Hemp and cannabis are becoming important agricultural products, and are being increasingly used in medicinal products, cosmetics, foods, oils, and textile fibers around the world. The cannabis market is growing rapidly, mainly due to the use as therapeutics for medicial treatments by the pharmaceutical industry.
Water in cannabis and hemp impacts the determination of the potency and must be determined accurately to calculate the correct delta-9 tetrahydrocannabinol (THC) content of the plant. Although both plant types share similar characteristics, cannabis contains a higher amount of THC compared to hemp. However, to be legally classified as hemp, the United States Department of Agriculture (USDA) set a limit for the total delta-9 tetrahydrocannabinol (THC) concentration. The limit is set to contain not more than 0.3 % THC on a dry- weight basis (see definition 7 CFR Part 990 Oct 2019).1 European Union guidlines currently define a limit ≤ 0.2% for the total THC concentration on a dry-weight basis for industrial hemp (status Jan. 2022).2
To calculate the dry-weight delta-9 THC concentration, an accurate analytical method must be employed for determining the exact water content. Currently most laboratories use loss on drying methods (LOD) which measure all volatile components by heating. This, however, can overstate the water content of the sample, which in turn would lead to an incorrect delta-9 THC concentration in the dry weight, resulting in wrong classification as hemp. This could potentially lead to penalties for a farmer or processor, or the forced destruction of their product.
The purpose of this application is to demonstrate a moisture determination method for hemp and cannabis flower which is selective for water, and will also provide rapid and accurate test results.
Three methods for determining water in hemp and cannabis were evaluated:
Loss on drying is a simple weighing-based technique that removes water by heating. The equipment needed is reasonably priced, but the method can be time consuming. Nowadays, this method is commonly used in several different industries. However, the loss on drying method is not specific for water and the test results obtained can include amounts of other volatile compounds too. Depending on the conditions chosen, this could lead to an incorrect water value, which in turn would affect the accuracy of the reported dry weight THC concentration result.
Karl Fischer titration based methods are simple to run, but the equipment is a bit more expensive than that used for LOD determination. The advantage of the Karl Fischer titration method is its specificity for water. As a result, the reported water value does not include amounts of other volatile compounds. Two methods based on the Karl Fischer titration were evaluated ― coulometry with a Karl Fischer oven and a direct volumetric titration with an external extraction. The coulometric method is best suitable for samples with low water content in the range of 10 ppm to 10,000 ppm (1%) or when only little sample material is available. In contrast, the volumetric titration is used for solid and liquid samples with water contents from 0,01% to 100%. Samples for direct volumetric Karl Fischer titration must be soluble in the Karl Fischer solvent. Organic plant materials like hemp and cannabis are not suitable for direct measurement, so either an external extraction with a suitable solvent or a Karl Fischer oven method is employed. The Karl Fischer oven method is ideal for this measurement as it completely evaporates the water from the sample and directly transfers it to the titrator. Both approaches of direct volumetric titration after external extraction and the Karl Fischer oven method have been examined. Water content of finely ground hemp flower samples was determined using the methods described below. For any hemp or cannabis analysis, proper preparation is important to provide a representative and well homogenized sample. A method of sample homogenization that is frequently applied in the industry is cryogenic ball milling. This thoroughly homogenizes the sample and leads to a particle size of 100 µm or smaller.
The water content of a hemp sample was determined by coulometric Karl Fischer titration combined with a Karl Fischer oven. A temperature ramp was run prior to the analysis for evaluating the optimum temperature at which the water is completely and efficiently released without decomposition of the sample. The optimal temperature for the sample used was determined to be 150 °C. Samples were weighed into sealed vials for use in the Karl Fischer oven. An empty vial was used as a blank to determine any water which may have adhered to the vial. The value obtained for the blank vial was subtracted from each sample’s value as determined by the instrument.
The sample was analyzed in quintuplicate, and the measured values were averaged to obtain the result (see results with 2 different coulometric reagents, CombiCoulomat and Anolyte, in Tables 5 and 6).
1 g hemp was extracted with 25 g of methanol by stirring in septum sealed vials. Different extraction conditions (extraction times and temperature) were applied. The water value of the methanol determined was used as a blank value for use in the final calculations. The solid hemp was allowed to settle, and an exact weight aliquot of the mixture (methanol/extracted water) was taken using a syringe, and injected into the titration cell of a volumetric Karl Fischer titrator. The exact sample weight was determined by back weighing. The titrator then measured the water content of the injected sample. The exact value of water content of the hemp sample was then calculated using the following equation:
W1= [W3 x (m1+m2) – W2 x m2]/m1
|Where:||W1 is the result in %|
|W2 is the % water of the methanol used for extraction|
|W3 is the % water determined for the extracted methanol aliquot|
|m1 is the mass of the sample extracted|
|m2 is the mass of the extraction methanol|
The results of the analysis were found to be insufficiently reproducible and highly dependent on the chosen extraction conditions. And hence are not presented here in detail. The external extraction technique was found to be disadvantageous in comparison to the KF oven technique and thereby cannot be recommended for water determination in hemp.
In this experiment, the water content of the sample was determined by loss on drying until a constant mass was reached. The sample was heated for 2 h at a temperature of 150 ˚C. The weight of the sample was determined before and after heating, to calculate the weight lost during the experiment - the loss on drying. The sample was analyzed in duplicate and the values were averaged to obtain the result (see results in Table 7).
The Karl Fischer oven method with coulometry was compared to the loss on drying method (see Table 8). Results for the latter (LOD) were about 35% higher than for the Karl Fischer oven method. This requires to consider the measument of other volatile compounds by the loss on drying method to avoid wrong water content results.
Water content of hemp samples was determined using Karl Fischer titration techniques (coulometry with oven; volumetric titration with external extraction) and compared to loss on drying.
The Karl Fischer coulometric titration in combination with a Karl Fischer oven provides reproducible results. It prevents an overestimation of water content caused by volatile compounds, as to be considered for the loss on drying method. In addition it requires only a small amount of sample and reagent. The volumetric Karl Fischer titration with external extraction did not produce reproducible results and is therefore not recommended. However, the volumetric Karl Fischer method in combination with a Karl Fischer oven can be employed, but since volumetry is not as sensitive as coulometry, the sample size needs to be increased to get reliable results.
Therefore, it is recommended to use the Karl Fischer oven method with coulometry for water determination in hemp and cannabis to achieve the most accurate results. This enables the exact and precise calculation of the dry weight delta-9 THC concentration.
See also our Cannabis Testing page at SigmaAldrich.com/cannabis