The current investigation has demonstrated a practical solventless method for concentrating phytocannabinoids and terpenoids as they appear in fresh inflorescences, thus providing less complex base materials for subsequent usage with the ability to maintain the acid cannabinoids or with the option to pursue subsequent decarboxylation to neutral cannabinoids. The degree of concentration of phytocannabinoids produced here is noteworthy in comparison to traditional sieving as reported in past research on hashish production with yield of 40% THC from the chemovar “Skunk #1” or a 50–55% THC putative upper limit yield with high-tech sieving [11]. Similarly, the achieved concentrations are similar to those realized in solvent extractions (vide infra).

The preservation of lower molecular weight monoterpenoids demonstrated is of particular interest in comparison to prior studies that document significant wastage with drying and curing [8]. Such findings have been corroborated in a more recent study in which fresh varieties expressed higher monoterpene content while dried samples demonstrated lower concentrations after a loss of the lower molecular weight compounds with lower boiling points [12]. While the differences between fresh and dried preparation terpenoid concentrations observed here are less prominent than in the study of Ross et al. [8], this can be attributed to the advanced drying technology regimen applied herein, employing low ambient temperature (vide infra, Methods Section 4.6).

The question remains whether such efforts to harness potential synergy of the cannabis entourage are necessary or desirable for the cannabis consumer, irrespective of whether such use be medical or adult. Two recent studies have demonstrated a lack of interaction among common terpenoids on CB₁ receptors [13,14] but without accounting for other mechanisms of action that might produce such synergy, whether by boosting therapeutic effects or reducing adverse events attributable to THC [2]. A recent study did show cannabimimetic effects by common cannabis terpenoids [15] but at concentrations at or above 100 µM. It is highly likely that consumers distinguish differences in the subjective experience that they experience from different cannabis chemovars [9]. This, in turn, may account for the thousands of varieties available in legal and clandestine markets. Contemporaneously, some objective evidence for corroboration of such synergistic entourage effects is emerging [9]. One example is the preliminary finding that limonene reduced anxiety from THC inhalation in a dose-responsive manner in humans in a randomized controlled trial setting [16]. Addition experiments underway may further elucidate phytocannabinoid/terpenoid interactions in randomized controlled trials.

The technique presented here may represent an advantage in that various secondary extraction techniques can exacerbate the problem of monoterpenoid wastage. Supercritical CO₂ extraction has been very attractive for its efficiency in preserving cannabinoids and its avoidance of solvent residues, but as it is normally performed in a single pass, this technique yields markedly reduced monoterpenoid fractions in favor of sesquiterpenoids, as illustrated (Figure 10), in which phytocannabinoid concentration is increased from 20.5 to 71.6% (Extraction ratio: 3.5×) from flower to extract, while terpenoid total rises only from 2.2 to 2.4% (Extraction ratio: 1.1×) and with wastage of limonene in particular. This is in marked contrast to the preservation and concentration of limonene in the dry ice kief process (Figure 7 and Figure 9). Given the ability of limonene to reduce THC-associated anxiety and to elevate mood [2,16], a loss of clinical efficacy is certainly possible. Similarly, myrcene concentrations reduced by supercritical CO₂ extraction (Figure 10) are enhanced in dry ice kief. Myrcene is noted to have anti-inflammatory and sedative effects that may be clinically valuable in various conditions [2].

Monoterpenoid losses have been similarly documented [17] in a study in which a supercritical CO₂ extracted concentrated cannabinoids and sesquiterpenoids while reducing the final values for monoterpenoids, with the conclusion that such products have significantly altered biochemical profiles than their flower source with different fragrances and flavors as well.

Such problems of CO₂ extraction may be avoided by eliminating the temptation to accomplish the task in one procedure; rather, separate dedicated passes for phytocannabinoid and terpenoid fractions may be preferable, with a subsequent mixture of the two runs. Another alternative was demonstrated with focused ultrasound extraction (FUSE) in which optimal conditions for extraction were ascertained to be 100 bar, 35 °C and 1 mL/min with no co-solvent for terpenes and with 20% ethanol as co-solvent for cannabinoids [18]. FUSE was more efficient than supercritical fluid extraction, with an 80% yield of essential oils (EOs) and cannabinoids (CBs) in the first pass.

Cold ethanol provides an alternative that actually is more efficient than CO₂ in terpenoid extraction. Simple filtration of the resulting material obviates the need for a separate “winterization” step in order to remove the waxy cell wall ballast from flowers (that adversely affects taste and impairs shelf-life) [1]. Tinctures result directly, or the material may alternatively be roto-vaped to purge ethanol for concentrate production and vaporization, etc.

Steam distillation is the primary method of producing essential oils (terpenoid fraction), sometimes without significant cannabinoid extraction. When derived from cannabis, these materials could be utilized as the source material to “improve” cannabis extracts to conform to desired specifications [9]. In the New York state cannabis program, it is mandatory that terpenoids in all cannabis products must be cannabis-derived.

A remaining issue for dry ice kief would be decarboxylation of the native acid cannabinoids to their neutral counterparts (THC, CBD and cannabigerol (CBG), etc.). Soxhlet extraction employs a closed system with heating that can achieve decarboxylation of the base material without terpenoid evaporation, but with the possibility of some bioconversion to other forms.

Kryo-Kief™ may represent a “value-added” proposition in that the process yields two end products: the dry ice kief itself and the remaining frozen-sifted flower by-product. As is evident from the analyses in the above figures, the latter still retains significant phytocannabinoid and terpenoid concentrations in material that is not tainted by water (as in water hash extraction) nor solvents. Thus, it may be useful feedstock for subsequent secondary extraction via other techniques. In other words, nothing is wasted after the production of the premium extraction. Key dictates of hashish production have been previously espoused by Clarke [11]: (1) hashish is for devotees with ample supplies of flowers, and (2) the finest quality is obtained by employing purest forms from the start. The approaches outlined herein seem to fulfil these criteria.

Dry ice has been previously employed for cannabis product preparation, but most often it is employed in conjunction with cannabis waste products, such as fan leaves and “shake”, and it is usually employed in direct content with plant material as an abrasive agent that not only dislodges trichomes but also leaf and non-glandular trichome detritus. This is in contrast to the use herein of fresh inflorescence exposed solely to CO₂ vapor. Dry ice treatment does not require equipment under high pressure, as in supercritical CO₂ extraction, does not require residual solvents with possible toxicity (e.g., butane) and can be stored in insulated containers for utilization in subsequent runs.

Certain statements on dry ice safety measures are necessary. Firstly, the dry ice employed must be of pharmaceutical grade and devoid of contaminants. The material employed herein was food grade dry ice but was for experimental purposes and not for human use. Secondly, dry ice is extremely cold and will produce severe freeze injuries to exposed skin unless proper gloves and other protective equipment are utilized. Finally, CO₂ vapor from dry ice is heavier than air and is a toxicant that can accumulate in an enclosed area with a risk of hypoxia, asphyxia and cardiac dysrhythmias unless the workspace is properly ventilated [19,20].

A limitation of this study was that single trials and comparisons were carried out on each of the four chemovars without repetition in order to ensure reproducibility. It is certainly possible that different dry ice exposure times or sifting duration may produce optimized extraction results. Future studies with comparison procedure runs may yield additional data that increase the efficiency and yields of the current methods employed.