Synthesis and characterization of solid 2-methoxycinnamylidenepyruvic acid

The 2-methoxycinnamylidenepyruvic acid (2-MeO-HCP) was synthesized and characterized for nuclear magnetic resonance (1H and 13C NMR), mass spectrometry (MS), Infrared spectroscopy (FTIR) and differential scanning calorimetry (DSC). The application of DSC for purity determination is well documented in literature and is used in the analysis of pure organic compounds. The molecular geometry and vibrational frequencies of 2-MeO-HCP have been calculated.


Introduction
Various procedures concerning the preparation of α-keto acids have been reported in the literature. 1 The α-keto acids are of continuing interest as model substrates of enzymes, intermediates in chemical syntheses, in the development of enzyme inhibitors and drugs and in other ways. 1 The aldol condensation products between cinnamaldehyde and pyruvic acid as, 4-dimethylamino and cinnamylidenepyruvic acids has also been described in the literature 2,3 .
Extensive experimental and theoretical investigations have focused on elucidating the structure and normal vibrations of organic compounds 4 .Thus, calculations of optimal molecular geometry and vibrational of 2-MeO-HCP were made.The calculated harmonic frequencies are usually higher than the corresponding experimental quantities, due to a combination of electron correlation effects and basis set deficiencies.It is well known that Hartree-Fock (HF) method tends to overestimate vibrational frequencies.However, density functional theory (DFT) calculations are reported to provide excellent vibrational frequencies of organic compounds if the calculated frequencies are scaled to compensate for the approximate treatment of electron correlation for basis set deficiencies and the anharmonicity. 5n this paper, the 2-MeO-HCP was investigated by means of infrared spectroscopy (IR), mass spectrometry (MS), nuclear magnetic resonance (NMR) spectroscopy and differential scanning calorimetry (DSC).The results allowed us to acquire information about these compounds in the solid state.The DSC purity determination method was established using melting point (122 ºC) observations revealing a relation between a substance's melting point and its purity 6 .
The optimisation of the DSC purity determination method was done by Blaine et al. 7 using a Nacional Institute of Standards and Technology (NIST) purity standard -phenacetin doped with p-aminobenzoic acid -in 1984.As a result of their investigations, the experimental parameters influencing purity results were assessed: specimen size, heating rate, level of impurity.An optimal specimen size (1.7 mg) and heating rate (0.5-2 °C min -1 ) were deduced.
The 2-methoxycinnamaldehyde, (CH 3 O-C 6 H 4 -(CH) 2 -CHO, 96%) predominantly trans, was obtained from Aldrich and sodium pyruvate (H 3 C-CO-COONa, 99%) was obtained from Sigma.2-methoxycinnamylidenepyruvic acid (2-MeO-HCP) was synthesized following the same procedure as described in literature, 8 with some modifications, as follows: an aqueous solution of sodium pyruvate (1.5 g per 10 mL) was added dropwise with continuos stirring to a methanolic solution of 2-methoxycinnamaldehyde (2.0 g per 50mL).Five millilitres of an aqueous sodium hydroxide solution 1.25 mol L -1 was added slowly while the reacting system was stirred and cooled in an ice bath.The rate of addition of alkali was regulated so that the temperature remained between 5 and 9 °C.The system was stirred at ambient temperature (~28 °C) for about 5 h.To the pale yellow solution was added dropwise with continuous stirring with a glass rod, twenty millilitres of chilled concentrated (12 mol L -1 ) hydrochloric acid.The system was left to stand for ca.16 h in a freezer (-6 °C) and the yellow orange precipitate (impure 2-MeO-HCP) was filtered, washed with distilled water to remove most of the unreacted aldehyde and secondary products and dried on Whatman nº 44 filter paper.The aqueous suspension of the impure acid was dissolved with 5 mL of aqueous sodium hydroxide solution 2 mol L -1 and filtered on Whatman nº 44 filter paper.The yellow solution was stirred with a glass rod and added slowly 20 mL of chilled concentrated (12 mol L -1 ) hydrochloric acid and left to stand for ca.16 h in freezer (-6 °C).The orange precipitate of 2-MeO-CP was filtered and washed with distilled water until elimination of chloride ions and dried on Whatman nº 44 filter paper and kept in a desiccator over anhydrous calcium chloride.(yield = 28,6%).

General methods
Infrared spectra for 2-MeO-HCP were run on a Nicolet Model Impact 400 FTIR instrument within the 4000-400 cm -1 range.The solid samples were pressed into KBr pellets.DSC curve were obtained with DSC Q10 from TA Instruments.The purge gas was an air flow of 50 mL min -1 .A heating rate 0.5 K min -1 for the DSC was adopted, with samples weighing about 1,5 mg.Aluminium crucibles with perforated covers were used for the DSC.
High resolution mass spectra with ESI ionization was measured on UltrO-TOF instrument (Bruker Daltonics) operating in positive mode.MeOH/H 2 O (4:1) was applied as solvent system.

Purity determination
The determination of purity is based on the assumption that an impurity will depress the melting point of a pure material whose melting is characterized by a melting point (T 0 ) and an enthalpy of fusion (∆H fus.).The effect of an impurity on T o of the 2MeO-HCP was determined by DSC method basing on the Van't Hoff equation ( 1). ( Where T s is the sample temperature at equilibrium (K), T o is the melting point of the pure component (K), R is the gas constant, x is the concentration of impurity (mole fraction) and F is the fraction molten at T s .
The obtained DSC curve exhibits the endothermic event corresponding to 2-MeO-HCP melting point (∆H fus = +23.36kJ mol -1 ).The value of purity found was confirming low impurity content.

Computational Strategy
In the present investigation, the employed quantum chemical approach to determining the molecular structures was Becke three-parameter hybrid method 9 using the Lee-Yang-Par (LYP) correlation functional 10 and the basis sets used for calculations were the 6-311G. 11,12The performed molecular calculations in this work were done using the Gaussian 03 routine. 13olecular structure of the compound could not be determined by the single crystal Xray diffraction technique, a geometry optimization was computed using the optimized algorithm of Berny. 14heoretical infrared spectrum was calculated using a harmonic field 15 based on C 1 symmetry (electronic state 1 A).Frequency values (not scaled), relative intensities, assignments and description of vibrational modes are presented.The calculations of vibrational frequencies were also implemented to determine an optimized geometry constitutes minimum or saddle points.The principal infrared-active fundamental modes assignments and descriptions were done by the GaussView 3.0 graphics routine. 16he optimized structure parameters of 2-MeO-HCP dimer calculated by ab initio DFT-B3LYP levels with the 6-311G basis set are listed in Table 2 in accordance with the atom numbering scheme given in Figure 2.
Optimized geometries and vibrational modes for studied molecular structures were compared with the experimental frequencies.The optimized geometric parameters (bond lengths and bond angles) obtained, see Table 2, did not can be compared with experimental parameters because the crystal structure of the title compound is not available until now.The Mulliken atomic charges, dipole moment and theoretically computed energies are gather in Table 3. Taking into account that the molecular geometry of the dimer alone in vapor phase may be different from that dimer in the solid phase, owing to extended hydrogen bonding and stacking interactions there is reasonable agreement between the calculated and experimental geometric parameters.As discussed by Johnson et al. 17 , DFT method predicts bond lengths which are systematically too long, particularly the CH bond lengths.The experimental and theoretical IR spectra of 2-MeO-HCP are given in Figure 3.The resulting vibrational frequencies for the optimized geometries and the proposed vibrational assignments are given in Table 4. Comparisons between theoretical and experimental IR spectra indicated that vibrations were more intense in experimental than theoretical spectrum.O-H stretching band is characterized by very broad band appearing near about 2500-3600 cm -1 .This may be due combined effect of intermolecular hydrogen bonding.The O-H in plane bending vibration occurs in general at 1440-1395 cm -1 .The O-H out of plane bending vibration occurs in 960-875 cm -1 . 18In 2-MeO-HCP, O-H stretching were assigned at 3309 and 3358 cm -1 , and in plane and out of plane bending vibrations were assigned at 1385 and 957 cm -1 , respectively.Theoretically computed values (3300, 1400 and 970 cm -1 ) were in agreement with experimental results.

Bond lengths(Å)
The main feature of carboxylic group is a single band observed usually in the 1700-1800 cm -1 region. 18This band is due to the C=O stretching vibration.Theoretically computed value of C=O (1750 cm -1 ) band show is agreement with experimental data (1760 cm -1 ).Experimental vibration frequencies were in agreement with that theoretically calculated, see Table 4.

Conclusion
In conclusion, 2-MeO-HCP purity was determined for DSC and structural elucidation was made by MS, NMR and IR.Theoretical infrared spectra was calculated and compared with experimental data.The difference between the observed and scaled wavenumbers values of most of the fundamentals is very small.Any discrepancy noted between the observed and the calculated frequencies may be due to the fact that the calculations have been actually done on a single molecule contrary to the experimental values recorded in the presence of intermolecular interactions.Therefore, the assignments made at higher levels of theory with only reasonable deviations from the experimental values, seem to be correct.

For 2 -
MeO-HCP group the vibrational modes are C-H stretch, C=O stretch, O-H stretch, O-CH 3 stretch, CH 3 O-C ar stretch, C-C stretch, C co -C coo stretch, C-C-C bending, CH 3 bending, C-H bending, COO bending and ring bending.