How to Think Like an Analytical Chemist

My mission in writing this series of Cannabis Analysis columns is to passionately advocate for more and better testing in the cannabis industry. The science whose increased use I am advocating for is analytical chemistry. Many of my previous columns on spectroscopy, chromatography, and other topics have been about analytical chemistry. It is about time I backed up a little and defined what analytical chemistry is, and to teach my readers how to think like one.

What is Analytical Chemistry?

Here is my definition of analytical chemistry:

“Analytical chemistry is the science of determining what chemical species are present in samples and in what amounts.”

Chemical species can consist of molecules, atoms, and ions (charged atoms or molecules). Note that I do not include things like protons, neutrons, quarks, etc. As someone once said: chemistry goes from the macroscopic world to the nucleus, and physics begins at the nucleus. I agree.

There are essentially two different types of chemical analysis captured by this definition. An analysis that determines what chemical species are present is called a qualitative analysis because the answer is not a number, but an identification. A quantitative analysis measures the amount of chemical species in a sample, by definition, producing a number. This will typically be a concentration such as weight percent, moles per liter, or parts per million.

Cannabis analysis then is the determination of what chemical species are present in what amounts in cannabis samples. We could argue what “cannabis samples” are, but I will cast a wide net and say they consist of cannabis plant material, its extracts, distillates, and final products. Essentially, anything with THC, CBD, or other cannabinoids will be considered a cannabis sample. Here are some of the analytical chemistry techniques in common use in cannabis labs and the things they are used to analyze for:

• Gas Chromatography (GC) – Cannabinoids and terpenes
• High Pressure Liquid Chromatography (HPLC) – Cannabinoids
• Infrared Spectroscopy – Cannabinoids and terpenes
• Gas-Chromatography-Mass Spectrometry (GC-MS) – Pesticides
• Liquid Chromatography-Mass Spectrometry (LC-MS) - Pesticides
• Atomic Spectroscopy – Heavy metals

These are not all of the techniques you will find in a cannabis analysis lab, but most labs will have some or all of these instruments. The vast majority of cannabis analyses are looking for some or all of cannabinoids, terpenes, pesticides, and heavy metals. Others and I have written about the use of these techniques in cannabis analysis, and they will continue to dominate my articles going forward (1-3).

At the beginning of the science of analytical chemistry over a hundred years ago, all analyses were done using what is called wet chemistry; literally analyses using beakers, flasks, test tubes, and solutions. As time went on, instruments and then computers were invented to perform these analyses, which is where we are today. Hence, an understanding of analytical chemistry involves familiarizing oneself with topics such as physics, optics, electronics, computers, computer programming, and engineering. This is why I love analytical chemistry; it allows me to masquerade as a chemist while learning about new things in all sorts of other interesting fields.

Because cannabis products go onto or into human beings, they must be tested before they reach the point of purchase to ensure they are safe. This testing can be done in-house by cannabis businesses, or by third party labs. You can’t be in this industry without at some point dealing with cannabis analyses, cannabis analysis reports, cannabis labs, or cannabis analyzers. The goal of this article, and perhaps some subsequent ones, will be to familiarize my readers with the science of analytical chemistry so you can make better use of this science in your work.

We can think of any chemical analysis as consisting of the parts shown in Equation 1:

These parts can be defined as the following,

Sample: The thing being analyzed.

• Analyzer: The instrument performing the analysis.
• Method: The set of instructions performed to complete an analysis, sometimes called a Standard Operating Procedure or SOP.
• Analyst: The human being performing the analysis. There exist automated analyzers and robots that perform some chemical analyses, but there is always a human being involved in preparing samples, setting up the analyzer, or at minimum turning the analyzer on and monitoring its work.
• Analysis: The results.

Defining some further terms is in order:

• Analyte: The thing whose presence or amount is being measured. In our industry analytes can be the concentration of THC in a marijuana bud, the amount of limonene in a vape pen formulation, or the parts per million of a pesticide in a tincture.
• Matrix: The nature of a sample. This includes the composition of the sample, the concentrations of the components present, and other variables such as pH, pressure, and temperature. For example, the matrix of a tincture could be described as being that of CBD, and MCT oil, at room temperature and pressure.
• Method Development: The art and science of developing a method to analyze for the chosen analytes(s) in a given sample matrix.

The Golden Rectangle of Chemical Analysis

As an analytical chemist, how do I go about choosing an instrument and method to perform the analysis of a sample I am handed? I use the Golden Rectangle of Chemical Analysis as seen in Figure 1.

The terms in Figure 1 have the following definitions in my parlance:

• Accuracy: This is something we have defined in previous columns (4). Briefly, accuracy is a measure of how far off you are from the true value. This necessitates there exists a way of determining the true value, which is provided by standard samples. Accuracy is NOT the same as precision, which is a measure of reproducibility as I have written before (4).
• Speed: This is a measure of how much time an analysis takes. In a perfect world, we would want all of our analyses to be done quickly to make the most efficient use of the analyst and analyzer. There are two things that determine the speed of a chemical analysis. One is how long it takes to prepare the sample, and the second is how long it takes to run the sample on an analyzer. For example, in a chromatographic analysis the speed would include how long it takes to work the sample up for analysis, and the amount of time it takes to effect the separation on a chromatograph.
• Cost: The presence of this variable in Figure 1 is a no brainer. Of course we want our analyses to be performed as inexpensively as possible. There are several things that determine the cost of an analysis. The first is the upfront cost of the instrument itself. The cost of analytical instruments can vary from tens of thousands of dollars for infrared spectrometers and chromatographs to hundreds of thousands of dollars for GC-MS and LC-MS systems. I once looked into starting my own cannabis analysis lab, and determined I would need $1 million just for the instruments. I didn’t bother opening a lab. The second part of the cost of an analysis is the maintenance and upkeep of analyzers. This can include the cost of repairs, spare parts, and consumables like chromatographic columns, pump seals, or spectroscopic light sources. The third part of an analysis cost is sample preparation. Particularly for chromatographic analyses, there are significant sample preparation costs including vials, vial covers, syringes, filters, and solvents. Lastly, there is the cost of the labor to analyze a sample, which includes the time needed to prepare a sample and the time to actually run the sample on an analyzer.
  • Many analytical instruments are expensive to maintain and even more expensive to fix when they break down. I strongly recommend most labs use what are called service contracts. This is a contract you sign with the manufacturer of your instrument that states in exchange for a fixed annual payment they will come by at least once a year, perform routine maintenance, and make any needed repairs at no additional cost. It works like an insurance policy in that it costs money every year, but if your instrument breaks you will be glad you have the insurance. I see some labs foolishly “self-insuring” by paying for maintenance and repairs as they go. This is silly because the cost of fixing analytical instruments is high. For example, labs must pay for the time and travel costs of getting repair people to their instrument. The instrument company starts the clock when the repair person leaves their house. Companies charge hundreds of dollars per hour for travel time, there is the cost of the travel, there is the time spent onsite, and then the cost of the repair parts. The latter can be thousands of dollars. Again, it is best to have a service contract.
• Representative Sampling: This is also something I have written about before (5). By definition, when we gather a sample, we are taking a part of a whole and then analyzing that part. A problem arises if the part is not the same as the whole, leading to sampling error (5). The solution to sampling error is to analyze as many samples as possible and average the results (5). For this to be practical, the analytical method must be fast, easy, and inexpensive. This aspect of analytical methods is somewhat dependent on the other things listed in Figure 1, but it is important enough to merit its own mention. Representative sampling is a particular problem when analyzing cannabis plant material, since it is a heterogeneous natural material and no two samples are alike.
• Sensitivity: This is a measure of the minimum amount of analyte that can be detected by an analyzer. This is essentially the same as a term you may see in a lab report, the limit of detection. Now, if you are lucky enough to have a large amount of sample then sensitivity is not an issue, but if you have very little sample or you are looking for an analyte in low concentration, an instrument’s sensitivity can become very important. In our industry, the need for sensitivity varies. For example, in marijuana bud samples the THC concentration is often present at levels of 10% or greater, so a potency analyzer does not need to be highly sensitive. On the other hand, there are state laws that require the measurement of certain pesticides at the part per billion (ppb) level. These analyses require a very
  sensitive analyzer.
• Specificity: There are over 10 million chemical substances known to humanity. Specificity is a measure of how well a method distinguishes between the analyte and all the other stuff that might be present in a sample matrix. This property of a method is particularly important when dealing with complex mixtures or mixtures of things that are similar, such as terpenes. For example, imagine we are trying to measure the potency of a marijuana bud. Recall that the Total THC in this sample is given by Equation 2:
  

• A method that determines just Total THC would not be as specific as a method that determined both THC and THCA.
• The six parameters listed in Figure 1 interact and there are tradeoffs between them. For example, high accuracy is nice to have but frequently comes at the cost of taking more time and being more expensive. Examining many samples to minimize sampling error is great, but that of course drives up the cost, time, and complexity of analyses. A specific method is lovely, such as having an analyzer that tells you about the Total THC in a hemp sample to determine legality. But if your interest is in knowing about the other cannabinoids present, which is called a cannabinoid profile, your very specific analyzer will not be of much use to you.
• Given the parameters in Figure 1, which one is most important, or how do we weigh the relative merits of one versus the other? The answer to this question, like the answer to many questions in life, is it depends. It depends on what you need to learn from a sample. That is, you have to put the sample into context. For example, if you are testing a marijuana bud to see if it meets your state’s allowable pesticide limits, the method will be different than if you want to determine the amount of CBD in a tincture. That is, chemical analyses take place within a context, and it is important to understand what that context is before applying the Golden Rectangle of Chemical Analysis to choose your method.

The Cult of Accuracy

It is my observation that for many people involved in chemical analyses, the rectangle in Figure 1 would consist only of one point, and the only parameter listed would be accuracy. Their obsession with obtaining the highest accuracy at all costs is what I call the cult of accuracy. Let me give you an example.

I once had an industrial client who hired me to develop a quantitative infrared spectroscopic method for them. Being a good analytical chemist, I asked them a bunch of questions to determine the context of the analysis (asking the right questions before an analysis to determine context is important, a good topic for a future column). I determined that to monitor the quality of their chemical process an accuracy of ±10% was needed. I subsequently developed a method that was accurate to ±5%. Pleased with myself, I presented the results to the client. Instead of the joy I thought I would engender in them, they rejected my method as not being “accurate enough”. They demanded an accuracy of ±1%. They gave no rational reason for this, just that they wanted the analysis to be as “accurate as possible.” Higher accuracy comes at the price of cost and speed. In this case, the infrared spectroscopic method I developed took 2 minutes and required no sample preparation, so it was fast, easy, inexpensive, and was simple enough that non-scientists could perform the analysis.

The gas chromatographic method my client ultimately chose was accurate but involved significant sample preparation costs in consumables including solvents, vials, vial caps, and filters. The method was slow because each sample required 10 minutes of sample preparation time plus an additional 10 minutes to run the sample on the GC. Worst of all, they had to hire an experienced analytical chemist to perform the method, maintain and run the instrument, adding further to cost. They sacrificed time and money on the altar of accuracy. There is no need to do this, which is why I teach the Golden Rectangle concept in every analytical chemistry course I teach.

The moral of the story is that context matters in analytical chemistry and armed with that knowledge and the Golden Rectangle of Chemical Analysis, you will be able to pick the best technique for any given sample.

References

1. Smith, B.C.. A Proposed Representative Sampling Plan for Hemp Grows. Cannabis Science and Technology. 2020, 3 (6), 24-38
2. Giese, M.W.; Lewis, M.A.; Giese, L.; andSmith, K.M. Method for the Analysis of Cannabinoids and Terpenes in Cannabis. Journal of AOAC International.2015, 98(6), 1503. DOI: 10.5740/jaoacint.15-116
3. Ruppel, T. D.; Kuffel, N. Cannabis Analysis: Potency Testing Identification and Quantification of THC and CBD by GC/FID and GC/MS, Perkin Elmer Application Note https://cdn.technologynetworks.com/tn/Resources/pdf/cannabis-analysis-potency-testing-identification-and-quantification-of-thc-and-cbd-by-gcfid-and.pdf (Accessed 2025-12-10).
4. Smith, B., Error, Accuracy, and Precision. Cannabis Science and Technology. 2018, 1(4), 12-16.
5. Smith, B., A Proposed Representative Sampling Plan for Hemp Grows. Cannabis Science and Technology. 2020, 3(6), 10-13.

About the Columnist

Brian C. Smith, PhD, is Founder, CEO, and Chief Technical Officer of Big Sur Scientific. He is the inventor of the BSS series of patented mid-infrared based cannabis analyzers. Dr. Smith has done pioneering research and published numerous peer-reviewed papers on the application of mid-infrared spectroscopy to cannabis analysis, and sits on the editorial board of Cannabis Science and Technology. He has worked as a laboratory director for a cannabis extractor, as an analytical chemist for Waters Associates and PerkinElmer, and as an analytical instrument salesperson. He has more than 30 years of experience in chemical analysis and has written three books on the subject. Dr. Smith earned his PhD on physical chemistry from Dartmouth College. Direct correspondence to: brian@bigsurscientific.com

How to Cite This Article

Smith, B., How to Think Like an Analytical Chemist, Cannabis Science and Technology, 2025, 8(6), 6-9.