An inorganic starch–iodine model: the inorganic–organic hybrid compound {(C₄H₁₂N₂)₂[Cu^II₄](I₂)}_n

Abstract

The dark-blue crystal color of {(C₄H₁₂N₂)₂[Cu^II₄](I₂)}_n, its mixture of I⁻, I₃⁻ and linear I₄²⁻ or linear I₅⁻ polyiodide species in a linear channel arrangement, its channel diameter of ∼5.5 Å and the helical arrangement of the hydrogen bonded {(C₄H₁₂N₂)₂[Cu^II₄]}⁺ supramolecular host around the channels agree with the description of the classical, yet structurally elusive, starch–iodine compound.

---

Polyiodides, I_n^m− have an extensive structural chemistry with a variety of cations.^1–5 The well-known, yet structurally elusive, colloidal starch–iodine complex,⁶ discovered by Colin and de Claubry in 1814,⁷ is believed to consist of polyiodides in a supramolecular host–guest compound.¹Starch which forms a very dark blue-black complex with iodine is a very sensitive test for iodine (in the likely presence of iodide).⁸Amylose from starch is responsible for this blue polyiodide complex. In this complex amylose forms a probably left-handed helical structure of 8 Å pitch with 6 glucose units per turn. The ∼5 Å wide central cavity of this helix may accommodate polyiodides which may consist of I₅⁻, I₃⁻ and I₂ units modulated as (I₂⋯I₃⁻)_n or (I₂⋯I⁻⋯I₂)_n.⁶ The average I⋯I distances of 3.1 Å between these units are shorter than van der Waals contacts and, therefore, charge delocalization can occur along the chain which explains the intense chromophore absorbing around 600 nm (Fig. S11, ESI†).^1,6,14

Here we present the dark-blue inorganic–organic hybrid compound {(C₄H₁₂N₂)₂[Cu^II₄](I₂)}_n (1) (Fig. 1) as a model for the structurally unknown starch–iodine complex. Compound 1 formed under hydrothermal conditions from comproportionation of Cu^IIO and Cu⁰ in a mixture of HI (57 wt%), I₂ and piperazine, C₄H₁₀N₂.† The crystal structure of 1 contains isolated tetraiodocuprate(I) anions, [Cu^II₄]³⁻, piperazinium dications, (C₄H₁₂N₂)²⁺ and what appears to be chains of equidistant iodine atoms along (1/4, 1/4, z and 3/4, 3/4, z) (Fig. 2).‡

Fig. 1 (a) Dark-blue crystal of compound 1. (b) Crystal degradation in water or alcohols through leaching of iodine, I₂ into solution. (c) Amorphous (from XRPD) residue before complete dissolution.

Fig. 2 Unit cell packing of 1. Selected distances (Å) and angles (°): Cu–I1 2.677(2), I2–I2⁶ 2.755(5), I2–I2^5′ 0.631(5), I1–Cu–I1³ 107.22(6), I1–Cu–I1⁵ 110.61(3), I2^5′–I2–I2⁶ 180; symmetry transformation: 3 = 1.5 −x, 0.5 −y, z; 5 = x, 0.5 −y, 0.5 −z; 5′ = x, 0.5 −y, 1.5 −z; 6 = y, x, 2 −z; 7 = 1.5 −x, y, 0.5 −z.

The shape of the electron density around the ideal position 1/4, 1/4, 1/4 (intersection of two 2₁ axes) of the I2 atoms in the channels of 1 (Fig. 2) suggests a statistical disorder within the iodine chain, thus yielding only an average distance of 3.39 Å. From the subsequent split position refinement of the I2 site probable I⋯I distances of 2.76, 3.39 and 4.02 Å can be deduced and also intermediate values of 3.07 and 3.70 Å if the center of the split position is considered as well (Fig. 3).

Fig. 3 Difference Fourier synthesis of the average structure calculated without the I2 atom (isolines with 5 e Å⁻³; dark points: position of the I2 atoms in the final split atom refinement).

However, the diffraction images of the studied crystal with the partially ordered iodine structure showed eight additional maxima in the vicinity of each Bragg peak that could be indexed by the introduction of a q-vector of 0.2, 0.2, 0.38. Based on the fact that a modulation vector with three components is required, not only an incommensurate ordering of the polyiodide species in one channel but in addition a correlation between the occupation of neighboring channels can be assumed.

From the charge balance in the sum formula of 1, {(C₄H₁₂N₂²⁺)₂[Cu^II₄³⁻](I₂⁻)}_n two iodine atoms share a single negative charge, (I₂⁻). The inter-iodine distance of 3.39 Å in the channels of 1 is indicated in the X-ray patterns by diffuse lines suggesting statistical disorder within the iodide chain, thus, yielding only an average distance (Fig. 2 and 3).‡ With respect to the tetrahedral tetraiodocuprate(I) moiety in 1 we note that we are not aware of any other structurally authenticated halocuprate with isolated metal tetrahedra.⁹ Edge- and corner-sharing polyhedra prevail in halometallates.¹⁰ In particular no isolated [CuI₄]³⁻ is known.¹¹

Compound 1 is shown by X-ray powder diffraction to be a pure phase and by ESR to contain only Cu(I) (Fig. S5 and S6 in ESI†). One of the best analytical methods to study polyiodide species is Raman-Fourier transform (FT) spectroscopy. The Raman-FT spectrum of 1 displays strong maxima at 109 and 168 cm⁻¹ (Table 1, Fig. 4) due to the symmetric stretching vibrations of the linear and symmetric I₃⁻ and I₄²⁻ or I₅⁻. The maxima at 86 and 140 are due to [CuI₄]³⁻,¹² with some possible contribution to the peak at 140 from slightly asymmetric I₃⁻. From the Raman study free iodine, I₂ or I₂ only loosely coordinated to iodide does not seem to be present (Table 1). Heptaiodide, I₇⁻ and higher polyiodides are excluded because they are not known in a linear arrangement. Thus, the iodide chains (I₂⁻)_n in 1 are described as incommensurate polyiodide chains of linear I₄²⁻ or I₅⁻, I₃⁻ and I⁻ units. While pentaiodide is better known in its more common bent forms,¹ examples of linear I₅⁻ have been reported (Table 1).

Fig. 4 Raman-FT spectrum of 1 with assignment.

Table 1 Raman bands and d(I–I) of linear polyiodides I₃⁻, I₄²⁻ and I₅⁻

Compound (reference)	d (I–I–I)/Å	ν/cm^−1	Compound (reference)	d (I–I)/Å	ν/cm^−1
a br = broad, s = strong, m = medium, w = weak. b mo2ttl = 3,5-bis(N-morpholinio)-1,2,3-trithiolate. c [15]aneO5 or [15]aneS5 = 1,4,7,10,13-pentaoxa or pentathiocyclopentadecane. d Bis(dimethylthio)tetrathia-fulvalene (DMT-TTF) bisannulated tetraoxatetrathiatetracosane. e [15]aneN4 = 1,4,8,12-tetraazacyclopentadecane. f TMA = trimesic acid. g [18]aneN2S4 = 1,4,10,13-tetrathia-7,16-diazacyclooctadecane.
Free I2^18	2.670	180s	[I⋯I–I⋯I]^2− = I4^2− (symm.) in [(C5H7N2Se–)2](I3)(1/2 I4)^20	2.819, 3.405	155s
Weakly bound I2 in [mo2ttl]2I16,^b^,^19 [Ag2([15]aneS5)2]I12,^c^,^3 [K([15]aneO5)2]I9^c^,^27	2.741, 2.755, 2.716	174s, 172brs, 180brs
[I–I–I]^− = I3^− (slightly asymm.) in [n-Bu4N](I3)^21	2.890, 2.950	108s	[I⋯I–I⋯I]^2− = I4^2− (symm.) in [Cd(NH3)4I2·I2]^22	2.793, 3.386	Not det.
[I–I–I]^− = I3^− (symm.) in [bis(DMT-TTF)](I3)(I5)^d^,^23	2.929	110s, (324w)	[I–I⋯I⋯I–I]^− = I5^− (symm.) in [bis(DMT-TTF)](I3)(I5)^d^,^23	2.785, 3.191	163m, (304 w)
I3^− in (1,1′-(propane-1,3-diyl) ferrocenium)3(I3)2(I5)^24	2.889, 2.967	Not det.	I5^− in (1,1′-(propane-1,3-diyl) ferrocenium)3(I3)2(I5)^22	2.800, 3.220	Not det.
[I–I–I]^− = I3^− (slightly asymm.) in Pd2Cl2([15]aneN4)](I3)2^e^,^25	2.871, 2.953	107s, 135w	[I–I⋯I⋯I–I]^− = I5^− (symm.) in (TMA·H2O)10 H^+I5^− ^f^,^26	2.740, 3.260	(75vw)^21, 163s
[I–I–I]^− = I3^− (slightly asymm.) in [Pd2Cl2([18]aneN2S4)]1.5I5(I3)2^g^,^27	2.904–2.929, 2.948–2.959	108s, 138w

---

Charged-balanced sum formulae of 1, {(C₄H₁₂N₂)₂[Cu^II₄](I₂)}_n can be described with I₄²⁻ or with mixtures of linear I₅⁻, I₃⁻ and I⁻:

I₄²⁻: {(C₄H₁₂N₂)₄[Cu^II₄]₂(I₄²⁻)}_n (2 ×1);

I⁻ and I₃⁻: {(C₄H₁₂N₂²⁺)₄[Cu^II₄³⁻]₂(I⁻)(I₃⁻)}_n (2 ×1);

I⁻, I₃⁻ and I₅⁻: {(C₄H₁₂N₂²⁺)₁₂[Cu^II₄³⁻]₆(I⁻)₄(I₃⁻)(I₅⁻)}_n (6 ×1).

Resonance Raman spectra of the starch–iodine complex display four characteristic maxima¹³ at 24, 52, 112 and 164 cm⁻¹ which have been assigned to stretching and bending vibrations of the polyiodide chain. The polyiodide species in the amylose–iodine complex is certainly not simply I₃⁻. It is more likely that the species in the starch–iodine complex are (I₂⋯I₃⁻)_n or (I₅⁻)_n.¹⁴

In (Me₄Sb)₃I₈ a superficially similar, linear ordered iodine atom chain structure has been proposed with slightly alternating I⋯I distances of 3.283(2) and 3.299(2) Å (cf. 3.39 Å in 1).¹⁵ From the charge balance and from the thermal ellipsoids presented in (Me₄Sb)₃I₈, the (I₈³⁻)_n chains are likely to be incommensurate chains of polyiodide species as in 1.

Even today, the exact structure of the starch– or amylose–iodine complex is not really known.⁶ An X-ray powder pattern of the amylose–iodine complex can be indexed using a hexagonal cell (γ = 120°) with a = b = 12.97, c = 7.91 Å (cf.1, tetragonal a = b = 12.98, c = 6.78 Å).¹⁶ As a model to the elusive starch–iodine complex crystalline Na⁺, Ba²⁺ and Cd²⁺ salts of α-cyclodextrin–iodide structures have been studied.¹⁷ The cyclodextrin for the structural model studies consists of α(1–4)-linked six-membered cyclic glucose rings. Yet, they lack the helical arrangement of amylose.

In 1 the N–H⋯I hydrogen bonding from the piperazinium dications to the tetraiodocuprate anion leads to helical arrangements of two interdigitated fourfold helices around each polyiodide chain. Adjacent polyiodide chains are surrounded by alternating right-handed (P) 4₁ or left-handed (M) 4₃ helices, which wrap around the channels in the c direction (Fig. 5). These helices enclose a channel diameter of ∼5.5 Å (inter-atomic distance) in 1, to be compared with the channel diameter of ∼5.2 Å in [Cd²⁺–α-cyclodextrin–polyiodide],¹⁴ which is regarded as a good model of the starch–/amylose–iodine complex.

Fig. 5 Helical arrangement in 1 shown in [100] and [001]. Hydrogen bonds as dashed lines (N)H1⋯I1 2.61, N⋯I1 3.51(2) Å, N–H1⋯I1 173.7°. The polyiodide chains are depicted as yellow columns.

When crystals of 1 are immersed in water or other polar solvents (methanol, ethanol) in the presence of air, yellow-brown iodine, I₂, leaches from them to give a near-colorless amorphous solid which eventually also dissolves (Fig. 1). The deep blue color of starch–iodine also disappears when air is bubbled through its colloidal solution under light due to the oxidation of I⁻ to I₂.⁸

The dark-blue crystal color of 1, its polyiodide species in a linear channel arrangement, the channel diameter, and the helical arrangement of the supramolecular host around the channels suggest that the description of the inorganic–organic hybrid compound {(C₄H₁₂N₂)₂[Cu^II₄](I₂)}_n (1) as the first inorganic–organic hybrid model for the classical starch–iodine compound is justified.

We thank M. Welschehold, S. Steimle, D. Gnandt, T. Engesser, M. Wendorff and Dr H. Müller-Sigmund for their help and DFG grant Ja466/41-1.

Notes and references

1. P. H. Svensson and L. Kloo, Chem. Rev., 2003, 103, 1649; A. J. Blake, F. A. Devillanova, R. O. Gould, W.-S. Li, C. Radek and M. Schröder, Chem. Soc. Rev., 1998, 3, 195; P. Coppens, in Extended Linear Chain Compounds, ed. J. S. Miller, Plenum Press, New York, 1982, pp. 333–356.
2. I. Pantenburg and I. Müller, Z. Anorg. Allg. Chem., 2004, 630, 1637; I. Pantenburg and K.-F. Tebbe, Z. Anorg. Allg. Chem., 2002, 628, 1780; K.-F. Tebbe and I. Dombrowski, Z. Anorg. Allg. Chem., 1999, 625, 167; R. Richter, O. Seidelmann and L. Beyer, Z. Anorg. Allg. Chem., 1999, 625, 511; S. A. El-Khalik, M. El Essawi, I. Dombrowski and K.-F. Tebbe, Z. Naturforsch., B: Chem. Sci., 1999, 54, 136; K.-F. Tebbe and R. Loukili, Z. Anorg. Allg. Chem., 1998, 624, 1175; A. J. Blake, V. Lippolis, S. Parsons and M. Schröder, Chem. Commun., 1996, 2207.
3. A. J. Blake, R. O. Gould, W.-S. Li, V. Lippolis, S. Parsons, C. Radek and M. Schröder, Inorg. Chem., 1998, 37, 5070.
4. F. Cramer, U. Bergmann, P. C. Manor, M. Noltemeyer and W. Saenger, Justus Liebigs Ann. Chem., 1976, 1169; L. Mohanambe and S. Vasudevan, Inorg. Chem., 2004, 43, 6421.
5. W. Bi, N. Louvain, N. Mercier, J. Luc and B. Sahraoui, CrystEngComm, 2007, 9, 298; C. A. Ilioudis and J. W. Steed, CrystEngComm, 2004, 6, 239; G. J. Reiß and J. S. Engel, CrystEngComm, 2002, 4, 155.
6. M. H. Baghal-Vayjooee, A. J. Colussi and S. W. Benson, J. Am. Chem. Soc., 1978, 100, 3215; W. Saenger, Naturwissenschaften, 1984, 71, 31.
7. J. J. Colin and H. G. de Claubry, Ann. Chim. (Cachan, Fr.), 1814, 90, 87.
8. R. J. Meyer, Gmelins Handbuch der Anorganischen Chemie, Verlag Chemie, Berlin, 1933, vol. 8, pp. 197ff, 323ff.
9. C. H. Arnby, S. Jagner and I. Dance, CrystEngComm, 2004, 6, 257; B. Scott, R. Willett, A. Saccani, F. Sandrolini and B. L. Ramakrishna, Inorg. Chim. Acta, 1996, 248, 73; S. Jagner and G. Helgesson, Adv. Inorg. Chem., 1991, 37, 1.
10. C.-L. Chen and A. M. Beatty, Chem. Commun., 2007, 76; N. Mercier, CrystEngComm, 2005, 7, 429; M. C. Aragoni, M. Arca, C. Caltagirone, F. A. Devillanova, F. Demartin, A. Garau, F. Isaia and V. Lippolis, CrystEngComm, 2005, 7, 544; N. Mercier and A. Riou, Chem. Commun., 2004, 844.
11. Cambridge Structure Database search, CSD Version 5.29 (November 2007) with two updates (January and August 2008).
12. D. E. Irish, L. Stolberg and D. W. Shoesmith, Surf. Sci., 1985, 158, 238; J. E. Potts, R. C. Hanson, C. T. Walker and C. Schwab, Solid State Commun., 1973, 13, 389.
13. H. Kim, Biopolymers, 1982, 21, 2083.
14. S. B. Sharp and G. I. Gellene, J. Phys. Chem. A, 1997, 101, 2192; X. Yu, C. Houtman and R. H. Atalla, Carbohydr. Res., 1996, 292, 129.
15. U. Behrens, H. J. Breuning, M. Denker and K. H. Ebert, Angew. Chem., Int. Ed. Engl., 1994, 33, 987.
16. R. E. Rundle and D. French, J. Am. Chem. Soc., 1943, 65, 1707.
17. M. Noltemeyer and W. Saenger, J. Am. Chem. Soc., 1980, 102, 2710.
18. A. Bondi, J. Phys. Chem., 1964, 68, 441.
19. P. Deplano, F. A. Devillanova, J. R. Ferraro, M. L. Mercuri, V. Lippolis and E. F. Trogu, Appl. Spectrosc., 1994, 48, 1236.
20. F. Bigoli, F. Demartin, P. Deplano, F. A. Devillanova, F. Isaia, V. Lippolis, M. L. Mercuri, M. A. Pellinghelli and E. F. Trogu, Inorg. Chem., 1996, 35, 3194.
21. M. Mizuno, J. Tanaka and I. Harada, J. Phys. Chem., 1981, 85, 1789.
22. K.-F. Tebbe and M. Plewa, Z. Anorg. Allg. Chem., 1982, 489, 111–125.
23. T. Akutagawa, Y. Abe, Y. Nezu, T. Nakamura, M. Kataoka, A. Yamanaka, K. Inoue, T. Inabe, C. A. Christensen and J. Becher, Inorg. Chem., 1998, 37, 2330.
24. T. Dong, H. Lin, M. Hwang, T. Lee, L. Tseng, S. Peng and G. Lee, J. Organomet. Chem., 1991, 414, 227.
25. A. J. Blake, W.-S. Li, V. Lippolis, S. Parsons and M. Schröder, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1998, 54, 299.
26. E. M. Nour, L. H. Chen and J. Laane, J. Phys. Chem., 1986, 90, 2841; F. H. Herbstein, M. Kapon and G. M. Reisner, Proc. R. Soc. London, Ser. A, 1981, 376, 301.
27. A. J. Blake, W.-S. Li, V. Lippolis, S. Parsons, C. Radek and M. Schröder, Angew. Chem., Int. Ed., 1998, 37, 293.
28. V. Petříček, M. Dušek and L. Palatinus, JANA2000, The crystallographic computing system, Institute of Physics, Praha, Czech Republic, 2000.

---

Footnotes

† Electronic supplementary information (ESI) available: Synthesis, elemental analyses, IR, full Raman, XRPD, ESR, TG–DTA, EDS–SEM, iodine–starch reaction. CCDC 708974. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/b820151j

‡ X-Ray structure determination: C₈H₂₄N₄CuI₆, 1001.25 g mol⁻¹, tetragonal, P4₂/nnm (134), Z = 2, 203(2) K on a Bruker ApexII CCD, 4.44°≤ 2θ≤ 51.68°, ω-scans, a = b = 12.9597(4), c = 6.7723(3) Å, V = 1137.43(7) ų, D_calc. = 2.923 g cm⁻³, crystal size 0.20 × 0.17 × 0.12 mm, F(000) 894, 3770 reflections measured, 590 independent (R_int = 0.0313), μ(Mo-K_α) (λ = 0.71073 Å), 29 refined parameters, R1 = 0.0631, wR2 = 0.1677 for 489 reflections with I > 2σ(I), R1 = 0.0720, wR2 = 0.1734 for all data, goodness-of-fit 1.111. All non-hydrogen atoms refined anisotropically.Due to the variable disorder in the channels (see structure description above), several crystals of 1 were tested and measured on different diffractometers (STOE-IPDS-II, Bruker APEXII CCD) and at different temperatures. The diffraction patterns of several other crystals of lower quality exhibit additional diffuse streaks along the c* direction and the analysis of their Bragg reflections reveals only a continuous tube of electron density along the 1/4, 1/4, z direction. Nevertheless, this also integrates to exactly two iodine atoms per channel. Only after several trials one crystal could be isolated that does not show diffuse scattering. However, the diffraction images of this crystal showed eight additional intensity maxima in the vicinity of each Bragg peak. These extra peaks could be indexed with the q-vector 0.2, 0.2, 0.38. The solution of the modulated structure in four-dimensional space will be the subject of subsequent studies.Currently, the structure was solved by direct methods (program SHELXS-97) using the Bragg reflections of the small tetragonal unit cell only. All non-hydrogen atom positions, including a single I atom at the Wyckoff site 4c representing the iodine species in the channels, were first refined anisotropically by full-matrix least squares methods on F² (SHELXL-97). Due to the strong anisotropy of the thermal parameters of this position, a detailed inspection of the electron density inside the channel was carried out using the program JANA2000 (see Fig. 3).²⁸ The shape of the electron density suggested the use of a split position for the I2 atom in the final refinement of the average structure. A free refinement of the occupation of the I2 position reveals exactly 2 iodine atoms per channel and thus a mean distance of the I atoms of 3.39 Å, which is in good agreement with the chemical analysis (Cu–I = 1 : 6, see ESI†) and the iodine species detected using Raman spectroscopy (see above).

---