Review: Rapid Analysis of Δ-9-Tetrahydrocannabinol in Hair Using Direct Analysis in Real Time Ambient Ionization Orbitrap Mass Spectrometry

Emily C. Lennert





tetrahydrocannabinol, THC, hair, drug, drugs of abuse, direct analysis in real time, DART, mass spectrometry, MS, orbitrap

Article Reviewed

Duvivier, W. F.; van Beek, T. A.; Pennings, E. J.M.; Nielen, M. W.F. Rapid analysis of Δ-9-tetrahydrocannabinol in hair using direct analysis in real time ambient ionization orbitrap mass spectrometry. Rapid Communications in Mass Spectrometry. 2014, 28, 682-690.


The opinions expressed in this review are an interpretation of the research presented in the article. These opinions are those of the summation author and do not necessarily represent the position of the University of Central Florida or of the authors of the original article.


Analysis of hair is often used in forensic and clinical toxicology, due to the fact that compounds such as drugs as other intoxicants are incorporated into hair after intake. Hair analysis holds several advantages over the analysis of blood or urine, in that sampling is non-invasive, simple, and compounds may be detected in the sample for months or years as opposed to days. Longitudinal segments, i.e. the length, of hair can be analyzed to establish approximate timelines based on an average growth rate of 1.3 cm/month. Current methods of analysis primarily involve digestion of a 1-3 cm hair segment, followed by analyte extraction and analysis by gas chromatography – mass spectrometry (GC-MS) or liquid chromatography – mass spectrometry (LC-MS). Due to the length of hair required, only a rough estimate of time of intake can be determined. The authors of this study present a new method of screening for Δ-9-tetrahydrocannabinol (THC) in hair by direct analysis in real time – mass spectrometry (DART-MS). DART-MS allows for fast, direct analysis in ambient conditions without the need for extraction or other means of sample preparation, making it an ideal technique for screening of samples.

A THC standard solution was prepared from cannabis extraction. Cannabis sativa leaves and tops were crushed with a mortar, and 4.6 g of crushed material was extracted in 250 mL of methanol for 30 min. The extract was filtered and stored in a glass bottle. The resulting stock solution was determined to have a concentration of 339 ppm THC, by LC-MS/MS analysis. Secondary stock solutions were then prepared by dilution with methanol. Concentrations ranged from 0.11 to 85 ppm THC, with a constant internal standard of 25 ppm quinine. Quinine also allowed for visualization of the spiked region of hair by fluorescence under UV light.

LC-MS/MS was used as a reference method (i.e. The current analytical method)to allow for validation of the DART-MS results. Samples were prepared for LC-MS/MS by extraction. Hair was first decontaminated with dichloromethane washes. Then, a 30-50 ng of sample was cut into small pieces and placed in 5.0 μL of 1 ppm THC-d3 in methanol. The sample was then incubated in 1.0 mL of NaOH for 10 min at 95˚C, then the pH was adjusted to 2 using acetic acid. After pH adjustment, liquid-liquid extraction using 3.0 mL of n-hexane:ethyl acetate (90:10) was performed. The organic phase, i.e. the layer containing the analyte, was separated by liquid-liquid extraction then dried under a stream of nitrogen at 40˚C. After drying, the sample was reconstituted in 100.0 μL of methanol and analyzed by LC-MS/MS.

Hair samples for DART-MS were prepared one of two ways. One set of samples was obtained from chronic cannabis users, i.e. cannabis users that had used cannabis at least three times per week for the previous six months. The second set of samples was prepared by spiking of decontaminated hair samples. Diluted secondary THC solutions were sprayed onto small sections of hair using a nebulizer. This allowed for the spiking of small zones on lengths of blank hair. Hair samples were prepared for DART-MS analysis by attachment to a stainless steel mesh screen, which was then placed in a holder that moved the sample in the x-direction, e.g. left to right, at a scan speed of 0.2 mm/s.

The effect of DART gas temperature was investigated to optimize the method. Using mesh screens and hairs spiked with 2.0 μL of THC secondary solution, samples were analyzed between 100˚C and 500˚C for mesh and between 100˚C and 300˚C for hair, at 50˚C intervals. Hair was not analyzed above 300 ˚C because it began to burn at the 300˚C mark. For mesh samples, optimum signal for the target analyte, m/z 315.2319, was obtained at 200 and 250˚C. Similarly, for hair samples, optimum signal for the target analyte was obtained at 200 and 250˚C. The authors selected a gas heater temperature of 250˚C as the optimal temperature for all subsequent experiments.

Spiked hair samples were analyzed by DART-MS to determine whether the hair scan concept was viable. Samples were prepared as described above, and were scanned along the length of the hair from scalp to tip end. Reproducibility was studied by spiking of three zones on a strand of hair, repeated on eight strands to study repeatability. The method was found to be very reproducible and repeatable, with relative standard deviations below 20% for the one-sided zone deviation and 26-27% for the intensity deviation. One-sided zone deviation is the measure of deviation in one side of the length of hair that is analyzed, i.e. half the length of hair exposed to analysis. See Table 1 within the study for average deviations and relative standard deviations. Based the one-sided zone deviation, a spot size can be calculated, i.e. two times the one-sided deviation. A spot size of approximately 5 mm was determined, which would allow for the establishment of a retrospective timeline with an accuracy of approximately +/- 2 weeks, assuming average hair growth of 1.3 cm/month. To determine the sensitivity and linearity of the DART method, a range of concentrations of secondary THC solution were spiked on mesh and hair. THC solutions between 0.85 and 85 ppm were spiked onto mesh and analyzed by DART-MS, resulting in a linear calibration curve with a R2 value 0f 0.998. Correction to the relative response to the internal standard resulted in a R2 value of 0.992. The R2 value is an indicator of the linearity of the calibration; a R2 of 1 corresponds to a perfectly linear calibration. Next, THC solutions between 3.4 to 34 ppm were spiked onto blank hair samples. The initial calibration resulted in a R2 value of 0.881, which improved to 0.991 upon correction to the internal standard. A detection limit of approximately 5 ng/mg hair was determined for THC detection by DART-MS. Compared to the Society of Hair Testing (SoHT) recommended cutoff value of 0.05 ng/mg for THC, the DART-MS method using a benchtop orbitrap MS is not sensitive enough. However, the authors theorize that with a more sensitive MS, a lower detection limit can be reached. Spiked strands of hair were also analyzed by LC-MS/MS to verify results of the DART-MS method.

Samples obtained from chronic cannabis users were tested to ensure that DART-MS was capable of detecting incorporated THC in the hair strand. THC was detected in six of eleven samples tested using the DART-MS can method. Samples were also tested by LC-MS/MS to determine the concentration of THC in the hair. Samples were found to contain between 5.8 and 8.37 ng THC/mg hair. These results confirmed the estimated detection limit of DART-MS, and suggest that the DART-MS scan method presented could be applied to screening for THC in the hair of chronic cannabis users. However, for compliance with the SoHT cutoff, a more sensitive MS will be required. Hair samples were also measured before and after decontamination to confirm that incorporated THC was being detected. THC was detected in both pre- and post-decontamination samples, with only slight reduction in THC levels detected. LC-MS/MS analysis showed a 12% decrease in the level of THC post-decontamination. This shows that the majority of detected THC in the chronic cannabis user samples was incorporated THC.

Scientific Highlights

  • THC was detected in hair using DART-MS for both spiked samples and real world samples obtained from chronic cannabis users.
  • The detection limit for THC using DART- orbitrap MS was determined to be 5 ng/mg hair. This detection limit is not low enough to meet SoHT guidelines for cutoff values.
  • With a spot size of approximately 5 mm, DART-MS may be used to establish retrospective timelines with an approximate accuracy of +/- 2 weeks.


Rapid screening methods are desirable to prevent backlog of casework for confirmatory analysis methods such as LC-MS.

Potential Conclusions

DART-MS may offer a rapid screening method for the detection of THC in the hair of chronic cannabis users.


Photo by Niklas Hamann on Unsplash.