The Essence of Modern HPLC: Advantages, Limitations, Fundamentals, and Opportunities

This article reexamines the fundamental concepts of high performance liquid chromatography (HPLC) to bring fresh insights to how we perform HPLC today. It reviews the prominent advantages that render HPLC indispensable and comments on its many perceived limitations. Several opportunities with far-reaching impacts in life science for the separation scientist are described.

This is the second installment of a new column in LCGC North America titled "Perspectives in Modern HPLC," which will be published every quarter and features fresh perspectives, innovative approaches, best practices, megatrends, and emerging opportunities in this ever-evolving field of separation science. The first installment in the April 2013 issue was devoted to new high performance liquid chromatography (HPLC) products introduced at Pittcon 2013 (1).

HPLC is a dominant analytical technique with "mature" technologies that have been widely practiced for five decades. Innovations such as ultrahigh-pressure liquid chromatography (UHPLC), liquid chromatography–mass spectrometry (LC–MS), two-dimensional liquid chromatography (2D-LC), chiral separations, core–shell columns, and novel stationary phases have helped drive HPLC to higher performance in diverse applications, yielding faster speed, higher resolution, greater sensitivity, and increased precision. The practice of HPLC is no longer limited to specialists or "chromatographers," but is now widely performed by students, chemists, biologists, production workers, and other novices in academia, research, and quality control laboratories. More than $4 billion of HPLC equipment, columns, and accessories were sold worldwide in 2012 (2).

There is no shortage of information on HPLC (3–9). Hundreds of books, thousands of articles, and millions of web citations are available. My goals are to reexamine the big picture of HPLC and its applications from a user's perspective. I will strive to find approaches to make it more productive or relevant. I am excited to be a new columnist for LCGC North America and promise to dig deeper and comment on ideas to make HPLC more exciting and less arduous for all practitioners. A listing of my tentative topics for 2013 and beyond can be found in the addendum.

In this installment, the essence of modern HPLC is discussed — first, by examining the reasons that make HPLC so ubiquitous; second, by reviewing the fundamental chromatographic principles on how they can guide the way we conduct HPLC separations; and finally, by commenting on some less-obvious opportunities with far-reaching impacts or immediate job prospects for separation scientists.

Table I: Advantages and limitations of HPLC

Why Is HPLC So Ubiquitous? Advantages and Perceived Limitations of HPLC

Why is HPLC so ubiquitous as practiced by thousands of practitioners around the world? Table I lists the advantages and perceived limitations of HPLC (3). The reader, without a doubt, has seen similar lists elsewhere. Here, I would like to discuss a few highlighted advantages (in bold) with an example on a stability study followed by a description of its perceived limitations.

Table II: Results of a three-month accelerated stability study in a new drug product formulation

Advantages of HPLC

The dominance of HPLC as a premier analytical technique is no accident. The most prominent advantage is its applicability to diverse analytes types, from small organic molecules and ions to large biomolecules and polymers. The successful coupling of HPLC to MS gave it an invincible edge as "the perfect analytical tool" — combining excellent separation capability with the unsurpassed sensitivity and specificity of MS. HPLC–MS is rapidly becoming the standard platform technology for bioanalytical testing (drugs in biological fluids), trace analysis for residues in food, forensic and environmental samples, and life science research (3–6). Finally, the excellent precision and robustness of HPLC with UV detection makes it an indispensable tool for quality control (QC). This last point is illustrated by a case study on stability evaluation of a pharmaceutical product shown in Figure 1 and Table II.

Figure 1: UHPLC chromatograms of (a) a retention marker solution and (b) a three-month stability sample (extract of a tablet kept in a stability chamber at 50 °C/75%RH). This is an example of a stability-indicating assay used extensively in the pharmaceutical industry to establish shelf life. Column: 100 mm × 3.0 mm, 2-μm dp ACE Excel 2 C18; mobile-phase A: 20 mM ammonium formate (pH 3.7); mobile-phase B: 0.05% formic acid in acetonitrile; flow: 0.8 mL/min; temperature: 40 °C; pressure: 450 bar; gradient: 5–15% B in 2 min, 15–40% B in 10 min, 40–90% B in 1 min; detection: UV absorbance at 280 nm; sample: tablet extract in 20% acetonitrile in 0.1 N HCl; injection volume: 3 μL.

Figure 1 shows chromatograms of a retention marker solution and a three-month stability sample of a drug tablet formulation. The retention marker solution contains the active pharmaceutical ingredient (API) spiked with its expected impurities and degradants. This type of testing is conducted routinely by pharmaceutical laboratories to establish shelf lives and storage conditions for API and drug products (5,6). The HPLC conditions use a multisegment gradient with ammonium formate buffer and acetonitrile. Peak designations shown in the chromatograms are API (SRR, absolute configuration for the drug molecule with three chiral centers); SRS and RRR (process impurities-diastereomers); M235, M416, and M399 (degradants designated by their MS parent ions); ketone (an oxidative degradant); and BHA (butyl hydroxyanisole, an antioxidant additive). The bottom chromatogram shows the extract of a tablet formulation kept in a stability chamber at 50 °C/75%RH for three months, indicating increased levels of degradants for M416, SRS, RRR, ketone, and M399 (data captured in the stability table in Table II). The chromatograms and the operating conditions are fairly unremarkable by today's standard though they serve to illustrate some of the less obvious strengths of HPLC in QC applications, such as

• the ability to quantitate all components (API and all related substances, including isomers);

• very precise retention times and peak areas using UV detection (<0.1% relative standard deviation [RSD] is routinely achievable for UHPLC versus 0.2–0.3% RSD for HPLC);

• highly reproducible assays (robustness) by different laboratories (with instruments from different vendors and columns from different batches);

• high-sensitivity assays for trace impurities ~0.01% in this assay (limit of quantitation of 0.05% or 0.10% is required by regulations);

• and all components readily identified by MS for assays with volatile mobile phases.

Table II is a stability report summarizing data from the three-month time point of this accelerated stability study of the oral tablet formulation under various storage and packaging conditions, indicating increased levels of degradants at 40 °C/75%RH and 50 °C/75%RH, particularly for hydrolytic degradant M399. The remarkable aspect lies in the exceptional quality of the stability data generated by HPLC, with its ability to track changes of drug impurities over time. Data are highly reproducible with a high degree of confidence and can be repeated by different labs. This high level of data reliability and reproducibility is taken for granted in HPLC applications for quality control to such an extent that it becomes mundane — a feat less achievable by capillary electrophoresis (CE), MS, or supercritical fluid chromatography (SFC).

Perceived Limitations of HPLC

Limitations of HPLC are rarely discussed and are, therefore, more interesting. "Perceived limitations" is the terminology used here since most have been mitigated by recent advances and are no longer real practical issues.

Lack of a Universal Detector

The lack of a universal detector for HPLC is often mentioned, although the UV–vis detector comes close to one for chromophoric compounds. Refractive index detection fits the bill, but suffers from low sensitivity and incompatibility with gradient elution. Evaporative light scattering detection (ELSD) was a contender, but was surpassed by charged aerosol detection (CAD). CAD uses a nebulizer with corona discharge detection and has better sensitivity (low ng) and ease-of-use than ELSD (3,10).

Mass spectrometry is becoming a universal detection method for ionic or ionizable compounds with incredible speed, sensitivity, and selectivity. The developments of triple-quadrupole MS-MS, high-resolution MS (for example, time-of-flight [TOF] and orbital trap), and hybrid MS (Q-TOF or ion trap–orbital trap) (11) in combination with UHPLC and 2D-LC have transformed our abilities to develop and perform bioanalytical assays, multiresidue analysis of complex samples, and life science research (9).

Less Separation Efficiency Than Capillary Gas Chromatography

Conventional HPLC has a practical peak capacity (P_c) of ~200 using columns with ~20,000 plates under gradient conditions — not particularly effective for very complex samples (3). The advent of UHPLC has extended P_c to 400–1000 range in a time span of ~60 min (9,12–16). 2D-LC can further increase P_c for comprehensive analysis of very complex samples in proteomics and metabonomics (9,16).

Relatively More Difficult for Novices

The bewildering number of HPLC modules, columns, mobile phases, and operating parameters renders HPLC difficult for the novice. Surprisingly, with a single-point control of the HPLC system by the data system, it becomes relatively easy to teach a new person to run an existing HPLC method. For example, just place the sample vial into the autosampler tray and the assay can be started with few mouse clicks with formal-looking reports automatically printed afterwards. Nevertheless, substantial experience and scientific judgment are needed to develop a new method, interpret a strange result, or to troubleshoot a problem. The good thing is that chromatographic principles are well documented and can easily be explained by more experienced colleagues in your laboratory. HPLC is complex, predictable, and not particularly complicated to a typical scientist with a strong chemistry background.

Still Arduous, Particularly for Regulated Testing

HPLC is versatile, quantitative, sensitive, and extremely precise. It can also be time-consuming and arduous, particularly for regulated analysis under good manufacturing practices (GMP). For instance, these are the steps in a typical operation: weighing reference standards; preparing samples and mobile phases; setting up the column and all modules; performing system suitability testing; injecting standards to calibrate the system followed by samples analysis; performing peak integration; reporting; reviewing; and sign-offs. Fortunately, most steps are automated by precision instruments for routine testing and are therefore highly reproducible. Compare it with spectroscopic analysis such as the identification of raw materials using a hand-held Raman spectrometer — point the laser to the sample, press a button and a pass or fail result with GMP documentation is available in seconds. One piece of advice: Don't use HPLC unless you have to quantitate analytes with high accuracy and precision.

HPLC Fundamentals and Insights on Performing HPLC Separations

Let's briefly review the fundamental principles to look for some fresh insights on how we perform HPLC analysis. First, the goal of most HPLC analysis is to quantitate analytes in a sample (mixture) by physically separating its components. It is useful to categorize samples as simple or complex since the analytical approach is quite different. In chromatography, three factors control the separation or resolution (R_s) of several components in the sample:

• Retention or having k values (retention factor) greater than one

• Selectivity (α) or differential migration of analytes in the column

• Separation power or column efficiency (N) — the ability of the column to separate many components in the chromatogram.

These factors are defined and explained in every HPLC textbook (3–6). For isocratic analysis in which the mobile phase composition remain constant, the relationship of R_s to k, α, and N are described by the resolution equation shown below (3,4).

Retention Selectivity Efficiency

Conventional wisdom leads us to the following "rules of thumb" for isocratic analysis.

First, k or retention factor is the ratio of retention times of the analyte to that of an unretained component. Keep k from 1 to 20 by adjusting mobile phase strength (% organic in reversed-phase LC). For potency or performance testing of the main component (for example, assays of drug substances or products, dissolution, or content uniformity testing), adjust the k to be ~1 and use a short column (length = 50 mm) for fast analysis (<2 min). For multicomponent analysis of a simple mixture, increase k by lowering the mobile phase strength until all components are retained and separated from each other. If there are four components and four distinct peaks are observed, then this can be a preliminary method condition. If there is a pair of coeluting or partially resolved peaks with R_s< 1.0–1.5, then the selectivity (α) of the two peaks or "critical pair" (α, which is the ratio of the two k values) should be adjusted.

Second, a selectivity value of 1.0 of the critical pair means coelution (that means interference of an analyte with another component), which precludes accurate quantitation because the method is not specific to that analyte. In HPLC, it is often easier to change the mobile phase (organic solvent type, buffer type or strength, and pH) because they can be continuously varied. The next step is to change column type or column temperature. In HPLC, adjusting or fine-tuning α is the main, and most time-consuming part of the method development process.

Finally, N or plate count is a measure of the separation power of the column and is proportional to column length and inversely proportional to particle size (d_p) (3–5). N can be reasonably low (for example, 5000 plates) for simple mixtures using a short column. Longer columns with higher N are preferred for more complex mixtures or for closely eluting analytes (for example, isomers). The practical maximum for N in conventional HPLC is ~20,000 plates for routine testing; equivalent to the N of a 150-mm-long column packed with 3-μm particles.

In isocratic analysis, the analyte peak or band is continuously broadened with higher retention times by the inherent chromatographic process (3–5). Values of k exceeding 20 are typically not feasible because the peaks would be too broad for quantitation and the gain in resolution becomes negligible. In reversed-phase LC, -log k is proportional to the solvent strength of the mobile phase or % organic solvent (%B). These relationships are well-behaved and very predictable (3,4).

Gradient analysis with increasing mobile phase strength is typically preferred for complex samples or mixtures with diverse polarities or for assays in which all components must be reported (such as impurity testing of pharmaceuticals). Gradient analysis yields higher peak capacity (P_c, defined as the number of peaks that can be resolved in chromatogram with an R_c value of 1.0; for example, P_c is ~200 for gradient vs. ~50–100 for isocratic) (2) and sharper peaks because peak widths are similar for all peaks irrespective of retention times. P_c is useful to measure performance under gradient conditions since one can only measure N isocratically. Gradient methods are more difficult to develop because retention and selectivity (and P_c) are affected by many factors, including initial and ending solvent strengths (%B) and gradient time (t_G), in addition to the typical mobile phase factors. Secondary factors are flow rate (F) and column temperature (3–5). Gradient analysis is less susceptible to extracolumn band broadening because sample band dispersion before the column and large injection volumes are inconsequential (for injection of samples in lower-strength diluents) — an important fact for UHPLC using columns with smaller internal diameters (9,13).

The advent of UHPLC (systems with low dispersion and pressure limits of 15,000 to 19,000 psi) together with the use of smaller internal diameter columns packed with sub-2-μm particles, accentuated the need for better understanding of chromatographic fundamentals such as R_s, k, N, α, particle size (d_p), column internal diameter (d_c), column void volume (V_m), peak volume, peak width, instrument bandwidth or system dispersion, flow-cell volume, and dwell volumes (3,9,13). Because UHPLC is becoming the modern standard HPLC platform, better understanding of these concepts will be helpful for efficient operation and method development and transfer (9,16).

In summary, the biggest strength of HPLC is its versatility for reliable quantitation of analytes in complex mixtures through physical separations of the analyte peaks from coeluted components. To effect separations, one must have retention (k), selectivity (α), and adequate plate counts (N). Retention is related to the partition coefficient of the analyte molecule between mobile and stationary phases. This partitioning process is repeated millions of times down the column to allow separation of analytes with minute differential migration (α). Selectivity (α) can be "tweaked" by changing column or mobile phase parameters. The unlimited number of combinations of columns, mobile phases, and controlling factors makes HPLC complex but gives endless possibilities for the quantitation of all or specific components in many sample types. HPLC works reliably in practice because of the gentle, predictable nature of the liquid phase chromatographic processes and the availability of precise instrumentation with efficient and reproducible columns. Very complex samples with thousands of analytes can be separated by "brute force" with UHPLC and 2D-LC coupled with UV, MS, or MS-MS (16). The complexity (versatility) of HPLC is its greatest strength and also its key weakness (laborious).

Opportunities for Separation Scientists

Today, I believe that biology and life sciences are the research areas that offer opportunities for separation scientists to make the greatest impact. Biology is the "Wild West" of the 21st century with a lot of fertile ground for scientific discovery. Unfortunately, most biologists are not experts in the versatile tools for discovery, HPLC or LC–MS (8), and separation scientists (mostly analytical chemists) don't usually have the intimate knowledge of the great biological problems (such as cell signaling and curing cancer or Alzheimer's disease). It would be ideal if scientists could straddle both analytical chemistry and biology to tackle pressing problems, such as the identification of disease biomarkers used for clinical diagnostics in personalized medicine (17–19). Many instrumentation and pharmaceutical companies are already investing heavily in this area (20), though new approaches are needed such as automated procedures to isolate the key analytes in these complex matrices (19). Here, 2D-UHPLC coupled with hybrid high-resolution MS can be a powerful generic tool — but only for those scientists with a good understanding of the problem and the analytical technologies.

My next comments are some immediate job opportunities for separation scientists in our recovering economies. In 2012, six of the 15 top selling drugs were monoclonal antibodies (mAb) (21). With hundreds of on-going mAb research projects as therapeutics, many job opportunities are available in the characterization and quality control of mAbs (22,23). However, analysts experienced in assessing the critical quality attributes of biological drugs are rare and graduate students are not trained in this area because it is not the funding source of their professors. So, there appears to be a disconnect between graduate training and job opportunities that goes beyond summer internships in the pharmaceutical industry. Perhaps a closer collaboration or partnership between academia and industry is the right solution.

Summary and Conclusions

In this installment, my first real column for "Perspectives in Modern HPLC," I have addressed the essence of modern HPLC by reviewing its advantages, limitations, and fundamental principles. HPLC is the dominant analytical technique because of its versatility, reproducibility, and wide applicability in research and quality control. HPLC is a complex technique because of its myriad combinations of modules, columns or mobile phases, and operating parameters. A deeper understanding of the principles is becoming more important for the effective use of UHPLC, the new standard platform of HPLC. Passionate separation scientists with expertise in LC–MS plus an in-depth understanding of the great problems in biology are in an excellent position to develop new approaches to make real impacts in life science for a better tomorrow.

Acknowledgments

The author is grateful to Drs. Sam Yang, Mohammad Al-Sayah, and C.J. Venkatramani, and Midco Tsang and Bob Garcia, Jr., of Genentech; Drs. Davy Guillarme of University of Geneva, and Ron Majors of Agilent Technologies; and Professors Kevin Schug of University of Texas at Arlington and Milton Lee of Brigham Young University for providing useful inputs and comments. The opinions expressed in this column are solely those of the author and bear no reflections on those of LCGC North America or other organizations.

Addendum

Agenda for 2013 and beyond for "Perspectives in Modern HPLC":

2013

• New HPLC product introductions at Pittcon 2013

• Essence of modern HPLC

• A three-pronged template approach for rapid HPLC method development

• Myths in UHPLC

Some potential Future Topics:

• Seven common faux pas in modern HPLC

• High-resolution UHPLC

• Analytical platform technologies

• Key equations in pharmaceutical analysis

• HPLC in drug discovery, development, and quality control

• Trends in modern food and environmental testing by HPLC

• LC–MS in clinical diagnostics

• Best practice of HPLC for characterization of biopharmaceuticals

Your ideas and inputs are solicited on areas deserving further investigation or discussions. Please send your comments and suggestions to: dong.michael_w@gene.com

References

(1) M.W. Dong, LCGC North Am. 31(4) 316–325 (2013).

(2) Market Analysis and Perspectives Report for Analytical and Life Science Instruments Industry (Strategic Directions Inc., Los Angeles, California, 2012).

(3) M.W. Dong, Modern HPLC for Practicing Scientists (Wiley, Hoboken, New Jersey, 2006).

(4) L.R. Snyder, J.J. Kirkland, and J. W. Dolan, Introduction to Modern Liquid Chromatography, 3rd ed. (John Wiley & Sons, Hoboken, New Jersey, 2009).

(5) Handbook of Pharmaceutical Analysis by HPLC, S. Ahuja and M.W. Dong, Eds. (Elsevier/Academic Press, 2005).

(6) HPLC for Pharmaceutical Scientists, Y.V. Kazakevich and R. LoBrutto, Eds. (Wiley, Hoboken, New Jersey, 2007).

(7) C.F. Poole, Essence of Chromatography (Elsevier Science, Amsterdam, The Netherlands, 2002).

(8) Chromatography: A Science of Discovery, R.L. Wixom and C.L. Gehrke, Eds. (Wiley, Hoboken, New Jersey, 2010).

(9) UHPLC in Life Sciences, D. Guillarme, J-L Veuthey, and R.M. Smith, Eds. (Royal Society of Chemistry, Cambridge, United Kingdom, 2012).

(10) M. Swartz, M. Emmanuel, A. Awad, and D. Hartley, "Advances in HPLC Systems Technology" supplement to LCGC North Am. 27(4), 40–48 (2009).

(11) Mass Spectrometry for Drug Discovery and Drug Development, W.A. Korfmacher, Ed. (Wiley, Hoboken, New Jersey, 2013).

(12) J.E. MacNair, K.C. Lewis, and J.W. Jorgenson, Anal. Chem. 69, 983–989 (1997).

(13) M.W. Dong, LCGC North Am. 25(7), 656–666 (2007).

(14) N. Wu and A.M. Clausen, J. Sep. Sci. 30, 1167–1182 (2007).

(15) D. Guillarme and M.W. Dong, Amer. Pharm. Rev., (2013) submitted.

(16) M.W. Dong, D. Guillarme, S. Fekete, R. Rangelova, J. Richards, D. Prudhomme, and N.P. Chetwyn, J. Chromatogr. A. submitted.

(17) L. Sannes, "Commercializing Biomarkers in Therapeutic and Diagnostic Applications – Overview," Insight Pharma Report, May 2011.

(18) A. Tessitore, A. Gaggiano, G. Cicciarelli, D. Verzella, D. Capece, M. Fischietti, F. Zazzeroni, and E. Alesse, Int. J. Proteomics 2013, 1–15 (2013).

(19) F.E. Regnier, LCGC North Am. 30(8), 622–623 (2012).

(20) M. Hollmer, "2012's Top 10 Diagnostics Companies," Fierce Medical Devices, November 27, 2012, www.fiercemedicaldevices.com/special-reports/top-10-diagnostics-companies.

(21) J.D. Carroll, "The 15 best-selling drugs of 2012," Fierce Pharma, October 9, 2012, http://www.fiercepharma.com/special-reports/15-best-selling-drugs-2012.

(22) T. Zhang, J. Zhang, D. Hewitt, B.Tran, X. Gao, Z.J. Qiu, M. Tejada, H. Gazzano-Santoro, and Y-H. Kao, Anal. Chem. 84, 7112–7123 (2012).

(23) S. Fekete, M.W. Dong, T. Zhang, and D. Guillarme, J. Pharm. Biomed. Anal. submitted.

Michael W. Dong is a senior scientist in Small Molecule Drug Discovery at Genentech in South San Francisco, California. He is responsible for new technologies, automation, and supporting late-stage research projects in small molecule analytical chemistry and QC of small molecule pharmaceutical sciences. He holds a PhD in analytical chemistry from the City University of New York and a certificate in Biotechnology from U.C. Santa Cruz. He has conducted numerous courses on HPLC/UHPLC, pharmaceutical analysis, HPLC method development, drug development process, and drug quality fundamentals. He is the author of Modern HPLC for Practicing Scientists and a co-editor of Handbook of Pharmaceutical Analysis by HPLC. He is a member of the editorial advisory board of LCGC.

Michael W. Dong