Determination of total 3‐chloropropane‐1,2‐diol (3‐MCPD) in edible oils by cleavage of MCPD esters with sodium methoxide

A method for the determination of total 3‐chloropropane‐1,2‐diol (3‐MCPD) in edible fats and oils was presented. 3‐MCPD was released from 3‐MCPD fatty acid esters by transesterification with NaOCH₃/methanol. After derivatization with phenylboronic acid, 3‐MCPD was determined by GC‐MS. Deuterium‐labeled 3‐MCPD was used as internal standard. In a model experiment, it was shown that acidic hydrolysis with methanol/sulfuric acid, which is normally used for the release of 3‐MCPD from its esters, can cause problems because under acidic conditions additional 3‐MCPD can be formed. No additional 3‐MCPD was formed using NaOCH₃/methanol for transesterification. Eleven samples of cold‐pressed and refined safflower oils were analyzed with this method. Levels of total 3‐MCPD were in the range from <100 up to 3200 µg/kg.     

Determination of 3‐MCPD esters in edible oil – methods of analysis and comparability of results

The discovery of fatty acid esters of 3‐chloropropane‐1,2‐diol (3‐MCPD) in edible oil products initiated food monitoring campaigns in many EU Member States. As the determination of 3‐MCPD esters was new to most laboratories, questions on the reliability of the produced analysis data were raised. In response to this, the Institute for Reference Materials and Measurements (IRMM) of the European Commission's Joint Research Centre (JRC) organised a proficiency test on the determination of 3‐chloropropane‐1,2‐diol esters (3‐MCPD esters) in edible oils. The aim of this proficiency test was to scrutinise the capabilities of official food control laboratories, private food control laboratories as well as laboratories from food industry to determine the 3‐MCPD esters content of edible oils. The study was carried out in accordance with “The International Harmonised Protocol for the Proficiency Testing of Analytical Chemistry Laboratories” and ISO Guide 43. The test materials dispatched to the participants were: refined palm oil, extra virgin olive oil spiked with 3‐chloropropane‐1,2‐dioleate and 3‐MCPD standard solution in sodium chloride. Altogether 41 laboratories from 11 EU Member States, Switzerland and Macedonia subscribed for participation in the study. The analysis task was to determine the 3‐MCPD esters content as total 3‐MCPD content of the test samples. Participants were free to choose their analysis methods. In total, 34 laboratories reported results to the organisers of the study. The performance of laboratories in the determination of 3‐MCPD esters in edible oils was expressed by z‐scores. About 56% of the participants performed satisfactorily in the determination of 3‐MCPD esters in palm oil and 85% for the spiked extra virgin olive oil test sample. The study revealed that the direct transesterification of the sample without the prior removal of glycidol esters might lead to strong positive bias.     

Generation of 3‐monochloro‐1,2‐propanediol and related materials from tri‐, di‐, and monoolein at deodorization temperature

Heating tests of pure tri‐, di‐, and monoolein (TO, DO, and MO, respectively) with and without the addition of tetrabutylammonium chloride as a chloride source at 240°C revealed the characteristic reactions that generate 3‐monochloro‐1,2‐propanediol‐related materials (3‐MCPD‐RM) in each acylglycerol. 3‐MCPD‐RM were formed mainly from DO and MO, with only a small amount from TO. Glycidyl ester was the predominant class of 3‐MCPD‐RM generated from both DO and MO, and was increased continuously throughout the heating period with comparable rates in both DO and MO, which also generated 3‐MCPD esters with chloride in a short completion time with an achieved level that was fourfold higher for MO than for DO. The production of free glycidol and 3‐MCPD was confirmed only in heated MO, but not from TO and DO, in a closed heating system, although these compounds were never detected in oils heated under simulated distillation conditions using a gas stream. In a closed system, both free glycidol and 3‐MCPD were increased throughout the heating period, which differed from the esters. Since an interesterification reaction, which produced free glycerol, was observed only in heated MO, free glycerol might be one of the precursors for those free forms. For clarification, further investigation is required.     

Influence of precursors on the formation of 3‐MCPD and glycidyl esters in a model oil under simulated deodorization conditions

To find new ways for reducing the potential of palm oil to form 3‐monochloropropane‐1,2‐diol (3‐MCPD) and glycidyl esters during refining it is helpful to know more about the influence of different precursors like diacylglycerols (DAGs) and monoacylglycerols (MAGs), lecithin, and chlorine containing compounds. After adding increasing amounts of the different precursors to a model oil obtained by removal of polar compounds from crude palm oil and heating the mixture under standardized conditions to 240°C for 2 h the contents of 3‐MCPD and glycidyl esters were analyzed according to the standard procedure of DGF C‐VI 18 (10). DAGs and MAGs were found to increase the potential of palm oil to form 3‐MCPD and glycidyl esters, but refined lecithin showed no influence. Sodium chloride as well as tetra‐n‐butylammoniumchloride (TBAC) led to higher contents of the esters. Whereas the addition of TBAC raised the amount of glycidyl esters as well as 3‐MCPD esters, sodium chloride largely raised the amount of 3‐MCPD esters. An addition of 5 mmol of sodium carbonate/kg model oil spiked with sodium chloride reduced the amount of glycidyl esters almost completely; the 3‐MCPD esters were reduced by 50%. About 1 mmol sodium hydrogen carbonate/kg oil reduced both 3‐MCPD and glycidyl esters almost completely. Practical applications : For the mitigation of the formation of 3‐MCPD esters and related compounds in refined edible oils, it is helpful to know more about the effect of different possible precursors. Using a broader data basis, it is possible to adopt the oil processing but especially the choice of the raw material to the demands of the market for lower contents of the esters in the refined products.     

Factors Impacting the Formation of 3-MCPD Esters and Glycidyl Esters During Deep Fat Frying of Chicken Breast Meat

The effect of the frying temperature, frying duration and the addition of NaCl on the formation of 3‐monochloropropane‐1,2‐diol (3‐MCPD) esters and glycidyl esters (GE) in palm olein after deep frying was examined in this study. The eight frying systems were deep‐fat frying (at 160 and 180 °C) of chicken breast meat (CBM) (with 0, 1, 3 and 5% sodium chloride, NaCl) for 100 min/day for five consecutive days. All oil samples collected after each day were analyzed for 3‐MCPD ester, GE, and free fatty acid (FFA) contents, specific extinctions at 232 and 268 nm (K₂₃₂ and K₂₆₈), p‐anisidine value (pA), and fatty acid composition. There was a significant (p < 0.05) decrease in the 3‐MCPD esters and a significant (p < 0.05) decrease in the GE with the increasing of the frying duration. There were significant (p < 0.05) increases in the 3‐MCPD esters formed when the concentration of NaCl increased from 0 to 5%. The addition of NaCl to the CBM during deep frying had no significant effect on the GE generation. The FFA contents, K₂₃₂ and K₂₆₈ and pA showed that all the frying oils were within the safety limit.     

Improvement of accuracy in quantification of 3‐monochloropropane‐1,2‐diol esters by Deutsche Gesellschaft für Fettwissenschaft standard methods C‐III 18

By Deutsche Gesellschaft für Fettwissenschaft (DGF) standard methods C‐III 18 for the determination of 3‐monochloropropane‐1,2‐diol (3‐MCPD), the minimum limit of detection was lower in the case of actual oil samples compared to the calibration samples. The problem was found to be lied in the low recovery of 3‐MCPD derivatives from the aqueous phase to the organic phase at the extraction step of the standard procedure. The substitution of the conventional solvent, n‐hexane, with n‐butanol, chloroform, and ethyl acetate increased the recovery to the relative extent of 5.6, 4.7, and 3.9, respectively. The modification contributed to improve the accuracy of the method, especially at lower concentration (<1 ppm) of 3‐MCPD. Practical applications: This paper provides the modification of DGF standard methods C‐III 18 in order to improve the accuracy to quantify 3‐MCPD at lower concentration. It might be important for estimation and control of our daily intake of 3‐MCPD, and for the product control in the fat and oil processing.     

Fatty acid esters of glycidol in refined fats and oils

Discrepancies in the analysis of 3‐chloropropane‐1,2‐diol (3‐MCPD) esters can be explained by the hypothesis that in some refined oils significant amounts of fatty acid esters of glycidol (glycidyl esters) are present in addition to 3‐MCPD esters. Glycidyl esters were separated from triacylglycerols by gel permeation chromatography (GPC) and detected by gas chromatography‐mass spectrometry (GC‐MS). Six samples of palm oil and palm oil‐based fats were analyzed by GPC and GC‐MS. In chromatograms of all samples, significant peaks, retention time and mass spectra in conformity with self‐synthesized glycidyl palmitate and glycidyl oleate were detectable. Quantification of individual glycidyl esters was not possible because of a lack of pure standards. Concentration of ester‐bound glycidol in different samples of fats and oils was estimated using an indirect difference method. Glycidyl esters could be detected only in refined, but not in crude or native, fats and oils. The highest concentrations were detected in palm oil and palm oil‐based fats. In a palm oil sample, glycidyl ester concentration varied according to different deodorization parameters, temperature, and time, while 3‐MCPD ester concentration was relatively constant, indicating that mitigation of glycidyl esters possibly may be achieved by optimizing refining parameters.     

Mitigation of 3‐MCPD and glycidyl esters within the production chain of vegetable oils especially palm oil

In 2007 the oil producing industry and the downstream food processing industry were worried by the announcement that fatty acid esters of 3‐MCPD and later in 2008 that glycidyl esters have been found in different types of vegetable oils after processing. The reason for the concern was that the German Federal Institute for Risk Assessment assumed in a first statement the complete degradation of the esters to free 3‐MCPD and glycidol both classified by the International Agency for Research on Cancer (IARC) as possible and probably, respectively, carcinogenic to humans. In the meantime a lot of research has been done on these heat‐induced compounds to mitigate their formation during oil processing and today the content in vegetable oils could be remarkably lower if all approaches were implemented. This paper summarizes the different mitigation strategies and shows their effect on lowering the amount of 3‐MCPD and glycidyl esters in refined vegetable oils.     

Surfactants from glucamines and ω‐epoxy fatty acid esters

The ring‐opening reaction of ω‐epoxy fatty acid methyl esters (9,10‐epoxy decanoic acid methyl ester, 10,11‐epoxy undecanoic acid methyl ester and 13,14‐epoxy tetradecanoic acid methyl ester) with N‐methyl glucamine or glucamine was studied. A new method (in methanol under reflux) leads to products, which are surfactants of linear‐ or γ‐type‐structure, in fair yields and purities. In a following step the ring‐opening products were saponificated by enzymatic catalysis or by conversion with sodium hydroxide leading to amphoteric surfactants. The surface‐active properties of the methyl esters and their corresponding acids or salts were studied at various pH‐values by measurement of surface tension of aqueous solutions and by examination of foaming properties. Depending on the chain length of the used epoxides, the structure of the obtained sugar‐based surfactant and the pH‐values, different surface‐active properties were observed. Several of the achieved products reduce the surface tension of aqueous solution down to 26—40 mN/m. Examination of foaming properties of the latter show rather poor or good foamers, depending on the surfactant used or the pH‐value of the solution.     

Chemical Synthesis and Gas Chromatographic Behaviour of γ‐Stearidonic (18:4n‐6) Acid

γ‐Stearidonic acid, 18:4n‐6, a potential product of β‐oxidation of arachidonic acid (20:4n‐6), was only recently positively identified in a living organism—a thermophilic cyanobacterium Tolypothrix sp., albeit at low levels, whilst some indirect evidence suggests its wider presence, e.g. in a unicellular marine alga. We have prepared 18:4n‐6 using an iodolactonisation chain‐shortening approach from 22:5n‐6 and obtained its ¹H‐, ¹³C‐, COSY‐ and HSQC NMR spectra, with 18:5n‐3 spectra also recorded for a comparison. The GC and GC‐MS behaviour of its methyl ester was also studied. Like another Δ3 polyunsaturated acid, octadecapentaenoic (18:5n‐3), 18:4n‐6 rapidly yields 2‐trans isomer upon formation of dimethyloxazoline derivative. On a polar ionic liquid phase (SLB‐IL100, 200 °C) the methyl ester could be mistaken for 18:3n‐3, while on methylsilicone phase (BP1, 210 °C) it eluted ahead of 18:3n‐6 and 18:4n‐3, suggesting that when present it may be easily misidentified during GC analysis of fatty acids.     

The Synthesis of trans‐Perhydroindolic Acids and their Application in Asymmetric Domino Reactions of Aldehyde Esters with β,γ‐Unsaturated α‐Keto Esters

(2S,3aR,7aS)‐Perhydroindolic acid, the key intermediate in the synthesis of trandolapril, and its trans‐isomers, were readily prepared. These proline‐like molecules are unique in that they contain a rigid bicyclic structure, with two hydrogen atoms trans to each other at the bridgehead carbon atoms. These molecules were used successfully as chiral organocatalysts in asymmetric domino Michael addition/cyclization reactions of aldehyde esters with β,γ‐unsaturated α‐keto esters. They proved to have excellent catalytic behavior, allowing for the synthesis of multi‐substituted, enantiomerically enriched hemiacetal esters. Under optimal conditions (using 10 mol% catalyst loading), a series of β,γ‐unsaturated α‐keto esters was examined with up to 99% de, ee and yield, respectively. Additionally, the enantiomerically enriched hemiacetal esters could be readily transformed into their corresponding bioactive pyrano[2,3‐b]pyrans (possessing a multi‐substituted bicyclic backbone).     

Dietary trans α‐linolenic acid does not inhibit Δ5‐ and Δ6‐desaturation of linoleic acid in man

Dietary trans monoenes have been associated with an increased risk of heart disease in some studies and this has caused much concern. Trans polyenes are also present in the diet, for example, trans α‐linolenic acid is formed during the deodorisation of α‐linolenic acid‐rich oils such as rapeseed oil. One would expect the intake of trans α‐linolenic acid to be on the increase since the consumption of rapeseed oil in the western diet is increasing. There are no data on trans α‐linolenic acid consumption and its effects. We therefore carried out a comprehensive study to examine whether trans isomers of this polyunsaturated fatty acid increased the risk of coronary heart disease. Since inhibition of Δ6‐desaturase had also been linked to heart disease, the effect of trans α‐linolenic acid on the conversion of [U‐¹³C]‐labelled linoleic acid to dihomo‐γ‐linolenic and arachidonic acid was studied in 7 healthy men recruited from the staff and students of the University of Edinburgh. Thirty percent of the habitual fat was replaced using a trans ‘free’‐ or ‘high’ trans α‐linolenic acid fat. After at least 6 weeks on the experimental diets, the men received 3‐oleyl, 1,2‐[U‐¹³C]‐linoleyl glycerol (15 mg twice daily for ten days). The fatty acid composition of plasma phospholipids and the incorporation of ¹³C‐label into n‐6 fatty acids were determined at day 8, 9 and 10 and after a 6‐week washout period by gas chromatography‐combustion‐isotope ratio mass spectrometry. Trans α‐linolenic acid of plasma phospholipids increased from 0.04 ? 0.01 to 0.17 ? 0.02 and cis ? ‐linolenic acid decreased from 0.42 ? 0.07 to 0.29 ? 0.08 g/100 g of fatty acids on the high trans diet. The composition of the other plasma phospholipid fatty acids did not change. The enrichment of phosphatidyl ¹³C‐linoleic acid reached a plateau at day 10 and the average of the last 3 days did not differ between the low and high trans period. Both dihomo‐γ‐linolenic and arachidonic acid in phospholipids were enriched in ¹³C, both in absolute and relative terms (with respect to ¹³C‐linoleic acid). The enrichment was slightly and significantly higher during the high trans period (P<0.05). Our data suggest that a diet rich in trans α‐linolenic acid (0.6% of energy) does not inhibit the conversion of linoleic acid to dihomo‐γ‐linolenic and arachidonic acid in healthy middle‐aged men consuming a diet rich in linoleic acid.     

Alteration of Wax Ester Content and Composition in Euglena gracilis with Gene Silencing of 3‐ketoacyl‐CoA Thiolase Isozymes

Euglena gracilis produces wax ester under hypoxic and anaerobic culture conditions with a net synthesis of ATP. In wax ester fermentation, fatty acids are synthesized by reversing beta‐oxidation in mitochondria. A major species of wax ester produced by E. gracilis is myristyl myristate (14:0‐14:0Alc). Because of its shorter carbon chain length with saturated compounds, biodiesel produced from E. gracilis wax ester may have good cold flow properties with high oxidative stability. We reasoned that a slight metabolic modification would enable E. gracilis to produce a biofuel of ideal composition. In order to produce wax ester with shorter acyl chain length, we focused on isozymes of the enzyme 3‐ketoacyl‐CoA thiolase (KAT), a condensing enzyme of the mitochondrial fatty acid synthesis pathway in E. gracilis. We performed a gene silencing study of KAT isozymes in E. gracilis. Six KAT isozymes were identified in the E. gracilis EST database, and silencing any three of them (EgKAT1‐3) altered the wax ester amount and composition. In particular, silencing EgKAT1 induced a significant compositional shift to shorter carbon chain lengths in wax ester. A model fuel mixture inferred from the composition of wax ester in EgKAT1‐silenced cells showed a significant decrease in melting point compared to that of the control cells.     

Application of Indirect Enzymatic Method for Determinations of 2-/3-MCPD-Es and Gly-Es in Foods Containing fats and Oils

Because of the potential health risks, fatty acid esters of 3‐chloro‐1,2‐propanediol (3‐MCPD‐Es), 2‐chloro‐1,3‐propanediol (2‐MCPD‐Es) and glycidol (Gly‐Es) in foods are drawing the attention of public health authorities. To assess applicability of the rapid indirect method developed earlier by using a Candida rugosa lipase for the analyses of refined fats and oils was applied to the analyses of various foods. Mayonnaise, vegetable oil margarine and fat spread could be analyzed with the hydrolysis condition of 30 min at room temperature. Analyses of 3‐MCPD‐Es in margarines and fat spreads containing milk fat could be analyzed by increasing the hydrolysis temperature to 40 °C. The results in a mayonnaise, four fat spreads and five margarines analyzed by the enzymatic method were 0.10–0.98 mg/kg for 3‐MCPD, 0.05–0.41 mg/kg for 2‐MCPD and 0.15–0.59 mg/kg for Gly, and correlated well with the results obtained by AOCS Cd 29a with Cd 30–15 with slopes of 0.99–1.13, and R²s of 0.87–0.99. Further, by adding a simple fat extraction step using a solvent mix at 60 °C, foods high in protein and carbohydrate, such as infant formulas, could also be successfully analyzed with >90 % recovery in 1 day. Because the enzymatic method requires only 30 min for hydrolysis, the method is considered suitable for routine analyses of 2‐/3‐MCPD‐Es and Gly‐Es in foods.     

Alcoholysis of waste fats with 2‐ethyl‐1‐hexanol using Candida antarctica lipase A in large‐scale tests

Commercial native lipase A from Candida antarctica was used to produce alkyl esters through the alcoholysis of (waste) fats with 2‐ethyl‐1‐hexanol. The process was carried out in batch stirred tank reactors (from 100 mL up to 3000 L). The content of alkyl esters in reaction mixtures was determined by gradient HPLC using an evaporative light scattering detector and the reaction progress was controlled by determining the ratio of the palmitic acid ester peak area to the oleic acid ester peak area in HPLC chromatograms. The results show that alcoholysis is the favoured reaction in presence of excess water and water‐insoluble alcohols in comparison with hydrolysis (fatty acid content <5%). The optimum amount of water for the alcoholysis was found to be 80–100% of the amount of fat. In the presence of low quantities of water both alcoholysis and hydrolysis are slow. Conversion rate increases with increasing temperature to 65–70 °C. Based on these results a large‐scale test to produce 3000 L of alkyl ester (to be used as lubricant coolant) was carried out. The experiments have proved that alcoholysis is completed after about 7–10 h depending on temperature.     

Microbial transformation of androst‐4‐ene‐3, 17‐dione by Bordetella sp. B4 CGMCC 2229

BACKGROUND: Microbial transformation of steroids has attracted widespread attention, especially the transformation of those steroids synthesized with difficulty by chemical methods. In this study, microbial transformation of androst‐4‐ene‐3, 17‐dione (AD) by Bordetella sp. B4 was investigated, and the effect of temperature on transformation was studied. RESULTS: Three metabolites were purified by preparative TLC and HPLC, and identified as androsta‐1,4‐diene‐3,17‐dione (ADD), 9α‐hydroxyandrost‐4‐ene‐3, 17‐dione (9α‐OH‐AD), and 3‐hydroxy‐9, 10‐secoandrost‐1, 3, 5‐triene‐9, 17‐dione (3‐OH‐SATD) by nuclear magnetic resonance imaging (NMR), Fourier transform infrared spectroscopy (FTIR) and mass spectroscopy (MS). It was first reported that the genus of Bordetella has the capability of AD degradation. Microbial transformation of AD was performed at 30 °C, 37 °C, 40 °C and 45 °C. The 9α‐OH‐AD yield reached a maximum within 16 h when the strain was cultivated in media with AD as sole carbon at 37 °C. Surprisingly, ADD was produced by the strain cultivated at 40 °C but not at 37 °C, which was different from previous reports. It was deduced that the alcohol dehydrogenase that catalyzed the transformation of AD to ADD may be temperature sensitive. CONCLUSION: Androst‐4‐ene‐3,17‐dione was converted into 9α‐hydroxyandrost‐4‐ene‐3, 17‐dione and other metabolites rapidly by Bordetella sp. B4. It is anticipated that the strain Bordetella sp. B4 CGMCC 2229 can be used in the steroids industry. Copyright © 2009 Society of Chemical Industry     

Three‐Component Staudinger‐Type Stereoselective Synthesis of C‐Glycosyl‐β‐lactams and their Use as Precursors for C‐Glycosyl Isoserines and Dipeptides. A Polymer‐Assisted Solution‐Phase Approach

A collection of 4‐(C‐galactosyl)‐ and 4‐(C‐ribosyl)‐β‐lactams featuring different substituents at C‐3 and N‐1 was prepared by combining in a one‐pot procedure a formyl C‐glycoside, a primary amine, and a substituted acetyl chloride in the presence of base (Staudinger‐type reaction). Sulfonyl chloride and aminomethylated resins were used in sequence to remove excess of components and by‐products. Two pure C‐glycosyl‐β‐lactams were effectively transformed into C‐glycosyl‐N‐Boc‐β‐amino‐α‐hydroxy esters (C‐glycosyl isoserines) and a C‐ribosyl dipeptide via base‐promoted heterocycle ring opening by methanol and L ‐phenylalanine methyl ester, respectively.     

Synthesis of poly(ester‐anhydride)s based on poly(ϵ‐caprolactone) prepolymer

The objective of this study was to prepare high molecular weight poly(ester‐anhydride)s by melt polycondensation. The polymerization procedure consisted of the preparation of carboxylic acid terminated poly(?‐caprolactone) prepolymers that were melt polymerized to poly(?‐caprolactone)s containing anhydride functions along the polymer backbone. Poly(?‐caprolactone) prepolymers were prepared using either 1,4‐butanediol or 4‐(hydroxymethyl)benzoic acid as initiators, yielding hydroxyl‐terminated intermediates that were then converted to carboxylic acid‐terminated prepolymers by reaction with succinic anhydride. Prepolymers were then allowed to react with an excess of acetic anhydride, followed by subsequent polycondensation to resulting high molecular weight poly(ester‐anhydride)s. Upon coupling of prepolymers, size exclusion chromatography analyses showed an increase from 3600 to 70,000 g/mol in number‐average molecular weight (M_n) for the 1,4‐butanediol initiated polymer, and an increase from 7200 to 68,000 g/mol for the 4‐(hydroxymethyl)benzoic acid‐initiated polymer. 4‐Hydroxybenzoic acid and adipic acid were also used as initiators in the preparation of poly(?‐caprolactone) prepolymers. However, with these initiators, the results were not satisfactory. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 81: 176–185, 2001     

Quasi‐alternating polyesteramides from ε‐caprolactone and α‐amino acids

Glycine‐ɛ‐caprolactone‐based and α‐alanine‐ɛ‐caprolactone‐based polyesteramides with a strong tendency to form alternating sequences (degree of randomness = 1.64 and 1.31) were synthesized by melt polycondensation of intermediate hydroxy‐ and ethyl ester‐terminated amides. These intermediates were synthesized by the reaction of equimolar amounts of ɛ‐caprolactone and glycine or L‐α‐alanine ethyl esters in mild conditions. The structure and microstructure of these polyesteramides are discussed on the basis of an in‐depth nuclear magnetic resonance study. Both polyesteramides are semi‐crystalline, but the glycine‐based one presents the highest melting enthalpy. This polyesteramide also exhibits higher Young's modulus and stress at break than its α‐ and β‐alanine counterparts. © 2016 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2016 , 133, 44220.