Chemical Changes During Hydrothermal Carbonization of Manure Derived from Free-Range Bred Chickens and Its Potential as Organic Fertilizer for Tomato, Lettuce and Sunflower Plants

Abstract

Hygienization by hydrothermal carbonization (HTC) of chicken manure (CM) at 250 °C allows its valorization as soil amendment or even organic fertilizer. To test if this hypothesis is also valid for feedstocks from free-range breeding, respective material of a small farm in southern Spain was comprehensively chemically characterized. The hydrochar of the manure collected from the ground of the farm was rich in mineral matter. After HTC, 68% of the organic carbon (C) was recovered, whereas 82% of the nitrogen (N) was lost most likely by volatilization and with the discarded process water. Despite this, 2.8% of the total N in the hydrochar was identified as inorganic N (N_i). Solid-state ¹³C and ¹⁵N NMR spectroscopy revealed aromatization of organic C and N, although alkyl C and amide N still contributed with 23% and 35% to the total organic C and N, respectively. The obtained distribution of N-forms indicated that enough N_i is plant-available for early plant growth, while the remaining N occurs in structures that can be slowly mobilized during advanced plant development. Low heavy metal concentrations suggest low phytotoxicity. Pot experiments with lettuce, sunflower, and tomato plants confirmed species- and dosage-dependent effects. A dosage of 3.25 t ha⁻¹ improved lettuce and sunflower yields, whereas a dosage of 6.5 t ha⁻¹ provided no additional growth benefits but caused phytotoxic reactions of the tomato plants. Our results support HTC as a strategy to valorize CM from free-range farms, although, due to the high variability of such materials, we recommend a thorough chemical characterization and phytotoxic tests before its application.

1. Introduction

The sustained growth of the global population is driving an increase in worldwide food and livestock feed consumption, which is unfortunately coupled with a surge in the production of organic waste. A key goal in the sustainable management of this waste is its conversion into value-added products such as soil amendments or fertilizers, which can be reintegrated into the ecological cycle while restoring soil physical and chemical properties [1]. Animal waste from livestock farming, such as manure, has drawn particular attention due to its high production rate and the significant environmental and ecological challenges it poses, including greenhouse gas emissions, water contamination, and soil pollution [2]. Beyond its environmental impact, the revalorization of manure is also of socioeconomic interest because of its fertilizing potential, given its content of essential nutrients for agricultural applications [3].

Composting is one of the most common manure recycling methods. However, it is often associated with malodors and potential soil contamination due to surviving pathogens or pharmaceutical residues [4]. Consequently, alternative treatments that ensure efficient hygienization and degradation of pollutant and toxic components are needed. Thermochemical treatments have emerged as promising options to transform organic waste into stable, sterilized materials. Currently, dry pyrolysis and hydrothermal carbonization (HTC) are the two main thermochemical processes used to produce solid carbonaceous materials from organic waste. Both processes occur under oxygen-depleted conditions. However, pyrolysis operates at higher temperatures (300–700 °C) with dry feedstocks, yielding in a solid product known as biochar [5], whereas HTC converts wet biomass into hydrochar under subcritical water conditions (150–350 °C, autogenous pressure) [6]. For wet feedstocks such as manure, HTC presents clear advantages, including lower production costs due to reduced energy requirements and the absence of a drying step [7]. This is of particular interest for large-scale producers that need to manage large volumes of manure. However, bearing in mind that poultry manure contains considerable amounts of nutrients that can be recycled for crop production, hydrochar offers a further advantage over biochar. This consists in its lower chemical recalcitrance, promoting faster nutrient remobilization and the stabilization of nutrients in more plant-available forms [3]. As reported by Paneque et al. [7], amendment of hydrochar from sewage sludge allowed a better plant growth than the corresponding biochar in particular during the first plant development stage. Compared to untreated sewage sludge amendment, less nitrogen (N) leached during the early plant development, and better plant growth was achieved during a second cropping period after addition of hydrochar [8]. Based on these results, it was concluded that hydrochar of N-rich feedstock provided a favorable balance between N that was quickly available for plant growth and N that was protected from leaching by its immobilization in organic forms with medium term residence times. This aspect may be of interest not only for fertilizer producers or large-scale farmers, but also for small-scale poultry breeders intending to perform sustainable management by enabling recycling of the chicken manure (CM) for fertilization purpose. However, here, one must note that properties of CM depend on poultry breeding conditions and are expected to be largely different for feedstocks derived from automatized large-scale producers and CM accumulating in small farmers practicing free range chicken breeding. Whereas the manure of the former is expected to be composed mostly of organic components, the latter will certainly provide a larger mineral fraction due to mixing of the excretions with the soil of the chicken run. Comparably, the feeding method of the poultry may affect the properties of the manure, since residual food, spread in the chicken run, can be integrated into the feedstock during its collection. Thus, it may be assumed that a hydrochar produced from CM of a free-range will contain less plant nutrients than a product derived from residues of a commercial large-scale chicken farm.

However, to evaluate if this nutrient reduction affects the suitability of the respective hydrochar as potential organic fertilizer or soil amendment, a detailed characterization of the product is essential. This includes recovery and distribution of nutrients among HTC phases [4,9,10,11,12], but also the determination of heavy metals (HMs), and other metals, as their transformations during HTC may affect the stability and interactions of other elements in the resulting hydrochar. Elements such as Ca, Mg, Fe, and Al have been reported to play critical roles in P immobilization and retention during HTC [13,14,15,16].

A parameter, affecting the suitability of a product as an efficient soil amendment and organic fertilizer, represents its biochemical recalcitrance. Commonly, this can be deduced from the quality of its organic matter. There is still a lack of knowledge since only few studies have explored changes in the organic carbon (C) and N composition of manure-type feedstocks caused by HTC using advanced spectroscopic techniques [17,18,19]. In addition, field or pot-scale experiments are required to assess the actual fertilizing performance and biological safety of the potential soil amendment [7,20,21,22].

Taking all these considerations into account the present work seeks to fill this knowledge gap and provide an orientation for evaluating if hydrochar produced from CM of a free-range management may be usable for improving soil fertility. Therefore, the main objectives of this study are to:

(i)

Conduct a comprehensive chemical characterization of hydrochar derived from CM from a small-scale chicken breeder, emphasizing the assessment of organic C and N forms by solid-state nuclear magnetic resonance (NMR) spectroscopy, alongside with the quantification of macro- and micro-nutrient as well as HM contents.

(ii)

Analyze the losses and recovery rates of different nutritive elements during the HTC process, calculated by taking the product of the hydrochar mass yield and the content of the analyzed element in the hydrochar, and normalizing it by the element’s content in the feedstock.

(iii)

Evaluate the resultant hydrochar in pot experiments to assess both its preliminary phytotoxicity and its potential suitability as a fertilizer on three different economically and agronomically important plant species: lettuce, sunflower, and tomato.

2. Materials and Methods

2.1. Feedstock

Fresh raw chicken manure (CM) was collected from a chicken yard located in Utrera (Seville, Spain) (37°13′16.7″ N–5°46′25.0″ W). Chickens were raised non-commercially in free range, intended solely for personal consumption. The diet of these animals was exclusively based on corn (C4 plant). The manure collection was conducted in the morning, selecting the freshest and most recent manure pile from the floor of the chicken run, and gathering it manually. The collected material was stored in plastic bags. During the collection process, care was taken to avoid extracting material that was not manure, such as plant residues or soil. However, inclusion of soil could not completely be avoided. According to SoilGrids [23], the soils in the area surrounding the chicken yard are classified as Vertisols (21%), with possible occurrences of Luvisols (20%), Cambisols (14%), and Regosols (11%). In addition, they show a loam to clay loam texture, with approximately 306 g kg⁻¹ of clay, 339 g kg⁻¹ of silt, and 355 g kg⁻¹ of sand in the top 30 cm. The estimated bulk density is 1.56 g cm⁻³, and the organic C content is approximately 1% corresponding to an estimated soil organic C stock of approximately 31 t ha⁻¹. Such characteristics are typical of mineral agricultural soils with moderate organic matter content and relatively dense structure. The manure sample was promptly dried to reduce biological activities. Subsequently, the dried manure underwent sieving, and the fraction ranging from 0.2 to 0.8 mm was isolated and preserved to produce hydrochar and for further analysis.

2.2. Production of Hydrochar

The collected and dried CM was subjected to hydrothermal treatment at the Leibniz-Institut für Agrartechnik und Bioökonomie e.V (ATB), Potsdam-Bornim (Germany), using an 18.75 L-stirred pressure reactor (Parr reactor 4555 series, Moline, IL, USA), fitted with an external resistance heater and internal sensors to monitor pressure and temperature. After mixing the CM with deionized water at a 1:5 ratio, the slurry was heated in the closed reactor at a rate of 2 °C min⁻¹ until 250 °C. Reaching the desired temperatures in the reactor required about 120 min. This temperature was held for 30 min. Throughout the process, there was constant stirring at 200 rpm. Subsequently, the heater was turned off and the reactor was left to cool for a period of approximately 20 h after which the HTC slurry was vacuum filtered using flat filter paper (ROTH Type 113A-110, 5–8 µm). The gaseous product and the process liquid were discarded. The resulting hydrochar was dried for 48 h at 60 °C before storage in hermetically sealed plastic bags.

2.3. Characterization of the Chicken Manure and the Produced Hydrochar

2.3.1. Chemical Composition

Prior to the analyses, all samples were oven-dried at 40 °C for 24 h to remove moisture that may have been adsorbed during storage. For the determination of the ash content of the samples, they were also heated at 105 °C for 1 h to remove additional moisture before increasing the temperature to 550 °C for 5 h to combust organic residues. The pH of the used feedstock and hydrochar was determined in the aqueous phase of a soil suspension (2.5 g of dry sample with 10 mL of deionized water) with a Crison pH-meter Basic 20 (Crison, Barcelone, Spain) after the solid phase had settled for 30 min. Subsequently, the electrical conductivity (EC) was measured in the filtered supernatant using a Crison EC-meter Basic 30+ (Crison, Barcelone, Spain). The maximal water holding capacity (WHC) was obtained according to the method of Veihmeyer and Hendrickson [24] by placing 6 g of each sample over a filter paper (Whatman 2) into a funnel. After saturation of the samples with distilled water and a 2 h resting period, the weight difference between the dry and moist sample was calculated considering the weight of the funnel and the filter paper, resulting in the maximal WHC. The latter is expressed as the retained water as percentage relative to the total dry weight (DW) of the sample.

Total C, organic C (C_org), inorganic C (IC), and N contents were measured using the Primacs SNC100 elemental analyzer (SKALAR, Breda, The Netherlands) and the Shimadzu TOC-V analyzer (Shimadzu, Kyoto, Japan). Elemental hydrogen (H) was determined using a LECO CHNS-932 (St. Joseph, MI, USA) from samples demineralized through four consecutive treatments with 10% (m/m) hydrofluoric acid (HF). Elemental oxygen (O) was calculated as the difference between the total sample weight and the sum of the percentages of ash content, C, N, H, P, and sulphur (S). P and S were measured using inductively coupled plasma-optical emission spectroscopy (ICP-OES; Varian 720-ES) (Varian Inc., Palo Alto, CA, USA).

The concentrations of dissolved organic C and N (DOC and DON) were quantified with a Shimadzu TOC-V analyzer (Shimadzu, Kyoto, Japan) in 0.5 M K₂SO₄ extracts from CM and its hydrochar.

The δ¹³C (‰; vs. VPDB) and δ¹⁵N (‰; vs. air-N₂) of both CM and the hydrochar were measured with a Flash HT Plus elemental analyzer (Thermo Fisher Scientific, Bremen, Germany) coupled to a Delta-V advantage isotopic ratio mass spectrometer (IRMS) (Thermo Fisher Scientific, Bremen, Germany) via ConFlow IV interfase (Thermo Scientific, Bremen, Germany).

Plant-available inorganic N (N_i) forms, such as NO₃⁻ and NH₄⁺, were extracted with 2 M KCl and determined colorimetrically [25,26] using an Omega SPECTROstar (BMG LABTECH GmbH, Ortenberg, Germany) spectrophotometer (software V6.20 Edition 2). The contents of total macro- (P, K, S, Ca and Mg) and micro-nutrient (Fe, Cu, B, Mn, Zn, Ni, and Mo) as well as other metal and HM contents (Na, Al, Ba, Li, Sr, As, Cd, Co, Cr, Hg, Pb, and V), were determined by ICP-OES (Varian 720-ES) (Varian Inc., Palo Alto, CA, USA). Available P was quantified with a multiparameter Bran-Luebbe autoanalyzer (Bran-Luebbe Analytics, Norderstedt, Germany) after extraction with sodium hydrogen carbonate at pH 8.5 [27]. Available K, extracted with ammonium acetate at pH 7, and water-soluble S contents were measured by ICP-OES (Varian 720-ES) (Varian Inc., Palo Alto, CA, USA) [28].

The recovery of each element or nutrient after HTC was calculated as the product of the hydrochar mass yield and the content of the analyzed element in the hydrochar, divided by the content of the element present in the feedstock.

2.3.2. Solid-State ¹³C and ¹⁵N Nuclear Magnetic Resonance Spectroscopy

Prior to solid-state NMR spectroscopy, and after decarbonization with HCl (10%), both CM and its hydrochar were demineralized with four consecutive treatments with 10% (w/w) hydrofluoric acid (HF) to enhance the organic matter concentration. Following the removal of the supernatant, the solid residue was washed with distilled water until the solution exceeded a pH of 5 and then it was subjected to freeze-drying [29].

Solid-state ¹³C NMR spectra were generated using a Bruker Avance III HD 400 MHz WB spectrometer (Bruker, Biospin GmbH & Co. KG, Ettlingen, Germany), equipped with a triple resonance broadband probe and using zirconium rotors with a 4 mm OD and KEL-F caps. The cross-polarization magic angle spinning (CP-MAS) technique was applied with a rotor spinning speed of 14 kHz. To prevent spin modulation of Hartmann-Hahn conditions, a ramped ¹H-pulse was applied during a 1ms-contact time. Preliminary testing confirmed that a pulse delay of 1 s was adequate to prevent saturation. The spectra were obtained with 5000 to 15,000 scans, and a line broadening between 50 and 100 Hz was applied. They were quantified according to Knicker et al. [30] using the software package MestReNova 10^® (Santiago de Compostela, Spain). Assessment of the relative ¹³C intensity distribution involved integrating specific chemical shift regions: alkyl C (0–45 ppm); N-alkyl/methoxyl C (45–60 ppm); O-alkyl C (60–110 ppm); aryl C (110–140 ppm); O/N-aryl C (140–160 ppm); carboxyl C (160–185 ppm); and carbonyl C (185–225 ppm). At the used spinning rate, the chemical shift anisotropy of crystalline structures is not fully averaged. This leads to the appearance of spinning sidebands at both sides of the parent signal at a distance determined by the spinning speed. For the studied hydrochar they occur mainly for the aromatic C and appear between 250 and 300 ppm and between −50 and 0 ppm. Since they contain intensity of the parent signal their relative contribution was added to that of the aromatic C region [30]. Calibration of the ¹³C chemical shift scale was performed using tetramethylsilane (0 ppm) with glycine as the reference (COOH at 176.08 ppm).

The recovery of C_org as a specific C group in hydrochar as a function of the treatment C_x (hc) was calculated according to Paneque et al. [18], as follows:

$${\mathrm{C}}_{\mathrm{x}}(hc)={\displaystyle \frac{C\left(x\right)\ast C\left(t\right)}{100}}$$

where C(t) represents the remaining C in the hydrochar as a percentage of the C content of the chicken manure (C_CM), and C(x) denotes the relative ¹³C intensity of a specific ¹³C chemical shift region.

The same spectrometer was used to acquire the solid-state ¹⁵N NMR spectra but using a 7 mm double resonance probe operating at 40.56 MHz. Here, a 1 ms contact time, a 90° H pulse width of 3.5 μs, and a pulse delay of 300 ms, were applied. The chemical shifts are referenced to the nitromethane scale (0 ppm) and were adjusted using ¹⁵N-labeled glycine (−347.6 ppm). The region assigned to amide N (−248 to −285 ppm) and pyrrole N (−150 to −248 ppm) were integrated. For the spectra, between 850,000 and 1 million scans were accumulated at a spinning speed of 6 kHz.

2.4. Plant Experiment Conditions and Measurements

Due to the physical and chemical properties of the studied hydrochar (low EC, high NO₃⁻ content providing a favorable inorganic to recalcitrant organic N ratio, and presumably low concentration of HMs), its potential as a fertilizer as well as its phytotoxicity was assessed. For this purpose, a greenhouse pot-experiment was carried out to examine the germination and early growth of lettuce (Lactuca sativa L. var. Batavia), sunflower (Helianthus annuus L.), and tomato (Solanum lycopersicum L.) in response to the application of hydrochar. For each plant, a total of 30 pots (7 × 7 × 9 cm) were arranged, each containing a substrate mixture with a peat to vermiculite proportion of 60:40 (v:v). Out of the 30 pots, 10 were left untreated as controls (CTR), 10 received an equivalent dose of 3.25 t ha⁻¹ of hydrochar (HC-3.25), and the remaining 10 pots were amended with a dose of 6.5 t ha⁻¹ (HC-6.5). The doses of 3.25 and 6.5 t ha⁻¹ of hydrochar correspond to 2.7 and 5.4% w/w, respectively, and were applied by mixing it into the top 3 cm. In total, 90 pots were set up. These low doses correspond to an addition of approximately 15.2 and 30.4 kg N ha⁻¹ and ensured avoidance of potential surplus fertilization if used in field applications of crops needing repeated fertilization during their total growing phase.

Four seeds were sown per pot, which were placed into a greenhouse with a 14 h/10 h light/dark photoperiod (250 μmol m⁻² s⁻¹) at 25 °C/20 °C respectively and 40–70% relative humidity. To ensure a minimum of 8 seeds per biological replicate for the statistical analysis of the germination, the seed development was monitored using 2 pots each. The germinated seeds of lettuce and sunflower plants were counted and recorded during the initial 7 days after sowing (DAS), whereas the data for the tomato seeds were recorded after 14 DAS due to a delayed germination of 7 days. Based on the methodology described in Ranal et al. [31,32], the following key germination parameters were assessed:

-

An estimate of the germinability: Germination percentage (G);

-

Mean Germination Time (MGT) corresponding to the time required for seeds to germinate or emerge;

-

Mean Germination Rate (MGR) representing the speed at which seeds germinate;

-

The time required for 50% of the seeds to germinate (T50);

-

Synchronization index (Z) describing the uniformity or synchronization in the timing of seed germination;

-

The uncertainty of the germination process (U) reflecting the degree of uncertainty associated with the distribution of the relative frequency of germination.

The vigor Index (VI), which determines the performance of seeds during germination and seedling emergence, was calculated as the product of G (%) by the hypocotyl length (cm).

After 7 DAS for lettuce and sunflower and 14 DAS for tomato, one germinated seedling from each pot was selected and left in the substrate (n = 10), while all other seedlings were harvested for an additional biomass analysis (taking the mean for each pot, n = 10). Within each treatment, the chosen seedlings displayed similar health and size characteristics, ensuring uniformity across the samples. The pots were watered daily avoiding any leaching. The trays underwent daily clockwise rotation to provide uniform light exposure and environmental conditions for all plants during the experiment. No pesticides or supplementary fertilizers were introduced to the substrate or plants before or during this research.

The hypocotyl and total plant height of lettuce and sunflower plants was measured daily until 7 DAS, whereas for tomatoes, the assessment was extended until 14 DAS due to the delayed germination. Additionally, the total height of all plants (n = 10) was measured at 15 and 20 DAS, and at 25 DAS for tomatoes. Similarly, the photosynthetic efficiency or quantum yield of photosystem-II (QY_PSII), a well-known marker for plant stress, was assessed using a portable fluorimeter (FluorPen FP-100; Photon System Instruments, Brno, Czech Republic) at 7, 14, and 20 DAS for lettuce and sunflower plants, and at 14, 20, and 25 DAS for tomato. Three readings were taken from each leaf and averaged for each determination (n = 10).

After 20 days of the experiment, lettuce and sunflower plants were harvested, oven-dried (72 h at 65 °C), and weighed to obtain the shoot dry weight (DW). For the tomato plants, the final harvest occurred at 25 DAS.

2.5. Statistical Analysis

Statistical analyses were performed using the STATGRAPHICS Centurion XIX software (version 19.6.03) (StatPoint Technologies, Warrenton, VA, USA). The Shapiro–Wilk test was used to verify the normality of the datasets. Normal distributed response variables were analyzed by t-Student test or one-way ANOVA, followed by Tukey’s Honestly Significant Difference test (HSD). When response variables were non-normal, Mann–Whitney U or Kruskal–Wallis followed by Mann–Whitney U tests were conducted.

3. Results and Discussion

3.1. Mass Yield and Ash Content

The HTC of CM at 250 °C with a residence time in the reactor of 0.5 h led to a 10.4% loss of the original dry weight, resulting in a remarkable higher mass yield compared to the values found in the literature for this type of feedstock subjected to similar carbonization temperatures and residence times [33,34,35,36]. This is best explained by the high ash content of the CM, exceeding 80% (

Table 1. Mass yield after the HTC of chicken manure (CM), the ash content, the elemental composition (C_org, N, O, H) of the feedstock and its respective hydrochar, as well as the recovery of these elements after the HTC treatment at 250 °C for 0.5 h. The values represent the mean of three replicates (n = 3) ± standard error (SE). The values followed by different letters indicate significant differences according to the statistical test used (T-Student or U Mann–Whitney). Levels of significance: p > 0.05 (“ns”, not significant differences); * p ≤ 0.05; *** p ≤ 0.001.

			Elemental Composition
	Mass Yield	Ash Content	Corg	N	O b	H
	% DM a	g kg−1	(%)	(%)	(%)	(%)
CM	-	850 ± 9	7.70 ± 0.19 a	2.35 ± 0.18 a	3.52 ± 1.02 a	1.04 ± 0.25
Hydrochar	89.6	916 ± 2	5.86 ± 0.74 b	0.47 ± 0.05 b	1.38 ± 0.21 b	0.61 ± 0.15
p-value			***	***	*	ns
Recovery (% DM)		96.0	68.3 ± 1.69	17.8 ± 1.97	49.9 ± 2.59	63.7 ± 2.52

^a DM: dry matter; ^b O content = 100 − Ash content + C + N + H + P + S.

). This high content deviates from what is typically observed in manure-like feedstock and resembles more the levels found in sewage sludge, where sand is prevalent among the residues [18]. In our case, residual soil which is expected to be part of CM collected from barn-raised chickens most likely contributed strongly to the high ash contents of the CM and subsequently of the hydrochar.

The preferential loss of organic matter with respect to the mineral phase during HTC represents a further reaction leading to a relative increase in the ash content to 916 g kg⁻¹ in the hydrochar (Table 1). However, the high ash recovery of 96% indicates that few ash components dissolved in and were lost with the process water.

3.2. Physical and Chemical Parameters of Chicken Manure and Hydrochar

The HTC of the feedstock slightly but significantly increased its pH from 7.21 to 7.92, whereas EC dropped drastically by 91%, from 3.83 to 0.33 mS cm⁻¹, indicating a considerable loss of charged groups or salts. Even lower values have been documented in other manure-based hydrochars [5,12]. This observation is best explained with the mobilization of ions during the HTC process and their leaching into the reaction solution that has been removed and discarded after the process [6,37]. Note that the thermal process promotes decarboxylation and deamination releasing volatile CO₂ and NH₃ [6,38].

Although a high WHC is not expected for the lignocellulosic-poor CM, the HTC process decreased this value from 108% (±6.63) to 32% (±1.44). As WHC is typically associated with porosity [1], our results suggest poor pore structure development during HTC.

3.3. Modifications in the Elemental and Nutrient Composition During HTC

3.3.1. Elemental and Isotopic Composition

In accordance with the high ash content, the total C_org content of the feedstock is 77 g kg⁻¹ (7.70%), which is low considering the type of raw material [1,4,33] (Table 1). The high content of IC in CM (10.6 ± 0.29 g kg⁻¹), is in line with the calcareous nature of the soils at the farm. Only 68% of total C_org of the feedstock survived the HTC treatment, which is higher than the value reported by Ghanim et al. [34], for poultry litter treated by HTC at 250 °C for 0.5 h (43%). In a subsequent work [35], the same authors observed that increasing the residence time to 2 h further reduced the C recovery to approximately 30%. Comparison of those data with ours highlights the variability in C retention as a function of feedstock characteristics and process conditions. During HTC, biomass components are hydrolyzed into soluble oligomers and monomers, which, along with the DOC already present in the feedstock, can be dissolved in the liquid phase of the process [39]. These soluble compounds can undergo further degradation by dehydration, dehydroxylation, decarboxylation, and condensation [18,40]. This is confirmed by the low recovery of O and H (Table 1) and the decrease in the atomic O/C and H/C ratios after HTC (from 0.34 to 0.18 and from 1.62 to 1.25, respectively). A rough evaluation of the atomic H/C ratio confirms that in average 1.25 protons are bound to one C indicating that the hydrochar does not possess a highly condensed aromatic structure.

In terms of losses, the N content in the hydrochar experienced a much more significant decrease than C as only 18% of this element remained in the hydrochar (Table 1). As a result, the hydrochar exhibited a low N content of only 4.7 g kg⁻¹. Comparable N contents of CM and its hydrochar as well as N-recovery rates were reported by Heilmann et al. [13]. Other authors claim higher losses of over 65% of N after HTC treatment of manure at 250–260 °C [1,4,33]. In accordance with the findings proposed by Li et al. [4], a substantial amount of N from the CM must have been converted into the liquid and gas phases rather than the solid phase.

The fact that HTC increased the C_org/N ratio from three to thirteen confirms the low thermal stability of some N-functional groups in CM (

Table 2. Total organic carbon (C_org), C_org/N ratios and inorganic N (N_i) contents (NO₃⁻-N and NH₄⁺-N) of chicken manure (CM) and its respective hydrochar, as well as the recovery rates after the HTC treatment at 250 °C for 0.5 h. The values represent the mean ± standard error (SE) for n = 9. The values followed by different letters indicate significant differences according to the statistical test used (T-Student or U Mann–Whitney). Levels of significance: p > 0.05 (“ns”, not significant differences); *** p ≤ 0.001.

	Corg/N	NO3−-N	NH4+-N	Ni of the Total N
	(w/w)	g kg−1	g kg−1	%
CM	3.36 ± 0.08 b	0.38 ± 0.02 a	0.17 ± 0.00 a	2.34 ± 0.18
Hydrochar	13.0 ± 0.75 a	0.08 ± 0.02 b	0.05 ± 0.01 b	2.78 ± 0.31
p-value	***	***	***	ns
Recovery (% DM)	-	19.4 ± 3.90	23.5 ± 3.07	106 ± 15.5

). This is expected since uric acid is the main organic N-form in the faecal remains of birds. Thus, this final ratio falls within the expected range for hydrochars from such feedstock [4,33] and the treatment conditions [36]. Note that more than 97% of the N content of feedstock and product is organic (Table 2). Even though the contribution of N_i is similar in both materials, there was no significant enhancement of plant-available NO₃⁻-N and NH₄⁺-N in the hydrochar after HTC. In fact, only up to 24% of these nutrients were retained after HTC. Note that despite the low recovery, the content of N_i in the hydrochar is consistent with other hydrochars, and even higher in the case of NO₃⁻-N (82 mg kg⁻¹) [1,18]. The NO₃⁻ content is particularly relevant for rapid plant fertilization. Thus, as reported by Paneque et al. [18], a favorable ratio between plant-available, fast-released N and an immobilized organic N that is slowly mobilized is a notable characteristic for a soil amendment that can act as a potential medium to long-term fertilizer. In our case, given the N_i contents observed, the hydrochar appears suitable to supply an initial pulse of readily available N at the onset of fertilization, while also contributing a pool of slow-release N for sustained medium- to long-term nutrition.

Another form of organic matter that is expected to be easily bio-accessible represents DOC and DON. As demonstrated in the study of Bargmann et al. [41], this can increase microbial activity and may prime the overall turnover of soil organic matter. Similarly, as described in the previously cited work, the potential immobilization of labile N in microbial biomass should also be considered. Thus, the contents of DOC and DON of soil amendment can be seen as an important parameter for estimating their fertilization efficiency. In CM the DOC and DON contents were 10.8 ± 0.46 g kg⁻¹ and 3.25 ± 0.11 g kg⁻¹, respectively, whereas these values accounted to 3.48 ± 0.11 g kg⁻¹ and 0.54 ± 0.02 g kg⁻¹ in the hydrochar.

The δ¹³C signature of CM is with −21.14‰ less negative than for soils with input of plants that follow the C3 photosynthetic pathway (−24 to −30‰ vs. VPDB) (Table S1), which is best explained by the fact that the chickens were fed with corn, a C4 plant. HTC shifted, the δ¹³C compositions towards −19.75‰ vs. VPDB, although without significant differences. It was proposed that changes in the isotopic signature of δ¹³C during charring may result from the preferentially loss of heavy [42] or light [43] C components through CO₂ emissions. It was also suggested that condensation and polymerization processes lead to more stable aromatic structures and heavier isotopically C due to the persistence of certain components [44]. However, the behavior of the atomic ratios during HTC cannot support such condensation and polymerization processes. The HTC treatment of CM resulted in a change in the δ¹⁵N values from 4.85 to 7.61‰ vs. air-N₂, which aligns with those of Reza et al. [45]. They proposed that a shift toward a heavier δ¹⁵N composition may be explained by the decomposition of amino acids and the incorporation of N into organic N-heterocyclic compounds through Maillard reactions.

3.3.2. Macronutrient and Micronutrient Analysis

The CM was characterized by a higher concentration of Ca compared to the other analyzed macro-nutrients (

Table 3. Concentration of macro-nutrients (P, K, S, Ca, and Mg) and extractable P, K, and S of chicken manure and its respective hydrochar, as well as the recovery rates after the HTC treatment at 250 °C for 0.5 h. The values represent the mean ± standard error (SE) for n = 3. The values followed by different letters indicate significant differences according to the statistical test used (T-Student or U Mann–Whitney). Levels of significance: *** p ≤ 0.001.

Element (g kg−1)	Chicken Manure	Hydrochar	p-Value	Recovery (% DM)
P	2.75 ± 0.10 a	1.04 ± 0.02 b	***	34.0 ± 0.62
Extr-P	0.59 ± 0.01 a	0.02 ± 0.00 b	***	2.55 ± 0.07
K	4.59 ± 0.18 a	0.60 ± 0.02 b	***	11.7 ± 0.33
Extr-K	4.60 ± 0.12 a	0.25 ± 0.01 b	***	4.95 ± 0.12
S	1.33 ± 0.05 a	0.24 ± 0.01 b	***	16.1 ± 0.32
Extr-S	0.49 ± 0.01 a	0.04 ± 0.00 b	***	7.28 ± 0.06
Ca	40.3 ± 1.31 a	32.1 ± 1.09 b	***	71.5 ± 2.42
Mg	1.84 ± 0.04 a	0.92 ± 0.02 b	***	44.8 ± 0.94

). These concentrations as well as those of S, fall within the ranges given in the literature for a manure-type raw material [1,4,9,33,46]. However, the values for P, K, and Mg can be low [1,9,21,33].The HTC treatment did not result in an enrichment of these macro-nutrients (Table 3). Conversely, losses were recorded for P (>65%), K, and S (both > 80%). This observation may be related to the high solubility of these elements in water, facilitating their loss with the liquid phase after HTC (Table 3). Because Ca can enhance P immobilization by complexing for example orthophosphate, a lower P loss during the process was expected [2]. Note that Ca has been documented as one of the key factors in controlling the solubility of P and, therefore, in its immobilization and prevention of loss during HTC [14,15].

Due to its high solubility, K losses of up to 90% from the solid phase were reported during HTC of chicken and cow manure [9,21,33]. In fact, in our study, 100% of K present in the CM was soluble (Table 3). Nevertheless, despite the low final concentration of K in the hydrochar, the ratio of soluble K to total K was 42.4% (Table 3). This indicates that HTC resulted in a stabilization of K. Similar to what was suggested for N, a balanced ratio between K that is rapidly plant accessible and immobilized K that is slowly released would be a desirable property for a soil amendment used also as a fertilizer. Concerning S, the literature reveals high variability with respect to recovery rates for manure-type feedstocks that were pyrolyzed under conditions similar to ours. Mau et al. [46], using poultry litter as feedstock, and Hejna et al. [36], applying CM, reported S recovery rates ranging from 65% to 75% following HTC at 250 °C or 240 °C, respectively, for 0.5 h. However, Li et al. [4] obtained a S recovery rate of less than 32% following HTC of CM or swine manure at 240 °C for 10 h. Therefore, the S losses recorded in our study resemble those obtained with longer mean residence times. The macro-nutrients that showed the highest recovery rates were Ca and Mg, which are among the elements that tend to concentrate most after HTC processing of manure-derived feedstocks [1].

In terms of micro-nutrients, the Fe content of CM exceeded those of Cu, B, Mn, and Zn (

Table 4. Concentration of micro-nutrients (Fe, Cu, B, Mn, Zn, Ni, and Mo) of chicken manure and its respective hydrochar, as well as the recovery rates after the HTC treatment at 250 °C for 0.5 h. The values represent the mean ± standard error (SE) for n = 3. The values followed by different letters indicate significant differences according to the statistical test used (T-Student or U Mann–Whitney). Levels of significance: ** p ≤ 0.01; *** p ≤ 0.001.

Element (mg kg−1)	Chicken Manure	Hydrochar	p-Value	Recovery (% DM)
Fe	6889 ± 263 a	4206 ± 76.3 b	***	54.7 ± 0.99
Cu	66.0 ± 1.88 a	31.5 ± 3.16 b	***	42.7 ± 4.28
B	10.0 ± 0.28 a	2.35 ± 0.11 b	***	20.9 ± 0.99
Mn	137 ± 4.36 a	60.7 ± 0.58 b	***	39.6 ± 0.38
Zn	184 ± 5.93 a	72.9 ± 2.04 b	***	35.5 ± 1.01
Ni	7.22 ± 0.32 a	3.87 ± 0.07 b	***	48.0 ± 0.93
Mo	1.39 ± 0.05 a	0.81 ± 0.09 b	**	51.8 ± 5.86

), a trend commonly reported for other manure-derived biomasses [1]. Whereas most of the analyzed micro-nutrients typically increase after an HTC treatment at temperatures near 250 °C [1,38], the hydrochar produced in our study does not conform to this pattern. Consistent with what was observed with macro-nutrients, the recovery values of micro-nutrients are predominantly below 50% (Table 4). Reza et al. [10], following HTC at 260 °C for 5 min, also observed a significant loss of these micro-nutrients in the solid product, although they used lignocellulosic biomass. Of note, the essential plant micro-nutrients Zn and Cu in the hydrochar do not exceed the concentration thresholds set by European regulations for HMs in solid organic fertilizers [47] or the limits outlined in the soil protection act [3].

3.3.3. Other Metals and Heavy Metals Analysis

Monitoring the fate of HMs as well as other potentially toxic metals such as Na, Al, Ba, Li, or Sr is crucial due to their potential hazardous effects on ecosystems and their impact on the food chain if used as a soil amendment. Compared to other elements, the HMs As, Cd, Co, Cr, Hg, Pb, and V have often been found to be more resistant against removal with the aqueous phase of the HTC process [3].

The concentrations of As, Pb, Cd, Hg, and Co in the CM (

Table 5. Concentration of other metals (Na, Al, Ba, Li, and Sr) and heavy metals (As, Cd, Co, Cr, Hg, Pb, and V) of chicken manure and its respective hydrochar, as well as the recovery rates after the HTC treatment at 250 °C for 0.5 h. The values represent the mean ± standard error (SE) for n = 3. The values followed by different letters indicate significant differences according to the statistical test used (T-Student or U Mann–Whitney). Levels of significance: p > 0.05 (“ns”, not significant differences); ** p ≤ 0.01; *** p ≤ 0.001.

Element (mg kg−1)	Chicken Manure	Hydrochar	p-Value	Recovery (% DM)
Na	1208 ± 31.3 a	231.4 ± 4.49 b	***	17.2 ± 0.33
Al	6031 ± 114 a	3425 ± 42.9 b	***	50.9 ± 0.64
Ba	26.3 ± 0.78 a	10.0 ± 0.17 b	***	34.1 ± 0.59
Li	4.95 ± 0.23 a	2.82 ± 0.07 b	***	51.0 ± 1.19
Sr	26.3 ± 0.58 a	15.1 ± 0.43 b	***	51.3 ± 1.47
As	3.42 ± 1.14 a	1.11 ± 0.57 a	ns	29.0 ± 14.9
Cd	0.14 ± 0.04 a	0.20 ± 0.03 a	ns	129 ± 19.4
Co	1.69 ± 0.13 a	0.96 ± 0.06 b	**	50.8 ± 3.31
Cr	22.0 ± 0.76 a	15.4 ± 0.26 b	***	62.6 ± 1.06
Hg	1.19 ± 0.29 a	0.87 ± 0.27 a	ns	65.4 ± 20.1
Pb	11.9 ± 0.60 a	6.63 ± 0.38 b	**	50.0 ± 2.88
V	13.9 ± 0.49 a	8.16 ± 0.18 b	***	52.7 ± 1.14

) fall within the range reported for pig and chicken manures by Luutu et al. [21]. In their study, they observed an enrichment of these elements following HTC at 260 °C and 1 h residence time. This was only observed for Cd in this study. In contrast, all other metals demonstrated a significant and effective removal, similar to the study conducted by Reza et al. [10], on HM removal by the HTC treatment of lignocellulosic material at 260 °C for 5 min. In our work, the recovery of HM and metals such as Al and Sr was slightly higher (around 50%) compared to the other metals studied (Table 5). This might be due to their restricted mobility, high boiling points, and the transformation of organic matter [1,3]. Conversely, Na emerged as the element with the highest losses during HTC, primarily due to its tendency to partially enter the aqueous phase owing to its high solubility in water [38].

Remarkably, within the hydrochar, none of the elements exceed the concentration thresholds established by European regulations for HMs in solid organic fertilizers, excluding those not encompassed by this framework [47]. Yet, when conducting risk assessments, the pollutant’s loading rate holds more significance than its concentration. Hence, a substantial impact of these concentrations of HM on the phytotoxicity of hydrochar is not anticipated.

3.4. Solid-State NMR Spectroscopy

3.4.1. Solid-State ¹³C NMR Spectroscopy

The solid-state ¹³C NMR spectrum of CM (Figure 1A) is primarily dominated by signals in the O-alkyl C chemical shift region (60–110 ppm), which embraces resonances of C in simple and complex carbohydrates, alcohols, and ethers [19]. Thus, O-alkyl C represented 50% of the total ¹³C intensity. Due to the presence of plant residues mixed within the manure, contributions of cellulose and the propanyl side chain of lignin derivatives are expected [48]. Indeed, the main signals observed around 72–75 ppm is best assigned to C2, C3, and C5 from cellulose, as well as C from xylans. Additionally, peaks at 62–65 ppm can be attributed to the amorphous and crystalline domains (C6) of cellulose. Signals at 83 and 88 ppm, likewise, arise from the non-crystalline and crystalline components of C4, whereas the signal at 104 ppm originates from the anomeric C1 of cellulose. The chemical shift region between 0 ppm and 45 ppm is assigned to alkyl C and contributes with 23% the total ¹³C intensity (Figure 1B; Table S2). This region exhibits a prevalence of methyl groups (CH3) and methylene (CH2) chains (peaks at 23, 30–32 and 40 ppm) as they occur in acetylated sugars, peptides, or lipids. The N-alkyl/methoxyl C region between 45 ppm and 60 ppm indicates the presence of C–N in amino acids or amino sugars but may also contain contribution of methoxyl C in lignin. Aryl C and phenolic C contribute with 7% and 3%, respectively (Figure 1B; Table S2), to the total ¹³C intensity. They resonate in the region between 110–160 ppm of the CM spectrum (Figure 1B), showing peaks at 128, 137 and 154 (C in N-heterocyclic and/or aromatic C of lignin like phenolic syringyl-like compounds) [48,49]. Finally, the carboxyl C region (160–185 ppm) comprises 7% to the total ¹³C intensity (Figure 1B; Table S2) and shows a peak at 172 ppm, attributable to C in carboxyl groups.

As anticipated, the HTC process led to decarboxylation, evident through the low recovery of carboxyl C of 2.5% (Figure 1B; Table S2). During this process, O-alkyl and N-alkyl/methoxyl C degraded, whereas aromatic C, seemingly formed through the dehydration and cyclization of carbohydrates and peptide material [18]. Thus, the relative contribution to total ¹³C intensity in both chemical shift regions experienced an increase from 7% to 21% and from 3% to 6%, respectively (Figure 1B; Table S2), after HTC treatment. This result is consistent with the observed decrease in the O/C and H/C ratios.

The signals in the O-alkyl and N-alkyl/methoxyl C region of the hydrochar spectrum peak at the same chemical shift as in that of the CM (Figure 1A), although their intensity differs from the former. The total ¹³C intensity for both chemical shift regions varied from 50% and 10%, respectively, in CM to 21% and 7% in hydrochar (Table S2). However, in the aryl C region amounting to 21% of the total ¹³C intensity (Table S2), the peak at 128 ppm evolves to the main signal of the region (Figure 1A). Similar to the increase in the aromatic C intensity, the contribution of the chemical shift region of alkyl C to the total ¹³C intensity increased from 23% to 40% (Table S2). An increase in this region following HTC was also reported by Cao et al. (swine manure; 250 °C, 20 h) [17] and Paneque et al. (sewage sludge; 260 °C, 0.5 h) [18].

Considering the C recovery of the hydrochar and, consequently, the different C groups after the HTC process, no net lipid synthesis or degradation occurred. Rather, a relative enrichment was observed, attributed to the loss of other compounds. Carbohydrates, notably impacted in this instance, are preferentially degraded during thermal processes, consequently causing a relative enrichment among all other classes of compounds [17].

3.4.2. Solid-State ¹⁵N NMR Spectroscopy

In the solid-state ¹⁵N NMR spectrum of CM, a signal at −260 ppm dominates the amide N chemical shift region (−248 to −285 ppm) (Figure 2A), contributing with 83% to the total ¹⁵N intensity (Figure 2B; Table S3). This intensity is mainly caused by peptide-N and, to a lesser extent, by amide-C in amino sugars. The ~10% contribution to the total ¹³C intensity recorded in the chemical shift region between 60 and 45 ppm in the ¹³C NMR spectrum of CM may be related to the N-alkyl C of these amides. However, given the possible presence of some plant residues in the CM, it is likely that, although to a small extent, the methoxyl C from lignin is also contributing to this signal.

The remaining 17% of the total ¹⁵N intensity in the ¹⁵N NMR spectrum of CM occurs in the pyrrole N region (−150 to −248 ppm) (Figure 2B; Table S3). Here, a small signal at −225 ppm appears and can be assigned to pyrrole-N [18] but may also be indicative for uric acid. Another anticipated peak at −346 ppm, corresponds to terminal amino N in peptides or amino sugars (Figure 2A).

After the HTC process, the pyrrole N region increases to 65% of the ¹⁵N NMR spectrum of the hydrochar (Figure 2A,B; Table S3), which is in accordance with the formation of N-heterocyclic aromatic structures such as indoles, imidazoles, and pyrroles during the carbonization process. Indeed, in the ¹⁵N NMR spectrum of the hydrochar, a distinct signal appears at −245 ppm, corresponding to the chemical shift of Indole N (Figure 2A).

HTC led to a reduction of the ¹⁵N intensity in the amide N region of the hydrochar spectrum, from 83% in that of the raw material to 35% in that of the hydrochar (Figure 2B; Table S3). Although the main peak at −260 ppm observed in the ¹⁵N NMR spectrum of the CM can still be identified in the one of hydrochar, the additional signal at −255 ppm in the latter evidences the chemical changes that have occurred during HTC. Despite lacking defined peaks assignable to carbazole N (around −262 ppm) and benzamide-type N (−275 ppm), these could potentially contribute to the shoulder of the amide N resonance [18].

3.5. Assessment of Phytotoxicity and Fertilizing Potential

3.5.1. Germination and Seedling Growth

For lettuce, the application of hydrochar had no impact on the studied germination parameters except for VI. Its value is higher for both hydrochar treatments compared to CTR (

Table 6. Germination parameters of lettuce, sunflower and tomato seeds between treatments. Germination (G), mean germination time (MGT), mean germination rate (MGR), time to 50% germination (T50), synchronization index (Z), uncertainty of germination process (U), and the Vigor index (VI). The values represent the mean of five replicates (n = 5) ± standard error (SE). Four seeds per pot were sown, resulting in 20 seeds per treatment. The values followed by different letters in columns indicate significant differences according to Tukey’s test. Levels of significance: p > 0.05 (“ns”, not significant differences); * p ≤ 0.05. “Homogeneous group” statistics was calculated through one-way ANOVA test.

Treatments		G (%)	MGT (Days)	MGR (Day−1)	T50 (Days)	Z (Unit Less)	U (Bit)	VI
Lettuce	CTR	87.5 ± 3.95	2.81 ± 0.15	0.36 ± 0.02	2.07 ± 0.17	0.42 ± 0.09	1.21 ± 0.21	46.1 ± 2.44 b
	HC-3.25	85.0 ± 2.50	2.66 ± 0.21	0.39 ± 0.03	1.70 ± 0.06	0.56 ± 0.11	0.91 ± 0.24	58.4 ± 2.58 a
	HC-6.5	85.0 ± 2.50	2.72 ± 0.19	0.37 ± 0.03	1.98 ± 0.16	0.43 ± 0.04	1.14 ± 0.11	60.8 ± 4.00 a
p-value		ns	ns	ns	ns	ns	ns	*
Sunflower	CTR	77.5 ± 2.50	3.57 ± 0.17 b	0.28 ± 0.01 a	3.02 ± 0.18 b	0.51 ± 0.13 a	1.21 ± 0.10 b	609 ± 27.6 ab
	HC-3.25	69.3 ± 6.27	4.51 ± 0.24 a	0.22 ± 0.01 b	3.97 ± 0.30 a	0.12 ± 0.05 b	1.93 ± 0.15 a	540 ± 44.9 b
	HC-6.5	77.5 ± 4.68	3.84 ± 0.19 b	0.26 ± 0.01 ab	2.83 ± 0.11 b	0.43 ± 0.08 a	1.14 ± 0.17 b	671 ± 25.5 a
p-value		ns	*	*	*	*	*	*
Tomato	CTR	83.3 ± 0.00 a	9.52 ± 0.56	0.11 ± 0.01	8.50 ± 0.89	0.26 ± 0.07 a	1.48 ± 0.22	185 ± 14.8 a
	HC-3.25	86.7 ± 6.24 a	9.64 ± 0.89	0.11 ± 0.01	8.98 ± 1.22	0.33 ± 0.09 a	1.35 ± 0.26	187 ± 15.8 a
	HC-6.5	70.0 ± 3.33 b	10.4 ± 0.52	0.10 ± 0.00	9.90 ± 0.58	0.03 ± 0.03 b	1.96 ± 0.13	116 ± 8.80 b
p-value		*	ns	ns	ns	*	ns	*

). The hydrochar-treated plants showed longer hypocotyl lengths compared to the CTR already during the first days of the germination period (Figure 3A). This aligns with the VI results, as this index depends on germination (G) and hypocotyl length at 7 DAS (Table 6). Only HC-3.25 treated plants were higher than the CTR plants at 20 DAS (Figure 3B). Zhang et al. [50] reported that during early growth stages, when lettuce plants exhibit high relative growth, the rate of N uptake increases over time. Consequently, our results may indicate a deficiency of available N during this early growth phase of the CTR plants, negatively impacting their growth.

In the case of sunflower plants, the HC-3.25 treatment caused delayed germination, as evidenced by the longer time required for seeds to emerge (MGT), the longer time needed until 50% of the seeds germinated (T50), and the degree of uncertainty associated with the distribution of the relative frequency of germination (U) (Table 6). Consequently, this treatment exhibited a lower speed of germination (MGR), and less synchronization of the process (Z) compared to the other treatments. Although HC-3.25 exhibited the lowest VI value, its value did not significantly differ from those of the other treatments (Table 6), thus indicating that this germination delay had no significant impact on seedling growth. However, the germination delay is not attributed to deleterious phytotoxicity of the hydrochar, as higher doses did not promote this effect. Moreover, far from having a negative effect, HC-3.25 treated plants were the only ones that were significantly higher at 20 DAS compared to those of CTR (Figure 3D).

In contrast, the application of the HC-6.5 dose to the pots with the tomato plants adversely affected the G percentage and Z, although this was not seen with the HC-3.25 dose. These results are consistent with those obtained by Suarez et al. [22], reporting that the addition of a 5% (DW) dose of hydrochar derived from garden and park waste affected tomato seed germination negatively, both in soil and peat-based growth media. The diminished VI of the HC-6.5 plants indicates a phytotoxic effect of this dose for this species. In fact, these plants experienced reduced hypocotyl and plant development throughout the experiment (Figure 3E,F). Considering that the physical (pH, EC) and chemical (nutrients and HMs) characteristics of the hydrochar do not reach values that could negatively affect germination and seedling growth, all these effects may be explained by the presence of toxic organic species such as furfural, furans, or phenolic compounds [21,22,51] that were not removed with the liquid after the HTC. Furthermore, the fact that the effects are only evident at HC-6.5 doses and in tomato plants suggests that this species is more sensitive to potential phytotoxins compared to the other two studied.

The documented effects of hydrochar on germination and growth range from inhibition to stimulation and may vary depending on the feedstock, hydrochar dosage, and the studied species [7,21,52]. As observed in other studies, both the hydrochar dosage and the tested plant species proved crucial in determining hydrochar’s impact on seed germination and early growth in this research.

3.5.2. Biomass and Efficiency of Photosystem II

In the early seedling stage (7 or 14 DAS), none of the studied plants showed significant reductions of the produced shoot dry biomass, although the HC-6.5 tomato plants began to exhibit a negative impact of this treatment (Figure 4C). At 20 DAS, the HC-3.25 treatment emerged as the one that yielded improved shoot dry biomass in lettuce and sunflower plants compared to CTR (Figure 4A,B). Hydrochar has previously been documented to enhance plant growth and productivity through the release of nutrients [7,53]. A suitable NO₃⁻ content, leading to a favorable inorganic N to organic N ratio such as that found in the used hydrochar, allows short- and medium-term fertilization, as also proposed by Paneque et al. [18]. However, it is worth to highlight that on the long-term, organic N components facilitating a slow release of available N is a further benefit of charred residues [54]. Remarkably, by increasing the application with HC-6.5 and hence the NO₃⁻ content, greater or similar plant biomass yields were not obtained in relation to HC-3.25. This may be due to the compensation of the positive effect of increased N-availability by an increase in potential phytotoxins that are non-deleterious but are slightly inhibitory on the growth of lettuce and sunflower plants [21]. This suggest that with lower doses, a better balance between both effects can be achieved. In fact, this may explain why the HC-6.5 sunflower plants exhibited the lowest QY_PSII values at 7 and 14 DAS, indicating a higher degree of physiological stress without compromising biomass generation compared to the CTR plants (Figure 4E). Conversely, at 14 DAS, among sunflowers, plants treated with hydrochar achieved the highest QY_PSII, aligning with the greatest shoot biomass production at 20 DAS. Paneque et al. [55] also observed that sunflowers treated with biochar exhibited higher efficiency of photosystem-II compared to those grown in unamended substrates, which was associated with increased growth and yield. However, the trend we observed in sunflowers was not replicated in lettuce plants treated with hydrochar, as no significant differences in QY_PSII compared to the CTR plants were recorded throughout the experiment (Figure 4D).

In line with our germination results and the study by Suarez et al. [22], HC-6.5 tomato plants experienced a drastic reduction of the shoot biomass at 25 DAS because of the physiological stress induced by this dose (Figure 4C). The plant stress was further evidenced by the drastic decline of QY_PSII recorded during the experiment (Figure 4F). The decrease in growth, photosynthetic efficiency, and consequently the inhibition of biomass generation is among the primary responses of plants to stressful situations [55]. This result, along with those obtained for germination and growth, highlights again the high sensitivity of this species to higher doses of hydrochar. This certainly must be considered when developing a soil application strategy for improving crop performance using hydrochar.

4. Conclusions

The present research represents a case study on the potential of processing CM from extensively barn-raised chickens into hydrochar through HTC and evaluates the suitability of the obtained hydrochar for fertilization purposes. The HTC pyrolysis was chosen since it provides hygienization concomitantly with sufficient stabilization of organic matter to provide a soil amendment with slow N/P/K-release fertilizer capacities. Since we evidenced a considerable loss of exactly those nutrients during HTC, we recommend the development of strategies for their recovery from the discarded process water to improve the recycling efficiency. Nevertheless, despite this loss, the N_i content of the obtained hydrochar should be sufficient for healthy early plant development. The remaining organic N, on the other hand is protected from leaching by incorporation in organic forms that can be steadily mobilized by microorganisms for its use by the growing plants. This equally applies to K reflecting a soluble-to-total K ratio that should allow a steady but slow release of this element during hydrochar aging in soils. These conclusions were supported by the pot experiments with tomato plants, lettuce and sunflower providing preliminary evidence of the growth-promoting capacity of the tested hydrochar in particular for lettuce, even at the relatively low dosage of 3.25 t ha⁻¹. Bearing in mind that transformation and chemical reactions of the organic fraction during HTC depend on the nature rather than on the quantity of the organic matter in the feedstock, it may be concluded that comparable favorable distributions of N and K forms can be achieved with CM of large-scale chicken farms. However, thresholds for maximal application dosages still must be evaluated to avoid surplus fertilization and to remain below the allowed fertilization limits provided by EU or other comparable regulations. Note that concentrations of plant-specific phytotoxins have also to be considered.

Overall, our findings indicate that HTC represents a promising strategy to valorize and stabilize CM, by converting it into a nutrient-bearing material with potentially reduced phytotoxicity. As shown in our study this does not only apply to material derived from large commercialized and automatized farms described in the literature but also to feedstock produced by small-scale free-range breeders. Since our work is understood as a first step approach, further research is certainly needed to validate the results under field conditions and over longer time scales, including a wider range of application doses and plant species, to assess nutrient dynamics and long-term agronomic effects.