Phase Transition and Polymorphic Behavior of ... - ACS Publications

Subscriber access provided by The University of British Columbia Library

Article

Phase transition and polymorphic behavior of binary systems containing fatty alcohols and peanut oil Fabio Valoppi, Sonia Calligaris, and Alejandro G. Marangoni Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.6b00145 • Publication Date (Web): 24 Jun 2016 Downloaded from http://pubs.acs.org on June 26, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Crystal Growth & Design is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

1

Phase transition and polymorphic behavior of binary systems containing fatty

2

alcohols and peanut oil

3

Fabio Valoppi†, Sonia Calligaris†, Alejandro G. Marangoni‡*

4

†

5

33100 Udine, Italy

6

‡

7

2W1

Dipartimento di Scienze Agroalimentari, Ambientali e Animali, Università di Udine, via Sondrio 2/a,

Department of Food Science, University of Guelph, 50 Stone Road East, Guelph, ON, Canada, N1G

8 9 10 11

Corresponding author

12

Alejandro G. Marangoni

13

Department of Food Science, University of Guelph, 50 Stone Road East, Guelph, ON, Canada, N1G

14

2W1

15

Phone: +1 519 824 4120 x 54340

16

Fax: +1 519 824 6631

17

E-mail: [email protected]

18 19 20 21 22 23

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

24

ABSTRACT: This study reports a detailed characterization of the phase transition behavior and

25

polymorphism of fatty alcohols with different chain lengths in peanut oil. Upon crystallization, fatty

26

alcohols nucleated into rotator phases, which then transformed into well-defined crystal polymorphs.

27

At concentrations greater than 30%, fatty alcohol crystals were in the monoclinic γ-form with a

28

lamellar thickness that decreased as the length of the carbon chain of the fatty alcohols decreased. At

29

concentrations lower than 30%, on the other hand, fatty alcohol crystals formed an orthorhombic β′-

30

form. In this case, two main crystal families with lamellar thicknesses were detected. In particular, the

31

thicker family range was from 5.95 to 4.96 nm moving from 1-docosanol to 1-hexadecanol while the

32

thinner family range was from 4.98 to 3.68 nm. The thicker crystal population progressively decreased,

33

while the thinner crystal population increased, suggesting an interconversion between these species.

34

The kinetics of these changes increased crystallizing the systems at 20 °C.. Ultra-small X-ray scattering

35

was used to characterize nanoscale structure and mesoscale crystal aggregation. It was found that

36

crystal aggregates were characterized by a diffuse surface with crystallites in the range 125-765 nm and

37

a mass fractal dimension for crystals aggregation of 2.26-2.7.


Page 3 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


38

1. INTRODUCTION

39

Fatty alcohols are amphiphilic molecules with an hydroxyl group attached to an aliphatic chain.1 The

40

presence of the polar group imparts the ability to self-assemble into different forms that have been

41

proposed for a variety of applications in different fields as (i) emulsifiers in biodiesel fuels,2,3 (ii) new

42

thermal energy storage materials,4,5 or (iii) hydrogenated and saturated fat replacers in foods.6-8

43

The number and type of structures that fatty alcohols could form depend on fatty alcohol chain length,

44

their concentration, solvent type as well as temperature and processing conditions.9 Thermal phase

45

transition and polymorphic behavior of pure fatty alcohols has been previously reported.5,10 It was

46

found that fatty alcohols have a lamellar crystalline structure with two possible polymorphs: β, γ.5,11

47

During cooling, another phase characterized by different polymorphic forms emerges from the melt:

48

the rotator-phase.5,11,12 The rotator phase is a solid state structure, in which the molecules are able to

49

rotate around their long axis.5,13 Further cooling induces the formation of monoclinic phases. The

50

crystal polymorph mainly depends on the number of carbon in the carbon chain: odd fatty alcohols

51

form β-phases, whereas even fatty alcohols structure themselves into γ-phases. The latter is packed

52

with a greater tilt angle than β-phase.5

53

Beside this information, very little and fragmented information can be found in the literature on phase

54

transition behavior of fatty alcohols in hydrophobic solvents, especially in food grade materials (e.g.

55

vegetable oils). Gandolfo et al.6 studied the capacity of mixtures of fatty alcohols and fatty acids in the

56

chain length range of 22-30 carbons to gel sunflower, soybean and rapeseed oil upon crystallization.

57

Moreover, Lupi et al.14 studied the structuring capacity of policosanol – a mixture of fatty primary

58

alcohols characterized by an aliphatic chain in the range of 20-36 carbon – in virgin olive oil.

59

According to Co and Marangoni15, fatty alcohol gels are formed by large crystals, probably arranged as

60

stacks of platelets. If fatty alcohols and fatty acids are mixed in a particular ratio, small needle-like

61

crystals are observed, and they are, maybe, responsible for the rheology of the final systems.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 27

62

Information found in the literature did not include the phase transition behavior of different fatty

63

alcohols in oil as affected by carbon chain length.

64

The aim of this work was to study the phase transition behavior of binary systems containing different

65

concentrations of fatty alcohols (1-docosanol, 1-eicosanol, 1-octadecanol, 1-hexadecanol, and 1-

66

tetradecanol) in peanut oil. Peanut oil was chosen as an example of edible oil widely used in different

67

food applications, containing 37-61% oleic acid, 24-43% linoleic acid, and around 10-14% saturated

68

fatty acids.16 Differential scanning calorimetry was applied to study the thermal events occurring

69

during cooling and further heating, whereas XRD analysis was employed to analyze polymorphism of

70

the systems upon static crystallization at 20 °C or

71

scattering was also applied as a novel non-destructive technique to study the characteristics of the fatty

72

alcohol aggregates in the matrix. USAXS is a new and powerful non-destructive analytical technique

73

able to characterize crystalline structures from the nanoscopic to microscopic level.17 With USAXS, it

74

is possible to investigate a range of length scale that are not accessible using classical Wide and Small-

75

angle X-ray scattering. Crystalline aggregates from 10 nm to more than 5 µm can be characterized.17

76

The application of USAXS to edible fats was introduced by Pink et al.18, Peyronel et al.19, and

77

Peyronel et al.20 These authors used computer simulations to predict and validate experimental USAXS

78

patterns on tristearin or tripalmitin in triolein at different concentrations.

79

2. EXPERIMENTAL SECTION

80

2.1. Materials. 1-dodecanol (C12OH, purity ≥ 98%), 1-tetradecanol (C14OH, purity ≥ 97.0%), 1-

81

eicosanol (C20OH, purity ≥ 98.0%), and 1-docosanol (C22OH, purity ≥ 98.0%), were purchased from

82

Sigma-Aldrich (Oakville, Canada). 1-hexadecanol (C16OH, purity ≥ 99.5%) and 1-octadecanol (C18OH,

83

purity ≥ 95.0%) were purchased from Fluka (Oakville, Canada) and Acros Organics (Ottawa, Canada),

84

respectively. Peanut oil was purchased in a local market.

7 °C and during storage. Ultra-small X-ray


Page 5 of 27

1 2 3 85 4 5 6 86 7 8 87 9 10 88 11 12 13 89 14 15 90 16 17 91 18 19 20 92 21 22 93 23 24 25 94 26 27 95 28 29 96 30 31 32 97 33 34 98 35 36 99 37 38 39 100 40 41 101 42 43 44 102 45 46 47 103 48 49 104 50 51 52 105 53 54 106 55 56 107 57 58 59 108 60


2.2. Sample preparation. Binary systems were prepared by mixing different concentrations of fatty alcohols in peanut oil. The mixtures were heated at 80 °C in an oven for at least 10 min until crystalline material was completely melted. Samples were then mixed for 3 – 5 min using a hot plate magnetic stirrer at 80 °C. The molten system was then transferred into containers for further analysis. All concentrations are expressed as mass percentage (% w/w). 2.3. Differential scanning calorimetry (DSC). DSC analysis was carried out using a Mettler Thermal Analysis DSC 1 (Mettler-Toledo, Inc., Mississauga, Canada). Heat flow and temperature calibration were achieved using indium (heat of fusion 28.45 J/g, m.p. 156.6 °C). Binary systems were prepared in according to the sample preparation section. Samples were cooled at room temperature and 6 – 8 mg of sample were carefully weighted in 40 µL aluminum DSC pans, closed with hermetic sealing. Samples were heated at 80 °C for 10 min, cooled from 80 to 5 °C at 5 °C/min, held at 5 °C for 10 min, and heated from 5 to 80 °C at 5 °C/min under nitrogen flow (20 mL/min). An empty pan was used as a reference in the DSC cell. The peak melting temperature (Tm) was taken as the minimum value of heat flow during transition. Total peak enthalpy was obtained by integration of the melting curve. The machine equipment program STARe ver. 12.10a (Mettler-Toledo, Inc., Mississauga, Canada) was used to plot and analyze the thermal data. Finally, Hildebrand equation (eq. 1) was used to model thermal data. ln ‫= ݔ‬

߂‫݂ܪ‬ ܴ

1

1

൬ܶ − ܶ ൰ ݂

(eq. 1)

݉

where x, ∆Hf, and Tf are the molar fraction or the weight fraction, melting enthalpy (J/mol), and the melting temperature (K) of the crystalline compound, respectively. R is the universal gas constant (8.314 J/mol K), and Tm is the melting temperature of the binary mixture. 2.4. Powder X-ray diffraction (XRD). Molten binary systems prepared as described in section 2.2 were poured in a glass sample holder with an area of 20×20 mm and depth of 1 mm. Samples were then statically crystallized at 7 or 22 °C (ambient temperature) from the melt. In both cases, samples were ACS Paragon Plus Environment


1 2 3 109 4 5 6 110 7 8 111 9 10 112 11 12 13 113 14 15 114 16 17 115 18 19 20 116 21 22 117 23 24 25 118 26 27 119 28 29 120 30 31 32 121 33 34 122 35 36 123 37 38 39 124 40 41 125 42 43 44 126 45 46 127 47 48 128 49 50 51 129 52 53 130 54 55 131 56 57 58 132 59 60

Page 6 of 27

hold at the selected temperature for at least 10 min to allow fatty alcohols crystallization. After cooling, sample excess was scraped from the glass sample holder using the edge of a spatula. Binary systems containing 5% fatty alcohols were stored at 25 °C and analyzed after 2 h, 1, 7, 14, 30, 54 days using a Rigaku Multiflex X-ray Diffractometer (RigakuMSC Inc., The Woodlands, USA), while samples containing 100, 70, and 30% fatty alcohols were analyzed after 1 hour of equilibration at 20 °C. Pure fatty alcohols were also analyzed after cooling the samples in situ from hot melt to a fixed temperature. X-rays were generated setting the copper source at 44 kV and 40 mA resulting in 1.54 Å X-ray wavelength. The setting for the divergence slit, scattering slit, and receiving slit, were 0.5°, 0.5°, and 0.3 mm, respectively. Temperature of samples during data collection were controlled using a water bath (Isotemp mod. 6200, Fisher-Scientific, Inc., Pittsburgh, USA) connected to a flat surface heat exchanger placed under the glass sample holder and a thermocouple placed underneath the glass sample holder was used to monitor the temperature during the entire experiment. Samples were scanned from 2θ equal to 1.1 to 35.0° at a scanning speed of 1° per min. Resulting diffraction patterns were processed and analyzed using MDI Jade v. 9.0.1 (Materials Data Inc., Livermore, CA, USA). In particular, a spline cubic baseline was mounted on diffraction pattern as background line and peaks were determined using a profile shape function (i. e. Pearson-VII, pseudo-Voigt, Gaussian, and Lorentzian) that allowed residual error of fit to be minimized. 2.5. Ultra-small angle X-ray scattering (USAXS). Molten binary systems were poured in a sample holder composed by a glass cover slip attached to a 1 mm thick press-to-seal silicon isolator with holes 8-9 mm in diameter (Grace Bio-Labs, Bend, USA). After the samples were statically crystallized at ambient temperature (22 °C), another glass cover slip was attached to the silicon isolator to avoid sample structural damaging. 1D-collimeted USAXS measurements were carried out at beam-line 9 ID at the Advanced Photon Source (APS), (Argonne National Laboratory, Illinois, USA). A Bonse-Hart instrument was used to


Page 7 of 27

1 2 3 133 4 5 6 134 7 8 135 9 10 136 11 12 13 137 14 15 138 16 17 139 18 19 20 140 21 22 141 23 24 25 142 26 27 143 28 29 144 30 31 32 145 33 34 146 35 36 147 37 38 39 148 40 41 149 42 43 44 150 45 46 47 151 48 49 50 152 51 52 153 53 54 154 55 56 57 155 58 59 156 60


collimate the X-rays and to collect the data.21,22 The basic set up for edible fat samples was describe by Peyronel et al.20, which can be summarized as a pair of collimating crystals, followed by a ion chamber, the isolator-sample holder mounted on a plate, a pair of analyzer crystals, and a photodiode detector. During experiments at the APS data were acquired using the Spec software for diffraction experiments. Data reduction and analysis were performed using the Irena IgorPro-based software23 following the procedure proposed by Peyronel et al.20. In particular, the Unified Fit model24,25 was used, which employs a non-linear regression analysis. The analysis consists of the identification of different regions (levels) in the q-space, which are characterized by two parameters: (i) a parameter P related to the slope of the curve which gives information about the internal structure of the scatterer via a fractal dimension, D,18 and (ii) a parameter Rg, which identifies the average radius of gyration of the scatterers giving rise to the structure with the fractal dimension D. Edible fat systems had shown a maximum of 3 slopes and two radii of gyration for level 1 and level 2 when Unified Fit is used.26 Through models, computer simulations18,27 and experiments17,19, the D value has been shown to be a reliable parameter. On the other hand, the Rg is the exponent in the formula28 and it should be considered only as an indicator of sizes, rather than a precise value. Having this in mind, Rg was used to calculate the average radius (R) of the scatterers as spheres (eq. 2).19 ହ

ܴ = ටଷ ܴ௚

(eq. 2)

2.6. Data analysis. Samples were tested at least in triplicate. Data were processed using GraphPad Prism v. 5.03 (GraphPad Software, Inc., La Jolla, USA). Linear regression analysis by least squares regression was performed and the goodness of fit was evaluated on the basis of statistical parameters of fitting (R2, p-value, standard error) and the residual analysis.



1 2 3 157 4 5 6 158 7 8 159 9 10 160 11 12 13 161 14 15 162 16 17 163 18 19 20 164 21 22 165 23 24 25 166 26 27 167 28 29 168 30 31 32 169 33 34 170 35 36 171 37 38 39 172 40 41 173 42 43 44 174 45 46 175 47 48 176 49 50 51 177 52 53 178 54 55 179 56 57 58 180 59 60

Page 8 of 27

3. RESULTS AND DISCUSSION

Figure 1A shows the crystallization curves of binary systems containing peanut oil and different percentages of C20OH, chosen as an example among fatty alcohols considered in this study. Two exothermic peaks were recorded during cooling from 80 to 5 °C at 5 °C/min of the pure C20OH fatty alcohol (Figure 1B) as well as mixtures containing from 90 to 40% fatty alcohol concentration (Figure 1A). It can be noted that peak 1 shifted to low temperature as the concentration of fatty alcohol decreased; while no changes in peak temperature was noted for peak 2. Below 40% C20OH, the peak 1 disappeared or appeared as shoulder of peak 2. This behavior was observed for all other considered fatty alcohols (C22OH, C18OH, C16OH, and C14OH). The only difference recorded was the change of the critical percentage at which two separated peaks were detectable for C14OH samples. In this case, below 60% content of the fatty alcohol only peak 2 was recorded (Figure S1, supporting information). As shown in Figure 1B, the crystallization temperature of peak 1 and 2 progressively decreased as fatty alcohol chain length decreased. As reported in the literature, fatty alcohols with both odd and even carbon chain length show similar crystallization behavior than n-alkanes and the following transitions during cooling have been reported: liquid (L) rotator phase (R′) solid (S).1,4,5,12 Thus, cooling pure fatty alcohols as well as binary systems from melt, a rotator phase was formed between peak 1 and 2, further cooling induced the conversion of rotator phase to solid phase. To confirm the presence of the rotator phase, pure fatty alcohols were analyzed using XRD.

Figure 2A and B shows respectively the SAXS and WAXS patterns recorded for pure C22OH, C20OH, C18OH, C16OH, and C14OH, at temperature in between peak 1 and 2 (65.8, 58.2, 52.0, 44.1, and 33.3 °C, respectively). It was found that pure C14OH and C16OH showed a main peak in the WAXS region (Figure 2B)at 4.18 Å that could be ascribed to a pseudo-hexagonal phase. This rotator phase


Page 9 of 27

1 2 3 181 4 5 6 182 7 8 183 9 10 184 11 12 13 185 14 15 186 16 17 18 19 20 21 187 22 23 188 24 25 26 189 27 28 190 29 30 191 31 32 33 192 34 35 193 36 37 38 194 39 40 195 41 42 196 43 44 45 197 46 47 198 48 49 199 50 51 52 200 53 54 201 55 56 57 202 58 59 203 60


corresponds to R′II as reported by Metivaud et al.1 On the other hand, pure C18OH, C20OH, and C22OH showed three peaks in the WAXS region that were associated to a pseudo-monoclinic rotator phase namely called R′IV as reported by Ventolà et al.10 Regarding the SAXS region (Figure 2A), a main peak (i.e. 50.6 Å for C20OH) corresponding to the d001, followed by its higher order reflections was found in each pattern. It can be noted that the d001 values increased linearly as the fatty alcohol chain length also increased (R2 > 0.99; Figure S2, supporting information), in agreement with the following equation: 2 10 ݀ = ൬ ൰ ∗ ‫ ܮܥ‬+ ൬ ൰ ݈ ݈

where l is the Miller index, and CL is the fatty alcohol chain length. Thus, it can be concluded that fatty alcohols formed a double layer structure (also confirmed by the reflexions relative position ratio of 1/2, 1/3, 1/4, 1/5, and 1/6) with a distance of approximately 10 Å in the -OH HO- group. This information allowed to calculate the carbon chain tilting angle, which resulted to be approximately 58 – 59°.

The dissolution behavior of pure fatty alcohols as well as fatty alcohols binary systems was then studied. Figure 3A shows the melting curves of selected systems containing 100, 50, and 10% C20OH. One main melting peak was recorded in all cases. Other fatty alcohols showed similar behavior (data not shown). The melting peak temperature of samples as a function of fatty alcohol concentration in peanut oil was reported in Figure 3B. As expected, peak temperature decreased as fatty alcohol concentration decreased. In particular, it was found that all systems showed a decrease of circa 10 °C from 100 to 20% fatty alcohol concentration, whereas about 15 °C decrease of the peak melting temperature was observed in the region 10 to 2% fatty alcohol concentration. Finally, it can be noted that the peak melting temperature increased with fatty alcohol carbon chain length. This in agreement with behavior observed for triacylglycerols (TAGs), waxes, and saturated fatty acids, diluted in liquid oil.29-31 ACS Paragon Plus Environment


1 2 3 204 4 5 6 205 7 8 206 9 10 207 11 12 13 208 14 15 209 16 17 210 18 19 20 211 21 22 212 23 24 25 213 26 27 214 28 29 215 30 31 32 216 33 34 217 35 36 218 37 38 39 219 40 41 220 42 43 44 221 45 46 222 47 48 223 49 50 51 224 52 53 225 54 55 226 56 57 58 59 60

Page 10 of 27

The solubility of a crystalline compound in liquid oil can be modelled using the Hildebrand equation (eq. 1), as described by Timms.32 A straight line in a ln x (mol fraction) vs 1/Tm plot is observed, it can be hypothesized that the solubility of the crystalline compound in oil is ideal.29 Figure 3C shows the Hildebrand plot for the binary systems considered. It can be noted that the Hildebrand equation cannot be used to describe the melting temperature dependence of the fatty alcohol concentration, since deviations were observed below 10% fatty alcohol concentration, where non-ideal behavior prevailed. Thus, the dissolution of fatty alcohol in peanut oil did not follow an ideal behavior probably due to the interactions between the solvent and fatty alcohols or the presence of different polymorphic forms.

Figure 4 shows changes in melting enthalpy as a function of fatty alcohol concentration. As expected, the melting enthalpy increased linearly as the amount of alcohol in the binary system increased (R2 > 0.99 and p < 0.05). Moreover, shorter chain fatty alcohols showed a lower melting enthalpy comparted to longer fatty alcohols on all the concentration rage. When more -CH2-groups are present in fatty alcohol carbon chain, molecules can establish more interactions and thus more energy is required to destroy the structure. Figure 5A shows the WAXS and SAXS patterns of binary systems containing C20OH at different concentrations after crystallization from the melt to 7 °C. The WAXS region (2θ ≥ 15°) of pure C20OH as well as mixtures containing more than 30% fatty alcohol showed peaks at 4.28, 4.10, 3.97, and 3.63 Å characteristics of the monoclinic γ-form, as discussed by Ventolà et al.5 and Metivaud et al.1 In the SAXS region (2θ < 8°), the d001 was found at 45.6 Å, , followed by its higher order reflection peaks in relative position ratios of 1/2, 1/3, 1/4, suggesting a lamellar structure. Such lamellar structure is typical of long chain hydrocarbon-like molecules such as monoglycerides, waxes and triglycerides.30,33,34 It can be noted that the lamellar thickness remained constant moving from 100 to 30% C20OH concentration


Page 11 of 27

1 2 3 227 4 5 6 228 7 8 229 9 10 230 11 12 13 231 14 15 232 16 17 233 18 19 20 234 21 22 235 23 24 25 236 26 27 28 29 30 31 32 33 34 35 36 37 38 237 39 40 238 41 42 43 239 44 45 240 46 47 241 48 49 50 242 51 52 243 53 54 244 55 56 57 245 58 59 60


(Figure 5A). Using the same approach previously described, the calculated carbon chain tilting angle of these systems was between 60 – 65°. On the other hand, 5% C20OH containing system displayed peaks at 4.15 and 3.69 Å in the WAXS region characteristic of orthorhombic β′-form. In the SAXS region, two peaks at 45.6 Å and 54.1 Å were observed. These peaks suggested that during cooling two crystal families were formed. Similar results were also found for binary systems containing C22OH, C18OH, and C16OH (Figure S3, supporting information).

Table 1. d001 values for the two crystal families found in the SAXS region in binary systems containing 5% C22OH, C20OH, C18OH, and C16OH. Fatty alcohol Crystal family 1 Crystal family 2 (Å)

(Å)

C22OH

59.5

49.8

C20OH

54.1

45.6

C18OH

48.5

41.2

C16OH

43.6

36.8

The lamellar thickness of systems containing 5% C22OH, C20OH, C18OH, and C16OH are reported in Table 1. It can be observed that, as expected, the lamellar thickness decreased as the length of the carbon chain of the fatty alcohols decreased. The trend was linear (R2 > 0.99 and p < 0.05). It is interesting to note that by changing the cooling procedure (from melt to 7 or 22 °C), no differences in XRD results were found (Figure S4, supporting information). The calculated carbon chain tilting angle for these systems was around 46 – 50°. It is interesting to note that the d001 values for both crystal families reported in Table 2 are different from those found in rotator phases (Figure 2), suggesting the formation of different crystals. This result can explain DSC data previously described.



1 2 3 246 4 5 6 247 7 8 248 9 10 249 11 12 13 250 14 15 251 16 17 252 18 19 20 253 21 22 254 23 24 25 255 26 27 256 28 29 257 30 31 32 258 33 34 259 35 36 260 37 38 39 261 40 41 262 42 43 44 263 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 264 60

Page 12 of 27

The only exception to the described behavior were samples containing C14OH at different concentrations (Figure 5B). First of all, the 5% C14OH system did not form a semi-solid material and thus was not analyzed. Systems containing 100, 70 and 30% C14OH showed the presence of both γ- and β′-forms. Two lamellar thicknesses of 39.5 and 32.6 Å were detected. The thicker one can be associated to the β′-form since it has a smaller tilt angle respect to the γ-form as described by Ventolà et al.5 Samples containing 5% fatty alcohols were then stored at 25 °C for up to 54 days in order to determine polymorphic stability. During storage, no changes in crystal structure were detected. However, significant modifications of the peaks’ intensity were recorded: the intensity of crystal family 1 progressively decreased with a corresponding increase in the intensity of crystal family 2 (Figure S5, supporting information). These results could be due to a polymorphic conversion between the two crystal families. The crystals belonging to the unstable family (greater d001 value) converted into more stable crystals (smaller d001 value), as represented in Figure 6. To determine if the cooling rate and/or the fatty alcohol carbon chain length affected the rate of conversion from crystal family 1 to crystal family 2, the time at which the intensity of family 1 peak matched family 2 peak intensity (t1:1) was determined.

Table 2. Time at which the intensity of family 1 peak matched family 2 peak intensity (t1:1) of binary systems containing 5% of different fatty alcohols cooled at 7 °C and 22 °C (ambient temperature) . Fatty alcohol

t1:1 (days) Cooling at 7 °C Cooling at 22 °C

C22OH

> 54

< 0.1

C20OH

10

10.5

C18OH

0.14

0.16

C16OH

1

< 0.1


Page 13 of 27

1 2 3 265 4 5 6 266 7 8 267 9 10 268 11 12 13 269 14 15 270 16 17 271 18 19 20 272 21 22 273 23 24 25 274 26 27 275 28 29 276 30 31 32 277 33 34 278 35 36 279 37 38 39 280 40 41 281 42 43 44 282 45 46 283 47 48 284 49 50 51 285 52 53 286 54 55 287 56 57 58 59 60


Table 2 shows the t1:1 for binary systems containing 5% different fatty alcohols cooled at 7 °C or 22 °C (ambient temperature). It can be noted that both cooling methodologies and fatty alcohol chain length affected the t1:1. The longer the carbon chain length the longer the t1:1 for samples cooled at 7 °C. The only exception was that of the binary system containing C18OH. This could be attributed to the lower purity of C18OH compared to the other fatty alcohols used in the study. Comparing the two cooling procedures applied, t1:1 of binary systems containing C18OH and C20OH was not affected by cooling methodology. On the other hand, systems containing C16OH and C22OH highlighted the influence of the cooling method. Cooling the samples to ambient temperature allowed these systems to form crystals in the more stable form (lower lamellar thickness). Finally, ultra-small X-ray scattering (USAXS) was applied to characterize the supramolecular structures formed by fatty alcohols in binary systems. Figure 7 shows the USAXS data for binary systems containing 5% C22OH (A), C20OH (B), C18OH (C), and C16OH (D) cooled at room temperature. To interpret data, the Unified Fit model was applied. This model describes the scattering of systems containing different structures idealized approximately as spherical objects combining the Guinier’s law and the Porod power law. The first law is reflected as a knee in a log (normalized scattering intensity) - log q plot, whereas the second law is reflected as a straight line (Figure 7). Spherical objects are characterized by an average radius of gyration (Rg), which can be calculated using Guinier’s law, while information on the internal structure of these objects is given by the exponent P of the Porod power law. Different structural levels can thus be found in these systems as described by Peyronel et al.17 The structural level can be described by its Rg and P parameters. In the present study, two structural levels characterized by P1, Rg1, and P2 parameters were detected (Figure 7 and Table 3).



1 2 3 288 4 5 6 289 7 8 9 10 11 12 13 14 15 16 17 18 19 290 20 21 291 22 23 24 292 25 26 293 27 28 294 29 30 31 295 32 33 296 34 35 36 297 37 38 298 39 40 299 41 42 43 300 44 45 301 46 47 302 48 49 50 303 51 52 304 53 54 305 55 56 57 306 58 59 60

Page 14 of 27

Table 3. Unified Fit model parameters obtained for binary systems containing 5% fatty alcohol cooled at 22 °C (ambient temperature). Standar error of fitting is also reported. Fatty alcohol

|P1|

Rg1 (nm)

(-)

R

|P2|

(nm)

(-)

C22OH

4.35 ± 0.04

148 ± 4

191 ± 5

2.35 ± 0.01

C20OH

4.30 ± 0.08

97 ± 10

125 ± 13

2.26 ± 0.01

C18OH

4.19 ± 0.02

550 ± 18

710 ± 23

2.60 ± 0.10

C16OH

4.66 ± 0.08 593 ± 119 766 ± 154 2.70 ± 0.30

With regards the first structural level, all systems studied showed a |P1| value greater than 4. This means that the surfaces of the aggregating particles are diffuse and a not well defined edge between solid and liquid phase is present. Rg1 was then used to calculate the radii of the spherical objects (R). It can be noted that binary systems containing C22OH and C20OH showed similar radii. These spheres were smaller than those formed by C18OH and C16OH (Table 3). Finally, in the second structural level it was found that the |P2| values were between 2 and 3. According to Peyronel et al.20, when 1 ≤ |P| < 3, the P value indicates the mass fractal dimension of the aggregates. In this case, these spherical objects aggregate (probably via reaction limited cluster-cluster aggregation, RLCA) to form a disordered distribution of mass embedded in a 3D space. 4. CONCLUSIONS This work was a systematic investigation of the phase transition behavior of fatty alcohols with increasing chain length from 14 to 22 in peanut oil. It was established that upon cooling fatty alcohols formed firstly a metastable rotator phase followed by transformation to a stable crystal polymorph, whose characteristics depended on fatty alcohol concentration and chain length. All fatty alcohols at concentration higher than 30% were in the monoclinic γ-form with lamellar thickness that decreased as the length of the carbon chain of the fatty alcohols decreased. On the other hand, an orthorhombic β′-


Page 15 of 27

1 2 3 307 4 5 6 308 7 8 309 9 10 310 11 12 13 311 14 15 312 16 17 313 18 19 20 314 21 22 315 23 24 25 316 26 27 317 28 29 318 30 31 32 319 33 34 320 35 36 321 37 38 39 322 40 41 323 42 43 44 324 45 46 325 47 48 326 49 50 51 327 52 53 328 54 55 329 56 57 58 330 59 60


form was detected at 5% fatty alcohol concentration. In that case, two crystal families showing different lamellar thickness values were detected. During storage, the crystal family of greater lamellar thickness converted to the family with lower thickness. This behavior is typical of less stable polymorphs converting to more stable ones. These changes were affected by the cooling procedure used during crystallization. Finally, ultra-small X-ray scattering was used to characterize crystal aggregates in the matrix. It was found that these aggregates were characterized by diffuse surface with crystallites in the range 125-765 nm and a mass fractal dimension for crystals aggregation of 2.26-2.7. Further research will be needed to fully elucidate the role of concentration, processing condition, and storage time on crystal aggregates using the USAXS technique. Results acquired are useful to design organogels containing fatty alcohols and an edible oil.

ASSOCIATED CONTENT Supporting Information. Crystallization curves, regression on peak positions, XRD patterns, and crystal families evolution over time for binary systems containing fatty alcohols and peanut oil. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] Notes The authors declare no competing financial interest.



1 2 3 331 4 5 6 332 7 8 333 9 10 334 11 12 13 335 14 15 336 16 17 18 337 19 20 338 21 22 339 23 24 25 340 26 27 341 28 29 342 30 31 32 343 33 34 344 35 36 345 37 38 39 346 40 41 347 42 43 44 348 45 46 349 47 48 350 49 50 51 351 52 53 352 54 55 353 56 57 58 59 60

Page 16 of 27

ACKNOWLEDGMENTS This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357. Authors want to thank Dr. Jan Ilavsky and Dr. Fernanda Peyronel for USAXS measurements and data analysis.

REFERENCES (1)

Metivaud, V.; Lefevre, A.; Ventolà, L.; Negrier, P.; Moreno, E.; Calvet, T.; Mondieig,

D.; Cuevas-Diarte, M. A. Chem. Mater. 2005, 17, 3302-3310. (2)

Dunn, R. O.; Bagby, M. O. J. Am. Oil Chem. Soc. 1995, 72, 123-130.

(3)

Dunn, R. O.; Schwab, A. W.; Bagby, M. O. J. Dispersion Sci. Technol. 1993, 14, 1-16.

(4)

Ventolà, L.; Calvet, T.; Cuevas-Diarte, M. A.; Oonk, H. A. J.; Mondieig, D. Phys.

Chem. Chem. Phys. 2004, 6, 3726-3731. (5)

Ventolà, L.; Ramírez, M.; Calvet, T.; Solans, X.; Cuevas-Diarte, M. A.; Negrier, P.;

Mondieig, D.; van Miltenburg, J. C.; Oonk, H. A. J. Chem. Mater. 2002, 14, 508-517. (6)

Gandolfo, F. G.; Bot, A.; Flöter, E. J. Am. Oil Chem. Soc. 2004, 81, 1-6.

(7)

Daniel, J.; Rajasekharan, R. J. Am. Oil Chem. Soc. 2003, 80, 417-421.

(8)

Schaink, H. M.; van Malssen, K. F.; Morgado-Alves, S.; Kalnin, D.; van der Linden, E.

Food Res. Int. 2007, 40, 1185-1193. (9)

Hyde, S.; Andersson, S.; Larsson, K.; Blum, Z.; Landh, T.; Lidn, S.; Ninham, B. W. In

The Language of Shape, the Role of Curvature in Condensed Matter: Physics, Chemistry and Biology; Hyde, S.; Andersson, S.; Larsson, K.; Blum, Z.; Landh, T.; Lidn, S.; Ninham, B. W., Eds.; Elsevier Science B. V.: Amsterdam, 1997.


Page 17 of 27

1 2 3 354 4 5 6 355 7 8 356 9 10 357 11 12 13 358 14 15 359 16 17 360 18 19 20 361 21 22 362 23 24 25 363 26 27 364 28 29 365 30 31 32 366 33 34 367 35 36 368 37 38 39 369 40 41 370 42 43 44 371 45 46 372 47 48 373 49 50 51 374 52 53 375 54 55 376 56 57 58 59 60


(10)

Ventolà, L.; Calvet, T.; Cuevas-Diarte, M. A.; Ramirez, M.; Oonk, H. A. J.; Mondieig,

D.; Negrier, P. Phys. Chem. Chem. Phys. 2004, 6, 1786-1791. (11)

Sirota, E. B.; Wu, X. Z. J. Chem. Phys. 1996, 105, 7763-7773.

(12)

Yamamoto, T.; Nozaki, K.; Hara, T. J. Chem. Phys. 1990, 92, 631.

(13)

Sirota, E. B.; King, H. E.; Shao, H. H.; Singer, D. M. J. Phys. Chem. Us 1995, 99, 798-

(14)

Lupi, F. R.; Gabriele, D.; Baldino, N.; Mijovic, P.; Parisi, O. I.; Puoci, F. Food Funct.

804.

2013, 4, 1512-1520. (15)

Co, E. D.; Marangoni, A. G. J. Am. Oil. Chem. Soc. 2012, 89, 749-780.

(16)

Carrín, M. E.; Carelli, A. A. Eur. J. Lipid Sci. Tech. 2010, 112, 697-707.

(17)

Peyronel, F.; Ilavsky, J.; Pink, D. A.; Marangoni, A. G. 2014, 26, 223-226.

(18)

Pink, D. A.; Quinn, B.; Peyronel, F.; Marangoni, A. G. J. Appl. Phys. 2013, 114,

234901. (19)

Peyronel, F.; Ilavsky, J.; Mazzanti, G.; Marangoni, A. G.; Pink, D. A. J. Appl. Phys.

2013, 114, 234902. (20)

Peyronel, F.; Pink, D. A.; Marangoni, A. G. Curr. Opin. Colloid In. 2014, 19, 459-470.

(21)

Ilavsky, J.; Zhang, F.; Allen, A. J.; Levine, L. E.; Jemian, P. R.; Long, G. G. Metall.

Mater. Trans. A 2013, 44A, 68-76. (22)

Long, G. G.; Jemian, P. R.; Weertman, J. R.; Black, D. R.; Burdette, H. E.; Spal, R. J.

Appl. Crystallogr. 1991, 24, 30-37. (23)

Ilavsky, J.; Jemian, P. R. J. Appl. Crystallogr. 2009, 42, 347-353.

(24)

Beaucage, G. J. Appl. Crystallogr. 1995, 28, 717-728.

(25)

Beaucage, G. J. Appl. Crystallogr. 1996, 29, 134-146.



1 2 3 377 4 5 6 378 7 8 379 9 10 380 11 12 13 381 14 15 382 16 17 383 18 19 20 384 21 22 385 23 24 25 386 26 27 387 28 29 388 30 31 32 389 33 34 390 35 36 391 37 38 39 40 392 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(26)

Page 18 of 27

Peyronel, F.; Quinn, B.; Marangoni, A. G.; Pink, D. A. J. Phy. Conden. Matter 2014, 26,

464110. (27)

Quinn, B.; Peyronel, F.; Gordon, T.; Marangoni, A.; Hanna, C. B.; Pink, D. A. J. Phy.

Conden. Matter 2014, 26, 464108. (28)

Beaucage, G.; Schaefer, D. W. J. Non-Cryst. Solids 1994, 172-174, 797-805.

(29)

Ahmadi, L.; Wright, A. J.; Marangoni, A. G. Eur. J. Lipid Sci. Tech. 2008, 110, 1014-

(30)

Blake, A. I.; Co, E. D.; Marangoni, A. G. J. Am. Oil Chem. Soc. 2014, 91, 885-903.

(31)

Inoue, T.; Hisatsugu, Y.; Yamamoto, R.; Suzuki, M. Chem. Phys. Lipids 2004, 127, 143-

(32)

Timms, R. E. Prog. Lipid Res. 1984, 23, 1-38.

(33)

Da Pieve, S.; Calligaris, S.; Co, E.; Nicoli, M. C.; Marangoni, A. G. Food Biophys.

1024.

152.

2010, 5, 211-217. (34)

Acevedo, N. C.; Peyronel, F.; Marangoni, A. G. Curr. Opin. Colloid In. 2011, 16, 374-

383.


Page 19 of 27

1 2 3 393 4 5 6 394 7 8 395 9 10 11 12 13 14 15 16 17 18 396 19 20 397 21 22 398 23 24 399 25 26 27 400 28 29 401 30 31 32 402 33 34 403 35 36 404 37 38 39 405 40 41 406 42 43 407 44 45 46 408 47 48 409 49 50 410 51 52 53 411 54 55 412 56 57 58 413 59 60


Table of Contents Graphic

Synopsis Thermal and polymorphic behavior of fatty alcohols mixed with peanut oil was studied. Fatty alcohols in binary systems nucleated into rotator phases, which then transformed into monoclinic γ-form for fatty alcohol concentrations greater than 30%, or orthorhombic β′-form for fatty alcohol concentration lower than 30%. The orthorhombic crystals showed two crystalline families and their ratio changed during storage. For Table of Contents Use Only Title: Phase transition and polymorphic behavior of binary systems containing fatty alcohols and peanut oil Authors: Fabio Valoppi, Sonia Calligaris, Alejandro G. Marangoni*



1 2 3 414 4 5 6 415 7 8 416 9 10 417 11 12 13 418 14 15 419 16 17 420 18 19 20 421 21 22 422 23 24 25 423 26 27 424 28 29 425 30 31 32 426 33 34 427 35 36 428 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 27

FIGURE LEGENDS

Figure 1. Crystallization curves of binary systems containing different amounts of C20OH (A) and different pure fatty alcohols (B) obtained at 5 °C/min. Figure 2. XRD patterns recorded for C22OH, C20OH, C18OH, C16OH, and C14OH at 65.8, 58.2, 52.0, 44.1, and 33.3 °C, respectively. Figure 3. Melting curves of systems containing 100, 50, and 10% C20OH (A) obtained at 5 °C/min, peak melting temperature as a function of concentration of fatty alcohols in peanut oil for all binary systems considered (B), and Hildebrand plot for binary systems (C). Figure 4. Melting enthalpy (∆Hm) as a function of fatty alcohol concentration. Figure 5. XRD patterns for binary systems containing C20OH (A) and C14OH (B) at different concentrations statically cooled at 7 °C. Figure 6. Schematization of possible fatty alcohol binary system crystals modifications during storage. Figure 7. USAXS data for binary systems containing 5% of C22OH (A), C20OH (B), C18OH (C), and C16OH (D) statically cooled at 22 °C (ambient temperature).


Page 21 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Crystallization curves of binary systems containing different amounts of C20OH (A) and different pure fatty alcohols (B) obtained at 5 °C/min. 65x25mm (600 x 600 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. XRD patterns recorded for C22OH, C20OH, C18OH, C16OH, and C14OH at 65.8, 58.2, 52.0, 44.1, and 33.3 °C, respectively. 64x23mm (600 x 600 DPI)


Page 22 of 27

Page 23 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. Melting curves of systems containing 100, 50, and 10% C20OH (A) obtained at 5 °C/min, peak melting temperature as a function of concentration of fatty alcohols in peanut oil for all binary systems considered (B), and Hildebrand plot for binary systems (C). 121x87mm (600 x 600 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. Melting enthalpy (∆Hm) as a function of fatty alcohol concentration. 56x38mm (600 x 600 DPI)


Page 24 of 27

Page 25 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. XRD patterns for binary systems containing C20OH (A) and C14OH (B) at different concentrations statically cooled at 7 °C. 63x23mm (600 x 600 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6. Schematization of possible fatty alcohol binary system crystals modifications during storage. 70x29mm (600 x 600 DPI)


Page 26 of 27

Page 27 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7. USAXS data for binary systems containing 5% of C22OH (A), C20OH (B), C18OH (C), and C16OH (D) statically cooled at 22 °C (ambient temperature). 119x83mm (600 x 600 DPI)


Phase Transition and Polymorphic Behavior of ... - ACS Publications

Recommend Documents