The objective of this study was to investigate the influence of molecular weight of kcarrageenan on its ice recrystallization inhibition (IRI) activity. To reduce the molecular weight, acid hydrolysis with 0.1 M trifluoroacetic acid (TFA) at 80 C or 0.1 M hydrochloric acid (HCl) at 60 C was performed. In addition, molecular weight was reduced by dialyzing kcarrageenan against demineralized water. It was demonstrated that IRI activity of kcarrageenan decreases with decreasing molecular weight, which is contrary to previous studies. It was shown in our study that the different results may be attributed to an aggregation of kcarrageenan molecules due to high NaCl concentration, originated from HCl and subsequent neutralization with NaOH. This aggregation causes a decrease in IRI activity. Furthermore, it was shown that the addition of a small amount of NaCl can lead to an increase in IRI activity of kcarrageenan
Journal of Food Engineering 209 (2017) 26e35 Contents lists available at ScienceDirect Journal of Food Engineering journal homepage: www.elsevier.com/locate/jfoodeng Influence of acid hydrolysis and dialysis of k-carrageenan on its ice recrystallization inhibition activity €nder b, Daniel Wefers b, Mirko Bunzel b, Volker Gaukel a Andreas Leiter a, *, Johanna Maila a KIT (Karlsruhe Institute of Technology), Institute of Process Engineering in Life Sciences, Section I: Food Process Engineering, Kaiserstrasse 12, 76131 Karlsruhe, Germany b KIT (Karlsruhe Institute of Technology), Department of Food Chemistry and Phytochemistry, Institute of Applied Biosciences, Adenauerring 20a, 76131 Karlsruhe, Germany a r t i c l e i n f o a b s t r a c t Article history: Received December 2016 Received in revised form April 2017 Accepted 11 April 2017 Available online 12 April 2017 The objective of this study was to investigate the influence of molecular weight of k-carrageenan on its ice recrystallization inhibition (IRI) activity To reduce the molecular weight, acid hydrolysis with 0.1 M trifluoroacetic acid (TFA) at 80 C or 0.1 M hydrochloric acid (HCl) at 60 C was performed In addition, molecular weight was reduced by dialyzing k-carrageenan against demineralized water It was demonstrated that IRI activity of k-carrageenan decreases with decreasing molecular weight, which is contrary to previous studies It was shown in our study that the different results may be attributed to an aggregation of k-carrageenan molecules due to high NaCl concentration, originated from HCl and subsequent neutralization with NaOH This aggregation causes a decrease in IRI activity Furthermore, it was shown that the addition of a small amount of NaCl can lead to an increase in IRI activity of k-carrageenan © 2017 Elsevier Ltd All rights reserved Keywords: Acid hydrolysis Dialysis Ice recrystallization inhibition Ions k-Carrageenan Molecular weight Introduction Freezing is a popular method for food preservation It extends the shelf life by slowing down chemical and enzymatic reactions as well as microbial reproduction (Gaukel, 2016) However, during storage and distribution of frozen products, especially under unfavorable temperature conditions, changes of ice crystal microstructure can affect the quality of frozen food products These changes are termed as “recrystallization” which is defined as changes in size, number, and shape of unique ice crystals while keeping the total volume of ice constant (Cook and Hartel, 2010) Recrystallization processes have a high impact on the appearance and on the texture of frozen food (Pham and Mawson, 1997) For instance, ice cream becomes coarse and unacceptable if ice crystals grow during storage and exceed a threshold detection size (Hartel, 1996, 2001) Recrystallization processes can be retarded by low and constant storage temperatures (Donhowe and Hartel, 1996) or by formulation, especially by the addition of ice recrystallization inhibitors Traditionally, hydrocolloids are added to inhibit * Corresponding author E-mail address: andreas.leiter@kit.edu (A Leiter) http://dx.doi.org/10.1016/j.jfoodeng.2017.04.013 0260-8774/© 2017 Elsevier Ltd All rights reserved recrystallization in ice cream (Adapa et al., 2000; Bahramparvar and Mazaheri Tehrani, 2011; Marshall et al., 2003; Miller-Livney and Hartel, 1997) However, despite a large amount of research, the exact recrystallization inhibition mechanism of hydrocolloids is still not understood in every detail (Bahramparvar and Mazaheri Tehrani, 2011; Leiter and Gaukel, 2016) One such hydrocolloid is k-carrageenan which is a linear, sulfated polysaccharide extracted from certain species of red seaweeds (Rhodophyceae) (McHugh, 2003) It is composed of an alternating disaccharide unit of 1,3-linked b-D-galactose-4-sulfate and 1,4-linked 3,6-anhydro-a-D-galactose (Rochas and Rinaudo, 1984) The weight average molecular weight of commercially available k-carrageenan ranges typically from 300 to 650 kDa (Hoffmann et al., 1996; Lahaye, 2001; Piculell, 2006; Prajapati et al., 2014) However, k-carrageenan is quite polydisperse Thus, most of the material is found in the range of 102e103 kDa but there is also a long tail on the low-molecular side of the distribution (Piculell, 2006) k-carrageenan can form thermoreversible gels upon cooling (Piculell, 2006) Therefore, the polysaccharide is used as gelling, thickening, and stabilizing agent especially in food products and in cosmetics and pharmaceutical formulations (Campo et al., 2009; McHugh, 2003; Rinaudo, 2008; van de Velde and De Ruiter, 2006) The molecular mechanism of gelation is still not A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 understood in detail A widely accepted model of the gelation mechanism is the “domain model” (Mangione et al., 2003; Morris et al., 1980) According to the “domain model”, k-carrageenan molecules exist in solution as unstructured random coils above a certain temperature A temperature reduction below this so called coil-helix transition temperature induces the formation of double helices Below the transition temperature, the intermolecular association between the double helices is limited to the formation of small independent domains involving a limited number of double helices However, when cations are present, double helices of different domains aggregate to enable long-range cross-linking (Campo et al., 2009; Morris et al., 1980) which can lead to the formation of a gel (Mangione et al., 2003) However, despite a lot of research in this area, it is still a matter of debate if double or single helices are formed (Ciancia et al., 1997; Rochas and Rinaudo, 1984; Schefer et al., 2015; Smidsrød and Grasdalen, 1982; Ueda et al., 1998; Viebke et al., 1995) In addition to its ability to form a gel, k-carrageenan shows strong ice recrystallization inhibition (IRI) activity in frozen sucrose skasolutions (Chun et al., 2012; Gaukel et al., 2014; Kamin rznicka et al., 2015; Leiter et al., 2017) Although several poDwo tential mechanisms for IRI activity of hydrocolloids have been suggested, the exact IRI mechanism of k-carrageenan is still not understood It is assumed that the IRI activity of certain hydrocolloids originates from a decrease in mobility of water molecules due to water binding or steric hindrance (Caldwell et al., 1992; MillerLivney and Hartel, 1997; Regand and Goff, 2002, 2003) However, by using pulsed field gradient (PFG) nuclear magnetic resonance (NMR) it was demonstrated that the addition of k-carrageenan to a sucrose solution does not significantly influence the self-diffusion coefficient of water (Gaukel, 2004) Furthermore, the formation of a gel is often discussed as a possible IRI mechanism of hydrocolloids (Bahramparvar and Mazaheri Tehrani, 2011; Goff et al., 1999; Regand and Goff, 2003) However, investigations on the relation of gelation and IRI activity of k-carrageenan showed that the formation of a k-carrageenan gel results in a reduction of its IRI activity (Leiter et al., 2016a) Moreover, it is assumed that certain hydrocolloids may retard the incorporation of water molecules into the ice crystal lattice by modifying the ice crystal/liquid interface (Martin et al., 1999) or by binding to the ice crystal surface (Gaukel et al., 2014; Martin et al., 1999; Sutton et al., 1996, 1997) In previous studies, it was shown that ice crystals in a k-carrageenan sucrose solution are more rectangular and elongated similar to ice crystal shapes found in sucrose solutions with ice-binding proteins (IBP) (Gaukel et al., 2014) In addition, it was shown that in solutions, where k-carrageenan exhibits IRI activity, more angular ice crystals are present, in contrast to rather round ice crystals in solutions, where k-carrageenan exhibits no IRI activity (Leiter et al., 2017) Therefore, it has been suggested that the IRI activity of k-carrageenan originates from an interaction with the ice crystal surface similar to IBPs or polyvinyl alcohol (PVA) (Budke and Koop, 2006; Davies, 2014; Inada and Lu, 2003) In a previous study we showed that IRI activity of k-carrageenan was significantly increased at pH adjusted by hydrochloric acid (Leiter et al., 2017) The increase in IRI activity was discussed to be possibly attributed to a reduction in molecular weight due to hydrolysis This led to the hypothesis that smaller k-carrageenan ska-Dwo rznicka et al molecules exhibit a higher IRI activity Kamin (2015) also showed a significant increase in IRI activity when the average molecular weight of k-carrageenan was reduced from 34,000 kDa to 2700 kDa by acid hydrolysis using hydrochloric acid However, the considerably higher weight average molecular weight in this study compared to the typical molecular weight of untreated k-carrageenan (300e650 kDa) suggests aggregation of k-carrageenan molecules Aggregation possibly occurred due to the 27 presence of Naỵ cations by neutralization with 0.1 M NaOH or due to a high ion content of the raw material In our previous papers (Leiter et al., 2016a, 2017), we already showed that aggregation of kcarrageenan helices is induced by the presence of cations, which leads to a reduction of IRI activity Thus, it is not sure whether the skaincreased IRI activity, which was observed by Kamin rznicka et al (2015), is really due to an increased IRI activity Dwo of hydrolyzed k-carrageenan molecules or just due to a reduction of aggregates Therefore, the objective of this study was to investigate if there is an increased IRI activity of k-carrageenan molecules with reduced molecular weight For this purpose, k-carrageenan was hydrolyzed by different methods and reduction of molecular weight was verified by size-exclusion chromatography (SEC) before IRI experiments Hydrolysis methods were adjusted to avoid additional salt in the k-carrageenan samples Thus, IRI experiments should not be influenced by aggregation of k-carrageenan molecules Firstly, k-carrageenan was hydrolyzed by 0.1 M trifluoroacetic acid (TFA), because TFA is volatile and evaporates during freezedrying Therefore, no subsequent neutralization of the acid is necessary and no ions are added to the polysaccharide preparation Secondly, molecular weight was reduced by dialyzing k-carrageenan against demineralized water and subsequent freeze-drying This process was described to induce autohydrolysis of the kcarrageenan (Hoffmann et al., 1996) For the examination of dif ska-Dwo rznicka et al (2015) ferences between the results of Kamin and the results provided in this study, k-carrageenan was also hydrolyzed with 0.1 M hydrochloric acid (HCl) and neutralized with NaOH In addition, rheological temperature sweep measurements were performed to detect possible aggregation of k-carrageenan molecules Materials and methods 2.1 Materials k-carrageenan was used as unstandardized extract from the red seaweed Eucheuma Cottonii produced by gel-press method in the Philippines It was provided by Eurogum A/S (Herlev, Denmark) and was used in this study without further purification Ion (major element is potassium with a mass fraction of 7.29%) and sulfate content (19.55%) of the unstandardized sample is described in previous papers (Leiter et al., 2016a, 2017) Because the composition of natural k-carrageenan may differ from batch to batch (van de Velde et al., 2002), k-carrageenan from the same batch was used for all experiments Sucrose was common household sugar from a local supplier All other chemicals were of analytical grade 2.2 Preparation of different k-carrageenan samples 2.2.1 Dialysis k-Carrageenan solution (1 mg mLÀ1) was prepared by dissolving the k-carrageenan extract in pure water at 60 C for 30 An aliquot (50 mL) of this k-carrageenan solution was dialyzed 36 h against L of demineralized water using a dialysis tubing (Carl Roth, Germany) with a molecular weight cut-off of 3500 Da The demineralized water was replaced every 12 h Subsequently, the dialyzed k-carrageenan solution was freeze-dried For an optimal comparison of the IRI activity of the dialyzed kcarrageenan with the untreated k-carrageenan, the same ion concentration has to be present in both solutions Therefore, a second dialyzed k-carrageenan sample, which has the same ion concentration as the untreated k-carrageenan, was prepared For this purpose, the polysaccharide solution was treated as described above, but the dialysis tubing was placed in L of 2.5 mM KCl 28 A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 solution after the dialysis against demineralized water to add Kỵ cations to the dialyzed k-carrageenan before freeze-drying The KCl solution was replaced every 12 h for 36 h The concentration of 2.5 mM KCl is the same as the total salt concentration in a mg mLÀ1 k-carrageenan extract solution (Leiter et al., 2016a, 2017) To obtain a sufficient amount of freeze-dried sample for the following experiments, this procedure was repeated several times and the freeze-dried samples were mixed and stored in a desiccator 2.2.2 Acid hydrolysis of k-carrageenan with trifluoroacetic acid k-carrageenan stock solutions (2 mg mLÀ1) for acid hydrolysis were prepared by dissolving k-carrageenan extract or dialyzed kcarrageenan (each 400 mg) in 200 mL of pure water at 60 C for 30 After cooling down to room temperature, 10 mL of the stock solution was mixed with 10 mL of 0.2 M trifluoroacetic acid (TFA) to get a final TFA concentration of 0.1 M The 0.1 M TFA k-carrageenan solution was heated to 80 C and kept at this temperature for min, 10 min, 20 or 60 The hydrolysates were cooled down immediately by placing the solutions in ice water The conditions of the TFA hydrolysis should cleave some glycosidic bonds, but only negligible degradation of 3,6-anhydro-a-D-galactose units (Stevenson and Furneaux, 1991; Yang et al., 2009) To obtain a sufficient amount of each sample, the method was performed with several aliquots of the stock solution Finally, the aliquots of same TFA treatment times were mixed, freeze-dried and stored in a desiccator 2.2.3 Acid hydrolysis of k-carrageenan with hydrochloric acid For HCl hydrolysis, 10 mL of the stock solution was mixed with 10 mL of 0.2 M HCl to get a final HCl concentration of 0.1 M The HCl/ k-carrageenan solution was heated to 60 C and kept at this temperature for 30 The hydrolysate was cooled down immediately by placing the solution in ice water Just as for the acid hydrolysis with TFA, these hydrolysis conditions should cleave some glycosidic bonds, but negligible degradation of 3,6-anhydro-a-D-galactose units (Sun et al., 2015) To avoid further hydrolysis, the solution was adjusted to pH with 0.1 M NaOH before freeze-drying In the results section, this freeze-dried sample is designated as sample B A non-hydrolyzed reference k-carrageenan sample, which contains the same amount of salt as the hydrolyzed k-carrageenan sample, was prepared by dissolving 20 mg of k-carrageenan in 20 mL of 0.1 M NaCl solution The pH of this solution was Therefore, the pH was adjusted to with 0.1 M NaOH The nonhydrolyzed reference sample solution was freeze-dried and designated as sample A To obtain a sufficient amount of each sample, the method was performed with several aliquots of the stock solution, and samples were stored as described in section 2.2.2 The calculated mass ratio of NaCl to k-carrageenan in both sample A and B is 5.844 Thus, the final NaCl concentration cNaCl in a 49% sucrose solution with a specific concentration cA/B of sample A or B can be calculated according to the following equation: cA=B ỵ 5:844 cNaCl ẳ 2300, Knauer, Germany) As helix formation and aggregation of kcarrageenan molecules affect the SEC separation (Ueda et al., 1998; Viebke et al., 1995), the column was heated at 60 C to maintain kcarrageenan molecules in the disordered coil form and to prevent aggregation (see Fig and section 3.1) Sodium phosphate buffer (0.05 M, pH 6) containing 0.2 M NaCl was used as eluent at a flow rate of 0.5 mL minÀ1 Samples were dissolved in the mobile phase (1 mg mLÀ1), and the injection volume was mL A calibration with dextran standards (2000 kDa, 1050 kDa, 400 kDa, 125 kDa, 45 kDa; Fluka, Switzerland) was performed for a rough estimation of the apparent molecular weight of the main peaks Mp of the elution profile The relationship between molecular weight of the main peak Mp (Da) and retention time t (min) followed equation (2): log Mp ¼ À0:00814  t þ 7:11634 R2 ¼ 0:9875 (2) 2.4 Recrystallization experiments 2.4.1 Sample preparation IRI activity of the different hydrolyzed k-carrageenan samples was studied in a 49% (w/w) sucrose solution prepared with demineralized water For this purpose, the appropriate amount of kcarrageenan sample (final concentration 0.146 mg mLÀ1, mg mLÀ1 or 6.844 mg mLÀ1) was dissolved in the sucrose solution by continuous stirring at 60 C for 30 After dissolving, the solution was cooled down to room temperature In order to investigate the influence of NaCl on the IRI activity of k-carrageenan, the appropriate amount of salt (final concentration 0.3 mM, 14.6 mM, 30 mM, 60 mM or 100 mM) was mixed with the k-carrageenan powder (final concentration mg mLÀ1) before adding the sucrose solution Then, the k-carrageenan-salt mix was dissolved in the sucrose solution by continuous magnetic stirring at 60 C for 30 2.4.2 Sample freezing and storage Sample freezing and storage was performed as previously described by Leiter et al (2016b) For ice recrystallization analysis, (1) 2.3 Analysis by size exclusion chromatography Polymer degradation was analyzed by size-exclusion chromatography (SEC) The system consisted of a K-500 pump (Knauer, Germany), a Degasys-1310 degasser (Knauer, Germany), a Sepharose CL-4B column (bed volume: 85 cm  1.6 cm, GE-Healthcare, Great Britain), and a refractive index (RI) detector (Smartline Fig Temperature sweep analysis (2 C minÀ1) of untreated k-carrageenan extract (ck-carr ¼ mg mLÀ1) dissolved in the mobile phase (0.05 M sodium phosphate buffer (pH 6) containing 0.2 M NaCl) used in size-exclusion chromatography analyses A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 sample solution (18 ml) was placed between two microscope cover slips (0.13e0.16 mm thick) stuck on an object slide The slide was covered with another cover slip and sealed with silicone The cover slips were used as spacers to allow ice crystal growth in all dimensions After drying of the silicone (1 h), samples were subjected to fast freezing by immersion in liquid nitrogen for a few seconds By using this freezing step, the aqueous solution was transformed into a glassy state After freezing, samples were stored at a constant temperature of À12 C ± 0.1 C in a small externally cooled storage chamber This procedure allows the samples to crystallize in a uniform way during equal heating conditions from the glassy state to the storage temperature In contrast to the temperature of a typical household freezer (about À18 C), a storage temperature of À12 C was selected to accelerate ice recrystallization processes and thus reduce the necessary storage time for a IRI characterization (Donhowe and Hartel, 1996) Under the described conditions, resulting ice volume fraction is approximately 22% according to the sucrose/water phase diagram from Riedel (1949) The storage chamber itself was placed in a cooled glove box with a temperature of À12 C ± C The temperature inside the storage chamber and glove box was recorded by thermocouples during the storage time of five days 2.4.3 Microscopy and image analysis During storage time, pictures of ice crystals were taken with a digital camera (Altra SIS20, Olympus, Japan) attached to a polarization microscope (BX41, Olympus, Japan) installed in the glove box Pictures were taken h and 96 h after freezing For evaluation of the pictures, the contour of an ice crystal was manually circumscribed on a computer with the software ImagePro Insight 9.1 (Media Cybernetics, USA) From the defined area of each ice crystal, the equivalent diameter was calculated as the diameter of a circle with the same area To get a representative distribution of ice crystal sizes, 300 to 400 ice crystals were analyzed per object slide and the arithmetic equivalent diameter was calculated For each experiment, two object slides were prepared and each experiment was performed twice Thus, mean ice crystal diameter x and standard deviations were determined from the arithmetic equivalent diameters of four object slides For determination of statistical significance of mean ice crystal diameters One-way ANOVA with Tukey’s test was carried out (alpha ¼ 0.05) In this study, the term “IRI activity” of k-carrageenan is used when mean ice crystal diameters in sucrose solution with kcarrageenan are significantly smaller than mean ice crystal diameters in a pure sucrose solution at a given storage time In addition, the smaller the mean ice crystal diameter in a solution with k-carrageenan, the higher its IRI activity is rated However, this is only true when k-carrageenan does not affect ice crystal nucleation To the best of our knowledge, the influence of k-carrageenan on nucleation was not investigated so far, and only limited data are available for nucleation from the glassy state in food systems (Hartel, 2001) However, in contrast to nucleation from the glassy state, Muhr and Blanshard (1986) showed that polysaccharides not significantly affect ice nucleation in a sucrose solution, when the system crystallizes from solution phase Thus, even if we cannot completely exclude that k-carrageenan affects ice nucleation from the glassy state, we assume that the initial ice crystal diameters are the same for all sucrose solutions in this study However, additional experiments are necessary in the future to prove this assumption 2.5 Rheological temperature sweep measurements Rheological measurements were performed on a Physica MCR 101 rheometer (Anton Paar GmbH, Austria) with a double gap geometry (DG 26.7) Temperature was controlled by the temperature 29 device C-PTD 200 with Peltier temperature control (Anton Paar GmbH, Austria) Coil-helix transition and aggregation was determined by a temperature sweep, which was performed similar as described previously (Leiter et al., 2016a) For temperature sweep analysis, k-carrageenan solutions were prepared as described in section 2.4.1 and kept at 60 C to maintain k-carrageenan in the disordered coil form The hot k-carrageenan solution was loaded in the measuring device preheated to 60 C To reduce water evaporation, the solution between double gap geometry and outer wall was coated with a thin layer of low viscosity paraffin oil Afterwards, the solution was held at 60 C for Storage modulus (G’) was determined during cooling (2 C minÀ1) to C and during reheating (2 C minÀ1) to 60 C after five minutes holding time at C Temperature sweep was performed at Hz within the linear viscoelastic region All rheological data shown are the mean of two experimental replicates Results and discussion 3.1 Determination of the coil-helix and helix-coil transition temperatures of k-carrageenan First, it was evaluated whether a column temperature of 60 C during the size-exclusion chromatography (SEC) allows for the analysis of k-carrageenan in the disordered coil form Hereby, it can be ascertained that the chromatograms reflect single k-carrageenan molecules instead of k-carrageenan aggregates For this purpose, rheological temperature sweep analysis with untreated k-carrageenan extract dissolved in the SEC mobile phase was performed In this experiment, aggregation is usually evident from a thermal hysteresis between the transition temperatures from coil to helix during cooling and from helix to coil during heating (Piculell, 2006) As shown in Fig 1, G0 increases rapidly at about 20 C during cooling, which is considered to be due to the coil-helix transition During heating, the decrease of G’, which is considered as the helix-coil transition, is shifted to a higher temperature of about 30 C Thus, a thermal hysteresis between the transition temperatures and aggregation of k-carrageenan molecules take place However, at a temperature of 60 C, k-carrageenan is in the coil form as the temperature is beyond the helix-coil transition temperature of about 45 C Thus, it can be concluded that the estimated molecular weight at a column temperature of 60 C represents single k-carrageenan molecules 3.2 Molecular weight distribution and ice recrystallization inhibition (IRI) activity of k-carrageenan after acid hydrolysis with 0.1 M trifluoroacetic acid (TFA) SEC chromatograms of untreated and hydrolyzed k-carrageenan are shown in Fig The hydrolysis with 0.1 M TFA was performed for different time periods For the untreated k-carrageenan, the molecular weight of the main peak Mp is about 1420 kDa according to equation (2) Besides, there is a typical long tail at the lowmolecular side of the distribution, which has already described in literature (Piculell, 2006) The second peak at a retention time of about 370 is most likely due to the ions, which are already present in the used k-carrageenan extract (see section 2.1) Hydrolysis with 0.1 M TFA results in a significant reduction of molecular weight for all hydrolysis times The main peak of molecular weight distribution is shifted from 1420 kDa to about 20.3 kDa Although the peaks of the k-carrageenan hydrolysates cannot be completely resolved, a continuous reduction of the molecular weight is evident from the chromatograms This is demonstrated by the increasing intensities of the peak at 20.3 kDa and the peak derived from low molecular weight compounds These results 30 A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 Fig Size-exclusion chromatograms of untreated k-carrageenan extract and freezedried k-carrageenan samples hydrolyzed with 0.1 M trifluoroacetic acid for different times The signal intensities are arbitrary, but the same scale is used for the refractive index signals of all curves show that hydrolysis of k-carrageenan with 0.1 M TFA leads to a reduction of molecular weight with increasing hydrolysis time In the next step, the IRI activity of the different k-carrageenan hydrolysates was investigated In Fig 3, the mean ice crystal diameters in frozen sucrose solutions containing hydrolyzed kcarrageenan samples after h and 96 h storage time are depicted In addition, the mean ice crystal diameters in a sucrose solution with and without untreated k-carrageenan are shown First of all, it can be concluded that recrystallization occurs in all solutions as indicated by an increase of the mean ice crystal diameters over storage time for all samples This is, as expected, mostly distinct in the sucrose solution without added k-carrageenan At both storage times, the mean ice crystal diameter in a solution with untreated k-carrageenan is significantly smaller than the mean ice crystal diameter in pure sucrose solution, showing the strong IRI activity of untreated k-carrageenan After a storage time of h only hydrolysis with 0.1 M TFA for 60 leads to a Fig Influence of different hydrolysis times with 0.1 M trifluoroacetic acid on ice recrystallization inhibition activity of k-carrageenan dissolved in a sucrose solution (wsuc ¼ 49% (w/w), ck-carr ¼ mg mLÀ1) Bars with the same small letter at the same storage time are not significantly different (P > 0.05) significantly larger mean ice crystal diameter compared to the untreated k-carrageenan Hydrolysis times shorter than 20 not exhibit a statistically significant influence on the mean ice crystal diameter during the first h of storage time Here, it is important to mention that the values of mean ice crystal diameters after h storage time are not the initial ice crystal sizes Due to experimental setup and sample preparation, it was not possible to take pictures before h storage time But even if we cannot completely eliminate the option that k-carrageenan affects ice nucleation, we assume that the initial ice crystal diameters are the same for all solutions (see section 2.4.3) After 96 h of storage time and, therefore, a longer time period for recrystallization processes, the influence of hydrolysis on the IRI activity of k-carrageenan becomes more clear It is apparent that the mean ice crystal diameters increase with increasing hydrolysis time However, the differences in mean ice crystal diameters of the hydrolyzed samples compared to the untreated sample are only statistically significant after hydrolysis times of 20 and 60 The pH value of all hydrolyzed k-carrageenan sucrose solutions was about Because the pH was lower than the pH of the untreated k-carrageenan sucrose solution (pH 6), it might be possible that not all TFA was evaporated during freeze-drying However, as all hydrolyzed k-carrageenan solutions have the same pH, the residual amounts of TFA in the different hydrolyzed k-carrageenan samples should be similar In addition, as there is no significant difference in IRI activity of untreated (pH 6) and hydrolyzed k-carrageenan (pH 4), an influence of non-evaporated residual TFA on IRI activity can be excluded As we used the same mass concentration of k-carrageenan samples (1 mg mLÀ1) in this recrystallization experiment, the molar concentration of k-carrageenan is significantly higher in solutions with hydrolyzed k-carrageenan due to the smaller molecular weight An influence on the ice content can be neglected as the molar concentration of the hydrolyzed k-carrageenan is too low to influence the equilibrium phase volume of ice Therefore, we can conclude that k-carrageenan molecules that have a lower molecular weight due to TFA hydrolysis exhibit a reduced IRI activity However, the hydrolyzed k-carrageenan samples still show IRI activity, because the mean ice crystal diameters of all hydrolyzed samples are significantly smaller than the mean ice crystal diameters in pure sucrose solution The question arises why the IRI activity of small k-carrageenan molecules is reduced Assuming that IRI activity of k-carrageenan originates from an interaction with the ice crystal surface, a conformational change of the k-carrageenan molecules could be a reason for the decreased IRI activity For example, Abad et al (2008) did not observe a conformational change from coil to helix of kcarrageenan molecules with a weight average molecular weight Mw of 34 kDa, which is in a similar range as the molecular weight of the used hydrolyzed k-carrageenan samples (see Fig 2) Therefore, the coil-helix transition was investigated In Fig the temperature sweep analysis of the untreated k-carrageenan extract and the hydrolyzed samples (5 and 60 min) dissolved in 49% sucrose solution are shown It is apparent that there is a coil-helix transition in solution with untreated k-carrageenan extract at a temperature of about 25 C In addition, a small thermal hysteresis and thus aggregation is detectable Aggregation is probably due to the presence of cations in the k-carrageenan extract (Leiter et al., 2016a, 2017) In contrast, no coil-helix transition and aggregation is detectable in both hydrolyzed k-carrageenan solutions These results are in agreement with those obtained by Abad et al (2008) Because there is no significant difference in IRI activity of untreated and hydrolyzed k-carrageenan (see Fig 3), it seems that coilhelix transition is not a necessary condition for IRI activity However, we cannot completely exclude a coil-helix transition as Rochas A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 Fig Temperature sweep analysis (2 C minÀ1) of untreated k-carrageenan extract (ck-carr ¼ mg mLÀ1) and k-carrageenan hydrolyzed with 0.1 M trifluoroacetic acid for or 60 in 49% (w/w) sucrose solution To distinguish between the data points of hydrolyzed k-carrageenan samples, only every second data point of the 60 hydrolyzed sample is shown et al (1990) detected a coil-helix transition even for k-carrageenan molecules with a Mw of about kDa Possibly, the transition temperature is lower than C as coil-helix transition temperature decreases with decreasing molecular weight (Meunier et al., 2001) Furthermore, measurements of G’ in this low range might be insufficiently sensitive to detect the coil-helix transition Therefore, further investigations on the relation between k-carrageenan conformation and IRI activity are necessary 3.3 Molecular weight distribution and IRI activity of k-carrageenan after dialysis Beside the acid hydrolysis with 0.1 M TFA, the molecular weight of k-carrageenan was reduced by using dialysis Hoffmann et al (1996) showed that molecular weight decreases with increasing dialysis time of a mg mLÀ1 k-carrageenan solution against Milli-Q water and subsequent freeze-drying It can be seen from the results shown in Fig that 36 h dialysis of the untreated k-carrageenan extract against demineralized water and subsequent freeze-drying leads to a significant reduction of the molecular weight The molecular weight of the main peak Mp is shifted from 1420 kDa to about 23.1 kDa In contrast, the molecular weight reduction is significantly lower, when k-carrageenan is additionally dialyzed against a KCl solution before freeze-drying However, this procedure also yields a lower molecular weight compared to the molecular weight of the untreated k-carrageenan extract The main peak of the k-carrageenan additionally dialyzed against a KCl solution is shifted from 1420 kDa to 219 kDa while a small molecule fraction remains in the range of 1420 kDa In addition, the molecular weight distribution is much broader if compared to the k-carrageenan sample dialyzed against demineralized water Hoffmann et al (1996) provided a possible explanation for the significantly lower molecular weight of the dialyzed k-carrageenan in comparison to the dialyzed k-carrageenan with the addition of Kỵ cations The authors showed that dialysis against demineralized water leads to a conversion of the k-carrageenan molecule into the acid form causing a pH drop In our case, the pH of the k-carrageenan solution in the dialysis tubing dropped from about to 31 Fig Size-exclusion chromatograms of untreated k-carrageenan extract, k-carrageenan dialyzed against demineralized water, and k-carrageenan dialyzed against demineralized water and subsequently dialyzed against 2.5 mM KCl before freezedrying The signal intensities are arbitrary, but the same scale is used for the refractive index signals of all curves during 36 h of dialysis According to Hoffmann et al (1996) this pH decrease already induces depolymerization of k-carrageenan In addition, they assumed that the concentration of the sample during freeze-drying will further lower the pH at a local level which causes an increased depolymerization They further supposed that if an excess of sulfate-bound cations, such as potassium, is present during freeze-drying, there is no acid form of the k-carrageenan molecule and thus additional depolymerization does not occur This assumption of Hoffmann et al (1996) is in agreement with our results that both the molecular weight and the IRI activity are unaffected when the untreated k-carrageenan was dissolved in demineralized water and freeze-dried with its initial cations (data not shown) Therefore, the molecular weight reduction of the kcarrageenan dialyzed against demineralized water and KCl is most likely derived from an autohydrolysis during the dialysis, whereas additional molecular weight reduction of the k-carrageenan dialyzed against demineralized water is mostly derived from an increased autohydrolysis during freeze-drying of the acid k-carrageenan form In Fig 6, the mean ice crystal diameters in frozen sucrose solutions with the different dialyzed k-carrageenan samples after h and 96 h of storage time are depicted Larger mean ice crystal diameters and thus a reduced IRI activity were observed for samples with a lower molecular weight The small reduction of the molecular weight of the k-carrageenan dialyzed against demineralized water and KCl leads to a small increase in the mean ice crystal diameter after 96 h compared to the untreated k-carrageenan However, this difference is not statistically significant In contrast, the mean ice crystal diameters in the k-carrageenan solution dialyzed against demineralized water are significantly larger than in the solution containing untreated k-carrageenan extract To further investigate whether smaller k-carrageenan molecules exhibit a reduced IRI activity, we hydrolyzed the dialyzed k-carrageenan with 0.1 M TFA for 60 to further reduce the molecular weight For this sample, the reduction of molecular weight leads to a total loss of IRI activity as there is no significant difference between the mean ice crystal diameters in a pure sucrose solution and in a solution containing the dialyzed and subsequently hydrolyzed k-carrageenan sample 32 A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 Fig Mean ice crystal diameters in frozen sucrose solutions with different depolymerized k-carrageenan samples (wsuc ¼ 49% (w/w), ck-carr ¼ mg mLÀ1) after h and 96 h of storage time at À12 C Molecular weight was reduced by dialyzing k-carrageenan against demineralized water and by hydrolyzing the dialyzed k-carrageenan with 0.1 M trifluoroacetic acid Bars with the same small letter at the same storage time are not significantly different (P > 0.05) Thus, these results are in good agreement with the results of the 0.1 M TFA hydrolysis Both experiments show that the reduction of molecular weight leads to a decrease in IRI activity This relationship between molecular weight and IRI activity was also found for polyvinyl alcohol (PVA) and for the ice binding protein AFGP (Budke et al., 2014; Congdon et al., 2013; Inada and Lu, 2003) Both PVA and AFGP exhibit strong IRI activity and bind to the ice crystal surface Budke et al (2014) postulated that, firstly, an increased molecular weight and thus an increased molecular size may result in a larger surface area covered per molecule and, secondly, larger molecules exhibit more potential moieties for the adsorption to the ice, which may lead to a faster and/or stronger adsorption Assuming that IRI activity of k-carrageenan also originates from an interaction with the ice crystal surface, this explanation is also conceivable for the influence of the molecular weight of k-carrageenan on its IRI activity However, these findings are ska-Dwo rznicka et al (2015) who contrary to the study of Kamin showed that k-carrageenans hydrolyzed by sulfuric and hydrochloric acid have an increased IRI activity Possibly, hydrolysis with sulfuric and hydrochloric acid results in a different molecular structure of degraded k-carrageenan with an increased IRI activity To evaluate this hypothesis, the effect of k-carrageenan hydrolysis with HCl on the IRI activity was also studied However, Naỵ cations from NaOH, which was used to neutralize the acid solution, can lead to aggregation It was already demonstrated that aggregation of k-carrageenan molecules decreases IRI activity (Leiter et al., 2016a) Therefore, aggregation was additionally analyzed by rheological temperature sweep measurements Fig Size-exclusion chromatograms of untreated k-carrageenan extract and kcarrageenan that was hydrolyzed with 0.1 M HCl for 30 min, neutralized with NaOH, and freeze-dried The signal intensities are arbitrary, but the same scale is used for the refractive index signals of all curves molecules (Leiter et al., 2016a, 2017) Therefore, the same ion concentration has to be present in the hydrolyzed k-carrageenan and the untreated k-carrageenan to allow a comparison of the IRI activity For this reason, we made a reference sample by adding the same NaCl amount that is present in the hydrolyzed k-carrageenan sample to the untreated k-carrageenan extract and freeze-dried this sample without hydrolysis In the following, this reference sample is termed sample A and the hydrolyzed k-carrageenan sample is termed sample B The ion content in sample A and sample B is equal and the influence of ions on IRI activity is the same in both samples Thereby, an investigation on the influence of the molecular weight on IRI activity is possible To investigate the influence of hydrolyzed k-carrageenan on IRI activity, freeze-dried samples A and B were dissolved in 49% sucrose solutions to obtain the same concentration (1 mg mLÀ1) as in the previous experiments However, due to the high amount of ions, the final k-carrageenan concentration in these sucrose solutions is only 0.146 mg mLÀ1 Therefore, sucrose solutions with a concentration of 6.844 mg mLÀ1 of sample A or B leading to a final k-carrageenan concentration of mg mLÀ1 were also investigated The final salt and k-carrageenan concentrations in the different sucrose solutions with sample A or B are listed in Table For solutions with a lower sample concentration (1 mg mLÀ1), the hydrolyzed k-carrageenan sample (sample B) shows slightly higher mean ice crystal diameters than k-carrageenan with NaCl (sample A) after h and 96 h (Fig 8a) However, the difference is only significant at h of storage time For higher sample concentrations (6.844 mg mLÀ1), no significant differences between the mean ice crystal diameters were observed for samples A and B 3.4 Molecular weight distribution and IRI activity of k-carrageenan after acid hydrolysis with 0.1 M HCl The hydrolysis with 0.1 M HCl leads to a significant reduction of molecular weight (Fig 7) The main peak is shifted from 1420 kDa to 262 kDa, and the second peak at a retention time of 370 strongly increases This strong increase is mainly due to the NaCl content in the freeze-dried sample, which originates from the HCl used for hydrolysis and the NaOH used for neutralization In our previous studies, we showed that the presence of Naỵ cations reduces IRI activity of k-carrageenan due to aggregation of the Table NaCl and k-carrageenan concentrations for samples A and B dissolved in 49% sucrose solution Concentration of sample A or B [mg mLÀ1] Final mass concentration of k-carrageenan [mg mLÀ1] Final mass concentration of NaCl [mg mLÀ1] Final molar concentration of NaCl [mmol LÀ1] 6.844 0.146 0.854 5.844 14.6 100 A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 33 Fig Mean ice crystal diameters in frozen sucrose solutions (wsuc ¼ 49% (w/w)) with untreated k-carrageenan extract, untreated k-carrageenan with NaCl (sample A), or kcarrageenan first hydrolyzed with 0.1 M HCl and then neutralized with NaOH (sample B) after h and 96 h of storage time at À12 C The same NaCl contents are present in sample A and B (a) ck-carr ¼ 0.146 mg mLÀ1, csample A or B ¼ mg mLÀ1 (b) ck-carr ¼ mg mLÀ1, csample A or B ¼ 6.844 mg mLÀ1 Bars with the same small letter at the same storage time are not significantly different (P > 0.05) (Fig 8b) Thus, hydrolysis of k-carrageenan with 0.1 M HCl at 60 C for 30 has no significant effect on its IRI activity However, the increase in concentration of sample A and B from mg mLÀ1 to 6.844 mg mLÀ1 and thus an increase in the k-carrageenan concentration from 0.146 mg mLÀ1 to mg mLÀ1 (see Table 1) leads to a significant increase in mean ice crystal diameters The mean ice crystal diameters in solutions with a sample concentration of 6.844 mg mLÀ1 are about 40 mm (Fig 8b) and are similar to the mean diameter of ice crystals in pure sucrose solution which is about 46 mm (see Figs and 6) Thus, there is hardly any IRI activity left in the solutions with large sample concentrations of sample A or B, independently whether or not k-carrageenan is hydrolyzed This result is rather surprising because increasing kcarrageenan concentrations in sucrose solutions usually results in increasing IRI activities (Gaukel et al., 2014) This was also observed for the untreated k-carrageenan solution, which exhibits a smaller mean ice crystal diameter at a higher concentration For example, the mean ice crystal diameter after 96 h storage time decreases from 16 mm (Fig 8a) to 10 mm (Fig 8b) when the concentration of the untreated k-carrageenan increases from 0.146 mg mLÀ1 to mg mLÀ1 in pure sucrose solutions Therefore, it is likely that the large salt concentration of 100 mM is responsible for the decreased IRI activity due to aggregation of k-carrageenan molecules To prove this hypothesis, a temperature sweep analysis was performed (Fig 9) It is evident that increasing the concentration of sample A and B from mg mLÀ1 to 6.844 mg mLÀ1 leads to an increase of aggregation In a sucrose solution with a low concentration of sample A (1 mg mLÀ1) there is a small thermal hysteresis and thus aggregation is detectable In this solution the NaCl concentration is 14.6 mM (see Table 1) In addition, the coil-helix transition is at about 10 C By increasing the concentration of sample A to 6.844 mg mLÀ1 and thus the NaCl concentration to 100 mM, the thermal hysteresis is significantly higher Consequently, the degree of aggregation increases Additionally, the coilhelix transition is shifted from 10 C to 30 C This is in agreement with previous studies showing that coil-helix transition is shifted to higher temperature with increasing cation concentration (Doyle et al., 2012; Rochas and Rinaudo, 1980) In a sucrose solution containing the hydrolyzed sample B at low concentration (1 mg mLÀ1) neither aggregation nor coil-helix transition are detectable in accordance with the results of the TFA hydrolysis shown in Fig However, by increasing the concentration of the hydrolyzed sample B to 6.844 mg mLÀ1 a large thermal hysteresis is detectable The lower coil-helix transition temperature of 20 C compared to the transition temperature of 30 C of sample A is in agreement with the results of Meunier et al (2001) who showed that coil-helix transition temperature decreases with decreasing molecular weight Finally, it is apparent that in both sucrose solutions with higher sample concentrations and thus higher NaCl concentrations the degree of aggregation is increased Thus, these results confirm the hypothesis that the reduced IRI activity at higher sample concentrations is due to a higher degree of aggregation 3.5 Influence of NaCl concentration on IRI activity Surprisingly, the IRI activity of sample A and B at low concentrations is significantly increased relative to untreated k-carrageenan (Fig 8a) For example, after 96 h the mean ice crystal diameters in sucrose solutions with low concentrated sample A or B Fig Temperature sweep analysis (2 C minÀ1) of freeze-dried sample of untreated kcarrageenan extract with NaCl (sample A) and k-carrageenan hydrolyzed with 0.1 M HCl for 30 and subsequent neutralization with NaOH (sample B) dissolved in 49% (w/w) sucrose solution (csample ¼ mg mL-1 or 6.844 mg mLÀ1) Both freeze-dried samples have the same mass ratio of NaCl to k-carrageenan 34 A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 skaPossibly, this explains the contrary results of Kamin rznicka et al (2015) In this study, the IRI activity of a Dwo neutralized untreated k-carrageenan sample was compared with a neutralized hydrolyzed k-carrageenan sample in sucrose solution However, in their study different concentrations of the untreated and hydrolyzed samples and thus different salt concentrations were used for recrystallization experiments (0.01% of the neutralized untreated k-carrageenan sample and 0.005% of the neutralized hydrolyzed k-carrageenan sample) Thus, the higher IRI activity was possibly derived from a lower degree of aggregation in the solution with hydrolyzed k-carrageenan due to the lower salt concentration and not from a hydrolysis of the k-carrageenan molecules Conclusion Fig 10 Mean ice crystal diameters after a storage time of 96 h at À12 C Influence of molar NaCl concentration on ice recrystallization inhibition activity of k-carrageenan dissolved in a sucrose solution (wsuc ¼ 49% (w/w), ck-carr ¼ mg mLÀ1) are significantly smaller than ice crystal diameters in sucrose solution with the same amount of untreated k-carrageenan (0.146 mg mLÀ1) without additional salt The mean ice crystal diameters in these solutions with sample A and B are nearly of the same size as ice crystal diameters in sucrose solution with a higher k-carrageenan concentration of mg mLÀ1 (Fig 8b) Possibly, the addition of a small amount of NaCl, which induces either no or only a low degree of aggregation, leads to an increased IRI activity of kcarrageenan For example, we showed in a previous study that the addition of NaCl up to a limiting value between 20 and 30 mM to a solution with the ice binding protein AFP III leads also to an increased IRI activity (Leiter et al., 2016b) To further verify the assumption that a small amount of salt increases the IRI activity of k-carrageenan, we added the same small amount of NaCl (14.6 mM) to the sucrose solution with the higher k-carrageenan concentration of mg mLÀ1 In Fig 10, the influence of the molar NaCl concentration on the mean ice crystal diameter in k-carrageenan sucrose solution after 96 h storage time is shown The mean ice crystal diameter in the kcarrageenan sucrose solution with a final NaCl concentration of 14.6 mM is significantly smaller than in pure k-carrageenan sucrose solution (P < 0.05) Whereas at higher NaCl concentrations the IRI activity is reduced due to a higher degree of aggregation as expected However, the addition of a very small NaCl amount (0.3 mM) does not influence IRI activity, which is in good agreement with the observation that the addition of 0.3 mM KCl also does not influence the IRI activity (Leiter et al., 2016a) Thus, there is probably only a small concentration range where IRI activity is increased However, the reason for this remains unclear and further investigations are necessary The results of Figs and 10 show that the salt concentration can influence the IRI activity in different ways A high salt concentration leads to aggregation and reduces IRI activity whereas a low salt concentration can improve the IRI activity Hence, the addition of salt (due to the utilization of a buffer system or the necessity of neutralization) should ideally be avoided for investigations on the influence of the molecular weight or the molecular structure of kcarrageenan on its IRI activity However, if this is not possible, recrystallization experiments should be performed at least at the same salt concentration In addition, as IRI activity is lost at a high degree of aggregation, salt concentrations should be selected carefully so that no or only a low degree of aggregation occurs In this study, the influence of the molecular weight of k-carrageenan on its IRI activity was investigated The molecular weight was reduced by acid hydrolysis or by dialysis against demineralized water Dialysis and subsequent freeze-drying of the acid form of kcarrageenan led to a significant reduction of the molecular weight of k-carrageenan whereas higher molecular weights were observed when sulfate-bound cations were present during freeze-drying It was demonstrated that IRI activity of k-carrageenan decreases with decreasing molecular weight Assuming that IRI activity of kcarrageenan originates from an interaction with the ice crystal surface, we suppose that larger molecules probably exhibit more potential moieties for an interaction with the ice crystal surface, which may lead to a faster and/or stronger adsorption Furthermore, we showed that the addition of NaCl can influence the IRI activity in different ways A high salt concentration leads to an aggregation of the k-carrageenan and reduces IRI activity whereas a small salt concentration can improve the IRI activity However, the reason for this increase in IRI activity is unclear As salt is present in almost all food products, future research should focus on possibilities to avoid aggregation in the presence of salt In addition, more research is necessary to provide a better understanding of the IRI mechanism of k-carrageenan For this purpose, more detailed studies on the influence of ions on IRI activity are required Acknowledgement The authors would like to thank Welding GmbH & Co KG and Eurogum A/S for providing the k-carrageenan extract References Abad, L., Okabe, S., Shibayama, M., Kudo, H., Saiki, S., Aranilla, C., Relleve, L., La Rosa, A de, 2008 Comparative studies on the conformational change and aggregation behavior of irradiated carrageenans and agar by dynamic light scattering Int J Biol Macromol 42 (1), 55e61 Adapa, S., Schmidt, K.A., Jeon, I.J., Herald, T.J., Flores, R.A., 2000 Mechanisms of ice crystallization and recrystallization in ice cream: a review Food Rev Int 16 (3), 259e271 Bahramparvar, M., Mazaheri Tehrani, M., 2011 Application and functions of stabilizers in ice cream Food Rev Int 27 (4), 389e407 Budke, C., Dreyer, A., Jaeger, J., Gimpel, K., Berkemeier, T., Bonin, A.S., Nagel, L., Plattner, C., DeVries, A.L., Sewald, N., Koop, T., 2014 Quantitative efficacy classification of ice recrystallization inhibition agents Cryst Growth & Des 14 (9), 4285e4294 Budke, C., Koop, T., 2006 Ice recrystallization inhibition and molecular recognition of ice faces by poly(vinyl alcohol) Chemphyschem a Eur J Chem Phys Phys Chem (12), 2601e2606 Caldwell, K., Goff, H., Stanley, D., 1992 A low-temperature scanning electron microscopy study of ice cream II Influence of selected ingredients and processes Food Struct 11 (1) Campo, V.L., Kawano, D.F., Silva, Dílson Braz da, Carvalho, I., 2009 Carrageenans: biological properties, chemical modifications and structural analysis e a review Carbohydr Polym 77 (2), 167e180 Chun, J.-Y., Kim, J.-M., Min, S.-G., 2012 Effect of whey protein isolate on ice A Leiter et al / Journal of Food Engineering 209 (2017) 26e35 recrystallization characteristics in whey protein isolate/k-carrageenan matrix Korean J Food Sci Animal Resour 32 (5), 627e634 Ciancia, M., Milas, M., Rinaudo, M., 1997 On the specific role of coions and counterions on k-carrageenan conformation Int J Biol Macromol 20 (1), 35e41 Congdon, T., Notman, R., Gibson, M.I., 2013 Antifreeze (glyco)protein mimetic behavior of poly(vinyl alcohol): detailed structure ice recrystallization inhibition activity study Biomacromolecules 14 (5), 1578e1586 Cook, K., Hartel, R.W., 2010 Mechanisms of ice crystallization in ice cream production Compr Rev Food Sci Food Saf (2), 213e222 Davies, P.L., 2014 Ice-binding proteins: a remarkable diversity of structures for stopping and starting ice growth Trends Biochem Sci 39 (11), 548e555 Donhowe, D.P., Hartel, R.W., 1996 Recrystallization of ice in ice cream during controlled accelerated storage Int Dairy J (11e12), 1191e1208 Doyle, J., Giannouli, P., Philp, K., Morris, E.R., 2012 Effect of Kỵ and Ca2ỵ cations on gelation of k-carrageenan In: Williams, P.A., Phillips, G.O (Eds.), Gums and Stabilisers for the Food Industry, vol 16 RSC Publishing, Cambridge, pp 158e164 Eng Gaukel, V., 2004 Untersuchungen zum Einfluss von Antigefrierproteinen auf die €hrend der Gefrierlagerung, dargestellt an ModRekristallisation von Eis wa €sungen für Eiskrem GCA-Verlag, Herdecke elllo Gaukel, V., 2016 Cooling and freezing of foods In: Smithers, G.W (Ed.), Reference Module in Food Science Elsevier, Amsterdam Gaukel, V., Leiter, A., Spieß, W.E., 2014 Synergism of different fish antifreeze proteins and hydrocolloids on recrystallization inhibition of ice in sucrose solutions J Food Eng 141, 44e50 Goff, H.D., Ferdinando, D., Schorsch, C., 1999 Fluorescence microscopy to study galactomannan structure in frozen sucrose and milk protein solutions Food Hydrocoll 13 (4), 353e362 Hartel, R.W., 1996 Ice crystallization during the manufacture of ice cream Trends Food Sci Technol (10), 315e321 Hartel, R.W., 2001 Crystallization in Foods Aspen Publishers, Gaithersburg, Md Hoffmann, R.A., Russell, A.L., Gidley, M.J., 1996 Molecular weight distribution of carrageenans: characterisation of commercial stabilisers and effect of cation depletion on depolymerisation In: Phillips, G.O., Williams, P.A., Wedlock, D.J (Eds.), Gums and Stabilisers for the Food Industry, vol IRL Press at Oxford University Press, Oxford, pp 137e148 Inada, T., Lu, S.-S., 2003 Inhibition of recrystallization of ice grains by adsorption of poly(vinyl alcohol) onto ice surfaces Cryst Growth & Des (5), 747e752 ska-Dwo rznicka, A., Antczak, A., Samborska, K., Lenart, A., 2015 Acid hyKamin drolysis of kappa-carrageenan as a way of gaining new substances for freezing process modification and protection from excessive recrystallisation of ice Int J Food Sci Technol 50 (8), 1799e1806 Lahaye, M., 2001 Chemistry and physico-chemistry of phycocolloids Cah Biol Mar 1e2 Leiter, A., Emmer, P., Gaukel, V., 2016a Influence of gelation on ice recrystallization inhibition activity of k-carrageenan in sucrose solution Food Hydrocoll (in press) Leiter, A., Gaukel, V., 2016 Food freezing: crystal structure and size In: Smithers, G.W (Ed.), Reference Module in Food Science Elsevier, Amsterdam Leiter, A., Ludwig, A., Gaukel, V., 2017 Influence of heating temperature, pH and ions on recrystallization inhibition activity of k-carrageenan in sucrose solution J Food Eng 195, 14e20 Leiter, A., Rau, S., Winger, S., Muhle-Goll, C., Luy, B., Gaukel, V., 2016b Influence of heating temperature, pressure and pH on recrystallization inhibition activity of antifreeze protein type III J Food Eng 187, 53e61 Mangione, M., Giacomazza, D., Bulone, D., Martorana, V., San Biagio, P., 2003 Thermoreversible gelation of k-Carrageenan: relation between conformational transition and aggregation Biophys Chem 104 (1), 95e105 Marshall, R.T., Goff, H.D., Hartel, R.W., 2003 Ice Cream, sixth ed Kluwer Academic/ Plenum Publishers, New York Martin, D.R., Ablett, S., Darke, A., Sutton, R.L., Sahagian, M., 1999 Diffusion of aqueous sugar solutions as affected by locust bean gum studied by NMR J Food Sci 64 (1), 46e49 McHugh, D.J., 2003 A Guide to the Seaweed Industry Food and Agriculture 35 Organization of the United Nations, Rome Meunier, V., Nicolai, T., Durand, D., 2001 Structure of aggregating k-carrageenan fractions studied by light scattering Int J Biol Macromol 28 (2), 157e165 Miller-Livney, T., Hartel, R.W., 1997 Ice recrystallization in ice cream: interactions between sweeteners and stabilizers J Dairy Sci 80 (3), 447e456 Morris, E.R., Rees, D.A., Robinson, G., 1980 Cation-specific aggregation of carrageenan helices: domain model of polymer gel structure J Mol Biol 138 (2), 349e362 Muhr, A.H., Blanshard, J.M.V., 1986 Effect of polysaccharide stabilizers on the rate of growth of ice J Food Technol vol 21 (6), 683e710 Pham, Q.T., Mawson, F.R., 1997 Moisture migration and ice recrystallization in frozen foods In: Erickson, M.C., Hung, Y.-C (Eds.), Quality in Frozen Food Chapman & Hall, New York, pp 67e91 Piculell, L., 2006 Gelling carrageenans In: Stephen, A.M., Phillips, G.O., Williams, P.A (Eds.), Food Polysaccharides and Their Applications, second ed CRC/Taylor & Francis, Boca Raton, FL Prajapati, V.D., Maheriya, P.M., Jani, G.K., Solanki, H.K., 2014 Carrageenan: a natural seaweed polysaccharide and its applications Carbohydr Polym 105, 97e112 Regand, A., Goff, H.D., 2002 Effect of biopolymers on structure and ice recrystallization in dynamically frozen ice cream model systems J Dairy Sci 85 (11), 2722e2732 Regand, A., Goff, H.D., 2003 Structure and ice recrystallization in frozen stabilized ice cream model systems Food Hydrocoll 17 (1), 95e102 €gen und Gefriertemperaturen von Fruchts€ Riedel, L., 1949 Brechungsvermo aften in Abh€ angigkeit von der Konzentration Z für Leb eForschung 3, 289e299 Rinaudo, M., 2008 Main properties and current applications of some polysaccharides as biomaterials Polym Int 57 (3), 397e430 Rochas, C., Rinaudo, M., 1980 Activity coefficients of counterions and conformation in kappa-carrageenan systems Biopolymers 19 (9), 1675e1687 Rochas, C., Rinaudo, M., 1984 Mechanism of gel formation in k-carrageenan Biopolymers 23 (4), 735e745 Rochas, C., Rinaudo, M., Landry, S., 1990 Role of the molecular weight on the mechanical properties of kappa carrageenan gels Carbohydr Polym 12 (3), 255e266 Schefer, L., Adamcik, J., Diener, M., Mezzenga, R., 2015 Supramolecular chiral selfassembly and supercoiling behavior of carrageenans at varying salt conditions Nanoscale (39), 16182e16188 Smidsrød, O., Grasdalen, H., 1982 Some physical properties of carrageenan in solution and gel state Carbohydr Polym (4), 270e272 Stevenson, T.T., Furneaux, R.H., 1991 Chemical methods for the analysis of sulphated galactans from red algae Carbohydr Res 210, 277e298 Sun, Y., Yang, B., Wu, Y., Liu, Y., Gu, X., Zhang, H., Wang, C., Cao, H., Huang, L., Wang, Z., 2015 Structural characterization and antioxidant activities of kappacarrageenan oligosaccharides degraded by different methods Food Chem 178, 311e318 Sutton, R.L., Cooke, D., Russell, A., 1997 Recrystallization in sugar/stabilizer solutions as affected by molecular structure J Food Sci 62 (6), 1145e1149 Sutton, R.L., Lips, A., Piccirillo, G., 1996 Recrystallization in aqueous fructose solutions as affected by locust bean gum J Food Sci 61 (4), 746e748 Ueda, K., Itoh, M., Matsuzaki, Y., Ochiai, H., Imamura, A., 1998 Observation of the molecular weight change during the HelixÀCoil transition of k-carrageenan measured by the SECÀLALLS method Macromolecules 31 (3), 675e680 van de Velde, F., De Ruiter, G.A., 2006 Carrageenan In: Steinbüchel, A (Ed.), Biopolymers Online Wiley, New York van de Velde, F., Knutsen, S.H., Usov, A.I., Rollema, H.S., Cerezo, A.S., 2002 1H and 13C high resolution NMR spectroscopy of carrageenans: application in research and industry Trends Food Sci Technol 13 (3), 73e92 €m, J., Piculell, L., 1995 Characterisation of kappa- and iotaViebke, C., Borgstro carrageenan coils and helices by MALLS/GPC Carbohydr Polym 27 (2), 145e154 Yang, B., Yu, G., Zhao, X., Jiao, G., Ren, S., Chai, W., 2009 Mechanism of mild acid hydrolysis of galactan polysaccharides with highly ordered disaccharide repeats leading to a complete series of exclusively odd-numbered oligosaccharides FEBS J 276 (7), 2125e2137 ... (Bahramparvar and Mazaheri Tehrani, 2011; Goff et al., 1999; Regand and Goff, 2003) However, investigations on the relation of gelation and IRI activity of k-carrageenan showed that the formation of a k-carrageenan. .. k-carrageenan concentrations for samples A and B dissolved in 49% sucrose solution Concentration of sample A or B [mg mLÀ1] Final mass concentration of k-carrageenan [mg mLÀ1] Final mass concentration of. .. distribution and ice recrystallization inhibition (IRI) activity of k-carrageenan after acid hydrolysis with 0.1 M trifluoroacetic acid (TFA) SEC chromatograms of untreated and hydrolyzed k-carrageenan