Postharvest Biology and Technology 35 (2005) 309317

MA storage of Cavendish bananas using silicone membrane and diffusion channel systems
Opal J. Stewarta,b, , G.S.V. Raghavanb , Kerith D. Goldena , Y. Gari epyb
a

Department of Basic Medical Sciences (Biochemistry Section), University of the West Indies, Mona Campus, Kingston 7, Jamaica b Department of Agricultural and Biosystems Engineering, McGill University, Macdonald Campus, 21, 111 Lakeshore Road, Ste-Anne-de-Bellevue, Quebec, Canada H9X 3V9 Received 18 March 2004; received in revised form 18 September 2004; accepted 17 October 2004

Abstract A study was conducted to evaluate the potential of the passive silicone membrane and diffusion channel systems to preserve the quality and extend the shelf-life of bananas. Cavendish bananas were stored for 42 days at 15 C under MA conditions using silicone membrane and diffusion channel systems. The respiration rate under MA at 15 C was estimated as 20% or 60% less, or 20% greater than that measured in regular atmosphere at the same temperature. Three different silicone membrane areas (50.29, 98.56 and 158.43 cm2 ) and three different diffusion channel lengths (4, 7 and 10 cm) were tested. Results showed that the estimation of a 60% reduction in respiration rate was most accurate. The smallest area of silicone membrane achieved gas levels of 3.5% CO2 /3% O2 in about 10 days while the shortest diffusion channel achieved 5% CO2 /3% O2 , in 1216 days. Fruit in these atmospheres remained unripe for 42 days, had harvest-fresh appearance, good colour, minimum mould and excellent marketability compared with controls and fruit stored in different gas compositions. In general, the silicone membrane system was found to be superior; it achieved stability more quickly than the diffusion channel system, maintained more stable gas levels throughout storage and had better physiological and sensory ratings. The diffusion channel system had higher CO2 levels that may have resulted in peel discoloration in some chambers and may have affected other quality attributes. The silicone membrane system offers an inexpensive and easy to use alternative to the traditional methods used for MA storage of bananas. 2004 Elsevier B.V. All rights reserved.
Keywords: Banana; Diffusion channel; Modied atmosphere; Musa sp; Postharvest; Silicone membrane

1. Introduction

Corresponding author. Present address: Science Department, Stewart Postharvest Solutions Ltd., 32 Southmoor Road, Jericho, Oxford OX2 6RD, UK. Tel.: +44 1865 510252; fax: +44 20 7900 276. E-mail address: opal stewart@yahoo.com (O.J. Stewart). 0925-5214/$ see front matter 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.postharvbio.2004.10.003

The main factors affecting commercial quality of bananas are concerned with appearance rather than with internal fruit quality. These include absence of bruises, blemishes and rots, a bright yellow peel colour and adequate shelf-life (Marriot, 1980). At every stage,

310

O.J. Stewart et al. / Postharvest Biology and Technology 35 (2005) 309317

from harvesting to retailing, breaking, bruising and scarring with attendant fungal infection and rotting by many different organisms, are ever present problems (Seymour et al., 1993). The presence of supercial blemishes, caused partly prior to harvest but mainly incurred during the postharvest period, is the main cause of the downgrading of bananas from premium quality to lower grades, resulting in large loss of fruit (Seymour et al., 1993). The search for methods to prolong the shelf-life of bananas has been the object of several studies, chief among which is the investigation into various MA systems (Marriot, 1980). Most MA systems are complex in nature and their extensive use has been hindered by the substantial costs incurred. As a result, the variety of techniques available to large-scale banana producers may not be readily accessible by farmers operating on a smaller scale (Marchal, 1998). For some, future storage lies in the ability to offer effective, low cost storage alternatives for shipping to more distant, yet available markets and to facilitate the use of slower, less expensive methods of transportation. Storage of bananas under MA has been primarily achieved using low-density polyethylene (LPDE) bags (Stiles, 1991; Rahman et al., 1995; Chamara et al., 2000). Shelf-life may be further increased by using an ethylene absorber such as potassium permanganate in combination with polyethylene lms (Scott and Roberts, 1966; Jiang et al., 1999; Chamara et al., 2000). Although the traditional MA packaging for bananas has been restricted to the use of plastic lm bags (Marchal and Nolin, 1990; Maneerat et al., 1997), other systems have been developed and commercially implemented. Banana ripening can be delayed for several weeks when ventilated in gas mixtures with low O2 and high CO2 (Mapson and Robinson, 1966). Under conditions of 1.52.5% O2 and 710% CO2 , the extension of the preclimacteric period is about six times that in regular atmospheres (Marriot, 1980). The use of a MA containing 23% O2 and 5% CO2 has been shown to extend the storage life of bananas 46 weeks (Ke and Hwang, 1987; Truter and Combrink, 1990), however, the extensive use of such MA systems has been hindered by the substantial costs incurred (Marchal, 1998). Two of the most commercially used techniques for altering storage gas composition are the silicone membrane and diffusion channel systems. The former is a

membrane, which is differentially permeable to respiration gases. It consists of a Tergal net (5254 g2 ) covered with a thin (90 m) and uniform layer of silicone rubber. At 1 Pa, its permeabilities to CO2 , O2 and C2 H4 are 80, 15 and 32 nmol1 m2 Pa1 , respectively. At a given temperature, the amount of gas diffusing through the silicone membrane will depend on the membrane area, its permeability and the gas partial pressure difference across the membrane. The area of membrane required was calculated using the equation below (Gari epy et al., 1986): A= respiration rate ms PCO2 CO2 (1)

where A is the area of silicone membrane in m2 , the respiration rate of the product stored under MA is measured in litres of CO2 consumed kg1 per day, ms is the mass of bananas stored in kg, PCO2 is the permeability of silicone membrane to CO2 in nmol1 m2 Pa1 and CO2 is the desired CO2 partial pressure difference across the membrane in Pa. The diffusion channel system is also capable of maintaining a specic gas composition in MA storage systems. Diffusion channels are open channels that provide a path for the diffusion of gases in and out of an airtight chamber (Ramachandra, 1995). Gases diffuse through these tubes at rates according to their length and cross-sectional area and gas concentration gradients between the chamber and ambient air (Ratti et al., 1997). This system is capable of maintaining different steady state compositions with O2 from 1.5% to 18% depending on the length of the diffusion channel and the mass of the stored product and is relatively insensitive to changes in barometric pressure and uctuations in the storage room temperature (Baugerod, 1980). The concentration of CO2 is almost 0% with the help of a chemical adsorbent such as hydrated lime (Ratti et al., 1997). The length of the diffusion channel is given as a function of O2 concentration, cross-sectional area, respiration rate and mass of produce in the equation below (Ratti et al., 1998): L = Ac C DO2 N2 (1 yO2 air ) ln respiration rate ms (1 yO2 chamber ) (2) where L is the length of diffusion channel in m, Ac is the cross-sectional area of the diffusion channel in m2 ,

O.J. Stewart et al. / Postharvest Biology and Technology 35 (2005) 309317

311

C is the total gas concentration in mol m3 , DO2 N2 is the diffusivity coefcient of oxygen in nitrogen in m2 s1 , yO2 is the oxygen mole fraction, the respiration rate of the product stored is measured in mg CO2 consumed kg1 h1 and ms is the mass of bananas in kg. The assumptions made for using this system are that the total pressure is constant at 101.33 kPa, all CO2 produced is scrubbed, the system is binary (only O2 and N2 are diffusing since all CO2 is scrubbed), the process is pure diffusion and the temperature is constant, the oxygen concentration in the chamber is uniform at any time, diffusion is one-dimensional and the diffusion coefcient of oxygen in nitrogen is constant. The purpose of this study was to evaluate the potential of the silicone membrane and diffusion channel systems to provide suitable conditions for optimum storage of Giant Cavendish bananas and to assess the quality of bananas stored using these systems.

held in place by threaded rods. At one end of the chamber, the acrylic lid served as a cover and was removable to facilitate opening and closing of the chambers. The cover was transparent, allowing for visual product inspection during storage. The silicone membrane or diffusion channel tube was installed onto the acrylic cover of each chamber. Neoprene gaskets were used at both ends of the chamber to ensure air tightness. Chambers were made airtight using a digital pressure indicator, DPI 601, with a resolution of 0.05% in the range 13500 kPa. The pressure inside the chamber was increased to 5 kPa of relative pressure using an air pump and this pressure maintained for 20 min. The chamber was considered air tight if no change in pressure was recorded over the 20-min period. 2.3. Measurement of respiratory gases Respiratory gases (O2 and CO2 ) within each chamber were measured during storage using a SRI 8610A gas chromatograph (SRI Instruments, California, USA) tted with a thermal conductivity detector. Oven temperature was set at 45 C and the detector temperature set at 120 C. Helium was used as the carrier gas at 10 kPa. Data were recorded and analysed using a Peak Simple Chromatography Data System, model #203 (SRI Instruments, California, USA). Two gas samples of 0.5 ml each were taken from each storage chamber and injected into the gas chromatograph. Prior to analysis, the instrument was calibrated with a standard of known concentration. 2.4. Measurement of respiration rate Respiration rate was measured at 15 C under regular atmosphere until the fruit were ripe and was used to estimate the likely respiration rate under MA. Storage chambers containing approximately 5 kg of bananas each were kept in the storage rooms. During measurement, chambers were closed for 3 h and the O2 consumed within this period measured. Measurements were made every 2 days during the storage period. The respiration rate was calculated using the equation below (Ratti et al., 1996): respiration rate = V dCO2 ms dt (3)

2. Materials and methods 2.1. Plant material and storage conditions Preclimacteric bananas (Musa acuminata, cv. Cavendish) were obtained from local importers at Courchesne Larose Ltd., (Montreal, Quebec). Fruit of commercial maturity, approximately 5 days from harvest on arrival, were allowed a pre-cooling time of 1218 h before the start of each experiment. Bananas (5 kg) were weighed and placed in each chamber, using three replicates per treatment. Fruit were stored at 15 C for 6 weeks before being induced to ripen by returning fruit to regular atmosphere and increasing the temperature to 23 C. All experiments were conducted in a walk-in cold room of dimensions 4.5 m 3.0 m and 2.1 m high. The temperature of the cold room, which was set at 15 C and maintained by a temperature control unit, uctuated between 14 C and 15.5 C. 2.2. Experimental chambers The laboratory scale storage chambers were each composed of a length of PVC pipe, 50 cm long and 25 cm internal diameter. The extremities of each unit were tted with a square sheet of acrylic 6 mm thick,

312

O.J. Stewart et al. / Postharvest Biology and Technology 35 (2005) 309317

where the respiration rate is measured in mg CO2 consumed kg1 h1 , dCO2 is the change in CO2 concentration in the chamber in mg l1 , dt is the time difference in hour, V is the free volume of the chamber in litres and ms is the mass of bananas stored in kg. 2.5. Storage of bananas using the silicone membrane and diffusion channel systems By comparison with the literature, the respiration rate under MA was estimated to be either 20% or 60% less, or 20% greater than the respiration rate at regular atmosphere. These estimations were used to calculate three areas of silicone membrane and three lengths of diffusion channels to t the three possible outcomes (Gari epy et al., 1986). The different areas of silicone membrane used were 50.3, 98.6 and 158.4 cm2 and the three different lengths of diffusion channel tubes used were 4, 7 and 10 cm. The smallest area of silicone membrane and diffusion channel corresponded to a 60% reduction in the respiration rate under MA while the largest area of silicone membrane and length of diffusion channel corresponded to 20% increase in respiration rate under MA. The medium area of silicone membrane and length of diffusion channel corresponded to a 20% reduction in respiration rate under MA. Each membrane/diffusion channel was installed onto separate laboratory scale chambers via a window. Each treatment consisted of three replicates. The gas concentration in each sealed chamber was established by ushing the containers with a mixture of 90% N2 , 5% O2 and 5% CO2 until the O2 concentrations reached 68.5% O2 . At this point the gas concentrations were expected to be stabilized and maintained by the silicone membrane and diffusion channel tubes themselves. For each experiment, three chambers held at regular atmosphere were used as controls. These chambers were not tted with silicone membrane windows or diffusion channels and were provided with perforations to ensure good air circulation throughout the experiment. 2.6. Evaluation of fruit quality After storage, unripe fruit were induced to ripen by holding in regular atmosphere at 23 C for 57 days.

Peel colour was measured using a Minolta Chroma Meter, model CR-300b (Minolta Inc., Canada) tristimulus colour analyser to measure chromaticity in L, a and b coordinates and the data converted to chroma and hue angle, respectively. Duplicate measurements were taken on the same side of each fruit and an average was taken. Pulp penetration force was determined on individual fruit using a Soiltest Penetrometer, model CL-700 (Soiltest Inc., Chicago, USA), tted with an 11 mm probe tip. Each fruit was tested on the same side twice and an average was taken. The value recorded for the pulp penetration force was the maximum force required for the fruit to yield to the tip of the probe and was recorded in Newtons (Medlicott et al., 1990). Total soluble solids and cumulative weight loss were measured as previously reported (Williams et al., 2003). Evaluation of total sugars was made using a spectrophotometric method according to Martin et al. (2002) and titratable acidity was measured as previously reported (Rahman et al., 1995). Cumulative weight loss during storage was calculated and expressed as percentage physiological weight loss on a fresh weight basis. 2.7. Sensory evaluation The taste panel consisted of 15 untrained individuals who were asked to evaluate the overall appearance, taste, texture, aroma, sweetness and pulp colour of representative samples after storage. Ratings were made on a Hedonic scale from 1 to 5 (1 = very bad, 2 = poor, 3 = fair, 4 = good, 5 = excellent). These values were converted to percentages by expressing the score obtained as a percentage of the maximum possible score (Larmond, 1977). For comparison, Cavendish bananas (of the same cultivar as those stored) were obtained from a retail outlet and included in the test as a hidden reference. Evaluation of the samples was carried out in triplicate for each storage condition, each sample served on a white plate, which reduced the likelihood of the background colour affecting the perception of colour. The order of presentation of the samples was randomised among the panellists and sessions. All evaluations were conducted in isolated sensory booths, illuminated under white light under standard conditions (Larmond, 1977).

O.J. Stewart et al. / Postharvest Biology and Technology 35 (2005) 309317

313

2.8. Statistical analysis Data collected for the assessment of quality before and after storage were subjected to variance analysis (ANOVA) using the SAS package. Treatments were compared at the P < 0.05 level to identify signicance between treatments. 3. Results and discussion Relative humidity (RH) levels at regular atmosphere were 7080%. However, within the sealed chambers, RH was close to 100% as shown by condensation. In all experiments, visual inspection showed that the incidence of bacterial and fungal rots was considerably reduced by storage under MA. This reduction was primarily due to the low O2 concentration within the storage chamber which slowed both produce and microbial respiration. The reduction of storage rots in MA has been demonstrated with a variety of produce such as cabbage and carrots (Reeleder et al., 1988). 3.1. Measurement of the respiration rate under MA The average respiration rate of green bananas at 15 C, regular atmosphere and 80% relative RH was 7.5 0.7 mg CO2 kg1 h1 . Fruit stored under these conditions (controls) required 29 days to reach a fully ripe stage without being induced to ripen. The respiration rate under MA was 2.8 0.3 mg CO2 kg1 h1 , approximately 40% of that in regular atmosphere. The decline in the respiration rate under MA may be attributed to the reduced level of O2 and increased CO2 levels within the chambers, which is a well known characteristic of storage under MA (Smock, 1979). 3.2. MA storage using silicone membranes Best storage with silicone membranes was achieved using an area of 50.29 cm2 representing a respiration rate 60% less than that in regular atmosphere at the same temperature. The changes in O2 and CO2 levels with time in these chambers was measured throughout storage (Fig. 1A). Gas levels of 3% O2 and 3.5% CO2 were achieved in 10 days and remained relatively stable throughout the storage period. On ripening, these fruit were found to be superior in quality to controls,

Fig. 1. Change in O2 ( ) and CO2 ( ) concentrations with time in chambers tted with silicone membranes ((A) area = 50.29 cm2 ; (B) area = 98.56 cm2 ; (C) area = 158.43 cm2 ).

store-bought references and some treatments with different gas compositions. Chambers tted with a membrane area of 98.56 cm2 achieved gas levels of about 2.5% CO2 and 6% O2

314

O.J. Stewart et al. / Postharvest Biology and Technology 35 (2005) 309317

(Fig. 1B) in about 12 days while chambers tted with the largest membrane area of 158.43 cm2 achieved gas levels of about 2% O2 and 710% CO2 after about 16 days (Fig. 1C). Quality evaluation showed that these fruit were inferior to controls as well as most other treatments (see Section 3.4). 3.3. MA storage using diffusion channels The shortest length of diffusion channel (4 cm) was found to be most efcient at achieving and maintaining stable gas compositions during storage, also indicating that the most accurate estimate of the respiration rate under MA was 60% less than the value measured at regular atmosphere. In these chambers, gas levels of 4% O2 and 6% CO2 were achieved in about 16 days (Fig. 2A). Chambers with 7 cm long diffusion channels had gas levels of 45% O2 and 78% CO2 (Fig. 2B) while chambers with 10 cm long diffusion channels had gas levels of 3% O2 and 89% CO2 (Fig. 2C) achieved in 15 and 17 days, respectively. While the O2 levels differed by only 12%, the CO2 levels reached 59% and were more detrimental to fruit quality than the level of O2 (24%). This is probably because in some instances, high levels of CO2 often have similar effects to those of anaerobiosis (Wills et al., 1989). Fruit in chambers with high CO2 levels (> 7%) suffered from skin discoloration and developed an ethanol atmosphere in the chambers, which indicated fermentation. These fruit were badly damaged and were not evaluated by the taste panel, although it is likely that such high levels of CO2 affected avour and texture. The inaccuracy in the estimation of the respiration rate under MA is very evident in Fig. 1B and C as well as is Fig. 2B and C. These systems had difculty achieving the pre-designed gas levels of 3% O2 and 5% CO2 and in some cases never achieved stability throughout the storage period. 3.4. Quality evaluation of MA-stored fruit As was expected, there were signicant differences between the quality attributes recorded for fruit before and after storage. There was an increase in chroma, total soluble solids, total sugars and titratable acidity while there was a decrease in hue angle and rmness (Tables 1 and 2).
Fig. 2. Change in O2 ( ) and CO2 ( ) concentrations with time in chambers tted with diffusion channels ((A) L = 4 cm; (B) L = 7 cm; (C) L = 10 cm).

O.J. Stewart et al. / Postharvest Biology and Technology 35 (2005) 309317

315

Table 1 Total soluble solids, rmness and colour parameters of bananas before and after storage using diffusion channels and silicone membrane systems Treatment Green (before storage) Ripe controls Membrane (50.29 cm2 ) Membrane (98.56 cm2 ) Membrane (158.43 cm2 ) Diffusion channel (4 cm) Diffusion channel (7 cm) Diffusion channel (10 cm) Reference (store-bought) Total soluble solid (% glucose) 2.8a 29.0b 28.1b 27.8b 29.1b 26.7c 24.1d 24.7d NM* Firmness (Ncm2 ) 14.6a 8.5b 9.0b 8.6b 8.9b 8.5b 8.6b 6.7c 8.4b Chroma 37.0a 56.8b 57.2b 50.8a 41.8a 42.0a 43.4a 45.8a 39.5a Hue angle 118.3a 88.8b 90.58b 92.14b 92.22b 91.40b 87.16b 88.72b 90.81b

NM* , not measured. All values are expressed as the mean of n 3; P < 0.05. Means with the same letters in a column are not signicantly different.

There were signicant differences in total soluble solids, rmness, chroma, total sugars and mass loss during storage between fruit stored using membranes and those stored using diffusion channels (Tables 1 and 2). All diffusion channel treatments were observed to have lower values of total soluble solids than membrane treatments and ripe controls. With respect to rmness, no signicant differences were observed with the exception of fruit stored using 10 cm long diffusion channels that showed signicantly less rmness. The inhibition of these processes may be due to the higher levels of CO2 observed in these chambers. MA storage has been reported to disrupt metabolism in fruit stored in CO2 -rich atmospheres (Brecht, 1980; Wills et al., 1989). Hue angle and titratable acidity did not show any signicant difference between the treatments using silicone membranes and diffusion channels. As expected, a signicant difference was observed between the fruit before and after storage but not as a result of the treatments applied.

For most treatments mass loss during storage under MA was signicantly lower (P < 0.05) than that in fruit stored in regular atmosphere (controls). Fruit stored in regular atmosphere suffered a mass loss of 12.5% while fruit stored using silicone membranes lost 23.5% and fruit stored using diffusion channels lost 47%. This was probably due to the air in regular atmosphere having less humidity (80%) than air in the sealed storage containers that achieved 100%. A number of studies have pointed to the fact that maintaining banana fruit in an atmosphere of low humidity results in dehydration with concomitant induction of the climacteric (Blankenship and Herdeman, 1995; Xue et al., 1996). In the case of the diffusion channels it is possible that moisture was being gradually lost from the chamber through the small open diffusion channels. This indicates the benecial effect and superiority of the membrane system for MA storage of bananas compared with diffusion channels. Peel colour, weight loss and storage life are the ripening characteristics most af-

Table 2 Mass loss during storage, total sugars and titratable acidity of bananas before and after storage using diffusion channels and silicone membrane systems Treatment Green (before storage) Ripe controls Membrane (50.29 cm2 ) Membrane (98.56 cm2 ) Membrane (158.43 cm2 ) Diffusion channel (4 cm) Diffusion channel (7 cm) Diffusion channel (10 cm) Reference (store-bought) Mass loss during storage (% Fwt+ ) NM 12.5a 2.9b,c 3.4b,c 2.3b,d 6.2c 4.4c,d 6.7c NM Total sugars (mM) 0.8a 17.3b 18.0b 15.6b,c 17.7b 14.5b,c 10.8c 11.0c NM Titratable acidity (103 mol H+ ) 15.6a 65.6b 61.1b 58.9b 55.9b 62.6b 68.4b 70.8b NM

NM, not measured. All values are expressed as the mean of n 3; P < 0.05. Means with the same letters in a column are not signicantly different.

316

O.J. Stewart et al. / Postharvest Biology and Technology 35 (2005) 309317

Table 3 Sensory rating of bananas stored using diffusion channels and silicone membrane systems Treatment Membrane (50.29 cm2 ) Membrane (98.56 cm2 ) Membrane (158.43 cm2 ) Diffusion channel (4 cm) Diffusion channel (7 cm) Diffusion channel (10 cm) Ripe controls Reference (store-bought) Aroma 77.4a 78.4a 79.8a 71.0a,b 56.1a,b NM 47.3b 77.0 Texture 73.5a 72.6a 69.3a 72.2a 47.0b NM 52.9a,b 68.6a,b Taste 74.1a 71.2a 67.4a,b 69.6a,b 50.2b NM 74.4a 54.6a,b Overall appearance 85.7a 81.1a 69.3a 84.1a 67.8a 18.70b 80.30a 86.97a Sweetness 77.1a 68.4a 66.6a 70.1a 52.0b NM 78.6a 56.2a Pulp colour 89.9a 72.1a 53.8b 82.0a 68.1a NM 71.7a 56.8b

NM, not measured. All values are expressed as the mean of n 3; P < 0.05. Means with the same letters are not signicantly different.

fected by humidity levels during storage (Blankenship and Herdeman, 1995). Weight loss during transport and storage can be of serious economic importance especially if bananas are sold by weight. Moisture loss of this form is usually from the peel and directly affects the acceptability of the fruit, because drying enhances the appearance of skin blemishes and gives the peel a smoky and streaky appearance (Wills et al., 1989). A similar trend was observed for chroma, hue angle and titratable acidity. Although lower values were observed for most of the diffusion channel treatments compared with membrane treatments and ripe controls, the effect was more pronounced in chambers tted with 7 cm and 10 cm long diffusion channels. 3.5. Sensory evaluation of MA-stored fruit Table 3 shows the scores from the sensory evaluation of fruit stored under MA using silicone membranes and diffusion channels. As with the quality attributes, the general trend was that fruit stored using silicone membranes were more acceptable than those stored using diffusion channels, with the exception of fruit stored in 4 cm long diffusion channels. Fruit stored with 10 cm long channels were not rated excepting on their overall appearance. The smallest area of silicone membrane and the shortest diffusion channel had similar ratings overall and rated well compared with the store-bought reference fruit. In spite of being stored in regular atmosphere, controls had good scores for overall appearance comparable with membrane-stored fruit and the storebought reference but had shorter shelf lives. While all MA treatments had to be induced to ripened, controls ripened naturally after 29 days at regular atmosphere.

4. Conclusions Quality evaluation of MA-stored fruit indicated that after 42 days at 15 C, best results were obtained using silicone membranes of area 50.29 cm2 and diffusion channels of length 4 cm. Bananas stored under these conditions had harvest-fresh appearance, good colour, minimum mould and excellent marketability conditions. Sensory evaluation revealed that fruit stored with the smallest area of silicone membrane and the shortest diffusion channel were of superior eating quality and suffered the least mass losses during storage compared to fruit stored using different areas of silicone membrane and lengths of diffusion channel as well as controls. MA-stored fruit had better texture, taste and sweetness than controls and store-bought reference. The silicone membrane and diffusion channel systems described in this study offer an inexpensive alternative to the complex systems that are commonly associated with MA systems. Other advantages include ease of manipulation and the possibility of marketing stored produce progressively without disturbing the atmosphere in the remaining storage containers.

Acknowledgements The authors are grateful to the Natural Sciences and Engineering Research Council (NSERC) and the University of the West Indies Research and Publication Fund for providing funding for these projects. We also thank Dr. Valerie Orsat and Mr. Tim Rennie for their assistance.

O.J. Stewart et al. / Postharvest Biology and Technology 35 (2005) 309317

317

References
Baugerod, H., 1980. Atmosphere control in controlled atmosphere storage rooms by means of controlled diffusion through air lled channels. Acta Hortic. 116, 179185. Blankenship, S.M., Herdeman, R.W., 1995. High relative humidity after ethylene gassing is important to banana fruit quality. Hort. Tech. 5, 150151. Brecht, P.E., 1980. Use of controlled atmospheres to retard deterioration of produce. Food Technol. 34, 4550. Chamara, D., Illeperuma, K., Galappatty, P.T., 2000. Effect of MA and ethylene absorbers on extension of storage life Kolikuttu banana at ambient temperature. Fruit 55, 361388. Gari epy, Y., Raghavan, G.S.V., Theriault, R., 1986. Controlled atmosphere storage of celery with the silicone membrane system. Int. J. Refrig. 9, 234240. Ke, L., Hwang, S.C., 1987. Postharvest handling of bananas in Taiwan. Taiwan Banana Research Institute, Annual report, pp. 4755. Jiang, Y., Joyce, D.C., MacNish, A.J., 1999. Extension of the shelflife of banana fruit by 1-methylcyclopropene in combination with polyethylene bags. Postharvest Biol. Technol. 16, 187193. Larmond, E., 1977. Laboratory Methods for Sensory Evaluation of Food. Food Research Institute, Research Branch, Canada Department of Agriculture, Ottawa, publication 1637. Maneerat, C., Tongta, A., Kanlayanarat, S., 1997. A transit model to predict O2 and CO2 concentrations in MA packaging of banana at various temperatures. In: Seventh International Controlled Atmosphere Research Conference, University of California, Davis, p. 109, abstracts. Mapson, L.W., Robinson, J.E., 1966. Relation between oxygen tension, biosynthesis of ethylene, respiration and ripening changes in banana fruit. J. Food Technol. 1, 215217. Marchal, J., Nolin, J., 1990. Fruit quality: pre and postharvest physiology. Fruit (Paris), 119122, Special issue. Marchal, J., 1998. An overview of post-harvest aspects of banana. Acta Hort. 490, 501509. Marriot, J., 1980. Bananasphysiology and biochemistry of storage and ripening for optimum quality. Crit. Rev. Food Sci. Nutr. 13, 4188. Martin, A.B., Cuadrado, Y., Guerra, H., Gallego, P., Hita, O., Martin, L., Dorado, A., Villalobos, N., 2002. Differences in the contents of total sugars, reducing sugars, starch and sucrose in embryogenic and non-embryogenic calli from Medicago arborea L. Plant Sci. 154, 143151. Medlicott, A.P., Semple, A.J., Thompson, A.J., Blackbourne, H.R., Thompson, A.R., 1990. Measurement of colour changes in ripen-

ing bananas and mangoes by instrumental, chemical and visual assessments. Trop. Agric. 62, 161166. Rahman, N., Yuquing, W., Thompson, K., February 2024, 1995. Temperature and MA packaging effects on the ripening of bananas. In: Proceedings of the International Conference on Postharvest and Postharvest Technologies for the Fresh Fruit and Vegetables, pp. 313321. Ramachandra, M., 1995. diffusion channels for broccoli storage (MSc Thesis). Department of Agricultural and Biosystems Engineering, McGill University, Macdonald Campus, Montreal, Canada, pp 3435. Ratti, C., Rabie, H.R., Raghavan, G.S.V., 1997. Diffusion channel design for MA storage of vegetables. NABEC-94, Ontario, paper 94340. Ratti, C., Rabie, H.R., Raghavan, G.S.V., 1998. Modelling MA storage of fresh cauliower using diffusion channels. J. Agric. Eng. Res. 69, 343350. Ratti, C., Raghavan, G.S.V., Gariepy, Y., 1996. Respiration rate model and MA packaging of fresh cauliowers. J. Food. Eng. 28, 297306. Reeleder, R.D., Raghavan, G.S.V., Monette, S., Gariepy, Y., 1988. Use of controlled atmospheres to control storage rot of carrots caused by Sclerotinia sclerotiorum. J. Int. Refrig. 12, 159 164. Scott, K.J., Roberts, E.A., 1966. Polyethylene bags to delay ripening of bananas during transport and storage. Aust. J. Exp. Agric. Anim. Husb. 6, 197199. Stiles, M.E., 1991. Scientic principles of controlled/MA packaging. In: Ooraikul, B., Stiles, M.E. (Eds.), MA Packaging of Food. Ellis Harwood Ltd., England. Truter, A.B., Combrink, J.C., 1990. Controlled and MA storage of bananas. Acta Hort. 275, 631637. Seymour, G., Taylor, J., Tucker, G., 1993. In: Seymour, G.B. (Ed.), Biochemistry of Fruit Ripening. Chapman and Hall, London, pp. 1834. Smock, R.M., 1979. Controlled atmosphere storage of fruit. Hortic. Rev. 1, 301336. Williams, O.J., Raghavan, G.S.V., Golden, K.D., Gariepy, Y., 2003. Postharvest storage of Giant Cavendish bananas using ethylene oxide and sulphur dioxide. J. Sci. Food. Agric. 83, 180186. Wills, R.B.H., McGlasson, W.B., Graham, D., Lee, T.H., Hall, E.G., 1989. Postharvest: an Introduction to the Physiology and Handling of Fruit and Vegetables. SP Professional Books, Hong Kong, pp. 6675. Xue, Y., Kubo, Y., Nakamura, R., 1996. Effects of humidity on the ripening characteristics of banana fruit. Nippon Shokuhin Kagaku kaishi 43, 541545.