Optimization of banana juice extraction using combination of enzymes

Abstract

Banana juice extraction was optimized using central composite rotatable design with four numerical factors was employed to design the experiments. The numerical factors were incubation temperature (30–50 °C), and incubation time (20–60 min), cellulase concentration (0–0.4 g/100 g banana) and pectinase concentration (0–0.4 g/100 g banana). The optimum condition for extraction of banana juice were incubation temperature of 36.5 °C, incubation time of 29.33 min, cellulase concentration of 0.34% and pectinase concentration of 0.35%. The corresponding values of yield, viscosity, lightness, acidity, reducing sugar content and total soluble solids at the optimum condition were 74.15%, 101.14 mPa s, 30.06, 2.72%, 171.28 mg/100 g juice and 12.10 °Brix, respectively. Response surface analysis showed that yield increased with incubation time, and cellulase and pectinase concentration, and with temperature it initially increased and then decreased. The processing parameters had an opposite effect on viscosity. Reducing sugar content was also affected by all processing parameters.

Introduction

Banana is the second largest fruit, after citrus, produced worldwide (Mohapatra et al. 2010). The world banana production was 148 million tonnes in 2016 of which India produced 29.1 million tonnes (FAOSTAT 2017).Various food products can be developed from banana, but it is primarily consumed fresh (Lee et al. 2006).

Banana is not suitable for cold storage and decomposition of ripe banana is rapid. After harvesting of banana, the juice can be extracted from ripe bananas and stored conveniently for longer duration. Conventionally banana juice is extracted by normal hydraulic pressing or centrifugation (Lee et al. 2006). The pectinaceous nature of the banana pulp makes juice extraction from banana very difficult (Sagu et al. 2014). Several methods have been tried to increase the yield of juice or to increase the processing efficiency of banana juice extraction. A mechanical juice extractor was developed by Kasozi and Kasisira (2005) to extract juice from peeled bananas by mixing with spear grass. This method reduced the time for juice extraction, but the yield efficiency was also low at 47% as compared to traditional method where efficiency was 69%. Another method using hot water was developed by Lee et al. (2006) for extraction of juice from banana with a maximum yield of 40%. All these methods require further processing to make it acceptable to the consumers as the banana juice obtained by these methods tend to be turbid, viscous and grey in colour. Enzymatic methods can be used to increase yield of juice from banana with acceptable quality.

Cell walls of fruits and vegetables are made up of various biomolecules like cellulose, pectin, hemicellulose, etc. which are difficult to degrade by traditional extraction methods. For maximizing the yield of juice from fruits or vegetables it is necessary to breakdown these complex biomolecules. For this purpose various treatment methods and enzymes can be used (Monitor 2007).

Enzymes, which can break these molecules into smaller units, can be used to increase the yield of juice and processing efficiency. Enzymatic methods also yield a juice which is clear and more attractive (Sharma et al. 2017). Pectinase has been utilized by many authors to increase the yield of banana juice (Sagu et al. 2014; Zaker et al. 2014; Tapre and Jain 2014; Kyamuhangire et al. 2002; Tadakittisarn et al. 2007; Egwim et al. 2013) or for clarification of banana juice (Barman et al. 2015). Other enzymes were also used by other authors for extraction of banana juice. Among the other enzymes used for extraction of banana juice were amylase (Zaker et al. 2014) and Macerax PM, which is a combination of pectinase, cellulase and hemicellulase, (Ibarra-Junquera et al. 2014). Extraction condition for banana juice extraction was not optimized by using combination of enzymes. Therefore, the objective of the present investigation is to optimize the processing conditions for enzymatic extraction of juice from banana using cellulase and pectinase based on yield and properties of the juice.

Materials and methods

Banana juice extraction

Banana (Musa AAB) samples, locally known as “Amritsagar” of even ripeness were purchased from local market of Tezpur University, Assam, India. A lot of 5 kg bananas were purchased at a time. All the bananas of a lot were peeled and cut into 3 mm thick slices. All the bananas of a single lot were sliced and mixed properly for one batch of 30 experiments, and from there 100 g of slices were used for one experiment. Like this 30 experiments were conducted in total as per experimental design. Hundred grams of the slices from the batch were mixed with 50 ml of distilled water and grinded for 2 min in motorized grinder (Philips HL1632) to make pulp. After grinding, enzymes of particular concentrations were mixed with the pulp and incubated for specified time and temperature according to the experimental design. Cellulase from Trichoderma viridae (Product No. C9422) with an activity of 3 U/mg and pectinase from Aspergillus niger (Product No. 17389) with an activity of 1.7 U/mg were procured from Sigma-Aldrich, USA. Juice was extracted from the pulp using a double fold muslin cloth (grade #90 with mesh size of 44 × 36 threads per square inch). The juice was then pasteurized at 80 °C for 5 min, put in sterilized plastic bottle and stored at 4 °C till further analysis.

Experimental design

A central composite rotatable design (CCRD) with four numerical factors was employed to design the experiments. The numerical factors were temperature of incubation (X₁), incubation time (X₂), cellulase concentration (X₃) and pectinase concentration (X₄). The temperature was varied from 30 to 50 °C, incubation time was varied from 20 to 60 min, and cellulase and pectinase concentration were varied from 0 to 0.4 g/100 g banana. A total of 30 experiments were performed (Table 1). The 30 experiments were selected on the basis that for CCRD there should be 2^k factorial points (in this case it is 16, core design), 2k axial or star points (in this case it is 08, outside the core) and several centre points (here it is 06) where k is the number of factors (here it is 04). Six experiments at the centre points of the design were performed to allow the estimation of pure error. All experiments were carried out in a randomized order to minimize the effect of external factors (Wanasundara and Shahidi 1998). All the experiments were carried out in triplicates.

Table 1.

Values of various responses for different experimental conditions

Sl. no.	Temperature (X1) (°C)	Time (X2) (min)	Cellulase (X3) (g/100 g banana)	Pectinase (X4) (g/100 g banana)	Yield (%)	Viscosity (mPa s)	Lightness (L value)	Acidity (%)	Reducing sugar (mg/100 g)	TSS (°Brix)
1	30	40	0.2	0.2	63.33	201	25.14	4.23	164.33	10.6
2	35	30	0.1	0.1	50.00	188	28.32	3.29	154.67	11.1
3	35	30	0.1	0.3	56.67	139	27.54	3.015	155.78	11.2
4	35	30	0.3	0.1	58.00	143	29.12	2.83	168.88	10.9
5	35	30	0.3	0.3	69.33	115	30.25	2.81	200.45	11.3
6	35	50	0.1	0.1	68.00	141	20.29	3.82	107.33	11.9
7	35	50	0.1	0.3	70.67	108	19.86	4.09	162.33	11.8
8	35	50	0.3	0.1	70.00	111	21.33	3.44	121.99	11.2
9	35	50	0.3	0.3	78.67	105	22.09	3.49	204.44	11.9
10	40	20	0.2	0.2	56.00	147	32.21	2.68	163.67	12
11	40	40	0	0.2	56.67	135	25.08	3.33	118.33	11.5
12	40	40	0.2	0	58.67	132	25.41	2.79	130.67	10.2
13	40	40	0.2	0.2	62.00	114	25.15	2.51	218	10.8
14	40	40	0.2	0.2	65.33	110	27.05	2.25	210.88	10.5
15	40	40	0.2	0.2	62.00	124	26.69	2.49	220.76	10.3
16	40	40	0.2	0.2	63.33	129	25.57	2.35	218.33	10.5
17	40	40	0.2	0.2	64.67	121	25.69	2.47	212	10.7
18	40	40	0.2	0.2	60.00	115	27.59	2.38	219	10.3
19	40	40	0.2	0.4	69.33	94	24.54	2.45	175.33	11.1
20	40	40	0.4	0.2	73.33	104	26.58	3.149	136.66	11.7
21	40	60	0.2	0.2	71.33	89	20.19	3.08	127.25	12.1
22	45	30	0.1	0.1	56.67	171	27.36	3.7	145.67	10.5
23	45	30	0.1	0.3	60.00	142	28.47	2.98	105.33	10.9
24	45	30	0.3	0.1	63.33	149	27.42	3.18	111.67	10.7
25	45	30	0.3	0.3	65.33	136	29.24	3.28	97.67	10.9
26	45	50	0.1	0.1	54.67	132	21.69	3.045	131.33	11
27	45	50	0.1	0.3	56.00	110	22.15	3.15	110	10.7
28	45	50	0.3	0.1	58.00	113	21.64	3.35	91.03	10.1
29	45	50	0.3	0.3	61.33	120	21.28	3.52	133	10.9
30	50	40	0.2	0.2	50.00	198	26.31	4.41	101.58	9.4

Yield

After the extraction of the banana juice it is measured in a weighing balance using a beaker.

$$Yield,\%=\frac{Amountofjuicerecovered}{Amountofpulptaken}\times 100$$

Amount of pulp taken = amount of banana taken + amount of water added

Viscosity

Viscosity of the extracted juice was determined using rotational viscometer (DV-79, LAB-KITS, Hong Kong) at 7.5 rpm.

Lightness

Lightness of the banana juice was measured using Colorimeter (Ultrascan VIS, Hunterlab, USA). Results were obtained in terms of L*(lightness) value ranging from 0 (black) to 100 (white).

Titrable acidity, reducing sugar content and total soluble solids (TSS)

Titrable acidity was measured as % malic acid by titrating 5 ml of juice with 0.1 N NaOH solution (AOAC 2000). Reducing sugar was determined colorimetrically using dinitrosalicylic acid (DNSA) reagent (Miller 1959). The TSS of the extracted juices was estimated using hand refractometer (0–32 °Brix).

Data analysis, optimization and validation of optimization result

Design Expert version 8 was used for analysis of data for responses and optimization. Experimental data were fitted to a second order polynomial model as follows:

$$Y={\beta }_o+\sum _{i=1}^3{\beta }_i{X}_i+\sum _{i=1}^3{\beta }_{\mathit{ii}}{X}_i^2+\sum \sum _{i<j=1}^3{\beta }_{\mathit{ij}}{X}_i{X}_j$$

where Y represents the response, β_o, is the constant, β_i, β_ii and β_ij are the regression coefficients and X_i and X_j are the independent variables in coded values.

Significant terms in the model were found by analysis of variance (ANOVA).

Optimization of the extraction process was done using desirability function. In order to determine the accuracy of the result and validate it, five experiments were performed at the optimization condition and % error between the experimental and predicted value (provided by the software) were determined. A % Error below 10% can be accepted for validation (Qi et al. 2009).

Response surfaces were generated to study the effect of interactions on the responses.

Results and discussion

Model fitting

The values of responses for different experimental combinations are shown in Table 1. Multilinear regression analysis of the data yielded second order polynomial equations for all the responses. Analysis of variance (ANOVA) was performed to determine the significant effects of the process variables on the various responses (Table 2). Model adequacy was checked by lack of fit test and by considering fitted R², predicted R², PRESS and adequacy precision. A non-significant (p >0.05) lack of fit, predicted R² comparable to fitted R², adequacy precision higher than 4, implies that the model fitted is adequate to predicting (Corzo et al. 2008; Erbay and Icier 2009).

Table 2.

Analysis of variance (ANOVA) for the fitted models of the various responses

Source	Yield		Viscosity		Lightness		Acidity		Reducing sugar		TSS
	SS	p value	SS	p value	SS	p value	SS	p value	SS	p value	SS	p value
Model	1305.4309	< 0.0001**	22,568.72	< 0.0001**	295.3533	< 0.0001**	8.881777	< 0.0001**	51,428.16	< 0.0001**	10.93617	< 0.0001**
X1: Temperature	220.0387	< 0.0001**	12.04167	0.5438 ns	0.324338	0.5492 ns	0.002017	0.7861 ns	9427.581	< 0.0001**	2.666667	< 0.0001**
X2: Time	196.4820	< 0.0001**	5370.042	< 0.0001**	276.2852	< 0.0001**	0.546017	0.0004**	956.47	0.0007**	0.201667	0.0487*
X3: Cellulase	298.4265	< 0.0001**	1683.375	< 0.0001**	3.912338	0.0503*	0.100363	0.0702 ns	363.0926	0.0186*	0.026667	0.4475 ns
X4: Pectinase	153.2676	< 0.0001**	2583.375	< 0.0001**	0.161704	0.6714 ns	0.041667	0.2283 ns	2123.461	< 0.0001**	0.666667	0.0014**
X21	66.9732	0.0005**	10,285.36	< 0.0001**	2.079003	0.1417 ns	6.471855	< 0.0001**	12,373.94	< 0.0001**	0.4725	0.0050**
X22	0.9632	0.5998 ns	28.00298	0.3585 ns	0.672324	0.3916 ns	0.43373	0.0010**	8999.429	< 0.0001**	3.986786	< 0.0001**
X23	7.4494	0.1572 ns	11.0744	0.5602 ns	1.701453	0.1809 ns	1.275268	< 0.0001**	14,015.49	< 0.0001**	1.981071	< 0.0001**
X24	2.0166	0.4508 ns	140.1458	0.0512 ns	5.875074	0.0198*	0.101227	0.0691 ns	7223.834	< 0.0001**	0.026786	0.4465 ns
X1X2	294.7231	< 0.0001**	0.5625	0.8950 ns	2.197806	0.1316 ns	0.551306	0.0004**	491.8415	0.0077**	0.4225	0.0072**
X1X3	6.2625	0.1925 ns	264.0625	0.0108*	2.949806	0.0844 ns	0.275625	0.0056**	1905.541	< 0.0001**	0.0025	0.8145 ns
X1X4	23.4014	0.0186 ns	217.5625	0.0185*	0.345156	0.5369 ns	0.008556	0.5777 ns	2596.667	< 0.0001**	0	1.0000 ns
X2X3	12.2325	0.0754 ns	189.0625	0.0264*	0.247506	0.6003 ns	0.021025	0.3864 ns	30.94141	0.4529 ns	0.1225	0.1153 ns
X2X4	3.3581	0.3328 ns	264.0625	0.0108*	0.507656	0.4552 ns	0.142506	0.0347*	2019.379	< 0.0001**	0	1.0000 ns
X3X4	8.0231	0.1428 ns	540.5625	0.0008**	0.558756	0.4338 ns	0.0529	0.1774 ns	1360.688	0.0001**	0.25	0.0305*
Residual	50.2668		468.0833		12.95924		0.396158		781.4003		0.6575
Lack of fit	31.0149	0.6399 ns	217.25	0.8782 ns	8.353642	0.5832 ns	0.346074	0.0918 ns	699.5696	0.0612 ns	0.449167	0.4977 ns
Pure error	19.25188		250.8333		4.6056		0.050083		81.83075		0.208333
Cor. total	1355.698		23,036.8		308.3126		9.277935		52,209.56		11.59367
R2	0.962881		0.979681		0.957967		0.957301		0.985033		0.943288
Adjusted R2	0.928236		0.960717		0.918737		0.917449		0.971065		0.890357
Predicted R2	0.84755		0.930001		0.822424		0.777374		0.920563		0.750967
Adequacy Precision	23.22516		28.56515		20.64924		17.06782		24.41392		19.58893

SS sum of squares, ns non significant

**Significant at α = 0.01; *significant at α = 0.05

It can be observed from Table 2 that for yield all the main factors and, X²₁, X₁X₂ and X₁X₄ were significant; for viscosity the linear terms X₂, X₃ and X₄, the quadratic terms X²₁, and all the interactive terms except X₁X₂ are significant; for lightness, only the term X₂ and X²₄ are significant; for acidity only one linear term i.e. X₂, all the quadratic terms except X²₄, and the interactive terms X₁X₂, X₁X₃ and X₂X₄ are significant; for reducing sugar all the linear, quadratic and interactive terms except X₂X₃ are significant; and for TSS the terms X₁, X₂ and X₄, the quadratic terms except X²₄, and the interaction of X₁X₂ and X₃X₄ are significant. All the models have non-significant lack of fit (p > 0.05), which is good, and also the adequacy precision for all the models were more than 4. The R² value of the models for yield, viscosity, lightness, acidity, reducing sugar content and TSS were 0.96, 0.98, 0.96, 0.96, 0.98 and 0.94, respectively which indicates that the models fitted well to the data. The adjusted R² values and predicted R² values (Table 2) were comparable to the R² values implying that the fitted models provided appropriate approximation of the true process.

The final equations in terms of coded factors are as follows:

$$\text{Yield}\text{of}\text{juice},{\text{Y}}_1=62.89-3.03{\text{X}}_1+2.86{\text{X}}_2+3.53{\text{X}}_3+2.53{\text{X}}_4-1.56{\text{X}}_1^2+0.19{\text{X}}_2^2+0.52{\text{X}}_3^2+0.27{\text{X}}_4^2-4.29{\text{X}}_1{\text{X}}_2-0.63{\text{X}}_1{\text{X}}_3-1.21{\text{X}}_1{\text{X}}_4-0.87{\text{X}}_2{\text{X}}_3-0.46{\text{X}}_2{\text{X}}_4+0.71{\text{X}}_3{\text{X}}_4$$ 1

$$\text{Viscosity}\text{of}\text{juice},{\text{Y}}_2=118.83+0.71{\text{X}}_1-14.96{\text{X}}_2-8.38{\text{X}}_3-10.38{\text{X}}_4+19.36{\text{X}}_1^2-1.01{\text{X}}_2^2-0.64{\text{X}}_3^2-2.26{\text{X}}_4^2-0.19{\text{X}}_1{\text{X}}_2+4.06{\text{X}}_1{\text{X}}_3+3.69{\text{X}}_1{\text{X}}_4+3.44{\text{X}}_2{\text{X}}_3+4.06{\text{X}}_2{\text{X}}_4+5.81{\text{X}}_3{\text{X}}_4$$ 2

$$\text{Lightness}\text{of}\text{juice,}{\text{Y}}_3=26.29+0.12{\text{X}}_1-3.39{\text{X}}_2+0.40{\text{X}}_3+0.08{\text{X}}_4-0.28{\text{X}}_1^2-0.16{\text{X}}_2^2-0.25{\text{X}}_3^2-0.46{\text{X}}_4^2+0.37{\text{X}}_1{\text{X}}_2-0.43{\text{X}}_1{\text{X}}_3+0.15{\text{X}}_1{\text{X}}_4-0.12{\text{X}}_2{\text{X}}_3-0.18{\text{X}}_2{\text{X}}_4+0.19{\text{X}}_3{\text{X}}_4$$ 3

$$\text{Acidity}\text{of}\text{juice,}{\text{Y}}_4=2.41-0.01{\text{X}}_1+0.15{\text{X}}_2-0.06{\text{X}}_3-0.04{\text{X}}_4+0.49{\text{X}}_1^2+0.13{\text{X}}_2^2+0.22{\text{X}}_3^2+0.06{\text{X}}_4^2-0.19{\text{X}}_1{\text{X}}_2+0.13{\text{X}}_1{\text{X}}_3-0.02{\text{X}}_1{\text{X}}_4+0.04{\text{X}}_2{\text{X}}_3+0.09{\text{X}}_2{\text{X}}_4+0.06{\text{X}}_3{\text{X}}_4$$ 4

$$\text{Reducing}\text{sugar}\text{content}\text{of}\text{juice,}{\text{Y}}_5=216.50-19.82{\text{X}}_1-6.31{\text{X}}_2+3.89{\text{X}}_3+9.41{\text{X}}_4-21.24{\text{X}}_1^2-18.11{\text{X}}_2^2-22.60{\text{X}}_3^2-16.23{\text{X}}_4^2+5.54{\text{X}}_1{\text{X}}_2-10.91{\text{X}}_1{\text{X}}_3-12.74{\text{X}}_1{\text{X}}_4+1.39{\text{X}}_2{\text{X}}_3+11.23{\text{X}}_2{\text{X}}_4+9.22{\text{X}}_3{\text{X}}_4$$ 5

$$\text{TSS}\text{of}\text{juice,}{\text{Y}}_6=10.52-0.33{\text{X}}_1+0.92{\text{X}}_2-0.03{\text{X}}_3+0.17{\text{X}}_4-0.13{\text{X}}_1^2+0.38{\text{X}}_2^2+0.27{\text{X}}_3^2+0.03{\text{X}}_4^2-0.16{\text{X}}_1{\text{X}}_2+0.01{\text{X}}_1{\text{X}}_3-0.09{\text{X}}_2{\text{X}}_3+0.13{\text{X}}_3{\text{X}}_4$$ 6

Effect of process variables on yield of juice

Figure 1 shows the effect of interaction of various parameters on yield of juice. It can be observed that yield of juice increases with time, concentration of cellulase and concentration of pectinase when the temperature is low. When the temperature was high, it was observed that yield of juice decreased with time of incubation. As the temperature in increased it can be seen that the yield of juice initially increased with increase in temperature and then started decreasing with a further increase in temperature. This indicates that the activity of the enzymes decreased with temperature after a certain point i.e. around 37 °C. The results were in agreement with the results obtained by Demir et al. (2001) and Sandri et al. (2011). When we consider the interaction of time and temperature it was seen that maximum yield was achieved when the temperature is lowest and time is maximum (Fig. 1a). When we consider the interaction of cellulase or pectinase with temperature it was observed that maximum yield was achieved at a temperature somewhere between maximum and minimum temperature and when the concentration of cellulase or pectinase is maximum (Fig. 1b, c), although the effect of temperature is not significant as can be observed from Table 2 also. This implies that the enzymes do not lose the activity within the temperature range selected in the present study. When considering the interaction of cellulase and time (Fig. 1d), pectinase and time (Fig. 1e) and cellulase and pectinase (Fig. 1f) it was detected that yield of juice increased with increase in all the parameters and maximum yield was obtained when both the parameter had maximum values. Further, it was observed that at higher concentration of the enzymes, yield of juice did not increase significantly with increase in incubation time, thereby making the interaction of the enzymes with time non-significant (Table 2). Cellulase was found to be more effective than pectinase in increasing the yield of juice.

Fig. 1.

Effect of various factors on yield of juice

Effect of process variables on viscosity of juice

It can be observed from Fig. 2 that the viscosity of the juice decreases with time, cellulase concentration and pectinase concentration. With temperature the viscosity initially decreased up to a certain point and then decreased with further increase in temperature. The results are in agreement with Sagu et al. (2014) who observed an initial decrease and then increase in viscosity with temperature for banana juice treated with pectinase. The reduction in viscosity after a certain temperature might be attributed to the reduction in pectinase and cellulase activity after that temperature. Demir et al. (2001) and Sandri et al. (2011) also obtained an optimum temperature of 30–40 °C for pectinase, and Sit et al. (2015) obtained an optimum temperature for cellulase activity near 40 °C. Viscosity was found to decrease with time as more and more of pectin and cellulose were degraded with time thereby reducing the viscosity. The viscosity of the juice also decreased with increase in concentration of cellulase and pectinase (Fig. 2f). From Fig. 2f, it can also be observed that pectinase was more effective in decreasing the viscosity compared to cellulase. This might be due to the fact that pectin acts as a gelling agent, which increases the viscosity of juice, and degradation of pectin is more essential compared to cellulose for reduction of viscosity, although destruction of cellulose is necessary for increasing the yield of juice. Even though, decrease in viscosity is crucial for increasing the yield of juice, but at the same time it a very low viscosity of the extracted juice might affect the mouth feel of the juice which is an important sensory attribute for acceptability of the juice.

Fig. 2.

Effect of various factors on viscosity of juice

Effect of process variables on lightness of juice

It is observed from Fig. 3 that lightness of the juice decreased with time of incubation, and was not much affected by temperature. With increase in concentrations of cellulase and pectinase the lightness slightly improved, although the improvement was not significant. The lightness of the extracted juice decreased with time. The decrease in lightness with time might be attributed to the enzymatic browning of the juice due to presence of polyphenol oxidase, which is naturally present in banana pulp. The more time the more enzymatic browning. The polyphenol oxidase enzyme is not deactivated till 50 °C and higher temperature is required for its deactivation, therefore the temperature range investigated in this work had no effect on lightness. The non-significant change in clarity with temperature and enzyme concentration obtained in the present study are in agreement with the results obtained by Sagu et al. (2014) for pectinase treated banana juice and Rai et al. (2004) for pectinase treated mosambi juice, where they observed a non-linear change in clarity with temperature, time and concentration of enzymes. From the present study it can be inferred that the activity of polyphenol oxidase might have negated the effect of cellulase and pectinase, which would have otherwise improved the lightness of the juice.

Fig. 3.

Effect of various factors on lightness of juice

Effect of process variables on acidity of juices

The change in acidity of the banana juice with the process variables is shown in Fig. 4. It is observed that acidity of the juice increased with time of incubation, and with temperature it initially decreased and then increased with temperature. Acidity was only slightly affected by concentration of pectinase and concentration of cellulase. The increase in acidity with time might be attributed to the release of galacturonic acid, a degradation product of pectin, and reducing sugar which is released by the degradation of cellulose (Sreekantiah 1975; Sharma et al. 2017). The initial decrease and then increase in acidity with temperature might be attributed to association of the degradation products with other components, which was favoured up to a certain temperature and then dissociation of them with further increase in temperature.

Fig. 4.

Effect of various factors on acidity of juice

Effect of process variables on reducing sugar content and brix of the juices

It was observed from Eq. 5 that the reducing sugar contents initially increased and then decrease with increase in every parameter as the quadratic parameters were significant for every parameter (Table 2). The decrease was less for pectinase concentration as compared to other parameters. This trend can be explained by the fact that although the reducing sugar content increased with increasing values of the parameters, after a certain value a further increase in reducing sugar content was linked to higher association of these molecules with other components of the juice. Therefore the optimum reducing sugar content was obtained at points in between the maximum and minimum values of the parameters. TSS increased with time, cellulase and pectinase concentration, but decreased with temperature (Eq. 6). The increase in TSS with time, cellulase and pectinase might be attributed to the higher amount of degradation products which were soluble, be it acidic or non-acidic, associated or non-associated. With temperature the enzymes gets deactivated thereby reducing the amount of soluble substances in the juice.

Optimization and validation of optimization result

Optimization of the parameters for juice extraction from banana was carried out to maximize the yield, lightness, reducing sugar content and TSS of the juice, and to minimize the viscosity and acidity. The processing conditions selected for optimization were temperature of incubation from 30 to 50 °C; incubation time from 20 to 60 min, cellulase concentration from 0 to 0.4% and pectinase concentration from 0 to 0.4%. A multi-objective optimization using desirability function was carried out. Optimization by desirability function combines all the responses into single measurement (Park and Park 1997). One solution with the highest desirability of 0.82 was obtained. The values of extraction parameters at the optimum condition were incubation temperature of 36.5 °C, incubation time of 29.33 min, cellulase concentration of 0.34% and pectinase concentration of 0.35%. The corresponding predicted values of yield, viscosity, lightness, acidity, reducing sugar content and TSS at the optimum condition were 74.15%, 101.14 mPa s, 30.06, 2.72%, 171.28 mg/100 g juice and 12.10 °Brix, respectively.

The optimization result was validated by conducting 5 independent experiments at the optimum processing condition. The average experimental values of yield, viscosity, lightness, acidity, reducing sugar content and TSS at the optimum condition were found to be 74.25%, 103.6 mPa s, 31.44, 2.69%, 173.94 mg/100 g juice and 12.19 °Brix respectively. The percentage error between the predicted and experimental values were calculated. The percentage errors for predicted and experimental values obtained for yield, viscosity, lightness, acidity, reducing sugar content and TSS are 0.13, 2.43, 3.08, 1.12, 3.23 and 0.71% respectively which is acceptable (Qi et al. 2009). As the experimental and predicted values of responses were within the acceptable range it can be said that the validation of the optimization result is successful.

Conclusion

The processing conditions for extraction of juice from banana using cellulase and pectinase were optimized in the present study. Optimization was carried out to maximize the yield, lightness, reducing sugar content and TSS of the juice, and to minimize the viscosity and acidity. The optimum conditions achieved were incubation temperature of 36.5 °C, incubation time of 29.33 min, cellulase concentration of 0.34% and pectinase concentration of 0.35%. The corresponding values of yield, viscosity, lightness, acidity, reducing sugar content and TSS at the optimum condition were 74.15%, 101.14 mPa s, 30.06, 2.72%, 171.28 mg/100 g juice and 12.10 °Brix, respectively. The study revealed that yield of juice increased while viscosity decreased with increase in concentration of cellulase and pectinase, and incubation time. With temperature yield increased and viscosity decreased with rise in temperature up to a certain temperature only, which shows that the optimum temperature for pectinase and cellulase activity is somewhere between the temperature ranges chosen. Although, the concentrations of the enzymes used were high in the present study, but at the same time being able to extract juice from a fruit like banana which is perishable and from which juice extraction is difficult, will open new ways for utilization of banana. Keeping in mind the nutritive properties of banana and current unavailability of banana juice in the market, a higher price can be sought for banana juice from the consumers, which might compensate the cost of production of banana juice. At the same time, further research is required to bring down the cost of production of banana juice like using immobilized enzyme or crude enzyme preparation. Thus, from the present study it can be concluded that combination of cellulase and pectinase could be used to get higher yield of juice with desirable quality from banana, which is otherwise difficult to extract.