versión On-line ISSN 0717-9707
J. Chil. Chem. Soc. v.53 n.3 Concepción sep. 2008
J. Chil. Chem. Soc, 53, N° 3 (2008) págs: 1588-1593
STUDY ON QSSR WITH THE ATOMIC IONICITY INDICES*
YAXIN WU1, CHANGMING NIE*1, RONGYAN WU2, SAIHONG JIANG1, SONGNIAN WEN1
1School of Chemistry and Chemical Engineering, University of South China, Hengyang, Hunan Province 421001
In this paper, we find that stretching vibration frequencies of the X=0 for series of R1R2C=0, (RO)2CO, RC02H, α-halogenated aldehydes, cross-RONO and R2NNO can be described by the ionicityindices of X and O atoms as follows: υ = aINI(O)+bINIX+c, where a, b, c are regression constants, and INI(X) , INI(O) represent the ionicityindices of X and O atoms. The efficiency of the model is verified by high correlation coefficients inthe range of 0.960-1.000. Similarly, the ultraviolet absorption energy of ketones can also be characterized by the ionicityindices of C and O as ΔE = aINI(O)+bINIX+c. The predicted results are in good agreement with the experimental ones. Furthermore, the good stability and powerful predictive ability of those models are proved by LOO method.
Keywords: stretching vibration frequency; ultraviolet absorption spectrum; X=0 bond; ionicityindex
Since the Wienerindex W introduced in 19471, the topologicalindices have performed an important role in quantitative structure-activity/property relationships (QSAR/QSPR) study. Many researchers gave much attention to investigations on QSAR, QSPR, QSRR, etc2-8. The topological chemistry and structuralinformationparametershavebeenwidelyappliedtobiopharmaceutics, environmental pollution control and many other fields9-17.
Bajaj Sanjay15 proposed a novel highly discriminating adjacency-cum-distance based topological descriptor, termed it as segmented eccentric connectivityindex and investigated the discriminating power of theindex in predicting the anti-HIV activity of 2-pyridinone derivatives. The discriminating power of the segmented eccentric connectivityindex was found to be superior to that of the distance based Wiener'sindex and adjacency based molecular connectivityindex. Roy Kunal16 proved the quantitative structure-toxicity relationship (QSTR) study to be a valuable approach in ecotoxicity estimations of acute and chronic toxicity to various organisms, and in fate estimations of physical/chemical properties, degradation, and bioconcentration. In the study, he modeled the inhibition of 41 substituted phenols on germination rate of Cucumis sativus with extended topochemical atom (ETA) indices, and indicated that ETA descriptors are sufficiently rich in chemical information to encode the structural features contributing significantly to the comparative inhibition activity of substituted phenols on germination rate of Cuicumis sativus. Dureja H.17 studied the relationship between the topochemicalindices and permeability of diverse series of compounds through blood-brain barrier and compared the three-topochemicalindices, Wiener's topochemical index, molecular connectivity topochemicalindex and eccentric connectivity topochemical index in his investigation. Accuracy of prediction was found to vary from a mínimum of 83% to a máximum of similar to 95% using these models.
Recently, the study on quantitative structure-spectrum relationship (QSSR) has attracted widespread attention.18-21 Zhongchen Cao18 studied polarizability effect of alkyl group on the stretching vibration frequency for X=0 bond. Khadikar P.19 estimatedthe characteristic vibration of carbonyl group by using the Szeged index. You Jinglin20 composed the novel topological index SIT to depict the Raman spectra of silicate glasses and their liquids. Zhou L. P.21 investigated the prediction of carbón-13 NMR chemical shifts of alkanes with rooted path vector, and got a good result. On the basis of previous work22-30, we put forward new QSSR models on the ground of the ionicityindices22 of X and O atoms, and use these models to study the law of change for infrared stretching vibration frequencies of X=0 and the ultraviolet absorption energy 0fC=O
Theories and Methods
1 Model Development
In this paper, we consider using the ionicityindices of atoms in X=0 to depict the change rule of the spectral property of compounds with X=0 and build a múltiple linear regression (MLR) model as follows:
where a, b, c are regression constant. P represents spectral property of compounds with X=O. The correlation coefficient (R), the Fischer ratio value (F), and the determinant constant (S) can be used to assess the quality of the models. INI(O)( and INIX are the ionicityindices of the atom O and X.
2 Ionicity indices
Authors ever pointed the ionicityindices (XINI)22 of atoms in molecule as
In the Equation (2), XA is PΔυling electronegativity scale, and XE is the atomic equilibrium electronegativity23 in molecule. XE is figured as
In the Equation (3), Σ1 is the sum of atoms or branching groups directly attached to the atom, and ΣXG is the sum of electronegativities of atoms or groups directly attached to the atom. XG is given by the following24:
As shown in Fig. 1, the group can be divided into k layers, and the left atoms of the dotted line labeled 1, 2, 3, ...k are known as the ground atoms. The right suffix of atoms / is the numeration of the ground atom. n11, n21 ...nkl are the sum of the ground atom / and other atoms or groups directly attached to the ground atoms in the grade are the sum of electronegativities of the ground atom / and other atoms or branching groups directly attached to the ground atoms inthe grade 1, 2, 3,...k.
RESULTS AND DISCUSSION
1 QSSR study for infrared spectrum
Scientists began to research into the infrared spectrum in the 20th century. Ordinarily, the infrared absorption frequency of the organism is in the range of 4000-625 cm-1, and contains plentiful structural information. The infrared absorption wave lengths of differentfunctional groups inthe organic molecules are different from each other. Affected by the chemical condition around the functional groups, the changing of the position, intensity and shape of the absorption band occurs. Application of the infrared spectrum has played a very important part in organic chemistry, biological chemistry and medicinal chemistry, etc. For instance, the infrared spectrum is regarded as a crucial deliberatedindex for more and more medicines by codex in various countries31. Above all, it is very useful and necessary to establish a convenient and exact method to predict the infrared spectrum information in virtue of the quantitative structure-spectrum relationship.
1.1 Relationships between ionicityindices and the values of stretching vlbration frequencies
There are many factors influencing on the displacement of the absorption bands, for example, the electronic effect, stereoscopic effect, coupling effect, bond strength, tautomery, hydrogen bond. Researchers found out that the change of the stretching vibration frequencies of the X=0 bond is mostly determined by the electronic and stereoscopic factors31. Many experimental results demonstrated that the variation of the eigen frequency of X=0 happens mainly because of the impacts of the substituents linked directly to the X atom. especially the affect of the electronic effect of groups. We believe that the affect can be described with ionicityindices of atom X and O. We replaced P in Equation (1) with the stretching vibration frequencies (v) of the X=0 and obtained Equation (5).
We made linear simulation of the stretching vibration frequencies of the X=0 for various series according to the Equation (5) using the known infrared spectrum data31. And the followings are obtained:
We calculated the stretching vibration frequencies of the X=0 for 35 compounds by the Equation (6)-(1 1), and listed the results on the Table 1.The results indicated that the stretching vibration frequencies of the X=0 were well correlated with the ionicityindices of the atom O and X, the correlation coefficient R was in the range of 0.960-1.000, and 5, the standard deviation was among 0.179-7.294. The relative errors between the calculated values and the experimental ones are under 0.6%.
The calculated values and the experimental values are depictured on Fig. 2. We can see that the calculated values fit closely with the experimental ones.
To further prove the stability of the model, we used the LOO method to test the Equation (5). In this method28, we took one sample from N samples, made the regression of the ionicityindices with the experimental values of the remaining N-1 samples, and obtained a regression equation for N-1 samples, then predicted the value of the sample taken-out by the regression equation. Using the same method, we got N values predicted. Then N values predicted were regressed with their experimental values, and the regression results such as regression constants, correlation coefficient Rcv standard error Scv could be used to test the stability and validity of the model. We tested the Equation (6) and Equation (9) by LOO method.
For series R1R2C=0:
The results showed that the method using Equation (5) to study the stretching vibration frequencies of the X=0 was rational and credible.
1.2 Relationshlps between ionicityindices and the values of stretching vibration frequencies change
The stretching vibration frequency of the C=0 for α-halogenated aldehyde becomes larger, becΔυse the a-H atom is substituted by the halogen atom with larger electronegativity. The influence order of the halogens is F>Cl>Br>I. In addition, the change of the stretching vibration frequencies of the C=0 (Δυ) has relation to the number of the substituted a-H atoms. In this paper, we studied the law of change of Δυ for α-halogenated aldehyde. According to Equation (5), we obtain
Equation (14)-Equation (15) we get
Based on Equation (16), We studied the relationship between the values of Δυ31 for 5 α-halogenated aldehydes and the ionicityindices of the atom O and C by using regression analysis, and gained
The experimental values and the calculated ones are listed on Table 2.
The result illustrated that Equation (17) better opened out the change of the stretching vibration frequencies of the C=0 for series α-halogenated aldehydes. In fact, the ionicityindex denotes the electronegativity variation between the neutral atom and the bonded atom in a molecule. The more largely electronegativity varíes, the higher ionicity level of the atoms is. Forming a molecule, the electronegativity variations of the atoms in diverse chemical conditions are different too. Therefore the ionicityindices of the atoms are different from each other. All above lead to the different stretching vibration frequencies of the X=0 (υ).
2 QSSR study for ultraviolet spectrum
The ultraviolet absorption spectrum is one of the four important spectrums in structural analysis for organic compound. Researchers are eagerly looking forward to describing the change rule of the ultraviolet absorption spectrum quantificationally.32 Because of various chromophore groups and aux chromous groups, it is difficult to construct a model which adapt to relate the ultraviolet absorption spectrum for all kinds of organic compounds. In this work, we study the quantitative structure-ultraviolet absorption spectrum relationship of 14 ketones in the same solvent simply.
As we known, the ketone is non-conjugated compound. For the carbonyl (C=0) of ketones (R1R2=0), the most important influence on ultraviolet absorption energy (ΔE1) is the properties of the substituents (R). We use the ionicityindices of atom C and O to denote the chemical condition for some ketones with different Rs, describe the change rule of ΔE for C=0 by using Equation (1) with ΔE instead of P, and obtain
We do the regression analysis with INI(O) , INI(C) and the experimental value of ΔE, and get
The experimental value of the ultraviolet absorption wavelength (λ) is from reference33. According to the Equation (20), we can obtain the energy of electrón transition of ultraviolet absorption (ΔE) for C=0.
Both the calculated values and the experimental values are Usted on Table 3. From the result, we find that most of the absolute errors from this work are less than that of ref 32.
From Fig.3, it is obviously that the calculated values accord well with the experimental ones.
The Equiation (19) was tested by LO method:
From the high Rcv it can be seen that the ultraviolet absorption energy (ΔE) have a good linear correlation with the INI(C) , INI(O) for ketones.
The method using the ionicityindex to study QSSR is successful. (1) The method is simple and accurate, and the ionicityindex is convenientto be gotten. (2) The ionicityindices character the X=0 with different chemical condition around in molecules, and realize the uniqueness token of the molecule. (3). Correlation coefficients of the models range from 0.960 to 1.000. It indicates that the ionicityindex is very useful in quantitative structure-spectrum. relationship study. (4) The good stability and strong predictive capability of these models are proved by LOO method.
We acknowledge the support from the Science and Technology Projects of Hunan Province (No.06FJ4104) and Technology Innovation Plans of Economy Commission of Hunan Province (No.283).
1. Wiener U, Am. Chem. Soc. 69, 17, (1947). [ Links ]
2. Cornwell, Edward and Cordano, Gianni, J. Chil. Chem. Soc. 48, 23, (2003). [ Links ]
3. Reza Ashrafi, Ali, Loghman, Amir, J. Chil. Chem. Soc. 51, 968, (2006). [ Links ]
4. Chunhui Lu, Weimin Guo, Xiaofang Hu, Chem. Phys. Lett. 417, 11, (2006). [ Links ]
5. Chunhui Lu, Weimin Guo, Chunsheng Yin, Anal. Chim. Acta. 561, 96, (2006). [ Links ]
6. Qingsong Wang, Juan Liu, Meirong Xu, Changjun Feng, J. Wuhan Univ. Tech. 28, 117, (2006). [ Links ]
7. Peng Zhou, Hu Mei, Feifei Tian, Zhiliang Li, Chinese J. Anal. Chem. 34, 25. 1096, (2006) [ Links ]
8. Pompe M., Randic Milán, J. Chem. Inf. Model. 46, 2, (2006). [ Links ]
9. Yan Chen, Changjun Feng, Chemistry, 69, 277, (2006) [ Links ]
10. Xihua Du, Yan Chen, Ziqiang Tang, Mingjin Li, Keying Cai, J. Nanjing Univ. Tech. 28, 98, (2006) [ Links ]
11. Viney Lather, Anil K Madan, J. Mol. Graph Model. 23, 339, (2005) [ Links ]
12. Yovani Marrero-Poncea, Alma Huesca-Guille'nb, Froyla'n Ibarra-Velarde, J. Mol. Struct. 111, 67, (2005) [ Links ]
13. Randic M., Zupan J., Vikic-Topic D., Plavsic D., Chem. Phys. Lett. 431, 375, (2006) [ Links ]
14. Estrada E., Comput. Biol. Chem. 27, 305, (2003) [ Links ]
15. Bajaj Sanjay, Sambi S. S., Gupta S., Madan A. K., QSAR Comb. Sci. 25, 813, (2006) [ Links ]
16. Roy Kunal, Ghosh Gopinath, QSAR Comb. Sci. 25, 846, (2006) [ Links ]
17. Dureja H, Madan A. K. Int. J. Pharm. 323, 27, (2006) [ Links ]
18. Cengzhong Cao, Chinese J. Org. Chem. 18, 546, (1998) [ Links ]
19. Khadikar P., Mandloi M., Shrivastava A., Phadnis A., Oxid. Commun. 26, 161,(2003) [ Links ]
20. You Jinglin, Jiang Guochang, Chen Hui, Xu Kuangdi, Rare Metals , 25, 431,(2006) [ Links ]
21. Zhou L. P., Sun L. L., Yu Y., Lu W., Li Z. L., J. Mol. Graph. Model. 25, 333, (2006) [ Links ]
22. Changming Nie, Zhonghai Li, Songnian Wen, Chinese J. Org. Chem. 21, 46, (2002) [ Links ]
23. Changming Nie, Guowen Peng, Fangzhu Xiao, Shan Li, Xiaomei He, Zhonghai Li, Congyi Zhou, Chinese J. Anal. Chem. 34, 1560, (2006) [ Links ]
24. Changming Nie, J. Wuhan Univ. 46, 176, (2000) [ Links ]
25. Congyi Zhou, Changming Nie, Shan Li, Songnian Wen, Guowen Peng, Zhonghai Li, Chinese J. Inorg. Chem. 23, 25, (2007) [ Links ]
26. Changming Nie, Yimin Dai, Songnian Wen, Zonghai Li, Chinese J. Chromatogr. 23, 1, (2005) 1592 [ Links ]
27. Changming Nie, Yimin Dai, Songnian Wen, Acta Chim. Sinica, 63, 1449, (2005) [ Links ]
28. Congyi Zhou, Changming Nie, Shan Li, Zhonghai Li, J. Comput. Chem. 28, 2413, (2007) [ Links ]
29. Congyi Zhou, Xi Chu, Changming Nie, J. Phys. Chem. B, 111, 34, 10174, (2007) [ Links ]
30. Congyi Zhou, Changming Nie, Bul. Chem. Soc. Japan, 80, 1504, (2007) [ Links ]
31. Jingxi Xie, The application of infrared spectroscopy to organic chemistry and medicinal chemistry, Science Press, Beijing, 1987. [ Links ]
32. Chenzhong Cao, J. Xiangtan Normal College, 19, 1, (1998) (in chinese) [ Links ]
33. Weast R C, Handbook of Chemistry and physics, 65th ed, CRC press, INC, 1984-1985. [ Links ]
(Received: September 12, 2007 -Accepted: April 29, 2008)
* e-mail: email@example.com