Thermodynamics-antioxidant activity relationships of some 4-benzylidenamino-4, 5-dihydro-1h-1,2,4-triazol-5-one derivatives: Theoretical evaluation

ABSTRACT

Density functional methods were used to predict the antioxidative efficiency of thirteen 4-benzylidenamino-4, 5-dihydro-1H-1,2,4-triazol-5-one derivatives in the gas phase and in the solution phase (water and benzene). Optimized geometries of molecules and reaction thermodynamic energies (enthalpies and reaction-free energies) of three main antioxidant mechanisms (hydrogen atom transfer, single electron transfer-proton transfer, and sequential proton loss electron transfer) were studied at B3LYP/6-31G (d,p) level. Solvent contributions to thermodynamic energies were computed employing integral equation formalism integral equation formalism polarized continuum model method. Obtained results revealed that the three main working mechanisms were endothermic, but not spontaneous especially in the gas phase. We found that the single electron transfer process from the anionic form was more preferable than that from the neutral form in the gas phase. The comparison of the ionization potentials of 4-benzylidenamino-4, 5-dihydro-1H-1,2,4-triazol-5-one derivatives to those of classical antioxidants (gallic acid, caffeic acid, ferulic acid, and ascorbic acid) indicated that the electron transfer mechanism was more predominant in the thirteen 4-benzylidenamino-4, 5-dihydro-1H-1,2,4-triazol-5-one derivative compounds. Thermodynamically, single electron transfer process from the anionic form was the most preferable mechanism in the gas phase. Solvent effect drastically modified thermodynamic energies of mechanisms. The proton transfer process was the thermodynamically favored mechanism as compared to other mechanisms in both solvents. It is worth mentioning that all the mechanisms were found not to be spontaneous in the solution phase except the proton transfer process.

KEYWORDS:

• 4-benzylidenamino-4
• 5-dihydro-1H-1
• 2
• 4-triazol-5-one derivatives
• Antioxidant activity
• HAT
• SET-PT
• SPLET

Introduction

The production of the excess of reactive species (reactive oxygen species [ROS], reactive nitrogen species [RNS], and reactive chlorine species [RCS]) in human organisms which react with the DNA, proteins, and other cellular components pose a health hazard.^[1] A significant destabilization of the DNA molecules is then expected at the side of its lesion.^[2,3] This destabilization results from a localized perturbation of the tacking forces responsible for adhesion with water molecules (throughout hydrogen bonds) and positive ions like Na⁺ surrounding the DNA double helix.^[3,4] In addition, ROS and RNS can attack various key cellular proteins like those participating in the DNA repair and cell cycle control.^[5] This latter phenomenon is usually underestimated in the analysis of chronic exposure to oxidative stress.^[1] The oxidative stress to the organism can be intensified beyond the basal level by a variety of endogenous and exogenous factors.^[6] Free radicals derived from circulating catecholamines are one possible endogenous source.^[7] The antioxidant defense system appears to be able to react adaptively in response to oxidative challenges, subject to regulation by the availability of other antioxidant compounds.^[8]

This fact is the justification of the considerable effort to find out of novel antioxidant compounds. The results obtained have shown that the antioxidant therapies have been recognized as strategies for preventing degenerative diseases.^[9–11] The applications of antioxidants as preservatives in the food industry^[12] and skin-protective ingredients in cosmetics^[13] have also been observed. For instance, Citrius flavonoids (dietary polyphenolics) have been detected to be more effective antioxidant in vitro than vitamin E and C, and, therefore, used against cancer cell lines.^[14–16] Phenolic compounds and its derivatives widely present in plants (vegetables, fruits, grains, and spices).^[17] play an important role in the nutrition and the diet. This is the plausible explanation of many experimental^[18,19] and theoretical^[20–27] studies on the antioxidant activities of phenolic acids (PhAs) observed. More recently, Yuzhen et al. succeeded, through an experimental and theoretical evaluation, to establish a relationship between three factors of selective PhAs and its derivatives: structure-thermodynamic-antioxidant activity.^[28] In the same vein, the 1,2,4-triazole derivatives have also been experimentally found out to be a powerful antioxidant with multiple biological activities (antibacterial, antifungal, antitubercular, anti-inflammatory, analgesic, antitumoral, antiviral, and anticonvulsant).^[29–34] However, to the best of our knowledge, no study regarding the comparison of the antioxidant activities of 1,2,4-triazole derivatives to those of PhAs. The survey of the literature showed that few molecules containing simultaneously the 1, 2, 4-triazol ones and 3,4-dihydroxyphenyl moieties have been synthesized.^[33,35] These synthesized chemicals were screened for their antioxidant properties. But this experimental estimation did not give any information on comparison of the contribution of each moiety to the antioxidant activity of these molecules. In our previous works on the 1,2,4-triazole derivatives, the emphasis was put on the theoretical assessment of the chelation of divalent cations by some 4-benzylidenamino-4,5-dihydro-1H-1,2,4-triazol-5-one derivatives (BDHTD) in various media.^[36] The investigation of the complexation of BDHTD by metal cations using the density functional method B3LYP indicated that that k²-O,N structures are most stable in the gas phase. A significant reduces of bond dissociation enthalpies (BDEs) observed when going from isolated BDHTD to complexes obtained confirms the enhancement of the antioxidant activity by the metal chelation.

It is noteworthy to mention that there are three generally accepted mechanisms of antioxidant action.^[20,37] For an antioxidant RX-H, single-step hydrogen atom transfer (HAT) mechanism is described according to Eq. (1; X = O, S, N, and C). A numeral thermodynamic parameter associated with HAT mechanism is bond dissociation enthalpy (BDE).

(1)

The second possible mechanism is single electron transfer-proton transfer (SET-PT) mechanism by which an antioxidant RX-H can deactivate a free radical. In the first step (single electron transfer), the electron leaves the antioxidant molecule RX-H (Eq. [2]). The ionization potential (IP) is the enthalpy required to withdraw its electron.

(2)

Subsequently, the proton is transferred to RX^• formed. The enthalpy change of the reaction (Eq. [3]) is called proton dissociation enthalpy (PDE).

(3)

The sequential proton loss electron transfer (SPLET) is another two-step mechanism described by the following reactions.

(4)

(5)

The reaction enthalpy of the first process is usually denoted proton affinity (PA) of the formed anion. The electron transfer enthalpy (ETE) represents the reaction enthalpy of the second step of the SPLET mechanism.

Here, we clarify the contribution of each moiety of BDHTD using DFT calculations. We have performed a systematic study of reaction thermodynamic energies (enthalpy and free energy) to HAT, SET-PT, and SPLET mechanisms for BDHTD. More precisely, five sets of thermodynamic parameters were estimated. Because of no experimental reaction thermodynamic energies for mechanisms of BDHTD are available; the authors integrated the substitution effect in the analysis. The solvation in two solvents (benzene and water) has been explored. From the thermodynamic point-of-view, the preferred mechanism was identified.

Computational details

All calculations were performed using Gaussian 09 program package.^[38] The geometries of each structure (RX-H, RX^•, RX⁻, and RX-H⁺) was optimized using density functional theory (DFT) with B3LYP^[39] functional without any constraints. Calculations were performed using 6-31G (d,p) basis set in the gas phase and solution phase. In the same vein, the triple zeta basis set 6-311++G(d,p) in which diffuse s and d polarization function was added on hydrogen atom^[40] was used for solvation enthalpies of hydrogen atom or charges. The optimized structures were confirmed to be real minima by frequency calculations. The enthalpies obtained were zero-point corrected with unscaled frequencies. The description of the thermodynamic characteristic of solvation was performed employing an integral equation formalism polarized continuum model (IEF-PCM) method.^[41] The authors have determined the enthalpies of all reactions studies (reactions in Eqs. [1–5]) from the calculated total enthalpies at 298.15 K and 1.0 atmosphere pressure:

(6)

(7)

(8)

(9)

(10)

The calculated gas-phase enthalpy change of proton is () = 6.197 kJ/mol and the gas-phase enthalpy of an electron () is 3.1351 kJ/mol.^[42] Hydrogen atom, proton, and electron solvation enthalpies for studied solvents (water and benzene) were taken from previous researches.^[23,43] From the calculated free energies, we have also determined the following quantities (bond dissociation free energy [BDFE], ionization potential free energy [IPFE], proton dissociation free energy [PDFE], proton affinity free energy [PAFE], and electron transfer free energy [ETFE]) for the reactions in Eqs. (1–5).

(11)

(12)

(14)

(15)

For electron (e⁻) and proton (H⁺) free energy, we employed the values of –3.72^[44] and –26.28 kJ/mol.^[45,46] Hydrogen atom, proton, and electron solvation free energies in studied solvents (water and benzene) were taken from the report by Fifen et al.^[47]

Results and discussion

Geometrical details

The investigation of the substituent effect has forced the author to use several derivatives (A–M) with electron donating or withdrawing substituent on the 1, 2, 4-triazol-5-one ring of the 4-benzylidenamino-4, 5-dihydro-1H-1, 2, 4-triazol-5-ones (A) as shown in Fig. 1. The hypothesis of configurations with both free meta O₁-H₁ and O₂-H₂ bonds of the catechol moiety is excluded because of the zero probability of presence of similar conformations of 3,4-dihydroxyphenylpyruvic acid.^[23] As observed in Fig. 1, the authors only take into account the O₂-H₂…O₁ intramolecular interaction that has showed to be stronger than O₁H₁….O₂ for 3,4-dihydroxyphenylpyruvic acid.^[23] The optimized parameters obtained in various media (gas, water, and benzene) labeled according to the convention given in Fig. 1, are listed in Table 1. Since no experimental data of the molecules have been reported so far, the authors have compared these geometrical parameters with related crystallographically characterized 1-acetyl-4-(p-chlorobenzylideneamino)-3-methyl-4, 5-dihydro-1H-1, 2, 4-triazol-5-one^[48] presented in Fig. 2. One finds that the theoretical bond lengths are in general slightly higher than experimental results. This minor deviation may result from the neglected solvent effect or a slight structure difference between experimental and the author’s data.

Figure 1. Numbering system used to designate specific atom of the 4-benzylidenamino-4, 5-dihydro-1H-1, 2, 4-triazol-5-ones. A and specification of electron donating or withdrawing substituent used.

Figure 2. Structure of 1-acetyl-4-(p-chlorobenzylideneamino)-3-methyl-4, 5-dihydro-1H-1, 2, 4-triazol-5-one used for geometrical comparison.

It can be seen from Table 1 that the substitution effect is minor on the bond distances in various media. But a sensible variation of dihedral angle values can be observed in ESI ‡, Tables 1–3. For instance, the difference of N1-N₂=C₃-C₅ value obtained for A and those of other molecules ranges from 0.6 to 6.4° in gas. This fact exhibits that the rotation about the N₂=C₃ axis leads to a slight staggering of these bonds. The results obtained (ESI ‡, Table 1) revealed that the planarity of the triazole ring is slightly deformed due to the deviation of N₁ ranging 3.6–7.5°. Whereas, the benzene ring in all the optimized structures are planar. This planarity of both rings showed that our optimized geometries is in good agreement with crystallographic data of other triazole derivatives.^[48] The N₃-C₇-N₁-N₂ values in the gas phase show that the atom N₃ is out of the plane defined by C₇, N₁ and N₂ (see Fig. 1, in which N₃ is drawn in the plane). Therefore, the authors concluded that the whole molecule (A–N) do not lie in the same plane. It follows from ESI ‡, Tables 2 and 3 that this fact remains the same in the solution phase. In the whole, the optimization processes were characterized by the formation of O₂-H₂…O₁ hydrogen bond. Such an intramolecular hydrogen bond has been observed in many previous investigations dealing with compounds containing the phenolic moieties.^[18–27] Recently, Yuzhen et al. demonstrated the important contribution of these intermolecular hydrogen bonds on the stability^[28] of the corresponding radicals through a comparison of energy difference of the optimized structures of radicals without hydrogen bond and those with hydrogen bond on the phenolic moiety. ESI ‡, Fig. 1 exhibits different X-H bond distances (X = O₁, O₂, C₃, and N₃) in various media. Higher values are obtained for C₃-H₃ bonds because of captodative effect in which, the cause of a synergistic weakening of the C-H bond the carbon center is the simultaneous substitution by an electron-withdrawing substituent and an electron-donating substituent.^[49] ESI ‡, Fig. 2a shows that the dipole moment of each molecule increases with the increase of the dielectric constant of the solvent. This conclusion corroborates earlier study^[50] that has shown the decreasing of the X-H bonds by the solvent’s polarity and the induced enhancement of the antioxidant activity of molecules. Fifen et al.^[23] have given the explanation that the electrostatic interactions between charged particles are all the weaker as the dipole moment of the molecular system increases. The plausible argument is the hypothesis of another electronic distribution resulting from the solvation of the molecule. These results are in the line with the observations of Clemens et al.^[51] on some organic chromophores. Accordingly, ESI ‡, Fig. 2b shows that the vibrational frequencies of the X-H mode decreases with the increasing of dielectric constant of the solvent. For instance, for I (R = Br), v(O₂-H₂), and v(O₁-H₁) frequencies are 3768 and 3841 cm⁻¹ in vacuum, respectively, compared to 3759 cm⁻¹ and 3834 cm⁻¹ in benzene. Such a trend shows that the identification in the infrared spectra by a red shift of the X-H stretching mode is probable. It is clear that this could be an identification of a good indication of the easier of the hydrogen bond.

Table 1. Relevant B3LYP/6-31G(d,p) geometrical parameters of BDHTD in various media.

	GAS					WATER					BENZENE
	Bond lengths				Bond angles	Bond lengths				Bond angles	Bond lengths				Bond angles
	N1-N2	C2=N3	C3-C5	H2…O1	O2 H2…H1	N1-N2	C2=N3	C3-C5	H2…O1	O2 H2…H1	N1-N2	C2=N3	C3-C5	H2…O1	O2 H2…H1
A	1.411	1.290	1.466	2.115	114.4	1.402	1.293	1.462	2.108	114.0	1.399	1.292	1.462	2.108	114.3
B	1.410	1.289	1.467	2.114	114.4	1.401	1.293	1.462	2.110	114.0	1.412	1.2901	1.466	2.115	114.2
C	1.409	1.289	1.466	2.114	114.4	1.401	1.293	1.462	2.112	113.9	1.398	1.292	1.463	2.109	114.3
D	1.396	1.291	1.463	2.110	114.6	1.402	1.293	1.462	2.112	113.9	1.399	1.292	1.462	2.110	114.2
E	1.409	1.289	1.467	2.114	114.4	1.405	1.293	1.461	2.109	113.9	1.401	1.293	1.462	2.108	114.2
F	1.408	1.289	1.467	2.114	114.5	1.401	1.293	1.462	2.111	113.9	1.398	1.292	1.462	2.111	114.2
G	1.417	1.289	1.465	2.114	114.3	1.406	1.294	1.460	2.108	113.9	1.417	1.290	1.463	2.113	114.1
H	1.399	1.291	1.461	2.109	114.4	1.404	1.294	1.460	2.110	113.9	1.401	1.293	1.461	2.110	114.1
I	1.399	1.291	1.462	2.110	114.4	1.403	1.294	1.460	2.109	113.9	1.401	1.292	1.461	2.109	114.2
J	1.406	1.291	1.464	2.112	114.5	1.399	1.293	1.462	2.110	114.0	1.405	1.292	1.463	2.112	114.2
K	1.416	1.288	1.467	2.115	114.5	1.404	1.291	1.462	2.111	113.9	1.416	1.289	1.466	2.114	114.2
L	1.412	1.290	1.465	2.113	114.3	1.406	1.294	1.460	2.109	113.9	1.403	1.293	1.461	2.109	114.1
M	1.406	1.290	1.463	2.110	114.4	1.409	1.292	1.461	2.110	113.9	1.408	1.291	1.462	2.109	114.1
Exp	1.378 ^[48]	1.271^[48]	1.447^[48]

Table 2. Relevant B3LYP/6-31G(d,p) geometrical parameters of BDHTD in various media.

Gas	BDE1	BDE2	BDE4	IP	PDE1	PDE2	PDE4	PA1	PA2	PA4	ETE1	ETE2	ETE4
A	312	351	380	705	924	962	991	1395	1422	1468	234	246	228
B	312	344	373	694	935	967	996	1400	1424	1478	229	237	212
C	312	344	374	691	937	969	999	1400	1424	1476	228	236	214
D	314	346	375	692	939	970	1000	1402	1426	1478	228	236	214
E	311	343	356	679	949	981	994	1397	1419	1445	231	241	228
F	311	342	354	672	956	987	998	1404	1426	1459	223	233	211
G	315	347	374	721	910	943	969	1392	1415	1450	240	249	240
H	316	349	375	718	914	947	973	1394	1416	1440	238	249	251
I	316	348	374	714	918	950	977	1395	1417	1440	237	248	251
J	312	343	344	673	956	986	988	1410	1429	1488	219	231	173
K	313	345	362	691	939	971	988	1403	1430	1475	227	232	203
L	313	345	378	718	912	944	977	1382	1404	1420	248	258	274
M	315	348	382	710	922	955	990	1394	1417	1423	238	271	276
	BDFE1	BDFE2	BDFE4	IPFE	PDFE1	PDFE2	PDFE4	PAFE1	PAFE2	PAFE4	ETFE1	ETFE2	ETFE4
A	279	315	345	697	893	929	958	1364	1388	1438	233	246	225
B	278	309	337	685	904	935	963	1369	1391	1448	227	236	208
C	279	310	337	683	907	938	965	1369	1392	1446	228	236	210
D	281	311	339	684	908	938	966	1371	1394	1448	228	236	209
E	276	306	321	670	917	948	962	1365	1385	1415	229	239	225
F	277	308	319	662	927	957	969	1371	1391	1430	224	235	207
G	281	313	339	714	878	910	936	1360	1382	1421	239	249	236
H	282	314	340	711	859	915	941	1362	1383	1411	238	250	248
I	283	314	340	708	886	918	943	1363	1382	1410	238	251	248
J	279	309	310	665	925	955	957	1379	1396	1458	219	231	171
K	280	311	326	683	908	940	955	1373	1397	1446	225	232	198
L	279	311	341	708	883	914	945	1352	1371	1391	245	258	268
M	282	314	347	703	890	923	955	1362	1383	1394	238	249	271

Table 3. B3LYP/6-31G(d,p) reaction thermodynamic energies in kJ/mol in gas-phase: reaction enthalpies and reaction free energies.

Benzene	BDE1	BDE2	BDE4	IP	PDE1	PDE2	PDE4	PA1	PA2	PA3	ETE1	ETE2	ETE4
A	317	343	380	608	116	142	179	409	427	465	315	323	322
B	314	339	370	598	123	148	179	406	425	470	314	320	306
C	317	342	374	599	124	150	181	411	429	475	312	320	305
D	317	343	373	599	124	150	180	410	429	472	313	320	308
E	316	342	359	595	128	153	170	409	428	452	314	321	314
F	317	342	354	587	136	162	174	412	431	462	311	318	298
G	316	343	372	618	104	131	160	400	418	445	322	331	333
H	319	346	375	618	107	134	163	403	421	439	322	331	342
I	319	345		616	109	135		404	422	439	321	330
J	314	338	339	575	145	170	170	412	428	479	308	317	266
K	314	340	324	597	124	149	133	408	428	468	313	318	263
L	319	346	381	621	104	131	166	399	417	427	327	336	360
M	319	345	383	620	105	132	169	401	423	424	324	329	365
	BDFE1	BDFE2	BDFE4	IPFE	PDFE1	PDFE2	PDFE4	PAFE1	PAFE2	PAFE4	ETFE1	ETFE2	ETFE4
A	284	309	346	546	-66	-42	-5	229	244	285	250	260	256
B	280	305	334	535	-60	-35	-6	225	243	291	250	257	239
C	286	310	340	537	-56	-32	-2	231	249	298	249	257	237
D	285	310	338	538	-57	-32	-5	232	250	299	249	256	234
E	284	308	326	533	-53	-29	-12	230	247	275	250	257	246
F	284	308	320	523	-43	-20	-8	234	248	285	246	255	230
G	283	309	338	556	-77	-52	-22	219	236	267	260	268	266
H	286	312	341	553	-72	-46	-17	223	240	259	258	267	276
I	285	312		555	-74	-48		224	239	260	256	268
J	283	306	306	514	-36	-13	-13	232	247	301	246	254	200
K	282	307	290	536	-59	-33	-50	225	248	292	252	255	193
L	284	309	343	557	-77	-53	-19	220	234	248	260	270	290
M	286	312	346	558	-77	-51	-17	229	243	246	252	264	295

^aUsed solvation enthalpies:^[23,43] = 6.4 kJ/mol, = 894 kJ/mol, = –7 kJ/mol.

Table 4. B3LYP/6-31G(d,p) reaction thermodynamic energies in kJ/mol in benzene^ab: reaction enthalpies and reaction free energies.

Water	BDE1	BDE2	BDE4	IP	PDE1	PDE2	PDE4	PA1	PA2	PA4	ETE1	ETE2	ETE4
A	329	338	379	443	66	76	116	205	218	245	304	301	314
B	322	338	370	439	63	79	112	206	219	253	296	300	298
C	322	338	371	438	64	80	113	206	219	254	296	299	298
D	322	338	371	438	64	80	113	206	219	253	296	299	298
E	320	336	357	439	61	77	98	203	216	236	297	300	302
F	322	338	351	428	74	90	103	206	216	245	296	302	286
G	324	341	374	452	52	69	103	203	215	232	301	307	323
H	324	341	374	451	53	70	103	203	214	225	302	308	330
I	324	341		450	53	71		203	214	225	301	307
J	321	337	335	408	94	110	107	207	219	259	295	299	256
K	322	339	357	442	61	77	95	206	219	253	297	300	284
L	324	341	380	451	53	70	109	203	214	215	301	307	345
M	304	341	382	452	32	69	111	203	215	212	281	306	351
	BDFE1	BDFE2	BDFE4	IPFE	PDFE1	PDFE2	PDFE4	PAFE1	PAFE2	PAFE4	ETFE1	ETFE2	ETFE4
A	296	304	344	392	-16	-8	32	125	135	165	251	249	259
B	288	304	335	387	-18	-2	29	126	135	172	243	249	244
C	289	305	336	387	-18	-2	28	126	138	173	243	247	243
D	288	305	336	388	-19	-3	29	125	137	173	243	248	243
E	289	305	324	389	-20	-4	15	126	138	158	243	247	246
F	289	304	318	376	-7	8	22	127	143	166	242	241	232
G	291	307	341	401	-30	-14	20	124	131	152	247	256	269
H	290	307	341	399	-29	-12	22	123	132	144	247	256	277
I	290	307		399	-29	-12		123	133	143	247	255
J	287	303	300	354	13	28	25	126	136	178	241	247	202
K	289	305	322	390	-20	-5	12	126	136	172	243	249	230
L	289	305	344	399	-30	-14	24	123	133	134	246	252	289
M	269	306	346	401	-52	-15	25	123	133	131	225	253	295

^aUsed solvation enthalpies:^[23,43] = –4 kJ/mol, = –1022 kJ/mol, = –105 kJ/mol.

^bUsed solvation free energy:^[47] = –27.7 kJ/mol, = –1104.5 kJ/mol, = –156.8 kJ/mol.

Gas-phase reaction thermodynamic energies

Calculated B3LYP/6-31G (d,p) reaction thermodynamic energies in the gas phase are compiled in Table 2. Noticeably, the BDEs of molecules studied can be arranged in the following order: BDE₁ < BDE₂ < BDE₄. It can be deduce that H₁ is easier than the others H atoms. This is in good agreement with prior works on phenolic molecules.^[21,23] The authors attributed it to the fact that H₁ is a free hydrogen atom bond, whereas H₂ is implicated in the O₂H₂…H₁ intermolecular hydrogen. Table 2 clearly shows that the substituent effect is more pronounced for BDE₄ values because of nearness of H₄ atom to substitution side. Nevertheless, the BDE values of H₁ and H₂ are not almost affected by various substitutions on 1,2,4-triazole-5-one moieties. This absence of influence of the activities of O₁H₁ and O₂H₂ groups is due to the fact that the catechol moiety is located further away from the substitution side. Therefore, for sake of simplicity, the authors focused the analysis of the influence of the substituent effect only on thermodynamic energies related to H₄ atom. Our results are in accordance with the literature^[26,52] which revealed that the introduction of electron-donating substituent or electron-withdrawing substituent on the position of hydrogen atom of A (R = H) leads respectively to a decreasing or an increasing of BDE₄ values. An increasing of the BDE₄ values while the number of carbon atom of the akyl substituent increases is noticeable. The minor augmentation of BDE₄ values observed is due to the number of carbon atoms less than five. This fact is consistent with earlier researches^[53,54] that demonstrated that the influence of the side-chain on the antioxidant is more pronounced when the number of carbon atoms of the akyl substituent is exceeding five. For that reason, the steric effect may be integrated in the analysis of the scavenging activity of compounds. The higher reduction of BDE₄ values observed for E and F compared to those of molecules with alkyl substituent (B, C, and D) is due to the extension of conjugation system with the corresponding substituent (CH=C[CH₃]₂ or C₆H₅). These values are also in good accordance with preceding researches that showed that the HAT mechanism is mainly governed by the resonance effect.^[55] This effect is highly enlightened by 1,1-diphenyl-2-picryl-hydrazyl (DPPH) scavenging activity values in N, N-dimethylformamide (DMF) of 0.025 and 0.5 mg/mL for B and E, respectively, experimentally obtained by Yüksek et al.^[34] The slight BDE₄ gap observed between E and F (2 kJ/mol) results from the superposition of resonance effect and the captodative effect^[49] related to the presence of C=C group (in the substituent CH=C[CH₃]₂). This presence of C=C group is shown to favourably stabilize the specific radical species to enhance the radical scavenging activities.^[28] The highest decreasing of BDE₄ is obtained for J (R = NH₂) is attributed to strong electron-donating nature of NH₂. Table 2 also shows that for halogen substituent, the decreasing of BDE₃ values is almost constant. Independently of the substituent used, the BDFE_i values (i = 1, 2, 3) obtained are positive. This enables the authors to conclude that the HAT mechanism is not spontaneous. Table 2 revealed that BDFEs are lower than BDEs. BDFE values are lower by 23–38 kJ/mol. The assumption used by other researchers^[27,43] that the absolute values of entropy term reach few tens kJ/mol is thus confirmed. Therefore, to analyze the HAT mechanism, free energy or enthalpy can be used as the criterion of the thermodynamically preferred process.

Table 2 also shows that IP values are significantly greater than the corresponding BDE values. The HAT mechanism is thermodynamically preferred in gas than SET mechanism. To date, neither experimental nor theoretical study of IP values of BDHTDs have been published. The comparison of the IPs of BDHTD to those of phenolics^[21] (792.9, 762.9, and 742. 8 kJ/mol obtained, respectively, for gallic acid, caffeic acid, and ferulic acid at the B3LYP/6-311++G[d,p] level) indicates that the ET mechanism is more predominant in BDHTD compounds (investigated in this work). The authors also compared the IP values of BDHTD to that of the ascorbic acid^[28] (803.8 kJ/mol obtained B3LYP/6-311++G[2d,2p] /B3LYP/6-31G[d,p]) and concluded that despite the basis set effect, the ET mechanism is also dominant in BDHTD than in ascorbic acid. Our B3LYP results revealed that for halogen substituent (F, Cl and Br), the IP values are higher in comparison to A(R = H). Similar results have observed for electron withdrawing group (COCH₃ and COOH), whereas IP values are lower for electron donating group (akyl group, C₆H₅, NH2, OCH₃ and CH=C[CH₃]₂) in comparison to A(R = H). This trend is good agreement with the findings of Mezsam Najafi^[26] obtained during the theoretical analysis of the antioxidant activity of ortho- and meta-substituted daidzein derivatives in the gas phase. Our results (see Table 2) reveal that the IP values are almost close to IPFE values because of the tendency of the entropic term TΔS to disappear. This fact is in better accordance with the results of Holtomo et al.^[27] on antioxidative potency of caffeic acid phenethyl ester-Fe²⁺ complexes. Table 2 also shows that despite the fact that the IP values calculated from the Koopman theorem (IP = –E_HOMO) are lower compared to those calculated using Eq. (7), the prediction of the ordering of the antioxidant activity for BDHTD using the HOMO eigenvalues is similar to that obtained using IP values calculated from the latter procedure. In the same vein, it can be observed from ESI ‡, Fig. 3S (a) that J (R = NH₂) provided the highest HOMO energy (–541 kJ/mol), followed by F (R = CH=C[CH₃]₂; –551 kJ/mol; ESI ‡, Table 4). This clearly shows that J has the strongest electron-donating capacity among the molecules studied. Such an ordering of the antioxidant activity is reliable with that obtained from BDE₄ values. This parallel has been evoked in prior researches that clearly illustrated the structure-activity relationships of antioxidants.^[23,28,56] The exploration of Table 2 shows that IP values are significantly lower than PDE values in the gas phase. The authors then conclude that the SET mechanism is more dominant than the PT mechanism in this phase. In comparison to A (R = H), it is noticeable that a reduction of PDE values is obtained for halogen substituent and electron withdrawing group (COCH₃ and COOH). But, an augmentation of PDEs is observed for electron donating group (akyl group, C₆H₅, NH₂, and CH=C[CH₃]₂). This trend is contrary to that observed for IPs. ESI ‡, Fig. 3b reveals that the differences between PDE values and those of PDFE are almost similar to those obtained in HAT mechanism. The IPFE and PDFE values obtained are positive pointing out that SET-PT mechanism is not spontaneous in the gas phase.

Figure 3. Thermodynamic energies of BDHTD (in kJ/mol) in gas phase for J molecule (R = NH₂) in water and benzene: BDE₁, IP, PDE₁, PA₁, and ETE₁.

The halogen substituent causes decrease in PA values in comparison to A (R = H). Differences are in 0–28 kJ/mol. In the same vein, PA values are lower than values of A by 1–48 kJ/mol for L(COCH₃) and M(COOH). Our B3LYP results also confirm that electron donating substituents increase PA values. It is worth noting that a charged molecule is more sensitive to the substituent effect than its neutral counterpart. In general, electron donating substituents stabilize the anion formed (RX⁻) but destabilize the parent structures, whereas electron donating groups have an opposite effect.^[26] Table 2 demonstrates that the calculated IPs and PAs of the BDHTD molecules are significantly higher than BDEs illustrating that HAT mechanism represents the most probable process in the gas phase. Such a finding agrees with that reported for polyphenolic deoxybenzoins^[21] and natural PhA derivatives.^[28] In comparison to PA, ETEs of BDHTD molecules are significantly lower than PAs. The presence of electron withdrawing group in A (R = H) lead to higher ETEs (Table 2), while that of the electron donating substituents increases ETE values. This is in line in with the results obtained from previous study on p-phenylenediamine^[43] and substituted daidzein derivatives.^[26] Table 2 also illustrated higher ETE₄ values obtained in the presence of the electron donating substituents. The authors then concluded that the electron transfer process is more favorable on the 1, 2, 4-triazol-5-one ring in the presence of the electron donating substituents. But, reverse results are obtained in the presence of halogen and electron withdrawing substituents. In both cases, ESI ‡, Fig. 4 lucidly exhibits the ETE₄ and ETFE₄ are close due to the tendency of the entropic term TΔS to disappear as observed for IPFEs. From the positive free energy values obtained, one can conclude that the SPLET mechanism is not spontaneous. The comparison of IP and ETE for each substituent (Table 2 and ESI ‡, Fig. 4) demonstrates that the latter is greatly lower than the former. Consequently, the SET process from the anionic form is more preferable than that from neutral form in the gas phase.

Reaction thermodynamic energies in water and benzene

Reaction thermodynamic energies obtained in water and benzene are presented in Tables 3 and 4, respectively. On the whole, BDE₄ values in the solution phase (benzene and water) are lower than that in the gas phase. This decrease is more pronounced in water by 2–13 kJ/mol. The plausible explanation of this observation is the formation of strong hydrogen bond OH₂…O₁ which is weakened more and more when the solvent polarity increases. In reality, the higher the solvent polarity, the easier the separation of charges.^[21,23] Although deviations in individual values can be found, the trend found for both solvents are analogous. In contrast to BDE_1-2 values are higher in water by 8–17 kJ/mol in comparison to their gas-phase homologs, exception made for M (R = COOH). BDFEs displayed in Tables 3 and 4 which are lower than the BDEs, as observed in the gas phase. On average, reaction free energies are higher by 33.1 and 36.6 kJ/mol than corresponding BDEs of BDHTD molecules for water and benzene, respectively. The BDFEs are positive denoting that the HAT mechanism is not spontaneous in the solution phase.

Our calculations showed that solvent effects stronger influence upon IPs than BDEs. The IPs in the solution phase, are lower than those obtained in the gas phase. The drop in IP varies from 84 to 103 kJ/mol in benzene. In water, the IPs are dramatically lower by 240–269 kJ/mol in comparison to those obtained in the gas phase. The average difference in IP is 257.5 kJ/mol in water and 94.8 kJ/mol in benzene. The authors attributed this relative order of IP (water < benzene < gas) to the great sensitivity of cation radicals to the polarity of solvent used. Figure 3 shows that PDEs are significantly decreasing when moving from the gas phase to solution phase for J (R = NH₂). This trend is identical for others molecules. The average differences between the solution phase and gas phase reach 857 and 817 kJ/mol for water and benzene, respectively. The significant drop related to the increase of solvent polarity is attributed to the high solvation enthalpies of proton.^[23,43] This fact is in conformity with that reported polyphenolic deoxybenzoins by Yunsheng et al.^[21] Compared to BDEs, the corresponding PDE values turn out to be drastically low in the solution phase, whatever the molecule analyzed. Tables 2–4 and Fig. 3 displayed IP values that are significantly higher than BDEs. The authors therefore concluded that PT is thermodynamically favored mechanism compared to HAT whereas SET is least preferred mechanism in the solution phase. Overall, the IPFE values are positive illustrating that the SET mechanism is not also spontaneous in the solution phase. But, the PDFE values are negative indicating that the PT mechanism is spontaneous in both solvents, with the exception of PDFE₄ values in water.

Proton affinities in the solution phase are significantly lower than corresponding gas-phase PAs. Differences are ranging: 1179–1225 kJ/mol (in water) and 988–1009 kJ/mol (benzene). The average shift is 1205 and 996 kJ/mol for water and benzene, respectively. These noteworthy drops are also related to the high solvation enthalpies of proton.^[43] Tables 2–4 and Fig. 3 showed that PA_n values are higher than PDE_n values (i = 1, 2, 4) in both solvents. This is in accordance with the fact that cations easily liberate proton than neutral systems. This results are consistent with previous studies on caffeic acid phenethyl ester,^[50] 3,4-dihydroxyphenylpyruvic acid,^[23] and other molecular system.^[56]

In the solution phase, electron transfer enthalpies are higher than corresponding gas-phase values. The differences between ETE in the gas phase and solution phase lie in the 60–157 and 5–112 kJ/mol range, respectively, for benzene and water. This shows that both solvents do not facilitate electron transfer from anionic compounds. Similarly to observation made in the gas phase, a comparison with the solution-phase IP values (Tables 2–4 and Fig. 3) revealed that the corresponding ETE values were greatly lower. Hence, the SET from the neutral form is less favored than that of the anionic form. Such a conclusion is in agreement with results obtained from prior studies.^[21,57] In benzene, the PAFE values are significantly higher than the corresponding PA values by 167–183 kJ/mol, the average shift is approximately 180 kJ/mol. The differences between PA and PAFE values lie in 77–86 kJ/mol intervals in water. On the whole, the solution-phase PAFE and ETE values obtained are positive illustrating that the SPLET mechanism is not spontaneous in both solvents.

Conclusion

In this article, the authors have presented the quantum chemical calculations of the thermodynamic energies (reaction enthalpies and reaction free energies) of antioxidant mechanisms of 13 BDHTD at B3LYP/6-31G (d,p) level in the gas phase and solution phase. The optimization processes were characterized by the formation of an intramolecular hydrogen bond in the phenolic moieties that affects the stability of the corresponding radicals. The variation of BDEs depends on three factors: the nature of the substituent group, the influence of the side chain for alkyl substituents, and the extension of conjugation system with the corresponding substituent (CH=C[CH₃]₂ or C₆H₅). The HAT mechanism was thermodynamically preferred in gas than SET mechanism. The comparison of the IPs of BDHTD to those of classical antioxidants (gallic acid, caffeic acid, ferulic acid, and ascorbic acid) indicates that the ET mechanism is the most predominant in the 13 BDHTD compounds studied, despite the basis set effect. Thermodynamically, the SET mechanism is more dominant than the PT mechanism in the gas phase. Moreover, the comparison of IP and ETE for each substituent demonstrates that the latter is greatly lower than the former indicating that the SET process from the anionic form is more preferable than that from neutral form in the gas phase. The positive reaction free energies obtained for all the three mechanisms in gas showed that these mechanisms are not spontaneous in gas. Solvent effects stronger influence upon IPs than BDEs. The IPs in the solution phase, are lower than those obtained in the gas phase. The order of IP (water < benzene < gas) is related to the great sensitivity of cation radicals to the polarity of solvent used. Our results illustrated that PT is thermodynamically favored mechanism compared to HAT, whereas SET is least preferred mechanism in the solution phase. Globally, the solution PAFE and ETFE values obtained are positive illustrating that the SPLET mechanism is not spontaneous in both solvents.

Nomenclature

BDE	=	bond dissociation energy
BDFE	=	bond dissociation free energy
BDHTD	=	benzylidenamino-4, 5-dihydro-1H-1,2,4-triazol-5-one derivatives
ETE	=	electron transfer enthalpy
ETFE	=	electron transfer free energy
HAT	=	hydrogen atom transfer
IEF-PCM	=	integral equation formalism polarized continuum model
IP	=	ionization potential
IP	=	ionization potential
PA	=	proton affinity
PAFE	=	proton affinity free energy
PDE	=	proton dissociation enthalpy
PDFE	=	proton dissociation free energy
PhA	=	phenolic acids
RCS	=	reactive chlorine species
RNS	=	reactive nitrogen species
ROS	=	reactive oxygen species
SET	=	single electron transfer
SET-PT	=	single electron transfer-proton transfer
SPLET	=	sequential proton loss electron transfer

Acknowledgments

The authors are grateful to Bénoît Champagne (Université Catholique de Louvain [UCL]; Belgium) and Olivier Holtomo (University of Bamenda; Cameroon).

Funding

This work was supported by the Ministry of Higher Education of Cameroon.

Additional information

Funding

This work was supported by the Ministry of Higher Education of Cameroon.