Please click on the links below for a brief overview of some of the research projects that the group is working on currently.
Atomic carbon is an exceptionally reactive species. The high energy of carbon atoms, and the many interesting intermediates they produce in their quest to satisfy tetravalency, offer a plethora of interesting and unique reactions that can be studied by experiment and theory. While there are a number of ways of generating atomic carbon, we employ the electric arc technique, which is a particularly useful tool for organic chemists. A photograph of our reactor is shown below.
Our interest in the chemistry of atomic carbon stems from the ability of these species to produce carbenes in their reactions with a variety of organic substrates. Common processes that produce carbenes include (a) insertions into bonds, (b) cycloaddition to bonds, and (c) deoxygenation reactions (Figure 1).
Figure 1: Common reactions of atomic carbon
In one recent project, we undertook the first systematic study of the reactions of arc-generated carbon atoms with acyl chlorides. We noted that these reactions appeared to proceed by two distinct mechanistic pathways depending on the nature of the alkyl group in the substrate. In the case of acetyl chloride (1), we have observed the formation of vinyl chloride (3) which we attribute to the intermediacy of methylchlorocarbene (2) resulting from deoxygenation of the substrate (Scheme 1a). This is analogous to the formation of dichlorocarbene in the reaction of phosgene (COCl2) with atomic carbon as reported by Skell and coworkers. When pivaloyl chloride (4) is used, however, the predominant product is t-buyl chloride (5) and only small amounts of the cyclopropyl derivative (6) is produced (Scheme 1b). While 6 likely issues from t-butylchlorocarbene, formation of 5 is explained by a radical chain reaction initiated by the abstraction of the halogen (Scheme 2). When the alkyl group in the substrate is changed to an isopropyl group, isopropyl chloride (7) and 1-chloro-2-methylpropene (8) are formed in comparable amounts which shows that both mechanistic pathways are competitive (Scheme 1c). This work seems to indicate that as the size of alkyl group increases the deoxygenation process is sterically encumbered and the chain reaction mechanism becomes more favorable.
We are now examining the reactions of atomic carbon with other organic susbstrates such as alkynes and dicarbonyl compounds. In addition to the experimental approach, we are also investigating these reactions using modern computational methods.
Acknowledgements: We thank Professor Phil Shevlin for generously donating time, equipment, and technical expertise for this project, and Professor Murray Campbell and Mr. Charles Jones for their assistance with the reactor assembly. We gratefully acknowledge support of this work by NSF (grant CHE-07719335) and the Colby College Division of Natural Sciences.
Another area of interest to our group is the syntheses of naturally occurring compounds that have biological activity and potential medicinal value. One of our earlier ventures into this field resulted in the synthesis of Montiporynes A and B which are isomeric, naturally occurring marine metabolites found in the hard coral Montipora sp. These metabolites are known to display significant in vitro cytotoxicity against several human solid tumor cells. [1, 2]
We are now working on the synthesis of two compounds (see Figure 1), both of which have been recently isolated from Piper Sanctum (Piperaceae), a plant that is commonly found in south central Mexico. The anise-scented leaves have been used as wrappers to cook a variety of dishes in the cuisine of the region. The plant, which is also known locally as acuyo or hierba santa, has been traditionally revered for its medicinal value. Tea prepared from the leaves of this plant, for example, has been used as a home remedy to combat stomach cramps, coughs, tuberculosis, asthma, bronchitis, and colds. These compounds seem to show significant activity against Mycobacterium Tuberculosis. 
Figure 1: Two of the compounds found in Piper Sanctum (Piperaceae) that have antimycobacterial activity.
We have recently initiated another bioorganic project in our laboratory aimed at the synthesis of cassiferaldehyde (1) and its analogs 2 and 3. Cassiferaldehyde was reported just this year as a constituent of the twigs of Cinnamomum cassia which is widely distributed in southern China, Myanmar, Laos, and Vietnam. It has inhibitory activity against tyrosinase. It has been shown recently that tyrosinase-inhibitors may offer a possible treatment for Parkinsons disease.  Our goal is to synthesize 1, 2, and 3 and determine their x-ray crystal structures. We will also collaborate with biochemistry colleagues to assay the activity of the synthetic analogs 2 and 3.
(1) Bae, B. H.; Im, K. S.; Choi, W. C.; Hong, J.; Lee, C.-O.; Choi, J. S.; Son, B. W.; Song, J.-I.; Jung, J. H. J. Nat. Prod. 2000, 63, 1511.
(2) Speed, T. J.; Thamattoor, D. M. Tetrahedron Lett. 2002, 43, 367.
(3) Mata, R.; Morales, I.; Pérez, O.; Rivero-Cruz, I.; Acevedo, L.; Enrique-Mendoza, I.; Bye, R.; Franzblau, S.; Timmermann, B. J. Nat. Prod. 2004, 67, 1961.
(4) Ngoc, T. M.; Lee, I.; Ha, D. T.; Kim, H.; Min, B.; Bae, K. J. Nat. Prod. 2009, 72, 1205.
(5) Tessari, I.; Bisaglia, M.; Valle, F.; Samorì, B.; Bergantino, E.; Mammi, S.; Bubacco, L.; J. Biol. Chem. 2008, 283, 16808.
It is well known that beta-substituents alter the stability and reactivity of carbenes and various functional groups have been investigated to understand this effect. The chemistry of beta-acetoxycarbene (2), however, has never been studied until the present work. As there appear to be potential complications in the synthesis of conventional precursors to 2 (for example, diazirines and diazo compounds), we synthesized the cyclopropaphenantherene-based, photochemical precursor 1 as shown in Scheme 1. Photolysis of 1 in benzene-d6 affords vinyl acetate (3) as the only product isomeric with 2 (Scheme 1). No adduct was observed when alkenes were added to trap the carbene.
The formation of vinyl acetate raises interesting mechanistic questions, and three different pathways were considered (Scheme 2). An obvious route to 3 is via the 1,2-H shift (red), but it is also possible to get 3 by shifting the acetoxy group, either directly via the alkyl oxygen (blue), or through a five-membered cylic process involving the carbonyl oxygen (green).
To gain some insight into this rearrangement, we generated 2-d1, in which the deuterium was attached to the divalent carbon (Scheme 3). Remarkably, this study indicated that although the 1,2-H shift was the major pathway accounting for about 90% of the product, the remaining 10% was produced by the acetoxy shift (Scheme 2). We then generated 2-d2, in which the methylene group was labeled (Scheme 4). Interestingly, this study indicated that although the 1,2-D shift was the major pathway accounting for about 82% of the product, the remaining 18% was produced by the acetoxy shift (Scheme 2). Thus it appears that a primary kinetic isotope effect plays a substantial role in the 1,2-H(D) shift mechanism.
While the deuterium labeling experiments clearly show that the acetoxy group does move, they do not provide any insight into how such a move occurs. In other words, these labeling experiments cannot distinguish between the alkyl oxygen and carbonyl oxygen shift.To address this issue we have recently prepared a precursor for 2 in which the carbonyl oxygen is partially labeled with 18O. Photolysis experiments are now underway. In addition, we are also investigating the behavior of 2 by modern computational methods. Results are forthcoming.
Acknowledgements: We thank Professors Christopher Hadad and R. B. Sunoj for helpful discussions. We gratefully acknowledge support of this work by NSF (grant CHE-07719335) and the Colby College Division of Natural Sciences.
Most chemistry students learn early on that recrystallization is an excellent technique for purifying solids. While this is generally true, strange phenomena can occasionally manifest themselves during the crystallization process. One such phenomenon, which has long gone unnoticed, but is now beginning to intrigue scientists, is called concomitant polymorphism. In this case, two or more different forms (polymorphs) of a single compound crystallize out of the same solution. This is a rare occurrence as polymorphs typically form under different conditions, and thus concomitant polymorphism has only been reported for a small number of compounds. In our recent investigation into the photolysis of bis(9-anthryl)cyclopropenone (1) into bis(9-anthryl)acetylene (2), we serendipitously discovered that 2 crystallizes as two distinct polymorphs (Scheme 1).
Remarkably, both polymorphs of 2 may be crystallized from the same dichloromethane solution. The photograph of a vial containing both forms of the compound is shown in Figure 1a. One form (2a), which has a distinct orange red coloration and an essentially planar molecular structure, is known in the literature. The other form (2b), which we discovered recently, has a pale yellow color and a twisted structure with the planes of the two anthracene rings forming an angle of about 67°. The crystal structure of 2b and the interesting manner it packs within the crystal are shown in Figure 1b and 1c respectively.
Dilute solutions of 2a and 2b in 1,4-dioxane and hexanes gave the same UV-vis spectrum indicating a common species in both solutions. Both forms of 2 also gave the same 1H NMR spectrum in CDCl3. Calculations using Hartree-Fock (HF) as well as Density Functional Theory (B3LYP) methods, using the 6-31G* basis set, suggest that the twisted form 2b is slight more stable that the planar form 2a. To the best of our knowledge, this is the first example of a pure hydrocarbon crystallizing concomitantly as conformational polymorphs.
Acknowledgements: We thank Professor Rebecca R. Conry for her assistance with the x-ray crystal structures, and the Colby College Division of Natural Sciences for financial support.