The Quantitative Determination of Phytocannabinoids in Hemp Oils Using HPLC with UV Detection

Published on: 
Cannabis Science and Technology, June/July 2018, Volume 1, Issue 2

Medical marijuana generally possesses high levels of the psychotropic tetrahydrocannabinol (Δ9-THC) and lower levels of the nonpsychotropic cannabidiol (CBD). Pain mitigation and reduced severity of nausea and seizures are just a few of the therapeutic benefits reported by medical cannabis patients. Conversely, hemp contains high levels of CBD and lower levels (generally less than 0.3%) of THC. CBD has been reported to reduce or eliminate pain, stress, depression, inflammation, and headaches.

To complicate matters, there is evidence that a combination of CBD, a host of other minor cannabinoids, and a complex array of terpenoids may provide the most benefit—called the entourage effect (1). Both free and esterified sterols and triterpenes have been identified in cannabis and hemp, among which β-sitosterol and β-amyrin have been shown to have antibacterial properties (2). CBD-rich oil has become increasingly popular and is administered via sublingual drops, gel capsules, or as a topical ointment.

The main source of CBD-rich oil is industrial hemp. Hemp is considered a rustic plant because it is frost resistant, adapts to poor soil, reproduces easily, and does not require chemical fertilizers, pesticides, herbicides, or fungicides to thrive. A hemp crop tends to resist mildew and requires less water than cotton. Hemp textiles are considered softer than cotton.

CBD oil is derived as concentrate from carbon dioxide or butane extraction of hemp, sometimes followed by steam distillation or ethanol distillation for purification. The Farm Bill of 2014 distinguishes hemp from marijuana, yet interpreting the law is difficult in that “CBD oil” may be classified as marijuana. Various forms of hemp oil may be purchased online.

The U.S. Food and Drug Administration (FDA) has issued warning letters against unproven claims to some companies that market products containing CBD. As part of these actions, the FDA has determined the cannabinoid content of some hemp products and many were found to contain levels of CBD that are very different from the label claim. It is important to note that such products are not approved by the FDA for the diagnosis, cure, mitigation, treatment, or prevention of any disease (3).

The literature indicates more than 90 phytocannabinoids in cannabis and hemp (4). Only a handful are common targets of analysis for medical potency because they are the most abundant. Like cannabis, hemp oil may be analyzed easily and effectively for its cannabinoid content. This work presents the quantitative determination of 11 important cannabinoids, including the predominate CBD (Figure 1), in various hemp oils. (See upper right for Figure 1, click to enlarge; Figure 1: Cannabinoids found in hemp and marijuana.)

Standard Curves

Using a comprehensive mixture of 11 cannabinoids (Shimadzu Part #220-91239-21; 250 µg/mL), standard curves were prepared for each target analyte with a minimum acceptable correlation coefficient (R2) of 0.999 over six standard levels. A linear dynamic range was established at 0.5–100 mg/L (ppm) in each analyte.


Instrument and Chromatographic Conditions

Instrument: Shimadzu LC-2030C UV (Cannabis Analyzer for Potency)

Column: Shimadzu NexLeaf CBX for Potency (150 mm x 4.6 mm, 2.7-μm dp)

Guard column: Shimadzu NexLeaf CBX Guard

Mobile-phase A: 0.09% phosphoric acid in water

Mobile-phase B: 0.09% phosphoric acid in acetonitrile

Gradient: B concentration 70% (initial) →95% (8 min)


Flow rate: 1.5 mL/min

Column temperature: 35 °C

Injection volume: 5 μL

Detection: 220 nm

Hemp Oil Sample Preparation

Hemp oils are typically rich in CBD, with relatively minor concentrations of other cannabinoids. All cannabinoid targets have a linear dynamic range, above which the detector response ceases to be linear with concentration. Accurate quantitation relies on the detector response to the analyte lying within the calibration range. Therefore, two dilution factors were used, depending on the quantitative goal. One dilution factor yielded the appropriate detector sensitivity to the array of minor cannabinoids. A second, higher dilution factor was established for the most accurate quantitation of the major CBD component so that it’s response was within the established quantitative dynamic range established for that analyte. In practice, it was found that the two approaches yielded quantitative values for CBD that agreed within 0.2%.

A. Quantitative Total Cannabinoids

  1. Add 400 µL of isopropanol to a 2-mL glass vial
  2. Add 10 µL of hemp oil sample and completely dissolve
  3. Agitate for 30 s
  4. Add 400 µL of methanol to the mixture
  5. Agitate for 30 s
  6. Filter the mixture through a 0.2-µm PTFE syringe filter into a high performance liquid chromatography (HPLC) vial
  7. (Note: Total dilution factor 81X)

B. Quantitative CBD Only

  1. Add 800 µL of methanol to a 2-mL glass vial
  2. Add 200 µL of the part A mixture
  3. Agitate for 30 s
  4. (Note: Total dilution factor 405X)

Five hemp oils were tested in this study; they were purchased from various mail-order vendors. The appearance and label information for three of the five appear in Figure 2, referenced as black, blue, and green. (See upper right for Figure 2, click to enlarge; Figure 2: Hemp oil 1, black label, label claim: 23 mg per serving; 100 servings per 100 mL; calculation of label claim: 23,000 µg/mL or 2.3%. Hemp oil 2, blue label, 500 mg per 30 mL, calculation of label claim: 16,666 µg/mL or 1.7%. Hemp oil 3, green label, 15 mg per 1 serving per 0.5 mL = 15 mg/0.5 mL, calculation of label claim: 30,000 µg/mL or 3.0%.) The two other samples tested but not pictured are referred to as red and yellow.

Qualitative Analysis of Hemp Oils

Chromatograms for hemp oils 1 (black), 2 (blue), and 3 (green) appear in Figure 3. (See upper right for Figure 3, click to enlarge; Figure 3: (a) Hemp oil 1 (black), total cannabinoids (81x dilution). (b) Hemp oil 1 (black), CBD only (405x dilution). (c) Hemp oil 2 (blue), total cannabinoids (81x dilution). (d) Hemp oil 2 (blue), CBD only (405x dilution). (e) Hemp oil 3 (green), total cannabinoids (81x dilution). (f) Hemp oil 3 (green), CBD only (405x dilution).) Peak labels appear for only those cannabinoids identified in the sample.

Quantitative Results Summary for Hemp Oils

The astute reader may correctly ask about the noncannabinoid constituents in the sample and how they may influence the chromatography. As the sample preparation presented here is essentially a simple dilution of the oil sample into injectable form, nothing is done to separate cannabinoids from terpenes, phenolics, or pigment compounds. Under the reversed-phase chromatography conditions used here, any pigment compounds emerge early in the chromatography.

The sum total terpene content of hemp and cannabis is 1–3% (2), comprising possibly hundreds of terpenoids. As for chemistry, many triterpenes and sesquiterpenes have similar polarities to the cannabinoids and are expected to retain on the analytical column to similar degrees as the cannabinoids. Within the detection limits of this work, any terpenoids are undetectable, essentially lost in the baseline.


Tables I and II summarize the quantitative findings for the samples studied. Table I reflects the accurate quantitation of CBD using the higher dilution factor (405X). (See upper right for Table I, click to enlarge; Table I: Summary of CBD quantitative determination for five hemp oils.) Table II reflects the quantitation of the comprehensive target list. (See upper right for Table II, click to enlarge; Table II: Summary of total cannabinoids quantitative determination.)

As a general sample observation, hemp oils 1 (black) and 2 (blue) exhibited a transparent, weak-yellow or green coloration. Our assumption was that each of these is a product of multistep purification after extraction; for example, carbon dioxide or butane extraction followed by steam distillation. Notably, hemp oil 3 (green) was opaque brown or green and gritty in appearance. It also had the most intense smell—a distinctly “earthy” odor. Accordingly, our assumption was that the sample was the result of crude extraction only, with no further refinement.

It is important to note that it has been reported in the literature that the whole plant can be more beneficial to the consumer because it contains not only the cannabinoids, but also an array of terpenes providing a synergistic whole plant benefit. The whole plant can also provide essential fatty acids, plant sterols for lowering cholesterol, and antioxidants chlorophyll and vitamin E.

Hemp oils 1 (black) and 2 (blue) showed high ratios of CBD to total cannabinoids, both at 92%, and the lowest quantity of other cannabinoids. This finding supported the assumption, along with transparency and color, that these oils were the more highly purified samples. Both samples also tested close to label claim at 95% and 92%, respectively.

Hemp oil 3 (green) revealed the highest content of CBD and total cannabinoids, yet exhibited the lowest ratio of CBD to total cannabinoids (59%). This observation is consistent with the assumption that its crude appearance reflected the least amount of post-extraction purification. Although its CBD percent of label claim tested the lowest (81%), this sample did contain the highest level of CBD compared to all other oils tested.

Hemp oils 4 (red) and 5 (yellow) tested higher than label claim at 122% and 200%, respectively. The observation is consistent with FDA findings for CBD products, perhaps calling into question the type and accuracy of testing used to justify label claims.

In summary, all samples contained less Δ9-THC than the amount allowed by law (0.3%). Also, all samples showed an array of other cannabinoids, but the minor component, THC-V, was not detected in any of the hemp oil samples. From a quality control point of view, two samples were within a reasonable range of the label claim at ±10%. One sample was well below label claim and two other samples were well above the label claims, one by as much as 200%. When purchasing CBD oils, one should consider

  • label claim,
  • actual concentration,
  • the quality control from batch to batch,
  • other cannabinoids of importance,
  • whole plant complexity, and
  • the selling price.


  1. E.B, Russo, Br. J. Pharmacol. 163, 1344–1364 (2011).
  2. C.M. Andre, J.F. Hausman, and G. Guerriero, “Cannabis Sativa: the Plant of the Thousand and One Molecules,” Frontiers in plant science, (2016).
  3. US Food and Drug Administration, Warning Letters: (FDA, Rockville, Maryland).
  4. M.A. ElSohly and D. Slade, Life Sci. 78, 539–548 (2005).

Craig Young, MS, is the HPLC product manager for Shimadzu Scientific Instruments in Columbia, Maryland. Bob Clifford, PhD, is the general manager for Shimadzu Scientific Instruments. Direct correspondence to:

How to Cite This Article

C. Young and B. Clifford, Cannabis Science and Technology 1(2), 38-43 (2018).