Elemental Content and Distribution in Various Willow Clones and Tissue Types

Abstract

Willows (genus Salix) are versatile plants with applications in construction, medicine, and biomass fuel in North America. Advances in breeding have improved willow clones for higher yields and pest resistance, but the chemical content and distribution across different plant parts remain poorly understood. This study examined the variation in chemical elements (carbon, hydrogen, nitrogen, sulfur, chlorine, and ash) across six willow clones (India, Jorr, Olof, Otisco, Preble, and Tora) and three tissue types (wood, bark, twigs). We also compared freeze-drying and oven-drying methods to assess their impact on chemical content. Freeze-dried samples generally exhibited higher carbon and hydrogen concentrations than oven-dried samples, with statistically significant differences primarily observed for carbon, while nitrogen showed no overall significant difference between drying methods. Chemical composition varied among clones, although no single clone consistently dominated across all chemical parameters. In contrast, pronounced tissue-type differences were observed: bark had higher nitrogen, carbon, sulfur, chlorine, and ash contents, whereas wood exhibited relatively higher hydrogen concentrations, with twigs showing intermediate values. These findings suggest that accounting for tissue-specific chemical differences can improve the selection and utilization of willow biomass and increase the accuracy of ecological assessments, including carbon storage estimates. The findings of this study indicate that oven-drying should remain in use within the bioenergy sector, whereas freeze-drying ought to become the preferred standard for carbon-accounting protocols.

1. Introduction

Willows (Salix spp.) belong to the family Salicaceae, which also includes poplars. Globally, there are approximately 450–520 willow species [1], with about 125 species native to North America [2]. Natural hybrids and multi-hybrid combinations further increase taxonomic complexity [3]. Willows exhibit diverse growth forms, ranging from prostrate dwarf shrubs to trees exceeding 40 m in height. They are primarily distributed across the temperate and arctic regions of the Northern Hemisphere, with some species occurring in subtropical and tropical zones [4]. Ecologically, willows occupy a wide range of habitats, from wetlands to upland sites.

Historically, willows have been utilized for various purposes, including basketry, construction of shelters and fences, and manufacture of arrow shafts and fish traps [4,5]. Traditionally, willow wood has served primarily as biomass fuel [6]. Biomass refers to renewable organic material derived from living organisms, and woody biomass constitutes a major fraction of the global biomass supply. The Sustainable Forest Development Act (Quebec, 2009) defines forest biomass as unmerchantable woody material generated through forest management or short-rotation plantations for energy production, excluding stumps and roots [7]. Willow bark and leaves possess medicinal properties, notably for treating rheumatism, due to the presence of salicin—a glucoside now synthesized as the active ingredient in aspirin [8]. Phenolic glycoside concentrations are highest in juvenile twigs, which have a greater bark proportion [9]. Bark typically constitutes about 19% of a 12-year-old willow stand [10] and approximately 14% in two-year-old shrub willow clones [11].

In northern latitudes, willows are well suited for bio-based value chains due to their high productivity, vigorous regrowth after harvest, and ease of cultivation as short-rotation coppice [12]. Breeding programs have produced hybrids with improved stem form, higher yields, and enhanced resistance to pests and diseases [13]. These hybrids exhibit up to 60% greater productivity, reduced rust incidence, and lower shoot damage from insects [14]. Furthermore, they perform well on marginal land with poor soils [12,15,16], making them attractive for biomass production without competing with food crops [17].

Perennial grasses are alternate sources of biomass in northern latitudes. Two perennial grasses that produce relatively high yields and are suitable for cold climates are reed canary grass (Phalaris arundinacea L.) [18] and switchgrass (Panicum virgatum L.), although the latter is better adapted to regions south of approximately 55° N latitude, extending from Saskatchewan to Nova Scotia and into the eastern United States [19]. Miscanthus, mainly Miscanthus giganteus cultivars, is a third grass that is often used for biomass production, but not as well suited to cold climates. It can be very productive, 15–30 Mg dry matter ha⁻¹ yr⁻¹ [20], but at northern latitudes the risk of winter mortality increases if soil temperature drops below approximately −8 °C and snow cover is minimal [21]. Perennial grasses generally contain higher ash concentrations and greater proportions of problematic inorganic constituents than short-rotation woody crops such as willow. Numerous studies have shown that grasses typically accumulate elevated levels of silicon, potassium, sodium, chlorine, and nitrogen, which are the primary drivers of slagging, fouling, and corrosion in thermochemical conversion systems [22,23,24].

In contrast, short-rotation coppice willow consistently exhibits lower ash contents and markedly lower concentrations of silica, alkali metals, chlorine, and nitrogen, resulting in more stable combustion and gasification behavior and reduced emissions-control requirements [25]. Although willows typically contain higher lignin concentrations than herbaceous feedstocks such as perennial grasses and cereal residues, this can be advantageous for thermochemical conversion because lignin is a major contributor to char formation [26]. In addition, willow’s comparatively low ash content reduces the loading of ash-forming inorganics that drive slagging, fouling, and corrosion and can impair gasification performance, supporting cleaner solid and gas product streams and simpler gas/solid cleanup [27].

Willow wood is light, with a specific gravity of 0.34–0.41, and is generally straight-grained, odorless, and tasteless. Although weak in bending and compression, it exhibits moderately high shock resistance [28]. Chemical composition varies among clones and environments. Baker et al. [3] reported cellulose contents ranging from 36% to 65% and lignin contents from 17% to 19% across 31 willow varieties. Serapiglia et al. [29] observed narrower ranges for cellulose (37.4–45.3%), hemicelluloses (31.1–34.9%), and lignin (20.3–23.2%) in 30 clones, noting significant clonal and site-specific variation. Ray et al. [30] demonstrated differences in composition, enzymatic saccharification, pretreatment response, and projected ethanol yields, highlighting willow’s potential as a lignocellulosic energy crop. Dou et al. [31] further characterized lignin structure, finding guaiacyl units predominant in bark, while syringyl and p-hydroxyphenyl units were more abundant in inner bark and wood.

In contrast, fewer studies have examined elemental composition. On an oven-dry mass basis, Stolarski et al. [32] reported carbon (C), hydrogen (H), and sulfur (S) contents of four clones ranging from 50.8–52.2%, 6.8–7.1%, and 0.03–0.05%, respectively. In a study of three willow clones, Personen et al. [33] reported C, H, and nitrogen (N) content by weight to range from 47.4% to 47.8%, 5.7% to 6.2%, and 0.1% to 0.2%, respectively. Personen et al. [33] also found that chlorine (Cl) content varied between the combined stem wood and bark samples (5.1 mg kg⁻¹ for Salix myrsinifolia and 21.6 mg kg⁻¹ for Salix schwerinii), while the chlorine content in wood samples alone was below 0.5 mg kg⁻¹. The elemental S content in the combined stem wood and bark ranged from 250 to 300 mg kg⁻¹, while in the stem wood alone, it ranged from 130 to 160 mg kg⁻¹ across the three clones. These findings underscore the need for further characterization of willow biomass to support its utilization in bioenergy and bioproduct applications.

Accurate estimation of carbon content (CC) in woody biomass is increasingly important with the implementation of carbon taxes and credits [34]. The common assumption that carbon constitutes 50% of a tree’s above-ground biomass introduces significant uncertainty. Because elemental composition varies by species, tissue type, age, sampling height, and growing conditions, detailed characterization is essential for refining forest carbon stock estimates [35,36,37]. Nitrogen concentration, for example, varies markedly from bark to pith and exhibits greater variability among woody tissues than carbon [36,38,39]. These concerns extend to other elements, including hydrogen, sulfur, chlorine, and ash, which are critical for biogeochemical modeling and energy conversion processes [39].

The major elements influencing wood calorific value are C, H, N, S, and oxygen, which can be quantified through ultimate analysis or predicted from proximate analysis [40,41]. The calorific value of a species, which measures the energy chemically stored in its biomass and can be transformed into usable energy [41], is crucial for its thermochemical conversion. In its simplest form, burning biomass can be considered the combustion of only carbon and hydrogen [42], with the calorific value increasing as the proportion of carbon and hydrogen rises [43]. In this idealized case, and assuming 100% efficiency, all carbon is oxidized to carbon dioxide, and all hydrogen is oxidized to water [42]. However, complete efficiency is theoretical, as volatile gases, known as hydrocarbons, are also produced from carbon and hydrogen. Unlike C and H, Cl and ash do not contribute to calorific value. Chlorine is problematic because it promotes corrosion and harmful emissions during combustion, while ash represents the non-combustible mineral fraction, reducing energy density and causing operational issues such as slagging and fouling [44].

The conventional oven-drying (OD) technique typically involves two stages: transferring water from the interior (high humidity) to the exterior (low humidity), followed by evaporation from the surface [45]. In contrast, the lyophilization or freeze-drying (FD) process consists of three stages: initial freezing, which crystallizes the free water; primary drying, which removes solid water via sublimation; and secondary drying, which eliminates most of the remaining moisture through desorption of adsorbed water [46]. The drying process affects the CC in woody material, with oven-drying resulting in lower CC compared to freeze-drying [35,36,37]. Mvolo et al. [36] found that freeze-dried H concentrations were significantly higher than those of oven-dried samples in white spruce (Picea glauca), while N concentrations did not differ significantly between freeze-dried and oven-dried samples.

Although willow (Salix spp.) has been studied for structural components and bioenergy potential, elemental composition across tissue types remains poorly documented. Existing research has focused primarily on carbon and, more recently, nitrogen and hydrogen in other species [35,36,47,48]. To our knowledge, no studies have examined variation in C, H, N, S, Cl, and ash content across tissues of willow clones grown in Canada or assessed the influence of drying methods.

Canada has an estimated 9.48 million hectares of marginal land suitable for energy crops, including 3.6 million hectares in Alberta [49,50]. With Canada’s commitment to net-zero emissions by 2050, willow offers significant potential for bioenergy and bioproduct development [51]. Understanding its chemical properties is essential for optimizing conversion processes and guiding breeding programs.

Wood properties are governed by chemical composition at three levels: (1) structural components of the cell wall (cellulose, hemicellulose, lignin) and associated molecules (extractives); (2) distribution of these components; and (3) their relative proportions across tissues [52]. A companion study examined structural components and their distribution in willow tissues [53]. The objective of this study was to quantify variation in elemental composition (C, H, N, S, Cl, and ash) across willow tissues (wood, bark, twigs) and to compare freeze-drying and oven-drying techniques.

2. Materials and Methods

2.1. Study Site

Willow wood samples were collected from two locations within a 340-hectare commercial willow plantation near Keoma, AB, Canada (Section 18: 51°13′8.61″ W, 113°29′11.65″ N; Section 27: 51°14′54.69″ W, 113°25′00.91″ N). The plantation is situated on well-drained Orthic Black Chernozem soils [54] formed on coarse-textured aeolian deposits. Surface soil textures (0–30 cm) range from loam to sand, and the landform varies from gently undulating terrain to low longitudinal dunes. According to the Canadian Land Suitability Rating System for Agricultural Crops [49], these soils are classified as Class 5, indicating very severe limitations for annual food crop production.

2.2. Planting Design

In spring 2016, stem cuttings (20 cm in length) from six willow clones (

Table 1. Samples description: Clones’ name; number of samples; planted area; number of stools sampled by clone; height and diameter averages and standard errors (in parentheses and in italic); and clones’ pedigree.

Clone Name	Number of Samples	Planted Area (ha)	Number of Stools Sampled	Average (+Stdev) Height (cm) †	Average (+Stdev) Diameter (mm) †	Pedigree of Clone
India ‡	15	1.11	3	561.7 (61.7)	59.8 (0.6)	S. dasyclados × S. dasyclados [56]
Jorr	15	1.08	3	342.9 (84.7)	24.3 (7.8)	S. viminalis × S. viminalis [56]
Olof	15	5.11	3	339.7 (79.4)	23.1 (5.2)	S. viminalis × (S. schwerinii × S. viminalis) [56]
Otisco	15	19.5	3	332.9 (63.0)	26.4 (9.5)	S. viminalis ‘SV2’ × S. miyabeana ‘SX64’ [57]
Preble	15	6.63	3	343.2 (51.7)	24.9 (5.0)	S. viminalis L. ‘SV2’ × (S. miyabeana ‘SX61’ × S. miyabeana ‘9970-037’) [56]
Tora	15	7.21	3	454.1 (63.7)	32.0 (5.4)	S. schwerinii × S. viminalis [56]

^† Height and diameter were measured from the tallest stem (plants are multi-stemmed). ^‡ Due to high mortality, only one plot was sampled.

) were planted using the conventional European offset double-row planting pattern [55]. This configuration consists of paired rows spaced 75 cm apart, with 150 cm alleys between double rows. Cuttings were spaced 60 cm within rows, resulting in an approximate planting density of 15,000 cuttings per hectare. Clones were planted in contiguous blocks of varying sizes (Table 1) distributed throughout the plantation. The plantation was not coppiced after the first growing season, contrary to standard practice.

2.3. Sample Collection and Preparation

In fall 2020, samples were collected from one contiguous planting block of each willow clone. For each clone, three stools were selected, and the tallest stem from each stool was harvested. Two tissue types—bark and stem wood—were sampled at two heights: ground level (0.3 m) and breast height (1.3 m). Additionally, twigs were collected at approximately 2.3 m to form a single branch sample per stool.

In total, 90 samples were collected for freeze-drying (6 clones × 3 stools × 2 tissues × 2 heights + 18 twigs), comprising 36 bark samples, 36 stem wood samples, and 18 twig (bark + wood) samples. A subset of 30 samples (6 clones × 1 stool × 2 tissues × 2 heights + 6 twigs) was collected for oven-drying. Thus, 120 samples were analyzed in this study (Table 1). In some cases, individual planting spots were empty due to stool mortality.

Harvested shoots from each clone were bundled, placed in 2-mil gauge plastic tubing, sealed at both ends, and stored at −3 °C until processing. All tissue samples were reduced to approximately half-matchstick dimensions, sealed in plastic bags, and stored in a freezer prior to grinding with a Wiley Mill (No. 20 mesh). Each sample was divided into two drying treatments:

• Freeze-drying: All 90 samples were freeze-dried at −50 °C for 7 days using an 8 L freeze-drying system (Labconco Co., Kansas City, MO, USA).
• Oven-drying: A subset of 30 samples was oven-dried at +70 °C for 3 days using a forced-air oven (Thermo Scientific, Lindberg/Blue M, Model MO1450A-1, Asheville, NC, USA).

2.4. Elemental and Ash Content Analyses

Powdered samples were analyzed for total carbon (C), hydrogen (H), nitrogen (N), and sulfur (S) using a LECO TruSpec CHNS Analyzer (LECO Instruments ULC, Mississauga, ON, Canada), calibrated with a certified EDTA standard. Quality assurance was maintained using certified reference materials and verification checks every 20 samples. Chlorine content was determined by mixing each sample with 0.75 M H₂SO₄ at a 1:10 ratio (sample: H₂SO₄) on a mechanical extractor for 1 h, followed by gravity filtration using Whatman No. 42 filter paper. The filtrate was analyzed using inductively coupled plasma optical emission spectrometry (ICP-OES; Spectro ARCOS FHM22, Kleve, Germany). Ash content was determined by combusting samples at 470 °C in a muffle furnace (Lindberg BF51842 PBC-1, Riverside, MI, USA) and calculated as:

$$\mathrm{Ash} \mathrm{content} (\%)={\displaystyle \frac{\mathrm{Weight} \mathrm{after} \mathrm{ignition}}{\mathrm{Weight} \mathrm{before} \mathrm{ignition}}}\times 100$$

This calculation was based on the published method for direct estimation of organic matter by loss on ignition [58], with a modified ignition temperature. Quality assurance included analysis of certified reference materials, duplicate measurements on every 10th sample, and instrument recalibration after each batch of 20 samples.

2.5. Statistical Analyses

Descriptive statistics were calculated for chemical constituents of willow tissues, including measures of central tendency (mean) and variability (standard deviation) within clones and tissue types. Data normality was assessed using the Shapiro–Wilk test, which indicated that the data were not normally distributed (p < 0.05). Consequently, non-parametric tests were employed to evaluate differences in chemical composition among groups.

For comparisons between two groups, the Wilcoxon rank-sum test was used. When more than two groups were compared, the Kruskal–Wallis test was applied, followed by Wilcoxon’s post hoc test for pairwise comparisons [59]. All statistical analyses were performed using R version 4.5.1 [60]. Statistical significance was determined at a 95% confidence level (p < 0.05) for all tests.

3. Results

3.1. Impact of Drying Technologies on Chemical Composition

Significant differences in the chemical composition of willow were observed between freeze-dried and oven-dried samples when tissues were pooled by clone. Freeze-dried samples exhibited higher C and H concentrations compared to oven-dried samples (

Table 2. Summary of chemical characteristics of Willow trees and standard deviations (in parentheses).

Clone	Tissue	FDN (%)	ODN (%)	FDC (%)	ODC (%)	FDH (%)	ODH (%)	FDS (mg/kg)	FDCl (mg/kg)	Ash (%)
India	Bark	1.09 (0.09)	1.09 (0.10)	50.4 (0.52) a	48.6 (0.07) b	5.86 (0.23)	5.68 (0.17)	792.0 (47.4)	11.6 (6.23)	5.00 (0.24)
	Twigs	0.90 (0.02)	1.10 (NA)	50.1 (0.31)	48.8 (NA)	6.21 (0.07)	6.06 (NA)	459.0 (NA)	52.9 (NA)	1.86 (NA)
	Wood	0.18 (0.04)	0.19 (0.00)	49.2 (0.19)	48.3 (0.14)	6.32 (0.05)	6.20 (0.04)	NA	25.8 (8.27)	0.77 (0.01)
Jorr	Bark	1.15 (0.14) a	1.55 (0.15) b	49.0 (0.53)	49.5 (NA)	5.87 (0.10)	5.95 (0.34)	899.0 (38.5)	142.0 (44.6)	3.92 (0.24)
	Twigs	0.94 (0.09)	0.86 (NA)	50.5 (0.90)	49.9 (NA)	6.20 (0.11)	6.13 (NA)	575.0 (NA)	69.1 (NA)	2.71 (NA)
	Wood	0.31 (0.03)	0.31 (0.01)	48.9 (0.36)	47.9 (0.14)	6.18 (0.04)	6.13 (0.00)	NA	21.5 (19.4)	0.68 (0.12)
Olof	Bark	1.24 (0.23)	1.33 (0.08)	50.5 (1.08)	49.8 (NA)	6.02 (0.18)	5.93 (0.20)	1014.0 (20.5)	42.8 (14.4)	2.92 (1.01)
	Twigs	0.82 (0.13)	0.88 (NA)	49.9 (0.30)	49.0 (NA)	6.11 (0.07)	6.06 (NA)	570.0 (NA)	43.3 (NA)	2.04 (NA)
	Wood	0.28 (0.02)	0.25 (0.00)	49.4 (0.35) a	47.7 (0.28) b	6.24 (0.05)	6.14 (0.07)	NA	21.9 (9.48)	0.72 (0.01)
Otisco	Bark	1.15 (0.12)	1.14 (0.06)	51.5 (0.94)	50.5 (0.14)	6.02 (0.21)	5.80 (0.08)	716.0 (26.2)	18.9 (11.2)	4.69 (0.33)
	Twigs	0.80 (0.06)	1.15 (NA)	50.7 (0.51)	50.4 (NA)	6.27 (0.02)	6.21 (NA)	354.0 (NA)	46.1 (NA)	2.48 (NA)
	Wood	0.25 (0.05)	0.29 (0.08)	49.6 (0.43)	48.7 (0.28)	6.26 (0.05)	6.15 (0.00)	NA	16.2 (5.30)	0.77 (0.06)
Preble	Bark	1.11 (0.08)	1.06 (0.14)	50.8 (0.72)	49.9 (NA)	5.97 (0.17)	5.79 (0.08)	676.0 (42.4)	50.6 (44.0)	4.80 (0.03)
	Twigs	0.69 (0.04)	0.89 (NA)	50.8 (0.35)	50.2 (NA)	6.28 (0.05)	6.20 (NA)	418.0 (NA)	22.7 (NA)	3.46 (NA)
	Wood	0.29 (0.02) a	0.26 (0.01) b	49.5 (0.35)	48.4 (0.42)	6.30 (0.12)	6.17 (0.06)	NA	18.6 (8.98)	0.89 (0.08)
Tora	Bark	1.33 (0.27) a	1.50 (0.04) b	50.5 (1.13)	49.2 (NA)	6.10 (0.20)	5.88 (0.22)	1224.0 (11.3)	85.4 (29.1)	5.13 (1.93)
	Twigs	1.08 (0.09)	1.01 (NA)	50.0 (0.00)	49.1 (NA)	6.17 (0.04)	6.15 (NA)	755.0 (NA)	96.5 (NA)	2.18 (NA)
	Wood	0.25 (0.08)	0.22 (0.03)	49.1 (0.39) a	47.7 (0.18) b	6.29 (0.06)	6.12 (0.02)	NA	26.5 (NA)	0.61 (0.04)

Note: Freeze- and Oven-dry sharing different letters are significant at p < 0.05. NA: Not Available.

; Figure 1). When analyzed by tissue and clone, differences between drying methods were generally weak. Significant differences in C were detected only for clones India, Olof, and Tora; and in N for clones Jorr, Preble and Tora.

3.2. Variation in Chemical Contents in Willow Clones

Variation in chemical composition was observed among the six willow clones under both freeze-dried (FD) and oven-dried (OD) conditions when tissues were pooled by clone (Figure 1; Table 2). Carbon concentrations were generally higher in FD samples (49.26–50.57%) compared to OD samples (48.41–49.76%). Hydrogen content showed minimal variation, with FD values of 6.06–6.19% and OD values of 5.96–6.06%. Nitrogen content ranged from 0.69% in India to 0.85% in Tora for FD samples, and from 0.71% in Preble to 0.92% in Jorr for OD samples. Sulfur content was highest in Tora (1067.67 mg kg⁻¹) and lowest in Otisco (595.00 mg kg⁻¹) for FD samples. Chlorine content also varied considerably, with Jorr exhibiting the highest value (79.42 mg kg⁻¹) and Otisco the lowest (23.28 mg kg⁻¹). Ash content ranged from 1.86% in Olof to 2.96% in Preble.

Otisco and Preble exhibited the highest freeze-dried carbon (FDC) values, followed by Olof, India, and Tora, while Jorr had the lowest. The Kruskal–Wallis test revealed significant differences in FDC among clones (χ² = 16.82, p = 0.01). Pairwise Wilcoxon comparisons indicated significant differences (p < 0.05) between India–Jorr, Olof–Jorr, Otisco–Jorr, and Preble–Jorr, with no significant differences among other clone pairs, including Tora.

For the remaining chemical parameters (FDN, ODN, FDH, ODH, ODC, FDS, FDCl, and ash), Kruskal–Wallis tests showed no significant differences among clones (p > 0.05; Figure 1). Despite the lack of statistical significance, some trends were evident. Tora had the highest FDN, followed by Jorr and Olof, while Otisco, Preble, and India exhibited lower values. FDH and ODH were uniform across clones. Jorr had the highest ODN, followed by Tora, Olof, Otisco, India, and Preble. Otisco exhibited the highest ODC, followed by Preble, India, Jorr, and Olof, with Tora lowest. For FDS, Tora ranked highest, followed by Olof, Jorr, India, Otisco, and Preble. Jorr had the highest FDCl, whereas Preble had the highest ash content, followed by Tora, Otisco, India, Jorr, and Olof.

3.3. Variation in the Chemical Content in Relation to Willow Tissues

Significant differences in chemical composition were observed among willow tissue types. Nitrogen content in freeze-dried (FD) samples was highest in bark (1.18%) and lowest in wood (0.26%), with oven-dried (OD) samples showing a similar pattern (bark: 1.28%, wood: 0.25%). Pairwise comparisons indicated significant differences in FDN and ODN between bark and wood, and between bark and twigs (p = 0.001).

Carbon content was highest in FD bark (50.43%) and lowest in FD wood (49.30%), with OD samples following the same trend. Pairwise comparisons revealed significant differences in FDC and ODC between bark and wood, and between twigs and wood (p = 0.001).

Hydrogen content (FDH and ODH) was relatively uniform across tissues, with FD values ranging from 5.97% (bark) to 6.27% (wood) and OD values from 5.84% (bark) to 6.15% (wood). Significant differences in FDH and ODH were detected between bark and twigs, and bark and wood (p < 0.05).

Sulfur content (FDS) was highest in bark (886.79 mg kg⁻¹) and lowest in twigs (521.83 mg kg⁻¹), with no data for wood. Pairwise comparisons showed significant differences between bark and twigs (p = 0.001).

Chlorine content (FDCl) varied considerably, with bark having the highest value (58.64 mg kg⁻¹) and wood the lowest (21.34 mg kg⁻¹). Significant differences were observed between bark and twigs (p = 0.01), while differences between bark and wood were not significant (p = 0.103).

Ash content was highest in bark (4.41%) and lowest in wood (0.74%). Pairwise comparisons indicated significant differences between bark and twigs, and bark and wood (p = 0.001).

4. Discussion

4.1. Elemental Content Values and Impact of Drying Method on Chemical Composition

The wood carbon content (ODC) observed in this study (47.7–48.3%) is comparable to values previously reported for willow (47.4–47.8%; [33]) but lower than those reported by Stolarski et al. [32] (50.8–52.2%). Similarly, wood hydrogen content (ODH) in this study (6.12–6.20%) falls between the ranges reported by Personen [33] (5.7–6.2%) and Stolarski et al. [32] (6.8–7.1%). Nitrogen content (ODN) in these samples (0.19–0.31%) was slightly higher than the 0.1–0.2% reported by Personen [33]. Chlorine content (FDCl) in this study (16.2–26.5 mg kg⁻¹) was markedly higher than the 0.5 mg kg⁻¹ reported for wood by Personen [33], but comparable to the upper range (21.6 mg kg⁻¹) reported for combined wood and bark samples.

The findings of this study indicate that freeze-drying is more effective than oven-drying in preserving the chemical composition of willow, particularly carbon and hydrogen [35,36,37,47,48]. Freeze-dried samples exhibited higher concentrations of these elements, likely due to better retention of volatile organic compounds such as hemicelluloses [36,61]. In contrast, oven-drying may cause thermal degradation of these compounds, potentially reducing the bioenergy potential of willow [61].

Nitrogen content showed minimal variation between drying methods, suggesting that nitrogen is relatively insensitive to drying conditions [36]. Although present in small amounts, nitrogen plays an important ecological role for microorganisms and insects inhabiting wood [38] and influences NO_X emissions during combustion [40,62].

The high sulfur content observed in freeze-dried samples raises potential concerns for thermochemical conversion, as elevated sulfur can negatively affect combustion efficiency and increase pollutant emissions [63].

4.2. Variation Across Willow Clones and Tissue Types

The chemical composition of willow varies across tissue types (bark, wood, twigs), influencing their suitability for bioenergy applications. Nitrogen content was highest in bark and lowest in wood, with twigs intermediate. This pattern reflects tissue metabolic activity: bark, being more metabolically active, requires higher nitrogen for growth and repair, whereas wood, primarily structural, stores less nitrogen [64,65]. Elevated nitrogen in biomass can reduce combustion efficiency and increase NO_X emissions, making bark less suitable for biofuel but potentially valuable for non-energy uses such as bioactive compound extraction [36].

Carbon content was also highest in bark, consistent with its role in biomass production. While bark’s high nitrogen may limit its bioenergy potential, its elevated carbon content could still make it viable if nitrogen-related emissions are managed. In contrast, wood, with lower nitrogen and moderate carbon, appears better suited for biofuel production [66]. Hydrogen content was relatively uniform across tissues, with wood slightly higher, contributing positively to its energy yield. However, higher sulfur and chlorine levels in bark pose challenges for combustion. Chlorine does not contribute to calorific value and, along with sulfur, can lead to acid formation, corrosion, and harmful emissions such as dioxins and furans. Wood’s lower sulfur and chlorine content makes it a more favorable candidate for thermochemical conversion where minimizing corrosive emissions is critical [40,41,44].

These findings align with earlier observations by Merrill and Cowling [38] and Mvolo et al. [36], who reported higher nitrogen concentrations in metabolically active tissues such as bark. This study extends that understanding by linking tissue-specific chemical variation to bioenergy and non-energy applications in willow clones, emphasizing the influence of nitrogen, sulfur, and chlorine on biomass suitability for energy production and value-added uses [64,65]. The non-detectable sulfur content in wood is consistent with previous studies, reflecting the limited metabolic activity of wood compared with more active tissues such as bark [6].

4.3. Implications for Bioenergy Production and Breeding

The chemical variability observed across willow clones and tissue types has important implications for bioenergy production and breeding strategies. Targeting clones with higher carbon content and lower sulfur and ash levels can optimize biomass for energy conversion, improving fuel quality and reducing pollutant emissions during combustion. Lower sulfur and ash content are particularly critical for minimizing SO₂ emissions and operational issues such as slagging and fouling, thereby enhancing the environmental sustainability of bioenergy systems.

Selective harvesting based on tissue composition offers another pathway to improve efficiency. For instance, prioritizing wood over bark for biofuel production can reduce emissions, as wood generally contains less sulfur and ash. Conversely, bark with its higher nitrogen content may be better suited for non-energy applications such as bioactive compound extraction, where nitrogen is advantageous rather than problematic.

These findings also highlight opportunities for breeding willow varieties tailored for bioenergy applications. Breeding programs could focus on traits such as higher carbon and hydrogen content, coupled with lower sulfur, nitrogen, and ash levels, to maximize energy yield while minimizing environmental impacts. Studies in SRC systems have demonstrated significant within-species variation among cultivars and clones in uptake of elements such as N, P, K, Ca, Mg and S when grown under identical soil and management conditions, suggesting that breeding of willows for attenuated or intensified elemental accumulation is possible [67,68,69]. Developing such optimized genotypes would ensure a more sustainable and efficient biomass resource for both thermochemical and biochemical conversion pathways.

5. Conclusions

This study examined variation in the chemical composition of willow across clones and tissue types, and explored its implications for bioenergy production and breeding. The results identify willow clones with higher carbon content and lower sulfur and ash levels, which could be more suitable for biofuel production, providing higher energy yields with fewer pollutant emissions. In contrast, clones with higher sulfur and chlorine content may be better suited for material applications such as bioactive compound extraction. These findings emphasize the importance of targeted harvesting strategies and highlight the potential for breeding willow varieties tailored to specific thermochemical and biochemical applications. By selecting clones based on key chemical traits including carbon, nitrogen, sulfur, and ash content, willow biomass can be optimized for efficient and sustainable bioenergy production while minimizing environmental impact. Because freeze-drying yields more accurate measurements of carbon content, it should become the preferred standard for carbon-accounting protocols. At the same time, oven-drying, already the most widely used method, should continue to be employed for large-scale industrial applications in the bioeconomy.