New Zealand Journal of Crop and Horticultural Science

ISSN: 0114-0671 (Print) 1175-8783 (Online) Journal homepage: http://www.tandfonline.com/loi/tnzc20

Approaches to optimising surface coatings for fruits Nigel H. Banks , Jonathan G. M. Cutting & Sue E. Nicholson To cite this article: Nigel H. Banks , Jonathan G. M. Cutting & Sue E. Nicholson (1997) Approaches to optimising surface coatings for fruits, New Zealand Journal of Crop and Horticultural Science, 25:3, 261-272, DOI: 10.1080/01140671.1997.9514015 To link to this article: http://dx.doi.org/10.1080/01140671.1997.9514015

Published online: 22 Mar 2010.

Submit your article to this journal

Article views: 204

View related articles

Citing articles: 16 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tnzc20 Download by: [37.44.207.44]

Date: 27 January 2017, At: 05:29

New Zealand Journal of Crop and Horticultural Science, 1997, Vol. 25: 261-272 0114-0671/97/2503-0261 $7.00/0 © The Royal Society of New Zealand 1997

261

Approaches to optimising surface coatings for fruits

NIGEL H. BANKS JONATHAN G. M. CUTTING SUE E. NICHOLSON Department of Plant Science and Centre for Postharvest and Refrigeration Research Massey University Private Bag 11 222 Palmerston North, New Zealand Abstract A mathematical model predicted that final water vapour permeance in surface-coated fruits would depend upon water permeance of the coating but not the proportion of pores blocked on the fruit surface. In contrast, predicted final oxygen (O2) permeance depended upon numbers of pores blocked but not O 2 permeance of the coating. Predicted variation in internal atmosphere composition caused by coatings that blocked different proportions of pores on the model fruit surface was consistent with data from two experiments on coated apples (Malus domestica Borkh.). A new equation was developed to characterise the relationship between internal carbon dioxide (CO2) and O2 levels resulting from different coating treatments. Two graphical approaches to assess surface coatings for fresh fruits are presented. In the first, a plot of water vapour permeance against internal O2 was used to identify the most suitable of three surface coatings for reducing water loss in 'Royal Gala' apples at 20°C. The second method used a plot of internal CO2 versus internal O2 in coated fruit to identify the crop's internal lower O2 limit (LOLi), which lies just below the optimum internal O 2 level for modified atmosphere effects. Coatings containing different concentrations of carboxymethyl cellulose produced internal O2 levels ranging from almost

H96074 Received 20 November 1996; accepted 27 June 1997

0 to 16 kPa in 'Granny Smith' apples at 20°C. The LOLi of these fruit was estimated using the new equation to be c. 0.8 kPa O2. Large fruit-to-fruit variability with some coating treatments indicated that uniformity of response may be as important as average response in selection of coatings. Risks cannot be separated from benefits when using surface coatings to gain modified atmosphere benefits, making their use to achieve modified atmosphere benefits more risk-laden than for other purposes. Keywords anaerobiosis; apple; internal atmosphere; lower oxygen limit; modified atmosphere; optimisation; permeance; surface coating; water loss; wax

INTRODUCTION Edible surface coatings are applied to fruits to improve cosmetic features, such as sheen or perceived depth of colour, to reduce deterioration by suppressing water loss or to achieve modified atmosphere benefits. Recently, there has been a substantial increase in research activity on surface coatings (Hagenmaier & Shaw 1992; Baldwin 1995; Hagenmaier & Baker 1995; Nussinovitch & Lurie 1995; Gontard et al. 1996; Mannheim & Soffer 1996), made from food grade ingredients, which must have appropriate physical and chemical characteristics to achieve desired effects. External and internal environments of a fruit after harvest are optimised when key aspects of deterioration (including development of disorders) are minimised. Optimisation of a surface coating involves selection of both type and amount of coating that results in maximum levels of benefits with acceptable levels of risks. The process of optimisation therefore involves choice based on comparison of some quantitative measures of benefits and risks. This paper deals with issues relating to optimisation of surface coating

New Zealand Journal of Crop and Horticultural Science, 1997, Vol. 25

262

treatments. It uses a mathematical model to develop key concepts required for the optimisation process and to make predictions of the effects of surface coatings on gas exchange of fruits. These predictions are assessed in the light of experimental data on internal atmosphere composition of apples (Malus domestica Borkh.). The experimental data are used to illustrate use of two dimensional plots involving internal atmosphere data for identifying coatings that have optimal effects on product gas exchange.

CONCEPTUAL AND MATHEMATICAL MODELS FOR OPTIMISATION In this section, essential concepts involved in optimisation of surface coatings are outlined, a preexisting mathematical model is used to develop predictions of the outcomes of treatment with surface coatings on exchange of O2, CO2, and water vapour, and a single equation is developed to explain the relationship between internal CO 2 partial pressure (P'CQ1, Pa) on internal O2 partial pressure (p'o2, ?a) resulting from the effects of surface coatings on internal atmosphere composition. Approaches to optimisation Benefits The most straightforward benefits resulting from a surface coating treatment include changes to the way the commodity looks or feels, such as enhancement of gloss or depth of colour or reduced greasiness of the fruit surface through modification of some physical attribute of the surface (Hagenmaier & Baker 1995). Other types of benefits are achieved directly or indirectly through modification of skin permeance to gases by coating. These, in turn, affect rate of transfer of gases or the magnitude of difference between internal and external atmosphere composition, as summarised by Fick's First Law of Diffusion for product at steady state: Ap.=

"j

.fruit

M_ A

(1)

where: A = surface area of fruit (m 2 ); Ap. = difference in partial pressures of gas j between internal and external atmospheres (Pa); M= weight

of fruit (kg); pjrmt = fruit skin permeance to gasy (mol/s m2 Pa); and r. = specific rate of transfer of gas j between internal and external atmospheres (mol/kg s). Thus, for a given difference in water vapour pressure between a fruit and its surrounding environment (A/?HiOPa) a surface coating with a low water vapour permeance can reduce rate of water loss. Other benefits derive from the modified internal atmosphere that can develop inside coated commodities. Processes that are affected by reduced O2 partial pressure and, to a more limited extent, those affected by increased CO2 partial pressure can be influenced by surface coatings in this way. Suppression of yellowing, loss of texture, and other aspects of metabolism are achieved either by direct effects of modified atmospheres on the processes involved or by inhibition of respiration, which limits availability of energy for deteriorative processes and conserves energy reserves of the tissue (Kader 1992). Risks By virtue of their position on the fruit, surface coatings inevitably have some effect on gas exchange of the treated commodity. They are therefore associated with some degree of risk of suffocation. To some extent this problem can be addressed through choice of materials with appropriate permeability characteristics. There have been considerable advances in knowledge of the physical and chemical characteristics of edible films in recent times (Hagenmaier & Shaw 1992; MartinPolo etal. 1992; Wong etal. 1992; Avena-Bustillos & Krochta 1993; Avena-Bustillos et al. 1994; Koelsch 1994; McHugh et al. 1994; Mc Hugh & Krochta 1994; Gontard et al. 1996; Mannheim & Soffer 1996). Such information will be important in the identification and development of materials suitable for application to fruits as surface coatings. However, regardless of permeability characteristics of the coating film per se, interaction of a surface coating with pores and cuticle on a fruit surface can result in dramatically different overall permeance for O2, depending on whether the pores are blocked or merely covered (Banks et al. 1993). Blockage of pores can result in very low values for skin permeance to O 2 which can depress p'Ol excessively and result in anaerobiosis, fermentation, development of off-flavours (Kester & Fennema

Banks et al.—Surface coatings for fruits 1986; Cohen etal. 1990; Hagenmeier& Shaw 1992; Mannheim & Soffer 1996) and uneven ripening (Smith et al. 1987; Meheriuk & Lau 1988; McGuire & Hallman 1995). These effects are exacerbated by elevated rates of respiration resulting from keeping produce at high temperatures, an effect which can be inferred from Equation 1 given that respiration is numerically equivalent to rate of oxygen transfer (r Oi , mol/kg s) for a fruit at, or close to, steady state. Separable mechanisms for benefits and risks Where the mechanisms involved in benefits and risks are to some extent distinct, plots of one or more measures of benefit against one or more indicators of risk provide an effective decision support tool. Cosmetic benefits are most readily separated from risk, essentially because they are not necessarily dependent upon modification of skin permeance for their achievement. For example, many wax coatings impart a worthwhile degree of gloss even when applied so diluted that they have minimal effect on gas exchange. To a limited extent, use of surface coatings to retard weight loss can be thought of in the same way because water is known to migrate through commodity surfaces via additional routes to those followed by the permanent gases (Banks et al. 1993). Unfortunately, attempts to substantially reduce weight loss may result in suffocation because heavier applications of coating can excessively impair exchange of the respiratory gases. In this paper, we use the model fruit system described by Banks et al. (1993) to explore the importance of permeance to water and O2 (p,', 0 and PQ^ mol/s m2 Pa) of, and pore blockage by, the coating in relation to risks and benefits from coating treatments. For coatings intended to reduce water loss, comparing relative effects on exchange of water vapour and other gases provides a straightforward basis for optimisation. Benefit can be expressed as reduction in Pt'uo whilst the principal risk arises from the potential for excessive depression of p'O2 in the coated commodity. Values approaching or below the internal lower O2 limit (LOL1, Beaudry et al. 1992; Yearsley et al. 1996) indicate that fruit are at risk from anaerobiosis. This approach is used to compare suitability of three surface coating materials for reducing weight loss in apples in the experimental study described below.

263 Common mechanism for benefits and risks For a coating intended to achieve modified atmosphere benefits, both benefit and risk are likely to derive from depression of p'O2. Again, LOU provides an indication of the point at which depression of p'O2 moves from benefit to risk and can be used to distinguish between near-optimal and non-acceptable coatings for the crop under prevailing conditions. This concept is critical for achieving modified atmosphere benefits, since super-optimal internal O2 partial pressure provides little benefit over storage in air, whereas sub-optimal levels can incur rapid loss of quality as a result of anaerobiosis. We propose, and demonstrate below, that approximate LOL's can be identified from plots of Pco2 versus p'O2 in fruit with widely differing values for skin permeance to gases, achieved by surface coatings. Predictions with pre-existing mathematical model Inputs The mathematical model described by Banks et al. (1993) was used to generate predicted values for internal atmosphere composition (p'O2 and PC02) and skin permeances to O2, CO2, and water vapour achieved with a tightly adhering coating that completely covered the cuticle but which only blocked a certain proportion of the pores on the fruit surface. Input data used in generating these predictions are shown in Table 1. Predictions The values of input variables used in the model produced predictions of permeances in non-coated fruit of c. 360 pmol/s m2 Pa for both O2 and CO2 and 80 000 pmol/s m2 Pa for H2O (highest values shown in Fig. 1). A theoretical, linear relationship between P^ o , and PQ, was generated by varying the number of pores blocked by coating in the model (Fig. 1). The slope of the relationship between P^o and PQ^ was slightly higher than that between PQ0^ and PQ, (see caption to Fig. 1) but the absolute magnitude of the change was orders of magnitude smaller than the absolute value of initial P ^ o ; relative changes in P y ^ 0 were negligible compared to relative changes in PQ, (F'g- !) • The model predicted that application of surface coatings with PH,0 values ranging from 20 nmol/s

New Zealand Journal of Crop and Horticultural Science, 1997, Vol. 25

264 2

m Pa (same as the fruit cuticle) to 2.0 x 104 mol/s m2 Pa (effectively no barrier to water movement = control) would depress permeance of the coated fruit to water vapour to an extent that varied inversely with permeance of the coating (Fig. 1) but that, in relative terms, this effect would be independent of the extent of pore blockage (Figs 1 and 2). Increased levels of pore blockage substantially depressed p'o2 in the model fruit system (Fig. 2).

This effect was associated with an increase in p'C02 (Fig. 3), though the magnitude of the negative slope of this relationship declined as p'O2 was depressed towards zero, becoming positive at levels of p'O2 less than 5 kPa before eventually swinging strongly negative at very low p'O2. Variation of PQ, f°r me coating from 0.1 to 10 times the value used in Fig. 2 generated plots of p'O2 versus pore blockage that were indistinguishable from that shown in Fig. 2 (data not shown).

Table 1 Data input values for key variables used in the fruit system model. Variable Area of pores on fruit surface Fruit surface area

Symbol

Cuticular permeance toO 2 Cuticular permeance toCO 2 Cuticular permeance toH 2 O Effective pore permeance toO 2 Effective pore permeance toCO 2 Effective pore permeance toH 2 O Maximum respiratory O2 uptake

Units

Source

—

0.05% A

m2

assumed

A

0.0166

m2

*.

2200

Pa

h

700

Pa

h

5

-

M Ph..

0.16 21

mol/s m2 Pa kg kPa

derived from mass; Clayton et al. (1995) after Banks et al. (1993) after Banks et al. (1993) after Banks et al. (1993) assumed assumed Nobel (1991)

Pco2

0

kPa

assumed

p ' coal

2.0 x 10"10

mol/s m2 Pa

r>' coat

8.0 x 10-'°

mol/s m2 Pa

6.4x10""

h

Fruit mass External O2 partial pressure External CO2 partial pressure Coating permeance toO 2 Coating permeance toCO 2 Coating permeance toCO 2

Value

PL

p ' coal "HiO

from 2.0 x lO"8 (low), mol/s m2 Pa 7 through 2.0 x 10~ (normal), 2.0 x 10-* (high), to 2.0 x 104 (control)

p ' cut

8.0 x 10-' 2

mol/s m2 Pa

after Hagenmaier & Shaw (1992) after Hagenmaier & Shaw (1992) after Hagenmaier & Shaw (1992); highest value = control (close to nil barrier effect) assumed

p 1 cut

6.4 xlO-' 1

mol/s m2 Pa

assumed

8

mol/s m2 Pa

assumed

7

mol/s m2 Pa

after Banks et al. (1993) after Banks et al. (1993) after Banks et al. (1993) after Dadzie et al. (1996)

•nro.

1

p cut -

p pores O2

8.0 x 107.0 x 10-

p- pores MX):

5.95 x 10-

7

mol/s m Pa

p'pores ' H:O

8.75 x 10-7

mol/s m2 Pa

'•of

2.5 x 10-7

mol/kg s

2

Banks et al.—Surface coatings for fruits

400

100 80

300

60 E 200 40

o Q.

100

20

o

o 100 200 300 2 PQ2 (pmol / s m Pa)

400

0

100

Fig. 2 Predicted changes in internal oxygen (p'o^, Pa) and water vapour permeance (P^ 0, mol/s m2 Pa) with variation in pore blockage of a model fruit system by a tightly adhering coating with attributes as described in Table 1.

20

Fig. 3 Predicted changes in internal carbon dioxide (p' COl , Pa) and respiratory quotient (RQ) with variation in internal oxygen (p'Oy, Pa) achieved by differing degrees of pore blockage of a model fruit system by a tightly adhering coating with "normal" attributes as described in Table 1.

Pa)

Fig. 1 Predicted relationships between p^0 and />„ (mol/s m 2 Pa;/> c ' o = 5.3 x KT11 +0.85 PQ ) and PH o and P£ (P^ 0 = l.o'x 1(T 8 + : 1.22 PQ J achieved by differing degrees of pore blockage of a model fruit system by a tightly adhering coating with "normal" attributes as described in Table 1.

265

40 60 80 pore blockage (%)

100

266

New Zealand Journal of Crop and Horticultural Science, 1997, Vol. 25

Discussion The values predicted for / ^ and P^o by t n e model for non-coated fruit were similar to those reported for 'Granny Smith' apples by Dadzie et al. (1996); that for P^ 0 was similar to values recently obtained in our laboratory for this cultivar (K. Maguire unpubl. data) and reported for apples by a number of workers (e.g., Smith 1933). The predicted slope of the relationship between PQO and PQ (0.85), was essentially identical to that "which would be expected from the relative diffusivities of the two gases in air if, in relative terms, the gas exchange potential of blocked pores was essentially eliminated. Although the absolute slope of the relationship between Pt\ 0 and PQ in the model fruit system was slightly greater, its negligible variation relative to the absolute magnitude of P^o underlines the large differences in relative contributions made by cuticle and pores to total /"H,Om t n e m ° d e l fruit system. The intercepts of both relationships represent values close to the permeance of coated cuticle for each gas. For PH2O, this w a s half of the value for noncoated fruit when cuticular and coating permeance were equal (20 nmol/s m2 Pa). These predictions confirm the importance of P ^ I 0 °f tne coating material in determining P^ 0 of the coated fruit. The data presented in Figs 1 and 2 demonstrate clearly that increased pore blockage resulted in increased depressions of p'O2. Interestingly, variation in both Po^ a n d p'O2 f° r coated fruit was affected only very slightly by varying permeance to O2 of the coating applied to the model system by two orders of magnitude (from the least to the most permeable materials described by Hagenmaier & Shaw 1992). This suggests that a pore blocked with currently known coating materials would allow very little gas exchange. Thus, problems with suffocating fruit are perhaps less likely to be solved by a search for more permeable materials than by identification of formulations that do not block pores. In summary, whereas changes to ^H,O achieved by coating would depend largely on the permeance to water of the coating layer overlying the cuticle, depression in p'o2would depend more on the proportion of pores blocked by the coating. Other evidence consistent with this proposition has recently been published for citrus (Mannheim & Soffer 1996). Given the range of ratios of permeances to H2O and O2 in coating materials used for fruits (between

c. 40 and 50 000; Hagenmaier & Shaw 1992) this places a major constraint on the use of surface coatings to reduce water loss in fruits until formulations that avoid pore blockage can be devised. Development of new model This section outlines the development of a straightforward, mechanistic description of the basis for dependence of Pco2 on P02 m coated fruits. Assuming uniform internal atmosphere composition and that />£Oi = 0 (where the superscript e denotes external), it follows from Equation 1 that: 'CO,

PCO2

Pco/mU

M

(2)

A

Given the empirical relationship between t o , a n d />o2recentry characterised by Dadzie et al. (1996), Equation 2 can be restated:

H

P'o2 2O,

101.325

PCO2 =

(3) with the left and right bracketed terms corresponding to the aerobic and anaerobic components of total CO2 production, respectively. Here: r ^ = maximum rate of CO2 production when O2 is not limiting (mol/kg s; this can be calculated from measured values for rco and p'02in air if Arj is known); k\ = a parameter equivalent to the Michaelis-Menten constant (Pa); p"" = total (atmospheric) pressure (Pa) and k2 and &3 are empirical constants. The constant 101.325 maintains consistency of units when gas composition is expressed as partial pressures (as recommended by Banks et al. 1995) rather than in atmospheres (as previously used by Banks et al. 1993). PQ is thought to be pore dominated (i.e., pores contribute the vast majority to total permeance of fruit skins to this gas; Banks et al. 1993). If Po of the cuticle (PQC"') is negligible, and assuming that the permeance of a blocked pore is effectively zero, total permeance to CO 2 (PQ0 , mol/s m2 Pa) of a fruit with a tightly adhering coating (Banks et al. 1993) would be related to the number of pores blocked and would be given by:

M A

Banks et al.—Surface coatings for fruits Pco^^+kyPo,

267 (4)

where: k$ = a parameter that represents coated cuticular permeance to CO2 (Paj"', mol/s m2 Pa); and k5 = a proportionality constant that adjusts for the difference in diffusivities of the two gases in open pores (theoretically, with a value of c. 0.85). This was shown to hold for the model fruit system in Fig. 1. Assuming that the respiratory quotient (RQ) of fruit when O2 is non-limiting is unity (Yearsley et al. 1996), Equation 1 and the aerobic component of Equation 3 can be used to calculate PQ : max

rr

O2

• n'

M A

P02

(5)

Substituting Equations 4 and 5 into Equation 3 and rearranging, it can be shown that for such coated fruit: Po2 + p'o2)

n'

Pco2

A

k4

M

max

P"" 1O1325-(A:2 + P02)

(6)

k

s-Po2

pbi -Po 2 )\kx + + Po2 )

In the experimental study described below, Equation 6 is shown to provide approximate fits to the relationships between p'C02and p'O2 indatasets obtained with coated apples.

EXPERIMENTAL STUDY This section outlines two experiments conducted with apples which produced data consistent with the predicted relationships between p'c02 and p'o2 described above and illustrate the utility of two dimensional plots of benefits and risks in the optimisation of surface coating treatments.

Materials and methods Experiment 1 Freshly harvested, mid season 'Royal Gala' apples {Mains domestica Borkh.; count 125; average weight 0.148 kg) were obtained from the Fruit Crops Unit, Massey University, Palmerston North, New Zealand and equilibrated to 20°C for

24 h. Groups of 12 randomly selected fruit were allocated to one of four surface coating treatments: control (no coating), polyethylene wax (undiluted "Citruseal"; Milestone Chemicals, Australia), a shellac-based coating (undiluted "Citrus Gleam"; Castle Chemicals, Australia), and carboxymethylcellulose (CMC; 2% in aqueous solution; low viscosity; BDH Chemicals, United Kingdom). Wetting agent ("Pulse"; 0.1% w/v; Monsanto, New Zealand) was added to each coating mixture to ensure even wetting of the fruit surface. Fruit were dipped for 1 s then allowed to dry on a rack. Rate of water loss was estimated from change in weight measured over a 24-h period commencing 18 h after treatment, in an airstream with average velocity of 3 m/s and water vapour pressure difference between fruit and airstream (Ap H i 0 ) of c. 1.2 kPa. Average ApHiOwas estimated using wet and dry bulb temperatures of the air and fruit surface temperature using standard psychrometric equations (Campbell 1977). Fruit surface area was estimated from weight using a regression equation (Clayton et al. 1995). Values of of each fruit were estimated using a re-arrangement of Equation 1. Internal atmospheres were sampled destructively by direct removal (Banks 1983) from the core cavity 24 h after P^ 0 determination. Sample O2 and CO2 contents were determined on 100 mm3 aliquot samples using an O2 electrode (Citicell C/S type, City Technology Ltd, London, United Kingdom) in series with a miniature infra-red CO2 transducer (Analytical Development Company, Hoddesdon, United Kingdom) with C^-free N 2 as carrier gas (flow rate 580 mm3/s). Ambient pressure data were collected with a pressure transducer (Barigo Electronic Altimeter, Barigo Barometerfabrik GmBH, D-7730 VillingenSchwenningen). Experiment 2 Mid-season harvested 'Granny Smith' apples (count 125; av. weight 0.149 kg) were obtained from a Hawkes Bay orchard and stored in air at 0°C for 2 weeks then equilibrated to 20 ± 1 °C for 48 h before experimentation. Sixty fruit were cannulated to enable repeated sampling of internal atmosphere composition (Banks 1983). Initial respiration rates were determined by measurement of the change in partial pressure of CO2 within 5.8 x 10"4 m3 opaque, sealed containers over 30 min. The experiment was

New Zealand Journal of Crop and Horticultural Science, 1997, Vol. 25

268

a completely randomised design with five treatments: control (not coated, 10 replicates) or coated with 1% (20 replicates), 2, 3, or 4% (10 replicates each) aqueous solution of CMC with wetting agent. Fruit were equilibrated for 4 days in the dark in air at 20°C before measurement of internal atmosphere composition from the cannulae. Data analysis Data on gas composition were converted to final format using formulae presented by Banks et al. (1995). Data were subjected to analysis of variance (ANOVA) and/or non-linear regression using the Statistical Analysis System (SAS 1990). Results Experiment 1 Variance of p'O2 data in Experiment 1 was heterogeneous between treatments (Table 2). However, apparent levels of significance were so high (P < 0.001 for comparing all treatments) that it was clear that there were substantial differences between the treatments for all variables measured and no transformations of data were used in the analysis. Shellac coating was the most effective treatment for decreasing weight loss, with final P ^ 0 33% less than control values (P < 0.001), compared to 28% for polyethylene, whilst the final value for fruit coated with CMC was similar to the controls (Table 2). However, shellac treatment also depressed average p'O2 to a much greater extent than polyethylene or CMC. Individual fruit values of p'O2 were much more variable in CMC-coated fruit than in the other treatments (Fig. 4A); p'Ol was uniformly low in shellac-treated fruit and somewhat depressed relative to controls in polyethylenetreated fruit.

10

Control CMC (2%) Polyethylene Shellac

control polyethylene CMC shellac

15

20

P i 2 (kPa)

Fig. 4 Plots of individual values for: A, water vapour permeance (P^ 0, mol/s m2 Pa); and B, internal carbon dioxide (p'C02, Pa) against internal oxygen (p'Oi, Pa) of 'Royal Gala' apples (Malus domestica) treated with three surface coating materials and kept at 20°C.

Although there was a wide spread of values for internal atmosphere composition in the four treatments, all p'C02 data lay on a consistent curve when plotted against p'O2 (Fig. 4B). As p'O2 was depressed by coating, p'C02was increased, but the magnitude of the slope of the relationship declined progressively so that a steady level of between 9 and 10 kPa p'C02 was reached for values of p'O2

Table 2 Internal oxygen (p' Ol ) and carbon dioxide (PcO2) partial pressures and permeance to water vapour (P^ 0 ) in 'Royal Gala' apples (Malus domestica) at 20°C—66 h after different coating treatments (means + SD; n = 12). (CMC = carboxymethyleellulose.) Treatment

O • • A

Pb2 (kPa)

(kPa)

18.4 + 0.39 8.2 ± 3.70 13.0 ±1.49 1.6 ±0.80

2.4 ± 0.46 8.2+1.78 6.2 ±0.58 11.5 ±1.24

Pco2

2

(nmol/s m2 Pa) 10.5 ±1.36 10.7 ±1.07 7.5 ± 0.99 7.0+ 1.17

Banks et al.—Surface coatings for fruits

269

Fig. 5 Plot of individual values for internal carbon dioxide (p' c02 , Pa) against internal oxygen (p'o^, Pa) of 'Granny Smith' apple (Malus domestica) fruit kept at 20°C demonstrating the potential to identify lower oxygen limits of apples using with carboxymethylcellulose surface coatings. Data points in the shaded zone are from fruit at risk of fermentation.

X A o O

control coated, 1% coated, 2% coated, 3% coated, 4%

20

between 7 and 3 kPa. Below 2 kPa p'O2, there was a modest upswing in values for p'COr Most values for shellac-treated fruit lay on or to the left of this point, whereas those for CMC-treated fruit were spread over a wide range which lay to the right and values for fruit in the polyethylene coating treatment were clustered still closer to control fruit values. Experiment 2 CMC coatings had highly significant effects on internal atmosphere composition of'Granny Smith' apples (Table 3). Depression of p'o, increased with increased concentration of CMC solution applied, though the extent of depression achieved by the 2% coating left little scope for further reduction with the higher CMC concentrations. In contrast, p'C02 increased progressively with CMC concentration

Table 3 Internal oxygen ( p g , ) and carbon dioxide ( PcoJ P a r t ' a ' pressures in 'Granny Smith' apples (Malus domestica) at 20°C—96 h after coating with 0 (n = 20), 1, 2, 3, or 4% (n = 10) carboxymethylcellulose (CMC) solutions (means ± SD). CMC concentration (%)

p'O2 (kPa) 14.1 ± 1.71 4.8 ±2.34 0.43 + 0.227 0.4410.152 0.33 + 0.063

p'O2

(kPa) 5.4 ± 0.60 9.6 ± 0.86 21.2+ 2.96 31 ±10.8 48.1 ± 5.34

in the coating solution (Table 3). There was a consistent relationship between the levels of p'C02 and p'O2 within the fruit (Fig. 5). Progressively greater depressions in p'Oi achieved by coating were linked to increases in p'COr The slope of this relationship in the region of 15 kPa p'Oi was close to —0.4, gradually flattening off as p'o, values approached 3—5 kPa O2. Below this level, there was some evidence of a decline in p'C02 values before, at levels of p'Oi below 1 kPa, p'CO2 increased dramatically. Non-linear regression ofp'CO2 on p'o^ using Equation 6 could only be made to produce significant fits to the data if several of the six parameters were fixed before analysis, presumably because of the large number of parameters involved. The fitted line shown in Fig. 5 was developed by fixing TQQ= 144.3 + 1.74 nmol/kg s (measured; n = 60), pe = 20.1 kPa (measured) and values of k2 = 0.6 kPa and £3 = 6 (slightly amended from values presented by Dadzie et al. (1996; 0.7 and 5, respectively) for improved visual fit). Values for k\, k4, and k5 were then fitted by non-linear regression (0.92 ±0.150 kPa, 51.6 ± 2.94 pmol/s m2 Pa and 0.844 ± 0.026, respectively). A sensitivity analysis of parameters k\,k2, £3, £4, and £5 indicated that there was quite a limited range (e.g., 10—20% of final values) for each parameter within which reasonable overall visual fit could be obtained. An estimate of LOU for the fruit (0.79 kPa) was obtained as the p'Ol value at which calculated RQ = 1.1 (Yearsley 1996) i.e., when:

270 0.1 -

New Zealand Journal of Crop and Horticultural Science, 1997, Vol. 25

=0

(7)

DISCUSSION The shapes of the relationships between p'C02 and />o, obtained experimentally (Figs 4B and 5) were very similar to the predictions made in Fig. 3, indicating that the model fruit system provides a reasonable approximation of reality. As pores became blocked, so p'c02 increased and />0, decreased in response to relative changes in transfer (related to production and consumption) of, and permeance to, the two gases. At levels of p'Ol very close to those found in air, RQ would remain constant at approximately unity and exchange of O2 and CO2 would be pore dominated. At this point, the magnitude of the slope of the line would be the inverse of the ratio of effective permeances of the pores to the two gases (—1.18). At all levels of p'o, developed in the model fruit system and experimental fruit, the slope was less negative than this because pores were never totally dominant. Indeed, they became progressively less so as more pores were blocked and the cuticular contribution to total gas exchange increased, thereby decreasing the ratio of PQI to P^o, • Further increases in pore blockage resulted in depression of p'o, to a point at which respiration was diminished and the slope of the line became increasingly positive, except in Experiment 1 in which a number of different coating materials were used. Finally, as fermentation became dominant at very low levels of p'o,, RQ became significantly greater than unity and p'COl increased dramatically. Fitting of Equation 6 to experimental data in Fig. 5 yielded further valuable insights into factors affecting the outcome of coating fruits. From the above, it is clear that permeance of the cuticle to CO2 has a major impact on the level of CO2 that would accumulate inside a fruit for a given depression of p 0 ,. The values of £4 generated by both the model and the visual fit of Equation 6 can only be regarded as tentative estimates (51.6—64 pmol/s m2 Pa) but do provide a useful starting point to explore further issues relating to increases in p'ccn in coated fruits. The value of £5 for Equation 6 obtained in Experiment 2 (0.844) was consistent

with predictions made with the model fruit system, supporting the suitability of the assumptions made in generating the predictions. Substantial deviation from this value would only be expected if the coating material was extremely permeable to gases, in which instance movement through the coating itself, rather than just elimination of a pore's function, would become a contributing factor. Gontard et al. (1996) have recently shown large variation with relative humidity in the absolute values of P^Oi to Z ^ , and the ratio of one to the other, in edible films; it seems likely that there would be interesting interactions between environmental humidity and the overall permeance of coated fruits to these two gases. The value of k\ required to obtain a close fit (0.92 kPa) was less than half that recently published by Dadzie et al. (1996; 2.2 kPa for 'Granny Smith') but increases of 20% caused marked deviation of the fitted line from the data. Further work with a greater focus on low levels of O2 than those used by Dadzie et al. (1996) would be worthwhile to confirm the shape of the respiration versus p'O2 curve to assist in predicting responses of apples to both controlled atmospheres and surface coating treatments. Parameters k2 and kj, influenced the steepness of rise in p'C02 at very low levels of p'o^ here values 15% higher or lower than those fitted caused visible deviation from the data. Alternative values would be required to explain the shape of the curve generated for the 'Royal Gala' fruit in Experiment 1, though this was not attempted as other factors such as variation in differential permeability of the coating materials could have contributed to the observed data. The estimate of LOU for the 'Granny Smith' apples at 20°C in this work (0.79 kPa) was within the 95% confidence interval of values for anaerobic compensation point reported for 'Cox's Orange Pippin' apples at 24°C (0.70-0.79 kPa; Yearsley et al. 1996) and slightly lower than estimates obtained on the basis of RQ (1.04-1.20 kPa; Yearsley et al. 1996). Use of surface coatings at a range of concentrations therefore provides the potential to identify LOU values without resort to resourceintensive controlled atmosphere experiments. It should be noted that this identification is confounded with variation in />c02when using surface coatings for this purpose. Such estimates

Banks et al.—Surface coatings for fruits of LOU may therefore be used most robustly for optimisation of surface coating treatments, though only small effects of elevated CO2 partial pressures of up to 8 kPa on LOU values have been observed in controlled atmosphere experiments (Yearsley unpubl. data). The lower O2 limit of the 'Royal Gala' apples in this study (c. 2.5 kPa O2) appeared to be considerably higher than that for 'Granny Smith' fruit. This could represent genuine differences in fruit physiology between the two cultivars, as Dadzie (1992) has observed that withintissue gradients in gas concentrations in 'Royal Gala' apples can be quite substantial; this could increase the apparent LOU of the bulk of the tissue if one area of the fruit was essentially anaerobic whilst others retained adequate O2. However, as noted above, the use of different coating materials with different permeability characteristics provides an alternative explanation for the different slope in the data for the shellac treatment and means that further work would be required to confirm the value ofLOUfor'Royal Gala'. Data on the average values for gas exchange variables in coated fruits, such as those presented Tables 2 and 3, provided a useful summary of the relative magnitude of overall effects of different treatments. From these, it was clear that shellac and polyethylene coatings were more effective in reducing water loss than CMC and that progressively higher concentrations of CMC coating achieved greater levels of modification of internal atmosphere composition. Given that p'O2 cannot fall below zero nor be higher than about 21 kPa, variance of p'Ol is likely to be lower at the extremes than near the middle of the range, a feature confirmed by Tables 2 and 3. Plots of P^o against p'Ol and p'co^ against p'on provided a visual tool for the rapid assessment of relative benefits (e.g., depression of ^,0) an(^ risks (depression of p'O2) of different surface coating treatments. In Experiment 1, CMC treatment provided essentially no benefit in terms of reduced potential weight loss but gave a highly unpredictable risk of suffocation. Shellac coating gave rise to predictable but unacceptably high risk whereas the risk of excessively low internal p'Ol values in polyethylene-coated fruit under the conditions of the experiment appeared quite small. None of the data points lay in the desirable high benefit, low risk lower right hand quadrant of Fig. 4A but for a given level of risk, polyethylene

271

coating had the greatest level of benefit. Fig. 5 demonstrated the potential to use plots of this kind to optimise surface coating treatments for modified atmosphere benefits by identifying a treatment concentration that reduced p'O2 to a level just above the LOU (shaded region of Fig. 5). However, in the instance of the CMC coating used for this exercise, this was hampered by high variability of p'Ol within a single treatment. Clearly, uniformity of response to coating would be of equal importance to average response in terms of selecting suitable coating treatments; plots of the individual data provide an ideal tool for this combined purpose. In conclusion, this study has made predictions of gas exchange characteristics in fruit treated with surface coatings using a model fruit system. These predictions were highly consistent with data obtained from experiments in which apples were treated with different surface coatings, in terms of the shapes of relationships between key attributes. The study emphasises the need to optimise surface coating treatments in terms of both selection of appropriate materials and identification of appropriate concentrations to apply to achieve desired benefits without excessive risks. It provides a conceptual framework for the assessment of benefits and risks in this optimisation process and some simple tools by which optimisation may be achieved. ACKNOWLEDGMENTS This research was funded by the Foundation for Research Science and Technology. It was carried out with facilities provided by the Lottery Sciences Commission. We thank Dussek Campbell Ltd for supply of Citrus Gleam coating. REFERENCES Avena-Bustillos, R. J.; Krochta, J. M. 1993: Water vapor permeability of caseinate-based edible films as affected by pH, calcium cross-linking, and lipid content. Journal of food science 58: 904-907. Avena-Bustillos, R. J.; Krochta, J. M.; Saltveit, M. E.; Rojas-Villegas, R. J.; Sauceda-Pérez, J. A. 1994: Optimization of edible coating formulations on zucchini to reduce water loss. Journal of food engineering 21: 197-214. Baldwin, E. A. 1995: Edible coatings for fresh fruits and vegetables: past, present, and future. Pp. 25-64 in: Edible coatings and films to improve food quality. Krochta, J. M.; Baldwin, E. W.; NisperosCarriedo, M. O. (ed.). Lancaster, Basel, Technomic Publishing Company.

272

New Zealand Journal of Crop and Horticultural Science, 1997, Vol. 25

Banks, N. H. 1983: Evaluation of methods for determining internal gases in banana fruit. Journal of experimental botany 34: 871-879. Banks, N. H.; Cleland, D. J.; Cameron, A. C.; Beaudry, R. M.; Kader, A. A. 1995: Proposal for a rationalized system of units for postharvest research in gas exchange. HortScience 30: 1129-1131. Banks, N. H.; Dadzie, B. K.; Cleland, D. J. 1993: Reducing gas exchange in fruits with surface coatings. Postharvest biology and technology 3: 269-284. Beaudry, R. M.; Cameron, A. C.; Shirazi, A.; DostalLange, D. 1992: Modified-atmosphere packaging of blueberry fruit: effect of temperature on package O2 and CO2 Journal of the American Society for Horticultural Science 117: 436-441. Campbell, G. S. 1977: An introduction to environmental biophysics. New York, Springer-Verlag. Clayton, M.; Amos, N. D.; Banks, N. H.; Morton, R. H. 1995: Estimation of apple fruit surface area. New Zealandjournal of crop and horticultural science 23: 345-349. Cohen, E.; Shalom, Y.; Rosenberger, I. 1990: Postharvest ethanol buildup and off-flavour in 'Murcott' tangerine fruits. Journal of the American Society for Horticultural Science 115: 775-778. Dadzie, B. K. 1992: Gas exchange characteristics and quality of apples. Unpublished PhD thesis, Massey University, Palmerston North, New Zealand. Dadzie, B. K.; Banks, N. H.; Cleland, D. J.; Hewett, E. W. 1996: Changes in respiration and ethylene production of apples in response to internal and external oxygen partial pressures. Postharvest biology and technology 9: 297-309. Gontard, N.; Thibault, R.; Cuq, B.; Guilbert, S. 1996: Influence of relative humidity and film composition on oxygen and carbon dioxide permeabilities of edible films. Journal of agricultural and food chemistry 44: 1064-1069. Hagenmaier, R. D.; Baker, R. A. 1995: Layered coatings to control weight loss and preserve gloss of citrus fruit. HortScience 30: 296-298. Hagenmaier, R. D.; Shaw, P. E. 1992: Gas permeability of fruit coating waxes. Journal of the American Society for Horticultural Science 117: 105-109. Kader, A. A. 1992: Modified atmospheres and lowpressure systems during transport and storage. Pp. 58-64 in: Postharvest technology of horticultural crops, 2nd. edition. Kader, A. A. (ed). Special publication 3311, Co-operative Extension, University of California, Division of Agriculture and Natural Resources.

Kester, J. J.; Fennema, O. R. 1986: Edible films and coatings: a review. Food technology 40(12): 47-59. Koelsch, C. 1994: Edible water vapor barriers: properties and promise. Trends in food science and technology, March 5: 76-81. Mannheim, C. H.; Soffer, T. 1996. Permeability of different wax coatings and their effect on citrus fruit quality. Journal of agricultural and food chemistry 44: 919-923. Martin-Polo, M.; Voilley, A.; Blond, G.; Colas, B.; Mesnier, M.; Floquet, N. 1992: Hydrophobic films and their efficiency against moisture transfer. 2. Influence of the physical state. Journal of agricultural and food chemistry 40: 413-418. McGuire, R. G.; Hallman, G. J. 1995: Coating guavas with cellulose- or carnauba-based emulsions interferes with postharvest ripening. HortScience 30: 294-295. McHugh, T. H.; Aujard, J. F.; Krochta, J. M. 1994: Plasticized whey protein edible films: water vapour permeability properties. Journal of food science 59: 416-419 & 423. McHugh, T. H.; Krochta, J. M. 1994: Sorbitol- vs glycerol-plasticized whey protein edible films: integrated oxygen permeability and tensile property evaluation. Journal of agricultural and food chemistry 42: 841-845. Meheriuk, M.; Lau, O. L. 1988: Effect of two polymeric coatings on fruit quality of 'Bartlett' and 'd'Anjou' pears. Journal of the American Society for Horticultural Science 113: 222-226. Nussinovitch, A.; Lurie, S. 1995: Edible coatings for fruits and vegetables. Postharvest news and information 6: 53N-57N. SAS 1990: Cary, NC, USA, SAS Institute Inc. Smith, S.; Geeson, J.; Stow, J. 1987: Production of modified atmospheres in deciduous fruits by the use of films and coatings. HortScience 22: 772-776. Smith, W. H. 1933: Evaporation of water from apples in relation to temperature and atmospheric humidity. Annals of applied biology 20: 220-235. Wong, D. W. S.; Gastineau, F. A.; Gregorski, K. S.; Tillin, S. J.; Pavlath, A. E. 1992: Chitosan-lipid films: microstructure and surface energy. Journal of agricultural and food chemistry 40: 540-544. Yearsley, C. W.; Banks, N. H.; Ganesh, S.; Cleland, D. J. 1996: Determination of lower oxygen limits for apple fruit. Postharvest biology and technology 8: 95-109.