Next Article in Journal
Anthocyanin Biosynthesis of White Asparagus Spears in Relation to the Pre-Harvest Exposure to Infrared and Sunlight Radiation
Previous Article in Journal
The Transcriptome and Metabolome Reveal the Mechanism by Which Melatonin Enhances Drought Tolerance in Platycrater argutae
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Horticulturae
Volume 11
Issue 9
10.3390/horticulturae11091090
horticulturae-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Soluble Eugenol Formulation for Managing Ball Moss on Ornamental Trees

by
Brianna Slade
1,
Kali Elftmann
2,†,
Heather Kirk-Ballard
1,‡ and
Zhijun Liu
2,*
1
School of Plant, Environmental and Soil Sciences, Louisiana State University AgCenter, Baton Rouge, LA 70803, USA
2
School of Renewable Natural Resources, Louisiana State University AgCenter, Baton Rouge, LA 70803, USA
*
Author to whom correspondence should be addressed.
†
Current address: Fremont-Winema National Forest, United States Forest Service, Klamath Falls, OR 97601, USA.
‡
Current address: Department of Horticulture, University of Georgia, Athens, GA 30602, USA.
Horticulturae 2025, 11(9), 1090; https://doi.org/10.3390/horticulturae11091090
Submission received: 8 August 2025 / Revised: 6 September 2025 / Accepted: 9 September 2025 / Published: 10 September 2025
(This article belongs to the Section Floriculture, Nursery and Landscape, and Turf)
Browse Figures
Review Reports Versions Notes

Abstract

Ball moss is an epiphytic, perennial monocot that attaches to many surfaces, including ornamental plants. Though not parasitic, ball moss can reduce the growth and health of host plants. Controlling ball moss has thus become necessary. Recommended methods include spraying baking soda or copper fungicide. This study was designed to validate the recommended methods and discover new, efficacious botanical ingredients in laboratory and field experiments. The efficacy of baking soda, but not the copper fungicide, was confirmed. However, baking soda blindly damages host plants and is not environmentally friendly. A screening study over several candidates (a monocot herbicide, eugenol, thymol, oleander extracts) selected eugenol from clove essential oil. In repeated laboratory studies, eugenol formulated into a soluble liquid (ESL) at 1% concentration achieved comparable lethal levels to 50% baking soda solution against ball moss. Efficacy was most apparent when applied in autumn. In the field trials, spraying ESL directly onto ball moss on live oak trees rather than broadcasting from the ground was efficacious. Possible mechanisms and limitations were discussed.

Graphical Abstract

1. Introduction

Ball moss (Tillandsia recurvata) is an epiphyte, or air plant, and a perennial monocot of the family Bromeliaceae. As the name suggests, it forms round, clustered growths that attach to various surfaces, including living tree branches, shrub twigs, or building structures. Ball moss produces its own food through photosynthesis, utilizing nutrients from the air and minerals from rainwater without parasitism.
Ball moss is listed as a native plant to Louisiana and other regions of the Southern United States, including Arizona, New Mexico, Texas, Georgia, Florida, Puerto Rico, and the Virgin Islands [1]. This perennial plant blooms in Baton Rouge, Louisiana in the spring [2]. Geographically, ball moss distributes southward to South America.
Ball moss is not true moss (non-vascular) as it is a flowering plant. It has leaves and photosynthetic pigments to fix solar energy using moisture and carbon dioxide in the air, minerals from rainwater, and other nutrients from earthly dust particles. Ball moss reaches reproductive stage in 4 to 5 years, during which it produces seeds mostly dispersed by the wind [3]. While it was previously thought that the species only utilized these structures for support without acting parasitically, a recent study on smooth mesquite trees (Prosopis laevigata) shows that ball moss can alter the anatomical structures of a host plant, leading to weakened hydraulic conductivity within the xylem and consequently inducing branch mortality [4].
The well-known southern epiphyte Spanish moss (Tillandsia usneoides) once draped live oak (Quercus virginiana) canopies throughout the southeastern United States, including Louisiana [5]. Today, much of that silvery fringe has been replaced by ball moss, which tolerates air pollution more than Spanish moss. Ball moss is now ubiquitous across the state. Many removal efforts stem from concerns over its negative impact on host-plant health, yet some still argue it is harmless. Meanwhile, ball moss has been shown to function as a bioindicator of air pollution, accumulating airborne oils and other contaminants [6].
On the LSU campus, where this study was conducted, ball moss infestation is a major problem. Infestation has become heavy and visible, with host trees typically including deciduous crape myrtles and evergreen live oaks. Methods have been explored to control ball moss infestation. Currently, two treatment methods, physical and chemical, are recommended. The physical treatment involves mechanical removal of ball moss by hand. The chemical method uses baking soda (sodium bicarbonate or NaHCO3) formulated at a 2:1 water to baking soda ratio [7]. However, these methods produced limited success. The physical method is clean and effective, but time-consuming and difficult to apply to all affected areas. The baking soda method seems to cause immediate injury to ball moss, but it creates unwanted effects to the surrounding area. For example, baking soda presumably introduces sodium into the soil, leading to compaction and other environmental issues. Baking soda also non-selectively injures host trees, as well as plants within their vicinities (such as shrubs and grasses), during the growing season, limiting treatment windows to deciduous periods. The current methods used to control ball moss are either impractical or problematic. Finding a method that is effective against ball moss, safer for the host and non-targeted plants, practical for applications, environmentally sound, and cost-effective remains an elusive goal.
Nature is a storehouse of phytochemicals for human medicines [8], veterinary medicines [9], organic pesticides [10], and specialty chemicals like rubber [11]. Some natural compounds such as eugenol and thymol, which are isolated from the essential oils of clove and thyme plants, are herbicidal [12,13]. Oleander is also phytotoxic [14]. In addition to these natural compounds, copper fungicide has fungicidal properties and is even recommended for controlling ball moss. Regardless of the primary ingredient, an herbicide controlling monocots (e.g., monkey grass) without harming dicots (e.g., crape myrtle or live oak) would be ideal for controlling ball moss infestation while simultaneously limiting harm to host plants and the environment. This study aimed to (1) validate the recommended method of baking soda or copper fungicide to control ball moss, (2) discover new compounds or new uses of a product to control ball moss, and (3) evaluate a promising candidate, eugenol, for its field efficacy against ball moss.

2. Materials and Methods

2.1. Screening Studies

Selection of ball moss as experimental materials: ball moss was collected by hand from southern live oak (Quercus virginiana) and crape myrtle (Lagerstroemia indica) trees, the two most prevalent ornamental trees with large quantities of ball moss on the LSU campus. Two characteristics determined whether the ball moss material was acceptable for experimental investigation: (1) it must be healthy and not previously treated by any methods (e.g., baking soda or prior mechanical removal) and (2) it must bear seed pods indicating it has reached the reproductive phase. The collected ball moss clusters were sorted by size (diameter of the ball) into a small, medium, or large group. Within each size group, ball moss received one of three treatments randomly: a natural compound at a single concentration, an untreated control (receiving only water), and a positive control (receiving baking soda). In total, six products and natural compounds were screened in separate experiments. For the screening experiments, two ball moss clusters of each size (two small, two medium, and two large) were placed into trays in groups of six (eighteen clusters total). Treatments were randomly assigned to each tray. To ensure uniform application, treatments were sprayed onto the clusters until the liquid was visibly dripping. Caution was taken to avoid drifting or splashing of the spray solutions onto adjacent clusters. Once spraying was complete, trays sat on a countertop in a laboratory room that was kept at approximately 25 °C (78 °F) without humidity control. Treatment effects were observed for two weeks. To accurately observe the response of ball moss to treatments, water was sprayed onto the clusters to rinse off treatment chemicals. This was especially useful to observe the effects of baking soda, which covered the ball moss with a blanket of white powder after the initial application solution had dried.
Candidate chemical agents: chemical agents/products included copper fungicide, a monocot herbicide, baking soda, oleander leaf or twig alcoholic extract, thymol (isolated from thyme essential oil), and eugenol (isolated from clove essential oil). The 3000-O copper fungicide (Certis, Columbia, MD, USA), Over the Top II herbicide (Fertilome, VPG, Bonham, TX, USA), and baking soda (Arm & Hammer, Church & Dwight Co., Inc., Ewing, NJ, USA) were purchased. The oleander solutions were prepared by extracting vegetative materials collected from a local ornamental oleander (Nerium oleander) plant. Following collection, the materials were air-dried under shade for three days, after which they were oven dried at 60 °C for a week. Once materials were dried, they were extracted with 70% aqueous ethanol at a 1:10 weight by volume ratio and placed on an orbital shaker to facilitate extraction for a week. Following this maceration extraction, the plant material was strained from the liquid extract, leaving the final product of alcoholic extracts ready for treatment applications. Thymol (99% pure, Acros Organics, Waltham, MA, USA) was either dissolved in 90% ethanol to form an alcoholic spray solution via the use of a co-solvent method or dispersed in a water solution containing stevia leaf extract (Steviva, Portland, OR, USA) via the use of a nano-suspension method summarized below. Eugenol (99% pure, Acros Organics, Waltham, MA, USA) spray solution preparation was identical to thymol preparation. The previously mentioned nano-suspension method was used to create a water-dispersible liquid of thymol or eugenol. Thymol or eugenol, along with the stevia leaf extract (acting as the dispersing agent), was processed into a soluble liquid (SL) concentrate using a method [15,16] based on the teaching of a patent invented at LSU [17]. The concentration of thymol or eugenol was 10% v/v in the SL, which was free of organic solvents, surfactants, emulsifying agents, or preservative chemicals. Concentrates were stored in a refrigerator until needed. Prior to immediate use, concentrates were allowed to equilibrate to room temperature for dilutions with deionized water. To account for the effect of ethanol as an extracting solvent, corresponding ethanol solutions without active ingredients were used as vehicle controls. Table 1 provides treatment information including the concentration of active ingredients (ai) in a product or formulation if known, solvent used for preparing a spray solution, and the ai concentration in the spray solution.

2.2. Dose Response Studies

The compound demonstrating the greatest efficacy against ball moss in the screening assays was subsequently evaluated in a dose–response study to establish the minimum active-ingredient concentration required to maintain injurious effect. Since eugenol in the soluble liquid (ESL) formulation showed the most potent efficacy in an array of screening studies, a series of concentrations were prepared by diluting the 10% ESL concentrate to 5%, 3%, and 1% eugenol, with water as the untreated control (0% eugenol).

2.3. Selection of a Candidate Product

Based on the dose response study, the most promising concentration of eugenol became the candidate product for further field studies. However, it is well known that a gap may exist between laboratory and field results. This translational gap is anticipated due to the differences in application environments. In laboratories, ball moss grows in less fluctuating temperatures and humidities than in fields. As a result, concentrations that are deemed effective in the laboratory may work less potently in the natural environment. Anticipating this gap, a higher concentration was used to offset potential loss in potency. Therefore, ESL at 3% eugenol was chosen for field application, despite the1% eugenol concentration being more effective. Our approach was to first demonstrate success in the field and later adjust the dosing to find the minimal concentration required to produce the same or similar results.

2.4. Field Studies

Crape myrtle trees (estimated to be around five years old) infested with ball moss were selected to evaluate the efficacy of 3% ESL. ESL was applied to ball moss in the spring, summer, fall, and winter (when trees dropped their leaves) to account for potential variations in chemical sensitivity related to developmental stage. Infested trees in spatial proximity were selected to minimize variations in micro-environmental conditions (e.g., shade, openness, topography). For each tree, three infested branches loaded with ball moss clusters were selected for treatment. Branch selection was decided based on the following features: branches contained similar number and size of ball moss clusters, clusters contained flowers (indicating they had reached the reproductive stage), and all clusters were in a similar relative position on the branch. Three treatments, 3% ESL (the candidate product), water (blank control), and 50% baking soda in water (positive control), were randomly assigned to a respective branch (one branch per treatment, three branches per tree total) and labeled using color-coded tape.
All three treatments in water were applied via spraying using handheld sprayers in mist mode (Solo 1L, Newport News, VA, USA). To avoid immediate rinsing and dispersing of applied treatments, spraying was timed to take place on dry days with no rain in the forecast for at least 24 h. This careful planning allowed the treatment chemicals ample time to be absorbed by the ball moss tissues. A minimum of two hours of dry weather was necessary to ensure this outcome. If no rain occurred during this period, the treatments were considered to be successfully applied. In cases in which selected branches overlapped or were in close proximity, measures had to be taken to prevent cross-contamination due to spraying drift. To do this, flat pieces of cardboard were held over neighboring clusters during application, ensuring that the clusters were only sprayed with their chosen treatment. Ball moss specimens were sprayed thoroughly until the treatment liquid was visibly dripping from them.
A two-week observation period followed spraying, during which visual photos were taken. After this period, branches bearing the treated ball moss clusters were removed and immediately brought to the laboratory. In the laboratory, ball moss clusters (six to twelve) were harvested by hand from each branch and placed into separate trays per treatment. Water was misted onto the clusters within all trays, which enhanced their natural green coloration to allow for effective injury assessment. Clusters with entirely green tissue were scored as 0% injury (no injury), while those completely blackened were scored as 100% injury. Partial injuries (e.g., 80%) were estimated by the proportion of blackened areas relative to the total cluster. Each cluster’s injury percentage was recorded, and these values were averaged to yield the mean injury for each treatment. Statistical analysis was conducted to detect the treatment differences using a single factor analysis of variance and t-test (Microsoft Excel, 2025, Version 2507). Treatment differences were declared significant at p ≤ 0.05 (Bonferroni corrected by the number of treatments).
The procedure detailed above was utilized for each season. Winter treatment was applied in December, when crape myrtle trees were leafless and dormant. Spring treatment was conducted in April, when leaves had recently emerged (about two weeks prior to treatment application) and appeared fresh and tender. Summer treatment was applied in July, when all leaves had aged to dark green. Fall treatment was conducted in October, before leaves would fall.
During the Fall Study, when leaves and some flowers of the trees remained, treatments were simultaneously applied to the host tree branches containing the treated clusters. Signs of phytotoxicity were examined two weeks after the treatments were applied. These signs included morphological changes (e.g., curly leaves), color changes, and necrosis.

2.5. Treatment Feasibility Trials

Only the candidate product ESL was used during the treatment feasibility trials. The baking soda and blank water treatments used in the laboratory and field studies were not involved. This was not a full-scale treatment trial, as it was conducted on one tree at a time. If the result was positive (i.e., damaged ball moss) and did not affect the appearances of the first ornamental tree, the trial proceeded onto the next tree. Two live oak trees on the LSU Baton Rouge campus were chosen by the Facility Services’ Landscape Horticulture division for these trials. Caution was made to ensure the feasibility trial did not cause apparent damage. Both trees were approximately 80 years old and had open and extensive crowns that covered a circular area of approximately 30 m. The trees appeared healthy, with green and flush leaves on the outside of the crowns. However, extensive ball moss infestation was found inside the canopy. These ball moss clusters resided on both live and deceased branches on the host tree. All clusters bore flowers, indicating that they had reached the reproductive stage. The two chosen trees were not adjacent.
Each tree was sprayed with the same 3% ESL product. The spraying method for each tree; however, was different. The first live oak tree was treated in the same fashion as the crape myrtle trees; liquids were sprayed at a close-range using hand sprayers until the product visibly began dripping from the clusters. The leaves on the trees were neither spray-targeted nor protected. The second live oak received treatment by a Stihl backpack sprayer (Stihl USA, Virginia Beach, VA, USA), which could hold 3.7 gallons in capacity and was powered by gasoline (Stihl Model# SR 430). The spraying was broadcasted from the ground. Like the crape myrtle sprayings, the spraying of both live oaks took place on days with no rain to allow time for optimum treatment penetration.
A two-week waiting period followed the spraying dates for each tree. After this period, ball moss specimens were mechanically removed from the trees and brought to the laboratory for analysis. These specimens were then examined for any injuries using the same method described in the outdoor studies.

3. Results

3.1. Screening Experiments

Significant differences (p = 3.41223 × 10−60) were found among the eleven treatments (DF = 10). Baking soda injured every ball moss cluster it contacted, resulting in a 100% mortality rate (Table 2). The validated baking soda method became the positive control in subsequent screening studies. None of the treatments other than
Baking soda achieved 100% injury rates. However, except for copper fungicide and herbicide, all treatments caused significant injuries when compared to the untreated water control. Copper fungicide only injured 7.2% of the treated ball moss, whereas the monocot herbicide was not effective at the recommended dose or 8 times the recommended dose (5.2% injury rate). Eugenol and thymol dissolved in 90% ethanol caused significant injury of 85.3% and 83.7%, respectively, but ethanol alone caused 80.5% injury, indicating compounding solvent effects. Eugenol and thymol formulated with a natural solubilizer and dispersed in water showed strong effects, with 93.7% and 91.0% injury rate observed in the treated ball moss, respectively. Oleander extracts prepared by 70% ethanol showed significant effects of 90.8% and 87.8% injury rates, respectively; these rates were higher than ethanol alone (80% injury rate), suggesting that unknown ingredients in these extracts increased their potency. The screening studies concluded the following rankings of effectiveness (from most effective to least effective): baking soda, eugenol soluble liquid (ESL), thymol soluble liquid, oleander extracts, eugenol and thymol in ethanol, and ethanol itself.

3.2. Dose Response Study

Screening studies identified the soluble-liquid formulation of eugenol (ESL) as the most effective treatment. Significant differences (p = 9.2290 × 10−34) were found among the six treatments (DF = 5). Baking soda as a positive control injured 91.5% of the ball moss clusters, whereas the untreated control group (Eugenol-0) showed 11.67% injury, a reflection of growth conditions rather than chemical-derived injury. ESL caused significant injury to ball moss, ranging from 87.5% to 98.17% injury (corresponding to a 10% to 1% ESL) (Figure 1). Baking soda covered the ball moss clusters in a white coating.
The ESL formulation caused the clusters to appear noticeably drier and darker, and the water-only control did not alter the morphology or grayish-green coloration of clusters (Figure 2 left). However, two weeks after treatment application, water misting revealed morphological and color alterations. The untreated control mostly retained its original morphology and green coloration (Figure 2 right). The ESL-treated ball moss shrank and desiccated. The 50% baking soda treatment was visibly harshest, rendering the ball moss collapsed in configuration and blackened.

3.3. Outdoor Studies

The screening experiments found that ESL was the most effective treatment candidate, while the dose response determined that 1% ESL resulted in the highest toxicity to ball moss. Translating these laboratory results to outdoor trials required accounting for the higher humidity in the field. Because ball moss typically exists in moister conditions outdoors and droplets may dilute upon deposition, we applied ESL at a 3% concentration to offset any in situ dilution of the treatment. Outdoor studies were conducted in four seasons on crape myrtle trees. In the spring when trees contained tender leaves, significant differences (p = 4.7110 × 10−15) were found in ball moss injury among the three treatments (DF = 2). ESL caused a significant (p = 5.7284 × 10−8) 53.7% injury to ball moss (Table 3) in comparison to the untreated control. However, this injury was significantly less (p = 2.5243 × 10−11) than that caused by baking soda. In the summer when trees contained mature leaves, ball moss injury was significantly (p = 4.7992 × 10−18) different among the three treatments (DF = 2). ESL-treated ball moss was significantly (p = 8.0 × 10−12) damaged (93.2% injury) compared to the untreated control. However, this injury was significantly less (p = 0.0052) than that caused by baking soda. In the fall when leaves were senescent, significant differences (p = 2.6184 × 10−28) were found in ball moss injury among the three treatments (DF = 2). ESL was potent and caused 98.8% injury, which was no difference (p = 0.0344, when Bonferroni corrected alpha is 0.0167) to the effect of the 50% baking soda treatment. In the winter when trees were leafless, ball moss injury was significantly (p = 3.4500 × 10−14) different among the three treatments (DF = 2). ESL produced 84.5% injury to ball moss, which was significantly (p = 9.4590 × 10−7) higher than the untreated control and not different (p = 0.0249, when Bonferroni corrected alpha = 0.0167) from the injury caused by baking soda. Regardless of season, baking soda at 50% w/v consistently blackened every ball moss cluster it contacted. Water as a blank control caused a baseline (approximately 5%) change. ESL was significantly more efficacious than the untreated control, but slightly less effective than baking soda. Substantial phytotoxicity to host trees was observed during the fall trial, during which the baking soda treatment visibly damaged host leaves (i.e., blackened coloration). ESL, on the contrary, caused minimal changes in the morphology and coloration of the host tree leaves and flowers.

3.4. Field Feasibility Trials

Two evergreen live oak trees were selected to receive 3% ESL treatments. One tree was treated in March, when new leaves were emerging among the mature leaves of previous years. ESL that was sprayed directly over ball moss (Figure 3a) caused significant injury (p = 4.4892 × 10−14) on branch one and p = 0.0096 on branch two) that surpassed that of the untreated control. One branch showed 99.1% injury, whereas the other showed only 47.6% (Table 4), suggesting that variability was caused due to coverage and bioavailability (more in discussions). Emerging new leaves of live oak trees were sensitive to ESL and exhibited leaf curling, while mature leaves were unaffected. The second live oak was treated in July. Instead of direct, head-on spraying, ESL was applied to the entire tree by broadcasting a mist from the ground (Figure 3b). Two weeks after the treatments, no visible changes were observed on ball moss or the host tree and the treatment difference was not significant between the 3% ELS and the untreated control (p = 0.7293 for branch one and p = 0.5789 for branch two), indicating no effect from broadcasting the same ESL (Table 4).

4. Discussion

Baking soda and copper fungicide are two chemicals recommended for controlling ball moss infestation [7]. This study confirmed the efficacy of baking soda but demonstrated that copper fungicide was ineffective against ball moss. Baking soda consistently and rapidly eradicated ball moss across seasons and diverse growth environments. However, using baking soda to control ball moss is problematic, and its efficacy could be short-lived. Re-emergence of new ball moss growth was observed, which could suggest that the treatments were partially bioavailable to only the top portions of the treated ball moss tissues. Baking soda is insoluble in water and easy to congregate, making deeper penetration harder. Operationally, constant agitation must be maintained to create a sprayable suspension. A flash pause in agitation could clot hoses. Additionally, applying a white powder coating to tree surfaces undermines the intended esthetic appeal of ornamental trees. Baking soda can also harm the environment as much as it degrades esthetic appeal. Leaching via rain to the ground harms both plants and the soil underneath treated host plants. Baking soda can harm host plants themselves as well, and thus extreme caution must be taken for use on evergreen or limited leafless seasons. These concerns are what impeded the use of otherwise very effective baking soda. Conversely, a soluble liquid product like the eugenol examined in this study overcomes these shortcomings. All ingredients in the ESL formulation are FDA-listed generally regarded as safe (GRAS). Since eugenol is unstable in the air, it is unlikely to accumulate in the environment.
Eugenol is abundant in clove essential oil, allowing it to participate in a plethora of bioactivities and therefore giving it a wide array of potential applications [18]. Eugenol was responsible for observed phytotoxicities from 2.5% clove oil solutions dispersed in 0.2% Tween 20 [19]. In a more recent study, eugenol was found to be one of the potential herbicidal natural compounds [20]. Formulating citronella [21], eucalyptus, or clove essential oils [22] with surfactants into nano-emulsion enabled the observation of herbicidal effects. The use of vinegar to disperse clove oil in water was effective in demonstrating herbicidal effects [23]. Preparing eugenol into a water-soluble powder with myo-inositol as a stabilizer was achieved, success of that would greatly aid homogenous dispersion of eugenol [24]. Ethanol-dissolved clove oil was herbicidal, but ethanol was a substantial contributor to the overall effects (in these experiments). This study demonstrated successful use of steviol glycosides including rubusoside for dispersing eugenol into a clear water solution. Rubusoside was shown to solubilize poorly water-soluble compounds by forming nano-micelles with the active compounds. Examples included the chemotherapeutic diterpenoid paclitaxel [25] and antifeedant compounds against aphids [15,16]. Eugenol-rich clove oil has been reported to possess versatile biological activities, most noticeably anti-foodborne [26] and lactamase-producing bacteria [27], as well as larvicidal effects [28,29]. The herbicidal effects displayed by clove oil were most relevant to this study. Clove bud oil at 5% concentration severely injured several weed species as fast as 5 days after treatment [13]. This study demonstrated that eugenol caused lethal injury to ball moss at a concentration as low as 1% when it was formulated into a soluble liquid by a botanical solubilizer. However, whether a lower concentration than 1% is still effective remains to be evaluated.
Although possible mechanisms behind the observed eugenol effect remain unknown, several are speculated. Firstly, clove oil was shown to disrupt cell membranes causing electrolyte leakage [30,31] and dysfunction of photosynthetic pigments [32]. Secondly, because eugenol was dispersed in water and possibly formed nano-micelles with the botanical solubilizer, as suggested by the nano-micellar paclitaxel [33], it is plausible that eugenol penetrated with the water stream into ball moss tissues in millions of nano-sized oil droplets, crossed watery cell walls, and became bioavailable to cell membranes. Thirdly, because cell membranes are composed of lipids (phospholipid) and are affinitive with oily eugenol, eugenol may soften and eventually “dissolve”, creating tiny “holes” in cell membranes. The report that clove oil can be used as a solvent [34] could support this speculation. Fourthly, the “desiccation to death” phenomenon observed on ball moss might be an indication of uncontrolled cell and tissue drying until the cells lost their vitality [35]. The slow drying process contrasts with the “dismantling” effect of baking soda, where ball moss collapsed abruptly. Ethanol is a penetration enhancer and causes drying via evaporation, which could explain why 70% or 90% ethanol alone was “effective.” Using a botanical solubilizer instead of ethanol is more practical and avoids the hazards associated with flammability. It is generally hypothesized that higher clove-oil concentrations cause greater disruption of cell membranes. While this hypothesis remains valid, the choice of solubilizer may also influence eugenol’s penetration into the tissue. The 10% ESL concentration might have reduced the amount of eugenol that was bioavailable to the cell membranes in ball moss compared to the 5%, 3%, and 1% ESL concentrations. This illustrates the possible role of formulating ingredients in altering efficacy, even though they are regarded as inert. The observed “glassification” appearances of ball moss clusters treated with 10% ESL, but not in those treated with 1% ESL, might support this speculation. Regardless, all of these are only speculative hypotheses that require further evidence.
It is recognized that younger tissues are more sensitive to chemical treatments because their surface chemistry is less complete. As tissues age, surface defense chemicals are built with cuticles and waxes to control water loss and shield the plant from chemicals [31]. Defense chemicals diminish as plants prepare for senescence. Interestingly, ball moss sampled in the spring exhibited lower sensitivity to clove oil than in the senescent fall, suggesting its phenology may not mirror that of its host. Further field trials are necessary to confirm this pattern. Our results indicate that the optimal timing for clove oil application is in autumn, before the host (e.g., crape myrtle) begins leaf abscission. Whether this timing holds true for other hosts (such as live oaks) remains to be determined. In any case, aligning treatment timing with host tissue hardiness may both minimize the phytotoxicity of host plants and maximize efficacy against ball moss.
It is worth noting that bioavailability may govern eugenol’s efficacy. The variable responses to the same eugenol treatment between the two branches of the same live oak tree could be a result of variability in bioavailability, with one being fully bioavailable while the other was partially bioavailable. Therefore, it is imperative to ensure full coverage and bioavailability during spraying operations. Direct spraying onto ball moss until the soluble liquid visibly drips ensured full bioavailability; broadcasting from the ground decreased treatment contact with ball moss, thus rendering the same product non-effective. Broadcasting is highly desirable for efficient coverage of large areas with ball moss infestation. Exploring effective and efficient delivery methods is thus warranted.

5. Conclusions

Though baking soda effectively controlled ball moss, it caused blanket phytotoxicity. Eugenol was found to be similarly effective without those adverse effects. Eugenol from 1 to 10% caused a significant injury to ball moss in laboratory studies. In outdoor and field trials, 3% eugenol in the form of a soluble liquid resulted in significant injury to ball moss that infested crape myrtle and live oak trees, achieving injury levels under some conditions comparable to the 50% baking soda treatment without unwanted problems. Treating ball moss in the fall was more effective than treatment in other seasons. Bioavailability was speculated to have governed the display of efficacy. Direct spraying onto the ball moss may have provided a fuller bioavailability, whereas broadcasting voided it. Therefore, ensuring full coverage to introduce the treatment’s active ingredients to ball moss tissues might be key to achieving optimal efficacy. Overall, ESL was a promising and safer alternative, but not yet fully comparable to baking soda under all conditions. Future studies can determine if a lower concentration than 1% ESL is still active, explore efficient ways of applying a bioavailable product, examine variations among host plants (whether they are deciduous or evergreen), test long-term control efficacy, quantify host phytotoxicity, and consider cost-effectiveness and scalability.

Author Contributions

Conceptualization, Z.L. and H.K.-B.; methodology, Z.L.; investigation, B.S. and K.E.; writing—original draft preparation, K.E., B.S. and Z.L.; writing—review and editing, Z.L. and H.K.-B.; supervision, Z.L.; project administration, Z.L.; funding acquisition, Z.L. and H.K.-B. All authors have read and agreed to the published version of the manuscript.

Funding

This project was funded in part by the A. Wilbert’s Sons Research Internships in Agriculture and Natural Resources Management at the LSU Agricultural Center. This material was based upon work that was funded by the National Institute of Food and Agriculture, U.S. Department of Agriculture, McIntire Stennis project under PG006074/CT0466.

Data Availability Statement

The data presented in this study are available on request from the corresponding author due to logistical and administrative restrictions.

Acknowledgments

We are grateful to Michael E. Salassi, A. Wilbert’s Sons LLC Endowed Professor of Ag & Natural Resources, for selecting our student researchers for two years in a row to receive the funds, which enabled the completion of seasonal studies. We are thankful to Ethan A. Mott and his crew at LSU Facility Services for providing the infested trees for outdoor studies and for assisting in and conducting the treatment trials on live oak trees. Gratitude goes to Ting Chen for assisting in the preparation of eugenol soluble liquid. We thank undergraduate student workers, Claire Bradley and Lark Davis, for assisting in some field experiments.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. USDA, NRCS. The PLANTS Database: Tillandsia recurvata. 2021. Available online: https://plants.usda.gov/plant-profile/TIRE (accessed on 10 July 2025).
  2. Bogren, R.; Singh, R. Ball Moss Cropping up in Louisiana. 2015. Available online: https://www.lsuagcenter.com/portals/communications/news/news_archive/2015/february/headline_news/ball-moss-cropping-up-in-louisiana (accessed on 10 July 2025).
  3. Singh, R. Ballmoss. 2016. Available online: https://www.lsuagcenter.com/~/media/system/b/c/f/b/bcfbd6b48789d9c998694670db4a1b8d/pub%203421%20ballmoss_laplantpath_revpdf.pdf (accessed on 10 July 2025).
  4. Pérez-Noyola, F.J.; Flores, J.; Yáñez-Espinosa, L. Is ball moss (Tillandsia recurvata) a structural parasite of mesquite (Prosopis laevigata)? Anatomical and ecophysiological evidence. Trees 2021, 35, 135–144. [Google Scholar] [CrossRef]
  5. USDA, NRCS. The PLANTS Database: Tillandsia usneoides. 2021. Available online: https://plants.usda.gov/plant-profile/tius (accessed on 8 July 2025).
  6. Rogers, C.; Bush, E. Using Tillandsia recurvata (Ball Moss) as a Biological Indicator to Monitor Air Pollution and Retain Oil Pollution. Open J. Air Pollut. 2022, 11, 62–69. [Google Scholar] [CrossRef]
  7. Kirk-Ballard, H. What’s That Hanging in My Tree? 2020. Available online: https://www.lsuagcenter.com/profiles/rbogren/articles/page1583511527092 (accessed on 10 July 2025).
  8. Kumar, A.; Nirmal, P.; Kumar, M.; Jose, A.; Tomer, V.; Oz, E.; Proestos, C.; Zeng, M.; Elobeid, T.; Sneha, K.; et al. Major Phytochemicals: Recent Advances in Health Benefits and Extraction Method. Molecules 2023, 28, 887. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
  9. Wynn, S.G.; Fougère, B.J. Veterinary Herbal Medicine: A Systems-Based Approach. Vet. Herb. Med. 2009, 2007, 291–409. [Google Scholar] [CrossRef]
  10. Souto, A.L.; Sylvestre, M.; Tölke, E.D.; Tavares, J.F.; Barbosa-Filho, J.M.; Cebrián-Torrejón, G. Plant-Derived Pesticides as an Alternative to Pest Management and Sustainable Agricultural Production: Prospects, Applications and Challenges. Molecules 2021, 26, 4835. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
  11. Cherian, S.; Ryu, S.B.; Cornish, K. Natural rubber biosynthesis in plants, the rubber transferase complex, and metabolic engineering progress and prospects. Plant Biotechnol. J. 2019, 17, 2041–2061. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
  12. Ahuja, N.A.; Batish, D.; Singh, H.P.; Kohli, R. Herbicidal activity of eugenol towards some grassy and broad-leaf weeds. J. Pest Sci. 2015, 88, 209–218. [Google Scholar] [CrossRef]
  13. Giannini, V.; Harris, J.R.; Todde, P.; McElroy, J.S. Concurrent weed growth suppression with essential oils and species-specific response to fractionated coconut oil. Ind. Crops Prod. 2022, 182, 114850. [Google Scholar] [CrossRef]
  14. Hagaggi, N.S.A.; Abdul-Raouf, U.M. Phytotoxic Interference of Culture Filtrates of Endophytic Bacteria Associated with Nerium oleander Leaf Against Seed Germination of the Invasive Noxious Weed Cenchrus echinatus. Curr. Microbiol. 2023, 80, 67. [Google Scholar] [CrossRef]
  15. O’Hara, F.M.; Liu, Z.; Davis, J.; Swale, D. Catalyzing systemic movement of inward rectifier potassium channel inhibitors for antifeedant activity against the cotton aphid, Aphis gossypii (Glover). Pest Manag. Sci. 2023, 79, 194–205. [Google Scholar] [CrossRef] [PubMed]
  16. O’Hara, F.; McComic, S.; Cremades, A.; Norris, E.J.; Liu, Z.; Davis, J.A.; Bloomquist, J.; Swale, D. Characterization of N-arylamide insecticides to control populations of the green aphid, Myzus persicae. Pestic. Biochem. Physiol. 2025, 212, 106459. [Google Scholar] [CrossRef]
  17. Liu, Z. Terpene Glycosides and Their Combinations as Solubilizing Agents. U.S. Patent 8,551,507,B2, 8 October 2013. [Google Scholar]
  18. Haro-González, J.N.; Castillo-Herrera, G.A.; Martínez-Velázquez, M.; Espinosa-Andrews, H. Clove Essential Oil (Syzygium aromaticum L. Myrtaceae): Extraction, Chemical Composition, Food Applications, and Essential Bioactivity for Human Health. Molecules 2021, 26, 6387. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
  19. Stokłosa, A.; Matraszek, R.; Isman, M.B.; Upadhyaya, M.K. Phytotoxic Activity of Clove Oil, Its Constituents, and Its Modification by Light Intensity in Broccoli and Common Lambsquarters (Chenopodium album). Weed Sci. 2012, 60, 607–611. [Google Scholar] [CrossRef]
  20. Torres-Pagán, N.; Muñoz, M.; Barbero, S.; Mamone, R.; Peiró, R.; Carrubba, A.; Sánchez-Moreiras, A.M.; Gómez de Barreda, D.; Verdeguer, M. Herbicidal Potential of the Natural Compounds Carvacrol, Thymol, Eugenol, p-Cymene, Citral and Pelargonic Acid in Field Conditions: Indications for Better Performance. Agronomy 2024, 14, 537. [Google Scholar] [CrossRef]
  21. Somala, N.; Laosinwattana, C.; Teerarak, M. Formulation process, physical stability and herbicidal activities of Cymbopogon nardus essential oil-based nanoemulsion. Sci. Rep. 2022, 12, 10280. [Google Scholar] [CrossRef] [PubMed]
  22. Almarie, A. Bioherbicidal Potential of Eucalyptus and Clove Oil and their Combinations on Four Weedy Species. Iraqi. J. Sci. 2021, 62, 1494–1502. [Google Scholar] [CrossRef]
  23. Evans, G.J.; Bellinder, R.R.; Goffinet, M.C. Herbicidal Effects of Vinegar and a Clove Oil Product on Redroot Pigweed (Amaranthus retroflexus) and Velvetleaf (Abutilón theophrasti). Weed Technol. 2009, 23, 292–299. [Google Scholar] [CrossRef]
  24. Koleva, I.Z.; Kamenova, K.; Petrov, P.D.; Tzachev, C.T. Water-Soluble Formulations of Curcumin and Eugenol Produced by Spray Drying. Pharmaceuticals 2025, 18, 944. [Google Scholar] [CrossRef]
  25. Liu, Z.; Zhang, F.; Koh, G.Y.; Dong, X.; Hollingsworth, J.; Zhang, J.; Russo, P.S.; Yang, P.; Stout, R.W. Cytotoxic and antiangiogenic paclitaxel solubilized and permeation-enhanced by natural product nanoparticles. Anticancer Drugs 2015, 26, 167–179. [Google Scholar] [CrossRef] [PubMed]
  26. Bai, J.; Li, J.; Chen, Z.; Bai, X.; Yang, Z.; Wang, Z.; Yang, Y. Antibacterial activity and mechanism of clove essential oil against foodborne pathogens. LWT 2023, 173, 114249. [Google Scholar] [CrossRef]
  27. Ginting, E.V.; Retnaningrum, E.; Widiasih, D.A. Antibacterial activity of clove (Syzygium aromaticum) and cinnamon (Cinnamomum burmannii) essential oil against extended-spectrum β-lactamase-producing bacteria. Vet. World 2021, 14, 2206–2211. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
  28. Osanloo, M.; Sedaghat, M.M.; Esmaeili, F.; Amani, A. Larvicidal Activity of Essential Oil of Syzygium aromaticum (Clove) in Comparison with Its Major Constituent, Eugenol, against Anopheles stephensi. J. Arthropod Borne Dis. 2018, 12, 361–369. [Google Scholar] [PubMed]
  29. Jyotsna, B.; Patil, S.; Prakash, Y.S.; Rathnagiri, P.; Kavi Kishor, P.B.; Jalaja, N. Essential oils from plant resources as potent insecticides and repellents: Current status and future perspectives. Biocatal. Agric. Biotechnol. 2024, 61, 103395. [Google Scholar] [CrossRef]
  30. Tworkoski, T. Herbicide effects of essential oils. Weed Sci. 2002, 50, 425–431. [Google Scholar] [CrossRef]
  31. Bainard, L.D.; Isman, M.B.; Upadhyaya, M.K. Phytotoxicity of clove oil and its primary constituent eugenol and the role of leaf epicuticular wax in the susceptibility to these essential oils. Weed Sci. 2006, 54, 833–837. [Google Scholar] [CrossRef]
  32. Synowiec, A.; Możdżeń, K.; Skoczowski, A. Early physiological response of broccoli leaf to foliar application of clove oil and its main constituents. Ind. Crops Prod. 2015, 74, 523–529. [Google Scholar] [CrossRef]
  33. Zhang, J.; Shu, J.; Stout, R.W.; Russo, P.S.; Liu, Z. Solubilization of Paclitaxel with Natural Compound Rubusoside toward Improving Oral Bioavailability in a Rodent Model. Pharmaceutics 2024, 16, 1104. [Google Scholar] [CrossRef]
  34. Liñán-Atero, R.; Aghababaei, F.; García, S.R.; Hasiri, Z.; Ziogkas, D.; Moreno, A.; Hadidi, M. Clove Essential Oil: Chemical Profile, Biological Activities, Encapsulation Strategies, and Food Applications. Antioxidants 2024, 13, 488. [Google Scholar] [CrossRef] [PubMed]
  35. Park, K.W.; Choi, S.H.; Ahn, J.Y.; Sohn, Y.G.; Kim, C.-G.; Lee, J.J. Herbicidal action of clove oil on cucumber seedlings. Weed Biol. Manag. 2011, 11, 235–240. [Google Scholar] [CrossRef]
Horticulturae 11 01090 g001
Figure 1. Dose response study over the effect of a eugenol soluble liquid (ESL) formulation on ball moss injuries. Injury to ball moss was expressed as percentage of non-green tissues per ball moss. Treatments include Eugenol-0 (no eugenol or water only), Eugenol-1 (1% ESL), Eugenol-3 (3% ESL), Eugenol-5 (5% ESL), Eugenol-10 (10% ESL), and Baking Soda (50% baking soda in water), a positive control. Each vertical bar represents standard error of the mean (n = 6). Different letters on top of each indicate significant differences (p ≤ 0.05) in t-tests with Bonferroni corrected alpha of 0.0083.
Figure 1. Dose response study over the effect of a eugenol soluble liquid (ESL) formulation on ball moss injuries. Injury to ball moss was expressed as percentage of non-green tissues per ball moss. Treatments include Eugenol-0 (no eugenol or water only), Eugenol-1 (1% ESL), Eugenol-3 (3% ESL), Eugenol-5 (5% ESL), Eugenol-10 (10% ESL), and Baking Soda (50% baking soda in water), a positive control. Each vertical bar represents standard error of the mean (n = 6). Different letters on top of each indicate significant differences (p ≤ 0.05) in t-tests with Bonferroni corrected alpha of 0.0083.
Horticulturae 11 01090 g001
Horticulturae 11 01090 g002
Figure 2. Physical appearances and color of ball moss immediately after treatment application (left) and two weeks after the treatment application (right). Water rinsing was used to enhance coloration. Baking soda was applied at 50% in water. ESL was tested at 1, 3, 5, and 10%.
Figure 2. Physical appearances and color of ball moss immediately after treatment application (left) and two weeks after the treatment application (right). Water rinsing was used to enhance coloration. Baking soda was applied at 50% in water. ESL was tested at 1, 3, 5, and 10%.
Horticulturae 11 01090 g002
Horticulturae 11 01090 g003
Figure 3. Treatment of ball moss infestation on live oak trees by eugenol soluble liquid. (a) A ball moss infested live oak tree receiving a direct mist-spray treatment of 3% ESL in March by a hand-held sprayer; (b) A ball moss infested live oak tree receiving a broadcast mist-spray treatment of 3% ESL in July from the ground.
Figure 3. Treatment of ball moss infestation on live oak trees by eugenol soluble liquid. (a) A ball moss infested live oak tree receiving a direct mist-spray treatment of 3% ESL in March by a hand-held sprayer; (b) A ball moss infested live oak tree receiving a broadcast mist-spray treatment of 3% ESL in July from the ground.
Horticulturae 11 01090 g003
Table 1. Chemicals screened for efficacy against ball moss.
Table 1. Chemicals screened for efficacy against ball moss.
TreatmentDescriptionActive Ingredients (ai)ai Concentration in ProductSolventai Concentration in Spray SolutionRole
Copper fungicide3000-O Copper FungicideCopper hydroxide46.1%Water0.68%New agent
HerbicideFertilomeSethoxydim18%Water1.07%New agent
Eugenol in ethanolEugenol 99% pure10%10%Ethanol (90%)10%New agent
Thymol in ethanolThymol 99% pure10%10%Ethanol (90%)10%New agent
Eugenol-SLEugenol soluble liquid (ESL)5%5%Water5%New agent
Thymol-SLThymol soluble liquid5%5%Water5%New agent
Oleander-LOleander leaf extractOleandrinUnknownEthanol (90%)UnknownNew agent
Oleander-TOleander twig extractOleandrinUnknownEthanol (90%)UnknownNew agent
Baking SodaBaking Sodasodium bicarbonate100%Water50%Positive control
Table 2. Ball moss injury caused by chemical treatments from screening studies. Values followed by different letters indicate significant differences at p ≤ 0.05 (Bonferroni corrected).
Table 2. Ball moss injury caused by chemical treatments from screening studies. Values followed by different letters indicate significant differences at p ≤ 0.05 (Bonferroni corrected).
ChemicalInjury (%) ± Standard Error of the Mean (n = 6)
Baking Soda100.00 ± 0.00 a
Copper fungicide7.2 ± 0.86 d
Herbicide5.2 ± 0.80 d
Eugenol in ethanol85.3 ± 1.48 bc
Thymol in ethanol83.7 ± 1.64 bc
Eugenol-SL93.7 ± 0.73 b
Thymol-SL91.0 ± 0.75 bc
Oleander-L ethanol90.8 ± 0.64 bc
Oleander-T ethanol87.8 ± 0.80 c
Ethanol 90%80.5 ± 0.99 c
Water6.8 ± 1.61 d
Table 3. Effects of eugenol in soluble liquid formulation (ESL) on ball moss related to seasons. Different letters following the mean injury percentages within each season indicate significant differences at p ≤ 0.05 (Bonferroni corrected).
Table 3. Effects of eugenol in soluble liquid formulation (ESL) on ball moss related to seasons. Different letters following the mean injury percentages within each season indicate significant differences at p ≤ 0.05 (Bonferroni corrected).
SeasonTreatmentInjury (%) ± Standard Error of the Mean
Spring
(n = 6)		3% ESL53.7 ± 1.35 b
Baking Soda100.0 ± 0.00 a
Water6.8 ± 2.68 c
Summer
(n = 6)		3% ESL93.2 ± 1.75 b
Baking Soda100.0 ± 0.00 a
Water5.5 ± 1.22 c
Fall
(n = 9)		3% ESL98.8 ± 0.49 a
Baking Soda100 ± 0.00 a
Water4.0 ± 1.66 b
Winter
(n = 8)		3% ESL84.5 ± 3.59 a
Baking Soda100.0 ± 0.00 a
Water6.7 ± 2.38 b
Table 4. Effects of eugenol in soluble liquid (ESL) formulation on ball moss in infested live oak trees. Different letters following the mean injury percentages within each branch of a tree indicate significant differences at p ≤ 0.05.
Table 4. Effects of eugenol in soluble liquid (ESL) formulation on ball moss in infested live oak trees. Different letters following the mean injury percentages within each branch of a tree indicate significant differences at p ≤ 0.05.
Live OakBranch NumberTreatmentInjury (%) ± Standard Error of the Mean (n = 12)
Tree 1—hand sprayOne3% ESL99.1 ± 0.40 a
Untreated10.7 ± 5.00 b
Two3% ESL47.6 ± 11.73 a
Untreated11.1 ± 3.72 b
Tree 2—broadcastOne3% ESL3.8 ± 0.55 a
Untreated4.2 ± 0.54 a
Two3% ESL3.3 ± 0.41 a
Untreated2.9 ± 0.34 a
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Slade, B.; Elftmann, K.; Kirk-Ballard, H.; Liu, Z. Soluble Eugenol Formulation for Managing Ball Moss on Ornamental Trees. Horticulturae 2025, 11, 1090. https://doi.org/10.3390/horticulturae11091090

AMA Style

Slade B, Elftmann K, Kirk-Ballard H, Liu Z. Soluble Eugenol Formulation for Managing Ball Moss on Ornamental Trees. Horticulturae. 2025; 11(9):1090. https://doi.org/10.3390/horticulturae11091090

Chicago/Turabian Style

Slade, Brianna, Kali Elftmann, Heather Kirk-Ballard, and Zhijun Liu. 2025. "Soluble Eugenol Formulation for Managing Ball Moss on Ornamental Trees" Horticulturae 11, no. 9: 1090. https://doi.org/10.3390/horticulturae11091090

APA Style

Slade, B., Elftmann, K., Kirk-Ballard, H., & Liu, Z. (2025). Soluble Eugenol Formulation for Managing Ball Moss on Ornamental Trees. Horticulturae, 11(9), 1090. https://doi.org/10.3390/horticulturae11091090

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop