Chemical, nutritive, fermentative and microbial composition of almond hull silage

ABSTRACT

Almond hulls are generally used as feed after drying. In this study, silage was made from almond hulls and quality and feed potential were investigated. In this context, chemical, fermentation, microbiological analyses and digestion–energy calculations were made. The total weight of the almond fruit was 10.47 ± 1.68 g, the hull weight was 6.56 ± 1.30 g and the percentage of the hull to the total fruit was 62.48 ± 6.57. In the almond hull silage samples, pH 5.75 ± 0.01, NH₃–N 112.5 ± 1.45 g kg⁻¹, lactic acid 63.79 ± 1.83 g kg⁻¹, acetic acid 22.94 ± 1.93 g kg⁻¹, propionic acid 28.27 ± 1.45 g kg⁻¹ and butyric acid 9.59 ± 0.88 g kg⁻¹ were determined. Yeast–mould and enterobacter were not detected and the lactic acid bacteria was 4.54 ± 0.04 log cfu g⁻¹. The most potassium 1212.50 ± 23.90 ppm, calcium 621.36 ± 23.91 ppm and magnesium 586.25 ± 21.43 ppm minerals were found. Crude protein was 92.5 ± 5.9 g kg⁻¹, crude ash 111.3 ± 1.1 g kg⁻¹, acid detergent fibre 282.7 ± 2.9 g kg⁻¹, neutral detergent fibre 394.3 ± 2.7 g kg⁻¹, total digestible nutrients 521.7 ± 8.5 g kg⁻¹, metabolic energy 1.85 ± 0.03 mcal kg⁻¹ and net energy lactation 1.11 ± 0.03 mcal kg⁻¹ were determined. It was concluded that although almond hull fermentation properties are partially desirable, they can be stored and used as silage.

KEYWORDS:

• Almond hull
• almond hull silage
• silage fermentation

Introduction

The use of alternative sources as feed for profitable and sustainable livestock is attracting attention and increasing its importance day by day due to reasons such as increasing food requirements with the population, increase in feed prices, climate change and supply-demand imbalance. In this sense, the green-looking outer hull, which is one of the by-products obtained after the almond harvest, has been used as feed for many years, especially by California farmers. Due to its high nutrient content, almond is seen as a valuable food worldwide and the interest in its cultivation is increasing day by day (Salgado-Ramos et al. 2022), the amount of by-products increases with increasing production and these by-products are considered an alternative source for feed and other sectors (Wang et al. 2021). According to the International Nut and Dried Fruit Council 2020 report, this increase has been 26% in the last 10 years. Minimization of waste products and evaluation of by-products play a key role in terms of sustainability and proper management (Barral-martinez et al. 2021).

Almond fruit consists of three basic parts: outer hull, inner shell and nut. Approximately 50% of the total fruit consists of the outer hull, 25% of the inner shell and 25% of the nut (Aguilar et al. 1984; Fadel 1999; Almond Board Of California 2018). According to the Almond Board of California data, 2233 billion kgs of almond hulls were produced in the 2020–2021 production season and it constitutes 49% of the total fruit weight (Almond Board of California 2021). According to these data, the almond purchased from the nut shop has about three times as many by-products and the outer hull of these products is about twice as much.

Almond hull has a major content in terms of structural carbohydrates such as ADF 32.20% ± 10.69 and NDF 38.66% ± 12.92 (Dairy One 2022), and non-structural carbohydrates such as glucose, fructose and sucrose (between 25% and 30% in total) (Offeman et al. 2014; Holtman et al. 2015). Alternative feed sources often have high fibre content (Wang et al. 2021), in this sense, the almond hull is also classified according to this content. According to the California Department of Food & Agriculture, to qualify as an almond hull, it should not contain more than 13% moisture, more than 15% fibre and more than 9% ash (CDFA 2013). According to the unofficial acceptance in the feed industry, almond hulls are considered prime hulls if their crude fibre content is less than 15% (Swanson et al. 2021).

Although almond hull is rich in structural and non-structural carbohydrates, its protein content (5.95% ± 2.89) is low (Dairy One 2022). A high fermentable carbohydrate content indicates that it can be used as a substitute for concentrated feed, and a high NDF ratio indicates that it can be considered as a substitute for roughage feed. In this respect, it can be considered two-way feed in its ration content. Californian farmers use almond hulls to reduce the amount of land to be cultivated for silage and to replace some of the silage. Almond hull can be considered as a product that can be mixed into corn silage due to the similarity of its energy values with corn silage, its high sugar content, its flavour to the food and its alternative to high-energy feeds (Oliveira 2021).

Almond hull is anatomically similar to the fleshy part of the peach and is used as a feed raw material in ruminant diets in different parts of the world (DePeters et al. 2020). They are used between 5% and 9% in the ration content of California dairy cattle farms and some of their needs for other feed sources are replaced in this way (Almond Board of California 2021) and it can be used safely up to 20% in ruminant diets (Swanson et al. 2021). Studies have been carried out in which almond hull is partially used instead of alfalfa in ruminate rations (Reed and Brown 1988; Rad et al. 2016), its effect on egg quality is examined (Wang et al. 2021) and nutrient contents and digestibility values are determined (Jafari et al. 2015; DePeters et al. 2020; Swanson et al. 2021; Swanson et al. 2021). In addition, although studies express that it reduces milk yield and dry matter consumption, it is recommended to be used in terms of low cost and evaluation of a by-product (Alibés et al. 1983; Williams et al. 2018). Studies on the use of almond hulls as feed continue to be carried out by scientists and industry stakeholders.

As a result, the almond hull can be dried during and after harvesting and stored for a long time as a raw material with low moisture content (Jafari et al. 2011). But when stored outside, it acts like a sponge, absorbing moisture and mould grows quickly. When it is damaged by rain, sugar content is lost and feed value decreases (Asmus 2015). In this respect, the method of preserving the almond hull as wet is a different issue. Most of the studies on the almond hull were done on dry material. In the light of this information, our hypothesis in this study is that almond hulls with a certain moisture content (60–70%) can be preserved as silage, and in this context, our aim is to evaluate the silage characteristics of almond hulls and their feed potential as silage.

Material and methods

Material collection and silage preparation

In this study, the outer hulls of almond fruit harvested from ‘Texas’ variety trees were evaluated (Figure 1). Almonds were collected from the orchard in Yazıbaşı Village (38°20′ N 35°34′ E) with an altitude of 1391 m, within the borders of the Develi district of Kayseri province in Turkey. Almonds were harvested by hand at the end of October 2021. To determine the amount and percentage of almonds and hulls, randomly 50 almond fruit and hulls were weighed.

Figure 1. The almond tree, wet almond hull (fresh), silage of almond hull (from left to right).

The cracked outer hulls of the fruit were separated from the interior by hand and collected in a container. Almond hulls were mixed and homogenized and transferred to polyethylene vacuum bags (Caso Professional Vacum Rolls, Arsberg, Germany) of 30 × 35 cm dimensions as 1 kg in five replications without any processing (chopping, etc.). The air in the polyethylene vacuum bags containing the sample was evacuated with a vacuum packaging machine (DZ-260/PD, SELES vacuum package device, Bursa, Turkey) and the mouth of the bags was closed automatically by heat treatment. The samples were left to ferment in an oxygen-free environment in the bags. These bags were stored in the laboratory at an average temperature of 20–25°C for 90 days (1 November–30 January).

Some of the almond hulls were dried naturally in the shade, and their nutritional values were also evaluated in the study.

Organic acid, pH, NH₃–N analyses

After the fermentation period (90 days), the sample bags were opened and immediately mixed with distilled water to contain a 20% (20 g sample + 80 ml distilled water) sample. The mixture was filtered through filter papers and the acidity of the filtrate was determined by a pH meter. For organic acid and NH₃–N analysis, 40 g of silage sample was shaken vigorously with 360 ml distilled water and filtered through the Whatman (no:1) filter paper. For NH₃–N analysis, 100 ml of filtrate was taken, put into the Kjeldahl wet burning tube placed in the distillation device (Gerhardt VAP 20, Königswinter, Germany) and 32% NaOH was added to it. The distillate was collected in a 50 ml of 2% boric acid solution using a distillation device, and this distillate was titrated with 0.1 N HCl to determine NH₃–N (Canpolat 2019). For organic acid analysis, some of these filtrates were stored in screw cap tubes at −20°C until the time of analysis (10 days). For lactic acid analysis, a 10⁻³ solution was prepared from the 10⁻¹ filtrate prepared for other analyses, using distilled water. 0.1 ml of solution was taken and 0.1 ml of copper sulphate and 6 ml of concentrated sulphuric acid were added to it and mixed with vortex. Then 0.1 ml of parahydroxybiphenyl was added to it, after vortexing, it was kept in boiling water for 1.5 min and cooled. The absorbance value was read in the spectrophotometer at 560 nm wavelength. The lactic acid content of the samples was determined by using the values read at 560 nm wavelength of the standard solutions containing lithium lactate (Canpolat 2019). The amount of acetic, propionic and butyric acid was determined by gas chromatography (GC 2010+ Shimadzu Corporation, Kyoto, Japan).

Chemical analyses and calculated values

Shade-dried almond hulls, evaluated in their natural state, were ground in a laboratory mill (IKA MF.10, Staufen, Germany) to pass through a 1 mm sieve and made ready for chemical analysis. Almond hull silages were ground after opening the vacuum bags and drying in an oven at 60 °C until they reached a constant weight (∼24 h) and put into ziplock bags for chemical analysis.

Crude protein (CP) was determined by the DUMAS method (AOAC 2006). This method, burning the sample in the furnace in the device (VELP NDA 701, Usmate Velate, Italy), reduces all nitrogen forms in it to elemental nitrogen by converting them to nitrogen oxide gases, and determining the amount of nitrogen by the thermal conductivity method, multiplying this amount by the protein factor and determining the crude protein value. Ether extract (EE) analysis was performed by the extraction method (ANKOM XT15, Macedon NY, USA) and petroleum ether was used as a solvent (AOCS 2004). Crude ash (CA) analysis was performed by burning the samples in a 550°C ash furnace (CARBOLITE ELF 11/6, Sheffield, UK) (AOAC 2005). Crude fibre (CF) analysis was carried out based on the detection of the burning part by boiling the defatted samples first in sulphuric acid, then in sodium hydroxide solution and finally burning the remaining mass (ISO 6865 2000). Acid detergent fibre (ADF) analysis was performed by boiling the sample in an acid detergent solution and neutral detergent fibre (NDF) analysis in a neutral detergent solution by determining the amount of remaining mass. Lignin analysis was carried out by determining the amount of the remaining samples after ADF analysis treated with concentrated (72%) sulphuric acid for a certain period (3 h) (AOAC 1997, 2022). An ANKOM²⁰⁰⁰ (Macedon NY, USA) analyzer was used for CF–ADF–NDF and lignin analyses. For acid detergent insoluble protein (ADICP) and neutral detergent insoluble protein (NDICP), CP analysis was performed on the basis of the method given above from the residues resulting from ADF and NDF analysis. Starch analysis was determined by the polarimetric method (ISO 10520 1997). Total sugar (glucose + fructose + maltose + sucrose) analysis was determined by the HPLC (Agilent, 1260, California, USA) method. For mineral contents, 0.5 g of samples was taken into vessel tubes, 10 ml HNO₃ and 2 ml HCl were added and 200°C was subjected to 15 min microwave (SPEEDWAVE, Jena, Germany) thawing at 1600 W. After the process, it was cooled to room temperature and the 0.2 µm syringe tip was filtered, and then the mineral amounts were determined in the ICP-MS (AGILENT 7500, California, USA) device (AOAC 2009).

Using the data obtained from chemical analysis, digestibility and energy parameters were calculated. These parameters are non-fibre carbohydrate (NFC), digestible dry matter (DDM), dry matter intake (DMI_BW %, body weight of animal), total digestible nutrients (TDN_1X), metabolic energy (ME) and net energy lactation (NEL) values. These parameters were calculated according to the formulas specified in Nutrient Requirements of Dairy Cattle (NRC 2001). These formulas are given below. In the formulas below, ‘td’ is truly digestible and ‘PAF’ is the abbreviation of processing adjustment factor. The PAF value was taken as 1. $\mathrm{NFC\%}=100\text{–}\left(\mathrm{CP}+\mathrm{EE}+\mathrm{CA}+\mathrm{NDF}\right)$ $\mathrm{TD}{\mathrm{N}}_{1\mathrm{X}}\mathrm{\%}=\mathrm{tdNFC}+\mathrm{tdCP}+\left(\mathrm{tdFA}\times 2.25\right)+\mathrm{tdNDF}\text{–}7$ $\mathrm{tdNFC}=0.98\times (100-\left[\left(\mathrm{NDF}-\mathrm{NDICP}\right)+\mathrm{CP}+\mathrm{EE}+\mathrm{CA}\right]\times \mathrm{PAF}$ $\mathrm{tdCPf}=C{P}^{[-1.2\times (\mathrm{ADICP}/\mathrm{CP}\left)\right]},\left(\mathrm{for}\mathrm{forage}\right)$ $\mathrm{tdFA}=\mathrm{EE}-1,\left(\mathrm{if}\mathrm{EE}<1\mathrm{less}\mathrm{than}\mathrm{tdFA}=0\right)$ $\begin{array}{rl}\mathrm{tdNDF}=& 0.75\times \left(\mathrm{NDFn}-\mathrm{ADL}\right)\times \left[1-\left(\mathrm{ADL}/\mathrm{NDFn}\right)\times 0.667\right],\\ & \left(\mathrm{NDFn}=\mathrm{NDF}-\mathrm{NDICP}\right)\end{array}$ $\mathrm{ME}\left(\mathrm{Mcal}/\mathrm{kg}\right)=\left(1.01\times \mathrm{DE}\right)-0.45,(\mathrm{if}\mathrm{EE}\le 3)$ $\begin{array}{rl}& D{E}_{1X}\left(\mathrm{Mcal}/\mathrm{kg}\right)=\left(\mathrm{tdNFC}/100\right)\times 4.2+\left(\mathrm{tdNDF}/100\right)\times 4.2\\ & +\left(\mathrm{tdCP}/100\right)\times 5.6+\left(\mathrm{FA}/100\right)\times 9.4-0.3\end{array}$ $\mathrm{N}{\mathrm{E}}_{\mathrm{L}}\left(\mathrm{Mcal}/\mathrm{kg}\right)=\left[0.703\times \mathrm{ME}\right]-0.19,(\mathrm{if}\mathrm{EE}\le 3)$ $\mathrm{DDM\%}=88,89-\left(0,779\times \mathrm{ADF}\right)$ $\mathrm{DM}{\mathrm{I}}_{\left(\mathrm{BW}\right)}\mathrm{\%}=120/\mathrm{NDF},(\mathrm{BW}:\mathrm{percent}\mathrm{of}\mathrm{body}\mathrm{weight})$

Microbiological analyses

As soon as the silage samples were opened, a 10⁻¹ dilution was prepared by mixing them with peptone water (10 g sample + 90 ml peptone water) and microbiological cultivations were made from this dilution using the spread plate technique. In this context, yeast–mould, enterobacteria and lactic acid bacteria were counted in the samples. The media prepared according to the manufacturer’s instructions were poured into Petri dishes. Potato Dextrose Agar (Merc, Darmstadt, Germany) was used as the medium for yeast–mould detection and was incubated for 5 days at 25 ± 1°C. For enterobacteria count, Violet Red Bile Agar W/Glucose (Condalab, Madrid, Spain) medium was used and incubation was carried out at 37 ± 1°C for 24 h. MRS Agar (Merc, Darmstadt, Germany) medium was used for the determination of lactic acid bacteria numbers and incubation was carried out at 37 ± 1°C for 72 h. At the end of the incubation periods, the colony-forming units were counted (using the ‘ImageJ 1.53k’ program public domain) and the results are shown in the table on a logarithmic basis.

Statistical analysis

Statistics, regarding the study, was performed with Minitab 16.1 software using a completely random one-way analysis of variance (ANOVA) procedure. All data were expressed as mean and standard deviation (mean ± stdsap) and Tukey’s family error rate test was used to determine the difference between samples. A P value greater than 0.05 was found to be significant for the difference.

Results and discussion

The weighing results of 50 samples taken randomly from almond fruits to determine the outer hull ratio are given in Table 1. The outer hull constitutes an average of 62.48% of the total weight. The results of this study differed from other studies because the fruit was weighed on a wet basis. Some studies stated that the amount of outer hull in dry almond fruit is 50% (Aguilar et al. 1984; Fadel 1999), and in some reports, 54% and 49% (Almond Board Of California 2018; Almond Board of California 2021). A review stated that the fresh hull ratio of almond fruit varies between 35% and 62%, according to the almond variety (Prgomet et al. 2017). As seen in the current and other studies, the ratio of the outer hull is at very significant levels and the results of the study are similar to the literature studies.

Table 1. Fresh amount of almond hull on total fruit (n = 50).

Indices	Results
Weight of whole fruit, g	10.47 ± 1.68
Weight of hull, g	6.56 ± 1.30
Percent of hull, %	62.48 ± 6.57

The chemical and nutritional content of fresh and silaged almond hulls are shown in Table 2. When the results are examined, it is seen that almond hull silage preserves its nutritional value when compared to its fresh form. The dry matter content of both samples was 298.4 and 310.6 g kg⁻¹ of the almond hull as fresh and silage, respectively. In the literature, the almond hull was generally evaluated in its dry form and no data were found regarding the dry matter percent of the fresh material. After that, the current study will focus on the silage value of the almond hull. However, the nutritional values of almond hulls in the fresh form before silage are also expressed in Tables 2 and 3. As well as, statistical comparisons are shown in the relevant table. Since there was no study in which an almond hull was evaluated as silage, comparisons were made over the literature results of the almond hull in its natural state.

Table 2. Chemical & nutritional composition of the almond hull as fresh and silage, g kg⁻¹ (dry matter basis), unless otherwise stated.

Indices	Fresh almond hull	Silage almond hull	SEM	P
Dry matter	298.4 ± 2.7	310.6 ± 3.3	3.02	0.000
Crude protein	79.6 ± 2.9	92.5 ± 5.9	4.63	0.002
Ether extract	9.1 ± 1.0	10.4 ± 0.5	0.77	0.025
Crude ash	113.0 ± 1.5	111.3 ± 1.1	1.30	0.077
Crude fibre	175.7 ± 2.2	183.4 ± 2.9	6.10	0.081
ADF	267.6 ± 3.8	282.7 ± 2.9	3.40	0.000
NDF	339.2 ± 3.9	394.3 ± 2.7	3.89	0.000
ADICP	27.3 ± 0.3	28.7 ± 0.8	0.59	0.005
NDICP	64.2 ± 0.3	74.4 ± 13.9	9.80	0.136
Lignin	111.4 ± 0.9	122.3 ± 4.3	3.13	0.001
Starch	58.9 ± 0.8	61.6 ± 0.8	0.28	0.001
Total sugar	5.2 ± 0.2	10.1 ± 0.3	0.27	0.000
NFC	459.1 ± 7.8	391.5 ± 7.2	7.52	0.000
TDN1X	551.4 ± 3.7	521.7 ± 8.5	6.54	0.000
DDM	680.5 ± 3.0	668.8 ± 2.3	2.65	0.000
DMI % (body weight%)	3.54 ± 0.04	3.04 ± 0.03	0.04	0.000
ME (Mcal/kg)	1.96 ± 0.01	1.85 ± 0.03	0.03	0.000
NEL (Mcal/kg)	1.19 ± 0.01	1.11 ± 0.02	0.02	0.000

DM: dry matter, ADF: acid detergent fibre, NDF: neutral detergent fibre, ADICP: acid detergent insoluble crude protein, NDICP: neutral detergent insoluble crude protein, NFC: non-fibre carbohydrate, TDN_1X: total digestible nutrition, DDM: digestible dry matter, DMI: dry matter intake of body weight, ME: metabolize energy, NE_L: net enegy lactation, SEM: Standard error means.

Table 3. Mineral composition of the almond hull as fresh and silage, ppm (dry matter basis).

Indices	Fresh almond hull	Silage almond hull	SEM	P
Calcium	532.62 ± 36.73	621.36 ± 23.91	30.99	0.002
Magnesium	478.00 ± 16.92	586.25 ± 21.43	19.31	0.000
Potassium	3234.10 ± 90.10	1212.50 ± 23.90	65.90	0.000
Phosphorus	<0.03	<0.03
Sulphur	8.15 ± 0.14	11.76 ± 0.18	0.17	0.000
Sodium	156.82 ± 9.72	165.80 ± 6.63	8.32	0.126
Manganese	3.45 ± 0.03	7.85 ± 0.14	0.10	0.000
Zinc	7.70 ± 0.18	9.14 ± 0.07	0.14	0.000
Iron	8.93 ± 0.08	11.15 ± 0.14	0.12	0.000
Aluminium	8.61 ± 0.12	8.00 ± 0.09	0.11	0.000
Silicon	2.50 ± 0.04	4.41 ± 0.10	0.08	0.000
Copper	13.21 ± 0.71	16.85 ± 0.11	0.10	0.000
Chromium	<0.001	<0.001
Cobalt	<0.005	<0.005
Nickel	0.31 ± 0.11	0.40 ± 0.11	0.01	0.000
Arsenic	<0.001	<0.001
Selenium	<0.003	<0.003
Molybdenum	<0.007	<0.007
Mercury	<0.002	<0.002

In the present study, the CP value of almond hull silage was 92.5 g kg⁻¹. CP values in studies for almond hull were 51.4–48.7 g kg⁻¹ (DePeters et al. 2020), 81.1 g kg⁻¹ (Elahi et al. 2017), 45.0 g kg⁻¹ (Swanson et al. 2021), 64.0 g kg⁻¹ (Williams et al. 2018), 32.7–26.5–32.0–23.2 g kg⁻¹ (Jafari et al. 2011) and 41.1 g kg⁻¹ (Calixto and Cañellas 1982). Although the CP value of an almond hull is low, it has different results. In the current study, the CP (92.5 g kg⁻¹) value was higher than the literature results. The mean values of CP were 59.5 g kg⁻¹ (Dairy One 2022) and 57.0 g kg⁻¹ (Feedipedia 2022) in two different feed libraries open to public access online. The CP value of the almond hull in the present study is similar to the CP value of corn silage. In addition, in a study conducted on horses fed with different amounts of the almond hull, it was determined that as the almond hull ratio increased, the CP digestion rate decreased (Clutter and Rodiek 1992).

The EE value of the silage, which was the subject of the study, was determined as 10.4 g kg⁻¹ and it was evaluated that the ether extract amount was low. In other studies, EE values were 29.6 g kg⁻¹ (Elahi et al. 2017), 37.0 g kg⁻¹ (Williams et al. 2018) and 4.4–9.1–8.4 g kg⁻¹ (Jafari et al. 2011). It can be seen that the results of each study differed from each other, and the results of the current study were compatible with the literature data.

The average crude ash value of the samples was 111.3 g kg⁻¹. In other studies, ash values are 114.8 g kg⁻¹ (Elahi et al. 2017), 59.0 g kg⁻¹ (Swanson et al. 2021), 81.2–86.1–128.3–62.7 g kg⁻¹ (Jafari et al. 2011) and 60.9 g kg⁻¹ (Calixto and Cañellas 1982). One report stated that the almond hull should not contain more than 9.0% ash at a moisture content of 13.0% (CDFA 2013). When this criterion is evaluated based on the dry matter, it means that the raw ash value over 103.5 g kg⁻¹ is not suitable in terms of quality criteria. In the current study, the ash value was slightly above the criterion. Some of the study results meet this criterion and some do not. The reason for the high ash value in some studies may be the contamination of the soil during the harvesting and drying of almond fruits. However, in the current study, since the almonds were collected by hand, there was no contamination with soil, so the ash value can be considered as an indicator of high mineral content.

The crude fibre ratio in this paper was 183.4 g kg⁻¹. In other studies, the CF value were 129.6–150.7 g kg⁻¹ (DePeters et al. 2020), 149.0 g kg⁻¹ (Swanson et al. 2021) and 106.0 g kg⁻¹ (Calixto and Cañellas 1982). The CF amount in the almond hull is not more than 150.0 g kg⁻¹ at 13% moisture, which is an important parameter in terms of quality (CDFA 2013). We can correct the value as 172.4 g kg⁻¹ on the basis of dry matter. The average CF value was expressed as 191.6 and 153.0 g kg⁻¹ in two different digital sources where the feed analysis values are indicated (Dairy One 2022; Feedipedia 2022). The CF value of the current study was determined as a value close to the quality criteria determined by CDFA. The difference between almond hull results may vary depending on the almond type, the amount of foreign matter mixed with the almond hull and the adhesion of the inner shell to the inner part of the outer hull in some species.

ADF and NDF values for this study were 282.7 and 394.3 g kg⁻¹, respectively. The study examining the effect of the almond hull used at different rates in horse diets showed that the rate of ADF digestion was not affected (Clutter and Rodiek 1992). In one study (DePeters et al. 2020), ADF and NDF ratios were 133.8–192.6 g kg⁻¹ for one variety and 158.9–220.7 g kg⁻¹ for another variety, respectively. In other studies, ADF and NDF were 296.6–598.3 g kg⁻¹ (Elahi et al. 2017), 149.0–238.0 g kg⁻¹ (Swanson et al. 2021) and 436.0–508.0 g kg⁻¹ (Williams et al. 2018). In a study in which the hulls of four different almond varieties were evaluated, it was stated that the ADF value ranged between 188.3 and 252.2 g kg⁻¹, and the NDF value between 280.5 and 326.4 g kg⁻¹ (Jafari et al. 2011). It can be seen that the results differ considerably from each other. ADF and NDF values were low in some studies and high in some studies. The results of current and other studies show that the quality of the almond hull as feed can be highly variable. The variability in these values also leads to the variability of the digestion rate.

ADICP and NDICP analysis results to determine the amount of nitrogen bound to cell wall components were 28.7 and 74.4 g kg⁻¹, respectively. No other study was found in which these values were determined in the almond hull. However, in an open-access online feed library, the ADICP and NDICP values for the almond hull are 18.5 and 23.0 g kg⁻¹, respectively (Dairy One 2022). In the present study, the results were higher than these values. It is considered that this is a difference due to the internal heating of the products, depending on the storage conditions.

The lignin value was 122.3 g kg⁻¹. The value of lignin in other studies: 76.3–86.9 g kg⁻¹ for two different varieties (DePeters et al. 2020), 72.0 g kg⁻¹ (Swanson et al. 2021) and 122.5–107.0 g kg⁻¹ in two different feed libraries that provide online access (Dairy One 2022; Feedipedia 2022). The variability of lignin value can be evaluated as the adhesion of the inner shell to the inner surface of the outer hull in some almond species. In some cases, the inner shell, which has a woody structure, cannot be completely cleaned from the outer hull. For this reason, both the lignin value and other nutritional values of the outer hull may vary, and its quality changes.

Starch content was determined as 61.6 g kg⁻¹. In different studies, starch contents were determined as 4.4–16.6 g kg⁻¹ (DePeters et al. 2020) and 9.3–26.0 g kg⁻¹ (Dairy One 2022; Feedipedia 2022). In this study, the starch content was found to be higher than in other studies. The total sugar content of almond hull silage was 10.1 g kg⁻¹. In a study on the almond hull, it was stated that the total sugar content ranged between 18% and 30% (Prgomet et al. 2017), and in another study it ranged between 30.6% and 42.0% (Offeman et al. 2014). The total sugar amount in this study does not comply with the literature. The reason is not fully understood. In this study, it was thought that a low fermentable carbohydrate ratio caused insufficient fermentation quality. However in the literature, it is stated that the total sugar content of the almond hull is high, so more studies should be done on this subject.

The values energy and digestibility were calculated as NFC – 391.5 g kg⁻¹; TDN_1X – 521.7 g kg⁻¹; DDM – 668.8 g kg⁻¹; DMI_BW% – 3.04%; ME – 1.85 Mcal kg⁻¹ and NEL – 1.11 Mcal kg⁻¹. In a similar study, two different types of almond hulls were expressed as 672.0–633.1 g kg⁻¹ for NFC, 694.6–670.0 g kg⁻¹ for TDN and 1.63–1.46 Mcal kg⁻¹ for NEL (DePeters et al. 2020). In other studies, NFC was 640.0 g kg⁻¹, NEL as 1.6 Mcal kg⁻¹ (Swanson et al. 2021), NFC as 330.0 g kg⁻¹ and TDN as 555.0 g kg⁻¹ (Alibés et al. 1983). In a table with more almond hull analysis results, TDN and NEL are 582.9 g kg⁻¹ and 1.33 Mcal kg⁻¹, respectively (Dairy One 2022). The results are partially similar to the literature data. This is due to the nutrient content that varies due to different reasons.

In the present study, the mineral contents of the almond hull as fresh and silage are given in Table 3. The order of mineral content (ppm) of almond hull silage was found as K > Ca > Mg > Na > Cu > S > Fe > Zn > Al > Mn > Si > Ni > P > Mo > Co > Se > Hg > Cr,As. The maximum amount of K was 1212.50 ppm, followed by Ca and Mg with 621.36 and 586.25 ppm, respectively. In another study, the highest mineral content of 2880 and 3450 ppm K was determined in two different types of almond hulls (DePeters et al. 2020). In different studies, the highest K was determined as 2500 ppm (Swanson et al. 2021) and 3370 ppm (Alibés et al. 1983). In this study, as in other studies, significant levels of K were detected in almond hulls. The high amount of K and its similarity to alfalfa may be a limiting factor in the transition period nutrition of cows. When the K values in this study and the literature are examined, the possibility of the almond hull and silage increasing the cationicity of the diet should be considered and the possibility that it may increase the risk of postpartum milk fever should be evaluated. Ca and Mg values were 621.36 and 586.25 ppm, respectively. In the literature, Ca (190.00–240.00 ppm) and Mg (90.00–110 ppm) were determined for two different types of almond hulls (DePeters et al. 2020), while in other studies Ca and Mg were 1080.00 and 220 ppm (Alibés et al. 1983) and Ca as 370.00 ppm (Jafari et al. 2011), respectively. It has been observed that the heavy metal contents are below the risky levels. Although the results in the literature and the results in the current study are different, the proportional amounts of minerals in the almond hull are similar and most K, Ca and Mg minerals are found in the hull.

Nutrient variations in almond hulls may due to the type of almond and the region where it grows, soil and climate, etc. The degree of purity, cleanliness and mixing ratio of the inner shell to the outer hull of the hulls should also be considered as another reason for the variation in results.

Almond hull silage microbiological analyses are given in Table 4. As a result, yeast–mould and enterobacteria were not detected in the samples. The number of lactic acid bacteria was 4.54 log cfu g⁻¹. The absence of yeast–mould and enterobacteria is a positive situation in terms of silage quality. Significant proliferation of lactic acid bacteria is considered to have a positive effect on the quality of almond hull silage but the silage acidity and organic acid results stated below do not support this situation sufficiently. In terms of microbiology, there is no similar data or literature study related to the almond hull and its silage.

Table 4. Microbiological values of almond hull silage, log cfu g⁻¹.

Indices	Results
Yeast	Not detected
Mould	Not detected
Enterobacteria	Not detected
Lactic acid bacteria	4.54 ± 0.04

The results of pH, NH₃–N, lactic acid, acetic acid, propionic acid and butyric acid for evaluating the fermentation profile are listed in Table 5. The average pH value of almond hull silage samples was 5.75 ± 0.01 and no similar study data were found in the literature. However, in the open-access feed library resource, the pH of the almond hull is 5.10. It was mentioned that the target pH level should be 4.30–5.00 in legume silages and the buffering effect of the high crude ash value (Kung et al. 2018). In the current study, the pH value could not be determined at the desired levels. It is considered that this may be due to the buffering effect of the high ash content of almond hull silage and the lower-than-expected total sugar content. The NH₃–N value was 112.5 ± 1.45 g kg⁻¹. No data were found related to this value studied in the almond hull and its silage. In a review study, it was stated that the NH₃–N value of legume silage should be between 100.0 and 150.0 g kg⁻¹ (Kung et al. 2018). Since the NH₃–N value was within the recommended range, it was evaluated that there was no significant degradation in the protein structure of the almond hull in the silage process. In terms of organic acid profile, the amount of lactic acid is 63.79 ± 1.83 g kg⁻¹, acetic acid is 22.94 ± 1.93 g kg⁻¹, propionic acid is 28.27 g kg⁻¹ ± 1.01 and butyric acid is 9.59 ± 0.88 g kg⁻¹. No data could be found in the sources to compare organic acid results. However, these results were compared with the values that should be in legume silages with <30–35% dry matter (lactic acid: 60.0–80.0, acetic acid: 20.0–30.0, propionic acid: <5, butyric acid: <5 g kg⁻¹) (Kung et al. 2018). The amounts of lactic and acetic acid were determined at the desired level, while the amounts of propionic and butyric acids were determined in undesired amounts. According to the fleig scoring, in which the silage quality is determined, the percentage distribution of lactic, acetic and butyric acid amounts in the total amount, almond hull silage is in the satisfactory silage quality group (Canpolat 2019). The presence of high amounts (>5.0 g kg⁻¹) of propionic and butyric acids can often be an indication of clostridial activity. Some clostridia bacteria have the ability to convert sugar to butyric acid, and some to convert lactic acid to butyric acid. This causes the pH level to be higher than expected (Kung et al. 2018). An increase in the amount of propionic and butyric acids may have occurred with the use of the small amount of sugar in the almond hull in the current study by clostridium. In the current study, acidity did not develop sufficiently and it does not bring the microbial flora and the organic acid profile to the desired level.

Table 5. Organic acid, pH and ammonia nitrogen values of almond hull silage, g kg⁻¹ (dry matter basis).

Indices	Results
pH	5.75 ± 0.01
NH3–N	112.5 ± 1.45
Lactic acid	63.79 ± 1.83
Acetic acid	22.94 ± 1.93
Propionic acid	28.27 ± 1.01
Butyric acid	9.59 ± 0.88

Although it is not exactly similar to this study, in another study, it was mentioned that value-added products can be produced by the fermentation of almond hulls. In this context, 0.55–0.47–0.37 g g⁻¹ of lactic acid was produced from almond hulls with different sugar content in the range of 5.5 < pH < 6.0 (Thomas et al. 2019). The fermentation pH value in this study is consistent with the pH value measured in the current study.

According to the results of a feed analysis table, which is expressed as ‘almond hull wet’ and shows that the DM ratio varies between 76.03% and 87.27%, the amounts of aflatoxin B1 and aflatoxin G1 are 0.587 and 0.237 ppm, respectively (Dairy One 2022). These results show that there may be a risk of mycotoxins in almond hulls depending on storage conditions and humidity. To prevent the mycotoxin formation in the fresh almond shell, making silage or adding it to silages such as corn silage may be important.

Conclusion

Many of the by-products that can not be consumed by humans are beneficial for cows. In this sense, almond hulls are already consumed by cows. However, in this study, the feed value was determined by making a silage of almond hull. In this paper, almond hull silage was found to be of medium quality, and this is considered to be due to the low fermentable carbohydrate content of the almond hulls used in the study. However, there is information in the literature indicating that the fermentable carbohydrate content of the almond hull is higher. As a result, the almond hull can be stored as silage while it has the appropriate moisture content, in conditions where it is not possible to dry or when it is desired to be stored by leaving it to fermentation. It is recommended to conduct more detailed (in vivo, in vitro) and different studies to support the results of this study and other studies.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support this study will be shared upon reasonable request to the corresponding author.