- Catégorie parente: AST Annales des Sciences et Technologie
Ridha Ahmedi and Touhami Lanez*
Chemistry department, University Centre of El-Oued, B.P.789, 39000, El-Oued, Algeria
Octanol/water partition coefficients logP of many selected substituted ferrocenes were predicted by calculation for the first time using the fragmental Rekker method. The calculated values of logP were compared with the experimental values. For estimation of the logP of the selected substituted ferrocenes, the average absolute error of logP is 0.08. Although the obtained values of logP by the proposed method are all very close to the experimental values, the values of logP obtained for ferrocene derivatives that do not contain a bulky group in their structure were significantly better than the results obtained for those containing a bulky group.
Keywords: logP, partition coefficient, ferrocene derivatives, liquid /liquid extraction,
Le coefficient de portage octanol/eau de quelques ferrocènes substitués a été prédit pour la première fois par l’adaptation de la méthode de Rekker. Les valeurs calculées de logP ont été comparés avec les valeurs expérimentales. La moyenne de l’erreur absolue de logP pour les dérivés ferrocèniques étudiés est de 0.08. Malgré que les valeurs obtenues de logP sont très voisines aux valeurs expérimentales, les valeurs de logP pour les ferrocènes non porteurs des groupements volumineux sont mieux que ceux porteurs des groupements volumineux.
Mots clés : logP, coefficient de partage, dérivés ferrocèniques, extraction liquid /liquid.
In recent years the n-octanol/water partition coefficient logP has become a key parameter in studies of the environmental fate of organic chemicals. Because of its increasing use in the estimation of many other properties, logP is considered a required property in studies of new or problematic chemicals. Although this surmounting interest in octanol/water partition coefficient measurementslays outover the past 90 years, no comprehensive articles of the partition coefficient of ferrocene derivatives have ever been published.In fact, despite the ferrocene itself, no value of partition coefficients has appeared in the literature. Among the many different theoretical methods for the calculation of logP of several simple aliphatic and aromatic compounds, which is described in literature,[1-4] no one of these estimation methods can be applied to organometallic compounds, such as ferrocene derivatives. The very rapid expansion of ferrocene chemistry during the last 50 years, notably in areas related to biology, medicine, catalysis and materials [5-11], led us to turn our attention to the n-octanol/water partition coefficient of ferrocene derivatives, the aim of the present paper is to present an extension of the Rekker method for the calculation of this very important parameter that quantifies the lipophilicity of these derivativesand connects between their structure and their biological activities. The calculation used for obtaining logP of ferrocene derivatives is based upon theextension of the Rekker method used for organic molecules [1, 12].
6. Calculation and validation of our model
We validated our model with ten different ferrocene derivatives (mainly selected from literature sources ) and we recommend carrying out the calculations in three decimals, with the final result rounded to two decimals.
1. Ferrocenes with saturated aliphatic hydrocarbon chains or functionalized aliphatic saturated chains as exemplified by N-(ferrocenyl)-isobutyamide
No correction is needed for this type of compounds; logP is obtained by the summation of the value of the fragmental constant of the four groups in the molecule (ferrocenyl, amide, CH and CH3).
1. Ferrocenes with ferrocenyl-aryl conjugation as exemplified by Phenylferrocene requires a correction of 1 Cm.
The conjugation require in general the application of 1 Cm, thus logP for Phenylferrocene can be obtained by the summation of the fragmental constant of the ferrocenyl and the phenyl groups which equal respectively to 2.456 and 1.902, a correction of 1 Cm = 0.219 should be added which correspond to ferrocenyl-aryl conjugation this gives a value of 4.58 for logP.
Ferrocenes with a basic fragment linked to two aromatic rings requires a correction of 1 Cm, example of this type of compound is N-[4-cyano-3-trifluoromethyl-phenyl]-ferrocenecarboxamide
In addition to the summing of each fragmental constant in the molecule a correction of 1 Cm should be added which correspond to the basic fragment linked to two aromatic rings. Ferrocenes linked to a direct heterocyclic ring or separated by one or more methylene groups requires a correction of 3 Cm; as shown for 4-(4’,4’-dimethyl-3’-ferrocenylethyl-5’-imino-2’-oxo-1’-imidazolidinyl)-2-trifluoromethyl-benzonitrile
logP is calculated as mentioned before, the correction of the basic fragment is replaced by the heterocyclic ring. logP = 5.17
The NCON group in this molecule can be regarded as HNCONH2 minus 3H, to obtain the fragmental constant for this group we subtract three fragmental constants of a hydrogen atom from the value of the fragmental constant of HNCONH2, the summing of the fragmental constant of all groups in the molecule and the addition of a corrective term of 3 cm which correspond to the existence of a heterocyclic ring in the molecule gives the value of 4.66 for logP.
2. 4-(4’,4’-dimethyl-2’,5’-dioxo-3’-ferrocenylmethyl-1’-imidazolidinyl)-2-trifluoromethyl-benzonitrile, logP is calculated as mentioned before. logP = 5.18
3. 4-(4’,4’-dimethyl-2’,5’-dioxo-3’-ferrocenylethyl-1’-imidazolidinyl)-2-trifluoromethyl-benzonitrile, logP is calculated as mentioned before. logP = 5.69
4. Ferrocenes with a basic fragment linked to two aromatic rings and resonance interaction as shown for N-[4-nitro-3-trifluoromethyl-phenyl]-ferrocenecarboxamide, this needs two corrections, the first is 3 Cm for the basic fragment and the second is 1 Cm for the resonance interaction.
After summing the fragmental constant of each group in the molecule, two corrections should be added, the first which equal to 1 Cm is for basic fragment (amide group) linked to two aromatic rings (ferrocenyl and phenyl), the second is equal to 3 Cm for the combination of two groups (nitro and amide) on a phenyl ring in para position which gives rise to a resonance interaction resulting in increased log P. logP = 4.45.
5. Ferrocenes with Hydrogen bonding requires a correction of 2 Cm as exemplified by 4-[4’,4’-dimethyl-2’,5’-dioxo-3’-ortho-hydroxymethyl-ferrocenylmethyl-1’-imidazolidinyl]-2-trifluoromethyl-benzonitrile.
In addition to the correction of 3 Cm for the heterocyclic ring, a correction of 2 Cm should be added for the hydrogen bond between the hydrogen of the hydroxyl group and the oxygen of the carbonyl group, as confirmed by X ray study, the effect of the two substuents on the ferrocenyl group is not considered when there is a hydrogen bond in the molecule. logP = 4.48.
6. Ferrocenes with electronic effect needs the application of a correction of 2 Cm as illustrated in 4-[4’,4’-dimethyl-2’,5’-dioxo-1’-imidazolidinyl-(3’-ortho-methoxymethyl-ferrocenylmethyl)]-2-trifluoromethyl-benzonitrile
We add 2 Cm for electronic effect of two substituents CH2OCH3 and on the same ring (cyclopentadienyl ring), logP = 5.11.
The experimental and calculated logP values for the ten ferrocene derivatives are presented in table2. These values are in good agreement between them.
Value for r2 of 0.976 was found for equation 8.
Due to the major importance of partition coefficients in related study in chemistry, we try successfully in this study to extend the rekker method for calculating octanol/water partition coefficients of ferrocene derivatives starting from characteristics of ferrocene compound. After having adapted the Rekker method for the calculation of partition coefficient, we became able, for the first time, to calculate the partition coefficient of ferrocene derivatives. Values of experimental and calculated logP for a series of ferocene derivatives are in good agreement. The results obtained for logP enable us to consider the process a solution for calculating partition coefficient for ferrocene derivatives and generalizing it to include all analogous complex compounds.
 Raimund Mannhold and Roeloff Rekker;Perspectives in Drug Discovery and Design; 18, 1–18 (2000).
 Renxiao Wang, Ying Fu and Luhua Lai; J. Chem. Inf.Comput.Sci, , 37, 615-621, (1997).
 William M. Meylan and Philip H.Howard ; Perspectives in Drug Discovery and Design,, 19,67–84, (2000).
 A. J. Leo; Hydrophobic Parameter Measurement and Calculation, Methods in Enzymologie, 202, 544-591, (1991)
 Halpern, J.; Pure Appl. Chem. 73, 209–220, (2001).
 Fish, R. H. & Jaouen, G. ; Organometallics 22, 2166–2177 (2003).
 Jaouen, G., Top. S., Vessieres, A. & Alberto, R.; J. Organomet. Chem., 600, 23-36, (2000).
 I. U. Khand, T. Lanez and P. L. Pauson; J. Chem. Soc. Perkin Trans. 1, 2075 (1989).
 T. Lanez and P. L. Pauson; J. Chem. Soc., Perkin Trans. 1, 2436 (1990)
10. T. Lanez, L. Bechki, B. Dadamoussa, M. Saidi and S. Benfardjallah; J. Soc. Alger. Chim. 13, N°2, 251 (2003).
 B. Terki, T. Lanez, S. Belaidi, H. Gornitzka and A. Ourari; Asian J. Chem. Vol.18, N° 3, (2006).
 R. F. Rekker and H. M. De Kort; European Journal of medicinal chemistry, 14, 479-488, (1979).
 A. Leo, C. Hansch and D.Elkins; chemical review, Volume 71, Number 6, (1971).
 A. J. Leo; MedChem, The Medicinal Chemistry Project, Pomona College, Claremont, CA 91711(2000).
 O. Payen; Synthèse d’anti-androgènes organométalliques non stéroïdiens et application au traitement du cancer de la prostate, Thèse de doctorat de l’université Pierre et Marie Curie (Paris VI).(2007)