2D Scans, More Fukui Options, & Max Concurrency
escape from scan-land; making More O'Ferrall–Jencks plots in Rowan; extending Fukui calculations to more levels of theory; managing large numbers of concurrent jobs
We’re excited to be releasing a few improvements to some of our workflows, detailed below.
Two-Dimensional Scans
Although some reactions only involve a single breaking or forming bond, many reactions involve multiple bonds that can change at different points in time. The most common way to analyze multiple changes in bonding is to make a More O’Ferrall–Jencks plot, which shows reaction progress and energy as a function of two reaction coordinates. Here’s the first example of such a plot (from 1970), used to show the different mechanistic possibilities for β-elimination:
Computed More O’Ferrall–Jencks plots are very useful for quickly assessing the viability of different mechanistic proposals. For instance, here’s a plot showing that a nucleophilic aromatic substitution reaction is concerted and not stepwise (as was previously thought):
We’re happy to be adding support for these plots to Rowan. Our scan workflow now supports two dimensions, making it easy to generate More O’Ferrall–Jencks plots at any level of theory: previously, this would have required a decent amount of custom scripting and orchestration. Our scan workflow also supports multiple reaction coordinates per dimension for complex cases, and includes the wavefront propagation algorithm for smoother scan surfaces.
Here’s what this looks like for a simple model system: addition of acetyl chloride to cyanide. Nucleophilic substitution at carbonyls ostensibly proceeds through a two-step process with a tetrahedral intermediate, but in many cases there’s evidence that the reaction is instead concerted (1, 2). In this case, that looks to be true—no stable minimum is visible on the 2D scan plot, and the minimum energy path instead just funnels straight to the product acyl cyanide.
More Levels of Theory for Fukui Indices
Historically, our Fukui index workflow only used xTB for partial charges and global electrophilicity computation. Many of our users have asked us for DFT-quality Fukui indices, and we’re happy to share that this is now possible. When submitting a Fukui index calculation, it’s now possible to choose the level of theory for the preoptimization step and the actual Fukui index calculations. If higher accuracy than our defaults is required, we recommend an xTB or NNP-based optimization method followed by a DFT Fukui index calculation.
Maximum Workflow Concurrency
In response to user feedback, we’ve added the ability to limit how many workflows run simultaneously, making it easier to manage workloads, avoid unexpected credit usage, and optimize how and when workflows are run. Workflow submission still works the exact same way for single and batch submission; this update only affects the behind-the-scenes mechanisms of workflow submission.
You can find and update the max concurrency setting on your account page.