CC BY-NC-ND
0King Saud University Journal of King Saud University - Engineering Sciences
www.ksu.edu.sa www.sciencedirect.com
ORIGINAL ARTICLE
Optimisation of spray drying operating conditions of Morinda citrifolia L. fruit extract using response surface methodology
Duduku Krishnaiah *, Awang Bono, Rosalam Sarbatly, Rajesh Nithyanandam, S.M. Anisuzzaman
Department of Chemical Engineering, School of Engineering and Information Technology, Universiti Malaysia Sabah, 88400 Kota Kinabalu, Malaysia
Received 5 June 2012; accepted 9 October 2012
KEYWORDS
Morinda citrifolia;
Microencapsulation;
Spray-drying;
Antioxidant;
Response surface
methodology
Abstract A conventional solvent extract of Morinda citrifolia L. fruit was spray dried using adjuvant maltodextrin (5 wt.%). Spray drying was carried out according to the D-optimal design, and the independent variables selected were temperature and Mcore/Mwall. The spray drying process was optimised by using response surface methodology (RSM) for four different responses: moisture content (MC), DPPH scavenging activity, total phenolic content (TPC), and total flavonoid (TF). The effects of temperature and of the core to wall material ratio were found to be significant for all responses. The optimal spray drying condition for maltodextrin as binding material was found to be 1:1.5 (Mcore/Mwall, volume ratio of M. citrifolia L. extract to additive solution) at 95 °C. The experimental values of the response variables correspond well to the predicted values. The microparticles obtained in this study represent an interesting food additive for incorporation into functional foods due to the presence of antioxidants.
© 2012 King Saud University. Production and hosting by Elsevier B.V. All rights reserved.
1. Introduction
Morinda citrifolia L. (Rubiaceae), noni a Polynesian medicinal plant has been used in the Pacific islands for more than
2000 years (Dixon et al., 1999; Pawlus et al., 2005). It is a small evergreen shrub or tree growing in tropical and subtropical areas worldwide. Originally native to southeastern Asia, the noni plant was spread to Australia, Hawaii, French Polynesia Islands, and other tropical areas through possible water-dispersal of buoyant seeds, or by being transported by early migrants or voyagers (Degener, 1929; Setchell, 1924). They are also referred as Indian Mulberry, Mengkudu, Hai Ba Ji, Ba Ji Tian, Nono or Nonu, Cheese fruit and Nhau (Deng et al., 2010). The bark, stem, root, leaf and fruit have been used traditionally as folk remedies for many diseases. Especially, as a popular ethno medicine among indigenous Polynesians, noni fruits were traditionally used for the improvement in health
* Corresponding author. Tel.: +60 88 320000x3059; fax: +60 320348.
E-mail address: krishna@ums.edu.my (D. Krishnaiah). Peer review under responsibility of King Saud University.
1018-3639 © 2012 King Saud University. Production and hosting by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/jjksues.2012.10.004_
for those with various health problems, such as cancer, infection, arthritis, diabetes, asthma and pain (Wang et al., 2002). It is primarily used to stimulate the immune system and thus to fight bacterial, viral, parasitic and fungal infections. It is also used to prevent the formation and proliferation of tumours, including malignant ones (Dixon et al., 1999; Earle, 2001). In 2002, the noni juice was also accepted in the European Union as a novel food (Dussossoy et al., 2011).
About 200 phytochemicals have been identified from the noni plant, including anthraquinones, flavonoids, and polysaccharides (Pawlus et al., 2005). The chemical constituents of M. citrifolia have been isolated from the fruits, such as fatty acid derivatives (Dalsgaard et al., 2006; Wang et al., 1999,2000). Recently Kim et al. (2010) isolated two novel fatty acid glucosides 1,6-di-O-octanol-ß-D-glucopyranose and 6-O-(ß-D-glucopyran-osyl)1-O-decanoyl-ß-D-glucopyranose from M. citrifolia L. fruit. Pharmacologically synergetic effects among the components in noni fruits may account for its diversified health benefits (Deng et al., 2010). In noni fruits, compounds such as scopoletin, rutin, ursolic acid, ß-sitosterol, asperuloside and damnacanthal are considered to be predominant (Wang et al., 2002; Levand and Larson, 1979; Pawlus and Kinghorn, 2007; Potterat, 2007). Noni fruit extract has also been claimed to have anti-inflammatory (Mckoy et al., 2002) and antioxidative activities (Ikeda et al., 2009) in several in vitro and in vivo test systems (Mahattanadul et al., 2011; Yang et al., 2010). In our recent study, the antioxidant potentials of various medicinal plant species were reviewed (Krishnaiah et al., 2011a). M. citrifolia L. fruit extract is a good source of natural antioxidants. Hence it is processed into microparticles by spray drying technique.
The microencapsulation technique of spray-drying is an effective way to protect drug or food ingredients against deterioration and volatile losses. The main objective of encapsulation is to protect the core material from adverse environmental conditions, such as undesirable effects of light, moisture and oxygen, thereby contributing to an increase in the shelf life of the product, and promoting a controlled liberation of the encapsulate (Shahidi and Han, 1993). The protective mechanism consists of the formation of a membrane wall that encloses droplets or particles of the encapsulated material. There are currently a variety of means available to prepare such microparticles (Fang and Bhandari, 2010). The choice of a particular method depends on the type of microparticles desired. The properties of the wall structures, and the micro-particle size and shape are important considerations. However, in the food and drug industry, spray-drying is still the most popular method for forming microparticles because it is very easy to industrialise and allows for continuous production (Su et al., 2008). Thus, many researchers have used the spray drying technique for the production of microparticles of fruit extract (Tonon et al., 2010; Saenz et al., 2009; Kha et al., 2010; Goula and Adamopoulos, 2010). In our previous study, optimal operating conditions of spray-dried M. citrifolia L. fruit extract using K-carrageenan and maltodextrin were determined (Krishnaiah et al., 2011b). In addition, optimisation of spray drying conditions of M. citrifolia L. fruit extract using K-carrageenan by response surface methodology was studied (Krishnaiah et al., 2009, 2011c).
For microencapsulation by spray-drying, maltodextrin is a good choice as wall material due to their low cost, bland flavour, high water solubility (up to 75%), and low viscosity when in solution (Turchiuli et al., 2005). Maltodextrins are
mainly used to reduce stickiness and agglomeration problems during storage, thereby improving product stability (Bhandari et al., 1993).
The response surface methodology (RSM) has been demonstrated to be a powerful tool for determining the effects of each factor and the interactions among them, and this allows process optimisation to be conducted effectively (Bas and Boyaci, 2007). The response surface procedures involve experimental strategy, mathematical methods, and statistical inference, which, when combined, enable users to make an efficient empirical exploration of the system in which they are interested (Myers, 1976). The experimental strategy enables the analyst to explore the response surface with equal precision, in any direction. The experimental design initially limits the region under investigation. Subsequent to the initial investigation, the experimental design enables the analyst to explore the response surface in a systematic manner in the direction that offers to be the most promising for improvement (Bono et al., 2004).
RSM can be applied to any system that has the following key elements: (1) a criterion of effectiveness, which is measurable on a continuous scale (extraction time), and (2) quantifiable independent variables (both controllable and uncontrollable) that affect the system's performance (such as the extraction process, solvent, and drying method). Given these conditions, RSM offers techniques for finding the optimum response of the system in an efficient manner (Powers, 1989). The major advantage of the RSM is that the amount of data needed for evaluation, analysis and optimisation significantly reduces the number of experiments required. RSM is a faster and more economical method for gathering research results than the classic one-variable at a time or full-factor experimentation. The Design-Expert 8.0.2 RSM software package has been used for this purpose. RSM generated the experimental design table using the D-optimal method.
The objective of the present study is to elucidate the effect of adjuvant maltodextrin in the spray-drying of M. citrifolia L. fruit extract and to find the optimal processing parameters of spray-drying to create microencapsulated powder M. citrifolia L. extract with the best free radical scavenging activity, the highest total phenolic content, the highest total flavonoid content, and the lower moisture content by applying RSM.
2. Materials and methods
2.1. Chemicals
Methanol and ethyl acetate (HPLC grade) were obtained from J.T. Baker (USA). Folin Ciocalteau reagent and tannic acid (TA) were purchased from Merck, Germany. The chemicals 2,2-diphenyl picryl hydrazyl (DPPH), aluminium trichloride (AlCl3), and sodium carbonate were obtained from Sigma Chemicals (St. Louis, USA). A total of 5 kg of M. citrifolia L. fruit that had not yet reached the ripening stage was obtained from Kota Kinablu, Sabah, Malaysia. All other chemicals were of reagent grade and were used without further purification.
2.2. Preparation of fruit extracts
M. citrifolia L. fruit was washed with tap water, then with distilled water. The fruit was peeled, and the core (pulp and seed) was cut into small species. The skin, pulp, and seed were
sun-dried for two days. Then, the sample was kept at 60 0C in a hot-air oven for one day to remove moisture content until a constant weight is obtained. The dried fruit was then finely powdered using a mixer. A total of 250 ml of ethyl acetate was added to 25 g of the powdered sample (10wt.%), and the extraction was performed in a water shaker bath at 35 0C for three days. The supernatant was then separated from the residue by filtration using Whatman No. 1 filter paper. The extracted solution was stored in a closed container and kept at 4 0C before being analysed.
2.3. Spray-drying of the M. citrifolia L. fruit extract
Maltodextrin (5 wt.%) was mixed with M. citrifolia L. fruit extract at different volume ratios (1:1, 1:1.6, and 1:4) as per the experimental design conditions of RSM, and then stirred to form an aqueous solution. The resulting mixture was then spray-dried using a lab plant spray-dryer SD-05 (pilot scale) with co-current flow (the spray-dried product and the drying air flow were in the same direction). The drying chamber had a diameter of 215 mm and a height of 500 mm. The main components of the system were the feed system of the M. citri-folia L. fruit extract consisting of a peristaltic pump, a fluid atomiser (inlet orifice diameter of 0.5 mm), and an air compressor, as well as a feed system for drying the gas, consisting of a blower and an air filter. Finally, a temperature control system and a product control system (cyclone) were also employed. The feed flow rate was kept at 315 ml/hr. The flow rate of the drying air was fixed at 60 m3/hr, and the atomising air remained at a pressure of 1.1 bar. After spray-drying, the powders were collected through a high efficiency cyclone in a glass container, transferred to a glass vial, and stored in desiccators at ambient temperatures.
Several spray-drying runs were carried out, as per the D-optimal experimental design, to investigate the effects of the inlet drying gas temperature for different volume ratios of M. citrifolia L. fruit extract to excipient. Further analyses were performed to determine the moisture content, the antiox-idant activity, the total phenolic content, and the total flavo-noid content of the microencapsulated powder.
2.4. Moisture content
The moisture content of the spray-dried microparticles was determined by the oven drying method (Rattes and Oliveira, 2007). Samples of the microparticles with predetermined masses were placed in an oven (Memmert), heated to 102 0C, and weighed on an analytical balance (Mettler Toledo PB153-S/FACT, Switzerland) until a constant mass was observed. The product moisture content was determined from the weight loss by averaging three measurements.
2.5. DPPH radical scavenging activity
DPPH is a stable free radical that reacts with compounds that are able to donate a hydrogen atom. Thus, the hydrogen donating abilities of spray-dried M. citrifolia L. fruit extract was determined from the change in the absorbance at 515 nm by the Blois (1958) method with slight modifications as that of Praveen and Awang, 2007. The modification is no pH adjustment due to the usage of organic medium, methanol.
For free radical scavenging measurements, samples in methanol solution were prepared by dissolving 10 mg of spray-dried powder in 30 ml of methanol and centrifuging for 10 min using a Sartorius Sigma 3-18 K centrifuge. Aliquots of supernatant were added to 3 ml of 0.025 g/l DPPH in methanol. The change in absorbance was measured after 40 min at room temperature using a 4802 UV-Vis double-beam spectrophotometer. Methanol was used as the reference. DPPH (0-100 mg/l) was used to obtain a standard calibration curve. All measurements were made in triplicate.
The DPPH radical scavenging activity was calculated as a percentage according to the following equation:
DPPH scavenging activity (%) =
Abs515sample
Abs515DPPH solution
2.6. Total phenolic content
The total phenolic content (TPC) was determined according to the Folin Ciocalteau method (Slinkard and Singleton, 1977), with slight modifications as that of Praveen and Awang, 2007. The modification is an estimation of total phenolic content in terms of tannic acid equivalent. Briefly, samples in methanol solutions were prepared by dissolving 10 mg of spray-dried powder in 30 ml of methanol and centrifuging for 10min using a Sartorius Sigma 3-18 K centrifuge. The supernatant of the sample extract (0.5 ml) was added to 2.5 ml of the 0.2 N FC reagent and allowed to react for 5 min. Then, 2 ml of 75 g/l sodium carbonate was added to the reaction mixture and diluted to 25 ml using distilled water. Finally, the reaction mixture was incubated for two hours at room temperature, and the absorbance was measured at 760 nm using a 4802 UV-Vis double-beam spectrophotometer. Methanol was used as the reference. Tannic acid (0-100 mg/l) was used to produce a standard calibration curve. The total phenolic content was expressed in mg of tannic acid equivalents (TAE/g of spray-dried powder).
2.7. Totalflavonoid content
The total flavonoid content was determined using the Dowd method, as adopted by Arvouet-Grand et al., 1994. A total of 5 ml of 2% aluminium trichloride (AlCl3) in methanol was mixed with the same volume of the extract solution (0.4mg/ml). Absorption readings at 415 nm using a UV-Vis double-beam spectrophotometer were taken after 10 min against a blank sample consisting of a 5-ml extract solution with 5 ml of methanol without AlCl3. The total flavonoid content was determined using a standard curve with catechin (0-100 mg/l) as the standard. The content is expressed as mg of catechin equivalents (CE/g of extract).
2.8. Experimental design
The response surface methodology (RSM) was applied using a commercial statistical package, Design-Expert version 8.0.2 (Statease Inc., Minneapolis, USA) to identify optimum levels of two variables of temperature (0C) and Mcore/Mwall (no unit) regarding four responses: moisture content, DPPH scavenging
Table 1 Factors and levels tested for the experimental design.
Parameter Temperature (°C) MCOre/MWall
Factor A B
Max. parameter 140 1
High level + 1
Average parameter 115 0.625
Medium level 0
Min. parameter 90 0.25
Low level -1
activity, TPC, and TF of spray dried M. citrifolia L. fruit extract. The coded and uncoded independent variables used in the RSM design are listed in Table 1. The experiments were designed according to the D-optimal design, as shown in Table 2. The order of experiments has been fully randomised. The data were analysed by multiple regressions using the least-squares method. A second order polynomial equation was used to express the responses as a function of the independent variables.
A quadratic model was used to express the responses as a function of independent variables, where A and B are coded values of temperature and Mcore/Mwall. The test of statistical significance was performed on the total error criteria, with a confidence level of 95%. The significant terms in the model were found by analysis of variance (ANOVA) for each response. The adequacy of the model was checked by calculating the R2 and adjusted-R2. The numerical and graphical optimisation techniques of the Design-Expert software were used for the simultaneous optimisation of the multiple responses. The desired goals for each variable and response were chosen. All of the independent variables were kept within range, while the responses were either maximised or minimised.
3. Results and discussion
3.1. Model fitting
The responses of MC, DPPH scavenging activity, TPC, and TF obtained from the experiments are listed in Table 2. The experimental data were used to calculate the coefficients of the quadratic equation, and Tables 3 and 4 summarise the ANOVA results for the significance of the coefficients of the models and regression coefficient, respectively. For any of the terms in the model, a large regression co-efficient and a small p-value would indicate a more significant effect on the respective response variables. ANOVA showed that the resulting quadratic model adequately represented the experimental data, with coefficients of multiple determinations (R2) of 1.00, 0.99, 0.91, and 0.97 for the responses of MC, DPPH activity, TPC, and TF for maltodextrin as wall material.
The coefficient of determination, R2, is the proportion of variation in the response attributed to the model rather than to random error. It has been suggested that a good-fitting model should have R2 no less than 80%. When R2 is close to unity, the empirical model is suitable for fitting the actual data. A lower value of R2 indicates that the model is inappropriate for explaining the relation between variables (Little and Hills, 1978; Mendenhall, 1975).
It should be noted that adding a variable to the model will always increase R2, regardless of whether or not the additional variable is statistically significant. Thus, a large value of R2 does not always imply the adequacy of the model. For this reason, it is appropriate to use an adjusted-R2 of over 90% to evaluate the model adequacy. Only for the TPC of maltodex-trin were the adj-R2 values found to be less than 0.90. Higher adj-R2 indicated that non-significant terms have not been included in the model.
Table 2 Experimental design conditions and results.
S. No. Two factors Responsive value
Factor 1 Factor 2 Factor 1 Factor 2 MC% DPPH scavenging TPC mg of TAE/g TF mg of
Temperature Mcore/Mwall A B activity% of SDP CE/ g of SDP
(°C) (no unit)
Uncoded Coded values AV PV AV PV AV PV AV PV
values
1 90.00 1.000 -1 1 11.60 11.39 18.58 18.92 33.00 34.50 35.00 36.23
2 140.00 0.625 1 0 6.20 5.81 11.90 11.87 27.00 28.51 30.00 29.91
3 90.00 0.625 -1 0 12.40 12.76 28.36 26.57 54.00 45.72 45.00 42.45
4 140.00 1.000 1 1 4.40 4.38 5.44 5.04 18.00 16.80 22.50 21.39
5 140.00 0.625 1 0 6.00 5.81 11.00 11.87 25.50 28.51 28.00 29.91
6 115.00 0.250 0 -1 9.00 8.60 13.80 13.05 27.00 24.83 32.50 32.51
7 140.00 0.250 1 -1 7.10 7.27 8.06 7.91 25.00 22.70 30.00 29.45
8 90.00 0.250 -1 -1 14.30 14.16 23.80 23.44 39.00 39.43 40.00 39.70
9 90.00 0.250 -1 -1 14.00 14.16 22.50 23.44 37.00 39.43 38.00 39.70
10 90.00 0.250 -1 -1 14.20 14.16 23.00 23.44 38.00 39.43 40.00 39.70
11 115.00 0.625 0 0 7.10 7.17 16.24 16.60 30.00 30.88 35.00 34.12
12 115.00 1.000 0 1 5.60 5.77 9.56 9.36 21.00 19.41 27.50 26.75
13 90.00 1.000 -1 1 11.50 11.39 18.50 18.92 32.00 34.50 36.00 36.23
14 115.00 0.625 0 0 7.00 7.17 16.00 16.60 28.00 30.88 32.50 34.12
15 140.00 1.000 1 1 4.20 4.38 5.20 5.04 18.00 16.80 21.00 21.39
16 140.00 0.250 1 -1 7.00 7.27 8.05 7.91 22.50 22.70 30.00 29.45
AV - actual values; PV - predicted values.
Table 3 Variance analysis for responses.
Sum of Squares Degrees of freedom Mean of squares F-value p-Value prob > F
Regression 191.38 5 38.28 492.05 <0.0001
Residual 0.78 10 0.08
Total 192.16 15
DPPH scavenging activity
Regression 737.57 5 147.51 218.22 <0.0001
Residual 6.76 10 0.68
Total 744.33 15
Regression 1152.97 5 230.59 19.06 <0.0001
Residual 120.97 10 12.10
Total 1273.94 15
Regression 598.71 5 119.74 57.78 <0.0001
Residual 20.72 10 2.07
Total 619.44 15
Table 4 Estimated coefficients for the experimental design and their VIF.
Co-efficient Value Standard error Confidence interval at 95% confidence VIF
Low High
Intercept 7.17 0.16 6.81 7.52
A-Temperature -3.48 0.08 -3.66 -3.29 1.04
B-Mcore/MWall -1.41 0.09 -1.60 -1.22 1.02
AB -0.03 0.09 -0.24 0.18 1.02
A2 2.11 0.17 1.74 2.48 1.06
B2 0.02 0.16 -0.33 0.37 1.09
R2 1.00
Adj-R2 0.99
DPPH scavenging activity
Intercept 16.60 0.47 15.54 17.65
A-Temperature -7.35 0.24 -7.89 -6.81 1.04
B-Mcore/MWall -1.85 0.25 -2.41 -1.29 1.02
AB 0.41 0.28 -0.21 1.03 1.02
A2 2.62 0.49 1.54 3.71 1.06
B2 -5.39 0.46 -6.42 -4.36 1.09
R2 0.99
Adj-R2 0.99
Intercept 30.88 2.00 26.44 35.33
A-Temperature -8.61 1.02 -10.89 -6.33 1.04
B-Mcore/MWall -2.71 1.06 -5.07 -0.34 1.02
AB -0.24 1.18 -2.86 2.38 1.02
A2 6.23 2.07 1.62 10.85 1.06
B2 -8.76 1.96 -13.12 -4.40 1.09
R2 0.91
Adj-R2 0.86
Intercept 34.12 0.83 32.28 35.96
A-Temperature -6.27 0.42 -7.22 -5.33 1.04
B-Mcore/MWall -2.88 0.44 -3.86 -1.90 1.02
AB -1.15 0.49 -2.23 -0.07 1.02
A2 2.06 0.86 0.15 3.97 1.06
B2 -4.49 0.81 -6.29 -2.68 1.09
R2 0.97
Adj-R2 0.95
Each of the experimental values was compared to the predicted value calculated from the model (Table 2). The results suggest that the models used in this study were able to identify the optimum operating condition of spray drying of M. citrifo-lia L. fruit extract.
3.2. Response surface analysis of moisture content
The response surface analysis (RSA) of the data in Table 2 demonstrates that the relationship between the moisture content and independent variables is quadratic, with a very good regression coefficient (R2 = 1.00). Eq. (1) shows the relationship between moisture content and the independent variables (temperature (A) and Mcore/Mwall (B)).
MC = 7.167 - 3.475A - 1.415B - 0.032AB + 2.114A2 + 0.017B2 (1)
Figs. 1 and 2(A) are the response surface plot and isopleths showing the effects of temperature and Mcore/Mwall on the moisture content. Temperature showed a negative linear effect and a positive quadratic effect on the moisture content (p < 0.0001; p < 0.0001). Mcore/Mwall showed a negative linear effect on the response (p < 0.0001). The moisture content decreased when Mcore/Mwall increased, which revealed that a higher ratio is favourable for obtaining a low moisture content at a lower temperature. At a higher temperature, the moisture content also decreased when Mcore/Mwall increased, which revealed that a higher ratio of 1.00 is favourable for obtaining a low moisture content. As the temperature increases at the lower ratio of Mcore/Mwall (more binding material), the moisture content decreases, which may be due to the temperature effect on the powder obtained. These results are consistent with other findings (Goula et al., 2004; Quek et al., 2007). However, in the experimental region, the lowest moisture
16.00 14.00
B: Mcore/Mwall (no unit) 040
100 A: Temperature (°C)
0.55 . 110
B: Mcore/Mwall (no unit) a4^^^ «юд; Temperature (°C)
0.25 90
'S га
га JE
[jj 30.00
B: Mcore/Mwall (no unit) o.4o д. Temperature (.C) B: Mcore/Mwall (no unit) o.4o
Figure 1 Variations in output experimental design with temperature and Mcore/MwaU (maltodextrin, (A) moisture content, (B) DPPH scavenging activity, (C) TPC, and (D) TF).
0.25 90
0.25 90
Moisture content (%
90 95 100 105 110 115 120 125 130 135 140 90 95 100 105 110 115 120 125 130 135 140
X1 : A: Temperature (°C) X2: B: Mcore/Mwall (no unit)
X1 : A: Temperature (°C) X2: B: Mcore/Mwall (no unit)
TPC (mg of TAE/ g of SDP)
TF (mg of CE/ g of SDP)
"I I \ г
95 100 105 110 115 120 125 130 135 140
X1: A: Temperature (°C) X2: B: Mcore/Mwall (no unit)
"I 1 I I I I r^ I
90 95 100 105 110 115 120 125 130 135 140
X1: A: Temperature (°C) X2: B: Mcore/Mwall (no unit)
Figure 2 Isopleths of output experimental design with temperature and Mcore/MwaU (maltodextrin, (A) moisture content, (B) DPPH scavenging activity, (C) TPC, and (D) TF).
content was observed for 1:1 at 140 °C. It was also clear that the experimental results of moisture content and the predicted values obtained using Eq. (1) are not significantly different (Fig. 3).
3.3. Response surface analysis of DPPH activity
The RSA in Table 2 demonstrates that the relationship between the DPPH scavenging activity and independent vari-
ables is quadratic, with a good regression coefficient (R2 = 0.99). Eq. (2) shows the relationship between DPPH scavenging activity and independent variables (temperature and Mcore/Mwall).
DPPH = 16.595 - 7.352A - 1.849B + 0.413AB + 2.625A2 - 5.391B2
Figs. 1 and 2(B) are the response surface plot and isopleths showing the effect of temperature and Mcore/Mwall on the
Figure 3 Comparison of the experimental results of dpph scavenging activity with those calculated via D-optimal design resulted equation.
0 2 4 6 8 10 12 14 16
Moisture content (%), (Experiment»])
Figure 4 Comparison of the experimental results of moisture content with those calculated via D-optimal design resulted equation.
DPPH scavenging activity. Temperature showed a negative linear effect and a positive quadratic effect on the DPPH scavenging activity (p < 0.0001; p < 0.001). Mcore/Mwall showed a negative linear effect and a negative quadratic effect on the DPPH scavenging activity (p < 0.0001; p < 0.0001). The DPPH scavenging activity first increased and then decreased when Mcore/Mwall increased, which revealed that a ratio of 0.85 is favourable for obtaining high DPPH scavenging activity at a lower temperature. At a higher temperature, the DPPH scavenging activity increased and then decreased when Mcore/ Mwall increased, which revealed that a ratio of 0.55 is favourable for obtaining high DPPH scavenging activity. As the temperature increases at the lower ratio of Mcore/Mwall (more binding material), the DPPH scavenging activity decreases. From the results it was clear that the activity is higher at a medium ratio and lower temperature, since the ingredients bind effectively to the wall material at these points. That is there is no leakage of active ingredients through the wall material and undesired materials are kept out (Gibbs et al., 1999; Mozafari et al., 2008). However, in the experimental region, the highest DPPH scavenging activity was observed as 1:1 at 90 0C. It was also clear that the experimental results of DPPH
0 10 20 30 40 50 60
TPC (mg of TAE/g of SDP), Experimental
Figure 5 Comparison of the experimental results of total phenolic content with those calculated via D-optimal design resulted equation.
and the predicted values obtained using Eq. (2) are not significantly different (Fig. 4).
3.4. Response surface analysis of total phenolic content
The RSA of the data in Table 2 demonstrated a good regression value (R2 = 0.91), and Eq. (3) shows the relationship between TPC and independent variables (temperature and Mcore/ Mwall).
TPC = 30.881 - 8.608A - 2.707B - 0.241AB + 6.235A2
- 8.761B2 (3)
Figs. 1 and 2(C) are the response surface plot and isopleths showing the effect of temperature and Mcore/Mwall on the total phenolic content. Temperature showed a negative linear effect and a positive quadratic effect on the TPC (p < 0.0001; p < 0.05). Mcore/Mwall showed a negative linear effect and a negative quadratic effect on the response (p < 0.05; p < 0.01). The TPC increased gradually and then slightly decreased when Mcore/Mwall increased, which revealed that a higher ratio of 0.85 is favourable for obtaining a high total phenolic content at a lower temperature. At a higher temperature, the total phenolic content first increased and then decreased when Mcore/Mwall increased, which revealed that the ratio of 0.55 is favourable for obtaining a high phenolic content. As the temperature increases at the lower ratio of Mcore/Mwall (more binding material), TPC decreases. From the results, it was clear that as with the DPPH scavenging activity, a higher ratio of 0.85 at a lower temperature resulted in maximum TPC. This is due to phenolic compounds also having diverse biological effects as those of antioxidants (Chung et al., 1998). However, in the experimental region, the highest total phenolic content was chosen as 1:1 at 90 0C. It was also clear that the experimental results of the total phenolic content are slightly lower than the predicted values (Fig. 5).
3.5. Response surface analysis of totalflavonoids
The RSA in Table 2 demonstrated a good regression value (R2 = 0.97) and Eq. (4) shows the relationship between TF and independent variables (temperature and Mcore/Mwall).
О 10 20 30 40 JO
TF (mg of CE/g of SDP), Experimental
Figure 6 Comparison of the experimental results of total flavonoids with those calculated via D-optimal design resulted equation.
Table 5 Experimental verification of predicted independent variables.
Response Maltodextrin
PV AVa
Moisture content (%) 11.67 11.17 ± 0.33
DPPH scavenging activity (%) 24.78 24.22 ± 0.43
TPC (mg of TAE/ g of SDP) 42.81 42.02 ± 0.41
TF (mg of CE/ g of SDP) 40.97 40.50 ± 0.60
a All the experiments were repeated three times.
TF = 34.118 - 6.271A - 2.881B - 1.149AB + 2.061A2
- 4.486B2 (4)
Figs. 1 and 2 (D) are the response surface plot and isopleths showing the effect of temperature and Mcore/Mwall on the total flavonoid. Temperature showed a negative linear effect and a positive quadratic effect on the TF (p < 0.0001; p < 0.05). Mcore/Mwall showed a negative linear effect and a negative quadratic effect on the TF (p < 0.0001; p < 0.001). Both independent variables showed a negative interaction effect on the response (p < 0.05). The TF increased gradually and then decreased when Mcore/Mwall increased, which revealed that a medium ratio is favourable for obtaining a high total flavonoid at a lower temperature. At a higher temperature, when Mcore/ Mwall increased, the total flavonoid increased and then decreased, which revealed that a ratio of 0.70 is favourable for obtaining a high flavonoid content. As the temperature increases at the lower ratio of Mcore/Mwall (more binding material), TF decreases. From the results, it was clear that a medium ratio of 0.70 at a lower temperature results in maximum TF. This shows the positive correlation between DPPH and TF (Krishnaiah et al., 2011b). However, in the experimental region, the highest total flavonoid was chosen as 1:2 at 90 0C. It was also clear that the experimental results of total flavonoid are slightly lower than the predicted values (Fig. 6).
3.6. Optimisation of spray drying operating condition
The spray drying condition would be optimum if the DPPH scavenging activity, TPC, and TF reached maximum values
and moisture content reached minimum values. The values of all the responses at operating conditions were converted to a desirability function. The desirability values of the minimum and maximum were configured as 0 and 1, respectively. The maximum desirability function obtained was taken as the optimum operating condition (Bono et al., 2004). The optimal spray drying condition for maltodextrin as the binding material was found to be 1:1.5 (Mcore/Mwall) at 95 0C.
3.7. Verification of predicted independent variables
The suitability of the model equation for predicting the optimum response values was verified using the optimal condition. The experimental values (Table 5) for maltodextrin are 11.17%, 24.22%, 42.02 mg of TAE/g of SDP and 40.50 mg of CE/g of SDP for the four responses of moisture content, DPPH scavenging activity, TPC, and TF, respectively.
4. Conclusions
With the advent of statistical techniques such as RSM and the availability of software packages for problem-solving, it is taking least ever time to optimise processes, products, and the design not only because of increased computer speed but also because the techniques reduce the amount of experimental data required for optimisation analysis. This has led to a better understanding of the process through the results represented in the form of graphs such as 3D surfaces and contours.
Many experiments have been conducted on spray drying of M. citrifolia L. fruit extract for different operating conditions: temperature (90-140 0C) and Mcore/Mwall (1:1 to 1:4). An experimental design has been constructed as per the RSM D-optimal design conditions in order to study the influence of the above parameters on moisture content, DPPH scavenging activity, TPC, and TF. RSM and the conventional graphical and desirability function methods have been effective at determining the optimum zone within the experimental region. From the response surface quadratic model, it was found that the spray drying conditions were significantly affected by temperature and Mcore/Mwall. At optimum spray drying condition of maltodextrin, moisture content, DPPH scavenging activity, total phenolic content, and total flavonoids were found to be 11.67%, 24.78%, 42.81 mg of TAE/g of SDP and 40.97 mg of CE/g of SDP respectively. This study also reveals that M. citrifolia L. fruit extract is a good source of antioxidants, phenolic compounds, and flavonoids.
Acknowledgements
The authors wish to acknowledge the financial support of MOSTI Malaysia. This work was carried out under the E-sci-ence Grant No: SCF0049-IND-2007.
References
Arvouet-grand, A., Vennat, B., Pourrat, A., Legret, P., 1994. Standardisation of an extract of propolis and identification of the major compounds. J. Pharm. Belg. 49, 462-468. Bas, D., Boyaci, I.H., 2007. Modelling and optimization. I. Usability of response surface methodology. J. Food Eng. 78 (3), 836-845.
Bhandari, B.R., Senoussi, A., Dumoulin, E.D., Lebert, A., 1993. Spray drying of concentrated fruit juices. Drying Technol. 11 (5), 1081— 1092.
Blois, M.S., 1958. Antioxidant determinations by the use of a stable free radical. Nature 181, 1199-1200.
Bono, A., Mun, H.C., Rajin, M., 2004. Statistical analysis of natural ingredient based lipstick formulation. In: Proceedings of the Symposium of Malaysian Chemical Engineers, Perak, SOMChe Universiti Teknologi Petronas, pp. 13-14.
Chung, K.T., Wong, T.Y., Huang, Y.W., LIN, Y., 1998. Tannins and human health: a review. Crit. Rev. Food Sci. Nutr. 38, 421-464.
Dalsgaard, P.W., Potterat, O., Dieterle, F., Pauluat, T., Kuhn, T., Hamburger, M., Nonioside, E.-H., 2006. New trisaccharide fatty acid esters from the fruit of Morinda citrifolia (Noni). Planta Med. 72, 1322-1327.
Degener, O., 1929. Ferns and Flowering Plants of Hawaii National Park. Honolulu Star-Bulletin Ltd., Honolulu, Hawaii, pp. 282-283.
Deng, S., West, B.J., Jensen, C.J., 2010. A quantitative comparison of phytochemical components in global Noni fruits and their commercial products. Food Chem. 122, 267-270.
Dixon, A.R., Mcmillen, H., Etkin, N.L., 1999. Ferment this: the transformation of Noni, a traditional polynesian medicine (Mor-inda citrifolia, Rubiaceae). Ecol. Bot. 53, 51-68.
Dussossoy, E., Brat, P., Bony, E., Boudard, F., Poucheret, P., Mertz, C., Giaimis, J., Michel, A., 2011. Characterization, anti-oxidative and anti-inflammatory effects of Costa Rican Noni juice (Morinda citrifolia L.). J. Ethnopharmacol. 133, 108-115.
Earle, J.E., 2001. Plantas medicinales en el trópico humedo. Editorial Guayacan, San Jose.
Fang, Z., Bhandari, B., 2010. Encapsulation of polyphenols - a review. Trends Food Sci. Technol. 21, 510-523.
Gibbs, B.F., Kermasha, S., Alli, I., Mulligan, C.N., 1999. Encapsulation in the food industry: a review. Int. J. Food Sci. Nutr. 50, 213-224.
Goula, A.M., Adamopoulos, K.G., 2010. A new technique for spray drying orange juice concentrate. Innov. Food Sci. Emerg. Technol. 11 (2), 342-351.
Goula, A.M., Adamopoulos, K.G., Kazakis, N.A., 2004. Influence of spray drying conditions on tomato powder properties. Drying Technol. 22 (5), 1129-1151.
Ikeda, Y., Wada, M., Nishigki, T., Nakamoto, K., 2009. Quantification of coumarin derivatives in Noni (Morinda citrifolia) and their contribution of quenching effect on reactive oxygen species. Food Chem. 113, 1169-1172.
Kha, T.C., Nguyen, M.H., Roach, P.D., 2010. Effects of spray drying conditions on the physicochemical and antioxidant properties of the Gac (Momordica cochinchinensis) fruit aril powder. J. Food Eng. 98 (3), 385-392.
Kim, H.-K., Kwon, M.-K., Kim, J.-N., Kim, C.-K., Lee, Y.-J., Shin, H.J., Lee, J., Lee, H.-S., 2010. Identification of novel fatty acid glucosides from the tropical fruit Morinda citrifolia L.. Phytochem. Lett. 3, 238-241.
Krishnaiah, D., Sarbatly, R.Ram., Mohan Rao, S., Nithyanandam, R.R., 2009. Optimal operating conditions of spray dried noni fruit extract using K-carrageenan as adjuvant. J. Appl. Sci. 9 (17), 30623067.
Krishnaiah, D., Sarbatly, R., Nithyanandam, R.R., 2011a. A review of the antioxidant potential of medicinal plant species. Food Bioprod. Process. 89 (3), 217-233.
Krishnaiah, D., Sarbatly, R., Nithyanandam, R.R., 2011b. Microen-capsulation of Morinda citrifolia L. extract by spray-drying. Chem. Eng. Res. Des. 90 (5), 622-632.
Krishnaiah, D., Sarbatly, R., Nithyanandam, R.R., 2011c. Optimization of spray for drying Morinda citrifolia L. fruit extract. J. Appl. Sci. 11 (13), 2276-2283.
Levand, O., Larson, H.O., 1979. Some chemical constituents of Morinda citrifolia. Planta Med. 36, 186-187.
Little, T.M., Hills, F.J., 1978. Agricultural Experimentation Design and Analysis. John Wiley, New York.
Mahattanadul, S., Ridtitid, W., Nima, S., Phdoongsombut, N., 2011. Effects of Morinda citrifolia aqueous fruit extract and its biomarker scopoletin on reflux esophagitis and gastric ulcer in rats. J. Ethnopharmacol. 134, 243-250.
Mckoy, M.L.G., Thomas, E.A., Simon, O.R., 2002. Preliminary investigation of the anti-inflammatory properties of an aqueous extract from Morinda citrifolia (Noni). Pharmacol. Soc. 45, 76-78.
Mendenhall, W., 1975. Introduction to Probability and Statistics, fourth ed. Duxbury Press, North Settuate, MA.
Mozafari, M.R., Khosravi-Darani, K., Borazan, G.G., Cui, J., Pardakhty, A., Yurdugul, S., 2008. Encapsulation of food ingredients using nanoliposome technology. Int. J. Food Prop. 11, 833844.
Myers, R.H., 1976. Response Surface Methodology. Edwards Brothers, Ann Arbor, MI.
Pawlus, A.D., Kinghorn, D.A., 2007. Review of the ethnobotany, chemistry, biological activity and safety of the botanical dietary supplement Morinda citrifolia (Noni). J. Pharm. Pharmacol. 59, 1587-1609.
Pawlus, A.D., Su, B.N., Keller, W.J., Kinghorn, A.D., 2005. An anthraquinone with potent quinone reductase-inducing activity and other constituents of the fruits of Morinda citrifolia (Noni). J. Nat. Prod. 68, 1720-1722.
Potterat, O., 2007. Morinda citrifolia (Noni) fruit - phytochemistry, pharmacology, safety. Planta Med. 73, 191-199.
Powers, R.F., 1989. Optimization models for logistics decisions. J. Bus. Logist. 10, 106-121.
Praveen, K.R., Awang, B., 2007. Antioxidant activity, total phenolic and flavonoid content of Morinda citrifolia fruit extracts from various extraction processes. J. Eng. Sci. Technol. 2 (1), 70-80.
Quek, S.Y., Chok, N.K., Swedlund, P., 2007. The physicochemical properties of spray-dried watermelon powder. Chem. Eng. Process. 46 (5), 386-392.
Rattes, A.L.R., Oliveira, W.P., 2007. Spray drying conditions and encapsulating composition effects on formation and properties of sodium diclofenac microparticles. Powder Technol. 171, 7-14.
Saenz, C., Tapia, S., Chavez, J., Robert, P., 2009. Microencapsulation by spray drying of bioactive compounds from cactus pear (Opuntia ficus-indica). Food Chem. 114, 616-622.
Setchell, W.A., 1924. American Samoa: Part I. Vegetation of Tutulia Island vol. 2. The Carnegie Institution of Washington, Washington, p. 45.
Shahidi, F., Han, X.Q., 1993. Encapsulation of food ingredients. Crit. Rev. Food Sci. Nutr. 33, 501-547.
Slinkard, K., Singleton, V.L., 1977. Total phenol analyses: automation and comparison with manual methods. Am. J. Enol. Viticult. 28, 49-55.
Su, Y.L., Fu, Z.Y., Zhang, J.Y., Wang, W.M., Wang, H., Wang, Y.C., Zhang, Q.J., 2008. Microencapsulation of Radix salvia miltiorrhiza nanoparticles by spray-drying. Powder Technol. 184, 114-121.
Tonon, R.V., Brabet, C., Hubinger, M.D., 2010. Anthocyanin stability and antioxidant activity of spray dried acai (Euterpe oleracea Mart.) juice produced with different carrier agents. Food Res. Int. 43, 907-914.
Turchiuli, C., Fuchs, M., Bohin, M., Cuvelier, M.E., Ordonnaud, C., Peyrat-Maillard, M.N., Dumoulin, E., 2005. Oil encapsulation by spray drying and fluidized bed agglomeration. Innov. Food Sci. Emerg. Technol. 6, 29-35.
Wang, M., Kikuzaki, H., Csiszar, K., Boyd, C.D., Maunakea, A., Fong, S.F.T., Ghai, G., Rosen, R.T., Nakatani, N., Ho, C., 1999.
Novel trisaccharide fatty acid ester identified from the fruits of Morinda citrifolia (Noni). J. Agric. Food Chem. 47, 4880-4882. Wang, M., Kikuzaki, H., Jin, Y., Nakatani, N., Zhu, N., Csiszar, K., Boyd, C., Rosen, R.T., Ghai, G., Ho, C.-T., 2000. Novel glycosides from noni (Morinda citrifolia). J. Nat. Prod. 63, 1182-1183. Wang, M.Y., West, B.J., Jensen, C.J., Nowicki, D., Chen, S., Palu, A., 2002. Morinda citrifolia (Noni): a literature review and
recent advances in Noni research. Acta Pharmacol. Sin. 23, 1127-1141.
Yang, J., Gadi, R., Paulino, R., Thomson, T., 2010. Total phenolics, ascorbic acid, and antioxidant capacity of Noni (Morinda citrifolia L.) juice and powder as affected by illumination during storage. Food Chem. 122, 627-632.
