clarification and stability enhancement of pear juice using loose

8/13/2019 Clarification and Stability Enhancement of Pear Juice Using Loose

1/6

Volume 3 Issue 6 1000162J Food Process Technol

ISSN:2157-7110 JFPT, an open access journal

Open AccessResearch Article

FoodProcessing & Technology

Vivekanand et al.,J Food Process Technol 2012, 3:6http://dx.doi.org/10.4172/2157-7110.1000162

Keywords:Loose nanofiltration; Permeate; Retentate; Flux; Volumeconcentration ratio (VCR); Molecular weight cut off (MWCO)

Introduction

Membrane filtration in the ruit juice industry has grown rom anovel approach into a reliable and economically attractive standardunit operation. Membrane processing has been used or concentrating,

clariying/ractionating and/or puriying ruit juices and or enhancingprocess efficiency and profitability. Tis technique has provided anexcellent alternative or manuacturing high quality ruit juices undermore hygienic conditions. Microfiltration (MF) & ultrafiltration (UF)are cross flow filtration processes, which retain particles in a widerange rom 10,000 Da to 500,000 Da. MF and UF have been widelyused or clarification o ruit juices such as pear and apple. However,studies suggest that UF juices are more susceptible to post bottling haze(PBH) than traditionally clarified juices [1]. PBH is due to incompleteelimination o all potential haze precursors, products o polyphenoland protein interactions. Tis suggests that MF and UF, due to largepore size, are unable to retain all constituents and coloured compoundspotentially responsible or imparting haze and brown colour ormation.Te use o smaller pore size or NF membranes may overcome this

problem.Te borderline between UF and NF is not very distinct and overlaps

in the range o 1000 to 5000 Da. ypically, Nanofiltration (NF) is amembrane separation process likely to reject molecules in the rangeo 100-5000 Da. Tis molecular weight cut off (MWCO) range is veryimportant in the ruit juice processing industry because most o thepolyphenolic compounds in this range usually permeate through theUF membrane and are ofen associated with deects like browning andhaze ormation. Membrane filtration with MWCO between 1000-5000Da may be applied to retain majority o potential haze precursors,coloured organic compounds and at the same time allowing sugars andacids to pass through, which are major constituents o ruit juice. Figure1 illustrated the characteristics o MF, UF, NF and Loose NF according

to their MWCO.NF applications have been developing over the years and in coming

years may provide a convenient and economical way or decolourising

and stabilising ruit juices (pear and apple juice). Generally, problemsassociated with NF include low flux rate, high rejection o sugars andrequirement o a large membrane surace area. Tese problems havelargely contributed to its limited application on a commercial scale.However, NF applications in the ruit juice industries are still in thedevelopmental phase. A considerable amount o research needs tobe done beore juice manuacturers will accept NF as a suitable unitoperation.

Recent advances in membrane manuacturing processes haveresulted in the availability o thin film (polyamide) composite membranes(a thin film o polyamide is cast on a support layer, which is usuallypolysulone) in the molecular weight cut off (MWCO) ranges o 1000-

*Corresponding author: Said Ajlouni, Department of Agriculture and Food Sy-

stems, Melbourne School of Land and Environment, The University of Melbourne,

Parkville, Victoria 3010, Australia, Tel: (03) 83448620; Fax: (03) 83445037; E-mail:

[email protected]

ReceivedMay 04, 2012; AcceptedJuly 10, 2012; PublishedJuly 15, 2012

Citation:Vivekanand V, Iyer M, Ajlouni S (2012) Clarication and Stability En-

hancement of Pear Juice using Loose Nanoltration. J Food Process Technol

3:162. doi:10.4172/2157-7110.1000162

Copyright: 2012 Vivekanand V, et al. This is an open-access article distributedunder the terms of the Creative Commons Attribution License, which permits

unrestricted use, distribution, and reproduction in any medium, provided the

original author and source are credited.

Abstract

This research explored the possible application of loose Nanoltration (NF) to enhance 19 stabil ity and shelf life

of pear juices. The results obtained from loose Nanoltration of sulte pre-treated (SPT) and sulphite un-treated pear

juices (SUT) showed similar patterns of decline in ux rate and increase in volume concentration ratio (VCR) over

time. The treated juice (loose NF permeate) showed a signicant (P< 0.05) increase in lightness (L*), improvement

from brownish tinge to clear light greenish tinge (a*) and an improvement in overall colour difference (Eab) values.

Additionally, loose NF permeates from SPT and SUT were much clearer than the control as indicated by 87-91%improvement in percent adsorption. Loose NF permeates compared to control showed only a minor increase in

the absorbance values when tested for stability at 80C. Such changes in the permeate characteristics indicated

signicant improvement juice quality. The retentate from loose NF revealed high sugar content (17-18Brix), was

dark brown in colour and contained more organic acids.

Clarification and Stability Enhancement of Pear Juice using LooseNanofiltrationV. Vivekanand1, M. Iyer2, and S. Ajlouni3*

1Technical Manager - Fats & Whey Fonterra Australia Pty Ltd, 327 Ferntree Gully Road, Mt Waverley, Vic 3149, Australia

2Program Manager Product Innovation Dairy Australia, Level 5, IBM Centre, 60 City Road, Southbank, Victoria, 3006, Australia

3Department of Agriculture and Food Systems, Melbourne School of Land and Environment, The University of Melbourne, Parkville, Victoria 3010, Australia

Loose NF 10-3

-10-2

100-5000

NF 10-3

-10-2

100-5000

UF 10-2

-10-1

10,000-500,000

MF 10-1

-101

10,000-500,000

Membrane Type Pore Size (m) Molecular Weight

Cut-off (Da)

Figure 1: A diagram comparing the characteristics of MF, UF, NF and

Loose Nano-ltration according to molecular weight cut off (Da) and mem-

brane pore size (m).
http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162


2/6

Citation:Vivekanand V, Iyer M, Ajlouni S (2012) Clarication and Stability Enhancement of Pear Juice using Loose Nanoltration. J Food Process

Technol 3:162. doi:10.4172/2157-7110.1000162

Page 2 of 6



5000 MWCO. Tese thin film composite (FC) polyamide membranesare also least tolerant to chlorine, which is generally specified in termso ppm-hours [2]. raditionally, UF membranes are manuactured

by casting in a single layer, and these membranes have comparativelyhigher tolerance to chlorine. Tereore, to differentiate the membraneswith these characteristics, the term loose Nanofiltration has been usedin order to reflect a separate class NF membranes. Tese MWCO rangesmay suit special applications to overcome post bottling haze (PBH) inthe ruit juice processing industry.

In the dairy industry, spiral wound is the most popular membraneconfiguration. Spiral wound membranes are very economical becauseo low cost and small floor requirements compared to other designs.However, the spiral wound membrane design has not gained asignificant acceptance by the ruit juice industry because pressing juicescontains high levels o suspended solids that cause immediate ouling ospiral membranes. Furthermore, the presence o these suspended solids

poses difficulties in cleaning o these spiral membranes.

For streams with high levels o suspended solids, tubular designshave been more widely used in the juice industry because they canhandle fluids with high-suspended solids, have lower tendency to ouland can be cleaned easily. UF clarified pear juices contain little or nosuspended matter and could be successully used as a eed or a looseNF process.

Loose NF could be used as a secondary membrane process in spiralwound designs or decolourising and stabilising UF clarified ruit

juices. Additionally, the improved design o the mesh spacers in spiralwound membranes has allowed handling o fluids with high solublesolids, and thus providing higher turbulence, lower tendency to ouland easier to clean membrane. Te objective o the present study was

to investigate the effect o loose NF on the quality characteristics o UFclarified pear juice.

Materials and Methods

Pear juice

UF clarified pear juice samples with and without pre-treatment withsulfite (oxidation o mash) and concentrated to 70Brix were kindlyprovided by a local ruit juice manuacturing company in Victoria.Te concentrated pear juice samples that were sulfite pretreated (SP)

were diluted to 13Brix and 18Brix and reerred to as pear juice withsulfite (PJS) and partially concentrated pear juice with sulfite (PCPJS),respectively. Te concentrated and sulphite untreated (SU) pear juice

was diluted to 13Brix only and reerred to as pear juice without sulfite(PJWS). All these diluted pear juice samples were pumped throughthe loose NF pilot plant (Figure 2) in different batches, and permeatesamples rom each batch were collected.

Loose nanofiltration system

Loose Nanofiltration (NF) was perormed in batches using apilot plant (Filtration system, Getzville, NY, USA) fitted with a 5000MWCO spiral wound membrane (GM,Tin film composite Polyamidemembrane rom Osmonics, area: 0.353 m2), with a chlorine toleranceo 1000 ppm days and was operated at 50C and 1240 5 kPa averagepressure (Figure 2). As a batch operation, permeate was collected in aseparate tank and retentate was recirculated into the eed tank. Loose

NF operation was terminated when the permeate flux reached 15-20 L/m2/hr afer 150 min o operation or PJS and PJWS and afer 105 minor PCPJS. Tese trials were perormed in triplicate or each type o

juice (PJS, PJWS and PCPJS). Water and sucrose solution (12Brix)were passed through the loose NF pilot plant in order to test the systemoperation parameters and both flux and VCR were evaluated.

Cleaning of membranes

Membrane cleaning was carried out using various cleaning

solutions at 45-50C. Membranes were cleaned by flushing the system

with water until it became clear (~5 136 min), and circulating 0.3% o

U 11 (Ultrasil 11, sodium hydroxide, Ecolab, Australia) or 20 minutes

ollowed by water rinse until clear or the pH was 8. Tis process was

ollowed by circulating 0.6% o U 56 (Ultrasil 56, enzyme cleaner,

Ecolab, Australia) or 25 min and then by water rinse until clear or

the pH was 7. Sanitisation was carried out by circulating a solution o

0.3% U MP (Ultrasil MP, acidic product, Ecolab, Australia) or 10 min

ollowed by water rinse until clear or the pH was 7. Finally 0.15% o U

11 was circulated again or 10 min and rinsed with water until clear or

pH was 7-8 and the system was ready to be used again. Water flux o

150-175 L/m2/h indicated that the system was clean and ready to be

used again. In some o the experimental runs U 11 was replaced by Easy

Advantage (Mildly alkaline detergent, Ecolab, Australia) to study any

changes in the recovery o the original flux.

Measurements, sampling and analysis

Flux, inlet and outlet pressure and temperature on the loose NF

system were recorded at 15 min intervals. Volume concentration ratiowas calculated as the ratio o litres o eed to the litres o retentate atthat particular time. Average pressure was recorded as a mean o theinlet and outlet pressure during loose nanofiltration. Measurementswere recorded and samples were collected at 15 minute intervals untilthe nanofiltration was discontinued. Samples were stored at 20C orurther analysis.

Te membrane resistance (RM), total membrane resistance (R)and irreversible ouling resistance (RF) were calculated using theollowing equations [3]:

RM = [P/ ( J)] (Equation 1)

where,

RM = Resistance due to membrane

P = Average pressure

12

5

10

7

9

6

4

3

7

8

Figure 2: Schematic diagram of loose Nanoltration pilot plant. [1: Feed tank;2: Pump: 3: Feed; 4: Inlet pressure gauge; 5: Spiral wound membrane hous-

ing; 6: Outlet pressure gauge; 7: Back pressure valve; 8: Retentate; 9: Perme-

ate; 10: Permeate tank].
http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162


3/6


Technol 3:162. doi:10.4172/2157-7110.1000162

Page 3 of 6



= Viscosity o sof water

J = Water flux

R= [P/ ( J)] (Equation 2)

where,

R = otal resistance (using stabilized permeate flow)


= Viscosity o permeate

J = Permeate flux

RF= [{P/ ( J)} - RM] (Equation 3)

where,

RF = Irreversible ouling

RM = Resistance due to membrane


= Viscosity o sof water

J = Water flux (afer ouling o membranes)

Te samples were analysed or pH, Brix, colour, clarity, titratable

acidity as described by Vivekanand et al. [4,5] Sugars (sucrose, glucose,

ructose, and sorbitol), organic acids (citric acid, malic acid and ascorbic

acid) and phenolic compounds (arbutin, 5-hydroxy methyl urural,

chlorogenic acid, caffeic acid, p-coumaric acid and rutin) in pear juice

samples were determined ollowing the method o Vivekanand et al. [5]

A Brookfield LV viscometer at 60 rpm with a UL adapter (BrookfieldEngineering Laboratories Inc., Stoughton, MA, USA) was used to

measure the viscosity o pear juice at 20C.

Colour of loose NF permeate

Colour characteristics o permeate were measured afer dilutingpermeate to 12Brix, so that a comparison could be made at the sameconcentrations. Colour was measured directly and indirectly usinga spectrophotometer (Shimadzu UV-1601, Japan) and a minoltachromameter,(Minolta, Model CR-300, Japan)

Stability test

Stability tests were conducted on all juice samples (permeates and

concentrates). Te juice concentrate samples were diluted to 12Brixand 50 mL o this juice sample was heated to 80C in a water bath(Laboratory water bath, Australia). Absorbance readings were recordedusing a spectrophotometer (Shimadzu, Kyoto, Japan) at 425 nm afer 0,1, 2, 3, 4 and 5 h o heat treatment.

Statistical analysis

A randomized split plot design was used in the study and all externalactors (juice samples, reezing, transportation and storage) were fixed.Each experiment was perormed in triplicate, and results rom analyseswere recorded in duplicate (except or HPLC analysis, which wasperormed using samples rom two experiments in duplicate). All theresults obtained were statistically analyzed using Statistical Package orSocial Science licensed through the University o Melbourne computer

network. Variations between and within treatments were evaluatedusing analysis o variance. Te means were separated using LSD at 95%confidence level.

Results and Discussion

Water flux

Te water flux o the loose NF membrane at different averagepressures and temperatures was determined (Figure 3). Te fluxincreased steadily with pressure. Higher temperature avoured higherflux rate o water. It has been stipulated that higher velocity o watermolecules at 50C was mostly due to the high shif in hydrogen bondswithin the water molecules thus reducing the viscosity.

Water flux is a measure o a membranes permeability when themembrane is new and there are no ouling materials present on thesurace o the membrane. Tereore, water flux provides a good estimateo the resistance offered by the membrane and thereore an indicationo how well the membrane has been cleaned.

Fluxes using sucrose solution

Most ruit juices have a soluble solids concentration in the rangeo 1222Brix, which are made up mainly o sugars [3] and to a smallextent o organic acids. Permeate flow rates o sugar solutions providean indication o the maximum attainable output o the filtered juiceunder ideal conditions i.e. no filtrable solids present. Te effect o

viscosity on the flux can also be estimated using such sugar solutions.

Sucrose solution (12Brix) was initially passed through themembrane using a batch operation to study its effect on operatingparameters (flux and volume concentration ratio) o the loosenanofiltration process. Te 12Brix sugar concentration was chosenbecause it could represent the common average soluble solidconcentration in most ruit juices. able 1 shows the effect on flux, VCRand sugar rejection during membrane filtration. Loose NF o sucrose

solution showed a sharp linear decrease in flux rate during the first 30min, ollowed by gradual and less severe decline in flux over the last 90min o the loose nanofiltration process. Te VCR increased two oldin the first 30 min and our old afer 120 min. Te concentration osugar increased to 13.5Brix in the retentate during the first 15 min

0

50

100

150

200

250

0 690 830 1035 1175 1250 1310

Pressure [kPa]

Flux[L/m2h]

50 C 20 C

Figure 3:Changes in water ux at different temperatures and pressures. Re-

sults represent mean values of three runs.

Time

(min)

Permeate Flux

(L/m/hr)

VCR Brix

Permeate Retentate

Start 135.7 0.00 1.0 0.00 12 0.00 12 0.00

15 118.7 0.00 1.5 0.03 12 0.00 13.5 0.00

30 101.8 0.00 2.0 0.00 12 0.00 13.5 0.00

45 101.8 0.00 2.4 0.00 12 0.00 13.5 0.00

60 93.3 0.00 2.8 0.00 12 0.00 13.5 0.00

75 93.3 0.00 3.2 0.00 12 0.00 13.5 0.00

90 84.8 0.00 3.5 0.00 12 0.00 13.5 0.00

105 84.8 0.00 3.8 0.00 12 0.00 13.5 0.00

120 76.3 0.00 4.0 0.00 12 0.00 13.5 0.00

Table 1: Loose Nanoltration of sucrose solution (12oBrix) at an average pressure

of 1241 kPa. Results represent mean values of three runs SD.
http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162


4/6


Technol 3:162. doi:10.4172/2157-7110.1000162

Page 4 of 6



and remained constant thereafer. Tese results suggest that duringthe initial 30 min, maximum ouling o the membrane occurred andhigh osmotic pressure offered by the increase in VCR and Brix value

(contributed by the total soluble solids). Te decline in flux may beattributed to the sucrose membrane interactions and concentrationpolarisation at the membrane surace, thereby increasing the overallresistance and finally leading to membranes ouling.

Pear juice processing

Flux: Te starting flux rate o PCPJS (18Brix) was approximately2 times lower than PJS (13Brix) and PJWS (13Brix). However, datarevealed that the permeate fluxes or PJS and PJWS were significantlyhigher (P


5/6


Technol 3:162. doi:10.4172/2157-7110.1000162

Page 5 of 6



there were no significant differences (P>0.05) between the three typeso permeate.

Similarly, the degree o redness to greenness was also reportedin terms o a* value (+a* 315 = red and a* = green). Tere was asignificant improvement (P0.05) in a* values were observed betweenthese permeates.

Te colour difference values (Eab) were in agreement with L* anda* values (able 3) and revealed significant (P < 0.05) improvementin the overall colour o various permeates (PJS, PJWS and PCPJS)compared to their controls.

Clarity was measured also as absorbance at 650 nm (able 3), andall permeates (PJS, PJWS and PCPJS) were ound to be 100% clearwhen compared with their control samples. Tere was a significant

reduction (P


6/6


Technol 3:162. doi:10.4172/2157-7110.1000162

Page 6 of 6



glucose and sorbitol contents. Tere was a reduction o 10%, 8% and13% in the 364 Brix content o the PJS, PJWS and PCPJS, respectively.

Loose NF permeates (PJS and PJWS) showed significant (P

0.05)changes were noted or copper, sodium and potassium (able 7).

Colour of loose NF retentate:Visual examination o the retentaterom all three trials revealed dark brown colour and viscous texture. Teretentate had a very high amount o coloured compounds and thereorehad a high absorbance and dark colour (able 8). Dark colour o theretentate was mainly due to concentration o coloured compounds.

pH and titratable acidity in loose NF retentate: Loose NFretentate samples (able 9) showed a minor decrease in pH value anda significant (P 0.05) differences in the fluxrate and VCR trend over time o both pear juices (PJS and PJWS). Teloose NF permeate showed a significant (P 0.05) betweenthe three types o permeate. Additionally loose NF permeates rompear juice with sulur (PJS) and without sulur (PJWS) were clear asindicated by 85-90% reduction in % adsorption. Loose NF permeatesrom PJS and PJWS showed only 10% and 7.0% reduction in the Brixcontent, respectively. Tere were no significant (P>0.05) changes in thepH values; however, there was minor reduction in the titratable acidity.

Te decline in titratable acidity could be attributed to the decrease inboth, citric and ascorbic acids o the permeate as our results revealed.Permeate stability during storage at 20C showed a minor increase in

the absorbance values when tested or stability at 80C. Consequently,it was concluded that loose NF could improve quality o pear juicepermeate. Te retentate rom loose NF was dark brown in colour andhad more sugar content (18-20Brix) and more organic acids, whichcould be reused during clarification process (UF) to recover more othese sugars and organic acids.

Acknowledgement

Grateful acknowledgement is extended to Mr. Tri Q. Huynh from Osmonics,

Minnesota, USA for supplying spiral membrane, Mr. Brandon Moore from Ecolab,

Australia for supplying all the cleaning and sanitation chemicals for the loose

Nanoltration membrane experiment and Mr Joe Pelle from Victoria University

for assistance during analysis on atomic adsorption unit at Victoria University,

Werribee campus.

References

1. Nagel CW, Schobinger U (1985) Investigation of the Origin of Turbidity in

Ultraltered Apple and Pear Juice Concentrate. Confructa Studien26: 16.

2. Cheryan M (1998) Ultraltration and Microltration Handbook. Technomic

Publishing Co. Lancaster.

3. Padilla-Zakour OI, McLellan MR (1993) Optimization and Modelling of Apple

Juice Cross Flow Micro-ltration with a Ceramic Membrane. J Food Sci 58:

369-374.

4. Vivekanand V, Iyer M, Ajlouni S (2001) Ultraltration of Partially Concentrated

Pear Juice. Fruit P rocessing 11: 256-263.

5. Vivekanand, Ajlouni S, Iyer M (2003) Quality Enhancement of UF Claried Pear

Juice using Adsorbent and Weak Base Resins at Different Temperatures. J

Food Sci 68: 333-338.

6. Wagner J (2001) Membrane Filtration Handbook. Osmonics Inc USA.

7. Heatherbell DA, Short JL, Strubi P (1977) Apple Juice Clarication by

Ultraltration. Confructa Studien22: 157.

8. Kirk DE, Montgomery MW, Kortekaas MG (1983) Clarication of Pear Juice by

Hollow Fiber Ultraltration. J Food Sci 48: 1663-1667.

9. Yu ZR, Chiang BH (1986) Passion Fruit Juice Concentration by Ultraltration

and Evaporation. J Food Sci 51: 1501-1505.

10. Girard B, Fukumoto LR (2000) Membrane Processing of Fruit Juices and

Beverages: A Review. Crit Rev Biotechnol 20: 109-75.

Retentate Abs Clarity L* a* Eab

Control

(Sulfurized)

0.3 0.03 0.0 0.00 -31.9 1.02 -5.3 1.9 42.1 0.44

Control

(Unsulfurized)

0.4 0.02 0.1 0.00 -33.9 0.52 -5.4 0.1 44.3 0.51

PJS 0.89 0.13 0.11 0.01 -36.53 0.52 -3.96 0.83 49.50 0.79

PJWS 1.34 0.06 0.13 0.01 -37.12 0.06 -4.66 0.05 50.46 0.40

PCPJS 0.71 0.03 0.09 0.01 -35.5 0.13 -2.57 0.14 47.88 0.31

Table 8: Colour characteristics of loose NF retentate. Results represent mean val-

ues of three runs standard deviations.

Retentate pH TA

Control (Sulfurized) 3.90 0.02 0.4 0.00

Control

(Unsulfurized)

3.90 0.01 0.3 0.00

PJS 3.84 0.02 0.74 0.00

PJWS 3.76 0.01 0.74 0.00

PCPJS 3.86 0.00 1.07 0.00

Table 9: pH and titratable acidity (TA) of loose NF retentate. Results represent

mean values of three runs standard deviation.

Submit your next manuscript and get advantages of OMICS

Group submissions

Unique features:

User friendly/feasible website-translation of your paper to 50 worlds leading languages

Audio Version of published paper

Digital articles to share and explore

Special features:

200 Open Access Journals

15,000 editorial team

21 days rapid review process

Quality and quick editorial, review and publication processing

Indexing at PubMed (partial), Scopus, DOAJ, EBSCO, Index Copernicus and Google Scholar etc

Sharing Option: Social Networking Enabled

Authors, Reviewers and Editors rewarded with online Scientifc Credits

Better discount for your subsequent articles

Submit your manuscript at: http://www.omicsonline.org/submission
http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162http://dx.doi.org/10.4172/2157-7110.1000162http://books.google.co.in/books/about/Ultrafiltration_And_Microfiltration_Hand.html?id=LpiuJVxJS_AC&redir_esc=yhttp://books.google.co.in/books/about/Ultrafiltration_And_Microfiltration_Hand.html?id=LpiuJVxJS_AC&redir_esc=yhttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1993.tb04277.x/abstracthttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1993.tb04277.x/abstracthttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1993.tb04277.x/abstracthttp://www.findanexpert.unimelb.edu.au/display/publication5680http://www.findanexpert.unimelb.edu.au/display/publication5680http://www.findanexpert.unimelb.edu.au/display/publication5680http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.2003.tb14161.x/abstracthttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.2003.tb14161.x/abstracthttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.2003.tb14161.x/abstracthttp://gendocs.ru/docs/30/29820/conv_1/file1.pdfhttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1983.tb05055.x/abstracthttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1983.tb05055.x/abstracthttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1986.tb13845.x/abstracthttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1986.tb13845.x/abstracthttp://www.ncbi.nlm.nih.gov/pubmed/10890454http://www.ncbi.nlm.nih.gov/pubmed/10890454http://www.ncbi.nlm.nih.gov/pubmed/10890454http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1986.tb13845.x/abstracthttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1983.tb05055.x/abstracthttp://gendocs.ru/docs/30/29820/conv_1/file1.pdfhttp://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.2003.tb14161.x/abstracthttp://www.findanexpert.unimelb.edu.au/display/publication5680http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2621.1993.tb04277.x/abstracthttp://books.google.co.in/books/about/Ultrafiltration_And_Microfiltration_Hand.html?id=LpiuJVxJS_AC&redir_esc=yhttp://dx.doi.org/10.4172/2157-7110.1000162

clarification and stability enhancement of pear juice using loose

Documents