For bench top scale experiments, a domestic microwave oven (Panasonic NN-SA661S), 1200W, 2450 MHz, 1.2 ft ³ was modified for continuous circulation of blackwater by drilling two holes into the back of the unit and inserting flared teflon feed through fittings that were then sealed with glue and microwave shielding adhesive tape to make inlet and outlet ports. Twenty-five feet of coiled Teflon tubing (ID 9.5 mm) was connected to the internal inlet and outlet ports with a teflon bracing spine to distribute the tubing evenly in the microwave. Two foot sections of prothane ii tubing (ID 7.95 mm) were used to connect the external inlet and outlet ports to an insulated reservoir ( Figure 1A). A total tubing volume of 0.6 L was used. Type K thermocouples were inserted into the inlet and outlet ports, as well as in the reservoir. Temperatures were recorded with a FLUKE 54II B thermometer. A 24V Jabsco pump capable of delivering 2.9 GPM was inserted between the reservoir and inlet port and a metering valve, to control flow, was placed between the outlet port and the reservoir ( Figure 1A). The reservoir was placed on a stir plate (Corning) and blackwater in the reservoir was agitated throughout microwave experiments to ensure a homogenous temperature. A total volume of 1.5 L of blackwater was re-circulated through the system at a flow rate of 1 LPM until the desired reservoir temperature was achieved. To ensure the integrity of the microwave compartment was not compromised a microwave leakage detector (EMF-810, Lutron Electronics) was used to confirm no electromagnetic radiation was detected outside of the unit during operation. The desired temperature and heating rate was achieved by controlling the microwave power setting.

For conventional thermal heating experiments a 2L glass beaker was placed on a combination stir/hotplate (Corning) and 1.5L of blackwater was heated under continuous agitation. A Type K thermocouple was placed in the center of the blackwater ( Figure 1B). Temperature was recorded with a FLUKE 54II B thermometer. The desired temperature and heating rate were achieved by controlling the analogue heat setting (1–10) on the hotplate.

To test for the applicability of microwave disinfection at a scale practical for implementation, a stack of 5 domestic microwave ovens (Panasonic NN-SA661S), 1200W, 2450 MHz, 1.2 ft ³ was constructed by placing one microwave per shelf on a 5 shelf rolling cart. Two holes were drilled into the back of each microwave and flared Teflon feed through fittings were inserted and sealed with glue and microwave shielding adhesive to create the inlet and outlet ports for each microwave. Five feet of coiled Teflon tubing (ID: 6.35 mm) was placed into the microwave compartment of each microwave. With the exception of the first and last microwaves in the stack, the outlet of the proceeding microwave was connected to the inlet of the next microwave in the series with the same Teflon tubing. The inlet of the first microwave was connected to an untreated blackwater reservoir and the outlet of the last microwave was connected to the treated blackwater reservoir using prothane ii tubing (ID 7.95 mm) ( Figure 1C). The total tubing volume was 0.54 L. For some experiments an insulated copper coil was inserted between the outlet of the last microwave and the treated blackwater reservoir to increase retention time at temperature. A 24V Jabsco pump capable of delivering 2.9 GPM and a metering valve were inserted between the untreated reservoir and inlet port of the first microwave in the stack ( Figure 1C). The desired temperature was achieved in a single pass through the stack by controlling the flow rate and microwave power setting. To ensure the integrity of the microwave compartments were not compromised a microwave leakage detector (EMF-810, Lutron Electronics) was used to confirm no electromagnetic radiation was detected outside of the units during operation.

This study was performed using blackwater obtained from a prototype toilet system, described elsewhere ²⁷ . The blackwater was composed of 25% urine in recycled flush water contaminated with feces and had a total solids (TS) content of approximately 0.7%. Simulated pit latrine sludge was made by mixing feces and urine to a TS content of 1–3%, in a 19 L polypropylene container. The simulated pit latrine sludge was then allowed to sit undisturbed for at least a month. The raw feces and urine were obtained from healthy anonymous volunteers 20–50 years in age. All feces and urine collection methods were reviewed and approved by RTI’s institutional review board. Blackwater or simulated sludge was used as collected or diluted with DI H ₂O as necessary to achieve the desired TS content. The initial temperature of blackwater, simulated sludge, and feces, prior to all experiments, was room temperature.

Three separate microwave or heating systems were used in this study ( Figure 1) and have been described previously in this publication ( sections 2.1 – 2.3). The conventional hotplate heating system ( Figure 1B) was used to determine the time and temperature kill dynamics of blackwater. The single microwave system ( Figure 1A) was used to determine if microwave radiation played a role in disinfection by comparing to conventional hotplate heating. This was accomplished by maintaining comparable heating rates in the two systems (~1°C/min). The single microwave system was also used to determine the effect of TS on microwave sterilization of blackwater.

Intact feces was exposed to microwave radiation by evenly distributing an intact fecal sample in the bottom of a sealed microwave safe plastic container. Temperature was verified by removing the sample from the microwave and inserting a type K thermocouple into the center of the fecal sample. Temperatures were recorded with a FLUKE 54II B thermometer.

The microwave stack system ( Figure 1C) was used to determine the feasibility of a single pass high throughput system (i.e. practical application). All experiments were performed at least twice to confirm results, except for the intact feces trial, which was performed once.

Lysongeny broth (LB) was used as culture media and consisted of 10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl in deionized water. The broth was sterilized at 121°C for 15 minutes and then cooled to a temperature allowing handling. Samples for a three tube most probable number (MPN) assay were collected using clean sterile pipettes and 15 mL centrifuge tubes. For MPN quantification, the FDA method for MPN from serial dilutions was used by serially diluting (10 ^-1 – 10 ^-8) each sample with LB and culturing 1mL of the diluted sample in a single well of a sterile 48-well plate ²⁸ . Each dilution, at each time, point was run in 3 individual wells. Samples were incubated at 37°C for 48 hours before being analyzed. Reported values are the average of at least 2 individual trials.

TS were determined according to the EPA method 1684 “Total, fixed, and volatile solids in water, solids, and biosolids”. Briefly, new clean aluminum weigh boats were heated at 103–105°C for an hour, cooled to room temperature and used immediately. The weight of the empty dish was recorded as W _dish . Next a 20–50 mL sample of blackwater was added to the dish and the weight recorded as W _total . The sample was then dried at 103–105°C overnight, cooled to room temperature and the weight recorded. The sample was then returned to the oven at 103–105°C and heated for an hour, cooled and re-weighed. This process was repeated until the weight change was less than 1 mg/L, this was recorded as W _dry . TS were determined as follows.

T o t a l s o l i d s ( % ) = ( W d r y   –   W d i s h W t o t a l   – W d i s h ) * 100

Redox potential (ORP), pH, and conductivity were all measured using a Myron L 6PFC ^E Ultrameter II (Myron L Company, Carlsbad CA) according to the manufacturer’s instructions. The conductivity cell and pH/ORP sensor wells were rinsed 3 times with the sampled blackwater at each time point, prior to data recording. The heating rate was calculated by dividing the change in temperature ( ΔT in °C) over the time period ( Δt in minutes) in which the change occurred.

For experiments involving comparison of two or more disinfection protocols over time at different temperatures, log-transformed MPN data were compared by two-way ANOVA with a Sidak’s multiple comparison test. For experiments comparing the effect of heating method (microwave versus hot plate) or solids content on disinfection at a given temperature, mean log reductions upon reaching that temperature were compared by Student’s t test. A p value of < 0.05 was considered statistically significant. Statistical calculations were performed with GraphPad Prism v 7.01.

Energy usage measurements were calculated by averaging the wall current draw for a single microwave running at 100% power for five minutes using Mastech MS 2138R inductive ammeter. For hotplate energy the total time needed to heat the blackwater to 95°C, on the highest analog heat setting (10), was used. For microwave stack energy the total time each microwave was required to be on, at full power, was used. The energy required to heat 1L of water from room temperature to 95°C was used to determine the energy efficiency of each system.

The time-temperature disinfection kinetics of blackwater conventionally heated on a hot plate under constant agitation are shown in Figure 2. The blackwater was obtained from the prototype toilet system as described in the Methods section 2.4. The average initial thermal heating rate and TS across all trials were 1.01 ± 0.15 °C/min, and 0.55 ± 0.16 % respectively. The average initial pH, conductivity, and ORP were 8.73 ± 0.12; 14651 ± 3631 µS/cm; and -224 ± 82 mV respectively. Heating resulted in complete disinfection (MPN below the disinfection threshold of 5/ml) upon reaching 95°C. Heating to either 70°C or 80°C resulted in a significant reduction in MPN, but did not reach the disinfection threshold. The reductions in MPN upon achieving 70°C and 80°C (5.2 and 5.4 log reduction, respectively) were not significantly different from each other suggesting that below 95°C a hotter temperature did not result in a faster kill ( Figure 2). Maintaining the blackwater at 70°C or 80°C for a further 100 minutes resulted in an additional 0.6 log and 0.9 log reduction, respectively; however, this difference was also not statistically significant ( Figure 2). Therefore, we propose, 95°C should be the target temperature for sterilization of blackwater.

Using the bench top single microwave system ( Figure 1A) we sought to determine if microwaves could affect disinfection through a non-thermal mechanism by comparing disinfection of microwave irradiated fecal contaminated blackwater at a sub-lethal temperature to that of conventionally heated blackwater ( Figure 3A). A sub-lethal temperature of 40°C was chosen because it is was not expected to affect cell viability and any disinfection could be attributed to a non-thermal mechanism provided by the microwaves. Initial blackwater conditions can be found in Table S1. A temperature of 70°C ( Figure 3B) was chosen to investigate if microwave radiation had a synergistic effect on disinfection at elevated temperature (i.e., whether microwaves plus heat provide enhanced disinfection over heat alone). Initial blackwater conditions can be found in Table S2. No reduction in MPN was seen under either condition (MVH or conventional heating) at 40°C. Heating to 70°C resulted in significant reductions in MPN in both conditions that were not significantly different from each other. The heating rates of both systems were maintained at approximately 1°C/min to ensure equal time-temperature exposure and that the only difference between samples was the addition of microwaves in the MVH system. Thus, our data show that microwaves provided no enhancement of disinfection under non-lethal or lethal temperature conditions when compared to conventional heating.

Figure 4 shows the effect of TS content on disinfection after MVH to 70°C of intact feces, high (2.3%), and low (0.19%) TS samples in the bench top microwave apparatus. Initial blackwater conditions can be found in Table S3. The bacterial counts in the low TS sample decreased an average of 3.8 log orders while the high TS sample decreased 3.7 log orders whereas intact feces had a 3.0 log decrease, with final MPN’s of 1.8 × 10 ³, 2.2 × 10 ⁴, and 3.2 × 10 ³ bacteria/mL respectively. The differences among these values are well within the confidence intervals of the MPN assay and thus no significant difference in disinfection was detected between samples of varying total solids contents. The high and low TS samples were heated at comparable heating rates using the bench top microwave system shown in Figure 1A. Because the intact fecal sample was too viscous to pass through the flow through system it was placed in a sealed plastic container in the central microwave cavity on high power. This resulted in the fecal sample having a higher heating rate (9.64°C/min) than the low and high TS samples. Interestingly, the intact fecal sample was dewatered by microwave heating, TS prior to microwave heating was 19.9% and increased to 73.4% post microwave treatment. A similar dewatering phenomenon was observed in a recent study by Maiwoo et al. 2016 ¹⁵ .

A five microwave, single-pass, flow-through, system was developed to simulate a practical microwave disinfection system. For example, one that could be mounted on the side of a vacuum tanker. The experimental goals for this system were to determine the fastest flow rate that would allow the blackwater to reach 95°C in a single pass and the total necessary energy expenditure, as well as the feasibility of field implementation. This temperature was chosen based on the data from Figure 2 showing complete blackwater disinfection at 95°C. Initially, an 18 L volume of tap water was passed through the system to determine the optimal flow rate (0.624 LPM) for achieving 95°C from the 5 microwave stack under full power. However, the heating profile of blackwater differed from pure water mostly due to the instability particles in the blackwater caused in the flow meter, resulting in unstable flow rates and temperature fluctuations and highlighting the need for caution when designing water treatment systems based off of simulated or surrogate effluents, a challenge this group has encountered and addressed previously ²⁷

Once the system was adapted for blackwater, a flow rate of 0.82 LPM was used with the average time in the microwave stack being 39.7 seconds. Under these conditions the blackwater remained above 80°C for 12 seconds and 95°C for 6 seconds, however this was not sufficient to achieve full disinfection (Trial 1, Table 1). For Trials 2–5 ( Table 1) a 50 ft length (2.4L) of coiled insulated copper piping was added to the output of microwave five to allow the blackwater to remain at an elevated temperature for an additional amount of time. The temperature drop from the inlet to outlet of the coil was 1°C. The coil increased the length of time the blackwater was above 80°C to 2.9 minutes and the time above 95°C to 2.8 minutes however, complete disinfection was still not achieved. This process of increasing the time above 95°C was continued up to 9.4 minutes ( Table 1). When sterilization was still not achieved, the length of copper tubing was increased to 100 ft, allowing the blackwater to remain at temperature for an additional 18.8 minutes (Trials 7 and 8) where full sterilization was achieved ( Table 1). When we compare these results to those obtained in Figure 2 we can infer the difference in the time-temperature kinetics is due to the difference in come-up time. The samples in Figure 2 were at a temperature above 80°C for approximately 15 minutes before 95°C was achieved, while the samples in Table 1 were heated rapidly using the full power of the microwave. These results tell us the advantage of rapid heating microwaves provide is limited because an extensive time at elevated temperature is still needed for sterilization; however, this limitation can be overcome by the addition of insulated tubing allowing the effluent to be exposed to the proper time-temperature conditions without additional energy input.

The time required by the hotplate to heat 1L of water on the highest setting was 48 minutes, while the 5 stack of microwaves needed only 2.14 minutes. Overall the wall plug power measured for the microwave system was higher than for the hotplate heating system, 1920 W vs 313 W respectively. However, due to the longer time required by the hotplate to heat 1L of blackwater, the total energy required was 32% greater than that used by the 5 stack microwave system ( Table 2). The throughput of the microwave system was 15 LPH. These results show a microwave heating system for the disinfection of blackwater has the potential to be an energy efficient field option, however throughput conditions need to be optimized.