By Gina Shaw
Milestones of gas chromatography in pharma
Editor’s Note: This article on the history and impact of gas chromatography in the pharmaceutical industry is the fourth article in a new series for Pharmaceutical Formulation and Quality. In “PharmaTools: Technologies That Changed Pharma and Biotech,” we look at various technologies such as gas chromatography that have played a key role and had an indelible impact on the pharma and biotech industries. In our next issue, we will examine the evolution of formulation technologies.
Dave Johnson, the veteran product manager for gas chromatography (GC) at Agilent, paints a vivid picture of the early days of gas chromatography technology. “Think of Grandpa Moses with his backyard still, doing his separation to distill the alcohol from the rest of what he’s mashing,” he said.
“That’s what the chemists in the 1950s did with this new gas chromatography technique. It was like in the old movies, with a boiling flask that has a coiled glass tube that starts dripping into another flask.”
Gas chromatography, like all forms of chromatography, takes its name from the Greek word chroma, meaning color, and graph, meaning writing. Russian botanist Mikhail Semenovich Tsvett used the word chromatogram to label his technique for separating different types of chlorophyll by trickling dissolved plant pigments through a tube packed with calcium carbonate powder. Different pigments stuck to the paper, with different strengths creating a series of colored bands.
As it turned out, color was not needed for “color writing.” Chromatography could also work to perform separations on colorless substances. In the early 1950s, two British biochemists, J.P. Martin and A.P. James, developed the idea of vaporizing the sample instead of just using a liquid.
“It got more sophisticated with different devices that could achieve separation after heating,” Johnson said. “The chemical companies, like DuPont, had glass blowers on staff who would work with scientists to identify what kind of separation was needed and to create the different configurations of glass that were the actual earliest gas chromatographs.”
Early Commercial Gas Chromatographs
Many believe that the first commercial gas chromatography was introduced by Perkin Elmer. According to chromatography pioneer Harold McNair, PhD, however, the first was actually the VPC Gas Chromatographic Apparatus. That apparatus was developed in 1954 by the now-defunct British company Griffin and George Ltd.
By the early 1960s, there were four main players in gas chromatography: Hewlett-Packard, Perkin Elmer, Varian, and Shimadzu.
Dr. McNair, who was instrumental in the start-up of the chromatography divisions at both Hewlett-Packard and Varian and also wrote the GC bible, Basic Gas Chromatography, built his own gas chromatograph as a graduate student at Purdue in 1957. “Nobody knew what the thing was,” Dr. McNair said.
But “the thing” turned out to be ideally suited for the pharmaceutical industry. “In the sense of separating residual solvents, pharma is one of the easiest things to do with gas chromatography,” Dr. McNair said.
Still, it had its limitations. “Pharmaceutical components have to be soluble in salt water. Gas chromatography in its early stages was too active for varipolar compounds,” Dr. McNair said. “When they started making fused silica capillary columns in the early 1960s, that was much better suited to pharma. But Perkin Elmer, the first developer of those columns, kept the patent so tight that no one could use them. A few years before Perkin Elmer’s patent ran out, Hewlett-Packard and Varian began promoting inert gas chromatography columns, which could handle pharma’s needs much better.”
A small company had also formed during that same time period in Wilmington, Delaware, one that would eventually become Agilent by way of Hewlett-Packard. Frank Martinez, a glass blower at DuPont who had worked with some of the earliest GC scientists, asked the company if he could make gas chromatographs at home and sell them back to DuPont and to other companies as well.
“So the first gas chromatographs made by Agilent were Frank Martinez’s while he was still employed at DuPont,” Johnson said. “He got so busy that he started to form his own company, F&M Scientific. Their first customer was the National Institutes of Health.”
In 1965, Hewlett-Packard took over F&M to extend its testing and measurement capabilities in chemical analysis; in 1999, nearly 25 years later, that division spun off to become part of Agilent Technologies.
Linear Temperature Programming Breakthrough
F&M’s key contribution to the field of gas chromatography, brought to market in the early 1960s, was something called linear temperature programming. “Instead of just boiling the flask with the liquid you’re trying to separate at one temperature, they found a way to increase the temperature over 20°C per minute, giving them more resolving power and the ability to get better separations of different components,” Mr. Johnson said.
“That was a big step forward,” he said. The first GC with linear temperature programming was dubbed Model 202, named after one of the main roads running through Wilmington.
“That’s really what ultimately made GC useful for pharma,” said Dr. McNair. “Inert fused silica capillary columns [which came later] combined with temperature programming, because pharma products are a little more higher weight than petroleum gases and light solvents, and usually more complex.”
The European manufacturers, however, were slow to adopt linear temperature programming. “By the time they started, around 1968, Hewlett-Packard was number one in the world for gas chromatography in terms of gross sales,” Dr. McNair said.
The first digital integrators for gas chromatography were developed in the late 1960s and early 1970s. “The main goals of gas chromatography are to identify what your compound is and say how much is in there. With linear temperature programming, if I’m starting at this temperature, say 200°, I ramp it up at 20°; if I get a peak on my graph paper where the signal goes up at a certain time, I can say it’s probably X compound.
“But what about how much of it? The way you answer that question is to take the signal plot on a piece of graph paper. The area under the curve of the signal is related to how much of the compound is in your sample,” he said.
Before the advent of the first digital integrator in 1969, scientists would literally take a pair of scissors and cut out the area under the plotted curve on the graph paper and then weigh it. “The only other option was to count the squares on the graph paper, which was too tedious,” Johnson said. “Digital integrators allowed you to mathematically calculate the area under the curve in real time and much more accurately.”
In the 1970s, Johnson said, scientists concentrated on making GC instruments that were more robust, rugged, and reliable for use by non-experts, the people who weren’t trying to do gas chromatography for a living but were making products and wanting to use GC as a tool. That included pharma.
In 1971, HP introduced the first in-house-designed GC instrument, the 5700. “It was for routine use,” Johnson said. “In 1973, they introduced the 5830, the first gas chromatograph controlled by a microprocessor, which allowed you to put in more features and interact with better data handling.”
A Big Step Forward
In 1979, an enormous step forward for GC and chromatography in general was introduced: fused silica capillary columns. “Up until then, all we ever had were glass columns. By then they were smaller than in the 1950s, but we were at the limit of how much separation they could really do,” Johnson said.
Invented at Hewlett-Packard, these much smaller columns had a phase coating on the inside that would facilitate the separation.
“This gave orders of magnitude of separating power to your sample. With a very complex mixture, you had never been able to see all the different peaks. Instead, you would just get a blob if you had something complex,” he said. “Now, you could separate complex mixtures much better and be able to see the individual peaks.”
Pharmaceutical companies were among the last industries to adopt this, because they did not want to make changes in all their methods. Finally, they recognized the huge impact this would have on their field, according to Johnson.
With the advent of fused silica capillary columns, gas chromatography became much more important for pharmaceutical research and development. “Residual solvents may not have needed as much resolution, but capillary columns allowed pharmaceutical researchers to get a finer and finer cut of what their reactions were and what components they were making. It was a great step forward for R&D,” said Johnson.
A key GC tool introduced in the 1990s was atmospheric pressure control. “With GC, you have a capillary column and then a carrier gas, like helium, to push your sample through. A big part of getting reproducible results is having very precise control of the pressure of your carrier gas,” he said.
In 1989, Hewlett-Packard came out with electronic atmospheric pressure control, a boon for the routine pharma lab with its ability to create an electronic record of both the pressure settings and actual pressures during each chromatographic analysis. “For the regulatory and compliance side, being able to go back and say, ‘Yes, my instrument was running at these pressures and they were recorded and manipulated electronically,’ gave a lot of credibility to the answer,” Johnson said.
During the last 10 years, the field has been moving toward fast GC, Dr. McNair said. “This results from a better understanding of the Van Dinckter equation: The retention time of a column is proportional to the column length. Shorter columns mean faster analysis, but they have fewer plates,” he explained.
“But if you use Van Dinckter analysis, with thinner film and a small column, you can put plates back, use a faster flow rate, and in some cases even change from helium to hydrogen, which is even faster. If you have thin films, you can double and even quadruple your flow rates.”
Creation of GC-MS Systems
Perhaps the most important development for gas chromatography in the pharmaceutical industry in recent years has been the marriage of GC to mass spectrometry (MS) to create GC-MS systems. Mass spectrometers were used as the detector in GC systems as early as the 1950s, but for decades the systems were too costly and unwieldy for most settings. As computers became more and more miniaturized, the 1990s heralded the era of true GC-MS.
“Whenever you get an impurity, you can tell what it is immediately. You don’t have to go back and scratch your head,” Dr. McNair said.
Another revolutionary change for gas chromatography, multidimensional GC, is just now coming into its own, according to Dr. McNair. “This means that you have two columns of different selectivity doing the same analysis. It’s only for complex samples, things that have maybe 30 or more peaks.
“The system takes a bundle of unresolved peaks and shunts them to a separate column with a separate detector and separates them there,” he said.
It’s difficult and expensive, but for certain uses, such as the complex samples often essential to pharmaceutical research and development, this “GC-GC” approach may also prove a key wave of the future.