Understanding Capillary LC - Part 1: Intro to Scale, Flow and Fit

Capillary-scale liquid chromatography (capillary LC) is reshaping how scientists think about separation performance, solvent consumption, and instrument design. But what exactly defines capillary-scale, and how does it differ from conventional analytical HPLC?

In this first installment of Understanding Capillary LC, Dr. Jim Grinias (Rowan University) joins Axcend's Dr. Sam Foster to walk through the fundamentals of capillary LC, from key physical parameters, such as column inner diameter and flow rates, to the instrumentation and workflow considerations unique to capillary-scale analysis.

We’ll explore the benefits of capillary LC, such as increased sensitivity in ESI-MS, lower solvent consumption, and minimized waste, as well as the tradeoffs and technical requirements for adopting capillary LC in your lab. The session will also touch on real-world applications and emerging opportunities where capillary LC has already been implemented.

Whether you're new to the technology or simply want a clearer understanding of how it compares to traditional HPLC, this session will provide the foundational knowledge needed to engage with capillary LC more confidently.

Key Takeaways

• Understand how capillary LC is defined and how it differs from analytical scale
• Learn about the flow rates, column sizes, and solvent requirements unique to capillary systems
• Explore the pros and cons of working at capillary scale
• Discover current and emerging applications for capillary LC

Webinar

Speakers

Samuel Foster, Ph.D.
Application Scientist
Axcend

Samuel Foster completed his Ph.D. in Pharmaceutical Chemistry from Rowan University in 2025. His research has focused on the development and application of capillary scale liquid chromatography instrumentation. He currently works at Axcend as an application scientist focusing on the development of chromatographic workflows for a variety of analyte classes including oligonucleotides, monoclonal antibodies, and drugs of abuse.

Jim Grinias, Ph.D.
Professor
Rowan University

Jim's research background focuses on the fundamental development of liquid chromatography (LC) columns in capillaries and microfluidic devices. LC columns are at the heart of many analytical separation techniques across pharmaceutical, environmental, and biomedical research projects. His early work focused on the physical structure of the packed chromatographic bed inside a fused silica capillary and led to strategies that could be used to pack more efficient columns in capillaries and also miniaturized microfluidic devices. Other interests have included understanding the physical processes beyond bed structure that impact column performance (included extra-column effects and frictional heating) and applying LC and mass spectrometry (MS) instrumentation to solve analytical problems in neuroscience and molecular physiology.

Transcript

Sam Foster:

Alright. Thank you all for being here. I'm Dr. Samuel Foster. I'm an application scientist here at Axcend. And I'm joined by my good friend and mentor, Dr. Jim Grinias from Rowan University. Jim, if you want to introduce yourself and say hi. Tell them a little, about yourself.

Jim Grinias:

Yeah, thanks again, Sam, for the invitation to join you today for this exciting conversation. I'm a professor of chemistry and biochemistry at Rowan University. I've been here about eight and a half years. And before that, spent some time doing capillary liquid chromatography work at both the University of North Carolina at Chapel Hill and University of Michigan. And thanks everyone for attending today. I see a lot of familiar names in the attendee list, so it's great to see everyone today and excited to have this chat with you, Sam.

Sam Foster:

Yeah, thank you. So for those who don't know Axcend, we are a capillary scale liquid chromatography instrument manufacturer. We're not going to be talking a ton about the Axcend specifically today. We're going to be talking more about capillary scale as a whole. And so I thought it would be interesting to start off with a little quiz for everyone. Don't worry. It won't be, too hard. But really what I want to just see is what everyone's, sort of experience is with capillary scale. So you should be seeing, a, a prompt pop up on your screen, trying to get a sense of, you know, who uses it. Am I talking with experts, or newcomers to the field? Well, we'll give that a couple of seconds to see how everyone goes. Alright. Seems that, the majority of you have heard a little bit about capillary LC. About 60%. And then we have about 22% that use it regularly and 17% that know nothing about it. So we have a pretty good mix. I'm happy to see that. So moving forward, we're going to be discussing what we'll talk about today. Really what I'm trying to give is an overview of capillary scale, understanding not only what it is, sort of the history where we've come from and where we're going. We'll briefly touch on some of the, differences between, say, an analytical or preparatory scale and capillary scale, give you a better sense of it. And then this is going to be an ongoing series. And so, we're going to then go more in-depth into a lot of those topics and differences to talk about instrument considerations, ways that, you know, you can, adapt your chromatography. And then finally, we'll finish off with just some current applications and what's going on in the space right now. So to start off with, chromatographic scale is typically defined by the inner diameter of your column. So, those of you who work with analytical scale, traditionally that means a 4.6mm inner diameter column. Typical, you know, operating conditions are a couple of milliliters a minute. Then you know, over the years, they switch to a three millimeter inner diameter for a solvent saver version. Narrow bore or sometimes they refer to it as UHPLC. That's going to be 2.1mm of inner diameter. Microbore is a millimeter to half a millimeter. And really then capillary scale and what we're going to be focusing on today is columns with sub millimeter inner diameters, typically about 0.1 to 0.3mm. And those are going to be run at a couple of microliters a minute. And so this is when we're mentioning capillary scale, we're talking about columns with, 0.1 to 0.3mm of inner diameter.

Jim Grinias:

And I will note that IUPAC, who is kind of in charge of the international naming of elements and all sorts of definitions in the chemistry world, they don't really have firm, strict definitions of these ideas. And so you'll often hear in the mass spectrometry community, you know, strict adherence to the words nano LC because they're using 75 micron ID capillaries operating at, you know, low nano liter per minute flow ranges. But, other people will use capillary broadly for any column that's packed, in a few silica capillary. So there's a lot of differences here. But typically we're talking about things that operate in the well under a microliter per minute up to, you know, 1 to 10, or maybe even a little higher than that, microliters per minute.

Sam Foster:

Yeah, absolutely. So moving forward, the history of capillary scale, this is a brief summary. You could teach a whole college course on this. I tried to break it down decade by decade and going over, sort of key aspects and advances that were made in each of those. So, really capillary LC started in the late 1970s. And, one of our founders, Milton Lee, was part of the initial, research on that. These columns were typically either open tubular or some, some loosely packed, columns. There, from the 1980s, they started getting more fully packed columns and really starting to flesh out the technology. Another key aspect here that we're going to come back to is that electrospray ionization coupled to mass spec was really demonstrated for its use in large molecules. And, that's going to become a staple of, one of the most common use cases for capillary scale. In the 1990s, capillary columns started to get packed with sub-2 micron particles and operated at extremely high, pressures. And we'll go into sort of why and how that matters. In the early 2000s, we start to see more widespread use in proteomics due to some sensitivity gains that come with ESI at low flows. And we'll go into why that happens. In the 2010s, we start to see some compact and portable capillary scale instrumentation become available. And that's really, sort of where Axcend as a company started. And now in the 2020s, we're looking at, sort of integration into more common workflows, starting to shift away from being exclusively an omics focus technique and starting to become, a more broadly used technique. So moving forward, I mentioned that capillary LC started in, the late 1970s. So this was one of if not, the first separations performed on the capillary scale. This was an open, tubular column. This was done by Novotny’s group, which Milton Lee, one of the scientific founders of Axcend, was in. Really this is just demonstrating, the, the initial phases and the first steps that were being taken. We could see very long separation times. But this is sort of where it all began.

Jim Grinias:

Yeah. So really, the foundational technology sort of came out of the ability to draw these, these very fine tube glass capillaries. So there's some use for them, you know, comparably in, in fiber optic cables when you have a hole in em there. But, I think most people who are familiar with capillary scale separations think of capillary GC columns and kind of coming out of Golay and his work, you know, affiliated with, PerkinElmer. So really the, the, the enabling technology that allowed all of this to happen is the creation of the few silica capillary with very small inner diameters. And that really, you know, instead of coating a, thin layer of stationary phase on the wall, like it's, you know, typically done in GC, basically could you do the same thing either comparably, you know, with an LC, relevant stationary phase or try to pack the same packing materials that we're used to in, you know, larger 4.6 down to maybe 2.1mm ID, columns and putting it down into something that's, you know, several orders of magnitude smaller in diameter.

Sam Foster:

So, yeah, no, thank you very much. So I mentioned that electrospray is going to be, sort of a big deal for capillary LC. The theory behind it is that you can, apply an electric field to a very thin, often, tapered cone of liquid, and that will break the sort of surface tension, allowing for a spray of droplets which then evaporate and get, large and small molecules into the gas phase. And really where capillary scale will benefit from this. And I'll show some examples later, is that when you're running at these smaller and lower flow rates, you get better, evaporation and better ionization efficiency. And so you will often find you get higher sensitivities, especially when you're working in very sample-limited situations. Specifically in the metabolomics and proteomics fields. And so, in the 1980s, this was sort of where it was first introduced and, started to gain some steam in the proteomics fields or in the omics fields. And so I wanted to mention that is this is going to become a foundational coupling technique to capillary scale. even to this day.

Jim Grinias:

And in general, the 1980s really was the explosion of capillary separations in general. So in addition to, you know, the last slide where Sam talked a little bit about, the kind of the very late 70s and the push for capillary LC. So in the early 80s, capillary electrophoresis, was first, demonstrated. And so building off of that, one of the challenges there is its applicability to neutral species. So people have started exploring things like, micellar electro kinetic chromatography. There was pushes for, capillary scale SFC coupled to FID detection. But I think one of the things that really allowed capillary LC to remain, you know, a dominant, capillary scale separation technique was its ease relative to some of those other, mentioned separation modes, how easy it is to couple it to electrospray and get LC-MS analysis. I think it really gives you a really broad, tight sets of analytes that you can, separate and analyze using this technique. So it's really push for, as we'll hear in a couple minutes about omics. But this sort of match made in heaven between capillary LC and ESI-MS is really what's pushed the field over the past few decades.

Sam Foster:

Yeah. Moving forward. And this is one that, I think Jim will be very familiar with. So the one of the benefits of capillary LC, as we'll discuss, is the ability to operate at extremely high pressures to produce very, very efficient separations, because you're able to use very small, often sub two micron particles in very long, you know, meter long columns at very high pressures. You can generate some, extremely high plate counts, hundreds of thousands of plates. And Jim's research advisor, Jorgensen, he was one of the founding members, along with, actually Milton Lee doing some of this, ultra high pressure work. And so, I wanted to give, a mention to that is really in the 1990s. And I'm sure Jim will offer far more history because he was sort of one of the leaders in this field, is, you know, doing these very high pressure separations, we’re able to really push the envelope as to what chromatography can produce from a more fundamental perspective.

Jim Grinias:

Well, yeah. So I think really in general, if you go back to the broad unified theory of chromatography, you know, especially as it relates to LC and thinking about particle packed beds as our, as our, you know, main stationary phase, the idea of pressure being the key limitation. Is what has really driven the technology in the field forward. And it's always a balancing act because as you go to higher and higher pressures, you do face a challenge. And we're going to hear about it a little bit, the generation of heat due to viscous friction. And so you do have to kind of balance the flow rates that you're going at with the pressures you're operating. And so it happens that as you go to really, really high pressures, you need to move down to the capillary scale to be able to maintain your separation efficiency. And so in this case, some of the separations shown here demonstrated go up to 100,000 psi. So well beyond, what's ever been really been commercially available in a HPLC instrument. But what it boils down to in some cases is an engineering challenge, because a lot of these components for pumping at really, really high pressures that existed. But the idea was when you have a capillary scale column, what do you need to do with it? So in general ferrule technology, which really found, you know, is the foundational underlying of how we make our fluidic connections, typically pinches, you know, a tube, whether it's a stainless steel tube, a capillary tube in one spot, when you think about how the fittings were made and they were homemade fittings for these initial demonstrations, it was taking that idea of instead of pinching, you know, a column to hold it in place, basically gripping it and having that hold pressure hold over, a longer distance. And that allowed these, these excessively high pressures, along with all the other metal fittings that were needed for the fluidic flow path. So you can see that, you know, if you go to these extremely high pressures with really long columns, really small particles, you can get close to GC like separation under LC conditions. It was a lot more challenging due to differences in diffusivity between the gas phase and the liquid phase. But, I think this really shows what's possible in terms of commercial usage. Really. The balancing act would never get to 100,000 psi, but they were able to based on some of these earlier results, about give or take eight years later, moved from 6000 PSI is kind of the standard HPLC pressure limit to somewhere in the 15 to 20,000 psi range, as the UHPLC pressure limit.

Sam Foster:

Yeah. Thank you. This is, some of the foundational work, and, Jim, I know you were a big part of that. Maybe not in the 1990s, but, and so here we have, some, some work. So in the 2000s, and really, you know, proteomics and metabolomics had been going on for, for quite some time. But in the 2000s, we really started to see sort of a boom of that. And so for a lot of these techniques we’re able to take some of those very efficient, separation techniques or, you know, not necessarily always UHPLC techniques, but very low flow techniques and perform analysis on, extremely small sample volumes and get, highly sensitive data here. This was just a review article showing off what we can get. And they were able to identify thousands of different compounds using these types of techniques. And so, while this had been going on in the background and it's really just a step in history, the omics fields, sort of were, were the most common use of capillary scale. I think when we were in the 2000s.

Jim Grinias:

And I think really over the past quarter century since the data came out, that's really been the continued commercial driver, of, of capillary LC until recently where, you know, accessibility to instruments has kind of increased. But generally over the past quarter century, it's really been using these columns for primarily proteomics, sometimes metabolomics, and some of the other omics fields, but primarily proteomics, trying to just get extremely high efficiency separations on very low sample volumes and then using really high powered mass spectrometers to do some, some really great detection work as well. So this has been a really enabling technology as it relates to biomedical research, especially in the world of proteomics. So I think this will continue to be one of the biggest uses, although a lot of the things we're gonna talk about today kind of move into the small molecule world.

Sam Foster:

Yeah. So moving forward now, I have sort of this history of compact and portable instrumentation. So not all of these are capillary scale. But really what I wanted to highlight was that, pardon me, starting around, you know, the 2010s to 2020 era, we start to get some more commercially available compact capillary LC systems, and really starting to see, a rise in using these systems, not just in the proteomics fields, but for more routine LC. Pardon me. And so, really, Axcend, the paper you can see there is from 2020 but Axcend themselves started in 2017, 2018 ish in producing their systems. And so really they aren’t the only capillary scale system out there. But in the 2010 to 2020 era is where we see a lot of those begin to be introduced and, setting the groundwork for, for where the field is today. So we're going to skip the 2020s for history, because we're going to be talking about that and a lot of what we're doing in the modern day. So there's going to be quite a bit of that. We're going to talk about later. Now we're going to get into some of the differences that typically are associated with capillary scale. So what I have on your screen now is these, are different column types drawn to scale and maintaining the same linear velocity across those three, at analytical scale, if we were to run at one milliliter a minute, then if we wanted to move to narrow bore and keep the same linear velocity, it'd be 0.2ml a minute. And so then down on the capillary scale we're at four microliters a minute. And so one of the biggest differences, and in fact the one that I think everyone hears and is like whoa is just how much of a solvent reduction you're at. It's several orders of magnitude in terms of total solvent used. And that includes injection volumes. That includes flow rates. So, really it's a dramatic shift in total solvent consumption. And I think that's, sort of the big takeaway that everyone hears.

Jim Grinias:

Yeah. And I think as green analytical chemistry continues to grow in interest and people think about what is the total footprint of our system, it you think about not only the, you know, maybe financial gain of this idea of, of our operating at lower solvents and also maybe because we generate less waste, it costs less to dispose of the waste, something like that. But I think the other thing to think about is when you're moving around less solvents, the, the transportation costs and the and the carbon, you know, footprint that's related to transporting these things. Overall, it really sort of, the entire lifecycle of these solvents really changes when you think about being able to go down to, you know, hundreds of times lower flow rates and still get, you know, somewhat comparable results.

Sam Foster:

Yeah. Moving on as well is when we're operating at these lower flow rates, we're going to also be generating less heat. So as Jim mentioned, one of the fundamental limitations is pressure, but also heat generation. At some point when you're operating at these high enough pressures, your viscous friction gets to a point where it's going to start to negatively impact your chromatographic performance. Whether that be some band broadening because of temperature differences, between the center of the column where, you know, heat is sort of generated and not dissipated and the edges, but also, in fact, some of these pressures, if we were to operate analytical scale columns at these same pressures, the heat generated is enough to take solvent from room temperature to boiling across the length of the column. And so, we really run into some, some fundamental issues there. And, by swapping to these lower columns, less heat is generated. And because less heat is generated, we're able to maintain chromatographic efficiency even when operating at these very extreme conditions to produce, higher efficiency separations.

Jim Grinias:

Yeah. I love this information. I, I frequently teach it whether it's in class or a short course. You know, 15 years of my life could be, you know, summed up in this one slide and I, I love it, but even the one micron is a little bit smaller than what's readily commercially available. Now, I think really, when you think about trying to maintain, you know, optimal efficiency, where's the, the minimum of your of your van Deemter curve or where are you operating at the best conditions as you go from a five micron particle down to a one micron particle? To keep the math easy, you know, nice round whole numbers, because we know as we go down in smaller diameters, we want to increase our linear velocity, the position where that minimum is going to be. We need to basically be operating at 125 times the pressure, to get that best separation. And we're going to see, you know, dramatically improved chromatographic efficiency if we do that. But again, as we go to these really high pressures, we need to consider this heat generation and just the destructive effect it has on our bandwidth, and our peaks will really start having some really weird peak shapes as you go to these, you know, viscous friction conditions. So you can see there as you go from, you know, 24W, you know, heat generation down to, you know, low milliwatt generation down to, you know, small capillary scale. I think that really is, eye opening, you know, quick mathematical back of the envelope calculation on, on what you can do. And, with capillary scale to get these really high efficiencies at really high pressures.

Sam Foster:

Yeah. Thank you. I think that, this one, we we've mentioned it a lot, but I always love this graphic because it really sums up just the differences. So this is sort of a quick summary of electrospray. So this is one of the primary benefits of low flow LC effectively what's happening is we're making these small droplets and then those droplets evaporate, getting your analyte into the gas phase. When you go to these lower flow sense or these lower flow, ESI states, what we see is that the droplet size shrinks and you're able to get more efficient ionization of just your analyte. You're also able to use smaller injection volumes because we've reduced everything down. And so, when you're doing something like single cell separations and things like that, where you're in a very sample limited environment, this is a very key technique. And so we can see the difference there in sensitivity between, nano ESI at 100 nanoliters a minute versus a thousand nanoliters a minute. And it's very, very clear where sort of our analyte peaks are. And so that's one of the primary benefits that's been leveraged, for, for low flow LC and, and something that's been a staple of capillary, for quite some time. Moving forward, another interesting piece of, capillary LC is that we can work with, sort of nontraditional column formats, things that maybe aren't practical at the analytical scale, become practical down at the capillary scale. So one of those primary ones and sort of one of the reasons why they initially started developing it was the idea of open tubular columns, sort of like we use in GC. It's sort of capillary that's coated with a thin layer of functionalized groups. What we see is that because the diffusivity of liquid is much lower than that of gas, we have to go to very small inner diameter columns in order to generate efficient separations. And so if you were to take, say, a 4.6 and run it as an open tubular column, you'd get incredibly poor, if any, sort of chromatographic, separation power. And so really by, by reducing down to these different sizes, we're able to unlock, techniques that otherwise wouldn't be practical or would provide very poor data.

Jim Grinias:

Yeah. I mean, at the end of the day and sometime in the space age and in the future, you know, there's no doubt theoretically, fundamentally, that the ultimate LC separation would be something, you know, a one, a single diameter tube, one micron or even smaller in diameter, you know, very, very thin coating of stationary phase. And, and having it be, you know, tens of meters long, just like we do in GC, the same types of general ideas. I mean, the big difference there is, again, you can get by with wider diameters because you have the gas mobile phase there versus the liquid here. But if we were able to, you know, have a great system that had zero dead volume and extremely high pressures to pump through that, and a detector sensitive enough to see the very, very, very small amounts of sample volume that we would be able to, then we would be able to load onto a column like that. That will be our ideal separation. And maybe decades down the line, we'll get there. But, you know, in terms of what's practically feasible at the at this stage, the packed beds remain, you know, sort of giving us the best of both worlds in terms of getting the amount of sample we need out of the column that matches, you know, our detector sensitivity, as well as, a good enough separation efficiency to, to separate most mixtures that we're looking at these days. We're always going to want more and more higher efficiency. But for what's practical right now, you know, sort of in the realm of we're going to talk about different column dimensions here in a little bit, but sub2 micron, you know, 10 to 15cm long columns, you know, same as we see in analytical scale. I think those kind of things will remain in capillary, just differences in diameter.

Sam Foster:

Another sort of interesting column format. I have a comparison here is a, sort of a pillar array column versus, I have a packed, particle column. And this was some work Jim did for his dissertation. The primary difference here, if we think about the van Deemter A term, is you have, a much more uniform bed, with pillar arrays, you're able to, you know, generate a very, homogenous bed. However, we aren't able to really operate at high flows. We have pretty limited, loading capacity. And so, while pillar arrays are great in terms of performance and reducing that A term, we do have to operate them at very, very low flow rates because of current, you know, manufacturing issues and loadability issues. And so, it offers some efficient chromatography while at the same time, you know, can really only be operated efficiently on the capillary scale. What we can see here, I mean, looking at the homogeneity of the different pillars, they all are, you know, almost identical. And we're able to make a very uniform bed versus if you look at something like a packed column over there on the right, you can see that there's a variety of different, particle sizes and, and sort of voids and not a perfectly homogenous bed like we'd see with pillar array. Another, sort of interesting format is that of monolith columns. So, this it's a little bit different from sort of that packed column that we saw and also a pillar array instead. This is a porous membrane that then gets functionalized. And so, those pores you have a little bit higher homogeneity, or a little bit higher heterogeneity. The pore sizes aren't always uniform. And you'll actually start to see some, slightly decreased performance from that of a packed bed. However, really what the key difference is, is that you see much lower, sort of back pressure generation and lower, pressures needed to operate. And so, monolithic columns can get pretty ridiculous in terms of length. I think the largest commercially available one is, two meters long. And so it's a stark difference from sort of your five, ten, 15 centimeter columns is that, you're able to you know, extend these out incredibly long to get really, really efficient separations, with, with reduced back pressures.

Jim Grinias:

I think this is another great morphology. In addition to the pillars mentioned on the aforementioned slide, I think the big thing is, as Sam mentioned before, how many are readily sort of commercially available. So a lot of times people doing work with monolith columns are generating their own columns. That can be a little bit of a trickier situation. And also generally, especially if you are buying the columns, what stationary phases are readily available. We'll talk about that a little bit going on. But in most cases you're limited to, you know, standard reverse phase operations and maybe in a couple niche cases being able to do some, some HILIC separation of polar compounds, but you don't get the quite the breadth of, of phases that you might see with, with a typical packed bed column.

Sam Foster:

So here, one of the key differences, and this is really going to be the topic of, our next webinar, So stay tuned for that. We'll be posting details soon, on registration dates and that, but that of solvent delivery. So when we're reducing these flow rates, it's not always a direct, 1 to 1. You can't just really take an analytical scale instrument and tell it to operate down at these, sort of, you know, low, low microliters a minute range. It often requires some, some very careful considerations. And so, the two most common separation or, solvent delivery types are that of syringe based or piston based. In piston based separations, that's more akin to sort of your traditional legacy LCs and analytical scale LCs. It typically operates by having two reciprocating pistons, out of phase with each other so that while one piston is filling, the other is dispensing solvent. And that leads to a pretty stable flow generation. There are some slight fluctuations with that if they aren't perfectly in phase, and that can lead to dips in your baseline. When we're looking at really low flow, you know, microliters a minute, these pumps have to actually be optimized for that by reducing your stroke volume, implementing pulse dampeners and sort of, updating it to limit the fluctuations of those pulses in order to, provide more stable low flow. On the other hand, and this is what's used for the Axcend, a syringe based system is a little bit different. Rather than having two pistons out of phase, we effectively have one piston that starts the run by filling itself. And then we'll stop, change the flow over and dispense the flow in a continuous fuller burst. And so this is called pulseless, oop pardon me, this is called pulseless flow. It's not necessarily pulseless in the sense that, you still will see slight fluctuations in your solvent delivery due to the steps of the motor. But it's considered pulseless in the sense that it doesn't necessarily need, the same pulse dampeners associated with piston based pumps. And so that can reduce your overall dwell time, and sort of allow you to produce, a little bit different data.

Jim Grinias:

So the, the really high pressures we talked about a few minutes ago, those use these really big pneumatic amplifier pumps that, you know, are kind of piston based in general, but in terms of solvent delivery, without having a refill step, you kind of want to stay within one, piston volume. So it in terms of designing the method, you want something that's a little bit more comparable to a syringe pump. Both of those. You're also because you're pushing one liquid, you're somewhat a little bit limited to isocratic separations, but you can it's a lot easier with syringe pumps than with the pneumatic amplifier pumps, basically have two running together and a small mixing valve. So when you do see some pulses, you know, at really low flow rates, it's pretty minute. And again we'll talk about this in on in later webinars. So not as big of a deal now but really my new due to motor steps I think the bigger thing is you know making sure you get good mixing and that's where you might see a little bit. But we see, you know, potential issues with mixing as it relates to a lot of different chromatographic, separations and different instrument platforms.

Sam Foster:

One of the key drawbacks of capillary scale, and this is something that, we'll be, we'll be discussing in future webinars in a little bit more detail is, that of extra column band broadening. So in analytical scale or higher scales, it's not so much a problem. It's still experienced there. But if you had, say, one microliter of, void volume, not doing much in your system, but you're running at a milliliter a minute, it's not necessarily going to have, a massive impact on your system. If you have that same, one microliter of void volume somewhere, maybe a loose fitting, but you're running at one microliter a minute. Well, all of a sudden, now that's an entire minute of time where your peak’s broadening you’re not necessarily getting, you know, optimal separation efficiency. And so, this was just showing off some of the, common ways that peaks can broaden, due to, you know, injection variations, tubing variations, flow cell variations. And so, we're going to be doing an entire, you know, one of these webinar installments on, considerations for that. There's ways to mitigate it using zero dead volume fittings, face sealing fittings, reduced inner diameter tubing. But, really, this is a large consideration is that when you go to capillary scale, you really have to optimize your system for it. And so that that's something that, we'll be taking a deep dive into to, to discuss how we actually go about doing that. Additionally, so we do see increased sensitivity when we're performing electrospray, when we go on to the, capillary scale and especially in, you know, sort of UV detection and these other, detection methods, we tend to be limited by, the path length of the column. So, as we discussed with sort of the, the extra column band broadening any additional volume that you're going to add to your system is going to start to broaden your peaks a little bit. Well, adding a detector is adding volume. And so it's a careful balancing act between how much volume do you add to your detector. Versus, you know, how much, do you negatively impact your chromatographic performance. And so, there were there are certainly concerns with detector sensitivity. There's ways to mitigate it and in fact perform, very sensitive separations even without, mass spec. But, it's certainly something that needs to be optimized specific to the system.

Jim Grinias:

I vividly remember the collection of this data. Sam was still a student here at Rowan University, and Elisabeth, one of his colleagues at Axcend was working with us as well on, on this data. And, oftentimes in research, you get really excited when you see what you expect to see. And Beer-Lambert Law, which, you know, describes how absorbance is related to molar absorptivity and path length of your of your flow cell and the concentration of a given species. When we multiply the path length by eight and keep everything else the same, we should expect eight times increase in signal because of the extra column effects that that Sam mentioned before. We saw 7.8. So really close. But getting exactly what you expect sometimes that you know is a great thing to see. And I remember being really excited when we, when we saw this that, you know, a simple change, but gave us exactly what we were looking for to see that, you know, big increase in signal, which helps a lot of, you know, in cases where you might be a little bit more, closer to a detection limit.

Sam Foster:

So one of the, the drawback this is becoming less and less so, but one of the primary drawbacks of capillary scale is it's still sort of gaining commercial viability and seeing a surge in commercial offerings. And so, there will be cases especially, you know, when you're, you're getting into it where if you're trying to translate a method, the column or the phase or the, the functionality that you're looking for may not always be, exactly available on the capillary scale. And so, there are some great methods for, for solving that. We, recently published a paper going over, sort of a summary of all the commercial offerings. But if what you're looking for isn't offered, they have a lot of alternatives. In fact, there's something called the column selectivity database run by, a good friend of mine, Dwight Stoll. And what you can do there is by using the hydrophobicity, hydrophobic subtraction model, you're able to input your column and see columns with similar, similar characteristics and similar functionalities. And so, even if your column isn't directly available, you can find a column with, similar or sometimes better functionalities and selectivity, for the analyte you're looking for. And so, it's, it's, a way of figuring out how to translate your methods to something else that, you know, will work. So taking a look at sort of what's going on in the modern day and how we're putting this to use now, this is a lot of work I did with Jim here at Rowan for, my, my PhD. It's being used all over the place in routine separations. Here we worked with, another group, for biopharmaceutical analysis, in which we were able to perform, a number of critical quality attribute analyses on, trastuzumab where we were able to see the intact, fragmented and, reduced, versions of this Mab. So these, would be sort of comparable to separations used in the pharmaceutical industry to verify that, you were able to, to produce the monoclonal antibody with the same, sort of bio similarity to what you're looking for.

Jim Grinias:

Yeah. I think the one thing I want to note on this is this was a demonstration of hooking up a compact capillary LC system directly up to, to a large mass spectrometer. But to get the really high resolution you need for some of these studies, those are the mass spectrometer you're going to be able to go with. But, very little was needed to do to modify the system. A little bit of excess broadening was, was seen because we used the standard source that's a little bit more compatible with analytical scale separations, but it was compatible with the flow rate. So basically you just had to kind of change one tube and just change, you know, where the LC eluent flow was coming from. And we're able to get, you know, really high mass spec resolution results and the results as it related to a 2.1mm ID column were almost identical. So I think this really shows, you know, a great use case where you can kind of stick a small LC in front of a big mass spec and get the results you're typically looking for. And I think that that's really like the key thing is and a lot of these, but a lot of these pharmaceutical and biopharmaceutical applications, you know, you're not at like really trace levels of analyte. You know, sensitivity is never really our challenge is just trying to maintain, you know, a great separation efficiency and then let the detectors handle some of that. And I think this is a great way we can do them. Yeah.

Sam Foster:

Another interesting use. And this sort of leans into the reduced form factor and compact, portable nature of some of these offerings, and especially the Axcend offering. We were able to do a panel of a number of drugs of abuse, and their you know, metabolites for, sort of in-field testing. So if you were at a crime site and needed to test the sample, and this is some ongoing work, we're actually, working on, on doing some really interesting stuff with that. So stay tuned. But, these were a number of different, relevant, illicit drugs, the benzodiazepines, cannabinoids, methadone, cocaine, opioids, heroin. And we were able to perform all of these at microliter a minute flow rates. And we were able to perform all of them in under about five minutes. So they're, they're it's not only efficient chromatography, but it's also portable chromatography. And I think that's another, really interesting case is being able to actually bring the LC to the point of need rather than, bringing the sample to a lab.

Jim Grinias:

Yeah. And this is ongoing work. Some of our current team members, John Angelina, continue to work on this. And I think that's one interesting thing we're looking at is, as we think more about real samples, things like urine or blood, being able to do some of the sample cleanup, one at a really small volume scale and but two in a fast and online format, you know, might enable, you know, complete analytical workflows that, again, generate less waste, but, you know, might provide a needed analytical solution, right, where the samples are being collected. And so it's kind of a continued dream of mine. We'll continue to work towards that. But definitely in terms of coupling sample preparation at the scale and automating that and then coupling directly to this instrument. As Sam mentioned, we’ve got some good stuff cooking right now, and we'll continue to work on that. Hopefully report on it sometime in the near future.

Sam Foster:

Yeah. Another really interesting application is that of, sort of automated reaction monitoring. So for, automated reaction monitoring, typically you want to make sure that you're, you know, within, a nearby location, you don't necessarily want to have long, durations between when you say withdraw and sample from a reaction flask and analyze it is it's not fully representative of the bulk solution. Additionally, you don't always want to be taking these reactions out of, say, a fume hood. They might be generating something hazardous. And so, but you also want to limit the amount of volume you're taking from that reaction. That way you aren't necessarily interfering with how the reaction is progressing or changing any conditions. And so, for that, the Axcend also has sort of capillary scale, sample handling systems that can perform filtration and sample introduction. Here we have a reaction. It's just an imine condensation where we take an aldehyde and an amine and mix them together to get an imine, you can see a representative chromatogram there. But looking at it over time, and this is really going to show where the power of the technique lies. Looking at it, over time, we were able to take, this reaction, set it up and monitor it about every 15 minutes over the course of 3.5 hours. And due to the, reduced sample volumes and the reduced solvent consumption, we were able to perform these separations over the course of 3.5 hours, taking under one milliliter of total solvent, across the entire analysis. So that's, mobile phase consumption, sample consumption, all of it, is under a milliliter for 3.5 hours of continuous operation. And what we can see, here is we get to watch as our product peak grows, as our reaction peak decreases. And so this can be used not only for monitoring and screening reactions, but also for, reaction optimization, as we can plot the kinetics and determine ways to, to optimize these reactions without committing to, large volume and large scale studies. Another one moving a little bit away from the Axcend and sort of, Jim’s dream and a little bit of my dream for the future is, methods of high throughput analysis using the capillary scale. And so because we're operating with these smaller inner diameter columns and we're trying to generate very high linear velocities to perform very fast separations, we're able to, say operate, you know, sub 10s separation times while still only taking microliters a minute, whereas, on the on the higher scales you'd be looking at, you know, 4 or 5ml a minute. And so the solvent consumption can start to get, pretty prohibitive, in terms of its expense. So here we just had an OEM pump, an injector valve. And then we're using the Axcend detector taken out of the box. But we're doing a pretty standard separation here. It's just a separation of thiourea, propiophenone and valerophenone isocratically. But by taking the column and cranking the flow rate to about 80 microliters a minute, we're able to get 4 second separations where we see, resolution between these three different components. And so this, you know, marks the beginning of, method for, for high throughput analysis with, low environmental impact.

Jim Grinias:

And I think even beyond the environmental impact, when you talk about these really high throughput separations, you need to if you're using analytical scale columns, you're typically operating at, you know, 3 to 5 or even if the pump allows for it, a higher milliliter per minute flow rates. And those aren't necessarily great in terms of compatibility with ESI for mass spec detection. The flow rates needed for these this kind of work is, you know, in the 50 to 100 microliter per minute flow range even at capillary so well beyond, you know, the flow range that we've talked about before, but something that's a lot more compatible with ESI-MS. So as you think about being able to do these high throughput separations, capillary. This is another spot where capillary scale you know can really shine. And so even though maybe you can go a little bit faster with modern instrumentation at the analytical scale, the ability to run these in parallel when you have capillary instruments and columns is great too. So even though this is on the order of 4 to 5, second separation, you know, we've done some things where we've run them, you know, 2 or 4 at a time in parallel. And so you can kind of get closer to that overall, you know, one second analytical duty cycle time, for these separations. So, really there's a lot of things needed. Better instrumentation, better columns, you know, with smaller particles, you know, or other column formats, that run at these really high flow rates. But definitely, I think this is a direction we're going to be spending a lot more time working on in the coming years as we try to really see where does LC fit in the world of high throughput screening? Not on the minutes scale, but on the low second scale.

Sam Foster:

Going forward with that, when you're operating that quickly. And so this was some work we did with our collaborators, at the University of Michigan and Bob Kennedy’s Group. When you're trying to analyze these samples so quickly, modern auto samplers, don't always have the cycle time to accommodate these, you know, low 4 or 5 seconds separations. And so, one of the methods for that is utilizing the low injection volumes of capillary scale by using droplet based injection systems where, you take an aliquot of your sample and, then move to an admissible oil, and that allows you to continually be withdrawing, skipping some of that cycle time, needed to say, dispense the syringe, move it over and fill up another syringe. You can just continually be sampling to form these small droplets and aliquots of sample. And then by toggling the injection valve, only when these analytes are actually in the, injection loop, we're able to, to perform very high throughput, separations. And so here, this was a, reaction monitoring work that we did, whereby, we screened a number of different reaction conditions for the reactions you see there in the bottom left. And this was, I believe seven second, separation times. And so, you can see there we're able to screen an entire 96 well, plate over the course of a couple of minutes. Just because we're able to, to push sample through very, very quickly, compatible with ESI and, you know, very sensitive in terms of being able to detect these, compounds as they're formed. So to wrap things up, we've really started to see capillary scale in the, in the 1970s, and it's only really grown from then. Typically capillary scale. The definition shifts, as Jim had mentioned. But typically we consider it anywhere between about 75 micron to half a millimeter in inner diameter. The reduced heat formation allows for, for very high pressure separations to be performed. The lower flow rates allow us to use the nontraditional column formats for, for, interesting and often, useful separation methods. The, the low flow associated with capillary scale offer sensitivity gains. And also just the low flow reduces the amount of solvent consumption. And so, it's, it's better for the environment as well.

Jim Grinias:

I was just going to chime in as well. Sorry I came off of mute. I think in general, you know, there are going to be some certain cases where standard technology, standard methodology, you know, can be used. But I think that there are a large number of chromatographic separations that the mixtures are not extremely complex, but you do need to have a separation step to help. And so there are ways we can try to speed those things up. They're not necessarily, you know, extremely, you know, low concentration samples. And so I think the capillary scale instrumentation we have available to us now makes a large number of separations that people have been running for various analytical purposes over the past few decades. A lot of those things, the technology is at a point now where we can use these low flow separations, get comparable results to what has been seen. And so for sort of a fit for purpose case, I think there's a lot of use case scenarios where capillary is the way to go. And that's something that, you know, I want to continue to, to kind of push for, with broadly within the field of analytical chemistry, is that really I think the time is now to continue to push this forward.

Sam Foster:

Yeah. With that, I, I have a references here. You'll see them cited throughout. And so, if anyone has any questions or wants to read more, we have those offered. And with that, thank you all for coming. I'm going to open the floor for questions here. But, if you have further questions or we don't get to yours, please feel free to scan the QR code. Visit us at our website. You know, reach out. I'm always happy to discuss this, and, Yeah. With that, I'll be looking in the chat for any questions people have. Thank you very much for attending. We thank you, Jim, for, taking the time out of your day to come speak with us.

Jim Grinias:

Yeah. No. Happy to chat. Just for the kind of the purposes of the audience Sam and I have been working together on various aspects of capillary LC for, you know, probably about seven years now, maybe longer and closer to eight years. So it's been really fun. And I think it's cool to see where some of the previous work is going and excited for where we're headed now. And I'm going to maybe chime in because I've seen a lot of the questions start flowing. And so there's a quick one that's sort of definition related about viscosity. So what's the units of viscosity? Poise more commonly centipoise with typical solvents that we're talking about. A poise is a Pascal second, I believe formally, yeah. Or maybe 0.1 Pascal second. So they're closely related to pressure and time, but I think the bigger thing, how does it relate to fluid resistance and flow. Obviously as viscosity goes up. If we flow it at a standard flow rate, as the viscosities goes higher, the pressure we need to flow at that given flow rate’s going to go higher as well. And so I think the big thing is generally as we think about designing gradient methods, we oftentimes see the pressure trail off as the amount of organic in our, mobile phase increases during the gradient. So, generally, acetonitrile and methanol, lower viscosity than water. And kind of as you have these mixtures of them, tends to be around 20% acetonitrile and water is kind of the max pressure for that one. And maybe closer to 50/50 mix of water, methanol, max pressure there. And then as you go, it goes down. So yeah.

Sam Foster:

What are the key considerations for selecting between different capillary column formats such as monolith, open, tubular, and pillar array. All right. Well there's a couple. So first and foremost you got to make sure that you're scaling your injection volumes properly. Depending on, on the even. And you see this even with standard packed bed format, you have to make sure that, that you're scaling your injection to, to not overload the stationary phase. So, you know, you have to typically reduce or increase it depending on the format you're using. You also have to scale your, your flow rates and pressures to that of, that can be accommodated by those formats. If you were to say, take like a pillar array, which, you know, doesn't always have the highest pressure input, 4000 bars through it like we've seen in some of these packed beds, it probably wouldn't end very well. And so, reading up on, on sort of the, the operating conditions and making sure that, you're, you're scaling your, your system and your conditions in order to, to stay within those.

Jim Grinias:

Yeah. And I think I'll just chime in as well. A lot of it is, you know, what's what specific stationary phase coatings are available for, for these different formats. You know, if you're if you're not going to make your own columns, what's readily available. And I think when it comes to readily available, you're looking primarily at the packed beds and pillar arrays. And I think the pillar arrays are great for a lot of proteomics applications. But, you know, for kind of the breadth of phases that are available, particle packed beds, you know, sort of sort of continue to, to give you the, the breadth of separation modes you might be interested in exploring.

Sam Foster:

Yeah. The next one here is how do you mitigate challenges of extra column band broadening and capillary scale LLC? So we'll be doing probably, an entire, webinar on that. So, so look out for that. Really the answer is reduce your extra column volume. That typically comes in the sense of, reducing the inner diameter of your connective tubing, but also reducing the length of your connected tubing, trying to get things more localized. There's also specific fitting types which offer, face sealing, which, which reduces, any potential for void volumes. And really just making sure that, you're optimizing your fluidic setup and not necessarily leaving gaps, in there. What trade offs exist between improving detection sensitivity and reducing path length on capillary, detection? Oh, in on-capillary detection.

Jim Grinias:

Yeah I think okay. So on capillary really relates to kind of the same points as related to extra column volumes that Sam was just talking about. If you think about on capillary detection, where there isn't really a flow cell, but it's, it's you're measuring the absorbance across the tubing diameter right directly adjacent to the, the outlet of the column. It's going to be bigger volume, higher broadening as that diameter of the tube increases. So you're going to see, you know, increasing bandwidth. But your path length will be higher. So you'll be able to see a lot more. So it's really an whether it's a flow cell or on capillary, That's really the balancing act. What is the volume of where the detection is taking place versus you know, what is the path length of that for absorbance detection? Obviously these things change as we talk about mass spec or other detection modes. But that's kind of primarily the balances, you know, trying our best to maintain peak shape. Versus how much volume does the flow solve? And I think that's one nice thing where these 0.3mm ID columns, it's a nice balance between, the column volume and its relation to the volume of the extra tubing and other components of the system versus, you know, what path lengths can you get away with? It's a little bit tougher. And that's why I think, again, you see primarily Mass Spec as the sole detection mode. Maybe a little bit electrochemical or some other, you know, niche techniques. When you go down to like 75 micron or smaller diameter columns. So that's really a big thing is what's the balance between the volumes and the path length.

Sam Foster:

Yeah. I think we probably have time for one more question. Is capillary chromatography sample preparation methods the same as with analytical chromatography? Yeah, pretty much. I mean, you won't need as much sample, typical injection volumes that we use on the capillary scale or in the 40 nano liter range. That being said, typically I make up my samples, you know, in one mil HPLC vials that are, traditional to sort of the higher scales. And so, there isn't necessarily a fundamental change in any workflow. You're just going to end up using less of that sample you prepared. And in fact, for, for some really interesting stuff, we've been working on some, solid phase extraction methodology on the capillary scale, that I think is really cool. So be on the lookout for that.

Jim Grinias:

Yeah, that was just kind of what I was going to chime in as well that, you know, in cases where you're not sample limited, like someone like a clinical sample of, of urine, for example, you're generally not going to have, you know, too much of an issue in terms of the, the collection volume. And so that being the case, depending on how much you load on the thing, if you use more standard workflows, an additional potential advantage, even if it takes a little bit longer, you can get some added pre concentration, you know, a little bit higher pre concentration than you might see on the analytical scale because you're injecting typically lower volumes onto the column than you are the bigger columns. You can sort of resuspend a dried down sample in a lower volume. And so you get a little bit of an advantage there. In sample limited cases, it really depends on, you know, how are how are you able to manipulate those very, very tiny sample volumes. So online is one way. And there are, you know, kind of people are developing different robotic approaches to that as well. But I think that that in general, that's going to be kind of the big things. And then another question came up about trap columns for capillary scale. They exist a little bit in the, in the idea of, of a trapping step, where maybe you're trying to, to load a lot of column and elute in a very small band. Really what they are oftentimes used for is proteomics applications where you're trying to get, you know, a lot of sample kind of concentrated down to a smaller sample band. So whether that's offline or online, but yeah, proteomics applications, it has really been the, the kind of big application for those.

Sam Foster:

Yeah. No thank you. That was an excellent answer. With that, I think we're coming up on our time here. If you have any other questions or want to reach out and learn more, please feel free to check the QR code or head to our website. And, thank you once again, Jim, for coming out.

Jim Grinias:

Thanks for having me, Sam. And thanks for everyone attending.

Sam Foster:

Yeah. And, make sure to tune in next time where we'll be talking about solvent delivery systems. Be looking for the details soon. Thank you everyone.