Iodine Mossbauer Investigation of the Starch

(1) (a) This work was supported by the National Science Foundation through .... (47) A. L. Andreassen, D. Zebeiman, and S. H. Bauer. J. Am. Chem. Soc...
14 downloads 0 Views 842KB Size
3322

Journal of the American Chemical Society

suitable crystals could be prepared, but this has not yet been achieved; the various forms of electronic spectroscopy seem to have the best prospects. It seems unlikely that the various types of kinetic experiments which have been devised will ever really draw such a fine distinction unambiguously. Examination of Figure 3 also suggests that the choice of names, “classical” and “nonclassical”, has given the discussion more impetus than its substance would justify.

References and Notes (1) (a) This work was supported by the National Science Foundation through Grants GP-31360X and CHE76-01181 to the University of Minnesota. (b) A preliminary report of this work has appeared as a communication; M. M. Kreevoy, T. M. Liang, and K.-C. Chang, J. Am. Chem. SOC.,99, 5207 (1977). (c) M. M. Kreevoy was the guest of the Physical Chemistry Laboratory, Oxford, while much of this work was in progress. He wishes to thank Drs. W. J. Albery and R . K. Thomas, of that laboratory, and Dr. J. C. Speakman of the University of Glasgow for stimulating and heipfui dlscussions of this work. (2) When A1 and A2 are the same, these substances have been called homoconjugates: when they are different, heteroconjugates. I. M. Kolthoff and M. K. Chantooni, Jr., J. Phys. Chem., 66, 1675 (1962); J. Am. Chem.

SOC.,85, 426 (1963). (3) I. Olovson and P. G. asson in ”The Hydrogen Bond-Recent Developments in Theory and Experiment”, Vol. 11, P. Schuster, G. Zundel, and C. sandorfy, Eds., North-Holland Publishing Co., Amsterdam, 1976, Chapter 8. (4) D. Hadfi and S. Bratos in ref 3, Chapter 12. (5) G. Zundel in ref 3, Chapter 15. (6) J. C. Speakman, Struct. Bonding(Berlin), 12, 141 (1972). (7) (a) V. Goid, Trans. Faraday Soc., 56,255 (1960);(b) A. J. Kresge, h e Appl. Chem., 8, 243 (1964). (6)L is used as a symbol for an atom that may be H, D, or T. The symbol D(L&) indicates that the deuterium which enters the equilibrium is taken from liquid water which contains both isomers, preferably in approximately equal amounts: H(L20) has the analogous significance for hydrogen. (9) J. F. Coetzee, Pfog. Phys. Ofg. Chem., 4, 45 (1967). ( I O ) A. Martinsen and J. Songstad, Acta Chem. Scand., Ser. A, 31, 645 11977\.

(11)

i.M. Kolthoff, M. K. Chantooni, Jr., and S.Bhowmik, J. Am. Chem. Soc.,

88, 5430 (1966).

(12) J. F. Coetzee and G. R. Padmanabhan, J. Phys. Chem., 69, 3193 (1965). (13) i C. Evans, Spectrochim.Acta, 16,994 (1960). (14) Sadtler Standard Spectra, Standard Grating Spectrum No. 23594. (15) T. M. Liang, Ph.D. Thesis, University of Minnesota, 1979, pp 146-148. (16) Reference 15, pp 55-57. (17) Reference 15, p 14. (18) M. M. Kreevoy and T. S . Straub, Anal. Chem.. 41, 214 (1969). (19) S.Nagakura, J. Chem. Phys., 23, 1141 (1955). (20) Homoconjugation constants appear to be similar in suifolane and acetonitrile.21f22

(21) J. F. Coetzee and R. J. Bertogzi, Anal. Chem., 45, 1064 (1973). (22) I. M. Kolthoff. M. K. Chantooni, Jr., and S.Bhowmik, J. Am. Chem. Soc.,

/ 102:lO 1 May 7 , I980

90, 23 (1968). (23) R. L. Benoit, A. L. Beauchamp, and R. Domain, horg. Nucl. Chem. Lett., 7, 557 (1971). (24) The PKHAvalues of benzoic acid, 4-nitrophenol, and acetic acid in acetonitrile are 20.4,2620.9,’l and 22.3,21respectively. (25) Relative acidities of structurally similar substances appear to be the same in sulfolane as they are in acetonitrile.21

(26) i. M. Kolthoff and M. K. Chantooni, Jr., J. Phys. Chem., 70, 856 (1966). (27) b)R. L. Schowen, frog. Phys. Org. Chem., 9, 275 (1972);(b) J. F. MataSegreda. S. Wirt, and R. L. Schowen, J. Am. Chem. Soc., 98, 5608 (1974). (28) V. Gold and C. Tomlinson, J. Chem. SOC.13,1707 (1971). (29) Sadtler Standard Spectra, Standard Grating Spectrum No. 8461. (30) R . Livingston, “Physic0 Chemical Experiments”. Macmiilan, New York, 1957, pp 22-30. (31) Tetraethylammoniumperchlorate has a dissociation constant of 5.6 X M in acetonitrile; J. F. Coetzee and G. P. Cunningham, J. Am. Chem. Soc., 87. 2534 11965). --(32) R. G. Jones a n d i R. Dyer, J. Am. ~hem.S O ~ . 95,2485 , (1973). (33) (a) J. L. Wood, J. Mol. Shuct., 13, 141 (1972); (b) R. L. Dean and J. L. Wood, ibid.. 26. 197 11975). (34) J. Laane; Appi. Spectrosc., 24, 73 (1970). (35) J. L. Wood in “Spectroscopy and Structure of Molecular Complexes”, J. Yarwood, Ed., Plenum Press, New York. 1973, pp 336-340. This artlcie I

also contains references to other papers on this subject.

(36) D. G. Truhlar, J. Comput. Phys., 10, 123 (1972). (37) M. M. Kreevoy, ”Isotopes In Organic Chemistry”, Vol. 2, E. Buncel and C. C. Lee, Eds., Eisevier, Amsterdam, 1976, p 16; however, the harmonic oscillator energy levels have been replaced with those of the indicated potential functions. (38) In the case of Vl the lowest two eigenvalues are separated by a tunneling frequency of 21 cm-‘. Since this is only a fraction of thermal energy at room temperature these levels would be very nearly equally populated. Their average energy is shown in Figure 3 and used in the calculation of 61.None of the other potential functions shown has closely spaced energy values. (39) (a) A. L. MacDonald, J. C. Speakman, and D. Hadfi, J. Chem. SOC.,Perkin Trans. 2, 825 (1972);(b) G. E. Bacon, C. R. Walker, and J. C. Speakman,

ibid., 979 (1977). (40) R. Attig and J. M. Williams, J. Chem. Phys., 68, 1389 (1977). This paper actually deals with D502+, but the oxygen-oxygen distance and IR spectrum are similar to those found in Speakman’s “type A” bicarboxylates.

(41) L. J. Altman, D. Launganl, G. Gunnarsson, H. Wennerstrom, and S.ForsBn. J. Am. Chem. SOC., 100,8269 (1978). (42) G. E. Walrafen in “Hydrogen-Bonded Solvent Systems”, A. K. Covington and P. Jones, Eds., Taylor and Francis, London, 1968, p 16. (43) B. S.Auk, J. Phys. Chem., 82, 844 (1978). (44) Reference 34, pp 360-363. (45) J. A. Popie, W. G. Schneider, and H. J. Bernstein, “High ResolutionNuclear Magnetic Resonance”, McGraw-Hill, New York, 1959, pp 221-223. (46) M. C. Flanigan and J. R. de la Vega, Chem. Phys. Lett., 21, 521 (1973). (47) A. L. Andreassen, D. Zebeiman, and S.H. Bauer. J. Am. Chem. Soc., 93, 1 148 (197 1). (48) M. M. Kreevoy and B. Ridi, unpublished results. (49) A. J. Gordon and R. A. Ford, “The Chemist’s Companion”, Wiley, New York, 1972, p 107. (50) J. C. Speakman, Spec. Period, Rep.: Moi. Shuct. Diffr. Methods. 3,87-9 1 (1975).

A Resonance Raman/Iodine Mossbauer Investigation of the Starch-Iodine Structure. Aqueous Solution and Iodine Vapor Preparations Robert C. Teitelbaum,’a-c*dStanley L. Ruby,Ib and Tobin J. Marks*l*J Contributionfrom the Department of Chemistry and the Materials Research Center, Northwestern University,Evanston, Illinois 60201, and the Physics Division, Argonne National Laboratory, Argonne, Illinois 60439. Received August 2, 1979

Abstract: The structure of the blue-black iodine complex of amylose (the linear, helical component of starch), prepared either from iodine and iodide in aqueous solution or from crystalline amylose and iodine vapor, has been studied by resonance Raman and iodine- 129 Mossbauer spectroscopy. In both cases it is concluded that the identity of the major chromophore is essentially the same: the pentaiodide (IS-) anion. For the material prepared from iodine vapor, the iodide required for 1s- formation is produced by hydrolysis or alcoholysis of iodine. The other product of this reaction, a hypoiodite, has been assigned in the iodine Mossbauer spectrum.

Historically, scientific interest in the interaction of the various fractions of starch with iodine has stemmed from the 0002-7863/80/ 1502-3322$01 .OO/O

striking color changes which accompany c~mplexation.~ Amylose, the linear starch fraction, which is comprised of

0 1980 American Chemical Society

Teitelbaum, Ruby, Marks

/

3323

Starch-Iodine Structure

Several spectrophotometric and potentiometric investigations in aqueous solution later identified the requirement of I- for complex f ~ r m a t i o n Although .~ I- might have a structureforming role not related to inclusion, it was also recognized that it might be incorporated within the helix as discrete I- ions or as 13-. A model for starch-iodine composed of linear chains of 13- (C) as in (benzamide)zH+I3- l o has also been pro-

-

-

-

- - - I - I - I_ _ _ - -1- I - I . - - - - I - I - I - - C

posedl0 and widely accepted. The similarity between amylose and the cyclodextrins’ (cycloamyloses) in structure and in the tendency to form inclusion led to new structural suggestions for starch-iodine based upon the crystal structures of (a-CD)2.LiI3.12-8H20 and (cu-CD)2.Cdo.s.I.212.26H20 (a-CD = cyclohexaamylose).I2 Either alternating arrays of 12 and 13- units (D) or symmetrical, linear 15- units (E) were considered to be plausible structures.12aThere is additional precedent in polyiodide structural chemistry for either motif.143’5 Figure 1. Schematic view of the starch-iodine structure. Shown is the amylose helix with the iodine chain in the center.

r

I

1

[I- 1 - - - 1 - - - 1 - 1 1 -. . . . . . .[ I - I - - - I - - - I - I ] E

A large battery of physicochemical techniques has been I directed at identifying the form(s) of iodine present in starch-iodine. These methods have included o p t i ~ a l , ~ ~ ~ ~ ~ ~ infrared,I6 electron paramagnetic r e ~ o n a n c e ORD/CD,’* ,~~ A iodine M o ~ s b a u e r , ’and ~ resonance Raman spectroscopy,20 chains of 1,4 linked a-D(+)-glucopyranose units (A),394forms as well as X-ray powder d i f f r a c t i ~ n , ~spectrophotometric ~~~~~*g an intensely blue-black adduct with aqueous iodine in the and potentiometric titrations? and intrinsic viscosity studies8 presence of iodide: “starch-iodine”. In contrast, the branched Many of these investigations have unfortunately suffered from fraction of starch, amylopectin, does not form such a complex. the intrinsic deficiencies of the particular method when applied Although starch-iodine has been known for a great many to such a problem, and/or from the lack of realistic model compounds and modern data analysis procedures. Thus, in year^,^,^ and few chemists indeed are unfamiliar with the use of this intensely colored classic indicator in qualitative and early (but important) iodine Mossbauer studies,I9 only quantitative analysis,6there is surprisingly little unambiguous structural possibilities B and C were considered, and no atinformation on the actual structure of the chromophore with tempt was made to determine relative iodine site populations. A,,, =600 nm. Structure C was chosen over B because nonequivalent iodine In a series of pioneering optical and X-ray diffraction sites were observed and because spectral parameters were ~ t u d i e s Rundle ,~ and his co-workers established that the iodine reminiscent of (ben~amide)zH+’~~I3-. An early resonance component of starch-iodine is present in a unidimensional Raman investigation of starch-iodine20 also only considered array within an amylose helix of diameter -13 A, period -8 possibilities B and C ; structure C was assigned with limited A, and with six glucose residues per turn. The structure is il- reference to appropriate model compounds and the symmetries lustrated in Figure l . The amylose employed for the bulk of of internal coordinate changes most likely to be resonant enthese investigations was the crystalline “V” form which is an hanced. The degree to which the aqueous and “V” amylose alcohol inclusion c o m p l e (amylose ~ ~ ~ ~ is~now ~ ~known to form preparations of starch-iodine might or might not be the same a great many such complexes, all having the included molecules has been an additional complication. The net result of these contained within the h e l i ~ ~ Rundle’s ~ , ~ ) . samples were prepared efforts is that, it has not been possible to unambiguously difby staining the solid “V” amylose with iodine vapor. The fact ferentiate among the various proposed structures, and there that ordinary amylose will not form the complex under these has been no definitive information on the identity of the iodine conditions, and that starch-iodine has been traditionally prespecies in starch-iodine. pared in aqueous solution, in the presence of iodide, has inWe recently reported resonance Raman and iodine-129 troduced a significant complication in understanding the Mossbauer spectroscopic results on the structure of starchstructure of starch-iodine. iodine prepared in aqueous solution.21A full range of model The actual nature of the iodine chromophore within the compounds was utilized,2’ and Mossbauer spectral analysis amylose helix has been the subject of considerable speculation was carried out by iterative computer simulation. I t was conand controversy. The earliest proposals were that iodine is cluded in that preliminary report that the predominant iopresent as discrete I 2 molecules which are “dissolved” in the dine-containing species present was Is-, as in structure E. This relatively hydrophobic interior of the amylose helix ( B).5b*8 polyiodide species formally requires the presence of both 12 and I- for formation, in agreement with experimental observa- - - I - I - - - - I - I _ _ _ _ I-I-. t i o n ~An . ~ important and unanswered question in developing B a complete and coherent description of starch-iodine is whether

3324

Journal of the American Chemical Society

/ 102:lO /

May 7 , 1980

or not the chromophore in the material prepared from “V” amylose and I2 vapor is really the same, and, if so, how this comes about. In this contribution we present a full discussion

of our spectral studies of starch-iodine, prepared7 by both aqueous and 12 vapor methods.

A.SOLUTION

Experimental Section Synthesis of Starch-Iodine. Aqueous Solution Method?a Granular potato amylose was purchased from Aldrich Chemical Co. A typical preparation involved the reaction with stirring of 1.8 g of amylose, 0.88 g (3.5 mmol) of triply sublimed 12, and 0.53 g (3.5 mmol) of NaI in 50 mL of deionized water. The resulting blue-black complex, which is formed immediately, was collected by centrifugation and washed with deionized water. Washing was then repeated until no I- could be detected in the water washings when treated with AgNO3 solution. The product was then freeze-dried. Synthesis of Starch-Iodine. Iodine Vapor Methods7Granular potato amylose was crystallized from H2O and I-butanol according to the procedure of Schoch.22In a typical crystallization, 0.50 g of granular amylose was dissolved in 150 mL of boiling deionized water, with constant stirring. To the boiling solution, 10.5 mL of 1-butanol was slowly added, and the solution allowed to cool to room temperature over a period of about 6 h. The “wet” crystalline “V” amylose was collected by suction filtration and washed repeatedly with analytical reagent grade methanol. The remaining methanol was removed by drying in a vacuum desiccator. Iodination was accomplished by placing a vial of the crystalline “V” amylose in a closed vessel in contact with iodine vapor. The characteristic color of the starch-iodine complex developed immediately. A sample prepared in this manner typically contained 5 1 0 % iodine by weight. Synthesis of Starch-Iodine-129. Aqueous Solution Method. The reagent 12912 was prepared by oxidation of an acidified Na1291solution (obtained from Oak Ridge National Laboratory) and subsequent extraction with pentane. The pentane was then evaporated in a stream of prepurified nitrogen to yield solid 12912. For a source of I-, the basic Na1291solution from Oak Ridge was neutralized with H2S04 and used without dilution. The Mossbauer spectroscopy sample of the amylose-iodine complex was prepared by combining 0.7 l g of amylose with 35 mg of 12912and 0.80 mL of the neutralized Na1291solution in 30 mL of deionized water. The product was collected by centrifugation and washed with water, and the same freeze-drying procedure as described above was employed for drying. Synthesis of Starch-Iodine-129. Iodine Vapor Method. The Mossbauer spectroscopy sample of the “V” amylose-iodine complex was prepared using the same procedure as for the unenriched sample (vide supra), except that 12912was used. The reagent 12912was prepared as described above. Raman Measurements. Laser Raman spectra were recorded with Kr+ (6471 A) or Ar+ (4579,4880,4965,5145 A) excitation usinga Spex 1401 monochromator and photon counting detection. The solid samples were studied in 5- or 12-mm Pyrex sample tubes spinning at 1200 rpm. A 180° backscattering geometry was employed. A number of scans were made of each sample (the initial at lowest laser power) to check for possible sample decomposition. Spectra were calibrated with the exciting line (VO) or laser plasma lines. Iodine-129 Mossbauer Spectroscopy. These measurements employed the apparatus described p r e v i o ~ s l y .Absorbers ~ ~ ~ ~ ~ ~were prepared by thoroughly powdering the iodine- 129 enriched sample and, where necessary, mixing it with an inert filler (boron nitride) to achieve complete filling of the sample container. The absorbers typically contained ca. 7 mg 1291/cm2.Both the source and absorber were cooled to 4.2 K during data collection. Typically, two to three sources were used, in sequence, to collect all the data for a given sample. Data collected from each source were summed to give the final spectra. Individual runs were checked for reproducibility. The spectrometer velocity was generated with a feedback-controlled vibrator using sinusoidal acceleration, and the velocity drive was calibrated with S7Fe foil. Mossbauer effect data processing and analysis employed the computer program GENF1T,23 which finds the best values of the parameters of isomer shift, quadrupole coupling constant, line width, populations, base line, and asymmetry parameter via nonlinear least-squares minimization of the difference between the observed and calculated spectra. Starting spectral parameters were based on literature data,19,24a*34-36 and care was taken to approach optimum fits from several directions so that converging on local minima would be

&VAPOR

S00!0

V00!0

30010

?.00!0

!00!0

O!O

WAKNUHBLR C!/CM>

Figure 2. Resonance Raman spectra ( v g 5145 A) of (A) starch-iodine prepared in aqueous solution; (B) starch-iodine prepared with iodine vapor. avoided. The goodness of fit is judged by the parameter “Misfit”, which has been previously defined by Ruby.2s The end sites in the pentaiodide ion showed somewhat larger line widths than the other iodine sites. This could be explained if the end sites have a slight distribution of quadrupole coupling constants. That is, not all end sites are exactly equivalent. This nonequivalence was incorporated in the fitting procedure as a velocity-dependent linebroadening function

rn= v ‘ / [ K ( V-~ &)I2+ r t r u e Z

(1)

rtruc

where is the actual line width, 6 is the isomer shift, V, is the velocity of line n, K is the broadening parameter, and F,, is the observed width of the nth line. This function broadens the lines furthest away from the center of gravity (6, isomer shift) to a greater extent than those lines close to 6. This has the same overall effect as a quadrupole distribution. Using this value of I?,,, reasonable line widths for all sites are obtained.

Results and Discussion The reaction of amylose with iodine and iodide in aqueous solution or the reaction of the “V” form of amylose with iodine vapor gives an intensely blue-black adduct, “starch-iodine”. In investigating the nature of the iodine chromophore in these materials, by resonance Raman and iodine- 129 Mossbauer spectroscopy, we apply the techniques and criteria which we have developed in our solid-state studies of iodine-oxidized mixed valence compounds.24 In this contribution we restrict our discussion to amylose-iodine stoichiometry ranges and preparative methodologies which would be considered typicaL7 Resonance Raman Studies. In Figure 2 are presented resoby the aqueous solution, I- pr~cedure,’~ and by the I2 vapor p r ~ c e d u r e Importantly, .~ the spectra are essentially identical, indicating that the scattering species are essentially identical. The spectra vary only modestly with exciting laser frequency ( v g 5145-4579 A) as illustrated in Figure 3. In an effort to determine whether any structural changes in the samples might have occurred upon drying, freshly prepared solid samples were subjected to high vacuum ( Torr) for various periods of time. There was no observable change in Raman spectra over the course of several hours. Indeed, even at temperatures as high as 70 “ C , only slight changes in the spectrum could be discerned in the course of 1 h (Figure 4),i.e., a slight relative

Teitelbaum, Ruby, Marks

/

3325

Starch-Iodine Structure

A

, 8.4965A

B

!~

I, in benzene

d 1-1-1 1-1-1 1-1-1 T

,

l

l

500

l

l

l

l

4 00

l

l

l

l

,

,

,

,

,

,

,

,

,

2 00

300

,

,

,

,

100

D

W A V E NU M BE R ( c m-I

l ,

Figure 3. Exciting frequency ( 8 0 ) dependence of the resonance Raman spectrum of the starch-iodine complex: (A) Ar+, 4579 A; (B) Ar+, 4965 A; (C) Ar+, 5145 A.

i

1-1

1

; -I I 1-1 I I-

E

I iI i

h ll

I

500

400

300

WAVENUMBER

200

100

(cm-1)

Figure 4. Effects of heating and vacuum on the resonance Raman spectrum of the starch-iodine complex (80 5145 A).

Figure 5. Resonance Raman spectra (5145-A excitation) of (A) starch (amylose)-iodine; (B) polycrystalline (trimesic acid.H20)lo*H+Is-; ( C ) 12 dissolved in benzene; (D) polycrystalline (benzamide)2H+I3-; (E) polycrystalline (phenacetin)zH+13-.12; (F) polycrystalline (a-cyclohexaamylose)2Li+13-~I~.8H20.

increase (ca. 15%) in the intensity of the 109-cm-' scattering. The interpretation of these changes will be deferred until after spectral assignments have been made. Interestingly, aqueous suspensions of starch-iodine are decolorized by heating to temperatures greater than 60 oC,26 The resonance Raman spectra of both starch-iodine samples (Figures 2 and 3) exhibit strong scattering a t 163 cm-l and considerably weaker transitions a t 109 and 56 cm-'. In addition, overtone bands are assigned at 322 (2 X 163) and 215 (2 X 109) cm-I. A combination band which is assigned at 272 (163 109) cm-l is assurance that the 163- and 109-cm-l fundamental transitions are not in separate lattices (Le., due to different compounds). In interpreting the starch-iodine

spectra, reference is made to model compounds of established structurs which are relevant to the differentiation of species B-E, as well as other possibilities. The spectra of most polyiodides can be understood by recognizing that 12 acts as a Lewis acid and that coordination to electron donors (Le., I-) results in popu,ation of molecular orbitals with 1-1 antibonding character, hence an increase in 1-1 bond length and a decrease in bond order and stretching force constant are o b ~ e r v e d . ~ ~ . ~ ~ The spectra and structures of most complex polyiodides can be understood in terms of aggregates of the simpler subunits 12, 13-, and I-.24,28 Figure 5 compares the resonance Raman spectrum of starch-iodine to spectra of various model compounds. In a

SOOrO

350'0

'100'0

3S0 0

300'0

250'0

USIENJWER

+