Mechanism
The primary mechanism for metal incorporation by co-condensation of silica precursors and metal salts was disclosed by Vinu et al.9 For example, Al was successfully incorporated into SBA-15 by increasing the pH of the gel above the aqueous isoelectric point for silica. At pH > 2 silica precursor species in solution possess a negative charge due to deprotonation, which favors interaction with cationic metal hydroxide species found at the elevated pH. Initially, we believed that DHP could increase the pH of the solution by way of NH3 release, promoting Fe incorporation in a similar manner.
To investigate this hypothesis, we prepared four control vessels filled with identical quantities of H2O, HCl, DHP, and (NH4)2Fe(SO4)2 as those used in preparation of our Fe-SBA-15 samples and monitored their pH. Upon introduction of HCl the pH of each vessel was acidified to 1.16. Upon equilibration, an amount of DHP ranging from 0.25 g to 2.00 g was added to each of the vessels. For introduction of 0.25, 0.50, or 1.00 g of DHP, the solution pH remained below 1.67. However, upon adding 2.22 g of DHP, the pH rose from 1.16 to 3.72. Addition of 0.55, 1.11, or 2.22 g (NH4)2Fe(SO4)2 further increased the pH to a maximum of 1.87. When mixing 4.44 g of iron salt, pH decreased from 3.73 to 3.59. While DHP is a suitable ammonia source in the agricultural industry, we did not see evidence of a significant increase in solution pH or bubbling in the form of NH3 (g). Since the pH of the gel predominately remained below 2 for all samples, yet significant metal incorporation was still observed we believe a different mechanism for Fe incorporation must occur compared to that of the “pH-adjusting” method.10, 11
ICP-AES and EDAX show P as the only elemental component in the material outside of Fe, Si, and O in the final MSN. Since direct interaction between SBA-15 and Fe seems unlikely under the higher acidity conditions, we were prompted to investigate the role of P. Similar to SBA-15 prepared here, porous phosphosilicates are often synthesized from TEOS and orthophosphate by HCl catalyzed sol-gel
practices .12, 13 It has been shown by 31P NMR, 29Si NMR, Raman, and FTIR that phosphoric acid is capable of co-hydrolyzing and co-condensing with TEOS during the formation of mesoporous materials.13 Here, Fe incorporation could have been due to the in situ formation of oligomeric iron polyphosphates. Biologically important polyphosphates such as ATP have been shown to have high affinity for metal chelation.14 Additionally, Huo, Stucky, and Schuth describe the higher affinity of polydentate silicate oligomers towards cations as mechanism for self-assembly in base catalyzed ionic templated systems.3 If negatively charged polyphosphates were to form in solution they could chelate iron species and co-condense with the forming silica network.
However, the formation of polyphosphates at low pH is unfavorable and we found by XPS, ICP-OES, and XRD that the nP:nFe was consistently 1:1, whereas polyphosphate species would produce nP:nFe > 1.15 Additionally, if a phosphosilicate material was forming, we would also expect the nP:nFe to exceed one since the excess of TEOS would afford abundant opportunities for phosphorous-silica interaction compared to phosphorous-iron. Another alternative could have been phosphates playing the role of counterions during the counterion mediated assembly process of SBA-15.4, 16 A variety of counterions have been investigated including NO3, Br, Cl, SO4, but not phosphates.6 Typically the counterion influences the degree of ordering of the MSN or the overall particle morphology. By delaying addition of our DHP until after sufficient induction period for the formation of S+X- by protonated P-123 (S+) and chloride anions (X-), surplus of H2PO4- could not participate as X- reducing the possibility of Fe enrichment at the silica-template interface through phosphate-iron complexation.