Daniel Zimmerli, associate scientist and lead of the Separation Science Point group, discusses how his team uses high-performance liquid chromatography (HPLC) and ultra-high performance (or pressure) LC (UHPLC) technologies for analytical and preparative work in the Chemistry Technologies and Innovation department at F. Hoffmann- La Roche. He shares his perspectives on the changes taking place in liquid chromatography-based separations, both with the instrumentation hardware and data analysis software. He advises chromatographers and lab managers to keep their eyes and ears open for new instruments being released in the marketplace, and to look six to 12 months into the future on the types of applications they will be working on when making their buying decisions.

Q: In which types of applications do you use chromatography?

A: I work in the Discovery Chemistry group at La Roche, and we do a lot of separation and purification to support early discovery work. We have two dedicated laboratories with a lot of chromatography instrumentation. We have 12 different HPLC instruments and one UHPLC for high-throughput work. We use HPLC for chiral separation and preparative work, and for separating biological samples we use UHPLC. On average, we run about 800 to 1,000 samples per instrument per day. For the preparative work, which involves around 1g to 10 g of sample, we average 3,000 samples per year using traditional HPLC.

Q: What types of changes have you seen in the chromatography market over the years?

A: I have worked for 12 years in this lab, and in the past few years I have seen traditional HPLC moving more to UHPLC. We have seen new machines and new columns in the market that have made separations faster and more high-throughput. We have seen sample run times go down from 30 minutes to one or two minutes, and we are much faster now than we were 10 years ago. We now have different column chemistries, and the particle size and dimensions of the columns have also changed. Previously, we worked at pressures of around 200 bar, but now we can go up to 1,500 bar with the new column packing. With faster run times we save a lot on solvents, too. One year ago when the cost of acetonitrile—our main solvent for reversephase chromatography—went up nearly five times, we saved a lot of money and solvent with our fast columns.

Q: How is the transition from HPLC to UHPLC?

A: The cost of UHPLC is about 20 to 25 percent higher than traditional HPLC, but in about three to six months you break even. You have to invest more to begin with, but depending on the number of samples you run, you can break even very quickly. Our lab works seven days a week, and we average about 2,000 samples on each machine per week.

Q: How often do you upgrade your instruments and technology?

A: We are always looking for machines that are fast, that we can easily adapt to our columns, and that are robust. Every year, we buy at least two or three new HPLC instruments. We either move the older HPLCs to other labs within La Roche or donate them to universities. Our department is Chemistry Technologies and Innovation, and we need to constantly upgrade to use the newest technology and use the best that’s out there on the market. We test a lot of machines during the year, keep our eyes open for what’s new on the market, and attend conferences to keep up with new releases. In our group, we are not expected to buy only from a certain vendor. In some other departments where they are controlled or under regulatory restrictions, they have to work with only one vendor and they change their machines only every five to seven years.

Q: What are your experiences with using the instrument software?

A: Now, there are some really good machines and columns coming in for the UHPLC market. But in most cases, in the beginning, they all have problems with their software. It takes nearly six months to a year before the software bugs can be worked out. Sometimes we are the first users on a new machine, and we have to give feedback to the company, and then updates are made. We have service contracts with every company, and most companies provide two-day user training. But we are very familiar with a lot of different instruments and software, so in our group we usually don’t need any special training. Most companies have good customer service, and they all have good technical support and are open to discussion.

Q: What are some of the limitations with using chromatography?

A: There has been no change in the preparative market in the past five to 10 years. Everything is for the analytical market; this is possibly because not many people work in the preparative market. We are hoping that we will be able to apply UHPLC technology to the preparative work. For the separation of some chiral molecules and peptides, we need hours to separate the compounds and we need a lot of solvent. For chiral molecules that take 10 minutes to separate in the analytical scale, it may take up to an hour in the preparative scale. Hopefully, companies will come out with new instruments in the preparative market in the next couple of years, and preparative work will get better, faster and cleaner with the use of supercritical fluid chromatography (SFC).

Q: What do you see happening in this field going forward?

A: I think the traditional HPLC market will stay but the percentage of use will go down every year. Maybe in the next 10 to 20 years everyone will upgrade to UHPLC. In most companies, there are labs that work in a controlled environment and have protocols validated for regulatory use with traditional HPLC instruments. For those labs it will take time, about three to five years, and money to adapt their protocols and transfer everything to UHPLC. Some departments at La Roche have 20 traditional HPLC instruments and they are changing three instruments every year to UHPLC analysis.

Q: What is your advice for lab managers looking to invest in HPLC or UHPLC?

A: When you are looking to buy a new instrument, you should know what you want to separate and how much resolution is needed. Look a little into the future, about six to 12 months ahead, to figure out exactly what you need. For a smaller lab with lower throughput, cheaper machines working at low pressure may suffice, depending on the application. All vendors offer a range of different machines, and there are instruments that are in the range between HPLC and UHPLC that are sometimes 20 percent cheaper. So keep your eyes open and speak to different people.