Thomas
Lang
,
Ernest
Graf
*,
Nathalie
Kyritsakas
and
Mir Wais
Hosseini
*
UMR UDS-CNRS 7140, Université de Strasbourg, Institut Le Bel, 4 rue Blaise Pascal, CS90032 67081 Strasbourg, France. E-mail: hosseini@unistra.fr; egraf@unistra
First published on 25th September 2012
The design and synthesis of two new Zn(II) and Pd(II) strapped-porphyrin based molecular turnstiles have been achieved. These molecules are based on a porphyrin backbone as a stator and a handle as a rotor. The junction between the two parts is ensured by covalent bonds using two opposite meso positions. Both compounds have been characterised in solution by NMR spectroscopy and in the solid state by X-ray diffraction on a single crystal. The dynamic of the two systems have been investigated in solution by 1- and 2-D NMR techniques such as ROESY. The switching between the open and closed states is described. Whereas for the Pd(II) complex, the handle freely rotates around the stator, for the Zn(II) analogue owing to the presence of a water molecule bound to the zinc atom and hydrogen bonded to the pyridyl moiety of the rotor, the rotation is blocked. The unlocking of the rotational movement was achieved using an external agent, mainly dimethylaminopyridine as a strong coordinating ligand.
Here, we report on the design, synthesis, characterization, both in solution and in the crystalline phase, and on the switching between the open and closed states of molecular turnstiles based on strapped-porphyrins19 composed of a free base meso tetraaryl porphyrin and its zinc20 or palladium complexes and a covalently attached handle bearing a pyridyl moiety as a monodentate coordinating site.
![]() | ||
Scheme 1 Strapped porphyrin 1, its Zn(II) and Pd(II) complexes 2 and 3 respectively, and the water complex 2·H2O and the assignment of H atoms. |
The operational design of the turnstile (Fig. 1) is based on: (i) free rotation (open state, Fig. 1a) of the rotor around the stator as a consequence of the absence of specific interactions between the two parts, (ii) introduction of the Zn(II) cation within the macrocyclic coordinating moiety of the strapped porphyrin 1 (Fig. 1b) affording the neutral complex 2 (Scheme 1). Owing to the propensity of Zn2+ cation to adopt mainly a square based pyramidal geometry when complexed by the tetraaza core of the porphyrin, the coordination of the pyridyl unit of the rotor should lock the rotation process leading thus to the closed state of the turnstile (Fig. 1b). In order to prove the latter statement, the Zn2+ cation may be replaced by Pd(II) (complex 3, Scheme 1) adopting the square planar geometry without extension of its coordination sphere. In that case, the locking process cannot take place and thus the rotation of the rotor around the stator should not be altered (Fig. 1c).
![]() | ||
Fig. 1 Schematic representation of strapped-porphyrins: (a) free base (open state), (b) Zn–porphyrin (closed state), (c) Pd–porphyrin (open state). |
The synthetic strategy adopted for the preparation of the strapped porphyrin 1 (Scheme 2) was based on the condensation of the dipyrrylmethane 1322 and the α, ω dialdehyde 11. The latter was prepared from the commercially available 3-hydroxybenzaldehyde 14, tetraethyleneglycol 4 and 2,6-pyridine diol 6.
![]() | ||
Scheme 2 |
Compound 4 was first converted at room temperature in the presence of a catalytic amount of pyridinium para-toluenesulfonate (PPTS) into its mono-DHP protected derivative 523 upon treatment with dihydropyrane (DHP) in CHCl3 (34% yield). The latter was then condensed in the presence of NaH in refluxing THF with the dichloro compound 724 affording the compound 825 in 54% yield. Compound 7 was prepared in 82% yield in THF at room temperature upon treatment of 6 with SOCl2. The bis-OTHP derivative 8 was then deprotected under acidic conditions affording the diol 9 in 88% yield. The latter was quantitatively activated into its dimesylate derivative 10. The handle 11 was obtained in the presence of potassium carbonate in 96% yield in refluxing CH3CN upon reaction with 3-hydroxybenzaldehyde 14.
Dipyrrylmethane 1322 was obtained in 63% yield upon reaction of the aldehyde 12 with pyrrole in the presence of TFA. Condensation, in a CH2Cl2–EtOH mixture and in the presence of TFA, of the dialdehyde 11 with dipyrrylmethane 13 followed by oxidation in THF and in the presence of Et3N using DDQ,26 afforded strapped-porphyrin 1 in 2% yield.
In order to increase the yield of the final step, another synthetic strategy was explored. The compound 11 was converted in 66% into its bisdipyrrylmethane derivative 15 by reaction with an excess of pyrrole in the presence of TFA. Unfortunately, this strategy did not enhance the yield of the porphyrin formation. Indeed, the condensation of 15 in the presence of TFA with the aldehyde 12 in a CH2Cl2–EtOH mixture followed by oxidation using DDQ26 in the presence of Et3N in THF afforded the porphyrin derivative 1 in 1% yield.
The Zn–porphyrin complex 2 was quantitatively obtained in a CH2Cl2–MeOH mixture and at room temperature, upon treatment of 1 with Zn(OAc)2. The Pd–porphyrin complex 3 was obtained in 66% yield in a refluxing CHCl3–MeOH mixture, upon addition of Pd(OAc)2 to a solution of the porphyrin 1.
The assignment of all hydrogen atoms of the strapped porphyrins 1–3 was achieved at 25 °C in CD2Cl2 by 1- and 2-D NMR COSY and ROESY experiments. Using the same technique, the dynamics of compounds 1–3 in solution (CD2Cl2) was also studied at 25 °C.
For both porphyrins 1 and 3, a splitting of hydrogen atoms of the benzonitrile moieties Ha, and Hb into Ha, , and Hb, Hb (Scheme 1) was observed. This differentiation, resulting from the slow rotation of the handle around the porphyrin core within the NMR time scale, is in agreement with a previous study performed on an analogue system.18 The binding of the Pd(II) cation by the tetraaza core of the porphyrin has almost no influence on the H atoms of the handle. Indeed, when compared to the parent compound 1, signals corresponding to Hr and Hs of the pyridyl unit are only slightly shifted by 0.03 ppm and 0.06 ppm respectively. The same observation holds for Hu and Hv atoms belonging to the tetraethyleneglycol units which are shifted by 0.08 ppm and 0.04 ppm, respectively (Fig. 2a and c). These observations indicate the free rotation of the handle around the porphyrin stator defining thus the open state of the turnstile.
![]() | ||
Fig. 2 A portion (2.1–9.1 ppm) of the 1H-NMR spectra (CD2Cl2, 400 MHz, 25 °C) of 1 (a), 2 (b) and 3 (c). * and ¤ correspond to residual CHCl3 and MeOH, respectively. For assignment of H atoms see Scheme 1. |
This statement was further confirmed by a 2-D ROESY experiment at room temperature which revealed only the expected correlations between Hk and Hj and Hg due to spatial proximities resulting from the atomic connectivity in the free strapped porphyrin 1 and its Pd complex 3 (Fig. 3a and b).
![]() | ||
Fig. 3 Portions of the 2-D ROESY correlations for 1 (a), 3 (b) and 2 (c and d). * and ¤ correspond to residual chloroform and methanol, respectively. For attribution of H atoms see Scheme 1. |
For the Zn–porphyrin turnstile 2, the C2v symmetry of the molecule is conserved. The 1H-NMR investigations revealed, as expected, a drastic upfield shift of Hu and Hv atoms by 1.32 ppm and 0.53 ppm, respectively, implying the locking of the handle and the positioning of this portion of the handle within the anisotropic cone of the porphyrin backbone. The observed splitting increase between Hb and signals is another argument in favour of a blocked rotation. However, the pyridine Hr and Hs hydrogen atoms were found to be downfield shifted by 0.35 ppm and 0.85 ppm, respectively (Fig. 2b). This observation is not in agreement with a direct binding of the Zn(II) cation located within the centre of the porphyrin stator by the N atom of the pyridyl unit of the handle. Indeed, if this was the case, one would expect upfield shifts of Hr and Hs signals. Thus, the closed state of the turnstile must result from an indirect interaction between the handle and the stator. The 2-D ROESY sequence at room temperature revealed, in addition to the expected correlations between Hk and Hj and Hg, as also observed for 1 (Fig. 3a) and 3 (Fig. 3b), new correlations between Hg of the meso aromatic groups and Hu, Hv and Hq belonging to the handle (Fig. 3c). Furthermore, it also revealed the presence of correlations between Hu, Hv and Hq and a water molecule (Fig. 3d). This observation is consistent with the coordination of a water molecule by the Zn cation and interacting with the handle through a H-bond with the pyridyl moiety affording the complex 2·H2O (Scheme 1). This proposal also explains why the proton signals of Hr and Hs of the pyridyl unit are slightly downfield shifted. This hypothesis is substantiated in the crystalline phase by the structural investigation on a single crystal presented hereafter.
The solid state structures of both compound 2 and 3 were studied by X-ray diffraction on single crystals obtained upon slow diffusion of pentane into a CH2Cl2 solution of 2 (Fig. 4) or 3 (Fig. 5).
![]() | ||
Fig. 4 Solid state structure of compound 2·H2O. One of the two tetraethyleneglycol units connecting the handle to the porphyrin is found to be disordered. Hydrogen atoms and solvent molecules are omitted for clarity. For bond distances and angles, see text. |
![]() | ||
Fig. 5 Solid state structure of compound 3. Hydrogen atoms and solvent molecules are omitted for clarity. One of the ethyleneglycol units is found to be disordered. For bond distances and angles, see text. |
For the Zn–porphyrin 2·H2O complex, the structural study revealed that the crystal (triclinic, P) in addition to the complex 2 contains water and dichloromethane molecules (Fig. 4). No noticeable deformation of the porphyrin backbone is observed. The zinc cation is located above the porphyrin mean plane (0.441 Å). The coordination geometry around the metallic centre is a slightly distorted square based pyramid. Its coordination sphere is composed of all four N atoms of the porphyrin core with Zn–N distances in the 2.063(3)–2.083(3) Å range and N–Zn–N angles of 160.02(12)–160.97(12)° (N–Zn–Ntrans), 87.67(11)–89.05(11)° (N–Zn–Ncis). The water molecule occupies the apical position with the Zn–O distance of 2.120(3) Å and N–Zn–O angles of 98.53(11)–100.49(11)°. The four meso substituents are not coplanar but tilted with respect to the porphyrin mean plane with CCCC dihedral angles of −67.5° and −71.2° for the aryl units of the bridging handle and of −80.2° and 59.7° for the benzonitrile groups. One of the two tetraethyleneglycol units connecting the handle to the porphyrin is found to be disordered. Interestingly, the water molecule is hydrogen bonded to the pyridine nitrogen atoms of the handle with a O–N bond distance of 2.915(4) Å and NHO angles of 165(4)°.
For the Pd–porphyrin complex 3, the crystal (triclinic, P) is composed of the porphyrin 3, pentane and H2O solvent molecules (Fig. 5). The latter occupy in the crystal the empty space between compounds 3 without any specific interactions. The porphyrin backbone is not distorted. The palladium cation, located almost in the centre of the porphyrin core, is tetracoordinated. The coordination geometry around the metallic centre is almost square planar with Pd–N distances in the 1.995(9)–2.024(10) Å range and N–Pd–N angles of 178.3(4)–178.6(4)° (N–Pd–Ntrans), 89.7(4)–90.6(4)° (N–Pd–Ncis). The meso substituents are tilted with respect to the porphyrin mean plane with CCCC dihedral angle of −65.6° and −57.6° for the aryl units of the bridging handle and −75.0° and 61.0° for the benzonitrile groups. One of the glycol units is found to be disordered over two positions.
In order to unlock the turnstile, 1–10 equivalents of para-dimethylaminopyridine (DMAP), a competitive ligand, was added to 2 and the process monitored by 1H-NMR in CD2Cl2 at 25 °C (Fig. 6). As expected, upon addition of one equivalent of DMAP (Fig. 6b), the open state of the turnstile was restored. Indeed, by considering the “free-base” strapped-porphyrin 1 as a reference (Fig. 6d), the presence of DMAP caused a shift of all proton signals towards those observed for 1. In particular, Hr and Hs hydrogen atoms of the pyridyl moiety were upfield shifted by 0.16 ppm and 0.65 ppm respectively, whereas Hu and Hv signals were downfield shifted by 1.30 ppm and 0.55 ppm respectively. Furthermore, Hb and signals were found to be less differentiated. Upon further addition of DMAP, no additional shift of signals of the turnstile was observed (see Fig. 6c for addition of 2 eq. of DMAP). Dealing with DMAP, due to the fast exchange dynamics, its signals appeared to be broadened. However, for the 1/1 mixture of DMAP/2, in agreement with the coordination of DMAP to the Zn cation, the signal corresponding to the CH3 groups was found to be upfield shifted by 0.53 ppm. Upon increasing the DMAP/2 ratio from 1 to 10, all signals of DMAP became narrower and downfield shifted.
![]() | ||
Fig. 6 Portion of the 1H NMR spectra (CD2Cl2, 300 MHz, 25 °C) of 2·H2O (a) in the presence of 0 eq. (a), 1 eq. (b) and 2 eq. (c) of DMAP and of the parent compound 1 (d). * corresponds to signals of coordinated DMAP. For assignment of H atoms see Scheme 1. |
Crystals of 2 and 3 were obtained in a crystallization tube (0.5 × 20 cm) at room temperature upon slow diffusion of pentane into a solution of the complex in CH2Cl2 through a buffer layer of pentane/CH2Cl2 (1/1).
Crystallographic data for
2: C138.5H131ClN14O23Zn2, M = 2525.81, triclinic, P, a = 13.5277(4) Å, b = 15.3111(7) Å, c = 17.7296(5) Å, α = 86.083(3)°, β = 67.796(2)°, γ = 77.863(3)°, U = 3323.7(2) Å3, Z = 1, μ = 0.454 mm−1, refls measured: 34
801, independent refls: 14
673 [Rint = 0.0240], final R indices [I > 2σ (I)]: R1 = 0.0782, wR2 = 0.2357, R indices (all data): R1 = 0.0969, wR2 = 0.2528, GOF on F2: 1.063.
Crystallographic data for
3: C148H156N14O23Pd2, M = 2711.67, triclinic, P, a = 14.8196(12) Å, b = 17.1292(15) Å, c = 17.1532(17) Å, α = 63.953(3)°, β = 84.648(3)°, γ = 64.883(3)°, U = 3518.9(5) Å3, Z = 1, μ = 0.328 mm−1, refls measured: 12
750, independent refls: 11654 [Rint = 0.0378], final R indices [I > 2σ (I)]: R1 = 0.1132, wR2 = 0.2997, R indices (all data): R1 = 0.1510, wR2 = 0.3255, GOF on F2: 1.416. Owing to the weak diffraction power of the crystal, the data collection was performed with a theta value of 25°.
1H-NMR (CD2Cl2, 400 MHz) δ (ppm): −2.94 (s, 2H, NHy), 2.88 (m, 4H, OCH2v), 3.00 (m, 4H, OCH2q), 3.14 (m, 4H, OCH2p), 3.34 (m, 4H, OCH2o), 3.51 (m, 4H, OCH2n), 3.57 (s, 4H, OCH2u), 3.64 (m, 4H, OCH2m), 3.89 (m, 4H, OCH2l), 4.31 (m, 4H, OCH2k), 6.15 (d, 2H, Pyr, 3J = 7.5 Hz), 6.46 (t, 1H, Pys, 3J = 7.5 Hz), 7.36 (dd, 2H, Arj, 3J = 8.5 Hz, 4J = 2.0 Hz), 7.69 (t, 2H, Ari, 3J = 8.0 Hz), 7.76 (m, 2H, Arg), 7.88 (d, 2H, Arh, 3J = 7.5 Hz), 8.08 (d, 4H, Ara, 3J = 8.5 Hz), 8.33 (m, 4H, Arb), 8.76 (d, 4H, β-pyrc, 3J = 4.5 Hz), 8.94 (d, 4H, β-pyrd, 3J = 4.5 Hz). 13C-NMR (CD2Cl2, 100 MHz) δ (ppm): 68.0, 69.7, 69.9, 70.0, 70.4, 70.5, 70.7, 71.0, 73.3, 112.0, 114.5, 119.1, 119.2, 120.7, 122.2, 127.7, 128.0, 130.7, 131.1, 132.3, 135.4, 136.3, 143.1, 147.0, 157.1, 157.7. MS (ESI) for C69H65N7O10: 1152.30 g mol−1. m/z (M + Na+): calculated = 1174.469; measured = 1174.497.
1H-NMR (CD2Cl2, 400 MHz) δ (ppm): 2.25 (s, 4H, OCH2u), 2.35 (m, 4H, OCH2v), 2.79 (m, 4H, OCH2p), 2.81 (m, 4H, OCH2q), 3.13 (m, 4H, OCH2o), 3.49 (m, 4H, OCH2n), 3.65 (m, 4H, OCH2m), 3.89 (m, 4H, OCH2l), 4.32 (m, 4H, OCH2k), 6.50 (d, 2H, Pyr, 3J = 7.5 Hz), 7.31 (m, 3H, Pys, Arj), 7.62 (t, 2H, Ari, 3J = 8.0 Hz), 7.68 (br, 2H, Arg), 7.77 (d, 2H, Arh, 3J = 7.5 Hz), 8.02 (m, 6H, Ara,b), 8.29 (d, 2H, Arb, 3J = 8.0 Hz), 8.77 (d, 4H, β-pyrc, 3J = 4.5 Hz), 8.95 (d, 4H, β-pyrd, 3J = 4.5 Hz). 13C-NMR (CD2Cl2, 100 MHz) δ (ppm): 68.6, 69.3, 69.6, 69.6, 70.0, 70.5, 70.6, 70.9, 71.0, 111.5, 114.4, 118.7, 119.5, 120.0, 121.1, 122.6, 127.6, 128.1, 130.5, 131.5, 132.8, 135.0, 135.4, 137.8, 144.4, 148.3, 149.6, 150.5, 156.3, 157.5.MS (ESI) for C69H63N7O10Zn: 1215.66 g mol−1. m/z (M + Na+): calculated = 1236.382; measured = 1236.400.
1H-NMR (CD2Cl2, 400 MHz) δ (ppm): 2.92 (m, 4H, OCH2v), 3.07 (m, 4H, OCH2), 3.20 (m, 4H, OCH2p), 3.37 (m, 4H, OCH2o), 3.53 (m, 4H, OCH2n), 3.65 (m, 8H, OCH2u,m), 3.89 (m, 4H, OCH2l), 4.30 (m, 4H, OCH2k), 6.18 (d, 2H, Pyr, 3J = 7.5 Hz), 6.52 (t, 1H, Pys, 3J = 7.5 Hz), 7.35 (dd, 2H, Arj, 3J = 8.0 Hz, 4J = 2.0 Hz), 7.68 (m, 4H, Arg,i), 7.82 (d, 2H, Arh, 3J = 7.5 Hz), 8.07 (m, 4H, Ara), 8.26 (m, 2H, Arb), 8.32 (m, 2H, Arb), 8.72 (d, 4H, β-pyrc, 3J = 5.0 Hz), 8.91 (d, 4H, β-pyrd, 3J = 5.0 Hz). 13C-NMR (CDCl3, 125 MHz) δ (ppm): 68.0, 69.7, 69.8, 70.0, 70.3, 70.5, 70.7, 70.9, 73.2, 111.5, 114.9, 118.5, 119.2, 119.2, 121.2, 121.5, 127.6, 128.0, 130.8, 130.9, 132.2, 134.7, 136.6, 140.8, 141.0, 142.2, 146.2, 156.5, 157.1. MS (ESI) for C69H63N7O10Pd: 1255.37 g mol−1. m/z (M + Na+): calculated = 1278.359; measured = 1278.358.
1H-NMR (CDCl3, 300 MHz) δ (ppm): 1.38–1.85 (m, 6H, CH2(THP)), 2.83 (br, 1H, OH), 3.40–3.85 (m, 18H, OCH2, OCH2(THP)), 4.58 (m, 1H, CH(THP)). 13C-NMR (CDCl3, 75 MHz) δ (ppm): 19.6, 25.6, 30.7, 61.7, 62.4, 66.8, 70.1, 70.7, 70.8, 73.1, 99.1. Anal.: calculated for C13H26O6: C = 56.10%; H = 9.42%; found: C = 56.53%; H = 9.23%.
1H-NMR (CDCl3, 300 MHz) δ (ppm): 1.47–1.87 (m, 12H, CH2 (THP)), 3.45–3.90 (m, 36H, OCH2k–v, OCH2 (THP)), 4.62 (m, 2H, CH (THP)), 4.68 (s, 4H, OCH2u), 7.39 (d, 2H, Pyr, 3J = 8.0 Hz), 7.72 (t, 1H, Pys, 3J = 8.0 Hz). 13C-NMR (CDCl3, 75 MHz) δ (ppm): 19.6, 25.6, 30.7, 62.4, 66.8, 70.4, 70.7, 70.8, 70.8, 74.0, 99.1, 120.1, 137.5, 157.9. Anal.: calculated for C33H57NO12: C = 60.07%; H = 8.71%; N = 2.12%; found: C = 59.95%; H = 9.07%; N = 1.88%.
1H-NMR (CDCl3, 300 MHz) δ (ppm): 2.83 (br, 2H, OH), 3.58–3.75 (m, 32H, OCH2k–v), 4.67 (s, 4H, OCH2u), 7.38 (d, 2H, Pyr, 3J = 8.0 Hz), 7.71 (t, 1H, Pys, 3J = 8.0 Hz). 13C-NMR (CDCl3, 75 MHz) δ (ppm): 61.8, 70.4, 70.5, 70.7, 70.8, 72.7, 74.0, 120.2, 137.6, 157.9. Anal.: calculated for C23H41NO10 0.5 C4H10O: C = 56.80%; H = 8.77%; N = 2.65%; found: C = 56.79%; H = 8.70%; N = 2.62%.
1H-NMR (CDCl3, 300 MHz) δ (ppm): 3.06 (s, 6H, CH3), 3.62–3.77 (m, 28H, OCH2v–l), 4.36 (m, 4H, OCH2k), 4.68 (s, 4H, OCH2u), 7.39 (d, 2H, Pyr, 3J = 8.0 Hz), 7.73 (t, 1H, Pys, 3J = 8.0 Hz). 13C-NMR (CDCl3, 75 MHz) δ (ppm): 37.9, 69.1, 69.4, 70.4, 70.7, 70.8, 73.9, 120.3, 137.8, 157.8. Anal.: calculated for C25H45NO14S2: C = 46.36%; H = 7.00%; N = 2.16%; found: C = 46.35%; H = 6.87%; N = 2.05%.
1H-NMR (CDCl3, 300 MHz) δ (ppm): 3.65–3.75 (m, 24H, OCH2v–m), 3.87 (m, 4H, OCH2l), 4.18 (m, 4H, OCH2k), 4.67 (s, 4H, OCH2u), 7.19 (m, 2H, Ar), 7.35–7.44 (m, 8H, Ar, Pyr), 7.69 (t, 1H, Pys, 3J = 8.0 Hz), 9.95 (s, 2H, CHO). 13C-NMR (CDCl3, 75 MHz) δ (ppm): 67.9, 69.7, 70.4, 70.7, 70.8, 71.0, 74.0, 113.1, 120.1, 122.2, 123.7, 130.2, 137.9, 157.9, 159.5, 192.2.
1H-NMR (CDCl3, 300 MHz) δ (ppm): 3.65 (m, 24H, OCH2v–m), 3.79 (m, 4H, OCH2l), 4.03 (m, 4H, OCH2k), 4.62 (s, 4H, OCH2u), 5.41 (s, 2H, CH), 5.90 (m, 4H, pyr.), 6.12 (q, 4H, pyr., 3J = 3.0 Hz), 6.66 (m, 4H, pyr.), 6.77 (m, 6H, Ar), 7.18 (dd, 2H, Ar, 3J = 9.0 Hz, 3J = 7.5 Hz), 7.35 (d, 2H, Pyr, 3J = 7.5 Hz), 7.69 (t, 1H, Pys, 3J = 7.5 Hz), 8.15 (br, 4H, NH). 13C-NMR (CDCl3, 75 MHz) δ (ppm): 44.1, 67.4, 69.9, 70.4, 70.6, 70.8, 70.9, 107.2, 108.4, 112.9, 115.2, 117.3, 120.3, 121.2, 129.7, 132.5, 143.9, 157.7, 159.1.
Footnotes |
† This article is included in the All Aboard 2013 themed issue. |
‡ CCDC 894032 and 894033. For crystallographic data in CIF or other electronic format see DOI: 10.1039/c2nj40657h |
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013 |