27th IAPRI Symposium on Packaging 2015
Optimisation of Packaging for Carrot Roots
L.) Stored at Different
Hanne Larsen 1* and Anne-Berit Wold 2
1 Nofima AS - Norwegian Institute of Food, Fisheries and Aquaculture Research,
Osloveien 1, NO-1430 Ås, Norway
2 Norwegian University of Life Sciences, Dept. of Plant Sciences, PB 5003, N-1432, Norway
*Corresponding author name. Email: email@example.com
Keywords: Optimal gas atmosphere, quality and shelf life, modelling, respiration rate, gas
permeability, number of perforations.
L.) are an important vegetable crop worldwide, and the most important field
vegetable produced and consumed in Norway. Carrots are vulnerable to water loss, and proper
packaging will prevent desiccation and hence prolong the shelf life. Today carrots are packaged in
laser perforated packages. After packaging, the carrots are stored for several days in chill cabinets or
at room temperature in the grocery stores until purchase. The consumer occasionally report about
paraffin odour and flavour of the carrots, which could be due to high ethanol content in the carrots as
reported by Seljåsen et al. . Short shelf life of carrots due to rotting and mould growth are an
increasing problem throughout the storage season. CO2 concentrations above 5% have been shown to
increase spoilage, and O2 concentrations below 3% are not well tolerated and generally results in
increased bacterial rot according to Suslow et al. . There is a need for determination of the optimal
Abstract: Carrot roots are vulnerable to water loss, and proper packaging prevents desiccation and
prolongs the shelf life from time of packaging until consumption. Today, Norwegian carrots are
packaged in laser perforated biaxially oriented polypropylene film (BOPP). The packages are displayed
at chill conditions in some grocery stores, and at room temperature in others. Previous packaging
experiments (unpublished results) have shown that the CO2 level in the carrot packages can be very
high (up to 40 %) after only 3 days of storage at room temperature. A range of fruit and vegetables
are injured at high CO2 concentrations.
Aim 1: Determination of optimal gas atmosphere in carrot packages. In an experiment at Nofima,
carrot roots were packaged in films with different perforations and stored at two storage conditions
for 15 days. Effect of packaging and storage conditions was analysed at the end of the storage period
using chemical analysis, descriptive sensory analysis and microbial registration. Carrots in packages
with high CO2 levels developed ethanol odour and taste and were more prone to microbial spoilage
during the storage period. Hence, it is important to have product adapted permeability of the
packages with sufficient number of perforations in order to avoid adverse impact on product quality.
Aim 2: Demonstrate the use of a simple model previously developed at Nofima modelling the gas
concentrations in packages for fruits and vegetables. The most important parameters are weight and
respiration rate of the produce, permeability of the film and perforations and the volume of the
package. The model was adapted to respiration rate data for different carrot varieties and data for
packages with different types of perforations in order to optimise the gas atmosphere in packages for
carrot roots stored at two storage conditions.
gas atmosphere in carrot packages at realistic storage conditions in order to minimise spoilage and
Knowledge of produce respiration rates and package transmission rate is two key factors in the choice
of appropriate packaging materials for different fruit and vegetables. According to Beaudry  and
Watkins  is the choice of product-optimised film crucial to avoid detrimental low levels of O2 and/or
high levels of CO2 that could induce anaerobic metabolism with possible off-flavour generation and
risk of anaerobic microorganism proliferation. In order to computerize the selection of the appropriate
packaging material for different fruit and vegetables without performing huge packaging experiments,
much effort is put into modelling of the gas exchange processes continuing inside the package during
storage. Gaining the data needed for input in the models can be cumbersome, and data found in
literature are often given in different units and are not stated at the desired storage conditions, e.g.
gas transmission rates are usually measured at 23 °C. Respiration data might also show great
variation due to cultivar, quality and season.
Larsen  describes a low cost methodology for package optimising for fruit and vegetables. The
methodology uses 1) a low cost gas analyser and commercial packages as “respiration chambers” in
order to measure the respiration rates for fruit and vegetables, 2) the same low cost gas analyser to
measure the gas transmission rates (gas TR) and CO2TR/OTR-ratio (permselectivity, commonly
denoted β) at different temperatures for whole packages with and without perforations and for the
single perforations and finally 3) integrates respiration rates and gas transmission rate data for the
whole package into a simplified predictive model using Microsoft Excel. The described procedure using
low cost equipment and commercial packages is an alternative method for laboratories, packaging
material producers, farmers and packaging houses to optimize their packages based on own
measurements under realistic storage temperatures.
Task 1: Determination of the optimal gas atmosphere in carrot packages at realistic storage conditions
in order to minimise spoilage and quality deterioration. The carrot quality was evaluated using
chemical analyses, descriptive sensory analyses and microbial registration at the end of storage.
Task 2: Demonstrate the use of a simplified prediction model modelling the gas concentrations in
packages for different carrot cultivars and packages with different types of perforations in order to
optimise the gas atmosphere in packages for carrot roots stored at two storage conditions.
2.1 Task 1: Determination of optimal gas atmosphere in carrot packages
2.1.1 Products and packaging
L. cv ‘Romance’) from the same batch and producer were washed, sorted and
packaged at a commercial packing house in pouches containing one kg of carrots. The packaging
materials were 1) 40 µm biaxially oriented polypropylene (BOPP) (ScanStore Packaging AS, Middelfart,
Denmark) with 13 laser perforations (approximately 100 µm in diameter; denoted “LP”), 2) 40 µm
BOPP (NNZ Scandinavia ApS, Odense, Denmark) with 11 warm needle perforations (approximately
800 µm in diameter; denoted “NP”), and 3) 40 µm BOPP (Tommen Gram, Levanger, Norway) with
approximately 560 warm needle perforations per 10 x 10 cm (“bread pouch”, denoted “BP”). The
packages were stored at chill storage conditions of 4 °C for 6 days followed by 6 °C for 9 days, or
retail storage conditions of 4 °C for 3 days, 20 °C for 3 days and 6 °C for 9 days.
2.1.2 Measurements of O2 and CO2 in the headspace and ethanol in the carrots
O2- and CO2 composition in the headspace of the carrot packages were measured at day 1, 3, 4, 5, 6,
7, 9, 12 and 14 using a CheckMate3 O2/CO2 analyser (PBI-Dansensor, Ringsted, Denmark). Gas
atmosphere samples were collected through a self-adhesive septum placed on the pouches.
Ethanol were analysed after 1, 7 and 15 days of storage. 100 g carrot were wrapped in commercial
housekeeping aluminium foil, and subsequently packaged in sealed vacuum pouches before freezing
at - 40 °C. The frozen carrot samples were thawed and homogenised prior to water extraction. The
extract were filtrated prior to injection into a gas chromatograph with a flame ionization detector
(FID). Five replicates were run for the LP-retail samples, and three replicates were run for all the
2.1.3 Sensory analysis and evaluation of diseases on carrots
Qualitative Descriptive Sensory Analysis was performed according to ISO 13299:2003(E) using a
trained sensory panel consisting of 10 assessors at Nofima (Ås, Norway). Prior to analysis, the
assessors agreed on 23 sensory attributes characterizing carrot. The assessors recorded their results
at individual speed on a 15 cm non-structured continuous scale. The data registration system
(EyeQuestion Software, Logic8 BV, Nederland) transformed the responses into numbers between 1
(low intensity) and 9 (high intensity).
After 15 days of storage at chill or retail temperatures, the number of roots showing visible symptoms
of different diseases was recorded and based on macroscopically physically visible symptoms they
were sorted into different categories. The results are presented in % diseased carrots after 15 days of
2.1.4 Statistical analysis
Analysis of variance (ANOVA) was performed for ethanol and % diseased carrots (significance level p
< 0.05) using general linear model in Minitab 17 Statistical Software (Minitab Inc., State College, PA,
USA) and Tukey’s multiple comparison test. Sensory data were analysed using general linear models
(Proc GLM) in SAS 9.4 (SAS Institute, Inc, Cary, NC, USA).
2.2 Task 2: Measurement of input data and modelling
2.2.1 Measuring respiration rate
The closed system methodology was used for measurement of the O2 consumption rates (O2CR) and
CO2 production rates (CO2PR) as described by Larsen  and Larsen et al.  using 500g carrot
(three replicates) of each cultivar. Respiration rate was measured at 4 °C and 22 °C for 9 cultivars
from north (Trøndelag) and south (Vestfold) in July, September, January and May. Respiration rate
was not measured for all combinations of cultivar, place of growth and sampling time (not all cultivars
were available for analysis at each time of sampling), and the analysis was performed for a random
batch of carrots at each time of sampling.
2.2.2 Measuring gas transmission rate
Oxygen transmission rate (OTR) and carbon dioxide transmission rate (CO2TR) in the packages was
measured using a modification of the Ambient Oxygen Ingress Rate (AOIR) method described by
Larsen et al. . The AOIR method measures the OTR of whole packages using a low cost gas
analyser. A later work by Larsen and Liland  demonstrates the determination of both O2 and CO2-
transmission rates and permselectivity at different temperatures for whole perforated and non-
perforated packages and the single perforations. In the present work we transferred the methodology
from Larsen and Liland  to a specially built gas transmission cell (276 ml volume), where a piece of
film was mounted on the top of the cell before flushing of the cell, gas sampling three times and
finally calculations of the O2 and CO2 transmission rates for films and perforations. The gas
transmission rates were measured at 4 °C and 22 °C.
2.2.3 Predictive modelling
Data from the respiration and transmission rate analyses were integrated into a simple predictive
model using Microsoft Excel, originally developed by Schlemmer and Allermann  and modified and
verified as described by Larsen . The input data was O2 consumption rates, CO2 production rates,
volume of the package, product weight, O2 and CO2 transmission rates for films and perforations and
number of perforations. Model data was compared to measured data for carrots (cv. ‘Romance’)
packaged in BOPP-pouches and stored at 4 °C for 14 days.
3 Results and discussion
3.1 Task 1: Determination of optimal gas atmosphere in carrot packages
The changes in O2- and CO2 composition in the headspace of the carrot packages was different at chill
and retail storage condition (Figure 1). O2- and CO2- concentrations were close to air atmosphere in
the needle and “bread” perforated packages stored at chill storage conditions for 14 days. In the laser
perforated packages the O2-concentration stabilised at 12 to 13 % and the CO2-concentration at 9 to
10 %. At retail storage conditions the “bread” perforated packages had O2- and CO2-concentrations at
air atmosphere levels, while the needle perforated packages showed a small decrease in
Figure 1: Changes in O2- and CO2 concentrations in carrot packages during
14 days chill and retail storage. Chill storage is 6 days at 4 °C followed by 9 days
at 6 °C, and retail storage is 3 days at 4 °C, 3 days at 20 °C and 9 days at 6 °C.
LP=laser perforated and NP=needle perforated. The O2- and CO2 concentrations in
BP=”bread” perforated packages were air atmosphere (not shown).
O2-concentration to approximately 16 % and an increase in CO2-concentration to approximately 5 %
during the three days storage at 20 °C. In the laser perforated packages the O2-concentration
decreased to 4 % at the last day of storage at 20 °C (day 6) and the CO2-concentration was
approximately 32 % at day 6. According to Beaudry  will the CO2 concentration be below 21 % in
packages relying on gas transmission through perforations as long as the respiration is aerobic and
the respiratory quotient (RQ) is one. Hence, a CO2 concentration far above 20 % in the laser
perforated packages at retail condition indicated that the metabolism in the carrots had changed from
aerobic to anaerobic mode, also verified by the ethanol detected in these packages (Table 1).
Sensory assessment revealed that carrots in laser perforated packages stored at retail conditions had
significantly higher scores for ethanol odour and flavour (Table 1) and a significantly higher ethanol
Table 1: Effect of different perforated films and storage temperatures on
sensory attributes, ethanol content and % diseased carrots for carrots stored
for 15 days. The Table only shows sensory attributes with p-values < 0.02.
(mg kg-1) **
382 ± 125b
111 ± 13b
130 ± 27b
1878 ± 481a
411 ± 152b
241 ± 150b
54 ± 18bc
20 ± 8a
29 ± 8a
73 ± 22c
40 ± 19ab
66 ± 18c
* Values followed with similar letters are not significantly different by Tukey’s test (p ≥ 0.05)
** Ethanol reference value (after 1 day of storage) was 257 mg kg-1.
This is in accordance with Seljåsen et al. , which found that carrots stored at 10 and 20 °C with
increased levels of CO2 and decreased levels of O2 had higher scores for ethanol flavour and odour
compared to carrots stored at 2 °C. Seljåsen et al.  have previously shown that sickeningly sweet
taste is correlated with ethanol content.
After 15 days of storage, the percentage of diseased carrots varied from 20 to 54 and from 40 to 73
for the carrots stored at chill and retail conditions, respectively (Table 1). Soft decay was the main
disease of the carrots, especially for the carrots in the laser perforated packages (> 90 %).
Percentage diseased carrots were highest in the laser perforated packages stored at retail conditions.
However, the percentage diseased carrots were not significantly higher in these packages compared
to the carrots stored in needle and “bread” perforated packages stored at retail conditions. During
packaging the laser perforated packages was part of a commercial production line undergoing strict
quality control discarding packages with minor spoilage, whereas the needle perforated and “bread”
perforated packages were a special delivery for this experiment and did not undergo the same strict
quality control. This could explain the none-significant differences between the different degrees of
perforation especially at retail conditions, where the high temperatures give close to optimal
conditions for decay.
Our conclusion from the packaging and storage experiment was that the overall carrot quality was
best maintained in needle perforated packages with atmosphere close to air, with no major weight
loss, no ethanol formation and the lowest incidences of storage diseases for both chill and retail
conditions. Our experiment verified findings in previous studies stating that a CO2 concentration above
5 % in carrot packages should be avoided. A level of 5 % CO2 was set as the maximum limit in task 2
in the calculation of optimal number of perforations for carrot packages.
3.2 Task 2: Modelling – measured input data and model results
3.2.1 Respiration rate of carrots
O2 consumption rates (O2CR) and CO2 production rates (CO2PR) were measured for nine carrot
cultivars, two places of growth, different times of year and at 4 °C and 22 °C (Figure 2).
Figure 2: Carrots with lowest and highest respiration rates measured at 4 °C and 22 °C.
Figure 2 shows that the early cultivar ‘Nelson’ had twice the respiration rates as the winter storage
cultivar ‘Triton’. The rest of the cultivars was at a similar level as ‘Triton’, with slightly higher
respiration rates in the autumn just after harvesting. The mean respiratory quotient (RQ =CO2PR/
O2CR) was 0.7 at all sampling times at 4 °C, but at 22 °C the RQ seemed to change slightly during
storage from a mean value of 0.9 just after harvesting increasing to 1.3 in January.
3.2.2 Gas transmission rate for packages and single perforations
Oxygen and CO2 transmission rates were measured for continuous films and for single perforations
Figure 3: Oxygen and CO2 transmission rates for continuous film (3a) and
film with laser and needle perforations (3b) measured at 4 °C and 22 °C.
Figure 3 demonstrates the differences in permselectivity (CO2TR/OTR = β) and temperature influence
between perforated and non-perforated materials. The permselectivity for the continuous film was 3,
whereas the permselectivity for the perforated materials was 0.9 for the single perforations. These
results are in accordance with the findings of other authors such as Larsen and Liland , Fonseca et
al.  and Gonzalez et al. . By an increase in temperature from 4 °C to 22 °C, the OTR and
CO2TR for the continuous film increased by a factor of 1.7 and 1.9, respectively. The similar factor for
the laser perforated film (Figure 3b) was 1.1 for OTR and CO2TR. Different behaviour for continuous
and perforated materials in response to temperature changes is also stated by Larsen and Liland 
and Beaudry . Hence, at similar O2 permeability performance, perforated packages are more
prone to high CO2 concentrations compared to non-perforated packages.
3.2.3 Verification of model
Measured respiration rate and gas transmission rate data were loaded into the predictive model as
described by Larsen . Model data was validated by comparing the predicted values to measured
data for carrots (cv. ‘Romance’) packaged in BOPP-pouches and stored at 4 °C for 14 days (Figure 4).
Figure 4: Oxygen and CO2 levels measured in carrot packages stored
at 4 °C (4a) and model data (4b).
Figure 4 shows that the predicted O2 and CO2 curves (Figure 4b) are relatively close to the exact O2
and CO2 concentrations measured in the carrot packages (Figure 4a). Larsen  also demonstrated a
acceptable correlation between predicted and measured data for broccoli florets and plum fruit.
However, one should be aware the decline in respiration rate as the O2 concentration approaches
approximately 4-5 %. Using a constant O2 consumption and CO2 production in this situation may
result in less accurate model data Larsen .
3.2.4 Practical use of predictive model
By using the developed model, the optimal number of perforations in packages for carrots with the
lowest and highest respiration rates at chill and retail storage conditions was calculated (Table 2).
Table 2: Number of laser and needle perforation needed in order to avoid CO2 levels
above 5 % in packages with 1 kg carrot with the lowest and highest respiration rate
stored at chill and retail storage conditions
Low respiration rate
High respiration rate
A high number of laser perforations is necessary for the early grown carrots with high respiration rate,
especially at retail storage conditions with a calculated number of 250 perforations. Laser perforations
are commonly placed along one single row at the package, and such a high number of laser
perforations is difficult to obtain in one single row. Twenty-two larger needle perforations was
calculated to be sufficient for the high respiration carrot at retail storage conditions, which is feasible
to have in one row. Hence, a package with 22 needle perforations will cover and be suitable for all the
carrot cultivars harvested both early and late in the season, and there will be no need for using a
“summer” and “winter” film which is commonly used today.
The overall carrot quality was best maintained in needle perforated packages with atmosphere close
to air, with no major weight loss, no ethanol formation and the lowest incidences of storage diseases
for both chill and retail conditions. Our experiment verified findings in previous studies stating that a
CO2 concentration above 5 % in carrot packages should be avoided. A CO2 concentration of 5 % was
hence defined as the upper limit in the calculation of optimal number or perforations for carrot
The input data in the prediction model was the lowest and highest respiration rates measured for nine
carrot cultivars, two places of growth, different times of year and at 4 °C and 22 °C. Gas transmission
rate was measured for the film and single perforations at 4 °C and 22 °C. The modelled data was
compared to exact data measured for packaged carrot, and the model showed acceptable fit.
By using the developed model, the optimal number of perforations in packages for carrots with the
lowest and highest respiration rates at chill and retail storage conditions was calculated. A high
number of laser perforations is necessary in carrot packages, especially for the early grown cultivar
with high respiration rate stored at retail conditions (a calculated number of 250 perforations per
package). By using larger needle perforations, the number of perforations could be reduced to 22.
We wish to thank Lågen Gulrot for support and packaging of the carrots, Trøndergrønt for delivering
carrot for respiration rate analyses, Liv Berge for assistance in microbial registration and sample
preparation for chemical analyses, Aud Espedal and Karin Solgaard for handling of samples for
analyses, Kristine S. Myhrer for running the sensory analysis and Kristian Hovde Liland for performing
statistical analysis. Agricultural Food Research Foundation (Oslo, Norway) is greatly appreciated
funding this research. Some of the work presented is performed as a part of the Norwegian project
“Grøntpakk”. We also want to thank the collaboration partners in this project and the Oslofjord
Foundation for funding the “Grøntpakk” project.
1. R. Seljåsen, H.L. Kristensen, C. Lauridsen, G.S. Wyss, U. Kretzschmar, I. Birlouez-Aragone and J.
Kahl, 2013, “Quality of carrots as affected by pre- and postharvest factors and processing”,
Journal of the Science of Food and Agriculture
; vol. 93, no. 11, pp. 2611-2626.
2. T. Suslow, J. Mitchell and M. Cantwell, 2002, “Carrot: Recommendations for Maintaining
Postharvest Quality”, http://postharvest.ucdavis.edu/producefacts/, access date: 2015.05.12.
3. R.M. Beaudry, 2000, “Aroma generation by horticultural products: What can we control?
Introduction to the workshop”,
, vol. 35, no. 6, pp. 1001-1002.
4. C.B. Watkins, 2000, “Responses of horticultural commodities to high carbon dioxide as related to
modified atmosphere packaging”,
, vol. 10, no. 3, pp. 501-506.
5. H. Larsen, 2015, “Low Cost Methodology for Package Optimising for Fruit and Vegetables”, XIth
Int. Controlled and Modified Atmosphere Research Conf.,
vol. 1071, pp. 327-334.
6. H. Larsen, A. Leufvén and M. Høy, 2011, “Respiration rate of cubed carrots (
relation to gas transmission through the packaging material”,
25th IAPRI Symposium on
, Berlin, Germany, 16-18 May, ISBN: 978-3-940283-31-3.
7. H. Larsen, A. Kohler and E.M. Magnus, 2000, “Ambient Oxygen Ingress Rate method - an
alternative method to Ox-Tran for measuring oxygen transmission rate of whole packages”,
Packag. Technol. Sci.,
vol. 13, no. 6, pp. 233-241.
8. H. Larsen and K.H. Liland, 2013, “Determination of O2 and CO2 transmission rate of whole
packages and single perforations in micro-perforated packages for fruit and vegetables”,
of Food Engineering
, vol. 119, pp. 271-276.
9. P. Schlemmer and H. Allermann, 2008, “Perforation of food packaging”
16th IAPRI World
Conference on Packaging
, Bankok, Thailand, 8-12 June.
10. R. Beaudry, 1999, “Effect of O2 and CO2 partial pressure on selected phenomena affecting fruit
and vegetable quality”,
Postharvest Biol and Technol.
, vol. 15, pp. 293-303.
11. R. Seljåsen, H. Hoftun, J. Selliseth and G.B. Bengtsson, 2004, “Effects of washing and packing on
sensory and chemical parameters in carrots (
J Sci Food Agric.,
vol. 84, pp.
12. R. Seljåsen, G.B. Bengtsson, H. Hoftun and G. Vogt, 2001, “Sensory and chemical changes in five
varieties of carrot (
L) in response to mechanical stress at harvest and post-
Journal of the Science of Food and Agriculture
, vol. 81, pp. 436-447.
13. S.C. Fonseca, F.A.R. Oliveira, I.B.M. Lino, J.K. Brecht and K.V. Chau, 2000, “Modelling O2 and
CO2 exchange for development of perforation-mediated modifed atmosphere packaging”,
of Food Engineering
, vol. 43, pp. 9-15.
14. J. Gonzalez, A. Ferrer, R. Oria and M.L. Salvador, 2008, “Determination of O2 and CO2
transmission rates through microperforated films for modified atmosphere packaging of fresh
fruit and vegetables”,
Journal of Food Engineering
, vol. 86, pp. 194-201.
15. R. Beaudry, 2008, “MAP as a basis for active packaging”, In: C.L. Wilson (ed),
Active Packaging for Fruits and Vegetable
, CRC Press, Taylor & Francis Group. Boca Raton,
Florida. pp. 31-56.