**2.1 Plant material**

Red pungent dried paprika fruits or, more precisely, pericarp (*Capsicum annuum* L., ssp. *microcarpum longum conoides*, convar. Horgos) used in this study were obtained from the Markova Ceshma region, Prilep, Republic of Macedonia. The pepper species was authenticated by Prof. Danail Jankulovski, Faculty of Agricultural Sciences and Food, Skopje, Republic of Macedonia. A voucher specimen (#1035) is deposited there. The dried pericarp was ground using Retsch ZM1 mill (Germany) and sieved (0.250 mm particle size). The paprika samples placed in dark glass bottles were stored at 4 C in refrigerator.

### **2.2 Extraction procedure**

The impact of three different solvents (ethanol, methanol and *n*-hexane) on the PCO yield, capsaicin and capsanthin content in it were explored using maceration by solid:liquid ratio 1:20 w/v. A 1 g paprika sample (0.0001 g accurately weighed) was used in preparation of single extract. Furthermore, for extraction parameter study at different temperature and time, the extraction was carried out in thermostatic water bath at a temperature of 30, 40, 50, 60 and 70 C, respectively with the exception of 70°C when ethanol was utilized. The effect of dynamic extraction time on the analyte of interest was followed during 60, 120, 180 and 300 min, respectively. After extraction for selected time and at maintained temperature, the solvent was removed under vacuum (rotary vacuum evaporator, type Devarot, Slovenia, 35 C, atm. pressure). Solvent traces were discharged by drying the sample at 40 C, 105 mPa (vacuum drier, Heraeus Vacutherm VT 6025, Langenselbold, Germany). Each extraction procedure was performed in duplicate under the same operating conditions.

### **2.3 Determination of pungent capsicum oleoresin yield**

Obtained PCOs were cooled in a desiccator and weighed. The steps of drying, cooling and weighing were repeated until the difference between two consecutive weights was smaller than 2 mg. The PCO yield was estimated according to dry matter weight in extracted quantity of red pungent paprika. The extract was transferred into a 100 mL volumetric flask and filled to 100 mL with ethanol (1st dissolution).

### **2.4 Determination of capsaicin content in pungent capsicum oleoresin**

The capsaicin content in the extracts was determined by reading of the absorbance at 282 nm. Actually, 0.5 mL of 1st dissolution was dissolved and filled up to 10 mL with ethanol and the absorbance was measured. The concentration of capsaicin was estimated from the standard curve for capsaicin given by the Eq. (1).

$$\mathbf{y} = \mathbf{9}.64\mathbf{x} + \mathbf{0}.005 \quad \mathbf{R} \mathbf{=} 0.9909 \tag{1}$$

where x = μg capsaicin/mL extract and y = absorbance.

### **2.5 Determination of capsanthin content in pungent capsicum oleoresin**

Pigments concentration in red pungent paprika extract was calculated using the extinction coefficient of the major pigment capsanthin (1%E460nm= 2300) in acetone (Hornero-Méndez et al., 2000).

### **2.6 Apparatus**

112 Mass Transfer in Chemical Engineering Processes

Rafajlovska, 1992; Kense, 1970; Rajaraman et al., 1981). Basically, PCO contains pigments carotenoids predominantly capsanthin (Giovannucci, 2002, Hornero-Méndez et al., 2000; Matsufuji et al., 1998) and not less than eight percent of total capsacinoids. Furthermore, beside the pigments, chemical entities such as flavors, taste agents, vitamins and fatty oil are also present in the PCO components profile (Howard et al., 1994; Vinaz et al., 1992). However, a survey of literature reveals that, generally, the most commonly employed and a preferred method for extraction of compounds present in plant matrices is the conventional solid-liquid extraction using organic solvents. In later studies, these conventional methods were improved, modified or rationalized by varying different operating parameters

The paprika oleoresins are produced by solvent extraction of dried, ground red pepper fruits, using a solvent-system compatible with the lipophilic/hydrophilic characteristics of the extract sought and subsequent solvent-system removal. The solvents most commonly used for paprika oleoresin extraction are trichloroethylene, ethylacetate, acetone, propan-2 ol, methanol, ethanol and *n*-hexane (Cvetkov & Rafajlovska, 1992; Hornero-Méndez et al.,

Although many studies have been published on the development and implementation of the different operating conditions for PCO recovery, little attention seems to have been given to the optimization of the various extraction variables (e.g. the appropriate solvent, temperature, dynamic extraction time, quantity of sample, etc.) nor has a systematic study for the optimization of the method been carried out. Therefore, in a situation, where multiple variables may influence the extraction yield, application of a response surface methodology (RSM) to optimize the extraction condition offers an effective technique for studying and optimizing the process and operating parameters (Acero-Ortega et al., 2005;

As part of our contribution to the studies on extraction methods for pungent red paprika we have carried out organic solvent extraction procedure under different conditions, resulting in optimized conditions for the matrix compounds from *Capsicum annuum* L. Hence, the principal goals were to study the influence of the solvent type, extraction temperature and dynamic time on pungent red paprika extraction efficiency expressed by PCO yield and capsaicin and capsanthin content in it and to establish mathematical models to predict system responses.

Red pungent dried paprika fruits or, more precisely, pericarp (*Capsicum annuum* L., ssp. *microcarpum longum conoides*, convar. Horgos) used in this study were obtained from the Markova Ceshma region, Prilep, Republic of Macedonia. The pepper species was authenticated by Prof. Danail Jankulovski, Faculty of Agricultural Sciences and Food, Skopje, Republic of Macedonia. A voucher specimen (#1035) is deposited there. The dried pericarp was ground using Retsch ZM1 mill (Germany) and sieved (0.250 mm particle size).

The impact of three different solvents (ethanol, methanol and *n*-hexane) on the PCO yield, capsaicin and capsanthin content in it were explored using maceration by solid:liquid ratio 1:20 w/v. A 1 g paprika sample (0.0001 g accurately weighed) was used in preparation of

C in refrigerator.

(Boonkird et al., 2008; Toma et al., 2001; Vinatoru, 2001; Wang & Weller, 2006).

2000; Kense, 1970).

Giovanni, 1983; Li & Fu, 2005; Montgomery, 2001).

The paprika samples placed in dark glass bottles were stored at 4

**2. Materials and methods** 

**2.2 Extraction procedure** 

**2.1 Plant material** 

The spectrophotometric measurements were carried out on a Varian Cary Scan 50 spectrophotometer (Switzerland) in 1cm quartz cells, at 25 C.

### **2.7 Statistical analysis**

The statistical analysis and evaluation of the data were performed using STATISTICA 8 (StaSoft, Inc., Tulsa, USA) software. A two-predictors non linear regression model was used to evaluate the individual and interactive effects of two-independent variables, extraction temperature (x1) and dynamic time (x2). The responses measured were PCO yield, capsaicin and major pigment capsanthin present in the PCO.

The second order model includes linear, quadratic and interactive terms thus, in the responses function (Y)-Eq. 2, xi and xj are predictors; 0 is the intercept; i are linear coefficients; ii are squared coefficients; ij are interaction coefficients and is an error term.

Extraction of Oleoresin from Pungent Red Paprika Under Different Conditions 115

bo (intercept) 3.153322 186.625700 117.141400 b1 0.260120\*\*\* - 0.262400 3.238900 b2 0.025702\*\* 0.063600 0.987800\*\*\* b12 - 0.000452 0.010700 - 0.023000

<sup>2</sup> - 0.000010 - 0.000500\*\* - 0.000800 b12 - 0.000286\*\* 0.004800\*\*\* - 0.012200\*\*\* *R2* 0.9795 0.9810 0.7890 adjusted R2 0.9722 0.9742 0.7138 *p* or probability 0.0000 0.0000 0.0002

Table 1. Regression coefficients, *R2*, adjusted R2 and *p* for three dependent variables for

The influence of extraction conditions on the PCO yield and capsaicin were presented by the coefficients of the second-order polynomials. As shown in Table 2, PCO yield was significantly affected by the positive linear effect (*p* < 0.001) of the temperature and the positive linear effect (*p* < 0.01) of the time. In this case, the temperature and time were relevant variables for the model. However, significant linear interaction between the temperature and time (*p* < 0.01) had a negative sign. Moreover, it was found that the influence in the second-order term for the both variables showed no significant effect (*p* > 0.05). These results suggest that the linear effect of the extraction temperature was the primary determining factor for PCO yield but there is no need for prolonged solid/liquid phase contact. The response surface and contour map were also developed to facilitate the visualization and latter, for predicting the optimum condition for PCO yield and capsaicin

Fig. 1b shows that the PCO yield increased as the temperature increased. As for the capsaicin content in PCO, the positive interaction among the independent variables (p < 0.001) significantly influenced the capsaicin content. It was also found that quadratic effect of extraction time is negative at p < 0.01. However, the linear term of temperature and time showed no significant effect on capsaicin content in ethanolic PCO. Hence, when analyzing the interactive effect of temperature and time on the extraction efficiency of capsaicin (Fig. 2) in the model developed for ethanol as extraction solvent, it was observed that extended time

Fig. 3 shows that owing to the capsanthin temperature liability (Ahmeda et al., 2002; Pérez-Gálvez et al., 2005; Schweiggert et al., 2007), capsanthin extraction in ethanolic medium should be performed at decreased temperature of about 40oC at most during extended time.

of extraction is not appropriate under increased temperature condition.

b2

in ethanol (Fig. 1).

Subscripts: 1 = temperature (°C); 2 = time (min);

pungent capsicum oleoresin obtained by ethanol.

**3.1.2 Influence of extraction temperature and time** 

\*Significant at 0.05 level; \*\*Significant at 0.0l level; \*\*\*Significant at 0.001 level.

Yield (%) Capsaicin (mg/100g) Capsanthin (mg/100g)

$$\mathbf{Y} = \boldsymbol{\beta}\_0 + \sum\_{\mathbf{i}=1}^{\mathbf{k}} \boldsymbol{\beta}\_{\mathbf{i}} \mathbf{X}\_1 + \sum\_{\mathbf{i}=1}^{\mathbf{k}} \boldsymbol{\beta}\_{\mathbf{i}} \mathbf{X}\_2 + \sum\_{\mathbf{i}>1}^{\mathbf{k}} \boldsymbol{\beta}\_{\mathbf{i}} \mathbf{X}\_i \mathbf{X}\_{\mathbf{j}} + \quad \text{ } \mathbf{z} \tag{2}$$

ANOVA was used to evaluate the significances of the coefficients of the models judged by computing the *F*-value at a probability (*p*) of 0.001, 0.01 and 0.05.

The influence of the predictors on the responses was also presented using 3-D mesh plots and contour maps.
