Abstract
The theoretical properties of nanobubbles (NBs), such as a negative surface charge and large interfacial surface area, allow for highly efficient gas transfer and stagnation time in water and may reduce the surface tension of NB-treated water sources. These properties make NBs unique candidates for addressing issues like root zone oxygen deficiency, common in conventional and hydroponic crop production. Therefore, the objectives of this research were to confirm the presence of NBs in treated water and determine how time and temperature affect dissolved oxygen (DO) retention in NB-oxygenated water. Two membrane-based NB injection systems were compared with a standard method of aeration (aquarium air stone) and untreated potable water to determine the effect of NB oxygenation on DO retention time and nanoscopic particle size and concentration of potable water sources. NB oxygenation of potable water generally resulted in a greater number of nanoparticles detected compared with untreated potable water. NB oxygenation increased initial levels of DO in potable water when compared with the standard air stone. NB oxygenation failed to increase DO retention time compared with a standard air stone, regardless of water temperature. NB oxygenation remains a method of efficiently oxygenating large volumes of water, although the NBs investigated in the study did not increase DO retention in a potable water source.
Providing ample oxygen to a plant root system is a constant concern in agricultural and horticultural production. Plant root systems may be exposed to oxygen-deficient conditions within the root zone if soils have poor porosity or become waterlogged (Ponnamperuma 1984). Roots of crops produced hydroponically require access to oxygen dissolved in a nutrient solution. In deep-flow-technique production systems, inadequate root aeration can be a problem, as oxygen diffusion can be drastically reduced in stagnant nutrient solutions (Morimoto et al. 1989). Oxygen deficiencies also can be a concern in other methods of hydroponic crop production, such as nutrient-film-technique systems (Gislerød and Kempton 1983; Jackson 1980). This problem may be compounded during the summer, as nutrient-solution oxygen solubility is reduced in greenhouses experiencing high-temperature extremes. As nutrient solution temperature increases, there is a direct relation to plant oxygen consumption and an inverse relation to the amount of oxygen dissolved in the solution (Al-Rawahy et al. 2019). Even in cooler climates such as Ontario, Canada, root zone DO levels in a commercial cucumber greenhouse were reported to be as low as 2 mg·L−1 during the summer months (Zheng et al. 2007).
Nanobubbles
Nanobubbles (NBs) are gas-filled cavities in a liquid that are found dispersed within solution (bulk NBs) or at the interface between a liquid medium and a solid substrate (surface NBs) (Wang et al. 2017). Nanobubbles exhibit several unique properties. Because of their small size, NBs have a large surface area per unit volume, with a corresponding concentration as high as 100 million to 10 trillion bubbles per milliliter of liquid (Atkinson et al. 2019). Nanobubbles allow for high gas dissolution rates in liquids due to higher internal pressure in the bubble than their environment and high stagnation in the liquid phase (Ushikubo et al. 2010). Additional studies suggest that NBs exhibit a long residence time in solution owing to a negatively charged surface (zeta potential). This surface charge prohibits the coalescence of bubbles, which is characteristic of larger bubbles, which coalesce and rise to the surface (Takahashi et al. 2007; Ushikubo et al. 2010). Nanobubbles have been reported to remain in an aqueous solution for weeks and even months (Azevedo et al. 2016; Duval et al. 2012).
The properties of NBs may allow for increased DO retention times in water and increased DO retention at elevated water temperatures, which is important for crop production in systems with seasonal and diurnal irrigation water temperature fluctuations. The objectives of our research were to confirm the presence of NBs in treated water and characterize how time and temperature affect the DO retention of NB-oxygenated water. It was hypothesized that NB oxygenation would result in more nanoscopic particles per ml of water and longer DO retention compared with traditional means of water oxygenation, regardless of water temperature.
Materials and methods
Dissolved oxygen study
Research was conducted at the University of Arkansas Rosen Alternative Pest Control Center in Fayetteville, AR, USA. All water originated from the same source, supplied by Beaver Water District (Lowell, AR, USA). Two proprietary membrane-based NB injection systems [Nano Bubble Technologies, Sydney, NSW, AU (NBT); Moleaer Inc., Torrance, CA, USA (Moleaer)] were compared with a control that consisted of a standard method of aeration [6-inch Top Fin Aquarium Stone, United Pet Group, Atlanta, GA, USA (Air Stone)].
Experimental units were collected in 1-gal glass jars with circular mouths that measured 2 inches in diameter, allowing for 12.5 square inches of water exposed to the atmosphere. Nanobubble-oxygenated treatments were cycled through respective NB injection systems, each delivering 1.5 L·min−1 of industrial-grade oxygen (Airgas, Radnor, PA, USA) to a water volume of 100 gal, for a period of 1 h at a rate of 12 gallons per minute. The 6-inch aquarium stone was placed directly into each 1-gal glass jar and was allowed to oxygenate using the same industrial-grade oxygen for 1 h. Containers were sealed with an airtight cap and transported to a growth chamber. Temperature-controlled growth chambers (Conviron E7, Winnipeg, MB, Canada) were set to maintain temperatures of 20, 30, and 40 °C. The DO content of each treatment was measured at 0 and 24 h, and then every 24 h until DO levels returned to that in untreated reference water. DO content was measured using a portable DO meter (Model HI98193; Hanna Instruments, Woonsocket, RI, USA) and a Clark-type polarographic DO probe with a polytetrafluoroethylene polymer membrane cap (Model HI764073, Hanna Instruments).
The experimental design was a randomized complete block with four replications of each water treatment, with two separate experimental runs of each growth chamber temperature. Blocking was used to account for any minor variations that may have existed within the growth chamber. Testing of temperature effects was not conducted due to output limitations of our NB injection systems and a lack of available growth chambers, which resulted in extended periods between generated batches of water that would have allowed for inconsistencies between batches placed in growth chambers of different temperatures. Therefore, all water treatment comparisons are only made within a given temperature.
No significant differences between experimental runs of the same temperature were detected. Therefore, data from each temperature were pooled for further analysis. To assess differences in the rate of DO loss between treatments with appropriate adjustment for the effect of time, an analysis of covariance was conducted with oxygenation treatment as a fixed effect and time as a covariate. This was followed by separate regression analyses to determine differences in initial DO concentrations and the subsequent rates of DO loss over time. The statistical software used was R (v. 4.3.2 R Core Team 2023) with the R packages Car (Fox et al. 2012), Tidyverse (Wickham 2019), and Lme4 (Bates et al. 2015), using the R-studio interface (R-Studio 2023.12.0 Build 369).
NanoSight analysis
To quantify and characterize the NBs present in treated water, nine replications of each NB-oxygenation method and a control of untreated water were analyzed at the Vanderbilt University Institute of Nanoscale Science and Engineering (Nashville, TN, USA) on three separate occasions using a NanoSight NS300 (Malvern Panalytical, Malvern, United Kingdom) equipped with a scientific complementary metal-oxide semiconductor camera and a red laser. Water used for NanoSight analysis was generated separately from water generated for the DO experiment. The Moleaer unit was no longer available on the third sampling, so the analysis only compared the NBT injection system to the control. NanoSight analysis determined the total concentration of nanoscopic particles per milliliter of water, the mean and standard deviation of nanoscopic particle size, and the D10, D50, and D90 of each sample, which signify the particle diameter below which 10%, 50%, and 90% of the total volume in the sample was contained. Data were subjected to analysis of variance in SAS (SAS version 9.4; SAS Institute, Cary, NC, USA) using PROC MIXED. Where appropriate, treatment means were separated using Fisher’s protected least significant difference (P = 0.05).
Results and discussion
Dissolved oxygen study
For all three growth chamber temperatures, the highest order interaction of time × treatment significantly affected the rate of DO loss from oxygenated water (Table 1). Nanobubble-oxygenated water treatments did not result in increased retention time of DO compared with the Air Stone treatment at any growth chamber temperature. In the 20 and 30 °C experimental runs, NBT contained the greatest initial DO concentration but had the greatest rate of DO loss compared with Moleaer and Air Stone, which did not differ from one another (Table 2, Fig. 1). In the 40 °C experimental run, Moleaer and NBT contained greater initial DO concentrations but displayed greater rates of DO loss compared with the Air Stone (Table 2, Fig. 1). Although NB treatments failed to increase DO retention time, they did achieve ∼11% to 22% greater initial DO concentrations compared with the Air Stone when averaged across all temperatures (Table 3).
Analysis of covariance testing the effects of time, oxygenation treatment, and their interaction on the rate of dissolved oxygen loss in oxygenated water sources in growth chambers at three different storage temperatures.
Slopes and intercepts for the linear regression of the rate of dissolved oxygen loss of three oxygenated water sources in growth chambers at three different temperatures.
Dissolved oxygen loss as affected by Oxygenation Method × Time for three methods of water oxygenation stored in growth chambers at 20, 30, and 40 °C.
Citation: HortTechnology 34, 6; 10.21273/HORTTECH05524-24
Combined means and standard deviations for initial dissolved oxygen concentrations achieved by three methods of water oxygenation and an untreated control.
The inability of NB-oxygenated water sources to retain DO longer than water oxygenated with the Air Stone may be because of a weak negative surface charge (zeta potential) on the NBs generated in this study. Zeta potential is a critical parameter when discussing bubble stability (Han et al. 2004, 2006; Hewage 2020). Nanobubbles created in pure water, free of electrolytes, have high absolute zeta potentials that create repulsion forces, reducing bubble coalescence. Because there is an inverse relationship between the electrolyte concentration of water and the magnitude of negative NB surface charge, increasing electrolyte concentration in water decreases the magnitude of the bubble zeta potential (Han et al. 2004, 2006). More specifically, cations with high valency have the potential to neutralize the bubble charge (Hewage 2020). Although the zeta potential of bubbles was not measured directly in our research, the presence of electrolytes in our water source may have reduced the absolute zeta potentials of generated NBs, resulting in reduced DO stability in the NB-oxygenated water.
Nanobubbles have been observed to be stable in solution for multiple weeks (Azevedo et al. 2016; Duvall et al. 2012). The stable NBs observed by Azevedo et al. (2016) were generated using air injection, rather than pure oxygen, in ultrapure water (filtered and deionized). The lack of electrolytes in the treated water likely played a role in the long-term stability of the NBs reported by Azevedo. Stable NBs observed by Duval et al. (2012) were reported as being generated in ultrapure water by a precise, vigorous shaking process in the presence of purified air. The ultrapure water contained added monovalent ions of known concentrations. Bubble stability was greatest in solutions with the lowest concentration of ions.
The lack of multivalent ions in the treated water investigated by Duval et al. (2012) also likely facilitated long-term bubble stability. The water used in our research originated as tap water from the Beaver Water District (Lowell, AR, USA), and pure oxygen was used to create NBs. The difference in gases and solutes between our experiment and those of Azevedo et al. (2016) and Duval et al. (2012) likely resulted in differences in bubble stability; therefore, oxygen loss occurred more rapidly than expected.
Because water used for irrigation and hydroponic nutrient solutions contains electrolytes not found in pure or deionized water, NB stability may be compromised in field applications of NB technology. In addition, DO loss from flowing NB-oxygenated water occurs rapidly, making the continuous generation of NBs necessary to maintain elevated levels of DO in treated water (Langenfeld and Bugbee 2021). The rapid loss of DO observed in flowing water by Langenfeld and Bugbee (2021) could also impact the use of NB-treated water for use in other agricultural and horticultural systems. The distance the NB-treated water must travel from an NB generator to reach its intended location, coupled with the atomization of the water flowing through an irrigation head and nozzle, could also result in rapid DO loss.
Although no statistical comparisons were made between temperatures, as expected, increased growth chamber temperature resulted in faster DO loss. Because the focus of this work was to investigate the retention of DO rather than the effect of water temperature at generation on oxygen dissolution levels, it should be noted that the water temperature at the time of generation was ∼15 °C, which is less than the growth chamber temperatures in which the water was placed. Because water holds less DO as it warms, it is reasonable to assume that had the water been generated at a temperature of 40 °C, a decrease in initial DO concentration for all water treatments would have been observed compared with water generated at 30 or 20 °C. Further investigation into the effect of water temperature on DO concentration at the initial generation of NB-oxygenated water will help with understanding the ability of NB injection systems to oxygenate water at elevated temperatures. Because irrigation water used in the horticultural and agricultural industries is subjected to diurnal and annual water temperature fluctuations, the ability to oxygenate water at high temperatures is important for the applicability of NB technology to the horticultural and agricultural industries.
NanoSight analysis
NanoSight analysis confirmed the existence of NBs in NB-oxygenated water. However, the number of nanoscopic particles detected was inconsistent between treatments and analyses (Table 4). Each NB treatment significantly increased nanoparticle concentration compared with Control water in only one of three analyses, even though nanoparticle concentrations were numerically greater for NB treatments in every analysis (Table 4). Mean nanoparticle size, standard deviation, and size distribution were inconsistently affected by NB oxygenation (Table 4). Detected particle size was generally greater and more variable in NB-oxygenated water compared with the Control but never differed between NB treatments. The size of NBs created through porous membranes is affected most notably by the flow rate of the water in which they are created (Ulatowski and Sobieszuk 2018). Increased particle size variability in NB treatments compared with the Control was surprising due to the generation of NBs occurring through a porous membrane of fixed pore size with water at a constant flow rate. Because proprietary technology was used to conduct this research, pore sizes of the specific membranes used were unavailable. It is possible, however, that a range of pore sizes in the membranes resulted in increased variability of nanoscopic particles compared with the Control.
NanoSight analysis of the mean, standard deviation, size distribution, and concentration of nanoscopic particles in potable water oxygenated using nanobubble injection systems and an untreated control.
It is important to note the limitations of our analyses confirming the presence of NBs in treated water. Because this research was designed to have direct applicability for practitioners who likely do not use ultrapure water for irrigation, potable water sources rather than ultrapure water sources were used for NanoSight analysis. Because ultrapure water sources were not used in our experiments, we cannot say with certainty that all particles detected were NBs, due to the difficulty of distinguishing between nanoparticles and NBs (Eklund 2019; Yasui et al. 2019). The limitation of our work is also a limitation for potential end users of NB technology. Detection limitations are a significant challenge for NB detection in nonpure water sources. Because irrigation water and hydroponic nutrient solutions are not ultrapure water sources, the inability to distinguish between nanoparticles and NBs is an important limitation for golf course superintendents or hydroponic producers wanting to confirm the functionality of their systems by the detection of NBs in treated water. This limitation should be strongly considered before purchasing and installing NB injection systems based on the current state of measurement technology.
Indirect measurements like the induced coalescence method, the light scattering method, and DO reverse estimation are the most accessible methods for monitoring the existence of NBs in treated water (Kim et al. 2018). The induced coalescence method requires applying irradiating ultrasonic waves to the water, which makes this method an additional challenge for users of NB technology (Kim et al. 2018). The light scattering method is the simplest method for determining the existence of sub-micron bubbles, but the scattering of light could be affected by turbid water sources used in golf course irrigation and hydroponic production (Kim et al. 2018). Induced coalescence and light scattering do not indicate the size or number of bubbles created, simply that bubbles of less than 1 µm are present in the water. Dissolved oxygen reverse estimation allows for bubble size estimation but requires monitoring the DO content of treated water under three different pressures, which can be a time-intensive process, not well suited for real-time water analysis (Kim et al. 2018). Due to the difficulty associated with DO reverse estimation, this method should not be considered for real-time analysis of NB-treated water. Kim et al. (2018) state that DO concentration analysis confirmed the presence of sub-micron bubbles, but it was unclear how the authors measured DO and how it confirmed the existence of sub-micron bubbles. Until efficient, reliable detection of NBs becomes accessible for impure water sources, confirming the existence of NBs in treated water will continue to hinder the widespread use of NB technology.
Conclusions
This research demonstrates that NB oxygenation of irrigation water sources can allow for excellent levels of oxygen dissolution in irrigation water, but other benefits, such as longer DO retention, may not be observed. Other than efficient gas dissolution, expectations of improved water performance should be tempered and considered before investing in a membrane-based NB injection system. A significant hurdle that currently limits the broad application of this technology is the ability to quickly and accurately measure the NB concentrations in generated water. Using the DO concentration of treated water to confirm NB injection system functionality remains the fastest and most efficient method but does not confirm the existence of NBs. Until these challenges are overcome, it will be almost impossible for a producer to assess if an NB injection system is functioning properly.
References cited
Al-Rawahy MS, Al-Rawahy SA, Al-Mulla YA, Nadaf SK. 2019. Influence of nutrient solution temperature on its oxygen level and growth, yield, and quality of hydroponic cucumber. J Agric Sci. 11(3):75–92. https://doi.org/10.5539/jas.v11n3p75.
Atkinson AJ, Apul OG, Schneider O, Garcia-Segura S, Westerhoff P. 2019. Nanobubble technologies offer opportunities to improve water treatment. Acc Chem Res. 52(5):1196–1205. https://doi.org/10.1021/acs.accounts.8b00606.
Azevedo A, Etchepare R, Calgaroto S, Rubio J. 2016. Aqueous dispersions of nanobubbles: Generation, properties, and features. Minerals Eng. 94:29–37. https://doi.org/10.1016/j.mineng.2016.05.001.
Bates D, Maechler M, Bolker B, Walker S, Christensen RHB, Singmann H, Bolker MB , et al. 2015. Package ‘lme4’. Convergence, 12(1):2.
Duval E, Adichtchev S, Sirotkin S, Mermet A. 2012. Long-lived submicrometric bubbles in very diluted alkali halide water solutions. Phys Chem Chem Phys. 14(12):4125–4132. https://doi.org/10.1039/c2cp22858k.
Eklund F. 2019. Nanobubbles in water–how to identify them and why they are stable (PhD Diss). Chalmers University of Technology, Gothenburg, Sweden.
Fox J, Weisberg S, Adler D, Bates D, Baud-Bovy G, Ellison S, Heiberger R, et al. 2012. Package ‘car’. Vienna: R Foundation for Statistical Computing. 16(332):333.
Gislerød HR, Kempton RJ. 1983. The oxygen content of flowing nutrient solutions used for cucumber and tomato culture. Sci Hortic. 20(1):23–33. https://doi.org/10.1016/0304-4238(83)90108-5.
Han MY, Ahn HJ, Shin MS, Kim SR. 2004. The effect of divalent metal ions on the zeta potential of bubbles. Water Sci Technol. 50(8):49–56. https://doi.org/10.2166/wst.2004.0486.
Han MY, Kim MK, Ahn HJ. 2006. Effects of surface charge, micro-bubble size, and particle size on removal efficiency of electro-flotation. Water Sci Technol. 53(7):127–132. https://doi.org/10.2166/wst.2006.216.
Hewage SDA. 2020. The stability of nanobubbles and its application in contaminated sediment treatment (PhD Diss). New Jersey Institute of Technology, Newark, NJ, USA.
Jackson MB. 1980. Aeration in the nutrient film technique of glasshouse crop production and the importance of oxygen, ethylene, and carbon dioxide. Acta Hortic. 98:61–78. https://doi.org/10.17660/ActaHortic.1980.98.5.
Kim S, Kim H, Han M, Kim T. 2018. Generation of sub-micron (nano) bubbles and characterization of their fundamental properties. Environ Eng Res. 24(3):382–388. https://doi.org/10.4491/eer.2018.210.
Langenfeld N, Bugbee B. 2021. Dissipation rates of oxygen nanobubbles in recirculating systems. Utah State University Techniques and Instruments Paper 19. https://digitalcommons.usu.edu/cpl_techniquesinstruments/19/.
Morimoto T, Masuda T, Nonami H. 1989. Oxygen enrichment in deep hydroponic culture improves growth of spinach. ecb. 27(3):97–102. https://doi.org/10.2525/ecb1963.27.97.
Ponnamperuma FN. 1984. Effects of flooding on soils, p 9–45. In: Kozlowski TT (ed). Flooding and plant growth. Academic Press, Cambridge, MA, USA. https://doi.org/10.1016/B978-0-12-424120-6.50007-9.
Takahashi M, Chiba K, Li P. 2007. Free-radical generation from collapsing microbubbles in the absence of a dynamic stimulus. J Phys Chem B. 111(6):1343–1347. https://doi.org/10.1021/jp0669254.
Ulatowski K, Sobieszuk P. 2018. Influence of liquid flowrate on size of nanobubbles generated by porous-membrane modules. Chem Process Eng. 39(3):335–345. https://doi.org/10.24425/122954.
Ushikubo FY, Enari M, Furukawa T, Nakagawa R, Makino Y, Kawagoe Y, Oshita S. 2010. Zeta-potential of micro-and/or nano-bubbles in water produced by some kinds of gases. IFAC Proc Ser. 43(26):283–288. https://doi.org/10.3182/20101206-3-JP-3009.00050.
Wang L, Wang X, Wang L, Hu J, Wang CL, Zhao B, Zhang X, Tai R, He M, Chen L, Zhang L. 2017. Formation of surface nanobubbles on nanostructured substrates. Nanoscale. 9(3):1078–1086. https://doi.org/10.1039/c6nr06844h.
Wickham H. 2019. Advanced r. Chapman and Hall/CRC.
Yasui K, Tuziuti T, Izu N, Kanematsu W. 2019. Is surface tension reduced by nanobubbles (ultrafine bubbles) generated by cavitation? Ultrason Sonochem. 52(2019):13–18. https://doi.org/10.1016/j.ultsonch.2018.11.020.
Zheng Y, Wang L, Dixon M. 2007. An upper limit for elevated root zone dissolved oxygen concentration for tomato. Sci Hortic. 113(2):162–165. https://doi.org/10.1016/j.scienta.2007.03.011.