2.1. Water Quality
The water pH showed almost stable values, next to 8 in both systems, without significant differences between the inflow water and outflow from the control tank (Figure 1
A). A slight difference in pH was observed in the ozonized system between the inflow water and the outflow water after the first stage of adjustment. On average, the pH registered in the inflow water was of 8.09 in contrast to that of outflow water of 8.23. This result, although not significant, indicates the remarkable effect of ozone with respect to the organic substance. The ozonation of waters containing organic matter produces biodegradable by-products such as organic acids, aldehydes, and ketoacids [20
]. Moreover, since the ozone generator is fed by air rather than oxygen, some nitrous or nitric acids from the nitrogen oxides can also be produced. These chemical compounds can significantly affect the water pH.
The redox potential showed initial values of 310 mV in both systems, reaching a peak of about 400 mV in the second forcing day to then assuming lower values fluctuating around 350 mV (Figure 1
B). The recorded values are perfectly in line with the measurements in the open-loop system, indicating that the oxidant value of the environment treated with ozone ensured similar conditions as determined by a continuous well water flow. The high redox potential measured in the closed-loop system is typically determined by the extreme reactivity of ozone that can oxidize a wide spectrum of compounds and produce highly reactive short-lived free radicals that can further react [21
The electrical conductivity (EC), although with comparable absolute values between the two systems, presented an opposite trend (Figure 1
C). The values were higher in the ozonized system with a starting value of 550 µS cm−1
and increased by 7% in 8 days. In the control, instead, the initial value of the EC was 520 µS cm−1
and decreased by about 3% during the experiment. The EC was higher in the system treated with ozone because it was a closed system where there is no water replacement, and then the water evaporation slightly increased the salts concentration. In the control, the continuous water flows allowed the constant maintenance of the EC level in the tank, increasing the leaching of those that are present in the soil fraction near the root system.
The turbidity in the tank with ozonized water highlighted a peak on the first day of around 150 Nephelometric Turbidity Units (NTU) after RTT containers placement; then, it declined in the following two days by 95% and by 99.3% until the end of the experiment with values of about 1.00 NTU (Figure 1
D). The measurements performed in inflow and outflow water showed similar values, perfectly overlapping. In the control, the inflow water was characterized, as is obvious, by constant physical–chemical parameters and an average turbidity throughout the duration of the test of about 0.50 NTU. Regarding outflow water, a peak of 150 NTU was registered on the first day with a fast reduction on the second day reaching values of 3.5 NTU; in the following experimental days, the turbidity settled down at 0.45 NTU (the same value of inflow water). The highest turbidity values were due to plant container immersion in the forcing pool that caused a water contamination with soil. The faster reduction of turbidity in the control system was linked to the continuous input of clean water, resulting in a dilution of suspended particles. In any case, we can say that the water turbidity in the two systems did not present significant differences after the third day of the forcing process. The pre-oxidation applied in this system confirmed what is reported in literature [12
]. The main possible points of chemical oxidant introduction are pre-oxidation, intermediate oxidation and final disinfection. Usually, pre-oxidation leads to the elimination of mineral compounds, color, turbidity and suspended solids, bad tastes and odors; in addition, this step partly degrades natural organic matter and inactivates microorganisms; finally, this treatment generally enhances the coagulation–flocculation–decantation step. The turbidity reduction with ozone application is also reported by other authors [22
] for surface waters with a turbidity decrease of 40%. Drinking water or swimming pool water are characterized by a 0.2–0.4 NTU in post-ozonation [21
]; then, our data showed really good water quality from the turbidity point of view considering that there is a soil contamination.
In the ozonized system, the dissolved oxygen (DO) reached values of 8.25 mg L−1
during the first experimental day and 9 mg L−1
by the third day to the end of the test (Figure 1
E) both for inflow and outflow water. In the first two days of the experiment, the two systems did not achieve a balance; the DO readings were variable, probably due to the high presence of organic particles still in the solution. Subsequently, in both systems, the concentration of dissolved oxygen was comparable, demonstrating the good functioning of ozone treatment and water recirculation.
Regarding water temperature (Figure 1
F), the results obtained in the two systems were affected, as expected, by the environmental conditions in which the test was held. The maximum temperature recorded in the system with ozone treatment was of 18.2 °C on the first day and the minimum value of 13.8 °C was recorded in the final stages of the experiment. For the control system, water temperature reached the maximum of 17.5 °C on the first day and then was stabilized near 15 °C. Temperatures recorded in the two systems are suitable for high ozone action. In fact, the reaction ability of ozone is significantly affected by the temperature. Usually, ozone depletion rates increase with increasing temperature [23
]. First-order kinetic plots indicate that, in general, ozone depletion is first-order in O3
concentration for at least three to four reaction half-lives. Rate constants derived from these plots show that O3
decay rates increase more than one order of magnitude for a temperature range of 5–35 °C. For 5 °C and 15 °C, an initial, fast reaction phase precedes the principal first-order phase. For reactions at 25 °C and 35 °C, there is little or no distinct initial phase and the entire ozone depletion appears to be first-order in (O3
]. From the agronomical point of view, the water temperature was higher than optimal values [15
] for the forcing process in both systems; the higher sensitivity to thermal variations of the environment was measured for the closed loop system. The traditional method allows a constant water temperature to be maintained due to the continuous water exchange. However, it should be considered that under optimum conditions of the radicchio forcing process, these thermal differences tend to fade since the forcing process takes place mainly during winter time. Moreover, a slight increase of temperature compared to optimal values does not affect the product quality, but simply anticipates the marketable stage.
The bactericidal effects of ozone have been studied and documented on a wide variety of organisms, including Gram-positive and Gram-negative bacteria as well as spores and vegetative cells [24
] both on food and water [28
]. The latter showed that treating water with ozone (1 mg L−1
) resulted in a significant reduction (93%) in the number of live cells after 16 min treatment. In our experiment, the amount of viable microorganisms in the water at the end of the experiment was on average 1.4 × 103
in the control and 1.5 × 105
in the ozonized system. This result, apparently unfavorable for the closed-loop system, is a good outcome if compared with the open-cycle system. In the latter, the viable microbial load is lower only by 100 times compared to a system with the same water recirculation that is continuously in contact with plants, soil and subjected to the considerable effect of concentration due to evaporation. Additionally, for this aspect, the ozone treatment effectively contributes to the maintenance of good water quality during the forcing process.
2.2. Production and Quality Aspects of RTT
In relation to production aspects, the last part of the experiment was aimed at the evaluation and comparison of the RTT marketable yield and waste after the forcing-process practice. The average marketable production per plant was of 0.19 kg and did not differ between systems (Figure 2
A). Plants forced in the ozonized tank showed an average waste of 0.31 kg (Figure 2
B), whereas the control plants showed an average waste of 0.25 kg per plant. The difference is apparently higher in the ozone treatment, but not statistically different. The average height of the marketable head did not differ between forcing systems and was of 0.14 m (Figure 2
C). Similar to what is found for the plant height, the head basal diameter also did not substantially vary in relation to the used system (Figure 2
Concerning the qualitative parameters of the commercial product, pH, EC, soluble solids, titratable acidity and total phenols were not affected by the forcing system (Table 1
Only the total antioxidant activity (TAA) content was significantly higher (14.1%) in plants forced in the ozonized system. The ozone effect on the antioxidants content in vegetables has been addressed in several studies [29
]. However, O3
is often employed directly on the product in gaseous form, with modified atmosphere [31
], or as a disinfection treatment in fresh-to-eat products [32
]. Limited are the studies that evaluate the use of ozone in the water where the product is dipped [34
] and, to our knowledge, there is no information about plant growing inside ozonized water. Some recent studies used ozone for nutrient solution sanitization in hydroponics [14
], but no reference is made to the product quality. Many fresh-to-eat food products showed an antioxidant increase after ozone treatment in a modified atmosphere [36
], whereas in terms of ozonized water, the answer seems to be also linked to the species used. Some experiments [34
] reported that treating broccoli with ozonized water did not alter phenols and antioxidants content. In this experiment, instead, the water treatment increased the amount of antioxidants in the whole plant and the commercial product. This effect can be connected to the high reducing power that could act at root level by increasing the stress to the plant. However, no root system damages were observed in this trial because the ozone surplus in the system was dispersed by the ozone gas exhauster unit. However, any residual concentrations of 0.2–0.3 mg O3
can cause root injury when immersing plant roots in ozonized water [38
For RTT, the increase of TAA can be a good result because of the nutritional value of the product, but could also affect the bitter taste perceived by the consumer. As known, in fact, many of the antioxidant components present in radicchio confer a bitter taste to the product [39
]. For this reason, the determination of sesquiterpene lactones (SLs) was carried out in order to evaluate whether the increase of total antioxidants is due to this class of compounds. In Table 2
, the main SLs values that determine the bitter taste in chicory are reported.
It is possible to observe that both forcing systems did not affect these qualitative traits, despite the content of SLs being slightly higher in plants treated with ozone. Consequently, it is possible to exclude a negative effect of the increasing content of TAA in the commercial product. In order to determine the general quality assessment of the product from the antioxidants point of view, the phenolic acids content was also measured (Table 3
Also in this case, the main phenolic acids in RTT were chicoric acid and chlorogenic acid [40
]. The antioxidant and radical scavenger activity of phenolic compounds is well documented, in particular, caffeic acid derivatives can act as anti-inflammatory [43
], skin photo damage protectors [44
] and chicoric acid showed anti-HIV-1 activity [45
]. The results displayed that the forced chicory in the ozonized system has a significantly higher content of chicoric acid compared to control plants. This could be the reason for the higher TAA in radicchio forced in ozonized water. Regarding sugars content in the marketable product, no significant differences were observed between treatments (Table 3
Overall, considering both water features and RTT quality, we can say that the ozone application is able to maintain a good balance in the system without significantly altering the quality of the product. These results may have an additional advantage since this pilot system could be applied to other vegetable crops that require similar treatment or nutrient solution recirculation systems. In this regard, witloof chicory presents considerable problems during forcing [19
]. The hydroponic forcing is a very specific type of culture system in which a nutrient solution recirculates in a closed system providing excellent conditions for disease outbreak [45
]. Such a system favors the fast growing and uniform spread of the pathogen, enforced by a genetically uniform host and a controlled physical environment [46
]. The primary concern of witloof chicory growers is to prevent the pathogen from entering the forcing system. The application of the pilot system shown in this trial could also be a useful tool for other growing contexts.