5–99.7 mg) and measured on an FTIR spectrometer (VECTOR 22, Bruker, USA) with a 4 cm− 1 resolution and 100 scans between wavenumbers of 4000 and 400 cm− 1 (Chun et al., 2004). To analyze C forms from the FTIR spectra, we subtracted the background of the KBr window, automatically corrected the baseline and Libraries smoothed the spectra, identified the peaks, and

normalized the spectra on KPT-330 ic50 a reduced portion of the wavenumbers (4000–500 cm− 1). Kubiena boxes were used to collect undisturbed blocks of unamended and amended soils during the incubation period to make thin sections. After air drying, vertically-oriented thin sections measuring 2.5 × 5 cm and 30 μm thick were prepared by Spectrum Petrographics (Winston, OR, USA). The thin sections were used to observe soil structures under a polarized microscope (AFX-II Type, Nikon Precision Instruments, Belmont, CA). The biochar sample was viewed by optical microscopy with reflected MAPK inhibitor light and then scanning electron microscopy (SEM) (Hitachi, S-3000N, Japan) to identify its micro-scale structure. A back-scattered electron image representing the mean atomic abundance in a back-and-white image was observed from the surface of the samples coated by Au. The mineral phases of the sample were identified using SEM and energy-dispersive spectroscopy (EDS) (Horiba, EMAX-ENERGY EX-200, Japan), with 15 kV and 180 pA for the acceleration voltage and beam current, respectively, in a vacuum of 25 Pa with

an Au coating. Analyzed points were selected using back-scattered electron images to avoid damaging samples. The soil erosion experiment was conducted using simulated rainfall equipment with 9.5 m in height (drop diameter is 2.5 mm and terminal velocity is 8.5 m s− 1), and all processes followed the ASTM-D7101 standard (America Standard Testing Materials, ASTM). Soil erosion processes are widely performed in the

field and laboratory using rainfall simulators and these simulations play an important role in controlling repeatable conditions and adjusting the required rainfall intensity (Tejada and Gonzalez, 2007). The erosion experiment simulated next a rainfall intensity of 80 mm h− 1 and a 10% slope gradient because this is the average slope gradient in the field. The rainfall experiment for all the treatment was in triplicate. The triplicate data were subjected to mean separation analysis using the 1-way ANOVA test at a significance of p = 0.05. The differences between mean values were identified using Duncan’s test. Pearson’s correlation coefficients were calculated to determine how the soil properties are related. Table 1 lists the properties of the soil and the biochar. The soil was very acidic (pH < 4.0) and had low levels of total organic carbon (TOC) (4.37%) and soil organic carbon (SOC) (< 2.0%), which is typical for soils in humid tropical regions. A low CEC might be the result of low organic matter content and low clay activity in the soil.