Optimization of the Industrial Production Process of Tunisian Date Paste for Sustainable Food Systems

Abstract

The production of date paste from second-grade date fruits is a fast-growing industrial activity which promotes more sustainable food systems. The industrial date paste process is mainly dependent on the thermal treatments of hydration and drying that precede flesh crushing. These thermal treatments are commonly performed industrially using steam hydration instead of water soaking and convective hot air drying, which are known to be energy-intensive operations leading to high greenhouse gas emissions. The objective of this work was to optimize, on the one hand, the operations of hydration and drying of dates at an industrial scale using a response surface Box–Behnken experimental design in order to reduce the energy consumption and, on the other hand, to assess the biochemical and microstructural properties of date paste produced under optimized conditions. Optimization was performed based on the measurements of sensory attributes, instrumental texture firmness, moisture content, water activity (a_w), and color parameters (L*, a*, b*), as well as on the energy savings related to the factors of hydration duration and temperature and time of drying. The optimal conditions to ensure the highest quality of the final product and the lowest energy consumption were 9.6 min of hydration at 80 °C and 3 h of drying at 52.28 °C. The biochemical analysis of the date paste produced under the optimized process showed that it is rich in dietary fibers (9.80 ± 2.10%) and total phenols (261 ± 6.2 mg gallic acid equivalent/100 g of extract). Furthermore, the studied sample exhibited a higher antioxidant potential than the raw date material as a result of the heat-inhibitory effect of oxidases. The obtained results suggest that date paste presents a good source of natural bioactive molecules and could potentially be considered as a functional food ingredient. SEM analysis showed that the microstructural properties of date paste produced under optimal conditions may promote its quality preservation during storage.

1. Introduction

In Tunisia, Palm tree (Phoenix dactylifera L.) cultivation and date production are important agricultural activities that strongly contribute to the economic growth and trade balance of the country. With a total area of 42,000 hectares and an annual production of 241,000 metric tons of dates, Tunisia is one of the world’s leading producers of dates [1].

Consequently, an important industrial activity, mainly operating in date packaging, has been developed. However, the latter activity faces a significant sustainability challenge, as more than 10% (w/w) of the production is discarded as waste or by-products [2]. Furthermore, in accordance with the UNECE DDP-08 standard [3], dates are classified into three categories, Extra, Grade, and Grade II, according to the percentage of defects that are tolerated per category.

The conversion of waste, by-products, and second-grade dates (Figure S1) into added-value products, properly utilized in various sectors, is an alternative option for more economically viable, environment-preserving, and sustainable activity. Currently, several products can be obtained from dates and derivatives, such as date powder, date syrup, and date paste [4]. Date paste is among the products obtained from second-grade varieties and by-products and is receiving significant interest for its nutritional quality and potential uses in the food industry [4].

From a nutritional point of view, date paste is a source of valuable nutrients, in particular through reducing sugars, mainly glucose and fructose, representing about 54–82% of the dry matter (DM), which provides considerable energy value. It also contains dietary fiber (13.1–15.8% total dietary fiber (TDF), 10.5–11.1% insoluble dietary fiber, and 2.3–4.7% soluble dietary fiber); minerals such as potassium (1118–1285 mg/100 g DM), magnesium (79–80 mg/100 g DM), phosphorus (76–88 mg/100 g DM) and calcium (24–37 mg/100 g DM); and other nutrients [4].

Date paste is widely employed in various formulations for sweet foods as an intermediate product. It has filler agent properties and can be a sugar substitute [5]. In effect, it is used in patisserie, biscuit-making, and the baking industry for a variety of purposes, including the preparation of creams, natural sweeteners, energy bars, and cookies [2,6].

It is also characterized by interesting techno-functional properties, namely the improvement of the stability of food formulations and the extent of the shelf life of bakery foods due to its bioactive components such as polyphenols [2]. Consequently, a growing demand for date paste by manufactures and retails of the food industry has been reported lately [7].

The date paste process is generally based on successive operations of hydration, drying, pitting, crushing, and packaging. The hydration step aims to tenderize date fruits, whereas drying allows for the adjustment of the moisture content and water activity at levels which ensure microbial stability. Date flesh obtained followed the pitting step is then typically crushed using twin screw extrusion equipment in order to produce date paste, which is subsequently packaged, generally as bars in polyethylene (HDPE) wrapping. This technology of date paste making has been described as simple in a few studies [6,7,8]. However, in the context of global warming and initiatives for reducing energy consumption, as well as the increase in end-user industries ensuring the highest quality of the final products, the optimization of the date paste process has become increasingly significant and requiring of meticulous attention. A Box–Behnken experimental design combined with the response surface methodology (RSM) is an efficient tool that can be employed for the optimization of industrial processes in the agri-food sector. A Box–Behnken design is an economical alternative to central composite designs, since it is a three-level fractional factorial design with desirable statistical properties. Nevertheless, there is limited research related to the optimization of the industrial date paste process using a Box–Behnken experimental design [9,10].

Given these circumstances, the present study aims to optimize the operations of hydration and drying of date paste using the response surface design (RSD) of Box–Behnken based on the characteristics of a reference sample obtained from second-grade Tunisian dates and produced following the industrial background and end-user requirements. The selected factors were the hydration time (0–0.5 h), drying temperature (50–70 °C), and dehydration time (3–7 h). Furthermore, the optimized sample was subjected to the characterization of the biochemical and microstructural properties and the evaluation of its antioxidant potential using DPPH (2,2-diphenyl 1-picrylhydrazyl) and ABTS (2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) radical scavenging assays.

The obtained results could be of great interest to producers and a driver for the enhancement of the sustainability of dates’ value chain through product valorization, industrial process improvement, and the reduction in their costs, as well as their environmental impacts.

2. Materials and Methods

2.1. Raw Material

Fresh second-grade Tunisian Deglet Nour dates were used in this study. Deglet Nour is the most abundant variety in Tunisia. Samples were harvested at the maturity stage in November 2022 from the date-producing oasis of the Regime Maatoug region in Kebili, southern Tunisia. The date samples were, first, characterized for their moisture content, water activity (a_w), and sucrose, glucose, and fructose contents and then stored in the company VACPA BOUDJEBEL’s cold rooms (Beni Khalled, Tunisia) at a temperature of 4 °C and a relative humidity (RH) of 68.5% prior to experiments.

2.2. Experimental Approach

For each process run performed in this study, 7700 kg of dates was hydrated inside industrial rooms, where live steam at 6 bar (gage pressure) was injected in order to reach air saturation. Hydration temperature control at 80 °C was ensured using steam regulating valves. The dates were then dried inside conventional convective industrial air dryers.

The reference sample was obtained using the following process conditions: hydration at 80 °C for 0.5 h and drying at 60 °C for 4 h.

The optimization of the industrial process was carried out with the aim of reducing energy consumption while keeping the same properties of the reference sample, namely its sensory attributes, moisture content, water activity (a_w), color parameters (L*, a*, b*), and instrumental texture firmness. The biochemical and microstructural characterization of the date paste produced under optimized conditions was then performed.

Sampling was carried out following the ISO 2859-1 [11] procedures in order to ensure the representativeness of each repetition.

2.3. Characterization of the Date Paste Reference Sample

2.3.1. Sensory Analysis

The descriptive sensory evaluation of the firmness and color of the date paste sample was performed by a group of 10 trained panelists on a 9-point structured scale, where varying ratings of “very high firmness and brown color lightness” (corresponding to 9) and “very weak firmness and color lightness” (corresponding to 1) were assigned. Before the sensory analyses, the panelists were trained on each attribute according to ISO 11036 [12].

2.3.2. Moisture Content (%H)

The moisture content of the date paste was determined according to the AOAC 972.20 method [13] using a Dried Fruit Moisture tester (Safe Food ALLIANCE DFA, Yuba City, CA, USA).

2.3.3. Water Activity (a_w)

Water activity was measured using a Lab Swif (Novasina, Switzerland) water activity meter.

2.3.4. Instrumental Firmness Analysis

One-cycle compression tests were performed using a Perten TVT 6700 texture analyzer (Perten Instruments Pty Ltd., Hägersten, Sweden). The firmness of date paste corresponds to the maximum force (N) recorded for a 10 mm cylindrical probe reaching the depth of 20 mm at a 0.5 mm/s speed.

2.3.5. Color Measurements

The color parameters were based on the CIE L*a*b* coordinates, where L* = lightness (where 0 = black, 100 = white), a* (+60* = redness and −60* = greenness), and b* (+60* = yellowness and −60* = blueness) were measured using a CR 300 colorimeter (Konika Minolta, Osaka, Japan). The measurements were performed on date paste bars under illuminant D65 and the CIE 1931 Standard Observer [14].

2.4. Optimization of the Date Paste Production Process

A Box–Behnken experimental design with three continuous factors was used to optimize and study the effect of the hydration time (A), drying time (B), and drying temperature (C) on the moisture content, water activity, firmness, and L*, a*, and b* color parameters of the date paste.

Table 1. Process variables and their levels in the Box–Behnken design for the optimization of the date paste process.

Variable	Variable Symbols		Level
	Coded	Uncoded	Coded	Uncoded
Hydration time (h)	xA	XA	−1	0
			0	0.25
			1	0.5
Drying time (h)	xB	XB	−1	3
			0	5
			1	7
Drying temperature (°C)	xC	Xc	−1	50
			0	60
			1	70

shows the coded and uncoded levels of the independent variables. The design was created and analyzed using Minitab software (v.19, Minitab Inc., State College, PA, USA). It consisted of seventeen experimental randomized runs, including five replicates at the center point (Table 1).

The factors and their levels were chosen according to the industrial background experience.

The best model for each response was determined using full quadratic models while applying backward elimination (α = 0.05). The model with the highest adjusted correlation coefficient was chosen. The regression coefficients were determined based on the analysis of variance (ANOVA) at a significance level of 0.05.

2.5. Characterization of Optimized Date Paste Sample

2.5.1. Biochemical Analysis

The ash content was measured using the gravimetry method on a 3 g sample after combustion in a muffle furnace from Nabertherm (Lilienthal, Germany) at 550 °C for 5 h. Protein was analyzed according to the Kjeldahl method, using a factor of 6.25 for the conversion of nitrogen to crude protein. Fat content was calculated based on weight loss by extraction for 8 h with petroleum ether in a Soxhlet apparatus (Falk Instruments; Triviglio, Italy). After extraction, the solvent was removed under reduced pressure at 40 °C. The residue was then dried for 2 h at 30 °C.

The total dietary fiber content was determined using the enzymatic–gravimetric method AOAC 985-29 [15]. Date paste samples were dried, then enzymatically digested with amylase, protease, and amylo-glucosidase to remove protein and starch. Four volumes of ethanol were added to precipitate the soluble dietary fibers. After centrifugation (1500 rpm, 5 min), the obtained supernatant was removed by vacuum aspiration, and the precipitate was rinsed with 78% ethanol. After a second centrifugation, the precipitate was washed with 95% ethanol and then with acetone. After the final centrifugation, the residue was dried at 40 °C and weighed. One sample was analyzed for protein, and the other was incinerated to determine the ash content. The total fiber content was determined based on the difference between the residue weight and the combined protein and ash contents of the date paste.

The total soluble sugar level of the date paste was quantified using the method of Dubois et al. [16]. Two aqueous extracts were obtained after incubation of the paste (1:10, w:v) in a water bath at 95 °C for 15 min. The extracts were collected after centrifugation (5000 rpm; 10 min). Subsequently, 1 mL of phenol (5%; w:v) was added to 1 mL of supernatant and 5 mL of concentrated sulfuric acid. Then, the mixture was incubated at 30 °C for 30 min. After cooling, the absorbance was measured at 490 nm. The total sugar content was calculated with reference to a calibration curve previously established with glucose [17]. The chromatographic separation method was performed following the method of Zhu et al. [18] with modification using an ICS 3000 and 5000 ion chromatograph (Thermo Fisher Scientific Dionex, Waltham, MA, USA). An SA10 CarboPac column (4 mm × 50 mm, i.d. 7.5 μm) and analytical column (4 mm × 250 mm, i.d. 7.5 μm) were used for separation of monosaccharides (glucose, fructose, and lactose) and disaccharides (maltose and sucrose). Detection was carried out using an electrochemical detector using pulsed amperometric mode.

2.5.2. Total Phenolics Quantification

A total phenols extract from an optimized date paste sample was prepared following the protocol described by Allouache et al. [17]. The date paste was mixed with ethanol 80% (1:10, M/V). Then, the mixture was shaken at room temperature at 250 rpm for 72 h. The solvent was removed every 24 h; and then, the obtained extracts were collected, filtered, concentrated using a rotary evaporator (RE-100; Le Mans, France) at 35 °C, and kept at +4 °C until use.

The total phenols content was determined colorimetrically using a Folin–Ciocalteau reagent as described by Al-Farsi et al. [19] with some modifications. Briefly, 100 μL of the extract was added to 500 μL of 10% Folin–Ciocalteau phenolic reagent and 1000 μL of distilled water. About 1500 μL of sodium carbonate (Na₂CO₃, 20%) was added to the tube after 10 min of incubation. After 2 h, the absorbance was measured in the dark at room temperature at 760 nm. Gallic acid (GA) was used to make the standard curve: y = 1.8517x + 0.0598, R² = 0.9986. The concentrations are expressed as milligrams of gallic acid equivalents (GAEs) per 100 g of fresh weight.

2.5.3. Determination of Antioxidant Activity

The radical scavenging test for the anionic DPPH was carried out as previously reported by Allouache et al. [17]. Accordingly, 2 mL of extract was added to an equal volume of DPPH–methanolic solution (10⁻⁴ mM). The mixture was then stirred and allowed to stand for 30 min in the dark at ambient temperature.

The absorbance was measured at 517 nm, and the result was expressed as micromoles of Trolox Equivalents/g of extract.

The cationic ABTS radical was used to assess the antioxidant activity as described by Allouache et al. [17]. Before the addition of 6.75 mg of potassium persulfate, the ABTS (38.40 mg) was diluted in water (10 mL). The solution was kept for 12 h in the dark at ambient temperature before use and then diluted in ethanol to obtain an absorbance of 0.7 at 734 nm. Then, 2 mL of extract was added to 2 mL of ABTS solution and incubated for 30 min in the dark at ambient temperature. The results were expressed as micromoles of Trolox Equivalents/g of extract.

2.6. Scanning Electron Microscopy

Date paste microstructure images were collected using a scanning electron microscope (SEM) (ESEM Quattro S, Thermo Fisher Scientific, MA, USA). The environmental SEM capability of Quattro S allows for performing analyses without special preparation or conductive coating. Measurements were performed at 10 kV accelerating voltage, 10⁻³ Pa absolute pressure (vacuum), and a working distance of about 10.7 mm. The cross-section of pores was determined using the surface measuring tool of Fiji (ImageJ v 1.54f) software.

2.7. Statistical Analysis

Measurements, apart from RSM optimization and scanning electron microscopy, were replicated three times for the characterization of the raw material and biochemical analyses of paste produced under optimized conditions, as well as four times for the reference sample. Results were expressed as mean ± standard deviation.

A t-test and one-way ANOVA statistical analysis were performed using IBM SPSS^® Statistics V21.

3. Results and Discussion

3.1. Raw Material Characterization

The moisture content, a_w, sucrose, glucose, and fructose of the second-grade dates used in the present study are around 23.20 ± 0.5% (w. b), 0.725 ± 0.005, 31.10 ± 3.5 g/100 g, 14.60 ± 2.10 g/100 g, and 13.40 ± 1.5 g/100 g, respectively. Sucrose, glucose and fructose are the main sugars that are responsible for the sweet taste of dates, as reported by Ibrahim et al. [20]. In addition, the total sugar content in dry basis were about 77.08 ± 3.8%, which is in accordance with the results obtained by Vilella [21], who reported that dates’ sugar contents ranged between 70.60 and 78.20% in dry basis.

3.2. Optimization of Date Paste Production Process

3.2.1. Characterization of Date Paste Reference Sample

Descriptive sensory evaluation of firmness and color are among the major parameters upon which the best quality of a final product is defined in practice by industry actors. The reference date paste exhibits moderate firmness (5.58 ± 1.26) and a light brown color intensity (3.84 ± 1.6). The moisture content, water activity, instrumental texture firmness, and color parameters (L₀*, a₀*, and b₀*) are shown in

Table 2. Instrumental characterization of reference date paste sample.

Parameters	Value
Moisture content (% w·b)	18.38 ± 0.95
aw	0.66 ± 0.01
Firmness (N)	2.52 ± 0.31
Color parameters
L0*	37.76 ± 2.14
a0*	10.23 ± 0.46
b0*	21.39 ± 0.91

.

The moisture content and water activity values found in this study are in agreement with those obtained by Ahmed et al. [22] and Hassna et al. [23], who reported moisture contents of 16.5, 21.5, and 20–23%, as well as water activity around 0.6. This finding confirms that the date paste is an Intermediate-Moisture food (IMF), as defined by Pinki et al. [24].

The instrumental measurement of firmness is correlated to its sensory evaluation [25] and therefore plays an important role in defining the best quality of a date paste. The firmness of the studied reference sample is lower than those reported in the literature. Indeed, Seyed et al. [26] studied the firmness of date paste produced from Iranian dates and found values ranging between 5.07 and 7.07 N. The difference could be attributed to the test conditions used for the characterization of textural properties and the difference in cultivars, as well as their origins. The color parameter results show that our reference sample has a lighter, redder, and more yellow color than the date paste studied by Sánchez-Zapata et al. [27], who found values of L*, a*, and b* around 32.44, 5.78, and 7.33, respectively.

3.2.2. Optimization of Date Paste Process Using RSD

The design matrix, including the experimental responses, is summarized in

Table 3. Design matrix and experimental responses for the date paste production process optimization using RSD.

N	Factors			Responses
	Hydration Time (h)	Drying Time (h)	Drying Temperature (°C)	Firmness (N)	Moisture Content (% w·b)	Water Activity	L*	a*	b*
1	0.25	5	60	2.89	20.00	0.65	31.98	6.57	19.39
2	0.25	7	50	2.46	20.60	0.68	57.71	4.77	42.86
3	0	5	50	3.13	20.80	0.64	38.27	7.17	21.86
4	0.5	7	60	2.52	18.20	0.61	21.71	6.88	23.20
5	0.25	5	60	2.51	19.84	0.65	29.65	6.12	23.55
6	0	5	70	2.04	14.40	0.64	35.09	6.72	24.30
7	0.25	5	60	3.33	20.10	0.65	28.02	6.32	23.84
8	0.5	5	50	2.09	15.20	0.65	33.85	7.93	24.20
9	0.25	5	60	3.07	19.60	0.64	30.17	6.05	24.85
10	0	7	60	3.53	17.80	0.61	42.30	6.24	24.52
11	0.5	3	60	2.49	17.00	0.66	42.40	7.05	22.69
12	0.25	7	70	4.23	13.20	0.60	37.52	6.77	22.82
13	0.5	5	70	3.42	22.40	0.67	38.55	6.45	21.99
14	0.25	5	60	3.30	18.91	0.64	27.76	5.98	23.44
15	0.25	3	50	2.02	17.60	0.67	35.90	7.64	22.14
16	0.25	3	70	2.05	20.20	0.66	41.54	7.55	25.05
17	0	3	60	2.10	17.40	0.64	41.86	6.49	23.49

.

The best models found for each response are shown in the following equations:

$${\mathrm{Y}}_{\mathrm{F}}=9.18-9.49{\mathrm{X}}_{\mathrm{A}}-0.87{\mathrm{X}}_{\mathrm{B}}-0.14{\mathrm{X}}_{\mathrm{C}}-3.33{{\mathrm{X}}_{\mathrm{A}}}^2-0.70{\mathrm{X}}_{\mathrm{A}}{\mathrm{X}}_{\mathrm{B}}+0.24{\mathrm{X}}_{\mathrm{A}}{\mathrm{X}}_{\mathrm{C}}+0.02{\mathrm{X}}_{\mathrm{B}}{\mathrm{X}}_{\mathrm{C}}$$

$${\mathrm{Y}}_{\mathrm{M}\mathrm{C}}=-23.70-73.20{\mathrm{X}}_{\mathrm{A}}+10.34{\mathrm{X}}_{\mathrm{B}}+0.95{\mathrm{X}}_{\mathrm{C}}-0.954{{\mathrm{X}}_{\mathrm{A}}}^2-0.95{{\mathrm{X}}_{\mathrm{B}}}^2-0.95{{\mathrm{X}}_{\mathrm{C}}}^2+1.36{\mathrm{X}}_{\mathrm{A}}{\mathrm{X}}_{\mathrm{C}}-1.36{\mathrm{X}}_{\mathrm{B}}{\mathrm{X}}_{\mathrm{C}}$$

$${\mathrm{Y}}_{\mathrm{W}\mathrm{A}}=0.87+0.10{\mathrm{X}}_{\mathrm{A}}+0.04{\mathrm{X}}_{\mathrm{B}}-0.04{\mathrm{X}}_{\mathrm{C}}-0.04{{\mathrm{X}}_{\mathrm{A}}}^2$$

$${\mathrm{Y}}_{{\mathrm{L}}^{\ast }}=226-5{{\mathrm{X}}_{\mathrm{A}}+3.80\mathrm{X}}_{\mathrm{B}}+6.62{\mathrm{X}}_{\mathrm{C}}+1.80{{\mathrm{X}}_{\mathrm{B}}}^2+0.07{{\mathrm{X}}_{\mathrm{C}}}^2-0.07{{\mathrm{X}}_{\mathrm{A}}{\mathrm{X}}_{\mathrm{B}}+0.79\mathrm{X}}_{\mathrm{A}}{\mathrm{X}}_{\mathrm{C}}-0.7{\mathrm{X}}_{\mathrm{B}}{\mathrm{X}}_{\mathrm{C}}$$

$${\mathrm{Y}}_{{\mathrm{a}}^{\ast }}=31.40-2.54{{\mathrm{X}}_{\mathrm{A}}-31.40}_{\mathrm{B}}-31.4{\mathrm{X}}_{\mathrm{C}}+6.75{{\mathrm{X}}_{\mathrm{A}}}^2+0.03{\mathrm{X}}_{\mathrm{B}}{\mathrm{X}}_{\mathrm{C}}$$

$${\mathrm{Y}}_{{\mathrm{b}}^{\ast }}=45.50+17.60{{\mathrm{X}}_{\mathrm{A}}+11.47\mathrm{X}}_{\mathrm{B}}-11.4{\mathrm{X}}_{\mathrm{C}}-11.47{{\mathrm{X}}_{\mathrm{A}}}^2+0.70{{\mathrm{X}}_{\mathrm{B}}}^2+0.02{{\mathrm{X}}_{\mathrm{C}}}^2-0.02{\mathrm{X}}_{\mathrm{B}}{\mathrm{X}}_{\mathrm{C}}$$

where Y_F is the firmness (N), Y_MC is the moisture content (%), Y_WA is the water activity, Y_L* is the L* color parameter, Y_a* is the a* color parameter, and Y_b* is the b* color parameter.

X_A is the hydration time (h), X_B is the drying time (h), and X_C is the drying temperature (°C).

Statistical Analysis of Fitted Data and Model Validation

The statistical parameters (

Table 4. Statistical parameters of firmness, moisture content, water activity, and color parameters models.

Statistical Parameter	Models
	Firmness (N)	Moisture Content (% w·b)	Water Activity	L*	a*	b*
p (model)	0.011 **	0.003 **	0.014 **	0.076	0.114	0.031 **
R2	81.09%	90.01%	74.42%	74.39%	58.28%	75.28%
Adjusted R2	66.38%	80.02%	59.08%	48.77%	33.25%	56.05%

** Significant at the 95% confidence level.

) show that the firmness, moisture content, water activity, and b* color parameter models are statistically valid. Yang et al. [28] reported that the more the b* values decrease, the more the browning of color increases.

The validity of all responses models was confirmed by plotting the residuals. Figure 1 illustrates an example of these residual plots for firmness.

Figure 1a,b show that the residual plots are normally distributed. They have homogenous variances and are randomly distributed (Figure 1c). They are independent from each other (Figure 1d) as well.

Furthermore, validation of the response models was performed based on the experimental values obtained under the process conditions used for the reference product.

The t-test statistical analysis showed that there is no significant difference (p > 0.05) between the experimental and predicted values for firmness, moisture content, and water activity. Moreover, all the results found for the date paste samples are on the 95% confidence interval (

Table 5. Experimental and predicted firmness, moisture content, water activity, and b*color parameter for date paste sample hydrated for 0.5 h and dried for 4 h at 60 °C.

Firmness (N)			Moisture Content (% w·b)
Experimental	Predicted	95% CI **	Experimental	Predicted	95% CI **
2.18	2.55	[2.05; 3.05]	17.0	18.94	[17.44; 20.45]
2.39			18.5
2.62			19.0
2.90			19.0
Water activity			b* Color Parameter
Experimental	Predicted	95% CI **	Experimental	Predicted	95% CI **
0.65	0.65	[0.63; 0.67]	21.08	19.86	[15.45; 24.28]
0.65			20.66
0.66			20.46
0.66			20.74
			22.68
			20.77
			22.25
			22.45

** 95% confidence interval for predicted response.

) and therefore confirm the validity of all response models.

Figure 2a–d show that the drying time (B) has the only significant effect (p < 0.05) on the date paste’s firmness and water activity, and it interacts significantly with the drying temperature (C) (p < 0.05) on all responses. Furthermore, the binary interaction between hydration time (A) and drying temperature (C) has a significant effect on the date paste’s firmness and moisture content (p < 0.05).

Optimal Process Conditions

Optimization was carried out based on the water activity, moisture content, firmness, and b* parameters set to the 95% confidence intervals corresponding to the characteristics of the reference product derived from Table 2. In order to reduce the energy consumption, the hydration time, as well as the temperature and duration of drying, were also taken into consideration.

Table 6. Criteria for responses and factors for process optimization.

Response	Goal	Lower	Target	Upper
b*	Target	20.62	21.39	22.15
Water activity	Target	0.65	0.655	0.66
Moisture content (%)	Target	16.87	18.38	19.88
Firmness (N)	Target	2.03	2.52	3.01
Factor
Hydration time (A)	Minimization
Drying time (B)	Minimization
Drying temperature (C)	Minimization

summarizes the optimization criteria used for optimization.

Figure 3 shows contour plots for the optimal hydration time, which is equal to 0.16 h (9.6 min).

Figure 3a demonstrates that drying at 50 °C should be carried out for less than 6.9 h in order to keep the firmness within the 95% confidence interval. For drying for more than about 4.4 h at temperatures less than around 53 °C, the moisture content does not exceed the upper limit (Figure 3b). Figure 3c shows that below a 4.4 h duration of drying, the water activity is in the desired range for temperatures below 54 °C. Figure 3d reveals that the drying temperature should not exceed 3.9 h.

Figure 4 shows drying temperatures and times (white region) that can be used to register a significant decrease in energy consumption along with keeping all parameters of interest within the desired ranges. The optimal solution, with a composite desirability of 0.72, is 0.16 h or 9.6 min of hydration and 3 h of drying at a temperature of 52.28 °C. The predicted values of the moisture content, water activity, firmness, and b* are about 17.62%, 0.65, 2.53 N, and 21.73.

3.3. Biochemical and Microstructure Analysis of Date Paste Sample Produced under Optimized Conditions

3.3.1. Chemical Composition of Optimized Date Paste

The chemical composition of the date paste was determined in order to assess its potential application as an ingredient in the elaboration of functional foods.

Date paste is considered to be a potential source of valuable nutriments [7] and a high-energy fruit due to its high sugar content. It additionally constitutes an important source of dietary fibers and minerals, especially potassium [29]. Date fruits could actually be considered a great source of phenolic compounds compared with other fruits [30]. Furthermore, the chemical composition of date paste produced under optimized conditions is shown in

Table 7. Chemical composition of date paste produced under optimized conditions.

Component	Value
Moisture content (g/100 g)	17.00 ± 0.5
Ash (g/100 g)	1.74 ± 0.19
Proteins (g/100 g)	2.50 ± 0.40
Fat (g/100 g)	<0.30
Total sugars (g/100 g)	68.50 ± 3.18
Glucose (g/100 g)	20.50 ± 2.5
Maltose (g/100 g)	<0.20
Sucrose (g/100 g)	24.60 ± 2.19
Fructose (g/100 g)	23.40 ± 3.50
Lactose (g/100 g)	<0.20
Total Dietary Fiber (g/100 g)	9.80 ± 2.10
Total phenolic content (mg GAE/100 g)	261.00 ± 6.20
Anionic DPPH scavenging capacity (mmol Trolox Eq/g Extract)	2.94 ± 0.05
Cationic ABTS scavenging capacity (mmol Trolox Eq/g Extract)	310.51 ± 1.85

.

For this study, it is worth mentioning that the t-test analysis confirmed the validity of our optimization procedure, since it showed that there is no significant difference (p > 0.05) between the experimental (17.00 ± 0.5% w·b) and predicted (17.62% w·b) moisture contents of the date paste produced under optimal conditions. During the present optimized process, the decrease in the average moisture content from 23.20% w·b (raw date material) to 17.00% w·b (date paste) is mainly due to the drying operation, which was longer than the hydration one. The moisture content and water activity play important roles in the stability of date paste, as their decrease extends the storage time of the product by preventing microbial growth [31].

The obtained ash content was around 1.74 ± 0.19%, which is in agreement with the values reported by Sánchez-Zapata et al. [27]. According to the literature, date and its processed products are considered a good source of minerals such as potassium, iron, sodium, and iodine [32].

Date paste presented a low amount of proteins of about 2.50 ± 0.4%. However, previous works reported that it contains 23 types of amino acids, some of which are not present in most popular fruits such as orange, apple, and banana [33]. Regarding the fat fraction, the studied date paste produced under the optimized concentration showed an amount below 0.30%, which is confirmed by most authors, who have reported that the fat content of date paste ranged between 0.2 and 0.7% [6].

The date paste was characterized by the predominance of sugars, which reached a level of 68.50 g/100 g of date paste, corresponding to 82.53% in dry basis. The total sugar values found in this study are in agreement with those obtained by Muñoz et al., who found that date paste contained sugars ranging from 54% to 82% [6]. The one-way ANOVA test revealed that there is no significant difference (p > 0.05) between the amounts of sucrose, glucose, and fructose. Compared to the raw material, date paste presents a similar (p > 0.05) amount of total sugar contents. Nevertheless, a significant (p < 0.05) decrease in sucrose concentration was observed during the process, which can be explained by its inversion to fructose and glucose during the treatments of hydration and drying, thus confirming that heat promotes this phenomenon of inversion, as reported by Boubekri et al. [34]. The date paste prepared using the optimized process contains a considerable amount of total dietary fibers 9.8 ± 2.1% (Table 6), which is slightly higher than commercial dates’ TDF content (9.25%) [7]. Rastegar et al. [35] found that the percentage of TDF was in the range of 6.4–11.5%. The difference is attributed to the variety and the degree of ripeness. Compared to other fruit by-products (apple, orange, and grapes), which have TDF contents ranging between 1.0 and 4.4% [27], the date and its by-products (date paste) could be considered a relatively rich source of TDF, which makes them suitable for the preparation of fiber-based foods and dietary supplements [36].

The date paste also showed a high amount of total phenolic compounds of about 261 ± 6.2 mg GAE/100 g (Table 6). Mansouri et al. [37] and Sánchez-Zapata [27] found that the total phenolic content ranged from 149 to 836 and 225 mg of GAE/100 g, respectively. However, Wu et al. [38] reported much higher total phenolic contents in two date varieties: 661 and 572 mg of GAE/100 g fresh weight in Deglet Noor and Medjoul varieties, respectively. Many factors, such as the growing and storage conditions, maturity, geographic origin, fertilizer, and soil type, might be responsible for these differences [39]. Date paste also has a high total polyphenol content when it is compared to other fresh fruits such as grapefruit, orange, plum, and strawberry [40], which could be considered a positive attribute, since they could be used as a source of dietary phenolic compounds, especially in arid and semi-arid regions.

3.3.2. Antioxidant Activity

The antioxidant activity of the date paste was evaluated using the anionic DPPH and the cationic ABTS radical scavenging assays. The studied date paste exhibits high antioxidant activity in the DPPH test, with a value of 2.94 ± 0.05 mmol Trolox Eq/g Extract. The processing of date paste led to an increase in the antioxidant activity. Indeed, the raw material showed a lower activity, with a value of 0.05 ± 0.0025 mmol Trolox Eq/g Extract.

Regarding the ABTS test, it was also observed that the antioxidant activity is considered high, as it reached 310.51 ± 1.85 mmol Trolox Eq/g Extract. This result is similar to the one found by AL Farsi and Lee [6], who mentioned that the date paste exhibited high antioxidant capacity, as determined by the ABTS method, with a value of 11.08 mmol Trolox Eq/g.

This finding may be explained by the heat treatment effect, which constitutes a promising processing technique that might efficiently inactivate polyphenol oxidases and peroxidases, therefore positively contributing to the stabilization of total phenolics [41]. As generally recognized in the context of antioxidants in various fruits and vegetables, date paste may contain bioactive compounds, including phenolic acids, flavonoids, anthocyanins, phytosterols, and carotenoids, as well as some minerals [42]. As mentioned above, these findings remain relative to the studied date variety, as the various cultivars of date fruit around the world may exhibit highly variable antioxidant capacity depending on different factors such as the geo-climatic conditions, cultivar, ripening stage, etc. Consequently, the present findings suggest that date paste presents a good source of natural antioxidants and could potentially be considered a functional food or functional food ingredient. However, it is very important to keep this variation in antioxidant capacity in mind when attempting to establish the suitability of date fruit as a nutraceutical. These properties make them a cost-effective alternative, suitable for utilization as functional food ingredients, effectively transforming waste products into economically viable food substrates [43].

3.3.3. Microstructure Characterization

Microstructure analysis enables us to look into several aspects related to foodstuffs, namely texture and transport phenomena that may occur during storage or processing [44]. Figure 5a–c show SEM images at magnifications of 3, 40, and 80 times, respectively.

The date paste sample presents pores ranging from about 2.7 to 30,633.1 μm² of the cross-section (Figure 5a,c). The knowledge of pores’ dimensions could be helpful in physics-based numerical modeling of moisture transport during storage and operations such as cooking. The pores of date paste produced under optimal conditions are predominantly small, and hence, as classified by Datta [45], formulations for numerical modeling of simultaneous heat and mass transfer could be based on the Darcy flow equation, with no significant internal evaporation for storage or with strong evaporation for operations such as cooking.

Moreover, the micrographs show that our final product presents a more uniform texture than the date pastes studied by Rashid et al. [46], which can be explained by the differences in raw material varieties and date paste making process. The predominance of pores with small sizes, along with a more uniform texture, could help to reduce moisture transfer during storage, allowing for a better preservation of the date paste’s quality.

4. Conclusions

The present study was conducted to optimize the second-grade date paste production process at an industrial scale in order to reduce energy consumption while complying with the manufacturing and end-user specifications, resulting in the highest quality of the final product. The optimization results revealed that the hydration duration at 80 °C, as well as the drying temperature and time, are about 9.6 min, 52.28 °C, and 3 h, respectively. This implies 0.34 h of hydration and 1 h of drying savings compared to the conventional production process. The date paste produced under optimized conditions has a moisture content of 17.00% ± 0.5 (w·b), indicating that the raw material undergoes a moisture loss phenomenon, which promotes the stability of the final product during storage. The date paste presents a high nutritional value, with high amounts of carbohydrates (68.50 ± 3.18 g/100 g), total dietary fibers (9.80 ± 2.10 g/100 g), and total phenols (261.00 ± 6.20 mg GAE/100 g). The antioxidant activity tested by the DPPH and ABTS radical scavenging assays showed that the paste sample is characterized by a considerable antioxidant potential. Finally, microstructural characterization showed that the pores’ cross-sections range from about 2.7 to 30,633.1 μm² with the prevalence of small pores.

As a continuation of this work, process optimization using an Artificial Neural Network (ANN) followed by a shelf-life study could also be conducted for comparison with the RSM approach of the present study and in order to characterize the final product during storage.