Tropical mobola plum (Parinari curatellifolia): a full characterization of wood and bark within the scope of biore�neries

Parinari curatellifolia is the main species used to produce charcoal in Angola. It is chemical, anatomical, and thermal properties were analyzed. The bark is dark grey, rough, and corky, and the wood is brown to yellow-red. Compared to wood, bark �bers presented lower length, lumen, and wall thickness. There is not much difference between height and cell numbers of rays. Sieve tube elements appear solitary or in small groups (2–3 cells), and vessels were of two diameter classes but with diffuse-porous. Bark density was lower than wood (505 kg.m − 3 vs. 580 kg.m − 3 ). The mean chemical composition from bark vs. wood of P. curatellifolia was ash (3.2% vs. 1.6%), total extractives (12.2% vs. 10.0%), lignin (42.4% vs. 28.4%), and suberin 5.4%. Families identi�ed by GC-MS from DCM extracts were predominated by fatty acids in wood and triterpenoid contents in bark. Bark and wood had higher antioxidant activity in FRAP and DPPH methods. The bark had a monomeric lignin composition richer in guaiacyl-units (25.9% vs. 22.5%) and lower syringyl-units (5.7% vs. 8.5%). Potassium was the most abundant mineral, while the least is cadmium in wood and bark. Regarding thermal properties, bark presented higher moisture content (9.0% vs. 8.0%), ash (3.33% vs. 1.61%), total volatiles (27.5% vs. 20.7%), lower �xed carbon (69.1% vs. 77.7%) and higher heating value (20.9 MJ/kg vs. 19.1 MJ/kg). According to these characteristics, both biomasses are interesting for developing more value-added products besides burning under the context of biore�neries.


Introduction
The mobola plum, scienti cally known as Parinari curatellifolia (Planch.ex Benth.), is also called uchia, mendonça, muchacha, cork tree, grysappel, loxa, uyengyo, and olonsha (Maharaj and Glen 2008;Orwa et al. 2009; San lippo 2014).This tree belongs to the Chrysobalanaceae family, previously classi ed under the Rosaceae order due to its physical characteristics (Carnevale Neto et al. 2013).It is a medium to large evergreen tree that can grow up to 20 meters tall, with a bare stem and a dense, roundish mushroomshaped crown (Orwa et al. 2009).This species is widely distributed in Angola and the central region of Africa (Palgrave 2002; San lippo 2014).P. curatellifolia is an indigenous fruit species, an important food source for many rural communities (Shoko et al. 2014), and one of the most important medicinal plants (Mawire et al. 2021).For this reason, studies on the bark are mostly related to its evaluation for phytochemistry and pharmacology (Carnevale Neto et al. 2013;Halilu et al. 2013; Omale et al. 2020).Uses in ethnomedicine included treating wound infections, cancer, pneumonia, fever, bacterial infections, and in ammation disorders (Gororo et al. 2016; Kundishora et al. 2020).The plant also has other different uses, i.e., the bark and leaf extracts can be used for tanning, and bark produces a pink-brown dye which is generally used in basket work (Maharaj and Glen 2008); seeds oil content was suitable for biodiesel production (Nabora et al. 2019).The wood is very durable, hard, and heavy and is used as charcoal or rewood purposes (Bolza and Keating 1972); despite its high silica content, which is problematic because it blunts the blades of saws and other tools (San lippo 2014), wood is often used for woodwork, mortars, manufacture of pounding blocks or poles (Maharaj and Glen 2008;Orwa et al. 2009), as well as boats due to their resistance to marine borers (Bolza and Keating 1972;Flora Malesiana 1989;Prance 2007).
Even though the importance of this species, studies on its bark and wood anatomy are scarce, and few or no detailed descriptions were found.In addition, it is known that structural and anatomical features deeply in uence the wood's performance as a material.To our knowledge, this is the rst detailed report on the bark anatomy of P. curatellifolia.Metcalfe and Chalk (1950) gave some information on the general bark anatomy of various genus of Chrysobalanaceae, including species of Parinari and Roth (1981) characterized the bark structure of some genus of the group with a focus on the anatomy of species belonging to the genus Licania and only occasionally gave remarks of P. excelsa and P. rodolphii.Wood anatomy of Chrysobalanaceae was described in Flora Malesiana (1989) but reports on wood anatomical of P. curatellifolia are also limited and set up with other species of the genus in InsideWood (2004); in Richter and Dallwitz (2000) various genus of Parinari were described however without any references for P. curatellifolia.In a recent study, Massuque et al. (2021) evaluated the effect of ber and vessel biometry on wood combustibility.
Charcoal production causes a sizeable environmental impact, like deforestation and forest degradation, and continuously grows in African rural communities (Miapia et al. 2021).Then, more knowledge is needed on the chemical properties of miombo wood species, emphasizing those less used for the wood industry and exploited for charcoal and rewood production (Lhate et al. 2010).This research seeks to provide a detailed description of the bark and wood anatomy of P. curatellifolia, including a comparison with other Chrysobalanaceae species.The study also examines the chemical composition of the wood and bark, including ash, extractives, suberin, lignin, and holocellulose.Furthermore, the research evaluates the monomeric lignin composition, lipophilic extracts composition, phytochemical pro le, and antioxidant activities of ethanol and water extracts.This is the rst time such comprehensive data has been gathered on P. curatellifolia that may contribute to its expansion management and use along with the traditional applications but more e ciently, keeping in mind the scope of biore neries and the implementation of Sustainable Development Goals of the United Nations.
2 Material and methods

Site characterization and sampling
Three trees of P. curatellifolia were collected at Cuima, the municipality of Caala, located in Huambo, Angola (13° 33.216'S and 15°36.956'E).The region is 1700 m a.s.l.(above sea level), it has a mean annual temperature of 20°C and rainfall that ranges from 1200 to 1600 mm per annum.The dominant soils are ferrosols, usually found at higher elevations, followed by uvisols, which are more common at lower elevations (Chiteculo et al. 2022), with a pH from 5.5 to 6.5 and low nutrient content (Delgado-Matas and Pukkala 2015).The annual relative humidity varies from 60-70% (Ndjamba et al. 2021).The mean tree total height and diameter at breast height (DBH) were respectively 8.5 m (± 1.7) and 14.8 cm (± 1.2); stem sectional discs of 3-4 cm thickness were taken from each tree at DBH.

General features
For observation and measurement, the transverse surfaces of the discs were polished with sandpaper, and two cross diameters were measured for bark thickness determination.Samples were observed with a handheld digital microscope Dino-Lite AM4113ZT.

Anatomical characterization
Bark samples were impregnated with polyethylene glycol (DP1500).Transverse and longitudinal microscopic sections of approximately 17µm thickness were prepared with a Leica SM 2400 microtome using Masking Tape 3M 101E, according to Barbosa et al. (2010).The sections were stained with a double chrysoidine-astra blue staining and mounted on Kaiser Glycerin.After 24 hours, the lamellas were submerged into xylol for 30 minutes to remove the adhesive, dehydrated in 96% and 100% ethanol, and mounted in Eukitt.Sudan 4 was used for selective staining of suberin.The bark samples were also observed by scanning electronic microscopy (SEM) with a TM3030Plus Tabletop Microscope (Hitachi) with different magni cations, and the images were recorded in digital format.
Wood samples were cut from the mature wood and softened in boiling water.Transverse and longitudinal sections approximately 17µm thick were prepared with a sliding microtome, stained with safranin, and mounted in Eukitt.
Additionally, bark and wood samples were macerated in a 1:1 solution of 30% H 2 O 2 and CH 3 COOH at 60°C for 48 hours and stained with astra blue.The length, width, and cell wall thickness of 40 bers (bark/wood) and the length and tangential diameter of 25 sieve tube members and vessel elements per individual were determined.In addition, 25 counts of vessel element numbers per mm 2 were made.All measurements were done according toIAWA (1989) andAngyalossy et al. ( 2016) recommendations for wood and bark and used Leica Qwin software.Analysis of variance (ANOVA) was applied to evaluate signi cant differences between trees for the anatomical variables of bark and wood.Differences were assessed with Duncan's post hoc tests.Statistical signi cance was set at p < 0.05.SigmaPlot® (Version 11.0, Systat Software, Inc., Chicago, IL, USA) was used for all statistical analyses.Light microscopy observations were made using Leica DM LA, and photomicrographs were taken with a Nikon Microphot-FXA., The terminology used in this work followed the Angyalossy et al. (2016) for the bark and IAWA (1989) for the wood.

Bark and wood density
The basic density of bark and wood was calculated using water displacement for green volume determination and oven-dry weight, according to TAPPI 258 om-2.

Chemical summative analysis
The 40-60 mesh size fraction of the wood and bark from P. curatellifolia included the determination of ash, extractives, suberin, acid-insoluble lignin (Klason), and acid-soluble lignin and the monomeric composition of polysaccharides following the methodology described by Neiva et al. (2018;Vangeel et al. (2021); Gominho et al. (2021).Ash content was determined by combusting approximately 1.5 g of the sample overnight at 525°C and weighing the residue, following TAPPI standard T211 om-93.Extractives were determined using extraction thimbles by successive Soxhlet extractions with dichloromethane (DCM, 6h), ethanol (16h), and water (16h) as mass loss of the oven-dried thimbles after each extraction.Suberin content was performed in the extractive-free material using methanolysis to depolymerize 1.5 g extractivefree bark as described in Malengue et al. 2023.
The Klason lignin and soluble lignin in the extractive-free material were measured using TAPPI standards T222 om-88 and UM250 om-83, respectively.The ash content of the Klason lignin was considered in making the necessary adjustments.The monosaccharide composition of the polysaccharide fractions was determined in the hydrolysis liquor resulting from the acid-insoluble lignin.The neutral monosaccharides and galacturonic acid were separated by high-pressure ion-exchange chromatography in a Dionex ICS3000 equipped with a PAD detector using a Carbopac PA10 (4×250 mm) column plus Aminotrap and the eluent was NaOH + CH 3 COONa with a ow of 1 mL/min at 25°C.Acetic acid was separated in a Waters 600 and measured with a UV/Vis detector at 210 nm, with a Biorad Aminex 87H HPX column (300×7.8mm), and the eluent was 10 mN H 2 SO 4 with a ow of 0.6mL/min at 30°C.All the chemical analyses were made in triplicate.

GC-MS analysis
The dichloromethane extracts (DCM) obtained by Soxhlet extraction were dried in aliquots under nitrogen and then in a vacuum oven with phosphorus pentoxide.1 mg of the extract was dissolved in 120 µL of pyridine and trimethylsilylated with 80 µL of bis(trimethylsilyl)tri uoroacetamide (BSTFA) and oven heated at 60°C for 30 min.The derivatized dichloromethane extracts were injected in an Agilent GC 7890A coupled to 5975C MSD with the following parameters: oven temperature program ranging from 50 to 380°C at varying rates; the volatiles were separated in a capillary column Zebron 7HG-G015-02 (Phenomenex, USA) with 30m x 0.25mm ID x 0.1µm ( lm thickness) at a ow constant rate of He (1 mL/min); the injector temperature was 380°C.The mass spectra of the compounds were compared to a GC-MS spectral library (Wiley, NIST, and personal library) to identify them as TMS derivatives.Semi-quantitative analysis was performed by expressing the relative proportions of the peaks in the total ion chromatograms as percentages.

Phytochemical pro le and antioxidant activities
The extracts were carried out following the methodology described by Miranda et al. (2016).Approximately 0.250 g of the sample from the wood and bark of P. curatellifolia were extracted with ethanol/water (80:20 v/v) for 1h at 40 C using an ultrasonic bath, with a solid-liquid ratio 1:10 (m/v).The extract resulting from the ltration of the sample was used to evaluate the phytochemical pro le: total phenolics (TPC, expressed as mg gallic acid equivalents (GAE)/g Extract), avonoids (FC), condensed tannins (CTC) contents both expressed as catechin equivalents (CE)/g Extract; all as through calibration curves (Neiva et al. 2018).
The antioxidant activity was attained following two methodologies described by Sánchez-Moreno et al.
(1998): i) FRAP (Ferric-reducing antioxidant power expressed (mM TEAC/g extract) and ii) DPPH (free radical scavenging activity of 2,2-diphenyl-1-picrylhydrazyl).The DPPH results were expressed as IC50 (extract concentration required for 50% DPPH inhibition), in terms of Trolox equivalents (TEAC) on a dry extract base (mg Trolox/mg extract) and as antioxidant activity index (AAI = nal concentration of DPPH in the control sample/IC50.The AAI considers the mass of DPPH and test sample, decreasing the concentration in uence of the DPPH solution used.The antioxidant activity was classi ed as poor if AAI < 0.5, moderate if 0.5 < AAI < 1, strong if 1 < AAI < 2, and very strong when AAI > 2 (Neiva et al. 2018).All the analyses were also made in triplicate, and results were expressed as average with standard deviation.

Analytical pyrolysis
Extractives-free samples from wood and bark of P. curatellifolia were previously milled in a Retsch MM20 mixer ball mill for 10 min, and around 0.10 mg of the sample was weighed.Each sample was pyrolyzed at 550°C (for 1 min) in a platinum coil Pyroprobe connected to a CDS 5150 valved interface linked to a gas chromatographer (Agilent 7890B) and with a mass detector (Agilent 5977B).The volatiles formed were separated in a fused-silica capillary column (ZB-1701: 60m x 0.25 mm i.d.x 0.25 µm lm thickness, (Phenomenex, USA).The chromatographic conditions used were described in Malengue et al. 2023.The compounds were identi ed by comparing their mass spectra with the Wiley, NIST2014 database and literature (Faix et al. 1990;Ralph et al. 1991).The total area of the chromatogram was obtained automatically, and the percentage area of each compound identi ed was calculated.Total carbohydrates derivatives (C) and total lignin derivatives (L) were summed, and then the relation between carbohydrates and lignin was calculated and presented as the C/L ratio.The monomeric composition was also determined as H:G:S relation.

Quantitative determination of mineral content
About 300 mg of dry matter was placed in digestion with aqua regia acid solution (a mixture of hydrochloric acid with nitric acid in the ratio of 3:1) for 90 minutes at 105°C (Gaudino et al. 2007).The ICP-OES analytical technique (Inductively Coupled Plasma-Optical Emission Spectrometry) was applied to determine the elemental composition from wood and bark of P. curatellifolia using the Thermo Scienti c TM iCAPTM 7400 ICP-OES analyzer (Thermo Fisher Scienti c, Bremen, Germany).

Proximate analysis and ultimate analysis
Proximate analysis was performed according to the ASTM E870-82 method.Ultimate analysis was determined following the ASTM standard method D5373-08 on an Elemental analyzer (Thermo Finnigan-CE Instruments Flash EA 1112 CHNS series).The oxygen content was obtained by subtracting from 100% the sum of (C, H, N, S, and ash) contents in percentage.Moisture content, ash content, total volatiles, and xed carbon were determined following the methodology described by Costa et al. (2023).

Data analysis
Statistical analyses of anatomical features were performed using the software R version 4.3.0.for Windows.Signi cance was tested using a one-way ANOVA and comparing mean values via the Tukey post hoc test.The level of signi cance was always set to p < 0.05.

General features
The bark of P. curatellifolia is dark grey, rough, and corky (Fig. 1a), and the wood is brown to yellow-red.The average bark thickness was 1.2 (± 0.2) mm, including the rhytidome and the phloem.The phloem was 0.5 (± 0.1) mm thick and comprised the conducting phloem distinct from the nonconducting phloem by different colors due to the amount of sclerenchyma tissue.Growth increments were detected in the rhytidome and observed even by the naked eye (Fig. 1b).

Bark anatomical characterization
The general anatomy of P. curatellifolia bark had some similarities with another genus of the Chrysobalanaceae studied by Roth (1981), i.e., in some Licania sp.concerning the abundance of mechanical tissue, especially in the form of the nodules of sclereids, bro-sclereids and reduced bers, type of rays and dilatation tissue.
The rhytidome is moderately developed, appearing as small scales (Fig. 1b); a structural pattern in the rhytidome is recognized with a variable number of undulated periderm layers (2-4) intercalated with phloemic tissue, which includes compact nodules of scleri ed tissues (Fig. 1b-e); in the periderm produced by the phellogen, the phellem cells (typical cork cells, Fig. 4e) with thin walls are relatively abundant and arranged regularly in radial rows, some with brown cell contents alternating with others colorless and curving slightly, forming discontinuous arching layers in cross-section (Figs.1c, 1e) and appearing somewhat radially strati ed.The phelloderm is poorly developed, with 2-4 cells sometimes ligni ed with content and distinguished from cortical-like cells by their neat radial alignment (Fig. 1f-g); the observation of the phellogen is complex (Fig. 1g).
Figure 1 The phloem tissue produced by the vascular cambium is non-storied, comprising the conducting and nonconducting phloem (Figs.2a, 2c).The conducting phloem near the vascular cambium represents a tiny part of the phloem.It includes the sieve tube elements with companion cells (conducting cells), the axial parenchyma (storage tissue), rays (storage/transversal conduction), and the bers (mechanical support) (Fig. 2a-b).The transition between conducting and nonconducting tissue is somewhat abrupt, marked by the collapse of sieve tube elements and the formation of prominent nodules of sclerenchyma cells (Fig. 2a,   2c).
Figure 2 In the transverse view, the sieve tube elements appear round to polygonal with thin and unligni ed cell walls dispersed between parenchyma cells and bers, solitary or in small groups (2-3 cells); they can distinguish from neighboring parenchyma cells by their somewhat larger diameter.The sieve tube elements presented a mean length of 423µm (± 17), ranging between 127-670µm and a tangential diameter of 25µm (± 9), ranging between 6-42µm.The companion cells are di cult to recognize in transverse and longitudinal sections.Sieve plates are scalariform compound oblique, with > 10 areas per plate equally spaced (Figs.2b,   4a).
In the conducting phloem, the axial parenchyma is relatively scarce in small layers near the sieve tube elements; parenchyma cells appeared rectangular or polygonal in the transverse section and have thin unligni ed walls, but in nonconducting phloem cells could enlarge or radially divided giving rise to a dilatation tissue between the bers and nodules of cluster sclereids; in the outer phloem beneath the recently formed periderm, a considerable amount of cells like cortical-cells appeared (Fig. 1f); strands of crystal-bearing axial parenchyma of up to 10 cells were found along the margins of the ber band (Fig. 3b) near the nodules of sclereids.These crystalliferous parenchyma strands accompanied by bers were a conspicuous bark feature observed in other barks (Carlquist 2002 The rays, when observed in a transverse section, followed a straight to an undulated direction in the initial phloem but soon distorted due to the collapse of the sieve tubes and early formation of nodules of clusters sclereids (Fig. 3c); rays are non-storied, mainly uniseriate (Fig. 3a) and homogeneous with procumbent cells (Fig. 3d); the cells composing the phloem rays had thickened walls and dark contents (Fig. 3a).Dilatation of rays is moderate toward the bark outside (Fig. 3c).
Figure 3 The sclerenchyma tissue was abundant and formed a high proportion of P. curatellifolia bark; sclerenchyma is primarily in the form of sclereids and ber-sclereids, and comparatively few bers developed.A structural pattern in the nonconducting tissue of this species is recognized, starting with the formation of round and conspicuous nodules of cluster sclereids arranged more or less parallel to the vascular cambium that progressively becomes radially aligned and diminish towards the periphery (Figs.1b, 3c).According to Roth The sclereids appeared isodiametric, although they attained various shapes and sizes with thickened and poly-lamellate walls and pitted, sometimes enclosing one prismatic crystal (Fig. 4c, 4d).Fiber-sclereids are similar to the bers but shorter and only distinguished in macerated bark (Fig. 4c).Fibers are arranged in either solitary or in the form of tiny groups of 2-5 cells wide (Fig. 2a); the mean ber length was 1245µm (± 27) ranging between 903-1625µm and the width value was 26µm (± 0.06) ranging between 18-36µm.The bers were slender with a narrow lumen and pitted (Figs.4b, 5c), thick-walled 7µm (± 0.19) ranging between 4-10µm.

Figure 4
Phenolic compounds were observed in rhytidome and nonconducting phloem by dark color staining in axial and ray parenchyma cells, cortical-like parenchyma, and sclereids (Figs.1c, 1f).Crystals, presumably of calcium oxalate, mainly were present as prismatic crystals and found in parenchyma cells and sclereids (Figs.3b, 4c); starch grains were also observed in our samples (Fig. 5a-b), as also mentioned by Halilu et al. (2008) in the bark of Parinari sp.Silica was also noticed (Fig. 3d), agreeing with records in the Flora Malesiana (1989) concerning the occurrence of silica in the phloem of Chrysobalanaceae.According to An and Xie (2022), few studies regarding phytoliths in bark exist.Still, the authors referred to studies in taxa from West Africa (Collura and Neumann 2017), where more than 90% of bark samples produced phytoliths, suggesting that silica production is concentrated in bark more than in wood.
Phenolic compounds and calcium oxalate crystals are common in the barks (Evert 2006) and, in conjunction with sclerenchyma, have evolved secondary defense functions (Franceschi et al. 2005).Figure 5 3.

Wood anatomical characterization
The general wood structure of P. curatellifolia resembles the structure of other Parinari species described in InsideWood (2004), i.e., P. anamensis, P. campertis, P. congensis, P. excelsa, concerning the type of wood porosity, vessel pits, axial parenchyma, the composition of the rays and presence of silica grains.
Figure 6 Axial parenchyma is non-storied in narrow bands up to 3 cells wide (Fig. 6b) and in strands of over 5 cells (Fig. 7a-b).The parenchyma cells were rectangular and pitted (Fig. 7c); spiral thickenings were not observed in our samples, agreeing with observations of Ter Welle (1975); this author mentioned the presence of spiral thickenings in most representative genera of the Chrysobalanaceae, but highlight the lack of spiral in the genus Parinari.Crystals were also absent.
Libriform bers were generally non-septate (Figs.6d, 7c) and aligned in a radial row when observed in the transverse section.Fibers were very pitted with bordered pits in radial and tangential walls (Figs.7b, 8b) and well de ned in individualized bers (Fig. 8c).Bordered pits are signaled by Feitosa et al. (2012) in bers of this species.

Figure 7
Rays were non-storied exclusively uniseriate (Fig. 7a-b) with a mean ray height of 440µm (± 22), ranging between 124-827µm and 4-38 number of cells high.The rays were heterogeneous, with body ray cells procumbent with mostly 2-4 rows of upright and square marginal cells and sometimes rays with procumbent, square, and upright cells mixed throughout the ray (Fig. 8a-b).Radial canals were absent.Dark deposits were common in ray cells, but crystals were absent.Abundant silica was observed in ray cells and aggregations as irregularly shaped or globular bodies (Fig. 8b-c).Plants differ in their abilities to accumulate silica, varying signi cantly between species (Currie and Perry 2007).Silica bodies in ray cells were mentioned in various species of Chrysobalanaceae, i.e., Licania leptostachya, and species of different families, i.e., Lauraceae, Dipterocarpaceae, being a diagnostic feature (IAWA 1989).The role of phytoliths, microscopic amorphous silica structures, against biotic and abiotic stress as mechanical barriers and their use as taxonomic tools is highlighted by Nawaz et al. (2019).Figure 8 Traumatic structures, i.e., pith eck (Fig. 9a), presence of traumatic tyloses (Fig. 9b), and numerous traumatic thin vessels with tangential alignment (Fig. 9c), were detected, probably due in response to biotic or abiotic stress effects, i.e., drought, re, insects, injury to the cambium that induces a disorganized formation of parenchymatous cells (IAWA 1964;Schweingruber 2007).According to Ceccantini (1996) and Mota et al. (2017), water de cits may cause pith ecks formation.Table 1 summarizes the cell dimensions in the bark and wood of P. curatellifolia in studied trees.For bark, the height of the rays, the length and tangential diameter of the sieve-tube elements, and the length and lumen of the bers were signi cantly different (p < 0. 05) between trees; also, for wood ANOVA revealed signi cant differences (p < 0. 05) concerning rays (height and nº cells), vessel elements (length and tangential diameter), and ber (length, width, and lumen) among trees.There are few records concerning the correlation between xylem and phloem cells.Most were correlated with changes in conduit diameter, i.e., sieve tube diameter was strongly correlated with vessel diameter (Jacobsen et al. 2018).Still, our results also show that trees with large vessels have narrow sieve tube elements.Unfortunately, studies concerning bark and wood cell size variations between trees in Parinari sp.
were not found in the literature.

Basic density
As expected, the basic density of wood was higher than bark (580 vs. 505kg m − 3 ).P. curatellifolia, wood density values reported reached 593 kg.m − 3 (Abbot et al. 1997) and 720 kg.m − 3 (Orwa et al. 2009), which is in accordance with our results.However, no studies on the density of the bark were found.

Chemical composition
Chemical summative analysis of wood and bark from P. curatellifolia is presented in Table 2.

Lipophilic extracts composition
The quantitative proportions of the chemical families identi ed by GC-MS in the DCM extracts of wood and bark are presented in Table 3.It was possible to identify about 92.2% (from wood) and 90.3% (from bark) of the total ion chromatogram area; there is a distinct difference in chemical composition between wood and bark.The wood revealed a high content of fatty acids (86.2%),where de hexadecanoic acid (C 16:0 ), 9,12octadecadienoic acid (C 18:2 ), 9-octadecenoic acid (C 18:1 ) and octadecanoic acid (C 18:0 ) respectively with 28.4%, 17.2%, 30.8%, and 5.1% were the predominant compounds.In turn, the bark is rich in triterpenoids compounds (67.7%),where betulinic acid (32.2%), corosolic acid (10.1%), ursolic acid (8.8%), and oleanolic acid (6.5%) were identi ed as the main compounds.It isn't easy to compare the results obtained since, to our knowledge, this is the rst study reporting the lipophilic pro le of P. curatellifolia.Various studies focus on the phytochemical screening of plant fractions from P. curatellifolia, such as leaves, barks, wood, seeds, and roots with different polar solvents.These studies have identi ed the presence of saponins, alkaloids, Table 3 The proportion of chemical families in the lipophilic dichloromethane extracts (as a percentage of total area) identi ed by GC-MS in wood and bark from P. curatellifolia.4).Generally, barks present high contents of total phenolics (Ferreira et al. 2018).
Regarding the antioxidant activities, bark also presented higher FRAP values than wood (4.7 vs. 0.9 mM TEAC/g extract) and the same DPPH (1410 vs. 459 mg TEAC/g extract, Table 4).The antioxidant capacity expressed by AAI value was 8.8 in bark and 2.8 for wood, corresponding to a very strong antioxidant classi cation since AAI > 2 (Neiva et al. 2018).Overall, both tissues are interesting to be used as feedstock for antioxidant extraction and bio-based product development.Other LDC -other compounds assigned to lignin but not with H, G, S-lignin units (e.g., catechol, orcinol).
Table 5 3.10.Proximate analysis and ultimate analysis 6 presents the proximate, ultimate analysis, higher value and mineral composition of P. curatellifolia samples.Generally, the bark had higher moisture content (9.0 vs. 8.0%), ash (3.3 vs. 1.6%), total volatiles (27.5 vs. 20.7%),and less xed carbon (69.1 vs. 77.7%)compared to wood.The wood values are similar to the literature, except for ash content of 6.8%, the total volatiles of 21.5%, and xed carbon of 71.7% (Massuque et al. 2021).The ultimate analysis presented more similar values between bark and wood, e.g., the ratio between hydrogen and carbon in bark and wood (0.12 vs. 0.13) and the ratio between oxygen and carbon (0.93 vs. (more G-units), and its proximate values (e.g., HHV) con rmed its good properties for energy, which is now the traditional use.Still, its characteristics favored the production of more value-added products under the scope of bio-energy, such as biocarbon (also known as charcoal and produced more e ciently), also bio-chemicals that could be attained from sustainable and managed forests aiming to preserve this species.Therefore, P. curatellifolia biomass could be integrated into a biore nery context to provide socioeconomic bene ts to rural communities seeking to ful ll the Sustainable Development Goals of the United Nations.

(
1981) the abundance of mechanical tissue, especially in the form of sclereids, adds the hard and brittle consistency to the barks and constitutes a mechanical barrier against the collapse of living cells; this provides mechanical support of the tissue in the phloem of both tropical (Machado et al. 2005; Baptista et al. 2013) and temperate species (Şen et al. 2011; Cardoso et al. 2015; Vangeel et al. 2021).

Table 1
Biometric data for bark and wood of P. curatellifolia after individualization (mean values of three threes).
aAverage values followed by the same letter are not signi cantly different (p < 0. 05).

Table 1
Jorge et al. (2000)ese (1974)when comparing the cell dimensions between bark and wood, the wood has larger cells than the bark (Table1); i.e., bark bers are shorter than those of the wood (on average, 1363.30µm in wood and 1245.47 in bark) agreeing withQuilhó et al. (2013)in Quercus faginea, but in opposition to the results found in tropical trees byParameswaran and Liese (1974)and in Eucalyptus globulus byJorge et al. (2000).

Table 3 3
.8. Phytochemical pro le and antioxidant activity of ethanol/water extracts The yield of the extraction ethanol:water (80:20, v/v), phytochemical pro le, and antioxidant activity are presented in Table4.The extraction yield was similar between wood (8.57%) and bark (8.29%) and slightly lower than the content of the polar extractives determined by sequential solvent extraction (9.6% for wood and 11.3% for bark, Table2).The Soxhlet apparatus can attribute this result.The sample is repeatedly brought into contact with fresh portions of solvent, and the system remains at a relatively high temperature, furthermore, no ltration is required after leaching, and sample yield can be increased by performing several simultaneous extractions (Luque de Castro and Priego-Capote 2010).

Table 4 Table 5
Quanti cation of the pyrolysis-derived compounds from wood and bark of P. curatellifolia (% of total pyrogram area).

Table 6
Proximate analysis, ultimate analysis present in wood and bark from P. curatellifolia.Mean values of three-replicate.

Table 6 4
(Benhura et al. 2013he mineral composition is similar in fruits and seeds(Benhura et al. 2013; CONCLUSION Parinari curatellifolia is well known within rural communities in Huambo-Angola for providing rewood, charcoal, and fruit.As far as we know, this is the rst time this species is characterized regarding anatomy and chemistry, particularly for bark tissue.In contrast, few studies were found on wood chemistry and none about its anatomy.Chemically, bark has considerable suberin, extractives, lignin contents, and high antioxidant activity compared to wood.Therefore, local communities should use bark extraction for traditional medicine and industrial purposes to increase the potential valorization of P. curatellifolia under the biore neries framework.The wood's chemical composition, particularly lignin content and composition (Menéndez and Curt 2013)er value of 18.5 MJ/kg(Massuque et al. 2020), a slightly lower than 20.0 MJ/kg from wood of B. spiciformis and B. utilis, the favorite species of rural community(Menéndez and Curt 2013).For the mineral composition, the most important minerals were potassium (2560.3 and 2387.8 mg/kg), calcium (6509.2 and 912.3 mg/kg), and magnesium (374.9 and 874.6 mg/kg), respectively