Waste Nutrient Solution as an Alternative Fertilizer in Curled Mallow Cultivation

To determine the feasibility of reusing waste nutrient solution as an alternative fertilizer for vegetable production, we investigated the growth and shoot nutrient content of curled mallows (Malva verticillata L.) irrigated with tap water (pH 7.8, EC: 0.5 dS·m), nutrient solution (pH 5.7, EC: 2.7 dS·m) and waste nutrient solution (pH 5.0, EC: 2.2 dS·m in average) collected from plant factories. Three waste nutrient solutions were applied in sequential order to a waste nutrient solution treatment and mineral compositions of irrigation solutions were analyzed. We measured the total number of leaves, fresh and dry weight, chlorophyll content (SPAD value) and total phenolic content of curled mallow shoots and analyzed shoot and soil nutrient content using inductively coupled plasma-optical emission spectrometer (ICP-OES). Curled mallows were harvested twice during the cultivation. Curled mallows irrigated with waste nutrient solutions had a similar fresh weight (yield), total phenolic content and number of leaves compared to those grown with fresh nutrient solution, and had higher fresh and dry weight, chlorophyll content (SPAD value) and percentage dry weight compared to those grown with tap water upon first and second harvest. The dry weight of curled mallows grown in waste nutrient solution was lower than that of plants grown in nutrient solution on first harvest, but there was no significant difference between the waste nutrient solution and nutrient solution groups on second harvest. Curled mallows irrigated with nutrient solution and waste nutrient solution showed similar mineral content. These results suggest that waste nutrient solution in curled mallow cultivation could be reused and provide more efficient and sustainable nutritional solutions that improve the productive yields of crops in the agriculture sector.


Introduction
In recent years, hydroponic systems have become increasingly popular in the greenhouse industry (Savvas & Gruda, 2018).The hydroponic crop production increased up to approximately 30,000 ha in 2015 (Simpkins, Jungers, & Stimmel, 2015) and the hydroponic market is expected to have annual growth rate of 6.5% from 2018 to 2023 (Mordor Intelligence, 2018).As a consequence of the development of the hydroponics industry, the amount of waste nutrient solution discharged from hydroponic systems has also increased.
Discharged waste nutrient solution generally contains high concentrations used to increase crop yield.Nutrient solutions in closed hydroponic systems are periodically dumped during crop cultivation to prevent accumulation of salt ions, which can cause serious growth disorders (Sánchez-Guerrero, Lorenzo, Medrano, Baille, & Castilla, 2009;Savvas, Meletiou, Margariti, Tsirogiannis, & Kotsiras, 2005;Savvas et al., 2007) and plant diseases that can reduce crop yield and even eliminate all crops at once (Badgery-Parker, 2002;Grewal, Maheshwari, & Parks, 2011).Zekki, Gauthier, and Gosselin (1996) also noted that prolonged recycling of nutrient solution can cause yield reduction in closed hydroponic systems.In addition, the volume and concentration of applied nutrient solution are generally higher than those required for plant growth in order to satisfy variability in irrigation equipment and to maximize crop yield (Grasselly et al., 2005;Rouphael & Colla, 2009 Inc., on a , and ter to minimize mineral loss, and water was added until the amount of solution in the flask reached 100 mL.The solution was filtered through Whatman No. 2 filter paper and the filtrate was analyzed by NICEM (Seoul National University, Seoul, Korea) using ICP-OES to determine the mineral composition of the plant sample.
The total amount of phenolic compounds in each plant sample was determined using the Folin-Ciocalteu assay (Slinkard & Singleton, 1977).First, 0.2 g of the fine dry powder was mixed with 12 mL of extraction solution (acetone:methanol:water:acetic acid = 40:40:20:1) in a 50 mL conical centrifuge tube for 10 s using a homogenizer (PT-MR2100, Kinematica AG, Lucerne, Switzerland).The tube was kept at 60 o C for 1 h and more extraction solution was added until the amount of solution in the tube reached 20 mL, followed by filtration.Subsequently, 1 mL of the filtrate was mixed with 1 mL of 10% Folin-Ciocalteu phenol reagent and 1 mL of 7.5% Na 2 CO 3 solution, in sequence.After incubation for 2 h at room temperature, absorbance was determined using a UV visible spectrophotometer (UV-2550, Shimadzu Corp., Kyoto, Japan) at 726 nm.The same procedure was repeated with garlic acid solutions (0, 0.1, 0.2 and 0.5 mg•mL -1 ) and a standard curve was obtained.

Statistical Analysis
SAS version 9.2 software (SAS Institute INC., Cary, NC, USA) was used for analysis of variance (ANOVA).Differences among treatment means were separated by the Fisher's least significant difference (LSD) test at P < 0.05.

Quality of Irrigation Solution
The mineral composition of nutrient solution, waste nutrient solution and tap water used for curled mallow irrigation are shown in Table 1.Waste nutrient solution generally has low EC and pH levels and high Na, Cl and SO 4 ion concentrations compared to regular nutrient solution (Kumar & Cho, 2014).In this experiment, all waste nutrient solutions applied to WNS treatment group had lower EC and pH levels and NO 3 , NH 4 , PO 4 and K concentration compared to nutrient solution, while Mg, Na, Cl and SO 4 concentrations were higher in waste nutrient solution.This corresponds with the results of Zekki et al. (1996) and Zhang et al. (2010), who reported Mg, Na, Cl and SO 4 concentration increased in the waste nutrient solution as the crops grew.Note.z Not detected.
The chemical composition of nutrient solutions is determined not only based on EC and pH, but also the relative ion proportions (Steiner, 1961).In this experiment, the cation and anion composition of three waste nutrient solutions were similar to that of nutrient solution (Figure 3).The position of each point within an equilateral triangle was determined based on cation (K, Mg and Ca) and anion (NO 3 , SO 4 and PO 4 ) ratios, and the length of the perpendiculars from the point to each side of the triangle corresponded to the proportion of each ion (De Rijck & Schrevens, 1998).The average Euclidean distance between the nutrient solution and waste nutrient solution was 0.07 for cation and anion composition, while that between nutrient solution and tap water was 0.47 and 0.63, respectively, in trilinear coordinates.This suggests that ion proportions of waste nutrient solutions from closed hydroponic spinach cultivation systems would not have negative effects on curled mallow cultivation in that the relative ion ratios of waste nutrient solutions were similar to those of 'Enshi' nutrient solution which ion balance was designed for various vegetables in all growth stages (Park & Kim, 1998).Note.z Means within the same column followed by the same letter are not significantly different at the P = 0.05 level of probability based on Fisher's least significant difference (LSD) statistics.
The fresh weight of curled mallows irrigated with waste nutrient solution (44.8 and 17.6 g) did not differ significantly from nutrient solution (49.1 and 16.3 g), but was 22.9 and 28.7% higher compared to those irrigated with tap water (36.5 and 13.7 g) upon first and second harvest (Figures 4C and 4H).The total yield (fresh weight per unit area on first and second harvest) of curled mallows in the WNS group was 4.5% lower compared to that in the NS group, but was 24.5% higher compared to the TW group.This suggests that crops cultivated with waste nutrient solution are economically equivalent to those cultivated with nutrient solution, and economically superior to the use of tap water.
The dry weight of the WNS group (5.3 g) was 17.4% lower than that of NS (6.4 g) upon first harvest, but there was no significant difference between the two treatments upon second harvest, which implies a relative increase in the dry weight of WNS compared to NS during the period between harvest stages (Figures 4D and 4I).The dry weight indicates the amount of surplus carbohydrates from photosynthesis and respiration, and the results suggest that curled mallows irrigated with waste nutrient solution can accumulate similar amounts of carbohydrates to those irrigated with nutrient solution by second harvest.As the nutrient uptake of plants can change with plant growth stage, the nutrient requirements might decrease with increasing plant age and thus the nutrient content of WNS could meet the requirements for plant growth after the first harvest.The curled mallows in the WNS group (5.3 and 2.3 g) had 40.5 and 51.7% higher dry weight compared to that of the TW group (3.8 and 1.5 g) upon first and second harvest, respectively.Choi et al. (2011a) reported similar results for fresh and dry weights of Chinese cabbage irrigated with waste nutrient solution, which were similar or slightly lower than those irrigated with nutrient solution and similar or higher than those irrigated with groundwater 26 and 56 DAT.According to Kim et al. (2000), poinsettia irrigated with waste nutrient solution had higher fresh and dry weights of shoot compared to nutrient solution-irrigated groups.Park et al. (2005) also reported that red pepper (Capsicum annum L.) irrigated with hydroponic wastewater had a similar yield to those irrigated with nutrient solution and a more than 20% higher yield compared to tap water.
The percentage dry weight of curled mallows differed significantly among the three treatment groups (Figures 4E and 4J).The percentage dry weight in the WNS group (0.12) was 9.8% lower compared to the NS group (0.13) and higher by 13.2% compared to the TW group (0.1) at the first harvest.Upon second harvest, the percentage dry weight of curled mallows in the WNS group (0.13) was 13.0% lower and 15.8% higher compared to the NS (0.15) and TW groups (0.11), respectively.This suggests that the percentage dry weight is positively correlated with SPAD value (Figures 4B and 4G) which generally correlates with photosynthesis.Jacobson (1945) reported a relationship between the percentage dry weight and photosynthesis, with higher rates of photosynthesis correlating with greater percentage dry weight in green leaves when compared leaves with chlorosis.

Nutrient Content of Soil
The nutrient content of the soil for WNS plants did not vary from that of NS plants in N and P at the first harvest (Table 3).Considering that the N and P concentrations of nutrient solution were higher than those of waste nutrient solution, similar residual mineral levels in the soil of the NS and WNS groups implies that NS plants absorbed more N and P compared to WNS plants.This could be confirmed based on the total nutrient content of curled mallow shoots, which was calculated by multiplying the average plant nutrient concentration (Table 2) and dry weight (Figures 4D and 4I), with the results showing 30.8 and 19.1% higher N and P content in the NS group compared to the WNS group, respectively.The soil nutrient content in the WNS group was 0.13%, 1.09 g•kg -1 and 0.91 g•kg -1 higher in N, P and K concentrations, respectively, compared to the TW group, but was 0.85 g•kg -1 lower in K concentration compared to the NS group on first harvest.The Ca concentration in the WNS group was lower than that in the NS group and higher than that in the TW group, while Mg concentration in the WNS group was the highest among the three treatments.
Table 3. Mean values (±standard error) of nutrient content in soil before cultivation and after the first and second harvest.
Treatment N (%) P (g•kg -1 ) K (g•kg -1 ) Ca (g•kg -1 ) Mg (g The soil mineral concentrations in the WNS group were lower compared to NS and higher compared to TW for all minerals analyzed upon second harvest.The N, P and K concentrations of soil in the WNS group were 0.25%, 3.53 g•kg -1 and 2.43 g•kg -1 lower compared to NS, but 0.11%, 2.44 g•kg -1 and 1.78 g•kg -1 higher compared to TW, respectively.All soil mineral concentrations in the NS and WNS groups increased upon second harvest compared to first harvest except for Mg concentration in WNS.The observed increases in N, P and K between the two harvest stages were 51.9, 176.7 and 85.6% in NS, while those in WNS were only 1.8, 22.6 and 38.4%, respectively.The difference in soil mineral concentration between the two harvest stages indicates the amount of residual soil nutrient content after mineral absorption by plants in the second cultivation period, and the above results demonstrated that more nutrients were provided to the NS group than the plants needed, causing the accumulation of excess minerals in the soil (Ju, Kou, Christie, Dou, & Zhang, 2007).The same was not seen for the WNS group.This suggests that using waste nutrient solution for irrigation instead of nutrient solution could decrease water pollution related to leaching of excess soil nutrients.

Nutrient Content of Plant Samples
Nutrient content per gram of curled mallows (leaf and stem tissues) irrigated with waste nutrient solution did not differ significantly from that of nutrient solution in N, P, K and Ca concentration upon first and second harvest (Table 2).This could be because waste nutrient solution with low EC and pH and high Na and Cl concentrations was buffered with minerals in the commercial soil.The commercial soil already contained a certain amount of minerals before irrigation and when waste nutrient solution was applied, minerals in both the solution and soil were mixed and provided to curled mallows.In addition, soil nutrients over the optimum concentration level could have affected plant nutrient content.When nutrients are provided in excess of the optimum concentration level, further mineral absorption is limited and plants cultivated within the nutrient sufficiency range contain similar mineral levels in their tissues (Campbell, 2000).In this experiment, even with a considerable excess of residual nutrients in the soil of the NS group on second harvest, curled mallows could not further absorb minerals due to having reached the optimum concentration level.Fallovo, Rouphael, Rea, Battistelli, and Colla (2009) showed similar results, in that the concentration of macronutrients in lettuce was limited even when the nutrient solution concentration exceeded a certain level.
The waste nutrient solution provided during the second cultivation period had lower EC and pH values compared to that during the first period, but nutrient content of curled mallows in the WNS and NS groups did not differ significantly between harvests.The average nutrient content in WNS upon second harvest was higher than that of NS for all analyzed ions, even those that were lower in upon first harvest.This could be due to a decrease in the optimum soil nutrient concentration between harvest stages, or by decreased nutrient uptake rate as the plants age (Warncke & Barber, 1974).Edwards andBarber (1976a, 1976b) Figure 4. and perce harvest.N with tap ).

Table 1 .
The pH, EC and ion concentrations of nutrient solution, waste nutrient solution and tap water.

Table 2 .
Nutrient content of curled mallows irrigated with nutrient solution, waste nutrient solution and tap water at the first and second harvest •kg -1 ) also stated that the maximum ion influx of N and P decreased with plant age.This suggests that less minerals will be needed as curled mallows age, and that use of nutrient solution during the first cultivation period and waste nutrient solution during the second