Sunflower seed processing affects nutrient utilization and performance in feedlot ewe lambs fed high-concentrate diets with or without chitosan

1 Introduction

The intensification of lamb production systems has increasingly relied on high-concentrate diets to maximize energy intake and improve growth performance. In this context, dietary lipids supplied through oilseeds are widely used to increase energy density. However, the success of this strategy depends on maintaining ruminal stability, as excessive or poorly regulated lipid availability may disrupt microbial activity and compromise fermentation efficiency. Unsaturated fatty acids, particularly oleic and linoleic acids, impair ruminal fiber utilization by inhibiting cellulolytic microorganisms. Maia et al. (2007) and Maia et al. (2010) demonstrated the direct inhibitory effects of polyunsaturated fatty acids on cellulolytic bacteria, whereas Ibrahim et al. (2021) reported a reduction in fibrolytic populations and protozoa. Fiorentini et al. (2015) associated highly unsaturated lipid sources with lower fiber digestibility.

Importantly, the impact of lipids on ruminal function is not determined solely by their dietary inclusion levels but also by their physical presentation within the feed matrix. In oilseeds, lipids are predominantly stored within intracellular compartments and are associated with structural components of the cell. Barletta et al. (2016) demonstrated that preserving seed integrity reduces the immediate exposure of lipids to ruminal microorganisms, thereby modifying the lipid availability kinetics during fermentation. In contrast, mechanical processing disrupts cellular structures and increases the accessibility of intracellular contents, potentially intensifying lipid–microbe interactions in the rumen. Therefore, the physical form of oilseeds is a key factor regulating lipid availability and its consequences for ruminal fermentation and microbial activity.

This aspect is particularly relevant under high-concentrate feeding conditions, where ruminal fermentation is inherently characterized by rapid degradation of carbohydrates, reduced pH, and a microbial ecosystem that is more susceptible to perturbation. Under these conditions, maintaining synchronization between fermentable energy and nitrogen supply is essential for optimizing the growth of microbes. Microbial protein synthesis is a major contributor to the metabolizable protein supply in ruminants, often accounting for 50–80% of the total amino acids absorbed, depending on diet composition and ruminal conditions, as reported by Lu et al. (2019) and more recently by Cai et al. (2025). Therefore, strategies that preserve microbial efficiency under challenging conditions may be more effective than those that attempt to correct imbalances after they have occurred.

Among feed additives, chitosan has been investigated as a potential modulator of ruminal fermentation in sheep. As a cationic biopolymer derived from chitin, its antimicrobial activity is associated with electrostatic interactions between positively charged amino groups and negatively charged microbial cell surfaces, which can alter membrane permeability and microbial metabolism (Kong et al., 2010). In ruminant systems, these antimicrobial properties have attracted attention because of their potential to modify ruminal fermentation and improve nutrient utilization. For example, He et al. (2024) reported changes in nutrient disappearance, fermentation characteristics, and ruminal microbial populations following chito-oligosaccharide supplementation, whereas Piboonkunsamlit et al. (2025) highlighted the broader potential of chitosan to improve feed efficiency and sustainability in ruminant production systems. However, the responses to chitosan supplementation are inconsistent. A meta-analysis conducted by Harahap et al. (2022) showed that improvements in performance and nutrient utilization occur only under specific dietary conditions, whereas experimental studies, such as those by Jacaúna et al. (2021); Pereira et al. (2018), and Magalhães et al. (2019), reported variable or limited responses depending on diet composition and inclusion level.

The interaction between chitosan and dietary lipids, particularly in high-concentrate systems, is a critical but underexplored area. Both unsaturated fatty acids and chitosan exerted antimicrobial effects on ruminal microorganisms, suggesting the possibility of overlapping or redundant mechanisms of action. In diets where lipid availability is increased, especially through the mechanical processing of oilseeds, the ruminal microbial ecosystem may respond differently to additional modulation with chitosan. Under these conditions, the additional effect of chitosan may be reduced or dependent on the prevailing dietary and ruminal environment. However, there is limited information on how oilseed processing interacts with chitosan supplementation to influence ruminal function and nutrient utilization.

Therefore, it remains unclear how the physical form of oilseeds affects nutrient utilization and microbial nitrogen production and whether it modifies the response to fermentation modifiers in high-concentrate diets. Addressing this gap is particularly relevant for production systems that aim to optimize biological efficiency while minimizing input costs and dietary complexity.

Based on this rationale, we hypothesized that the physical form of sunflower seeds would affect nutrient utilization, estimated microbial nitrogen production, and animal performance, and that the response to chitosan supplementation would depend on the processing of sunflower seeds. Growing ewe lambs were selected because they represent an important animal category in replacement and meat production systems and provide a relatively homogeneous physiological group for evaluating dietary responses during growth. Accordingly, this study evaluated the effects of sunflower seed processing (whole vs. ground) and chitosan supplementation on nutrient utilization, ingestive behavior, nitrogen metabolism, estimated microbial nitrogen production, growth performance, and descriptive economic indicators in feedlot ewe lambs.

2 Materials and methods

2.1 Locations, animals, and experimental periods

The experiment was conducted at the Laboratory of Research in Ruminant Nutrition and Feeding and Animal Nutrition Laboratory of the Universidade Estadual de Santa Cruz, Ilhéus, Bahia, Brazil (14°47′ S, 39°17′ W; altitude: 57 m). The regional climate is classified as Af (humid tropical) according to Köppen, with an average annual temperature of 24.5 °C and annual rainfall of approximately 1,700 mm.

Thirty-two growing ewe lambs (initial body weight: 15.12 ± 3.57 kg; approximately 75 days old) were used in this study. Prior to the beginning of the trial, the animals were clinically evaluated, weighed, identified with neck collars, dewormed with albendazole (5 mL/100 kg BW; Biozen^®, Brazil), and vaccinated against rabies and clostridial diseases (2 mL/animal), according to manufacturer recommendations.

Ewe lambs were housed in individual suspended pens (1.0 m²) equipped with slatted floors, individual feeders, and automatic drinkers, allowing precise measurement of individual feed and water intake throughout the experiment.

The experimental design was a completely randomized design in a 2 × 2 factorial arrangement with two factors: sunflower seed processing (whole or ground) and chitosan inclusion (0 or 4 g/kg DM). Before allocation, the animals were stratified by initial body weight and then randomly assigned to one of the four treatments (n = 8 per treatment), ensuring balanced distribution among the groups. Each animal was considered an experimental unit in this study. A sensitivity analysis performed using PROC POWER in SAS version 9.4 (SAS Institute Inc., Cary, NC), with average daily gain as the primary response variable, indicated that the adopted sample size provided 80% power to detect a minimum difference of 32 g/d, based on the residual variance estimated from the statistical model.

The experimental period lasted 59 days, consisting of 17 days of dietary adaptation, followed by 42 days of data collection. The experimental phase was divided into two consecutive 21-day intervals for the operational organization of sample collection and experimental monitoring, with treatments and management conditions maintained constant throughout the 42-day period.

2.2 Experimental diets

The experimental diets were formulated to be isonitrogenous and to meet the nutritional requirements of growing ewe lambs according to Exigências Nutricionais de Caprinos e Ovinos—BR-Caprinos & Ovinos (BR-Caprinos & Ovinos, 2024), targeting an average daily gain of 300 g/day.

The diets comprised 25% forage and 75% concentrate (dry matter basis), with sorghum silage as the forage source. The concentrate consisted of ground corn, soybean meal, sunflower seeds (whole or ground), a mineral mixture, sodium bicarbonate, and limestone. Diets were formulated to provide similar crude protein and energy concentrations across treatments, differing only in sunflower seed processing and chitosan inclusion levels. The ingredients and chemical compositions of the diets are presented in Table 1.

Table 1. Ingredient and chemical composition of the experimental diets according to chitosan inclusion level.

Neutral detergent fiber corrected for ash and protein.

*Guaranteed levels: zinc, 3,800 mg/kg; sodium, 147 g/kg; manganese, 2,000 mg/kg; cobalt, 15 mg/kg; copper, 590 mg/kg; sulfur, 18 g/kg; selenium, 20 mg/kg; iodine, 50 mg/kg; chromium, 20 mg/kg; molybdenum, 300 mg/kg; calcium, 110 g/kg (maximum 135 g/kg); fluorine, maximum 870 mg/kg; and phosphorus, 87 g/kg.

Sunflower seed processing did not alter the formulated ingredient proportions or analyzed chemical composition; therefore, whole- and ground-seed diets are not presented separately.

The four experimental treatments were as follows: (1) whole sunflower seeds without chitosan (0 g/kg DM), (2) whole sunflower seeds with chitosan (4 g/kg DM), (3) ground sunflower seeds without chitosan (0 g/kg DM), and (4) ground sunflower seeds with chitosan (4 g/kg DM).

Sunflower seeds used in the whole-seed treatments were offered intact, without mechanical processing. For the ground-seed treatments, sunflower seeds were processed using a hammer mill without a sieve. The purpose of this process was to mechanically disrupt seed integrity rather than produce a standardized particle size distribution. The particle size distribution of the processed sunflower seed fraction was not directly measured. However, the particle size distribution was determined for the complete diets according to Zanotto et al. (2016) using Granucalc^® software, and the results are presented in Table 2. Because these measurements refer to complete diets rather than specifically to sunflower seed particles, the degree of seed disruption and lipid exposure from the seed matrix could not be directly quantified.

Table 2. Particle-size distribution (% retained) of experimental diets containing whole or ground sunflower seeds and chitosan at 0 or 4 g/kg DM.

SS, sunflower seed processing; GSD, geometric standard deviation; GMDi, geometric mean diameter.

Chitosan was supplied as a top-dress once daily during the morning feeding, previously mixed with approximately 50 g of concentrate to ensure complete ingestion by ewe lambs. The chitosan used had a degree of deacetylation of 86.3%, bulk density of 0.33 g/mL, pH of 7.9, and molecular weight lower than 150 kDa (Polymar^®, Fortaleza, Ceará, Brazil). The 4 g/kg DM inclusion level was selected based on previous biologically responsive doses in ruminants and corresponded to an estimated intake of approximately 155–160 mg/kg BW per day (Pereira et al., 2018; Magalhães et al., 2019; Dias et al., 2020).

Feed was offered ad libitum twice daily (08:00 and 16:00 h), allowing for approximately 10% refusals to ensure unrestricted intake. Orts were collected and weighed daily to determine the individual feed intake.

2.3 Laboratory analysis

Samples of feed, refusals, and feces were thawed at room temperature, oven-dried in a forced-air oven at 60 ± 5 °C for 72 h, ground to pass a 1-mm screen, and stored in airtight containers until further analyses.

All chemical analyses were performed in duplicate. Dry matter, ash, ether extract, and crude protein contents were determined according to the standard procedures described by the Association of Official Analytical Chemists (AOAC, 2016). The crude protein content was calculated as the total nitrogen content multiplied by 6.25.

Neutral detergent fiber corrected for ash and protein (NDFap) and acid detergent fiber (ADF) were determined using heat-stable α-amylase and autoclave procedures, as described by Detmann et al. (2025).

Indigestible neutral detergent fiber (iNDF) was determined as an internal marker following the in situ incubation procedures described by Reis et al. (2017) and dos Santos et al. (2025). For this purpose, approximately 2 g of each sample (feed, refusals, and feces) was weighed into non-woven fabric bags (100 g/m²; 5 × 5 cm) and incubated in the rumen for 288 h. After incubation, the residues were recovered, washed, dried, and analyzed for dry matter and NDF content.

Total carbohydrates were estimated according to Sniffen et al. (1992), and non-fibrous carbohydrates were calculated as described by Hall (2003). The total digestible nutrients and gross energy were estimated using equations adapted from da Cruz et al. (2021), and the digestible energy was derived from the total digestible nutrient content of the diet.

2.4 Nutrient intake and apparent digestibility

The feed offered and refusals were recorded daily for each ewe lamb to determine the nutrient intake. Representative samples of the feed and refusals were collected throughout both 21-day collection intervals. Samples collected within each interval were pooled by animal and stored at −10 °C. The resulting interval composites were subsequently combined proportionally to obtain one final composite sample per animal, representing the complete 42-day data collection phase for chemical analysis.

Fecal samples were collected directly from the rectum at 0, 2, 4, 6, and 8 h after feeding on days 17 and 21 of each collection interval to account for the diurnal variation in fecal excretion. Samples were initially pooled by animal within each interval, stored at −10 °C, and subsequently combined to obtain one final composite fecal sample per animal representing the complete 42-day data collection period.

Dry matter intake (DMI) was calculated as the difference between the feed offered and refusals and expressed on a dry-matter basis.

Fecal dry matter excretion was estimated using indigestible neutral detergent fiber (iNDF) as an internal digestion marker. The iNDF concentration in the feed, refusals, and feces was determined as described in Section 2.3 using long-term in situ ruminal incubation (288 h). The use of iNDF as an internal marker is based on its resistance to microbial degradation and consistent recovery across the gastrointestinal tract, providing a reliable estimate of fecal output.

The apparent digestibility coefficients of dry matter, organic matter, crude protein, ether extract, non-fibrous carbohydrates, and neutral detergent fiber corrected for ash and protein (NDFap) were calculated using the following equation:

Apparent digestibility (%)=100×[(nutrient intake−fecal nutrient output)/nutrient intake]

where fecal nutrient output was estimated from the total fecal dry matter excretion and nutrient concentration in feces.

This approach allowed the estimation of apparent digestibility under practical feeding conditions without total fecal collection, ensuring a reliable assessment of nutrient utilization across treatments.

2.5 Ingestive behavior

Ingestive behavior was evaluated to characterize feeding, rumination, and idling activities over a continuous 24-h period. Observations were conducted using a time-sampling methodology at 10-min intervals, following the procedures described by Carvalho et al. (2004). This interval has been widely adopted in ruminant behavior studies as a reliable compromise between observational resolution and operational feasibility, allowing for an accurate estimation of time budgets without continuous recording.

Ewe lambs were visually monitored by trained observers who underwent calibration to ensure consistency in the activity classification. The behavioral categories were defined as follows: feeding (active ingestion of feed), rumination (regurgitation and rechewing of cud), and idling (absence of feeding or rumination activity). During the nocturnal period, artificial lighting was maintained, and the animals had previously adapted to these conditions to avoid any behavioral changes.

Rumination dynamics were further characterized by directly measuring ruminal bolus processing. For each ewe lamb, the number of chews per bolus and time spent ruminating per bolus (s/bolus) were recorded using a digital stopwatch. Measurements were obtained during three representative periods of the day (10:00–12:00, 14:00–16:00, and 18:00–20:00 h) to capture the postprandial and intermediate rumination patterns, respectively. Within each period, three boluses per animal were evaluated, resulting in nine observations per animal.

The total daily time spent on each activity was calculated by summing the number of observations assigned to each behavioral category and multiplying them by their sampling interval. Feeding and rumination efficiencies were calculated as the ratio between dry matter intake (DMI) or NDFap intake and the respective time spent on feeding or rumination, expressed as g/h.

Total chewing time was defined as the sum of the feeding and rumination times, as proposed by Polli et al. (1996). Additional behavioral parameters, including the number of activity bouts and average duration per bout, were derived from the frequency and temporal distribution of the observed events.

This approach allowed for a detailed characterization of ingestive behavior patterns and their relationship with dietary treatments.

2.6 Water balance

Water intake and excretion were evaluated during the last two days of each 21-day collection interval. Measurements from both intervals were combined to calculate one mean value per animal for each water balance variable over the 42-day data collection period.

The supplied water intake was determined individually for each ewe lamb as the difference between the volume of water offered and refusals recorded over a 24-h period. To account for potential water losses unrelated to animal consumption, control buckets containing the same initial volume of water were placed within the experimental facility but outside the reach of the animals. The variation in water volume in these control buckets was monitored, and evaporation losses were considered negligible (<1% of the total volume) and, therefore, not included in the calculations.

Dietary water intake was estimated based on the moisture content of the feed offered and refusals, calculated as the difference between fresh and dry matter. The total water intake was calculated as the sum of the supplied and dietary water intakes:

Total water intake=supplied water intake+dietary water intake

Fecal water excretion was estimated from the total fecal output and fecal dry matter concentration using the relationship between fresh and dry fecal mass. Urinary water excretion was estimated based on the total urinary volume, as described in Section 2.7.

The water balance was calculated as follows:

Water balance=total water intake−(fecal water excretion+urinary water excretion)

All variables were expressed in absolute terms (L/d) and relative to dry matter intake or body weight gain, allowing the evaluation of water use efficiency across treatments.

This approach enabled the assessment of water dynamics in relation to nutrient intake and animal performance.

2.7 Nitrogen balance, urea excretion and estimated microbial nitrogen production

Nitrogen balance and estimated microbial nitrogen production were evaluated to characterize nitrogen utilization and microbial protein supply, respectively.

Spot urine samples were collected from each ewe lamb on day 18 of each 21-day collection interval, approximately 4 h after morning feeding, to represent postprandial metabolic conditions. Samples were obtained during spontaneous urination and immediately acidified by diluting 10 mL of urine in 40 mL of 0.036 N sulfuric acid (H₂SO₄) to prevent nitrogen loss via volatilization. The samples collected in the two intervals were subsequently combined by animal to obtain one final composite urine sample representing the complete 42-day data collection phase and stored at −10 °C until analysis.

The daily urinary volume was estimated using creatinine as an internal marker, assuming a constant excretion rate of 17.05 mg/kg BW, as previously validated for sheep under similar physiological conditions (Pereira et al., 2018). The total urinary volume (L/day) was calculated as the ratio of the estimated daily creatinine excretion to the concentration in the spot urine samples.

The urinary concentrations of urea, creatinine, and uric acid were determined using commercial colorimetric kits (Bioclin^®, Belo Horizonte, Brazil). Total urinary nitrogen was quantified using the Kjeldahl method (AOAC, 2016). Allantoin, xanthine, and hypoxanthine concentrations were determined using enzymatic and colorimetric assays, as described by Chen and Gomes (1992).

Total purine derivative excretion was calculated as the sum of allantoin, uric acid, xanthine, and hypoxanthine. Absorbed purines (mmol/day) were estimated from purine derivative excretion using the model proposed by Chen and Gomes (1992), assuming a recovery rate of 0.83 mmol. Microbial nitrogen production (g/day) was estimated from the absorbed purines using a fixed ratio of purine nitrogen to total nitrogen in the microbial biomass (0.116).

The nitrogen balance was calculated as the difference between nitrogen intake and total nitrogen excretion (fecal + urinary nitrogen), allowing the estimation of nitrogen retention.

This integrated approach, which combines urinary purine derivatives and nitrogen balance, provided estimates of microbial nitrogen production and overall nitrogen utilization under experimental conditions.

2.8 Performance

Ewe lambs were weighed at the beginning and end of the 42-day data collection phase after a 16-h feed withdrawal, with free access to water to minimize variation in body weight associated with gut fill. Initial body weight was recorded on day 1, and final body weight was recorded on day 42.

Average daily gain (ADG; g/d) was calculated as:

ADG=(final body weight−initial body weight)/42

Feed efficiency (FE; g gain/kg DM intake) was calculated as:

FE=ADG(g/d)/DMI(kg/d)

All performance variables were calculated individually, with each ewe lamb considered the experimental unit.

2.9 Bioeconomic assessment

A bioeconomic evaluation was conducted to assess the relationship between animal performance and feeding costs across treatments, following the approach described by Rodrigues et al. (2013).

Feed costs were calculated based on the proportion of each ingredient in the diet and their respective market prices, expressed on a dry matter basis. Ingredient prices were obtained from the regional market (Bahia, Brazil) in February 2020 and converted to US dollars (US$) using the exchange rate at the time of the study.

The cost of the diet (US$/kg DM) was calculated as:

Diet cost=∑(ingredient proportion×ingredient price)

Daily feeding cost (US$/animal/day) was calculated as:

Daily feeding cost=DMI×diet cost

The total feeding cost (US$/animal) over the 42-day data collection phase was obtained by multiplying the daily feeding cost by the number of feeding days.

Revenue was estimated based on total body weight gain during the experimental period and the market price of live animals, using a fixed value of US$ 1.50 per kg of body weight:

Revenue=total body weight gain×price per kg BW

Net margin (US$/animal) was calculated as:

Net margin=Revenue−Total feeding cost

Additionally, the feed cost per unit of weight gain (US$/kg gain) was calculated to evaluate production efficiency across the treatments.

All economic indicators were expressed on a per-animal basis and compared descriptively under the adopted price assumptions.

Economic indicators were calculated deterministically from treatment mean dry matter intake and body weight gain using fixed ingredient prices and live-animal market value; therefore, the results are presented as scenario-specific descriptive estimates.

2.10 Statistical analysis

Data were analyzed using a completely randomized design in a 2 × 2 factorial arrangement, with two sunflower seed processing methods (whole or ground) and two levels of chitosan inclusion (0 or 4 g/kg DM). For each response variable, one final value per animal was used in statistical analysis, and the animal was considered the experimental unit. Period was not included in the model because no independent period-specific analytical observations were generated.

The statistical model included the fixed effects of sunflower seed processing (SS), chitosan supplementation (CHI), and their interaction (SS × CHI):

Yijk=μ+SSi+CHIj+(SS×CHI)ij+eijk

where Y_ijk is the dependent variable, μ is the overall mean, SS_i is the effect of sunflower seed processing (i = whole or ground), CHI_j is the effect of chitosan inclusion (j = 0 or 4 g/kg DM), (SS × CHI)_ij is the interaction effect, and e_ijk is the residual error.

Initial body weight was tested as a covariate and included in the model when significant (P ≤ 0.05).

Prior to analysis, the data were evaluated for normality of residuals using the Shapiro–Wilk test and for homogeneity of variances using Levene’s test.

Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., 2016), using the PROC GLM procedure.

When significant interaction effects were detected (P ≤ 0.05), they were interpreted accordingly; otherwise, main effects were considered. Statistical significance was set at P ≤ 0.05. All results are presented as least-squares means.

3 Results

3.1 Nutrient intake and digestibility

No interaction between sunflower seed processing and chitosan supplementation (CHI) was detected for nutrient intake or digestibility (P ≥ 0.118; Table 3). Nutrient intake was not affected by sunflower seed processing or CHI supplementation (P ≥ 0.125).

Table 3. Nutrient intake and apparent digestibility in ewe lambs fed diets containing sunflower seeds (whole or ground) and chitosan (0 or 4 g/kg DM).

SS, sunflower seed (whole vs. ground); CHI, chitosan; SEM: standard error of the mean; NDFap: neutral detergent fiber corrected for ash and protein; NFC, non-fiber carbohydrates; TDN, total digestible nutrients; DM, dry matter.

Sunflower seed processing affected nutrient digestibility, with lambs fed whole seeds showing greater apparent digestibility of dry matter, organic matter, non-fibrous carbohydrates, and total digestible nutrients than those fed ground seeds (P ≤ 0.007). In contrast, the apparent digestibility of ether extract, crude protein, and NDFap were not influenced by sunflower seed processing (P ≥ 0.082).

Chitosan supplementation did not affect intake or digestibility variables (P ≥ 0.138).

3.2 Ingestive behavior and rumination

Significant interactions between sunflower seed processing and chitosan supplementation (CHI) were observed for rumination efficiency of dry matter (DM; P = 0.014) and neutral detergent fiber (NDFap; P < 0.001), as well as for total rumination time (P = 0.001; Table 4). The unfolding of these interactions revealed that the impact of sunflower seed processing depended on the presence of the additive (Figures 1A, B).

Table 4. Ingestive behavior, chewing activity, and feeding patterns of ewe lambs fed diets containing sunflower seeds (whole or ground) and chitosan (0 or 4 g/kg DM).

SS, sunflower seed (whole vs. ground); CHI, chitosan; SEM, standard error of the mean; NDFap, neutral detergent fiber corrected for ash and protein. NCD, number of chews per day; TCT, total chewing time (h/day); NCB, number of chews per bolus; CTB, chewing time per bolus (s/bolus). Feed and rumination efficiencies are expressed as g of DM or NDFap per hour.

In the absence of CHI, lambs fed whole sunflower seeds showed greater DM (+22.2%) and NDFap (+29.8%) rumination efficiencies than those fed ground seeds. Conversely, when CHI was supplemented at 4 g/kg DM, no differences between seed processing methods were detected (P ≥ 0.233).

A consistent interaction pattern was observed for the rumination time (Figure 1C). Without CHI, lambs fed ground seeds spent 24.1% more time ruminating than those receiving whole seeds (421 vs. 339 min/d; P = 0.024). However, this difference was eliminated with CHI inclusion (P = 0.270). Regarding the diurnal pattern, the bunk attendance remained similar across the treatments, with peaks occurring immediately after fresh feed delivery (Figure 2).

For variables in which no interaction was observed, SS significantly influenced feeding efficiency (Table 4). Lambs fed whole sunflower seeds showed greater feeding efficiency for DM (265.1 vs. 198.0 g/h; P = 0.004) and NDFap (58.5 vs. 43.3 g/h; P = 0.004) than those fed ground seeds. No independent effects of CHI were observed on feeding efficiency (P ≥ 0.232). Other behavioral parameters, including chewing activity, weight per bolus, and idle time, were not influenced by sunflower seed processing, chitosan, or their interaction (P ≥ 0.105; Table 4).

3.3 Water intake and balance

No interaction between sunflower seed processing and chitosan supplementation (CHI) was observed for water intake or balance variables (P ≥ 0.235; Table 5).

Table 5. Water intake and water balance of ewe lambs fed diets containing sunflower seeds (whole or ground) and chitosan (0 or 4 g/kg DM).

SS, sunflower seed (whole vs. ground); CHI, chitosan; SEM, standard error of the mean.

The water intake variables included drinking water intake, water intake from feed, total water intake, and water intake expressed per unit of dry matter intake. Water excretion includes urinary and fecal water loss. Water balance was calculated as the total water intake minus the total water excretion.

Sunflower seed processing significantly affected water intake per unit of dry matter intake. Lambs fed ground sunflower seeds showed greater water intake per kg DM than those fed whole seeds (2.87 vs. 2.50 L/kg DM; P = 0.033), representing an increase of approximately 15%.

Conversely, water intake from the diet was higher in lambs fed whole sunflower seeds (0.51 vs. 0.47 L/d; P = 0.008), corresponding to an approximately 8% increase.

No effect of SS was observed on total water intake, urinary water excretion, fecal water excretion, or overall water balance (P ≥ 0.370).

Chitosan supplementation did not affect any of the water-related variables (P ≥ 0.232).

3.4 Nitrogen metabolism and estimated microbial nitrogen production

No interaction between sunflower seed processing and chitosan supplementation (CHI) was observed for nitrogen metabolism or estimated microbial nitrogen production variables (P ≥ 0.115; Table 6).

Table 6. Nitrogen balance, plasma urea concentrations, and estimated microbial nitrogen production in ewe lambs fed diets containing sunflower seeds (whole or ground) and chitosan (0 or 4 g/kg DM).

SS, sunflower seed processing (whole vs. ground); CHI, chitosan; SEM: standard error of the mean; CP, crude protein; TDN, total digestible nutrients.

Nitrogen intake and fecal, urinary, and total nitrogen excretion were not affected by sunflower seed processing or chitosan supplementation (P ≥ 0.142). Lambs fed whole sunflower seeds showed greater nitrogen absorption (16.94 vs. 15.74 g/d; P = 0.021), nitrogen retention (10.55 vs. 9.19 g/d; P = 0.010), plasma urea nitrogen concentration (22.7 vs. 17.5 mg/dL; P < 0.001), estimated microbial nitrogen production (2.84 vs. 2.01 g/d; P = 0.033), and microbial efficiency (30.35 vs. 23.22 g CPmic/kg TDN; P = 0.019) than those fed ground seeds. Chitosan supplementation did not affect these variables (P ≥ 0.441).

3.5 Growth performance

No interaction between sunflower seed processing and chitosan supplementation (CHI) was observed for performance variables (P ≥ 0.764; Table 7).

Table 7. Growth performance and feed efficiency of ewe lambs fed diets containing sunflower seeds (whole or ground) and chitosan (0 or 4 g/kg DM).

SS, sunflower seed (whole vs. ground); CHI, chitosan; SEM, standard error of the mean. FE, feed efficiency, expressed as grams of weight gain per kilogram of dry matter intake. WIEWG, water intake efficiency per unit of body weight gain (L/kg). IBW, initial body weight, included as a covariate (mean = 15.12 ± 3.57 kg).

Sunflower seed processing affected growth performance. Lambs fed whole sunflower seeds showed greater final body weight (25.8 vs. 24.2 kg; P = 0.015), average daily gain (254.1 vs. 215.2 g/d; P = 0.015), and feed efficiency (313.2 vs. 290.1 g gain/kg DM intake; P = 0.028) than those fed ground seeds.

In contrast, the water intake per unit of body weight gain was lower in lambs fed whole sunflower seeds (8.2 vs. 10.0 L/kg; P = 0.018), representing a reduction of approximately 18%.

Chitosan supplementation did not affect the growth performance or efficiency variables (P ≥ 0.231).

3.6 Economic Indicators

Economic indicators were assessed descriptively (Table 8). Chitosan supplementation increased diet cost from an average of US$ 0.215 to US$ 0.270/kg DM. Under the price assumptions used, whole sunflower seeds without chitosan showed the highest descriptive net margin (US$ 7.87 per animal), whereas ground sunflower seeds with chitosan showed the lowest descriptive net margin (US$ 4.57 per animal). The cost-to-revenue ratios were 48.43 and 64.91%, respectively.

Table 8. Economic indicators of ewe lambs fed diets containing sunflower seeds (whole or ground) and chitosan (0 or 4 g/kg DM).

Average values (US$) per ton of dry matter in Feed: 144.64 US$ (sorghum silage); US$ 150.60 (grain corn); US$ 216.99 (soybean meal); US$ 559.16 (urea); US$ 21,428.57 (chitosan); US$ 68.54 (limestone); US$ 450.94 (mineral mixture) and (bicarbonate) US$ 1,435.79.

Feed conversion multiplied by the cost of the diet.

Cost of weight gain multiplied by total weight gain.

Total weight gain multiplied by the price received.

Revenue minus total feed cost. Values were originally calculated in Brazilian Real and converted to US dollars using the average exchange rate at the time of the experiment.

4 Discussion

4.1 Physical form of sunflower seeds as a determinant of nutrient utilization

Whole sunflower seeds increased the apparent digestibility of dry matter, organic matter, non-fiber carbohydrates, and total digestible nutrients compared with ground sunflower seeds, despite similar nutrient intake. Because the diets had similar chemical composition, these responses are better interpreted in relation to differences in the physical form of sunflower seeds and the particle-size profile of the complete diets. The intact seed matrix may have influenced the pattern of lipid exposure during digestion, providing a coherent interpretation for the greater nutrient utilization observed with whole seeds. Direct confirmation of this mechanism would require measurements of sunflower seed particle size, lipid release, ruminal fermentation, and microbial community composition.

In high-concentrate diets, ruminal conditions are typically characterized by a high supply of rapidly fermentable carbohydrates, reduced pH, and lower abundance of fibrolytic microorganisms. Under these conditions, the ruminal ecosystem becomes more susceptible to additional dietary factors such as lipids. Vargas et al. (2023) reported that variations in ruminal pH and substrate availability can alter fermentation characteristics, whereas Li et al. (2024) observed diet-associated shifts in the ruminal microbiome under high-concentrate feeding systems. Within this context, the biological response to dietary lipids depends not only on their inclusion level but also on their accessibility to rumen microorganisms.

The processing of oilseeds modifies this accessibility. Grinding increases the exposure of intracellular components, including lipids, whereas whole seeds maintain structural barriers that delay microbial access. As discussed by Huang et al. (2021), intact oilseeds can modulate the kinetics of lipid availability in the rumen, whereas Kubelková et al. (2018) reported that increased lipid accessibility alters fermentation patterns. Therefore, the lower digestibility of organic matter and non-fiber carbohydrates observed in lambs fed ground seeds may be consistent with differences in the interaction between intracellular lipids and digestive processes.

A key aspect of this interaction is the availability of unsaturated fatty acids. When more accessible in the ruminal environment, these compounds can alter microbial activity and fermentation dynamics, particularly among the microbial groups involved in fiber degradation, as discussed by Ibrahim et al. (2021). Likewise, Sun et al. (2022) highlighted that dietary lipids can influence the composition and activity of ruminal microbiota, with responses depending on the fatty acid profile and availability. Therefore, differences in lipid accessibility could influence microbial efficiency and nutrient utilization even when total lipid intake remains unchanged. In the present study, altered exposure of intracellular components represents one plausible interpretation, but it was not directly measured.

Conversely, the structural integrity of whole seeds may have moderated the exposure of intracellular components during digestion. This interpretation is consistent with the greater apparent nutrient digestibility, nitrogen retention, estimated microbial nitrogen production, and growth performance observed in ewe lambs fed whole seeds. However, because ruminal fermentation, lipid release, and digesta kinetics were not directly measured, these responses cannot be attributed conclusively to differences in ruminal lipid accessibility or microbial activity.

4.2 Effects of sunflower seed processing on nutrient digestibility and energy utilization

Ether extract digestibility did not differ between whole and ground sunflower seeds (853.1 vs. 819.4 g/kg), indicating that total-tract lipid disappearance was not significantly affected by seed processing. This pattern suggests that the differences observed in animal performance were not explained by total-tract lipid disappearance alone and may instead have involved changes in the pattern or site of nutrient digestion. However, because ruminal and post-ruminal digestion were not evaluated separately, the specific site at which these responses occurred cannot be established.

In contrast, the higher digestibility of non-fiber carbohydrates (830.4 vs. 780.4 g/kg) and organic matter in the ewe lambs fed whole seeds reflects greater apparent utilization of fermentable substrates. In high-concentrate diets, microbial degradation of carbohydrates depends on the interaction between substrate availability and microbial activity, both of which can be influenced by the presence and accessibility of dietary lipids in the rumen. Previous studies have shown that lipid supplementation can modify fermentation patterns without necessarily reducing overall digestibility, depending on the interaction between lipids and ruminal microorganisms (Vargas et al., 2020b). Therefore, differences in the exposure of intracellular seed components provide one plausible explanation for the observed carbohydrate digestibility responses, although effects related to particle size, passage rate, and site of digestion cannot be excluded.

Importantly, the effect of seed processing should not be interpreted as equivalent to the inclusion of free oils. Lipids in oilseeds remain physically associated with plant tissues and intracellular structures, even after mechanical processing. Consequently, grinding increases the exposure of intracellular contents and the accessibility of lipids to ruminal microorganisms, but does not necessarily result in the immediate availability typically observed with liquid fat sources. Studies evaluating oilseed-based lipid supplementation have shown that the biological responses to dietary lipids depend not only on the lipid supply itself but also on the form in which lipids are presented within the diet (Gandra et al., 2016; Dias et al., 2020). Therefore, the observed differences in carbohydrate digestibility are consistent with altered exposure of intracellular seed components as one possible explanation, rather than providing direct evidence of physicochemical interference by lipids with feed particles.

Increasing the surface area through grinding may increase the exposure of intracellular components to digestive and microbial processes. Previous studies have associated greater lipid accessibility with changes in ruminal fermentation and microbial responses (Vargas et al., 2020a). Thus, the lower non-fiber carbohydrate digestibility and estimated microbial nitrogen production observed with ground seeds may be consistent with altered nutrient–microbe interactions. However, differences in particle-size distribution, digesta kinetics, energy–nitrogen synchronization, and site of digestion may also have contributed, and the underlying mechanism cannot be confirmed from the measurements obtained in this study.

4.3 Nitrogen metabolism and estimated microbial nitrogen production

Estimated microbial nitrogen production increased by 41% (2.84 vs. 2.01 g/d), and estimated microbial efficiency was 31% greater (30.35 vs. 23.22 g CPmic/kg TDN) in lambs fed whole sunflower seeds than in those fed ground seeds. These responses were accompanied by greater N absorption (+8%; 16.94 vs. 15.74 g/d) and N retention (+15%; 10.55 vs. 9.19 g/d), despite similar N intake, indicating more favorable nitrogen utilization with whole seeds.

Plasma urea concentration was also greater in ewe lambs fed whole sunflower seeds, suggesting a greater flux of nitrogen through the urea pool. In ruminants, plasma urea reflects the balance among ammonia absorption, hepatic urea synthesis, gastrointestinal urea recycling, and urinary excretion (Nichols et al., 2022). Therefore, the greater plasma urea concentration should not be interpreted independently as evidence of improved nitrogen-use efficiency.

Microbial protein synthesis depends on the synchronization between fermentable energy and available nitrogen. Greater exposure of intracellular lipids after grinding may alter microbial membrane function and metabolic activity, thereby reducing the efficiency of nitrogen incorporation into microbial biomass (Kubelková et al., 2018; Ibrahim et al., 2021). Similarly, Mirzaei-Alamouti et al. (2021) reported shifts in microbial community structure and fermentation pathways in response to lipid supplementation, whereas Vesga et al. (2024) observed transient reductions in microbial protein synthesis during adaptation to high-lipid diets.

The greater estimated microbial N production and N retention observed with whole seeds may be associated with differences in the exposure of intracellular seed components and nitrogen utilization. However, contributions from energy–nitrogen synchronization, digesta passage, post-ruminal nitrogen utilization, and metabolic nitrogen partitioning may also have influenced these responses.

4.4 Ingestive behavior and water intake responses

In the absence of chitosan, ewe lambs fed whole sunflower seeds spent 24% less time ruminating (339 vs. 421 min/d) and exhibited greater rumination efficiency (g DM or NDFap/h) than those fed ground seeds. Despite the lower time of rumination, nutrient digestibility was higher in animals fed whole seeds, indicating that chewing activity was not a limiting factor for digestion under the conditions of this study.

Although differences in ingestive behavior are often associated with the physical characteristics of the diet, including particle size and structural integrity, substrate fermentability and digesta kinetics may also influence rumination responses. In high-concentrate diets, fermentable carbohydrate supply and diet physical structure can jointly affect chewing activity and digestive processes (Zhang et al., 2017; Wang et al., 2020). Because ruminal fermentation and digesta passage were not directly measured, the relative contribution of these factors cannot be established.

This framework helps explain the observed reduction in rumination time in the ewe lambs fed whole seeds. Lower rumination time may reflect differences in diet physical form, particle-size distribution, digesta passage, or substrate degradation associated with sunflower seed processing. Conversely, the greater rumination time observed with ground seeds may have been associated with differences in diet structure, particle distribution, or digesta kinetics. These alternatives cannot be separated with the measurements available.

Water intake also differed between seed-processing methods. Lambs fed ground sunflower seeds showed greater water intake per unit of dry matter intake (2.87 vs. 2.50 L/kg DM). This response may have been influenced by differences in feed physical characteristics, nutrient utilization, digesta dynamics, or metabolic demand. However, because ruminal osmolarity, fermentation acids, drinking-event patterns, and digesta passage were not measured, its physiological basis cannot be established. High-concentrate diets are characterized by intense microbial fermentation and rapid substrate turnover (Zhang et al., 2017; Wang et al., 2020). Therefore, the greater water intake observed in animals fed ground seeds may reflect differences in feed physical characteristics, nutrient utilization, digesta dynamics, or metabolic demand.

The greater water intake per unit of weight gain observed with ground seeds (10.0 vs. 8.2 L/kg gain) resulted from the combined differences in water intake and body weight gain. Accordingly, this variable should be interpreted as a descriptive measure of water use relative to growth rather than as direct evidence of altered ruminal homeostasis or water-regulatory mechanisms.

4.5 Chitosan responses under high-concentrate dietary conditions

The dose evaluated was within a biologically relevant range. Based on average intake and body weight, 4 g/kg DM provided an estimated 155–160 mg/kg BW per day, which was comparable to the 136–150 mg/kg BW used in studies reporting responses in nitrogen utilization, estimated microbial nitrogen production, or ruminal fermentation in lambs and buffaloes (Pereira et al., 2018; Magalhães et al., 2019; Dias et al., 2020). However, because only one inclusion level was evaluated, a dose-dependent response cannot be excluded.

Positive responses to chitosan have been reported under other dietary conditions. Pereira et al. (2018) observed increased estimated microbial protein supply and nitrogen retention in feedlot lambs at doses close to those used in the present study, whereas Magalhães et al. (2019) reported changes in nitrogen utilization in lambs receiving processed cottonseed. Dias et al. (2020) also observed changes in ruminal fermentation and nutrient metabolism in buffaloes fed whole raw soybean. In contrast, the present study detected no additional effects of chitosan on nutrient digestibility, nitrogen metabolism, estimated microbial nitrogen production, or growth performance. These contrasting responses indicate that dose alone does not determine chitosan efficacy and that forage-to-concentrate ratio, lipid source, seed physical form, animal category, and product characteristics may modify the biological response.

The absence of additional measurable productive responses under the conditions evaluated may reflect the dependence of chitosan efficacy on ruminal environment, diet composition, additive dose, and interactions with other dietary components.

Chitosan is a cationic biopolymer whose antimicrobial activity is primarily mediated by electrostatic interactions with negatively charged microbial cell membranes, leading to disruption of membrane integrity and interference with nutrient transport, particularly in gram-positive bacteria (Kong et al., 2010). However, the magnitude of this effect is strongly influenced by ruminal conditions, including pH, substrate availability, and interactions with dietary components.

High-concentrate diets are generally associated with rapid carbohydrate fermentation, lower ruminal pH, and greater dominance of acid-tolerant and amylolytic microbial populations. Under these circumstances, the scope for additional modulation by antimicrobial additives may be limited. Moderate lipid inclusion can also alter microbial populations and fermentation pathways without necessarily improving overall efficiency (Vargas et al., 2020a, Vargas et al., 2020b; Ibrahim et al., 2021). Thus, the absence of a measurable response to chitosan may have been related to the prevailing dietary environment, although a dose-dependent effect cannot be ruled out.

This interpretation is also biologically consistent with a possible interaction between chitosan and dietary lipids. Unsaturated fatty acids, particularly when more accessible, can exert antimicrobial effects on ruminal microorganisms, including protozoa and fibrolytic bacteria. Mirzaei-Alamouti et al. (2021) reported shifts in microbial community structure following lipid supplementation, whereas Vesga et al. (2024) observed transient reductions in microbial efficiency during adaptation to high-lipid diets. Therefore, overlapping antimicrobial effects of dietary lipids and chitosan represent one possible explanation for the limited marginal response to chitosan, although this mechanism was not directly evaluated.

The physicochemical behavior of chitosan may also influence its effectiveness. Although mildly acidic conditions may increase chitosan protonation and cationic charge, this does not necessarily result in greater biological efficacy because enhanced nonspecific binding to dietary proteins, fiber, and fatty acids may reduce the fraction available for direct interaction with microbial cells (Kong et al., 2010), particularly in lipid-containing diets.

Responses to chitosan are therefore likely to depend on the combined effects of dose, product characteristics, diet composition, and ruminal conditions. The variable responses reported at 10 g/kg DM by Kirwan et al. (2021) and at 14 g/kg diet in association with unsaturated fatty acid sources by Gordiano et al. (2023) further indicate that higher inclusion levels do not necessarily produce consistent benefits across dietary contexts.

The interaction observed for rumination behavior suggests that chitosan may have exerted secondary effects, particularly with ground sunflower seeds. However, these responses did not translate into measurable improvements in nutrient utilization or animal performance at the inclusion level evaluated.

4.6 Integrative perspective: linking nutrient utilization and animal performance

Collectively, the responses observed for nutrient digestibility, nitrogen utilization, and animal performance showed a consistent biological pattern. Lambs fed whole sunflower seeds exhibited greater digestibility of non-fiber carbohydrates (830.4 vs. 780.4 g/kg; +6.4%) and organic matter, which was accompanied by a greater estimated microbial nitrogen production (2.84 vs. 2.01 g/d; +41%) and greater nitrogen retention (10.55 vs. 9.19 g/d; +15%). These responses indicate more efficient overall utilization of dietary energy and nitrogen.

This pattern of greater nutrient utilization was accompanied by improved animal performance. Whole-seed diets increased average daily gain (254.1 vs. 215.2 g/d; +18%) and improved feed efficiency (+8%), despite similar dry matter intake across treatments. The performance advantage observed with whole sunflower seeds was therefore associated with greater nutrient utilization rather than greater nutrient intake.

From a mechanistic perspective, the observed pattern is consistent with possible differences in the exposure of intracellular seed components during digestion. The structural integrity of whole seeds may have moderated this exposure and contributed to the observed nutrient-utilization responses. However, lipid release, ruminal fermentation, microbial activity, and digesta kinetics were not measured; therefore, this explanation remains inferential.

Behavioral and physiological responses further support this interpretation of the results. Lambs fed whole seeds showed lower water intake per unit of gain and reduced rumination time; these responses may have been influenced by differences in diet structure, digesta kinetics, nutrient utilization, and growth rate.

Chitosan supplementation did not produce additional measurable effects on nutrient utilization or animal performance under the conditions evaluated.

Under the price assumptions used, lambs fed whole sunflower seeds without chitosan showed the highest descriptive net margin, reaching US$ 7.87 per animal, which was 72.2% greater than that observed for ground sunflower seeds with chitosan. Under the adopted calculation framework, this difference resulted from the combination of lower estimated feeding costs and greater observed weight gain.

The economic relevance of chitosan supplementation depends on whether its additional cost is accompanied by measurable productive benefits. Piboonkunsamlit et al. (2025) emphasized that the practical viability of chitosan should be evaluated in relation to its effects on feed efficiency and animal performance. In the present study, the absence of improvements in productive responses resulted in lower descriptive net margins for diets containing chitosan under the price conditions evaluated.

These results indicate that the performance advantage of whole sunflower seeds was associated with greater nutrient utilization rather than greater nutrient intake. Under the bioeconomic scenario evaluated, using whole sunflower seeds without chitosan provided the most favorable combination of biological performance and feeding cost.

4.7 Limitations and future directions

Chitosan was evaluated at a single inclusion level using one product with specific physicochemical characteristics; therefore, the absence of additional productive responses should be interpreted within the dose, product, and dietary conditions evaluated, and a dose-dependent response cannot be excluded.

Although particle-size distribution was determined for the complete diets, the particle-size distribution of sunflower seed particles themselves and lipid release from the seed matrix were not directly quantified. These measurements would have provided a more direct basis for linking sunflower seed processing to seed disruption, lipid exposure, and nutrient utilization.

This study also did not include direct measurements of ruminal pH, volatile fatty acid profiles, ammonia concentrations, fatty acid flow, or microbial community composition. Therefore, the proposed mechanisms should be interpreted as biologically supported inferences based on apparent digestibility, estimated microbial nitrogen production, nitrogen balance, and performance responses. Alternative contributions from diet particle-size distribution, digesta passage, site of digestion, nutrient synchronization, and post-ruminal nutrient utilization cannot be excluded.

The economic indicators were deterministic, scenario-specific estimates based on fixed ingredients and live-animal prices. Confidence intervals and sensitivity analyses across alternative market scenarios were not performed; therefore, the economic ranking may change under different price conditions.

Future studies should directly evaluate ruminal fermentation and microbial structure to confirm whether seed processing alters lipid–microbe interactions and nutrient use in high-concentrate diets. Measurements of pH, VFA, ammonia dynamics, biohydrogenation intermediates, and microbial populations using 16S rRNA sequencing or metagenomics would help to clarify these mechanisms. Estimated microbial nitrogen production and urinary nitrogen excretion were derived from spot urine samples using creatinine-based estimation of daily urinary volume. These estimates rely on the assumption of relatively constant creatinine excretion and may be affected by individual variation and the representativeness of the sampling time.

Further studies should determine whether the physical form of oilseeds affects methane emissions, hydrogen partitioning, and response to rumen-modulating additives. This would help establish whether the structural modulation of oilseed lipids can complement or replace additive-based strategies in feedlot systems.

5 Conclusion

In high-concentrate diets for feedlot ewe lambs, sunflower seed processing affected nutrient utilization and animal performance. Feeding whole sunflower seeds, rather than ground seeds, increased the apparent digestibility of organic matter and non-fiber carbohydrates, nitrogen retention, estimated microbial nitrogen production, average daily gain, and feed efficiency, without affecting dry matter intake. Chitosan supplementation at 4 g/kg DM did not provide additional measurable benefits in nutrient intake, digestibility, nitrogen metabolism, or animal performance under the conditions evaluated. Under the price assumptions used, whole sunflower seeds without chitosan showed the highest descriptive net margin. These findings support the use of whole sunflower seeds as a practical feeding strategy in the high-concentrate diets evaluated.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. During the preparation of this work, the authors used Paperpal for language refinement and editing support. Mendeley was used for reference organization and standardization. After using these tools and services, the authors reviewed and edited the content as needed and took full responsibility for the content of the publication.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Femi Fawole, University of Ilorin, Nigeria

Reviewed by: Hani M. El-Zaiat, Sultan Qaboos University, Oman

Seyed Morteza Vaghar Seyedin, University of Birjand, Iran