Small changes to molecular structures can transform how reactions occur, but studying reaction mechanisms directly is difficult. An imaging technique that provides direct insights into competing mechanisms might improve matters.
For decades, chemists have picked apart chemical reactions to understand each of the steps involved. Knowing the precise mechanism of a reaction — such as the order in which bonds are made or broken — allows us to predict the outcome, as well as to design molecules and materials, and to discover new chemistry. The determination of a reaction mechanism usually involves piecing together a variety of information from indirect observations, requiring both guesswork and sleuthing. Writing in Nature Communications, Carrascosa et al.1 report their use of a technique called velocity-map imaging to directly visualize not just a single mechanism, but two competing mechanisms.
The authors started by considering a substitution reaction in which group X, attached to a carbon atom, is replaced by group Y (introduced into the reaction as a negative ion, Y−; Fig. 1). The process involved is called the SN2 mechanism, and is one of the most studied mechanisms in organic chemistry. The reaction boils down to a competition between X and Y to determine which can form the stronger bond to the carbon atom. If Y can form a stronger bond, then it will replace X.
The SN2 mechanism was first characterized2 in detail by the British chemist Christopher Ingold in the 1930s. This required considerable detective work and was based on two key observations, which were made for many variants of the substitution reaction under a wide range of conditions. First, the reaction rate depends on the concentrations of the two reactants. This means that both must be involved in the slowest step in the mechanism. Second, the geometry of the product molecule is always inverted in relation to that of the organic reactant. This tells us that the bonds from Y and X to the carbon atom form and break at the same time, rather than in two steps, and also that Y− must approach the carbon atom from the side opposite to X. Meanwhile, the three substituents attached to the carbon atom that are not directly involved in the reaction undergo a geometric inversion similar to an umbrella turning inside out.
Ingold's methods were rather roundabout, but workers from the same laboratory as Carrascosa et al. revealed3 in 2008 that more-direct insight can be obtained by studying the SN2 mechanism in the gas phase. In this approach, the two reactants are prepared as separate beams of gaseous molecules, which are crossed so that the reaction occurs at the point of intersection. Using molecular beams allows both the energy available for the reaction and the velocities of the reactants to be carefully controlled. The directions in which the products scatter can then be measured using velocity-map imaging, and the resulting images provide a direct 'fingerprint' of the reaction mechanism.
Carrascosa and colleagues have now used this approach to study a series of substitution reactions. They began with a simple case in which the iodine atom of methyl iodide (CH3I) is replaced by chlorine. The authors observed that the iodide ions (I−) produced in the reaction are scattered almost exclusively in the same direction as that in which the incident chloride ions (Cl−) were travelling, as would be expected for the SN2 mechanism (Fig. 2a). Moreover, the reaction products fly away nearly as quickly as is allowed energetically, which indicates that any kinetic energy of the incoming Cl− beyond the amount required for the reaction is transformed almost directly into the kinetic energy of the product I− — very little of this energy is dispersed into vibration or rotation of the products.
Having established the fingerprint for an SN2 mechanism, Carrascosa et al. recorded images for similar processes involving different reactants — which is when things became more interesting. For example, when just one of the hydrogen atoms in methyl iodide is replaced with a methyl group (forming ethyl iodide, CH3CH2I), it should still be possible for the SN2 mechanism to occur. However, the authors found that the scattering distribution for this reaction (Fig. 2b) looks nothing like the distribution recorded for the reaction with methyl iodide: the products scatter in the opposite direction, and the most intense scattering is much closer to the centre of the image.
This example highlights a common problem in chemistry: often, more than one type of reaction becomes possible as soon as the chemical complexity of the reactants is increased beyond that of the simplest case. In the case of ethyl iodide and chlorine, the competing reaction is an E2 elimination, in which Cl− pulls a hydrogen atom away from the methyl group to form hydrogen chloride (HCl), and the resulting rearrangement of electrons in the organic reactant leads to the formation of a double bond between the two carbon atoms, eventually yielding ethene (CH2=CH2) as I− is ejected. As in the SN2 mechanism, the process takes place in a single step, with simultaneous bond formation and breakage.
Carrascosa et al. went on to study several similar reactions under different conditions, to investigate competition between the SN2 and E2 mechanisms in detail. Their method of direct visualization allowed them to identify at least two subtypes of the E2 mechanism, depending on whether the reactant halogen ion approached the organic reactant from the same side as the departing halogen ion, or the opposite side.
In common with any gas-phase method, the authors' approach does not provide direct information about how solvent molecules affect the reaction mechanism — in practice, organic reactions are almost always performed in solution. However, if experiments such as those of Carrascosa et al. were combined with studies of the same reactions in solution, the results might provide a new way to separate the solvent effects from reactant effects.
There will also be a limit — yet to be established — to the complexity of reactions that can be studied using the authors' technique. This is partly because interpreting the data becomes more difficult as molecular size increases, but also because preparing large molecules in the gas phase is a considerable challenge. For reactions that are too complex for a full study to be performed, useful insights might still be gained by looking at simplified model systems. Nevertheless, the new work paves the way for an exciting array of mechanistic studies. For example, this approach could be used to study how reaction mechanisms alter in response to steric effects, which occur when bulky groups of atoms block access to certain parts of a molecule, or to investigate whether reactions that are thought to occur through several steps actually do so. Footnote 1
Carrascosa, E. et al. Nature Commun. 8, 25 (2017).
Ingold, C. K. Structure and Mechanism in Organic Chemistry 310 (Cornell Univ. Press, 1953).
Mikosch, J. et al. Science 319, 183–186 (2008).