Dana Ciccone, chief executive officer of Steep Hill Hawaii, has been a patient advocate and leader in cannabis education in Hawaii, as well as a member of the Hawaii Medical Marijuana Dispensary Task Force, an organization formed by the University of Hawaii College of Social Sciences Public Policy Center to develop regulations for the state. “We are proud not only to be the first cannabis lab to be licensed in the State of Hawaii, but also now the first lab to achieve ISO certification as well,” says Ciccone. “Industry businesses, medical professionals, state regulators, and patients can be confident that our lab and its testing standards will operate to the highest international standards.”
According to the press release, the laboratory will offer services for testing cannabinoid profiles (potency), terpenes, pesticides, heavy metals, biological screening, and residual solvents, testing for 17 Cannabinoids and 43 terpenes. The release states they are locally owned and operated, providing testing services for not just industry businesses, but in-state card-holding patients as well.
“This is a turning point for the industry – we have moved very quickly to raise the industry standards in Hawaii to internationally recognized certification,” says Ciccone. “I am very proud our scientific team for the professionalism and hard work they put in to achieve this certification.”
Earlier this week, SC Labs issued a press release announcing they achieved ISO/IEC 17025:2005 accreditation for the cannabinoids panel at their Santa Cruz location.
“We are thrilled to announce our ISO accreditation as this is one of our most important achievements over the past seven years of serving the cannabis industry and demonstrates our commitment to serving our clients with integrity,” says Jeff Gray, co-founder and chief executive officer of SC Labs. ISO 17025 accreditation represents an international standard for a laboratory’s technical competence in producing accurate test results.
“Being accredited to this International Standard demonstrates our robust quality system, technical competence, the calibration and suitability of our instrumentation and our ability to produce precise and accurate test data,” says Gray. “For clients, it enhances their confidence in our services and their choice in a business partner, provides them with additional legal defensibility in complying with upcoming regulations, and enhances the integrity of their products based on SC Labs results.”
SC Labs is currently expanding in California, growing their Southern California and Santa Cruz locations, and adding field offices throughout the state, according to the press release.
A certified reference material (CRM) is generally recognized as providing the highest level of traceability and accuracy to a measurement. A CRM designed specifically for cannabis testing and tailored to state-specific testing regulations could help laboratories better ensure the safety of their products.
The fact that a certificate accompanies a reference material does not qualify it as a CRM. The reference material must be produced in accordance with ISO Guide 34 specifications by an accredited manufacturer. Adam Ross, key account manager and organic specialist at LGC Standards, says accreditation is a big part of bringing legitimacy to cannabis testing. “For a laboratory to receive an ISO 17025 accreditation, they must purchase their RMs from an ISO 17025 manufacturer. The best option is to purchase an ISO Guide 34 manufactured CRM,” says Ross. “It is particularly important for testing requirements, such as potency, pesticides, etc., where quantitation is expected, to use properly certified quantitative reference materials.” LGC Standards, a 175-year-old company, is one of those manufacturers that invested the time and money to achieve ISO Guide 34 accreditation and offers a spectrum of CRMs for cannabis testing.
The major advantage to using a proper CRM is an increased level of credibility. Auditors recognize the value of using a CRM which can add to the integrity of the results produced. The regular use of certified reference standards along with proper training, methodology and instrumentation, will facilitate a result that has the least amount of uncertainty and is more defendable. “The regular use of certified reference standards will help ensure products that go to market are safe to consume,” says Ross.
With regard to potency analyses, Ross has some key insights to help a laboratory better utilize CRMs. “My advice? Don’t mix the cannabinoids; labs analyzing by GC/FID have discovered that some of the cannabinoids will co-elute. Also, they have a short shelf life when mixed together,” says Ross. “Cannabinoid analysts should use GC/MS or LC/MS for their analysis or analyze the cannabinoids individually,” says Ross.
So what happens if a cannabis lab uses non-certified reference materials? Labs might save money in the short term. CRMs are slightly more expensive than a non-certified reference material, but will increase the defensibility of a lab’s data. Using a reference material created in-house or from a non-accredited vendor can lead to less-than-accurate results. A non-certified reference material has a greater chance of being made incorrectly. The publication of incorrect data damages the credibility of the testing lab and could lead to legal action against the lab from damaged parties.
One of the major challenges for the cannabis testing industry is the variation in state-to-state regulations. Ross says that Oregon’s regulations are pretty comprehensive and that other states should look to the Oregon Environmental Laboratory Accreditation Program (ORELAP) for guidance. According to Ross, ORELAP would like to see higher quality standards with legitimate traceability. Utilizing CRMs the correct way will help laboratories achieve greater accuracy.
Here are some tips for using CRMs appropriately:
Always bring your standards to room temperature before making a dilution.
Matrix matched calibration standards provide more accurate quantitation. Prepare standards in the solvent from extracted blank matrices.
Always bracket your analytical runs with continuing calibration verification standards. Proving that your instrument remained calibrated during the run gives your data more credibility.
Analytical chemists purchase CRMs for three primary uses in the testing lab:
To calibrate the instrument that will be used to perform the testing
To confirm the instruments continuing calibration throughout the analytical process
For analytical quality control or “spikes”
Typically, labs will spike known concentrations of the analytes of interest into a control sample and regular samples with the intent of testing analytical efficiency. Recoveries of analytes from the spiked control sample tell the chemist how well the analytical method is working. The spiked samples (matrix spikes) demonstrate to what extent the sample matrix (the consumable being tested) is influencing the results of the analytical procedure.
CRMs could be described as the nexus between cannabis testing results, the human element and the instrumentation used in an analysis. By using a cannabis-specific CRM, the cannabis testing community can demonstrate tangible improvements in accuracy and legitimacy.
The AOCS Annual Meeting is an international science and business forum on fats, oils, surfactants, lipids and related materials. The American Oil Chemist’s Society (AOCS) is holding their meeting in Orlando, Florida from April 30 to May 3, 2017. Last year’s meeting included discussions on best practices and the pros and cons of different extraction techniques, sample preparation, proficiency testing and method development, among other topic areas.
Posters on display for the duration of the Annual Meeting will discuss innovative solutions to test, preparing samples, discovering new compounds and provide novel information about the compounds found in cannabis. David Egerton, vice president of technical operations at CW Analytical (a cannabis testing laboratory in Oakland, CA), is preparing a poster titled Endogenous Solvents in Cannabis Extracts. His abstract discusses testing regulations focusing on the detection of the presence of solvents, despite the fact that endogenous solvents, like acetone and lower alcohols, can be found in all plant material. His study will demonstrate the prevalence of those compounds in both the plant material and the concentrated oil without those compounds being used in production.
The conference features more than 650 oral and poster presentations within 12 interest areas. This year’s technical program includes two sessions specifically designed to address cannabinoid analytics:
Lab Proficiency Programs and Reference Samples
How do you run a lab proficiency program when you cannot send your samples across state lines? What constituents do you test for when state requirements are all different? Are all pesticides illegal to use on cannabis? What pesticides should be tested for when they are mostly illegal to use? How do you analyze proficiency results when there are no standard methods? Learn about these and other challenges facing the cannabis industry. This session encourages open and active discussion, as the cannabis experts want to hear from you and learn about your experiences.
The need for high-quality and safe products has spurred a new interest in cannabinoid analytics, including sample preparation, pesticide, and other constituent testing. In this session, a diverse group of scientists will discuss developing analytical methods to investigate cannabis. Learn the latest in finding and identifying terpenes, cannabinoids, matrix effects, and even the best practice for dissolving a gummy bear.
Cynthia Ludwig, director of technical services at AOCS, says they are making great progress in assembling analytical methods for the production of the book AOCS Collection of Cannabis Analytical Methods. “We are the leading organization supporting the development of analytical methods in the cannabis industry,” says Ludwig. “Many of the contributors in that collection will be presenting at the AOCS Annual Meeting, highlighting some of the latest advances in analyzing cannabis.” The organization hopes to foster more collaboration among those in the cannabis testing industry.
In addition to oral and poster sessions, the 2017 Annual Meeting will feature daily networking activities, more than 70 international exhibitors, two special sessions, and a Hot Topics Symposia which will address how current, critical issues impact the future of the fats and oils industry.
The Hoban Law Group filed a petition on behalf of three clients against the DEA in the U.S. Court of Appeals for the Ninth District on January 13th, according to a press release. The clients represented by Hoban Law Group in the suit are Hemp Industries Association, RMH Holdings, LLC and Centuria Natural Foods, Inc. The companies are based in California, Colorado and Nevada respectively and are all active in the legal hemp trade. The press release says RMH Holdings “sources its products from industrial hemp lawfully cultivated pursuant to the Agricultural Act of 2014 (also known as the Farm Bill).”
In December, the DEA published a ‘Final Rule’ that classifies cannabis-derived extracts, such as CBD oil, in their own category with a code number to “better track these materials and comply with treaty provisions.” The announcement by the DEA ultimately serves to make any cannabis extract a Schedule 1 narcotic. “Extracts of marihuana will continue to be treated as Schedule I controlled substances,” says the document.
Bob Hoban, managing partner of Hoban Law Group says the action is clearly beyond the DEA’s authority. “This Final Rule serves to threaten hundreds, if not thousands, of growing businesses, with massive economic and industry expansion opportunities, all of which conduct lawful business compliant with existing policy as it is understood and in reliance upon the Federal Government,” says Hoban.
The lawsuit states that they want a judicial review of the DEA’s actions “on the grounds that the Final Rule is (1) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law, e.g. the CSA, the Farm Bill, and the DEA’s regulations; (2) contrary to constitutional right, power, privilege, or immunity; (3) in excess of statutory jurisdiction, authority, or limitations; and, (4) without observance of procedure required by law.” The suit also claims that the ‘Final Rule’ conflicts with other federal laws like the Data Quality Act, Regulatory Flexibility Act and Congressional Review Act.
According to Garrett Graff, associate attorney at Hoban Law Group, the entire Cannabis genus is not unlawful and the DEA is overstepping its authority. “As the Ninth Circuit found in 2003 and 2004 there are certain parts of the plant like the stalk and seed that are congressionally exempted from the Controlled Substances Act and thus the DEA’s rulemaking authority,” says Graff. “By creating a drug code for ‘marihuana extract’, the DEA is saying that they are a controlled substance, but that goes against a number of existing laws.”
The definition of ‘marihuana extract’ under the ‘Final Rule’ also references extracts containing one or more cannabinoids, which goes beyond the realm of cannabis altogether, according to Graff. “The DEA and many other sources have acknowledged and confirmed that cannabinoids can be derived from other varieties of flowers, cacao and other sources, making it virtually impossible to distinguish which cannabinoids would be subject to this drug code,” says Graff. “The DEA’s rule effectively makes the presence of cannabinoids a determinative factor of a controlled substance, which is inconsistent with what Congress has said.”
The petition filed is essentially the initiation or commencing of a lawsuit. Graff says their case is rooted in statute. “We hope to accomplish a striking of the rule, permanent injunction of the rule and for the DEA to engage in the appropriate processes and procedures when making rules in the future,” says Graff. “Alternatively, an amendment to the rule to make the definition of ‘marihuana extract’ consistent with existing law and reflect those portions and varieties of the plant which are in fact lawful could be considered.” It may still be roughly 30 days before the DEA responds with briefing and possibly an oral argument to follow on the various issues surrounding the petition, says Graff. The Ninth Circuit petition, including briefings and hearings, is likely to take at least several months.
As mentioned in Part 1, the physiological effects of cannabis are mediated by a group of structurally related organic compounds known as cannabinoids. The cannabinoids are biosynthetically produced by a growing cannabis plant and Figure 1 details the biosynthetic pathways leading to some of the most important cannabinoids in plant material.
The analytical measurement of cannabinoids is important to ensure the safety and quality of cannabis as well as its extracts and edible formulations. Total cannabinoid levels can vary significantly between different cultivars and batches, from about 5% up to 20% or more by dry weight. Information on cannabinoid profiles can be used to tailor cultivars for specific effects and allows end users to select an appropriate dose.
Routine Analysisvs. Cannabinomics
Several structurally analogous groups of cannabinoids exist. In total, structures have been assigned for more than 70 unique phytocannabinoids as of 2005 and the burgeoning field of cannabinomics seeks to comprehensively measure these compounds.¹
Considering practical potency analysis, the vast majority of cannabinoid content is accounted for by 10-12 compounds. These include Δ9-tetrahydrocannabinol (THC), cannabidiol (CBD), cannabigerol (CBG), Δ9-tetrahydrocannabivarian (THCV), cannabidivarin (CBDV) and their respective carboxylic acid forms. The cannabinoids occur primarily as carboxylic acids in plant material. Decarboxylation occurs when heat is applied through smoking, vaporization or cooking thereby producing neutral cannabinoids which are more physiologically active.
Potency Analysis by HPLC and GC
Currently, HPLC and GC are the two most commonly used techniques for potency analysis. In the case of GC, the heat used to vaporize the injected sample causes decarboxylation of the native cannabinoid acids. Derivatization of the acids may help reduce decarboxylation but overall this adds another layer of complexity to the analysis² ³. HPLC is the method of choice for direct analysis of cannabinoid profiles and this technique will be discussed further.
A sample preparation method consisting of grinding/homogenization and alcohol extraction is commonly used for cannabis flower and extracts. It has been shown to provide good recovery and precision² ³. An aliquot of the resulting extract can then be diluted with an HPLC compatible solvent such as 25% water / 75% acetonitrile with 0.1% formic acid. The cannabinoids are not particularly water soluble and can precipitate if the aqueous percentage is too high.
To avoid peak distortion and shifting retention times the diluent and initial mobile phase composition should be reasonably well matched. Another approach is to make a smaller injection (1-2 µL) of a more dissimilar solvent. The addition of formic acid or ammonium formate buffer acidifies the mobile phase and keeps the cannabinoid acids protonated.
The protonated acids are neutral and thus well retained on a C18 type column, even at higher (~50% or greater) concentrations of organic solvent² ³.
Detection is most often done using UV absorbance. Two main types of UV detectors are available for HPLC, single wavelength and diode array. A diode array detector (DAD) measures absorbance across a range of wavelengths producing a spectrum at each point in a chromatogram while single wavelength detectors only monitor absorbance at a single user selected wavelength. The DAD is more expensive, but very useful for detecting coelutions and interferences.
Chemical Constituents of Marijuana: The Complex Mixture of Natural Cannabinoids. Life Sciences, 78, (2005), pp. 539
Development and Validation of a Reliable and Robust Method for the Analysis of Cannabinoids and Terpenes in Cannabis. Journal of AOAC International, 98, (2015), pp. 1503
Innovative Development and Validation of an HPLC/DAD Method for the Qualitative and Quantitative Determination of Major Cannabinoids in Cannabis Plant Material. Journal of Chromatography B, 877, (2009), pp. 4115
Rebecca is an Applications Scientist at Restek Corporation and is eager to field any questions or comments on cannabis analysis, she can be reached by e-mail, email@example.com or by phone at 814-353-1300 (ext. 2154)
Election Day 2016 resulted in historic gains for state level cannabis prohibition reform. Voters in California, Maine, Massachusetts and Nevada chose to legalize adult use of Cannabis sp. and its extracts while even traditionally conservative states like Arkansas, Florida, Montana and North Dakota enacted policy allowing for medical use. More than half of the United States now allows for some form of legal cannabis use, highlighting the rapidly growing need for high quality analytical testing.
For the uninitiated, analytical instrumentation can be a confusing mix of abbreviations and hyphenation that provides little obvious information about an instrument’s capability, advantages and disadvantages. In this series of articles, my colleagues and I at Restek will break down and explain in practical terms what instruments are appropriate for a particular analysis and what to consider when choosing an instrumental technique.
Potency analysis refers to the quantitation of the major cannabinoids present in Cannabis sp. These compounds are known to provide the physiological effects of cannabis and their levels can vary dramatically based on cultivation practices, product storage conditions and extraction practices.
The primary technique is high performance liquid chromatography (HPLC) coupled to ultraviolet absorbance (UV) detection. Gas chromatography (GC) coupled to a flame ionization detector (FID) or mass spectrometry (MS) can provide potency information but suffers from issues that preclude its use for comprehensive analysis.
Pesticide Residue Analysis
Pesticide residue analysis is, by a wide margin, the most technically challenging testing that we will discuss here. Trace levels of pesticides incurred during cultivation can be transferred to the consumer both on dried plant material and in extracts prepared from the contaminated material. These compounds can be acutely toxic and are generally regulated at part per billion parts-per-billion levels (PPB).
Depending on the desired target pesticides and detection limits, HPLC and/or GC coupled with tandem mass spectrometry (MS/MS) or high resolution accurate mass spectrometry (HRAM) is strongly recommended. Tandem and HRAM mass spectrometry instrumentation is expensive, but in this case it is crucial and will save untold frustration during method development.
Residual Solvents Analysis
When extracts are produced from plant material using organic solvents such as butane, alcohols or supercritical carbon dioxide there is a potential for the solvent and any other contaminants present in it to become trapped in the extract. The goal of residual solvent analysis is to detect and quantify solvents that may remain in the finished extract.
Residual solvent analysis is best accomplished using GC coupled to a headspace sample introduction system (HS-GC) along with FID or MS detection. Solid phase microextraction (SPME) of the sample headspace with direct introduction to the GC is another option.
Terpene Profile Analysis
While terpene profiles are not a safety issue, they provide much of the smell and taste experience of cannabis and are postulated to synergize with the physiologically active components. Breeders of Cannabis sp. are often interested in producing strains with specific terpene profiles through selective breeding techniques.
Both GC and HPLC can be employed successfully for terpenes analysis. Mass spectrometry is suitable for detection as well as GC-FID and HPLC-UV.
Heavy Metals Analysis
Metals such as arsenic, lead, cadmium, chromium and mercury can be present in cannabis plant material due to uptake from the soil, fertilizers or hydroponic media by a growing plant. Rapidly growing plants like Cannabis sp. are particularly efficient at extracting and accumulating metals from their environment.
Several different types of instrumentation can be used for metals analysis, but the dominant technology is inductively coupled plasma mass spectrometry (ICP-MS). Other approaches can also be used including ICP coupled with optical emission spectroscopy (ICP-OES).
Rebecca is an Applications Scientist at Restek Corporation and is eager to field any questions or comments on cannabis analysis, she can be reached by e-mail, firstname.lastname@example.org or by phone at 814-353-1300 (ext. 2154)
When a cannabis sample is submitted to a lab for testing there is a four-step process that occurs before it is tested in the instrumentation on site:
It is ground at a low temperature into a fine powder;
A solution is added to the ground powder;
An extraction is repeated 6 times to ensure all cannabinoids are transferred into a common solution to be used in testing instrumentation.
Once the cannabinoid solution is extracted from the plant matter, it is analyzed using High Pressure Liquid Chromatograph (HPLC). HPLC is the key piece of instrumentation in cannabis potency testing procedures.
While there are many ways to test cannabis potency, HPLC is the most widely accepted and recognized testing instrumentation. Other instrument techniques include gas chromatography (GC) and thin layer chromatography (TLC). HPLC is preferred over GC because it does not apply heat in the testing process and cannabinoids can then be measured in their naturally occurring forms. Using a GC, heat is applied as part of the testing process and cannabinoids such as THCA or CBDA can change form, depending on the level of heat applied. CBDA and THCA have been observed to change form at as low as 40-50C. GC uses anywhere between 150-200C for its processes, and if using a GC, a change of compound form can occur. Using HPLC free of any high-heat environments, acidic (CBDA & THCA) and neutral cannabinoids (CBD, THC, CBG, CBN and others) can be differentiated in a sample for quantification purposes.
Near infrared (NIR) has been used with cannabis for rapid identification of active pharmaceutical ingredients by measuring how much light different substances reflect. Cannabis is typically composed of 5-30% cannabinoids (mainly THC and CBD) and 5-15% water. Cannabinoid content can vary by over 5% (e.g. 13-18%) on a single plant, and even more if grown indoors. Multiple NIR measurements can be cost effective for R&D purposes. NIR does not use solvents and has a speed advantage of at least 50 times over traditional methods.
The main downfall of NIR techniques is that they are generally less accurate than HPLC or GC for potency analyses. NIR can be programmed to detect different compounds. To obtain accuracy in its detection methods, samples must be tested by HPLC on ongoing basis. 100 samples or more will provide enough information to improve an NIR software’s accuracy if it is programmed by the manufacturer or user using chemometrics. Chemometrics sorts through the often complex and broad overlapping NIR absorption.
Bands from the chemical, physical, and structural properties of all species present in a sample that influences the measured spectra. Any variation however of a strain tested or water quantity observed can affect the received results. Consistency is the key to obtaining precision with NIR equipment programming. The downfall of the NIR technique is that it must constantly be compared to HPLC data to ensure accuracy.
At Eurofins Experchem , our company works with bothHPLC and NIR equipment simultaneously for different cannabis testing purposes. Running both equipment simultaneously means we are able to continually monitor the accuracy of our NIR equipment as compared to our HPLC. If a company is using NIR alone however, it can be more difficult to maintain the equipment’s accuracy without on-going monitoring.
What about Terpenes?
Terpenes are the primary aromatic constituents of cannabis resin and essential oils. Terpene compounds vary in type and concentration among different genetic lineages of cannabis and have been shown to modulate and modify the therapeutic and psychoactive effects of cannabinoids. Terpenes can be analyzed using different methods including separation by GC or HPLC and identification by Mass Spectrometry. The high-heat environment for GC analysis can again cause problems in accuracy and interpretation of results for terpenes; high-heat environments can degrade terpenes and make them difficult to find in accurate form. We find HPLC is the best instrument to test for terpenes and can now test for six of the key terpene profiles including a-Pinene, Caryophyllene, Limonene, Myrcene, B-Pinene and Terpineol.
Quality systems between different labs are never one and the same. Some labs are testing cannabis under good manufacturing practices (GMP), others follow ISO accreditation and some labs have no accreditation at all.
From a quality systems’ perspective some labs have zero or only one quality system employee(s). In a GMP lab, to meet the requirements of Health Canada and the FDA, our operations are staffed in a 1:4 quality assurance to analyst ratio. GMP labs have stringent quality standards that set them apart from other labs testing cannabis. Quality standards we work with include, but are not limited to: monthly internal blind audits, extensive GMP training, yearly exams and ongoing tests demonstrating competencies.
Maintaining and adhering to strict quality standards necessary for a Drug Establishment License for pharmaceutical testing ensures accuracy of results in cannabis testing otherwise difficult to find in the testing marketplace.
Important things to know about testing
HPLC is the most recommended instrument used for product release in a regulated environment.
NIR is the best instrument to use for monitoring growth and curing processes for R&D purposes, only if validated with an HPLC on an ongoing basis.
Quality Systems between labs are different. Regardless of instrumentation used, if quality systems are not in place and maintained, integrity of results may be compromised.
GMPs comprise 25% of our labour costs to our quality department. Quality systems necessary for a GMP environment include internal audits, out of specification investigations, qualification and maintenance of instruments, systems controls and stringent data integrity standards.
David Goldstein, co-founder and chief executive officer of PotBotics, launched a medical cannabis recommendation engine called PotBot with the goal to better inform patients to target their conditions with more accurate recommendations based on scientific research. “This is a tool to help move the market away from the thousands of strain names that are mainly just marketing or branding indicators,” says Goldstein. The medical application is designed to inform patients on peer-reviewed data, research on the treatment of their ailments with cannabis and the specific cannabinoids that are necessary for treating their condition. They began development on PotBot in October of 2014, launching the beta version to 400 users in November of 2015. On April 20th, 2016, Goldstein launched officially in the Apple Store, and the program will be available on Android in July.
Rather than focusing on strain names, PotBot focuses on the cannabinoid values to help patients gain an understanding of the correlation between which compounds might best target their condition. “This is a great tool for patients trying to familiarize themselves with what strains might work best,” says Goldstein. “For example, insomnia patients generally need cannabis with higher CBN levels, so we first educate the patient on cannabinoid ranges to shoot for and what strains might help. PotBot would recommend the strain Purple Urple because it is an indica found to have higher CBN values,” adds Goldstein. The program goes into great detail with the patient’s preferences including everything down to consumption methods so they know why it might recommend certain strains.
The recommendation tool is accessible via kiosks at dispensaries, on a desktop version for the computer as well as on the Apple Store for iPads and iPhones. “I do not see it as a way of replacing budtenders, rather supplementing them with knowledge,” says Goldstein. PotBot is designed as a tool to supplement the budtender’s understanding of cannabis, so the budtender does not need to know everything off the top of their head or recommend strains based on anecdotal information, according to Goldstein.
Goldstein’s team at PotBotics performed extensive research prior to launching PotBot, spending two years doing strain testing to develop the program. “There is currently no regulatory body [for strain classification] so we took it upon ourselves to work with the best testing laboratories for truly robust analyses and properly vetted growers to get the most valid data,” says Goldstein. “The current strain classification system and nomenclature is rather unscientific so we focus on cannabinoid values and soon we will be able to incorporate terpene profiles in the recommendation.” Moving away from the common focus on taste, smell and other qualitative values, they focus on medical attributes of cannabinoid profiles because they have the most peer-reviewed research available today.
As an OEM, the company designed the tool to work with each dispensary’s inventory, to provide recommendations for strains that a patient can access on site, however anyone can access the recommendation tool for free at PotBot.com. Goldstein’s company and their mission represent an important development in the cannabis industry; this could begin a key transition from thousands of understudied strain names to a more scientific and calculated method to treating patients’ conditions with cannabis.
Despite the title, this article is not about weight loss – it is about generating valid analytical data for quantitative analyses. In the last installment of The Practical Chemist, I introduced instrument calibration and covered a few ways we can calibrate our instruments. Just because we have run several standards across a range of concentrations and plotted a curve using the resulting data, it does not mean our curve accurately represents our instrument’s response across that concentration range. In order to be able to claim that our calibration curve accurately represents our instrument response, we have to take a look at a couple of quality indicators for our curve data:
correlation coefficient (r) or coefficient of determination (r2)
back-calculated accuracy (reported as % error)
The r or r2 values that accompany our calibration curve are measurements of how closely our curve matches the data we have generated. The closer the values are to 1.00, the more accurately our curve represents our detector response. Generally, r values ≥0.995 and r2 values ≥ 0.990 are considered ‘good’. Figure 1 shows a few representative curves, their associated data, and r2 values (concentration and response units are arbitrary).
Let’s take a closer look at these curves:
Curve A: This represents a case where the curve perfectly matches the instrument data, meaning our calculated unknown values will be accurate across the entire calibration range.
Curve B: The r2 value is good and visually the curve matches most of the data points pretty well. However, if we look at our two highest calibration points, we can see that they do not match the trend for the rest of the data; the response values should be closer to 1250 and 2500. The fact that they are much lower than they should be could indicate that we are starting to overload our detector at higher calibration levels; we are putting more mass of analyte into the detector than it can reliably detect. This is a common problem when dealing with concentrated samples, so it can occur especially for potency analyses.
Curve C: We can see that although our r2 value is still okay, we are not detecting analytes as we should at the low end of our curve. In fact, at our lowest calibration level, the instrument is not detecting anything at all (0 response at the lowest point). This is a common problem with residual solvent and pesticide analyses where detection levels for some compounds like benzene are very low.
Curve D: It is a perfect example of our curve not representing our instrument response at all. A curve like this indicates a possible problem with the instrument or sample preparation.
So even if our curve looks good, we could be generating inaccurate results for some samples. This brings us to another measure of curve fitness: back-calculated accuracy (expressed as % error). This is an easy way to determine how accurate your results will be without performing a single additional run.
Back-calculated accuracy simply plugs the area values we obtained from our calibrators back into the calibration curve to see how well our curve will calculate these values in relation to the known value. We can do this by reprocessing our calibrators as unknowns or by hand. As an example, let’s back-calculate the concentration of our 500 level calibrator from Curve B. The formula for that curve is: y = 3.543x + 52.805. If we plug 1800 in for y and solve for x, we end up with a calculated concentration of 493. To calculate the error of our calculated value versus the true value, we can use the equation: % Error = [(calculated value – true value)/true value] * 100. This gives us a % error of -1.4%. Acceptable % error values are usually ±15 – 20% depending on analysis type. Let’s see what the % error values are for the curves shown in Figure 1.
Our % error values have told us what our r2 values could not. We knew Curve D was unacceptable, but now we can see that Curves B and C will yield inaccurate results for all but the highest levels of analyte – even though the results were skewed at opposite ends of the curves.
There are many more details regarding generating calibration curves and measuring their quality that I did not have room to mention here. Hopefully, these two articles have given you some tools to use in your lab to quickly and easily improve the quality of your data. If you would like to learn more about this topic or have any questions, please don’t hesitate to contact me at email@example.com.