BACKGROUND:

Wastewater is an inevitable byproduct of our existence. Globally, we produce 359 billion cubic meters of wastewater per year (Jones et al., 2021). 58% of that wastewater is eventually treated in wastewater treatment facilities, whose primary purpose is the removal of nitrogen and phosphorus, dangerous diseases such as cholera, and heavy metals like zinc, cadmium, nickel and copper (Jones et al., 2021; Delaware Health and Social Services Division of Public Health, 2014). These pollutants are extremely harmful to aquatic ecosystems. The Gowanus Canal is an example of the perils of untreated wastewater: it is regularly inundated with raw wastewater overflows, and is one of the least healthy bodies of water in the United States (Riverkeeper n.d.; The City of New York DEP and Aecom, 2021). The importance of wastewater treatment is unquestionable. However, traditional methods of treatment have significant limitations, calling their use into question.

Traditional wastewater treatment plants (WWTPs) are expensive, energy intensive, and present sustainability challenges (Kesari et al.,2021). Most crucially—and most concerningly—traditional WWTPs face an environmental Catch-22 of sorts: cleaner effluent (with a higher percentage of pollutants removed) requires more energy, and thus, more carbon emissions. For one pollutant, nitrogen, Mohsenpour et al. (2021) found the energy cost to be 6.74 kilowatt hours per kilogram of nitrogen removed from wastewater. Since nitrogen harms ecosystems by causing eutrophication (dense plant growth killing other life in the ecosystem), the high energy cost of removing nitrogen in traditional wastewater treatment puts local ecological interests in conflict with broader climate goals (USGS, 2019; Conley et al., 2009; Ngatia et al., 2019; Mohsenpour et al., 2021; Foley, et. al., 2010). Developed countries, which tend to prioritize their local environments, spend as much as 3% of total electricity generated on wastewater treatment (Chae and Kang, 2013; Plappally and Lienhard, 2012; Wang et al., 2016). In the U.S., this electricity consumption caused an estimated 11.5 million tonnes of CO2 emission (Rothausen and Conway, 2011; US Department of Energy, 2016; Chisholm, 2013). However, this figure represents only the direct carbon impact of wastewater treatment: indirect emissions and toxic byproducts further impact the climate.

Another drawback of wastewater treatment is that, even though water is cleaned, the process requires increased infrastructure, chemical consumption, and the creation of toxic byproducts like sludge, which is made up of organics and minerals, including heavy metals (Foley et al., 2010; Mohsenpour et al., 2021; Seyssiecq et al., 2003). All of these drawbacks indirectly contribute to increasing CO2 emissions, especially sludge, which needs to be transported offsite and has limited applications (Wang et al., 2012; Foley et al., 2010; Gasco et al., 2005; Mohsenpour et al., 2021). While it can be repurposed as fertilizer, sludge has the potential to contaminate soil, causing further ecological harm (Mohsenpour et al., 2021). Furthermore, sludge production is positively correlated with energy consumption. Wang et al. (2012) used a linear regression to model the energy use of wastewater treatment plants as a function of sludge production, and found a R2 value of .9045. This proposed model shows a direct relationship between sludge production and energy consumption, both of which contribute to greater carbon emissions. 

In addition to carbon, traditional wastewater treatment also releases nitrous oxide (Guisasola et al., 2008; Zhang et al., 2008; Wunderlin et al., 2012). Nitrous oxide is 320 times more harmful to the climate than carbon dioxide, as nitrous oxide reacts with the ozone layer, further diminishing it (Intergovernmental Panel on Climate Change, 2007). Because the ozone layer absorbs harmful radiation from the sun, nitrous oxide emissions allow more radiation to enter our atmosphere and thus more warming (WMO 2014; Ravishankara et al., 2009). Wastewater treatment is responsible for 2.8% of global nitrous oxide emissions, and those emissions grew by 44% between 1990 and 2014 (Intergovernmental Panel on Climate Change, 2007; US EPA, 2016). Although nitrous oxide emissions are a principle failure of traditional methods, they are not the only harmful byproduct. Sludge and insufficiently purified effluent can be directly dangerous to human health through the consumption of contaminated crops (Gastco et al., 2005; Alghobar et al., 2014).

Wastewater and sludge by-products are often repurposed for agricultural purposes to increase water availability and to improve soil nutrient content through fertilization.  However, this practice often contaminates the soil and crops. For example, Alghobar et al. found that crops irrigated with treated wastewater contained higher levels of heavy metals compared to those irrigated with groundwater (2014). Gasco et al. found that sludge could leach lead and nickel into soil at levels above drinking water guidelines (2005). Furthermore, these dangerous byproducts are likely to become more prevalent as the wastewater treatment industry is projected to grow 6.5% annually until 2025 (Meticulous Market Research Pvt. Ltd., 2020). Evidently, continual reliance on traditional methods is not sustainable.

Algae wastewater treatment is a promising alternative to traditional methods as it doesn’t produce nearly as many toxic byproducts and addresses many of the pitfalls of traditional treatment. In algae wastewater treatment, nutrients and metals are absorbed as algae grows (Dalrymple et al., 2013; Salama et al., 2019; Mohsenpour et al., 2021). Specifically, wastewater is used as media for algae growth since it is rich in nutrients such as nitrogen, phosphorus, ammonia, and heavy metals, most of which promote algal health (metals such as copper, nickel, and iron promote health, while lead and cadmium are toxic to algae) (Dalrymple et al., 2013; Salama et al., 2019). Besides consuming these nutrients, algae are able to remove various toxic organic compounds—via adsorption, biodegradation, and accumulation, which are natural processes algae have evolved in response to harmful environmental stressors—minimizing pollution from effluent (Leng et al., 2020; Mohsenpour et al., 2021). Additionally, algae in wastewater production captures carbon dioxide. Almomani et al. reported that the average carbon dioxide biofixation rate—a measure of CO2 converted (captured) by plants—of mixed algae cultures in wastewater is 0.460 grams carbon per liter (of culture) per day (2019). Eucommia ulmoides, a small tree, has an approximate rate of .4 - 1 grams carbon per meter tree height per day (height measured from root depth); therefore, 15 L of culture is roughly equivalent in carbon fixation rate to one Eucommia ulmoides (Miyauchi et al., 2019). Algae is thus an efficient carbon sequester that could meaningfully contribute to lowering greenhouse gasses. Algae can also be harvested for biofuels, thereby generating energy and revenue (via the sale or use of fuel) (Mohsenpour et al., 2021; Arun et al., 2020). These properties all contribute to algae’s potential to have lower costs and be more effective than traditional methods. However, many studies, though not all, have found algal methods to be more expensive, which has hindered its popularity (Mohsenpour et al., 2021; Gouvei et al., 2016). Thus, in order for algae wastewater treatment to be widely used, it must become more economically competitive than traditional methods.  

Although the literature is divided on the current cost efficiencies of algae wastewater treatment, researchers agree that it could have significantly lower costs than traditional WWTPs, if optimized (Lundquist et al., 2015; Mohsenpour et al., 2021). Much of the literature has been focused on optimizing aeration, which oxygenates the cultures promoting algae growth, and affects cost because it is energy intensive (Liu et al., 2020; Mohsenpour et al., 2021). This narrow focus on aeration has left a large knowledge gap in other areas, notably, the effect of light wavelength. Algae are photosynthetic, so light conditions are a crucial factor to their growth, and since algae growth is the mechanism to purify wastewater, light is critical in the context of wastewater treatment. Thus, better light conditions can partially address the fiscal competitiveness of algal methods. However, there is a lack of research surrounding the mechanism for, and the effect of, light wavelength on algae growth. This research is crucial to improving algae wastewater treatment because light wavelength tolerance is highly specific to plant type. 

Plant growth performance differs under various light wavelengths. This is because chlorophyll, and other photosynthetic pigments found in plants, differentially absorb certain wavelengths of light. A seminal paper published by McCree et al. used quantum yield of CO2 assimilation, a widely used measure of photosynthetic efficiency, to investigate the efficacy of various light wavelengths on plant growth (1971). McCree and colleagues observed that red light (defined as in between 600-700 nanometers, or nm) was most efficient for 22 species of common crop plants, followed by green light (500-600 nm), then blue light (400-500 nm) (McCree 1971; Liu and van lersel Marc, 2021; Youshi et al., 2008). However, different light wavelengths provide different advantages to plants. In their review on greenhouse lighting, Singh et al. (2015) noted that red light is necessary for photosynthesis, blue is crucial for the synthesis of chlorophyll, chloroplast development, and other functions, and green can penetrate the canopy, or dense algae cultures, to drive photosynthesis (Klein, 1992). However, more research is needed to fully understand this relationship as plant growth conditions are not consistent across different species with some species even showing opposite growth responses to light wavelengths. 

Illustrating this point, Stutte et al. (2009) found an increase of biomass yield in lettuce when wavelength was increased from 660 to 690 nm (i.e. far red); but Yorio et al. (2001) noted radishes, spinach, and lettuce needed blue light for higher biomass. The literature is limited on green light, but Kim et al. (2004) and Novičkovas et al. (2010) have shown positive responses in lettuce and cucumber, while Das et al. (2011) found it ineffective in Spirulina. Singh concluded their review stating that, although optimizing light can reduce costs, the physiological processes and mechanisms determining plant response to differing light wavelengths are not fully understood (2015).  In short, despite differing responses, light wavelength is likely to have a significant effect on growth across all plants, for unknown reasons. Consequently, optimal algal light conditions are important to achieve ideal algae WWTP outcomes. 

Like previous studies investigating light wavelength effects on the growth of common crops, the literature remains conflicted on the ideal light conditions for algae. While Chojnacka et al. found that optimal illumination is necessary to maximize photosynthetic rate and nutrient removal efficiency economically, the specific ideal light conditions are dependent on the algae type (2004). Das et al.studied red, blue, white, and green lights and found that blue light achieved the highest growth rate in Nannochloropsis sp., while Wang et al. found red light to be most effective in Spirulina (2011); (2007). Das reported a growth rate of 0.64 x 106 cells per milliliter per day (d-1)  in blue light vs. 0.51 d-1 in red light—determined by measuring Optical Density680 (OD680, where the subscript refers to the wavelength), a standard method for measuring algae growth via spectrophotometer. In stark contrast, Wang et al. documented a growth rate of 0.44 d-1 at OD560 for red light, and 0.12 d-1 for blue light (2007). Although the results vary widely, it’s important to note that both studies found significant differences in algae growth and performance based on the lighting conditions, suggesting that light wavelength is extremely important to algal performance, and consequently, the economic viability of algae wastewater treatment.
Wang investigated the economic efficiency using the equation  Eeff=Cn-C0kTP   ,  where Cn is the biomass concentration (g dry weight per L) of algae harvested on the nth day, k the charge per unit of power supply ($/W-1), T the time since inoculation of cultures (days) and P is power (W). Unsurprisingly, considering their growth findings, Wang found red light, at an intensity of 1500-3000 μmol m-2 s-1, is most efficient at 70 -110 grams per liter per dollar (g L-1) $-1 , while all other wavelengths did not crack 70 (g L-1) $-1. While Wang’s work is pertinent to Spirulina, economic efficiency of other algaes is not well established. Further research on economic efficiency and light wavelength is necessary to determine even more efficient strains of algae. Chlorella vulgaris, a widely used algae for wastewater treatment, is a promising candidate to improve economic viability and the benefits of alternative wastewater treatments. However, more research is needed to optimize the growth and culturing of this strain.

Yan et al. (2013) recently determined that Chlorella vulgaris growth in wastewater is optimized with red light by using multiple photobioreactors under different lighting conditions and measured phosphorus and nitrogen removal from wastewater.  Specifically, they determined that 75% of nitrogen was removed under red light vs. only 47% by blue. This is not an isolated result: Ge et al. (2013), also investigating Chlorella vulgaris growth and nutrient removal under varying light wavelengths, found similar results, showing 81% of nitrogen removed under red light vs. 14% under blue. Building on the work of Yan and Ge et al., Zhao et al. (2015) investigated five different ratios of red to blue light and found that Red (7): Blue (3) was most effective for nitrogen removal, with 56% of TN removed, and that 7:3 achieved a higher growth rate than monochromatic light conditions, with .321 g-1 L-1 d-1, and .267 g-1 L-1 d-1, respectively. Given the mechanisms behind light wavelength and growth—especially that blue light is crucial for the synthesis of chlorophyll, chloroplast development, and other functions — it’s logical that optimizing the ratios of red to blue light may prove to be an effective strategy (Singh et al., 2015). However, other work has shown that supplementing red and blue light with green light enhances growth in seedlings and lettuce. For example, Kim et al. measured a 30% increase in lettuce leaf area when adding 24% green light compared to red and blue alone, and Folta found that seedlings grown under RGB light were 25% taller than those grown under RB (2004); (2004). Green is often un-rigorously dismissed as photosynthetically inefficient, without considering the physiological benefits beyond photosynthesis (confirmed by the data above (Singh et al., 2015)).  This conclusion is also consistent with Klein’s 1992 discovery that green light is more penetrative than other light types and reaches lower canopy leaves, a mechanism which potentially benefits algae growth by penetrating dense algal cultures. These findings warrant further investigation into the implications for algae WWTP, where there has been no literature on using RGB ratios. 

The present study aims to build upon the work of Zhao et al., while considering the implications of Kim et al., in order to evaluate the effect of a mixture of light wavelengths on the growth and nutrient removal of Chlorella vulgaris in synthetic wastewater. This study has four arms: a positive control red (7): blue (3), a control white light, a negative control red (1): blue (9) and an experimental arm with red (5.25): blue (2.25): green (2.4), with each arm containing three biological replicates. These arms are derived from Zhao, who found red (7): blue (3) to be most effective with a growth rate of 0.321 ± 0.08 / d, and red (1): blue (9) to be least effective (0.276 ± 0.08 / d), and the Kim et al., who found small proportions of greenlight, specifically 24%, to be most effective (2015; 2004); the experimental arm maintains a 7:3 ratio between red and blue while containing 24% green light. Synthetic wastewater—used by both Yan and Ge et al.—is a safe alternative to real wastewater, which contains dangerous pathogens such as E. coli and cholera, as well as fecal matter and antibiotics (Delaware Health and Social Services Division of Public Health, 2014; Leng et al., 2020). Synthetic wastewater contains the nutrients that are most problematic in real wastewater (aforementioned nitrogen and phosphorus), has a consistent formulation, and offers a suitable model to avoid the dangers of handling untreated wastewater (Yan et al., 2013; Ge et al., 2013; Mohsenpour et al., 2021; Ye et al., 2009). Thus, although synthetic alternatives are used (in particular, the recipe is adapted from a standardized OECD 1996 paper), the findings associated with this study are applicable to real life wastewater treatment. 

While Zhao used the standard method’s ascorbic acid method to determine total phosphorus levels (American Public Health Association, 1995), I was unable to measure nutrient content for this study because no suitable assay was identified. However, since algae growth is the mechanism of nutrient removal, tracking growth provides sufficient measurement of performance. Zhao used filtration to measure algae growth, which previous SRD students found to be unreliable and frustrating in the classroom setting. Last year I isolated algae pellets from the supernatant and found their mass, but that method proved unreliable (R2 value of .05 for linear regression); given that, I tried estimating growth using a Secchi Stick. While Secchi Stick growth approximations are not suitable for published research, the estimates produced a regression that was sufficient to track culture growth. However, since the algae cultures were never dense enough to measure with a Secchi Stick, I instead measured their optical density at OD600 using a benchtop spectrophotometer. This method is accepted in the literature and suitable for published research; it also proved to be the most efficient in the classroom setting. I anticipated the red (5.25): blue (2.25): green (2.4) would prove to be most effective for growth and nutrient removal because of Singh’s analysis of the benefits of ratios of red and blue to chloroplasts (red is necessary for photosynthesis, blue is crucial for the synthesis of chlorophyll, chloroplast development, and other functions), Zhao’s work showing a 7:3 ratio to be most ideal for wastewater treatment, Kim’s work showing growth boost to plants under red:blue ratios when 24% green is introduced, and finally Klein’s finding of green light’s penetrative properties. 

The wide ranging applications of algae wastewater treatment—extremely clean effluent, carbon capture, and biofuel production—make it one of the most important emerging technologies. Optimizing light wavelength is an important first step to the mass adoption of algal treatment, because optimal illumination will help address the fiscal inviability of algal methods, the main factor hindering their popularity. Adoption of algal treatment methods will improve local scale ecosystem health while addressing macroscale climate concerns. If the results of this study are conclusive, green light will be established as an important factor to optimize algae growth and, ultimately, sewage treatment outcomes. Future work could build on this conclusion, exploring the optimal wavelengths of green combined with fixed ratios of other light colors. As discussed earlier, Kim et al. (2004) found 24% green light with red and blue light was most effective for lettuce. In a later study, Jonkan et al. (2012) investigated green wavelengths specifically and found 510 nm to produce the best lettuce growth, but work on the most ideal mixtures and wavelengths in algae wastewater treatment has been limited. Because of the physiological relationship between light wavelength and growth—red is necessary for photosynthesis, blue is crucial for the synthesis of chlorophyll, chloroplast development, and other functions, and green has penetrative properties (Singh et al., 2015; Klein, 1992)—and the large difference in performance between the nuances of light ratios throughout the entire plant lighting literature, future work should focus on refining the ratios and precise wavelengths in RGB mixtures to produce the best treatment outcomes, with respect to nutrient removal and economic efficiency. 

METHODS & PROCEDURE:

Overview

The present study expands on several seminal studies regarding light wavelength, algae wastewater treatment, and plant growth. Three papers are important to emphasize. First, Zhao et al. (2015), whose study investigated five different ratios of red to blue light and found that some ratios of light were more effective than monochromatic light conditions for algae wastewater treatment. Given the mechanisms behind light wavelength and growth—especially that blue light is crucial for the synthesis of chlorophyll, chloroplast development, and other functions—it’s logical that optimizing the ratios of red to blue light may prove to be an effective strategy for treatment (Singh et al., 2015). The second paper, Kim et al. (2004), measured a 30% increase in lettuce leaf area in lettuce when adding 24% green light compared to a dichrome red : blue ratio alone. Green light is often prematurely dismissed as photosynthetically inefficient without considering the physiological benefits beyond photosynthesis, confirmed by the data above (Singh et al., 2015). Green light may be particularly effective for algae growth, based on the third paper by Klein (1992). Their team discovered that green light is more penetrating than other light types and reaches lower canopy leaves. Thus, green light could potentially probe deeper into algal culture containers and thus enhance growth. 

The present study aims to build upon the work of Zhao et al., while considering the implications of Kim et al. and Klein, to evaluate the effect of a mixture of light wavelengths on the growth and nutrient removal of Chlorella vulgaris in synthetic wastewater. Algae growth occurs in tandem with nutrient removal, as the wastewater acts as a growth media for the algae and is purified as the algae grows and “feeds” on the nutrients (Dalrymple et al., 2013). Nutrient removal, particularly of phosphorus and nitrogen, is the primary goal of wastewater treatment, and thus phosphorus ppm is a dependent variable in our study, along with growth, as both track the purification of the wastewater. (Yan et al., 2013; Ge et al., 2013; Mohsenpour et al., 2021; Ye et al., 2009). Nitrogen was not measured because no suitable assays for the SRD classroom were found. This study has four arms, which are four different light ratios: a positive control red (7): blue (3), a negative control white light, a negative control red (1): blue (9) and an experimental arm with red (5.25): blue (2.25): green (2.4), with each arm containing three biological replicates. These arms are derived from Zhao, who found red (7): blue (3) to be most effective at demonstrating the correlation that certain mixtures of light wavelength will improve algae growth, and red (1): blue (9) to be least effective, and Kim et al., who found small proportions of greenlight, specifically 24% of total light, to be most effective (2015; 2004); the experimental arm maintains a 7:3 ratio between red and blue while containing 24% green light, thus combining the two findings to potentially optimize growth. White light is standard and widely used for plant growth, and demonstrates no correlation between manipulating light ratio and improved growth, as white light is a consistent ratio of the whole visible light wavelength spectrum (Singh et al 2015). Light intensity, measured by Lux meter (Dr. Meter Digital Lux Meter Model LX1330B), was kept consistent at 750 ± 250 lux. 

Chlorella vulgaris Algae was ordered (Algae Research Supply ID: ACCv-01000) and then cultivated in new cultures before being inoculated into wastewater cultures (see section “Inoculation”). The wastewater cultures were stored in 500 mL plastic erlenmeyer flasks (Product ID: 726828), which was similar to Ge et al’s (2013) study of C. vulgaris. in wastewater treatment, who used 1 L erlenmeyer flasks. 25 mL samples, poured directly from the culture flasks into freestanding 50 mL tubes (ID:41121700), were taken daily for five days (starting three days after inoculation, to give time for growth, a timetable supported by my preliminary data last year which found little growth in the days immediately after inoculation). When counting the initial three growth days, this matches the eight day period used in Ge et al.’s study, and is longer than Zhao’s six day period, giving ample time to observe growth of cultures (2013; 2015).  Daily growth measurements were taken via OD600 optical density measurements taken using a spectrophotometer (see section “Growth Measurement”). Qualitative observations and photos were also recorded daily in a lab notebook and a qualitative data organizer-chart, noting perceived growth, color, optical density, or abnormalities. 

Safety

This study involves the use of chemicals and heavy metals. Thus PPE (labcoat, goggles, gloves, and close toed-shoes) was worn throughout my research, and safety practices were informed by the MSDS(s) of chemicals used (Carbamide/Urea, MSDS; NaH2PO4, MSDS; KH2PO4, MSDS; CaCl2, Number: MSDS; MgSO4, MSDS; Sodium Hydroxide Solution, MSDS; Hydrochloric Acid Solution, MSDS) provided by chemical/material makers. Eyewear, a lab coat, and gloves reduced chemical exposure to skin, eyes, and hands in accordance with MSDS instructions. In particular, strict safety procedures regarding eyewear, lab coat, and gloves were followed during phosphorus measurement (ID: HI706-25; Reagent A MSDS, Reagent B MSDS), to reflect the potential dangers of the measurement reagent chemicals. When the soldering iron was used, a safety protocol was followed, and hand washing was taken seriously to minimize lead exposure (Hudson Valley Community College, n.d.). 

Initial Culturing 

Chlorella vulgaris was ordered from Algae Research Supply (ID: ACCv-01000), and grown, in the original container with the cap removed, under 3000 ±1000 lux white light in a Johnny Seeds Full-Size Seedling Light Cart with 640 Watt Lights (ID: 7026). This initial algae was the stock culture. This light intensity was chosen based on Carolina Biological’s algae care guide (James, 2012). 

After several days of growth (3-10 days), new cultures, referred to as “Secondary Cultures,” were inoculated using the original stock culture, distilled water, and Algae Nutrient Media (ID: ANF2-01000), at dilutions of 1 mL / L of both algae and media. Culture and Media were measured using a P1000 volumetric pipette, while distilled water was measured with a 1000 mL graduated cylinder. This approach is an algae care best practice consistent with the care guide provided by Algae Research Supply and Carolina’s algae care guide (James, 2012). Secondary cultures were grown under the same conditions as the stock culture. The secondary cultures were important to the experimental design as they allowed the algae to adjust to SRD classroom conditions before being used for wastewater treatment, and allowed algae to grow rapidly in the presence of nutrients, ensuring healthy cultures before inoculation into wastewater. Additionally, creating secondary cultures created more algae, serving an auxiliary role in case of unexpected death or procedural mistakes. 

Light Ratios

A major adaptation to the SRD classroom was the construction of LED light panels and a shelving unit to control the lighting conditions. This differed from Ge and Zhao, who both used incubators. Because incubators can cost thousands of dollars and use considerable space for a niche application, it was not feasible to order one for thise study. Furthermore, the model used by both Ge and Zhao, the SPX-400I-G made by Boxun Industry & Commerce Co., is not available on their website. 

The panels were constructed using standard home-LED strips (Amazon ID:B091YY2PSB) and styrofoam blocks (Amazon ID:B0858KJWDF), where the strips were mounted onto the blocks with 3M Double Sided Mounting Tape (Amazon ID:B0B4MTX8RQ), and the strips were powered and controlled by their accompanying bluetooth/remote enabled power sources, which allowed their color and intensity to be controlled with the included remote (Amazon ID:B091YY2PSB). Proportions of the lights were set to different colors to achieve the ratios. In order to achieve those proportions, multiple colors needed to be activated at once. Thus, the LED strips were cut and then soldered at their banded connection points, such that on certain strips only the “B” (Blue diode) channel was connected, and on others only the “R” (Red diode) channel was connected.  For the novel lighting condition, one strip required the “G” green diode to be connected. Disconnecting and reconnecting via soldering allowed certain colors to be isolated so that two strips plugged into the same power source could each have their own color by setting the powersource to violet (a combination of red and blue diodes, where the incorrect/extra color would be unable to proceed past the cut/solder point and thus the diodes light only one color). Nine panels were constructed: 2 panels with 63 “red” diodes, 1 panel with 96 “red” diodes (with the 96th covered by tape), 3 panels of 54 “blue” diodes, 1 panel of 54 “blue” diodes and 18 “red” diodes, 1 panel with 42 “blue” diodes (the 42nd is covered) and 45 green diodes, and 2 panels with 90 “white” diodes. See photo of a completed panel below. 

The photo shows the layout of the LED strip on the styrofoam block, where the strip was arranged tightly with hairpin turns at the edge of the block. The tight hairpin turns allowed up to 100 diodes to fit on a single panel, while loose turns might afford enough space for approximately 70. The turns created tension which loosened the strips requiring them to be secured to the foam with 3M mounting tape before and after each turn (with tape placed directly underneath the final diode before the turn and the first diode after). Care was taken to ensure that all diodes were completely flat on the surface of the block so as to obtain the most consistent lighting possible. Thus, the hairpin turns were positioned so that no diodes fell on the turn, which may have obstructed or misdirected them. The researcher also took pains to ensure that the cut/solder points were located such that they had low tension, so that those relatively fragile points would experience minimum stress. This was achieved by placing the cut/solder point either off the panel or on the panel surrounded by tape. 

It is important to note that not all of the LED panels looked exactly like the one pictured. They varied in number of diodes (as shown in the list of panels earlier in this subsection), and some panels featured more than one color. In the case of those panels, the second color was started from the opposite corner of the first, and laid out in the same manner as the first. This is shown in the diagram below, of the green/blue panel. 

The panels were grouped according to the table below to create the light ratios

Shelving Unit

As referenced in the overview, the study has four arms and thus four light conditions. 180 diodes were used per condition, with either two or three panels per condition, as shown in the figure below (the diodes per panel/distribution of diodes is described under “Light Ratios,” the previous subsection). Two TiFFCOFiO Power Surge protectors (ID: B09256HSB4) were used to power the light panels. The panels were connected to power sources (each powersource powering two panels), and the power sources were plugged into the surge protectors according to the figure. The panels were fixed to the shelves above, facing downwards, using soft copper wire (ID:B003B91BTG). An awl was used to poke two holes near the edge of the panel, which the wire was threaded through and tied to the shelf above. This was repeated on the opposite side of the panel to ensure it was secure and level (parallel to the shelves). On some panels, this was supplemented by duct tape for extra support, if it was needed for the panel to remain level and snug to the shelf above. 

Although the number of diodes was consistent, the lux varied slightly: across the shelves it was 750 ± 250 lux. This variation is not ideal and is potentially a confounding factor in the experiment. Unfortunately, I was unable to mitigate this issue because it seems inherent to the design of the panels, and because of time constraints I was not able to address the root cause. 

The second part of the adaption from the incubator was the shelving unit, which held all the cultures and lighting conditions across its four shelves. The shelving unit was completely covered with tinfoil to eliminate any light from leaking out, and, crucially, to prevent any overhead/ambient white light from leaking in. Light escaping lowers light intensity and would impact the light ratios, causing a potentially confounding factor in the experiment.The unit also created four independent lighting environments inside via tinfoil, which was also important to minimize confounding factors and systematic error, because the light ratios could dilute/influence each other if not sufficiently separated. This was achieved by covering each shelf floor in tinfoil, which, in conjunction with the foil on the outside, completely separated the four shelves and thus the four light conditions. See figure below of the shelving unit.  

Then, a 4x concentrated wastewater solution was created according to the following table, again using the P1000 pipette and Flinn Analytical balance (with the balance used to measure the carbamide, which was transferred with the same method as for the previous solutions, while the P1000 was used to transfer the solutions 1-4). The solution was created in a 1 L volumetric flask and stored in that flask after being sealed with parafilm. 

After the solution was homogenized by swirling and inverting the contents in the 1 L flask, it was diluted using 3 L of distilled water and stored in an eight gallon water jug sealed with parafilm. Then, the pH was calibrated using a Hanna pH Meter (ID: HI9810412; Hanna Instruments, n.d.). The pH meter was calibrated using the provided calibration solutions and the manual was followed for operation. The contents of the flask were transferred to a large gray bucket, and the solution was stirred manually while it was titrated with either Sodium Hydroxide Solution, 0.1 M, (ID: S0149)(MSDS) or Hydrochloric Acid Solution, 0.1 M, (ID:H0014)(MSDS). Based on the initial reading, the wastewater was adjusted using the acid if it was above 6.65, and with the hydroxide if it was below 6.35 (neither was added if the wastewater was already within the range). One drop of solution was added at a time via a generic pipette. The meter was given time to adjust to the new reading between drops, which were repeated until the pH was within the range. Titration was important to allow the algae to survive, and present in the work of Zhao and Ge et al. 

Inoculation 

As mentioned earlier, the study had four arms with three biological replicates each, and thus 300 mL of wastewater was transferred into each of the twelve plastic erlenmeyer flasks (Product ID: 726828). The wastewater was measured using a 1000 mL graduated cylinder. Using the P1000 pipette, 2 mL of algae derived from the secondary culture was inoculated into each flask. 

Growth Measurement

Zhao used filtration to measure algae growth, which previous SRD students found to be unreliable and frustrating in the classroom setting. Previous work attempted to isolate algae pellets from the supernatant and determine their mass, but that method proved unreliable (R2 value of .05 for linear regression); given that, I measured optical density at OD600 using a benchtop spectrophotometer. The spectrophotometer is a sophisticated and reliable instrument and proved efficient for the classroom. Samples were shaken, transferred to cuvettes, and then measured in the spectrophotometer. The cuvettes were wiped with paper towels before measurement to obtain accurate results. 

Data Analysis

All data was recorded in Google Sheets. Growth data was analyzed using linear and exponential regressions. Average growth rates and phosphorus measurements were evaluated across different light conditions using a two tailed independent t-test. P values < .05 were regarded as significantly different. 

RESULTS AND DISCUSSION:

Figure 1 shows growth on the y axis (measured via optical density at OD600)  vs. time (days) on the x axis (days)  for all four arms over four days. Sampling began four days after inoculation and was conducted daily. The positive and negative control arms did not demonstrate the expected correlation, with the negative control (.318) far outperforming the positive control (.214). Additionally, the RGB group was the worst performing group (.122), which was the opposite of the hypothesis. These results were inconsistent with Zhao’s findings as 7:3 (.214) underperformed 1:9  (.318) and white (.219) (Zhao found 0.321 ± 0.08 g/d for 7:3, vs.0.276 ± 0.08 g/d and .276± 0.08 g/d for white), and inconsistent with Ge and Yan’s findings as the predominantly red colored light group 7:3 underperformed monochrome white, whereas Ge and Yan both showed monochrome red outperforming white (Yan:  248 mg/L under red vs. 111 mg/d under white. Ge: 120 mg/d  vs. 207 mg/L).

This data suggests an invalid approach and thus neither supports nor refute the hypothesis. The positive control arm, derived from Zhao’s data showing a growth rate 0.321 ± 0.08 g/d for the red 7: blue 3 ratio, showed absorbance of .214, a marked underperformance of the negative control arm red 1: blue 9’s absorbance of .318. The negative control was also derived from Zhao who found a rate of 0.276 ± 0.08 g/d; thus the expected relationship was a ratio of 1.2:1 between the growth of 7:3 and 1:9 and the actual result was .673:1.

The controls did not yield the expected outcomes suggesting significant error. In the experiment, there is always possible random error when making the wastewater solution, which has the potential to significantly impact growth as the concentration of nutrients in the solution is low thus any deviation from the recipe is significant. Error with the pipette such as air bubbles and other inconsistencies between measurement, perhaps even miscounting the number of pipette transfers, could have significantly affected nutrient content and thus growth. Error when inoculating algae with the P1000 pipette might also significantly impact growth, as varying levels of inoculum will exponentially differ in growth over time. 

While the erlenmeyer flasks were autoclaved, the use of plastic erlenmeyer flasks with varying level of opacity was a source of systematic error (SoE). As the flasks aged, they developed marks which may have impacted the ability of light to filter through to algae. Furthermore, some flasks displayed significantly more aging than others, with some being completely yellowed and others remaining near-pristine. Additionally, the algae occasionally remained extremely stuck to the bottom of certain flasks, and occasionally one flask in a particular arm would perform markedly worse than the other replicates for no apparent reason. For example, flask b in trial 3 (one of two trials averaged into the data above) had absorbance of .047, compared to an average of .216 for the other two replicates. This may have been influenced by the design of the light panels, which did not cast completely even light but instead often lit sections of separately colored light. This too may have influenced data in trials two and three (the data averaged here) where the flasks were never rotated, compared to the successful trial where the flasks changed position daily.  

Figures 2a-2c show growth, on the y axis (measured via optical density at OD600) vs. time (days) on the x axis for all four arms over four days. Sampling began four days after inoculation and was conducted daily. The data replicated Zhao’s result that red 7: blue 3 caused higher growth than red 1: blue 9 and monochrome white, with the present study yielding total absorbance (the study’s assay for growth) of .299 vs. .203 vs. .257, and Zhao finding .321 g/d vs. 276 g/d vs. 0.267 ± 0.05 (2015). This is also consistent with Ge et al and Yan et al’s findings that red light causes higher growth than blue light (Ge found growth of 248 mg/L under red vs. 94 mg/L under blue, while Yan found 120 mg/d vs. 43 mg/d),as the predominantly red ratio outperformed the predominantly blue ratio (.299 vs. .203). The data also partially replicates their finding that red light performed better than white light (Ge found growth of 248 mg/L under red vs. 207 mg/L under white, while Yan found 120 mg/d vs. 111 mg/d), as 7:3 and RGB both outperformed white light (.299 and .347 vs. .257). While Red 1: Blue 9 underperformed white light, Zhao’s showed 1:9 at .276± 0.08 g/d and white at .267± 0.06. However, Yan and Ge both showed blue underperforming white (Yan: 111 mg/d vs. 43 mg/d. Ge:  207 mg/L vs. 94 mg/L), so my result is not entirely unexpected. In sum, replicating these findings thus supports the experimental design as it shows that the research was able to independently verify accepted knowledge, suggesting that confounding variables and SoE were sufficiently minimized. 

The experimental RGB ratio outperformed to the 7:3, encouraging as that was the previous best known ratio in the literature. Specifically, the RGB ratio had 16% higher absorbance than the 7:3 ratio, which compares unfavorably to Kim’s 30% increase in leaf area in lettuce when R:B is supplemented with green, but is still a measurable and valuable improvement.  However, it is important to note that this performance gap was only visible after excluding an outlier, in which growth was significantly lower. This can be seen when viewing the second figure, where 7:3 outperformed RGB by 13%, which in turn had a higher growth rate than white and 1:9. This is a 29 point swing between outlier inclusion and exclusion. Because the performance is tied to the outlier, this is only a promising indication of the potential of the RGB ratio, and limited conclusions regarding its efficacy in relation to 7:3 can be drawn. Figure 1c shows the growth data from the first figure analyzed using exponential regression. Exponential regression was chosen because it models algae growth, which is exponential. This underscores the importance of small differences in growth rate, as the gap in total growth increases exponentially over time. The R2 values, a measurement of how well the curve predicts the data, ranged from .91 - .99, where the RGB had the highest value and white had the lowest value. A higher R2 value in this context represents a more normal and thus healthier growth pattern. Therefore the RGB curve value of .99 represents a positive finding that supports my hypothesis that RGB is the ideal ratio for algae wastewater treatment.

That considered, more data is necessary to fully determine the validity of the hypothesis. First, data on nutrient removal, particularly of nitrogen and phosphorus, is important to evaluate treatment performance. The acquisition of an advanced water quality photometer could aid future SRD students in a wide array of water research including the measurement of nutrient content in wastewater. While the present study based experimental procedure on the relevant literature, there are adjustments that future research should take advantage of to reduce sources of systematic error (SoE). Future research should use glass erlenmeyer flasks instead of plastics, as algae often stuck to the plastic, which potentially affects the growth measurement because optical density is not a reliable indicator of growth if the algae is not diffused throughout the solution. Additionally, the plastic became discolored increasing its opacity, potentially impacting the amount of light reaching the algae. Correcting these SoE might give future researchers the opportunity to collect more data showing performance increase of cultures under the RGB ratio compared to cultures under the 7:3 ratio, which would concretely validate the hypothesis. Validating this conjecture—especially with research that replicated the findings in a larger, more realistic treatment setting, and research using a more sophisticated lighting method—would be a significant step towards discovering the ideal lighting conditions for algae wastewater treatment, which in turn would be significant to the wider adoption of algal methods. Building on those conclusions, the validity of green light in R:B ratios for other plant species and horticultural contexts more broadly could be further studied, which would represent a huge paradigm shift in the literature which has widely dismissed greenlight. 

This shift could bring the opportunity to explore the optimal wavelengths of green combined with fixed ratios of other light colors. As discussed earlier, Kim et al. (2004) found 24% green light with red and blue light was most effective for lettuce. In a later study, Jonkan et al. (2012) investigated green wavelengths specifically and found 510 nm to produce the best lettuce growth, but work on the most ideal mixtures and wavelengths in algae wastewater treatment has been limited. Because of the physiological relationship between light wavelength and growth—red is necessary for photosynthesis, blue is crucial for the synthesis of chlorophyll, chloroplast development, and other functions, and green has penetrative properties (Singh et al., 2015; Klein, 1992)—and the large difference in performance between the nuances of light ratios throughout the entire plant lighting literature, future work should focus on refining the ratios and precise wavelengths in RGB mixtures to produce the best treatment outcomes, with respect to nutrient removal and economic efficiency. 

SUPPLEMENTAL INFO AND FIGURES:

Lab Notebook:

Lab Notebook

References:

Alghobar, M. A., Ramachandra, L., & Suresha, S. (2014). Effect of sewage water irrigation on soil properties and evaluation of the accumulation of elements in Grass crop in Mysore city, Karnataka, India. American Journal of Environmental Protection, 3(5), 283-291.

Almomani, F., Judd, S., Bhosale, R. R., Shurair, M., Aljaml, K., & Khraisheh, M. (2019). Integrated wastewater treatment and carbon bio-fixation from flue gases using Spirulina platensis and mixed algal culture. Process Safety and Environmental Protection, 124, 240-250.

APHA-AWWA-WPCF: Standard Methods for the Examination of Water and Wastewater. Washington DC: American Public Health Association; 1995:17–185.

Arun, S., Sinharoy, A., Pakshirajan, K., & Lens, P. N. (2020). Algae based microbial fuel cells for wastewater treatment and recovery of value-added products. Renewable and Sustainable Energy Reviews, 132, 110041.

Chae, K. J., & Kang, J. (2013). Estimating the energy independence of a municipal wastewater treatment plant incorporating green energy resources. Energy Conversion and Management, 75, 664-672.

Change, I. C. (2014). Mitigation of climate change. Contribution of working group III to the fifth assessment report of the intergovernmental panel on climate change, 1454, 147.

Chisholm, A. (2013). A blueprint for carbon emissions reductions in the UK water industry. London: CIWEM.

Chojnacka, K., & Noworyta, A. (2004). Evaluation of Spirulina sp. growth in photoautotrophic, heterotrophic and mixotrophic cultures. Enzyme and microbial technology, 34(5), 461-465.

The City of New York Department of Environmental Protection & Aecom USA Inc. (2015, June 1). June 2015 Combined Sewer Overflow Long Term Control Plan for Gowanus Canal. NYC.gov. Retrieved December 12, 2021, from https://www1.nyc.gov/assets/dep/downloads/pdf/water/nyc-waterways/gowanus-canal/gowanus-canal-ltcp-201506.pdf

Conley, D. J., Paerl, H. W., Howarth, R. W., Boesch, D. F., Seitzinger, S. P., Havens, K. E., ... & Likens, G. E. (2009). Controlling eutrophication: nitrogen and phosphorus. Science, 323(5917), 1014-1015.

Dalrymple, O.K., Halfhide, T., Udom, I. et al. Wastewater use in algae production for generation of renewable resources: a review and preliminary results. Aquat. Biosyst. 9, 2 (2013). https://doi.org/10.1186/2046-9063-9-2

Das, P., Lei, W., Aziz, S. S., & Obbard, J. P. (2011). Enhanced algae growth in both phototrophic and mixotrophic culture under blue light. Bioresource Technology, 102(4), 3883-3887.

Delaware Health and Social Services Division of Public Health. (2014, September). Frequently Asked Questions Sewage. SEWAGE. Retrieved May 28, 2022, from https://dhss.delaware.gov/dph/files/sewagefaq.pdf

Foley, J., De Haas, D., Hartley, K., & Lant, P. (2010). Comprehensive life cycle inventories of alternative wastewater treatment systems. Water research, 44(5), 1654-1666.

Folta, K. M. (2004). Green light stimulates early stem elongation, antagonizing light-mediated growth inhibition. Plant Physiology, 135(3), 1407-1416.

Gascó, G., Lobo, M. C., & Guerrero, F. (2005). Land application of sewage sludge: A soil columns study. Water SA, 31(3), 309-318.

Ge et al. Purifying synthetic high-strength wastewater by microalgae chlorella vulgaris under various light emitting diode wavelengths and intensities. Journal of Environmental Health Science and Engineering 2013 11:8.

Gouveia, L., Graça, S., Sousa, C., Ambrosano, L., Ribeiro, B., Botrel, E. P., ... & Silva, C. M. (2016). Microalgae biomass production using wastewater: treatment and costs: scale-up considerations. Algal Research, 16, 167-176.

Greenberg, A. E., Clescerl, L. S., & Eaton, A. D. (Eds.). (1992). Standard Methods: For the Examination of Water and Wastewater (18th Edition ed.). American Public Health Association.

Guisasola, A., de Haas, D., Keller, J., & Yuan, Z. (2008). Methane formation in sewer systems. Water Research, 42(6-7), 1421-1430.

Hanna Instruments. (2018). Instruction Manual : HI706 Phosphorus High Range. Hanna Instruments. https://www.hannainst.com/hubfs/product-manuals/IST706_05_18-1.pdf?hsLang=en

Johkan, M., Shoji, K., Goto, F., Hahida, S. N., & Yoshihara, T. (2012). Effect of green light wavelength and intensity on photomorphogenesis and photosynthesis in Lactuca sativa. Environmental and Experimental Botany, 75, 128-133.

Jones, E. R., van Vliet, M. T., Qadir, M., & Bierkens, M. F. (2021). Country-level and gridded estimates of wastewater production, collection, treatment and reuse. Earth System Science Data, 13(2), 237-254.

Kesari, K.K., Soni, R., Jamal, Q.M.S. et al. Wastewater Treatment and Reuse: a Review of its Applications and Health Implications. Water Air Soil Pollut 232, 208 (2021). https://doi.org/10.1007/s11270-021-05154-8

Kim, Hyeon-Hye, Gregory D. Goins, Raymond M. Wheeler, and John C. Sager. "Green-light supplementation for enhanced lettuce growth under red-and blue-light-emitting diodes." HortScience 39, no. 7 (2004): 1617-1622.

KLEIN, R. M. (1992). Effects of green light on biological systems. Biological Reviews, 67(2), 199-284.

Leng, L., Wei, L., Xiong, Q., Xu, S., Li, W., Lv, S., ... & Zhou, W. (2020). Use of microalgae based technology for the removal of antibiotics from wastewater: a review. Chemosphere, 238, 124680.

Liu, J., Pemberton, B., Lewis, J., Scales, P. J., & Martin, G. J. (2020). Wastewater treatment using filamentous algae–a review. Bioresource Technology, 298, 122556.

Liu, J., & Van Iersel, M. W. (2021). Photosynthetic physiology of blue, green, and red light: Light intensity effects and underlying mechanisms. Frontiers in plant science, 12, 328.

Lundquist, T., Woertz, I., Hutton, M., Spierling, R., Blackwell, S., & Crowe, B. (2015, June 24th). Wastewater Reclamation and Biofuel Production Using Algae. Wastewater Reclamation and Biofuel Production Using Algae. Retrieved December 12, 2021, from https://www.energy.gov/sites/prod/files/2015/07/f24/lundquist_bioenergy_2015.pdf

McCree, K. J. (1971). The action spectrum, absorptance and quantum yield of photosynthesis in crop plants. Agricultural Meteorology, 9, 191-216.

Meticulous Market Research Pvt. Ltd. (2020, 06 18). Water and Wastewater Treatment Market Worth $211.3 Billion by 2025, Growing at a CAGR of 6.5% from 2019- Global Market Opportunity Analysis and Industry Forecasts by Meticulous Research. GlobeNewswire. https://www.globenewswire.com/news-release/2020/06/18/2050151/0/en/Water-and-Wastewater-Treatment-Market-Worth-211-3-Billion-by-2025-Growing-at-a-CAGR-of-6-5-from-2019-Global-Market-Opportunity-Analysis-and-Industry-Forecasts-by-Meticulous-Researc.html#:~:

Miyauchi, T., Machimura, T., & Saito, M. (2019). Estimating carbon fixation of plant organs for afforestation monitoring using a process‐based ecosystem model and ecophysiological parameter optimization. Ecology and Evolution, 9(14), 8025-8041.

Mohsenpour, S. F., Hennige, S., Willoughby, N., Adeloye, A., & Gutierrez, T. (2021). Integrating micro-algae into wastewater treatment: A review. Science of the Total Environment, 752, 142168.

Ngatia, L., Grace III, J. M., Moriasi, D., & Taylor, R. (2019). Nitrogen and phosphorus eutrophication in marine ecosystems. Monitoring of marine pollution, 1-17.

Novičkovas, A., Brazaitytė, A., Duchovskis, P., Jankauskienė, J., Samuolienė, G., Virsilė, A., ... & Zukauskas, A. (2010, August). Solid-state lamps (LEDs) for the short-wavelength supplementary lighting in greenhouses: experimental results with cucumber. In XXVIII International Horticultural Congress on Science and Horticulture for People (IHC2010): International Symposium on 927 (pp. 723-730).

OECD: Guidelines for Testing of Chemicals Simulation Tests-Aerobic Sewage Treatment. Technical Report. Paris: Organisation for Economic Co-operation and Development (OECD); 1996:19–142. 

Plappally, A. K. (2012). Energy requirements for water production, treatment, end use, reclamation, and disposal. Renewable and Sustainable Energy Reviews, 16(7), 4818-4848.

Ravishankara, A. R., Daniel, J. S., & Portmann, R. W. (2009). Nitrous oxide (N2O): the dominant ozone-depleting substance emitted in the 21st century. science, 326(5949), 123-125.

Riverkeeper. (n.d.). GOWANUS CANAL. Riverkeeper. Retrieved December 12, 2021, from https://www.riverkeeper.org/wp-content/uploads/2009/09/Gowanus_Fact_Sheet.pdf

Rothausen, S. G., & Conway, D. (2011). Greenhouse-gas emissions from energy use in the water sector. Nature Climate Change, 1(4), 210-219.

Salama, E. S., Roh, H. S., Dev, S., Khan, M. A., Abou-Shanab, R. A., Chang, S. W., & Jeon, B. H. (2019). Algae as a green technology for heavy metals removal from various wastewater. World Journal of Microbiology and Biotechnology, 35(5), 1-19.

Seyssiecq, I., Ferrasse, J. H., & Roche, N. (2003). State-of-the-art: rheological characterisation of wastewater treatment sludge. Biochemical Engineering Journal, 16(1), 41-56.

Singh, D., Basu, C., Meinhardt-Wollweber, M., & Roth, B. (2015). LEDs for energy efficient greenhouse lighting. Renewable and Sustainable Energy Reviews, 49, 139-147.

Stutte, G. W., Edney, S., & Skerritt, T. (2009). Photoregulation of bioprotectant content of red leaf lettuce with light-emitting diodes. HortScience, 44(1), 79-82.

U.S. Department of Energy. (2017). Electric Power Annual 2016. Independent Statistics & Analysis, Washington.

US Greenhouse Gas Inventory Report: 1990-2014 | US EPA. (2020, September 10). US Environmental Protection Agency. Retrieved May 4, 2022, from https://www.epa.gov/ghgemissions/us-greenhouse-gas-inventory-report-1990-2014

USGS. (2019). Nutrients and Eutrophication | US Geological Survey. USGS. Retrieved June 1, 2022, from https://www.usgs.gov/mission-areas/water-resources/science/nutrients-and-eutrophication

Wang, C. Y., Fu, C. C., & Liu, Y. C. (2007). Effects of using light-emitting diodes on the cultivation of Spirulina platensis. Biochemical Engineering Journal, 37(1), 21-25.

Wang, H., Yang, Y., Keller, A. A., Li, X., Feng, S., Dong, Y. N., & Li, F. (2016). Comparative analysis of energy intensity and carbon emissions in wastewater treatment in USA, Germany, China and South Africa. Applied Energy, 184, 873-881.

Wang, X., Liu, J., Ren, N. Q., & Duan, Z. (2012). Environmental profile of typical anaerobic/anoxic/oxic wastewater treatment systems meeting increasingly stringent treatment standards from a life cycle perspective. Bioresource technology, 126, 31-40.

WMO (2014). Scientific Assessment of Ozone Depletion: 2014. No. 55, 416 pp.

Wunderlin, P., Mohn, J., Joss, A., Emmenegger, L., & Siegrist, H. (2012). Mechanisms of N2O production in biological wastewater treatment under nitrifying and denitrifying conditions. Water research, 46(4), 1027-1037.

Yan, C., Zhang, L., Luo, X., & Zheng, Z. (2013). Effects of various LED light wavelengths and intensities on the performance of purifying synthetic domestic sewage by microalgae at different influent C/N ratios. Ecological Engineering, 51, 24-32.

Ye, F., & Li, Y. (2009). Enhancement of nitrogen removal in towery hybrid constructed wetland to treat domestic wastewater for small rural communities. Ecological Engineering, 35(7), 1043-1050.

Youshi Tazoe, Yuko T. Hanba, Tsuyoshi Furumoto, Ko Noguchi, Ichiro Terashima, Relationships Between Quantum Yield for CO2 Assimilation, Activity of Key Enzymes and CO2 Leakiness in Amaranthus cruentus, a C4 Dicot, Grown in High or Low Light, Plant and Cell Physiology, Volume 49, Issue 1, January 2008, Pages 19–29, https://doi.org/10.1093/pcp/pcm160

Yorio, N. C., Goins, G. D., Kagie, H. R., Wheeler, R. M., & Sager, J. C. (2001). Improving spinach, radish, and lettuce growth under red light-emitting diodes (LEDs) with blue light supplementation. HortScience, 36(2), 380-383.

Zhao, Y., Sun, S., Hu, C., Zhang, H., Xu, J., & Ping, L. (2015). Performance of three microalgal strains in biogas slurry purification and biogas upgrade in response to various mixed light-emitting diode light wavelengths. Bioresource technology, 187, 338-345.

Zhang, L., De Schryver, P., De Gusseme, B., De Muynck, W., Boon, N., & Verstraete, W. (2008). Chemical and biological technologies for hydrogen sulfide emission control in sewer systems: a review. Water research, 42(1-2), 1-12.