Granulation compared to co-application of biochar plus mineral fertilizer and its impacts on crop growth and nutrient leaching

Mechanized biochar field application remains challenging due to biochar’s poor flowability and bulk density. Granulation of biochar with fertilizer provides a product ready for application with well-established machinery. However, it’s unknown whether granulated biochar-based fertilizers (gBBF) are as effective as co-application of non-granulated biochar with fertilizer. Here, we compared a gBBF with a mineral compound fertilizer (control), and with a non-granulated biochar that was co-applied at a rate of 1.1 t ha−1 with the fertilizer in a white cabbage greenhouse pot trial. Half the pots received heavy rain simulation treatments to investigate nutrient leaching. Crop yields were not significantly increased by biochar without leaching compared to the control. With leaching, cabbage yield increased with gBBF and biochar-co-application by 14% (p > 0.05) and 34% (p < 0.05), respectively. Nitrogen leaching was reduced by 26–35% with both biochar amendments. Biochar significantly reduced potassium, magnesium, and sulfur leaching. Most nitrogen associated with gBBF was released during the trial and the granulated biochar regained its microporosity. Enriching fertilizers with biochar by granulation or co-application can improve crop yields and decrease nutrient leaching. While the gBBF yielded less biomass compared to biochar co-application, improved mechanized field application after granulation could facilitate the implementation of biochar application in agriculture.


Gas adsorption of biochar and fertilizer
CO2 adsorption isotherms were recorded at 273.15 K (cryoTune 195, 3P Instruments GmbH& Co. KG, Odelzhausen, Germany) in a relative pressure range of 3 x 10 -5 to 3 x 10 -2 and at 77 K for N2 isotherms in a relative pressure range of 1.2 x 10 -3 to 9.9 x 10 -1 .
Samples were outgassed at 30 °C1 under vacuum for a minimum of 12 hours.At this temperature, outgassing mass losses were in the range of the water contents of the samples measured with Karl Fischer titration (see Table S1 for further details) and no precipitated salt was visible at the top of the measurement cell in case of NPK and gBBF, which was the case when outgassed at e.g., 90 °C.

Scanning Electron Microscopy (SEM)
For scanning electron microscopy (SEM), a gBBF granule and its cross section (after preparation with a scalpel) were mounted on an aluminum stub using adhesive carbon tape and sputtered with gold on the outer surface.Secondary electron imaging was performed in a Jeol JSM-6610LV (Joel Ltd., Tokyo, Japan).Electron dispersive X-ray (EDX) mapping was done with a 10 minute measurement time, 10 kV acceleration voltage, and a 20 mm 2 silicon drift detector (X-Max, Oxford Instruments, Oxford, United Kingdom).

Repeated extractions
For the repeated extractions, samples were weighted into 100 mL Erlenmeyer flasks (equivalent of 0.4 g of N per flask), 50 mL of a 0.0125 M CaCl2 solution were added and shaken at 125 rpm on a horizontal shaker for 1h.Subsequently, the extracts were filtered through a filter paper.The filter paper including the decanted solid fractions of biochar and fertilizer was returned into the sample flask with another 50 mL of fresh extractant.
The procedure was repeated twice, with two hours and 45 hours extraction times, respectively, summing up to a total extraction time of 48h.Liquid extracts were filtered to <0.45 µm (Chromafil PES Xtra 45/25, Machery-Nagel, Düren, Germany) and analyzed for total nitrogen (TN) content.
Extracted TN from the sample during each extraction step was calculated as follows, where

Incubation experiment
Released TN during the incubation experiment was calculated as follows: where -Nreleased,i is the mass fraction of extracted TN after sampling i in relation to the initial total N contained in the solid sample (in %), -cN,i is the TN concentration in the filtered extract from sampling i, -Vtot is the total incubation volume (=100 mL) -m is the mass of the fertilizer weighed into the bottle -ßN is the TN content in the fertilizer (w/w) -VW is the volume of de-ionized water added to the bottle after sampling (=5mL), -i is the number of total samples taken, starting with i=0.

Greenhouse trial maintenance
The pots were arranged in a randomized block design on a greenhouse table with eastwest orientation (located at 48°29'01.7"N7°56'25.9"E)with an average temperature of 22±6 °C, an average humidity of 71±24 % and a CO2 concentration of 498±51 ppm (Figure S2) and artificial lighting.In each pot, automated drip irrigation with 4 watering spots per pot was fixed (NetBow™, Netafim Ltd., Tel Aviv, Israel) which was operated daily with tap water.Once per week, the position of the pots was randomly re-arranged within each block and pots were manually weighed, watered to 65% of the maximum water holding capacity (WHC, cf. Figure S3 for results), and the automated water dose was adjusted, if necessary.Maximum WHC were measured in triplicates in 1L pots after saturation of the individual soil-biochar and -fertilizer mixtures in a water bath for 16h and drainage of surplus water on a dry sand bed for 48h, with subsequent calculation of stored water mass per mass of dry matter mixture of soil, biochar and fertilizer.Two weeks after sowing, the seedlings were reduced to one plant per pot and an additional amount of 240 g soil (dry matter equivalent) was added to each pot on top to stabilize the remaining seedlings.

Biomass analysis
The dried and milled aboveground biomass (120 mg) was digested in a mixture of Frickenhausen, Germany) using KNO3 for the calibration in the range of 0 to 200 mg NO3 -L -1 .The microtiter plates were incubated for 60 minutes at 60 °C (ThermoMixer C, Eppendorf SE, Hamburg, Germany), after the reaction liquids from the kit were added to the sample.
TN in the leachates and extracts was quantified on a TOC-VCPN equipped with the total nitrogen measurement unit TNM-1 (Shimadzu Corporation, Kyoto, Japan).The calibration curve was obtained by a mixture of ammonium sulphate and potassium nitrate in the range of 0-100 mg N L -1 (N was present to 80% as NO3-N and to 20% as NH4-N in the standard).

Supplementary figures and tables
Table S1 Water content determined with Karl Fischer (KF) titration for the different samples and mass losses during outgassing at 30°C for 16 hours before gas adsorption measurements.gBBF: granulated biochar-based NPK fertilizer.NPK: granulated mineral NPK fertilizer.An outgassing temperature of 30 °C was chosen, since a higher outgassing temperature (90 °C) lead to salt precipitates at the top of the sample cell in case of the BBF and NPK samples, which would have disturbed the sample morphology before the adsorption experiments.Measurements were performed at 77 K after outgassing the samples for 12 hours at 303.15 K.

Sample
The adsorbed volume of N2 was transformed to standard temperature conditions.

-
Nextracted,i is the mass fraction of extracted N during extraction step i in relation to TN initially contained in the solid sample (in %), -cN,i is the TN concentration in the filtered extract from extraction step i, -VE,i is the volume of extractant added to the sample for extraction step i, -VER,i-1 is the remaining volume of extractant in the flask from the extraction step before, -cN,i-1 is the TN concentration in the extractant from extraction step before, mS is the mass of the N containing solid fertilizer and -ßN is the TN content in the solid fertilizer

Figure S2
Figure S2 Greenhouse climate during the 115 days of the pot trial: Temperature (a), relative humidity (b) and ambient CO2 concentration logged in 10-minute intervals (SD800, Extech Instruments, Nashua, USA).

Figure S3 Figure S4
Figure S3 Water holding capacities (WHC) of the different soil mixtures, expressed as ratio of the dry mixture of soil and fertilizer in %.Data are presented as means ± standard deviation.Different letters above error bars indicate significant differences (one-way ANOVA, p<0.05,Tukey's post hoc test).

Figure S5
Figure S5Pots fixed on bottom-sealed pipe sockets before conduction of precipitation events for nutrient leaching from each individual pot.

Figure S6 Figure S7
Figure S6 Aboveground biomass harvested by cutting the cabbage plants below the lowest leaf base (a) and cabbage head separated from marketable, unfolded cabbage leaves (b).

Figure S8
Figure S8 CO2 adsorption isotherms of the collard-milled biochar (<1 mm) (a) the granulated biochar-based fertilizer (b) and the granulated NPK fertilizer (c, two replicated measurements).Measurements were performed at 273.15K.Samples were outgassed at 303.15 K under vacuum for 12h before measurements.The adsorbed volume of CO2 was transformed to standard temperature conditions.

Figure S10
Figure S10 Pore size distribution of collard-milled biochar (<1 mm, a), the granulated biochar-based fertilizer (gBBF, b) and the granulated mineral fertilizer (NPK, c): First derivate of the pore volume by pore width (dV/dd, red) and cumulative pore volume (black).Data is based on CO2 adsorption experiments performed at 273.15 K after outgassing of the samples at 30°C for 12 h.

Figure S11 Figure S12
Figure S11 Scanning electron microscope image of the surface of a granulated biochar-based NPK fertilizer (gBBF).The full length of the scale bar represents a length of 1 mm.'Spektrum' indicates spots manually picked for elemental analysis by electron dispersive x-ray spectroscopy.

Figure S13 Figure S14
Figure S13 Energy dispersive spectroscopy (EDX) mapping on an interface of a granulated biochar-based NPK fertilizer (gBBF).The original scanning electron microscope image and the overlay of the EDX mapping can be found in Figure S12.Mapped elements are carbon (C), oxygen (O), sulfur (S), phosphorus (P), potassium (K), magnesium (Mg) and nitrogen (N).Images were obtained with an accelerating voltage of 10 kV.

Figure S15
Figure S15 Energy dispersive spectroscopy (EDX) mapping on an interface of a granulated biochar-based NPK fertilizer (gBBF).The original scanning electron microscope image and the overlay of the EDX mapping can be found in Figure S14.Mapped elements are carbon (C), oxygen (O), sulfur (S), phosphorus (P), potassium (K), magnesium (Mg) and nitrogen (N).Images were obtained with an accelerating voltage of 15 kV.

Figure S16
Figure S16 Release of nitrogen (N) from the fertilizer granules and the mixture of the NPK fertilizer with milled biochar in non-shaken distilled water during 9 days.Error bars show the standard deviation (n= 3).Released N is presented as percentage of total N in the respective fertilizer mixtures.NPK: pure NPK fertilizer granule; gBBF: granulated biochar-based fertilizer; BC(HM)+NPK: loose mixture of hammer milled biochar (<12 mm) and pure NPK fertilizer; BC(CM)+NPK: loose mixture of collard milled biochar (< 1mm) and pure NPK fertilizer.For the loose mixture of BC and NPK, the materials were separately weighted into the incubation bottle before addition of the distilled water.

Figure S18
Figure S18 Soil urease activity in soil samples taken in redundantly prepared pots after 70 days and after harvest.For the measurement after 70 days, five redundantly prepared pots were sampled with a soil core sampler.The samples at day 116 were obtained during harvest after separation of the cabbage rootstocks from the soil.Data is presented as mean ± standard deviation (n=5).NPK: granulated NPK fertilizer.gBBF: granulated biochar-based NPK fertilizer.B+NPK: co-application f biochar and granulated NPK fertilizer.

Figure S19
Figure S19 Extractable nitrogen (N) species in the soil samples taken after 36 days (panel a) and 70 days (panel b) after sowing.The pots were redundantly prepared and on the specified days, one soil sample per pot was taken with a soil core sampler through the whole pot depth.These pots did not receive leaching events.Extractable total N, NO3 --N and NH4 + -N was measured in filtered 0.0125M CaCl2 extracts (1+4; m+V).The difference between total N and the two measured mineral N species was defined as organic N (Norg).Data are presented as mean ± standard deviation (n=5).The data were tested for significant differences by one-way analysis of variance for each N fraction individually (p<0.05,Tukey's post-hoc test, indicated by different letters above error bars).

Figure S20
Figure S20 Nitrogen use efficiency (NUE) of all fertilized plants based on the N uptake in the aboveground biomass with or without leaching events.Data is presented as mean ± standard deviation (n=5).NPK: pure NPK fertilizer.gBBF: biochar-based NPK fertilizer granule.B+NPK: Milled biochar applied to the soil in combination with pure NPK fertilizer.The nonsignificant p value was obtained from two-way analysis of variance (p<0.05).

Figure S21
Figure S21 CO2 adsorption isotherms recorded for the pristine biochar (pristine BC), the pristine granulated biochar-based fertilizer (pristine gBBF) and soil aged gBBF granules (three replicates labelled as 'soil incubated gBBF'), either sampled from pots without (a) or with (b) leaching events during cultivation of white cabbage.Measurements were performed at 273.15K.Samples were outgassed at 303.15 K under vacuum for 12h before measurements.The adsorbed volume of CO2 was transformed to standard temperature conditions.

Figure 22
Figure 22 Pore size distribution of pristine collard-milled biochar (BC, a), soil-aged gBBF granules sampled after harvest of cabbage plants in pots without leaching events (b-d) and soil-aged gBBF granules sampled in pots including leaching events (e-g).The first derivate of the pore volume by pore width (dV/dd, red) and cumulative pore volume (black) is displayed.Data is based on CO2 adsorption experiments performed at 273.15 K after outgassing of the samples at 30°C for 12 h.

Table S4
Table S2 Main characteristics of the soil used in the greenhouse study.Concentrations are given based on dry matter, were applicable.Particle size distribution and texture of the soil used for the greenhouse study.

Table S5
Treatments prepared for the pot trial including the individual biochar application quantities and nutrient dosages.Biochar application quantities per hectare were calculated assuming a planting density of 40.000 cabbage plants ha-1and a concentrated application of the biochar in the root-zone of the plants.n.a.: not applicable.Treatments with ID 4, 5, 6 and 7 received two leaching events for nutrient leaching at each 30 L m -2 precipitation after 35 days and 63 days of plant cultivation.

Table S6
Recovery of phosphorus (P), potassium (K), magnesium (Mg), sulfur (S), calcium (Ca), copper (Cu), manganese (Mn) and zinc (Zn) from a certified reference material after microwave digestion and quantification with inductively coupled plasma optical emission spectroscopy (ICP-OES).The reference material was a rye grass listed as ERM-CD281 as sample no.1010 by the European Commission.Errors indicate ± standard deviation of replicated (n=5) digestions.

Table S8
Total volume of leachates and total amount of phosphorus (P) leached from the pots in the two precipitation events.Data is presented as

Table S10
Statistical results of two-way analyzes of variance for nutrient uptake in aboveground cabbeg biomass.Factors: 'Leaching' (no leaching /incl.leaching), 'Fertilizer type' (NPK/gBBF/B+NPK) and the interaction of both individual factors.Additionally, the block effect is presented 'Block'.