Polysaccharide Coordination Complexes ... - ACS Publications

Article pubs.acs.org/IC

Mössbauer Spectroscopic Characterization of Iron(III)− Polysaccharide Coordination Complexes: Photochemistry, Biological, and Photoresponsive Materials Implications Hendrik Auerbach,† Giuseppe E. Giammanco,‡ Volker Schünemann,† Alexis D. Ostrowski,‡ and Carl J. Carrano*,§ †

Department of Physics, University of Kaiserslautern, 67663 Kaiserslautern, Germany Department of Chemistry, Bowling Green State University, Bowling Green, Ohio 43403 United States § Department of Chemistry and Biochemistry, San Diego State University, San Diego, California 92182-1030, United States ‡

S Supporting Information *

ABSTRACT: While polycarboxylates and hydroxyl-acid complexes have long been known to be photoactive, simple carboxylate complexes which lack a significant LMCT band are not typically strongly photoactive. Hence, it was somewhat surprising that a series of reports demonstrated that materials synthesized from iron(III) and polysaccharides such as alginate (poly[guluronan-comannuronan]) or pectate (poly[galacturonan]) formed photoresponsive materials that convert from hydrogels to sols under the influence of visible light. These materials have numerous potential applications in areas such as photopatternable materials, materials for controlled drug delivery, and tissue engineering. Despite the near-identity of the functional units in the polysaccharide ligands, the reactivity of iron(III) hydrogels can depend on the configuration of some chiral centers in the sugar units and in the case of alginate the guluronate to mannuronate block composition, as well as pH. Here, using temperature- and field-dependent transmission Mössbauer spectroscopy, we show that the dominant iron compound detected for both the alginate and pectate gels displays features typical of a polymeric (Fe3+O6) system. The Mössbauer spectra of such systems are strongly dependent on temperature, field, size, and crystallinity, indicative of superparamagnetic relaxation of magnetically ordered nanoparticles. Pectate and alginate hydrogels differ in the size distribution of the iron oxyhydroxy nanoparticles, suggesting that in general smaller nanoparticles are more reactive. Potential biological implications of these results are also discussed.

■

environment.7−9 Although the exact mechanism(s) for the photochemical degradation of these complexes is complex, the iron complex absorbs visible light, forming a short-lived excited state which decays nonradiatively by an electron transfer from the carboxylate ligand to the metal center.10 The iron is reduced and the carboxylate radical loses CO2, forming a carbon-centered radical which is able to undergo further chemistry.11 As indicated above, while polycarboxylates and hydroxy-acid complexes have long been known to be photoactive, simple carboxylate complexes which lack a significant LMCT band are not typically strongly photoactive. Hence it was somewhat surprising that a series of reports demonstrated that materials synthesized from iron(III) and polysaccharides such as alginate (poly[guluronan-co-mannuronan]) or pectate (poly[galacturonan]) formed photoresponsive materials that convert from hydrogels to sols under the influence of visible light.12−17 The iron(III)−alginate system in particular showed relatively efficient photochemistry, with quantum yields of around 10%.15 These materials have numerous potential applications in areas

INTRODUCTION The role of iron(III) complexes in the photodegradation of organic compounds has been known for a long time and has important applications in water pollution treatment.1,2 Simple low-molecular-weight soluble iron(III) hydroxo species appear to undergo photoreduction by UV light to produce iron(II) and the very reactive hydroxyl radical, which can then react with organic substrates via radical reactions, while solid-state iron(III) oxides and oxyhydroxides undergo photoinduced electron hole pair formation, leading to their functioning as semiconductor photocatalysts.3 “Simple” iron(III) polycarboxylate and hydroxy-acid complexes are among the most photoactive species known.4,5 Here visible or near-UV light is absorbed into a LMCT band of the complex, leading to reduction of iron(III) to iron(II) with a concomitant oxidation and decarboxylation of the organic component. For example, photolysis of iron(III) oxalate complexes is thought to be a major pathway for the formation of hydrogen peroxide in atmospheric waters.6 Iron complexes of bacterial siderophores such as vibrioferrin that contain an αor β-hydroxy acid moiety are also known to undergo photochemistry that ultimately converts biologically unavailable iron(III) into more bioavailable iron(II) in the marine © 2017 American Chemical Society

Received: May 5, 2017 Published: September 15, 2017 11524

DOI: 10.1021/acs.inorgchem.7b00686 Inorg. Chem. 2017, 56, 11524−11531

Article

Inorganic Chemistry

Figure 1. Mössbauer spectra of iron alginate hydrogel at pH 4.5 measured at T = 77 K (a) and at T = 6 K and an external field of B = 20 mT (b) and B = 5 T (c) parallel to the γ-ray beam. The solid lines are simulations with the parameters given in Table 1.

polysaccharides remains unknown, the characterization of which by Mössbauer spectroscopy is the focus of this report.

such as photopatternable materials, materials for controlled drug delivery, and tissue engineering.13−17 Despite the similarity of the functional units in the polysaccharide ligands, the reactivity of iron(III) hydrogels seems to depend on the configuration of some chiral centers in the sugar units and in the case of alginate the guluronate to mannuronate block composition, as well as pH.15 In a seemingly unrelated biological context, we have recently reexamined the extensive and powerful binding of iron to the surface of both the filamentous brown algae Ectocarpus siliculosus and the giant kelp Macrocystis pyrifera, determined its binding constants, and showed that the carboxylate groups from alginate were likely the biological ligands for iron.18,19 A biological role for this phenomenon as an iron buffer was proposed. While both algae showed surface binding of iron(III), the magnitudes of that binding were surprisingly different, with Ectocarpus seemingly binding more than 100 times as much iron as Macrocystis.20 This was despite the fact that, irrespective of the vastly different tissue morphologies of these two species, their cell walls are quite similar on a molecular level with alginate comprising 30−40% of the dry mass of both. However, in light of the above, a hypothesis to account for the unexpected difference between Macrocystis and Ectocarpus presents itself. Thus, it seems likely that any iron(III) bound to the carboxylate groups of alginate on the surface of Macrocystis blades, some 85% of which float on the surface of the ocean and thus are exposed to strong sunlight, could be expected to undergo photochemistry resulting in photoreduction with formation of soluble iron(II), which could then be lost to solution. Such photochemistry for the benthic Ectocarpus would be much less likely, thus maintaining coordination of iron(III) to the alginate. While some work has been done in identifying the actual photoreactive species for simple iron(III) complexes,10 the exact nature of the iron(III) coordinating species and hence the origin of the photoreactivity for iron bound to complex

■

EXPERIMENTAL SECTION

Low-viscosity sodium alginate from brown algae (Lot A 112), 1,10phenanthroline, and hydroxylamine hydrochloride were purchased from Sigma-Aldrich and used as received. Poly-D-galacturonic acid 95%, molecular weight 25000−50000 g/mol (Lot 81325), was purchased from Sigma-Aldrich and prepared as the sodium salt by neutralization with NaOH. Hydrogen peroxide, 30% by weight, and hydrogen chloride, 36.5−38%, were purchased from BDH Chemicals and used as received. Preparation of 57FeIII Solution. A 7.8 mg portion (0.137 mmol) of 57Fe0 powder was dissolved in 1 mL of 11.65 M HCl, at 96 °C for 4 h. The solution was cooled, and then 1.37 mmol of H2O2 (16 mL, 30% w/v) was added. After 5 min, the flask was immersed in an oil bath at 160 °C to evaporate the liquids. The remaining solid was dissolved in 1 mL of concentrated HCl, and 1.37 mmol of H2O2 (16 mL, 13% w/ v) was added again before evaporating the solution. This process was repeated twice before dissolving the total solids in 2 mL of deionized water. Complexation with 1,10-phenanthroline was used to evaluate the composition of the stock. Briefly, 11.8 mL of the stock solution was mixed with 2 mL 0.5 M acetate buffer, pH 4.0, and 1 mL of 0.1% w/v 1,10-phenanthroline in water. The solution was diluted to 10 mL and allowed to react for 15 min. The absorbance was measured at 510 nm and compared to calibration with FeCl2 standards in the range of 0.9−1.3 mM. According to this method, the 57Fe stock was >99.5% ferric iron. The total iron concentration was determined by treating the sample with 1 mL of 10% w/v hydroxylamine hydrochloride for 20 min before adding the phenanthroline. This method yielded a total [Fe] = 68.7 mM. Preparation of 57FeIII−Polyuronate Samples. For samples in solution, the iron to carboxylate molar ratio was fixed at 1:56. Briefly, 10 mg (0.0505 mmol) of sodium polyuronate was dissolved in 0.4 mL of deionized water. Next, the polysaccharide solution was slowly mixed with 0.5 mL of 1.8 mM 57FeIII with vigorous vortexing. A 0.5 M NaOH solution (10−20 μL) was used to adjust to pH 4.5 before diluting to 1 mL. For samples in the gel state, the iron to carboxylate ratio was adjusted to 1:4.9. In a general experiment, 16.4 mg (0.08 mmol) of sodium polyuronate was dissolved in 0.4 mL of deionized water. Next, 11525


Article

Inorganic Chemistry

Table 1. Mössbauer Parameters As Obtained from the Analysis of the Mössbauer Spectrum of Iron Alginate Hydrogel Displayed in Figure 1ba component Sb δ (mms−1) ΔEQ (mms−1) η Bhfc (T) Γd (mms−1) I1:I2:I3e D (cm−1) E/D Axx,yy,zz/μNgN (T)

α, β, γ (deg) areaf (%)

1

2

3

4

5

5/2 0.51 0.66 1.0

2 1.43 −3.25 1.0

0.53 0

0 0.53 −0.65 0.0

0.53 0

0.45

0.65

55.8 0.60;0.70;0.90 4.0:3.0:1.5

0.75

8.00

0.5 ± 0.2 0.33 −20.5 ± 2; −20.5 ± 2; −20.5 ± 2;

7±2 0.33 0.8 ± 0.3; −28.8 ± 2; 13.3 ± 1.5 24.8, 70.8, 82.6 8±2

19 ± 4

22 ± 2

41 ± 4

10 ± 2

All simulations shown in Figure 1 have been analyzed with one consistent data set, except some changes in line widths Γ, line ratio I1:I2:I3, area, and hyperfine field Bhf as given in the corresponding footnotes. bThe spin-Hamiltonian simulations of components 1 and 2 at T = 77 K were performed with an external field of B = 44 μT orthogonal to the γ-ray, which corresponds to the strongest part of the earth’s magnetic field in Kaiserslautern, Germany. cBhf at 6 K and 5 T: 51.3 T. dΓ at 77 K: component 1, 0.40 mms−1; component 2, 0.55 mms−1; component 3, Γ1,2,3 = 0.80, 0.80, 1.00 mms−1; component 4, 0.48 mms−1; component 5, 2.30 mms−1. Γ at 6 K and 5 T: component 1,0.70 mms−1; component 2, 0.42 mms−1; component 3, Γ1,2,3 = 0.85, 0.54, 0.63 mms−1; component 4, 0.48 mms−1. eI1:I2:I3 for component 3 at 6 K and 5 T: 3.7:1.3:2.7. fArea at 6 K and 5 T: component 3, 60 ± 2%; component 5, 0% a

the polysaccharide solution was slowly mixed with 0.5 mL of 34 mM 57 III Fe with vigorous vortexing. The gel was formed quickly after the pH of the solution was adjusted to 4.5 by adding 100 μL of 0.5 M NaOH. Samples were prepared in the dark or under red light and stored at 77 K until the irradiation experiments were performed. Transmission Mö ssbauer Spectroscopy (TMS). The Mössbauer spectra were recorded in transmission geometry using a constant acceleration spectrometer operated in conjunction with a 512-channel analyzer in the time-scale mode (WissEl GmbH). The detector consisted of a proportional counter, and the source contained 57Co diffused in Rh with an activity of 1.4 GBq. The spectrometer was calibrated against α-iron at room temperature. For measurements at 77 K, samples were placed in a continuous flow cryostat (OptistatDN, Oxford Instruments). Field-dependent conventional Mö ssbauer spectra were recorded with a closed-cycle cryostat from CRYO Industries of America, Inc. equipped with a superconducting magnet as described earlier.21 The field direction of the field-dependent measurements was parallel to the γ-ray beam. Spectral data were transferred from the multichannel analyzer to a PC for further analysis, employing the public domain program Vinda running on an Excel 2003 platform.22 The spectra were analyzed by least-squares fits using Lorentzian line shapes, Γ, with line intensities, I, unless mentioned otherwise. Spectral components attributed to paramagnetic iron(II) and iron(III) centers as well as diamagnetic dinuclear iron(III) centers were simulated by means of the spin-Hamiltonian formalism23 (see also the Supporting Information) Photolysis Experiments. The photolysis experiments were carried out in a dark room with 8 LEDs (P = 6 mW; Power Light Systems GmbH) with a wavelength of λ 405 nm and Γfwhm = 20 nm. As a power supply a programmable dc power supply (Model 382280) from EXTECH was used. The distance between the sample and the LEDs was approximately 10 cm. Before irradiation samples were thawed to room temperature in the dark. Irradiation with light was at room temperature for 45 min. After the photolysis the samples were immediately refrozen at 77 K for TMS analysis.

stay below the gelation point were insufficient to achieve adequate TMS spectra over reasonable time periods. Hence, only the hydrogels were studied further. The Mössbauer spectrum at 77 K and 0 T applied field of an iron−alginate hydrogel prepared from 17 mM 57Fe and 83 mM alginate at pH 4.5 and kept in the dark is shown in Figure 1a. The experimental spectrum shows a strong central quadrupole doublet with additional weaker magnetic features in the wings. Lowering the temperature to 6 K and applying an external field of 20 mT (Figure 1b) showed only slight changes in the spectral pattern: namely, an increase in the magnetic background with respect to the still-dominating central doublet. The fact that the spectral contribution of the central doublet decreases with decreasing temperature is indicative of the presence of superparamagnetic iron containing nanoparticles. The relatively sharp line observed at +3 mms−1 originates from a doublet with a high isomer shift >1 mms−1 characteristic for a ferrous high-spin FeII species. The complicated structure of the magnetic feature at both 77 and 6 K and the presence of a doublet even at 6 K clearly shows that the Mössbauer spectrum reflects a superposition of several iron species (see Figure S1 in the Supporting Information). Increasing the external field further to 5 T (Figure 1c) gave rise to a better-resolved spectrum in which it is clear that the magnetic spectral pattern could not be simulated by a single magnetic sextet, as can be seen from the shoulders at −6.5 and 7.3 mms−1 (see Figure S1c). In addition the intensity at around 0.5 mms−1 also points to the presence of a diamagnetic iron species. On the basis of these observations, all of the spectra displayed in Figure 1 could only be satisfactorily fit assuming five iron species: namely, two paramagnetic iron species (denoted in the following as species 1 and 2), two superparamagnetic nanoparticle-like species (species 3 and 5), and the diamagnetic iron species 4. The Mössbauer parameters of iron alginate hydrogel so obtained are outlined in Table 1. In the following we discuss the corresponding spectral assignments.

■

RESULTS Although we initially wished to study both the iron alginate and pectate sols and hydrogels, the concentrations of iron needed to 11526


Article

Inorganic Chemistry

Figure 2. Mössbauer spectra of iron−alginate hydrogel at pH 4.5 after irradiation with light for 45 min measured at T = 77 K (a) and at T = 6 K and an external field of B = 20 mT (b) and B = 5 T (c) parallel to the γ-ray beam. The solid lines are simulations with the parameters given in Table S2 in the Supporting Information.

At 6 K species 1 has an isomer shift of δ = 0.51 mms−1 and a quadrupole splitting of ΔEQ = 0.66 mms−1. These parameters are typical for high-spin ferric iron with octahedral N/O coordination.24 Species 1 comprises 10% of the total signal and displays a magnetic structure which has been simulated using the spin Hamiltonian formalism for a rhombic, paramagnetic, S = 5/2 species assuming a rhombicity parameter of E/D = 0.33 and a zero-field splitting D = 0.5 cm−1. Although it is minor, this iron species contributes to the Mössbauer spectra, as can be seen from Figure S2 in the Supporting Information. Paramagnetic high spin ferric ions with E/D = 0 have an axial ligand field, a situation commonly observed in iron porphyrins,25,26 whereas E/D = 0.33 reflects a rhombic ligand field. Both the observed isomer shifts and the quadrupole splittings are consistent with the mononuclear iron carboxylate complex proposed here.27 The presence of such a single ion rhombic paramagnetic S = 5/2 species with E/D = 0.33 is not unexpected. Such iron species are often observed in complex iron-containing systems such as heterogeneous iron catalysts28 vacuoles from fermenting Saccharomyces cerevisiae29 and iron siderophores30 leading to g = 4.3 signals in electron paramagnetic resonance (EPR) spectroscopy. Species 2 shows a doublet with δ = 1.43 mms−1 and |ΔEQ| = 3.25 mms−1 at a low field of 20 mT which changes to a broad ill-defined magnetic structure at an external field of 5 T (8% of the total spectral area). These parameters are typical for high spin ferrous iron24 with octahedral N/O coordination, which results from either traces of iron(II) in the initial 57Fe solution or some adventitious photochemistry (vide infra). Therefore, the spectral signature of species 2 can also be analyzed by the spin-Hamiltonian formalism. In fact, it is possible to reproduce this pattern by means of a spin-Hamiltonian simulation for a paramagnetic S = 2 species assuming a rhombicity parameter of E/D = 0.33 and a zero field splitting D = 7 cm−1. It should be noted that the full parameter set of species 2 was determined via analysis of an almost fully photolyzed alginate sample, the

Mössbauer spectra of which consisted almost entirely of species 2 (see Figure S3 and Table S1 in the Supporting Information). The observed D value is slightly lower than those observed for iron(II) hexaquo complexes in the solid state, where D values of 11−15 cm−1 have been determined by high-field EPR.31 Species 3 displays a magnetic sextet with δ = 0.53 mms−1 and a magnetic hyperfine field Bhf = 55.8 T at 6 K and 20 mT, the spectral line shape of which has been simulated by using simple Lorentzian lineshapes with the intensities I1:I2:I3 and the line widths Γ1,2,3 given in Table 1. On the basis of its Mössbauer parameters species 3 likely represents an iron oxo-hydroxo nanoparticle phase.32 Species 5 has the same isomer shift as species 3 but shows a broad magnetic feature at 6 K and a field of 20 mT. A welldefined magnetic sextet is observed when the electronic relaxation time τel is much longer than the Larmor precession time τL of the magnetic moment of the 57Fe nucleus in the magnetic hyperfine field generated by the iron’s electron shell. When the electronic relaxation time approaches τL, line broadening is observed. When τel ≈ τL, even more broadened spectral features occur with the magnetic splitting depending on the electronic relaxation time together with a partial collapse of the magnetic splitting into a central line.33 Therefore, we believe that species 5 represents smaller nanoparticles which are still superparamagnetic at 6 K and have a relaxation time comparable to the Larmor precession time τL which in a field of 55.8 T for a nucleus in its ground state (I = 1/2) is ∼10 ns and for its excited 14.4 keV nuclear state is ∼20 ns (see also the Supporting Information). The application of a large external field is expected to slow the relaxation rate of such particles.33 Indeed, increasing the external field to 5 T gives a spectrum where the intermediate relaxing species 5 has disappeared, all of whose intensity goes into that for species 3, showing clearly that 3 and 5 are the same relaxing nanoparticles/aggregates differing only in size and hence relaxation rate. Together 3 and 5 represent about 60% of the total area. 11527


Article

Inorganic Chemistry

Figure 3. Mössbauer spectra of iron pectate hydrogel at pH 4.5 measured at T = 77 K (a) and at T = 5 K and an external field of B = 20 mT (b) and B = 5 T (c) parallel to the γ-ray beam. The solid lines are simulations with the parameters given in Table S4 in the Supporting Information.

Finally, the doublet representing species 4 has δ = 0.53 mms−1 and |ΔEQ| = 0.65 mms−1 at 6 K in low field and constitutes ca. 22% of the total area. On the basis of the fact that at high field 4 could be simulated with the spinHamiltonian formalism by assuming a diamagnetic species which only shows a magnetic splitting due to an external field of 5 T (see simulation of 4 in Figure 1c) and that its isomer shift is characteristic for a ferric high-spin species, we conclude that 4 represents a diamagnetic complex likely to be a hydroxo-/carboxylato-bridged dimer similar to those occurring in the subunits R2 of class Ic ribonucleotide reductase and the R2-like ligand-binding oxidases (R2lox).34 Photolysis of the iron−alginate hydrogel at 405 nm for 45 min brought about a number of changes to the Mössbauer spectra (Figure 2). The spectral analysis itself, however, could be performed with the same species 1−5 as discussed above, the parameters of which are displayed in Table 1. In most cases only the relative proportions of the iron species change after photolysis (for details see Table S2 in the Supporting Information). The most obvious change was an increase in the amount of species 2, i.e. the high spin iron(II), from ca. 8% to 32% of the total signal. The isomer shift and quadrupole splitting of this iron(II) species are virtually identical with those reported for the Fe(H2O)62+ ion,35 indicating that after photoreduction the iron(II) is no longer coordinated to the alginate, consistent with the breakdown of the iron(III) crosslinked hydrogel into an un-cross-linked sol. Beyond this, the composition of the partially photolyzed mixture, as determined by fitting the 6 K and 5 T spectra, indicate little change in the amount of species 1, i.e. the mononuclear iron(III) complex, and a small reduction in the diamagnetic species 4, with the largest reduction being in the amount of the magnetically ordered iron nanoparticles, species 3(5), suggesting that the last species is the primary photoactive species. Raising the pH of a sample of the iron(III)−alginate hydrogel from 4.5 to 8.0 resulted in a color change from pale yellow to red-brown and a conversion of the gel to a sol.

Mössbauer spectra are displayed in Figures S4 and S5 in the Supporting Information, and parameters are given in Table S3 in the Supporting Information. The Mössbauer parameters are very similar to those reported for γ-FeOOH,36 indicating that the polysaccharide carboxylate coordination cannot compete with iron(III) hydrolysis at this pH. As expected, photolysis at 405 nm for up to 75 h showed no photoreduction of the iron(III) (see Figure S6 in the Supporting Information). Since previous work indicated that iron(III)−alginate sols were significantly more photoreactive than the corresponding iron(III)−pectate sols despite the near-identity of the functional units in the polysaccharide ligands, which differ only in the configuration of some chiral centers in the sugar units, we hoped to compare the iron(III) coordination modes between the two.14 Figure 3 shows the Mössbauer spectra of an iron− pectate hydrogel prepared from 17 mM 57Fe and 83 mM pectate at pH 4.5 and kept in the dark at various temperatures and external magnetic fields. The spectra for the pectate gels were first fit by assuming the same five species with very similar Mössbauer parameters as for the alginate gels. However, this led to a misfit of the Mössbauer spectrum obtained at 5 K and 20 mT (data not shown). Therefore, the new species 6, exhibiting a broad magnetic structure, had to be introduced (Figure 3a,b). Under low-field conditions and 5 K (Figure 3b) species 6 comprises 41% and species 5 11% of the total spectral area. Both components disappear in a field of 5 T (Figure 3c), and all their intensity goes into species 3, which then comprises 82% of the area at 5 T (Table S4 in the Supporting Information, footnote Δ). As pointed out above, this behavior is typical for superparamagnetic agglomerates, the relaxation time of which is in the range of the nuclear moments precession time τL at low fields. The application of a high external field slows the relaxation time and a magnetic sextet is observed. The final parameter set is given in Table S4. In comparison to the alginate sample the pectate sample has comparable amounts of species 1 (9 vs 10%), and species 2 (4 vs 8%) but a higher 11528


Article

Inorganic Chemistry

Figure 4. Mössbauer spectra of iron pectate hydrogel at pH 4.5 after irradiation with light for 45 min measured at T = 77 K (a) and at T = 6 K and an external field of B = 20 mT (b) and B = 5 T (c) parallel to the γ-ray beam. The solid lines are simulations with the parameters given in Table S5 in the Supporting Information.

polymeric (Fe3+O6) system. The Mössbauer spectra of such systems are strongly dependent on temperature, field, size, and crystallinity, reflecting superparamagnetic relaxation of magnetically ordered nanoparticles.20 Such iron oxo/hydroxo nanoparticles typically show superparamagnetic splitting (doublet− sextet transition) at temperatures below 77 K and can be, in addition, considerably broadened due to surface effects in small nanoparticles (because the surface atoms have a different chemical environment than in the bulk) and/or amorphous character of the mineral (because of the variable chemical environment). Larger and/or more crystalline particles typically show superparamagnetic blocking at higher temperature in comparison to smaller/more amorphous particles. On this basis, species 3 represents larger or more highly crystalline nanoparticles, species 5 small or less crystalline nanoparticles, and species 6 nanoparticles intermediate in size and/or crystallinity. As indicated above, previous work suggested that iron(III)− alginate sols were more photoreactive than the corresponding iron(III)−pectate sols. Hence, we hoped to see differences in the iron(III) coordination modes between the two to account for this observation. However, in our hands we saw no difference in photoreactivity between the two gels, with similar amounts of iron(II) being formed after 45 min of photolysis at 405 nm (a net increase in iron(II) of 24% for alginate vs 24% for pectate). The same trend was also true for longer irradiation times. These observations are likely the result of (a) the much high concentrations needed for the TMS as compared to the solution optical spectral studies of quantum yield even when enriched with 57Fe and (b) differences between sol and gel where we were past the gelation point for both species. Indeed, preliminary unpublished results indicate that the differences in photoreactivity between alginate and pectate are found to decrease markedly with increasing iron concentration and on converting from sol to gel.

percentage of species 3 (30 vs 19%) and a lower percentage of species 4 (5 vs 22%) and species 5 (11 vs 41%). Upon photolysis under the same conditions as for the alginate (Figure 4) the Mössbauer spectra were again fit to the same set of species but with an increase in the hexaaquo iron(II), species 2, consistent with photoreduction. Beyond the increase in iron(II) the spectral fitting showed a significant decrease in species 3(5,6) and essentially no change in species 1 and 4 after irradiation with light for 45 min. That only the percentage of magnetically ordered nanoparticulate species 3(5,6) changes upon photolysis again suggests that it is the primary photoactive species. Long-term irradiation (480 min) yields a relative spectral contribution of the FeII species 2 of 76% (see Figures S7 and S8 and Table S6 in the Supporting Information).

■

DISCUSSION The occurrence of several forms of iron in complex systems such as iron−alginate or −pectate hydrogels is indeed expected. These are shown to be crystalline and/or amorphous iron oxide/hydroxide phases occurring in different sizes from the nano- to the micrometer regime exhibiting superparamagnetism, polynuclear complexes such as diamagnetic μ-oxo-bridged diiron sites, and noninteracting isolated paramagnetic iron sites, all likely coordinated to alginate or pectate carboxylate groups. By means of field- and temperature-dependent Mössbauer spectroscopy, it is possible to disentangle at least the isolated paramagnetic and diamagnetic species from magnetically ordered phases and superparamagnetic nanoparticles via a description of the former by means of the spin-Hamiltonian formalism which we have included in our analysis. The dominant iron compound detected for both the alginate and pectate hydrogels prior to photolysis is species 3(5,6), which accounts for between 60−80% of the absorption area (Table 1 and Table S4 in the Supporting Information). The Mössbauer spectrum of this species displays features typical of a 11529


Article

Inorganic Chemistry Notes

One clear difference in the iron chemistry between the alginate and pectate gels was noted, however, in that the size or crystallinity of the magnetically ordered species differs between the two with alginate consisting of 19% “large/crystalline” particles and 41% “small/amorphous” particles while the pectate had 30% “large”, 41% “medium”, and 11% “small”. Upon partial photolysis the data are consistent with the magnetically ordered species 3(5,6) being the primary photoactive centers. For the alginate gels it is species 5 that decreases most markedly upon photolysis, while for pectate it is species 6, suggesting that in general the small to medium nanoparticles reacted more quickly. However, it should be noted that, if photolysis is allowed to go to completion (ca. 480 min), almost all the iron(III) species except 1 ultimately disappear and are converted into iron(II) species 2 (see Figures S3, S7, and S8 in the Supporting Information). The results presented here also have potential biological relevance. Previous work has established that the iron on the surface of macro brown algae such as Ectocarpus siliculosus is in the form of polymeric iron/oxo/hydroxo nanoparticles attached to the surface via carboxylate groups from the alginate. Mössbauer spectroscopic parameters for this surfacebound iron (δ = 0.49 mms−1, ΔEQ = 0.65 mms−1, and Bhf = 45 T) are very similar to those found for species 3(5) in this study.20 Thus, the present results provide support to the idea that iron bound to the surface of macroalgae through the alginate groups that predominate in their cell walls could be photoreactive. It also provides a potential rationale for the observation that the benthic alga Ectocarpus has far more iron bound to its cell surface than the compositionally similar but physiologically and environmentally different Macrocystis pyrifera (giant kelp), the majority of which floats on the sea surface and is exposed to strong sunlight. Alternatively, the different species of macro brown algae (Ectocarpus vs Macrocystis) are known to have different ratios of mannuronate (M) vs guluronate (G) in their native alginate structure.39 Such compositional variance can account for large structural changes in the alginate, where M-rich blocks form a 2:3 helix vs the Grich blocks that form a 3:1 helix.37 These differences in helical structure have been shown to coordinate to divalent metal ions differently;38 thus, it is likely that changes in M and G composition could result in different types of coordinated iron(III) species as well. Indeed, changes in the G and M composition in the alginate affect the efficiency of the photochemical reaction, at least in the sol phase.15

■

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS V.S. and H.A. acknowledge the support by the research initiative NANOKAT and by the German Science foundation (DFG) via SFB/TRR88 3MET. C.J.C. thanks the German Academic Exchange Service (DAAD) for a 2015 Faculty Visit Research grant. We thank several anonymous reviewers for helpful comments.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b00686. Spin-Hamiltonian formalism in Mössbauer spectroscopy, experimental errors of Mössbauer parameters, calculation of the Larmor precession of the 57Fe nucleus, and additional Mössbauer data (PDF)

■

REFERENCES

(1) Nansheng, D.; Tao, F.; Shizhong, T. Photochemical degradation of dyes in aqueous solutions containing Fe(III)-hydroxy complex. Chemosphere 1996, 33, 547−557. (2) Xu, P.; Zeng, G. M.; Huang, D. L.; Feng, C. L.; Hu, S.; Zhao, M. H.; Lai, C.; Wei, Z.; Huang, C.; Xie, G. X.; Liu, Z. F. Use of iron oxide nanomaterials in wastewater treatment: A review. Sci. Total Environ. 2012, 424, 1−10. (3) Feng, W.; Nansheng, D. Photochemistry of hydrolytic iron (III) species and photoinduced degradation of organic compounds. Chemosphere 2000, 41, 1137−1147. (4) Prihod’ko, R. V.; Soboleva, N. M. Photocatalysis: Oxidative Processes in Water Treatment. J. Chem. 2013. (5) DeGraff, B. A.; Cooper, G. D. On the photochemistry of ferrioxalate system. J. Phys. Chem. 1971, 75, 2897−2902. (6) Zuo, Y.; Hoigne, J. Photochemical decomposition of oxalic, glyoxalic and pyruvic acid catalyzed by iron in atmospheric waters. Atmos. Environ. 1994, 28, 1231−1239. (7) Amin, S. A.; Green, D. H.; Küpper, F. C.; Carrano, C. J. Vibrioferrin, an Unusual Marine Siderophore: Iron binding, Photochemistry, and Biological Implications. Inorg. Chem. 2009, 48, 11451− 11458. (8) Amin, S. A.; Green, D. H.; Küpper, F. C.; Hart, M.; Sunda, W.; Carrano, C. J. Bacterial-Algal Interactions: Siderophore Based Photochemical Iron ″Sharing″. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 17071−17076. (9) Barbeau, K.; Rue, E. L.; Bruland, K. W.; Butler, A. Photochemical cycling of iron in the surface ocean mediated by microbial iron(III)binding ligands. Nature 2001, 413, 409−413. (10) Pozdnyakov, I. P.; Wu, F.; Melnikov, A. A.; Grivin, V. P.; Bazhin, N. M.; Chekalin, S. V.; Plyusnin, V. F. Russ. Chem. Bull. 2013, 62, 1579. (11) Abrahamson, H. B.; Rezvani, A. B.; Brushmiller, J. G. Photochemical and spectroscopic studies of complexes of iron(III) with citric acid and other carboxylic acids. Inorg. Chim. Acta 1994, 226, 117−127. (12) Narayanan, R. P.; Melman, G.; Letourneau, N. J.; Mendelson, N. L.; Melman, A. Photodegradable Iron(III) Cross-Linked Alginate Gels. Biomacromolecules 2012, 13 (8), 2465−2471. (13) Bruchet, M.; Melman, A. Fabrication of patterned calcium crosslinked alginate hydrogel films and coatings through reductive cation exchange. Carbohydr. Polym. 2015, 131, 57−64. (14) Giammanco, G. E.; Ostrowski, A. D. Photopatterning the Mechanical Properties of Polysaccharide-Containing Gels Using Fe3+ coordination. Chem. Mater. 2015, 27 (14), 4922−4925. (15) Giammanco, G. E.; Sosnofsky, C. T.; Ostrowski, A. D. LightResponsive Iron(III)−Polysaccharide Coordination Hydrogels for Controlled Delivery. ACS Appl. Mater. Interfaces 2015, 7 (5), 3068− 3076. (16) Giammanco, G. E.; Carrion, B.; Coleman, R. M.; Ostrowski, A. D. Photoresponsive Polysaccharide-Based Hydrogels with Tunable Mechanical Properties for Cartilage Tissue Engineering. ACS Appl. Mater. Interfaces 2016, 8 (23), 14423−14429. (17) Machida-Sano, I.; Ogawa, S.; Ueda, H.; Kimura, Y.; Satoh, N.; Namiki, H. Effects of Composition of Iron-Cross-Linked Alginate Hydrogels for Cultivation of Human Dermal Fibroblasts. Int. J. Biomater. 2012, 2012, 1.

AUTHOR INFORMATION

Corresponding Author

*E-mail for C.J.C.: [email protected]. ORCID

Alexis D. Ostrowski: 0000-0002-3207-1845 Carl J. Carrano: 0000-0002-7447-4087 11530


Article

Inorganic Chemistry

with calcium ions: a molecular modeling investigation. Carbohydr. Res. 1999, 317, 119. (38) Agulhon, P.; Markova, V.; Robitzer, M.; Quignard, F.; Mineva, T. Structure of Alginate Gels: Interaction of Diuronate Units with Divalent Cations from Density Functional Calculations. Biomacromolecules 2012, 13, 1899. (39) Küpper, F. C. Unpublished personal communication.

(18) Miller, E. P.; Böttger, L. H.; Weerasinghe, A. J.; Crumbliss, A. L.; Matzanke, B. F.; Meyer-Klaucke, W.; Küpper, F. C.; Carrano, C. J. Surface Bound Iron: A Metal Ion Buffer in the Marine Brown Alga Ectocarpus siliculosus? J. Exp. Bot. 2014, 65, 585−594. (19) Böttger, H. H.; Miller, E. P.; Andresen, C.; Matzanke, B. F.; Küpper, F. C.; Carrano, C. J. Atypical Iron Storage in Marine Brown Algae: a Multidisciplinary Study of Iron Transport and Storage in Ectocarpus siliculosus. J. Exp. Bot. 2012, 63, 5763−5772. (20) Miller, E. P.; Auerbach, H.; Schünemann, V.; Carrano, C. J. Surface Binding, Localization and Storage of Iron in the Giant Kelp Macrocystis pyrifera. Metallomics 2016, 8, 403−411. (21) Janoschka, A.; Svenconis, G.; Schünemann, V. A closed cyclecryostat for high-field Mössbauer spectroscopy. J. Phys.: Conf. Ser. 2010, 217, 012005. (22) Gunnlaugsson, H. P. Spreadsheet based analysis of Mössbauer spectra. Hyperfine Interact. 2016, 237, 79. (23) Trautwein, A. X.; Bill, E.; Bominaar, E. L.; Winkler, H. Ironcontaining proteins and related analogs  complementary Mössbauer, EPR and magnetic susceptibility studies. Struct. Bonding (Berlin, Ger.) 1991, 78, 1. (24) Schünemann, V.; Winkler, H. Structure and dynamics of biomolecules studied by Mössbauer spectroscopy. Rep. Prog. Phys. 2000, 63, 263. (25) Peisach, J.; Blumberg, W. E.; Ogawa, S.; Rachmilewitz, E. A.; Oltzik. The Effects of Protein Conformation on the Heme Symmetry in High Spin Ferric Heme Proteins as Studied by Electron Paramagnetic Resonance. J. Biol. Chem. 1970, 245, 3342. (26) Yonetani, T.; Drott, H. R.; Leigh, J. S., Jr.; Reed, G. H.; Waterman, M. R.; Asakura, T. Electromagnetic Properties of Hemoproteins. J. Biol. Chem. 1969, 245, 2998. (27) Dziobkowski, C.; Wrobleski, J.; Brown, D. Magnetic Properties and Mossbauer spectra of several iron(III)-dicarboxilic acid complexes. Inorg. Chem. 1981, 20, 671−678. (28) Pérez Vélez, R.; Ellmers, I.; Huang, H.; Bentrup, U.; Schünemann, V.; Grünert, W.; Brückner, A. Identifying active sites for fast NH3-SCR of NO/NO2 mixtures over Fe-ZSM-5 by operando EPR and UV−vis spectroscopy. J. Catal. 2014, 316, 103−111. (29) Cockrell, A. L.; Holmes-Hampton, G. P.; McCormick, S. P.; Chakrabarti, M.; Lindahl, P. A. Mössbauer and EPR Study of Iron in Vacuoles from Fermenting Saccharomyces cerevisiae. Biochemistry 2011, 50, 10275−10283. (30) Carrano, C. J.; Drechsel, H.; Kaiser, D.; Jung, G.; Matzanke, B.; Winkelmann, G.; Rochel, N.; Albrecht-Gary, A. M. Coordination Chemistry of the Carboxylate Type Siderophore Rhizoferrin: The Iron(III)Complex and Its Metal Analogs. Inorg. Chem. 1996, 35, 6429−6436. (31) Telser, J.; van Slageren, J.; Vongtragool, S.; Dressel, M.; Reiff, W. M.; Zvyagin, S. A.; Ozarowski, A.; Krzystek, J. High-frequency/ high-field EPR spectroscopy of the high-spin ferrous ion in hexaaqua complexes. Magn. Reson. Chem. 2005, 43, S130−S139. (32) Mørup, S. Mössbauer Effect in Small Particles. Hyperfine Interact. 1990, 60, 959−973. (33) Mørup, S. Magnetic Relaxation Phenomena. In Mössbauer Spectroscopy and Transition Metal Chemistry; Gütlich, P., Bill, E., Trautwein, A. X., Eds.; Springer-Verlag: Berlin, 2011. (34) Kutin, Y.; Srinivas, V.; Fritz, M.; Kositzki, R.; Shafaat, H. S.; Birrell, J.; Bill, E.; Haumann, M.; Lubitz, W.; Högbom, M.; Griese, J. J.; Cox, N. Divergent assembly mechanisms of the manganese/iron cofactors in R2lox and R2c proteins. J. Inorg. Biochem. 2016, 162, 164− 177. (35) Lakatos, B.; Korecz, L.; Meisel, J. Comparative study on the mössbauer parameters of iron humates and polyuronates. Geoderma 1977, 19, 149. (36) Oh, S. J.; Cook, D.; Townsend, H. Characterization of iron oxides commonly formed as corrosion products on steel. Hyperfine Interact. 1998, 112, 59. (37) Braccini, I.; Grasso, R. P.; Pérez, S. Conformational and configurational features of acidic polysaccharides and their interactions 11531


Polysaccharide Coordination Complexes ... - ACS Publications

Recommend Documents