Physico-chemical properties of waste derived biochar from community scale faecal sludge treatment plants

Biochar preparation

The faecal feedstocks for the preparation of the biochars used in this study were sourced from three different faecal sludge and septage processors in India: Narsapur in Andhra Pradesh, Warangal in Telangana and Wai, Maharashtra (Figure 2). Warangal and Narsapur treatment plants currently have a capacity of 15m³ per day, whereas the Wai treatment plant has a capacity of 70 m³ per day. FS collected from septic tanks is delivered to each processing plant where it is stored in holding tanks for the homogenization of the sludge. Tide Technocrats Private Limited have several community scale faecal sludges and septage processors which sanitize faecal waste and dewaters the sludge (5–10% moisture content) using solar energy in drying beds. Solar drying was managed on-site and expedited by spreading the sludge in a 10 mm layer. The sludge was pyrolyzed into biochar using a flame temperature operating range of 550–750°C. The process relies on autothermal operation, thus a limited supply of oxygen flows through an air fan into the main reaction chamber to allow for partial oxidation. process at the community scale faecal sludge and septage processors are outlined in Figure 1. Three 5kg biochar samples were collected from each processor in September 2018.

Figure 1. Flow diagram of waste through a Tide Technocrats Community Scale Faecal Sludge and Septage Processor adapted from (Tide Technocrats, n.d).

Figure 2. Photo of the community scale faecal sludge and septage treatment plant in Wai, Maharashtra (Tide Technocrats, n.d.).

Characterization of biochars

The biochars characterized were collected from the pyrolyzer.

Chemical analysis. Elemental C, N, S and H abundances were determined at Environmental Geosciences, University of Vienna, Austria using an elemental analyzer (Vario MACRO, Elementar).

The elemental composition (C, H, N, S and O) and ash content of the biochars were used to calculate the molar element ratios H/C, C/N, and O/C. The amount of oxygen in the samples was calculated from the subtraction of total percentage carbon, hydrogen, nitrogen, and sulphur and ash content from 100 (Castan et al., 2019).

Proximate analyses. Moisture and ash content of the three biochars were determined in triplicate by methods adapted from the literature (ASTM D 1762-84, 2011; Enders et al., 2012).

Crucibles and covers were cleaned by heating at 750°C for 6 hours and then cooling to 105°C. This procedure volatilized residual material on the crucibles. The crucibles were transferred to desiccators and cooled to ambient temperature. The mass of crucibles and crucible covers were recorded to 0.1mg and masses determined for all samples. Approximately 1.0g of biochar was added into each crucible. For the moisture determination the crucibles and covers were heated at 105°C for 18 hours and then transferred to desiccators whilst hot. The covers were removed briefly in order to safely remove the crucibles and covers from the oven. After cooling to ambient temperature, the mass of crucibles, covers and sample were recorded to 0.1mg for all samples.

For ash determination the covered crucibles with 105°C dry biochar was placed in the furnace. The covers were adjusted so that they were askew to allow air flow into the crucibles, while reducing the possibility of physical losses. The samples were heated from ambient to 750°C at a rate of 2°C per minute. The furnace was programmed to hold the temperature at 750°C for 6 hours then allowed to cool down to ~130°C. Crucible lids were adjusted to sit flush when the temperature of 105°C was reached. The crucibles were then removed from the furnace before placing in desiccators and left to cool to ambient temperature. The mass of each crucible and crucible cover with sample was recorded to 0.1mg.

Chars were ground to <850µm in a pestle and mortar to enhance representativeness of the sample and sieved to >149µm as this lessens physical losses upon rapid heating (Enders & Lehmann, 2015).

pH and electrical conductivity. The pH of biochar samples was measured by suspending 5.0g (ground to <2mm) biochar in deionised water in a 1:10 ratio (Singh et al., 2017). After 1 hour of shaking, suspensions were allowed to stand for 30 minutes before pH measurements were taken using a Voltcraft soil pH meter calibrated using pH 7 and pH 10 buffers. Electrical conductivity (EC) was measured on the same samples using a calibrated Whatman CDM 400 EC meter. The analyses of pH and EC were performed in triplicate.

FT-IR analysis. Fourier transform infrared (FTIR) spectra were used to identify the surface organic functional groups present in the biochar. High ash content in sludge-derived biochars leads to a high mineral content with bands in the infrared spectrum arising at similar wavenumbers to organic functional groups. To elucidate the different groups present, FTIR spectra of ashed biochars and de-ashed (acid washed) biochars were also generated.

Acid washing biochar to remove ash content (Klasson et al., 2009; Lima et al., 2016; Thomas Klasson et al., 2014) was achieved with 0.1 M HCl at a ratio of approximately 50:1 (v/w). Samples were shaken in a Uniwist 400 at 180 rpm for 2 hours before being filtered and washed with deionised water until a pH of 7 was reached. Samples were oven-dried at 80°C overnight.

The samples were gently ground using a pestle and mortar and analyzed using a Perkin Elmer Spectrum 2 FTIR spectrophotometer applying the Attenuated Total Reflectance (ATR) method with a diamond crystal. The resulting spectra were an average of 16 scans obtained in the range from 400 to 4000 cm⁻¹ with a spectral resolution of 2 cm⁻¹ for biochars and 4 cm⁻¹ for acid washed and ashed biochars.

Surface area. The BET (Brunauer, Emmett, and Teller) method is frequently used to determine the total surface area and pore size of materials. The BET analysis was conducted using the NOVA 2200e surface area and pore size analyzer (Quantachrome Instruments). The BET specific surface area of the three biochar samples were determined using two methods: N₂ as adsorptive gas at 77 K and CO₂ at 273 K.

Prior to these measurements, 200mg – 300mg of biochar (<2mm) were heated to 130°C under vacuum for a minimum of 4 hours. Then the samples were transferred to the instrument and outgassed at 105°C for a minimum of 4 hours following standard protocols. Samples were analyzed in triplicates.

For N₂ isotherms and CO₂ isotherms the BET equation was used to determine the specific surface areas from six points in the pressure region P/P₀ = 0.01–0.30 (Brunauer et al., 1938). For N₂ the pore size-distributions in the pressure region P/P₀ = 0.01–0.98 were ascertained using the built-in Density Functional Theory (DFT) model assuming slit-like pores. DFT considers micropore filling process, the development of the adsorbed film thickness, and importantly capillary condensation and evaporation, thus it can model hysteresis in the adsorption/desorption mesopore region of the isotherm.

For CO₂ isotherms the pore size-distribution, the cumulative pore volume (μPV) and the cumulative surface area (μSSA) in the pressure region P/P₀ = 0.001–0.030 were determined using the built-in Grand Canonical Monte Carlo (GCMC) simulation, again, assuming pores were slit-shaped.

The benefit of using of DFT and Monte Carlo simulation methods is that they provide a combined micro-mesopore analysis.

Measurement of zeta potential. Zeta potential measurements were undertaken according to methods reported in the literature (Samsuri et al., 2013; Yuan et al., 2011). The zeta potential values were determined by weighing 0.045g of 63µm sieved biochar into a 250ml conical flask and adding 180 mL of 0.1M NaCl solution to each flask. Five suspensions were prepared for each biochar at pH values between 5.0 – 9.0 with the pH of each suspension adjusted using HCl . Suspensions were then dispersed ultrasonically for 30 minutes at 30 ± 1 °C in a bath-type sonicator at a frequency of 40 kHz and a power of 300 W. The samples were then left to stand for 72 hours before being measured with a Malvern Zeta Sizer Nano. In total 15 suspensions were prepared and each suspension was measured a minimum of three times.

X-ray diffraction (XRD). The X-ray diffraction (XRD) analysis of the chars was conducted on a Bruker D8 Discover XRD. This was operated at 40 kV and 40 mA and the data collected over a 2θ range of 20–70° using the Cu-Kα radiation at a scan rate of 2° min⁻¹. The main phase peaks were identified by comparing the observed XRD patterns to the standards compiled by the Crystallography Open Database (COD) (Downs & Hall-Wallace, 2003; Gražulis et al., 2009; Gražulis et al., 2015, Gražulis et al., 2012; Merkys et al., 2016; Quirós et al., 2018).

SEM/EDX. SEM–EDX analysis offers detailed imaging data about the morphology and surface texture of individual particles, with characterization of the elemental composition of the analyzed volume. Scanning electron microscope (SEM) analysis was performed using a Hitachi TM3000 SEM fitted with a Bruker X-ray energy dispersive spectrometry (EDS). The two modes of operation in SEM analysis utilized here were backscattered electron imaging (BSE) and energy dispersive x-ray EDX. Biochar particles used were in the size range 150 µm– 850 µm. Prior to analysis samples were spread onto double-sided carbon tape and mounted on a SEM stub.

Cation exchange capacity. Cation exchange capacity measurements were performed in University of Santiago de Compostela, Spain, by the summation method of the exchangeable base cations of Ca, Mg, Na, K, and Al. NH₄Cl 1M (25ml) was added to the biochar sample (5g) and shaken manually before being left to stand overnight (16 hours). The following day, 75 ml of NH₄Cl 1M was added and then filtered using quantitative, low ash filter paper. (Peech et al., 1947). Ca, Mg and Al were measured by PerkinElmer PinAAcle 500 Atomic Absorption Spectrometer and Na, K, were measured by Atomic Absorption Spectophotometer with an Emission Flame.

Statistical analysis

Electrical conductivity, pH, moisture content, ash content, CEC, surface area and pore volume measurements were all performed in triplicate and experimental uncertainty given by standard deviation.

Proximate analyses EC, pH and elemental analyses

All biochars collected from the pyrolyzer had high ash contents (Nicholas, 2022). Warangal biochar (WGL_BC) recorded the highest ash content at 88.3% and Narsapur biochar (NSP_BC) and Wai biochar (WAI_BC) had lower ash contents at 67.0% and 62.3% respectively (Table 1.). Warangal biochar (WGL_BC) also had the lowest moisture content at 0.98% in comparison with 2.15% and 3.08% for Narsapur biochar (NSP_BC) and Wai biochar (WAI_BC) respectively. Measured pH values were high for all three biochars (11.86 – 12.45). The measured high ash content is consistent with the literature (Gold et al., 2018; Koetlisi & Muchaonyerwa, 2017; Liu et al., 2014). The initial feedstock of sewage sludge is high in ash and sewage sludges have been found to contain very high concentrations of Si (19–58%), Ca (5.1–7.4%), and P (3.4–4.9%) (Zielińska et al., 2015). Ash content of faecal sludge is also high and has been measured at 17.0 wt.%, significantly higher than measured ash content of sawdust at 0.8% (Liu et al., 2014). Ash content of faecal sludge has been found to be higher than that of sewage sludge (Koetlisi & Muchaonyerwa, 2017). Several reasons for this high ash content have been suggested including loss of volatile solids from latrine waste due to long storage times in the latrine (Zuma et al., 2015), ingress of grit and sand from poorly lined containment structures (Niwagaba et al., 2014) and in community toilets a higher rate of disposal of polluting waste can lead to higher ash contents (Barani et al., 2018).

Increases in pH due to increases in ash content in biochars derived from sewage sludge feedstocks have been previously reported (Hossain et al., 2011; Liu et al., 2014). The general alkaline character of biochar likely results from the increase in quantities of alkali salts (Na, K) and salts of alkaline elements (Ca, Mg) during the pyrolysis process (Singh et al., 2010).

WGL_BC recorded the most alkaline pH value (12.45) and the largest EC value (Table 1.) which is due to the higher ash content recorded for WGL_BC (Rehrah et al., 2014). There could be several reasons why the WGL biochar had a significantly higher ash content. Digestion during storage in onsite sanitation technologies can play a part in the high ash content of FS biochar (Gold et al., 2018) as well as contamination of FS by sand and grit caused by poorly lined containment structures (Niwagaba et al., 2014). A recent study investigating biochar from the same treatment facilities in India observed that sintered mineral depositions had to be removed from the reactor on a weekly basis (Krueger et al., 2020). Therefore, it is likely that ash concentrations from biochars from these types of treatment plants will fluctuate over time.

A high ash content of biochar could be useful with regards to its end-use a soil amendment. Increased crop growth with a highly alkaline (12.1), high ash biochar treatment of acidic soil has been previously reported (Smider & Singh, 2014) The authors deemed this was a result of the release of nutrients from the ash in the biochar itself and the biochar’s liming effect. It has been proposed that this liming effect is one of the main processes influencing the enhanced plant growth seen on biochar addition to soils (Jeffery et al., 2011). Altering soil pH is one of several mechanisms by which biochar can improve soils and increase agricultural productivity. Therefore, highly alkaline biochars could be of benefit to acidic soils are responsible for the severe limitation of crop agriculture worldwide. Currently only a small fraction of acidic soil is used for arable crops globally but approximately 50% of the earth’s potential arable lands are acidic (von Uexküll & Mutert, 1995).

The elemental composition (Table 1.) shows a relatively low percentage of carbon within the samples, 21–23% for NSP_BC and WAI_BC and a very low 8% for WGL_BC which is consistent with the measured ash content. Pyrolysis generally concentrates carbon in the biochar with an increase in C content relative to the feedstock frequently reported. However, most studies on sewage sludge (SS) –derived biochar show a decrease in the percentage of C in the final product relative to the feedstock (Agrafioti et al., 2013; Khan et al., 2013). FS- and SS–derived biochars generally have low total C concentrations in comparison with cellulose derived biochars (Tomczyk et al., 2020). This is due to the high ash content in the original feedstock of faecal and sewage sludge. The measured carbon concentrations in these biochars are consistent with carbon contents reported in the literature for faecal sludge biochar 27.4– 34.9% (Gold et al., 2018), 17.2 – 34.1% (Krueger et al., 2020), 6.5 – 11.1% (Koetlisi & Muchaonyerwa, 2017), and 19.5% (Woldetsadik et al., 2018).

Fourier transform infrared (FTIR) spectroscopy

FTIR spectra indicated that all three sludge biochars have a complex chemical bond structure with both organic matter and mineral compounds evident within the biochar. The FTIR spectra of all three biochars were similar with the same functional groups present on the surface; an indication of the homogenous nature of faecal sludge (Figure 3).

Figure 3. Fourier-transform infrared spectra (FTIR) of the three biochars NSP, WGL and WAI with dotted lines representing the main absorption (cm^-1) peaks.

High ash content in sludge-derived biochars leads to a high mineral content with bands arising on the spectrum at similar wavenumbers to organic functional groups. For example, a broad peak in the 1000–1200cm^-1 region can arise due to several functional groups such as inorganic and organic silicon, phosphorus compounds, as well as C-O stretching and sulphate groups (Coates, 2004)

Low intensity peaks evident in the 3800cm⁻¹ –3600cm⁻¹ region relate to OH group vibrations within mineral matter (Hossain et al., 2011) which indicates the presence of clay type compounds within the biochar (Table 2). Two peaks at 2980cm^-1 and 2890cm^-1 indicate asymmetric and symmetric aliphatic ν(CH) from terminal –CH₃ groups respectively (Socrates, 2001). However, these CH bands disappear at high temperatures due to demethylation and dehydration (Zhang et al., 2015) therefore in biochar pyrolyzed at 550 – 750°C the peaks are negligible.

Small peaks in the 2700–2100 region could be due to P-OH groups in phosphorus acids and esters which produce one or two broad bands (Stuart, 2004).

A peak at 1424cm⁻¹ in the biochar spectra (Figure 3.) corresponds to asymmetric stretches of carbonate groups, which correlates with the small peak at 874 cm⁻¹ due to the out-of-plane bending for CO₃²⁻ (Zhao et al., 2013). This could indicate the presence of calcite (calcium carbonate) in the sample. The presence of carbonate was verified as the FTIR spectrum of the acid washed biochar showed no clear peaks at 1424 cm⁻¹ or 874 cm⁻¹ confirming that acid washing removed carbonates from the sample (Figure 5.).

Another interesting difference between the ashed (Figure 4.) biochars and deashed biochars is a broad trough between 3400cm^-1 and 2500 cm^-1. This implies the presence of O-H in carboxylic acids however there is only a very weak intensity peak at ~1700 cm^-1 which could correspond to C=O in carboxylic acids. Other possible groups responsible for peaks within the 3400cm^-1 and 2500 cm^-1 region include ν(OH) from sorbed water and hydrogen bonded OH (Keiluweit et al., 2010).

Figure 4. FTIR spectra of ashed NSP biochar, ashed WAI biochar, and ashed WGL biochar.

Figure 5. FTIR spectra of deashed (acid washed) NSP, WAI and WGL biochar.

The low intensity peak in biochar between 1540 and 1650 could be indicative of C=O stretching vibrations for amides (Calderón et al., 2006), aromatic C=C stretching and carboxylate anion vibrations (Deacon & Phillips, 1980). The peak in the deashed biochar at 1580 cm⁻¹ to 1600 cm⁻¹ is indicative of a carboxylate ion, the conjugate base of a carboxylic acid (Deacon & Phillips, 1980; Ellerbrock & Gerke, 2021).

This peak was not evident in the ashed biochar (Figure 4.). It’s been suggested that a reduction in inorganics by acid demineralization allows previously hidden carbon to emerge so increasing the amount of acidic functional groups (Lou et al., 2011). In the ashed biochar there are very visible peaks ~1450cm^-1 indicative of a carbonate stretch (CO₃ ^2-) whereas as the peaks in the acid washed samples are much less visible indicating some carbonate salts within the ash content have been successfully removed by acid demineralization.

The very broad band in the range 1200–970cm^-1 is indicative of several functional groups. Inorganic and organic silicon and phosphorus compounds, as well as carbohydrates and sulphates can contribute to this broad peak (Wen et al., 2007). Sewage chars are known to contain high phosphorus levels suggesting that the peaks observed in 1200–950cm^-1 band arise from P containing functional groups such as asymmetric and symmetric stretching of PO₂ and P(OH)₂ in phosphate (Jiang et al., 2004). Si-O asymmetric stretching could also be present between 1000–1100cm^-1 (Falaras, 1999) as well as symmetric C-O stretching of ethers.

A peak at 462–464 cm⁻¹ evident in both biochar and acid washed biochar is indicative of bending vibration of Si-O-Si (459–463 cm⁻¹) (Qian & Chen, 2013). In the ashed biochar this peak seems to shift to a lower wavenumber 456cm^-1. It is possible the signals at 462–464 cm⁻¹ relate to bending vibration of Si-O-Si (459–463 cm⁻¹) and the signal at lower wavelength in the ashed biochar at 452cm⁻¹ relates to Si-O rocking (Shahrokh Abadi et al., 2015). A weak intensity signal at 1984 cm^-1 is evident in the ashed biochar but not in the deashed samples. This signal could indicate metal – carbonyl bonds, typically terminal M-CO bonds occur at 2125 - 1850 cm^-1. A quartz doublet at 796cm^-1 and 780cm^-1 is evident in the ashed biochar sample (Farmer, 1974).

There are more signals recorded in the 900-400 cm^-1 region for the ashed biochar than the deashed biochar which relate to clay minerals associated with biochar. Bands below 600 cm^–1 can be caused by stretching inorganic compounds such as KCl and CaCl₂ (Hossain et al., 2011).

The oxygen containing functional groups (OCFGs) present on biochars surface such as C=O groups determine its cation exchange capacity (CEC) (Banik et al., 2018). It is this property that enables biochars to adsorb cationic nutrients such as NH₄⁺, Ca²⁺, K⁺ within the soil and increases soils nutrient retention capability. The lack of C=O groups present in WGL_BC could affect its ability to retain nutrients and therefore its suitability as a soil amendment.

Surface area

The shape of the isotherms indicate a Type II isotherm, however, Type II isotherms are generally typified by a lack of hysteresis and no saturation at P/P₀ near to 1; typical of nonporous and macroporous adsorbents (Thommes et al., 2015). A deviation from a true Type II isotherm can be described as a pseudo-type II isotherm. These isotherms are associated with delayed capillary condensation due to the small degree of pore curvature and non-rigidity of the aggregate structure of the adsorbent. (Sing & Williams, 2004).

Hysteresis is present in all isotherms and can be classified as either type H3 hysteresis loop or type H4 hysteresis loop according to International Union of Pure and Applied Chemistry (Thommes et al., 2015). Hysteresis is caused by capillary condensation and is typical of mesoporous materials. H3 and H4 loops do not tend to close until equilibrium pressure is at or close to saturation pressure. H3 type is typical for loose aggregates of plate-like particles and in porous materials typical of pore networks containing macropores not entirely filled with condensate. H4 type loops suggest presence of slit-shaped pores including pores in the micropore region and plate-like particles with spaces between the parallel plates (Mokaya & Jones, 1995) and are common with activated carbons. H4 hysteresis loops are commonly observed with more complex materials consisting of both micropores and mesopores.

Adsorption and desorption N₂ isotherms for all biochars (Figure 6.) showed low surface areas of between 3.52 – 12.07m²g^-1 (Table 1) consistent with results reported in the literature for sewage sludge biochars which have low surface areas due to high ash content (Agrafioti et al., 2013; Bagreev et al., 2001; Schimmelpfennig & Glaser, 2012). It has been postulated that high ash contents reduce surface area by filling or blocking access to the biochar micropores (Song & Guo, 2012).

Figure 6. N₂ adsorption and desorption isotherms of WAI, WGL and NSP biochars (P/P₀= Relative pressure, V (ADS) cc/g = Volume of adsorption cc/g).

The low nitrogen uptake of all three biochars can be characteristic of materials with small ultra-micropores that are close to the kinetic diameters of nitrogen, since molecules cannot overcome the activation energy for passing through the pores at cryogenic temperatures (Kim et al., 2011). To investigate this potential microporosity further CO₂ adsorption isotherms at 273K were recorded for the three biochar samples (Figure 7).

Figure 7. CO₂ adsorption and desorption isotherms for WAI, WGL and NSP biochars (P/P₀= Relative pressure, V (ADS) cc/g = Volume of adsorption cc/g).

The CO₂-based BET specific surface areas (S_BET, μSSA) and pore volume (μPV) values were significantly larger than the N₂-derived BET specific surface area (S_BET) and pore volume (TPV) values signifying that kinetic limitations with N₂ physisorption were present for all biochars and there is some degree of microporosity present.

NSP BC showed the largest surface area measured with CO₂ and the lowest with N₂ indicating a more microporous structure whereas WGL BC had the highest N₂ SSA and lowest CO₂ µSSA signifying a slightly less microporous and more mesoporous structure. The greatest pore size distribution at pore diameters 4–15Å was recorded for NSP_BC also indicating it had more of a microporous structure than WGL_BC which recorded relatively sparse pore size distributions in this region (Figure 8b). In the mesoporous region, (16–150 Å), WGL_BC pore size distributions were much greater than both NSP_BC and WAI_BC confirming WGL_BC has a more mesoporous structure (Figure 8a).

Figure 8.

Pore volume weighted pore size-distribution derived from a) N ₂ (mesopore region 20–500Å) and b) CO₂ (micropore region<20Å) for NSP, WAI and WGL biochars.

The values obtained demonstrate the complex pore network within the biochar, even though the surface area values are generally low compared to other biochars there is still a degree of both microporosity and mesoporosity within the biochars. Low surface area biochars may be unsuitable for use as soil amendments as the water holding capacity is relatively low and the low porosities are not conducive to promoting soil microbial growth (Ishii & Kadoya, 1994; Thies & Rillig, 2009), which play an important role in nutrient cycling (Lambers et al., 2008). The surface area could be increased by increasing the pyrolysis temperature (Song & Guo, 2012; Tomczyk et al., 2020). The fast pyrolysis of municipal sludge biochar at temperatures 500 – 900 °C showed that increasing temperatures resulted in a greater microporous network within the biochar (Chen et al., 2014). Previous work has shown that the greatest enhancement of sewage sludge biochar porosity occurred between 400 – 600°C (Bagreev et al., 2001). However heavy metal concentration in biochars generally increase with pyrolytic temperature (Lu et al., 2013). This is because heavy metals do not volatilize, so their concentration within the biochar increases with pyrolysis temperature (Chanaka Udayanga et al., 2019; Hossain et al., 2011; Wang et al., 2021).

The larger surface areas (with CO₂) and more microporous structure of NSP_BC and WAI_BC relates to their lower ash content of 67.0% and 62.3% respectively. The lowest surface area measured (with CO₂) was that of WGL_BC which recorded the highest ash content of 88.3%. These findings support the notion that lower surface areas relate to ash filling or blocking access to the biochar micropores (Song & Guo, 2012).

Zeta potential

Zeta (electrokinetic) potential signifies the net charge between the surface plane and slip plane of a colloidal particle (Hiemenz & Rajagopalan, 1997). Zeta potential values yield information about the external surface charges of biochar particles in solution and indicates the sorption and nutrient holding characteristics of the biochar in soil. Negatively charged surfaces are unlikely to sorb negatively charged ions such as phosphate but are more likely to sorb positive cations such as heavy metal ions and ammonium ions.

The zeta potential values for all three biochar samples were negative in the pH range 5.0–9.5, revealing that negative charges are carried on the surface of the biochar particles (Figure 9). FTIR spectra revealed the existence of oxygen containing functional groups (–COO⁻ and–OH) on the biochars surface which can contribute considerably to surface charge of the biochars. The negative zeta potentials of all three biochars in the pH range 5.0 – 9.5 support this interpretation.

Figure 9. Zeta potentials of WGL_BC, WAI_BC and NSP_BC at pH values from 5-9.5.

At acidic pH, the zeta potentials of the biochar samples became less negative, indicating that the association of –COO⁻ and –O⁻ with H⁺ reduced the negative charge of the biochars. With increasing pH, the zeta potential of WAI_BC and NSP_BC biochars become more negative due to increasing deprotonation of the biochar surface functional groups (Yuan et al., 2011). However, at pH above 7 there was an increase in zeta potential for WGL_BC from -23.4mV to -7.7mV indicating a decrease in negative surface charge. WGL_BC contains the highest ash content of all the biochars (Table 1), and it is likely this that contributes to the increase in zeta potential values at pH>7. The mechanism by which surface charge increases at high pH values cannot be explained by deprotonation of the surface functional groups in the case of WGL_BC. The higher ash content indicates some other mechanism occurring. Zeta potential of fine coal tailings containing several ash-forming minerals showed a similar trend which the authors attributed to the presence of alumina and silicate particles, which result in lower negative zeta potential values. They also noted that varying zeta potential values at high pH could be attributed to the binding of more cations such as Ca²⁺ (Kumar et al., 2014). Positively charged calcium monohydroxide ions on the biochar surface would to some degree neutralize the negative surface charges resulting in less negative zeta potential values. (Liu et al., 2002). It is possible that at pH values >9 WGL_BC would have positive zeta potential values and thus more likely to sorb negatively charged ions such as nitrate or phosphate.

X-ray diffraction (XRD)

X-ray diffraction (XRD) analysis of the biochars revealed that mineral components in the crystal form were present in all three biochars (Figure 10). Quartz was identified as the predominant crystalline phase with the highest peak at 2θ around 26.6° (d = 3.33 Å) in NSP and WGL biochars. WAI biochar exhibited a more intense peak relating to CaSO₄ (anhydrite). Quartz, sylvite, calcite, calcium sulphate, albite were the most common phases identified. These minerals are formed during pyrolysis due to a reaction between CO₂ and alkaline-earth metals and alkaline oxyhydroxides.

Figure 10. XRD patterns of NSP, WAI and WGL biochar (Qu= Quartz, Al=Albite, Ca=Calcite, An = Anhydrite, Sy = Sylvite, Wh = Whewellite, Ank=Ankerite, Th= Thermonatrite).

Previous research has shown sewage sludge biochar to have a turbostratic structure where the carbon fraction is dominated by disordered graphitic crystallites (Srinivasan et al., 2015; Uchimiya et al., 2011). This is in discordance with the XRD results which show a lack of C (002) diffraction peaks (2θ = 15-30°) and C (101) diffraction peaks (2θ = 40-50°) due to amorphous carbon structures and graphite structures respectively. The biochars studied here do have a very high ash content and the lack of these peaks could be as a result of interference of high-intensity quartz peaks. Studies have shown that the high content of minerals, specifically quartz can affect the structural characterization of biochar carbon fraction (Feng et al., 2015).

The difference in mineral composition between the three biochars could be due to possible contamination of FS by sand and grit caused by poorly lined containment structures (Niwagaba et al., 2014). The containment structures at each location would have to be investigated to reach a definitive conclusion. Overall, the mineralogical composition of the biochars is in agreement with their high ash contents.

Scanning electron microscopy (SEM) with energy dispersive x-ray analysis (EDX)

SEM analysis reveals a complex porous structure evident in all biochars (Figure 11). The porous structure of biochars strongly resembles the cellular structure of the original feedstock (Fuertes et al., 2010; Yao et al., 2011). In the case of faecal sludge, cellular macroporous structures arise from undigested fibrous vegetable matter. The morphology of the biochar is honeycomb-like with cylindrical and slit like holes clearly observable. This porous structure can provide a specialized environment for the colonization of microbes (Thies & Rillig, 2012). This increase in mycorrhizal fungi contributes to increased mineralization of recalcitrant soil organic matter, ultimately improving soil and plant health (Anderson et al., 2011; Zimmerman et al., 2011). SEM images also show all three biochars have high ash content with EDX confirming the presence of mineral elements (Figure 12). The SEM images clearly showed a high presence of clay mineral particles/ash (white/grey) with a smaller amount of biochar particles present (black).

Figure 11.

SEM micrograph of (a) Original WGL biochar, (b) Original NSP biochar, (c) Original WAI biochar.

Figure 12. SEM-EDX map for all elements distribution across the area highlighted in image and associated energy dispersive X-ray (EDX) quantification of biochar.

Visually WGL biochar had a higher ash content which is concurrent with the ash percentage from proximate analyses. EDX results on the biochar particles themselves revealed high volumes of carbon and oxygen (Figure 12). Also present were silicon, calcium aluminium, potassium, magnesium, phosphorus, and sodium all of which are beneficial to plant health.

Cation exchange capacity

Cation exchange capacity (CEC) enables biochars to adsorb cationic nutrients such as NH₄⁺, Ca²⁺, and K⁺. It is thought this characteristic of biochar results predominantly from formation of carboxylic functional groups during oxidation (Cheng et al., 2006).

There was a large variation in CEC values with WGL biochar (WGL BC) the highest CEC at 129.3 cmolKg^-1 and NSP biochar the lowest CEC at 41.9 cmolKg^-1 (Table 1). Fresh biochars from lignocellulosic biomass generally have lower CEC, with manure-based biochars exhibiting higher CEC values (Tag et al., 2016). In the literature CEC values for biochar are highly variable, commonly ranging from 6 cmol₍₊₎ Kg⁻¹ (Munera-Echeverri et al., 2018) to 36.3 cmol₍₊₎ Kg⁻¹ (Song & Guo, 2012) to as high as 304 cmol₍₊₎ Kg⁻¹.

Yuan et al. (2011) proposed that high ash content biomass creates high CEC biochars and that K, Na, Ca, Mg, and P in the feedstock would promote formation of O-containing acidic functional groups such as carboxylic, and phenolic groups on biochar surface during pyrolysis and thus, result in higher CEC (Gaskin et al., 2008). However, FTIR analysis showed a lack of acidic functional groups such as phenolic groups in these biochars. It is possible that the high ash content of these biochars could contribute to methodological problems in determining CEC (Graber et al., 2017). There is a large range of CEC values reported in the literature and measurements are often poorly reproducible (Munera-Echeverri et al., 2018). FTIR shows that there are carbonates and silicates present in these biochars which would result in the release of base cations and interference with the sum of measured exchangeable base cations (Munera-Echeverri et al., 2018). WGL biochar records the highest CEC value (129.3 ± 2.3 cmol.kg^-1) and the highest ash content of all three biochars implying that it is the high ash content that is responsible for the high CEC value.