International Journal of Food Properties

ISSN: 1094-2912 (Print) 1532-2386 (Online) Journal homepage: https://www.tandfonline.com/loi/ljfp20

Prediction of the Mould-Free Shelf Life of Muffins

D. Doulia , G. Katsinis & F. Rigas

To cite this article: D. Doulia , G. Katsinis & F. Rigas (2006) Prediction of the Mould-
Free Shelf Life of Muffins, International Journal of Food Properties, 9:4, 637-650, DOI:
10.1080/10942910600853824

To link to this article: https://doi.org/10.1080/10942910600853824

Published online: 18 Apr 2007.

Submit your article to this journal

Article views: 5582

View related articles

Citing articles: 1 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ljfp20
International Journal of Food Properties, 9: 637–650, 2006
Copyright © Taylor & Francis Group, LLC
ISSN: 1094-2912 print / 1532-2386 online
DOI: 10.1080/10942910600853824

PREDICTION OF THE MOULD-FREE SHELF LIFE

OF MUFFINS

D. Doulia, G. Katsinis and F. Rigas

Laboratory of Organic Chemical Technology, NTUA, Polytechnioupolis, Zografou
Campus, Athens, Greece

In this article, the effect of commonly used ingredients in muffins (sugar, glycerol, potas-
sium sorbate, acidic agent, salt), with and without ethanol surface spraying and pasteuriza-
tion on the shelf life, was examined. Citric acid, sodium diacetate, and tartaric acid were
used as acidic agents. The investigation was based on a statistical design of experiments.
Using simple first order factorial designs, reliable models were constructed revealing the
effects of the selected ingredients, and their interactions on the shelf life. The effectiveness
of preservation ranked high to low as: (combination of citric acid, potassium sorbate, etha-
nol, and pasteurization), (combination of citric acid, potassium sorbate, and ethanol),
(combination of citric acid and potassium sorbate), (combination of sodium diacetate and
potassium sorbate), (combination of tartaric acid and potassium sorbate).

Keywords: Acidic agents, Ethanol spraying, Mould, Muffin, Pasteurization, Shelf life.

INTRODUCTION
Baked goods, containing principally flour are greatly affected by ingredients and
additives included in a recipe.[1] Cake containing oils is, in the initial batter stage, a very
complex emulsion of oil in an aqueous phase containing flour, starch, sugar, eggs, milk,
and minor ingredients (baking powder, emulsifier, salt, etc). High ratio cakes made with
liquid vegetable oil are aerated in the aqueous phase, and the form of stability is provided
by egg yolk components and added surfactants.[2] Each basic ingredient in cakes performs
a particular function.
Flour and eggs are fundamental structure builders providing strength and structure
to the cake. They also function as water binders because their principal components (such
as starch, protein, sugars, fat) have pronounced moisture-absorbing properties Starch,
when added to the batter in amounts of 1 to 5 % increases batter viscosity and improves
the volume symmetry, grain and texture of the resultant cake.[3] Sugar is the major flavor-
ing, which further acts as a water binder (very effective at binding moisture) and tender-
izer by diluting the flour proteins. Glycerol, a humectant, has the ability to absorb a great
quantity of moisture and confers an increased soft-eating character to the crumb of cake.

Received 22 August 2005; accepted 9 May 2006.

Address correspondence to D. Doulia, 1 Laboratory of Organic Chemical Technology, NTUA, Polytech-
nioupolis, Zografou Campus, Athens GR-157 80, Greece. E-mail: [email protected]

637
638 DOULIA, KATSINIS, AND RIGAS

Vegetable oil is the primary tenderizer, flavoring and moistener of cake baking. Milk and
water make the crumb texture finer and reduce lightness.
In addition, milk solids enhance flavor and function as water binder.[4] The func-
tions of monoglycerides can be attributed to the reduction of the surface tension of oil pro-
viding a better dispersion of the oil phase and resulting in improved cake texture, crumb
softness, and volume. They also decrease starch gelatinization in cakes leading to a better
cake structure with improved tenderness.[5] Baking powder is a leavening agent produced
by the mixing of an acid-reacting material (such as acid calcium phosphate or sodium acid
pyrophosphate) and sodium bicarbonate. It is used to liberate carbon dioxide gas causing
the necessary aeration and opening cake structure.[6]
Acidic agents are also frequently used in cakes. Citric acid CH2(COOH)-
COH(COOH)-CH2(COOH) has not been used as a basic antimicrobial agent, however, it
has activity against some moulds and bacteria. Citric acid and tartaric acid
(CHOHCOOH)2 can be used for adjusting the pH. Sodium diacetate has been recom-
mended for use in baked goods, bread, candy, cheese spreads, meats, sauces and more,
due to its microbial inhibitory action. It reduces the incidence of rope and growth of mould
by decreasing the dough pH. It is said to have a preservative effect disproportionately
large for its content of acetate.[7,8] Salt could affect significantly the Mould Free Shelf Life
(MFSL), due to its powerful water-binding properties. However, there is a limit to the salt
quantity because of its strong effect in flavor.[9] Potassium sorbate is a principal microbial
and mould inhibitor extensively used in bakery products. It is effective at pH level up to
about 6. At higher pH levels its effectiveness decreases significantly.[7,10,11,12] Ethanol, a
strong bactericide has recently been used for its effective preservative action in bread. The
addition of ethanol at levels 0.5 to 3.5 (g ethanol/100 g of cake) leads to a substantial
extension of the shelf life of baked products. For a cake recipe to be correctly constructed,
a good balance between the ingredients is needed.[12–14]
Spoilage from mould growth is much more likely to occur in cakes. The spoilage
organoleptically or by pathogenic microbial growth is relatively small in cakes because of
the nature of the ingredients used in cake making.[3,4] Although mould and mould spores
are destroyed by the heat of the baking process, due to thermal inactivation, post-baking
contamination occurs from the mould spores present in the atmosphere, during subsequent
handling operations such as cooling, finishing, and wrapping.[10,15] It is evident that shelf
life prolongation of cake is of great importance for the productivity and profitability of a
company. Similar preservation methods to those used for bakery products can be used for
muffins (type of cake). Mould inhibitors (such as propionates, sorbates, and ethanol),
modified atmosphere packaging, pasteurization, freezing, etc. are mentioned in the rele-
vant literature.[3,16,17,18]
Shelf life extension of cake and other baked products may be achieved using single
ingredient or process change, or a combination of many alternative changes according to
food legislation, ingredient availability and cost, consumer acceptance, and social trends.
Manipulation of ingredients, which slow the rate of moisture loss during storage, such as
sugar, glycerol, salt and preservatives, is a powerful tool, which could lead to a significant
inhibition of cake microbial growth and therefore, sensible extension of mould-free shelf
life could be achieved.[3,4,9,19]
This article aims to prioritize strategies that may be used in cake shelf life prolonga-
tion. Since muffin is a multicomponent mixture made by batter and additional optional
ingredients (such as emulsifiers, fat, glycerol, milk, preservatives, ethanol, etc.), it is very
difficult to investigate the influence of a single component on the shelf life in combination
 PREDICTION OF THE MICROBIAL SHELF LIFE OF MUFFINS 639

with the behavior of other components of the mixture. For that reason, five factorial exper-
iments were designed.[20] Particularly, the effect of some selected common ingredients of
cake and their interactions with other factors involved, on the mould-free shelf life was
investigated in three experimental designs. The selected ingredients were sugar, glycerol,
potassium sorbate, and an acidic agent (citric acid, sodium diacetate, and tartaric acid).
The approved best recipe composition was investigated further (in higher degree experi-
mental designs, by spraying the muffins with ethanol (0.5–1.5 g/100 g of muffin) and by
applying spraying with ethanol and pasteurization simultaneously. This method facilitates
the investigation by reducing significantly the number of required experiments, thus sav-
ing time and money. In addition, the models obtained may be used for predictions on the
probable mould-free shelf life of muffins, within the variation intervals of factors (optimi-
zation parameter). In addition, complementary measurements of pH (a second optimiza-
tion parameter), total microbial count and moisture were performed.

MATERIALS
Food grade basic (wheat flour, starch, sunflower oil, sugar, salt) and optional ingre-
dients were used. The preservative used was potassium sorbate (Cheminova) and the
acidic agents were citric acid (ADM), tartaric acid (ADM), and sodium diacetate (Boërni-
ger). Ethanol was 96° food grade (Merck). Monoglycerides were used as emulsifier (Con-
dea). Glycerol was food grade of 99.5% purity (Elton, Chemicals). The packaging material
was a laminated film of 30 μm oriented polypropylene (OPP) and 21 μm polyvinyl
dichloride PVdC (Mobil).

METHODS
Muffin Making Procedure
Muffin samples were prepared by mixing the chosen required and optional ingredi-
ents at proportions included in the formulae given in Table 1. The values of variables
(design factors) are given in Table 2. At first the liquid ingredients were mixed for 15 min-
utes at the first speed of the mixer. Then, the solid ingredients were added and the batter is
mixed for 2 minutes at the same speed. Accurately predetermined quantity of the mix (70 g)
was deposited in paper-lined tins for muffins and placed in pans. The muffins were baked
in an air oven at 180°C for 20 minutes. Then, they were cooled for 60 minutes at ambient
temperature, removed from the baking pan, wrapped in plastic films, thermo sealed, and
stored at room temperature. Five repetitions were performed for each experiment. The con-
trol was muffin prepared simultaneously (five repetitions) in the same oven under identical
conditions with those of experimental design and its composition is given in Table 1. In the
case of ethanol addition in muffin, the appropriate quantity of ethanol was sprayed on the
whole surface of the muffin samples before wrapping. Then, the pouches were thermo
sealed. When pasteurization was applied, the wrapped muffin samples were heated in an
oven at 140°C for 45 minutes giving a core temperature of about 65–70°C.

Examination of Muffin
Testing of the relative microbial free shelf life. The muffin samples were
examined for visible signs of microbial growth on the crust every day. The microbial
640 DOULIA, KATSINIS, AND RIGAS

Table 1 Composition a of cake samples in experimental design.

Ingredient 1st 2nd 3rd 4th 5thc Control

Flour 80 80 80 80 80 80
Starch 20 20 20 20 20 20
Water 75 75 75 75 75 75
Baking powder 3 3 3 3 3 3
Sunflower oil 33 33 33 33 33 33
Emulsifier 1 1 1 1 1 1
Eggs 30 30 30 30 30 30
Milk 12 12 12 12 12 12
Salt 1.5 1.5 1.5 variable variable 1.5
Sugar variable variable variable variable variable 75
Glycerol variable variable variable variable variable
Potassium Sorbate variable variable variable variable variable
Citric acid variable 0 0 variable variable
Sodium diacetate 0 variable 0 0 0
Tartaric acid 0 variable 0 0
Ethanol b 0 0 0 variable variable
a b c
g/100 g of (flour + starch); g/100 g of muffin; muffin samples were pasteurized.

Table 2 Coded and natural values of design factors.

Design factors Coded (−1,0, + 1) and natural levels a

Symbol Ingredient −1 0 +1 Jj

x%1 Sugar 60.0 90.0 120.0 30.00

x% 2 Glycerol 1,2,3 design 1.0 4.00 7.0 3.00
4,5 design 0.5 1.25 2.0 0.75
x%3 Potassium sorbate 0.1 0.55 1.0 0.45
x% 4 Citric acid 0.1 0.55 1.0 0.45
x% 5 Sodium diacetate 0.1 0.55 1.0 0.45
x%6 Tartaric acid 0.1 0.55 1.0 0.45
x% 7 Salt 1.0 4.00 7.0 3.00
x%8 Ethanol b 0.5 1.00 1.5 0.50
a b
g/100 g of (flour + starch); g/100 g of muffin.

shelf life is defined as the period in days in which the spoilage caused by microorgan-
isms was first observed. The shelf life was expressed in relation to the corresponding
control.
Microbial analysis. Microbial analysis of bread samples has been carried out in
triplicate. One g of the upper surface of a muffin sample was added to 10 ml of 0.85%
sterilized saline, giving a dilution of 1:10. Sometimes, further dilution of 1:100, 1:1000, or
1:10,000 was needed. After homogenizing this solution, 1 ml was inoculated on plates,
which contained tryptone glucose agar extract (Oxoid Ltd, UK) for total plate count and
Sabouraud dextrose agar (Oxoid Ltd, UK) for yeast and moulds count. Then, the plates
were incubated at 35°C, for 2 days for total plate count, and at 27°C for 5 days for yeast
and moulds count. Finally, the number of colonies were counted and calculated by weight
of muffin (Cfu/g muffin).
 PREDICTION OF THE MICROBIAL SHELF LIFE OF MUFFINS 641

Measurement of pH. An electronic pH-meter was used (704, O Metrohm,

Suisse). After calibration, using standard solutions at pH = 4 and pH = 7, each muffin sus-
pension (10 g of ground muffin were added to 100 g distilled water and the dispersion was
homogenized using a magnetic stirrer) was measured.
Moisture content determination. Water loss was determined by placing a
weighed quantity of a part of muffin sample at temperature 130°C in an air oven for two
hours, and it was expressed as moisture content: [(initial weight of muffin portion- weight
of muffin portion after two hours)/initial weight] × 100. Triplicates were performed for
each sample.[21]

Statistical Design of Experiments

In this article, five first order factorial designs[20] were applied with four or six
design factors in each statistical experiment (3 designs with N = 24 = 16 experiments each
and 2 designs with N = 26 = 64 experiments each). The coded and natural values of design
factors are shown in Table 2. Five repetitions were performed for each experiment trial
(design point). The coded values of the design factors are defined as:

x% j − x% jo
xj = (1)
Jj

where, xj is coded value of the factor, x% j is natural value of the factor, x% j0 is natural value
of the basic level, Jj is variation interval and j is the number of the factor. Both the rela-
tive shelf life (in relation to the corresponding control) of muffin samples and pH were
selected as optimization parameters (yi and pHi, respectively). Faulty observations were
determined with the aid of “Student’s” t-test. The homogeneity of variances of trials was
tested with Cochran's test and Fischer ratio. After calculating the factors coefficients
from the results obtained, the insignificant factors were discarded using the estimated
confidence interval. Then, the adequacy of the models was tested by the aid of the
Fischer ratio.

RESULTS AND DISCUSSION

The values of real and relative bread MFSL regarding to the levels of selected ingre-
dients of bread (sugar, salt, glycerol, potassium sorbate, and acidic agent) with and with-
out ethanol surface spraying are given in Tables 3–6. Five different factorial experiments
were designed and carried out to investigate the effects of the design factors on the MFSL
(optimization parameter).

y1 = f ( x1 , x2 , x3 , x4 ), y2 = f ( x1 , x2 , x3 , x5 ), y3 = f ( x1 , x2 , x3 , x6 ),
y4 = f ( x1 , x2 , x3 , x4 , x7 , x8 ), y5 = f ( x1 , x2 , x3 , x4 , x7 , x8 )

Where yi (i = 1 … 5) is the shelf life relative to control (optimization parameter) and xj

(j = 1 … 8) is the design factor shown in Table 2. The difference between the two last
642 DOULIA, KATSINIS, AND RIGAS

Table 3 Shelf life of muffin with citric acid as acidic agent (first design).

Shelf life b Relative shelf

Trial no. x1 Sugar a x2 Glycerol a x3 K Sorbate a x4 Citric acid a (days) lifec (y1)

Control 75 0 0 0 7.3 1.00

1 120 7 1.0 1.0 31.8 4.36
2 120 7 1.0 0.1 26.8 3.67
3 120 7 0.1 0.1 13.1 1.79
4 120 1 0.1 0.1 12.7 1.74
5 120 7 0.1 1.0 20.0 2.74
6 120 1 0.1 1.0 18.4 2.52
7 120 1 1.0 1.0 19.3 2.64
8 120 1 1.0 0.1 17.7 2.42
9 60 7 1.0 1.0 14.4 1.97
10 60 1 1.0 1.0 10.7 1.47
11 60 1 0.1 1.0 9.3 1.27
12 60 1 0.1 0.1 8.2 1.12
13 60 1 1.0 0.1 10.2 1.40
14 60 7 0.1 1.0 10.6 1.45
15 60 7 0.1 0.1 8.7 1.19
16 60 7 1.0 0.1 10.4 1.42
Basic level 90 4 0.55 0.55 14.6 2.00
a
Natural values of factors (g/100 g flour + starch); b Mean, ± SD was in the range 0.5–1; c ± SD was in the
range 0.08 – 0.18.

Table 4 Shelf life of muffin with sodium diacetate as acidic agent (second design)

Shelf life b Relative

a a a a
Trial no. x1 Sugar x2 Glycerol x3 K Sorbate x4 Na diacetate (days) shelf life c (y1)

Control 75 0 0 0 8.2 1.00

1 120 7 1.0 1.0 33.8 4.12
2 120 7 1.0 0.1 27.7 3.38
3 120 7 0.1 0.1 14.4 1.76
4 120 1 0.1 0.1 13.9 1.70
5 120 7 0.1 1.0 18.4 2.24
6 120 1 0.1 1.0 16.9 2.06
7 120 1 1.0 1.0 19.9 2.43
8 120 1 1.0 0.1 18.1 2.21
9 60 7 1.0 1.0 16.0 1.95
10 60 1 1.0 1.0 11.4 1.39
11 60 1 0.1 1.0 10.0 1.22
12 60 1 0.1 0.1 7.9 0.96
13 60 1 1.0 0.1 11.4 1.39
14 60 7 0.1 1.0 11.7 1.43
15 60 7 0.1 0.1 8.0 0.98
16 60 7 1.0 0.1 11.6 1.41
Basic level 90 4 0.55 0.55 15.6 1.90
b c
a
Natural values of factors (g/100 g flour + starch); Mean, ± SD was in the range 0.5–1.1; ± SD was in the
range 0.08–0.19.
 PREDICTION OF THE MICROBIAL SHELF LIFE OF MUFFINS 643

Table 5 Shelf life of muffin with tartaric acid as acidic agent (third design).

Shelf lifeb Relative

Trial no. x1 Sugara x2 Glycerola x3 K Sorbatea x5 Tartaric acida (days) shelf lifec (y1)

Control 75 0 0 0 8.3 1.00

1 120 7 1.0 1.0 33.6 4.05
2 120 7 1.0 0.1 26.0 3.13
3 120 7 0.1 0.1 14.5 1.75
4 120 1 0.1 0.1 14.0 1.69
5 120 7 0.1 1.0 19.2 2.31
6 120 1 0.1 1.0 16.4 1.98
7 120 1 1.0 1.0 20.3 2.45
8 120 1 1.0 0.1 18.6 2.24
9 60 7 1.0 1.0 15.6 1.88
10 60 1 1.0 1.0 11.3 1.36
11 60 1 0.1 1.0 10.3 1.24
12 60 1 0.1 0.1 8.1 0.98
13 60 1 1.0 0.1 11.5 1.39
14 60 7 0.1 1.0 11.9 1.43
15 60 7 0.1 0.1 8.5 1.02
16 60 7 1.0 0.1 11.6 1.40
Basic level 90 4 0.55 0.55 15.2 1.83
a
Natural values of factors (g/100 g flour + starch); bMean, ± SD was in the range 0.5–1.1; c± SD was in the
range 0.08 – 0.20.

Table 6 Shelf life of muffin with citric acid and ethanol (fourth design).

x3 K x4 Citric Shelf lifec Relative

a a
Trial no. x1 Sugar x2 Glycerol Sorbatea acida x7 Salt a
x8 Ethanolb
(days) shelf life (y1)

Control 75 0 0 0 1.5 0 8.5 1.0

49 60 2.0 1.0 1.0 7 0.5 >50 >5.88
50 60 2.0 1.0 1.0 1 0.5 >50 >5.88
51 60 2.0 0.1 1.0 7 0.5 22 2.6
52 60 2.0 1.0 0.1 7 0.5 >50 >5.88
53 60 0.5 1.0 1.0 7 0.5 >50 >5.88
54 60 2.0 0.1 1.0 1 0.5 16 1.9
55 60 2.0 1.0 0.1 1 0.5 >50 >5.88
56 60 0.5 1.0 1.0 1 0.5 >30 >3.53
57 60 2.0 0.1 0.1 7 0.5 21 2.47
58 60 0.5 0.1 1.0 7 0.5 14 1.65
59 60 0.5 1.0 0.1 7 0.5 >50 >5.88
60 60 2.0 0.1 0.1 1 0.5 15 1.76
61 60 0.5 1.0 0.1 1 0.5 30 3.53
62 60 0.5 0.1 1.0 1 0.5 10.2 1.20
63 60 0.5 0.1 0.1 7 0.5 11.0 1.29
64 60 0.5 0.1 0.1 1 0.5 9.0 1.06
Basic level 90 1.25 0.55 0.55 4 1.0 >80 >9.41
a b c
Natural values of factors (g/100 g flour + starch); g/100 g of muffin; Mean.
644 DOULIA, KATSINIS, AND RIGAS

designs is the pasteurization, which was applied only for the muffin samples of the 5th
design.
After processing these results discarding the insignificant factors, the following
models for the MFSL were derived, expressed in relative values of significant factors:

y1 = 2.07 + 0.66 x1 + 0.25 x2 + 0.35 x3 + 0.23 x4

+ 0.15 x1 x2 + 0.19 x1 x3 + 0.10 x1 x4 + 0.19 x2 x3 + 0.08 x2 x4
− 0.04 x3 x4 + 0.15 x1 x2 x3 + 0.04 x2 x3 x4 − 0.06 x1 x3 x4 (2)

y2 = 1.91 + 0.57 x1 + 0.24 x2 + 0.37 x3

+ 0.19 x5 + 0.14 x1 x2 + 0.18 x1 x3 + 0.19 x2 x3
+ 0.09 x2 x5 + 0.14 x1 x2 x3 + 0.05 x2 x3 x5 (3)

y3 = 1.89 + 0.56 x1 + 0.23 x2 + 0.34 x3 + 0.19 x6

+ 0.13 x1 x2 + 0.17 x1 x3 + 0.15 x2 x3 + 0.10 x2 x6
+ 0.11 x1 x2 x3 + 0.05 x2 x3 x6 (4)

The models of the fourth and fifth design could not be expressed by the relative
equations y4 and y5, since there were no results for the majority of muffin samples, which
showed unlimited shelf life. The above equations are a useful tool for the estimation of
the relative mould-free shelf life. Before using these equations, the coded values of fac-
tors must be calculated from Eq. (1) and data obtained from Table 2. Since the models
are not linear, the relationship between the relative mould-free shelf life and the factors
involved is more complicated, indicating that muffin MFSL is affected not only by the
single factors but also by their interactions. The signs are generally positive, which
means that the increase of any of these factors causes positive effect, leading eventually
to an increase of the MFSL. This effect of the ingredients examined (sugar, salt, potas-
sium sorbate, ethanol, acidic agents) is in accordance with that observed for bread and
bakery products.[3,9,10,18,22]
The addition of the acidic agents at concentration ranging from 0.1 to 1 g/100 g of
flour and starch was generally beneficial at extending muffin MFSL. It was observed in
almost all cases an increase of MFSL with regard to the control, while the other three fac-
tors content remained in the selected interval range (Tables 3–5). This behavior of acidic
agents could be attributed to the reduction of pH in the range 5.5–6, where potassium sor-
bate is more effective. In addition, these agents particularly Na diacetate and citric acid are
considered mould and rope growth inhibitors.[3] It is evident that citric acid had a more
significant inhibiting effect on microbial activities in comparison to Na diacetate or
tartaric acid (Fig. 1). The higher positive synergistic effect of citric acid is, also,
reflected in the higher values of coefficients of almost all the factors engaged (sugar,
glycerol, potassium sorbate, citric acid), and the optimization parameters in the models
(comparison of the parameter y1 with y2 and y3, respectively). Sodium diacetate and tar-
taric acid showed more similar inhibiting efficiency in the muffin MFSL, with Na diace-
tate being slightly more effective than tartaric acid. So, closer values of coefficients of
almost all the factors involved (sugar, glycerol, potassium sorbate, acidic agent) and of
 PREDICTION OF THE MICROBIAL SHELF LIFE OF MUFFINS 645

Figure 1 Effect of acidic agent on the relative shelf life of muffin (sugar, glycerol, and K sorbate fixed at their
maximum levels 120 g, 7 g, and 0.55 g, respectively, by 100 g of flour and starch).

the optimization parameters in the models (comparison of the parameter y2 with y3) were
observed. The rank of the effectiveness of the acidic agents is illustrated in Fig. 1, when
the rest factors are fixed at their maximum level. The interactions of citric acid, sodium
diacetate and tartaric acid with potassium sorbate are shown in Figs. 2–4 respectively. It
should be noticed the significant contribution of the three other factors, namely sugar,
glycerol, and potassium sorbate to the improvement of MFSL, which is reflected in the
values of their coefficients in the relative equations. Since the microbial inhibitory
action of these ingredients has been established, their positive influence on MFSL was
expected.[3,7,10,13,19]
The surface spraying of the muffins with ethanol (0.5–1.5 g/100 g of muffin)
caused significant prolongation of their shelf life. Among the 320 muffin samples (64
statistical experiments), only 80 (25%) showed microbial spoilage and consequently
definite MFSL The MFSL of these samples (with minimum sugar content and sprayed
with ethanol 0.5 g/100g of muffin) are given in Table 6. Fifty percent of the total sam-
ples (160 samples sprayed with ethanol 1.5 g/100 g of muffin) had MFSL higher than
70 days. The rest—25% of the total samples (80 samples with maximum sugar content
and sprayed with ethanol 0.5 g/100 g of muffin)—showed MFSL higher than 50 days.
The beneficial effect of ethanol has been proved for a variety of baked
goods.[3,16,18,22,23]
The 360 pasteurized and sprayed with ethanol (0.5–1.5 g/100 g of muffin) muffin
samples had extremely high MFSL. The MFSL of the majority of the samples (90%) was
higher than 100 days, indicating that the combination of pasteurization with ethanol
646 DOULIA, KATSINIS, AND RIGAS

Figure 2 Dependence of the relative shelf life of muffin (contour lines) on citric acid and potassium sorbate con-
tents with sugar and glycerol fixed at their basic levels (90 g and 4 g, respectively, by 100 g flour and starch).
Control shelf life 7.3 days. Coded values 1 to + 1 correspond to 0.1 to 1 g/100 g of flour and starch.

addition in this recipe was extremely effective. Both methods, pasteurization and the
use of chemical preservatives, have been extensively used in bakery products for
MFSL extension.[6,18,22] The effectiveness of the investigated methods is depicted in
Fig. 5.
The five different factorial designs were used also to investigate the effects of the
(same as above for the MFSL) design factors on the pH of the muffins (Tables 1 and 2).

pH1 = f ( x1 , x2 , x3 , x4 ), pH 2 = f ( x1 , x2 , x3 , x5 ), pH3 = f ( x1 , x2 , x3 , x6 ),
pH 4 = f ( x1 , x2 , x3 , x4 , x7 , x8 ), pH 5 = f ( x1 , x2 , x3 , x4 , x8 ) (5)

After processing these results discarding the insignificant factors, the following lin-
ear models for the pH were derived, expressed in relative values of significant factors:
pH1 = 6.63 − 0.73 x4 (6)

pH 2 = 7.08 − 0.40 x5 (7)

pH3 = 6.56 − 0.74 x6 (8)

 PREDICTION OF THE MICROBIAL SHELF LIFE OF MUFFINS 647

Figure 3 Dependence of the relative shelf life of muffin (contour lines) on Na diacetate and potassium sorbate
contents with sugar and glycerol fixed at their basic levels (90 g and 4 g, respectively, by 100 g of flour and
starch). Control shelf life 7.3 days. Coded values 1 to + 1 correspond to 0.1 to 1 g/100 g of flour and starch.

The effect of acidic agent on the muffin pH is given in Fig 6. The pH in the fourth
and fifth design remained almost constant at 7.5 and 7.35, respectively. This fact does not
mean that some of the selected factors had no effect on the pH, but the models were very
complicated due to the number of interactions (57 interactions). Thus, it was considered
that a further investigation was worthless.
As far as the total microbial analysis is concerned, the results agree with the mould
observations. None or extremely few colonies for yeast and mould were counted in each
muffin at time less than its MFSL. The microbial number started to rise at times, exceed-
ing the MFSL.

CONCLUSION
The mould free shelf life of muffins can be predicted using the equations derived by
simple first order factorial designs. The factors investigated were sugar, glycerol, potas-
sium sorbate, acidic agent, salt, ethanol surface spraying, and pasteurization. The models
revealed the effect of the selected ingredients and their interactions on the shelf life. The
combination of citric acid (an acidic agent), potassium sorbate (a conventional preserva-
tive), ethanol spraying of muffin surface, and pasteurization was proven to be the more
effective preservation method, followed by the combination of citric acid, potassium
648 DOULIA, KATSINIS, AND RIGAS

Figure 4 Dependence of the relative shelf life of muffin (contour lines) on tartaric acid and potassium sorbate
contents with sugar and glycerol fixed at their basic levels (90 g and 4 g, respectively, by 100 g flour and starch).
Control shelf life 8.3 days. Coded values 1 to + 1 correspond to 0.1 to 1 g/100 g of flour and starch.

Figure 5 Shelf life of muffins containing potassium sorbate as preservative with different methods.
 PREDICTION OF THE MICROBIAL SHELF LIFE OF MUFFINS 649

Figure 6 Effect of acidic agent on the pH of muffin (sugar, glycerol, and K sorbate fixed at their basic levels
90 g, 4 g, and 0.55 g, respectively, by 100 g flour and starch).

sorbate, and ethanol spraying. The rank of acidic agents effectiveness in MFSL was citric
acid > sodium diacetate > tartaric acid.

REFERENCES
1. Singh, N.; Gupta, S.; Sohdi, N. S.; Singh, R.P.; Effect of additives on dough and cookie making
properties of flour. Intern. Journal of Food Properties 2002, 5 (3), 547–562.
2. Krog, N.; Rilsom, T.; Larsson, K. Applications in the Food Industry: I. In Encyclopedia of
Emulsion Technology Applications, Vol. 2; Becher, P.; Eds.; Marcel Dekker: New York, 1985;
321–365.
3. Pyler, E.J. Baking Science and Technology, Vol. II; Sosland Publishing Company: Kansas City,
1988; 210–236.
4. Bennion, E.B.; Bamford, G.S.T. The Technology of Cake Making; Bent, A.J.; Eds.; Blackie
Academic and Professional: London, UK, 1997; 252–270, 386–390.
5. Shepherd, I.S.; Yoell, R.W. Food Emulsions; Marcel Dekker: New York, USA, 1976; 215 pp.
6. Stauffer, C.E. Functional Additives for Bakery Foods; AVI Book: New York, USA, 1990;
189 pp.
7. Davidson, P.M. Control of Microorganisms with Chemicals. In Control of Foodborn Microor-
ganisms; Juneja, V.K.; Sofos, J.N.; Eds.; Marcel Dekker: New York, 2002; 165–190.
8. Matz, S.A. Technology of the Materials of Baking; Elsevier Science Publishers: Amsterdam,
Netherlands, 1989; 260–262.
9. Cauvain, S.P; Young, L.S. Bakery Food Manufacture and Quality; Blackwell Science: Oxford,
UK, 2000; 188–201.
10. Doerry, W.T. Breadmaking Technology; American Institute of Baking: Manhatttan, USA, 1995;
156–159.
650 DOULIA, KATSINIS, AND RIGAS

11. Sofos, J.N. Sorbic Acid. In Natural Food Antimocrobial Systems; Naidu, A.S.; Eds.; CRC Press:
New York, 2000; 637–659.
12. Grundy, J.G. Preservatives. In Baked Goods Freshness, Technology, Evaluation and Inhibition
of Staling; Ronald, E.H.; Henry, F.Z.; Eds.; Marcel Dekker Inc: New York, 1996; 189–204.
13. Pateras I.M. Bread Spoilage and Staling. In Technology of Breadmaking; Cauvain, S.P.; Young,
L.S.; Eds.; Aspen Publishers: Gaithersburg, MD, 1999; 240–244.
14. Seiler, D.A.L.; Russel, N.J. Ethanol as a Food Preservative. In Food Preservatives; Russel, N.J.;
Gould G.W.; Eds.; Blackie and Son Ltd.: New York, 1991; 153–171.
15. Yang, T.C.S. Ambient Storage. In Food Storage and Stability; Taub, I.A.; Eds; CRC Press: New
York, 1998; 436–438.
16. Legan, J.D. Mould spoilage of bread: the problem and some solutions. International Biodeterio-
ration and Biodegradation 1993, 32, 33–53.
17. Karaoglu, M.M.; Kotancilar, H.G.; Gurses, M.; Microbiological characteristics of part-baked
white pan bread during storage. Intern. Journal of Food Properties 2005, 8 (2), 355–365.
18. Doulia, D.; Katsinis, G.; Mougin, B. Prolongation of the microbial shelf life of wrapped part-
baked baguettes. International Journal of Food Properties 2000, 3 (3), 447–457.
19. Doulia, D.; Rigas, F.; Katsinis, G. Multiparametric investigation of mould-free shelf life of
bread via factorial design. International Journal of Food Properties 2000, 3 (3), 363–374.
20. Biles, W.E.; Swain, J.J. Optimization and Industrial Experimentation; Wiley-Inerscience: New
York, 1980.
21. International Association for Cereal Chemistry. ICC-Standard No. 110/1; Verlag Moritz
Schafer: Detmold, 1960.
22. Matz, S.A. Bakery Technology; Elsevier Science Publishers: Amsterdam, 1989; 135–167.
23. Seiler, D. A. L. Bakery Products. In Principles and Applications of Modified Atmosphere
Packaging of Foods; Blakistone, B.A.; Blackie.; Eds; Academic and Professional: London,
1998, 135–157.