Biochar production

Rice husks obtained from a commercial mill in Cuautla, Morelos, from the spring-summer 2019 harvest, were used to make the BC. The technique used to make it was hydrothermal carbonization (HTC) at 200 °C with 10 % citric acid as a catalyst (^{Velázquez-Maldonado et al., 2019}).

Treatments

The individual materials used were rice husk biochar (BC) and commercial peat moss (T; Sunshine mix 3), which were mixed under different proportions: 20:80, 40:60, 60:40 and 80:20 % (v/v), respectively. The treatments were designated as: T1 = peat moss (control), T2 = BC, T3 = 20:80 % mix, T4 = 40:60 % mix, T5 = 60:40 % mix and T6 = 80:20 % mix.

Physical characterization

Granulometry. In an electric sieve shaker (MONTINOX®), with sieves (FIC®) of number 8, 10, 12, 16, 20 and 50 (2.38, 1.68, 1.41, 1.15, 0.86 and 0.24 mm opening, respectively), a sample composed of 800 cm³ of each treatment was placed for 3 min; subsequently, the material retained on each sieve was weighed and the percentage by particle size was calculated.

Bulk density (BD). For this determination, 232-mL polystyrene permeameters were used. Samples were saturated with running water for 24 h, placed in the permeameters and dried in an oven at 65 °C until constant weight (^{Gayosso-Rodríguez, Villanueva-Couoh, Estrada-Botello, & Garruña, 2018b)}.

Total porosity (TP), aeration porosity (AP) and water-holding porosity (WHP) were determined using the procedure described by ^{Landis, Tinus, McDonald, and Barnett (1990)}.

Nutrient concentration

N was determined for the individual materials (biochar and peat moss) by the micro Kjedhal method and P with the vanadate-molybdate-yellow method. Total K, Ca and Na content was obtained by means of the flamometry technique, and total Mg by atomic absorption spectrophotometry according to the official Mexican standard PROY-NOM-021-RECNAT-2000 (^{Diario Oficial de la Federación [DOF], 2000}).

Seedling growth

The experiment was carried out in June 2019 in a tunnel-type greenhouse belonging to the Faculty of Agricultural Sciences of the Universidad Autónoma del Estado de Morelos, located in Cuernavaca, Morelos, Mexico (18° 58’ 51” N and 99° 13’ 55” W, at 1,866 m a. s. l.). Temperature and relative humidity were monitored with an environmental data logger (U12, HobWxo^®). The average greenhouse temperature was 28.5 °C and the average relative humidity was 65 %. Treatments were evaluated in American-type ‘Thunderbird’ (Seminis^®) cucumber seedlings in 200-cavity polystyrene germination trays, with each cavity having a capacity of 20.5 mL. Seedlings were irrigated with water purified by reverse osmosis.

At 23 days after planting, stem length (ST), fresh weight of aerial biomass (FWAB), fresh weight of root biomass (FWRB), leaf area (LA), dry weight of aerial biomass (DWAB), dry weight of root biomass (DWRB) and relative chlorophyll content were recorded. FWAB and FWRB were obtained with a scale (Ohaus®) and LA with a leaf area meter (LI-3100C, LI-COR®, USA). For DWAB and DWRB, a circulating air oven (Pro1002498, Luzeren^®) was used at 70 °C until constant weight, and a portable SPAD meter (502 Plus, Minolta^®) was used to determine chlorophyll content (SPAD readings).

Experimental design and statistical analysis

In the first stage, a completely randomized design with three replicates was used to determine the physical and chemical properties of the substrate mixtures evaluated. In the second stage, a randomized block experimental design with six replicates and ten seedlings as the experimental unit was used. To ensure normality, the data expressed as a percentage were transformed with the square root of the arcsine. With the exception of the data on substrate physical property variables, data were subjected to analysis of variance and, when there were statistical differences, a Tukey's comparison of means test (P ≤ 0.05) was performed. Likewise, to determine if the physical properties of the substrates had a relationship with seedling growth, a Pearson correlation analysis was performed with 18 pairs of values and significant correlations (P ≤ 0.05) were reported, this by means of the SAS program (^{SAS Institute, 2004}).

Physical properties

In relation to the cumulative percentages of particle size up to 0.86 mm, T3 (20:80 mix) presented the highest value (63.31 %), while T1 (peat moss) had the lowest value (41.44 %) (Table 1). In sizes from 0.86 to 2.38 mm, the highest distribution of particles was concentrated in T2 (BC) with 55.26 %, and T3 had the lowest distribution (32.38 %). The highest percentage of particles larger than 2.38 mm was found in T1 (7.74 %), and the lowest value was in T2 (0.50 %). It is important to note that reports on BC particle size are scarce; however, the values of the present study contrast with those reported by ^{Pérez-Salas, Tapia-Fernández, Soto, and Benjamin (2013)} in beechwood (Gmelina arborea) biochar, since they obtained a greater distribution of particles from 0.24 to 0.84 mm with 57 %, followed by 23 % of particles larger than 2 mm and 20 % of particles smaller than 2 mm. This may be due to the composition of the plant material used and the biochar production process.

^{Cabrera (1999)} states that the components of substrates or mixtures should be made up of particles with sizes from 0.5 to 4 mm, with a percentage ≤ 20 % for sizes ≤ 0.5 mm, ≥ 60 % in sizes from 0.5 to 2 mm and ≤ 20 % in sizes > 2 mm. ^{Gayosso-Rodríguez, Borges-Gómez, Villanueva-Couoh, Estrada-Botello, and Garruña (2018a)} point out that percentages higher than 20 % in particles with sizes ≤ 0.5 mm affect the aeration capacity in substrates because it decreases with particle size.

In general, it was observed that the particle distribution was affected by the combination of both materials. That is, as the proportion of BC decreased, the cumulative percentages of particles < 0.86 mm and > 2.38 mm increased, while particles in the range of 0.86 to 2.38 increased as the proportion of BC increased.

TP increased with decreasing BC content in the mixture. Treatments T1, T3 and T4 were significantly different (P ≤ 0.05) from the rest of the treatments in TP and WHP (Table 2). Although no statistical differences were observed among T5, T6 and T2, the last presented the lowest values ​​of TP and WHP (76.25 and 63.23 %, respectively). The results of the present study were higher than those reported by ^{Webber, White, Spaunhorst, Lima, and Petrie (2018)}, who in mixtures of peat moss (Sun Gro Horticulture) and sugarcane bagasse biochar (25:75, 50:50 and 75:25 %) observed that the pore space ranged from 73.13 to 76.79 %, obtaining the largest pore space with the lowest biochar mixture. ^{Webber, White, Spaunhorst, and Petrie (2017)} report that the pore space ranged from 59.98 to 64.56 % with mixtures of peat moss (Sun Gro Horticulture) and sugarcane bagasse ash (25:75, 50:50 and 75:25). Regarding AP, the results ranged from 12.87 to 15.62 %, with no significant differences (P ≤ 0.05) among treatments.

In relation to TP in organic substrates, ^{Morales-Maldonado and Casanova-Lugo (2015)} state that it should be greater than 85 %. In this study, treatments T3 and T4 comply with that recommendation, which generates a balance between the water-air ratio (AP and WHP). On the other hand, treatments T1, T3 and T4 had significant differences in WHP compared to T2 and T6. In this regard, ^{Webber et al. (2017)} point out that decreasing the percentages of sugarcane bagasse ash mixed with peat moss (Sun Gro Horticulture) (75:25 % peat moss:ash) results in the highest pore space, water saturation and field capacity values. In contrast, ^{Webber et al. (2018)} report that the 75:25 % peat moss:sugarcane bagasse biochar mixture increased pore space, but decreased water saturation and field capacity properties in the substrates.

Water holding in a substrate is not only determined by particle size, but also by the arrangement, shape and compaction of the particles, since they generate different types of pores (^{Gutiérrez-Castorena, Hernández-Escobar, Ortiz-Solorio, Anicua-Sánchez, & Hernández-Lara, 2011}). For this reason, particles > 1 mm favor the formation of larger pores (^{Morales-Maldonado & Casanova-Lugo, 2015}), and, in this study, peat moss was the material with the highest amount of particles > 1 mm, so increasing the proportion of peat moss in the mixture also increased the TP. Large pores allow the accommodation of small intra- and interparticle particles, which generates pores that contribute to water conservation (^{Anicua-Sánchez et al., 2009}).

Most biochar-related research focuses on moisture content, ash, fixed carbon, volatility, and surface area as physical properties (^{Ding et al., 2017}; ^{Herrera et al., 2018}; ^{Rodríguez, Lemos, Trujillo, Amaya, & Ramos, 2019}). However, ^{Webber et al. (2018}, ²⁰¹⁷⁾ report physical properties such as total porosity, water saturation and field capacity in sugarcane bagasse biochar and sugarcane bagasse ash mixed with peat moss (Sun Gro Horticulture) at different percentages (0, 25, 50, 75 and 100 %).

Regarding BD, differences (P ≤ 0.05) were found among treatments; T2 and T6 had the highest BD with 0.20 g·cm^-3, while peat moss (T1) showed the lowest value with 0.11 g·cm^-3. ^{Pratiwi, Hillary, Fukuda, and Shinogi (2016)} obtained results of approximately 0.18 g·cm^-3 in rice husk biochar. On the other hand, ^{Alburquerque et al. (2014)} report a BD of 0.19, 0.25, 0.66, 0.72 and 0.74 g·cm^-3 in different biochars made from wheat straw, pine woodchips, olive-tree pruning, olive stone, and almond shells, respectively. ^{Webber et al. (2018)}, in sugarcane bagasse biochar, obtained a low BD (0.11 g·cm^-3); however, peat moss (Sun Gro Horticulture) had a BD of 0.11 g·cm^-3, equal to that found in this study. It is important to consider BD since, in addition to the effect it can have on plant growth, it can result in increased transportation and handling costs (^{Cabrera, 1999}). A low BD is desirable to facilitate handling and transport of germination trays (^{Bracho, Pierre, & Quiroz, 2009}).

Nutrient concentration

Differences (P ≤ 0.05) were found in the nutrient concentration of peat moss and BC in N, K, Ca, Mg and Na (Table 3). It was observed that the highest concentration of nutrient elements was found in peat moss, with the exception of P, which had no significant differences. These results may be due to the chemical composition of peat moss, since agricultural dolomite is added to it, which provides Ca and Mg and increases the availability of nutrients such as N, P and Ca (^{Calva & Espinosa, 2017}).

The nutrient content of the BC was higher than that reported by ^{Velázquez-Maldonado et al. (2019)} (with values of 0.32 % N, 504 mg·kg^-1 P, 1,117 mg·kg^-1 K and 983 mg·kg^-1 Mg), except for Ca (10,988 mg·kg^-1). Although the rice husks used in both studies were extracted from the same region of Cuautla, they were harvested in different years, which may have influenced the nutrient content of the biochar. On the other hand, the results obtained were lower than those reported by ^{Cho et al. (2017)}: 12,050 mg·kg^-1 P, 15,800 mg·kg^-1 Ca, 10,380 mg·kg^-1 Mg and 7,340 mg·kg^-1 Na, except for the content of N and K. In this case, the biochar was made using a wood roaster at temperatures ranging from 200 to 250 °C.

The differences in the nutrient concentrations of the different biochars could be influenced by the technique and temperature used in their preparation. In this regard, ^{Bethancourt, James, Villarreal, and Marin-Calvo (2019)} state that by increasing the temperature from 714 to 935 °C, in the gasification technique, the nutrient concentration of biochar increased from 0.30 to 0.50 % in N, from 6,000 to 10,000 mg·kg^-1 in P, from 8,000 to 10,000 mg·kg^-1 in K and from 236.7 to 524.0 mg·L^-1 in Mn. Therefore, it can be said that the properties of biochars are also affected by production techniques, raw materials, heat ranges, temperature, reactor pressure and the use of catalysts (^{Bento et al., 2019}; ^{Escalante-Rebolledo et al., 2016}; ^{Huang & Gu, 2019}).

Seedling growth

The growth of cucumber seedlings in treatments T1 and T3 was higher (P ≤ 0.05) than in the rest of the treatments (Table 4), with an increase of 81.65 and 84.81 % in SL, of 136.61 and 119.44 % in FWAB, of 106.96 and 105.90 % in LA, and of 166.99 and 145.08 % in DWAB, respectively. On the other hand, seedlings under the T2 treatment presented the lowest values in all variables evaluated.

^{Cho et al. (2017)} found greater seedling height and higher root, stem and leaf dry weight in Zelkova serrata with 20 % rice husk biochar mixed with soil plus fertilization. These authors attributed their results to the physical and chemical properties of the biochar. ^{Araméndiz-Tatis, Cardona-Ayala, and Correa-Álvarez (2013)} report, in eggplant seedlings, that when using three mixtures of raw rice husk, in different combinations with alluvium, vermicompost and poultry manure (50:50:0:0, 40:40:20:0 and 40:40:0:20 %, respectively), the treatments with the highest rice husk proportion recorded the lowest values in growth and biomass production variables.

In DWRB, the highest value was obtained with treatment T3 (51 mg) and the lowest value with T2 (23.67 mg). These results may be due to the physical and chemical properties of BC (Tables 1, 2 and 3). Regarding SPAD units in leaves, no significant differences were found.

Correlation analysis

Table 5 summarizes the variables with the greatest association, where positive correlations (P ≤ 0.01) were detected between TP-SL, TP-FWAB, TP-LA, TP-DWAB, WHP-SL, WHP-FWAB, WHP-LA and WHP-DWAB. It was observed that treatments T1 and T3 had higher TP, so they had greater water-holding capacity and lower density, characteristics that favored the growth and development of the seedlings. Likewise, a positive correlation (P ≤ 0.05) was detected between AP-DWAB.

On the other hand, negative correlations (P ≤ 0.01) were detected in BD-SL, BD-FWAB, BD-LA, BD-DWAB and BD-DWRB (Table 5). Treatments T1 and T3 had low values in BD, but produced plants with greater SL, FWAB, DWAB and DWRB. This indicates that a BD of 0.20 g·cm^-3 has a negative impact on plant growth and development. In this regard, ^{Gayosso-Rodríguez, Borges-Gómez, Villanueva-Couoh, Estrada-Botello, and Garruña-Hernández (2016)} point out that the density of a substrate is diverse, and that porosity and water movement depend on it.