Food and Packaging Interactions II - American Chemical Society

Chapter 13 Determination of Flavor-Polymer Interactions by Vacuum-Microgravimetric Method Ann M. Roland and Joseph H. Hotchkiss

Downloaded by UNIV OF PITTSBURGH on May 3, 2015 | http://pubs.acs.org Publication Date: September 13, 1991 | doi: 10.1021/bk-1991-0473.ch013

Institute of Food Science, Stocking Hall, Cornell University, Ithaca, NY 14853

A method to quantify the sorption of aroma compounds by food-contact polymers was developed. The method consisted of a microgravimetric balance which was contained in a vacuum chamber. Polymer samples were placed on the balance in the sealed chamber and out gassed to 10 torr. Small amounts of aroma compounds (d-limonene or d,1-linalool) were admitted to the evacuated chamber and the mass increase in the polymer resulting from compound sorption as well as the pressure within the vessel recorded by computer. Additional aroma compound was admitted to the chamber after equilibrium was achieved and a new equilibrium masspressure combination recorded. Sorption isotherms were constructed for each compound-polymer combination from individual equilibria. Solubility coefficients as well as predictions of the amount of aroma compound sorbed at a given aroma concentration (i.e., partial pressure) were made. Results indicated that this technique may be a useful predictor of individual flavor-polymer interactions at flavor concentrations and amounts which are found in foods. A hypothetical prediction of limonene sorption by a cereal box liner was made using data generated by this technique. -4

Sorption of food aromas by packaging materials i s factor i n q u a l i t y a l t e r a t i o n during storage (2). Changes i n both i n t e n s i t y and character of f l a v o r mixtures r e s u l t i n g from exposure to polymers can be detected by the human nose (2). These changes are due to absorption of c e r t a i n organic compounds that make up complex aromas by food-contact polymers. This phenomena has been termed "scalping." No standard method f o r p r e d i c t i n g the amount of sorption occurring f o r any polymer-flavor combination has been developed. Several sorption studies have u t i l i z e d sorption c e l l methods where 0097-6156/91A)473-0149$06.00A) © 1991 American Chemical Society

In Food and Packaging Interactions II; Risch, S., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1991.


150

FOOD AND PACKAGING INTERACTIONS

the penetrant i s dissolved or suspended i n a l i q u i d and brought into contact with polymer. Essential o i l s , pure compounds (2), or complex mixtures such as orange j u i c e are used (3, 4). The polymer and l i q u i d are either placed i n an inert container (2, 3), or the l i q u i d i s placed i n a polymeric container (4). Sorption i s measured by quantitative evaluation of the aroma compound(s) remaining a f t e r equilibrium or by quantitative headspace evaluation. In d i r e c t weighing methods, specimens are exposed to saturated vapor, withdrawn, weighed, and returned at i n t e r v a l s . A l t e r n a t e l y , samples immersed i n a l i q u i d penetrant are removed, b l o t t e d , weighed and returned. For example, the d i f f u s i o n of organic vapors into polystyrene and other polymers has been studied using t h i s method (5-8). The l i q u i d penetrant method was recently used by A i t h a l et a l . (9) to study aromatic penetrants i n polyurethane membranes. Short comings of current methods include: the presence of an a i r b a r r i e r , disturbances during the weighing, and heating of the specimen by condensation of solvent vapor when weighing i s done i n environments at temperatures lower than the l i q u i d penetrant (6). Measurements at high or saturated vapor pressures may not r e f l e c t sorption from foods where aromas are well below saturation. Such high penetrant concentrations can r e s u l t i n p l a s t i z a t i o n of the polymer which changes i t s sorptive properties. Recent constant vapor concentration methods employ a vapor generating/dilution system. A constant concentration of penetrant vapor stream i s produced by bubbling N through l i q u i d penetrant and d i l u t i n g with untreated N , and passing i t over the polymer (20). The polymer i s continuously weighed. Mohney et a l . (22) recorded the sorption of d-limonene vapor i n high density polyethylene/sealant laminant f i l m u t i l i z i n g t h i s method. Limonene vapor concentrations of 0.3-7.0 ppm (wt/v) were used. For the laminant and a d-limonene vapor concentration of 1.5 ppm (wt/v), a s o l u b i l i t y c o e f f i c i e n t (S) of 7.6 mg/(gxppm) was found. Some methods involve r e l a t i v e l y large amounts of aroma components because of the l i m i t e d s e n s i t i v i t y of detection methods. High amounts act as i n f i n i t e reservoirs as compared to amounts of aroma compounds present i n foods. Other system components such as aqueous media, or solvents used to disperse aroma compounds may a f f e c t the p a r t i t i o n i n g . There i s an added consideration of disturbances caused by the vapor stream decreasing the p o t e n t i a l s e n s i t i v i t y of mass determinations. Our objective was to determine s p e c i f i c flavor-polymer interactions at penetrant l e v e l s found i n the headspace of foods (often ppb concentrations) using f i n i t e amounts of aroma. We required a method that was quantitative and p r e d i c t i v e . 2

2

Materials and Methods Sorption Apparatus. A Cahn 2000 Electrobalance, control unit, weighing unit and vacuum chamber were used (Cahn Instruments, C e r r i t o s , CA; Figure 1). J o i n t s were sealed with o-rings and Apiezon L high vacuum grease (Apiezon Products Limited, London, England). The electrobalance stand was mounted on four SM-1 Stabl-


13.

Determination of Flavor-Polymer Interactions

ROLAND & HOTCHKISS

151

Levi pneumatic mounts (Barry Controls, Burbank, CA). The balance chamber was i s o l a t e d from the pumping system by a 30 cm LN trap. The chamber was closed o f f from the pump and trap by a valve after pump down. Internal pressure was continuously measured (type 600 Barocel Pressure Sensor, Datametrics/Dresser Industries, Inc., Wilmington, MA; 10 Vdc, 10.000 mmHg f u l l range). Ultra-Torr vacuum f i t t i n g s (Cajon Co. Macedonia, OH) and Teflon high vacuum valves were used for a l l connections. Sample temperature was continuously recorded (Type J thermocouple). Mechanical (Sargent-Welch Skokie, IL model 1402F) and o i l d i f f u s i o n pumps were used. The vacuum pump o i l , Plasma O i l 80 (CVC Products, Inc. Rochester, NY), i s rated at a vapor pressure of 8 χ 10' mmHg at 25°C. The d i f f u s i o n pump o i l , Convoil 20 (CVC Products, Inc. Rochester, NY), i s rated at a vapor pressure of 10" to 10" mmHg at 25°C. Mass was recorded i n m i l l i v o l t s (mV) with a f u l l scale reading of 9995 /ig and a s e n s i t i v i t y of 1 /ig. A thermostated 23 + 3.5°C insulated room housed the balance. A thermostated water bath (VWR Model 1145, VWR S c i e n t i f i c , San Francisco, CA) c i r c u l a t e d temperature c o n t r o l l e d water v i a tubing around the vacuum b o t t l e and sample tube which was held i n a water f i l l e d Dewar f l a s k . A constant sample temperature of 23.5 ± 0.5 °C was maintained. A metal encased i o n i z i n g unit (Staticmaster; Nuclear Products Co. E l Monte, CA) controlled s t a t i c e l e c t r i c i t y inside vacuum chamber. 2


6

6

8

Polymer sample. Polyethylene ( S c i e n t i f i c Polymer Products Ontario, New York, catalog #560, density 0.92 g/cm ,softening point 107°C) was hot pressed between Mylar sheets (103-105°C; Loomis Engineering Co. , Caldwell, NJ; 10,000 l b / i n ) . The sample was cooled a t room temperature. Samples (Table I) were pierced with a 0.01 mm diameter hangdown wire. 3

2

Table I: Polyethylene f i l m samples Aroma compound d-Limonene d,l-Linalool

Thickness 1.8 mil (0.046 mm) 2.0 mil (0.051 mm)

2

Area(cm ) 10.0 15.0

Aroma Compounds. (+)-Limonene (97%) (l-methyl-4-isopropenyl-lcyclohexene) was obtained from A l d r i c h Chemical Company (Milwaukee, WI). d , l - L i n a l o o l (95-97%) (3,7-dimethyl-l,6 octadien-3-ol) was obtained from Sigma Chemical Company (St. Louis, MO). Sorption experiments were c a r r i e d out as follows: Liquid penetrant was placed i n the penetrant vessel (Figure 1), frozen i n LN and outgassed through the vacuum pumping system at 4xl0" mmHg. The penetrant valve was closed. The f i l m sample was suspended from one o f the arms of the balance. The thermocouple was positioned 5

2


152

FOOD AND

PACKAGING INTERACTIONS

within 1 cm of the sample. The mass was determined and the balance mechanically and e l e c t r o n i c a l l y tared. The chamber was evacuated (4 xlO" mmHg) u n t i l a constant polymer mass was obtained. The valve i s o l a t i n g the main chamber from the pumping system was then closed. The penetrant vessel valve was then b r i e f l y opened and closed to admit a small f i n i t e amount of aroma into the glass vacuum v e s s e l . Once the equilibrium mass was attained a second dose of vapor was admitted. A new equilibrium sorption mass was then measured at the higher aroma vapor concentration. This procedure was repeated at increasing vapor concentrations with a single polymer sample (Figure 2). For determinations at the saturation vapor pressure the penetrant valve was l e f t open. Temperature, pressure and mass were recorded every 2 sec by a DT2805 data a c q u i s i t i o n board, DT707-T screw terminal panel with cold j u n c t i o n (Data Translation, Inc. Marlboro, MA) and a Leading Edge PC computer. The i d e a l gas r e l a t i o n s h i p was used to determine penetrant concentration i n the vapor and conversions made according to Barton


5

(12).

Results The mass (Mt) of penetrant sorbed per u n i t weight of penetrant free polymer at equilibrium was measured as a function of time and pressure. Mt approached a l i m i t i n g value (M.) at higher vapor pressures i n d i c a t i n g sample saturation. The r a t i o Mt/M* was p l o t t e d against t to obtain a sorption curve (Figures 3A, 3B; 13). Sorption curves at saturation vapor pressures (Figure 3A) are s i m i l a r to curves obtained using constant vapor pressure systems (II). However, when f i n i t e amounts of penetrant were admitted, equilibrium mass overshoots followed by loss of mass to equilibrium was observed (Figure 3B). A sample temperature r i s e and decrease of 0.1 to 0.3°C accompanied the mass overshoot (Figure 4). The pressure inside the chamber also increased sharply when the penetrant was introduced (Figure 4). Equilibrium s o l u b i l i t y c o e f f i c i e n t s (S) were c a l c u l a t e d from the equilibrium concentrations of penetrant i n the polymer and vapor phase according to: 1 / 2

S - Cp/C^ where C Cy p

(1) - equilibrium s o l u b i l i t y i n the polymer (/ig/mg) - equilibrium concentration i n the vapor phase (ppm)

Figure 5 compares the s o l u b i l i t y c o e f f i c i e n t s of l i n a l o o l and limonene as a function of vapor concentration. At equal vapor concentrations l i n a l o o l has a higher s o l u b i l i t y than limonene. The equilibrium vapor concentration may also be expressed as a r a t i o of the penetrant's p a r t i a l pressure to i t s saturation vapor pressure. When t h i s i s p l o t t e d against mass gained, a sorption


13. ROLAND & HOTCHKISS


DIAGRAM OF VACUUM MICROBALANCE APPARATUS

;

TEMPERATURE CONTROLLED

1

\


GLASS VACUUM V E S S E L

Τ m TEMPERATURE

Figure 1:

Figure 2:

Sorption Apparatus.

Mass gain f o r d-limonene sorption by polyethylene. F i n a l vapor pressure (mmHg): a 0.101, b 0.154, c 0.293, d 0.413, e 0.549, f 0.684, g 0.752 .


153


154


M

t

0

100

200

300

400 1

Time Figure 3:

/

2

500

600

700 1 / 2

(seconds

800

900

)

Sorption of d-limonene by polyethylene at 23.5°C; A [limonene] - 40.5>:10~ mol/L (sat); Β [limonene]^ - 5.5xl0" mol/L. 6

eq

6

τ 25.5

600 τ

m ο

500»

x

400-

pressure

+ 24.5 /—^

I

. 300-

C

cn

ο ο

+ 23.5

temperature

ο + 22.5 §. Ε Η+ 21.5

o

5

200-

(Λ CO

4)

QI

100 +

mass

1

1

ι

2

3

4

>

5

ι

6

ι

7

ι

8

ι

9

ι

20.5

10

Time (days)

Figure 4:

^

Mass gain, pressure, and sample temperature f o r dlimonene and polyethylene.


13.

ROLAND & HOTCHKISS


isotherm i s constructed for each (Figure 6).

penetrant/polymer

combination


Discussion The mass overshoot occurred with each penetrant i n l e t except at saturation pressures (Figure 2). Similar overshoots i n sorption have been observed by Rogers (14) with n-hexane and polyethylene. The cause of such behavior has not been f u l l y explained. I t may simply be that opening the penetrant valve allows f o r a large temporary increase i n e f f e c t i v e penetrant pressure i n the chamber with a r e s u l t i n g overshoot i n mass gain. I t i s also possible that the penetrant condenses on the sample then equilibrates with the vapor phase to reach an equilibrium mass. However, Rogers (14) and Feughelman (25) suggest that a transformation of the polymer may occur during sorption. A large number of weak interchain bonds could be broken by the i n i t i a l inrush of penetrant. Some of these bonds could be reformed l a t e r causing the exclusion of previously sorbed penetrant. An a l t e r n a t i v e hypothesis suggests that under conditions where d i f f u s i o n into t h i n specimens of high penetrant a f f i n i t y Is rapid, the heat of sorption may cause a large increase i n l o c a l temperature. The heat of condensation could also cause an increase i n sample temperature (16). Using a system of alkanes and polycarbonate Chen and Edin (17) demonstrated that d i s s o l u t i o n (sorption) i s exothermic and that the heat of solution i s l i n e a r l y r e l a t e d to the heat of condensation. The temperature increase from heat of sorption may l o c a l l y a f f e c t the polymer by thermal expansion or melting of c r y s t a l l i n e regions. As the temperature returns to ambient, c r y s t a l l i n e or semi-ordered regions may reform and a portion of the penetrant excluded from specimen (14). The higher solubility of linalool than limonene in polyethylene may be caused by i n two factors. L i n a l o o l i s a more l i n e a r , less bulky, molecule than limonene which would f a c i l i t a t e i t s a b i l i t y to move into the polymer matrix. Also, i t s higher b o i l i n g point (198°C l i n a l o o l ; 175°C limonene) i s i n d i c a t i v e of i t s a b i l i t y to condense and remain within the matrix. B o i l i n g point has been correlated to a higher s o l u b i l i t y by previous workers (18-20). Zobel experimentally determined the s o l u b i l i t y of limonene but not l i n a l o o l i n polypropylene. However, we calculated the s o l u b i l i t y of l i n a l o o l i n polypropylene using the constants derived by Zobel (20) for alcohol vapor i n polypropylene at 25°C (log S - -10.584 + (0.0286 χ T); where T- b o i l i n g point K). The s o l u b i l i t y derived for l i n a l o o l vapor was 770 g/Nm versus Zobel's experimentally determined value of 510 g/Nm for limonene suggesting that l i n a l o o l has a higher solubility than limonene i n polypropylene. Therefore, our experimental results i n polyethylene would l i k e l y agree with the order of s o l u b i l i t i e s f o r Zobel's determinations i n polypropylene. S o l u b i l i t y c o e f f i c i e n t s derived from the l i n e a r portion of sorption isotherms may be useful approximations of the loss of v o l a t i l e s from a packaged food. S o l u b i l i t y c o e f f i c i e n t s (S) were determined as follows :


155

156


6.0

j / /

5.0-4.0-

/ 3.0• Limonene

2.0-


•

Linalool

1.0-

2

3

4

Vapor Concentration Figure 5:

5 (ppm)

Solubility coefficients f o r d-limonene and d , l l i n a l o o l i n polyethylene as a function of vapor concentration of penetrant at 23.5°C.

Partial Vapor Pressure (P/Psat) Figure 6:

Sorption isotherms for d-limonene and i n polyethylene at 23.5°C.

d,l-linalool


13.

ROLAND & HOTCHKISS

Determination ofFlavor-Polymer Interactions

S - Cp/C. where

157

(2) - concentration of aroma i n the polymer at equilibrium (/ig/mg). - concentration of aroma i n the vapor at equilibrium (P/P, ) · at

and C

p

χ Ρ - aroma sorbed by package (/ig)

(3)

where Ρ

- Mass of polymer package (mg)

Solving equation 2 for C and substituting into equation 3 and solving f o r C i n terms of S: p


p

aroma sorbed ^ g ) - S χ

χΡ

(4)

Using t h i s relationship, the equilibrium sorption o f limonene from a flavored dry product by a polyethylene cereal l i n e r may be predicted given the following Information: 2

Ρ - 9060 mg PE (48 μπι; 45.3 g/m; 16 oz. package) S - 5.0 (Mg/mg)/(P/P ) (Figure 6 f o r limonene) Cv - 0.1 (10%) (estimated percentage of saturation i n headspace) aroma sorbed ^ g ) - 5.0 χ 0.1 χ 9060 - 4530 μ eat

β

The mass of aroma sorbed by the package i s estimated to be 4530 μg, assuming the aroma concentration i n 448 g (16 oz.) of dry product Is approximately 30 /ig/g (21). This represents a loss of 29% of the o r i g i n a l 13,400 μg of limonene present. This example i s summarized i n Table II along with calculations for limonene and l i n a l o o l at estimated vapor concentrations i n a f r u i t flavored dry product, (I.e. a breakfast c e r e a l ) . The analysis assumes a constant aroma vapor pressure i n the package and that the p a r t i t i o n i n g between the cereal and the vapor i s very rapid i n comparison to the vapor-polymer ( i . e . , the aroma from the product replaces the sorbed aroma as i t i s taken up by the package). This maintains a constant concentration of aroma i n the vapor a t the expense o f the aroma content i n the product. L i n a l o o l , usually present at 1/10 the concentration of limonene (21), may have a greater a p o t e n t i a l f o r a f f e c t i n g aroma due to sorption by a polyethylene package than limonene. The higher s o l u b i l i t y of l i n a l o o l coupled with i t s higher sensory impact (at l e a s t 2 f o l d that of limonene; 22) increases the e f f e c t of l i n a l o o l sorption. Conclusions Differences i n sorption p o t e n t i a l combined with differences i n concentrations and sensory impact of aroma compounds i l l u s t r a t e s that sorption can a l t e r the o v e r a l l flavor p r o f i l e (not j u s t intensity) of a packaged product. S o l u b i l i t y c o e f f i c i e n t s (S) may be



b

a

Ρ - 9060 mg Polyethylene (48 i n 16 oz. (448 g) dry product

2

109 22 μπι; 45.3 g/m ;

0.5 % 0.1 %

2.4

d,1-Linalool

4530 453 226

D

C χ Ρ

8

3

30

( Hg/g)

Aroma Cone. i n Product

16 oz. pkg)

( Mg)

(% of Sat.)

10 % 1 % 0.5 %

p

Aroma Sorbed by Package

Vapor Cone.

5.0

1 p

Low Cone. Solubility Coefficient / ll/mg I / s.t ' S

d-Limonene

Aroma Compound

134

13400

( eg)

6

Aroma Mass i n Product

16%

81%

33% 3% 2%

Aroma Mass Sorbed (%)

J

|

|

I I : Example of sorption of the f l a v o r compounds limonene and l i n a l o o l i n a product by a polyethylene package l i n e r

Hypothetical Fruit-Flavored Dry Product

Table



13. ROLAND & HOTCHKISS

Determination ofFlavor-Polymer

Interactions

159

useful i n p r e d i c t i n g the potential of d i f f e r e n t types of polymeric packaging materials sorb s p e c i f i c compounds. I t may be possible to compensate f o r losses by reformulation. A l t e r i n g the s p e c i f i c polymers used f o r a given food may also lead to improved food quality. The microgravimetric vacuum method provides an accurate means of obtaining sorption isotherms and s o l u b i l i t y c o e f f i c i e n t s f o r flavor-polymer combinations at penetrant concentrations which are representative of food systems. The apparatus i s r e l a t i v e l y simple and requires minimal operator attention once the experiments are started. Consecutive determinations may be made on the same sample at d i f f e r e n t levels of penetrant. The vacuum method minimizes disturbances to the weighing apparatus thus providing greater p o t e n t i a l for highest accuracy determination at low penetrant levels where there are smaller mass increases because of sorption.

Literature Cited 1. 2. 3.

4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

15. 16.

Farrell, C.J. Ind.& Eng. Chem. Res. 1988, 10(10), 1946-1951. Kwapong, O.Y., Food-package Interactions: The Sorption of Aroma Compounds by Polymeric Materials; M.S. thesis, Cornell University, Ithaca, NY. 1986. Hirose, K.; Harte, B.R.; Giacin, J.R.; Miltz, J.; Stine, C. In Food and Packaging Interactions; Hotchkiss, J.H., Ed.; American Chemical Society: Washington, DC, 1988, Ch. 3; pp 2841. 1988. Durr, P.; Schobinger, U.; Waldvogel, R. Alimenta. 1981, 20, 91-93. Crank, J.; Park, G.S. Trans. of the Faraday Soc. 1949, 45, 240-249. Park, G.S. Trans. of the Faraday Soc. 1950, 46, 684-697. Park, G.S. Trans. of the Faraday Soc. 1951, 47, 1007-1013. Laine, R.; Osbourn, J.O. J. Appl. Polymer Sci. 1971, 15, 327-339. Aithal, U.S.; Aminabhavi, T.M.; Cassidy, P.E. In Barrier Polymers and Structures; Koros, W.J., Ed.; American Chemical Society: Washington, DC, 1990, Ch. 19; pp 351-376. Baner, A.L.; Hernandez, R.J.; Jayaraman, K.; Giacin, J.R. Current Technol. in Flexible Packaging. 1986, ASTM STP 912, 49-62. Mohney, S.M.; Hernandez, R.J.; Giacin, J.R.; Harte, B.R.; Miltz, J . J. Food Sci. 1988, 53(1), 253-257. Barton, A.F.M. CRC Handbook of Solubility Parameters and Other Cohesion Parameters; CRC Press, Inc.: Boca Raton, FL, 1983. Fujita, H. 1961. Adv. Polymer Sci. 1961. 3, 1-47. Rogers, C.E. In Physics and chemistry of the organic solid state; Fox, D.; Labes, M.M.; Weissberger, Α., Eds.; Interscience Publishers A Div. of John Wiley & Sons, Inc.: New York, NY, 1965, Ch 6, Vol 2. Feughelman, M. J. Appl. Sci. 1959, 2(5), 189-191. Crank, J.; Park, G.S. In Diffusion In Polymers; Crank, J.; Park, G.S., Eds.; Academic Press: New York, NY, 1968, Ch 1, pp 1-39.


160

17. 18. 19. 20. 21.


22.


Chen, S.P.; Edin, A.D. Polymer Eng. Sci. 1980, 20(1), 40-50. van Amerongen, G.J. J. Polymer Sci. 1950, 5(3), 307-332. Barrer, R.M.; Barrie, J.Α.; Slater, J . J. Polymer Sci. 1958, 27, 177-197. Zobel, M.G.R. Polymer Test. 1985, 5(2), 153-165. Furia, T.E.; Bellanca, N. Fenaroli's Handbook of Flavor Ingredients; CRC Press, Inc.: Cleveland, OH, 1975, 2nd Ed., Vol. 1. Buttery, R.G.; Black, D.R.; Guadagni, D.G.; Ling, L.C.; Connolly, G.; Teranishi, R. J. Agric. Food Chem. 1974, 22, 773-777.

RECEIVED

June 20, 1991


Food and Packaging Interactions II - American Chemical Society

Recommend Documents