Hi. My name is JR Cobb. I work as an intern at JMP, and I also work as an undergraduate research assistant at UNC Chapel Hill. Today, I'm going to be talking about how we have used Graph Builder in JMP to help characterize the hydrogenation capabilities of this particular Iridium complex as part of my research group's larger goal of exploring cation-controlled catalysis that is without the presence of lithium, this catalyst is catalytically inactive. Then, if we add one equivalent of lithium, we get a catalytically active complex that can do various transformations.

Previously, this catalyst has been shown to summarize alkenes. But in this particular work, we're going to be exploring how hydrogenation performance varies when we modulate different variables. First of all, we need to understand why we're hydrogenating at all. The real answer has to do with the fact that adding hydrogen across a double bond has a lot of promise for introducing novel functional groups that make molecules a lot more interesting. For example, this pi system right here between a carbon and a nitrogen is called a nitrile.

If we hydrogenate that, we get an amine. We can similarly reduce a ketone to an alcohol, an amine to an amine. We can do a lot of transformations with hydrogenation. It makes molecules that are more functional or have different functionality than they were otherwise, which is, of course, what we want to do in chemistry.

In this particular work, we're going to be exploring the hydrogenation of alkenes, and the vision is for this hydrogenation to be selective. This red pathway on the right hand side shows complete and non-selective hydrogenation. What that means is that this molecule that has both a nitrile and a ketone is getting reduced completely.

We're getting amine and we're getting an alcohol. But what we would prefer, and this has to do with catalyst design as well as the conditions for the reaction is to only reduce one of these bots because that gives us more control over the transformations that we can do. For example, we want to only reduce the nitrile or only reduce the alcohol, but not both.

In this work, we'll also explore ways, that we can make sure that our hydrogenation is selective for the alkenes that we want to reduce and not for other molecules. This summarizes the vision and how we were able to set up these hydrogenations. So, again, here is our catalytically inactive complex, and we want to see no reaction here. This is an example with an alpha beta unsaturated ketone. We want no catalysis without lithium added to the solution.

When we add lithium, this is where we want to see only the reduction of this double bond right here, leaving this carbon oxygen double bond alone. This table briefly summarizes the reaction conditions that we used.

A smaller variant of high throughput experimentation, 18 vials at a time, were used where we loaded the catalyst, the lithium activator, if we use it, a substrate solution, an inhibitor to prevent polymerization, and an internal standard all went into a gas chromatography vial. This was hydrogenated at 10 bar of hydrogen for an hour. Then we monitored reaction progress and results using a proton as well as HSTC and MR spectroscopy.

The issue with this is the fact that we have a lot of things that we want to understand about how this catalyst works. We only knew that this catalyst could hydrogenate styrene. That was the only baseline that we had. We had all of these questions about the performance of this catalyst. This means that we had a lot of experimentation to do.

We wanted to see what are the impact of electronic donating and withdrawing groups on substrates. How does the substitution of a substrate impact the yield that we get? Does the actual geometry of the substrate impact the yield? We had all of these things that we wanted to answer, and this naturally gives us the need for visualization. We ended up with over 70 rows of data, and each one of these rows has a correspondence to an NMR spectrum.

While this spectrum, whether it's 1D proton or this 2DHS QC, it gives us a lot of information. But even once we've compiled the results into a big table, it's really hard to look at this and actually get a meaningful understanding of what kind of hydrogenation this catalyst is capable of. This is really where we employed Graph Builder a lot.

We wanted to create visualizations that would help us get a handle on this data without really having to wrangle through it, spectrum by spectrum. This is a messy, but good initial example of centralizing all of our yield data. The ideal was really high conversion from the alkene to the corresponding alkane. This is basically every run that we did in this particular work.

Immediately, we get some insights. First of all, the size of the circle has to do with the temperature that we perform the hydrogenation at. The little circles are room temperature and the larger circles are 60 degrees Celsius. The color of the circle has to do with the activator that we used. This blue color has correspondence to a salt called lithium BARF, which I'll talk a little bit more about later, and we can keep going from there.

The purple, the green, the red, and the blue, all just mean different salts were added or that no salt was added at all. Immediately, we see that this Ortho-vinyl anisole substrate had really high hydrogenation at room temperature. We know right off the bat that anisole derivatives are gonna be favorable to hydrogenate by this catalyst.

Similarly, at 60 degrees Celsius with the addition of lithium BARF, we see that cyclic alakeans have a really high propensity to get hydrogenated. Whereas some bulkier substrates like this Tetraphenylethylene and Tetramethylethylene. It doesn't really matter if we do it hot or cold. It doesn't really matter if we add lithium or not. These substrates are very challenging to hydrogenate.

This helped us get a handle on what kinds of substrates we should be choosing, but this table is still messy. It's still hard to extract variable by variable impacts on hydrogenation yield. But knowing that, that anisole derivative had a fairly high hydrogenation yield, we wanted to investigate other styrene derivatives.

For example, by adding one methyl group to styrene, which is a molecule here in the middle, we can get alpha-Methylstyrene. This is called the alpha carbon, or we can get Trans-β-methylstyrene by adding it on the other side of that double bond. We were curious what the geometry could impact. What we saw is that styrene has high hydrogenation yields because this double bond of concern is just hanging out without any steric interference. Whereas alpha-Methylstyrene with this methyl group here, was very difficult to hydrogenate, and we only achieved 35% yields. Whereas if we put the methyl group on the other side, we got in the teens percent yields.

This was getting closer to what we saw in our original styrene runs. We also postulated that because a lithium cation binds to the macrocycle that we could get a relatively negatively charged methoxy group to coordinate that lithium and direct the double bond closer to the Iridium center.

We wanted to do an experiment with having the methoxy group right next to the vinyl group and having the methoxy group across from the vinyl group. What we actually saw was a pretty profound difference in the yield of ortho-vinyl anisole versus paravinyl anisole. This was a really interesting study of how the geometry of the substrate all with the same molecular weight in each group still impacted the yield.

Similarly, there are two ways that we can really increase the rate of hydrogenation. That's to heat things up which this has to do with the four substrates that we did at an elevated temperature. This was 2-Cyclohexene-1 own, which is that one that has a double bond and a ketone. This was our selectivity study, as well as the same cyclic alkenes that we talked about previously. What we saw is that save for Tetramethylene, which still had a relatively low yield in this purple, everything else saw higher yields at 60C.

That's how we learned that higher temperatures were more favorable for hydrogenation. Additionally, we knew that lithium BARF, which is short for this long anion here. We knew that that would coordinate to the macrocycle and engender higher activity. But we were curious if a two plus or three plus cation could bind even stronger to the macrocycle, and increase yields even more.

And, unfortunately, using lithium triflate and magnesium triflimide, we didn't see that it was all that much better, than the lithium bar from we actually saw was rather a lot worse with lithium triflate being just as bad as no activator used at all. The final thing that we wanted to do was to use Steremal B1 parameters, which are these quantifications of Steric bulk that we are employing to assess how crowded the chemical environment is around a double bond.

In this case, the higher that the stairwell parameter is, the more bulky the olefin is. Therefore, we postulated it would be the most difficult for this substrate to bind to the Iridium center. This is pretty much exactly what we saw at lower summative Steremal B1 parameters. We saw higher hydrogenation yields, especially with lithium bar for higher temperatures.

As we got on over to the high sixes for the Steremal B1 parameter, we saw really, really low yield. All of these together, were able to be visualized by JMP, and we learned all of these lessons about how this catalyst behaves and hydrogenation schemes. We also have a lot of future directions. We can do combinatorial chemistry using DOE, to make new catalysts that we haven't even been able to make before. We can do screening runs to understand how they perform in hydrogenation.

We can also do principal component analysis by bringing in more continuous numeric variables and really just using JMP to its fullest to help us understand how the catalyst in our group can behave in this new realm of hydrogenation that we haven't actually explored before. JMP is going to make all of that possible, and we're really excited about that.

A really quick thanks to the Miller Group at UNC, which is the lab that this research was carried out in, the University of North Carolina Department of Chemistry and the Mass Spectrometry Lab, where all of these reactions took place. This was all done on a grant from the National Science Foundation, and that's what I got.