REGULAR PAPERS

Solvent Effect On The¹H NMR Spectra Of A Pyridinocalix(4)Arene Derivative And Its Protonation Constants In Methanol

Danil de Namor, A. F.¹; Aguilar Cornejo, A. F.¹; Chahine, S.¹; Nolan, K. B.²

¹ Thermochemistry Laboratory, Chemistry Division, School of Biomedical and Molecular Sciences, University of Surrey, Guildford, Surrey, GU2 7XH, UK.
Fax: +44 1483 876851; Email: a.danil-de-namor@surrey.ac.uk
² Department of Chemistry, Royal College of Surgeons in Ireland, St Stephen’s Green, Dublin 2, Ireland.

Received March 29, 2004. In final form June 29, 2004
Dedicated to Prof.  Pedro J.Aymonino on the occasion of his 65th birthday

Abstract
The solvent effect on the ¹H NMR spectra of 5, 11, 17, 23 tetra-tert-butyl-25, 26, 27, 28-tetra-[2-(4-pyridyl) methoxy]calix(4) arene, 1c in a wide range of solvents with different dielectric constants was investigated and the results are discussed.  It is shown that the ¹H NMR spectrum of this macrocycle is sensitive to the nature of the solvent. In all solvents the ligand shows a distorted ‘cone’ conformation..The aromatic protons show the most significant deshielding effect in acetonitrile, nitromethane and pyridine relative to chlroform, which may be attributed to the interaction of these solvents (acetonitrile,nitromethane) through their methyl groups with the hydrophobic cavity of the ligand.  For pyridine, p-p interactions between the pyridyl groups and this solvent may occur.
The protonation constants of this ligand in methanol were derived from potentiometric titration data in this solvent.  The results are compared with those for  structural isomers of 1c as well as with those previously reported for lower rim calix(4)arene derivatives containing aliphatic and alicyclic amines as pendant arms.  Protonation constant data show that the affinity of the latter to interact with the proton is greater than that of the former macrocycle in methanol.
Final conclusions are given.

Resumen
El efecto del solvente en los espectros ¹H RMN de 5, 11, 17, 23 tetra-tert-butilo-25, 26, 27, 28-tetra-[2-(4-piridil)metoxi]cálix(4)areno, 1c, en un amplio rango de solventes con distintas constantes dieléctricas fue investigado y los resultados son discutidos. Se muestra que el ¹H RMN espectro de este macrociclo es sensible a la naturaleza del solvente. En estos solventes la conformación del ligando corresponde a un cono distorsionado. Los desplazamientos químicos mas pronunciados de  los protones aromáticos con respecto a cloroformo  son observados en acetonitrilo, nitrometano y piridina. En los dos primeros solventes estos desplazamientos pueden ser atribuídos a una interacción entre los grupos metilos de estos solventes con la cavidad hidrofóbica del ligando. En piridina los resultados sugieren la posibilidad de interacciones del  tipo π­π. Las constantes de protonación de este ligando en metanol fueron derivadas de datos obtenidos de titulaciones potenciométricas en este solvente. Los resultados son comparados con aquellos correspondiente al de un isómero estructural de 1c como también con aquellos reportados en la literatura para derivados de  calix(4)arenos conteniendo aminas alifáticas y alicíclicas en el borde inferior del macrociclo. Los valores de las constantes de protonación muestran que la afinidad  de estos últimos para interaccionar con el protón es mucho mayor que aquellos correspondiente a los piridino-calix(4)arenos in metanol.  Conclusiones finales son dadas.

Introduction

     Calixarenes [1-5] are excellent platforms for the design of receptors for binding ions and molecules.  Depending of the nature of the functional groups at the lower and upper rim, calixarene derivatives show different and interesting complexing abilities.
     Pyridinocalix(4)arenes (1a, 1b, 1c) have received considerable attention in recent years.   Three structural isomers of pyridinocalix(4)arenes are known.

1a                               1b                                1c

                

     The ability of these ligands to interact with hard and soft metal cations is largely dependent on the position of the nitrogen atoms of the pyridyl rings with respect to the ethereal oxygens.  Thus in previous papers we have reported the solution properties of 2-pyridinocalix(4)arene (1a) and its structural isomer (1b) [7-9].  It was shown that while the former interacts with alkali-metal cations to form 1:1 (metal cation-ligand) complexes, the latter is unable to enter complexation with these cations.  This was attributed to an increase in the distance between the pyridyl nitrogen and the ethereal oxygen in moving from 1a to 1b.  X-ray crystallographic evidence of alkali-metal complexes of 1a have been reported [8]. As a continuation of this research the work goals of this paper are:

(i) To assess the effect of the solvent on the proton chemical shifts of 1c by carrying out ¹H NMR measurements in various solvents at 298 K given that the presence of a hydrophobic cavity in calix(4)arene derivatives may permit interaction with the solvent molecules.  It is therefore of considerable interest to investigate the extent in which the different donor-acceptor abilities of various solvents as well as other properties such as dielectric constant, dipole moments and other factors lead to shift changes in the ¹H NMR signals of 4 pyridinocalix(4)arene that are proportional to the extent of the interaction.

(ii) Interaction of 1c with the proton by ¹H NMR studies in CD₃OD followed by the determination of the protonation constants in the same solvent at 298.15 K.

Experimental Part

     The compound 5,11,17,23 – tetra-tert-butyl-25,26,27,28-tetra-[2-(4-pyridyl) methoxy]calix(4)arene, 1c, was synthesised and characterised as described elsewhere [10].

¹H NMR measurements

     This technique was used to investigate the interaction of a variety of solvents with 1c and to study the conformational changes that this ligand undergoes in these media. ¹H NMR measurements were conducted at 300 MHz in the following deuterated solvents, chloroform (CDCl₃), acetone (CD₃COCD₃), nitromethane (CD₃NO₂), benzene (C₆D₆), pyridine (C₅D₅N), methanol (CD₃OD), nitromethane (CD₃NO₂) and dimethylsulfoxide (d₆-DMSO) (all purchased from Aldrich), at 298 K. Solutions of the sample of interest were prepared in 1 cm³ in the appropriate deuterated solvent and placed in the 5 mm NMR tube using TMS as internal reference. ¹H NMR data were processed using the NUTS program [11].  Chemical shifts (d ppm) and difference in the chemical shifts (Dd = d – d_CDCl3 ppm) with respect to CDCl₃ (as a reference solvent) were recorded. 

Determination of the protonation constants of 1c in methanol at 298.15 K. 
These were determined by potentiometry as discussed elsewhere [12].

Results and Discussion

(i)        Interaction of 1c with the solvent: ¹H NMR studies.

     The interaction that occurs between a solvent molecule and the hydrophobic cavity of calyx(4)arene derivatives was described as ‘Allosteric effect’.[6] It is this kind of pre-organization of the hydrophilic cavity that contribute to the ability of the ligand to recognise selectively between its guests. This solvent-ligand interaction is best reflected by the difference in the chemical shifts of the hydrophobic cavity protons and those of the methylene bridging carbon in one solvent relative to another (reference solvent).   Chemical shifts data for 1c in the different solvents investigated are given in Table 1 at 298 K According to these data, the difference in the chemical shifts between the axial (H-5) and equatorial (H-6) protons of the methylene bridging carbon (Dd_ax-eq = d_ax. – d_eq. ppm), suggest that in all cases 1c exhibits a distorted ‘cone’ conformation (Dd_ax-eq = 1.21, 1.17, 1.18, 1.24, 1.24, 1.12, 1.14 and 1.13 ppm in C₆D₆, CDCl₃, C₅D₅N, CD₃COCD₃, CD₃OD, CD₃CN, CD₃NO₂ and d6-DMSO, respectively).

Table 1. Chemical shifts in ppm for the proton signals of 1c in different solvents at 298 K

Solventa	Dielectric constant	H-1	H-2	H-3	H-4	H-5b	H-6c	t-But
C6D6	-2.27	8.52	6.98	6.95	4.63	4.17	2.96	1.21
CD3Cl	4.81d	8.47	7.22	6.78	4.85	4.12	2.95	1.08
C5D5N	13.2	8.68	7.44	7.15	5.02	4.47	3.29	1.27
CD3COCD3	20.7	8.43	7.37	6.92	4.98	4.30	3.06	1.12
CD3OD	32.6	8.34	7.42	6.91	4.94	4.30	3.06	1.11
CD3CN	36.7	8.43	7.20	7.18	4.81	4.14	3.02	1.18
CD3NO2	38.6	8.45	7.30	7.19	4.93	4.28	3.11	1.16
d6-DMSO	46.7	7.26	6.29	5.81	3.79	3.16	2.03	1.45

a Abbreviations used; benzene C₆D₆; chloroform; CDCl₃ pyridine, C₅D₅N; acetone; CD₃COCD₃; methanol CD₃OD, acetonitrile; CD₃CN, nitromethane, CD₃NO₂; dimethylsueforide; d₆-DMSO
^bH-5 = H axial            ^bH-6 = H equatorial    ^d determined at 293.15 K

     On the other hand, a quick inspection of the chemical shifts for H-1 and 2, as well as those of the hydrophilic cavity, indicate that the nature of the solvent (protic or aprotic) [14, 15] is affecting both cavities.
     Taking chloroform as the reference solvent, chemical shift changes (Dd ppm) are calculated and these are shown in Table 2.

Table 2: Chemical shifts changes in ppm for the proton signals of 1c in different solvents taking
chloroform as the reference solvent at 298 K.

Solventa	H-1	H-2	H-3	H-4	H-5b	H-6c	Ht-But
C6D6	0.05	-0.24	0.17	-0.22	0.05	0.01	0.13
C5D5N	0.21	0.22	0.37	0.17	0.30	0.34	0.19
CD3COCD3	-0.04	0.15	0.14	0.13	0.13	0.11	0.04
CD3OD	-0.13	0.20	0.13	0.09	0.13	0.11	0.03
CD3CN	-0.04	-0.02	0.40	-0.04	-0.03	0.07	0.10
CD3NO2	-0.02	0.08	0.41	0.08	0.11	0.06	0.08
d6-DMSO	-1.21	-0.93	-0.97	-1.06	-1.01	-0.92	0.37

     A pronounced deshielding effect is observed for the H-3 proton (except in d6-DMSO), which suggest that these solvents interact with the hydrophobic cavity of 1c. Furthermore, the tert-butyl group protons show a deshielding effect in C₆D_6, CD₃CN and d6-DMSO (Dd H-3= 0.13, 0.19, 0.10 and 0.37 ppm), while this effect is less pronounced in the other solvents (Dd H-3 » 0.05 ppm). Therefore, this indicates that d6-DMSO might be forming an exo-complex with 1c (Dd H-3 = -0.97 ppm while Dd H-_t-Bu = 0.37 ppm).
On the other hand, more bulky solvents (C₆D₆ and C₅D₅N) than d6-DMSO, have produced a deshielding effect for both protons H-3 and H-_t-Bu, suggesting that the interaction between the solvent and the hydrophobic cavity is of the p-p type which can be the result of exo- or endo-complex formation with 1c.
     The deshielding effect observed for the H-5 and 6 in the solvents investigated (except d6-DMSO) relative to CDCl₃ and the Dd_ax-eq (discussed above), indicate that the hydrophobic cavity of 1c does not undergo a pronounced conformational change in these solvents with respect to CDCl₃ at 298 K. This might be due to the rigidity of the hydrophobic cavity caused by the bulky pendant arms (pyridine group) in the hydrophilic cavity, being incapable of getting closer to each other due to the electrostatic effect.
An interesting observation is that on increasing the dielectric constant of the solvent, the signal of the H-1 proton of the pyridine ring (see Table 1) shifts upfield to a smaller extent that the signal of the H-2 proton. 
     The shielding effect observed for H-1 (except C₆D₆ and C₅D₅N) suggest that not all solvents interact with the pyridyl groups in the hydrophilic cavity. On the other hand, a pronounced deshielding effect (H-1,2 and 4) is recorded in C₅D₅N suggesting a p-p interaction between the pyridyl groups and this solvent. For C₆D₆ slight deshielding effect is observed for H-1 and a pronounced shielding effect for H-2 and H-4 indicating that no interaction is taking place  between this solvent and the hydrophilic cavity of 1c.[16].
     The chemical shift changes of 1c in CD₃OD and CD₃COCD₃ relative to CDCl₃ show a deshielding effect for the H-2 and the H- 4 signals. This suggests an interaction between these solvents and the pendant arms at the lower rim of 1c. In acetone which is a dipolar aprotric solvent with a moderate dielectric constant, broadening of the signals for H-1 and H-2 occurs.  The H-1 signal shifts upfield but the H-2 signal shifts downfield compared to C₆D₆ and CDCl₃.  This may indicate specific interactions at sites away from the nitrogen of the pyridine ring.  In d6-DMSO which is a solvent with moderate dielectric constant and a high donor number [13,14], shielding effects are observed for all the protons of the hydrophilic cavity suggesting that no interaction is taking place with the pendant arms of 1c relative to CDCl₃ at 298 K.
In summary the conclusions are as follows,
i) In all solvents the ligand shows a distorted ‘cone’ conformation.
ii) The aromatic protons (H-3) show the most significant deshielding effect in CD₃CN, CD₃NO₂ and C₅D₅N relative to CDCl₃, which may be attributed to the interaction of these solvents (CD₃CN, CD₃NO₂) through their methyl groups with the hydrophobic cavity of the ligand. This interaction has been observed for various calixarene derivatives and acetonitrile [18,19.]. In C₅D₅N a p-p interaction between the pyridyl groups and this solvent is likely to occur..

(ii)     Protonation constants of 1c in methanol at 298.15 K

     The potentiometric titration curve of 1c with a methanolic solution of tetramethylammonium hydroxide is shown in Fig 1 as a plot of E/volts against the base/ligand ([OH-]/[lc]) ratio.  Also included in this figure is its first derivative. 

Figure 1. Potentiometric titration of 1c with tetramethyl ammonium hydroxide in methanol at 298.15 K

     The experimental data indicate that the inflection point in (i) or the maximum in (ii) corresponds to the following stoichiometry.

1c (MeOH) + 4 H+ (MeOH) ® H41c4+ (MeOH)

(1)

Potentiometric data were analysed by the use of the MINIQUAD program [20,21].  The calculated protonation constants of 1c in methanol are shown in Table 3 and these are referred to the processes described by eqs 2-5. 

1c (MeOH) + H+ (MeOH) [H1c]+ (MeOH)	(2)
[H1c]+ (MeOH) + H+ (MeOH)  [H21c]2+ (MeOH)	(3)
[H21c]2+ (MeOH) + H+ (MeOH)  [H31c]3+ (MeOH)	(4)
[H31c]3+ (MeOH) + H+ (MeOH)  [H41c]4+ (MeOH)	(5)

Table 3.Protonation constants of 1c in methanol at 298.15 K (expressed as log Kp) and overall protonation constants (expressed as log b) in the same solvent.

 

log KP1	5.91	Log b1	5.91
log KP2	5.35	Log b2	11.26
log KP3	5.03	Log b3	16.29
log KP4	4.70	Log b4	20.99

The overall protonation constants, log b corresponds to the following processes.

1c (MeOH) + H+ (MeOH)  [H1c]+ (MeOH)	(6)
1c (MeOH) + 2H+ (MeOH)  [H21c]2+ (MeOH)	(7)
1c (MeOH) + 3H+ (MeOH)  [H31c]3+ (MeOH)	(8)
1c (MeOH) + 4H+ (MeOH)  [H41c]4+ (MeOH)	(9)

     As the difference between the first and fourth protonation constants is only 1.21 units, it is reasonable to assume that the pyridine groups behave independently from each other with respect to protonation, otherwise protonation of one pyridine residue would markedly affect the basicity of another and give rise to significant pK differences [12].
     If the protonation constants of 1c are compared with those of 1b (Table 4), it is noted that the oxygen of the ether group plays a key role in the acid-base properties of these derivatives.  Thus the latter derivative is less basic than the former due to the close proximity of the pyridine nitrogen to the electron withdrawing ethereal oxygen.  When the protonation constants of 1c are compared to the corresponding values of alkylamino calix(4)arenes (1d-1g) (Table 4) dramatic changes in basic properties are observed. 

    

Table 4: Protonation Constants of lower rim pyridinocalix(4)arenes (1b,1c), aliphatic (1d, 1e) and alicyclic (1f, 1g,) amino calix(4)arene derivatives in methanol at 298.15 K

Derivative	log KP1	log KP2	log KP3	log KP4
1b	5.15	4.52	4.11	3.59
1c	5.91	5.35	5.03	4.70
1d	9.40	8.44	8.24	7.67
1e	9.48	8.96	8.69	8.10
1f	9.35	9.01	8.72	8.33
1g	7.49	6.93	6.93	5.98

     Thus while dimethylaminocalix(4)arene, 1d, has a value of 7.67 for log K_p4, that for 1c is 4.70.  The increased basicity of the alkylamines relative to pyridino calixarenes (1c, 1b) parallels that observed for similar non-calixarene derivatives (eg alykylamines vs. pyridines) and may be attributed to interaction of the pyridine nitrogen lone pair with the aromatic sextet [10].  In the case of the morpholine calix(4)arene, 1h, the influence of the ring oxygen decreases the basic properties as expected.  Species distribution curves as a function of pH were obtained for 1c using a MINIQUAD computer program.  As the pH changes from 3.5 to 7, the composition of the solution changes from one containing protonated forms of the ligand to one containing the unprotonated form.  At a pH @ 4.7, the major species in solution is [H₃1c]³⁺ while at pH 5.2 is [H₂1c]²⁺ and at pH 5.6 it is [H₁c]⁺.
     The complexation ability of 1c for metal cations in non-aqueous media (acetonitrile and methanol) investigated by ¹H NMR, conductimetry, potentiometry and calorimetry is under investigation.[22]

Conclusions

     From the above results it is concluded that:
(i)  The variation in the chemical shifts of 1c with the solvent will have implications on the thermodynamics of complexation of this ligand with metal cations particularly if the solvent is hosted in the hydrophobic cavity of the ligand and as a result it may exert an allosteric effect on the hydropholic cavity which may lead to  a re-organisation of the hydrophilic and thus unexpected recognition ability.
(ii) The position of the ethereal oxygen relative to the pyridyl nitrogen affects the basicity of the ligand. As expected calix(4)arene amino derivatives containing alicyclic and aliphatic amino groups ( 1d –1g) show a much higher basicity than pyridino calix(4)arenes.

References

[1]        C.D. Gutsche in Calixarenes, Monographs in Supramolecular Chemistry, ed. J.F. Stoddart, Royal Society of Chemistry, Cambridge, 1989        [ Links ]

[2]        Calixarenes: A Versatile Class of Macrocyclic Compounds; Ed. J. Vicens and V. Böhmer, Kluwer, Dordrecht, The Netherlands, 1991.         [ Links ]

[3]        Calixarenes, 50^th Anniversary Commemorative Volume.  J. Vicens, Z. Asfari, J.M. Harrowfield, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1994.         [ Links ]

[4]        C.D. Gutsche, Calixarenes Revisited.  Monographs in Supramolecular Chemistry.  Ed. J.F. Stoddart, The Royal Society of Chemistry, 1998.         [ Links ]

[5]        Calixarenes 2001, Eds Z. Asfari, V. Böhmer, J.M. Harrowfield, J. Vicens, Kluwer Academic Publishers, Dordrecht, The Netherlands, 2001.         [ Links ]

[6]        A.F. Danil de Namor, R Cleverley and M.L. Zapata-Ormachea, Chem.Rev.; 1998, 98, 2495; see also references therein.         [ Links ]

[7]        A.F. Danil de Namor in Calixarenes 2001.  Eds. Z. Asfari, V. Böhmer, J.M. Harrowfield, J. Vicens, Kluwer Academic Publishers, Dordrecht, The Netherlands 2001, Chapter 19.         [ Links ]

[8]        A.F. Danil de Namor, E.E. Castellano, L.E. Pulcha Subezar, O. E. Piro and O. Jafou, Phys. Chem. Chem. Phys, 1999, 1, 285.         [ Links ]

[9]        A.F. Danil de Namor, Coord. Chem. Rev.; 1999, 190-192, 283.         [ Links ]

[10]      A.F. Danil de Namor, O.E. Piro, L.E. Pulcha Salazar, A.F. Aguilar Cornejo, N.A. Al Rawi, E.E. Castellano and F.J. Sueros Velarde, J. Chem. Soc., Faraday Trans, 1998, 94, 3097        [ Links ]

[11]      “NUTS” NMR software for Windows 95.  Acorn NMR Inc; Fremont, California, USA.         [ Links ]

[12]      A.F. Danil de Namor, F.J. Sueros Velarde and M.C. Cabaleiro, J. Chem. Soc.; Faraday Frans, 1996, 92, 1731.         [ Links ]

[13]      C.D. Gutsche, Aldrichimica Acta, 1995, 28, 3.         [ Links ]

[14]      A.K. Covington and T. Dickinson, Physical Chemistry of Organic Solvents Systems, Plenum Publishing Company Limited.  London 1973.         [ Links ]

[15]      D. Popovych and R.P. Tomkins, Nonaqueous Solution Chemistry, John Wiley & Sons, Inc., New York, 1981.         [ Links ]

[16]      T.J. Batterham, NMR Spectra of Simple Heterocycles; John Wiley & Sons, Inc., New York, 1973.         [ Links ]

[17]      J.P. Kintzinger and J.M. Lehn, Mol.Phys; 1968, 14, 133.         [ Links ]

[18]      A.F. Danil de Namor, Pure Appl. Chem, 1993, 65, 193.         [ Links ]

[19]      A.F. Danil de Namor, E. Gil, M.A. Llosa Tanco, D.A. Pacheco Tanaka, L.E. Pulcha Salazar, R.A. Schulz and J. Wang; J. Phys. Chem.; 1995, 99, 16781.         [ Links ]

[20]      A. Sabatini, A. Vacca and P. Gans, Talanta, 1974, 21, 53.         [ Links ]

[21]      A. Sabatini and A. Vacca, Coord. Chem Rev; 1992, 120, 389.         [ Links ]

[22]     A.F.Danil de Namor, A. Aguilar-Cornejo, R. Soualhi, M. Shehab, K.B. Nolan, N.Ouazzani and L. Mandi, J. Phys. Chem 2005, 109, 14795Links ]mso-ansi-language:EN-US'>.