Spectroscopy Versus Chromatography for Potency Analysis

Published on: 
Cannabis Science and Technology, November/December 2019, Volume 2, Issue 6

Columns | <b>Cannabis Analysis</b>

An introduction to these techniques and an analysis of their advantages and disadvantages for potency analysis.

Both spectroscopy and chromatography have been used for decades to measure the concentrations of molecules in samples, and now both techniques have been used to measure cannabinoid profiles in marijuana and hemp based samples. This column briefly introduces these techniques to the novice, and then using the concepts of speed, cost, and accuracy, analyzes the advantages and disadvantages of each technique for cannabis potency analysis.

Both spectroscopy and chromatography have been used for decades to measure the concentrations of molecules in samples, and now both techniques have been used to measure cannabinoids in marijuana and hemp plant material, oils, extracts, and distillates (1–10). I feel the need to write this column because of misconceptions I hear stated by laypeople and experts in the cannabis industry. I have heard PhD scientists say that spectroscopy is “not quantitative.” This is of course nonsense since an equation called Beer’s Law (see “Spectroscopy” section further on) tells us spectroscopy is quantitative, and a shock to me considering I have written a book on quantitative spectroscopy (11). On the other hand, I have heard people berate chromatography by saying it “doesn’t work.” Chromatography does work for cannabinoid analysis as has been proven in the peer-reviewed literature (2,3 and references therein), and in the right hands produces accurate numbers. To try and clear the air, this column briefly introduces these techniques to the novice, and then using the concepts of speed, cost, and accuracy, analyzes the advantages and disadvantages of each technique for cannabis potency analysis.

The Golden Triangle of Chemical Analysis

The goal of any chemical analysis method is to obtain the greatest accuracy, with the fastest speed, at the lowest cost. These three criteria comprise what has been called the golden triangle of chemical analysis, as seen in Figure 1 (for the definition of accuracy please see my previous column [1]).

Note that the criteria are located at the corners of the triangle because they are too often mutually exclusive. For example, many techniques may be accurate but might be slow or expensive to use. On the other hand, techniques that are fast and inexpensive are frequently not as accurate as other available technologies. This is seen when comparing laboratory testing to field testing. Many laboratory instruments may require special utilities like cooling water or nitrogen gas, may be sensitive to temperature, humidity, or vibrations, may be too large to be taken outside the laboratory, or require a trained technician or scientist to be utilized. Thus, these instruments are not necessarily fast or inexpensive to operate, but they will often have the highest accuracy. On the other hand, field testing instruments tend to be fast, inexpensive, and operable by laypeople, but are not as accurate as laboratory testing because of the challenges of performing chemical analyses outside the laboratory. This is reflected in Figure 1 where lab testing is near the accuracy corner, while field testing is listed at the bottom near speed and cost.

Introduction to Chromatography

Chromatography is used to separate a mixture into its individual components. Once the components have been separated, the amount of each component can then be quantified. The word chromatography means “color writing” in Greek because in its original application chromatography was used to separate plant pigments into their separate colors which were viewed during the separation process.

In chromatography the sample to be analyzed is typically in solution. If a liquid needs to be analyzed the sample preparation may be simply a matter of “dilute and shoot” as chromatographers like to say, which means the sample must be diluted enough so that it is in the concentration range where the instrument can quantitate it. So, for example, to prepare a cannabis extract for chromatography, dissolving an appropriate amount in a solvent and diluting it with a further known quantity of solvent may be the only sample preparation needed.

However, the sample preparation for solids, such as cannabis plant material, is more challenging. The sample must be weighed, ground, have solvent added, agitated to promote extraction of the cannabinoids, filtered, and then diluted to the concentration range where the instrument can quantitate it (2). (Note: There have been many cannabis plant material sample preparation methods published, and even more in use. From my experience of monitoring analyses at many cannabis analysis laboratories, the method in reference 2 is best, and I would like to see it become the industry standard.)

Once the sample is in solution it is injected onto a “column” which is essentially a tube filled with a “stationary phase” and a “mobile phase.” The mobile phase can be a gas such as nitrogen or helium, which gives rise to gas chromatography (GC), or a solvent which gives rise to liquid chromatography (LC). The stationary phase is often silica particles that may or may not have a coating on them. The column can be as simple as a glass tube, or as modern as a thin steel pipe capable of withstanding high pressure, hence the technique of high performance liquid chromatography (HPLC). The mobile phase flows through the column carrying the sample molecules with it. For LC gravity can be used to encourage fluid flow, in gas chromatography changes in pressure are used, and in HPLC a pump is used to force fluid through the column.

The separation process for chromatography is illustrated in Figure 2.

In step 1 in Figure 2 the sample is loaded onto the beginning of the column. In LC this may involve simply pouring the sample into the top of the column. In GC the sample is injected into a heated port to vaporize it and it is then swept onto the column. In HPLC the sample is injected and the pumped flow of the liquid mobile phase carries the sample to the column. In the column the molecules in the sample will adhere to the stationary phase as seen in step 2. The pink, orange, and green bands represent three different types of molecules.

The critical element here is that different molecules will adhere to the stationary phase with different strengths or affinities. As mobile phase flows through the column, molecules with a weak affinity for the mobile phase will move through the column faster than molecules with a weak affinity. This is illustrated in step 3 in Figure 2, where the three molecular types have physically separated inside the column. You can almost think of chromatography as a molecular race. The column is the racetrack, the molecules move through it at different speeds, and cross the finish line at different points in time. In this case, the finish line is the end of the column. In step 4 in Figure 2 the green molecules have won the race and leave or “elute” from the column first. It has been separated from the other mixture components and is thus purified. Molecules with stronger affinities for the column will move more slowly and elute later. This is seen in step 5 in Figure 2 as the orange molecules leave the column in 2nd place. It is the difference in affinity for a stationary phase that chromatography uses to separate mixtures into their components.

As different batches of purified molecules leave the column, they can have their molecular structures determined and their concentrations measured. It is this latter measurement that will tell us, for example, the tetrahydrocannabinol (THC) content in an extracted marijuana bud. For quantitation, any number of different techniques can be used such as refractive index or the amount of light absorbed by the sample. For the determination of cannabinoids via HPLC, an ultraviolet-visible (UV-vis) detector is often used (2) (absolute proof that spectroscopy is quantitative since it is widely used as a chromatographic detector). A disadvantage of GC is that since the sample is heated, the acid form of the cannabinoids decarboxylate, which means GC struggles to quantitate these cannabinoids in samples. This may explain why today many laboratories are using HPLC for cannabinoid analysis.



Spectroscopy is the study of the interaction of light with matter (4). Light can be thought of as a wave, and an example of a light wave is seen in Figure 3, which is a plot of light wave amplitude versus time. (Light is properly called electromagnetic radiation [4], but for simplicity it will be called light here.)

To the left in Figure 3 the light wave starts off at zero amplitude. The wave goes up, comes back down, and crosses zero a second time. The wave then goes down, comes back up, and crosses zero a third time. What I have just described is called the cycle of a wave as seen in Figure 3. The distance forward travelled by the wave during one cycle is called its wavelength and is designated by the Greek letter lambda, λ. The wavelengths of light can vary from meters to billionths of a meter, and include radio waves, microwaves, infrared light or heat, the visible light that we can see, ultraviolet light that causes sunburns, X-rays, and so on. The fundamental measurement made in spectroscopy is to measure a spectrum. A spectrum can be a plot of the amount of light absorbed by a sample versus some property of light such as wavelength. An instrument that measures a spectrum is called a spectrometer. The infrared spectrum of tetrahydrocannabinolic acid (THCA) is seen in Figure 4.

The y-axis in Figure 4 is in absorbance units, and the height of the peaks is determined by the concentrations of molecules in the sample. The x-axis is plotted in wavenumber (which is related to wavelength). The peak positions are determined by the structures of the molecules present. For the analysis of cannabinoids in samples, infrared (IR) light has typically been used (5–10). It has been used to measure THC, cannabidiol (CBD), and other cannabinoids in marijuana plant material (5,6), hemp (7), cannabis extracts (8,9), and distillates (10).

To use light to measure concentrations in samples we take advantage of Beer’s Law, which is shown in equation 1.

A = εlc                     [1]

where A is the amount of light absorbed by a sample, ε is the absorptivity, l is the pathlength, and c is the concentration.

The absorbance reading is the peak height or area of a peak in a spectrum. The pathlength is the thickness of sample seen by the infrared light beam, and c is concentration which is of course what we wish to determine. The absorptivity, ε, is the proportionality constant between absorbance and concentration, and is explained in more detail elsewhere (11). To use Beer’s Law the size of the peaks in a spectrum are correlated to concentration using standard samples of known concentration.

To determine cannabinoids in dried cannabis plant material by IR spectroscopy the only sample preparation typically needed is to grind the sample (5–7). For cannabis extracts, oils, and distillates one can use a sampling method called attenuated total reflection (ATR) (8–10,12) where the liquid is simply spread on a window. It typically takes about 2 min to analyze a sample by IR spectroscopy (12).

Spectroscopy Versus Chromatography: Accuracy, Speed, and Cost

In a previous column, I discussed error, accuracy, and precision (1). Chromatography is an example of a primary method because chromatographs are calibrated using pure standard materials. Spectrometers give secondary analyses because they are typically calibrated using actual cannabis samples and reference cannabinoid data measured on the same samples via chromatography. When combining measurements to determine a concentration, the different sources of error combine via the laws of error propagation to determine the error in the final analysis (13). Long story short, chromatographic methods typically have excellent accuracy because they are primary methods. The accuracy of spectroscopic methods is typically not as good because the chromatographic reference data have error, the spectroscopic analysis adds its own error, and when the two are combined the total error for a spectroscopic analysis is generally greater than that for a chromatographic analysis (13).

However, remember that the “Golden Triangle” of chemical analysis seen in Figure 1 contains three parameters-accuracy, speed, and cost-and the true worth of an analytical method is based not just on accuracy but on all three of these criteria combined. When it comes to speed chromatographic analyses begin to suffer. It can take 5–10 min to prepare cannabis plant material, and another 5–10 min to run the sample through the chromatograph (2,3), which means it can take upwards of 20 min to analyze one sample.

Lastly, let’s talk about cost. The purchase price of an instrument can vary depending upon the make, model, features, and whether it is purchased new or used. An important, but I believe often overlooked, aspect of cost is the cost per analysis-the cost to analyze one sample. Chromatography, particularly for plant material, involves extensive sample preparation. This means consumables such as vials, syringes, filters, and solvent are needed to prepare samples. For HPLC there is the additional cost of environmentally sound solvent disposal. In my experience, the cost of consumables per sample in chromatography can run upwards of $10/sample. Although claims have been made that laypeople can prepare samples for and run a chromatograph, there is a reason many state regulations require highly trained technicians to run chromatographs (15)-it takes skill to operate them. The need then for highly trained people to perform chromatographic analyses drives up the cost per sample. Assuming it takes 10 min total per sample, and given that skilled technicians earn a good wage, the labor cost per sample with chromatography can be upwards of $15. The total cost per analysis then for chromatography can be $25 or more. Chromatographic instruments also are generally not used in the field because of the need for utilities such as a nitrogen tank for GC, their bulk, and the need for a skilled user.

Spectroscopy on the other hand does not suffer from these problems. For cannabis liquids such as oils, extracts, distillates, and tinctures there is no sample preparation (8–10). For dried cannabis plant material the only sample preparation is grinding, which takes 1 min and can be performed with an inexpensive coffee grinder (6,7). Anyone can do this. There is no need to extract, shake, filter, and then dilute the sample so the consumables cost per analysis for spectroscopy is $0. The labor cost is minimal since anyone can run a sample in about 2 min. In total then the cost per analysis for spectroscopy is about $0. Field portable cannabis analyzers now exist that are fast and inexpensive to use (14).

Spectroscopy and Chromatography

In the big picture chromatography is accurate, but is not fast or inexpensive to use. Spectroscopy is not as accurate as chromatography, but is generally faster and has a lower cost per analysis. What are we to do with this information? Which technique then is better for potency analysis? The answer is neither. As always, one should use the right tool for the right job. In cases where high accuracy is a must, such as for compliance testing or for measuring reference data to be used to calibrate a spectrometer, chromatography should be preferred. For situations such as field testing, or in-house testing, where speed and low cost are important, spectroscopy should be preferred.


We have reviewed how two common cannabis potency methods, chromatography and spectroscopy, work. The two methods were evaluated using the criteria of accuracy, speed, and cost. Chromatography is accurate but can be slow and expensive to use. Spectroscopy is typically not as accurate as chromatography, but is quicker with a lower cost per analysis. Because of its accuracy, chromatography should be preferred for in-laboratory use for applications such as compliance testing. Because of its speed and low cost, spectroscopy should be preferred for in-house or field testing.


  1. B.C. Smith, Cannabis Science and Technology 1(4), 12–16 (2018).
  2. M.W. Giese, M.A. Lewis, L. Giese, and K.M. Smith, J. AOAC Int. 98(6), 1503 (2015).
  3. T. Ruppel and M. Kuffel, “Cannabis Analysis: Potency Testing Identification and Quantification of THC and CBD by GC/FID and GC/MS,” PerkinElmer Application Note (2013).
  4. B.C. Smith, Fundamentals of Fourier Transform Infrared Spectroscopy, 2nd Edition (CRC Press, Boca Raton, Florida, 2011).
  5. C. Sánchez-Carnerero Callado, N. Núñez-Sánchez, S. Casano, and C. Ferreiro-Veraa, Talanta190, 147–157 (2018).
  6. B.C. Smith, M. Lewis, and J. Mendez, “Optimization of Cannabis Grows Using Fourier Transform Mid-Infrared Spectroscopy,” PerkinElmer Application Note (2016).
  7. B.C. Smith, Cannabis Science and Technology2(6), 28–33 (2019).
  8. B.C. Smith, Terpenes and TestingNov.-Dec., 48 (2017).
  9. B.C. Smith, Terpenes and TestingJan.-Feb., 32 (2018).
  10. B.C. Smith, P. Lessard, and R. Pearson, Cannabis Science and Technology2(1), 48–53 (2019).
  11. B.C. Smith, Quantitative Spectroscopy: Theory and Practice (Elsevier, Boston, Massachusetts, 2002).
  12. B.C Smith, Fundamentals of Fourier Transform Infrared Spectroscopy, 2nd Edition (CRC Press, Boca Raton, Florida, 2011).
  13. D. Shoemaker and C. Garland, Experiments in Physical Chemistry, 2nd Edition (McGraw Hill, New York, New York, 1967).
  15. California Bureau of Cannabis Control Regulations, Section 5719.

About the Columnist

Brian C. Smith, PhD, is Founder, CEO, and Chief Technical Officer of Big Sur Scientific in Capitola, California. Dr. Smith has more than 40 years of experience as an industrial analytical chemist having worked for such companies as Xeros, IBM, Waters Associates, and Princeton Instruments. For 20 years he ran Spectros Associates, an analytical chemistry training and consulting firm where he improved their chemical analyses. Dr. Smith has written three books on infrared spectroscopy, and earned a PhD in physical chemistry from Dartmouth College.

How to Cite This Article

B.C. Smith, Cannabis Science and Technology2(6), 10-14 (2019).