Associated Data
Abstract
Nanoemulsions of a terpene-rich by-product (TP) from commercial cannabidiol (CBD) production were successfully formulated and characterized. An enriched terpene distillate (DTP) was also obtained from steam distillation of TP and used for making nanoemulsions. The effects of formulation conditions including the hydrophilic lipophilic balance (HLB) value of the surfactant, TP and surfactant content, and sonication time on the properties of the emulsions were analyzed. The optimal formulation conditions were determined as surfactant HLB value of 13, TP content of 5 wt% in water, surfactant content of 2 times of TP, and sonication time of 1.5 min. A scaled-up production of the optimal nanoemulsion was also achieved using a microfluidizer and the effect of pressure and number of passes on emulsion properties was determined. The stability of the nanoemulsions was evaluated and the DTP nanoemulsion was determined to be the most stable. The nanoemulsions with desirable properties were then selected and evaluated for their insecticidal activity against the legume pest, Callosobruchus maculatus, with nanoemulsion of neem oil made under the same conditions as a control. Both TP and DTP nanoemulsions were found to exhibit excellent insecticidal activity, and the latter had the highest efficacy against the Callosobruchus maculatus.
1. Introduction
Although the Cannabis sativa L. (hemp, marijuana) plant produces more than 400 non-cannabinoid compounds and 125 phytocannabinoids (phenolic terpenes, their analogs, and transformation products) [1], the phytocannabinoid compounds have been the primary focus of the producers, consumers, and researchers [1,2]. In the United States, classification of C. sativa as industrial hemp requires that concentration of (−)-trans-Δ [9]-tetrahydrocannabinol (THC) and its precursor, tetrahydrocannabinolic acid not equal or exceed 0.3%. Industrial hemp strains containing high concentrations of CBD chemotypes have been selected for commercial extraction of CBD. Non-cannabinoid components reported in C. sativa included phenols, flavonoids, and terpenes [1]. During industrial processing, many of these compounds are separated from CBD and are discarded as part of production waste. A recent report on the chemical composition of these wasted resources from commercial CBD production revealed some to be composed of up to 40 wt % sesquiterpenoids [3]. Such products can serve as an abundant and inexpensive source of terpenes, and the utilization of terpene by-product can offset the cost of CBD and reduce environmental impact associated with its disposal.
Terpenes are responsible for plant aromas, can impact insect behavior [4,5], and are active against human and plant pathogens [[6], [7], [8]]. Terpenes play a central role in plant defense [9,10] and are elevated by plants under stress. For example, leaves and flowers of C. sativa plants infested by the two-spotted mite (Tetranychus urticae) had higher concentrations of cannabinoids and terpenes than those of non-infested plants [10]. The reported monoterpenes and sesquiterpenes in C. sativa include linear, monocyclic or bicyclic hydrocarbons and oxygenated hydrocarbons (terpenoids). Many of the biopesticides registered with the US Environmental Protection Agency (EPA) have high concentrations of terpenes found in hemp [11]. Essential oils (EOs) isolated from CBD chemotypes by hydrodistillation have shown to be toxic to pest species. An EO containing myrcene (27.5%), limonene (14.0%), and β-caryophyllene (7.6%) was found to be as repellent to mosquitoes as the insecticide DEET (N,N-diethyl-meta-toluamide) and was toxic to larvae [5]. Another EO preparation that contained (E)-caryophyllene (45.4%), myrcene (25.0%) and α-pinene (17.9%) was highly toxic to the peach-potato aphid (Myzus persicae) and the housefly (Musca domestica), moderately toxic to tobacco cutworm (Spodoptera littoralis) larvae, but not toxic to non-target invertebrates, Asian lady beetle (Harmonia axyridis) and earthworms (Eisenia fetida) [12].
Although the use of terpenes discarded in hemp processing as an environmentally friendly biopesticide for insect control is a promising application, their commercial use as biopesticides is limited due to their high lipophilicity, high volatility, and propensity to thermal and oxidative degradation [13], which translates to reduced efficacy. Formulation of nanoemulsions using terpene as the lipophilic phase can be an effective methodology to better deliver the bioactivity and overcome these limitations. In the past five years, significant effort has been devoted to nanoemulsion formulation of plant EOs to serve as a safer alternative to synthetic pesticides [14,15]. Promising insecticidal activities have been demonstrated for EOs, particularly those rich in monoterpenes and sesquiterpenes [[16], [17], [18]]. However, TP as a waste stream from CBD production is an important resource of terpenes that has been overlooked.
Nanoemulsions are effective pesticide delivery systems due to their high kinetic stability, low viscosity, and increased availability of the active compounds [[19], [20], [21]]. The process of nanoemulsification can improve the physicochemical stability of pesticides [22], reduce the use of organic solvents, and enhance insecticidal efficacy from improved spreading, deposition, and permeation of active components to the target site [[23], [24], [25]]. Thus, enhanced control over the release of active compounds translates to a substantial reduction in the amount of synthetic pesticide used [26,27]. Nevertheless, the potential adverse effects of synthetic pesticides on non-targeted organisms, in addition to their environmental accumulation [[28], [29], [30]] and likelihood in promoting insect resistance [[31], [32], [33]], have created a niche exquisitely suited for nanoemulsion-based biopesticides.
In the present work we have focused on the utilization and fabrication of terpene-rich hemp waste streams into nanoemulsion biopesticides. The objectives of this study were 1) to formulate nanoemulsions from crude TP by-product and after distillation to obtain DTP; 2) to characterize the composition of DTP; 3) to evaluate the effect of surfactant HLB, terpene concentration, and surfactant content on the physiochemical properties and stability of the nanoemulsions; and 4) to evaluate the insecticidal activity and efficacy of the resultant biopesticide nanoemulsions in preventing Callosobruchus maculatus infestation of mung bean (Vigna radiata) seeds. We have chosen this model insect, commonly called bean beetle or cowpea weevil, because it is commercially available, easy to rear in the laboratory, and previously has been used to evaluate insecticidal properties of terpene-containing EOs [34,35]. Additionally, there is a significant need for the development of a biopesticide to control this beetle [36], because it causes significant economic losses in pulse crops (e.g., cowpea, mung bean, and adzuki bean) that are important nutritional food sources in Africa and Asia [37], and currently there are limited control measures for this insect [36].
2. Materials and methods
Cannabis Terpene Mix 1 and Cannabis Terpene Mix 2 standards, representing a combined total of 42 different common Cannabis terpenoids, were purchased from SPEXCerti-Prep (Metuchen, NJ) as 100 μg/mL solutions in methanol. Tween 80, Span 80, methyl stearate (>99%) and solvents were used as received unless otherwise noted and purchased from ThermoFisher Scientific (Hampton, NH). Neem oil (NO) was purchased from Plantonix (Ashland, OR) as a positive control in the insect bioassays.
2.1. Sample collection
The TP by-product was provided by a local CBD manufacturer. It was collected as a viscous brown distillate generated from the distillation of a winterized CBD extract. Cultures of C. maculatus were obtained from Carolina Biological Supply (Burlington, NC). Organic mung bean seed (Vigna radiata L.) (Banyan Botanicals, Albuquerque, NM) was used in routine culture maintenance and for insecticidal experiments. Cultures and all experiments were maintained at 28 °C.
2.2. Steam distillation of TP to obtain DTP
The composition of the TP by-product was previously determined and quantitated [3]. To obtain DTP, 50 g of TP by-product was added into a 1 L round-bottomed flask containing deionized water (500 mL). The flask was connected to a standard short-path distillation apparatus and the contents were steam distilled at 100 °C for 6 h. The collected pale-yellow distillate was dried over Na2SO4 and stored under darkness. For chromatographic analysis, an aliquot of DTP was diluted in freshly distilled pentane to a concentration of 50 μg/mL.
2.3. Gas Chromatography−Flame ionization detection (GC−FID) analysis
Terpenes in DTP were quantitated using an Agilent HP6890 GC with FID system (Agilent Technologies, Santa Clara CA). Separation was achieved using a nonpolar HP-5 (5%-phenyl methylpolysiloxane, 30 m × 0.25 mm i.d., 0.25 μm film thickness) capillary column (J&W Scientific, Folsom, CA) with a helium carrier gas flow rate of 1.5 mL/min and 1:20 split ratio. The temperature program consisted of a 1 min hold at 40 °C, followed by 6 °C/min ramp to 250 °C held for 10 min, with both injector and detector temperatures at 250 °C. Individual six-point calibration curves (5–100 μg/mL) were constructed from Cannabis Terpene Mix 1 and Cannabis Terpene Mix 2 reference standards. Linear correlation coefficients ranged from 0.9797 to 0.9989 and results were reported as the average of two analytical replicates. Methyl stearate was used as the internal standard and relative percentages are reported on a weight basis for each analytical sample. Additional identity confirmation was achieved by a linear retention index (RI) relative to a homologous series of n-alkanes (C9–C26) and comparing against NIST-compiled values for the identical stationary phase.
2.4. Formulation of nanoemulsions
Emulsions were formulated using the methods reported by Nirmal et al. [21] and Abbas et al. (2015) with modifications. Both TP and DTP were formulated into nanoemulsions. To have a parallel comparison of their insecticidal activity without the interference of formulation ingredients and conditions such as surfactant content, only TP was used for emulsification optimization, whereas DTP and NO nanoemulsions were formulated at the determined optimal conditions for TP. The NO nanoemulsion was used as a positive control for bioassays.
For TP nanoemulsion formulation, TP as the oil phase and the surfactant system were first vortexed to obtain a mixture. A coarse emulsion was prepared by homogenizing the oil with surfactant and deionized water at 5000 rpm for 1 min. Subsequently, a fine emulsion was prepared by ultrasonication for 1 min using a high-intensity ultrasonic dismembrator (Model 500, Fisher Scientific, Hampton, NH) equipped with a 1.27 cm diameter probe made of titanium alloy operating at 50% amplitude. To avoid localized heating of the sample, the dismembrator was operated with 5 s pulses (5 s ON and 7 s OFF).
To determine the optimal HLB, surfactant systems with HLB values ranging from 10 to 15 were prepared by mixing Span 80 (HLB 4.3) and Tween 80 (HLB 15) at the following wt % ratios: 46.7:53.3 (HLB 10), 37.4:62.6 (HLB 11), 28:72 (HLB 12), 18.7:81.3 (HLB 13), 9.3:90.7 (HLB 14), 0:100 (HLB 15) [38]. The effect of surfactant content on emulsion properties was assessed at 10, 20, and 30 wt% with TP concentration (relative to the aqueous phase, not including the surfactant) fixed at 10 wt %. To determine the effect of TP concentration, 5, 10, and 15 wt % of TP (relative to the aqueous phase, not including the surfactant) was used at the determined optimal surfactant content. Ultrasonication time was assessed at 0.5 1, 1.5, 2, 2.5 and 3 min in order to produce the smallest particle with minimum energy input. Each emulsion formulation was replicated two times, and one sample was used from each replicate for characterizations. The detailed composition of the emulsions was summarized in Table 1.
Table 1
Composition of the nanoemulsions.
| Emulsion Code | TP content (g) | De-ionized water (g) | Surfactant content (g) |
|---|---|---|---|
| 10 wt% TP 1:1 | 1.0 | 9.0 | 1.0 |
| 10 wt% TP 1:2 | 1.0 | 9.0 | 2.0 |
| 10 wt% TP 1:3 | 1.0 | 9.0 | 3.0 |
| 5 wt% TP 1:2 | 0.5 | 9.5 | 1.0 |
| 15 wt% TP 1:2 | 1.5 | 8.5 | 3.0 |
| 5 wt% DTP 1:2 | 0.5 | 9.5 | 1.0 |
| 5 wt% DTP 1:3 | 0.5 | 9.5 | 1.5 |
| 5 wt% NO 1:2 | 0.5 | 9.5 | 1.0 |
2.5. Characterization of nanoemulsion properties
Droplet size distribution and zeta-potential of the nanoemulsions were determined by using a Zetasizer Nano-ZS (Malvern Panalytical, UK) at 25 °C at a 1:20 dilution with deionized water. Mean values are reported based on three separate replicates. Turbidity was determined using a Biomate 5 UV–Vis spectrophotometer (ThermoScientific Woburn, MA) at 600 nm as absorbance (Abs600), and it was expressed as percent transmission using deionized water as reference. Mean values are reported based on two separate replicates.
Stability of the nanoemulsions during storage was also evaluated adapting the method reported by Nirmal et al. [21] with modifications. Emulsions were first stored at 25 °C and creaming assessed visually for 10 d storage during the initial screening for optimal formulation conditions. The emulsions that did not undergo creaming were then characterized for changes in droplet size during a 30 d storage at 25 °C and 40 °C. Changes in nanoemulsion droplets were monitored by optical microscopy. One drop (4 μL) of the nanoemulsion was placed on glass slide, and a cover slide slipped over to produce a thin film. Samples were visualized using BX41 polarized light microscope (Olympus, Tokyo, Japan) a ×40 magnification.
The viscosity of optimal nanoemulsions was measured using a Discovery HR-2 hybrid rheometer (TA Instruments, New Castle, DE). Approximately 25 mL of the nanoemulsion were loaded into a concentric cylinder geometry attachment and equilibrated at room temperature for 1 min before measurement. The shear rate was set from 0.01 to 1000 s−1, and data points were recorded every second. The apparent viscosity of the emulsion was recorded, plotted against shear rate, and compared at the specific shear rate (950 s−1). Mean values are reported based on two separate replicates.
2.6. Scale-up of the optimal emulsions using a microfluidizer
To examine a more industrially feasible means of formulation, TP nanoemulsion was prepared using the optimal conditions identified and microfluidization was applied to create the emulsion. Coarse TP emulsions (100 mL) were first prepared by homogenizing (at 5000 rpm) TP, surfactant, and water mixture. This coarse emulsion was then processed in a water-cooled LM-20 Microfluidizer (Microfluidics, Westwood, MA) by employing different number of passes (2, 3, and 4) and pressures (10,000, 20,000, 30,000 psi). Ten milliliter aliquots from each treatment were collected for characterization. Mean values are reported based on two separate replicates.
2.7. Insect bioassays
Optimal TP, DTP, and NO nanoemulsions were evaluated for insecticidal activity by a bioassay designed to assess the effect of direct contact on numbers of viable eggs, development period, and adult emergence. Twenty healthy V. radiata seeds were submerged in the nanoemulsion treatment for 5 min and subsequently transferred to a Petri dish (60 mm) with a filter paper disc (10 mm) to absorb excess liquid. Five dishes were used for each treatment and the treatments were as follows: 1) emulsion control (surfactant in deionized water); 2) 5 wt % NO 1:2 (positive control); 3) 5 wt % DTP 1:2; 4) 5 wt % DTP 1:3; 5) 5 wt % TP 1:2. The ratio indicates the relative content of TP and emulsifier.
Two adult pairs of C. maculatus less than 2 d old were allowed free access to the treated seeds for 2 d and then removed. Numbers of viable eggs on the treated seeds were counted after one week. After this duration of time, viable eggs are opaque white indicating that the larvae had hatched, deposited frass into the eggshell, and burrowed into the bean [37]. Emergence of adults from nanoemulsion treated seeds was monitored three times per week until no new adults emerged (26–42 days) for two consecutive observation periods. The developmental period of C. maculatus was defined as the number of days between the oviposition (adult removal) and adult emergence [39]. Mean values are reported based on three separate replicates.
2.8. Statistical analysis
For data from the emulsion formulation, one-way analysis of variance was carried out and differences between pairs of means were compared using a Tukey test. The significant level was set at 0.05. Data from the bioassay were analyzed using two-way analysis of variance and differences between pairs of means were compared using a Tukey test at 0.05 significance level. Analyses were conducted using SAS 9.4 TS1M6 for Windows 64× (SAS institute Inc., Cary, NC). The economic loss due to emergence was calculated by % hatch rate × % bean with viable eggs.
3. Results and discussion
3.1. Composition of TP and DTP
The major components in DTP include the sesquiterpenoids (−)-trans-Caryophyllene (28.70 ± 3.43 wt %), cis-nerolidol (9.95 ± 1.07 wt %), α-humulene (7.37 ± 0.89 wt %), and α-bisabolene (7.25 ± 0.78 wt %). Monoterpenes accounted for less than 6 wt % for both distillates, although steam distillation of TP resulted in a slight enrichment of these compounds. TP contains a comparatively lower concentration of these components due to the commercial distillation conditions which result in a cannabinoid enrichment [3]. Only TP, however, contained cannabinoids. Evidently, DTP is a comparatively richer source of sesquiterpenes and terpenoids.
Steam distillation of crude TP to yield a refined DTP was performed to have a point of comparison. TP is a readily available by-product from CBD production and often treated as waste, whereas the quantity of DTP as how we produced from distillation was very limited. Even though the composition of these two products is different, all emulsions must be formulated using the same conditions for a parallel comparison of their insecticidal effect without the interference of factors such as HLB and surfactant content. Therefore, the nanoemulsion optimization was only performed on TP and then the conditions were extended to DTP and NO.
3.2. The required HLB for fabrication of TP nanoemulsions
The effect of surfactant mixtures with different HLB values on the droplet size distribution and stability of the nanoemulsions throughout 10 d storage at 25 °C is presented in Fig. 1a and b. TP concentration was fixed at 10 wt %. The initial droplet size distribution was not significantly different among treatments although creaming was observed after 10 d storage for HLB 10, 11, 12, and 15 (Fig. 1b). Nanoemulsions with HLB 13 and 14 had characteristic bluish tinge although turbid. The cloudy appearance is in agreement with the droplet size distribution which included large droplets in the size range of 1–10 μm. Another set of the emulsions was then formulated with HLB of 13 and 14 for further evaluation. The droplet size distribution of the two emulsions were found to be similar, although the emulsion made using with HLB 13 presented a smaller mean droplet size (140.2 ± 1.8 nm) compared to that with HLB 14 (156.9 ± 3.1 nm). Therefore, HLB of 13 was selected as the optimal value for subsequent formulations.
(a) Droplet size distribution of freshly made 10 wt % TP nanoemulsions formulated at the specified HLB value. (b) appearance of emulsions formulated at the specified HLB value after 10 d storage at 25 °C. Instances of creaming are indicated using boxes.
The HLB value of the surfactant system used to fabricate nanoemulsions significantly affects the mean droplet size, a predictor of emulsion stability [40,41]. Other workers have reported optimal surfactant HLB values for various oils. A wheat bran oil-in-water emulsion with lowest droplet size was obtained using a 37.4 wt % Span 80 and 62.6 wt % Tween 80 mixture corresponding to HLB 11. The optimal surfactant system for evening primrose seed oil was reported to be a mixture of 22 wt % Span 80 and 78 wt % Tween 80 corresponding HLB 12 [42]. Similarly, for rosemary EO, an HLB 15 was determined to be optimal [41]. These reports indicate the required HLB value for oils and EOs varies from 11 to 15. The optimal HLB 13 for TP is in agreement with the reported range.
3.3. Effect of surfactant content, TP content and sonication time on droplet size and turbidity of emulsions
To further reduce the droplet size, the effect of surfactant content, TP content and sonication time were investigated. The surfactant content used during emulsification is known to be an important factor affecting properties of the emulsions [43]. Fig. 2a shows its effect on droplet size distribution when TP content in water was fixed at 10 wt%. Increasing the surfactant content from 1 to 2 times of the TP content resulted in the disappearance of large diameter droplets and reduced the average droplet size from 141.3 to 64.7 nm. It was also noticed that droplet size distribution became narrower as the surfactant content increased. This is likely due to higher surfactant content leading to more effective reduction of interfacial tension at the oil-water interface [44]. However, further increasing the surfactant content to 3 times of TP content did not result in significant improvement. Others have reported similar findings [45]. It is likely that the equilibrium surface tension was achieved when surfactant content of 2 times of TP was used. It also indicates that surfactant content of 2 times of TP was sufficient to provide the droplet surface coverage needed, therefore, further increase of surfactant content beyond this point had limited effect on further reducing the droplet size. Therefore, 2 times of TP content was selected as the optimal surfactant content for subsequent formulations.
Effect of (a) TP (10 wt %)-to-surfactant ratio, (b) TP concentration, (c) ultrasonication time on droplet size of TP and NO nanoemulsion, and (d) mean droplet size of selected nanoemulsions throughout 30-day storage at 25 °C.
Fig. 2b displays the effect of TP content of 5, 10 and 15 wt % on droplet size at the optimal surfactant concentration. The mean droplet size increased with increasing TP content. The smallest particle (36.5 nm) was obtained at 5 wt % TP content with a concomitant shift to optical transparency which is in agreement with the work reported by others [46]. The larger droplet size observed at the higher TP content can be attributed to dynamic droplet formation and coalescence due to insufficient surfactant coverage of the droplets or frequent collision [46]. A wider droplet size distribution was also observed at higher TP content, suggesting the co-existence of large micelles and nano-sized oil droplets [21]. These observations indicated that the TP content played a critical role in the preparation of stable emulsions, concurring with conclusions reported by others [44]. Therefore, a 5 wt % TP that resulted in the smallest droplet size was selected as the optimal for the later study.
The ultrasonication time required to attain the smallest droplet size of TP nanoemulsion was determined to be 1.5 min, with increasing time proved to be either ineffective or even detrimental by slightly increasing the droplet size distribution (Fig. 2c). The droplet size of nanoemulsions depends on cavitation, turbulence and shear forces generated by the ultrasonicator. Smaller droplets of emulsions were formed with the increase in ultrasonication time [47]. Moreover, temperature of the emulsions increases due to the increase of energy input which can lead to reduction in viscosity and emulsion stability. In addition, ultrasonication can positively affect the adsorption rate of the surfactant to the surface of particles and thus decreases the droplet size distribution [48]. However, increasing the ultrasonication time past a certain threshold can induce recoalescence due to “over-processing” leading to formation of large droplets, which has been comprehensively reviewed by Jafari et al. [49]. Evidently, judicious optimization of ultrasonication time is not only important for fabricating stable emulsions but also for reducing energy costs. With the optimal formulation conditions determined, a DTP nanoemulsion with a mean droplet size of 34.7 nm was successfully fabricated for subsequent comparison.
An NO nanoemulsion was likewise formulated under the aforementioned conditions. Increasing ultrasonication time similarly decreased the droplet size. However, NO required a much longer ultrasonication time (2.5 min) compared to TP (1.5 min) to produce emulsions with similar droplet size. This may be attributed to its different chemical composition leading to different interfacial behavior. In addition, the apparent viscosity of NO was observed to be higher than TP, which may raise the energetic threshold needed to disrupt and break larger droplets into smaller ones [50].
3.4. Scaled-up fabrication of nanoemulsions by microfluidization
A scaled-up production of TP nanoemulsion at the optimal conditions (HLB 13, 5 wt % TP in water, surfactant content of 2 times of TP) was performed by microfluidization and the effect of operating pressure and number of passes on percent transmittance (i.e. turbidity) and droplet size was investigated. Turbidity and droplet size decreased with increasing pressure and number of passes as shown in Table 3. Inside the microfluidizer chamber, droplets are exposed to a high shear and complex mixing patterns that reduce the droplet size [51]. In general, all experimental conditions produced fine oil-in-water nanoemulsions with a mean droplet size of less than 100 nm. However, increasing the operating pressure and number of passes became less effective after a certain threshold. Increasing the pressure from 10,000 to 20,000 psi at three passes, for example, reduced the mean droplet size from 86.8 to 62.4 nm. Further increasing the pressure to 30,000 psi only reduced the droplet size to 55.5 nm. After three passes at 30,000 psi, the mean droplet size decreased from 68.6 to 55.5 nm, while a fourth pass only further reduced it to 51.1 nm.
Table 3
Effect of microfluidization pressure and number of passes on droplet size and transmittance of TP nanoemulsions.
| Pressure (psi) | Number of passes
|
||
|---|---|---|---|
| 2 | 3 | 4 | |
| Droplet size (d, nm) | |||
| 10,000 | 92.0 ± 1.1aA | 86.8 ± 0.1bA | 82.6 ± 0.4cA |
| 20,000 | 73.0 ± 0.3aB | 62.4 ± 1.2bB | 57.0 ± 0.4cB |
| 30,000 | 68.6 ± 0.1aC | 55.5 ± 0.1bC | 51.1 ± 0.4cC |
| Transmission (%) | |||
| 10,000 | 69.8 ± 0.1cC | 78.3 ± 0.1bC | 82.2 ± 0.1aC |
| 20,000 | 87.1 ± 0.0cB | 92.2 ± 0.5bB | 94.7 ± 0.1aB |
| 30,000 | 88.1 ± 0.5bA | 95.9 ± 0.1aA | 95.3 ± 0.0aA |
Upper case letters indicate significant differences caused by different pressures with the same number of passes. Lower case letters indicate significant differences caused by the number of passes at the same pressure. Mean values denoted with the same letter are not statistically different at α = 0.05.
Turbidity paralleled the trend observed for droplet size. This behavior can be attributed to coalescence and rupturing of particles as a result of the extreme environment inside the microfluidizer’s interaction chamber. Mahdi Jarari et al. [52] reported that the optimal conditions to produce a d-limonene oil-in-water nanoemulsion by microfluidization required a pressure of 10,000 psi and two passes. It was observed that further increasing the pressure was not beneficial and favored coalescence. Many others [51,53] have also reported similar observations. In our work, coalescence due to over-processing has not been observed under the tested conditions. The optimal microfluidization conditions were determined to be a pressure of 30,000 psi and four passes. TP represents a complex, highly viscous mixture of terpenoids and cannabinoids; it is therefore unsurprising that high pressures and multiple passes were required for nanoemulsion fabrication.
3.5. Stability and viscosity of the selected nanoemulsions
The optimal TP nanoemulsion (HLB 13, 1:2 TP-to-surfactant ratio, 5 wt % TP and 1.5 min ultrasonication time) and NO nanoemulsion (HLB 13, 1:2 NO-to-surfactant ratio, 5 wt % NO and 2.5 min ultrasonication time) were evaluated for stability throughout 30 d storage at 25 and 40 °C. In addition, a DTP nanoemulsion (HLB 13, 1:2 DTP-to-surfactant ratio, 5 wt % DTP and 1.5 min ultrasonication time) and the scaled-up TP emulsion fabricated by microfluidization (HLB 13, 1:2 TP-to-surfactant ratio, 5 wt % TP, 30,000 psi and four passes) were likewise subjected to the identical storage condition for evaluation.
At 25 °C, all the emulsions were stable, and the mean droplet size only slightly increased after 30 d storage (Fig. 2d). The DTP nanoemulsion had the least increase in mean droplet size. At 40 °C, significant changes were observed among the emulsions, especially for the TP nanoemulsions as shown in Fig. 3. TP nanoemulsions fabricated via ultrasonication and microfluidization became significantly more turbid after 30 d storage, suggesting coalescence. The percent transmittance of TP nanoemulsions produced by ultrasonication and microfluidization significantly decreased from 81.5 to 11.6% and 95.3 to 27.2%, respectively. The elevated storage temperature increased the thermal energy and therefore increased collision frequency. In contrast, DTP and NO nanoemulsions had less pronounced increase in turbidity. The percent transmittance of DTP and NO nanoemulsions decreased from 94.0 to 92.0% and 86.5 to 76.5%, respectively. These observations suggest that compositional homogeneity of the oil phase plays an important role in particle coalescence.
Appearance and microscopic observation (40× magnification) of selected nanoemulsions during the 30-day storage at 40 °C. Abbreviation: Sonic for sonication, and Mic for microfluidization.
The change in mean droplet size for TP nanoemulsions was monitored as an indicator of stability. As the mean droplet size increased, a concomitant increase turbidity was noted after 30 d storage at 40 °C. The mean droplet sizes for TP nanoemulsion produced via ultrasonication and microfluidization significantly increased from 36.5 to 157.5, and 51.1–180.2 nm, respectively. In contrast, the mean droplet size of DTP and NO nanoemulsions only increased from 34.7 to 44.1 nm and 54.4–61.1 nm, respectively. The visible increase in turbidity of both TP and NO nanoemulsions suggested the presence of much larger particles undetectable by our droplet size analyzer which has a 10 μm upper limit of detection. Optical microscopy of these emulsions revealed a significant number of large droplets in TP nanoemulsions after 30 d storage at 40 °C, whereas, almost none was observed in the DTP nanoemulsion (Fig. 3). A few of large species were likewise observed for the NO nanoemulsion.
Zeta-potential as another indicator of emulsion stability was also determined. The NO nanoemulsion had a much higher zeta-potential (−20.7 mV) compared to TP nanoemulsions (−8.3 mV) as shown in Table 4, and this may attribute to the observed higher stability. The DTP nanoemulsion had a lower zeta potential (−6.8 mV) compared to NO but was determined to be more stable than both NO and TP nanoemulsions. When the difference of zeta-potentials between various emulsions is small, zeta-potential is sometimes not a useful parameter for assessing stability. As was reported by Roland et al. [54], no correlation between zeta-potential (−43.1 to 50.2 mV) and overall stability was observed. In fact, the most visually stable emulsions in their study exhibited the lowest zeta-potential. In our case, other factors such as mean droplet size and compositional homogeneity may also have significantly affected the nanoemulsion stability.
Table 4
Zeta potential and viscosity of the selected nanoemulsions.
| Nanoemulsion | Zeta-potential (mV) | Viscosity (Pa s × 10−3) |
|---|---|---|
| 5 wt % TP (1:2)a | −8.3 | 5.8 |
| 5 wt % TP (1:2)c | −8.3 | 5.2 |
| 5 wt% DTP (1:2)a | −6.8 | 5.2 |
| 5 wt % NO (1:2)b | −20.7 | 6.1 |
Lastly, as an important parameter for determining a nanoemulsion’s application method, the viscosity of the optimal TP, DTP, and NO nanoemulsions at 25 °C was determined. Table 4 shows that all nanoemulsions have low viscosities, similar to that of water (4.2 × 10−3 Pa s). Although the TP nanoemulsion produced by microfluidization and DTP nanoemulsion have slightly lower viscosities than the TP and NO nanoemulsions produced by ultrasonication, the overall differences are minor and suggest all nanoemulsions are suitable to be applied by spraying.
3.6. Insecticidal activity of the terpene nanoemulsion via direct contact
In our preliminary experiments, volatiles from treated seed did not affect either egg production or adult C. maculatus emergence, therefore only the effect of direct contact was evaluated. The percentage of seeds with viable eggs was higher for the control than for either of the DTP treatments (Fig. 4). Nevertheless, both DTP treatments were not significantly different from either the NO (positive control) or 5 wt % TP nanoemulsions (F = 6.89; df = 4; p = 0.0063) (Fig. 4a), although they are numerically lower. This may be attributed to the higher total terpenoid concentrations and smaller droplet size in the DTP nanoemulsion which likely resulted in better dispersion of the active compounds to the seed surface. For all treatments, seeds had fewer viable eggs than the control (F = 7.84; df = 4; p = 0.0047) (Fig. 4b). Viable eggs were chosen in this experiment because they represent the impact on larvae development. This measure, however, did not allow for the determination of impact on egg hatching. With respect to the percentage of viable eggs that became adults, none of the treatments appeared to have an effect.
Effect of selected nanoemulsions on viable eggs of C. maculatus on (A) percentage of mung bean seed with eggs, and (B) mean number of eggs/seed. Oil-to-surfactant ratio denoted in parentheses for each treatment. Means denoted with the same lower case letter are not significantly different at α = 0.05. Error bars represent standard deviation of three trials.
Adult emergence was analyzed by two methods: 1) complete – all dishes were included in the analysis; and 2) no-zero – dishes with no beans on which there were viable eggs were excluded from the analysis. In the complete analysis, numbers of adults that emerged were different from the control only for the 5 wt % DTP (1:3) treatment (F = 4.29; df = 4; p = 0.0280) (Fig. 5A), but in the no-zero analysis both 5 wt % TP and 5 wt % DTP (1:3) were different from control (F = 11.57; df = 4; p = 0.0014) but not from each other (Fig. 5B). TP and DTP were equally effective in reducing the number of adults as NO. Nonetheless, TP and DTP do not contain azadirachtin, the key insecticidal compound in NO that may lead to safety concerns [55,56]. These results highlight the potential for utilization of either the crude by-product TP or its refined counterpart as effective biopesticides. The better performance of 5 wt % DTP (1:3) nanoemulsion can be attributed to the higher surfactant concentration and its smaller droplet size which may have resulted in enhanced dispersion and penetration of the active compounds [23,57,58].
Effect of selected nanoemulsions on adult emergence of C. maculatus (A) complete data set, (B) no-zero data set (dishes with no eggs were not included), and (C) estimates of losses based on adult emergence and percentage of beans with viable eggs. Oil-to-surfactant ratio is denoted in parentheses for each treatment. Means denoted with the same letter are not significantly different at α = 0.05. Error bars represent standard deviation of three trials.
(−)-trans-Caryophyllene (28.70 ± 3.43 wt %), cis-nerolidol (9.95 ± 1.07 wt %), α-humulene (7.37 ± 0.89 wt %), and α-bisabolene (7.25 ± 0.78 wt %) were the most abundant compounds in DPT (Table 2). It is therefore reasonable to surmise that these compounds contributed to the insecticidal activity against C. maculatus. Sesquiterpenoids have previously been used as antifeedants in other stored-product pests [59]. Because C. maculatus adults do not feed and only larvae are affected by antifeedants, reductions in viable egg counts may have been due to reduction in larval feeding. The EO from Hyptis suavenolens (containing 8 wt % (−)-trans-caryophyllene) has previously been reported to be active against bean beetles in repellency, fumigant toxicity, and contact toxicity studies [34]. The EO from Tephrosia vogelii, containing 4.6 wt % (−)-trans-caryophyllene and 8.5 wt % cis-nerolidol, was reported to have no effect on adult emergence or percentage of beans attacked by beans beetles. However, the EO from Tephrosia densiflora, containing 45 wt % (−)-trans-caryophyllene but no cis-nerolidol, was reported as effective against adult emergence at all tested concentrations and against percentage of attacked beans at the highest tested concentration [35].
Table 2
Chemical composition and relative percentagesc of terpene-rich by-product (TP) and enriched terpene distillate (DTP).
| Compound | TP (wt %) | DTP (wt %) | Compound | TP (wt %) | DTP (wt %) |
|---|---|---|---|---|---|
| Monoterpenesb | Terpenoids | ||||
| Myrcenea | 0.32 ± 0.00 | nd | Dehydro-1,8-cineole | 0.16 ± 0.04 | 0.31 ± 0.17 |
| Thuja-2,4(10)-diene | 0.91 ± 0.01 | 1.09 ± 0.14 | p-Cresol | nd | 0.15 ± 0.01 |
| β-Pinenea | 0.35 ± 0.01 | 0.63 ± 0.09 | Linaloola | tr | 0.02 ± 0.01 |
| Sabinenea | 0.20 ± 0.00 | 0.56 ± 0.06 | Endo-fenchol | nd | 0.14 ± 0.00 |
| (+)-3-Carenea | 0.18 ± 0.01 | 0.15 ± 0.00 | cis–p-menth-2-en-1-ol | 0.17 ± 0.01 | 0.20 ± 0.00 |
| α- Terpinenea | nd | 0.11 ± 0.16 | isoborneola | tr | 0.07 ± 0.10 |
| p-Cymenea | 3.99 ± 0.06 | 2.06 ± 2.05 | neo-menthol | nd | 0.08 ± 0.11 |
| β-Phellandrene | 0.17 ± 0.00 | 0.24 ± 0.00 | α-Terpineol | nd | 0.16 ± 0.16 |
| (Z)-Ocimenea | 0.57 ± 0.02 | 0.30 ± 0.42 | 1,8-Cineole | 0.61 ± 0.04 | 0.72 ± 0.08 |
| (E)-Ocimene | 0.81 ± 0.01 | 0.87 ± 0.13 | trans-Nerolidola | 0.28 ± 0.03 | 0.37 ± 0.28 |
| γ-Terpinene | 0.09 ± 0.00 | 0.67 ± 0.04 | cis-Nerolidola | 3.87 ± 0.39 | 9.95 ± 1.07 |
| Sesquiterpenes | Caryophyllene oxidea | 1.90 ± 0.19 | 3.00 ± 0.27 | ||
| α-Copaene | tr | nd | Guaiola | 2.76 ± 0.28 | 2.19 ± 0.47 |
| α-Ylangene | tr | 0.19 ± 0.01 | Humulene epoxide II | 0.55 ± 0.00 | 0.82 ± 0.19 |
| β-Copaene | nd | 2.20 ± 0.25 | 10-epi-γ-Eudesmol | 1.17 ± 0.93 | 1.86 ± 0.15 |
| β-Cubebene | 0.09 ± 0.01 | 0.17 ± 0.17 | γ-Eudesmol | 0.26 ± 0.01 | 0.67 ± 0.15 |
| γ-Gurjunene | 0.17 ± 0.01 | 0.94 ± 0.06 | Cubenol | 0.46 ± 0.35 | 0.15 ± 0.02 |
| α-Cedrene | tr | 0.11 ± 0.15 | α-Muurolol | 0.45 ± 0.35 | 0.48 ± 0.01 |
| (−)-trans-Caryophyllenea | 4.37 ± 0.46 | 28.70 ± 3.43 | β-Eudesmol | 0.90 ± 0.14 | 0.55 ± 0.03 |
| α-trans-Bergamotene | 0.29 ± 0.02 | 0.48 ± 0.50 | α-Eudesmol | 1.35 ± 0.09 | 0.89 ± 0.06 |
| α-Humulenea | 1.23 ± 0.12 | 7.37 ± 0.89 | Caryophyllenol II | 0.52 ± 0.05 | 0.30 ± 0.01 |
| (E,E)-β-Farnesene | 0.29 ± 0.03 | 1.89 ± 0.21 | Cedrola | nd | 0.27 ± 0.00 |
| allo-Aromadendrene | 0.11 ± 0.01 | 0.38 ± 0.01 | Intermedeol | 1.43 ± 0.15 | 0.81 ± 0.13 |
| β-Chamigrene | tr | 0.27 ± 0.02 | α-Bisabolola | 1.18 ± 0.12 | 0.63 ± 0.06 |
| β-Selinene | 0.51 ± 0.05 | 3.01 ± 0.30 | Eudesma-4(15),7-dien-1-β-ol | nd | 1.85 ± 0.09 |
| Valencenea | 0.93 ± 0.09 | 4.08 ± 0.46 | (2Z,6Z)-Farnesol | 0.42 ± 0.03 | 0.12 ± 0.12 |
| α-Muurolene | nd | 0.27 ± 0.01 | Isophytol | nd | 0.47 ± 0.66 |
| (E,E)-α-Farnesene | 0.37 ± 0.03 | 2.08 ± 0.19 | Cannabinoids | ||
| β-Bisabolene | 0.18 ± 0.01 | 0.78 ± 0.13 | Cannabidiol | 52.49 ± 0.79 | nd |
| γ-Cadinene | 0.14 ± 0.00 | 0.34 ± 0.48 | Cannabichromene | 1.10 ± 0.02 | nd |
| α-7-epi-Selinene | 0.23 ± 0.01 | 0.48 ± 0.24 | Δ [8]-Tetrahydrocannabinol | 1.02 ± 0.01 | nd |
| δ-Cadinene | 0.23 ± 0.02 | 1.86 ± 0.21 | Δ [9]-Tetrahydrocannabinol | 1.66 ± 0.02 | nd |
| α-Cadinene | 0.40 ± 0.05 | 1.13 ± 0.12 | Cannabigerol | 1.84 ± 0.01 | nd |
| α-Bisabolene | 2.91 ± 0.29 | 7.25 ± 0.78 | Other | ||
| β-Calacorene | 0.09 ± 0.00 | 0.31 ± 0.02 | 1-Octen-3-one | 0.28 ± 0.00 | 0.31 ± 0.02 |
| 6-Methyl-5-hepten-2-one | 0.24 ± 0.01 | 0.26 ± 0.01 | |||
| Hexanoic acid | nd | 0.19 ± 0.02 | |||
| (E)-Octen-2-al | nd | 0.15 ± 0.01 | |||
| 1-Octen-3-yl acetate | nd | 0.11 ± 0.16 | |||
| Octanoic acid | nd | 0.11 ± 0.16 | |||
| Vanillin | nd | 0.33 ± 0.04 |
Two quality losses can be attributed to C. maculatus infestation of mung beans. The first, reduced quality of beans due to the presence of a larvae, can be estimated as the percentage of beans with viable eggs (Fig. 4a) since those beans contained at least one larvae, but not all larvae fully developed into an adult. The second type of loss, beans that would be rejected in quality control due to the holes created by emerging adults (% loss), was estimated by correcting the percentage of hatches with percentage of beans with eggs. This estimate was lower for 5 wt % TP (1:3) and 5 wt % DTP (1:3) than for the control, although not significantly different from the other treatments (p = 0.0200) (Fig. 5C). This suggests that TP and DTP can significantly reduce bean rejects in quality control due to damages caused by emerging adults and improve the sustainability and economics of stored-products such as mung beans. More research is required to evaluate the effect of TP and DTP concentration on the insecticidal activity and germination of mung bean seed. In addition, the key component in TP that’s responsible for the insecticidal activity shall be confirmed in a future study.
4. Conclusion
Both TP and DTP contained high concentrations of terpenoids with well-documented insecticidal activity and are promising inexpensive alternatives to synthetic pesticides and NO for pest control of stored-products. TP and DTP nanoemulsions can be fabricated by sonication and scaled-up quantities can best be made by microfluidization. In the present work, an oil concentration of 5 wt %, oil-to-surfactant ratio of 1:2, and either 1.5 min sonication time at 50% amplitude or microfluidization using 4 passes at 30,000 psi, were identified as the optimal emulsification conditions. All nanoemulsions were stable at 25 °C although only the DTP nanoemulsion was stable and remained translucent at 40 °C throughout the 30-day storage. Both TP and DTP nanoemulsions exhibited promising insecticidal activity against C. maculatus in mung beans. The number of insect eggs per bean, percentage of beans with eggs, and the estimated percent seed loss due to adults hatched were significantly reduced after treating the beans with the TP nanoemulsions.
Author contribution statement
Tao Fei: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.
Kimberly Gwinn: Conceived and designed the experiments; Performed the experiments; Wrote the paper.
Francisco M.A. Leyva-Gutierrez: Performed the experiments; Analyzed and interpreted the data.
Tong Wang: Conceived and designed the experiments; Analyzed and interpreted the data; Wrote the paper.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors gratefully acknowledge assistance of Xiaocun Sun who performed the statistical analysis of insect bioassay data. This work is supported by Hatch/Multi-state project [accession number: 1023982].
References
