Tag Archives: removal

Solvent Remediation – The Last Step for Safe, Clean Hemp Extraction

By Tom Bisbee
No Comments

Botanical extraction is not specific to cannabis and hemp, and it is anything but new. Rudimentary forms of plant extraction have existed throughout history and evolved with high-tech equipment and scientific procedures for use in pharmaceuticals, dietary supplements and botanicals.

In food production, examples of hydrocarbon extraction processes are commonplace. Nut, olive and vegetable oil production use solvents to extract the oils. Decaffeinated coffee uses hydrocarbon extraction to remediate the caffeine, and making sugar from beets, or beer from hops, also requires solvents.

As such, the FDA has set guidelines for the amount of residual solvents considered safe for consumers to ingest. Yet, without FDA guidance in cannabis and hemp, many products aren’t being tested against these standards, and consumers will ultimately pay the price.

Understanding solvent remediation technology and processes

If we use ethanol extraction as an example, the extraction process is relatively simple. First, we soak the biomass in denatured or food-grade ethanol, ending up with a final solution that is 90-95% solvent. Then, we perform a bulk removal of the solvents, which takes out most, but not all, of the solvent. The next and final step should be to strip the remaining solvents from the extract entirely.

Stripping remaining solvents in bulk requires the right equipment.

But, in order to do so effectively, you need the right equipment, and unfortunately, this is where many producers fall short. Many producers use a vacuum oven to apply heat while reducing the headspace pressure to lower the solvent’s boiling point and evaporate it off.

However, it’s a static environment in a vacuum oven, which means the material is stagnant. So, the process may effectively remove the solvents close to the surface, but solvents deep inside the material tend to get trapped without some type of agitation or mixing.

The appropriate final step to complete solvent remediation is wipe-film distillation, which feeds small volumes into a column, which is then wiped into a very thin film and heated under vacuum pressure. Although the equipment necessary is costly, this last step removes any residual solvents from the product to create a safe, effective and consumable product.

Residual solvents present huge risks

As stated, many of the same solvents used in cannabis and hemp extraction have been considered safe in food production for decades. Reviewing chemical data sheets, many of the acceptable limits on solvents were determined for ingestion, which is fine for edibles and tinctures, but many cannabis and hemp products are intended for inhalation or vaporization.

Just a few of the dozens of various products types on the market today, extracted with a variety of different solvents

Unfortunately, some solvents can have negative health impacts, especially for those using cannabis or hemp for medical purposes or with compromised immune systems. Plus, as a therapeutic and recreational substance, consumers may be consuming more than the recommended amount, as well as using the products several times a day. Unfortunately, long-term exposure or repeated inhalation of these residual solvents hasn’t been thoroughly researched.

For example, inhaling ethyl alcohol (ethanol) can irritate the nose, throat and lungs. Extended exposure can cause headaches, drowsiness, nausea, vomiting and unconsciousness. Repeated exposure can affect the liver and nervous system.

In the food industry, hexane is approved for extracting spices or hops, and this solvent is widely used in cannabis and hemp extraction. However, if used in an inhalable product, chronic exposure to hexane could be detrimental, with symptoms including numbness in the extremities, weakness, vision problems and fatigue.

Consumers deserve transparency

In the industry’s earliest days, companies were tight-lipped about their processes, the chemicals they used and how they removed them. Everyone thought they had the “secret sauce” and didn’t want to share their approach. Today, companies are more open about what they use, how they process it and providing that necessary transparency.

Lack of quality and consistent regulations in these industries creates confusion for the consumers and loopholes for producers. Some producers test for everything under the sun, and some producers know exactly which labs will pass their products, regardless of test results.

While the regulatory bodies are distracted by the amount of THC that might linger in products, getting sick is overshadowed by the risk of getting high. In the meantime, consumers are left to their own devices to determine which products are safe and which are not.

Although testing mandates and regulations will help clean up the industry, until then, consumers need to demand full-panel COAs that not only show cannabinoid potency but also accurately display the test results for residual solvents, pesticides and heavy metals.

Pesticide Remediation by CPC

By Arpad Konczol, PhD
No Comments

Like any other natural product, the biomass of legal cannabis can be contaminated by several toxic agents such as heavy metals, organic solvents, microbes and pesticides, which significantly influence the safety of the end products.

Let’s just consider the toxicological effects. Since cannabis products are not only administered in edible forms but also smoked and inhaled, unlike most agricultural products, pesticide residue poses an unpredictable risk to consumers. One example is the potential role of myclobutanil in the vape crisis.

Unfortunately, federal and state laws are still conflicted on cannabis-related pesticides. Currently, only ten pesticide products have been registered specifically for hemp by the U.S. Environmental Protection Agency. So, the question arises what has to be done with all pf the high-value, but also contaminated cannabis, keeping in mind that during the extraction processes, not only the phytocannabinoids get concentrated but the pesticides as well, reaching concentrations up to tens or hundreds of parts per million!

Currently, there are three different sets of rules in place in the regulatory areas of Oregon, California and Canada. These regulations detail which pesticides need to be monitored and remediated if a certain limit for each is reached. Because the most extensive and strict regulations are found in Canada, RotaChrom used its regulations as reference in their case study.

Centrifugal Partition Chromatographic (CPC) system

To illustrate that reality sometimes goes beyond our imagination, we evaluated the testing results of a THC distillate sample of one of our clients. This sample contained 9 (!) pesticides, of which six levels exceeded the corresponding action limits. The most frightening, however, regarding this sample, is that it contained a huge amount of carbofuran, a category I substance. It is better not to think of the potential toxicological hazard of this material…

The CPC-based purification of CBD is a well-known and straightforward methodology. As the elution profile on the CPC chromatogram of a distillate shows, major and minor cannabinoids can be easily separated from CBD. At RotaChrom, this method has been implemented at industrial-scale in a cost effective and high throughput fashion. In any case, the question arises: where are the pesticides on this chromatogram? To answer this, we set ourselves the goal to fully characterize the pesticide removing capability of our methodologies.

Our results on this topic received an award at the prestigious PREP Conference in 2019. The ease of pesticides removal depends on the desired Compound of Interest.

Here is a quick recap on key functionalities of the partition chromatography.

  • Separation occurs between two immiscible liquid phases.
  • The stationary phase is immobilized inside the rotor by a strong centrifugal force.
  • The mobile phase containing the sample to be purified is fed under pressure into the rotor and pumped through the stationary phase in the form of tiny droplets (percolation).
  • The chromatographic column in CPC is the rotor: cells interconnected in a series of ducts attached to a large rotor
  • Simple mechanism: difference in partition

Let’s get into the chemistry a bit:

The partition coefficient is the ratio of concentrations of a compound in a mixture of two immiscible solvents at equilibrium. This ratio is therefore a comparison of the solubilities of the solute in these two liquid phases.

The CPC chromatogram demonstrates the separation of Compounds of Interest based on their unique partition coefficients achieved through a centrifugal partition chromatography system.

CPC can be effectively used for pesticide removal. About 78% of the pesticides around CBD are very easy to remove, which you can see here:

In this illustration, pesticides are in ascending order of Kd from left to right. CBD, marked with blue, elutes in the middle of the chromatogram. The chart illustrates that most polar and most apolar pesticides were easily removed beside CBD. However, some compounds were in coelution with CBD (denoted as “problematic”), and some compounds showed irregular Kd-retention behavior (denoted as “outliers”).

If pesticides need to be removed as part of THC purification, then the pesticides that were problematic around CBD would be easier to remove and some of the easy ones would become problematic.

To simulate real-world production scenarios, an overloading study with CBD was performed, which you can see in the graph:

It is easy to see on the chromatogram that due to the increased concentration injected onto the rotor, the peak of CBD became fronting and the apparent retention shifted to the right. This means that pesticides with higher retention than CBD are more prone to coelution if extreme loading is applied.

To be able to eliminate problematic pesticides without changing the components of the solvent system, which is a typical industrial scenario, the so-called “sweet spot approach” was tested. The general rule of thumb for this approach is that the highest resolution of a given CPC system can be exploited if the Kd value of the target compounds fall in the range of 0.5-2.0. In our case, to get appropriate Kd values for problematic pesticides, the volume ratio of methanol and water was fine-tuned. Ascending mode was used instead of descending mode. For the polar subset of problematic pesticides, this simple modification resulted in an elution profile with significantly improved resolution, however, some coelution still remained.

In the case of apolar pesticides, the less polar solvent system with decreased water content in ascending mode provided satisfactory separation.

Moreover, if we focus on this subset in the three relevant regulatory areas, the outcome is even more favorable. For example, myclobutanil and bifenazate, dominant in all of the three regulatory regions, are fully removable in only one run of the CPC platform.

Based on these results, a generic strategy was created. The workflow starts with a reliable and precise pesticide contamination profile of the cannabis sample, then, if it does not appear to indicate problematic impurity, the material can be purified by the baseline method. However, if coeluting pesticides are present in the input sample, there are two options. First, adjusting the fraction collection of the critical pesticide can be eliminated, however the yield will be compromised in this case. Alternatively, by fine-tuning the solvent system, a second or even a third run of the CPC can solve the problem ultimately. Let me add here, that a third approach, i.e., switching to another solvent system to gain selectivity for problematic pesticides is also feasible in some cases.

In review, RotaChrom has conducted extensive research to analyze the list of pesticides according to the most stringent Canadian requirements. We have found that pesticides can be separated from CBD by utilizing our CPC platform. Most of these pesticides are relatively easy to remove, but RotaChrom has an efficient solution for the problematic pesticides. The methods used at RotaChrom can be easily extended to other input materials and target compounds (e.g., THC, CBG).

extraction equipment

THC Remediation of Hemp Extracts

By Darwin Millard
1 Comment
extraction equipment

Remediation of delta-9 tetrahydrocannabinol (d9-THC) has become a hot button issue in the United States ever since the Drug Enforcement Agency (DEA) released their changes to the definitions of marijuana, marijuana extract, and tetrahydrocannabinols exempting extracts and tetrahydrocannabinols of a cannabis plant containing 0.3% or less d9-THC on a dry weight basis from the Controlled Substances Act. That is because, as a direct consequence, all extracts and tetrahydrocannabinols of a cannabis plant containing more than 0.3% d9-THC became explicitly under the purview of the DEA, including work-in-progress “hemp extracts” that because of the extraction process are above the 0.3% d9-THC limit immediately upon creation.

The legal ramifications of these changes to the definitions on the “hemp extracts” marketplace will not be addressed. Instead, this article focuses on the amount of d9-THC that is available in the plant material prior to extraction and tracks a “hemp extract” from the point it falls out of compliance to the point it becomes compliant again and stresses the importance of accurate track-n-trace protocols at the processing facility. The model developed to support this article was intended to be academic and was designed to follow the d9-THC portion of a “hemp extract” through the lifecycle of a typical CO2-based extract from initial extraction to THC remediation. A loss to the equipment of 2% was used for each step.

Initial Extraction

For this exercise, a common processing scenario of 1000 kg of plant material at 10% cannabidiol (CBD) and 0.3% d9-THC by weight was modeled. This amount, depending on scale of operations, can be a facility’s total capacity for the day or the capacity for a single run. 1000 kg of plant material at 0.3% d9-THC has 3 kg of d9-THC that could be extracted, purified, and diverted into the marketplace. CO2 has a nominal extraction efficiency of 95%, meaning some cannabinoids are left behind in the plant material. The same can be said about the recovery of the extract from the equipment. Traces of extract will remain in the equipment and this little bit of material, if unaccounted for, can potentially open an operator up to legal consequences. Data for the initial extraction is shown in Image 1.

Image 1: Summary Data Table for Typical CO2-based Extraction of Phytocannabinoids

As soon as the initial extract is produced it is out of compliance with the 0.3% d9-THC limit to be classified as a “hemp extract”, and of the 3 kg of d9-THC available, the extract contains approx. 2.8 kg, because some of the d9-THC remains in the plant material and some is lost to the equipment.

Dewaxing via Winterization and Solvent Removal

Dewaxing a typical CO2 extract via winterization is a common process step. For this exercise, a wax content of 30% by weight was used. A process efficiency of 98% was attributed to the wax removal process and it was assumed that 100% of the loss can be accounted for in the residue recovered from the equipment rather than in the removed waxes. Data for the winterization and solvent recovery are shown in Image 2 and 3.

Image 2: Summary Data Table for Typical Winterization of a CO2 Extract
Image 3: Summary Data Table for Solvent Removal from a CO2 Extract

Two things occur during winterization and solvent removal, non-target constituents are removed from the extract and there is compounded loss from multiple pieces of process equipment. These steps increase the concentration of the d9-THC portion of the extract and produce two streams of noncompliant waste.

Decarboxylation & Devolatilization

Most cannabinoids in the plant material are in their acid form. For this exercise, 90% of the cannabinoids were considered to be acid forms. Decarboxylation is known to produce a mass difference of 87.7%, i.e. the neutral forms are 12.3% lighter than the acid forms. Heat was modeled as the primary driver and a process efficiency of 95% was used for the conversion rate during decarboxylation. To simplify the model, the remaining 5% acidic cannabinoids are presumed destroyed rather than degraded into other compounds because the portion of the cannabinoids which get destroyed versus degrade into other compounds varies from process to process.

Devolatilization is the process of removing low-molecular weight constituents from an extract to stabilize it prior to distillation. Since the molecular constituents of cannabis resin extracts vary from variety to variety and process to process, the extracts were assumed to consist of 10% volatile compounds. The model combines the decarboxylation and devolatilization steps to account for complete decarboxylation of the available acidic cannabinoids and ignores their weight contribution to the volatiles collected during devolatilization. Destroyed cannabinoids result in an amount of loss that can only be accounted for through a complete mass balance analysis. Data for decarboxylation and devolatilization are shown in Image 4.

Image 4: Summary Data Table for Decarboxylation and Devolatilization of a CO2 Extract

As the extract moves along the process train, the d9-THC concentration continues to increase. Decarboxylation further complicates traceability because there is both a known mass difference associated with the process and an unknown mass difference that must be calculated and justified.


A two-pass distillation was modeled. On each pass a portion of the extract was removed to increase the cannabinoid concentration in the recovered material. Average data for distilled “hemp extracts” was used to ensure the model did not over- or underestimate the concentration of the cannabinoids in the distillate. The variables used to meet these data constraints were derived experimentally to match the model to the scenario described and are not indicative of an actual distillation. Data for distillation is shown in Image 5.

Image 5: Summary Data Table for Distillation of a Decarboxylated and Devolatilized Extract

After distillation, the d9-THC concentration is shown to have increased by 874% from the original concentration in the plant material. Roughly 2.2 kg of the available 3 kg of d9-THC remains in the extract, but 0.8 kg of d9-THC has either ended up in a waste stream or walking out the door.

Chromatography – THC Remediation Step 1

Chromatography was modeled to remove the d9-THC from the extract. Because there are several systems with variable efficiency rates at being able to selectively isolate the d9-THC peak from the eluent stream, the model used a 5% cut-off on the front-end and tail-end of the peak, i.e. 5% of the material before the d9-THC peak and 5% of the material after the d9-THC peak is assumed to be collected along with the d9-THC. Data for chromatography is shown in Image 6.

Image 6: Summary Data Table for d9-THC Removal using Chromatography

After chromatography, a minimum of three products are produced, compliant “hemp extract”, d9-THC extract, and noncompliant residue remaining in the equipment. The d9-THC extract modeled contains 2.1 kg of the available 3 kg in the plant material, and is 35% d9-THC by weight, an increase of 1335% from the distillation step and 11664% from the plant material.

CBN Creation – THC Remediation Step 2

For this exercise, the d9-THC extract was converted into cannabinol (CBN) using heat rather than cyclized into d8-THC, but a similar model could be used to account for this scenario. The conversion rate of the cannabinoids into CBN through heat degradation alone is low. Therefore, the model assumes half of the available cannabinoids in the d9-THC extract are converted to CBN. The entirety of the remaining portion of the cannabinoids are assumed to convert to some form of degradant rather than a portion getting destroyed. Data for THC destruction is shown in Image 7.

Image 7: Summary Data Table for THC Destruction through Degradation into CBN

Only after the CBN cyclization step has completed does the product that was the d9-THC extract become compliant and classifiable as a “hemp extract.”

Image 8: Summary Data Table for Reconciliation of the d9-THC Portion of the Hemp Extract

Throughout the process, from initial extraction to the final d9-THC remediation step, loss occurs. Of the 3 kg of d9-THC available in the plant material only 2.1 kg was recovered and converted to CBN. 0.9 kg was either lost to the equipment, destroyed in the process, attributable to the mass difference associated with decarboxylation, or was never extracted from the plant material in the first place. All of these potential areas of product loss should be identified, and their diversion risk fully assessed. Not every waste stream poses a risk of diversion, but some do; having a plan in place to handle waste the DEA considers a controlled substance is essential. Without a track-n-trace program following the d9-THC and identifying the potential risk of diversion would be impossible. The point of this is not to instill fear, instead the intention is to shed light on a very real issue “hemp extract” producers and state regulators need to understand to protect themselves and their marketplace from the DEA.