Interplay of Chiral Auxiliary Ligand and Azide Bridging Ligand during

Jul 30, 2014 - United States. •S Supporting ..... them, with further confirmation by using solid state CD .... Sayajirao University of Baroda, Vadod...
0 downloads 0 Views 2MB Size
Download PDF
Communication pubs.acs.org/crystal

Interplay of Chiral Auxiliary Ligand and Azide Bridging Ligand during the Coordination Network Formation with Copper(II) Hemant M. Mande,† Prasanna S. Ghalsasi,*,† and Navamoney Arulsamy‡ †

Department of Chemistry, Faculty of Science, The Maharaja Sayajirao University of Baroda, Vadodara, India Department 3838, Department of Chemistry, University of Wyoming, 1000 E., University Avenue, Laramie, Wyoming 82071-2000, United States

‡

S Supporting Information *

ABSTRACT: Self-assembly formation of a bridging ligand and chiral auxiliary ligand with a metal center, in general, gets dominated by the former interaction. We report here that introduction of weak noncovalent interactions, away from the chiral center, can manifest molecular chirality in an auxiliary ligand into the self-assembled structure, a strategy useful for observing a magnetochiral dichroism effect. Structural details of azido bridged copper(II) coordination networks with enantiomerically pure, racemic, and admixtures of chiral benzylamine derivatives are discussed.

C

metal-auxiliary ligand complexation, and (c) the interionic and noncovalent interaction among auxiliary ligands. However, the literature is mostly focused on designs based on the first two interactions toward controlling the structures.4 One will naively expect a significant contribution of the third interaction when the ligand is chiral, chiral auxiliary ligand (CAL). But the impact of the chirality of ligands in the formation of coordination networks has not received the required attention.4 Although metal complexes with CALs are extensively reported in the literature, the CAL−CAL interactions are not consciously investigated. To the best of our knowledge, this is the first report where an inherent chirality in a ligand is “effectively” used to modify the overall self-assembly process and coordination network structure. This strategy will help in designing compounds with magnetochiral dichroism (MChD) effects, yet a challenge in molecular materials and chiral recognition of substrates. The present investigation highlights the self-assembly driven coordination complex formation reaction between Cu2+ as a metal center with azide as a bridging ligand and derivatives of

hirality is one of the most promising and unique properties of a molecule and plays a vital role in living systems and modern technology. Chiral organic molecules are used as auxiliary ligands in the synthesis of multifunctional molecular magnets to obtain noncentrosymmetric structures.1 However, in most cases, the structural chirality does not significantly influence the bulk properties such as magnetism, chiral recognition.2 One of the reasons could be that the chirality in the ligand does not directly act as a “superexchange” or “bridging” between two metal centers. This is because the most common bridging ligands are small, such as cyanide, azide, or hydroxyl, making it challenging to incorporate chirality in them.3 Therefore, a commonly adopted strategy is to use a chiral auxiliary ligand (CAL) with a chiral center near the metal center. But this strategy actually hinders “free interaction” or the self-assembly process between two adjacent chiral centers present on auxiliary ligand, making it difficult to transfer molecular chirality into the coordination network.1,2 A one-pot method is commonly adapted for the synthesis of coordination networks where the self-assembly process dictates the metal ions geometry and hence overall dimensionality. During the formation of self-assembly and crystallization of coordination networked structures, three types of interactions compete, namely, (a) metal-bridging ligand interaction, (b) © 2014 American Chemical Society

Received: June 30, 2014 Revised: July 20, 2014 Published: July 30, 2014 4254

dx.doi.org/10.1021/cg500951b | Cryst. Growth Des. 2014, 14, 4254−4257

Crystal Growth & Design

Communication

α-methylbenzylamine as CAL.5 It was observed that when achiral benzylamine was used as an auxiliary ligand (AL), it gets attached to Cu(II) with a distorted square planar structure which in turn is a part of a two-dimensional (2-D) sheet formed by two μ-1,1- (end-on, EO) and two μ-1,3- (end-to-end, EE) coordinated azide molecules between a distorted square pyramidal Cu2+ center.6 An EO azide bridge is observed between two Cu2+ cations leading to ferromagnetic behavior of this compound. Even after the introduction of chirality at the pro-chiral center, insertion of a methyl group or ethyl group at a benzylic position, the overall structure of the compound remains similar to that of the achiral compound.7 This indicates the chiral center present near the metal center “does not influence” the self-assembly process, and the final structure is driven by metal-bridging ligand interactions only. The major difference in the achiral and chiral compound is that the latter compounds exhibit characteristic circular dichoism (CD) spectra, although chirality is not observed in the structure. To make a CAL effective, we inserted a hydrogen bond acceptor group at the para-position on the chiral methyl benzylamine ligand. As per our expectation, chirality is observed in the crystal structure in the form of a onedimensional (1-D) helical chain, with the presence of only EEazide bridged networked structures, as discussed below. To observe the interplay between two ligands, we supported the present work by using an enantiomerically pure, racemic mixture and admixture of CALs. Thus, to understand the roles played by CAL and bridging ligands, three separate sets of experiments were planned. In all these experiments, the ratio of metal/bridging ligand to CAL was kept 1:10:0.5 in mole equivalents. Experiment I was devised to understand the effect of substitution away from the chiral center in enantiomerically pure CAL. Here, we used two derivatives, total four, of methyl benzyl amines as CAL 4chloro-α-methylbenzylamine (4-Clmba: two enantiomers R and S as AR and AS respectively), and 4-methyl-α-methylbenzylamine (4-Memba: two enantiomers R and S as CR and CS), to obtain complexes in the form of single crystals as I-AR [Cu((R)-4-Clmba)2(N3)2]n and I-AS [Cu((S)-4Clmba)2(N3)2]n, I-CR [Cu((R)-4-Memba)2(N3)2]n and ICS[Cu((S)-4-Memba)2(N3)2]n, respectively. Experiment II was devised to understand the self-assembly behavior in a racemic mixture of CAL over their enantiopure form. Here we carried out two experiments separately on racemic mixtures of 4-chloro-α-methylbenzylamine (ARS) and α-methylbenzylamine (BRS). Experiment III was undertaken to understand the competition between two CALs during the process of crystallization. Here we used an equivalent mixture of enantiomerically pure (R)-4-chloro-α-methylbenzylamine (AR) and (R)-α-methylbenzylamine (BR) as CAL. All the final products were characterized by elemental analysis, spectroscopic measurements, and single crystal X-ray diffraction data. IR spectra for all of the new compounds exhibit strong absorptions in the 2085 to 2050 cm−1 region due to the azide ligands. UV−vis spectra and CD-spectra were measured as KBr pellets, exhibiting a band in the 450−500 nm region due to the presence of d−d transition. Crystallographic structural data for I-AR, I-AS, I-CR, and I-CS reveal very similar features for the compounds, where compounds I-AR and I-As, I-CR, and I-CS are a pair of enantiomers. These compounds crystallize in the chiral space group P21. Surprisingly, these structures are quite different from that of the closely related [Cu3((R/S)-mba)2(N3)6]n7 and do

not contain EO-binding azide bridges, proving a change in the self-assembly process. As a representative example, the structure of I-AR is shown in Figure 1. The crystal structure shows it to

Figure 1. (a) Coordination sphere in I-AR showing the EE-(N3−N4− N5) and EO- (N6−N7−N8) modes of binding of the azide ligands. Thermal parameters are drawn at 50%. Hydrogen atoms are omitted and not all symmetric equivalent atoms are labeled for clarity; (b) 1D helical chains of I-AR and I-AS.

be a neutral helical chainlike molecule in which the CuII ion is five nitrogen coordinated in the form of a distorted square based pyramid, CuN5. The coordination sphere is composed of two CAL nitrogen atoms, a unidentate azide nitrogen atom and two EE-binding azide nitrogen atoms. CAL occupies two trans sites of the basal plane, and the other two trans sites are occupied by a bridging and unidentate azide ligands. A second bridging azide occupies the apical site. In other words, one end of the bridging azide coordinates to a copper center through the equatorial site, while the other end coordinates to a neighboring copper center axially. The bridging of the azide through axial and equatorial sites leads to a helical shape to the chain. This in turn helped molecular chirality get expressed by showing two different handedness in a 1-D chain; both enantiomers show the exact opposite helical chain. The axial Cu−NEE‑azide bond distances at 2.353(3) Å are significantly longer than the equatorial Cu−NE‑Eazide bond distance, 2.005(3) Å. The latter distances are also comparable to the Cu−Namine and the equatorial Cu−NEO‑azide bond distances. As could be expected, the N−N bond distances in the EE-bound azide are nearly equal at 1.173(5) and 1.167(4) Å, whereas those in the EO-bound azide are unequal. The copper atom is 0.185 Å above from the mean plane N1− N3−N2−N8. Each CuII center is linked to another CuII center by EE azide bridges to two neighbors leading to a neutral network. The helices are generated around the crystallographic 21 screw axes with a neighboring Cu···Cu distance of 5.305− 5.353 Å. In a pitch of helical chain the Cu1a···Cu1b distance is 7.286−7.427 Å. The distance between two pitches of adjacent helical chain in the ab-plane is 11.569−11.777 Å and 12.709− 12.912 Å in bc-plane. For EE bridges, Cu1−N6−N7 angle is 128.43−135.05° and the Cu1−N8b-N7b is 127.86−133.94°; the EE azide is quasilinear (N6−N7−N8, 176.07−176.46°). For EO, Cu1−N3−N4 angle is 120.95−122.16°; the EO azide is also quasi linear (N3−N4−N5, 177.58−177.76°). We also observed two more weak interactions in the form of helical chains between, (a) EO-azide with CAL amine hydrogen 4255

dx.doi.org/10.1021/cg500951b | Cryst. Growth Des. 2014, 14, 4254−4257

Crystal Growth & Design

Communication

isomers gets resolved into separate crystals and crystallized to give 1-D (I-AR and I-AS) helical azido networked compound. This means only conglomerate crystals were formed, due to the presence of noncovalent interactions. The supramolecular interchain weak Cl---N interaction between the halogen of CAL and a terminal nitrogen atom of the EO azide stabilized the helical chains during crystallization, which hindered insertion of opposite enantiomeric ligand and rotation of phenyl ring. As crystallized bunch of crystals, without selection, i.e., bulk, showed no CD activity indicating conglomerate nature of crystals in a 1:1 ratio. (Figure S5c,d, Supporting Information). Similarly, interestingly, a racemic mixture of CAL-BRS crystallized to give three different types of crystals, which means along with conglomerate crystals (Figure S10a,b, Supporting Information), racemic crystals (Figure S11a, Supporting Information) were also obtained as confirmed by CD spectra (Figure S5a,c, Supporting Information). A general formula for these compounds was [Cu3(mba)2(N3)6]n (Tables S1 and S2, Supporting Information). The bond distances and bond angles of racemic compound I-BRS are varied from P212121 acentric crystals I-BR and I-BR (Tables S7 and S8, Supporting Information). The conglomerates (I-BR and I-BS) crystallized into the P212121 acentric space group, while the racemic (I-BRS) crystallized into the Pbca space group. Here Cu(II) was found to exist in three crystallographically different environments, two with square pyramidal in which each one coordinated with five EO azide bridge and one with square planar geometry coordinated with nitrogen of two ligands and two EE azide bridge. Self-assembly formation of compound IBR and I-BS formed neutral two-dimensional (2-D) brick-wall layers with a repeating azido-bridged eight-membered copper brick. A distance between two such neutral 2-D brick-wall layers is 8.275 Å in an enantiomerically pure ligand, which in reality allows the rotation of CAL’s phenyl ring. Close inspection of the crystal structure, as shown in Figure 3, shows that the dihedral angle between two CAL varies in chiral (15.85°) and in racemic (0°) compounds. Because of this rotation, the racemic compounds exist in rare perfect square planar geometry attached to both enantiomers of CAL, BR and BS.9 In short, BRS after crystallization showed no noncovalent interactions (hydrogen or van der Waals bonding) making it easy for the rotation and insertion of opposite enantiomer during racemic coordination network formation. That means in this structure the bridging ligand leads the crystallization process. In Experiment III, both enantiomerically pure AR and BR were mixed together in a 1:1 ratio and used as an admixtured CAL. Here, we obtained only two types of crystals (I-AR and IBR), and no mixing of CALs was observed during the formation of crystals. The crystallization of two crystals is in quite the same ratio. The crystal obtained at high temperature (50 °C) and low temperature (∼0 °C) also showed the same result. That means the self-assembly process, although driven by CAL and bridging ligand, maintains the expression of individuality of the ligand during the overall crystallization process. Compounds (I-A and I-C) were also studied for magnetic interactions using a SQUID magnetometer (Figure S11, Supporting Information). The value of χmT (0.08751 emu·K· mol−1) was nearly equal to the value expected for single copper(II) ion (μeff = 0.83 μB, s = 1/2 and g = 2.00). The χ−1 vs T plots obey the Curie−Weiss law with a negative Weiss constant in the range of θ = −130 to −161 K. The high negative value of θ and decreasing values of χmT suggest strong

(N5b---N1H and N5---N1aH, 3.037 Å) to form an opposite helical chain with respect to the main helical chain (Figure S2A, Supporting Information); (b) EE-azide nitrogen and CAL amine hydrogen (N7bH---N6, 2.294 Å) via copper which is collinear with a 21 screw axis (Figure S2B, Supporting Information). The 1-D helices of compounds are further stabilized via interchain supramolecular interactions. In the case of I-AR and I-AS interchain weak Cl---N interaction, between the halogen and terminal nitrogen atom of the EO-azide (distance 3.239 Å, Figure S3, Supporting Information) and for I-CR and I-CS, the helix is further stabilized via CH---π interactions7 (Figure S4, Supporting Information). Thus, when enantiopure CAL AR, AS, CR, and CS were used the self-assembly process forced the bridging ligand to be only in EE-mode to communicate between adjacent metal centers for the formation of 1-D helical chain network leading to truly transforming chirality in the molecule to the bulk structure. This 1-D helical chain network is further complemented by observation of Cl---N and CH---π interactions among CAL. These structural features are well complemented by observation in CD measurements of ligands in methanol (Figure S1a−c, Supporting Information) and compounds in KBr pellets. Figure 2 shows positive Cotton effects at λmax =

Figure 2. Solid state CD spectra of I-AR and I-As as KBr pellets (inset shows a comparison of optical activity of I-AR with its CAL AR).

232 and 503 nm for I-AR (R-isomer), with exactly opposite signs for I-AS (S-isomer) at the same wavelength, thus confirming the optical activity and enantiomeric nature of IAR and I-AS. In all the compounds, the sign of the Cotton effect of the ligand is completely transformed in the compounds over all the UV−visible range (Supporting Information, Figure S1). The inset in Figure 2 supports the transfer of chirality of free ligand AR into I-AR. We would like to point out that in the literature, normally the comparison of CD spectra of ligand with that of the final compounds is not discussed. We observed only the shape variation in the Cotton effect in the d−d transition region for all compounds. Thus, there is a correlation between the chirality of the CAL used and the final crystallized structure after self-assembly.8 In Experiment II, racemic ligand-ARS and ligand-BRS were used as CAL. Our results are based on selecting 10 random crystals and carrying out single crystal X-ray measurements on them, with further confirmation by using solid state CD measurements. When racemic ligand (±)-4-chloro-α-methyl benzylamine (ARS) is used for complexation, R as well as S 4256

dx.doi.org/10.1021/cg500951b | Cryst. Growth Des. 2014, 14, 4254−4257

Crystal Growth & Design

Communication

support. The authors thank the DST-PURSE Single Crystal Xray Diffraction facility at the Faculty of Science, The Maharaja Sayajirao University of Baroda, Vadodara. Authors thank Prof. K. Inoue, Hiroshima University Japan, for valuable discussion and magnetic measurements.

■

Figure 3. (a) I-BRS, red circles indicate the chiral centers; (b) dihedral angle between two CAL of I-BRS-0.00°.

antiferromagnetic interactions between neighboring Cu(II) ions through the EE-azide bridges, as seen in the crystal structure. In conclusion, we have shown by a proper substitution in a molecule, away from the chiral center, one can enhance or trigger chiral ligand−chiral ligand interaction in a self-assembly driven one-pot crystal formation reaction to design multifunctional molecular magnets. These CAL−CAL interactions will not only play an important role in crystal growth but might help in observing novel physical behavior, particularly magnetochiral dichroism effects.

■

ASSOCIATED CONTENT

S Supporting Information *

Complete synthetic procedure and characterization data and figures showing supramolecular Cl---N interactions and supramolecular C−H···π interactions. Crystallographic data table and cif files of I-AR, I-AS, I-CR, and I-CS. Crystal parameters, ORTEP diagrams of I-AR, I-AS, I-CR, and I-CS, H-bonding data, magnetic data. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

(1) Train, C.; Gruselle, M.; Verdaguer, M. Chem. Soc. Rev. 2011, 40, 3297. (2) (a) Crassous, J. Chem. Soc. Rev. 2009, 38, 830. (b) Crassous, J. Chem. Soc. Rev. 2012, 48, 9669. (c) Crassous, J. Chem. Commun. 2012, 48, 9687. (d) Kawamoto, T.; Suzuki, N.; Ono, T.; Gong, D.; Konno, T. Chem. Commun. 2013, 49, 668. (e) Li, Z.-Z.; Yao, S.-Y.; Wu, J.-J.; Ye, B.-H. Chem. Commun. 2014, 50, 5644. (3) (a) Chorazy, S.; Nakabayashi, K.; Imoto, K.; Mlynarski, J.; Sieklucka, B.; Ohkoshi, S.-I. J. Am. Chem. Soc. 2012, 134, 16151. (b) Pardo, E.; Train, C.; Lescouëzec, R.; Journaux, Y.; Pasa, J.; Ruiz-P, C.; Delgado, F. S.; Ruiz-Garcia, R.; Lloret, F.; Paulsene, C. Chem. Commun. 2010, 46, 2322. (c) Ferrando-Soria, J.; Cangussu, D.; Eslava, M.; Journaux, Y.; Lescouë zec, R.; Julve, M.; Lloret, F.; Pasán, J.; RuizPérez, C.; Lhotel, E.; Paulsen, C.; Pardo, E. Chem.Eur. J. 2011, 17, 12482. (4) (a) Zheng, X.-D.; Hua, Y.-L.; Xiong, R.-G.; Ge, J.-Z.; Lu, T.-B. Cryst. Growth Des. 2011, 11, 302. (b) Wen, H.-R.; Tang, Y.-Z.; Liu, C.M.; Chen, J.-L.; Yu, C.-L. Inorg. Chem. 2009, 48, 10177. (c) Inoue, K.; Imai, H.; Ghalsasi, P. S.; Kikuchi, K.; Ohba, M.; O̅ kawa, H.; Yakhmi, J. V. Angew. Chem., Int. Ed. 2001, 40, 4242. (d) Magnetism: Molecules to Materials V; Miller, J. S.; Drillon, M., Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2005. (e) Ribas, J.; Monfort, M.; Solans, X.; Drillon, M. Inorg. Chem. 1994, 33, 742. (f) Maji, T. K.; Mukherjee, P. S.; Chaudhuri, N. R.; Mostafa, G.; Mallah, T.; CanoBoquera, J. Chem. Commun. 2001, 1012. (g) Mukherjee, S.; Mukherjee, P. Acc. Chem. Res. 2013, 46, 2556. (5) Most of the azide mediated molecular magnet procedures use NaN3 in quite large excess, minimum 10 times, in order to have high nuclearity or nucleation in the final complex. (6) Zhen, S.; Jing-Lin, Z.; Gao, S.; Song, Y.; Che, C.-M.; Fun, H.-K.; You, X.-Z. Angew. Chem., Int. Ed. 2000, 39, 3633. (7) (a) Gu, Z.-G.; Xu, Y.-F.; Yin, X.-J.; Zhou, X.-H.; Zuo, J.-L.; You, X.-Z. Dalton Trans. 2008, 5593. (b) Gu, Z.-G.; Song, Y.; Zuo, J.-L.; You, X.-Z. Inorg. Chem. 2007, 46, 9522. (8) (a) Li, H.-Y.; Jiang, L.; Xiang, H.; Makal, T. A.; Zhou, H.-C.; Lu, T.-B. Inorg. Chem. 2011, 50, 3177. (b) Zheng, X.-D.; Jiang, L.; Feng, X.-L.; Lu, T.-B. Dalton Trans. 2009, 6802. (c) Ou, G.-C.; Jiang, L.; Feng, X.-L.; Lu, T.-B. Inorg. Chem. 2008, 47, 2710. (d) An, H.-Y.; Wang, E.-B.; Xiao, D.-R.; Li, Y.-G.; Su, Z.-M.; Xu, L. Angew. Chem., Int. Ed. 2006, 45, 904. (9) Desjardins, S. R.; Penfield, K. W.; Cohen, S. L.; Musselman, R. L.; Solomon, E. I. J. Am. Chem. Soc. 1983, 105, 4590.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS P.S.G. and H.M.M. thank UGC-DAE consortium for Scientific Research, Kalpakkam Node (CRS-K-04/19) for funding this research programme. P.S.G. thanks to Department of Science and Technology, New Delhi (SR/S1/IC-43/2009), for financial 4257

dx.doi.org/10.1021/cg500951b | Cryst. Growth Des. 2014, 14, 4254−4257