See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/274640824

Fat replacement in shortbread cookies using an emulsion filled gel based on

inulin and extra virgin olive oil

Article in LWT · September 2015

DOI: 10.1016/j.lwt.2015.03.063

CITATIONS READS

82 1,168

5 authors, including:

Mariagrazia Giarnetti Vito Michele Paradiso

Università degli Studi di Bari Aldo Moro Università del Salento
27 PUBLICATIONS 341 CITATIONS 169 PUBLICATIONS 2,823 CITATIONS

SEE PROFILE SEE PROFILE

Francesco Caponio Carmine Summo

Università degli Studi di Bari Aldo Moro Università degli Studi di Bari Aldo Moro
276 PUBLICATIONS 5,864 CITATIONS 242 PUBLICATIONS 4,164 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Great Teachers,Mentors and Contemporaries View project

Agro Business View project

All content following this page was uploaded by Antonella Pasqualone on 23 November 2021.

The user has requested enhancement of the downloaded file.

 LWT - Food Science and Technology xxx (2015) 1e7

Contents lists available at ScienceDirect

LWT - Food Science and Technology

journal homepage: www.elsevier.com/locate/lwt

Fat replacement in shortbread cookies using an emulsion filled gel

based on inulin and extra virgin olive oil
Mariagrazia Giarnetti, Vito Michele Paradiso*, Francesco Caponio, Carmine Summo,
Antonella Pasqualone
Department of Soil, Plant and Food Sciences, University of Bari Aldo Moro, Via Amendola 165/A, 70126 Bari, Italy

a r t i c l e i n f o a b s t r a c t

Article history: The European Food Safety Agency recommends low intakes of saturated fatty acids. This study evaluated
Received 7 October 2014 the replacement of 50% and 100% of butter in shortbread cookies with an emulsion filled gel (EFG) based
Received in revised form on inulin and extra virgin olive oil (EVOO), and investigated its impact on product characteristics (volatile
9 March 2015
profile, fracture behaviour and crumb pore size, sensory properties). Cookies containing both 50% and
Accepted 15 March 2015
Available online xxx
100% EFG had thinner pore walls than the control cookies, without EFG. The total substitution of butter
with EFG gave products with higher number of small-sized pores and lower level of fracture stress, with
poorer sensory properties than control cookies. The replacement of 50% of the butter fraction of
Keywords:
Emulsion filled gel
shortbread with EFG based on inulin and EVOO did not substantially affect cookie microstructure,
Fat replacers compared to control. Cookies with 50% butter replacement showed comparable fracture properties, and
Image analysis were well accepted by consumers, resulting a product with potentially healthier properties (19% less total
Fracture analysis fat, 39% less saturated fats) respect to control cookies.
Inulin © 2015 Elsevier Ltd. All rights reserved.

1. Introduction cholesterol/low density lipoprotein (LDL) concentrations

Fat mimetics or replacements in food products are highly
desirable as dietetic alternatives for reducing calories, fat, and
cholesterol intake (Haque & Ji, 2003). However, it is often difficult
to preserve desirable sensory properties of the food product with
regards to breaking strength, crumb texture, moisture content, and
mouthfeel (Zahn, Pepke, & Rohm, 2010).
Shortbreads are biscuits characterized by being brittle due to
their high quantities of fat (Manohar & Rao, 1999), containing high
level of saturated fatty acids (SFA) (Caponio, Summo, Delcuratolo, &
Pasqualone, 2006). Their recipe can vary widely: fat can range be-
tween 20 and 60% (as percentage of weight of flour), and sucrose
can range between 25 and 55% (Baltsavias, Jurgens, & van Vliet,
1999).
Recently, the European Food Safety Agency recommended that
intakes of SFA should be as low as possible, on the basis of the
relationship between dietary SFA intake and increased blood
* Corresponding author. University of Bari ‘‘Aldo Moro’’, Department of Soil, Plant
and Food Sciences, Food Science and Technology Unit, Via Amendola 165/A, I-70126
Bari, Italy. Tel.: þ39 (0)80 5442272.
E-mail address: vito.paradiso@uniba.it (V.M. Paradiso).
(European Food Safety Agency, 2010). Hence, the use of fat mi-
metics, shortenings, and emulsifiers with a healtier fatty acid pro-
file is an alternative for developing new biscuits recipes.
Inulin is a polydisperse b(2e1) fructan. The presence of b(2e1)
bond prevents it from being digested like a typical carbohydrate
and is responsible for its reduced energetic value and dietary fibre
behaviour (Niness, 1999). When thoroughly mixed with water or
another aqueous solution, inulin forms a particle gel network
resulting in a white creamy structure with a short spreadable na-
ture, which can easily be incorporated into foods to replace fat
(Paradiso et al., 2015).
The use of inulin as fat replacer in biscuits formulation was re-
ported by several authors coming to conflicting results. Zoulias,
Oreopoulou, and Tzia (2002) examined the effect of carbohy-
drate- or protein-based fat mimetics, used to replace up to 50% fat,
on the textural properties of cookies. In particular, the addition of
inulin as fat replacer resulted in more tender low-fat cookies.
Devereux, Jones, Mccormack, and Hunter (2003) investigated on
the consumer acceptability of low-fat foods containing inulin and
oligofructose. The results indicated that inulin had a noticeable
influence on the texture of baked goods, showing that it has not the
ability to “shorten” gluten strands in baked flour products as fat
does. Cookies containing less fat tend to have higher moisture

http://dx.doi.org/10.1016/j.lwt.2015.03.063
0023-6438/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Giarnetti, M., et al., Fat replacement in shortbread cookies using an emulsion filled gel based on inulin and
extra virgin olive oil, LWT - Food Science and Technology (2015), http://dx.doi.org/10.1016/j.lwt.2015.03.063
2 M. Giarnetti et al. / LWT - Food Science and Technology xxx (2015) 1e7
content and to be less crisp and less appreciated by consumers
(Drewnowski, 1997). Rodríguez-García, Laguna, Puig, Salvador, and
Hernando (2013) tested five formulations with different short-
ening/inulin proportions. At higher fat replacement levels, flour
was more available for hydration and gluten formation, leading to
harder biscuits. The authors concluded that shortening may be
partially replaced, up to 20%, by inulin.
The use of emulsion filled gels (EFG) offers a range of tools to
engineer foods without compromising the organoleptic properties
of the product (Sala, de Wijk, van de Velde, & van Aken, 2008). Our
recent paper (Paradiso et al., 2015) was aimed to setting up the
production of EFG based on inulin and extra virgin olive oil (EVOO).
The aim of this work was to replace butter with an EFG based on
inulin and EVOO in order to obtain shortbread cookies with a
healthier fatty acid profile and with sensory properties close to
those of full-fat ones.
2. Materials and methods
2.1. Raw materials
Commercially available refined soft wheat flour (type 00, ac-
cording to Italian regulations, having particle size similar to that of
US pastry flours), leavening agent (sodium bicarbonate and tartaric
acid), butter, EVOO, fresh whole eggs and food grade sodium
chloride were purchased from retailers. Inulin (Orafti® HPX, with
degree of polymerization  5 accounting for not less than 99.5% of
total, Beneo-Orafti SA, Oreye, Belgium) and soy lecithin were kindly
provided by Eigenmann & Veronelli SpA (Milan, Italy). All reagents
were purchased from Sigma Aldrich (Milan, Italy), unless otherwise
stated.
2.2. Preparation of the EFG
The preparation of the EFG was made according to a previous
paper (Paradiso et al., 2015). The amounts of the ingredients were
selected to obtain a homogeneous EFG, with the appearance of a
white-yellowish cream, and consistency similar to that of com-
mercial spreadable cheeses (Paradiso et al., 2015). The formulation
was: EVOO (37% w/w), inulin (19% w/w), soy lecithin (2% w/w), and
water (42% w/w). Homogenization was performed by means of
high power ultrasound using a 200 W transducer (Sonopuls HD
3200, Bandelin Electronic, Berlin, Germany) with 6 mm diameter
tapered tip (KE 76, Bandelin Electronic, Berlin, Germany) for 5 min.
2.3. Preparation of shortbread cookies
Three shortbread cookie typologies were prepared by using the
same quantity of all the ingredients and different butter:EFG pro-
portions: 100:0 (EFG-0), 50:50 (EFG-50), and 0:100 (EFG-100)
Table 1
Cookies formulations (g).
EFG-0 EFG-50
Flour 300 300
Buttera/EFG 130 130
of which Fat/oil 106.6 77.3
Water 13.4 34
Inulin e 12.4
Soy lecithin e 1.3
Sucrose 100 100
Egg 120 120
Leavening agent 8 8
Sodium chloride 2 2
a
82% fat content was considered for butter.
Please cite this article in press as: Giarnetti, M., et al., Fat replacement in shortbread cookies using an emulsion filled gel based on inulin and
extra virgin olive oil, LWT - Food Science and Technology (2015), http://dx.doi.org/10.1016/j.lwt.2015.03.063
(Table 1). The dough, sheeted to a thickness of 5 mm and cut into
rectangular shape (60 mm  45 mm), was baked in an air circu-
lation oven at 180  C for 25 min. The trays were filled with all three
types of cookies according to the Latin-square design (6  6) to
minimize any effect of tray location. Two independent replicate
baking experiments were carried out (12 cookies per type in each
experiment). Cookies for the consumer test were prepared in
further baking procedures.
2.4. Chemical characterization of cookies
Moisture content of cookies was determined according to the
AACC method 44-15A (AACC, 2000). Fat was extracted and deter-
mined by Soxhlet apparatus using diethyl ether as solvent. Fatty
acid composition was determined according to Pasqualone,
Paradiso, Summo, Caponio, and Gomes (2014).
2.5. Image analysis
Images of the shortbread cookies section (10-mm thick) were
captured using a digital camera DCM-TZ8 (Panasonic, Newark,
USA). A central field of view (FOV) (35 mm  6 mm), capturing the
majority of the area of each section, was evaluated for each image.
Images were analysed by ImageJ software (National Institutes of
Health, Bethesda, USA) using the Otsu's algorithm for setting the
threshold. Crumb grain was characterized by enumerating the
pores present in eight preselected dimensional classes (class
1 < 0.005 mm2; 2: 0.005e0.010 mm2; 3: 0.010e0.015 mm2; 4:
0.015e0.020 mm2; 5: 0.020e0.025 mm2; 6: 0.0.25e0.030 mm2; 7:
0.030e0.035 mm2; 8 > 0.035 mm2). The number of pores and the
area occupied by each class (expressed as percentage of the total
number of pores and of the total pore area, respectively) was
evaluated.
The cookie images were also analysed by grey-level mathe-
matical morphology. Crumb images were considered to contain
grey level information from pixels, of which the darkest individuals
belong to pore and the brightest belong to pore wall. It is possible to
characterise crumb grain of cookies by applying successively two
basic morphological operators (erosion or dilation) of increasing
size and measuring at each step the sum of grey levels (ero-
sionedilation curves), as follows. Erosion and dilation, basic
transformations in mathematical morphology (Devaux, Robert,
Melcion, & Le Deschault de Monredon, 1997), were applied to the
image through a structuring element of a given size (in pixel) and
shape (usually squares or oriented lines). Twelve erosion and
twelve dilation steps were applied using squared structuring ele-
ments, starting from 3 pixels  3 pixels squares. Erosion modifies
the grey level values by giving to each pixel the minimum value
observed within the surrounding window defined by the struc-
turing element. It consenquently causes a size reduction of bright
objects and a decrease of the sum of the grey level in the image.
Bright objects for which at least one dimension is smaller than the
structuring element are removed. Dilation, on the opposite, applies
to each pixel the maximum value observed within the structuring
EFG-100 element, causing an increase of bright object size and, therefore, a
300
reduction of dark objects. Progressive erosion and dilation steps
130 based on structuring elements of increasing size cause bright and
48.1 dark objects, respectively, progressively disappear. This allows to
54.6 obtain erosionedilation curves of the sum of grey level as a function
24.7
of erosion and dilation steps. A complete granulometric texture
2.6
100 characterisation is obtained by merging the curves assessed for
120 erosion and dilation steps. The left part of the curves (dilation)
8 represents grey-level variation caused by pore size, from the largest
2 to the smallest size. The right part (erosion) represents grey level
variations caused by pore walls thickness, from the smallest to the
 M. Giarnetti et al. / LWT - Food Science and Technology xxx (2015) 1e7
largest thickness. The higher the peak, the higher the number of
small objects (Devaux et al., 1997). The data were obtained by
measuring crumbs in ten different images (five from each baking
experiment) for each formulation.
2.6. Determination of fracture properties
The fracture properties of the shortbread cookies were studied
by three-point bending tests at room temperature. A Texture
Analyzer (Z1.0 TN, Zwick GmbH & Co. KG, Ulm, Germany) was used
with 1000 N load-cell. The distance between the support bars was
4 cm and probe travelling speed was set at 5 mm/min; the cookies
were placed on supports with their top surface down (Baltsavias,
Jurgens, & van Vliet, 1997). The downward movement was
continued till the cookie broke. At least eight cookies were tested
per cookie formulation. Fracture stress (N/m2) values were recor-
ded and the average value reported.
2.7. Sensory evaluation
2.7.1. Descriptive sensory analysis
Quantitative Descriptive Analysis (QDA) was performed by a
panel of ten assessors. The general guidelines for the selection,
training, and monitoring of assessors (International Standard
Organization, 2003) were used. Each assessor participated in four
1-h training sessions where the panel developed a lexicon of 11
attributes through consensus (Table 2). All three types of samples
were evaluated by scoring the intensity of each attribute on an
unstructured 10 cm scale (0 ¼ low intensity, 10 ¼ very strong in-
tensity). The sample presentation order was randomized between
and within panelists. All samples were coded by a three-digit
alpha-numeric code. Sparkling water was served for cleansing the
palate between samples (Laguna, Varela, Salvador, Sanz, & Fiszman,
2012). Samples were analysed in duplicate.
2.7.2. Acceptance testing
A total of 120 consumers (students and employees at University
of Bari; age 18e55, mean 25.3, 59 males and 61 females) were
recruited for the consumer test. The samples, encoded using a
three-digit alpha-numeric code, were offered to each consumer
according to a randomized complete block design to avoid the
“order effect” (MacFie, Bratchell, Greenhoff, & Vallis, 1989). Each
consumer also received a glass of sparkling water to cleanse the
Table 2
Sensory descriptors used for shortbread cookies.
Attribute Description
Odor
Caramel Odor reminiscent produced when caramelizing sugar
without burning it
Grain Odor reminiscent of freshly harvested wheat or rolled oats
or wheat
Buttery Odor of butter fat type
Flavor
Grain Favour of grains of wheat or wheat flakes
Buttery Flavor reminescent of butter
Taste
Sweetness Taste caused by sugar and other sweeteners
Saltiness Taste characterized by a solution of sodium chloride
Texture
Consistency Resistance of the structure of the cookie during mastication
Crunchiness Ease of breaking at first bite
Crush Ease of biscuit breaking into pieces during chewing and
texture of the pieces
Dryness Degree of dryness or humidity perceived in the mouth
when the cookie crumbles
Please cite this article in press as: Giarnetti, M., et al., Fat replacement in shortbread cookies using an emulsion filled gel based on inulin and
extra virgin olive oil, LWT - Food Science and Technology (2015), http://dx.doi.org/10.1016/j.lwt.2015.03.063
3
Table 3
Fats and moisture contents in cookies (g/100 g).
EFG-0 EFG-50 EFG-100
Fats 22.8 18.4 12.2
of which Saturated 15.8 9.6 2.1
Unsaturated 7.0 8.8 10.1
Moisture 3.7 4.5 7.1
palate between samples. Consumers were asked to express an
overall acceptability judgment on a continuous 10 cm-line
anchored at the left end to the judgment “I do not like it at all” and
at the right end to “I like it very much”.
2.8. Volatile compounds analysis
The analysis of volatile compounds was carried out according to
Di Mattia et al. (2014), apart the use of a 50/30 mm Divinylbenzene/
Carboxen/Polydimethylsiloxane (DVB/CAR/PDMS) fibre (Supelco,
Bellefonte, USA). The amounts of volatile compounds were
expressed as integrated peak area. Three samples from each baking
experiment were analysed per cookie formulation.
2.9. Statistical analysis
One-way analysis of variance (ANOVA) e followed by Fisher's
test (LSD) for multiple comparisos at p < 0.05 e was carried out on
the experimental data using the XLStat software (Addinsoft SARL,
New York, USA).
3. Results and discussion
3.1. Chemical characterization of cookies
In 50% butter replaced cookies the total fat content decreased by
19%, respect to control, while in fully butter-replaced the decrease
was 46% (Table 3). The decrease of saturated fats was even more
pronounced, being by 39% and 87% in EFG-50 and EFG-100
respectively. According to European regulations (Official Journal
of European Union, 2006), EFG-50 cookies could be labelled as
“reduced saturated fat” as compared to control cookies (EFG-0), and
EFG-100 could be labelled as both “reduced fat” and “reduced
saturated fat”, showing a saturated fat content (2.1 g/100 g) very
near to the threshold provided for “low-saturated fat” claim (1.5 g/
100 g).
The different amounts of water added when replacing butter
with EFG produced an increase in the final moisture levels in
cookies. Moisture, in fact was 3.7 g/100 g in EFG-0 cookies, and
increased at 4.5 and 7.1 g/100 g in EFG-50 and EFG-100 respectively.
3.2. Image analysis
With respect to the image analysis of the section of cookies, the
number of pores and the total percent area occupied by the pores of
each dimensional class for EFG-0, EFG-50, and EFG-100 are re-
ported in Fig. 1A and B, respectively. A significantly greater number
of small pores were present in EFG-100 samples (<0.005 mm2),
which accounted for more than 10% of the total pore-area. An
opposite trend was observed for pores in the 0.005e0.01 mm2 and
>0.035 mm2 dimensional classes. Larger pores (>0.035 mm2) were
found in all samples but, while they represented less than 58% in
EFG-100 crumb, they accounted for more than 75% of the total
pore-area in the EFG-0 and EFG-50 cookies. The differences in the
microstructure can be attributed to the role played by fats during
the formation of the cellular structure. Fats, in fact, decrease cell
4 M. Giarnetti et al. / LWT - Food Science and Technology xxx (2015) 1e7

wall permeability to gases and slow down permeability changes
induced by temperature (Daniels & Fisher, 1976), leading to greater
expansion. The expansion of gas cells in cookies would have been
enhanced as the total fat content of dough increased. It can be
hypothesized that also the different water amounts influenced pore
size: higher water content in the dough would have led to a more
homogeneous distribution of the leavening agent in the dough it-
self. This agrees with Yang et al. (2012) who observed similar
microstructure differences: larger number of small pores were
found in biscuits with a hydrophylic flavour solvent in the formu-
lation, that would have better dispersed the leavening agent added
to the dough.
Erosion-dilation curves (Fig. 2) exhibit one single peak, indi-
cating that the main variations in the grey-level were found in the
very first steps of erosion and dilation. The two parts of the curves
that correspond to the erosion and dilation steps appear to be quite
symmetrical. The curves showed that the higher the content of EFG,
the narrower and the higher the curve peak. As grey-level varia-
tions were normalized, the higher the middle peak, the higher the
number of smaller objects observed (Lassoued, Babin, Della Valle,
Devaux, & Re guerre, 2007). The erosion or dilation steps can be
related to the particle size: one pixel has a size of about 0.038 mm.
As the particles are eroded on every side, one erosion step makes
the diameter decrease by 2 pixels (i.e. 0.076 mm). According to the
obtained data, shortbread cookies with partial or total substitution
of fat with EFG showed thinner pore walls; moreover, more small
sized objects (pores) were found in the section of EFG-100 short-
bread cookies, with a concomitant decrease of the number of larger
ones, in agreement with results obtained from pore size analysis.
Morphological results showed that crumb of EFG-50 cookies was
very similar to that of control cookies (EFG-0).
3.3. Fracture properties
Measurement of the fracture stress (Table 4) showed that total
butter replacement in EFG-100 produced a significant decrease
(p < 0.05) of the stress value, which correlates with hardness. For
50% butter replacement by EFG no significant difference with the
control was found. The change in textural properties observed in
EFG-100 cookies could be due to the modification of microstruc-
ture, discussed above, induced by the total fat replacement.
Rodríguez-García et al. (2013) progressively replaced shortening by
adding increasing amounts of inulin to shortdough biscuits. They
found an increase of fracture stress as increasing amounts of
shortening were replaced, but at the highest level of replacement
(40%) they observed a decrease of this parameter due to the
insufficience of fat which, in turn, determined scarce matrix
lubrication and insufficient development of a linked structure.
Similar effects could have been induced by total replacement of
butter by EFG. Moreover, higher moisture could have contributed to

Fig. 1. Number of pores as percentage of total number (A) and area of pores as percentage of total area (B) for the selected eight dimensional classes for cookies in which the fat was
substituted by an emulsion filled gel based on inulin and extra virgin olive oil. Levels of fat substitution: 0% (EFG-0, dark grey); 50% (EFG-50, pale grey), and 100% (EFG-100, white).
Different letters mean a significant difference at p < 0.05.

Please cite this article in press as: Giarnetti, M., et al., Fat replacement in shortbread cookies using an emulsion filled gel based on inulin and
extra virgin olive oil, LWT - Food Science and Technology (2015), http://dx.doi.org/10.1016/j.lwt.2015.03.063
 M. Giarnetti et al. / LWT - Food Science and Technology xxx (2015) 1e7 5

‘caramel odor’, ‘buttery odor’, ‘buttery flavour’, ‘crunchiness’ and

‘crush’. The control cookies (EFG-0) generally showed the most
intense attributes, in particular resulted crispy and crushable. The
EFG-50 shortbread cookies were not distinguishable for ‘sweet-
ness’, ‘consistency’ and ‘dryness’, whereas showed weaker in-
tensities of the descriptors ‘buttery flavour’ and ‘buttery odor’, as
well as ‘caramel odor’. Lower ‘crunchiness’ and ‘crush’ were
perceived in these biscuits, probably due to the higher residual
moisture more than to micro-structural causes. The EFG-100
cookies exhibited the less intense ‘caramel odor’, ‘buttery odor’
and ‘buttery flavour’. Also textural sensory properties of EFG-100
samples were perceived as impaired, according micro-structural
and fracture results.

3.5. Volatile compounds analysis

The majority of volatile compounds of cookies (Table 6) were

Fig. 2. Erosion-dilation curves for crumbs of cookies in which the fat was substituted
by an emulsion filled gel based on inulin and extra virgin olive oil. Levels of fat sub-
stitution: 0% (EFG-0, continuous line); 50% (EFG-50, dashed line), and 100% (EFG-100,
dotted line). Step units are given in mm, one step corresponding to a variation of
2  0.038 mm.
the drop of fracture stress in EFG-100 cookies. On the other hand,
50% replacement of butter by EFG seemed not to significantly affect
the matrix properties. This could mean that the EFG, at that
amount, effectively acted as fat replacer, contributing to the linked
structure of the continuous matrix based on lipids and sugars, as
described by Baltsavias et al. (1997) and Chevallier, Colonna,
Buleon, and Della Valle (2000). Other authors (Forker, Zahn, &
Rohm, 2011; Laguna, Primo-Martín, Varela, Salvador, & Sanz,
2014; Sudha, Srivastava, Vetrimani, & Leelavathi, 2007) observed
increases of fracture stress when replacing shortenings, due to
gluten development. In our case, no increase was observed, either
with 50% or with 100% fat replacement, suggesting that the water
contained in the EFG was not available for gluten development and
that EFG was able to surround and isolate gluten and starch,
inhibiting the formation of gluten network and breaking the
protein-starch continuity (Ghotra, Dyal, & Narine, 2002; Laguna
et al., 2012).
3.4. Descriptive sensory analysis
produced during baking as a result of thermal reactions. The vol-
atile profiles of EFG-0 cookies were characterized by the predom-
inance of compounds deriving from butter heating. In particular,
methyl ketones were the most abundant compounds and derived
from the decarboxylation of b-keto acids produced from tri-
acylglycerides by heat treatment (Kinsella, Patton, & Dimick, 1967;
Nawar, 1985; Parks, Keeney, Katz, & Schwartz, 1964). The EFG-
0 cookies presented significantly higher contents of 2-butanone,
2-heptanone, 2-nonanone and 2-undecanone, accountable for
higher buttery and sweetness perception (Kato, 2003) in sensory
evaluation. The levels of these compounds significantly decreased
as butter replacement increased, weakening also the odor and
flavour intensity. On the other hand, as the amount of EFG
increased and butter fat was replaced by olive oil, autoxidation of
unsaturated fatty acids became the predominant source of lipid-
derived volatiles. In particular, hexanal, deriving from autoxida-
tion of linoleic acid via 13-hydroperoxide formation (Frankel, 1991),
was significantly higher in EFG-100 cookies.
Maillard reaction was another important source of volatile
compounds. They included: (i) the Strecker degradation aldehyde
3-methyl-butanal; (ii) furan derivatives, such as furfural, obtained
from the decomposition of 3-deoxyosones (Belitz, Grosch, &
Schieberle, 2009) and reported as the most important caramel-
like odorant of wheat bread crust (Schieberle & Grosch, 1994), g-
butyrolactone, and 2-(5H)-furanone; (iii) pyrazines, that
Table 5
Sensory evaluation scores (mean value ± SD), followed by Fisher's (LSD) test for
multiple comparison, of short-bread cookies (n ¼ 2).
Attribute EFG-0 EFG-50 EFG-100

Significant differences (p < 0.05) were observed among the Odor

samples in 9 out of 12 sensory attributes evaluated (Table 5). No Caramel 5.6 ± 2.1a 4.0 ± 2.1b 1.6 ± 2.0c
Grain 3.0 ± 2.4a 2.3 ± 2.2a 1.9 ± 2.0a
differences in the descriptors ‘grain odor’ and ‘grain flavour’, and in
Buttery 7.2 ± 1.6a 4.6 ± 1.9b 1.2 ± 1.4c
‘saltiness’ were observed. The strongest differences were in Flavor
Grain 2.8 ± 2.7a 2.6 ± 2.1a 2.7 ± 2.0a
Buttery 6.0 ± 1.8a 3.8 ± 1.8b 1.7 ± 2.0c
Table 4 Taste
Effect of fat type on the fracture stress of the shortbread cookies Sweetness 5.2 ± 2.1a 4.6 ± 1.9ab 3.7 ± 2.2b
(n ¼ 8). Saltiness 0.6 ± 0.7a 0.4 ± 0.6a 0.3 ± 0.5a
Texture
Samples Fracture stress (kN/m2)
Consistency 4.8 ± 1.8b 5.3 ± 1.6ab 5.8 ± 1.2a
EFG-0 10.69 ± 2.35a Crunchiness 7.5 ± 1.0a 6.3 ± 1.3b 5.3 ± 1.8c
EFG-50 9.41 ± 0.56a Crush 7.9 ± 0.4a 6.4 ± 0.9b 5.2 ± 1.2c
EFG-100 2.70 ± 1.06b Dryness 6.5 ± 1.3a 6.1 ± 0.9ab 5.5 ± 1.8b

EFG-0: shortbread cookies made with butter; EFG-50: short-
bread cookies made with 50% of replacement of butter with
emulsion filled gel; EFG-100: shortbread cookies made with
100% of replacement of butter with emulsion filled gel. Different
letters mean a significant difference at p < 0.05.
EFG-0: shortbread cookies made with butter; EFG-50: shortbread cookies made
with 50% of replacement of butter with emulsion filled gel; EFG-100: shortbread
cookies made with 100% of replacement of butter with emulsion filled gel. Scores
range: 0: low intensity; 10: very strong intensity. Same letter in a row do not differ
significantly (p < 0.05).

Please cite this article in press as: Giarnetti, M., et al., Fat replacement in shortbread cookies using an emulsion filled gel based on inulin and
extra virgin olive oil, LWT - Food Science and Technology (2015), http://dx.doi.org/10.1016/j.lwt.2015.03.063
6 M. Giarnetti et al. / LWT - Food Science and Technology xxx (2015) 1e7

Table 6 3.6. Acceptance testing

Peak areas (mean value ± SD; total ion counts  106) of the volatile compounds
extracted by SPME from cookies samples and Fisher's test (LSD) for the multiple
comparisons (n ¼ 6).
A significant difference in liking was detected between short-
bread cookies when hedonic scores were averaged over all 120
Compound LRIa EFG-0 EFG-50 EFG-100 consumers. The box-plot (Fig. 3) shows a wider dispersion of scores
2-Butanone 1008 4.49 ± 3.63a 2.14 ± 1.47b 0.00 ± 0.00b for EFG-50 cookies (from 0.15 to 10) than for the others. Fully
3-Methyl-butanal 1035 0.00 ± 0.00c 2.45 ± 0.52b 6.86 ± 1.83a substituted cookies did not meet consumers' acceptance: 25% of
2,3-Butanedione 1267 1.02 ± 0.18c 1.44 ± 0.10b 2.37 ± 0.19a
preference scores for EFG-100 were higher than 4.2, whereas 25% of
Hexanal 1385 3.61 ± 1.66b 4.78 ± 0.50a 5.05 ± 0.50a
2-Heptanone 1268 12.50 ± 3.46a 6.22 ± 0.77b 0.00 ± 0.00c preference scores for EFG-50 (median value ¼ 6.5) were higher
Limonene 1297 2.09 ± 1.33a 2.42 ± 0.42a 2.92 ± 0.68a than 7.9. EFG-50 cookies showed therefore an appreciable accep-
Methyl-pyrazine 1773 1.05 ± 0.66c 1.82 ± 0.16b 3.46 ± 0.15a tance level, though lower than that of reference cookies (median
Octanal 3598 1.13 ± 0.54a 0.72 ± 0.11b 1.09 ± 0.28ab
value ¼ 8.1). This could be due to the fact that 50% butter
3-Hydroxy-2-butanone 6296 0.00 ± 0.00b 2.44 ± 0.48a 0.00 ± 0.00b
2-Propanone 1318 4.05 ± 1.19b 12.90 ± 0.83a 0.00 ± 0.00c
replacement allowed to keep micro-structural properties quite
2,5-Dimethyl-pyrazine 1336 1.61 ± 0.54a 0.00 ± 0.00b 0.00 ± 0.00b comparable to those of reference butter cookies. They suffered only
2-Nonanone 2410 4.98 ± 1.60a 2.20 ± 0.36b 0.00 ± 0.00c a slight variation in fracture and textural sensory properties,
Nonanal 4900 3.23 ± 3.07a 2.18 ± 0.49a 3.45 ± 1.01a probably due to higher moisture, and a decrease in intensity of the
Trimethyl-pyrazine 1408 0.00 ± 0.00c 0.52 ± 0.09b 0.83 ± 0.19a
odor and flavour characteristics typically conferred by butter.
Acetic acid 1558 1.39 ± 7.18a 1.93 ± 0.18a 1.86 ± 0.57a
Furfural 1712 1.16 ± 1.64a 0.49 ± 0.11a 0.72 ± 0.14a Forker et al. (2011) obtained fat-reduced shortdough biscuits, using
Decanal 1715 0.73 ± 0.79a 0.00 ± 0.00b 0.00 ± 00b corn fibre and lupin extracts, which gained greater preference from
Benzaldehyde 1569 1.13 ± 1.03a 0.92 ± 0.29a 1.11 ± 0.18a consumers than reference biscuits. The achieved fat reduction was
2-Undecanone 1602 0.99 ± 0.25a 0.58 ± 0.13b 0.00 ± 0.00c 28.6%. Laguna et al. (2014) observed good consumer acceptance, in
Benzoic acid 1661 0.00 ± 0.00c 0.21 ± 0.23b 0.39 ± 0.14a
g-Butyrolactone 1715 0.68 ± 0.44a 0.50 ± 0.35a 0.52 ± 0.04a
one case even higher than reference, for biscuits with 15% fat
5H-Furan-2-one 2120 0.76 ± 0.25a 0.92 ± 0.30a 0.98 ± 0.44a replacement by inulin or hydroxypropyl methylcellulose.
Hexanoic acid 1913 0.35 ± 0.27a 0.38 ± 0.20a 0.23 ± 0.04a Rodríguez-García et al. (2013), reported that consumers perceived
Dimethyl-sulfone 1904 0.82 ± 0.56b 2.31 ± 3.73a 2.06 ± 2.98a slight differences between control biscuits and 20% fat-replaced
EFG-0: shortbread cookies made with butter; EFG-50: shortbread cookies made ones, and indicated that this level of freplacement by inulin can
with 50% of replacement of butter with emulsion filled gel; EFG-100: shortbread give place to good quality shortdough biscuits. Taranco n, Salvador,
cookies made with 100% of replacement of butter with emulsion filled gel. Different
and Sanz (2012) replaced 100% shortening of biscuits by sunflower
letters within a row mean significant difference at p < 0.05.
a
LRI, Linear Retention Index obtained on HP-Innowax capillary column.
oilewaterecellulose ether emulsions (containing 52% sunflower
oil), obtaining lower consumer acceptability than reference bis-
cuits, though with differences not higher than 1.1/9 points (5.2
versus 6.3 on a 9-point hedonic scale). Our results show that good
significantly affect the aroma of bakery products such as bread, results could be achieved also by using EFG, at fat replacement
cakes, cookies etc. (Pozo-Bayo  n, Ruíz-Rodríguez, Pernin, & Cayot, levels of 50%.
2007) and derive, during the Maillard reactions, from amino-
reductones (Zamora & Hidalgo, 2005). The main difference 4. Conclusions
observed for Maillard compounds among the tested samples
regarded the levels of the Strecker aldehyde 3-methly-butanal, that The total replacement of butter by EFG allowed a reduction of
was not detectable in EFG-0 cookies, while significantly increased total fat content of 46%, and of satured fat content of 87% respect to
as butter was replaced by EFG. The ordinary pathway of formation control. A “reduced fat” label claim can be used for this type of
of Strecker aldehydes involves the reaction of deoxyosones with cookies, on the basis of the European Regulation 1924/2006
aminoacids (leucine as regards 3-methly-butanal). Nevertheless, (Official Journal of the European Union, 2006); the sensory prop-
Cremer, Vollenbroeker, and Eichner (2000) proved that the for- erties of these cookies were, however, unsatisfactory and would
mation of Strecker aldehydes in low moisture systems does not have to be improved. The EFG-50 cookies were well appreciated by
only occur via ordinary Strecker degradation but also via alternative
reaction pathways branching off directly from the Amadori rear-
rangement product fructose-leucine. In fat substituted cookies,
inulin would provide additional free fructose to promote the for-
mation of the fructose-leucine adduct, enhancing the consequent
formation of 3-methly-butanal. As a consequence of the increased
formation of Strecker aldehydes, also some pyrazines were found in
higher amounts in substituted cookies, being aminoreductones
products of Strecker degradation. As regards furan compounds, no
clear differences were observed. It should also be considered that
lipid oxidation and Maillard reaction are interrelated, and that the
former can promote or prevent the latter, and produce compounds
different from those that would form in absence of lipids (Zamora &
Hidalgo, 2005; 2011). It is presumable that butter is “free” in the
EFG-0 cookies and can easily interfere with Maillard reaction, un-
like the lipid fraction entrapped in the emulsion filled gel matrix,
that is less available to compete with carbohydrates in the Maillard
reaction.
Fig. 3. Box-plot and whisker diagram showing the median values of the preference
Finally, low levels of hexanoic acid, previously observed in scores (n ¼ 120) given to shortbread cookies in which the fat was substituted by an
cookies (Pasqualone et al. 2015) and deriving from further oxida- emulsion filled gel based on inulin and extra virgin olive oil. Levels of fat substitution:
tion of hexanal, were found. 0% (EFG-0); 50% (EFG-50), and 100% (EFG-100).

Please cite this article in press as: Giarnetti, M., et al., Fat replacement in shortbread cookies using an emulsion filled gel based on inulin and
extra virgin olive oil, LWT - Food Science and Technology (2015), http://dx.doi.org/10.1016/j.lwt.2015.03.063
 M. Giarnetti et al. / LWT - Food Science and Technology xxx (2015) 1e7 7

consumers; these samples were characterized by a fat content Kinsella, J. E., Patton, S., & Dimick, P. S. (1967). The flavor potential of milk fat. A
review of its chemical nature and biochemical origin. Journal of the American Oil
reduced by 19% and a saturated fat content reduced by 39%. This
Chemists' Society, 44, 449e454.
kind of shortbread cookies, according to European Regulation, can Laguna, L., Primo-Martín, C., Varela, P., Salvador, A., & Sanz, T. (2014). HPMC and
be labelled as “reduced saturated fat” as compared to control inulin as fat replacers in biscuits: sensory and instrumental evaluation. LWT e
cookies (EFG-0). Food Science and Technology, 56(2), 494e501.
Laguna, L., Varela, P., Salvador, A., Sanz, T., & Fiszman, S. M. (2012). Balancing texture
This study showed therefore that the replacement of 50% of the and other sensory features in reduced fat short-dough biscuits. Journal of
butter fraction of shortbread with EFG based on inulin and EVOO Texture Studies, 43, 235e245.
Lassoued, N., Babin, P., Della Valle, G., Devaux, M. F., & Re guerre, A. L. (2007).
allows to obtain a product with potentially healthier properties,
Granulometry of bread crumb grain: contributions of 2D and 3D image analysis
due to inulin and EVOO, similar to control cookies as regards at different scale. Food Research International, 40, 1087e1097.
microstructure and fracture, and well accepted by consumers. MacFie, H. J. H., Bratchell, N., Greenhoff, K., & Vallis, L. V. (1989). Designs to balance
Further investigation on timeetemperature parameters in baking, the effect of order of presentation and first-order carry-over effects in hall tests.
Journal of Sensory Studies, 4, 129e148.
in order to reach optimal moisture, and on recipe formulation, in Manohar, R. S., & Rao, P. H. (1999). Effect of emulsifiers, fat level and type on the
order to improve odor and flavour, would allow to correct these rheological characteristics of biscuits dough and quality of biscuits. Journal of
major drawbacks and reach a further improvement of consumers' the Science of Food and Agriculture, 79, 1223e1231.
Nawar, W. W. (1985). Chemistry of thermal oxidation of lipids. In D. B. Min, &
acceptance also at higher fat replacement levels. T. H. Smouse (Eds.), AOCS monograph: Vol. 15. Flavor chemistry of fats and oils (pp.
39e60). Champaign, IL: American Oil Chemist' Society.
References Niness, K. R. (1999). Inulin and oligofructose: what are they? Journal of Nutrition,
129, S1402eS1406.
Official Journal of the European Union. (2006). Regulation No 1924/2006 on nutrition
AACC International. (2000). Approved methods of analysis (11th ed.). St. Paul, MN,
and health claims made on foods. L.404 of December 20th.
USA: AACC International.
Paradiso, V. M., Giarnetti, M., Summo, C., Pasqualone, A., Minervini, F., & Caponio, F.
Baltsavias, A., Jurgens, A., & van Vliet, T. (1997). Factors affecting fracture properties
(2015). Production and characterization of emulsion filled gels based on inulin
of short-dough biscuits. Journal of Texture Studies, 28, 205e219.
and extra virgin olive oil. Food Hydrocolloids, 45, 30e40.
Baltsavias, A., Jurgens, A., & van Vliet, T. (1999). Fracture properties of short-dough
Parks, O. W., Keeney, M., Katz, I., & Schwartz, D. P. (1964). Isolation and character-
biscuits: effect of composition. Journal of Cereal Science, 29, 235e244.
ization of the methyl ketone precursor in butter fat. Journal of Lipid Research, 5,
Belitz, H. D., Grosch, W., & Schieberle, P. (2009). Food chemistry. Berlin: Heidelberg:
232e235.
Springer-Verlag.
Pasqualone, A., Bianco, A. M., Paradiso, V. M., Summo, C., Gambacorta, G.,
Caponio, F., Summo, C., Delcuratolo, D., & Pasqualone, A. (2006). Quality of the lipid
Caponio, F., et al. (2015). Production and characterization of functional biscuits
fraction of biscuits. Journal of the Science of Food & Agriculture, 86, 356e361.
on, a, & Della Valle, G. (2000). Physicochemical be- obtained from purple wheat. Food Chemistry, 180, 64e70.
Chevallier, S., Colonna, P., Bule
Pasqualone, A., Paradiso, V. M., Summo, C., Caponio, F., & Gomes, T. (2014). Influence
haviors of sugars, lipids, and gluten in short dough and biscuit. Journal of
of drying conditions on volatile compounds of pasta. Food and Bioprocess
Agricultural and Food Chemistry, 48, 1322e1326.
Technology, 7, 719e731.
Cremer, D. R., Vollenbroeker, M., & Eichner, K. (2000). Investigation of the formation
Pozo-Bayo n, M. A., Ruíz-Rodríguez, A., Pernin, K., & Cayot, N. (2007). Influence of
of Strecker aldehydes from the reaction of Amadori rearrangement products
eggs on the aroma composition of a sponge cake and on the aroma release in
with alpha-amino acids in low moisture model systems. European Food Research
model studies on flavored sponge cakes. Journal of Agricultural and Food
and Technology, 211, 400e403.
Chemistry, 55, 1418e1426.
Daniels, D. G. H., & Fisher, N. (1976). Release of carbon dioxide from dough during
Rodríguez-García, J., Laguna, L., Puig, A., Salvador, A., & Hernando, I. (2013). Effect of
baking. Journal of the Science of Food and Agriculture, 27(4), 351e357.
fat replacement by inulin on textural and structural properties of short dough
Devaux, M., Robert, P., Melcion, J., & Le Deschault de Monredon, F. (1997). Particle
biscuits. Food and Bioprocess Technology, 6, 2739e2750.
size analysis of bulk powders using mathematical morphology. Powder Tech-
Sala, G., de Wijk, R. A., van de Velde, F., & van Aken, G. A. (2008). Matrix properties
nology, 90, 141e147.
affect the sensory perception of emulsion-filled gels. Food Hydrocolloids, 22,
Devereux, H. M., Jones, G. P., Mccormack, L., & Hunter, W. C. (2003). Consumer
353e363.
acceptability of low fat foods containing inulin and oligofructose. Journal of Food
Schieberle, P., & Grosch, W. (1994). Potent odorants of rye bread crust e differences
Science, 68, 1850e1854.
from the crumb and from wheat bread crust. Zeitschrift für Lebensmittel
Di Mattia, C. D., Paradiso, V. M., Andrich, L., Giarnetti, M., Caponio, F., & Pittia, P.
Untersuchung und Forschung A, 198, 292e296.
(2014). Effect of olive oil phenolic compounds and maltodextrins on the
Sudha, M. L., Srivastava, a. K., Vetrimani, R., & Leelavathi, K. (2007). Fat replacement
physical properties and oxidative stability of olive oil o/w emulsions. Food
in soft dough biscuits: Its implications on dough rheology and biscuit quality.
Biophysics, 9, 396e405.
Journal of Food Engineering, 80(3), 922e930.
Drewnowski, A. (1997). Why do we like fat? Journal of the American Dietetic Asso-
Tarancon, P., Salvador, A., & Sanz, T. (2012). Sunflower oilewaterecellulose ether
ciation, 97, S316eS322.
emulsions as trans-fatty acid-free fat replacers in biscuits: texture and
European Food Safety Agency. (2010). Scientific opinion on dietary reference values
acceptability study. Food and Bioprocess Technology, 6, 2389e2398.
for fats, including saturated fatty acids, polyunsaturated fatty acids, mono-
Yang, N., Fisk, I. D., Linforth, R., Brown, K., Walsh, S., Mooney, S., et al. (2012). Impact
unsaturated fatty acids, trans fatty acids, and cholesterol. European Food Safety
of flavour solvent on biscuit micro-structure as measured by X-ray micro-
Agency Journal, 8, 1461e1568.
Computed Tomography and the distribution of vanillin and HMF (HPLC). Eu-
Forker, A., Zahn, S., & Rohm, H. (2011). A combination of fat replacers enables the
ropean Food Research and Technology, 235, 1083e1091.
production of fat-reduced shortdough biscuits with high-sensory quality. Food
Zahn, S., Pepke, F., & Rohm, H. (2010). Effect of inulin as a fat replacer on texture and
and Bioprocess Technology, 5(6), 2497e2505.
sensory properties of muffins. International Journal of Food Science and Tech-
Frankel, E. N. (1991). Recent advances in lipid oxidation. Journal of the Science of
nology, 45, 2531e2537.
Food and Agriculture, 54, 495e511.
Zamora, R., & Hidalgo, F. J. (2005). Coordinate contribution of lipid oxidation and
Ghotra, B. S., Dyal, S. D., & Narine, S. S. (2002). Lipid shortening: a review. Food
Maillard reaction to the nonenzymatic food browning. Critical Reviews in Food
Research International, 35, 1015e1048.
and Nutrition, 45, 49e59.
Haque, Z. U., & Ji, T. (2003). Cheddar whey processing and source: II. Effect on non-
Zamora, R., & Hidalgo, F. J. (2011). The Maillard reaction and lipid oxidation. Lipid
fat ice cream and yogurt. International Journal of Food Science and Technology, 38,
Technology, 23(3), 59e62.
463e473.
Zoulias, E. I., Oreopoulou, V., & Tzia, C. (2002). Textural properties of low-fat cookies
International Standard Organization (ISO) 13299. (2003). Sensory analysis e Meth-
containing carbohydrate- or protein-based fat replacers. Journal of Food Engi-
odology e General guidance for establishing a sensory profile.
neering, 55, 337e342.
Kato, Y. (2003). Chemical and sensory changes in flavor of roux prepared from
wheat flour and butter by heating to various temperatures. Food Science and
Technology Research, 9, 264e270.

Please cite this article in press as: Giarnetti, M., et al., Fat replacement in shortbread cookies using an emulsion filled gel based on inulin and
extra virgin olive oil, LWT - Food Science and Technology (2015), http://dx.doi.org/10.1016/j.lwt.2015.03.063
View publication stats