Skip to main content

Research Description

Research Description

The research we conduct unites fundamental studies with real-world applications in the general areas of environmental chemistry, metal and metalloid chelate compounds, and solid-state materials. The research combines aspects of inorganic, organic, analytical, biological, and physical chemistry and consequently utilizes a broad range of analytical techniques. Examples of these techniques are listed in the sidebar, Research Techniques. Much of the research is collaborative and conducted with students and faculty at other institutions and with scientist in industry and national laboratories. (Some representative references are provided; the numbers (#) refer to the publication list on this website under Academics.)

I. Environmental Chemistry

The synthetic dithiol compound, BDTH2*, has emerged as an exceptional reagent for removing mercury and other soft divalent metals from water through the formation of covalent metal-sulfur bonds (Figure below). BDTH2 has good thermal stability, solubility in a wide range of organic solvents, does not easily form disulfide bonds, and has no odor. BDTH2 has sufficient solubility in water (400 – 2000 parts-per-million, ppm) to address most aqueous metal-contamination problems. BDTH2 has the remarkable ability, when used in a 10% excess, to precipitate Cd, Hg, and Pb from water to below detectable limits (low parts-per-billion, ppb). Furthermore, the BDT-metal precipitates are impervious to leaching except under extremely acidic or basic conditions. The covalent sulfur-metal bonds in the BDT-metal compounds are exceedingly stable and release metal only under extremely acidic and basic conditions (#161, Fuel 89 (2009) 1326). *The abbreviation “BDTH2” is derived from the common name “BenzeneDiamidoEthaneThiol” (IUPAC nomenclature, N,N’-Bis(2-mercaptoethyl)isophthalamide).

BDTH2 irreversibly precipitates mercury under a wide range of laboratory conditions, from groundwater on-site at former chlor-alkali facilities, contaminated soils outside a natural gas pressure-monitoring station (#129), and from gold mining effluent at pH 12 (#114, Env. Sci. Technol. 36 (2002) 1636 and #124, Ind. Eng. Chem. Res. 42 (2002) 5278. BDTH2 precipitates a variety of divalent metals from acid mine drainage (#120) and lead from lead-battery recycling effluent (#113, Ind. Eng. Chem. Res. 41 (2002) 1579). Arsenic has been of particular interest over the past few years. We have conducted field studies of agricultural lands where arsenic-containing poultry litter has been applied and we’ve prepared of a filtration unit capable of removing arsenite from tap water. 

Continuing research is elucidating the fundamental properties of BDTH2 and developing and understanding of the structure, bonding, and reactivity of metals and metalloids with BDTH2. Derivatives of BDTH2, and new compounds distinct from BDTH2, are being designed with properties targeted for specific industrial applications. Cysteamine, a component of the BDTH2 molecule, is a biological molecule and related to the amino acid, cysteine. Mercury forms linear compounds with cysteamine and cysteine (Figure 1(b); #109, Polyhedron 40 (2002) 225), but compounds with cysteamine and heavy metal halides have very diverse, multinuclear structures (Cd: #134; Pb: #146; Hg: #141, #149, Inorg. Chem. 45 (2006) 2112 and #150, Inorg. Chem. 45 (2006) 7261). The type of halide has a structure-directing influence in the mercury compounds (Figure 2; #140, Inorg. Chem. 44 (2005) 5753, #141, #149,Inorg. Chem. 45 (2006) 2112, #150, Inorg. Chem. 45 (2006) 7261). We use structures like these to better understand the toxicity, mobility, and speciation of metals and metalloids in biological systems.

Water purification technologies that we or others develop need to have the capability of detecting and quantifying the contaminants being removed. For example, a household water filtration system that removes arsenic or heavy metals should be able to “turn-off” when the filter has reached capacity or when the filter has been compromised or does not perform properly. To address this need we have partnered with Quansor, Inc. to develop new sensors for the real time detection of inorganic, biological, and organic contaminants in water. The proprietary Quansor Monitor system has the capability of determining the levels of specific contaminants as the water is passing through the system, in “real-time”. The Quansor Monitor is designed so that household or industrial water-quality information can be transmitted wirelessly to a computer or cell phone. We are specifically responsible for the sensor chemistry which we can immediately test and demonstrate on the Quansor Monitor installed in our laboratory.

II. Metal and Metalloid Chelate Compounds

Over the past two decades we have developed an extensive series of group 13 chelate compounds (Chem. Rev. 101 (2001) 37), many of which are cationic, that are catalysts for oxirane and lactide oligomerization and polymerization (#110, J. Chem. Soc. Dalton Trans. (2002) 410). The two most common structures for the compounds are binuclear (salhen(tBu)[BBr2]2) and mononuclear (salen(tBu)AlBr) (Figure 3) where salen is a tetradentate chelate). We have made a wide variety of metal and metalloid salen compounds in different stoichiometries and with different structures. Indeed, there really are no limitations to what a student could work on in this area by varying the type of ligand, the ligand substituents, metal or metalloid, reaction stoichiometry, and conditions. For example, we’ve made chelate compounds with the Group 2 elements in liquid ammonia and zinc compounds with the salen ligand derivatized to be water-soluble as potential CO2 capture agents (an extension of our much earlier work with zinc chelates; Inorg. Chim. Acta 277 (1998) 157).

Figure 3. The Crystal Structures of Salhen(tBu)(BBr2)2 (above) and Salen(tBu)AlBr (below)

Two specific projects will serve as examples of the research we are now conducting in this area. The first is the characterization of the monomeric, oligomeric and polymeric aluminum chelate-containing compounds that result from phosphate ester dealkylation (Section A).  In the second project we use aluminum chelates to deactivate nerve agents and pesticides through dealkylation (Section B and example in Figure 4). Studies on actual nerve agents and pesticides are conducted by collaborators at the Edgewood Chemical and Biological Center. We use non-toxic, much less harmful nerve agent and pesticide simulants and model compounds at UK. The ultimate goal of this project is to create water-soluble compounds that can decontaminate objects, such as vehicles or aircraft and their interior areas, by using the compound like soap and water (Figure 5).

Figure 4. Deactivation of Nerve Agent VX through Dealkylation and Covalent Bonding. (SAB = Salen Aluminum Bromide)

Figure 5. Water-Soluble Derivative of Aluminum Chelate (SAB-1) for the Decontamination of Objects Exposed to Organophosphate Chemical Weapons.


A. Aluminum Chelate Compounds through Dealkylation

            In the course of exploring the chemistry of chelated boron cations we discovered a unique reaction where salenAlX compounds (X = Cl and Br) induced cleavage of all three alkyl groups in trialkyl phosphates, (RO)3P=O (#112, J. Am. Chem. Soc. 124 (2002) 1864). Subsequently, we determined that the mono-metallic aluminum chelates, and in particular, the salenAlBr compounds had the most utility in dealkylating a broad range of phosphates in short periods of time (#142, J. Am. Chem. Soc. 128 (2006) 1147). We were quickly able to confirm the identity of the first two compounds, (a) and (b), in the proposed reaction sequence (Figure 6).

Figure 6. Complete Dealkylation of Trialkyl Phosphate to Produce Three New Inorganic Aluminum Phosphates (a), (b), and (c).

One example of compound (a) is the first structurally-characterized terminal aluminum phosphinate (Figure 7; #147, Inorg. Chem. 45 (2006) 3970). We had previously established that aluminum-phosphinates form dimeric and polymeric compounds (# 111, Inorg. Chem. 41 (2002) 558) but in this example the bulky salen(tBu) ligand and a moderately sterically encumbered phoshphinate led to the isolation of the unique monomer. Our current goals are to isolate further examples of the twice-dealkylated products (b), and possibly of greater interest, the fully dealkylated product (c). These compounds should be soluble in organic solvents suitable for characterization by X-ray crystallography and would be a new class of inorganic aluminum phosphate. However, it is unlikely that the structures of the compounds will be as simple as represented in Figure 6(c) with three very bulky salen(tBu)Al units surrounding a central phosphate. Rather, the resulting compounds could have extended structures and will probably be ionic. Whatever new structures we discover will certainly posses unique bonding and reactivity.

Figure 7. Single Dealkylation of Phosphinate to Produce Terminal Aluminum Phosphate and the Crystal Structure of the Compound.

B. Nerve Agent Deactivation

            We have demonstrated that the dealkylation reaction can be employed to dealkylate problematic nerve agents and environmental contaminants such as pesticides, both which possess phosphate ester (P-O-C) bonds (#142, J. Am. Chem. Soc. 128 (2006) 1147 and #156, New J. Chem. 32 (2008) 783). For nerve agents such as Sarin a key feature of the reaction is that the phosphate ester is cleaved to leave a covalent B-O-P or Al-O-P linkage in the final product (Figures 4 and 5 above). Thus, the breakage of a single bond allows for the deactivation of the entire chemical weapon agent, obviating the need for further treatment of the by-products. This makes the technology imminently suitable for use in gas masks where the compounds can be introduced adsorbed onto activated carbon, for example. For bulk chemical weapons destruction the dealkylation reaction can be made catalytic by the addition of BBr3, to regenerate the B-Br or Al-Br bonds and release the decomposed nerve agent as a covalent boron or aluminum phosphate (#112, J. Am. Chem. Soc. 124 (2002) 1864). For a review of chemical weapon agent destruction and detection see #165, Chem. Rev. 111 (2011) 5345.

IV. Solid-State Materials

Molecular precursors containing oxophilic metals can be converted to nanoparticulate metal oxides through ambient hydrolyses. This takes place due to the pre-existing metal-oxygen bonds that preordain the conversion of the molecule into the metal-oxide material (Section A). This “soft” materials synthesis technique is being explored for the preparation of nanomaterials with molecular functionality on the surface, and to create a wide variety of “mixed-metal” oxide materials with transition metals (Section B.) and lanthanides (Section C). Additionally, understanding metal-oxygen bonding provided a new means for reducing the oxidation of molten aluminum during secondary aluminum production (Section D).

A. Molecular Precursors to Nanoparticulate Alumina

We have created a new class of tetradentate compounds having the appropriate Al:O stoichiometry for the formation of Al2O3 (Organomet. 18 (1999) 976). The molecules have a tri-diamond shape that looks like the emblem of the Mitsubishi™ Company and so we refer to them as “Mitsubishi™ Molecules”(Coord. Chem. Rev. 210 (2000) 1). The compounds hydrolyze in organic solvents open to air to produce nanoparticulate alumina (#135, Z. Anorg. Allg. Chem. 631 (2005) 2937). In this reaction the Mitsubishi™ molecule acts as a template for the formation of alumina (Figure 8(a)), rather than the hydrated forms of alumina that results when organoaluminum compounds are hydrolyzed.  In collaboration with an industrial sponsor we demonstrated that the compounds deposit a corrosion resistant covering layer of alumina on heated metal substrates under anaerobic conditions.

We have used the nanoparticulate alumina to create an alumina-Pepsin composite material (Figure 7(b)) that is heterogeneous but maintains the activity found in the native enzyme (#126, Nano Lett. 3 (2003) 55). However, the alumina-Pepsin composite is easier to handle and isolate and is more thermally stable than the native enzyme. Alumina nanoparticles with attached biological components could be used in clinical diagnostics, nanosensors, localized delivery of biopharmaceuticals, and biological recognition systems. Furthermore, the “soft” ambient hydrolysis reaction may allow the introduction of the molecular functionality into the precursor molecule before the formation of the nanomaterial (Figure 7(c)). The resulting particle sizes could be controlled through changes in solvent, concentration, and temperature. We are exploring the potential of this methodology for the preparation of metal oxide-biological composite materials with transition metals and lanthanide elements. The materials would combine the biological function with the metal-based magnetic, luminescent, and other unique properties.

Figure 8. Ambient Hydrolysis of a "Mitsubishi Molecule" to Nanoparticulate Alumina (Al2O3) (a), Derivatization of the Nanoparticles with Phosphorylated Pepsi (b), and Derivatization of the Mitsubishi Molecule with Subsequent Hydrolysis and Formation of Nanoparticles with Surface Functionality (c).

B. Mixed-Metal Compounds

Gallium can be introduced into the structure of a Mitsubishi™ molecule (Figure 9(a)). Gallium atoms are slightly smaller than aluminum, due to the d-orbital contraction, and occupy the peripheral four-coordinate sites in the structure with the larger aluminum atom in the central six-coordinate site. This compound could be used as a single-sources precursor to nanoparticulate AlxGayOz materials if it undergoes hydrolytic decomposition like the all-aluminum compounds (Organomet. 18 (1999) 976).

We have also made Mitsubishi™ molecules that have one or two surface chloride groups (Figures 9(b) and 9(c))#155, J. Chem. Soc. Dalton Trans. (2008) 1037). We have combined the halide-containing compounds with transition metal anions with the expectation that a salt elimination reaction will take place to form compound where the transition metal is coordinated by the oxygens of the Mitsubishi™ molecule.  In the partial crystal structure (terminal groups are disordered Cl and Me) of the  resulting compound the aluminum and iron atoms are bridged by oxygen atoms in a lattice that looks like it could be easily converted into the AlxFeyOz solid state material (Figure 9(c)).

Figure 9. Clockwise from top left, Mitsubishi(tm) Molecules with Terminal, Four-Coordinate Gallium Atoms, One Chloride, and Two Chlorides, and the Partial Structure of an Al-Fe Compound.


C. Molecular Precursors to Lanthanide-Aluminum Oxides

Combination of AlMe3 with M(OiPr)3 forms the structurally characterized compounds shown below (with M = Y (left) and Yb (right); (#136, Main Group Chemistry 4 (2005) 3). When the compounds are hydrolyzed under ambient conditions an amorphous solid state material forms. The materials are likely to be nanoparticulate but further characterization has not been attempted. Heating of the solids causes the formation of crystalline Yb3Al5O12 and Y3Al5O12 (YAG). These reactions indicate the possibility that almost any molecular precursor containing oxophilic metals can be hydrolyzed to new or metastable phases of aluminum-metal-oxide materials. The low, possibly ambient, temperatures needed to cause hydrolysis would be conducive to the formation of nanoparticulate materials and metastable material phases.

Figure 10. Cluster Compounds Containing Aluminum and Yttrium (left) and Aluminum and Ytterbium (right) Prepared by Combining Mitsubishi Molecules with Lanthanide Alkoxides.

D. Prevention of Dross Formation in Molten Aluminum

            In secondary aluminum refining, such as the  recycling of aluminum cans (made from Al-Mg alloys), molten aluminum (mp 667°C) is oxidized at a relatively slow rate until a point is reached where accelerated “break-away” oxidation occurs resulting in loss of up to ~50% of the aluminum metal as dross (aluminum oxides). The break-away oxidation occurs when there is sufficient MgAl24 (spinel) present to act as a conduit for oxygen into the subsurface of the molten aluminum. Spinel forms through cation replacement in the solid MgO (mp 2852°C) present in the melt. Literature from several decades ago revealed that adding small amounts of boron (and beryllium) reagents to molten aluminum dramatically inhibited the formation of aluminum oxide (dross). The boron also reduced the amount of spinel that formed on the surface of the molten aluminum. In our own experiments ppm levels of boric acid were sufficient to dramatically reduce the oxidation of aluminum. We hypothesized that the added boron acts to prevent the transformation of MgO to MgAl24 by forming one or more covalent B-O-Mg bonds to the corners of the MgO crystallites. This would account for the ability of such low amounts of boron to prevent “break-away” aluminum oxidation. It would take only eight boron atoms connected to the corners of an MgO cube by B-O-Mg covalent bonds to prevent the entire cube of MgO from converting into spinel (#163, Main Group Chem. 9 (2010) 193 and Book Chapter 11).  This covalent surface coating effect” would also explain how our synthetic dithiol compounds prevented metal leaching from sulfide minerals (# 154, Main Group Chem. 6 (2007) 169).