Solvent-Mediated Reversible Structural ... - ACS Publications

1 downloads 0 Views 4MB Size Report
folding and unfolding of the HgI2 single-stranded helical coordination polymers. Page 2 of 38. ACS Paragon Plus Environment. Crystal Growth & Design. 1. 2. 3.
Article pubs.acs.org/crystal

Solvent-Mediated Reversible Structural Transformation of Mercury Iodide Coordination Polymers: Role of Halide Anions Pradhumna Mahat Chhetri, Xiang-Kai Yang, and Jhy-Der Chen* Department of Chemistry, Chung-Yuan Christian University, Chung-Li, Taiwan, R.O.C. S Supporting Information *

ABSTRACT: Solvothermal reactions of HgX2 with 2,2′-(1,2phenylene)-bis(N-pyridin-3-yl)acetamide, L, in ethanol afforded [Hg(L)X2]n (X = Cl, 1; Br, 2; I, 3), which are isostructural one-dimensional (1D) zigzag chains, while layering reactions of a ethanoic solution of L with a methanoic solution and an acetonitrile solution of HgI2, respectively, gave 1D helical chains [Hg(L)I2·MeOH]n, 4, and [Hg(L)I2· MeCN]n, 5. In marked contrast to 1 and 2, the iodidecontaining 3 is able to exhibit reversible structural transformation with 4 and 5 by adsorption and desorption of methanol and acetonitrile, suggesting the importance of N−H---X and Hg---X interactions in the evaluation of structural transformation. Moreover, complexes 4 and 5 exhibit reversible crystal to crystal transformation triggered by solvent exchange. Complexes 3−5 represent a unique example that the solvents show a significant effect on folding and unfolding of the HgI2 single-stranded helical coordination polymers.



INTRODUCTION The investigation in the rational design and synthesis of novel coordination polymers (CPs) continues to be an intense area of research activity during recent years as it extends the range of potential applications in the areas such as magnetism, luminescence, catalysis, and gas storage and sensing.1−3 Much effort has been devoted to the construction of CPs with helical structures to mimic the biological helices,4 because many DNA and proteins possess specific ordered structures such as a righthanded double helix and single α-helix. One of the most intriguing features of CPs is the versatile structures that lead to structural transformation by the influence of various internal and external factors such as heat, light, mechanochemical energy, solvation/desolvation, and association and dissociation of some chemical species to the center metal by changing coordination number and counteranion.5−7 Despite some recent progress, the ability to predict and control the structural transformation remains an elusive goal, and much more work is required for elucidating the details of such interesting phenomena. A lot of coordination compounds have been reported either in reversible or irreversible structural transformation;8,9 however, less mercury compounds have been explored for such a property. Mahmoudi et al. reported that the dimeric [Hg2(μ-bpdb)I4] (bpdb = 1,4-bis(2-pyridyl)-2,3-diaza-1,3-butadiene) polymerized on heating to form irreversibly a twodimensional (2D) coordination polymer [Hg2(μ-bpdb)(μI)2I2]n.10 Mobin et al. prepared one-dimensional (1D) polymeric chain [(Cl)Hg(μ-Cl)2(hep-H)]n (hep-H = 2-(2hydroxyethyl)pyridine) which can be reversibly transformed to a 2D network as a function of the temperature,11 and Hou et al. © 2017 American Chemical Society

have shown that the 2D [Hg3(μ-quinoxaline)2(μ-SCN)6]n can be reversibly transformed to the three-dimensional (3D) [Hg(μ-quinoxaline)(μ-CN)2]n by a solid state anion replacement.12 We have recently reported two pairs of 1D Hg(II) supramolecular isomers involving 1D helical and sinusoidal chains by using the flexible N,N′-di(3-pyridyl)adipoamide ligand which were able to carry out irreversible structural transformation under simple heating or hydrothermal conditions.13 Mercury is a soft base and has good affinity with N and S rather than other metal ions in the same group in periodic table. A variety in structures can be observed by the influence of the halide anions even though it is difficult to predict which anions give similar or different structures.14−20 Sometimes HgCl2 and HgBr2 play the same structural role but HgI2 is different,21−23 while in other cases HgBr2 and HgI2 play the same role but HgCl2 is distinct.21,22,24,25 The structural diversity of the Hg(II) halide complexes depends on the identity of the spacer ligand17 and experimental conditions such as temperature and solvents as well.19 The flexible bis-pyridyl-bis-amide (bpba) ligands which possess different -(CH2)n- skeletons have been used as the spacer to construct various coordination polymers during recent years. Structures of different dimensions with or without participation of auxiliary polycarboxylate ligands have been investigated.26 Two amide groups (−NH−CO−) present in them, which play important roles as the abundant potential hydrogen bond sites. To investigate the structure-directing role Received: May 28, 2017 Revised: July 12, 2017 Published: August 10, 2017 4801

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809

Crystal Growth & Design

Article

Table 1. Crystal Data for Complexes 1−5 compound formula fw crystal system space group a, Å b, Å c, Å α, deg β, deg γ, deg V, Å 3 Dcal, g cm−3 F(000) Z μ(Mo Kα), mm−1 range (2θ) for data collection, deg reflections collected independent reflections data/restraints/parameters quality-of-fit indicatorc final R indices [I > 2σ(I)]a,b R indices (all data)

1

2

3

4

5

C20H18Cl2HgN4O2 617.87 monoclinic P21/c 12.025(2) 18.959(3) 9.1192(16) 90 97.886(10) 90 2059.4(6) 1.993 1184 4 7.758 4.99−56.70

C20H18Br2HgN4O2 706.79 monoclinic P21/c 12.0882(4) 19.3558(6) 9.0541(2) 90 96.876(2) 90 2103.21(11) 2.232 1328 4 11.142 3.99−56.66

C20H18HgI2N4O2 800.77 monoclinic P21/c 12.3425(5) 20.0881(9) 8.9811(4) 90 95.840(3) 90 2215.19(17) 2.401 1472 4 9.759 3.31−56.67

C21H22HgI2N4O3 832.81 monoclinic Cc 14.2815(2) 18.5108(3) 9.6452(1) 90 98.846(1) 90 2519.49(6) 2.196 1544 4 8.588 3.63−56.69

C22 H21HgI2N5O2 841.83 monoclinic Cc 13.5788(1) 19.9068(2) 9.4786(1) 90 94.716(1) 90 2553.49(4) 2.19 1560 4 8.473 3.64−56.65

18247 5042 [R(int) = 0.0853] 5042/0/270 1.002 R1 = 0.0522, wR2 = 0.1189 R1 = 0.0850, wR2 = 0.1364

20743 5229 [R(int) = 0.0509] 5229/0/262 1.025 R1 = 0.0388, wR2 = 0.0752 R1 = 0.0695, wR2 = 0.0839

21883 5489 [R(int) = 0.0811] 5489/0/262 1.002 R1 = 0.0610, wR2 = 0.1168 R1 = 0.1346, wR2 = 0.1429

21850 5465 [R(int) = 0.0287] 5465/2/281 1.071 R1 = 0.0242, wR2 = 0.0555 R1 = 0.0264, wR2 = 0.0564

22720 5639 [R(int) = 0.0288] 5639/2/290 1.002 R1 = 0.0227, wR2 = 0.0482 R1 = 0.0251, wR2 = 0.0490

a R1 = ∑∥Fo| − |Fc∥/∑|Fo. bwR2 = [∑w(Fo2 − Fc2)2/∑w(Fo2)2]1/2. w = 1/[σ2(Fo2) + (ap)2 + (bp)], p = [max(Fo2 or 0) + 2(Fc2)]/3. a = 0.0660, b = 0, 1; a = 0.0343, b = 0.8730, 2; a = 0.0602, b = 0, 3; a = 0.0217, b = 0.5779, 4; a = 0.0107, b = 0, 5. cQuality-of-fit = [∑w(|Fo2| − |Fc2|)2/Nobserved − Nparameters)]1/2.

7.29−7.33(m, 4H); 7.23−7.21(m, 2H); 3.83 (s, 4H); Figure S1. IR (KBr disk, cm−1): 3309(w), 3253(w), 3226(m), 3172(m), 3114(m), 3069(w), 2979(m), 2860(m), 1694(m), 1669(s), 1607(s), 1553(s), 1483(s), 1411(m), 1353(m), 1288(m), 1240(m), 1213(m), 1188(m), 1147(m), 1104(w), 1048(w), 1027(w), 957(w), 885(w), 805(m), 753(m), 718(m), 700(m), 625(w), 578(w), 543(w); Figure S2. Synthesis of [Hg(L)Cl2]n, 1. Mercury(II) chloride (0.027 g, 0.10 mmol), L (0.035 g. 0.10 mmol) and 7 mL of ethanol were sealed in a 23 mL Teflon-lined stainless steel autoclave, which was then heated under autogenous pressure at 120 °C for 48 h followed by slow cooling at a rate of 2 °C h−1 to RT. Slow cooling of the reaction system afforded needle shape colorless crystals suitable for singlecrystal X-ray diffraction. Yield: 0.037 g (60%). Anal. calcd for 1, C20H18Cl2HgN4O2 (MW = 617.87): C, 38.87; N, 9.06; H, 2.94%. Found: C, 38.26; N, 8.86; H, 2.94%. IR (cm−1): 3343(s), 3182(m), 3123(m), 3070(s), 3045(m), 2916(m), 2360(m), 1706(s), 1664(s), 1598(s), 1585(s), 1537(s), 1482(s), 1455(s), 1426(s), 1411(s), 1363(s), 1344(s), 1324(s), 12969s), 1280(s), 1244(s), 1210(s), 1196(s), 1185(s), 1166(s), 1124(s), 1102(s), 1048(s), 1026(m), 971(m), 958(m), 939(m), 919(m), 861(m), 806(s), 758(s), 732(m), 401(s), 634(m), 620(m), 570(m), 554(m), 492(w), 458(w), 420(w). An alternative method for 1: A 20 mL ethanoic solution of L (0.069 g, 0.20 mmol) was prepared and a 20 mL of ethereal solution of HgCl2 (0.054 g, 0.20 mmol) was layered down slowly without disturbance on top. After 3 weeks, needle-shaped colorless crystals of 1 were generated. Yield: 0.094 g (76%). The structure of this product obtained by layering was verified by using powder X-ray pattern as the one prepared by using a solvothermal reaction, Figure S3. Synthesis of [Hg(L)Br2]n, 2. Prepared as described for 1, except that HgBr2 (0.036 g, 0.10 mmol) and L (0.035 g. 0.10 mmol) were used. Yield: 0.037 g (52%). Anal. calcd for C20H18Br2HgN4O2 (MW = 706.77): C, 33.99; N, 7.93; H, 2.57%. Found: C, 34.17; N, 7.99; H, 2.6%. IR (cm−1): 3452(s), 3438(s), 3349(s), 3239(s), 3115(s), 3066(s), 2915(s), 1960(w), 1916(w), 1822(w), 1707(s), 1683(s), 1680(s), 1598(s), 1584(s), 1535(s), 1481(s), 1448(s), 1428(s), 1425(s), 1361(s), 1341(s), 1323(s), 1285(s), 1279(s), 1244(s),

of the bpba ligand based on the rigid angular ditopic group in the formation of helical, mercury halide-containing coordination polymers that may or may not be affected by the halide ions to construct the structure as well as to impact its properties, we have replaced the flexible −(CH2)n− skeleton of the bpba ligand with the 1,2-phenyl-bis(methylene) group (−CH2−Ph−CH2−) to prepare 2,2′-(1,2-phenylene)-bis(Npyridin-3-yl)acetamide (L), which were reacted with the Hg(II) halide salts. Herein, we report the synthesis, structures, and structural transformation of five Hg(II) coordination polymers containing L. The effect of the halide anion on the structural transformation is also discussed.



EXPERIMENTAL SECTION

General Procedures. Elemental analyses were performed on a PE 2400 series II CHNS/O or an elementar Vario EL-III analyzer. The 1H NMR spectrum was recorded on a Bruker Avance II 400 MHz FTNMR. IR spectra (KBr disk) were obtained from a JASCO FT/IR-460 plus spectrometer. Powder X-ray diffraction was carried out using a Bruker D2 PHASER diffractometer with CuKα (λα = 1.54 Å) radiation. Emission spectra were recorded on a Hitachi F-4500 spectrometer. Materials. The reagent 1,2-phenylenediacetic acid was purchased from ACROS Co. 3-Aminopyridine, pyridine, tripenylphosphite, and mercury(II) halide salts were purchased from Alfa Aesar Co. Synthesis of L. 3-Aminopyridine (1.88 g, 20 mmol) solution in 10 mL of pyridine was mixed with 25 mL of pyridine solution of 1,2phenyldiacetic acid (1.94 g, 10 mmol) and was made a homogeneous mixture by stirring for 20 min. Triphenyl phosphite (6.2 g, 20 mmol) was added and stirred gently for 2 h. The mixture was then refluxed for 24 h. After air cooling to room temperature, it was washed with deionized water several times to get a pure form of L. Yield: 2.95 g (85%). Anal. calcd for C20H18N4O2 (MW = 346.38): C, 69.34; N, 16.17; H, 5.23%. Found: C, 70.27; N, 15.9; H, 4.55%. 1H NMR (DMSO-d6, δ): 10.38(s, 2H); 8.72(d, 2H); 8.23(d, 2H); 7.99 (d, 2H); 4802

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809

Crystal Growth & Design

Article

Table 2. Selected Bond Distances (Å) and Angles (deg) for Complexes 1−5 Hg−N(1) Hg−N(4A) Hg−X(1) Hg−X(2) ∠N(1)−Hg−N(4A) ∠N(1)−Hg−X(1) ∠N(1)−Hg−X(2) ∠N(4A)−Hg−X(1) ∠N(4A)−Hg−X(2) ∠X(1)−Hg−X(2)

1

2

3

4

5

2.4628(4) 2.4653(4) 2.3685(4) 2.3529(4) 88.950(3) 91.815(5) 109.152(5) 96.835(5) 96.835(3) 155.496(5)

2.4756(1) 2.4581(1) 2.4745(1) 2.4624(1) 88.874(0) 92.751(1) 108.032(1) 98.305(0) 97.685(1) 153.927(1)

2.5181(81) 2.473(103) 2.6212(9) 2.6227(8) 87.6(3) 96.3(2) 106.7(2) 100.5(2) 100.3(2) 149.44(3)

2.4204(68) 2.4287(56) 2.6417(8) 2.6317(6) 90.4(2) 104.15(13) 101.77(14) 100.61(14) 104.98(15) 143.13(2)

2.4528(53) 2.3966(42) 2.6678(6) 2.6465(5) 96.51(15) 101.39(11) 102.12(12) 102.29(12) 104.16(11) 142.037(19)

1208(s), 1194(s), 1185(s), 1166(s), 1123(s), 1100(s), 1046(s), 1026(s), 970(s), 958(s), 939(m), 916(m), 874(w), 861(w), 805(s), 757(s), 730(s), 700(s), 671(s), 632(s), 607(s), 569(s), 552(s), 514(m), 506(m), 492(w), 458(w), 410(w). An alternative method for 2: A 20 mL ethanoic solution of L (0.069 g, 0.20 mmol) was prepared and a 20 mL methanoic solution of HgBr2 (0.072 g, 0.20 mmol) was layered down slowly without disturbance on top. After 3 weeks, needle shape colorless crystals of 2 were generated. Yield: 0.024 g (34%). The structure of this product obtained by layering has been verified by using powder X-ray pattern as the one prepared by using solvothermal reaction, Figure S4. Synthesis of [Hg(L)I2]n, 3. Prepared as described for 1, except that HgI2 (0.045 g, 0.10 mmol) and L (0.035 g, 0.10 mmol) were used. Yield: 0.056 g (70%). Anal. calcd for C20H18HgI2N4O2 (MW = 800.77): C, 29.997; N, 6.997; H, 2.266%. Found: C, 30.348; N, 7.028; H, 2.322%. IR (cm−1): 3356(s), 3237(m), 3175(m), 3119(m), 3055(m), 2909(m), 1958(w), 1915(w), 1817(w), 1711(s), 1654(s), 1597(m), 1582(s), 1535(s), 1481(s), 1455(m), 1422(s), 1362(m), 1341(m), 1322(m), 1294(s), 1277(m), 1244(m), 1208(m), 1194(m), 1183(m), 1166(m), 1122(m), 1099(m), 1045(m), 1025(m), 968(m), 958(m), 939(m), 914(m), 899(w), 860(w), 805(m), 784(w), 756(m), 730(m), 701(m), 630(m), 595(w), 570(m), 559(m), 551(m), 513(w), 492(w), 457(w), 410(w). Synthesis of [Hg(L)I2.MeOH]n, 4. A 20 mL methanoic solution of HgI2 (0.091 g, 0.20 mmol) was prepared and layered on top of a 20 mL of ethanoic solution of L (0.069 g, 0.20 mmol). After 3 weeks, colorless column-shaped crystals of 4 were generated. Yield: 0.11 g (66%). Anal. calcd for C21H22HgI2N4O3 (MW = 832.81): C, 30.29; N, 6.73; H, 2.66%. Found: C, 30.03; N, 6.78; H, 2.61%. IR (cm−1): 3900(w), 3851(w), 3836(w), 3799(w), 3748(w), 3733(w), 3709(w), 3688(w), 3647(w), 3585(w), 3355(w), 3242(m), 3183(m), 3130(m), 3101(m), 3068(m), 2923(m), 2350(w), 1917(w), 1667(s), 1603(s), 1541(s), 1481(s), 1455(m), 1415(s), 1341(s), 1325(m), 1288(s), 1244(m), 1194(m), 1180(m), 1124(m), 1101(w), 1051(m), 1023(m), 961(m), 939(m), 909(m), 878(m), 807(m), 772(m), 727(m), 698(m), 643(m), 632(m), 575(m), 516(m), 458(w), 446(w), 432(w), 410(w) . Synthesis of [Hg(L)I2·MeCN]n, 5. A 20 mL acetonitrile solution of HgI2 (0.091 g, 0.20 mmol) was prepared and layered on top of a 20 mL ethanoic solution of L (0.069 g, 0.20 mmol). After 1 week, colorless column-shaped crystals of 5 were generated. Yield: 0.11 g (65%). Anal. calcd for C22H21HgI2N5O2 (MW = 841.83): C, 31.38; N, 8.31; H, 2.51%. Found: C, 31.32; N, 8.20; H, 2.27%. IR (cm−1): 3326(w), 3237(w), 3179(w), 3124(w), 3062(w), 2924(w), 2295(w), 2262(w), 1923(w), 1702(m), 1660(s), 1604(m), 1543(s), 1482(s), 1411(s), 1346(s), 1310(w), 1295(m), 1244(w), 1194(m), 1155(m), 1122(w), 1099(w), 1049(w), 968(w), 805(m), 779(w), 731(m), 700(m), 633(w), 573(w), 515(w), 408(w). X-ray Crystallography. The diffraction data for complexes 1 − 5 were collected on a Bruker AXS SMART APEX II CCD diffractometer at 22 °C, which was equipped with a graphite-monochromated MoKα (λα = 0.71073 Å) radiation.27 Data reduction was performed by use of well-established computational procedures.28 The structure factors were obtained after Lorentz and polarization, followed by empirical absorption correction based on “multi-scan”. The positions of some of

the heavier atoms were located by the direct or Patterson method and the remaining atoms were found in a series of alternating difference Fourier maps and least-squares refinements. The hydrogen atoms were added by using the HADD command in SHELXTL. Basic information pertaining to crystal parameters and structure refinement is summarized in Table 1. Selected bond distances and angles are listed in Table 2.



RESULTS AND DISCUSSION Crystal Structures of 1−3. Isomorphous crystals of 1−3 conform to the monoclinic crystal system and space group P21/ c. Figure 1a depicts a representative drawing showing the coordination environment about the metal ion. The Hg(II) ion is in a distorted tetrahedral geometry (τ4 = 0.67, 1; 0.69, 2; 0.73, 3),29,30 which is coordinated with two pyridyl nitrogen atoms and two halide anions. The X−Hg (X = Cl, 1; Br, 2; I, 3) bond angle is much greater than the corresponding N−Hg−N bond angle, and the X−Hg−X angle of 3 is significantly smaller than those of 1 and 2, while the Hg−N distances in 1−3 are similar. Moreover, while the Hg−Cl distances are significantly shorter than the corresponding Hg−N distances and the Hg− Br and Hg−N distances are similar, the Hg−I distances are significantly larger. The X−Hg−X angles in 1−3 that are much greater than the tetrahedral value indicate that halide anion is a better donor toward Hg(II) than nitrogen atom of the pyridyl ring, and the wider X−Hg−X angles in the sequence from X = I to X = Cl can be ascribed to the increase in s-orbital character of the hybrid orbital used by mercury ion in the Hg−X bond.31 Complexes 1−3 show zigzag 1D linear chains having N−H--O hydrogen bonds between amine hydrogen atoms and carbonyl oxygen atoms within a chain, Figure 1b. Another amine hydrogen atom in the same ligand is involved in the N− H---X hydrogen bond with the halide ion from adjacent chain. Besides, short contacts between mercury and halide ions are also observed within different layers, and the distance between these two atoms is less than sum of van der Waals radii of respective atoms, Figure 1c. Details of N−H---O, N−H---X, and Hg---X interactions are listed in Table 3. A series of double C−H---π interactions resulting from methylene hydrogen atoms to the centroids of the phenyl rings and from the pyridyl hydrogen atoms to the centroids of pyridyl rings can be observed in 1−3. Moreover, the C−H---π interactions resulting from the pyridyl hydrogen atoms to the centroids of pyridyl rings form a continuous fashion like face-to-edge and edge-toface, Figure 1d. Table S1 shows the C−H---π distances. The typical cutoff distance for H---π distance is 3.05 Å on the basis of Pauling’s value for the half thickness of a phenyl ring (1.85 Å)32 and the van der Waals radius of H (1.20 Å).33 As shown in Table 3, the H---Cl (2.62 Å), H---Br (2.78 Å), and H---I (3.03 Å) distances of the N−H---X (X = Cl, 1; Br, 2 4803

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809

Crystal Growth & Design

Article

and I, 3) hydrogen bonds are shorter than their corresponding van der Waals contacts which are 2.95, 3.05, and 3.18 Å (van der Waals radius for H = 1.20, Cl = 1.75, Br = 1.85, I = 1.98, and Hg = 1.98 Å),34 and the differences are 0.33, 0.27, and 0.15 Å, respectively, indicating much weaker N−H---I hydrogen bonds. Moreover, while the Hg---Cl distances of 3.4637(5) and 3.4224(5) Å and Hg---Br distances of 3.7051(7) Å are shorter than their corresponding van der Waals contacts of 3.73 and 3.83 Å, respectively, the Hg---I distance of 4.1134(10) Å is outside its upper limit of 3.96 Å, indicating the absence of Hg--I interaction. For the qualitative assessment of the relative strength of various hydrogen bonds, hydrogen bond distances can be evaluated in a common scale by using normalized distance function [RHX = d(H---X)/(rH + rX)] suggested by Lommerse et al. on the basis of van der Waals radii, rH and rX, which takes account of the different sizes of the halogen atoms.35,36 Accordingly, the normalized distances RHX (Å) for H---X contacts in 1−3 are calculated as 0.888, 0.911, and 0.953, respectively, indicating that the normalized distance of 3 is greater which also reflects a weaker hydrogen bond as compared to 1 and 2. Crystal Structures of 4 and 5. Crystals of both 4 and 5 conform to the monoclinic space group Cc. Figure 2a depicts a representative drawing showing the coordination environment about the metal ion. The Hg(II) ion adopts a distorted tetrahedral geometry (τ4 = 0.79 for 4 and 0.80 for 5), which is coordinated with two pyridyl nitrogen atoms and two chloride anions, forming a 1D helical chain, Figure 2b. A span of complete helical chain is 14.27 and 13.58 Å, respectively, for 4 and 5, involving two mercury atoms and one L ligand. The 1D helical chains of 4 are connected by two kinds of hydrogen bonds. In the first case the methanol molecules act like bridging ligands for O−H---O [H---O = 1.92 Å, ∠O−H---O = 179.4°] hydrogen bonds linking the amide oxygen atoms and N−H---O [O---H = 2.03 Å; ∠N−H---O = 155.6°] hydrogen bonds originating from the amine hydrogen atoms. In another case the interchain N−H---O [H---O = 2.02 Å; ∠N−H---O = 166.6°] hydrogen bonds between amide oxygen and amine hydrogen atoms can be shown, Figure 2c. In marked contrast, the helical chains of 5 are interlinked through only interchain N−H---O [H---O = 1.93 Å; ∠N−H---O = 179.2°] hydrogen bonds, and the acetonitrile molecules interact with the amine hydrogen atoms through N−H---N [H---N = 2.20 Å; ∠N−H---N = 148.0°] hydrogen bonds, Figure 2d. Ligand Conformation. In the case of bpba ligands with the flexible -(CH2) n- skeleton, the ligand conformation is determined by the C−C−C−C torsion angle as well as the orientations of CO and N−H groups:26 (a) The A and G conformations are given when the C−C−C−C torsion angle (θ) is 180 ≥ θ > 90° and 0 ≤ θ ≤ 90°, respectively, where A stands for “anti” while G for “gauche” conformation. (b) If orientation of CO or N−H group is on the same direction, the conformation is said to be “cis”, and if it is on the opposite direction, arrangement is assigned as “trans”. (c) Because of the differences in the orientations of the pyridyl nitrogen atom and

Figure 1. (a) A representative drawing showing coordination environment about the Hg(II) ion for 1−3. Symmetry transformations used to generate equivalent atoms: (A) x, −y + 3/2, z − 1/2. (b) 1D zigzag chain showing intrachain N−H---O hydrogen bonds. (c) A drawing showing the Hg---X (bright green) interaction, N−H---X (turquoise) hydrogen bonds and N−H---O (red) hydrogen bonds. (d) A drawing showing the C−H---π interactions.

Table 3. N−H---O, N−H---X, and Hg---X Bond Distances (Å) and Angles (deg) for Complexes 1−3

1 2 3

N−H---O

∠N−H---O

N---O

N−H---X

∠N−H---X

N---X

Hg---X

2.05 2.04 2.04

166.9 168.5 172.4

2.8984(4) 2.892(5) 2.90(1)

2.62 2.78 3.03

175.8 175.6 171.2

3.4751(5) 3.634(4) 3.878(9)

3.4637(5), 3.4224(5) 3.7051(7) 4.1134(10)

4804

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809

Crystal Growth & Design

Article

measuring their powder X-ray diffraction patterns. Figures S5− S9 show that the powder patterns of these five complexes match quite well with those simulated from single-crystal X-ray data, indicating the bulk purities of these complexes. Isostructural crystals of zigzag 1−3 without solvent association were synthesized by using solvothermal process. Attempts to prepare crystals through layer diffusion at room temperature afforded only HgI2 helical coordination polymers 4 and 5, while layering reactions of HgCl2 and HgBr2 with L resulted in 1 and 2, respectively, as prepared by using the solvothermal reactions. Getting two different types of structures individually from two processes in the case of HgI2 attracts us to investigate the structural transformation. To investigate the structural transformation, crystals of 3 were first kept separately in variable solvents (relative polarity range 0.3−1.0) such as water, methanol, 1-propanol, acetonitrile, and methylene chloride. In methanol and 1-propanol, crystals of 3 dissolved, while in other solvents, the crystals remained. Moreover, only methanol and acetonitrile were able to alter crystal structures which were structurally characterized as 4 and 5, while water, 1-propanol, and methylene chloride have no effect on 3 as verified by the powder X-ray patterns, Figure S10. These experiments enlightened and inspired us to repeat the processes for 1 and 2, but structures are stable especially toward methanol and acetonitrile along with other solvents, Figure S11 and Figure S12. The reversible structural transformations observed in 3−5 are depicted in Figure 3, vide infra. Reversible Transformations by Solvent Desorption/ Adsorption. The thermal gravimetric analyses (TGA) of 4 and 5 were investigated to determine temperature ranges for the solvent removal. The TGA curve of 4, Figure S13, shows a total weight loss of 4.0% in 90−140 °C, presumably due to the removal of the cocrystallized methanol solvent (calcd 3.8%), while that of 5, Figure S14, displays a total weight loss of 5.0% in 90−132 °C, corresponding to the removal of the cocrystallized CH3CN solvent (calcd 4.8%). Accordingly, complexes 4 and 5 were heated at 150 °C to confirm the solvent removal in the following desolvation reactions. Crystals of 3 were kept in methanol, which dissolved into solvent, and after 2 weeks the solvent slowly evaporated on exposure in air to generate crystals of 4 which were confirmed by the powder XRD pattern, Figure 4, and unit cell of the crystal. Crystals of 4 were then heated at 150 °C for 2 h to remove crystallized methanol solvents to afford the original 3. These efforts justified reversible structural transformation between 3 and 4. This process was repeated by starting with 4 in the reversed process to verify the reversibility of 4 individually where the matched powder XRD patterns with 3 conform the reversibility of 3 and 4, Figure 5. Similarly, the powder pattern of 5 after being heated at 150 °C for 2 h matched that of 3, which can reabsorb acetonitrile solvent to retain its structure, and corroborated the reversible crystal to crystal transformation between 3 and 5, Figure 6. Reversed process was also performed starting with 3, which on solvation with acetonitrile, afforded 5 and on desolvation recovered 3, established the reversible transformation, Figure 7. Reversible Transformation by Solvent Exchange. Unlike previous transformation where the solvation and desolvation were performed, the structural transformations between 4 and 5 can be carried out by solvent exchange. On the other hand, the transformation between 4 and 5 can also be proceeded by desolvation followed by the solvation with

Figure 2. (a) A representative drawing showing the coordination environment about the Hg(II) ion for 4 and 5. Symmetry transformations used to generate equivalent atoms: (A) x + 1, y, z for 4 and x − 1, y, z for 5. (b) 1D Helical chain for 4 and 5. (c) A drawing showing the interchain hydrogen bonding in 4; red dashed lines represent N−H---O (from methanol) and O−H---OC hydrogen bonds and green dashed lines represent N−H---OC hydrogen bonds. (d) A drawing showing interchain hydrogen bonding in 5; red dashed lines represent N−H---OC hydrogen bonds and blue dashed lines represent N−H---N−C hydrogen bonds.

amide oxygen atom, three more conformations anti-anti, synanti and syn-syn are also given. In the L ligand, the flexible -(CH2)n- skeleton has been replaced by the more rigid 1,2phenyl-bis(methylene) group (−CH2−Ph−CH2−). Only rules (b) and (c) are used to define the conformation. Accordingly, the conformations of L are cis syn-anti in 1−3 and trans anti-anti in 4 and 5, respectively. Table S2 lists the ligand conformations and corresponding angles of complexes 1−5. Structural Transformation. Complexes 1−5 provide a unique opportunity to investigate the structural transformations due to solvent adsorption/desorption and exchange in the Hg(II) halide coordination polymer containing L. To confirm the transformations in 1−5, we first checked their purities by 4805

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809

Crystal Growth & Design

Article

Figure 3. A scheme showing the reversible structural transformations in 3−5. Red bar represents the carbonyl group and the blue bar represents the N−H group of L. Red dashed line represents N−H---OC hydrogen bond, and the blue dashed line denotes O−H---OC and N−H---O (from methanol) hydrogen bonds in 4 and N−H---N hydrogen bond in 5.

Figure 5. Powder XRD patterns showing structural transformation between 3 and 4, starting from 4. (a) Simulation of 4, (b) as synthesized, (c) crystals of 4 heated at 150 °C, (d) simulation of 3, and (e) adsorption of MeOH by 4 after heated at 150 °C.

Figure 4. Powder XRD patterns showing structural transformation between 3 and 4, starting from 3. (a) Simulation of 3, (b) as synthesized, (c) crystals of 3 in MeOH, (d) simulation of 4, and (e) 3 heated at 150 °C after adsorption of MeOH.

file,37 which explores the intermolecular interactions in the crystal structure based on the electron densities of the spherical atoms.38 Color codes are used to predict the strength of the interactions by mapping the dnorm functions onto the Hirshfeld surface, where the red regions represent the shorter contacts and negative dnorm values and blue regions represent longer contacts and positive dnorm values, while the white regions indicate the contacts close in length to the van der Waals limit.38,39 The 3D dnorm surfaces can be resolved into 2D fingerprint plots which show the contributions of the intermolecular interactions. This tool is efficient to confirm the hydrogen bonding and atom−atom interaction around the Hirshfeld surface within a given compound as well as to compare interaction in isomorphous crystals.40−42

appropriate solvents. In these occasions, crystals of 4 and 5 were kept in acetonitrile and methanol, respectively, without dissolution for 3 weeks, which were then verified by the powder X-ray patterns. Figures 8 and 9 confirm the reversible crystal-tocrystal transformation due to solvent exchange between 4 and 5. On the basis of the reversible transformations observed for 3, 4, and 5, these complexes can be regarded as a unique example that the solvents show a significant effect on folding and unfolding of the HgI2 single-stranded helical coordination polymers. Hirshfeld Surface Analysis. Hirshfeld surface analysis was carried out on CrystalExplorer 3.1 by importing a crystal CIF 4806

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809

Crystal Growth & Design

Article

Figure 6. Powder XRD patterns showing structural transformation between 3 and 5, starting from 5. (a) Simulation of 5, (b) as synthesized, (c) crystals of 5 heated at 150 °C, (d) simulation of 3, and (e) adsorption of CH3CN by 5 after being heated at 150 °C.

Figure 9. Powder XRD patterns showing structural transformation between 4 and 5, starting from 4. (a) Simulation of 4, (b) as synthesized, (c) crystals of 4 in MeCN, and (d) simulation of 5.

complexes is the distance of closest and effective interaction in di and de for the N−H---X and Hg---X interactions. As shown in Figure 10, we notice the de and di distances mapped for N−H--X and Hg---X interactions in 3 are different from those in 1 and 2, and the Hg---X distance of 3 is much longer than 4 Å (de + di). These much weaker N−H---I and Hg---I interactions can be one of the factors that assists 3 for structural transformation. Roles of Halide Anion and Noncovalent Interactions. A notable question is why the reversible structural transformation was observed for the HgI2 complexes, but not for the HgCl2 and HgBr2 ones. For complexes 1−3, the forces involved in maximizing the Hg−N bond strengths are of the same magnitude; i.e., the Hg−N distances in complexes 1−3 are about the same. However, the fact that the X−Hg−X angle of 3 is significantly smaller than those of 1 and 2 suggests that the identity of the halide atom is one of the main structuredetermining factors. As shown in Table 3 and verified by the Hirshfeld surface analysis, the low electronegativity and large size of iodide anion inevitably affect the N−H---X and Hg---X distances. The weaker N−H---I hydrogen bond and absence of the Hg---I interaction in 3 facilitate the amide carbonyl group to rotate freely to change the ligand configuration through the folding/unfolding process (adjacent Hg−Hg distance from 9.98 to 14.27 Å) and convert reversibly intrachain N−H---O to interchain N−H---O hydrogen bonds during the structural transformation in the presence of effective methanol and acetonitrile solvents, Figure 11. The methanol and acetonitrile solvents that mediate the structural transformation interact with the amide groups of the L ligands through the N−H---O and O−H---O hydrogen bonds in 4 and N−H---N hydrogen bonds in 5, respectively. The effect of the halogen atoms on the structural diversity has been reported. It has been shown that the structures of organic crystals of chloro and bromo compounds in a halogen series generally show similar packing, whereas the iodo derivatives are usually different. The chloride and bromide anions which play the same structural effect can most probably be ascribed to their similar polarizations.43,44 In the series of Zn(II) complexes with N,N′-di(4-pyridyl)adipoamide ligands, the chloro and bromo complexes form double-stranded helical chains, whereas the iodo one shows sinusoidal chains, suggesting that the structure-determining factors are the size and electronegativity of the halide anion.45

Figure 7. Powder XRD patterns showing structural transformation between 3 and 5, starting from 3. (a) Simulation of 3, (b) as synthesized, (c) crystals of 3 in CH3CN, (d) simulation of 5, and (e) 3 heated at 150 °C after adsorption of CH3CN.

Figure 8. Powder XRD patterns showing structural transformation between 4 and 5, starting from 5. (a) Simulation of 5, (b) as synthesized, (c) crystals of 5 in MeOH, and (d) simulation of 4..

Exploration of full fingerprint plots of 1 −3 by using Hirshfeld surface analysis shows similar patterns for strong interaction such as N−H---O, N−H---X, and H---π interactions, Figure S15. However, the difference noticed among these 4807

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809

Crystal Growth & Design

Article

Figure 10. Full fingerprints of (a) 1, (b) 2, and (c) 3 and fingerprint plots of (d) Hg---Cl, (e) Hg---Br, and (f) Hg---I contacts.



CONCLUSIONS In summary, five Hg(II) CPs with the new L ligand have been synthesized and structural characterized, in which the HgI2 complexes 3−5 show reversible structural transformations induced by the solvent adsorption/desorption and exchange. Although 1−3 are isostructural, only the iodide-containing 3 is able to transform into varied structures. Methanol and acetonitrile solvents assist reversible structural transformation due to their ability in exchanging intrachain N−H---O hydrogen bonds in 3 and interchain N−H---O hydrogen bonds in 4 and 5. The larger size and smaller electronegativity of the iodide anion that result in weaker N−H---I hydrogen bond and absence of the Hg---I interaction facilitate the rotation of the amide carbonyl group and lead to the structural transformation, while the stronger N−H---X (X = Cl and Br) and Hg---X interactions in 1 and 2 limit such rotation and thus make the transformation inaccessible.

Figure 11. Change in ligand conformation from cis syn-anti to trans anti-anti during the structural transformation.

Emission Properties. Figure S16 shows emission spectra of L in the solution and solid states. Accordingly, the L ligand exhibits emission at 318 nm upon excitations at 295 nm in the solid state, while emissions at 398, 359, 360, and 358 nm upon excitations at 325, 305, 310, and 308 nm were observed in MeOH, EtOH, DMF, and DMSO, respectively. The emission spectra of L are solvent dependent, which suggests that a polar excited state capable of forming exciplex with different solvents is involved in the emission.46 However, no detectable levels of emission intensity can be found for 1−5, which may be ascribed to the fast nonradiative decay of the absorbed light by the complex in the solid state.47



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.cgd.7b00742. NMR and IR spectra (Figure S1 and Figure S2). Powder X-ray patterns (Figures S3−S12). TGA curves (Figure S13−S14). Fingerprint plots (Figure S15). Normalized emission spectra (Figure S16). C−H---π distances (Table 4808

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809

Crystal Growth & Design

Article

(22) Mahmoudi, G.; Bauza, A.; Gurbanov, A. V.; Zubkov, F. I.; Maniukiewicz, W.; Rodriguez-Dieguez, A.; Lopez-Torres, E.; Frontera, A. CrystEngComm 2016, 18, 9056−9066. (23) Wang, H.; Wang, P.; Huang, C.; Chang, L.; Wu, J.; Hou, H.; Fan, Y. Inorg. Chim. Acta 2011, 378, 326−332. (24) Hsu, W.; Li, Y.-S.; He, H.-Y.; Chen, K.-T.; Wu, H.-S.; Proserpio, D. M.; Chen, J.-D.; Wang, J.-C. CrystEngComm 2014, 16, 7385−7388. (25) Hsu, W.; Yang, X.-K.; Chhetri, P. M.; Chen, J.-D. Polymers 2016, 8, 137. (26) Thapa, K. B.; Chen, J.-D. CrystEngComm 2015, 17, 4611−4626. (27) APEX2, V2008.6; SAD ABS V2008/1; SAINT+ V7.60A; SHELXTL V6.14; Bruker AXS Inc.: Madison, Wisconsin, USA, 2008. (28) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112. (29) Yang, L.; Powell, D. R.; Houser, R. P. Dalton Trans. 2007, 955− 964. (30) Okuniewski, A.; Rosiak, D.; Chojnacki, J.; Becker, B. Polyhedron 2015, 90, 47−57. (31) Persson, I.; Sandström, M.; Goggin, P. L.; Mosset, A. J. Chem. Soc., Dalton Trans. 1985, 1597−1604. (32) Malone, J. F.; Murray, C. M.; Charlton, M. H.; Docherty, R.; Lavery, A. J. J. Chem. Soc., Faraday Trans. 1997, 93, 3429−3436. (33) Bondi, A. J. Phys. Chem. 1964, 68, 441−451. (34) Batsanov, S. S. Inorg. Mater. 2001, 37, 871−885. (35) Lommerse, J. P. M.; Stone, A. J.; Taylor, R.; Allen, F. H. J. Am. Chem. Soc. 1996, 118, 3108−3116. (36) Brammer, L.; Bruton, E. A.; Sherwood, P. Cryst. Growth Des. 2001, 1, 277−290. (37) Wolff, S. K; Grimwood, D. J.; McKinnon, J. J.; Turner, M. J.; Jayatilaka, D.; Spackman, M. A. University of Western Australia, 2012. (38) McKinnon, J. J.; Jayatilaka, D.; Spackman, M. A. Chem. Commun. 2007, 3814−1816. (39) Spackman, M. A.; Jayatilaka, D. CrystEngComm 2009, 11, 19− 32. (40) Martin, A. D.; Britton, J.; Easun, T. L.; Blake, A. J.; Lewis, W.; Schroder, M. Cryst. Growth Des. 2015, 15, 1697−1706. (41) Guzman-Percastegui, E.; Alvarado-Rodríguez, J. G.; CruzBorbolla, J.; Andrade-Lopez, N.; Vazquez-García, R. A.; NavaGalindo, R. N.; Pandiyan, T. Cryst. Growth Des. 2014, 14, 3742−3757. (42) Mahmoudi, G.; Zangrando, E.; Bauza, A.; Maniukiewicz, W.; Carballo, R.; Gurbanov, A. V.; Frontera, A. CrystEngComm 2017, 19, 3322−3330. (43) Dey, A.; Jetti, R. K. R.; Boese, R.; Desiraju, G. R. CrystEngComm 2003, 5, 248−252. (44) Saha, B. K.; Jetti, R. K. R.; Reddy, L. S.; Aitipamula, S.; Nangia, A. Cryst. Growth Des. 2005, 5, 887−899. (45) Hsu, Y.-F.; Hsu, W.; Wu, C.-J.; Cheng, P.-C.; Yeh, C.-W.; Chang, W.-J.; Chen, J.-D.; Wang, J.-C. CrystEngComm 2010, 12, 702− 710. (46) Singh, D.; Ghosh, S. K.; Baruah, J. B. J. Heterocycl. Chem. 2010, 47, 199−206. (47) Turro, N. J. Modern Molecular Photochemistry; University Science Books, CA, 1991.

S1). Corresponding angles and conformations (Table S2) (PDF) Accession Codes

CCDC 1551073−1551077 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Jhy-Der Chen: 0000-0003-2785-6741 Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We are grateful to the Ministry of Science and Technology of the Republic of China for support. REFERENCES

(1) Tiekink, E. R. T.; Vittal, J. J. Frontiers in Crystal Engineering; John Wiley & Sons, Ltd.: England, 2006. (2) Batten, S. R.; Neville, S. M.; Turner, D. R. Coordination Polymers: Design, Analysis and Application; Royal Society of Chemistry: Cambridge, 2009. (3) Li, J.-R.; Sculley, J.; Zhou, H.-C. Chem. Rev. 2012, 112, 869−932. (4) Constable, E. C. Chem. Soc. Rev. 2013, 42, 1637−1651. (5) Ke, S.-Y.; Wang, C.-C. CrystEngComm 2015, 17, 8776−8785. (6) Zeng, M.-H.; Hu, S.; Chen, Q.; Xie, G.; Shuai, Q.; Gao, S.-L.; Tang, L.-Y. Inorg. Chem. 2009, 48, 7070−7079. (7) Tripathi, S.; Srirambalaji, R.; Patra, S.; Anantharaman, G. CrystEngComm 2015, 17, 8876−8887. (8) Vittal, J. J. Coord. Chem. Rev. 2007, 251, 1781−1795. (9) Kole, G. K.; Vittal, J. J. Chem. Soc. Rev. 2013, 42, 1755−1775. (10) Mahmoudi, G.; Morsali, A. Cryst. Growth Des. 2008, 8, 391− 394. (11) Mobin, S. M.; Srivastava, A. K.; Mathur, P.; Lahiri, G. K. Dalton Trans. 2010, 39, 8698−8705. (12) Hou, Y.; Masoomi, M. Y.; Bagheri, M.; Morsali, A.; Joo, S. W. RSC Adv. 2015, 5, 81356−81361. (13) Thapa, K. B.; Hsu, Y.-F.; Lin, H.-C.; Chen, J.-D. CrystEngComm 2015, 17, 7574−7582. (14) Morsali, A.; Masoomi, M. Y. Coord. Chem. Rev. 2009, 253, 1882−1905. (15) Mahmoudi, G.; Khandar, A. A.; Zareba, J. K.; Bialek, M. J.; Gargari, M. S.; Abedi, M.; Barandika, G.; Van Derveer, D.; Mague, J.; Masoumi, A. Inorg. Chim. Acta 2015, 429, 1−14. (16) Ma, T.; Wang, Y.; Wang, F.; Li, F. Acta Crystallogr., Sect. E: Struct. Rep. Online 2012, 68, m367. (17) Masoumi, A.; Servati Gargari, M. S.; Mahmoudi, g.; Miroslaw, B.; Therrien, B.; Abedi, M.; Hazendonk, P. J. Mol. Struct. 2015, 1088, 64−69. (18) Pansanel, J.; Jouaiti, A.; Ferlay, S.; Hosseini, M. W.; Planeix, J.M.; Kyritsakas, N. New J. Chem. 2006, 30, 71−76. (19) Mahmoudi, G.; Khandar, A. A.; White, J.; Mitoraj, M. P.; Jena, H. S.; Der Vooort, P. V.; Qureshi, N.; Kirillov, A. M.; Robeyns, K.; Safin, D. A. CrystEngComm 2017, 19, 3017−3025. (20) Yuan, Z.-Z.; Luo, F.; Song, Y.-M.; Sun, G.-M.; Tian, X.-Z.; Huang, H.-Z.; Zhu, Y.; Feng, X.-F.; Luo, M.-B.; Liu, S.-J.; Xu, W.-Y. Dalton Trans. 2012, 41, 12670−12673. (21) Rana, L. K.; Sharma, S.; Hundal, G. Cryst. Growth Des. 2016, 16, 92−107. 4809

DOI: 10.1021/acs.cgd.7b00742 Cryst. Growth Des. 2017, 17, 4801−4809