The effect of native and modified konjac on the physical attributes of pasteurized and UHT-treated skim milk

lable at ScienceDirect

International Dairy Journal 21 (2011) 790e797

Contents lists avai

International Dairy Journal

journal homepage: www.elsevier .com/locate/ idairyj

The effect of native and modified konjac on the physical attributesof pasteurized and UHT-treated skim milk

John T. Tobin a,b, Sinead M. Fitzsimons a, Alan L. Kelly b, Mark A. Fenelon a,*

a Food Chemistry and Technology Department, Teagasc Food Research Centre, Moorepark, Fermoy, Co. Cork, Irelandb School of Food and Nutritional Sciences, University College Cork, Ireland

a r t i c l e i n f o

Article history:Received 8 November 2010Received in revised form25 April 2011Accepted 26 April 2011

* Corresponding author. Tel.: þ353 25 42355.E-mail address: [email protected] (M.A. Fen

0958-6946/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.idairyj.2011.04.008

a b s t r a c t

The aim of the current study was to examine the influence of addition of mechanically modified konjacglucomannan, on the physical properties of pasteurized and ultra-high-temperature (UHT) treated skimmilk. The effect of adding modified konjac (0e0.12%, w/w) on the viscosity and stability of skim milk,following pasteurization or UHT treatment, was investigated. It was possible to add modified konjac atlevels up to 0.12% (w/w) without phase separation, as compared to <0.03% (w/w) native konjac. Additionof modified konjac increased the viscosity of pasteurized and UHT skim from w2.0 to 3.6 mPa s, withUHT samples having increased stability to sedimentation during simulated storage. The addition ofmodified konjac had little effect on particle size or colour of the skim milk post heating. Addition ofmodified konjac could provide a mechanism to enhance the physical properties of both pasteurized andUHT-treated skim milk.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Polysaccharides are often added to protein solutions to enhanceor control viscosity. However, such additions to milk systems canresult in separation whereby a casein-enriched phase is formed,above a critical polysaccharide concentration (Tuinier, tenGrotenhuis, & de Kruif, 2000). Normally on mixing of twobiopolymers in solution, i.e., a protein and polysaccharide, one ofthe following scenarios may occur (Grinberg & Tolstoguzov, 1997;de Kruif & Tuinier, 2001; Tolstoguzov, 1991):

(i) Segregation, where the biopolymers are thermodynamicallyincompatible, leading to separation into a protein-enrichedphase and a polysaccharide-enriched phase.

(ii) Co-solubility, where the protein and polysaccharide aremiscible.

(iii) Association/complexation, whereby the polysaccharide mayadsorb onto one or more protein particles (complexcoacervation).

When a polysaccharide is added to skim milk, the resultingmixture can be described as a ternary system consisting of protein,

elon).

All rights reserved.

polysaccharide, milk salts and water (Syrbe, Bauer, & Klostermeyer,1998). These systems are complicated, as changes in ionic strengthand biopolymer concentrations can affect thermodynamiccompatibility, with high molecular weight polysaccharides tendingnot to bemiscible in colloidal protein solutions, resulting in complexcoacervation or phase separation. The chemical structure of poly-saccharides appears to be the factor most strongly affecting thephase separation of proteins and polysaccharides (Grinberg &Tolstoguzov, 1997). A reduction in the molecular weight of a poly-saccharidemayprevent phase separation,with the decreasing chainlength of the polymer moving the phase boundary towards higherpolymer concentrations. Tuinier et al. (2000) speculated that higherviscosity, phase-stable milk systems were achievable usingdegraded guar gum, with decreasing polymer chain length, thanwere achievable using native guar.

Previously Lagoueyte and Paquin (1998) used microfluidizationto reduce the molecular weight of xanthan gum. A microfluidizer isa high pressure homogenizer patented by Cook and Lagace (1985)and described by Paquin and Giasson (1989). This high energyhomogenization apparatus incorporates a high pressure intensifierpump which delivers product to an interaction chamber at speedsof up to 400 m s�1, generating pressures in the range 20e275 MPa.In the interaction chamber, the product is subjected to turbulence,cavitation and shear, which elicit structural changes, such asreducing particle size in emulsions (McCrae, 1994; Pouliot, Paquin,Robin, & Giasson, 1991; Strawbridge, Ray, Hallett, Tosh, & Dalgleish,1995), or reductions in molecular weight of polysaccharides.

J.T. Tobin et al. / International Dairy Journal 21 (2011) 790e797 791

Konjac glucomannan is a storage polysaccharide extracted fromthe tubers of the plant Amorphophallus konjac, which is native toareas of China and eastern Asia. Konjac is composed of D-mannoseand D-glucose sugars in the ratio of 1.6:1, with b-(1e4) linkages(Fang & Wu, 2004). Katsuraya et al. (2003) reported the degree ofbranching to be approximately 8%, through b-(1e6)-glucosyl units,while Ratcliffe, Williams, Viebke, and Meadows (2005) found themolecular weight of commercial konjac to be in the range of9.0 � 1.0 � 105 g mol�1. Konjac can interact synergistically withother polymers, such as kappa carrageenan and xanthan gum(Fitzsimons, Tobin, & Morris, 2008; Penroj, Mitchell, Hill, &Ganjanagunchorn, 2005).

Possible health benefits from the ingestion of konjac couldresult from the high viscosity of solutions of the polymer, i.e.,delayed stomach emptying (satiety), more gradual adsorption ofdietary sugar, and binding of konjac to bile acids in the gut, carryingthem out of the body in the faeces thus decreasing the cholesterolcontent (Li, Xia, Wang, & Xie, 2005). Al-Ghazzewi, Khanna, Tester,and Piggot (2007) found that UHT milk containing hydrolysedkonjac had greater prebiotic properties than UHT milk containinginulin, while Vuksan et al. (1999) found that addition of konjac tothe diet of diabetics had the potential to improve glycaemic control,blood lipid profile and also systolic blood pressure.

The aim of the present work was to study the physical attributesof pasteurized and UHT-treated skim milk systems, containingkonjac glucomannan either in its native state or post mechanicalmodification by high pressure microfluidization.

2. Materials and methods

2.1. Materials

Raw full-fat milk was supplied by the Dairy Production ResearchCentre (Moorepark, Fermoy, Co. Cork, Ireland). Konjac gluco-mannanwas supplied by Deoxy Ltd. (Glounthane, Co Cork, Ireland)under the product name Amalan. Stock solutions of 3% (w/w)konjac were prepared in de-ionised water. Refrigerated raw milkwas collected on themorning of the production run (approximately30 L), heated to 55 �C in a jacketed vessel, followed by separationusing an Armfield disc bowl centrifuge (Armfield, Ringwood, UK).The skimmedmilk was then re-circulated through the centrifuge toensure the lowest possible fat content. Five levels of konjac wereadded to skim milk, i.e., 0, 0.03, 0.06, 0.09 and 0.12% (w/w). Full-fatpasteurized homogenized milk for comparative purposes wassourced locally.

2.2. Microfluidization and heat treatment

Stock solutions of konjac were prepared using a MicrofluidizerProcessor M-110 EH (Microfluidics, Newton, Massachusetts, USA).Initially, 10 g of konjac powder was dispersed in 485 g of de-ionisedwater, followed by high shear processing using a diamond inter-action chamber. Thesemixtures were circulated for 10min at a flowrate of 415 mL min�1 and a pressure of 170 MPa, followed byaddition of 5 g of konjac with a further 10 min of high sheartreatment, to yield a stock solution of 3% (w/w) modified konjac.Mixtures of skimmilk and modified konjac were heat treated usinga MicroThermics Lab heat exchanger (MicroThermics, Raleigh,North Carolina, USA). Ultra-high-temperature (UHT) treatment wascarried out at a flow rate of 1 L min�1 with a preheating temper-ature of 95 �C, followed by final heat treatment at 142 �C, witha holding time of 3 s. Pasteurizationwas carried out at a flow rate of1 L min�1 at 72.8 �C, with a holding time of 15 s. All samples werecooled to approximately 5 �C post heat treatment. All konjac stock

solutions and subsequent heat treatments of skim milk konjacpreparations were carried out in triplicate.

2.3. Molecular weight analysis

The molecular weight distribution of native (1%, w/w) andmechanically modified (3%, w/w) konjac was analyzed using gelpermeation chromatography with multiple-angle laser light scat-tering detection (GPCeMALLS). Samples were prepared in 50 mM

sodium chloride so that they contained 2.5 mg mL�1 konjac thenfiltered and injected onto the column at a flow rate of 0.5 mL min�1

(200 mL). Separation was achieved using two columns in series,a Suprema 3000 and 30000 (Polymer Standards Service GmbH,Mainz, Germany). Two detectors were used in the determination ofthe molecular weight distribution, a Dawn DSP Laser Photometerand an Optilab DSP Interferometric Refractometer (Wyatt Tech-nology, Santa Barbara, California, USA). Results were interpretedusing Astra 4.90.07 software package (Wyatt Technology). Analysisof molecular weight was carried out in duplicate.

2.4. Sedimentation rates and phase separation behaviour

Phase separation of skim milk containing native konjac andsedimentation rates of UHT skim milk containing modified konjac(0e0.12%, w/w) were analyzed using a Lumifuge 116 (Lum GMBH,Berlin, Germany) stability analyzer. The Lumifuge is an analyticalcentrifuge that operates on the basis of continuousmeasurement oflight transmitted through the sample over the length of themeasurement cell. Measurements of phase separation (skim milkcontaining native konjac) were carried out at 46 � g for 1.5 h. UHTsamples were centrifuged at 1140 � g for 3.6 h, simulatingapproximately 6 months storage under conditions of normalgravity. Sedimentation rates were analyzed using the softwarepackage SepView 4.1 (Lum GMBH) which calculated the timecourse of the separation from themeasurement results. For analysisof phase separation behaviour of skim milk containing nativekonjac, a range (position in mm on the x-axis) of 91e113 mm wasused while, for UHT samples, a range of 91e112 mm and a trans-mission % of 55 was chosen as the position on the y-axis that wasmost representative of the movement of the interface over themeasurement time (Tobin et al., 2010).

2.5. Viscosity measurement

Viscosity measurements were carried out on full-fat milk andskim milk/konjac mixtures post processing, using an AR-G2rheometer (TA Instruments, Crawley, West Sussex, UK) equippedwith a parallel-plate geometry (diameter 60 mm). Samples werepre-sheared at 500 s�1 for 1 min, allowed to equilibrate for 1 min,sheared from 0.01 to 800 s�1 over 2 min, held at 800 s�1 for 1 minand sheared from 800 to 0.01 s�1 over 2 min; the same procedurewas followed for comparison of native and modified konjac, buta maximum shear rate of 500 s�1 was employed. All measurementswere carried out at 20 �C. For analysis of the relationship betweenviscosity and temperature for full-fat milk, a constant shear rate of800 s�1 was used, within the range 5e37 �C. Flow characteristics ofsamples are represented by plots of shear stress (Pa) and viscosity(Pa s) as a function of shear rate (s�1), konjac concentration (%, w/w)and temperature (�C).

2.6. Particle size analysis

Particle size was measured by dynamic light scattering usinga Zetasizer Nano system (Malvern Instruments Ltd, Malvern, Wor-cestershire, UK). Measurements were carried out at 22 �C, with

Mechanically modified konjac

0

Volume (mL)

10 20 30-0.04

0.00

0.08

0.12

R.I

. res

pons

e (V

olts

)

0.04

Native konjac

Fig. 1. The effect of microfluidization on the molecular mass distribution of konjac asrepresented by the variation in the refractive index elution profile by multiple-anglelaser light scattering.

Table 1Weight-averagedmolecular weight (Mw), number-averagedmolecular weight (Mn)and polydispersity determined by GPC/MALLS for native and modified konjac.

Samplesa Mwg mol�1

Mng mol�1

PolydispersityMw/Mn

Native konjac A 1.02 � 106 6.78 � 105 1.50Native konjac B 1.06 � 106 7.04 � 105 1.51Native average 1.04 � 106 6.91 � 105 1.51

Modified konjac A 2.64 � 105 2.41 � 105 1.10Modified konjac B 2.85 � 105 2.59 � 105 1.10Modified average 2.75 � 105 2.50 � 105 1.10

a A and B signify replicate analyses of individual samples of konjac.

J.T. Tobin et al. / International Dairy Journal 21 (2011) 790e797792

a fixed scattering angle of 173�. Samples were diluted (1:10) andplaced in disposable polystyrene cuvettes. The cumulative methodwas used to find the mean average (z-average), or the size ofa particle that corresponded to the mean of the intensitydistribution.

2.7. Reversed-phase high performance liquid chromatography

Samples of pasteurized skimmilk containing 0.12% (w/w) nativekonjac, following phase separation, and pasteurized skim with noadded konjac, were analyzed for levels of protein by an adaptationof themethod of Mounsey and O’Kennedy (2009). Samples (200 mL)were resuspended in 3800 mL dissociating buffer (20 mM bis-Trispropane containing 7 M urea and 5 mL mercaptoethanol, final pH7.5) followed by filtration (0.22 mm). Separation of samples wasachieved using a Poroshell 300SB-C18 column (Agilent Technolo-gies, Santa Clara, California, USA) attached to a Agilent 1200Separation Module (Agilent Technologies). The elution buffers usedwere (A) 90:10% (v/v) acetonitrile:water containing 1% (v/v) tri-flouroacetic acid and (B) 10:90% (v/v) acetonitrile:water containing1% (v/v) triflouroacetic acid. The elution gradient employed rangedbetween 26 and 100% of buffer (A) to buffer (B). The injectionvolume was 5 mL, column temperature was 35 �C, flow rate was0.5 mL min�1 and the proteins were resolved at a detectionwavelength of 214 nm. Peak areas were integrated by the AgilentChemstation software package (Agilent Technologies), while indi-vidual protein identification was ascertained from elution times ofknown standards.

2.8. Colorimetric measurements

Colour measurements were carried out using a tristimuluscolour analyzer, Chromameter model CR300 (Konica MinoltaSensing Europe, Central Milton Keynes, UK), and expressed usingCommission Internationale d’Eclairage L*, a*, and b* colour spaceco-ordinates.

2.9. Statistical analysis

All trials were carried out in triplicate and the subsequent datasets were subjected to one way ANOVA using MINITAB� 15 (MinitabLtd, Coventry, UK) statistical analysis package. The effect of differentlevels of konjac additiononviscosity, particle size, sedimentation rateand colour of both pasteurized and UHT-treated skimmedmilk wereinvestigated. Means with significant differences were comparedusing Fisher’s individual error rate with significance at p < 0.05. Thestatistical significance of differences betweenmeans for viscosity andparticle size between pasteurized and UHT-treated samples wasanalyzed by means of a paired T-test (confidence interval of 0.95)(Minitab Ltd).

3. Results and discussion

3.1. Molar mass and viscosity analysis of nativeand modified konjac

The effect of microfluidization on the molar mass distribution ofkonjac was investigated. The results are presented graphically inFig. 1, and also in terms of changes in molecular mass distributionin Table 1. A four-fold reduction in the weight-averaged molecularweight (Mw) of konjac was achieved through high shear treat-ment. Native konjac samples had an average Mw of1.04 � 106 g mol�1, while modified konjac had an average Mw of2.75 � 105 g mol�1 (Table 1). The values obtained for the nativekonjac were similar to the weight-averaged molecular weight

range of samples analyzed by Ratcliffe et al. (2005), who found themolecular weights of four commercial konjac samples to be9.0 � 1.0 � 105 g mol�1. An interesting observation was thelowering of the polydispersity, Mw/Mn, where Mn equals thenumber-averaged molecular weight of the modified konjac ascompared with the native konjac. This indicates that, aftershearing, the konjac molecules are of a more uniform size distri-bution than those of native konjac.

The plot of viscosity against shear rate (Fig. 2) shows thedecrease in the intrinsic viscosity of 1.0% (w/w) native konjaccompared with 1.0% (w/w) modified konjac. There was a ten-folddecrease in the apparent viscosity from 0.47 Pa s for 1.0% (w/w)native konjac to 0.045 Pa s (shear rate of 500 s�1) for 1.0% (w/w)modified konjac (measured at 20 �C). The decrease in both theapparent viscosity and the reduction in the molecular weight, postmicrofluidization correlate with the work of Lagoueyte and Paquin(1998), who found similar effects for solutions of xanthan gumfollowing microfluidization. Penroj et al. (2005) stated that thedeacetylation of unbuffered konjac solutions resulted in decreasedpH. However, in the current work decreases in viscosity, attributedto reductions in the Mw, are unlikely to coincide with removal ofacetyl groups as a decrease in pH was not observed.

3.2. Interaction between skim milk and native konjac

Native konjac was added to skimmilk (0.07%, w/w, fat, pH 6.9) atlevels ranging from 0 to 0.12% (w/w), to investigate the thermo-dynamic compatibility between the protein and polysaccharide.Visual observations, following overnight storage, concluded that

0.01

0.10

1.00

10.00

0 100 200 300 400 500 600

Shear rate (s-1)

Vis

cosi

ty (

Pa s

)

Fig. 2. Profile of viscosity(log) against shear rate for 1.0% (w/w) native konjac (A) and1.0% (w/w) modified konjac (>), in distilled water.

Fig. 3. Representative Lumifuge profiles of phase separation of pasteurized skim milk contai

J.T. Tobin et al. / International Dairy Journal 21 (2011) 790e797 793

phase separation occurred at levels as low as 0.03% (w/w), konjacwhile, at higher concentrations (0.09e0.12%, w/w), the separationwas even more pronounced; as demonstrated by the LumiFugeprofiles (Fig. 3A and B). As previously reported, exceeding a criticalpolysaccharide concentration causes phase separation (Tuinieret al., 2000). In the current work this separation was probablydue to a segregative interaction between the protein and thepolysaccharide, resulting in an effective attraction between theproteins through a depletion mechanism, whereby polymers ofdissimilar shape and structure may segregate, leading to, in thisinstance, a reduction of the konjac concentration near the proteinparticles (Tuinier et al., 2000). The separation profiles for skimmilkcontaining 0.09 and 0.12% (w/w) konjac (Fig. 3A and B), showthe separation of the protein and konjac over time into two distinctphases, i.e., a polysaccharide-enriched, protein-depleted aqueousphase (position 90e107 mm) and a protein-enriched, poly-saccharide-depleted aqueous phase (position 107e114 mm).Reverse-phase high performance liquid chromatography (RP-HPLC)results (Fig. 4) show the phase partitioning of the milk proteins inthe presence of native konjac (0.12%, w/w). It was found that thecaseins were concentrated in the sediment layer at w3.2 times theconcentration present in a control skim milk (no konjac), and thatthe protein-enriched sediment phase contained w24.3 times asmuch casein as the protein-depleted aqueous phase.

ning (A) 0.09% (w/w) and (B) 0.12% (w/w) native konjac during centrifugation at 46 � g.

0

100

200

300

400

500

600

2 7 12 17 22

Time (min)

Abs

orpt

ion

(mA

U)

Lower sediment phase

Skim milk

Upper aqueous phase

κ -CN

α S1-CN

β -CN

α S2-CN β -lac

Fig. 4. Reverse-phase high performance liquid chromatography profiles of proteins inpasteurized skim milk and the upper aqueous and lower sediment layers of phase-separated pasteurized skim milk containing 0.12% (w/w) native konjac.

0.0

0.5

1.0

1.5

2.0

2.5

0 200 400 600 800 1000

Shear rate (s -1 )

) a P ( s s e r t s

r a e h S

Fig. 5. Profile of shear stress against shear rate for pasteurized/homogenized full-fatmilk (A), pasteurized skim milk (6) and pasteurized skim milk containing 0.03%(w/w), modified konjac (>) at 20 �C. Traces are the mean of data from triplicateanalyses.

J.T. Tobin et al. / International Dairy Journal 21 (2011) 790e797794

3.3. Interaction between skim milk and modified konjac

In contrast to the interaction between native konjac and skimmilk, modified konjac was found, both visually andmicroscopically,to be stable to phase separation at addition levels up to 0.12% (w/w;micrographs not shown). This reduction in thermodynamicincompatibility may be attributed to the reduction in the weight-averaged molecular weight of the konjac (Table 1) as a result of thehigh fluidic shear and pressure generated during the micro-fluidization process. Consequently, a greater number of polymerchains are required to cause phase separation, due to a reduction inhydrodynamic volume compared with the native konjac molecules(Tuinier et al., 2000).

3.4. Viscosity measurements

The addition of modified konjac to skim milk significantly(p< 0.05) increased the apparent viscosity to a level greater than orequal to that of full-fat milk as measured at 20 �C (Table 2). Profilesof shear stress plotted against shear rate for both pasteurized(Fig. 5) and UHT-treated (Fig. 6) skim show, for comparativepurposes, that the addition level that most closely related to theshear stress profile for full-fat milk at 20 �C was 0.03% (w/w)modified konjac. As the temperature of full-fat milk decreases, theviscosity increases due to changes to the crystalline structure of the

Table 2Apparent viscosity, particle size and sedimentation rates of pasteurized and UHT-treated

Samplesb Pasteurized viscosity(mPa s)

UHT viscosity(mPa s)

Full-fat 2.48b

Skim 1.99ac 2.30a

Skim þ 0.03% (w/w) konjac 2.27bc 2.56a

Skim þ 0.06% (w/w) konjac 2.65bd 2.89ba

Skim þ 0.09% (w/w) konjac 3.10e 3.19cb

Skim þ 0.12% (w/w) konjac 3.69f 3.61c

a Values within a column not sharing a common superscript differ significantly (p < 0b Full-fat: commercial pasteurized and homogenized full-fat milk.

fat; at low temperatures, a higher level of addition of konjac may berequired to achieve the desired viscosity. The profiles of viscosityplotted against konjac concentration (%, w/w) show a linear rela-tionship for both pasteurized (R2 ¼ 0.98) and UHT (R2 ¼ 0.99) skimmilk containing modified konjac in the range 0e0.12% (w/w), withthe viscosity profile of full-fat in the temperature range 5e37 �Cincluded for comparative purposes (Fig. 7). Comparison of theapparent viscosity of pasteurized and UHT-treated samples(0e0.12%, w/w, konjac) as measured at 800 s�1 (Table 2), shows thedifference between the means not to be significantly different(p > 0.05).

3.5. Sedimentation analysis

The stability of both pasteurized and UHT milk samples con-taining konjac was investigated, during simulated storage (w6months). However, since pasteurized samples were phase stable atall levels of added modified konjac and due to the relatively shortshelf-life of pasteurized milk, only the results for UHTmilk samplesare discussed here. It has been shown that UHT-treated milks tendto be unstable during storage which can limit their use (Dalgleish,1992). The tendency of casein micelles, undergoing simulatedstorage in UHT milk, to sediment under accelerated gravity wascalculated (Table 2). The addition of modified konjac significantly(p < 0.05) decreased the movement of casein micelles towards thebottom of the sample cell, from a rate of 0.128 mm day�1 for UHTskim milk to 0.098 mm day�1 for samples containing 0.12% (w/w)

skim milk containing 0e0.12% (w/w) modified konjac.a

Pasteurized particlesize (nm)

UHT particlesize (nm)

UHT sedimentationrate (mm day�1)

180.27a 224.05a 0.128a

180.63a 221.68a 0.117ba

180.37a 229.42a 0.109bc

179.00a 236.01a 0.105bc

189.43a 226.38a 0.098c

.05).

0.0

0.5

1.0

1.5

2.0

2.5

0 200 400 600 800 1000

Shear rate (s-1)

Shea

r st

ress

(Pa

)

Fig. 6. Profile of shear stress versus shear rate for pasteurized/homogenized full-fatmilk (A), UHT skim milk (6) and UHT skim milk containing 0.03% (w/w) modifiedkonjac (>) at 20 �C. Traces represent the mean of data from triplicate analyses.

J.T. Tobin et al. / International Dairy Journal 21 (2011) 790e797 795

modified konjac. Dalgleish (1992) estimated the sedimentationbehaviour for large and small casein micelles in milk, underquiescent conditions, to be 0.186 and 0.071 mm day�1, respectively,which is consistent with the results reported in this study. Althoughthere is movement towards the bottom of the Lumifuge tube, forUHT-treated samples, over a simulated 6-month storage period,a solid sediment was not observed, as indicated by the continuedtransmission of light through the lower portion of the sample cell(Fig. 8A and B; position 112e114 mm). Initially during the sedi-mentation process, the area of greatest optical density wasobserved at the 114 mm position, typical of pellet formation.However, as the sedimentation process continued, the opticaldensity at 114 mm decreased over time, possibly due to particlesdiffusing from an area of high concentration to an area of lower

y = 14.1x + 1.894

R2 = 0.9812

y = 10.833x + 2.26

R2 = 0.9933

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

0 0.02 0.04 0.06

Konjac

Vis

cosi

ty (

mP

a s)

0 5 10 15 2

Tempera

Fig. 7. Relationship between apparent viscosity and konjac concentration for pasteurized (temperature for full-fat milk (-). Traces represent the mean of triplicate analyses.

concentration. This could be as a result of the polydisperse sizedistribution of casein micelles, which dictates the rate of sedi-mentation and diffusion, in conjunction with the relatively lowcentrifugal forces applied (Dalgleish, 1992). This area of relativelylow optical density, attributed to redispersion of the casein micellesby diffusion, appears to be caused by micelles moving in a flotationdirection (i.e., moving from right to left along the x-axis) encoun-tering micelles moving in a sedimentation direction (i.e., movingfrom left to right along the x-axis) forming an area of high opticaldensity at the 112 mm position along the x-axis rather than thepredicted position of 114 mm at the bottom of the sample cell. Theseparation profile for UHT-treated skim milk containing 0.12%(w/w)modified konjac (Fig. 8B) showed that the presence of konjacsignificantly (p < 0.05) decreased the rate of movement of thedispersed phase. This resulted in increased stability of the UHT-treated skim milk towards sedimentation, most likely due to theincrease in the intrinsic viscosity of the system (Table 2), ratherthan any interaction or weak electrostatic bond between konjacand the casein micelle, for which an increased particle size mightbe observed.

3.6. Particle size analysis and colour measurement

UHT treatment of skim milk samples caused a significant(p< 0.05) increase in caseinmicelle size (paired T-test) compared tothe sizes obtained for pasteurized samples (Table 2). This increase inparticle size could be caused byheat-denaturedwheyproteins, suchas b-lactoglobulin, complexing with casein micelles, along withpossible deposition of calciumphosphate onto themicelles (Singh &Fox, 1987; Wahlgren, Dejmek, & Drakenberg, 1990). However,addition of increasing levels of modified konjac did not significantly(p < 0.05) effect the particle size of either UHT or pasteurizedsamples post heat treatment, indicating that konjac had notadsorbed onto the casein micelle in the pH range employed (pH6.8e6.9).

Colour measurements were carried out to compare the effect ofthe addition of modified konjac to skim milk solutions. The resultsin Table 3 showed that full-fat milk had a whiter colour than skim

0.08 0.1 0.12 0.14

(%, w/w)

0 25 30 35 40

ture (°C)

A) and UHT (>) skim milk (20 �C), and relationship between apparent viscosity and

Fig. 8. Representative Lumifuge profiles of sedimentation rates of UHT-treated skim milk containing (A) 0% (w/w) and (B) 0.12% (w/w) modified konjac, during simulated storage(1140 � g).

J.T. Tobin et al. / International Dairy Journal 21 (2011) 790e797796

milk, as indicated by the higher L* value; also, full-fat milk hada positive b* value (indicating yellowness). In contrast, a negativevalue (indicating blueness) was obtained for skim milk. Addingincreasing levels of mechanically modified konjac did not

Table 3Colorimetric measurements of pasteurized and UHT-treated skim milk containing0e0.12% (w/w) modified konjac.a

Samplesb L* a* b*

PasteurizedFull-fat 86.30 �3.42 7.44Skim 70.86a �3.93a �6.23a

Skim þ 0.03% (w/w) konjac 82.70a �4.39a �3.93a

Skim þ 0.06% (w/w) konjac 72.38a �4.46a �3.89a

Skim þ 0.09% (w/w) konjac 72.13a �4.05a �4.68a

Skim þ 0.12% (w/w) konjac 71.24a �4.28a �4.80a

UHTSkim 73.46a �3.21a �3.09a

Skim þ 0.03% (w/w) konjac 74.23a �3.60a �2.07a

Skim þ 0.06% (w/w) konjac 73.98a �3.60a �2.04a

Skim þ 0.09% (w/w) konjac 73.89a �3.54a �2.10a

Skim þ 0.12% (w/w) konjac 73.66a �3.52a �2.13a

a Values within a column not sharing a common superscript differ significantly(p < 0.05).

b Full-fat: commercial pasteurized and homogenized full-fat milk.

significantly (p> 0.05) effect the colour values of either pasteurizedor UHT samples.

4. Conclusion

Addition of native konjac to skim milk caused phase separationat levels of 0.03% w/w. However, high shear modification of thekonjac reduced the molecular weight of the polysaccharide, whichaltered its thermodynamic behaviour in the presence of protein,thus allowing it to be added to skim milk as a viscosity-enhancingagent without the occurrence of macroscopic phase separation.Addition of modified konjac increased the stability to sedimenta-tion of UHT-treated skim milk during simulated storage, withoutsignificantly affecting either the colour or particle size post heattreatment. In conclusion, addition of mechanically modified (lowmolar mass) konjac could provide a mechanism to enhance boththe viscosity and stability of colloidal protein preparations atconcentrations more suitable for processing, compared to nativekonjac.

References

Al-Ghazzewi, F. H., Khanna, S., Tester, R. F., & Piggot, J. (2007). Potential use ofhydrolysed konjac glucomannan as a prebiotic. Journal of the Science of Food andAgriculture, 87, 1758e1766.

J.T. Tobin et al. / International Dairy Journal 21 (2011) 790e797 797

Cook, E. J., & Lagace, A. P. (1985). Apparatus for forming emulsions. U. S. Patent,4,533,254.

Dalgleish, D. G. (1992). Sedimentation of casein micelles during the storage of ultra-high temperature milk products - a calculation. Journal of Dairy Science, 75,371e379.

Fang, W., & Wu, P. (2004). Variations of konjac glucomannan (KGM) from Amorpho-phallus konjac and its refined powder in China. Food Hydrocolloids, 18, 167e170.

Fitzsimons, S. M., Tobin, J. T., & Morris, E. R. (2008). Synergistic binding of konjacglucomannan to xanthan on mixing at room temperature. Food Hydrocolloids,22, 36e46.

Grinberg, V. Y., & Tolstoguzov, V. B. (1997). Thermodynamic incompatibility ofproteins and polysaccharides in solutions. Food Hydrocolloids, 11, 145e158.

Katsuraya, K., Okuyama, K., Hatanaka, K., Oshima, R., Sato, T., & Matsuzaki, K. (2003).Constitution of konjac glucomannan: chemical analysis and 13C NMR spec-troscopy. Carbohydrate Polymers, 53, 183e189.

de Kruif, C. G., & Tuinier, R. (2001). Polysaccharide protein interactions. FoodHydrocolloids, 15, 555e563.

Lagoueyte, N., & Paquin, P. (1998). Effects of microfluidization on the functionalproperties of xanthan gum. Food Hydrocolloids, 12, 365e371.

Li, B., Xia, J., Wang, Y., & Xie, B. (2005). Grain-size effect on the structure andantiobesity activity of konjac flour. Journal of Agricultural and Food Chemistry, 53,7404e7407.

McCrae, C. (1994). Homogenization of milk emulsions: use of microfluidizer. Journalof the Society of Dairy Technology, 47, 28e31.

Mounsey, J. S., & O’Kennedy, B. T. (2009). Stability of b-lactoglobulin/micellar caseinmixtures on heating in simulated milk ultrafiltrate at pH 6.0. InternationalJournal of Dairy Technology, 62, 493e499.

Paquin, P., & Giasson, J. (1989). Microfluidization as an homogenization process forcream liqueur. Lait, 69, 491e498.

Penroj, P., Mitchell, J. R., Hill, S. E., & Ganjanagunchorn, W. (2005). Effect of konjacglucomannan deacetylation on the properties of gels formed from mixtures of

kappa carrageenan and konjac glucomannan. Carbohydrate Polymers, 59,367e376.

Pouliot, Y., Paquin, P., Robin, O., & Giasson, J. (1991). Etude comparative de l’effect deal microfluidisation et de l’homogeneisation sur la distribution de la taille desglobules de gras du lait de vache. International Dairy Journal, 1, 39e49.

Ratcliffe, I., Williams, P. A., Viebke, C., & Meadows, J. (2005). Physicochemicalcharacterisation of konjac glucomannan. Biomacromolecules, 6, 1977e1986.

Singh, H., & Fox, P. F. (1987). Heat stability of milk: role of b-lactoglobulin in thepH-dependent dissociation of micellar k-casein. Journal of Dairy Research, 54,509e521.

Strawbridge, K. B., Ray, E., Hallett, F. R., Tosh, S. M., & Dalgleish, D. G. (1995).Measurement of particle size distribution in milk homogenized by a micro-fluidizer: estimation of populations of particles with radii less than 100 nm.Journal of Colloid and Interface Science, 171, 392e398.

Syrbe, A., Bauer, W. J., & Klostermeyer, H. (1998). Polymer science concepts in dairysystems - an overview of milk protein and food hydrocolloid interaction.International Dairy Journal, 8, 179e193.

Tobin, J. T., Fitzsimons, S. M., Kelly, A. L., Kelly, P. M., Auty, M. A. E., & Fenelon, M. A.(2010). Microparticulation of mixtures of whey protein and inulin. InternationalJournal of Dairy Technology, 63, 32e40.

Tolstoguzov, V. B. (1991). Functional properties of food proteins and role of protein -polysaccharide interaction. Food Hydrocolloids, 4, 429e468.

Tuinier, R., ten Grotenhuis, E., & de Kruif, C. G. (2000). The effect of depolymerisedguar gum on the stability of skim milk. Food Hydrocolloids, 14, 1e7.

Vuksan, V., Jenkins, D. J. A., Spadafora, P., Sievenpiper, J. L., Owen, R., Vidgen, E., et al.(1999). Konjac-mannan (glucomannan) improves glycemia and other associ-ated risk factors for coronary heart disease in type 2 diabetes. Diabetes Care, 22,913e919.

Wahlgren, M., Dejmek, P., & Drakenberg, T. (1990). A 43Ca and 31P NMR study of thecalcium and phosphate equilibria in heated milk solutions. Journal of DairyResearch, 57, 355e364.