Investigation of naproxen drug using mass spectrometry, thermal analyses and semi-empirical molecular orbital calculation

2.1 Materials

All chemicals used were of analytical reagent grade (AR), and of highest purity available. They included naproxen (NAP, M.wt = 230.26 g mole), as an authentic sample which was kindly supplied by the Egyptian Drug Control Authority (EDCA), Cairo (Egypt).

2.2 Mass spectrometry (MS)

Electron ionization (EI) mass spectrum of naproxen is obtained using Thermo Finnegan TRACE DSQ quadruple mass spectrometer with electron multiplier detector equipped with GC–MS data system at the Cairo University micro-analytical center. The direct probe (DP) for solid material was used in this study. The sample was put into a glass micro vial, by a needle (≈1 μg max), the vial was installed on the tip of the DP containing heating cable and inserted into the evacuated ion source. The sample was ionized by an electron beam emitted from the filament, the generated ions being effectively introduced into the analyzer by the focusing and extractor lenses system. The MS was continuously scanned and the obtained spectra were stored. Electron ionization mass spectra were obtained at ionizing energy value of 70 and 12 eV, ionization current of 60 μA and vacuum is better than 10⁻⁶ torr.

2.3 Thermal analyses (TA)

The thermal analyses of naproxen drug were performed using conventional thermal analyzer (Shimadzu system of DTA-50 and 30 series TG-50) at the Cairo University micro-analytical center. The mass losses of 5 mg sample and heat response of the change of the sample were measured from room temperature up to 600 °C. The heating rate, in an inert argon atmosphere, was selected as 10 °C min^–1. These instruments were calibrated using an indium metal as a thermal stable material. The reproducibility of the instrument reading was determined by repeating each experiment more than twice.

2.4 Computational method

The MO calculations were performed using semi-empirical molecular orbital calculation. The method used in these computations is the parametric method (PM3) described by Stewart (1989). The default criteria for terminating all optimizations were increased by a factor of 100 (keyword PRECISE). Vibrational frequencies were computed for the studied structures (keyword FORCE) so as to check whether the newly designed geometries are local minima. All the molecular orbital calculations were carried out at the restricted Hartree–Fock level (RHF) for the neutral molecule of naproxen while the unrestricted Hartree–Fock level (UHF) was carried out for its cation by using the PM3 method followed by full optimization of all geometrical variables (bond lengths, bond angles, and dihedral angles), without any symmetry constraint. All structures were optimized to a gradient norm 0.01–0.05, using the eigenvector following the EF routine (Baker, 1986). All the semi empirical MO calculations were performed with the MOPAC2000 software package (Stewart, 1999) implemented on an Intel Pentium IV 3.0 G Hz computer.

3.1 Thermal analysis

The TA data of naproxen (NP) are illustrated in Fig. 2a and b. It is clear from TG/DTG curves (Fig. 2a) that, this compound decomposed completely within the temperature range of 80–480 °C (mass loss = 99.0%). The main mass loss of this compound occurs at 266.93 °C as shown by DTG curve (Fig. 2a). From DTA curve (Fig. 2b) it is clear that thermal decomposition of NAP occurs in two main endothermic regions and one exothermic region: 150–170, 210–280 and 400–480 °C which cannot easily be described by TG technique. These endothermic and exothermic mass losses required energy values of 255.41, 10.67 and 371.45 kJ mol⁻¹ at 157.0, 270.0 and 452.0 °C respectively. Therefore, the proposed thermal decomposition of NAP can be carried out in three consecutive steps as shown in Scheme 1.

Close

Figure 2

Thermal analyses of standard NAP drug: (a) TG/DT and (b) DTA.

Export to PPT

Close

Scheme 1

The proposed thermal decomposition pathway of NA.

Export to PPT

3.2 Mass spectral (MS) fragmentation of Naproxen drug

The electron ionization (EI) mass spectrum for NAP drug measured at 70 and 12 eV was recorded (Fig. 3a and b) and investigated. The spectrum of NAP drug at 70 eV (Fig. 3a) is characterized by many competitive and consecutive pathways, thus forming many intense fragment ions (Scheme 2). The main fragmentation pathways for NAP after ionization of neutral molecule at 8.764 eV consist of three principal pathways (path 1–3) as rationalized in Scheme 2. The signal that appears at m/z = 230 (RI = 100%) refers to the appearance of the main molecular ion [C₁₄H₁₄O₃]⁺. The high intensity reflects the stability of the molecular ion of NAP. The fragment ion at signal m/z = 185 (Scheme 2, path 1) represents the second most prominent ion [C₁₃H₁₃O]⁺ (RI = 90.42%) and is mainly due to the rupture of COOH molecule from main molecular ion. Also, its high stability is due to the presence of the lone pairs of electrons of oxygen atom with two aromatic rings. Two important fragment ions are observed in the mass spectra at m/z = 170 (RI = 26.65%) and at m/z = 154 (RI = 12.85%). These fragment ions may be due to the rupture of CH₃ and OCH₃ radicals, respectively. The fragment ion observed at m/z = 139 may be due to the formation of [C₁₁H₇]⁺. Comparison of TA (Scheme 1) and MS path 1 (Scheme 2) refers to the coincidence between TA scheme and MS of NAP drug metabolites (vitro fragments).

Close

Figure 3

Mass spectrum of NAP at: (a) 70 eV and (b) 12 eV.

Export to PPT

Close

Scheme 2

Proposed mass fragmentation of NAP drug in cationic form at 70 eV.

Export to PPT

At the low energy value of 12 eV (Fig. 3b), the base peak is observed at molecular ion m/z = 230, IR = 100%. It is noted that the peaks corresponding to the secondary process (m/z = 185) were observed at RI < 10%. Comparison of the data in Fig. 3a to that in Fig. 3b, refers to the fact that 70 eV is sufficient energy for fragmentation of NAP (mole mass 230) to give its daughter fragment ion [C₁₃H₁₃O]⁺ (m/z = 185, RI = 90.42%), but 12 eV the lower energy is insufficient to give this fragment from its parent drug molecule. This also confirms the proposed mass Scheme 2, which indicates that the path 1 is the only possible way to form [C₁₃H₁₃O]⁺ at 70 eV. This is also confirmed by the full scan MS/MS spectrum (Fig. 4) (da-fang et al., 2003) of NAP that showed a molecular ion at m/z = 229.1 of very low intensity, whereas its daughter fragment ion [C₁₃H₁₃O]⁺ (m/z = 184.9, RI = 100%) appeared as base peak.

Close

Figure 4

Full scan MS/MS spectrum of m/z = 230 of NAP (Aresta et al., 2006).

Export to PPT

3.3 Computational molecular orbital (MO) calculation

Molecular orbital (MO) calculation gives valuable information about the structure and reactivity of the molecules, which actually are used to support the experimental evidences (Fahmey et al., 2001; Zayed et al., 2005, 2006, 2007, 2010, 2012a,b). The most important parameters calculated using MO calculation includes bond orders, bond length, charge distribution, bond strain, and heat of formation and ionization energy. In the present work, the calculations have been carried out on NAP neutral molecule (related to TA decomposition) and charged molecular ion (related to MS fragmentation) which is used for prediction of weakest bond rupture to follow the fragmentation pathways in both techniques.

Fig. 1 shows the numbering system of NAP skeleton that helps in ordering the calculated parameters Table 1 presents the values of bond length (Å), bond order and bond strain (kcal mol⁻¹) of NAP drug in both neutral and in cationic forms. One can conclude the following from Table 1:

1. Small differences in bond length in NAP system upon ionization indicate that no appreciable change in the geometries upon ionization.

2. The lowest bond order (important for prediction of primary site of cleavage) observed at bond C17–C21 for both neutral (0.919) and positive species (0.932).

3. Upon ionization the stability of the molecule decreased by 180.53 kcal mol⁻¹ (ΔH_f (−97.79) −  $\mathrm{\Delta }{H}_{\text{f}}^{+}$ (92.74)).

The charge distribution on different atoms (C and O) and heats of formation; ΔH_f (kcal mol⁻¹) for neutral and charged NAP species are summarized in Fig. 5. Significant changes in the electron distribution with given system often takes place during the ionization.

Close

Figure 5

Charge distribution on different atoms for NAP: (a) neutral molecule and (b) molecular cation.

Export to PPT

3.4 Correlation between thermal analysis (TA) decomposition and MO-calculation

As indicated, the determination of initial bond rupture would be an important first step in using this calculation in predicative manner. MO calculation and PM3 procedure revealed that the C17-C12 bond has the lowest bond order at 0.919 and large bond length = 1.521 Å and bond strain = 0.032 kcal mol⁻¹. Experimental TA curves (Fig. 2) of naproxen reveal that the first weight loss equals to 19.56%. This weight loss corresponds to the rupture of COOH (lowest bond order) at temperature 157 °C of range 150–170 °C, followed by the rupture of C3–O13 bond (bond order = 1.037, bond length = 1.364 Å, bond strain = 0.032 kcal mol⁻¹). The second weight loss of 13.65% occurs at temperature in the range 180–280 °C. This weight loss is mainly due to rupture of OCH₃ (O + CH₃). This wide range of temperature may be due to slow fragmentation with small half life time (Pourmortazavi et al., 2008), since in TA decomposition of the molecules is continuously energized and determined by gas evolution and the distribution of energy can be described by energy (Hosseini et al., 2005).

On the other hand, the electrostatic repulsion between C17 (−0.012) and C21 (−0.386) for neutral molecule facilitates the rupture of this bond (C17–C12), than the second rupture of C3 (0.100) and O13 (−0.187) atoms, indicating that the COOH loss is more easy than O–CH₃ loss.

3.5 Correlation between mass spectral (MS) fragmentation and MO calculations of charged molecule

The scope of this investigation is restricted to a search for prediction of the first and subsequent bond ruptures during the course of fragmentation of NAP drug in MS technique. The subsequent fragmentation in MS is determined to a large extent by the initial bond rupture of molecular ion (Zayed et al., 2007; Stewart, 1989, 1999; Baker, 1986). A number of mass spectrometric techniques have utilized helping in rationalized the correct pathways of the molecules, among which are: threshold measurement and metastable abundance ratios (Cooks et al., 1973). On the other hand, computational process can provide important information which can be used successfully in description of primary site of cleavage. The theoretical data obtained can be considered valuable for MS; because they were studied in gas phase species, which can be handled much more easily by quantum chemistry (Stewart, 1989, 1999; Baker, 1986). Mass spectrum of NAP reveals (Scheme 2) three competitive and consecutive fragmentation pathways.

PM3 procedure on cationic form (Table 1) reveals that the C17–C21 bond is the first site of bond cleavage (lowest bond order = 0.932, large bond length = 1.520 Å, bond strain = 0.037 kcal mol⁻¹). This bond cleavage accompanied by rupture of COOH molecule forms the fragment ion [C₁₃H₁₃O]⁺ at m/z = 185 and relative intensity = 90.42%. Further loss is a rupture of CH₃ molecule from this fragment at m/z = 170 [C₁₂H₁₀O]⁺ (Scheme 2, path 1). On the other hand, the electrostatic repulsion between the charge localized on C17 (−0.057) and C21 (−0.392) atom (Fig. 5) facilitates the loss of COOH molecule.

The MO calculation data of fragment ion [C₁₃H₁₃O]⁺ at m/z = 185 are shown in Table 2. These data refer to the ordering of its bond strength; which is given by C9–C10 > C1–C2 > C3–C4 > C8–C17 > C7–C8 > C6–C7 > C4–C5 > C2–C3 > C5–C6 > C6–C1 > C10–C5 > C8–C9 > C3–O13 > C17–C20 > O13–C19. This means that, the first ruptured bond is O13–C19 leading to the formation of fragment ion m/z = 170 [C₁₂H₁₀O]⁺ via the loss of CH₃ group. It was followed by the rupture of the bond C17–C20; which leads to the formation of m/z = 154 (RI = 12.85%) via the loss of CH₃O and finally to the rupture of the bond C3–O13 leading to the formation of fragment ion [C₁₁H₇]⁺ that observed at m/z = 139. This means that all fragment ions are coming from the daughter m/z = 185 not from the parent drug m/z = 230. This is also is in good agreement with the proposed thermal and mass schemes (1 and 2) and MS/MS data (da-fang et al., 2003).

3.6 Correlation between TA and MS

It is important to make a discussion between results of TA and MS of NAP, to see the behavior of the drug in both techniques. This comparison shows the agreement between mass path 1 and TA. In both TA and MS techniques, it is proved that the C17–C21 bond is the first site of rupture. In TA, COOH rupture is followed by OCH₃ (O + CH₃) (Scheme 1). In MS fragmentation it is initiated by COOH rupture and followed by CH₃ group rupture in path 1 (Scheme 2). The obtained fragment ions in MS of m/z = 230, 185, 170, 154 and 139, that are confirmed by TA. This can be considered as a vitro metabolites of NAP drug; which are very similar to vivo urinary metabolites of naproxen obtained by liquid chromatography–electro-spray mass spectrometry (Aresta et al., 2006).