Synthesis of Novel Phosphorus-Containing Derivatives of 1,3,4-Trimethylglycoluril via the Birum–Oleksyszyn Reaction

This work presents the synthesis of a new compound, 1-[aryl-(diphenylphosphono)methyl]-3,4,6-trimethylglycolurils, via the interaction of benzaldehyde and its mononitro- and monohydroxyderivatives with 1,3,4-trimethylglycoluril and triphenylphosphite. By varying the reaction conditions and the catalysts, the obtained product yields ranged from satisfactory to good. The diastereomers formed during the reaction were separated by semipreparative HPLC on the C18 stationary phase. The isolated diastereomers were characterized by 1H, 13C, and 31P NMR, and the structures of the diastereomers were confirmed using a single-crystal X-ray crystal structure analysis and quantum chemical calculations.


Introduction
Glycoluril derivatives have attracted the attention of researchers due to their wide range of applications and, above all, due to their biological activities, particularly their nootropic, neurotropic, anxiolytic [1,2], and antibacterial properties [3,4].Additionally, glycoluril derivatives are widely used as "building blocks" to synthesize supramolecular compounds such as bambus[n]urils [5,6], cucurbit[n]urils [7,8], and molecular clips [9,10].Special attention has been paid to studying methods for synthesizing new glycoluril derivatives and their structural analogs and precursors as well as to the characterization of their properties [11][12][13][14][15]. Individual phosphorus-containing glycoluril derivatives are described in the literature [16,17], where their catalytic activity in some three-component reactions has been shown.Phosphorylated glycoluril derivatives were investigated as flame-resistant additives for rubber [18] and as moleculars tweezer exhibiting high binding properties to aliphatic amines [19].However, these reports are sporadic, and to date, only a relatively small number of synthetic approaches have been developed for preparing phosphoruscontaining glycoluril derivatives [16][17][18][19], and the main regularities of the formation of stereoisomers of the phosphonate derivatives of glycoluril have not been established either.We previously found [20] that the interaction of 1,3,4-trimethylglycoluril with aliphatic aldehydes and triphenylphosphite in acetonitrile using methanesulfonic acid as a catalyst led to the formation of 1-[1-(diphenylphosphono)alkyl]-3,4,6-trimethylglycoluril.
It is noteworthy that the a empt to synthesize phosphonate derivatives of 1,3,4-trimethylglycoluryl 4a was unsuccessful under typical conditions used to prepare diphenyl 1-aminophosphonates [21], namely, with glacial acetic acid as a solvent.Thus, according to the NMR analysis of the reaction mixture, the triphenylphosphite 3 interaction with benzaldehyde 2a in glacial acetic acid with 1,3,4-trimethylglycoluril 1 (80 °C, 2 h) led to the formation of target compound 4a with a low yield of 21%.In addition, the 31 P NMR spectra of the reaction mixture contained clear signals within 3.40-−0.57ppm, indicating both partial and complete hydrolysis of the ester groups for compound 3 as well as the formation of α-hydroxyphosphonate 6a ( 31 P δ = 15.99 ppm) as a by-product (Scheme 2) (Supplementary Materials, Figure S1).It is noteworthy that the attempt to synthesize phosphonate derivatives of 1,3,4trimethylglycoluryl 4a was unsuccessful under typical conditions used to prepare diphenyl 1-aminophosphonates [21], namely, with glacial acetic acid as a solvent.Thus, according to the NMR analysis of the reaction mixture, the triphenylphosphite 3 interaction with benzaldehyde 2a in glacial acetic acid with 1,3,4-trimethylglycoluril 1 (80 • C, 2 h) led to the formation of target compound 4a with a low yield of 21%.In addition, the 31 P NMR spectra of the reaction mixture contained clear signals within 3.40-−0.57ppm, indicating both partial and complete hydrolysis of the ester groups for compound 3 as well as the formation of α-hydroxyphosphonate 6a ( 31 P δ = 15.99 ppm) as a by-product (Scheme 2) (Supplementary Materials, Figure S1).
The authors carried out the selection of catalysts, solvents, temperatures, and reaction times to establish the optimal conditions for synthesizing the target product 4a because the typical conditions for the synthesis of α-aminophosphonates proved to be inefficient in obtaining phosphonate derivatives of 1,3,4-trimethylglycoluryl 1.
A series of model syntheses of compound 4a was carried out to select the optimal reaction conditions considering the fact that Lewis acids and organic Brønsted acids tend to show a catalytic effect in the Birum-Oleksyszyn reaction [22,23].Therefore, the reaction was carried out in anhydrous acetonitrile; AlCl 3 , BF 3 •Et 2 O, SnCl 4 , and TiCl 4 were used as Lewis acids, whereas trifluoroacetic acid, methanesulfonic acid, and trifluoromethanesulfonic acid were used as Brønsted acids in addition to glacial acetic acid (Table 1).The amount of catalyst equaled 10 mol% of the total mass of the reaction mixture, and the process was carried out at room temperature.The reaction flow was monitored by HPLC.Scheme 2. The probable formation pathways for by-products 5a, 6a, and 7a in the synthesis of compound 4a.
The authors carried out the selection of catalysts, solvents, temperatures, and reaction times to establish the optimal conditions for synthesizing the target product 4a because the typical conditions for the synthesis of α-aminophosphonates proved to be inefficient in obtaining phosphonate derivatives of 1,3,4-trimethylglycoluryl 1.
A series of model syntheses of compound 4a was carried out to select the optimal reaction conditions considering the fact that Lewis acids and organic Brønsted acids tend to show a catalytic effect in the Birum-Oleksyszyn reaction [22,23].Therefore, the reaction was carried out in anhydrous acetonitrile; AlCl3, BF3⸱Et2O, SnCl4, and TiCl4 were used as Lewis acids, whereas trifluoroacetic acid, methanesulfonic acid, and trifluoromethanesulfonic acid were used as Brønsted acids in addition to glacial acetic acid (Table 1).The amount of catalyst equaled 10 mol% of the total mass of the reaction mixture, and the process was carried out at room temperature.The reaction flow was monitored by HPLC.  1 Isolated yield. 2 With the addition of Ac2O as a dehydrating agent. 3CH3COOH as a solvent.
Table 1 shows that the use of glacial acetic acid as a solvent resulted in a relatively low yield of the target product 4a (23%).The maximum concentration of the la er in the Scheme 2. The probable formation pathways for by-products 5a, 6a, and 7a in the synthesis of compound 4a.    1 Isolated yield. 2 With the addition of Ac 2 O as a dehydrating agent. 3CH 3 COOH as a solvent.
Table 1 shows that the use of glacial acetic acid as a solvent resulted in a relatively low yield of the target product 4a (23%).The maximum concentration of the latter in the reaction mixture was reached within 60 min according to HPLC (Table 1, entry 10).However, with glacial acetic acid as a catalyst and acetonitrile as a solvent, the content of 4a remained relatively low, although it increased with time (Table 1, entries 1 and 2).
If trifluoroacetic, methanesulfonic, and trifluoromethanesulfonic acids (i.e., organic acids stronger than acetic acid) are applied as catalysts, the content of the target product 4a in the reaction mixture increases by up to nine times (Table 1, entries 3, 4, and 5).When trifluoroacetic acid is applied within a comparable reaction time, the analytical yield of 4a becomes almost six times higher than in the case of synthesis with glacial acetic acid.
For methanesulfonic acid, the maximum yield for trimethylphosphonate glycoluril 4a equaled 91% (82% isolated preparatively).This yield was achieved within 30 min under the selected reaction conditions.An increase in the reaction time when using methanesulfonic acid led to a decrease in the concentration of the target product 4a, which might indicate hydrolysis and the formation of the reaction by-products.
The application of trifluoromethanesulfonic acid also leads to a notable increase in the content of the target substance 4a for a relatively short reaction time (Table 1, entry 5), but the analytical yield of compound 4a is only 38%.We assume that trifluoromethanesulfonic acid not only promotes the main reaction but also catalyzes the side reactions, namely, the hydrolysis of the initial triphenylphosphite 3 and of the target compound 4a.Furthermore, trifluoromethanesulfonic acid might promote the formation of α-hydroxyphosphonate 6a (Scheme 2).To verify this assumption, we conducted an additional experiment and added acetic anhydride to the reaction mixture with trifluoromethanesulfonic acid as a catalyst in order to chemically bind the water formed during the reaction.Thus, we established that the practical yield of the target compound 4a sharply increased by up to 75% for a comparable reaction time, which implicitly confirms the assumption (Table 1, entry 6).
In the course of the further selection of the conditions to prepare compound 4a, we studied the effect of solvents on the target product yield under the same conditions (room temperature, methanesulfonic acid as a catalyst, 10 mol%).Table 2 shows the results obtained.Despite a rather low solubility of the initial 1,3,4-trimethylglycoluril 1 in nonpolar solvents at room temperature, the analytical yield of the phosphonate glycoluril 4a is relatively high and varies from 29 to 59% (Table 2, entries 1-3).At the same time, the maximum analytical yield of 91% (82% being a preparative yield) is achieved in an aprotic medium polar solvent, i.e., acetonitrile.The use of a more polar aprotic solvent, DMSO, does not lead to the formation of the product 4a in notable amounts according to HPLC.If methanol is applied as a reaction solvent, the highest analytical yield is achieved in 60 min and amounts to only 8%.
It has been experimentally established that an increase in the reaction temperature within 40 to 70 • C-with acetonitrile as a solvent and methanesulfonic acid as a catalyst-led to a decrease in the yield of compound 4a.Thus, at 40 • C, the maximum analytical yield of product 4a was 64% for 20 min.At 70 • C, the maximum analytical yield of 4a equaled 50% for 20 min.
In sum, the optimal conditions for the three-component Birum-Oleksyszyn reaction in the case of 1,3,4-trimethylglycoluril 1 (Scheme 1) appear to be the use of acetonitrile as a solvent (room temperature) and the use of methanesulfonic acid as a catalyst (10 mol% of the total reaction mixture).A decrease in the amount of catalyst to 5 mol% or its increase to 20 mol% results in decreases in the yield of the desired product 4a to 62% and 54%, respectively.

Separation and Identification of Diastereomers of Phosphonate Derivatives of 1,3,4-Trimethylglycoluriles 4a and 4a
The individual diastereomers 4a and 4a were preparatively separated and isolated by HPLC with a C18 stationary phase.The diastereomers were then characterized by NMR and single-crystal X-ray diffraction analysis (SCXRD).Obviously, the order of elution for diastereomers 4a and 4a depends on the mutual arrangement of the substituents around the asymmetric C7 centers (Scheme 1).The highest retention is characterized by the 4a isomer with the most sterically accessible hydrophobic groups.

X-ray Crystal Structure Analysis of the Diastereomers 4a and 4a
Compound 4a crystallizes in a monoclinic crystal system, space group C2/c.The asymmetric unit contains one 4a molecule and one and a half acetonitrile solvate molecules (Figure 1a).The unit cell contains eight formula units.The crystallographic space group is centrosymmetric; therefore, both enantiomeric forms of compound 4a crystallize as a true racemate.The configuration of the stereocenters (3a, 6a, CHPO 3 ) in compound 4a is as follows: R,S,R (and enantiomeric S,R,S) (Figure 1).asymmetric unit contains one 4a′ molecule and one and a half acetonitrile solvate molecules (Figure 1a).The unit cell contains eight formula units.The crystallographic space group is centrosymmetric; therefore, both enantiomeric forms of compound 4a′ crystallize as a true racemate.The configuration of the stereocenters (3a, 6a, CHPO3) in compound 4a′ is as follows: R,S,R (and enantiomeric S,R,S) (Figure 1).
The molecules of 4a′ of the same enantiomeric form are joined into homochiral supramolecular chains by short CH•••O contacts involving the oxygen atoms of the phosphonate group and the hydrogen atoms of the phenyl (d(O3-H22) = 2.60 Å) or glycoluril fragments (d(O2-H10) = 2.48 Å, Figure 2a).The second type of CH•••O contacts involves the carbonyl oxygen atom and aliphatic hydrogen atoms with the distances d(O4-H7) = 2.43 Å and d(O4-H10) = 2.40 Å, respectively.The chains are oriented along the crystallographic axis b.The enantiomeric chains of opposite stereochemistry are joined in the crystal packing by π-π stacking interactions between the phenyl groups of the phosphonate ester moiety with the distance of 3.295 Å between the planes of the rings (Figure 2b).Compound 4a crystallizes in a monoclinic centrosymmetric space group, P2 1 /c.In contrast to compound 4a , there are no solvate molecules in the crystal structure, and the asymmetric unit consists of one molecule (Figure 1b).The unit cell contains four formula units.Similarly to compound 4a , the enantiomeric pair of 4a crystallizes as a true racemate.The configuration of the stereocenters (3a, 6a, CHPO 3 ) is S,R,R (and R,S,S).Therefore, compounds 4a and 4a differ in the absolute configuration of the carbon atom connected to the phosphonate group, in accordance with the two possible directions of the 1,3,4-trimethylglycoluril addition to the carbonyl group of the aldehyde.
Similarly to compound 4a , the enantiomers of 4a of the same configuration are joined into supramolecular chains oriented along the crystallographic axis b (Figure 3a),  Compound 4a″ crystallizes in a monoclinic centrosymmetric space group, P21/c.In contrast to compound 4a′, there are no solvate molecules in the crystal structure, and the asymmetric unit consists of one molecule (Figure 1b).The unit cell contains four formula units.Similarly to compound 4a′, the enantiomeric pair of 4a″ crystallizes as a true racemate.The configuration of the stereocenters (3a, 6a, CHPO3) is S,R,R (and R,S,S).Therefore, compounds 4a′ and 4a″ differ in the absolute configuration of the carbon atom connected to the phosphonate group, in accordance with the two possible directions of the 1,3,4-trimethylglycoluril addition to the carbonyl group of the aldehyde.
Similarly to compound 4a′, the enantiomers of 4a″ of the same configuration are joined into supramolecular chains oriented along the crystallographic axis b (Figure 3a),  Compound 4a″ crystallizes in a monoclinic centrosymmetric space group, P21/c.In contrast to compound 4a′, there are no solvate molecules in the crystal structure, and the asymmetric unit consists of one molecule (Figure 1b).The unit cell contains four formula units.Similarly to compound 4a′, the enantiomeric pair of 4a″ crystallizes as a true racemate.The configuration of the stereocenters (3a, 6a, CHPO3) is S,R,R (and R,S,S).Therefore, compounds 4a′ and 4a″ differ in the absolute configuration of the carbon atom connected to the phosphonate group, in accordance with the two possible directions of the 1,3,4-trimethylglycoluril addition to the carbonyl group of the aldehyde.
Similarly to compound 4a′, the enantiomers of 4a″ of the same configuration are joined into supramolecular chains oriented along the crystallographic axis b (Figure 3a Further studies were focused on establishing the stereochemical features of obtaining phosphonate derivatives of 1,3,4-trimethylglycoluryl 4b-f with the participation of nitroand hydroxy derivatives of benzaldehyde 2b-f (Scheme 1) using HPLC, NMR spectroscopy, and quantum chemical calculations.
The X-ray crystal structures and NMR results for diastereomers 4a and 4a were compared with the NMR spectroscopy data for compounds 4b-f.Thus, for diastereomer 4 , the signals of the methine protons of 1 H CH-CH bond were recorded by two doublets within a narrow region of 5.11-5.37ppm, while for compound 4 , the signals of these protons were recorded within a broader region of 4.95-5.51ppm.The 31 P spectra for the type 4 diastereomers are characterized by a signal shift towards the weak field relative to the signal of the type 4 compounds.The nature of the substituent in the metaand para-position does not significantly affect the final ratio of the diastereomers of products 4b-e.However, we observed a notable decrease in the yield of reaction products with a para-substituent (the yields for 4d and 4e are 40% and 36%, respectively) compared to the reaction products with a meta-substituent (yields of 4b and 4c are 60% and 67%, respectively).In the case of compound 4f, only one diastereomer was isolated from the reaction mixture and assigned to the 4f structure.According to HPLC-MS, the application of 2-hydroxybenzaldehyde under the conditions of interest did not allow for the detection of any hydrolysis products of compound 4f or any ions with the mass that could be assigned to this compound.
Comparing the positions of diastereomeric structures for the phosphonates of 1,3,4trimethylglycoluriles (4a-f) on the energy profile (Table S1 Supplementary Materials) indicates a larger thermodynamic stability of the 4 -type structures regardless of the nature and position of the substituent in the ring.This correlates with the experimental data on ratio of 4 :4 stereoisomers (Scheme 1).The changes in the electron energy and enthalpy of the isomeric transition of 4 -type structures to 4 -type structures range from 2.85 to 3.58 kcal/mol and from 2.92 to 3.93 kcal/mol, respectively (Table S2 Supplementary Materials).
2.2.3.The Diastereomers of the Phosphonate Derivatives of 1,3,4-Trimethylglycolyrils (4a-f): General Regularities in the NMR Spectra NMR spectroscopy was applied to reliably identify the diastereomers of phosphonate derivatives of 1,3,4-trimethylglycolyrils 4a-f.Table 3 presents the results of comparing the experimental and pre-calculated values for the chemical shifts of the hydrogen atoms in bridging the methine groups of the glycoluril fragment (3a and 6a) and the hydrogen atom bonded to the carbon atom in the CHPO 3 group.The correlations between the experimental and pre-calculated δ values for all of these atoms can be found in Supplementary Materials (Figure S52).Table 3 shows that the δ for the hydrogen atom bonded to the carbon atom in the CHPO 3 group, as well as the methine protons of the bridging CH-CH (3a, 6a) group, can indicate the configuration of the phosphonate substituent in structure 4. Thus, the signals of the protons in the carbon atom with a phosphonate substituent (CHPO 3 ) of type 4 (RSR, SRS) are shifted towards lower values compared to the signals of the enantiomers of type 4 (SRR, RSS).For example, the experimental pre-calculated chemical shifts equal 5.77 and 5.33 ppm, respectively, for the hydrogen atoms bonded to the carbon atom of the phosphonate group (CHPO 3 ) in the RSR and SRS enantiomers of compound 4a .However, these shifts equal 5.85 and 6.57 ppm, respectively, for the hydrogen atoms of the SRR and RSS enantiomers of compound 4a .This pattern persists for the unsubstituted and OH-substituted synthesized glycoluril phosphonates 4a, 4c, and 4e.The experimental signals for similar hydrogen atoms in the indicated pairs of diastereomers of the nitro derivatives (4b, 4d, and 4f) have identical chemical shifts, while the pre-calculated chemical shifts sustain the trend.At the same time, for δ 31 P, we observed a significant discrepancy between the pre-calculated values and the experimental ones.This latter fact might be related to the calculation methodology.
A comparison of the experimental and pre-calculated δ values for the hydrogen atoms bonded to the carbon atom in the phosphonate substituent (CHPO 3 ) and for the bridging CH-CH (3a, 6a) group demonstrates the significant discrepancy in these values for the enantiomeric pairs of the SRR and RSS (4 ) types and indicates the values' precision for the RSR and SRS (4 ) pairs.For example, the experimental δ of the carbon proton (CHPO 3 ) and CH-CH protons (3a, 6a) in the compound of group 4a equal 5.85, 5.50, and 5.00 ppm, whereas the pre-calculated ones equal 6.57, 5.20, and 2.74 ppm.This discrepancy might be caused by the fact that the most stable pre-calculated conformations of structure 4a (SRR and RSS) contain the phenyl rings located relative to the glycoluril fragment, thus creating anisotropy cones and shielding the farthest methine proton in the bridging group (CH-CH).At the same time, the hydrogen atom in the methine group of the phosphonate substituent remains in the de-shielding region (Figure 4).This effect is not observed in the experimental 1 H NMR spectra, which might be linked to the impossibility of tracking the solvating effect of the solvent using ab initio methods and to the respective change in the structure conformations in the actual solution.The observed effect of phenyl rings on the SRR-and RSS-type enantiomers is retained for the OH and NO 2 derivatives (4b -4f ).
The 2D correlation experiment (NOESY) was performed for compounds 4a and 4a .The interactions between the protons of the 3a, 6a, and CHPO 3 sites were identified for compound 4a , while these interactions were not detected for compound 4a (Supplementary Materials, Figures S35 and S36).
Int. J. Mol.Sci.2023, 24, 17082 9 of 16 5.00 ppm, whereas the pre-calculated ones equal 6.57, 5.20, and 2.74 ppm.This discrepancy might be caused by the fact that the most stable pre-calculated conformations of structure 4a″ (SRR and RSS) contain the phenyl rings located relative to the glycoluril fragment, thus creating anisotropy cones and shielding the farthest methine proton in the bridging group (CH-CH).At the same time, the hydrogen atom in the methine group of the phosphonate substituent remains in the de-shielding region (Figure 4).This effect is not observed in the experimental 1 H NMR spectra, which might be linked to the impossibility of tracking the solvating effect of the solvent using ab initio methods and to the respective change in the structure conformations in the actual solution.The observed effect of phenyl rings on the SRR-and RSS-type enantiomers is retained for the OH and NO2 derivatives (4b″-4f″).The 2D correlation experiment (NOESY) was performed for compounds 4a′ and 4a″.The interactions between the protons of the 3a, 6a, and CHPO3 sites were identified for compound 4a′, while these interactions were not detected for compound 4a″ (Supplementary Materials, Figures S35 and S36).

Materials and Methods
All reagents (Acros Organics, Merck, Rahway, NJ, USA) were used as purchased, unless indicated otherwise.Solvents for HPLC (PanReac AppliChem, Darmstadt, Germany) were used for the reactions, benzene was dried over molecular sieves 3A for 48 h.1,3,4trimethylglycoluril was prepared according to a known procedure [24] and used as a racemate. 1

General Procedure for the Synthesis of Compounds 4a-4f
Aldehyde 2 (2 mmol), triphenyl phosphite 3 (2 mmol), and methanesulfonic acid (10 mol%) were added to the suspension of 1,3,4-trimethylglycoluril 1 (2 mmol) in dry acetonitrile (4 mL) in an argon atmosphere.The mixture was stirred for 1 h at room temperature and then distilled on a rotary evaporator.The residue was dissolved in 10 mL of toluene and washed with 5% aqueous Na 2 CO 3 solution (3 × 10 mL).The toluene solution was then washed with water (3 × 10 mL).Upon washing, the organic layer was dried over Na 2 SO 4 and concentrated on a rotary evaporator.Concentrated viscous oil was purified by preparative HPLC with C18 stationary phase and water-acetonitrile (60:40) mobile phase.

NMR Spectroscopy
1 H, 13 C, and 31 P NMR spectra were recorded using the Bruker AVANCE III HD 400 MHz NMR spectrometer (Brucker BioSpin GmbH, Ettlingen, Germany) with the PA BBO 400S1 BBF-H-D-05 Z SP probe head and the BCU temperature control unit, PLC on TTY1 of ELCB 1 autosampler, and TopSpin 3.

X-ray Crystallography
The X-ray diffraction data were collected at 150 • K with the Bruker D8 Venture diffractometer (Bruker Optik GmbH, Ettlingen, Germany) (0.5 • ωand ϕ-scans, fixedχ three-circle goniometer, CMOS PHOTON III detector, Mo-IµS 3.0 microfocus source, focusing Montel mirrors, λ = 0.71073 Å MoK α radiation, N 2 -flow thermostat).Data reduction was performed via the APEX 3 suite.The crystal structure was solved using the ShelXT [25] and was refined using the ShelXL [26] programs assisted by the Olex2 GUI [27].The atomic displacements for the non-hydrogen atoms were refined with harmonic anisotropic approximation.Hydrogen atoms were located geometrically and refined in the riding model.Table 4 represents the crystallographic parameters for compounds 4a and 4a .

HRMS
The analysis was performed using the HPLC-MS method in the electro-spray ionization mode using the Agilent QTOF-6550 (Agilent Technologies, Santa Clara, CA, USA) connected with Agilent 1260 Infinity II (Agilent Technologies, Santa Clara, CA, USA).The samples were analyzed with the Poroshell 120 EC-C18 column 2.1 × 100 mm, 2.7 µm particle size.The column temperature was set as 30 • C (±0.4 • C).The chromatographic analysis was performed in isocratic mode.The 0.05% formic acid (Honeywell Fluka, Charlotte, NC, USA) aqueous solution-acetonitrile mixture (60:40) was applied as a mobile phase.The flow rate was 0.5 mL/min.The samples were analyzed in a positive ionization mode with a scan range of 300-900 Da.Capillary and nozzle voltages were set to 2000 and 100 V, respectively.Drying gas temperature was 200 • C, and sheath gas temperature was 250 • C. The drying gas flow was 13 L/min, and the sheath gas flow was 11 L/min.The samples were preliminarily dissolved in acetonitrile (1:1000, v/v).Figures S37-S47 (Supplementary Materials) show the preparative elution profiles of the compounds 4a-4f.funding acquisition, V.P.T., S.I.G., A.A.B. and V.S.M.All authors have read and agreed to the published version of the manuscript.

Figure 1 .
Figure 1.Molecular X-ray crystal structures for compounds 4a (a) and 4a (b).Thermal ellipsoids are drawn for 50% probability level.Solvate acetonitrile molecules in panel (a) are omitted for clarity.The molecules of 4a of the same enantiomeric form are joined into homochiral supramolecular chains by short CH•••O contacts involving the oxygen atoms of the phosphonate group and the hydrogen atoms of the phenyl (d(O3-H22) = 2.60 Å) or glycoluril fragments (d(O2-H10) = 2.48 Å, Figure 2a).The second type of CH•••O contacts involves the carbonyl oxygen atom and aliphatic hydrogen atoms with the distances d(O4-H7) = 2.43 Å and d(O4-H10) = 2.40 Å, respectively.The chains are oriented along the crystallographic axis b.The enantiomeric chains of opposite stereochemistry are joined in the crystal packing by π-π stacking interactions between the phenyl groups of the phosphonate ester moiety with the distance of 3.295 Å between the planes of the rings (Figure 2b).Compound 4a crystallizes in a monoclinic centrosymmetric space group, P2 1 /c.In contrast to compound 4a , there are no solvate molecules in the crystal structure, and the asymmetric unit consists of one molecule (Figure1b).The unit cell contains four formula units.Similarly to compound 4a , the enantiomeric pair of 4a crystallizes as a true racemate.The configuration of the stereocenters (3a, 6a, CHPO 3 ) is S,R,R (and R,S,S).Therefore, compounds 4a and 4a differ in the absolute configuration of the carbon atom connected to the phosphonate group, in accordance with the two possible directions of the 1,3,4-trimethylglycoluril addition to the carbonyl group of the aldehyde.Similarly to compound 4a , the enantiomers of 4a of the same configuration are joined into supramolecular chains oriented along the crystallographic axis b (Figure3a), but the short CH•••O contacts of the phosphonate groups involve only the hydrogen atoms of the phenyl rings (d(O2-H6) = 2.53 Å and d(O3-H15) = 2.65 Å).The carbonyl oxygen atoms form only one type of short CH•••O contacts with the hydrogen atoms of the glycoluril fragment (d(O4-H9) = 2.65 Å).In the crystal packing, the enantiomers of 4a are joined only by C-H•••π (d(C19-H25C) = 2.76 Å) contacts, and no other specific supramolecular interactions were found.

Figure 4 .
Figure 4. Location of anisotropy cone of phenyl rings in the structure of SRR diastereomer of 4a″.

Figure 4 .
Figure 4. Location of anisotropy cone of phenyl rings in the structure of SRR diastereomer of 4a .

Table 1 .
The effect of a catalyst on the maximum analytical yield of 4a in the reaction mixture at room temperature in acetonitrile.

Table 2 .
The effect of a solvent on the maximum analytical yield of the target product 4a (methanesulfonic acid as catalyst (10 mol%), room temperature).

Table 3 .
The comparison of experimental and pre-calculated chemical shifts for hydrogen atoms of the 3a, 6a, and CHPO 3 carbon atom of the compounds 4a