The Pennsylvania State University The Graduate School
THE POTENTIAL, SAFETY, AND ENVIRONMENTAL IMPACTS OF GROWING
DUCKWEED IN AN ECOLOGICAL WASTEWATER TREATMENT SYSTEM AS A SOURCE
OF PROTEIN FOR LIVESTOCK OR FISH
A Dissertation in Environmental Engineering by Benjamin J. Roman
© 2020 Benjamin J. Roman
Submitted in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
December 2020
The dissertation of Benjamin J. Roman was reviewed and approved by the following:
Rachel A. Brennan Associate Professor of Environmental Engineering Dissertation Adviser Chair of Committee
John M. Regan Professor of Environmental Engineering
Joshua D. Lambert Professor of Food Science
Charles T. Anderson Associate Professor of Biology
Patrick Fox Department Head of Civil and Environmental Engineering
ii
ABSTRACT
Duckweed (Lemnaceae) are small, floating aquatic plants that have rapid growth rates and high protein contents when grown on nutrient-rich water, making them an ideal candidate to be used as a protein source. When combined with ecological wastewater treatment systems, duckweed can simultaneously remove unwanted nutrients from wastewater and produce a large quantity of protein-rich biomass that can be used for livestock or fish feed. The benefits of ecological wastewater treatment systems and the utilization of duckweed as a protein supplement are individually well documented, but no studies have investigated the potential of combining the two processes to recover duckweed grown in a wastewater treatment system as a protein source. In this dissertation, wastewater-grown duckweed was evaluated for its protein production potential, food safety, environmental impacts, and protein quality using a combination of laboratory-scale experiments and modeling.
Duckweed grown in a pilot-scale ecological wastewater treatment system (the Penn State Eco-
MachineTM) was first evaluated for its growth rate and protein content. It was determined that wastewater- grown duckweed can produce over 10 tonne/ha-yr of protein, which is 5-10 times higher than common land-grown crops (soybean, oats, etc.). In addition, after drying for 48 hours at 40oC or 4 hours at 60oC
(which is required for long-term storage), it was found that that no E. coli remained throughout the biomass. Thus, these studies suggest that wastewater-grown duckweed can be below regulatory limits for microbial food safety when dried at the appropriate temperature and duration.
In addition to evaluating the safety of dried plant biomass, counts of coliform bacteria and E.coli were measured on water samples collected throughout the Penn State Eco-MachineTM and after UV disinfection to determine if the treated effluent from the Eco-MachineTM is of sufficient quality to be effectively UV disinfected and used as irrigation water (≤126 CFU E.coli / 100mL). The results show that the Penn State Eco-MachineTM can reduce 3 logs of E.coli and 4 logs of coliform, and that remaining coliforms and E.coli are below detection after the UV disinfection system, suggesting that these systems are capable of producing effluent that is of sufficient quality to be used for irrigation water.
iii
A life cycle assessment was performed with the aim of quantifying the environmental impacts of the Penn State Eco-MachineTM treating municipal wastewater while concomitantly producing animal feed
(derived from duckweed) and irrigation water (derived by UV disinfection of the treated water). The results show that Eco-MachinesTM that do not require heating (i.e., in tropical/equatorial regions) consume about a third of the energy and produce half of the greenhouse gas emissions as conventional wastewater treatment systems (not including the GHG that are produced during the treatment process, such as NOx,
CH4, etc.). This would also have the added benefit of being in climatic region that is ideal for duckweed growth, further increasing the benefits of the system. In addition, increasing the growth area of duckweed by the use of vertical farming within the Eco-MachineTM improved the overall beneficial impact of the system. These results show that coupling feed/irrigation water production with ecological wastewater treatment systems creates a net benefit to human health and the environment.
Finally, a study measuring the body growth, food intake, and final organ and adipose tissue mass of male CF-1 mice fed a diet where 10% or 25% of the control protein (casein) was replaced with duckweed protein (DWP) was conducted to determine if duckweed protein is similar in quality to casein.
The average growth rates of the mice fed the three diets over a 30 day period were not significantly different: 0.21 g/day for the control diet; 0.24 g/day for the 10% DWP diet; and 0.25 g/day for the 25%
DWP protein diet. The daily food intake of both DWP diets was 6.5 – 8.0% higher than the control diet, but feeding efficiency (body weight gain/food intake) did not differ among diets. The relative weight of the liver, spleen, kidneys, heart, and epidydimal fat, and the colon length were not significantly different between treatment groups. The results from this study indicate that replacement of up to 25% dietary casein with DWP has no adverse effects on the growth rate and final organ and adipose tissue weights of laboratory mice.
iv
TABLE OF CONTENTS LIST OF FIGURES ...... viii LIST OF TABLES ...... x ACKNOWLEDGEMENTS ...... xii Chapter 1 - Introduction ...... 1 Humans and agriculture ...... 1 Global food crisis ...... 1 Combining wastewater treatment with food production ...... 3 Safety concerns ...... 3 Life cycle assessment ...... 4 Duckweed as a protein source ...... 5 References ...... 10 Chapter 2 – A beneficial by-product of ecological wastewater treatment: an evaluation of wastewater-grown duckweed as a protein supplement for sustainable agriculture ...... 13 Abstract ...... 13 Introduction ...... 14 Materials and Methods ...... 16 Experimental setup ...... 16 Duckweed harvesting, growth rate, and protein analysis ...... 17 Water sampling and analyses ...... 17 Metals analysis ...... 18 Statistical analysis ...... 18 Results and Discussion ...... 18 Wastewater and duckweed biomass characterization ...... 18 Duckweed protein trends ...... 23 Duckweed protein economic value ...... 26 Conclusion ...... 27 Acknowledgements ...... 27 References ...... 28 Chapter 3 – Pathogen removal in the Penn State Eco-MachineTM and identification of links between microbiomes and microbiological water safety ...... 30 Introduction ...... 30 Materials and methods ...... 32 The Penn State Eco-MachineTM ...... 32
v
Water sampling and analyses ...... 32 Sample process prior to pathogen enrichment ...... 33 Enrichment and isolation of presumptive Salmonella spp...... 34 Enrichment and isolation of presumptive Listeria spp...... 34 DNA extraction for PCR confirmation ...... 35 PCR confirmation of Salmonella spp...... 35 PCR confirmation of the general Listeria monocytogenes and Listeria spp...... 35 Enumeration of Escherichia coli and coliforms...... 36 Results and discussion ...... 36 Water quality ...... 36 Pathogen removal in the Penn State Eco-MachineTM ...... 40 Duckweed coliform and E.coli inactivation by drying ...... 41 Indicators for common pathogens ...... 42 Conclusions and future work ...... 43 Acknowledgements ...... 43 References ...... 44 Chapter 4 – Coupling ecological wastewater treatment with production of livestock feed and irrigation water provides net benefits to human health and the environment: a life cycle assessment ...... 47 Abstract ...... 47 Introduction ...... 48 Methods ...... 50 Description of the Penn State Eco-MachineTM and inventory analysis ...... 50 Duckweed characteristics ...... 53 Life cycle impact assessment (LCIA) ...... 54 Sensitivity analysis ...... 54 Results and discussion ...... 54 Midpoint characterization ...... 54 Damage categories ...... 58 Sensitivity analysis of heating/greenhouse requirements ...... 64 Sensitivity analysis on vertical farming options to increase duckweed growth area ...... 65 Conclusions ...... 68 Acknowledgements ...... 69 References ...... 70
vi
Chapter 5 – Duckweed protein supports the growth and organ development of mice: a feeding study comparison to conventional casein...... 74 Abstract ...... 74 Introduction ...... 75 Materials and methods ...... 78 Preparation of duckweed and diet formulation ...... 78 Mouse feeding experiment ...... 79 Statistical analysis ...... 79 Results and discussion ...... 79 Duckweed protein quality ...... 79 Body weight gain and relative organ weights ...... 81 Body weight gain and relative organ weights ...... 82 Food intake and feeding efficiency ...... 86 Conclusion ...... 86 Acknowledgements ...... 87 References ...... 88 Chapter 6 – Conclusions, significance, and future work ...... 93 References ...... 96 Appendix A – Duckweed grown on dairy cow manure - theoretical protein production and nitrogen and phosphorous removal in the Chesapeake Bay Watershed ...... 97 N and P removal from the Chesapeake Bay Watershed ...... 98 Protein requirement met, and N and P removal based on number of animals and growth area ...... 99 Techno-economic analysis of a theoretical 1000 dairy cow herd with 0-20 acres of plant growth area...... 100 Conclusions and future work ...... 101 Supplemental information ...... 104 Appendix B – Chapter 2 additional files ...... 107 Appendix C – Chapter 3 additional files ...... 116 Appendix D – Chapter 4 additional files ...... 117 Appendix E – Chapter 5 additional files ...... 119
vii
LIST OF FIGURES Figure 2.1. (A) Plan-view schematic of the Eco-MachineTM. Red square around the tanks indicates the four tanks from which water was diverted for this experiment (arrows indicate the direction of flow). Green outline on Aerobic 3 tank indicates where duckweed used to inoculate each growth tray was harvested from; (B) Plan-view photograph of the four, hydraulically separated duckweed growth trays. . 17
Figure 2.2. Crude protein content versus ammonium (A), total nitrogen (B), and dissolved oxygen (C) from each duckweed growth tray (n=8 per tray)...... 24
Figure 2.3. Photographs of typical duckweed roots in each growth tray (photo credit: Michael Shreve). 25
Figure 2.4. Comparison of growth rate, crude protein content, and protein yield of duckweed and common land-grown forage crops. (*Kennely et al., 1995; +Masuda & Goldsmith, 2009; Othis study). ... 27
+ - Figure 3.1. Water quality conditions (NH4 , NO3 , TN, and COD) throughout the Penn State Eco- MachineTM (n=18 collected from June 2019 – March 2020) ...... 38
Figure 3.2. Concentration of coliform and E.coli through the Penn State Eco-MachineTM (n=19 for Truck, Clarifier, and Pond; n=5 for UV)...... 40
Figure 3.3. E. coli counts by absence (0) or presence (1) of Salmonella, Listeria spp., and Listeria monocytogenes in three stages of the Penn State Eco-MachineTM (n=19 per sampling location). All samples taken after UV disinfection had no presence of Salmonella or Listeria (not shown)...... 42
Figure 4.1. Schematic of the Penn State Eco-MachineTM and scope of the LCA analysis. Solid arrows indicate the flow of wastewater through the treatment system. Dashed arrows indicate products. The dashed line is the system boundary...... 52
Figure 4.2. Midpoint category impacts of the Penn State Eco-MachineTM. Net impact value per million liter (ML) treated for each category is listed above (positive values represent detrimental impacts and negative values represent beneficial impacts)...... 55
Figure 4.3. Damage assessment of three phases of the Penn State Eco-MachineTM: a) construction; b) operation; c) products; and d) total (see Table 4.1 for inventory; positive values represent detrimental impacts and negative values represent beneficial impacts)...... 63
Figure 4.4. Impacts of the Eco-MachineTM with propane heating, natural gas heating, no heating, and no heating and no greenhouse per million liters (ML) wastewater treated at the four damage categories: a) human health; b) ecosystem quality; c) climate change; and d) resources. Net impact is shown adjacent to the bar in each plot (positive values represent detrimental impacts and negative values represent beneficial impacts).Sensitivity analysis on vertical farming options to increase duckweed growth area ...... 65
Figure 4.5. Vertical farming sensitivity analysis on the impact of the number of vertical trays for duckweed growth in each of the four damage categories: a) human health; b) ecosystem quality; c) climate change; and d) resources. The net impact for each damage category are shown adjacent to the bars (positive values represent detrimental impacts and negative values represent beneficial impacts)...... 67
viii
Figure 4.6. Vertical farming weighted impacts of damage categories based on the number of vertical trays used for duckweed growth. Positive values represent detrimental impacts and negative values represent beneficial impacts...... 68
Figure 5.1. (A) Body mass of mice fed three diets over four weeks: control (casein protein); 10% DWP (casein + 10% duckweed protein); and 25% DWP (casein + 25% duckweed protein). (B) Average growth rate of mice fed three different diets. (C) Organ percentage of body mass from the mice fed three diets after 4 weeks (n=10 per diet; 95% CI)...... Error! Bookmark not defined.
Figure 5.2. A) Food intake rates and B) feeding efficiency of the three diets (n=16 per diet for food intake; n=10 per diet for feeding efficiency; 95% CI)...... Error! Bookmark not defined.
Figure A.1. Theoretical % N and P removed from entering the Chesapeake Bay from manure based on duckweed or soybean growth area (assumes a 7-month growing season for duckweed and one harvest per year for soybean)...... 98
Figure A.2. Comparison of duckweed (DW) and soybean (SB) for fraction of protein requirements met for a 1000 dairy cow herd, with N- and P- uptake from manure fertilizer by both crops based on number of AU per duckweed/soybean growth area (assumes a 7-month growing season for duckweed and one harvest per year for soybean)...... 100
Figure A.3. Duckweed vs. soybean production on equivalent land areas: A) techno-economic analysis; B) fraction of protein requirements met for a 1000 dairy cow herd, with N- and P- uptake from manure fertilizer by the two crops. Analysis assumes a 7-month growing season in NE USA, and a crop value of $360/MT...... 101
Figure B.1. Duckweed protein content results from Cumberland Valley Analytical Services...... 113
ix
LIST OF TABLES
Table 1.1. Amino acid composition (g/100g protein) of various duckweed species from each of the five genera and of common plant-based meals used for livestock and fish...... 6
Table 1.2. In-depth characterization of the nutritional quality of Wolffia microscopia (adapted from Appenroth et al., 2017; FA = fatty acids)...... 6
Table 2.1. Average water quality characteristics from the influent, effluent, and within each duckweed growth tray, and duckweed biomass characteristics throughout the experiment (n=18; ± one standard deviation). Trays are named for the Eco-MachineTM tanks from which their flow is derived...... 20
Table 3.1. Additional water quality parameters for the Penn State Eco-MachineTM from June 2019-March 2020 (n=18 per location ± the standard deviation)...... 39
Table 4.1. Inventory used in the life cycle assessment of the Penn State Eco-MachineTM (positive values represent material consumed and negative values represent material produced)...... 53
Table 4.2. Damage assessment results of the three phases of the Penn State Eco-MachineTM per million liters (ML) treated. The Monte-Carlo analysis standard deviation is shown for each phase at each damage category (95% CI; positive values represent detrimental impacts and negative values represent beneficial impacts)...... 64
Table 4.3. Duckweed vertical growth tray inventory data used in the sensitivity analysis...... 66
Table 5.1. Composition of diets used in this study...... 78
Table 5.2. Proximate analysis and amino acid composition of the duckweed protein isolate used in this study...... 80
Table 5.3. Essential amino acid (EAA) composition of protein isolate from duckweed used in this study and other common plant-protein sources. Values are represented in g per 100 g raw material...... 80
Table 5.4. Mass of protein isolate required to meet the daily recommended essential amino acid (EAA) intake for an adult human by the World Health Organization. Values are represented in g per 70 kg body weight per day. Bolded and underlined values indicate the limiting essential amino acid...... 81
Table 5.5. Review of duckweed feeding studies published since 1990 (adapted from Sonta et al., 2019). Bolded Positive/Negative/Neutral in the result column indicates the effect on including duckweed in the diet compared to the conventional diet...... 84
Table A.1. Characteristics of dairy cow manure, duckweed and soybean used for this analysis...... 99
Table B.1. Harvested duckweed masses, densities, and growth rates over the course of the experiment...... 108
Table B.2. Duckweed biomass metal concentrations from ICP-MS performed by Penn State Laboratory for Isotopes and Metals in the Environment...... 114
x
Table C.1. Plate counts of total coliforms, fecal coliforms, and E.coli from duckweed grown in the Penn State Eco-MachineTM at different drying temperatures...... 116
Table D.1. SimaPro Characterization of the Penn State Eco-MachineTM operated for 30 years using IMPACT 2002+ V2.15...... 117
Table E.1. Body weight of mice over the course of the study...... 119
Table E.2. Food intake rate of each group over the course of the study (DWP = duckweed protein) ..... 120
Table E.3. Organ weights of each mouse...... 121
xi
ACKNOWLEDGEMENTS
The financial support of the Office of the Physical Plant, Department of Civil and Environmental
Engineering, and Institutes of Energy and Environment Seed Grant Program at Penn State are gratefully acknowledged.
Mostly, I would like to thank my advisor, Rachel Brennan, for her intellectual and moral support throughout this process. I can confidently say that I would not be here if you hadn’t inspired me to become involved with research at the Eco-MachineTM during the final semester of my Bachelor’s degree.
Thank you.
I would also like to thank my committee members for their patient advice and guidance throughout my graduate research, as well as the faculty, staff, and fellow graduate students of the Civil and Environmental Engineering Department for their support.
xii
Chapter 1 - Introduction Humans and agriculture
It is generally believed that modern humans evolved around 200,000 years ago, and that the population remained low (< 1 million) until the advent of farming around 10,000 BCE. The invention of agriculture was the first catalyst for rapid human population growth, leading to a global population of 170 million by the year 1 CE For the next ~1,800 years, population steadily grew to around 1 billion people. Over the last two centuries, the industrial revolution, improvements in public health, and intensive agriculture (the
‘green revolution’) enabled the human species to sky rocket its population to 7.8 billion people today, and it is projected the global population will reach 11.2 billion by the year 2100 (UN, 2019). Although necessary for supporting the global population, agriculture is arguably the most powerful force that humanity has unleashed onto the planet – roughly one-half of the Earth’s arable land has been transformed for agriculture, and over 75% of that area is used for grazing and growing feed grains for livestock (UN FAO, 2019). Agriculture has many negative effects on the environment, including producing greenhouse gas (GHG) emissions, releasing nutrients into natural water bodies and causing eutrophication, and depleting water and soil nutrients. In fact, food production is responsible for 25% of
GHG emissions globally, and is tied with electricity and heat production as the largest GHG emitter by any one sector (Poore & Nemecek, 2018).
Global food crisis
As population increases and developing regions become more affluent, the demand for all food products will inevitably rise. It is projected that by the year 2027, the demand for food will outweigh the supply, leading to a global shortage of 214 trillion Calories (kcal) per year, which equates to the same energy as
379 billion Big Macs per year – more than McDonald’s has sold in its entire 65-year existence (Gro
Intelligence, 2017). When distributed over the entire global population, this leads to a shortage of about
75 Calories per day per person (~3% of daily recommended Calorie intake). However, the predicted
Calorie shortage will not be spread evenly across the globe. Rapid population growth over the past decades in China, India, and many African countries has caused these nations to have large ‘Calorie gaps’
– the number of Calories these nations demand is much higher than what they produce, forcing them to import food from mainly North America, Europe, and South America. However, the agricultural output of these ‘Calorie rich’ nations is becoming stagnant and will not be able to meet the growing Calorie demand of the developing world. Strategies for combatting this looming food crisis include: changing consumption patterns (eating less meat); reducing food waste; and increasing agricultural yields.
The production of meat, especially beef, is extremely taxing on the environment: if all cattle on
Earth were their own nation, they would be the 3rd largest GHG emitter, only behind China and the United
States (Ranganathan et al., 2016). The obvious solution to reduce the impacts of meat production is to consume less meat. This is starting to occur in developed regions around the world. For example, a study done in the UK found that one-in-eight of their citizens are vegetarian or vegan, and over 60% of those have changed their diet within the last five years (Waitrose and Partners, 2019). Food waste is another issue leading to a global food crisis. In the United States, 30-40% of the food supply is wasted at the retail and consumer levels, corresponding to roughly 130 billion pounds (59 billion kg) and $161 billion worth of food wasted (US FDA, 2020). In fact, if just 15% of the food waste in the United States were recovered, it could feed 25 million people. In addition, food waste is the single largest category of material placed in landfills in the United States, comprising 22% of municipal solid waste (US EPA,
2020), and exacerbates environmental concerns related with landfills, such as GHG emissions, groundwater pollution, and dangers to human health (Lisk, 1991).
While these strategies for combating food shortage seem simple, they require individuals from
Calorie surplus regions to change their behavior on behalf of Calorie deficit regions, which is unlikely to occur and is therefore not practical for overcoming the food crisis. Increasing agricultural yields by commercializing small-scale farms in the developing world (essentially leapfrogging the green revolution reforms in the 1950-1960s), especially in India and Africa, is currently the most realistic solution for overcoming a global food shortage. In addition, developing new technologies/strategies to increase the
2 efficiency of agriculture and livestock production in the developed world will assist in providing food to a growing population.
Combining wastewater treatment with food production
One strategy for increasing food production efficiency is linking wastewater treatment with agriculture by utilizing photosynthetic plants to recover waste nutrients and convert those nutrients into high-protein biomass. This biomass can then be harvested and used for animal feed or crop fertilizer. However, most wastewater treatment systems are not designed to incorporate the production of beneficial by-products.
Conventional wastewater treatment systems use a series of aerated and anoxic tanks to manipulate bacteria to remove organics and nutrients from the water (Holmes et al., 2019). Nitrogen (N) is a major nutrient of concern in wastewater treatment, and entire systems are designed with the goal of removing nitrogen from the wastewater. Urea (CH4N2O) is the main source of nitrogen in wastewater, and it is
+ easily broken down into ammonium (NH4 ) in the early stages of wastewater treatment facilities. The
+ - NH4 is then oxidized to nitrate (NO3 ) in a metabolic process called nitrification that occurs under aerobic conditions. Although there are many different configurations, typically the wastewater is then put into
- anaerobic conditions where the NO3 is converted to nitrogen gas (N2) in a process called denitrification.
The N2 is allowed to escape into the atmosphere, essentially wasting a valuable source of nutrients.
Ecological wastewater treatment systems (e.g., constructed wetlands, Eco-MachinesTM) use the same principles as conventional wastewater treatment, but utilize eukaryotic, multicellular organisms (plants, snails, fish, etc.) in addition to microbes to imitate a natural ecosystem. These systems are capable of treating wastewater to the same extent as conventional systems, but with fewer energy and chemical inputs.
Safety concerns
A major concern surrounding the use of wastewater nutrients for food production is the risk of contamination by wastewater-borne pathogens and metals. Treated wastewater has been effectively used to irrigate crops with no significant health concerns regarding the presence of microbial contamination or
3 metals in the harvested plant tissue (Almuktar et al., 2015; Forslund et al., 2010; Kiziloglu et al., 2008; and Saffari & Saffari, 2013), but little work has been done to examine the risks of using plants grown hydroponically in a wastewater treatment system as a food source. However, pathogens associated with plant biomass have been shown to rapidly inactivate during drying ≥ 60oC (Lee & Kaletunc, 2002), which is already typically done in order to preserve the quality of the biomass. Thus, it is likely that wastewater- grown plants that are dried at ≥ 60oC would have a low risk for microbial contamination. Metals that are present in wastewater (commonly Al, Cd, Cr, Cu, Pb, Hg, Ni, Ag, and Zn) have been shown to concentrate up to 100 or 1000 times into plants, posing a potential health risk in utilizing these plants for food. However, the extent of metal uptake largely depends on the plant species and the growth medium
(Tangahu et al., 2011). Thus, it is necessary to examine specific plants on specific waste streams to accurately characterize potential health risks. After confirming the safety of using a wastewater-grown plant for food, the environmental benefits/impacts of these wastewater-to-food systems should be examined.
Life cycle assessment
Life cycle assessment (LCA) is a technique used to evaluate the potential environmental impacts in each step of a product or process by considering the material inputs, processing, disposal, and emissions/waste that a product or process uses during its life. LCA studies have been conducted on wastewater treatment systems (Dixon et al., 2003; Hospido et al., 2004; Li et al., 2013; and Tabesh et al.,
2019), ecological wastewater treatment systems (Hendrickson et al., 2015), and sustainable protein alternatives (Oonincx and de Boer, 2012; Halloran et al., 2016; Gnansounou and Raman, 2016; and van
Oirschot et al., 2017). The economic and environmental benefits of ecological wastewater treatment systems compared to conventional treatment systems are well documented – Eco-MachinesTM are cost competitive with conventional systems up to 2.3 million L/day (600,000 gal/day) when built in a greenhouse, and up to 3.8 million L/day (1 million gal/day) without a greenhouse, largely due to removing the need for heating (U.S. EPA, 2001). Additionally, an LCA conducted on a Living Machine
4
(similar to an Eco-Machine) for treating wastewater from an office building found that these ecological systems can reduce GHG emissions and energy consumption by 90% and 10%, respectively
(Hendrickson et al., 2015). To date, no study has been conducted to determine the additional economic and environmental benefits of coupling ecological wastewater treatment with sustainable protein production.
Duckweed as a protein source
One ideal plant for bridging the divide between ‘waste’ nutrients and valuable biomass is duckweed.
Duckweed is a floating aquatic plant that is the anatomically simplest and smallest flowering plant in the world. Duckweeds grow on every continent except Antarctica, and there are 37 species belonging to five genera: Landoltia, Lemna, Spirodela, Wolfia, and Wolffiella (Xu et al., 2014; Cui & Cheng, 2015). All of the species have flat, oval fronds from 1mm-1cm in diameter that float on the surface of the water, and some have root-like structures that assist in stabilizing the plant and obtaining nutrients (Leng, 1999).
They are capable of growing on water depths ranging from only a few millimeters to three meters and can tolerate a large range of water quality conditions. When grown under ideal conditions (namely water temperature, pH, nutrient concentrations, and light intensity), duckweeds can double their biomass in 16-
48 hours, and have a protein content up to 45% by dry matter (DM), making them a model candidate for sustainable protein production.
Duckweed has been studied in depth for its potential to sequester nutrients from waste streams, and its growth rate, protein content, and N and P uptake rates are well documented, establishing that duckweed is capable of producing a vast quantity of protein (Leng, 1999; Cheng et al., 2002; Xu & Shen,
2011; Ge et al., 2012; Xu et al., 2012; Sims et al., 2013; and Yu et al., 2014). The quality of duckweed protein has also been investigated - the amino acid composition of the five duckweed genera are similar, and are comparable to common plant-based meals used for livestock fodder (Table 1.1). Thus, based solely on amino acid composition, duckweed protein appears to be of sufficient quality to be used as a protein supplement for animal feed, and has been characterized by the World Health Organization as a
5
‘high quality’ protein source (WHO, 2007). In addition to protein, duckweed is a source of fatty acids, macro and micro elements, and phytosterols (Table 1.2).
6
Table 1.1. Amino acid composition (g/100g protein) of various duckweed species from each of the five genera and of common plant-based meals used for livestock and fish. Duckweed speciesa Common mealsb Amino acid Spriodela Landoltia Lemna Wolffiella Wolffia Soybean Peanut Rice Corn polyrhiza punctata minor hyalina microscopica Cysteine 0.8 1.1 0.9 1.0 1.2 1.9 1.6 1.3 1.7 Methionine 1.6 1.6 1.6 2.0 1.6 1.7 1.0 3.0 2.4 Asparagine 7.8 8.1 8.2 7.3 10.4 ND ND ND ND Threonine 4.3 4.1 4.0 4.2 4.7 3.9 1.6 3.8 3.0 Serine 4.1 4.0 4.1 4.3 4.7 4.2 ND ND ND Glutamine 9.6 9.5 9.8 10.5 10.9 18.4 17.7 ND ND Glycine 4.3 4.5 4.6 5.0 4.7 ND 5.0 ND ND Alanine 5.4 5.3 5.1 6.0 7.8 ND ND ND ND Valine 4.4 4.6 4.6 4.8 4.9 5.3 4.4 6.2 5.1 Isoleucine 3.3 3.5 3.7 3.9 3.7 6.0 4.6 5.2 4.9 Leucine 6.8 7.3 7.3 8.0 7.7 8.0 6.7 8.2 15.3 Tyrosine 3.1 3.1 3.7 3.8 3.3 4.0 4.4 5.7 2.3 Phenylalanine 4.0 4.5 4.4 5.1 4.2 5.3 5.1 5.0 5.6 Lysine 4.2 4.1 5.0 5.8 5.7 6.8 3.0 3.2 1.9 Histidine 1.6 1.6 1.5 1.7 1.7 2.9 2.1 1.7 2.1 Arginine 4.7 4.7 4.8 4.7 5.2 7.3 11.3 7.2 3.3 Proline 3.5 4.1 3.8 3.7 3.6 5.0 ND ND ND Tryptophan ND ND ND ND ND 1.4 1.0 1.3 0.5 aAppenroth et al., 2017, bLi et al., 2011, ND = not determined.
Table 1.2. In-depth characterization of the nutritional quality of Wolffia microscopia (adapted from Appenroth et al., 2017; FA = fatty acids). Fatty acid distribution (%) Macro elements Micro elements Heavy metals Phytosterols (g/kg DW) (mg/kg DW) (mg/kg DW) (50 mg/g lipid extract) Saturated FA 25.1 ± 0.1 Ca 6.0 Fe 240 Cd 0.40 Campesterol 18% Monounsaturated FA 3.8 ± 0.1 K 83 Mn 755 Pb 0.24 Stigamasterol 15% Polyunsaturated FA 71.1 ± 0.2 Mg 3.1 Cu 3.52 Hg 0.04 Sitosterol 53% Sum n3 44.2 ± 0.2 Na 0.3 Zn 30.8 As 0.05 ∆5-Avenasterol 11% Sum n6 26.8 ± 0.0 P 7.0 I 0.75 ∆5,24(25)-Stigmastadienol 1% n6/n3 ratio 0.61 ± 0.0 Se 0.06 Cycloartenol 2%
6
Many in vivo duckweed feeding studies have been conducted on various animals (dairy cattle, pigs, sheep, goats, poultry, and fish) to determine how duckweed protein compares to conventional feed sources for animal growth and survival (Haustein et al., 1994; Kabir et al., 2005; Ngamsaeng et al., 2004;
Khanum et al., 2005; Moss, 1999; Van et al., 1997; Damry et al., 2001; Reid, 2004; El-Shafai et al., 2004;
Tavares et al., 2008; and Effiong et al., 2009). However, the results from these feeding studies are inconsistent, making it difficult to establish duckweed as a reliable protein source. The confounding results found in these studies are largely due to the wide range of nutritional characteristics that duckweeds can display depending on the nutrient content of their growth medium, with protein contents ranging from 15-45% and fiber content ranging from 5-30% (Leng et al., 1995). In addition, many of these studies had poorly characterized animal diets and duckweed growth conditions, making it impossible to determine the reasons for the inconsistencies observed.
This dissertation addresses the key knowledge gaps preventing the establishment of integrated wastewater-food production systems utilizing duckweed by critically evaluating duckweed grown in a pilot-scale ecological wastewater treatment system (the Penn State Eco-MachineTM) by quantifying protein yields, food safety, economic and environmental impacts, and protein quality. The following four chapters are the foundation of this work, and are briefly described below:
Chapter 2: A beneficial by-product of ecological wastewater treatment: an evaluation of wastewater-grown duckweed as a protein supplement for sustainable agriculture.
This chapter establishes the quantity of protein that duckweed can produce when grown on wastewater from a pilot-scale ecological wastewater treatment system. In addition, this chapter quantifies metal uptake into wastewater-grown duckweed, and provides a basis for the ideal location to grow duckweed in these types of systems.
This manuscript has been published in Ecological Engineering:
Roman, B., and Brennan, R.A. 2019. A beneficial by-product of ecological wastewater
treatment: an evaluation of wastewater-grown duckweed as a protein supplement for 7
sustainable agriculture. Ecological Engineering: X, 1:100004.
https://doi.org/10.1016/j.ecoena.2019.100004
Chapter 3: Pathogen removal in the Penn State Eco-MachineTM and identification of links between microbiomes and microbiological water safety.
Chapter 3 focuses on pathogen removal throughout the Penn State Eco-MachineTM as well as pathogen removal on wastewater-grown duckweed biomass from drying. This research is a portion of a larger manuscript in progress that involves linking the microbiome of various stages of the Penn State Eco-
MachineTM to the presence/absence of common pathogens found in wastewater.
This manuscript is still in preparation with our collaborators:
Yan, R., Roman, B., Brennan, R.A., and Kovac, J. (2020) Pathogen removal in the Penn
State Eco-MachineTM and identification of links between microbiomes and
microbiological water safety. (In preparation).
Chapter 4: Coupling ecological wastewater treatment with production of livestock feed and irrigation water provides net benefits to human health and the environment: a life cycle assessment
Chapter 4 is a life cycle assessment of the Penn State Eco-MachineTM treating municipal wastewater while producing animal feed derived from duckweed and irrigation water derived from UV disinfected
Eco-MachineTM effluent. The results show that when these systems are operated in a region that does not require heating, the produce beneficial impacts to human health, ecosystem quality, and climate change.
This manuscript has been submitted:
Roman, B., and Brennan, R.A. 2020. Life cycle assessment of an Eco-MachineTM
indicates beneficial environmental impacts for upcycling wastewater nutrients into
protein-rich duckweed. Journal of Environmental Management. (In review).
8
Chapter 5: Duckweed protein supports the growth and organ development of mice: a feeding study comparison to conventional casein.
The fifth chapter is a feeding study comparing duckweed protein to casein to determine if duckweed protein supports the body growth and organ development of laboratory mice. The results show that when casein was replaced with duckweed protein up to 25%, there were no adverse effects on mice body growth and organ development.
This manuscript has been submitted:
Roman, B., Brennan, R.A., and Lambert, J.D. 2020. Duckweed protein supports the
growth and organ development of mice: a feeding study comparison to conventional
casein. Journal of Food Science. (In review).
Chapter 6: Conclusions, significance, and future work.
The final chapter concludes the dissertation and lays out potential future work, namely that studies should be conducted within agricultural operations to fully realize wastewater-grown duckweed as a protein supplement for livestock.
9
References
Almuktar, S.A.A.A.N.; and Scholz, M. (2015) “Microbial contamination of Capsicum annuum irrigated with recycled domestic wastewater treated by vertical-flow wetlands.” Ecological Engineering, 82:404- 414. Appenroth, K.J., Sree, K.S., Bohm, V., Hammann, S., Vetter, W., Leiterer, M., and Jahreis, G. 2017. Nutritional value of duckweeds (Lemnaceae) as human food. Food Chemistry, 217:266-273. https://doi.org/10.1016/j.foodchem.2016.08.116 Ashbey, E., and Wangermann, E. 1949. Senescence and rejuvenation in Lemna minor. Nature, 164:187. Cheng, J., Landesman, L., Bergmann, B. a, Classen, J.J., Howard, J.W., Yamamoto, Y.T., 2002. Nutrient Removal from swine lagoon liquid by Lemna minor 8627. Trans. ASAE 45, 1003–1010. Cheng, J.J., Stomp, A.M., 2009. Growing duckweed to recover nutrients from wastewaters and for production of fuel ethanol and animal feed. Clean Soil Air Water, 37:17-26. https://doi- org.ezaccess.libraries.psu.edu/10.1002/clen.200800210 Cui, W., Cheng, J.J., 2015. Growing duckweed for biofuel production: A review. Plant Biol. 17, 16–23. doi:10.1111/plb.12216 Dixon, A.; Simon, M.; and Burkitt, T. (2003) “Assessing the environmental impact of two options for small-scale wastewater treatment: comparing a reedbed and an aerated biological filter using a life cycle approach.” Ecological Engineering, 20:297-308. https://doi.org/10.1016/S0925-8574(03)00007-7 Ellis, E.C., Klein Goldewijk, K., Siebert, S., Lightman, D., & Ramankutty, N. 2010. Anthropogenic transformation of the biomes, 1700 to 2000. Global Ecology and Biogeography, 19(5):589-606. Forslund, A.; Ensink, J.H.J.; Battilani, A.; Kljujev, I.; Gola, S.; Raicevic, V.; Jovanovic, Z.; Stikic, R.; Sandei, L.; Fletcher, T.; and Dalsgaard, A. (2010) “Faecal contamination and hygiene aspect associated with the use of treated wastewater and canal water for irrigation of potatoes (Solanum tuberosum).” Agricultural Water Management, 98:440-450. Ge, X., Zhang, N., Phillips, G.C., Xu, J., 2012. Growing Lemna minor in agricultural wastewater and converting the duckweed biomass to ethanol. Bioresour. Technol. 124, 485–488. Gnansounou, E.; and Raman, J.K. (2016) “Life cycle assessment of algae biodiesel and its co-products.” Applied Energy, 161:300-308. https://doi.org/10.1016/j.apenergy.2015.10.043 Gro Intelligence, 2017. How can we avoid a food crisis that’s less than a decade away. Published donline at gro-intelligence.com. Retrieved from: https://gro-intelligence.com/insights/articles/214-trillion-calories Halloran, A.; Roos, N.; Eilenberg, J.; Cerutti, A.; and Bruun, S. (2016) “Life cycle assessment of edible insects for food protein: a review.” Agronomy for Sustainable Development, 36:57. https://doi.org/10.1007/s13593-016-0392-8 Holmes, D.E.; Dang, Y.; and Smith, J.A. (2019) “Chapter four – nitrogen cycling during wastewater treatment.” Advances in Applied Microbiology, 106:113-192. https://doi.org/10.1016/bs.aambs.2018.10.003 Hendrickson, T.P.; Nguyen, M.T.; Sukardi, M.; Miot, A.; Horvath, A.; and Nelson, K.L. (2015) “Life- cycle energy use and greenhouse gas emissions of a building-scale wastewater treatment and nonpotable
10 reuse system.” Environmental Science and Technology, 49”10303-10311. https://doi.org/10.1021/acs.est.5b01677 Hospido, A.; Moreira, M.T.; Fernandez-Couto, M.; and Feijoo, G. (2004) “Environmental performance of a municipal wastewater treatment plant.” International Journal of Life Cycle Assessment, 9(4):261-271. https://doi.org/10.1007/BF02978602 Kiziloglu, F.M.; Turan, M.; Sahin, U.; Kuslu, Y.; and Dursun, A. (2008) “Effects of untreated and treated wastewater irrigation on come chemical properties of cauliflower (Brassica olerecea L. var. botrytis) and red cabbage (Brassica olerecea L. var. rubra) grown on calcareous soil in Turkey.” Agricultural Water Management, 95:716-724. Lee, J.; and Kaletunç, G. (2002) “Evaluation of the heat inactivation of Escherichia coli and Lactobacillus plantarum by differential scanning calorimetry.” Applied and Environmental Microbiology, 68:5379- 5386. Leng, R.A., 1999. Duckweed: a tiny aquatic plant with enormous potential for agriculture and environment. Food & Agriculture Organization, Animal Production and Health Division. Rome, Italy. Leng, R.A., Stombolie, J.H., and Bell, R. (1995). Duckweed – a potential high-protein feed resource for domestic animals and fish. Livestock Research for Rural Development, 7(1). Li, X.; Rezaei, R; Li, P.; and Wu, G. (2011) “Composition of amino acids in feed ingredients for animal diets.” Amino Acids, 40:1159-1168. Li, Y.; Luo, X.; Huang, X.; Wang, D.; and Zhang, W. (2013) “Life cycle assessment of a municipal wastewater treatment plant: a case study in Suzhou, China.” Journal of Cleaner Production, 57:221-227. https://doi.org/10.1016/j.jclepro.2013.05.035 Lisk, D.J. (1991) “Environmental effects of landfills.” Science of the Total Environment, 100:415-468. Oonincx, D.G.A.B.; and de Boer, I.M.J. (2012) “Environmental impact of the production of mealworms as a protein srource for humans – a life cycle assessment.” PLoS ONE (7)12: e51145. https://doi.org/10.1371/journal.pone.0051145 Ranganathan, J., Vennard, D., Waite, R., Lipinski, B., Searchinger, T., and GlOBAGRI-WRR model authors. 2016. Shifting diets for a sustainable food future. World Resources Institute. Installment 11 of “Creating a Sustainable Food Future. Roman, B., and Brennan, R.A. 2019. A beneficial by-product of ecological wastewater treatment: an evaluation of wastewater-grown duckweed as a protein supplement for sustainable agriculture. Ecological Engineering: X, 1:100004. https://doi.org/10.1016/j.ecoena.2019.100004 Roser, M. 2013. Future population growth. Published online at OurWorldInData.org. Retrieved from: https://ourworldindata.org/future-population-growth Saffari, V.R.; and Saffari, M. (2013) “Effect of treated municipal wastewater on bean growth, soil chemical properties, and chemical fractions of zinc and copper.” Arabian Journal of Geosciences, 6:4475- 4485. Sims, A., Gajaraj, S., Hu, Z., 2013. Nutrient removal and greenhouse gas emissions in duckweed treatment ponds. Water Res. 47, 1390–1398. doi:10.1016/j.watres.2012.12.009
11
Tabesh, M.; Masooleh, M.F.; Roghani, B.; and Motevallian, S.S. (2019) “Life-cycle assessment (LCA) of wastewater treatment plants: a case study of Tehran, Iran.” International Journal of Civil Engineering, 17:1155-1169. https://doi.org/10.1007/s40999-018-0375-z Tangahu, B.V., Abdullah, S.R.S.; Basri, H., Idris, M., Anuar, N., and Mukhlisin, M. 2011. A review on heavy metals (As, Pb, and Hg) uptake by plants through phytoremediation. International Journal of Chemical Engineering, 939161. United States Environmental Protection Agency. 2001. The Living Machine® wastewater treatment technology. An evaluation of performance and system cost. Office of Water (4204). EPA-832-R-01-004. United States Environmental Protection Agency. 2020. Sustainable management of food. Published online at epz.gov. Retrieved from: https://www.epa.gov/sustainable-management-food United States Food and Drug Administration. 2020. Food loss and waste. Published online at fda.gov. Retrieved from https://www.fda.gov/food/consumers/food-loss-and-waste van Oirschot, R.; Thomas, J.B.E.; Grondahl, F.; Fortuin, K.P.J.; Brandenburg, W.; and Potting, J. (2017) “Explorative environmental life cycle assessment for system design of seaweed cultivation and drying.” Algal Research, 27:43-54. https://doi.org/10.1016/j.algal.2017.07.025 Waitrose and Partners, 2019. Food and Drink Report 2018-19. Published online at globalagriculture.org. WHO (2007) “Protein and amino acid requirements in human nutrition.” Geneva: WHO. Report of a joint FAO/WHO/UNU Expert Consultation. World Health Organisation Technical Reports Series no. 935. Xu, J., Cheng, J.J., Stomp, A.M., 2012. Growing Spirodela polyrrhiza in Swine Wastewater for the Production of Animal Feed and Fuel Ethanol: A Pilot Study. Clean - Soil, Air, Water 40, 760–765. doi:10.1002/clen.201100108 Xu, J., Shen, G., 2011. Growing duckweed in swine wastewater for nutrient recovery and biomass production. Bioresour. Technol. 102, 848–853. doi:10.1016/j.biortech.2010.09.003 Xu, Y., Ma, S., Huang, M., Peng, M., Bog, M., Sree, K.S., Appenroth, K.J., Zhang, J., 2014. Species distribution, genetic diversity and barcoding in the duckweed family (Lemnaceae). Hydrobiologia 743, 75–87. doi:10.1007/s10750-014-2014-2 Yu, C., Sun, C., Yu, L., Zhu, M., Xu, H., Zhao, J., Ma, Y., Zhou, G., 2014. Comparative analysis of duckweed cultivation with sewage water and SH media for production of fuel ethanol.
12
Chapter 2 – A beneficial by-product of ecological wastewater treatment: an evaluation of wastewater-grown duckweed as a protein supplement for sustainable agriculture
Roman, B., and Brennan, R.A. 2019. A beneficial by-product of ecological wastewater treatment: an evaluation of wastewater-grown duckweed as a protein supplement for sustainable agriculture. Ecological
Engineering: X, 1:100004. https://doi.org/10.1016/j.ecoena.2019.100004
Abstract
Ecological wastewater treatment systems that incorporate aquatic plants (like duckweed) have the potential to recover nutrients and produce high-quality protein, simultaneously alleviating two global issues: hunger and lack of sanitation. Although protein production by duckweed in simple lagoon systems and laboratory trials has been reported, its physiology throughout more complex wastewater treatment systems containing a range of different environmental conditions has not been examined. In this study, a duckweed co-culture (Lemna japonica/minor and Wolffia columbiana) was grown on wastewater from four different stages of a pilot-scale ecological treatment system. Contrary to the literature, the protein content of duckweed did not consistently increase with increasing aqueous nitrogen concentrations, but rather appeared to also be dependent on chemical and microbial interactions. This study indicates that with proper management, duckweed grown in ecological wastewater systems can sustainably produce protein at rates exceeding those of common land-grown forage crops (10.1 tonne ha-1 yr-1).
13
Introduction
In the developing world, up to 90% of domestic wastewater is discharged without any treatment, releasing valuable nutrients into aquatic systems, leading to poor water quality, eutrophication, and dead zones
(FAO UN, 2015). Often in similar locations, roughly 780 million people suffer from protein-energy undernourishment, due to scarcity of high-quality food (Swaminathan et al., 2012). As human population grows and developing countries become more affluent, the demand for animal-derived proteins is escalating: the Food and Agriculture Organization (FAO) estimates that dairy and meat consumption will increase by 82% and 102%, respectively, between 2000 and 2050 (Boland et al., 2013). In addition, protein-rich fodder to support livestock growth is often a limiting factor for meat production (Van Huis et al., 2013), and many regions with minimal arable land and water scarcity are being forced to import fodder, further decreasing the sustainability of local foods (USDA, 2013). There are also significant environmental concerns associated with the production of animal proteins, including land-use, greenhouse gas emissions, and degraded water quality, which all require careful consideration and proper management (McAllister et al., 2011).
Modern wastewater treatment technologies and intensive farming methods may at first appear to provide a solution to our growing sanitation and protein needs, but these are typically restricted to developed areas, require large amounts of energy, and often release excess nutrients and untreated contaminants into aquatic systems. Ecological wastewater treatment systems (ex., constructed wetlands;
Eco-MachinesTM) utilize diverse life forms, typically housed in a series of ponds or tanks, to clean wastewater to the same, or better, effluent quality as conventional wastewater treatment systems (ex., activated sludge plants). Compared to conventional technologies, ecological wastewater treatment systems have a smaller energy and chemical footprint, making them well suited for use in small or developing communities (US EPA, 2002). Moreover, certain aquatic plants grown in these systems (ex., duckweed) can sequester nutrients from wastewater, producing high-protein biomass that can be harvested and reused in sustainable agriculture.
14
The aquatic plants of the subfamily Lemnoideae (common name: duckweeds) require only sunlight to treat contaminated water, and simultaneously produce a concentrated source of protein and nutrients that can easily be harvested and reused for food production. Lemnoideae have been shown to grow rapidly on the surface of municipal, dairy, swine, industrial, and aquaculture wastewaters, reducing chemical oxygen demand (COD) and removing substantial amounts of nitrogen (N) and phosphorous (P) (Frederic et al., 2006; Adhikari et al., 2015; Cheng et al, 2002; Chaiprapat et al., 2003). Nitrogen uptake by duckweed is believed to be the most
+ critical factor affecting its protein content, with a preference of ammonium (NH4 ) over nitrate
- (NO3 ) for synthesizing amino acids (Landesman et al., 2005).
The general composition of different duckweed species are quite similar to each other, consisting of macro-nutrients (ex., Ca, Na, K, Fe, Mg) and low fiber and lignin contents (Culley
& Epps, 1973). However, the protein content of duckweed species can range from 15 to 45% by weight, depending on the quality of the water in which they are grown (Chantiratikul et al., 2010).
When grown in nutrient-rich environments like wastewater, the resulting high protein content of
Lemnoideae biomass can yield revenue as fodder for livestock and fish (Ansal et al., 2010;
Mohedano et al., 2012; Fang, 2013; Zhao et al., 2014). Predicting the protein content of duckweed grown in different environments is necessary for effective implementation in various locations with different water quality characteristics. Previous lab and full-scale lagoon studies have examined the effect of nutrient concentrations on duckweed growth and protein content
(Chaiprapat et al., 2003; Cheng & Stomp, 2009; Mohedano et al., 2012); however, an analysis of how duckweed is affected when grown on various stages of a more complex wastewater treatment system has yet to be investigated. The purpose of this study was to examine how N concentrations and N speciation affect the growth and protein content of duckweed grown in four different stages of an ecological wastewater treatment system.
15
Materials and Methods
Experimental setup
Experiments were conducted in a pilot-scale ecological wastewater treatment system (Eco-MachineTM) located in a 50 m2 greenhouse at The Pennsylvania State University (University Park, PA). This system operates year-round on primary influent municipal wastewater (following rag and grit removal), which is delivered several times per week from the local treatment plant to an underground outdoor holding tank
(11.35 m3, 4 d hydraulic residence time (HRT)). The wastewater is pumped from the holding tank (with a typical dissolved oxygen (DO) < 0.5 mg/L) into the greenhouse intermittently to achieve a flow rate of
2.65 m3/day, and then flows by gravity through components of the system in the following order (Figure
2.1A): closed Anaerobic tank (1.7 m3; 15 hr HRT); closed Anoxic tank (1.7 m3; 15 hr HRT); three open
Aerobic tanks (each 3.79 m3; 34 hr HRT); a clarifier (1.4 m3; 13 hr HRT); a horizontal subsurface flow wetland (44.6 m3; 17 day HRT); and a final display pond (1.8 m3; 16 hr HRT). The clarifier, horizontal subsurface wetland, and display pond were not part of this study. The Anaerobic and Anoxic tanks are identical and closed to the atmosphere. Wastewater from the Aerobic 3 tank, which is high in nitrate, is recycled to the Anoxic tank at 50% of the influent flow rate to achieve denitrification. Aerobic 1 and 2 are aerated periodically and have floating rafts vegetated with mature purple taro plants (Colocasia esculenta). Aerobic 3 is not aerated, and the water surface is covered with duckweed previously identified as a co-culture of Lemna japonica/minor and Wolffia columbiana (Calicioglu and Brennan, 2018).
Peristaltic pumps (Masterflex L/S 07528-10, Gelsenkirchen, Germany) were used to pump wastewater out of the sequential Anaerobic, Anoxic, Aerobic 1, and Aerobic 2 tanks at 27 mL/min (24 hr
HRT) and into the head of four separate shallow growth trays (each 115 x 33 x 10 cm; Figure 2.1B). The same pumps were used to pump wastewater out of the opposite end of each tray – the trays were hydraulically separated and not connected together. The trays were named for the tanks from which their water was supplied. The experiment was operated from January to May 2017, with daylight hours over this period ranging from 10 to 14 hours.
16
A Recycle Influent from Effluent Holding Anaerobic Anoxic Aerobic 1 Aerobic 2 Aerobic 3 Clarifier to tank Wetland
B
Figure 2.1. (A) Plan-view schematic of the Eco-MachineTM. Red square around the tanks indicates the four tanks from which water was diverted for this experiment (arrows indicate the direction of flow). Green outline on Aerobic 3 tank indicates where duckweed used to inoculate each growth tray was harvested; (B) Plan-view photograph of the four, hydraulically separated duckweed growth trays.
Duckweed harvesting, growth rate, and protein analysis
To start the experiment, each tray was filled with water from its respective tank, and then one square foot of duckweed harvested from the Aerobic 3 tank was added. A two-week acclimation period (with continuous flow) was allowed before sampling was initiated. During the experiment, half of the duckweed
(based on surface area) was harvested from each tray once per week using a net, rinsed with tap water, and dried at 45oC in an Econotherm Lab Gravity Convection Oven (Precision Scientific, Winchester, IL) until a constant weight was achieved (2 – 3 days). The dried duckweed was stored at room temperature in an air tight desiccator for a maximum of six months, until it was ground into a powder and analyzed for crude protein content (Cumberland Valley Analytical Services, Hagerstown, MD).
Water sampling and analyses
Temperature, conductivity, dissolved oxygen (DO), pH, and oxidation-reduction potential (ORP) within each tray were determined weekly on site using a YSI 556 multi-probe system (YSI Inc., Yellow Springs,
OH). At the same sampling events, water samples were taken from the influent and effluent tubing of each duckweed tray, collected into 50 mL centrifuge tubes, placed on ice in a cooler, and transported to
+ the lab. NH4 was measured using an Orion Star Series portable meter and electrode (Thermo Scientific,
17
Waltham, MA). Chemical oxygen demand (COD) was measured according to Standard Methods (5220
D; Clesceri et al., 1998). Total nitrogen (TN) was determined using a Shimadzu TOC-VCSH/CSN
- 3- analyzer (Shimadzu, Columbia, MD). After filtering (0.45 μm), NO3 , phosphate (PO4 ), and sulfate
2- (SO4 ) were quantified using a Dionex ICS-1100 ion chromatograph equipped with an AS-18 column and
30 mM potassium hydroxide eluent (Dionex, Sunnyvale, CA). All analyses were performed within two hours of collection.
Metals analysis
Duckweed dried at 45oC for 48 hours was ground using a mortar and pestle and digested in 70%
OmniTrace nitric acid in a MARS 6 microwave digestion system for 25 min (CEM Corporation,
Matthews, NC). Digested samples were diluted to 3% nitric acid and analyzed using inductively coupled plasma mass spectrometry (ICP-MS) at the Penn State Laboratory for Isotopes and Metals in the
Environment (LIME). Metals targeted in the analysis were: Aluminum (Al); Arsenic (As); Cadmium
(Cd); Chromium (Cr); Copper (Cu); Iron (Fe); Lead (Pb); Nickel (Ni); Silver (Ag); and Zinc (Zn).
Statistical analysis
SAS 9.4 was used to conduct a repeated measure analysis of covariance (ANCOVA) with an autoregressive lag 1 covariance structure to determine if a significant difference existed between the duckweed protein content, growth rate, and nutrient removal observed in each tray. Linear regressions were also conducted for duckweed protein content versus ammonia, total nitrogen, and dissolved oxygen in each tray to determine if a significant trend was observed.
Results and Discussion
Wastewater and duckweed biomass characterization
In general, the wastewater characteristics entering the Anaerobic tray (Table 2.1) were similar to that of low-strength domestic wastewater, with the exception of COD. The lower than average COD values entering the Anaerobic tray can be attributed to removal in the holding tank of the Eco-MachineTM: an
18 average of 36.7% COD removal was observed from the wastewater delivery truck to the holding tank, which is comparable to COD removal in septic tanks (US EPA, 2002). Temperature, conductivity, and pH stayed relatively constant through each tray, while DO and ORP were higher in the Aerobic trays due to aeration of their respective tanks, consistent with trends observed in conventional wastewater treatment.
19
Table 2.1. Average water quality characteristics from the influent, effluent, and within each duckweed growth tray, and duckweed biomass characteristics throughout the experiment (n=18; ± one standard deviation). Trays are named for the Eco-MachineTM tanks from which their flow is derived. Duckweed Growth Trays Parameter Anaerobic Anoxic Aerobic 1 Aerobic 2 Influent Effluent Influent Effluent Influent Effluent Influent Effluent COD (mg L-1) 114 ± 21 72.7 ± 20 70 ± 18 61.7 ± 22 39 ± 9.4 33.7 ± 12 38 ± 14 35.4 ± 15 TN (mg L-1) 32.9 ± 8.1 25.3 ± 7.9 25.7 ± 7.8 20.4 ± 7.2 25.6 ± 7.2 20.6 ± 7.5 22.9 ± 7.1 17.7 ± 7.8 + -1 NH4 -N (mg L ) 24.2 ± 9.1 17.6 ± 7.0 18.3 ± 7.0 12.9 ± 6.7 17.8 ± 6.8 11.0 ± 6.2 9.1 ± 5.4 5.1 ± 4.3 - -1 NO3 -N (mg L ) 0.0 ± 0.0 1.0 ± 1.5 0.0 ± 0.1 2.3 ± 2.0 1.6 ± 1.3 4.7 ± 4.4 8.8 ± 3.9 8.8 ± 4.4 3- -1 PO4 -P (mg L ) 3.2 ± 1.2 3.0 ± 1.2 2.9 ± 1.1 3.1 ± 0.9 2.9 ± 1.0 3.3 ± 0.9 3.3 ± 1.0 3.2 ± 0.7 2- -1 SO4 (mg L ) 18.3 ± 6.9 25.5 ± 5.7 31.7 ± 10.5 29.6 ± 4.8 31.7 ± 4.3 30.3 ± 3.5 30.6 ± 3.7 30.2 ± 3.6
Temp. (oC) 19.6 ± 3.4 19.5 ± 3.1 20.0 ± 3.3 20.0 ± 2.6 Cond. (mS/cm) 2.3 ± 0.9 2.3 ± 0.7 2.2 ± 0.5 2.1 ± 0.5 pH 7.3 ± 0.2 7.4 ± 0.2 7.4 ± 0.2 7.5 ± 0.2 DO (mg L-1) 0.5 ± 0.1 1.0 ± 0.6 1.7 ± 0.9 2.9 ± 1.4 ORP (mV) -197 ± 79 -15 ± 113 110 ± 70 151 ± 37
Duckweed growth rate 8.0 ± 4.0 7.0 ± 3.6 8.2 ± 3.9 6.9 ± 3.5 (g m-2 d-1) Duckweed protein content 37.0 ± 1.9 37.4 ± 2.3 37.0 ± 1.6 36.0 ± 2.2 (% DM) Protein yield 2.9 ± 1.4 2.6 ± 1.2 3.0 ± 1.5 2.5 ± 1.3 (g m-2 d-1)
Fe (mg kg-1) 408 ± 100 397 ± 82 298 ± 66 199 ± 30 Al (mg kg-1) 84.4 ± 24 73.3 ± 32 52.5 ± 33 65.7 ± 64 Zn (mg kg-1) 44.1 ± 8.8 42.2 ± 8.2 49.8 ± 9.8 62.8 ± 23 Cu (mg kg-1) 19.5 ± 8.7 15.5 ± 5.3 12.2 ± 3.4 14.3 ± 3.5 Pb (mg kg-1) 2.7 ± 1.5 2.6 ± 1.5 2.4 ± 1.3 3.1 ± 1.6 Ni (mg kg-1) 2.5 ± 1.5 1.8 ± 0.4 0.99 ± 0.2 0.89 ± 0.3 Cr (mg kg-1) 0.37 ± 0.1 0.36 ± 0.1 0.25 ± 0.1 0.27 ± 0.2 As (mg kg-1) 0.23 ± 0.06 0.25 ± 0.06 0.18 ± 0.05 0.15 ± 0.07 Cd (mg kg-1) 0.03 ± 0.03 0.02 ± 0.01 0.02 ± 0.01 0.03 ± 0.01 Ag within the duckweed biomass was below detection limit for all samples.
20
Nitrogen removal in the duckweed trays was similar to that measured in other studies (Zhao
+ et al., 2014; Mohedano et al., 2012). As expected, TN and NH4 concentrations decreased from the
- Anaerobic to the Aerobic trays, but interestingly, NO3 production did not correspond directly with
+ NH4 removal (Table 1). Previous studies have shown simultaneous nitrification/denitrification occurring in duckweed growth ponds, and have speculated the occurrence of denitrification in the benthic layer (Zimmo et al., 2003). Although the duckweed trays used in this study did not have a sediment layer, wastewater solids did accumulate on the bottom of each tray over the course of the
- experiment, coincident with decreasing NO3 concentrations in the effluent. These observations suggest that as solids accumulated, an anoxic zone was formed at the bottom of the trays that provided a niche environment for denitrifying bacteria.
TN removal was similar for all trays (20-25%). In ecological wastewater treatment systems, a combination of N transformation mechanisms are responsible for N removal, including:
+ + mineralization of organic nitrogen (ON) to NH4 ; sedimentation of ON; regeneration of NH4 from
+ - sediment; decay of plant biomass to ON; volatilization of NH3; nitrification of NH4 to NO3 ;
- + - denitrification of NO3 to nitrogen gas (N2); plant uptake of NH4 and NO3 ; and microbial uptake of
+ - NH4 and NO3 (Kadlec & Wallace, 2008). It is likely that a combination of these processes was contributing to N removal in the duckweed trays, making it difficult to attribute N removal to any one mechanism; however, the % TN removal by duckweed uptake in each tray was estimated to be:
13.8% for Anaerobic; 16.6% for Anoxic; 16.9% for Aerobic 1; and 13.2% for Aerobic 2, based on the calculations described in Eq. 1 – 3: