported uncorrected to the nearest °C. Optical rotations were measured using an Optical Activity Ltd. AA-10 Series Automatic Polarimeter, with a path length of 1 dm and with concentration (c) quoted in g/100 mL. Nuclear Magnetic Resonance (NMR) spectroscopy experiments were performed in deuterated solvents at 298 K (unless otherwise stated) on either a Bruker 500 MHz spectrometer or an Agilent ProPulse 500 MHz spectrometer. 1H, 13C, 11B, and 19F NMR chemical shifts (δ) are quoted in parts per million (ppm) and are referenced to either the residual solvent peak or tetramethylsilane (TMS) when possible.2111B NMR spectra were referenced directly, using the lock signal, to external BF3.Et2O (0 ppm). Coupling constants (J) are given in Hz. In those cases where 13C signals could not be observed by one-dimensional (1D) NMR spectroscopy due to low solubility, adjacent quadrupolar nuclei, or lack of adjacent 1H nuclei, then chemical shifts were measured indirectly from two-dimensional (2D) 1H–13C HMBC experiments.1915N NMR chemical shifts were measured indirectly from 2D 1H–15N HMBC spectra using a Bruker Avance 500 MHz spectrometer.2215N NMR experiments were carried out at 50 mM sample concentrations. 15N NMR spectroscopy was carried out in CDCl3 containing 50 mM nitromethane internal standard. The MeNO215N resonance was used as the reference for 15N chemical shifts, setting δN(MeNO2) = 0.00 ppm.23 Infrared (IR) spectra were recorded using a PerkinElmer Spectrum 100 Fourier-transform infrared (FTIR) spectrometer fitted with a Universal ATR FTIR accessory, with samples run neat and selected absorbances quoted as ν in cm–1. High-resolution mass spectrometry (HRMS) results were acquired on an externally calibrated Bruker Daltonics maXis HD ultrahigh-resolution-time of flight (UHR-TOF) mass spectrometer coupled to an electrospray source (ESI-TOF) or an Agilent QTOF 6545 with Jetstream ESI. In most cases, molecular ions were detected either in positive mode as their protonated, sodiated, or ammonium adduct forms, or in negative mode as deprotonated or acetate adduct species.

Additional details of materials, reagents, and instrumentation are provided in the Supporting Information. Additionally, full synthetic procedures for the preparation of all compounds reported herein, along with full characterization data including NMR spectra, can also be found in the Supporting Information.

Three-Component Derivatization of α-Methylbenzylamine 1 with 2-FPBA and (R)-BINOL

Enantiopure (R)- or (S)-α-methylbenzylamine 1 in CDCl3 (1.0 mL, 0.10 M containing ∼6 mM TMS internal standard) was added to 2-FPBA (15 mg, 0.10 mmol, 1.0 equiv) and (R)-BINOL (31.5 mg, 0.11 mmol, 1.1 equiv). The reaction mixture was stirred for 10 min at rt before an aliquot (650 μL) was removed, and NMR spectra of the resultant iminoboronate esters 2a/2b were acquired.

“One-Pot” Stepwise Three-Component Assembly of Sulfinamide 3, 2-FPBA, and Pinanediol

2-FPBA (0.12 mmol, 1.2 equiv) and anhydrous MgSO4 (200 mg) were added to a stirred solution of Ellman’s sulfinamide 3 (0.1 mmol, 1.0 equiv) in CDCl3 (1.0 mL, ∼6 mM TMS internal standard). The reaction mixture was stirred for 1 h at rt, and pinanediol (22 mg, 0.13 mmol, 1.3 equiv) was then added. The reaction was stirred for a further 10 min before being filtered, and the NMR spectra of a 650 μL aliquot were then acquired.

Three-Component Stepwise Derivatization of tert-Butylsulfinamide 3 with 2-FPBA and BINOL

Ellman’s sulfinamide 3 (1.0 mL, 0.1 M in CDCl3 with ∼6 mM TMS) of known enantiopurity was added to a mixture of 2-FPBA (15 mg, 0.10 mmol, 1.0 equiv) and enantiopure (R)- or (S)-BINOL (variable amount per sample). The resulting solution was stirred for 1 h at rt, with a 650 μL aliquot then removed and its NMR spectra acquired.

Scalemic and racemic samples of Ellman’s sulfinamide 3 were prepared by combining different amounts of enantiopure solutions of (R)- and (S)-sulfinamide 3 in CDCl3. Samples required for concentration screening experiments were prepared directly from 100 mM solutions by dilution, as required.

1H–15N HMBC spectra were acquired by using 50 mM solutions of the desired compound in CDCl3 containing 50 mM MeNO2 as an internal standard.

Results and Discussion

Screening experiments revealed that complexation of (rac)-SIBA 4 with enantiopure BINOL in CDCl3 at a 100 mM concentration produced 1H NMR spectra displaying broad and partially overlapped imine resonances.19 This result contrasted with the complexation reactions of (rac)-SIBA 4 with other diols (previously reported by us),19 which gave 1H NMR spectra of SIBE complexes displaying sharp and well-resolved imine resonances in most cases (see Table S1).1,19 Further investigations revealed that treatment of (rac)-SIBA 4 with (R)-BINOL gave 1H NMR spectra whose imine chemical shifts and ΔδH values varied according to the stoichiometry of (R)-BINOL used (Figure 2a and Table S3). For example, 1H NMR spectra of (rac)-SIBA 4 treated with ≤40 mol % (R)-BINOL displayed a single unresolved 1H NMR imine peak (Figure 2a, entries 1–4), while 60–150 mol % (R)-BINOL loadings produced increasingly differentiated 1H NMR imine peaks, whose averaged chemical shift value drifted incrementally upfield from δH 9.120 to δH 9.023 ppm (Figure 2a, entries 5–7).

Figure 2.

Figure 2

Open in a new tab

Expanded imine region of the 1H NMR spectra (500 MHz, CDCl3) of: (a) (R)-SIBA 4 (100 mM), recorded in the presence of increasing amounts of (R)-BINOL from 0% (top) to 150% loading (bottom); (b) (rac)-SIBA 4 and 100 mol % (R)-BINOL in CDCl3, acquired at decreasing concentrations from 100 mM (top) to 1.0 mM (bottom). Chemical shifts referenced to TMS as an internal standard (∼6 mM).

These (R)-BINOL concentration-dependent increases in imine ΔδH values were initially unexpected since the 1H NMR spectra of other SIBA-diol complexes did not display concentration-dependent imine chemical shift variations.1,19 This led us to conclude that treatment of SIBA 4 with BINOL does not produce IBEs like with other diols. Instead, a change in reaction manifold occurs whereby BINOL serves as a hydrogen-bonding CSA to differentiate the SIBA enantiomers (Figure 1c). Further evidence for BINOL acting as a hydrogen-bonding CSA for 4 is also evident from the gradual downfield shift of the BINOL–OH resonance as the concentration of BINOL increases, and the broader imine resonances produced in the 1H NMR spectra of these BINOL systems compared to typical SIBE/IBE CDA assemblies, also typical of a hydrogen-bonding system.

A review of the literature has revealed significant precedent for the use of BINOL as a hydrogen-bonding CSA to determine the er values of chiral amines, sulfinimines, alcohols, sulfoxides, acids, amino alcohols, and alkaloids.24−32 In order to confirm that BINOL was functioning as a CSA in our case, we next carried out dilution experiments on solutions of (rac)-SIBA 4 and 1 equiv of (R)-BINOL in CDCl3, which revealed significant variation in the chemical shifts of the diagnostic imine as the sample was diluted (Figure 2b and Table S4).33,341H NMR spectroscopic analysis at concentrations decreasing from 100 to 50 mM gave partially resolved imine peaks centered at δH 9.056 and 9.037 ppm, respectively (Figure 2b, Entries 1 and 2), while samples recorded at concentrations of ≤25 mM displayed only a single unresolved imine singlet at δH ≤ 8.973 ppm (Figure 2b, Entries 3–7). Corresponding upfield chemical shift drift of the imine peaks of enantiopure (SS)/(RS)-SIBA 4 complexed to (R)-BINOL was also observed on sample dilution (see Figure S1 and Table S5).

The concentration-dependent changes in chemical shift values and imine peak resolution in these complexation reactions are all consistent with greater noncovalent interactions occurring between the SIBA enantiomers and BINOL at greater BINOL loadings and higher sample concentrations. Further evidence that BINOL was acting as a CSA in CDCl3 was obtained by treating SIBA 4 (varying er)35 with 100 mol % BINOL in CD3CN. These reactions produced concentration-independent 1H NMR spectra displaying a single unresolved imine peak at δH 8.73 ppm, which is consistent with the more polar coordinating CD3CN solvent preventing H-bonding interactions from forming between the SIBA enantiomers and BINOL, thus not inducing enantiodiscrimination (see Figure S4).

To investigate the structural reasons for this change in complexation manifold, 11B and 15N NMR spectroscopic analyses on samples of SIBA (RS)-4, SIBE 5a, [SIBA (RS)-4·(R)-BINOL], and BINOL-derived IBE 2a in CDCl3 were then carried out to determine which systems contained N→B coordination bonds (Figure 3, see Table S10 for full data set, including other diastereomers). δN chemical shift values were determined indirectly through analysis of 1H–15N HMBC spectra (Tables S10 and S11).22,36 SIBA (RS)-4 in CDCl3 exhibited 11B and 15N NMR resonances at δB = 28.6 ppm and δN = −67.0 ppm that are consistent with (RS)-4 containing a planar sp2-hybridized boron atom and a noncoordinated imine nitrogen atom. The absence of an N→B coordination bond in (RS)-4 in the solid state was confirmed by X-ray crystallographic analysis (see Figures 4a and S61). A mixture of SIBA (RS)-4 and 150 mol % (R)-BINOL in CDCl3 gave comparable chemical shift values of δB = 28.8 ppm and δN = −68.5 ppm, consistent with any [SIBA (RS)-4·(R)-BINOL] complexes formed containing noncoordinated sp2-hybridized boron and nitrogen atoms. Pinanediol-derived SIBE (RS,R)-5a (strong [M + H]+ HRMS peak at m/z 388.2118) gave chemical shift values of δB = 30.5 ppm and δN = −53.3 ppm, once again consistent with the presence of noncoordinated sp2-hybridized boron and nitrogen atoms. Conversely, BINOL-derived (α-R,R)-IBE 2a gave significantly upfield chemical shift values of δB = 12.7 ppm and δN = −118.1 ppm, consistent with the presence of coordinated sp3-hybridized boron and nitrogen atoms. The presence of intramolecular N→B coordination bonds in BINOL-derived (α-S,S)-IBE 2a and (α-S,R)-IBE 2b in the solid state was also confirmed by X-ray crystallographic analysis (see Figures S58 and S59).1,37,38

Figure 3.

Figure 3

Open in a new tab

11B and 15N NMR chemical shifts of selected SIBA, SIBE, and IBE complexes. See Tables S10 and S11 for the full data set and benchmarking studies. δN is referenced to MeNO2.

Figure 4.

Figure 4

Open in a new tab

(a) X-ray crystal structure of SIBA (RS)-4 showing intermolecular hydrogen-bonded ladders formed by the boronic acid moiety of one SIBA unit with the sulfoxide oxygen of another SIBA unit. (b) Hydrogen-bonding networks present in the X-ray crystal structures of (RS)-m-tolyl methyl sulfoxide complexed to BINOL. (Reproduced from ref (43) with permission from The Japanese Chemical Society). (c) Evidence of hydrogen bonding between BINOL and (RS)-4 in solution from 1H NMR NOE interactions between the protons of the tert-butyl group of SIBA 4 and the aromatic protons of BINOL (see Figure S57 for spectra). H-bonds are represented by broken lines.

The absence of a stabilizing intramolecular N→B coordination bond in pinanediol-derived SIBE 5a is likely due to steric congestion and/or the electron-withdrawing N-sulfinyl group preventing electron density from being donated from its imine lone pair into the empty p orbital of the boron atom. This contrasts with BINOL-derived (α-R,R)-IBE 2a whose structure contains an intramolecular N→B coordination bond that generates sp3-hybridized boron and nitrogen atoms.

This absence of an intramolecular N→B coordination bond in pinanediol-derived SIBE 5a also provides an explanation for the inability of BINOL to form SIBE complexes. BINOL-SIBE formation would require incorporation of the conformationally restricted diol unit of BINOL (hindered rotation around the biaryl bond) into a strained seven-membered cyclic boronate ester ring. Furthermore, the absence of a stabilizing N→B bond means that this ring system would also need to accommodate a planar sp2-hybridized boron atom, thus introducing further strain into an already strained ring system. Therefore, it follows that the extra strain energy required to incorporate the diol unit of BINOL and a planar sp2 boron atom into a SIBE complex means that BINOL prefers to act as a H-bonding CSA to produce BINOL-coordinated SIBA complexes.24−31

Further evidence that BINOL does not readily react with aryl boronic acids to produce cyclic boronate esters was acquired by carrying out complexation reactions of m-tolylboronic acid and 2-FPBA with pinanediol and BINOL (1.50 equiv) in CDCl3. 1H NMR spectroscopic analysis of these simple complexation reactions revealed that only pinanediol was capable of producing cyclic boronate esters under these conditions, with BINOL showing no reaction with either boronic acid substrate (Scheme S1).39−41

Crystals of BINOL–SIBA complexes could not be obtained, however, X-ray crystal analysis of (RS)-SIBA 4 revealed the presence of strong intermolecular hydrogen bonds between the sulfinyl oxygen atom of one SIBA unit and the boronic acid protons of another to produce well-defined 4-membered Ar–B(OH)2···–O-+S-R hydrogen-bonding assemblies (Figure 4a). Previous X-ray crystallographic studies by Toda et al. have also shown that intermolecular hydrogen-bonding interactions are formed between the phenolic units of BINOL and the sulfinyl oxygen of analogous (RS)-m-tolyl methyl sulfoxide, which combine to produce hydrogen bond networks containing “infinite zigzag chains” (Figure 4b).42−44

Further evidence for strong intermolecular H-bonding interactions between BINOL and SIBA (R)-4 in CDCl3 was obtained from selective 1D 1H NOE spectroscopic studies on equimolar BINOL–SIBA mixtures in CDCl3. Selective excitation of the tert-butyl protons of SIBA (R)-4 revealed significant NOE interactions with all six 1H NMR aromatic environments of (R)-BINOL. These NOE interactions are consistent with a hydrogen bond between the sulfinyl oxygen of a SIBA unit (acceptor) and a phenolic group of a BINOL unit (donor) serving to hold the SIBA tert-butyl group and the BINOL aromatic protons in close proximity (Figure 4c).

Having established that BINOL was acting as a hydrogen-bonding CSA, we next explored its potential for determining the ers of scalemic samples of SIBA 4. Scalemic samples of SIBA 4 (10% er increments from (SS)-4 to (RS)-4) were treated with 1.50 equiv (maximum solubility level)45 of (R)-BINOL at a 100 mM concentration in CDCl3. Examination of these 1H NMR spectra revealed er-dependent variations in the chemical shifts of the diagnostic imine resonances of these scalemic samples (Figure 5a and Table S6). Treatment of (SS)-4 and (RS)-4 with (R)-BINOL gave 1H NMR spectra displaying imine signals with essentially identical chemical shift values at δH = 8.999 and δH = 9.000 ppm, respectively (Figure 5a, entries 1 and 11). This is in contrast to the chemical shift values observed for (rac)-SIBA 4 and (R)-BINOL, which displayed resolved imine peaks at higher δH = 9.015 and δH = 9.030 ppm values, respectively (Figure 5a, entry 6). Increasing the er of the SIBA analyte from 60(S):40(R) to 90(S):10(R) er resulted in the chemical shift of the minor (R)-imine resonance shifting incrementally downfield from δH 9.035 to 9.045 ppm, and the major (S)-imine resonance moving incrementally upfield from δH 9.017 to 9.009 ppm (Figure 5a, entries 5–2). These er-dependent changes in chemical shift produce a corresponding increase in ΔδH value from −0.018 ppm for a racemic sample to −0.036 ppm for a 90(S):10(R) er sample (Figure 5a, cf. entries 6 and 2). However, regardless of their overall ΔδH values, integration of fully resolved major and minor imine peaks of each of these scalemic samples enabled their er values to be accurately determined.

Figure 5.

Figure 5

Open in a new tab

Expanded imine regions of the 1H NMR spectra (500 MHz, CDCl3, 100 mM) of SIBA 4 of varying er (100% (S)-4 to 100% (R)-4 in 10% er increments from top to bottom) in the presence of: (a) 1.50 eq. loading of (R)-BINOL; (b) 1.50 equiv loading of (S)-BINOL; and (c) 1.00 equiv loading of (R)-BINOL. Chemical shifts referenced to TMS as an internal standard (∼6 mM). Partially resolved minor imine peaks are highlighted with arrows.

Conversely, acquisition of 1H NMR spectra of (R)-BINOL complexes of scalemic SIBA samples of 40(S):60(R) and 30(S):70(R) er revealed increasing overlap of the diastereomeric imine peaks, which coalesced into a single peak at ≤20(S):80(R) er (Figure 5a, Entries 7–10), meaning that their er levels could not be determined. However, repeating the 1H NMR study using (S)-BINOL (opposite CSA enantiomer) allowed this er-dependent imine overlap issue to be resolved (Figure 5b and Table S7). Acquisition of a new set of 1H NMR spectra using (S)-BINOL as the CSA produced a new series of 1H NMR spectra that displayed er-dependent variations in the chemical shifts of the major and minor imine peaks that mirrored those produced previously using (R)-BINOL (cf. Figure 5a,5b). This means that (R)-BINOL could be used as a CSA to accurately determine the ers of SIBA samples ranging from 50(S):50(R) to >99(S):1(R) er, while (S)-BINOL could be used to accurately determine the ers of SIBA samples ranging from 50(S):50(R) to <1(S):99(R) er. Detection limits of this CSA system were determined by treating scalemic SIBA 4 samples ranging from 95(S):5(R) to >99(S):1(R) er with 1.50 equiv of (R)-BINOL, whose corresponding 1H NMR spectra all gave accurate er values (up to 99.5(S):0.5(R) er, i.e., a 99% ee detection limit, Figures 6 and S6). Examination of the 1H NMR spectra produced using (S)- and (R)-BINOL revealed that they were not exactly mirrored, with very small variations likely due to minor changes in impurity profiles or moisture content subtly affecting the hydrogen-bonding networks that are responsible for enantiodiscrimination in this system.

Figure 6.

Figure 6

Open in a new tab

Expanded imine regions of the 1H NMR spectra (500 MHz, CDCl3, 100 mM) of scalemic SIBA 4 with high 95(S):5(R)-99.5(S):0.5(R) er levels in the presence of 1.50 equiv (R)-BINOL. Chemical shifts referenced to TMS as an internal standard (∼6 mM). Measured integral values are within the accepted margins of error. See Figure S6 for enhanced spectra containing insets of minor imine signals.

It was subsequently found that reducing the amount of (R)-BINOL used in the er-dependent variation studies of SIBA 4 to one equivalent resulted in a remarkable crossover event occurring in the chemical shift values of the major and minor imine peaks as the er of the SIBA analyte was varied (Figure 5c and Table S8). As seen previously, variation in the chemical shifts of the 1H NMR imine peaks was observed for er values between 90(S):10(R) and 50(S):50(R) er, with minor imine peaks of the (RS)-SIBA enantiomers appearing downfield of the major imine peaks of the (SS)-enantiomers (Figure 5c, Entries 2–6). Once again, incremental changes from 50(S):50(R) to 30(S)70(R) er resulted in increasing overlap of the major and minor imine peaks, which coalesced into a single broad imine peak at 30(S):70(R) er (Figure 5c, Entry 8). However, derivatization of SIBA samples of ≤20(S):80(R) er gave 1H NMR spectra where the major imine peaks of the (RS)-enantiomers were now upfield to the minor imine peaks of their (SS)-enantiomers (Figure 5c, Entries 9 and 10, imine peaks partially overlapped). This means that an er-dependent switch in the ΔδH sign of the imine resonances occurs in this system, with (R)-BINOL producing a large negative ΔδH value of −0.030 ppm for a 90(S):10(R) er sample (Figure 5c, entry 2), but a small opposing positive ΔδH value of +0.007 ppm value for a 10(S):90(R) er value (Figure 5c, entry 10). The results of this er determination study were verified by carrying out complementary complexation studies using (S)-BINOL as the CSA, which gave the expected mirrored results (see Figure S5 and Table S9 for details).

The multicomponent intermolecular hydrogen-bonding networks that are proposed in these BINOL–SIBA complexes provide a simple explanation to understand the er-dependent variations in 1H NMR imine chemical shift values that occur as the enantiopurity of the SIBA analyte is varied. The chemical shift values of the imine protons of each BINOL-coordinated SIBA enantiomer will be determined by the cumulative intramolecular and intermolecular shielding and deshielding effects that they experience. These intermolecular hydrogen-bonding networks are likely formed from interactions between the phenol groups of BINOL units and the sulfinyl and boronic acid groups of the SIBA analyte. This means that different scalemic SIBA samples will contribute different relative amounts of (R)- and (S)- enantiomers to their respective hydrogen bond networks, resulting in variable intermolecular imine shielding/deshielding effects being present in different scalemic samples.

We believe these concentration- and er-dependent variations in chemical shift values have important implications for how hydrogen-bonding-CSA-based NMR protocols should be bechmarked, with a lack of appreciation for these effects potentially leading to incorrect er values or absolute configurations being assigned. For example, the use of (R)-BINOL as a CSA for a SIBA sample of 10(S):90(R) er produces a 1H NMR spectrum containing a single imine peak that could be misinterpreted as evidence that this SIBA sample was enantiopure. In contrast, the 1H NMR spectrum of the same SIBA sample 10(S):90(R) er in the presence of (S)-BINOL displays a set of well-resolved major and minor imine peaks, whose imine integral ratio is an accurate reflection of the enantiopurity of the sample (cf. Entries 10 of Figure 5a,5b). Alternatively, comparison of the 1H NMR spectra of SIBA samples of 90(s):10(R) er and 10(S):90(R) er in the presence of 1.00 equiv BINOL (Figure 5c, cf. Entries 2 and 9) reveals that their minor imine peaks are more deshielded in both cases. This could easily be misconstrued to mean that the same SIBA enantiomer is present in excess in both samples, with this example clearly illustrating the potential pitfalls of using the sign of ΔδH values to assign the absolute configuration of chiral analytes in hydrogen-bonding CSA systems.

Given the findings described herein, we now recommend that new hydrogen-bonding-CSA-based NMR protocols used to determine the ers of chiral analytes should be repeated using both enantiomers of the CSA to generate duplicate sets of 1H NMR spectra at a fixed concentration and stoichiometry. These complementary 1H NMR spectra should then be compared to check for the presence of any er-dependent variations in diagnostic chemical shifts, thus ensuring that the correct CSA enantiomer is chosen to determine accurate er values or absolute configurations. Furthermore, the dramatic variations in chemical shift observed in this hydrogen-bonding CSA NMR study clearly illustrate the importance of reporting NMR sample concentration to ensure the accuracy and reproducibility of er determination protocols. We would also like to note that these variations can also occur to a lesser extent in CDA protocols that produce derivatized species that are capable of forming hydrogen-bonding networks or other noncovalent interactions, so similar precautions should also be taken in these cases.46−48

Conclusions

This study reports that the enantiomers of SIBA 4 coordinate to BINOL (acting as a CSA) in CDCl3 to produce diastereomeric hydrogen-bonded BINOL–SIBA complexes whose imine protons are distinguishable in their 1H NMR spectra. The inability of BINOL to produce a three-component SIBE complex (like other diols) is due to the difficulty of incorporating its conformationally restricted diol unit and a planar sp2 boron atom into a strained seven-membered boronate ester ring. Additionally, SIBAs and SIBEs were found to lack the N→B bond typical of IBEs, which further disfavors the incorporation of BINOL. The use of BINOL as a CSA to determine the ers of scalemic SIBA 4 samples revealed concentration- and er-dependent variations in diagnostic imine 1H NMR chemical shift values that can result in imine peak overlap for selected er values. This imine peak overlap issue can be resolved by repeating the protocol using the opposite enantiomer of BINOL as the CSA. This adaptation enables BINOL to be used to accurately determine the enantiopurity levels of scalemic samples of SIBA 4 (and hence Ellman’s sulfinamide 3) up to 99.5:0.5 er (99% ee).

Due to the potential complications highlighted throughout this work, we recommend that hydrogen-bonding-based CSA protocols used to determine er values of scalemic analytes should be repeated using both CSA enantiomers to identify whether er-dependent variations in diagnostic chemical shifts need to be accounted for, particularly in those cases where single resonances are observed (risk of peak overlap) or absolute configuration is being determined (possible signal crossover). We also re-emphasize the importance of carefully reporting analyte concentrations and CSA stoichiometries in experimental protocols to ensure that optimal CSA conditions can be reproduced effectively, and any risk of misassignment minimized.

Acknowledgments

The authors would like to dedicate this work to John S. Fossey (April 29, 1977 – April 15, 2022), a close friend and colleague who will be sorely missed. This work was funded by the EPSRC through a studentship in the Centre for Doctoral Training in Catalysis (EP/L016443/1) and by the Leverhulme Trust via a Leverhulme Trust Research Project Grant (ORPG-9125). T.D.J. wishes to thank the Royal Society for a Wolfson Research Merit Award and the Open Research Fund of the School of Chemistry and Chemical Engineering, Henan Normal University for support (2020ZD01). Characterization facilities were provided by the Material and Chemical Characterisation Facility (MC2) at the University of Bath (10.15125/mx6j-3r54).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.3c01613.

  • Experimental procedures, synthetic details, characterization, NMR spectra, and X-ray crystal structures (PDF)

The authors declare no competing financial interest.

Supplementary Material

ac3c01613_si_001.pdf (4.4MB, pdf)

References

  1. Groleau R. R.; James T. D.; Bull S. D. The Bull-James Assembly: Efficient Iminoboronate Complex Formation for Chiral Derivatization and Supramolecular Assembly. Coord. Chem. Rev. 2021, 428, 213599 10.1016/j.ccr.2020.213599. [DOI] [Google Scholar]
  2. Groleau R. R.; Powell M. E.; Evans C. D.; Fordred P. S.; Bull S. D.; James T. D.. Diastereomeric Derivatization for Spectroscopy. In Molecular Sciences and Chemical Engineering; Elsevier, 2022. [Google Scholar]
  3. Hamaguchi N.; Okuno Y.; Oe Y.; Ohta T. A Simple Quantitative Chiral Analysis of Amino Acid Esters by Fluorine-19 Nuclear Magnetic Resonance Using the Modified James–Bull Method. Chirality 2019, 31, 34–40. 10.1002/chir.23028. [DOI] [PubMed] [Google Scholar]
  4. Kelly A. M.; Pérez-Fuertes Y.; Arimori S.; Bull S. D.; James T. D. Simple Protocol for NMR Analysis of the Enantiomeric Purity of Diols. Org. Lett. 2006, 8, 1971–1974. 10.1021/ol0602351. [DOI] [PubMed] [Google Scholar]
  5. Kelly A. M.; Bull S. D.; James T. D. Simple Chiral Derivatisation Protocols for NMR Analysis of the Enantiopurity of 1,2-Diphenylethane-1,2-Diamine and N-Boc-Cyclohexane-1,2-Diamine. Tetrahedron: Asymmetry 2008, 19, 489–494. 10.1016/j.tetasy.2008.01.026. [DOI] [Google Scholar]
  6. Powell M. E.; Kelly A. M.; Bull S. D.; James T. D. A Simple Chiral Derivatisation Protocol for 1H NMR Spectroscopic Analysis of the Enantiopurity of O-Silyl-1,2-Amino Alcohols. Tetrahedron Lett. 2009, 50, 876–879. 10.1016/j.tetlet.2008.12.013. [DOI] [Google Scholar]
  7. Chaudhari S. R.; Suryaprakash N. Three-Component Chiral Derivatizing Protocols for NMR Spectroscopic Enantiodiscrimination of Hydroxy Acids and Primary Amines. J. Org. Chem. 2012, 77, 648–651. 10.1021/jo202334d. [DOI] [PubMed] [Google Scholar]
  8. Tickell D. A.; Mahon M. F.; Bull S. D.; James T. D. A Simple Protocol for NMR Analysis of the Enantiomeric Purity of Chiral Hydroxylamines. Org. Lett. 2013, 15, 860–863. 10.1021/ol303566k. [DOI] [PubMed] [Google Scholar]
  9. Chaudhari S. R.; Suryaprakash N. Simple and Efficient Methods for Discrimination of Chiral Diacids and Chiral Alpha-Methyl Amines. Org. Biomol. Chem. 2012, 10, 6410–6419. 10.1039/c2ob25599e. [DOI] [PubMed] [Google Scholar]
  10. Oliveira S. S.; Cunha R. L. O. R.; Silva M. S. 77Se and 125Te NMR Spectroscopy for Enantiopurity Determination of Chalcogen Amines. Tetrahedron Lett. 2016, 57, 4556–4559. 10.1016/j.tetlet.2016.08.091. [DOI] [Google Scholar]
  11. Fossey J. S.; Anslyn E. V.; Brittain W. D. G.; Bull S. D.; Chapin B. M.; Le Duff C. S.; James T. D.; Lees G.; Lim S.; Lloyd J. A. C.; Manville C. V.; Payne D. T.; Roper K. A. Rapid Determination of Enantiomeric Excess via NMR Spectroscopy: A Research-Informed Experiment. J. Chem. Educ. 2017, 94, 79–84. 10.1021/acs.jchemed.6b00355. [DOI] [Google Scholar]
  12. Brittain W. D.; Chapin B. M.; Zhai W.; Lynch V. M.; Buckley B. R.; Anslyn E. V.; Fossey J. S. The Bull–James Assembly as a Chiral Auxiliary and Shift Reagent in Kinetic Resolution of Alkyne Amines by the CuAAC Reaction. Org. Biomol. Chem. 2016, 14, 10778–10782. 10.1039/C6OB01623E. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Pérez-Fuertes Y.; Kelly A. M.; Johnson A. L.; Arimori S.; Bull S. D.; James T. D. Simple Protocol for NMR Analysis of the Enantiomeric Purity of Primary Amines. Org. Lett. 2006, 8, 609–612. 10.1021/ol052776g. [DOI] [PubMed] [Google Scholar]
  14. Yeste S. L.; Powell M. E.; Bull S. D.; James T. D. Simple Chiral Derivatization Protocols for 1H NMR and 19F NMR Spectroscopic Analysis of the Enantiopurity of Chiral Diols. J. Org. Chem. 2009, 74, 427–430. 10.1021/jo8019187. [DOI] [PubMed] [Google Scholar]
  15. Metola P.; Anslyn E. V.; James T. D.; Bull S. D. Circular Dichroism of Multi-Component Assemblies for Chiral Amine Recognition and Rapid Ee Determination. Chem. Sci. 2012, 3, 156–161. 10.1039/C1SC00496D. [DOI] [Google Scholar]
  16. Pushina M.; Farshbaf S.; Shcherbakova E. G.; Anzenbacher P. A Dual Chromophore Sensor for the Detection of Amines, Diols, Hydroxy Acids, and Amino Alcohols. Chem. Commun. 2019, 55, 4495–4498. 10.1039/C9CC01051C. [DOI] [PubMed] [Google Scholar]
  17. Mirri G.; Bull S. D.; Horton P. N.; James T. D.; Male L.; Tucker J. H. R. Electrochemical Method for the Determination of Enantiomeric Excess of Binol Using Redox-Active Boronic Acids as Chiral Sensors. J. Am. Chem. Soc. 2010, 132, 8903–8905. 10.1021/ja103462x. [DOI] [PubMed] [Google Scholar]
  18. Chatterjee S.; Anslyn E. V.; Bandyopadhyay A. Boronic Acid Based Dynamic Click Chemistry: Recent Advances and Emergent Applications. Chem. Sci. 2021, 12, 1585–1599. 10.1039/d0sc05009a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Groleau R. R.; Chapman R. S. L.; Ley-Smith H.; Liu L.; James T. D.; Bull S. D. A Three-Component Derivatization Protocol for Determining the Enantiopurity of Sulfinamides by 1H and 19F NMR Spectroscopy. J. Org. Chem. 2020, 85, 1208–1215. 10.1021/acs.joc.9b02473. [DOI] [PubMed] [Google Scholar]
  20. Lawrence K.; Flower S. E.; Kociok-Kohn G.; Frost C. G.; James T. D. A simple and effective colorimetric technique for the detection of boronic acids and their derivatives. Anal. Methods 2012, 4, 2215–2217. 10.1039/C2AY25346A. [DOI] [Google Scholar]
  21. Guzman A. L.; Hoye T. R. TMS is Superior to Residual CHCl3 for Use as the Internal Reference for Routine 1H NMR Spectra Recorded in CDCl3. J. Org. Chem. 2022, 87, 905–909. 10.1021/acs.joc.1c02590. [DOI] [PubMed] [Google Scholar]
  22. Kawęcki R.; Bednarek B.; Sitkowski J. Interaction of Sulfinimines with Boron Lewis Acids. J. Chem. Soc., Perkin Trans. 2 2001, 2, 1400–1403. 10.1039/b100465o. [DOI] [Google Scholar]
  23. Staun S. L.; Sergentu D. C.; Wu G.; Autschbach J.; Hayton T. W. Use of 15N NMR spectroscopy to probe covalency in a thorium nitride. Chem. Sci. 2019, 10, 6431–6436. 10.1039/C9SC01960J. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Redondo J.; Capdevila A.; Latorre I. Use of (S)-BINOL as NMR Chiral Solvating Agent for the Enantiodiscrimination of Omeprazole and Its Analogs. Chirality 2010, 22, 472–478. 10.1002/CHIR.20766. [DOI] [PubMed] [Google Scholar]
  25. Redondo J.; Capdevila A.; Ciudad S. Determination of the Enantiomeric Purity of the Antiasthmatic Drug Montelukast by Means of 1H NMR Spectroscopy. Chirality 2013, 25, 780–786. 10.1002/chir.22213. [DOI] [PubMed] [Google Scholar]
  26. Klika K. D.; Budovská M.; Kutschy P. Enantiodifferentiation of Phytoalexin Spirobrassinin Derivatives Using the Chiral Solvating Agent (R)-(+)-1,1′-Bi-2-Naphthol in Conjunction with Molecular Modeling. Tetrahedron: Asymmetry 2010, 21, 647–658. 10.1016/j.tetasy.2010.03.035. [DOI] [Google Scholar]
  27. Yuste F.; Sánchez-Obregón R.; Díaz E.; García-Carrillo M. A. Enantiodifferentiation of the Antitumor Alkaloid Crispine A Using the NMR Chiral Solvating Agents (R)- and (S)-BINOL. Tetrahedron: Asymmetry 2014, 25, 224–228. 10.1016/j.tetasy.2013.12.006. [DOI] [Google Scholar]
  28. Borowiecki P. Enantiodifferentiation of Promethazine Using (S)-(−)-BINOL as the NMR Chiral Solvating Agent: Determination of the Enantiomeric Purity and Performance Comparison with Traditional Chiral HPLC. Tetrahedron: Asymmetry 2015, 26, 16–23. 10.1016/j.tetasy.2014.11.018. [DOI] [Google Scholar]
  29. Yi J.; Du G.; Yang Y.; Li Y.; Li Y.; Guo F. Chiral Discrimination of Natural Isoflavanones Using (R)- and (S)-BINOL as the NMR Chiral Solvating Agents. Tetrahedron: Asymmetry 2016, 27, 1153–1159. 10.1016/j.tetasy.2016.09.002. [DOI] [Google Scholar]
  30. Ardej-Jakubisiak M.; Kawecki R. NMR Method for Determination of Enantiomeric Purity of Sulfinimines. Tetrahedron: Asymmetry 2008, 19, 2645–2647. 10.1016/j.tetasy.2008.11.023. [DOI] [Google Scholar]
  31. K. S. N.; Kumar T. N.; Singam S. S. R.; Chaudhari S. R. Selective Homodecoupled 1D-1H NMR Experiment for Unravelling Enantiomers. Anal. Chem. 2022, 94, 10299–10303. 10.1021/acs.analchem.2c01020. [DOI] [PubMed] [Google Scholar]
  32. Treatment of Ellman’s sulfinamide (rac)-3 with (R)-BINOL in CDCl3 did not result in enantiodiscrimination in the 1H NMR spectra.
  33. Comparable concentration-dependent variation in the imine chemical shift values of SIBA (RS)-4 and in the absence of BINOL in CDCl3 were also observed, see Figure S2.
  34. No significant variation in the imine chemical shift value of SIBA (RS)-4 was observed in the presence of excess sulfinamide (RS)-3, see Figure S3.
  35. Whilst ee values are often used to describe the enantiopurity of a chiral analyte, we have mostly used er values in this paper for clarity, as they more clearly describe the ratios of analyte enantiomers. For a discussion of the merits of using er values to describe stereoisomer composition levels, see:; Gawley R. E. Do the Terms “% ee” and “% de” Make Sense as Expressions of Stereoisomer Composition or Stereoselectivity?. J. Org. Chem. 2006, 71, 2411–2416. 10.1021/jo052554w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. 15N NMR chemical shift values should prove useful for exploring the presence/absence of intramolecular N→B bonding in other complexes. Consequently, the δN values of some noncoordinated imines, sulfinimines are provided for benchmarking purposes in Table S11.
  37. Arimori S.; Pérez-Fuertes Y.; Yeste S. L.; Kelly A. M.; Powell M. E.; Groleau R. R.; Kociok-Köhn G.; Bull S. D.; James T. D.. CSD Communication 2022 10.5517/ccdc.csd.cc2bhcqy. [DOI]
  38. Arimori S.; Pérez-Fuertes Y.; Lozano Yeste S.; Kelly A. M.; Powell M. E.; Groleau R. R.; Kociok-Köhn G.; Bull S. D.; James T. D.. CSD Communication 2022 10.5517/ccdc.csd.cc2bhcrz. [DOI]
  39. No reports of well characterized planar BINOL boronic esters containing sp2-hybridised boron atoms could be found in the CCDC database. Although some reports of BINOL-containing boronic esters were found, these publications either do not report diagnostic δB values, or report δB values that are consistent with complexes containing sp3-hybridised boron atoms (See Figure S7 for details).
  40. For previous reports of tetrahedral non-IBE BINOL-boronates, see:; Garg S.; Unruh D. K.; Krempner C. Synthesis, Structures and Catalytic Activity of Some BINOL Based Boronates and Boronium Salts. Dalton Trans. 2021, 50, 5044–5049. 10.1039/D1DT00842K. [DOI] [PubMed] [Google Scholar]
  41. For previous reports of tetrahedral non-IBE BINOL-borates, see:; Cros J. P.; Pérez-Fuertes Y.; Thatcher M. J.; Arimori S.; Bull S. D.; James T. D. Non-Linear Effects Operate and Dynamic Ligand Exchange Occurs When Chiral BINOL-Boron Lewis Acids Are Used for Asymmetric Catalysis. Tetrahedron Asymmetry 2003, 14, 1965–1968. 10.1016/S0957-4166(03)00352-5. [DOI] [Google Scholar]
  42. Toda F.; Mori K.; Sato A. A Determination Method of Absolute Configuration. Bull. Chem. Soc. Jpn. 1988, 61, 4167–4169. 10.1246/bcsj.61.4167. [DOI] [Google Scholar]
  43. Toda F.; Tanaka K.; Mak T. C. W. Mutual optical resolution of bis-β-naphthol and sulfoxides. Absolute configuration and crystal structure of a 1:1 molecular complex. Chem. Lett. 1984, 13 (12), 2085–2088. 10.1246/cl.1984.2085. [DOI] [Google Scholar]
  44. Toda F.; Mori K.; Okada J.; Node M.; Itoh A.; Oomine K.; Fuji K. New Chiral Shift Reagents, Optically Active 2,2′-Dihydroxy-1,1′-Binaphthyl and 1,6-Di(o-Chlorophenyl)-1,6-Diphenylhexa-2,4-Diyne-1,6-Diol. Chem. Lett. 1988, 17, 131–134. 10.1246/cl.1988.131. [DOI] [Google Scholar]
  45. Use of > 150 mol % BINOL loadings produced samples that were unsuitable for 1H NMR spectroscopic analysis due to precipitation of BINOL from CDCl3 on standing (see Table S3).
  46. Baumann A.; Wzorek A.; Soloshonok V. A.; Klika K. D.; Miller A. K. Potentially Mistaking Enantiomers for Different Compounds Due to the Self-Induced Diastereomeric Anisochronism (SIDA) Phenomenon. Symmetry 2020, 12, 1106 10.3390/sym12071106. [DOI] [Google Scholar]
  47. Mitra A.; Seaton P. J.; Assarpour R. A.; Williamson T. Unprecedented concentration dependent chemical shift variation in 1H-NMR studies: A caveat in the investigations of molecular recognition and structure elucidation. Tetrahedron 1998, 54, 15489–15498. 10.1016/S0040-4020(98)01015-1. [DOI] [Google Scholar]
  48. Klika K. D. Use of sub-stoichiometric amounts of chiral auxiliaries for enantiodifferentiation by NMR; caveats and potential utility. Tetrahedron: Asymmetry 2009, 20, 1099–1102. 10.1016/j.tetasy.2009.03.036. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ac3c01613_si_001.pdf (4.4MB, pdf)

Articles from Analytical Chemistry are provided here courtesy of American Chemical Society

RESOURCES