Inhibition of ammonia and hydrogen sulphide as faecal sludge odour control in dry sanitation toilet facilities using plant waste materials

On-site dry sanitation facilities, although cheaper than wet sanitation systems, suffer from high malodour and insect nuisance as well as poor aesthetics. The high odour deters users from utilizing dry sanitation toilets as an improved facility leading to over 20% open defecation in Sub-Saharan Africa. To address this malodour concern, this study first assessed odour levels, using hydrogen sulphide (H2S) and ammonia (NH3) as indicators, on two dry sanitation facilities named T1 and T2. The potential of using biomass (sawdust, rice husk, moringa leaves, neem seeds), ash (coconut husk, cocoa husk) or biochar (sawdust, rice husk, bamboo) as biocovers to remove or suppress odour from fresh faecal sludge (FS) over a 12-day period was investigated. Results showed that the odour levels for H2S in both T1 (3.17 ppm) and T2 (0.22 ppm) were above the threshold limit of 0.05 ppm, for unpleasantness in humans and vice versa for NH3 odour levels (T1 = 6.88 ppm; T2 = 3.16 ppm; threshold limit = 30 ppm limit). The biomasses exhibited low pH (acidic = 5–7) whereas the biochars and ashes had higher pHs (basic = 8–13). Basic biocovers were more effective at H2S emission reduction (80.9% to 96.2%) than acidic biocovers. The effect of pH on suppression of NH3 was determined to be statistically insignificant at 95% confidence limit. In terms of H2S and NH3 removal, sawdust biochar was the most effective biocover with odour abatement values of 96.2% and 74.7%, respectively. The results suggest that biochar produced from locally available waste plant-based materials, like sawdust, can serve as a cost-effective and sustainable way to effectively combat odour-related issues associated with dry sanitation facilities to help stop open defecation.

www.nature.com/scientificreports/ houses with wet toilet facilities, the majority of the natives use dry on-site toilet systems including shared facilities. This is because many of the houses are old structures built without toilet facilities. So inhabitants are compelled to use the nearest available public toilets. Some dwellers share public toilets intended for public schools. For this research, two public, dry on-site toilets (VIP latrines) were considered. The first public toilet (TI), which is located close to the Ayeduase market has depth beyond 2.5 m and houses ten (10) squat holes; one-half dedicated to each sex with an inter-squat hole concrete-partition-separation. An average of seventy-five (75) people use the toilet daily with a quarterly de-sludging frequency every year. The second public toilet (T2), is located close to the Ayeduase school, has a depth of almost 2 m and equipped with twelve (12) squat holes-two sets of five squat holes placed back to back on opposite sides of a dividing wall, with one set assigned to males and the other set to females. Every squat hole within a set is separated from each other by a concrete partition. The remaining two squat holes were adjacent to the five squat holes and contained in enclosed rooms. Over ninety (90) residents and school children patronize this toilet facility and are de-sludged every other week. Both TI and T2 were fitted with vent pipes at heights exceeding 500 mm above the roof of the superstructure.
Determination of on-site odours from VIP latrines. To determine the degree to which odour was a nuisance in the use of VIP latrines, direct on-site measurements of H 2 S and NH 3 concentrations, representing odour, were carried out in the enclosures of T1 and T2 without the need for gas collection. Odour readings were taken from both T1 and T2, consistently for ten (10) days; three times daily: morning (6:30-7:30), afternoon (12:30-13:30) and evening (17:30-18:30) in greenwish mean time (GMT). The odour-causing gases were detected and quantified via direct air measurements in the enclosures of T1 and T2 using an aeroqual potable gas analyser (series 200, New Zealand).

Sampling protocol of FS for laboratory-based experiments. FS from both T1 and T2 was sampled
from the surface of the pile of excreta beneath the pit pedestal. At the time of sampling, FS level in TI and T2 was, respectively, about 2 m and 10 cm, away from the squat hole. The sampling was undertaken by scraping off the top of the excreta with a one-meter-long ladle-like tool, to obtain a representative "fresh" faecal matter sample. The samples were collected into a tightly capped, 2000 ml plastic bucket and transported to an environmental laboratory in the civil engineering department of Kwame Nkrumah University of Science and Technology (KNUST) for analysis.

Acquisition and production of biocovers for odour mitigation. Locally available materials includ-
ing agricultural waste were used as potential biocovers for the mitigation of odour release from FS. Seven (7) biomasses were obtained for this experiment: sawdust (SD) from Celtis Mildbraedii, rice husk (RH), moringa leaves (M), neem seeds (NS), cocoa husk (CH), coconut husk (C N H) and bamboo (B) (See Fig. 1). Some of these biomasses were used, as is-sawdust, rice husk, moringa leaves, neem seeds-or were thermally treated via pyrolysis and ashing operations.
For pyrolysis, only biomasses from rice husk, bamboo and sawdust were utilized as feedstock. The desired biomass was weighed and fed into the pyrolysis reactor with pyrolysis conditions of 400 °C for 90 min of sustained pyrolysis. After pyrolysis, the produced biochar was left to cool in the reactor for about 30 min before transferring into tightly capped vials to be stored for further experiments.
For ash production, only biomasses from the cocoa husk and coconut husk were used. Each biomass was fed into a kiln (HT13T7, Kiln and Furnace Limited, Keele St. Tunstall, Stoke-On-Trant) where ashing was undertaken at a temperature of 700 °C. After ashing, the kiln was switched off and allowed to cool to within 60 °C and 80 °C. The ashes were collected and stored in tightly capped glass vials to be used for further experiments. Note that all acquired and produced biocovers were ground to within 210 to 75 µm and used for further experiments. Acronyms of B for biomasses, BC for biochars and A for ash were attached to the feedstocks for each process to help in identifying the thermal condition employed. Consequently, the obtained biocovers were tagged as SD-B, RH-B, M-B, and NS-B for biomasses of sawdust, rice husk, moringa powder and neem seeds powder, respectively. Tags were also assigned to sawdust biochar (SD-BC), rice husk biochar (RH-BC) and bamboo biochar (B-BC). Biocovers from the ash were also tagged for cocoa husk ash (CH-A) and coconut husk ash (C N H-A).
Evaluating the effect of additive application as biocovers for malodour mitigation. The experimental setup is shown in Fig. 2. The experimental design is a completely randomized design with two replicate measurements. The desired mass of each additive-biomass from sawdust, powdered rice husk, powdered moringa leaves and neem cake; ash from the cocoa husk and coconut husk and; biochar from sawdust, rice husk and bamboo-corresponding to 5wt.% (1:20 w/w) was determined and transferred into a 500 ml conical flask containing 300 g of fresh FS. Care was taken to ensure complete coverage of the FS in the conical flask with the applied biocovers. A control sample, which contained only FS without additives, was also include to facilitate the determination of the odour-removal efficiencies of the biosolid additives. Each conical flask was fitted with a single-perforated tightly fitting cork connected with a latex tubing that ends in 500 ml air-bag for gas trapping for further analysis. Analysis of the trapped gases was undertaken every three (3) days for 12 days. To prevent leakages, all openings around the corks were sealed with a sealant. The experiments were performed in two replicates. The performance of the biocover was evaluated based on the per cent reduction in odour-NH 3 or H 2 S-using Eq. (1).
(1)    www.nature.com/scientificreports/ where C 1 is the odour of the control sample (faecal sample without biocover) at a gas sampling time; C 2 is the odour of the biocover-applied faecal sample at that same gas sampling time.
Characterization experiments and statistical analysis. The fresh FS samples were analysed for chemical oxygen demand (COD-dichromate approach) 54 and biochemical oxygen demand (BOD) using the Winkler method. Both the FS and biosolid additives (biocovers) were analysed following the standard methods for water and wastewater analysis for the determination of total organic carbon (TOC), fixed solids (ash), total volatile solids (TVS) and total solids (TS) 54 . TKN content for FS and biosolid additives (biocovers) was determined following the EN 13342 standard 55 . Also, pH was determined using a calibrated pH meter (Palintest micro 800 Mult, Singapore). The trapped gases in the airbags from the experimental setup in "Evaluating the effect of additive application as biocovers for malodour mitigation" section were analysed qualitatively and quantitatively for H 2 S and NH 3 gases-representative of the malodorous gases-using the biogas 5000 gas analyser. Also, a comprehensive statistical analysis [two-way and one-way analysis of variance (ANOVA)] using the data analysis add-in in Microsoft® Excel was performed. The effect of two independent variables (biocover type and duration of biocover application) on suppression of odour (response variables: H 2 S and NH 3 ) were investigated using the two-way ANOVA without replication function. The one-way ANOVA was, however, employed to assess the statistical differences between the applied biocover types (biomass, biochar, ash) on the overall odour suppression for the entire duration of the experiment. A confidence level of 0.05 was chosen as the basis to either reject or fail to reject the null hypothesis of no statistically significant difference for the comparison. The Tukey-Kramer multiple comparison test was utilized for specificities in cases where the null hypothesis was rejected.
Consent to participate. All authors confirm participatory consent.

Consent for publication.
All authors accept to publishing in this journal.

Results and discussions
On-site odour evolution. Variations Fig. 3a. Generally, the highest H 2 S concentration in the public toilet occurred during the mornings and evenings ( Supplementary Fig. S1a); expectedly a consequence of the most patronized times in the day. Other factors such as user conduct and effectiveness of cleaning activities may have played a role. From Fig. 3a, it was evident that, except for day 6, daily averages of H 2 S was higher in T2 than T1, with some concentrations as high as 26.8 (day 5), 34.0 (day 8) and 39.2 (day 3) times that of T1. This could be attributed to the depth of the sludge in the pit of T1 (less than one-third of the pit depth) and T2 (almost filled to the brim), which is a consequence of the patronage frequency and cleaning activities.
Consequently, more H 2 S will escape from the pit into the privy room in the case of T2 than T1, even though both facilities were fitted with vent pipes. Also, H 2 S is heavier (density of 1.36 kg/m 3 ) than air (density of 1.225 kg/m 3 ) and as such usually lingers at the base of the latrine 56 even when fitted with vent pipes. Similar observations have been made by other authors who ascribed the observations to a large number of users 57 . Clearly odour in both T1 and T2 was detectable [> 0.005 ppm 20 ]. Except for day 1 (0.021 ppm) and 7 (0.018 ppm) for T1, the daily H 2 S averages exceeded the guideline value of 0.05 ppm 8,24 for all toilet facilities investigated, which will inevitably elicit complaints from users and residents, and hamper patronage of the toilet facilities.
Variations in NH 3 concentration on public toilets. The concentration of NH 3 was another component of odour that was measured in the two public latrines. Supplementary Fig. S1b shows the NH 3 concentration detected at different times in the day, during the 10-day experiment in T1 and T2. There was no observable trend for NH 3 evolution based on the sampling times. However, the daily averages ( Fig. 3b) showed a higher evolution of NH 3 for T1 than for T2 (except for day 4). Except for day 2, 3, 4 and 5, most of the daily NH 3 released were within the detectable threshold of 4 to 8 ppm for humans for T1 (day 1 = 4.90 ppm; day 6 = 7.04 ppm; day 7 = 4.70 ppm; day 8 = 4.88 ppm; day 9 = 6.88 ppm; day 10 = 5.49 ppm). For T2, however, none of the readings went beyond the detectable threshold for humans. It is noteworthy that all measured NH 3 concentrations were below the threshold of unbearableness and irritation for humans [(10 min exposure at 30 ppm-slight irritation; 10 min to 2 h exposure at 50 ppm-moderate irritation to the eyes, nose, throats and chest) 58 ]. According to Strande and Brdjanovic 57 , factors such as diet, climate, type of toilet facility, number of users among others influence odour release. Differences in the NH 3 measurement for T1 and T2 were attributed to the level of FS in the pit. Level of FS in TI and T2 were respectively, about 2 m and 10 cm away from the squat hole suggesting a poor maintenance regime especially for T2 since desludging should be undertaken when the sludge is about 50 cm from the slab. As such, NH 3 which is less dense (0.73 kg/m 3 ) than air (1.23 kg/m 3 ) at 15 °C at sea level, is more likely to escape easily into the atmosphere for T2 since the FS, in this case, was much closer to the squat hole; this may explain why less NH 3 was measured in T2 compared to T1. As such, the gas detector potentially recorded less NH 3 concentration in the privy room. For T1 because the sludge depth in the pit was less than one-third the depth of the pit, enough room was available for NH 3 to linger in the pit, which accounted for the reported higher NH 3 concentrations. Generally, it can be observed that NH 3 concentrations recorded in both toilets were higher than that for H 2 S in the latrine (Fig. 3). The NH 3 and H 2 S results in this study were in agreement with that obtained by Obeng et al. 8 Table 1 shows the characteristics of FS. The quotient of COD to BOD (COD/BOD ratio) obtained in this study was two (2); which is low compared to others reported in the literature. For instance, FS with a COD/BOD ratio of 5 and 6 was reported by Strande and Brdjanovic 57 and Jeuland et al. 59 for public toilets, respectively. Strande and Brdjanovic 57 intimated that the higher value of 5 was indicative of the slow degradation of organic matter. Besides, it is also known that the characteristics of FS vary depending on parameters such as diet, type of toilet technology, climate and the type of cleansing material utilised 8 . These can thus explain the variations of the COD/BOD ratio. Moisture content in FS obtained from the public toilet was high (80.5%) as expected. Strande and Brdjanovic 57 reported a similar result of high moisture content (83%) for dry VIP latrine sludge. Moisture content values ranging from 53 to 92% have been reported 60 . The weather condition was identified as a contributing factor   61 although FS is usually rich in nitrogen, changes in the expected nitrogen content in FS is largely subject to the diet of the user. For example, a highly proteinaceous diet will result in higher nitrogen content in FS 57 .

Characterization of biocovers employed as odour-reducing additives. The physicochemical
properties of the different biocovers employed as odour-reducing additives in this study are presented in Table 2. Moisture content for all biocovers was relatively low (< 5% for ash and biochars; within 8.6 to 11.2% for biomass) except for SD-B (30.8%). Less moisture content tends to hinder the growth of organisms, therefore, additives with less moisture content play a significant role in reducing bacterial growth 63 and consequently, potentially lessening odour-production and release.
Interestingly, the contents of fixed solids were observed to be inversely proportional to the volatile solids. It has been established that, upon increasing temperature for the determination of volatiles, the volatile matter is driven off, burnt away leaving ash, or fixed solids. C N H-A and CH-A both had higher fixed solids (respectively, 95.8% and 84.2%) compared to the other biocovers. This is because of the higher temperature they were subjected to, for ashing to take place; thus essentially eliminating the existing volatiles and leaving behind the fixed solids. Also, notice that the fixed solids in the biochars (RH-BC = 39.6%; SD-BC = 10.2%) were higher than that of their corresponding biomasses (RH-B = 15.9%; SD-B = 1.9%). Reasons are similar to those described earlier on in the text.
Carbon content was also generally high for biochars (B-BC = 53.7%; RH-BC = 35%; SD-BC = 52.1%) and biomasses (NS-B = 52.8%; M-B = 52.3%; RH-B = 48.8%; SD-B = 56.9%), and extremely low for ash biocovers (C N H-A = 2.4%; CH-A = 9.2%) for reasons attributed to the extent of thermal treatment each biocover precursor material underwent. That notwithstanding, SD-B's high carbon content may be a consequence of the source of the biomass, Celtis Mildbraedii; which is woody 64 . Similar results of the high carbon content of woody biomass have been reported by Sewu et al. 35 .
For C/N ratio, CH-A recorded the highest value of 125 followed by RH-B at 103. Moringa and neem seed powder recorded the lowest C/N ratio of 17 and 13, respectively suggesting higher nitrogen contents relative to carbon contents in these materials. A high C/N ratio has been reported to have an impact on the reduction of odour levels in compost 65 .

Evaluation of the odour-reduction/removal performances of the applied biocovers. Effect of biocovers as additives on H 2 S reduction.
The effect of biocovers on the suppression or inhibitions of H 2 S evolution from FS for each studied sampling day over 12 days are shown in Fig. 4a. Results show that apart from the inherent properties of biocovers-its efficacy on H 2 S reduction was time-dependent. Generally, biocovers with pHs in the acidic range required more time to be as effective as biocovers in the basic range. In addition, there was a general decline in H 2 S evolution with FS ageing. H 2 S was released most when FS was freshest at the time of sampling (day 3). This was particularly the case for the control sample and the biomasses (acidic biocovers). The basic biocovers, except for C N H-A, rather showed the most release of H 2 S on sampling day 6; with a trend consistently following the order: 6th day > 3rd day > 9th day > 12th day (in terms of concentration of H 2 S released). Furthermore, the application of basic biocovers led to a dramatic decrease in H 2 S evolution on the first sampling day (day 3); which was impressive. For instance, except for C N H-A, decreases were over 80% from the control value of 1245 ppm to 37 ppm for B-BC (97%); 198 ppm for RH-BC (84.1%); and 24 ppm for both SD-BC and The results of the overall per cent reduction in H 2 S over the entire study period of 12 days are shown in Fig. 4b. It is evident that, generally, all biocovers performed well (over 55%) in mitigating H 2 S release. However, the extent of H 2 S reduction (%) was greater in the basic biocovers than in the acidic biocovers. Basic materials are known to reduce H 2 S release best since the increase in pH converts H 2 S to sulphides; essentially trapping it within the FS 20 . That notwithstanding, amongst the basic biocovers, biochars were more effective at diminishing H 2 S evolution than ash with a near-complete reduction in H 2 S evolution [biochars: 96.2% (B-BC and SD-BC), 81.6% (RH-BC); ash: 80.9% (C N H-A) and 89.1% (CH-A)]. This contrary result of better reduction of H 2 S for biochar than ash, despite the higher pH of ash, may be due to the high surface area, surface functionality and porosity of biochar. Consequently, biochar may adsorb H 2 S thus limiting its release into the atmosphere. The carbon contents may likely be another reason, as the highest performing biocovers [biochar = 96.2% (B-BC and SD-BC); ash = 89.1% (CH-A)] also exhibited the highest carbon contents within the category of biochar (B-BC = 53.7%; SD-BC = 52.1%) and ash (CH-A = 9.2%) for the basic biocovers. Nevertheless, the higher performance of basic biocovers with high carbon content than acidic biocovers with comparable carbon contents for H 2 S reduction suggests that the earlier reasoning about surface area, pH, porosity and surface functionality may better explain   66 and typically find suitable niches in the applied biocovers. Porous and high surface area biocovers have been shown to provide suitable niches for microbial attachment and growth, and consequent inorganic/organic volatile compounds adsorption 66,67 . Thus, as the diffusion of the evolved H 2 S from the faecal sludge is slowed due to the physical barrier effect of the biocovers, contact with the sulphur-metabolizing bacteria such as Ochrobactrum, Paracoccus, Comamonas and Pseudomona 68 increases thereby yielding better H 2 S attenuation results, especially in the case of biochars, particularly SD-BC. Of the three groups of biocovers used, biochars have the most porosity and surface area hence the observed results.
Effect of biocovers on NH 3 reduction. The effect of biocovers on the suppression or inhibitions of NH 3 evolution from FS for each studied sampling day over the 12 days are shown in Fig. 5a. Generally, there seem not to be a clear trend in NH 3 suppression with time given the utilized biocovers. No trend consistent with pH values, C/N ratio or carbon content was found in this study. Nevertheless, contrasting results were observed for biomasses and their corresponding biochars. For instance, whilst RH-B exhibited a general gradual decline in NH 3 evolution with time [6th day (4.5%v/v) < 9th day (1.7%v/v) < 12th day (0.3%v/v)], its biochar (RH-BC) rather displayed an enhancement in NH 3 evolution with time [3rd day (6.3%v/v) > 6th day (6.8%v/v) > 9th day (8.7%v/v)]. www.nature.com/scientificreports/ In addition, compared to the control, the decline in NH 3 evolution was drastic and rapid for all sampling days over the 12 days with the application of SD-BC [3rd day (3.8%v/v); 6th day (1.8%v/v); 9th day (1.3%v/v); 12th day (0.2%v/v)]. A similar observation was also seen for the SD-BC precursor (SD-B) except on day 6; where a heightened NH 3 evolution rather occurred. The results of the overall percentage reduction in NH 3 over the entire study period of 12 days are shown in Fig. 5b. It is evident that majority of the acidic biocovers reduced the emission of NH 3 much better than the basic materials. In fact, except for SD-BC, all basic biocovers were not just extremely poor at attenuating NH 3 , but rather facilitated its release when compared to the control sample; by 58% (B-BC), 1.8% (RH-BC), 68.7% (C N H-A) and 49.1% (CH-A). Conversely, however, except for M-B, the acidic biocovers were good attenuators of NH 3 evolution. Particularly, RH-B (64.8%) was the most effective amongst the acidic biocovers at attenuating NH 3 evolution, followed by NS-B (17.4%) and SD-B (12.5%). Interestingly, not only was SD-BC the only effective biocover amongst the basic biocovers, it was also the most effective (74.7%) amongst all the investigated biocovers in this study in attenuating NH 3 evolution. The plausible reason for this observation was attributed to microbes-related (such as Rhodanobacter, Gemmatimonas, Flavisolibacter, and Sphingomonas) biotransformation and/or biodegradation due to the potentially higher surface area and porosity of SD-BC, thus yielding a more conducive environment for microbial attachment and growth 66,69 . The above reasoning was considered the most plausible for the superb performance of SD-BC, because except for the moisture content, its physicochemical properties were comparable to that of B-BC, yet B-BC exhibited a contrary result on NH 3 performance. According to Atia et al. 20 the application of the biocovers reduces emissions of NH 3 and other odorous gases in two ways: (1) physically limiting the emissions of NH 3 and other gases; (2) creating a biologically active zone on the top of the covers where the emitted NH 3 and other gases are aerobically decomposed by microorganisms. The effectiveness of different covers or odour-reducing material in mitigating H 2 S and NH 3 emissions vary and it is also dependent on the quantity of the materials added as a cover. In theory, the effective suppression of odour is influenced by the pH which creates an unfavourable environment for microbial growth, and the physical masking ability of the additives 20,25 . Statistical analysis of the effect of biocover type and duration of application on the suppression of odour from FS. The results of the applied statistical analytical tools generated by Microsoft® Excel are shown in Table 3. From the two-way ANOVA, the computed F values for the source of the variations [biocover source = 3.78 (H 2 S), 3.39 (NH 3 ); duration of application = 5.04 (H 2 S), 4.76 (NH 3 )] were all, greater than that of the F crit values (2.36 for biocover source; 3.01 for the duration of application). In addition, the P-values were lower than the level of significance at 0.05. From the aforementioned results, it was evident that the effects of the independent variables (biocover source and duration of application) on the suppression of both H 2 S and NH 3 evolution were statistically significant. These deductions were made based on two criteria: F value and the P-value. Values of F greater than reference F crit, and P-values lesser than the set level of significance at 0.05 (95% confidence limit) are indicative of a significant contribution to the variation by the group under investigation. Also, it was evident from the one-way ANOVA that variations in the means between the biocover type (biomass, biochar, ash) were statistically significant for H 2 S [F (26.34) > F crit (5.148); P-value (0.0011) < 0.05] and insignificant for NH 3 [F (1.956) < F crit (5.148); P-value (0.223) > 0.05]. Consequently, for the H 2 S, Tuker-Kramer multiple comparison test was used to evaluate which pair or combination of biocover type was the source of the Table 3. Results of the applied statistical analysis tools for the interpretation of odour suppression data (%). NB: Biocover source = (NS-B, M-B, RH-B, SD-B, B-BC, RH-BC, SD-BC, C N H-A, CH-A); Biocover type = (biomass, biochar, ash); duration of experiment, days = (3, 6, 9, 12); C.R. = critical range; df = degree of freedom; F = determined from experimental data using the F-test; F crit = F statistic obtained from the F-distribution; P-value = probability value; S = significant (absolute difference between mean odour suppression efficiencies is significant); NS = not significant (absolute difference between mean odour suppression efficiencies is not significant); n.a. = not applicable. www.nature.com/scientificreports/ variations for suppression of H 2 S evolution from FS. Clearly, the significant variations arose with biomass (low pH) and biochar/ash (higher pH) pairs suggesting that for effective H 2 S suppression, the pH of the biocovers is essential.
Economic analysis and feasibility studies. Determining the cost associated with using particular technologies is essential to the scale-up and ultimate success of the technology in a real application. Consequently, cost analysis was determined based on (1) accessibility of the raw materials for the biocover production; (2) processing/production of the biocovers; (3) performance of biocovers on H 2 S and NH 3 evolution; and (4) longevity of biocover. On the raw material accessibility, all materials were obtained as waste, thus the only cost likely to be incurred will be truck delivery and tipping fee, which hypothetically will be the same for all raw materials when factors such as proximity to the biocover production site and end-use (application) site remain the same. For the processing, however, the cost will follow the trend: biomass > biochar > ash. The main operation in the biomass processing was drying or used as-is. For biochar and ash, however, temperatures of 400 °C and 700 °C were used, respectively. Using a performance criterion of above 50%, it was apparent that from Figs. 4b and 5b, only RH-B (biomass: acidic) and SD-BC (biochar: basic) both exhibited H 2 S (56.6% and 96.2%, respectively) and NH 3 (64.8% and 74.7%, respectively) attenuation above the 50% threshold, for potential adoption in real application. However, for environmental considerations, RH-B may itself, decompose and release greenhouse gases (GHG) into the atmosphere owing to the abundance of labile carbon fractions in the biomass 70 . Conversely, SD-BC (biochar) which has a long carbon half-life contains aromatized carbons, which are unavailable for microbial consumption and can persist, therefore, for longer times (years) 71 thus increasing its overall utility, lessening application frequency and improving climatic conditions via carbon sequestration. Moreover, the pyrolysis process is considered a closed system and can be self-sustaining, where the other pyrolysis products (pyrolytic gases and bio-oil) are reutilized as fuel for running the process 72 , thus leading to drastic cost reduction. From the above analysis, employing SD-BC as biocover for H 2 S and NH 3 mitigation is feasible on a cost and removal efficiency basis.

Conclusions
This study assessed the potential of employing plant waste materials, biochar and ash as biocovers to attenuate foul odour evolution from faecal sludge in a laboratory setting. Results showed that odour levels, assessed with H 2 S and NH 3 as indicators, in the two public latrines were above the perceptible threshold of 0.005 ppm (H 2 S) and 0.05 ppm (NH 3 ) for humans and peaked in the mornings and evenings-correlated with patronage times.
Comparing the odour-causing substances, H 2 S and NH 3, only the former was above the threshold of unbearableness/annoyance of 0.05 ppm (H 2 S) and 30 ppm (NH 3 ) to humans in the toilets investigated. Characterization studies showed that the biomasses were acidic whereas the biochars and ashes were basic. Odour-suppression results showed that generally, high pH biocovers were more effective at suppressing H 2 S evolution from FS. The effect of pH on the suppression of NH 3 was determined to be statistically insignificant at 95% confidence limit. The per cent H 2 S and NH 3 reduction values were the highest for biocover from sawdust biochar; 96.2% and 74.7%, respectively. These results suggest that waste and readily available resources such as sawdust biomass, when converted to biochar, can serve as an effective tool to attenuate odour evolution from fresh faecal sludge in dry sanitation public toilets.

Data availability
Authors declare that data can be available upon request from the corresponding author.