Effects of pyrolysis temperature on the hydrologically relevant porosity of willow biochar
Effects of pyrolysis temperature on the hydrologically relevant porosity of willow biochar
Hyväluoma, Jari;Hannula, Markus;Arstila, Kai;Wang, Hailong;Kulju, Sampo;Rasa, Kimmo
2018-10-17 00:00:00
Eects of pyrolysis temperature on the hydrologically relevant porosity of willow biochar 1 2 3 4 5 1 Jari Hyv aluoma, Markus Hannula, Kai Arstila, Hailong Wang, Sampo Kulju, and Kimmo Rasa Natural Resources Institute Finland (Luke), Tietotie 4, FI-31600 Jokioinen, Finland BioMediTech Institute and Faculty of Biomedical Sciences and Engineering, Tampere University of Technology, Tampere, Finland University of Jyvaskyla, Department of Physics, P.O. Box 35, FI-40014 University of Jyvaskyla, Finland Biochar Engineering Technology Research Center of Guangdong Province, School of Environment and Chemical Engineering, Foshan University, Foshan, Guangdong 528000, China; and Key Laboratory of Soil Contamination Bioremediation of Zhejiang Province, Zhejiang A & F University, Hangzhou, Zhejiang 311300, China Natural Resources Institute Finland (Luke), Latokartanonkaari 9, FI-00970 Helsinki, Finland (Dated: October 18, 2018) Biochar pore space consists of porosity of multiple length scales. In direct water holding appli- cations like water storage for plant water uptake, the main interest is in micrometre-range porosity since these pores are able to store water that is easily available for plants. Gas adsorption mea- surements which are commonly used to characterize the physical pore structure of biochars are not able to quantify this pore-size range. While pyrogenetic porosity (i.e. pores formed during pyrolysis process) tends to increase with elevated process temperature, it is uncertain whether this change aects the pore space capable to store plant available water. In this study, we characterized biochar porosity with x-ray tomography which provides quantitative information on the micrometer-range porosity. We imaged willow dried at 60 C and biochar samples pyrolysed in three dierent tem- peratures (peak temperatures 308, 384, 489 C, heating rate 2 C min ). Samples were carefully prepared and traced through the experiments, which allowed investigation of porosity development in micrometre size range. Pore space was quanti ed with image analysis of x-ray tomography images and, in addition, nanoscale porosity was examined with helium ion microscopy. The image analysis results show that initial pore structure of the raw material determines the properties of micrometre- range porosity in the studied temperature range. Thus, considering the pore-size regime relevant to the storage of plant available water, pyrolysis temperature in the studied range does not provide means to optimize the biochar structure. However, these ndings do not rule out that process tem- perature may aect the water retention properties of biochars by modifying the chemical properties of the pore surfaces. I. INTRODUCTION the permanent wilting point [5]. Therefore, the charac- terization of biochars aimed at soil amendment purposes should be focused on the micrometre-scale pores that pro- Pyrolysis biochars are carbon-rich porous materials vide water storage capacity and fast release of water to produced by heating biomass in the absence of oxygen. plants. Porosity and pore size distribution of biochar depend on the raw material selection and process conditions (e.g., Physical characteristics of biochars are often studied by type of pyrolysis device, heating rate, maximum tem- gas adsorption techniques accompanied by the Brunauer- perature and holding time at maximum temperature). Emmett-Teller (BET) modelling to determine the spe- Application of biochars to soils can alter and improve ci c surface area [6] or Barrett-Joyner-Halenda (BJH) soil physical and chemical properties, such as hydraulic modelling to determine the pore size distribution [7]. properties including hydraulic conductivity and water re- Pore space analysis based on these methods and mod- tention capacity (e.g. Refs. [1{3]). Biochar pore sizes els is limited to pores smaller than 300 nm, whereby can cover a wide range of scales. For direct water hold- gas adsorption measurements do not tell much about the ing eects, the minimum pore diameter of interest is ap- porosity in the size range that is important for biochars proximately 200 nm, which corresponds to the permanent that are used as soil amendments to improve the water wilting point with matric potential -1.5 MPa [4]. Water holding properties of soil [8]. In spite of this limitation, stored in pores smaller than this is not plant available. in many cases BET analysis still is the only characteriza- In addition, water stored in submicrometre pores above tion method used for physical pore space even for biochar the permanent wilting point is not readily available for produced for soil amendment purposes. Mercury intru- plants. For example, it has been reported that dry mat- sion porosimetry is able to quantify the micrometre-scale ter production of plants may virtually cease far before porosity of porous materials. Unfortunately the feasibil- arXiv:1810.07417v1 [cond-mat.mtrl-sci] 17 Oct 2018 2 ity of the method is reduced by the assumptions used in interpretation of the results, which do not hold for arbitrary porous materials [9]. Numerous studies have concluded that elevated pyrol- ysis temperature leads to increased BET speci c surface area (see, e.g., [10{15]). Microstructural evolution of biochars during carbonization was studied by Kercher and Nagle [16], who explained that the increase of spe- ci c surface area results from the creation of nanometre- range pyrogenetic pores due to the growth of high-density turbostratic crystallites. The total porosity on the other hand does not have a clear dependence on the production temperature. Gray et al. observed that total porosity re- mained relatively stable with increasing process temper- ature, which suggests that pyrogenetic nanoporosity does FIG. 1: Samples for imaging were prepared according to an- not have signi cant volume even though it can provide nual rings of the willow stem wood. One of the segments major part of the surface area [13]. (1) was dried at 60 C and three of them were in addition Biochar structure is summarized in the model proposed pyrolysed at (2) 308, (3) 384 or (4) 489 C. by Gray et al. [13]. In this model the pore space of biochar consists of stable residual porosity resulting from the cellular structure of the raw material, pyrogenetic micrometre porosity. HIM also allows for direct imaging nanoporosity, and aliphatic functionality on the walls of of non-conductive samples as the beam induced charg- the residual porosity. Residual pore structure is consid- ing can be neutralized with an electron
ood gun. These ered stable and independent of the pyrolysis tempera- imaging techniques allows us to get direct evidence on the ture and forms majority of the total pore volume. Py- signi cance of process temperature on the porosity that rogenetic nanopores are increasingly created as process is relevant in applications based on water holding capac- temperature increases, but even at higher temperatures ity of biochar and also more generally in any application they form only a minor part of the total pore volume. relying on the micrometre-range porosity of biochar (e.g. Aliphatic functionality is volatilized at higher tempera- microbial habitats [18]). In addition, we are able to com- tures which results in reduced hydrophobicity. pare the importance of pore size regimes probed by gas It is thus likely that feedstock selection is of major adsorption techniques in biochar research targeting to importance when one is interested in water holding ap- soil amendment applications. plications of biochar such as soil amendments. Cellular structure of the feedstock accounts for major part of pore volume and these pores typically are in the size range ca- II. METHODS pable to store readily plant-available water. Pyrogenetic pores are increasingly created with elevated process tem- A. Sample preparation perature, but water in these pores is too tightly bound to be available for plants and the total volume of these pores is low. However, it is not clear if process temper- In a previous study [19] we found that pore charac- ature contributes to water holding also by altering the teristics of ostensible same biochar may vary remarkably physical pore characteristics, e.g. due to shrinkage in the due to the heterogeneity of the raw materials used in pore size [17], or only by changing the surface chemistry biochar production. In order to minimize variation in and consequently the contact angle at the pore walls [13]. the pore space properties originating from raw material Characterization of porosity development during py- heterogenity, following sample preparation approach was rolysis is still mostly based on indirect measurements used. Fresh (moisture content 54.40.3 %) willow (Salix and conceptual models rather than on direct evidence. schwerinii Amgunskaja') was used as raw material. The In this paper we report results of an imaging study bimodal pore size distribution of willow makes it an at- where micrometre-scale porosity of biochars is directly tractive raw material for a study considering changes is imaged with x-ray computed microtomography. The pore structure. The stem wood was rst peeled and then three-dimensional image analysis results obtained for wil- cut into thin slices with thickness ca. 5 mm and diameter low biochars pyrolysed in dierent temperatures are com- ca. 2 cm. Slices were then dried at T = 60 C for 48 h. pared against gas adsorption measurements. Further in- One of the thin slices was selected for imaging (Fig. 1). formation about biochar porosity is obtained with helium The core of the samples was removed by drilling a hole ion microscopy (HIM), which is a novel imaging tech- (diameter 8 mm) and the part remaining was divided in 4 nique within the family of scanning beam microscopes. segments. The outermost part of each segment was also HIM results in images with very high resolution and large removed so that the actual samples were approximately eld of depth, thus giving accurate information of sub- 5 mm wide. 3 B. Pyrolysis (ash content, SFS-EN 13039). The total concentrations of C, H, S and N were determined with an elemental analyzer (Flash EA1112, Thermo Finnigan, Italy), and One of the segments was only dried and three seg- the O content was calculated by subtracting C, H, N, S, ments were in addition pyrolysed in varying tempera- and ash contents from the total. The BET surface area of tures. Pyrolysis was carried out with a bench-scale batch- the biochar was measured with N sorption analysis at 77 type pyrolysis device with maximum sample size up to K with a surface analyzer (TriStar II 3020, Micromeritics 500 g. The device has external electric heating system Instrument Corporation, USA) after degassing at 300 C. programmed to elevate temperature at speed of approx- imately 2 C min . The actual temperature was mea- sured inside the pyrolysis vessel. The device is equipped D. X-ray microtomography and image analysis with water jacket cooler and gasbag for gas collection. Before processing, the air tightness of the device was checked and the system was thoroughly
ushed with ex- The 3D structural characterization of the biochars cess of N to secure oxygen-free conditions. The holding 2 was conducted with X-ray microtomography. Zeiss time at peak temperature was 90 min to guarantee com- MicroXCT-400 device (Pleasanton, CA, USA) was used plete charring. The peak temperatures measured inside to acquire 1600 projection images evenly distributed in the pyrolysis chamber were 308, 384 and 489 C. Yields full 360 degrees with exposure time of 4 seconds. The for biochar and liquid fractions in dierent process tem- selected magni cation resulted in pixel size of 1.13 m. peratures were weighted and gas volume at room tem- Source voltage was adjusted to 40 kV and source current perature was measured using a gas meter (Ritter). The to 250 A. Filters were not used. 3D reconstruction was samples that were used in the imaging were placed in cru- performed with device manufacturers XMReconstructor cibles and buried within the bulk sample (300 g). Pur- software. pose of this setup was to ensure that the sample segments Obtained grey-scale images were ltered and seg- could be traced after pyrolysis, and that comparisons be- mented as described in a previous paper [19]. As in that tween the segments are meaningful. After pyrolysis the paper, a fully automatic segmentation algorithm based samples for x-ray tomography were taken from the seg- on a modi ed version of Otsu's method [22] was used ments. Sample size in x-ray tomography was ca. 1 mm. to avoid operator-dependent bias in the results. This Imaged samples contain reasonable number of pores and method is based on the work of Hapca et al., who added a they thus represent the pore structure of each segment. pre-classi cation step to the standard Otsu method [23]. The experimental setup was designed to compare samples In our variant, the average of the threshold values given taken from a single slice of willow and thus the purpose by the standard Otsu method and the modi ed method of the samples is not to represent willow in general. More by Hapca et al. is used, which corresponded very well representative study containing for example several raw with visually selected threshold value. materials would require signi cantly greater number of Images were analysed for porosity, speci c surface area, samples, which is not currently realistic in studies using pore size distribution, and structural anisotropy. A de- x-ray tomography (cf. e.g. recent work by Berhanu et tailed account of the methods used for these analyses is al. [20]). Also the samples for HIM imaging were taken given in [19]. Here we only brie
y mention that speci c from the same segments. surface area was determined using Minkowski functionals The rationale behind the selected process conditions as described by Vogel et al. [24] and pore size distribu- is the following. The non-pyrolysed sample serves as a tion by method based on mathematical morphology and control and provides information of the original cellu- especially morphological opening (see [25]). Structural lar structure of the raw material. According to Yang anisotropy was quanti ed as the degree of anisotropy et al. [21], in pyrolysis hemicellulose degrades mainly in D which was determined using the grey-scale gradi- temperature range 220-315 C, cellulose in range 315-400 ent structure tensor [26]. D values close to unity de- C, and lignin in wider range 160-900 C. Thus the low- scribe isotropic pore structure and higher values indicate est pyrolysis temperature demonstrates situation where stronger structural anisotropy. In [19], D was deter- hemicellulose is degraded to pyrolysis products. At the mined for several biochars derived from dierent feed- second highest temperature the major part of cellulose is stock and biochars with an anisotropic pore structure deformed, and in the highest temperature, also consider- resulting from vascular cell structure had D > 20. able fraction of lignin has been exposed to thermochem- In addition to the above-mentioned analyses, three- ical conversation (see [21] for the pyrolysis curves). dimensional shapes of individual pores were analysed with the method described by Claes et al. [27]. Here all connected pore voxels were de ned as a single pore C. Sample characteristics and the moment of inertia tensor was computed for each set of pore voxels. The principal moments of inertia were Before laboratory analysis, all samples (dried willow then determined by diagonalizing the moment of inertia and biochars) were ground and sieved through a 2 mm tensor. The pore shape can be quanti ed by calculating sieve. Samples were analysed for loss of ignition at 550 C the three principal dimensions of an approximate ellip- 4 soid using the principal moments of inertia determined These results, together with visual inspection of the for the pore. These principal dimensions L, I and S are pore structure (Fig. 2), suggest that the inherent vas- used to characterize the pore shape, L being the longest cular cell structure of willow determines the micrometre- dimension of the pore, I the longest dimension perpen- scale pore structure of the biochar. Pyrolysis can modify dicular to L, and S the dimension perpendicular to L and inherent pore structure of raw material to some extent. I . For tubular pores like the vascular pores in willow, the These changes can be obtained already in the tempera- interesting quantity is the ratio S=I which quanti es the ture typical for torrefaction, however, the pore structure cross-sectional shape of the pores. For circular pore cross remained relatively stable over the pyrolysis temperature section, this ratio is close to unity while for pores with regime studied here. High degree of anisotropy (high at cross section S=I approaches zero. D value), both in wood and biochar samples, indicates straight pipe-like form of the pore system with poor con- nectivity of the pore system in lateral directions. On the other hand, pipe-like structure ensures good connectivity E. Helium ion microscopy of the internal porosity to the surfaces of biochar parti- cles. HIM imaging was performed with a Carl Zeiss Orion Closer inspection of the pore size distributions (Fig. NanoFab device. The helium ion current was 0.2-0.3 pA 3) shows that changes observed in the total porosity and and the beam energy was 30 keV. Imaging of the 60 C SSA arise from the shrinkage of biochars. This means and 308 C samples required charge neutralization with that pore size distribution shifts toward smaller pore sizes an electron
ood gun between single line scans. Dwell when sample undergoes pyrolysis process. As can be seen time with these samples was 0.5 s and a line averaging in the dierence plots (panels on the right-hand side in of 128 times was used. Use of the
ood gun was necessary Fig. 3), most considerable change in the pore size distri- for charge neutralization as the electrical conductivity bution takes place between dried sample and lowest py- of samples were so low that signi cant charging eects rolysis temperature (308 C). Pore size distribution was were observed in the imaging. Samples pyrolysed at 384 shifted slightly towards smaller pore sizes between 308 C and 489 C did not require the use of the
ood gun and 384 C. Between the two highest pyrolysis temper- due to their higher conductivity. The dwell time with atures (384 and 489 C) a tiny shift towards larger pore these samples was 10 s and the line averaging was 4 sizes was observed. The dierences between dierent py- times. Pressure in the measurement chamber during the rolysis temperatures are most probably consequences of imaging was 1-210 Torr. the accuracy of imaging and pore-size distribution analy- sis and do not re
ect true changes in the pore structure. The cross-sectional shape of the pores was found to be III. RESULTS AND DISCUSSION dierent for dried willow and biochars (Table II. Lower S=I value for biochars indicate that shrinkage of the raw material in pyrolysis is anisotropic whereby the cross- Basic properties of the dried willow and biochars are sectional shape of the pores is more elongated in biochars shown in Table I. The clear decrease in O/C and H/C than in non-pyrolysed willow. Visual inspection of pore ratios with increasing pyrolysis temperature indicates shapes supports this nding (Fig. 2). Shape analysis changes in the chemical composition and structural ar- was done separately for large and small pores and the rangement of the initial organic matter. The biochars results for both pore types are consistent. It seems that produced at 384 and 489 C had O/C ratio less than for small pores, the dierence in pore shape may partly 0.2 which indicate high degree of carbonization. The result from connections formed between adjacent pores high H/C ratio of the biochar produced at 308 C in- (Fig. 2). However, as the imaging resolution is fairly dicates existence of non-condensed aromatic structures close to the thickness of pore walls, this conclusion is (e.g. lignin) [28]. When applied in soil, decrease in O/C uncertain. and H/C ratios indicates higher stability against microbi- ological degradation thus improving carbon sequestration The yield data presented in Fig. 4 explain why there are major dierences between the dried sample and the potential of biochar. lowest pyrolysis temperature. In this rst step as much The image analysis data (Table 2) indicate that py- as 57 % of initial mass of the dried raw material was con- rolysis increased micrometre-scale porosity (9-16 %) and verted into gases and liquids. The additional mass loss speci c surface area (7-13 %) of biochars compared to from the lowest pyrolysis temperature to the highest one those of dried willow stem wood. Note that the start- is only 15 %. However, it is notable that at the highest ing point in our study is dried willow which has already pyrolysis temperature roughly three-fourths (72 %) of to- shrunken as compared to fresh willow. Change of pyrol- tal raw material weight is lost (i.e. converted into gases ysis peak temperature in the range 308-489 C had only and liquids), but still the micrometre-scale pore structure minor in
uence on the porosity and speci c surface area. of biochar retains relatively well its initial geometry and All D values are at high level irrespectively of the treat- size. ment (reference data for typical D values determined for various biochar structures can be found from [19]). The next question is that how does weight loss of 72 % 5 FIG. 2: 3D visualizations of the x-ray tomographic images. Size of each visualized subsample is 280 m280 m280 m. Control 0.09 0.08 0.07 0.06 0 10 20 30 40 50 60 0.05 0.05 T = 308 C vs. Control 0.04 0.04 0.03 0.03 0.02 0.02 0.01 0.01 0.0 0.0 T = 308 C 0.09 -0.01 0.08 -0.02 0.07 -0.03 0.06 -0.04 0.05 0.05 T = 384 C vs. T = 308 C 0.04 0.04 0.03 0.03 0.02 0.02 0.01 0.01 0.0 0.0 T = 384 C 0.09 -0.01 0.08 -0.02 0.07 -0.03 0.06 -0.04 0.05 0.05 T = 489 C vs. T = 384 C 0.04 0.04 0.03 0.03 0.02 0.02 0.01 0.01 0.0 0.0 T = 489 C 0.09 -0.01 0.08 -0.02 0.07 -0.03 0.06 -0.04 0.05 -0.05 0 10 20 30 40 50 60 0.04 Pore diameter [ m] 0.03 0.02 0.01 0.0 0 10 20 30 40 50 60 Pore diameter [ m] FIG. 3: Pore size distributions on left and dierences of two adjacent distributions on right. Difference Volume distribution Volume distribution Volume distribution Difference Difference Difference 6 TABLE I: Elemental composition, ash content , H/C and O/C atomic ratios, and BET surface area of dried and pyrolysed willow stem wood samples. Dried Pyrolysed 60 C 308 C 384 C 489 C C [wt-%] 44.6 67.9 75.9 83.7 H [wt-%] 5.78 4.72 4.00 3.18 N [wt-%] 0.15 0.27 0.31 0.37 S [wt-%] 0.02 0.01 0.01 0.01 O [wt-%] 48.9 25.9 18.2 8.3 H/C ratio 1.54 0.83 0.63 0.45 O/C ratio 0.82 0.29 0.18 0.07 Ash [wt-%] 0.51 1.19 1.60 4.44 2 1 BET [m g ] n.d. 3.96 7.47 6.82 TABLE II: Results of 3D image analysis. Porosity, speci c surface area (SSA), degree of anisotropy (D ) and shape ratio S=I determined for dried willow and biochars pyrolysed at dierent temperatures. For S=I ratio, mean value and standard deviation were calculated separately for large and small pores. 2 3 Sample Porosity [-] SSA [mm mm ] D [-] S=I (large pores) [-] S=I (small pores) [-] Dried willow 0.32 98 88 0.770.12 0.770.14 Pyrolysis at 308 C 0.36 105 48 0.610.20 0.630.0.21 Pyrolysis at 384 C 0.35 110 82 0.450.15 0.610.24 Pyrolysis at 489 C 0.37 111 30 0.600.14 0.620.22 aect the biochar quality? BET results (Table I) indicate ysed samples. very low surface area for all studied biochars. In line with High-resolution HIM images show that x-ray tomogra- general assumption, there is a slight increase in measured phy visualization exaggerates the wall thickness between surface area with increasing peak pyrolysis temperature. the smaller diameter pores. The wall thickness is close The minimal increase of surface area in the temperature to the resolution of x-ray tomography, whereby segmen- regime considered here is expected as previous studies tation of walls becomes inaccurate. While resolution of have shown that greater increase happens at higher tem- x-ray tomography was not able to resolve pores in sub- perature (see e.g. [12, 29]). BET surface areas show a micrometre size range which partially contributes to the slight increase in surface from lowest to highest process porosity able to store plant available water, HIM im- temperature, which is in principle similar than that seen ages ensure that the studied biochar did not contain such in image analysis results (Table II). However, it must be porosity. noted that these two techniques probe dierent pore size The pyrolysis temperature mainly aects chemical regime and thus the results are not comparable. properties instead of physical ones (Table 1). Thus, when biochar use as soil amendment is considered, selection Gas adsorption results are supported by observations of process temperature should be judged on the basis made with HIM. HIM images of the four sample types of chemical measures. For example, acidic functional are shown in Fig. 5. In spite of the very high resolu- groups on biochar surface contribute to their wettability tion of HIM images, signi cant dierences in the surfaces and water retention capacity [8]. The relationship be- of cellular pores cannot be observed indicating that the tween capillary pressure across the water meniscus (p), mass loss in pyrolysis generates structures mainly at sub- pore diameter (d) and contact angle () is described by nanometre length scales. This is in agreement with the the Young-Laplace equation, theory presented in [7], which suggests that additional porosity would result from structural defects in carbon. 4
cos Comparison of the HIM images and the x-ray tomogra- p = ; (1) phy visualizations show similar vascular pore structures of two dierent size groups, namely one of about 5 m in where
is the surface tension of water [4]. Thus increas- diameter and another 30-50 m in diameter. Highest res- ing pyrolysis temperature can alter the water retention olution HIM images also show slightly more sub-micron properties either by changing the pore size distribution roughness in the pore walls of the 60 C dried sample (i.e., pore diameter in Eq. (1)) or surface wettability via compared to the smoother inner surfaces of the pyrol- surface chemistry (contact angle). According to Young- 7 308 C 384 C 489 C 60 50 30 25 0 0 Biochar Liquid Gas FIG. 4: Yields for biochar, liquid and gas fractions in dierent process temperatures. Left panel shows the fraction of initial raw material weight converted to biochar and liquid and right panel the volume of produced gas. FIG. 5: Helium ion microscopy images of studied samples in four dierent image sizes. From left to right, dried willow and 2 2 biochars pyrolysed at increasing temperature. Size of the images increase from 44 m to 500500 m from bottom to the top. Fraction of initial raw material weight [%] Volume [L] 8 Laplace equation, reduced wettability (increased ) de- development at nanometre size range but only minor ef- creases the pore volume which is able to store water at a fect on the density of surface functional groups [42]. given soil-water potential. Our results do not necessarily indicate that pyrolysis temperature would not have any in
uence on the wa- IV. CONCLUSIONS ter retention properties of biochars and biochar-amended soils. However, if such eects exist, the explanation would be in the modi ed surface chemistry rather than Our results indicate that at pyrolysis temperatures be- in the physical properties related to modi ed pore struc- low 500 C and with low heating rate (2 C min ), the ture. Hydrophobicity of biochar decreases with increas- initial pore structure of raw material (in this work wil- ing pyrolysis temperature [30] and biochar can be either low) determines the micrometre-scale porosity of biochar. hydrophobic or hydrophilic. Considering the pore size regime relevant for plant water It is also important to note that pyrolysis temperature uptake, pyrolysis temperature cannot be used to optimize has an eect on the cation exchange capacity of biochar the micrometre-scale pore system of biochar. Also, the [31], and the change in this property may also depend nanoscale porosity of biochar remained at low and vir- on the process temperature [32, 33]. As cation exchange tually constant level when subjected to process temper- capacity aects soil fertility, it should be also accounted atures less than 500 C, which is typical for torrefaction for when optimizing process parameters for production and slow pyrolysis. of biochars for soil amendment purposes. Considering the development of engineered biochars Concerning soil amendment use of biochar, our study for agronomic applications with speci c target to improve is limited to so-called direct eects of biochar, where soil water holding capacity, following conclusions can be biochar pores directly act as storage space for water [34]. drawn: (1) If aim is to increase the soil water holding Biochar may in addition also have indirect eects, as capacity at speci c matric potential regime directly via biochar application may aect the water retention prop- biochar pores, selection of appropriate raw material with erties of soil by improving soil structure, aggregation and desired pore size distribution is essential; (2) In such ap- aggregate stability [34{38]. Also the surface chemistry of plication it should, however, be veri ed how much tor- biochar may have direct or indirect eects. Biochar con- refaction or low temperature pyrolysis modi es pore size tact angle aects directly the water retention in biochar distribution via shrinkage of raw material; (3) If poros- particles (as described by Eq. (1)). In addition, depend- ity remains at satisfactory range, process conditions can ing on the chemical characteristics, biochar amendment be optimized for desired chemical properties of biochar may result to formation of water repellent coatings on (e.g. surface functional groups, hydrophobicity, carbon soil aggregates [39] or remove hydrophobic compounds sequestration potential), quality of derived pyrolysis liq- from the soil [40] which aects the water repellency of uids, and/or energy balance of the entire process. the biochar amended soil [41]. In this work we considered only relatively low process temperatures due to device-speci c limitations. Consid- Acknowledgments ering future work, similar approach should be applied to temperature range covering also higher pyrolysis temper- atures as well as to raw materials with dierent chemical This project has received funding from the European properties. The eects of heating rate were not consid- Unions Horizon 2020 research and innovation programme ered here, but it has been found to in
uence the porosity under grant agreement No 637020 - MOBILE FLIP. [1] A. Basso, F. Miguez, D. Laird, R. Horton, M. West- 2624(200208)165:4<468::AID-JPLN468>3.0.CO;2-G gate. Assessing potential of biochar for increasing water- [5] O.T. Denmead, R.H. Shaw. Availability of soil water to holding capacity of sandy soils. GCB Bioenergy 5, 132- plants as aected by moisture content and meteorolog- 143 (2013), doi:10.1111/gcbb.12026. ical conditions. Agronomy Journal 54, 385-390 (1962), [2] H. Herath, M. Camps-Arbestain, M. Hedley. Eect of doi:10.2134/agronj1962.00021962005400050005x. biochar on soil physical properties in two contrasting [6] S. Brunauer, P. Emmett, E. Teller. Adsorption of Gases soils: An Al sol and an Andisol. Geoderma 209210, 188- in Multimolecular Layers. Journal of American Chemical 197 (2013), doi:10.1016/j.geoderma.2013.06.016. Society 60, 309-319 (1938), doi:10.1021/ja01269a023. [3] H. Blanco-Canqui. Biochar and Soil Physical Proper- [7] E. Barrett, L. Joyner, P. Halenda. The Determination ties. Soil Science Society of America Journal 81, 687-711 of Pore Volume and Area Distributions in Porous Sub- (2017), doi:10.2136/sssaj2017.01.0017. stances. I. Computations from Nitrogen Isotherms. Jour- [4] J. Bachmann , R.R van der Ploeg. A review on recent de- nal of American Chemical Society 73, 373-380 (1951), velopments in soil water retention theory: interfacial ten- doi:10.1021/ja01145a126. sion and temperature eects. Journal of Plant Nutrition [8] W. Suliman, J.B. Harsh, I.A-L. Nehal, A-M. Fortuna, I. and Soil Science 165, 468-478 (2002), doi: 10.1002/1522- Dallmeyer, M. Garcia-Prez. The role of biochar porosity 9 and surface functionality in augmenting hydrologic prop- [23] S.M. Hapca, A.N. Houston, W. Otten, P.C. Baveye. New erties of a sandy soil. Science of the Total Environment local thresholding method for soil images by minimizing 574, 139-147 (2017). grayscale intra-class variance. Vadose Zone Journal 12 [9] H. Giesche. Mercury porosimetry: A general (practical) (2013) , doi:10.2136/vzj2012.0172 overview. Particle & Particle Systems Characterization [24] H.J. Vogel, U. Weller, S. Schluter. Quanti cation of 23, 9-19 (2006), doi:10.1002/ppsc.200601009. soil structure based on Minkowski functions. Com- [10] J. Park, I. Hung, Z. Gan, O.J. Rojas, K. Hun Lim, S. puters and Geosciences 36, 1236-1245 (2010), doi: Park. Activated carbon from biochar: In
uence of its 10.1016/j.cageo.2010.03.007. physicochemical properties on the sorption characteris- [25] M. Hilpert, R. Glantz, C.T. Miller. Calibration of tics of phenanthrene. Bioresources Technology 149, 383- a pore-network model by a pore-morphological anal- 389 (2013), doi:10.1016/j.scitotenv.2016.09.025. ysis. Transport in Porous Media 51, 267-285 (2003), [11] . Zhang, C. You. Water holding capacity and absorption doi:10.1023/A:1022384431481. properties of wood chars. Energy & Fuels 27, 2643-2648 [26] Z. Tabor, E. Rokita. Quantifying anisotropy of trabecu- (2013), doi:10.1021/ef4000769. lar bone from gray-level images. Bone 40, 966972 (2007), [12] C.E. Brewer, V.J. Chuang, C.A. Masiello, H. Gonner- doi:10.1016/j.bone.2006.10.022. mann, X. Gao, B.n Dugan, L.E. Driver, P. Panzacchi, K. [27] S. Claes, J. Soete, V. Cnudde, R. Swennen. A three- Zygourakis. New approaches to measuring biochar den- dimensional classi cation for mathematical pore shape sity and porosity. Biomass and Bioenergy 66, 176-185 in complex carbonate reservoir rocks. Mathematical (2014), doi:10.1016/j.biombioe.2014.03.059. Geosciences 48, 619-639 (2016), doi:10.1007/s11004-016- [13] M. Gray, M.G. Johnson, M.I. Dragila, M. Kleber. Wa- 9636-z. ter uptake in biochars: The roles of porosity and hy- [28] S. Schimmelpfenning, B. Glaser. One step forward toward drophobicity. Biomass and Bioenergy 61, 196-205 (2014), characterization: some important material properties to doi:10.1016/j.biombioe.2013.12.010. distinguish biochars. Journal of Environmental Quality [14] W. Li, K. Yang, J. Peng, L. Zhang, S. Guo, H. 41, 1001-1013 (2012), doi:10.2134/jeq2011.0146. Xia. Eects of carbonization temperatures on charac- [29] S. Zhao, N. Ta, X. Wang. Eect of temperature teristics of porosity in coconut shell chars and acti- on the structural and physicochemical properties of vated carbons derived from carbonized coconut shell biochar with apple tree branches as feedstock mate- chars. Industrial Crops and Products 28, 190-198 (2008), rial. Energies 10, 1293 (2017); doi:10.3390/en10091293, doi:10.1016/j.indcrop.2008.02.012. doi:10.3390/en10091293. [15] A. Yadav, K.B. Ansari, P. Simha, V.G. Gaikar, A.B. [30] T. Kinney, C.A. Masiello, B. Dugan, W.C. Hocka- Pandit. Vacuum pyrolysed biochar for soil amendment. day, M.R. Dean, K. Zygourakis, R.T. Barnes. Hydro- Resource-Ecient Technologies 2, S177-S185 (2016), logic properties of biochars produced at dierent tem- doi:10.1016/j.ret.2016.11.004. peratures. Biomass and Bioenergy 41, 34-43 (2012), [16] A.K. Kercher, D.C. Nagle. Microstructural evolution doi:10.1016/j.biombioe.2012.01.033. during charcoal carbonization by X-ray diraction [31] B. Liang, J. Lehmann, D. Solomon, J. Kinyangi, J. Gross- analysis. Carbon 41, 15-27 (2003), doi:10.1016/S0008- man, B. O'Neill, J. Skjemstad, J. Thies, F. Luiza, J. Pe- 6223(02)00261-0. tersen, E. Neves. Black carbon increases cation exchange [17] L. Pulido-Novicio, T. Hata, Y. Kurimoto, S. Doi, S. capacity in soils. Soil Science Society of America Journal Ishihara, Y. Imamura. Adsorption capacities and related 70, 1719-1730 (2006), doi:10.2136/sssaj2005.0383. characteristics of wood charcoals carbonized using one- [32] J. Yuan, R. Xu, H. Zhang. The forms of alkalis in the step or two-step process. Journal of Wood Science 41, biochar produced from crop residues at dierent temper- 48-57 (2001), doi: 10.1007/BF00776645. atures. Bioresource Technology 102, 3488-3497 (2011), [18] L.S. Schnee, S. Knauth, S. Hapca, W. Otten, T. Eickhors. doi:10.1016/j.biortech.2010.11.018. Analysis of physical pore space characteristics of two py- [33] X. Gai, H. Wang, J. Liu, L. Zhai, S. Liu, T. Ren, H. rolytic biochars and potential as microhabitat. Plant and Liu. Eects of feedstock and pyrolysis temperature on Soil 408, 357-368 (2016), doi: 10.1007/s11104-016-2935- biochar adsorption of ammonium and nitrate. PLoS ONE 9. 12, e113888 (2014), doi:10.1371/journal.pone.0113888. [19] J. Hyv aluoma, S. Kulju, M. Hannula, H. Wikberg, A. [34] F. Verheijen, S. Jeery, A.C. Bastos, M. van der Velde, I. K alli, K. Rasa. Quantitative characterization of pore Diafas. Biochar application to soils: A critical scienti c structure of several biochars with 3D imaging. Environ- review of eects on soil properties, processes and func- mental Science and Pollution Research, in press (2017), tions. EUR 24099 EN, Oce for the Ocial Publications doi:10.1007/s11356-017-8823-x. of the European Communities, Luxembourg (2010). [20] S. Berhanu, M. Hervy, E. Weiss-Hortala, H. Proudhon, [35] S.P. Sohi, E. Krull, E. Lopez-Capel, R. Bol. A review M. Berger, A. Chesnaud, M. Faessel, A. King, D. Pham of biochar and its use and function in soil. Advances Minh, A. Villot, C. Gerente, A. Thorel, L. Le Coq, A. Nz- in Agronomy 105, 47-82 (2010), doi:10.1016/S0065- ihou. Advanced characterization unravels the structure 2113(10)05002-9. and reactivity of wood-based chars. Journal of Analyti- [36] H. Soinne, J. Hovi, P. Tammeorg, E. Turtola. Eect cal and Applied Pyrolysis 130, 79-89 (2018). of biochar on phosphorus sorption and clay soil ag- [21] H. Yang, R. Yan, H. Chen, D. Lee, C. Zheng. Character- gregate stability. Geoderma 219-220, 162-167 (2014), istics of hemicellulose, cellulose and lignin pyrolysis. Fuel doi:10.1016/j.geoderma.2013.12.022. 86, 1781-1788 (2007),doi:10.1016/j.fuel.2006.12.013. [37] S.-G. Lu, F.-F. Sun, Y-.T. Zong. Eect of rice husk [22] N. Otsu. A Threshold Selection Method from Gray- biochar and coal
y ash on some physical properties of Level Histograms. IEEE T Syst Man Cyb 9, 62-66 expansive clayey soil (Vertisol). Catena 114, 37-44 (2014), (1979),doi:10.1109/TSMC.1979.4310076. doi:10.1016/j.catena.2013.10.014. 10 [38] D. Wang, S.J. Fonte, S.J. Parikh, J. Six, K.M. Scow. doi:10.1016/j.geoderma.2016.09.018. Biochar additions can enhance soil structure and the [41] I.L. Hallin, P. Douglas, S.H. Doerr, R. Bryant. The eect physical stabilization of C in aggregates. Geoderma 303, of addition of a wettable biochar on soil water repellency. 110-117 (2017), doi: 10.1016/j.geoderma.2017.05.027. European Journal of Soil Science 66, 1063-1073 (2015), [39] A. Piccolo, G. Pietramellara, J.S.C. Mbagwu. Use doi:10.1111/ejss.12300. of humic substances as soil conditioners to increase [42] B. Zhao, D. O'Connor, J. Zhang, T. Peng, Z. Shen, aggregate stability. Geoderma 75, 267-277 (1997), D.C.W. Tsang, D. Hou. Eect of pyrolysis temperature, doi:10.1016/S0016-7061(96)00092-4. heating rate, and residence time on rapeseed stem de- [40] I.L. Hallin, P. Douglas, S.H. Doerr, I. Matthews, rived biochar. Journal of Cleaner Production 174, 977- R. Bryant, C. Charbonneau. The potential of 987 (2018). biochar to remove hydrophobic compounds from model sandy soil. Geoderma 285, 132-140 (2017),
http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.pngCondensed MatterarXiv (Cornell University)http://www.deepdyve.com/lp/arxiv-cornell-university/effects-of-pyrolysis-temperature-on-the-hydrologically-relevant-4TdABlp0OW
Effects of pyrolysis temperature on the hydrologically relevant porosity of willow biochar