Characterization of stereoisomeric 5-(2-nitro-1-phenylethyl)furan-2(5H)-ones by computation of 1H and 13C NMR chemical shifts and electronic circular dichroism spectra

Article abstract of DOI:10.1002/mrc.5073

In the present work, we describe the preparation of two diastereomers from the enantioselective Michael addition of furan-2(5H)-one to (E)-(2-nitrovinyl)benzene catalyzed by a dinuclear Zn-complex. The relative configurations of the diastereomeric products were assigned by comparing nuclear magnetic resonance (NMR) experimental chemical shift data with those computed by density functional theory (DFT) methods. Corrected mean absolute error (CMAE) and CP3 analyses were used to compare the data sets. The absolute configuration of each diastereomer was initially assigned by analysis of electronic circular dichroism (ECD) data, which was consistent with that of the known X-ray crystallographic structure of the product of a related reaction, namely, (R)-5-((R)-1-(4-chlorophenyl)-2-nitroethyl)furan-2(5H)-one.

Full text of DOI:10.1002/mrc.5073

Alvarenga Elson (Orcid ID: 0000-0002-7333-0006)
Characterization of stereoisomeric 5-(2-nitro-1-phenylethyl)furan-2(5H)-ones by
1
computation of H and 13C NMR chemical shifts and electronic circular dichroism
spectra
Dayane T. Lopesa, Thomas R. Hoyeb, Elson S. Alvarengaa*
a Department of Chemistry, Universidade Federal de Viçosa, Viçosa, MG 36571-900, Brazil
b Department of Chemistry, University of Minnesota, Minneapolis, MN 55455, USA
* Correspondence to: Elson S. Alvarenga, Department of Chemistry, Universidade Federal de
Viçosa, Viçosa, MG 36571-900, Brazil. E-mail: elson@ufv.br
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/mrc.5073
This article is protected by copyright. All rights reserved.

Abstract
In the present work we describe the preparation of two diastereomers from the
enantioselective Michael addition of furan-2(5H)-one to (E)-(2-nitrovinyl)benzene catalyzed
by a dinuclear Zn-complex. The relative configurations of the diastereomeric products were
assigned by comparing NMR experimental chemical shift data with those computed by DFT
methods. Corrected mean absolute error (CMAE) and CP3 analyses were used to compare the
data sets. The absolute configuration of each diastereomer was initially assigned by analysis
of electronic circular dichroism (ECD) data, which was consistent with that of the known X-
ray crystallographic structure of the product of a related reaction, namely (R)-5-((R)-1-(4-
chlorophenyl)-2-nitroethyl)furan-2(5H)-one.
Keywords
DFT, MAE, CP3, ECD, NMR, Michael addition
HIGHLIGHTS
Relative configuration determination by computation of NMR properties.
Absolute configuration analysis by electronic circular dichroism (ECD).
1. Introduction
One challenge for synthetic chemists is achieving the preparation of compounds having
high levels of excess of one stereoisomer among others. Chiral auxiliaries or asymmetric
This article is protected by copyright. All rights reserved.

catalysts often drive the new methodologies used to perform enantioselective and/or
diastereoselective reactions. The need for such methods is often motivated by the differences
in biological activity possessed by different stereoisomers.1–3
Conjugated γ-lactones have been used as the Michael donor upon activation to engage a
suitable acceptor.6 It should be noted that lactones are found in many biologically active
compounds and are, additionally, precursors for the further synthesis of a range of other such
compounds.7
NMR spectroscopy is the primary tool used to assign the constitution and relative
configuration of new chemical structures. However, unequivocal assignment of structure can
sometimes be challenging. In these instances, quantum chemical calculation of NMR
properties can be helpful as an auxiliary tool. Different protocols can be used to compare
experimental data with that computed for various candidate structures; among them is CP3,
an approach that results in a statistic parameter for comparison of two sets of experimental vs.
two sets of computed data of NMR chemical shifts.8
Another significant aspect of structure determination is assigning the correct absolute
configuration (AC) to enantioenriched compounds. For crystalline compounds, X-ray
diffraction can often allow for the determination of AC.9,10 However a suitable single crystal
is not always attainable, making the use of NMR-based techniques such as Mosher’s ester
analysis an alternative.11 In addition, quantum chemical calculations have emerged in the last
decade as a tool to predict with good accuracy spectroscopic parameters,12–18 including
chiroptical properties like electronic circular dichroism (ECD) spectra.19–22
The ECD technique can be employed to determine the absolute configuration of
compounds containing suitable chromophores that absorb in the UV-visible spectrum. The
experimental ECD spectra associated with quantum mechanical calculations using the
appropriate functional and basis set have been used to establish the absolute configurations
with good accuracy, and non-racemic samples of chiral compounds have been studied by this
method.23–26
As part of a program directed towards the synthesis of bioactive compounds, we were
interested in preparing substances with potential insecticidal and herbicidal activities.
Lactone compounds are known to present diverse biological activities. Therefore, we decided
to use the known4 Michael addition of furan-2(5H)-one (1) to (E)-(2-nitrovinyl)benzene (2) to
prepare a compound that could be further converted into various derivatives. From this
reaction we obtained two diastereomers, one known and one new, whose relative and
This article is protected by copyright. All rights reserved.

absolute configurations were determined by comparison of the experimental with the
computed NMR and ECD data. This analysis is the subject of the current manuscript.
2. Materials and Methods
2.1 General
THF was dried by refluxing over sodium wire in the presence of benzophenone as
indicator and oxygen and water scavenger. The THF was distilled and stored over molecular
sieves 4Å overnight. The reaction was followed by thin layer chromatography coated in
silica-gel plates in a ultraviolet chamber at 254 nm27 and subsequent staining with vanillin.
Column chromatography was performed over silica gel (60-230 mesh). NMR spectra were
recorded on a Bruker AVANCE III nanobay 400 MHz using deuterated chloroform as
solvent. Both proton and carbon chemical shifts are reported relative to internal TMS ( =
0.00 ppm). Carbon chemical shift assignments are based on analysis of the HMBC and
HSQC spectra. The term “nfom” in a proton resonance is used to indicate a non-first order
multiplet.
The experimental ECD curves were recorded on a J-815 Jasco spectropolarimeter. Each
compound was analyzed at a concentration of 0.1 mg/mL in a solution of acetonitrile at
ambient temperature in a 10 mm long cell from 200–500 nm. The data were saved in the .txt
format and then processed in Origin 2017 using the option: analysis > signal processing >
smooth > open dialog and selecting the FFT (fast Fourier transform) filter method and a
window of 136.
2.2 Procedure for preparation of the diastereomeric Michael adducts.
Products 4a and 4b were prepared using a modification of the reported methodology.4
The (S,S)-enantiomer of the bis-prophenol ligand (400 mg, 0.6 mmol, 1 equiv) was added to a
flask under nitrogen atmosphere, and anhydrous THF (6.2 mL) was added followed by slow
addition of a 1.0 M solution of Et2Zn in hexanes (1.2 mL, 1.2 mmol, 2 equiv). The resulting
mixture was stirred at rt for 20 min and subsequently used, assuming it to be a 0.1 M solution
of catalyst 3.
trans-β-Nitrostyrene (2, 0.950 mg, 6.3 mmol, 1 equiv) was added to a round-bottomed
flask containing dry THF (18.7 mL) under nitrogen atmosphere. The freshly prepared
solution of catalyst 3 in THF (6.2 mL, 0.63 mmol, 0.10 equiv) and furan-2(5H)-one (1, 0.9
mL, 0.01 mmol, 2 equiv) were successively added to the reaction mixture. The mixture was
stirred at room temperature for 22 h and diluted with EtOAc (50 mL). The reaction mixture
This article is protected by copyright. All rights reserved.

was filtered through a pad of Celite® and washed with water (62.5 mL). The aqueous layer
was separated and extracted with EtOAc (3 x 62.5 mL); the combined organic layers was
dried (MgSO4), filtered, and concentrated. Purification by flash silica gel column
chromatography (hexanes:EtOAc 1.5:1) gave, in order of elution, isomer 4a (or 4a-ent), a
mixture of 4a (or 4a-ent) and 4b, and isomer 4b. After 24 hours, crystals were formed from
the mixture in diethyl ether, which were separated to leave, predominantly, the diastereomer
4b. The final isolated yields of pure 4a and 4b were (349.1 mg, 23%) and (208.7 mg, 14%)
respectively.
(R)-5-((S)-2-Nitro-1-phenylethyl)furan-2(5H)-one (4a)
Rf = 0.48 (hexane:EtOAc 1.5:1); yellow solid (mp 119.9-122.7 °C); 13C NMR (100 MHz,
CDCl3) δ 171.6 (C2), 154.0 (C4), 134.5 (Phipso), 129.6 (Phm), 129.1 (Php), 127.9 (Pho), 122.8
(C3), 83.0 (C5), 76.8 (C2’), and 48.1 (C1’); 1H NMR (400 MHz, CDCl3) δ 7.44-7.34 (m, 3H,
H5’, H6’ and H7’), 7.25-7.30 (dd, J = 7.6, 2.0 Hz, 2H, H4’ and H8’), 7.18 (dd, J = 5.7, 1.7
Hz, 1H, H4), 6.18 (dd, J = 5.7, 1.7 Hz, 1H, H3), 5.21 (ddd, J = 9.0, 1.7, 1.7 Hz, 1H, H5), 4.95
(dd, J = 13.3, 5.3 Hz, 1H, H2’a or H2’b), 4.82 (dd, J = 13.3, 9.0 Hz, 1H, H2’a or H2’b), and
3.62 (ddd, J = 9.0, 9.0, 5.3 Hz, 1H, H1’); m/z (EI, 70 eV) 233 (0.16 M+, C12H11NO4), 186
(15), 150 (36), 104 (99), and 55 (100%); IR (ATR, thin film) ν 3092 (Csp2-H), 2922 (Csp3-H),
1788 and 1749 (C=O), 1602 (C=C), 1548 (N=O), 1156 and 1086 (=C-O-C), and 893 (C-N)
cm-1.
(R)-5-((R)-2-Nitro-1-phenylethyl)furan-2(5H)-one (4b)
13
Rf = 0.39 (hexane:EtOAc 1.5:1); colorless crystals (mp 152.2-153.4 °C); C NMR (100
MHz, CDCl3) δ 171.8 (C2), 153.4 (C4), 132.5 (Phipso), 129.2 (Phm), 128.8 (Php), 128.3 (Pho),
1
123.1 (C3), 81.9 (C5), 75.8 (C2’), and 46.0 (C1’); H NMR (400 MHz, CDCl3) δ 7.27-7.35
(m, 4H, H4, H5’, H7’ and H6’), 7.11-7.19 (nfom, 2H, H4’ and H8’), 5.93 (dd, J = 5.8, 2.0
Hz, 1H, H3), 5.38 (ddd, J = 3.3, 2.0, 1.6 Hz, 1H, H5), 4.98 (dd, J = 13.7, 7.6 Hz, 1H, H2’a or
H2’b), 4.82 (dd, J = 13.7, 7.6 Hz, 1H, H2’a or H2’b), and 4.05 (ddd, J = 7.6, 7.6, 3.2 Hz, 1H,
H1’); m/z (EI, 70 eV) 233 (0.15 M+, C12H11NO4), 186 (13), 150 (38), 104 (93), and 55
(100%); IR (ATR, solid state) ν 3106 (Csp2-H), 2919 (Csp3-H), 1772 and 1737 (C=O), 1600
(C=C), 1546 (N=O), 1167 and 1105 (=C-O-C), and 884 (C-N) cm-1.
2.3 Computation
Conformational search. A molecular mechanics “Conformer search” was used to identify
a suitable set of starting candidate structures for further DFT geometry optimization. This
initial search was performed using the optimized potentials for liquid simulations
This article is protected by copyright. All rights reserved.

(OPLS3)28,29 parameter set as implemented in Maestro 2018-1 (Maestro version 11.5.011).30
The OPLS3 force field includes systematic generation of torsional parameters previously
missing from earlier versions of this force field. The “Energy window for saving structures”
was kept at 5.02 kcal mol-1, “100 steps per rotatable bond” and the “Maximum number of
steps” was 1000.
The number of conformers found for (R)-5-((S)-2-nitro-1-
phenylethyl)furan-2(5H)-one (4a) and (R)-5-((R)-2-nitro-1-phenylethyl)furan-2(5H)-one (4b)
were 10 and 12, respectively. As an internal check on the computational methodology, the
enantiomeric (S)-5-((R)-2-nitro-1-phenylethyl)furan-2(5H)-one (4a-ent) and (S)-5-((S)-2-
nitro-1-phenylethyl)furan-2(5H)-one (4b-ent) were also computed in the same fashion and,
as they should, gave rise to an entirely analogous set of 10 and 12, mirror image conformers.
Chemical shift calculations. All candidate conformers from the molecular mechanics
conformational search were subjected to geometry optimization and frequency calculation
using Gaussian 1631 at the M06-2X32/6-31+G(d,p) level of theory. NMR shielding tensors
were calculated for each optimized conformer using B3LYP33 functional and 6-311+G(2d,p)
basis set. The Boltzmann weighting factors were calculated from the relative free energies
obtained from the frequency calculations. The shielding constants were Boltzmann averaged
for each nucleus of each isolated conformer. Solvation was addressed using integrated
equation formalism polarized continuum model (IEFPCM)34 during optimization and
shielding constants calculations. Chemical shifts were then calculated according to 훿푐푥푎푙푐
=
표
푥
휎 − 휎
6 , where 훿푐푥푎푙푐 is the calculated shift for nucleus x (in ppm), σx is the shielding
표
⁄
1− 휎 10
constant for nucleus x and σo is the shielding constant for the proton in tetramethylsilane
13
(TMS), which was obtained using the same level of theory (1H = 31.8816; C = 183.7262).
Some systematic differences between experimental and calculated chemical shifts were then
reduced by using linear regression. This was done in MS Excel for all the experimental (y
axis) versus computed (x axis) chemical shifts. The linear regression was performed for
chemical shifts grouped into a set of (5) Csp2-bound and (4) Csp3-bound protons (Tables S1-
S4). The linear regression was done also without separating Csp2-bound and Csp3-bound
protons. The 1H NMR chemical shifts of Csp2-bound and Csp3-bound protons were all grouped
together for the linear regression (Tables S26-S29).
CMAE analysis. The corrected mean absolute error (CMAE) is the average difference
between the calculated (δcalc) chemical shift, corrected by linear regression, and the
experimental (δexp) chemical shift for each of the unique protons in the candidate pairs of
This article is protected by copyright. All rights reserved.

structures (Σn|δcalc − δexp|/n). Better r2 fits were seen when the corrected shifts were separately
binned into a set of (5) Csp2-bound and (4) Csp3-bound protons as opposed to including all (9)
protons in a single regression set.
MAE analysis. Mean absolute error (MAE) data are those from the uncorrected chemical
shifts. No linear correction was carried out on the chemical shifts.
CP3 analysis. The computed chemical shifts used in the CMAE calculation were also
used in CP3 analyses. CP3 analyses were performed with uncorrected chemical shifts (no
linear correction, Figure S19-S22) and after linear correction (Figure S23-S25). The
1
13
calculated and experimental H and C NMR chemical shifts for both of the diastereomeric
structures were transferred to the CP3 analysis tool located at http://www-
jmg.ch.cam.ac.uk/tools/nmr/CP3.html.
Electronic Circular Dichroism (ECD) calculations. The geometry was optimized for all
(10 or 12) candidate conformers using Gaussian 1631 at the APFD35/6-311+G(2d,p) level of
theory. The electronic spectrum was also computed using this basis set with acetonitrile as
solvent (IEFPCM).34 For the calculated spectra the results were processed in GaussView 1631
software. All six conformers of each candidate structure were opened as a single new
molecule group. In the “Results” dialog the “UV-vis” option was selected; a plot of the
weighted ECD spectrum was created by selecting “mixture spectra in the plot menu”. In the
“mixture editor” dialog, under “Weight”, “assign Boltzmann population weights” was
selected. In the “combination region,” “curve” was selected from the “plot style” drop-down
menu. In the “plot style” column, the option none was selected for each conformer. “Ok”
returns one to the spectrum; in the “plot” drop-down menu, in properties the scale the “UV-
vis” option was deselected leaving only the ECD spectrum showing. In “Properties” the
“fixed range” for the x axis was set from 225 to 280 nm and y axis from -8 to 8 (units of
10-40 esu2cm2) for structures R,R and S,S (4b and 4b-ent). For candidate structures R,S and
SR (4a and 4a-ent) the x axis was set to the same wavelength of 225 and 280 nm and to -4 to
4 (units of 10-40 esu2cm2) for the y axis. “Save data” was used to save the spectrum as a
*.txt file, which was opened in Origin to combine with the experimental curve(s) to then
superimpose on the computed spectra.
3. Results and discussion
Lactones are of interest because of their biological activity as well as their uses in
synthetic chemistry. Among the latter is a variant of the Michael reaction in which a
This article is protected by copyright. All rights reserved.

butenolide anion acts as a nucleophile that adds to an appropriately electrophilic Michael
acceptor. Asymmetric catalysts have been used to promote this reaction. In the study reported
here, we used the method of Trost and Hitce4 to effect the addition of furan-2(5H)-one (1) to
trans-β-nitrostyrene (2) in the presence of the zinc (S,S)-bis-Pro-Phenol-based complex 3.
This afforded two diastereomeric products: the faster eluting compound (on silica gel) was
presumed to be either (R)-5-((S)-2-nitro-1-phenylethyl)furan-2(5H)-one (4a) or its
enantiomer (4a-ent); the slower eluting isomer had NMR spectral data that were essentially
identical with those reported for the known isomer4 (R)-5-((R)-2-nitro-1-phenylethyl)furan-
2(5H)-one (4b) (Fig. 1).
Figure 1: Asymmetric Michael reaction of trans-β-nitrostyrene (2) and furan-2(5H)-one (1),
catalyzed by the asymmetric zinc complex 3. Candidate structures 4a (R,S), 4a-ent (S,R), 4b
(R,R), and 4b-ent (S,S) for the two diastereomeric products.
The relative and absolute configurations of 4b were originally assigned based on the
consistency of its NMR spectral data within a series of related compounds made by this
methodology, one of whose structures was verified by a single crystal x-ray diffraction
analysis.4 Having both (4b) and (either 4a or 4a-ent) diastereomers in hand, we further
analyzed these isomers using computational NMR methods. More specifically, the proton and
carbon chemical shifts were computed for each and compared with the experimental values.
Subsequently, all four stereoisomers were further studied by comparing their experimental
This article is protected by copyright. All rights reserved.

and computed ECD spectra to confirm (for 4b) and establish (for 4a) their absolute
configuration.
All 1H and 13C NMR signals were assigned for both candidate structures by interpretation
of the NMR spectral data. However, the relative configurations of the isolated diastereomers
could not be definitively assigned based only on the interpretation of their NMR spectra. To
achieve that, we computed the NMR chemical shifts for the (R*,S*) and (R*,R*)
diastereomers (i.e., 4a and 4b) in order to compare these to the experimental values for each
compound. For the computations, the geometries for the set of conformations, identified by
an initial molecular mechanics conformational search,36 for each of 4a and 4b were optimized
at the M06-2X/6-31+G(d,p) level of theory and then subjected to NMR chemical shift
calculation using the B3LYP functional and 6-311+G(2d,p) basis set. The shifts for each
diastereomer were Boltzmann averaged according to the M06-2X energies of the full set of
conformers. The experimental proton chemical shifts for each of the two diastereomers as
well as the computed shifts (linearly corrected, see Supporting Information (SI)) for each of
the isomeric 4a and 4b are shown in Table 1. The experimental carbon chemical shifts for
each of the two diastereomers as well as the computed shifts after linear correction (SI) for
each of the isomeric 4a and 4b are shown in Table 2.
The corrected mean absolute error (CMAE) and the CP3 statistical analysis were then
used to identify the better match between the experimental and computed data sets. CP3 is
calculated according to Equation S1 (SI) and it is based on comparing differences in
calculated chemical shifts with differences in experimental data for equivalent atoms. Some
systematic errors in calculated data are cancelled by comparing the differences between two
calculated shifts and two experimental chemical shifts. A large positive value for CP3 points
to a correct assignment, whereas a large negative value indicates poor agreement. Nuclei
which are not useful for discriminating between structures are automatically given a low
weighting. The probability that the assignment combination is correct (4a = R,S, 4b = R,R) is
calculated by the Bayes’ theorem (Equation S2, SI).37
The results of these approaches are summarized in Table 3. Both CMAE and CP3
analyses of the proton data clearly indicate a better correlation between the experimental and
computed shifts for the pairs of structures 4aexpt vs. 4acalc and 4bexpt vs. 4bcalc (cf. green
highlighted values).
This article is protected by copyright. All rights reserved.

1
MAE analysis was carried out for the H NMR chemical shifts without linear
1
correction (Tables S30-S33). CMAE analyses were performed for the H NMR after linear
correction of the chemical shifts without separating the Csp2-bound and Csp3-bound protons
(Tables S26-S29).
1
CMAE calculation after linear regression of the H NMR chemical shifts of
4aexp+4bexp vs. 4acalc+4bcalc and 4aexp+4bexp vs. 4bcalc+4acalc was carried out (Tables S34-
S35). Linear correction done using the new Python scripts created by Willoughby-Hoye
1
13
13
1Hslope = -1.0767; Hintercept= 31.9477;
C
= -1.0522;
C
= 181.2412 followed by
slope
intercept
CMAE calculation was also evaluated. All MAE and CMAE analyses were consistent except
for the method employing intercet-slope from the literature.36 Incorrect matching 4bexp ≠ 4acalc
presented smaller CMAE (0.11 ppm) than the correct matching 4bexp=4bcalc (0.18 ppm).
However, except for the approach intercept-slope from the literature, the outcome of the other
calculations agreed as summarized in Table 4. The literature intercept-slope was created
using a large set of molecules with a wide range of chemical shifts, and the molecules in the
present work contain atoms with NMR shielding constant properties (through-space effects
and anisotropy) different from those in the literature. Therefore, the use of literature
intercept-slope tends to inadequately correct the chemical shifts in the present work.
The CMAE analysis of the carbon shift data (after linear correction; Tables S7-S10)
13
1
is far less clear. Previous work has shown that C shifts are sometimes less reliable than H
in identifying the correct relative configuration within sets of diastereomers.38 The two lowest
CMAE fits for the carbon data are both to the same calculated data set (i.e., that for 4bcalc; cf.
blue vs. gray highlighted CMAE values). Since it is, of course, impossible for both
diastereomers to have the same relative configuration(!), we also looked at the use of a
“match ratio,”39 an approach that recently has been statistically validated for identification of
the better of two possible pairwise correlations for over 200 pairs of diastereomers.38 In that
approach, greater weight is placed on any pair of CMAE values having a higher match ratio.
Applying that concept here leads to the correct conclusion, but the difference in the match
ratio values (1.34 vs. 1.21) is quite small and that criterion alone does not lend sufficient
13
confidence in guiding the assignment of relative configuration. Linear regression of the C
NMR chemical shifts of 4aexp+4bexp vs. 4acalc+4bcalc and 4aexp+4bexp vs. 4bcalc+4acalc
followed by CMAE analysis of the corrected shifts are summarized in Table 4 (cf. green
highlighted values). Linear regression was carried out in MS Excel and transferred to the SM
(Tables S36-S37). The assignment of relative configuration is more conclusive when linear
This article is protected by copyright. All rights reserved.

13
correction of the calculated C NMR chemical shifts is done for the entire set of chemical
shifts (4a+4b) than when it is carried out separately.
However, the CP3 analysis method for the carbon data is definitive, just as it was for
the proton data sets.
After using the above NMR methods to validate the relative configurations of the two
diastereomers as (4a or 4a-ent) and (4b), we decided to examine the ability of ECD
calculations to validate the known absolute configuration of 4b and then, if encouraging, to
assign the absolute configuration of the second diastereomer. The experimental ECD
spectrum (dashed) of the isomer 4b was compared to the calculated electronic spectra for
each of the enantiomers 4b and 4b-ent (Fig. 2, panel A). For each isomer, the calculated
spectrum was merged from the Boltzmann-weighted spectra from the six individual, lowest
energy conformers. We judged there to be a sufficiently good match (cf. red vs. dashed
spectra) to warrant using the same approach for the comparison of the computed ECD spectra
for 4a and 4a-ent with the experimental spectrum for the second diastereomer (Fig. 2, panel
B). In this case, there was a very close similarity between the blue and dashed spectra,
leading us to confidently assign the absolute configuration of the second isomer as 4a. In
other words, the major enantiomer in each of these two Michael addition products has arisen
from approach of the electrophilic nitrostyrene to the same enantiotopic face of C5 in the
enolized furanone nucleophile (cf. Fig. 1).
A)
B)
This article is protected by copyright. All rights reserved.

Figure 2: Panel A) Experimental electronic circular dichroism (ECD) spectrum of 4b (dotted
line) and the calculated spectra of the (R,R)- and (S,S)-enantiomers 4b (red) and 4b-ent
(orange). Panel B) ECD spectrum of the minor isomer (dotted line) and the calculated spectra
for the (R,S)- and (S,R)-enantiomers 4a (blue) and 4a-ent (green).
4. Conclusion
The enantioselective Michael addition of furan-2(5H)-one (1) to trans--nitrostyrene (2)
provided two diastereomeric products 4a and 4b. The relative configurations of these two
1
diastereomers were distinguished by comparing the experimental H NMR chemical shift
values with the shifts computed for each of the two candidate structures. The validity of using
the computed ECD spectrum to discern the correct absolute configuration for this class of
compound was established using the known (R,R)-stereoisomer 4b as a benchmark. This
approach was then used to assign the absolute configuration of the second diastereomer as the
(R,S)-isomer 4a.
Supporting Information
Further details of the NMR assignments, CMAE analyses, and CP3 analyses. The DFT
1
energies and optimized geometries for the individual conformers as well as copies of the H
and 13C 1D and 2D (HSQC and HMBC) NMR spectra for the isomers 4a and 4b may be
found in the online version of this article at the publisher’s website.
Acknowledgements
This article is protected by copyright. All rights reserved.

We are grateful to Conselho Nacional de Desenvolvimento Científico e Tecnológico
(CNPq), Fundação do Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG), RQ-MG
(Rede Mineira de Química), Coordenação de aperfeiçoamento de Pessoal de Nível Superior
(CAPES) – Finance Code 001 and the National Institute of General Medical Sciences of the
U.S. Department of Health and Human Services (R35 GM127097) for financial support.
Thanks to Victor Rubim Otati for recording the electronic CD spectra. This work was carried
out in part using software and hardware resources made available through the University of
Minnesota, Supercomputing Institute (MSI).
This article is protected by copyright. All rights reserved.

References
1.
2.
3.
4.
5.
6.
A. Calcaterra, I. D’Acquarica. J. Pharm. Biomed. Anal. 2018, 147, 323-340. DOI:
10.1016/j.jpba.2017.07.008.
C. Bertucci, D. Tedesco. J. Chromatogr. A. 2012, 1269, 69-81. DOI:
10.1016/j.chroma.2012.09.046.
S. Superchi, P. Scafato, M. Gorecki, G. Pescitelli. Curr. Med. Chem. 2018, 25, 287-
320. DOI: 10.2174/0929867324666170310112009.
B. M. Trost, J. Hitce. J. Am. Chem. Soc. 2009, 131, 4572-4573. DOI:
10.1021/ja809723u.
K. R. Rohit, S. M. Ujwaldev, K. K. Krishnan, G. Anilkumar. 201885-102. DOI:
10.1002/ajoc.201700491.
G. Chen, G. Liang, Y. Wang, P. Deng, H. Zhou. Org. Biomol. Chem. 2018, 16, 3841-
3850. DOI: 10.1039/c8ob00773j.
7.
8.
B. Mao, B. L. Feringa. 2017. DOI: 10.1021/acs.chemrev.7b00151.
S. G. Smith, J. M. Goodman. J. Org. Chem. 2009, 74, 4597-4607. DOI:
10.1021/jo900408d.
9.
S. Parsons. Tetrahedron: Asymmetry. 2017, 28, 1304-1313. DOI:
10.1016/j.tetasy.2017.08.018.
10. A. Linden. Tetrahedron: Asymmetry. 2017, 28, 1314-1320. DOI:
10.1016/j.tetasy.2017.07.010.
11. A. Cimmino, M. Masi, M. Evidente, S. Superchi, A. Evidente. J. Pharm. Biomed.
Anal. 2017, 144, 59-89. DOI: 10.1016/j.jpba.2017.02.037.
12. R. P. Oliveira, A. J. Demuner, E. S. Alvarenga, L. C. A. Barbosa, T. de Melo Silva. J.
Mol. Struct. 2018, 1152, 337-343. DOI: 10.1016/j.molstruc.2017.09.065.
13. A. Amat, C. Miliani, A. Romani, S. Fantacci. Phys. Chem. Chem. Phys. 2015, 17,
6374-6382. DOI: 10.1039/C4CP04728A.
14. E. E. Zvereva, A. R. Shagidullin, S. A. Katsyuba. J. Phys. Chem. A. 2011, 115, 63-69.
DOI: 10.1021/jp108057p.
15. A. R. Aguiar, E. S. Alvarenga, R. P. Oliveira, V. M. T. Carneiro, L. G. Moura. J. Mol.
Struct. 2018, 1165, 312-317. DOI: 10.1016/j.molstruc.2018.04.013.
16. J. L. C. Sousa, A. M. S. Silva, I. Alkorta, J. Elguero. Magn. Reson. Chem. 2019, 57,
512-521. DOI: 10.1002/mrc.4892.
17. V. M. T. Carneiro, A. R. Aguiar, E. S. Alvarenga. J. Mol. Struct. 2020, 1212, 128157.
DOI: 10.1016/j.molstruc.2020.128157.
This article is protected by copyright. All rights reserved.

18. M. G. Teixeira, E. S. Alvarenga. Magn. Reson. Chem. 2016, 54, 623-631. DOI:
10.1002/mrc.4411.
19. G. A. Suárez-Ortiz, C. M. Cerda-García-Rojas, M. Fragoso-Serrano, R. Pereda-
Miranda. J. Nat. Prod. 2017, 80, 181-189. DOI: 10.1021/acs.jnatprod.6b00953.
20. G. Pescitelli, L. Di Bari, N. Berova. Chem. Soc. Rev. 2011, 40, 4603. DOI:
10.1039/c1cs15036g.
21. J. T. Vázquez. Tetrahedron Asymmetry. 2017, 28, 1199-1211. DOI:
10.1016/j.tetasy.2017.09.015.
22. G. E. Tranter, D. D. Le Pevelen. Tetrahedron: Asymmetry. 2017, 28, 1192-1198. DOI:
10.1016/j.tetasy.2017.08.019.
23. B. D. Zhou, J. Ren, X. C. Liu, H. J. Zhu. Tetrahedron. 2013, 69, 1189-1194. DOI:
10.1016/j.tet.2012.11.050.
24. J. Ren, G.-Y. Li, L. Shen, G.-L. Zhang, L. A. Nafie, H.-J. Zhu. Tetrahedron. 2013, 69,
10351-10356. DOI: 10.1016/j.tet.2013.10.004.
25. Y. Sun, J. Liu, L. Li, C. Gong, S. Wang, F. Yang, H. Hua, H. Lin. Bioorganic Med.
Chem. Lett. 2018, 28, 315-318. DOI: 10.1016/j.bmcl.2017.12.049.
26. J. Ren, T. L. Mistry, P. C. Su, S. Mehboob, R. Demissie, L. W. M. Fung, A. K. Ghosh,
M. E. Johnson. Bioorganic Med. Chem. Lett. 2018, 28, 2074-2079. DOI:
10.1016/j.bmcl.2018.04.052.
27. E. S. de Alvarenga, W. A. Saliba, B. G. Milagres. Quim. Nova. 2005, 28, 927-928.
DOI: 10.1590/S0100-40422005000500034.
28. E. Harder, W. Damm, J. Maple, C. Wu, M. Reboul, J. Y. Xiang, L. Wang, D. Lupyan,
M. K. Dahlgren, J. L. Knight, J. W. Kaus, D. S. Cerutti, G. Krilov, W. L. Jorgensen, R.
Abel, et al. J. Chem. Theory Comput. 2016, 12, 281-296. DOI:
10.1021/acs.jctc.5b00864.
29. W. L. Jorgensen, J. Tirado-Rives. J. Am. Chem. Soc. 1988, 110, 1657-1666. DOI:
10.1021/ja00214a001.
30. Schrodinger Release 2018-1: Maestro, Schrodinger, LLC, New York.
https://www.schrodinger.com/citations/.
31. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M.
Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P.
Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding,
F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe,
This article is protected by copyright. All rights reserved.

V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. J.
Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R.
Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R.
L. Martin, K. Morokuma, O. Farkas, J. B. Foresman, and D. J. Fox, Gaussian, Inc.,
Wallingford CT, 2016.Gaussian 16, Revision B.01.
32. Y. Zhao, D. G. Truhlar. Theor. Chem. Acc. 2008, 120, 215-241. DOI: 10.1007/s00214-
007-0310-x.
33. A. D. Becke. J. Chem. Phys. 1993, 98, 5648-5652. DOI: 10.1063/1.464913.
34. A. V. Marenich, C. J. Cramer, D. G. Truhlar. J. Phys. Chem. B. 2009, 113, 6378-6396.
DOI: 10.1021/jp810292n.
35. A. Austin, G. A. Petersson, M. J. Frisch, F. J. Dobek, G. Scalmani, K. Throssell. J.
Chem. Theory Comput. 2012, 8, 4989-5007. DOI: 10.1021/ct300778e.
36. P. H. Willoughby, M. J. Jansma, T. R. Hoye. Nat. Protoc. 2020, 03216, 41596. DOI:
10.1038/s41596-020-0293-9.
37. S. G. Smith, J. M. Goodman. J. Org. Chem. 2009, 74, 4597-4607. DOI:
10.1021/jo900408d.
38. M. G. Chini, R. Riccio, G. Bifulco. European J. Org. Chem. 2015, 2015, 1320-1324.
DOI: 10.1002/ejoc.201403569.
39. D. J. Marell, S. J. Emond, A. Kulshrestha, T. R. Hoye. J. Org. Chem. 2014, 79, 752-
758. DOI: 10.1021/jo402627s.
This article is protected by copyright. All rights reserved.

Table 1. Experimental vs. computed 1H NMR chemical shifts of diastereomers 4a and 4b.
H Experimental
H Calculated correcteda
“R,S”
4a or 4a-ent
 (ppm)b
“R,R”
4b4
 (ppm)b
“R,S”
4a
 (ppm)c
“R,R”
4b
 (ppm)d
Proton
3
4
5
6.18 (dd, 5.7, 2.7)
7.18 (dd, 5.7, 1.7)
5.21 (dt, 9.0, 1.7)
5.93 (dd, 5.8, 2.0)
7.31 (m)
6.18
7.21
5.18
3.61
4.84
4.97
n/a
5.93
7.30
5.31
4.02
4.86
5.03
n/a
5.38 (m)
1’ 3.62 (ddd, 9.0, 9.0, 5.4) 4.05 (ddd, 7.6, 3.2)
2’a
2’b
4.82 (dd, 13.3, 9.0)
4.95 (dd, 13.3, 5.3)
n/a
4.82 (dd, 13.7, 7.6)
4.98 (dd, 13.7, 7.6)
n/a
Phipso
Pho
7.28 (m)
7.15 (m)
7.35
7.36
7.34
7.25
7.27
7.27
Phm
Php
7.40 (m)
7.31 (m)
7.40 (m)
7.31 (m)
b
aAfter linear correction (see SI). Data in parentheses following the shift value is the multiplicity and J values
c
d
(in Hz) for each proton. Data taken from “H Calc R,S Corrected” Table S1. Data taken from “H Calc R,R
Corrected” Table S4.
This article is protected by copyright. All rights reserved.

Table 2. Experimental vs. computed 13C NMR chemical shifts of diastereomers 4a and 4b.
H Experimental
H Calculated Correcteda
4a or 4a-ent
4b4
 (ppm)
4a
 (ppm)b
4b
 (ppm)c
Carbon
 (ppm)
2 (C=O)
171.6
122.8
154.0
83.0
171.8
123.1
153.4
81.9
171.9
121.0
158.2
82.8
171.9
121.5
157.8
81.0
3
4
5
1’
48.1
46.0
49.9
48.1
2’
76.8
75.8
77.3
76.2
Phipso
Pho
Phm
Php
134.5
127.9
129.6
129.1
132.5
128.3
129.2
128.8
135.3
126.8
127.2
127.0
133.0
126.7
127.6
127.0
aAfter linear correction (see SI). bData taken from “ALC” Table S7. cData taken from “ALC” Table S8.
This article is protected by copyright. All rights reserved.

Table 3. Methods used for comparison of experimental and calculated chemical shift data.a,b
pairwise
comparisons
match
ratio
CP3 all
datac
dCMAEproton eCMAEcarbon
CP3proton CP3carbon
4aexpt vs. 4acalc
4aexpt vs. 4bcalc
4bexpt vs. 4bcalc
4bexpt vs. 4acalc
0.03 ppm
0.06 ppm
0.04 ppm
0.09 ppm
1.53 ppm
1.26 ppm
1.48 ppm
1.98 ppm
98.3%
1.7%
100%
0%
100%
0
1.21
98.3%
1.7%
100%
0%
100%
0
1.34
aThe chemical shifts used in the CP3 analyses were not corrected by linear regression.
bThe CP3 analyses were performed before and after assignment of the signals. The
probabilities were identical in both situations (print screens of the CP3 analyses are shown in
the SI).
cCP3 analysis was performed with proton and carbon data.
dThe linear regression was performed for chemical shifts grouped into a set of (5) Csp2-
bound and (4) Csp3-bound protons (Tables S1-S4).
eAfter linear correction (ALC) of the calculated 13C NMR chemical shifts (Tables S7-S10).
This article is protected by copyright. All rights reserved.

Table 4. MAE and CMAE analyses calculated before and after linear correction of the
calculated 1H and 13C NMR chemical shifts, respectively.
Pairwise
comparisons
4aexp vs.
4acalc
aCMAEproton bCMAEproton cMAEproton dCMAEproton eCMAEproton fCMAEcarbon
0.03 ppm
0.06 ppm
0.04 ppm
0.09 ppm
0.05 ppm
0.16 ppm
0.08 ppm
0.11 ppm
0.29 ppm
0.35 ppm
0.28 ppm
0.33 ppm
0.07 ppm
0.15 ppm
0.07 ppm
0.15 ppm
0.09 ppm
0.24 ppm
0.18 ppm
0.11 ppm
1.51 ppm
1.93 ppm
1.51 ppm
1.93 ppm
4aexp vs.
4bcalc
4bexp vs.
4bcalc
4bexp vs.
4acalc
aThe linear regression was performed for chemical shifts grouped into a set of (5) Csp2-
bound and (4) Csp3-bound protons (Tables S1-S4).
bThe linear regression was done without separating Csp2-bound and Csp3-bound protons.
The 1H NMR chemical shifts were all grouped together for the linear regression (Tables S26-
S29).
cWithout linear correction of the calculated 1H NMR chemical shifts (Tables S30-S33).
1
dThe linear correction was done using the H NMR chemical shifts of 4a+4b. The CMAE
was calculated using (4a+4b)calc vs. (4a+4b)exp. and (4b+4a)calc vs. (4a+4b)exp. (Tables S34-
35)
eThe linear correction was done using the new Python scripts created by Willoughby-Hoye
13
1Hslope = -1.0767; 1Hintercept= 31.9477;
C
= -1.0522; 13Cintercept= 181.2412;36
slope
fThe linear correction was done using the 13C NMR chemical shifts of 4a+4b. The CMAE
was calculated using (4a+4b)calc vs. (4a+4b)exp. and (4b+4a)calc vs. (4a+4b)exp. (Tables S36-
37)
This article is protected by copyright. All rights reserved.