Available online at www.sciencedirect.

com

Infrared spectroscopy: an underexploited analytical tool

for assessing physicochemical properties of food
products and processing☆
Ana Borba 1 and Andrea Gómez-Zavaglia 2 ]]
]]]]]]
]]

Infrared spectroscopy is a useful tool to assess the physical and Introduction

chemical properties of different compounds. However, its use in Infrared spectroscopy is a powerful tool, widely em-
Food Science and Technology is mainly descriptive and/or ployed in fundamental sciences. However, its use in
combined with chemometric analysis. The aim of this review applied research (e.g. Food Science and Technology) is
was to underline available approaches for the analysis generally restricted to descriptive analyzes or to the as-
of physical and chemical properties associated with food sessment of adulterations and/or quantification of food
products and processing. components, when combined with chemometrics. This
review sheds light on the physicochemical aspects be-
In this context, this review covers recent applications of infrared hind the spectra, for stimulating a better exploitation of
spectroscopy in Food Science and Technology, focusing on this technique.
physical–chemical properties of food products and ingredients,
including macromolecules, and alterations arising from Fundamentals of infrared spectroscopy and
processing and storage. spectral analysis tools
Vibrational spectroscopy refers to the interaction of the
Overall, the work underlines the powerfulness of infrared infrared radiation with matter by absorption, emission, or
spectroscopy, a currently underexploited technique in the field, reflection. The infrared region of the electromagnetic
whose use should be stimulated for rapid and accurate spectrum (13333–10 cm−1) can be divided into near-
determinations both in academic and industrial contexts. (Near infrared (NIR): 13333–4000 cm−1), mid- (Mid in-
frared (MIR): 4000–400 cm−1), and far- (Far-infrared
Addresses (FIR): 400–10 cm−1) infrared, the first two being of
1
CIEPQPF - Department of Chemical Engineering, University of greater interest in Food Science (Table S1).
Coimbra, P-3030-790 Coimbra, Portugal
2
Center for Research and Development in Food Cryotechnology (CCT-
CONICET La Plata, UNLP), RA-1900, Argentina Infrared spectra arise from molecular vibrations occur-
ring when electrons are promoted from the fundamental
Corresponding author: Andrea Gómez-Zavaglia (angoza@qui.uc.pt) to an excited state and can be obtained in different
modes (transmission and reflectance). In transmission
Current Opinion in Food Science 2023, 49:100953 mode, the infrared light goes through a sample and is
collected on the other side of it. The transmission
This review comes from a themed issue on Food Physics &
Materials Science
techniques are excellent choices for the study of solids,
liquids, and gases, being the reference method for
Edited by Andrea Gomez-Zavaglia
quantitative analysis. However, when samples are not
For complete overview of the section, please refer to the article thick enough, any light frequency that is not absorbed
collection, “Food Physics & Materials Science 2023”
by the sample will be transmitted through it to the de-
Available online 18th October 2022 tector. Alternatively, reflection techniques can be em-
https://doi.org/10.1016/j.cofs.2022.100953 ployed. Reflection is the process by which the
2214–7993/© 2022 Elsevier Ltd. All rights reserved.
electromagnetic radiation collected by an interface
medium/sample, returns either at the boundary between
two media (surface reflection) or at the interior of a
medium (volume reflection). It is used to analyze li-

☆
Given the role as Guest Editor, Andrea Gomez-Zavaglia had no involvement in the peer-review of this article and has no access to information
regarding its peer-review. Full responsibility for the editorial process for this article was delegated to Alejandro Gregorio Marangoni, Editor in
Chief.

www.sciencedirect.com Current Opinion in Food Science 49 (2023) 100953

2 Food Physics & Materials Science

Abbreviations Transform spectroscopy

UV Ultraviolet
FTIR Fourier transform infrared spectroscopy ν stretch/ stretching bond
ATR-FTIR Attenuated total reflectance Fourier δ deformation / bending deformations
transform infrared spectroscopy Tg glass transition temperature
MIR Mid Infrared T temperature
NIR Near Infrared WTC wavenumber-temperature-coefficient
FIR Far Infrared DFT Density Functional Theory
FT Fourier transform aw water activity
DRIFT Diffuse Reflectance Infrared Fourier DM degree of methylation

quids, solids, gels, or coatings, it is convenient for qua- food processes [3]. As a polar molecule, water generates
litative analysis, and frequently used for quantitative intense bands, providing information about processes
ones. The FTIR reflection techniques include the involving water gains/losses (e.g. dehydration/rehydra-
Internal Reflection Spectroscopy [Attenuated Total tion, rubbery/amorphous states, and gelatinization).
Reflection (ATR)], the Reflection Spectroscopy
[Specular Reflection], and the Combination of Internal Infrared spectroscopy enables the study of samples in
and External Reflection [Diffuse Reflection different physical states (e.g. condensed phases, liquid,
(DRIFTs)] [1,2]. solid, and gaseous) with almost no preparation. Spectral
acquisition takes few minutes, and provides information
In infrared spectroscopy, functional groups show char- about all the constituents of complex systems (food ma-
acteristic vibrations, mechanically independent from the trices) in an environmentally friendly way [5]. However,
rest of the molecule. Hence, characteristic frequencies instrumental noise and bands' overlapping are the main
remain relatively unchanged, regardless the molecule challenges for spectral interpretation. Instrumental noise
the group is in, which facilitates the structural elucida- can be overcome by mathematically removing the instru-
tion of the studied compounds (Figure 1) [1,2]. The ment response function (digital smoothing process) and
origin of bands are changes in the dipolar moments, suppressing the noise. Peak-fitting and Fourier self-de-
therefore, the most polar functional groups (e.g. hydro- convolution are classical methods enabling the resolution
xyls, carboxylates, and carbonyls, among the most im- of overlapping bands. The former allows the estimation of
portant ones) are those giving rise to the most intense band parameters (e.g. position, width, and area) and the
features in the spectra (Table S1) [3,4]. Infrared spec- latter enhances resolution by narrowing the component
troscopy is also sensitive to certain polar intermolecular bands (Figure 2). Both methods have been widely em-
interactions (e.g. hydrogen bonds), which leads to no- ployed to investigate biological systems, mainly for de-
ticeable band shifts, very useful to understand different termining the secondary structure of proteins [6–11].
However, because of the well-established shortcomings of
these tools (e.g. excessive smoothing leading to artifacts,
Figure 1
overfitting), researchers and technicians should be thor-
oughly trained to obtain reliable results. This includes not
only the acquisition of spectra with as high as possible S/N
ratio to reduce the need for smoothing to the maximum
but also the use of appropriate mathematical tools to cri-
tically interpret the obtained results [1,2].

Applications of infrared spectroscopy in Food

Science
Water
Water molecules are widely present in food processing
(hydration/dehydration) and explain chemical alterations
and physical transformations of several products. They
Current Opinion in Food Science strongly absorb at 3300 and 1638 cm−1 (MIR), and at
6935 cm−1 and 5176 cm−1 (NIR) [12••].
Regions of the fundamental vibrational spectrum with the main
characteristic group frequencies. Water strongly impairs greater complexity to the spectra
(adapted from [1]). because some water bands (e.g. bending modes) strongly

Current Opinion in Food Science 49 (2023) 100953 www.sciencedirect.com

 IR spectroscopy for assessing food properties Borba and Gómez-Zavaglia 3

Figure 2

Current Opinion in Food Science

Unresolved (left) and peak-fitted (right) infrared spectra of the protein amide I region. The main secondary structure assignments (α, β structures) are
indicated in the fitted spectra.
(adapted from [8]).

overlap with the bands arising from compounds largely Carbohydrates

present in food products (e.g. amide I from proteins, Polysaccharides are polymeric carbohydrate macro-
νC]O from lipids). For this reason, the investigation of molecules in which monosaccharide units are covalently
water solutions and/or hydrated compounds deserves linked by glycosidic bonds, in linear or branched con-
careful analysis in FTIR analysis. figurations. The diversity of monomeric units and
linkage types highly increases their structural diversity.
Using FTIR to investigate solid or dehydrated products Polysaccharides' spectra can be divided into the fol-
does not have, in principle, such challenges. The water lowing regions [14••]:
present in food products either along processing or
during storage usually establishes hydrogen bonds with 4000–2500 cm−1 region: Characterized by a strong and
other macrocomponents (e.g. proteins, carbohydrates, broad band at 3600–3000 cm−1, corresponding to ν(OH)
and lipids), leading to noticeable downshifts of certain mode and arising from several hydroxyl groups with
bands (e.g. amide I, carbonyl, carboxylate groups, etc.), different conformations and positions in the carbohy-
which facilitates the understanding of such processes drates' rings. Water ν(OH) vibrational modes also give
[see sections Carbohydrates, Proteins, andFats and oils]. rise to bands in this region, which sometimes makes
difficult the analysis of these bands. The ν(CH2) and
In addition, the loss or gain of water molecules is asso- ν(CH3) (symmetrical and asymmetrical) also absorb in
ciated with textural changes in solid products such as this region (3000–2500 cm−1), but their intensity is ra-
bread and cookies, among others [13]. Therefore, FTIR ther low.
is a suitable tool to evaluate storage changes associated
with water gains/losses. Furthermore, structural changes Double-bond stretching region (1800–1500 cm−1):
associated with the water removal during dehydration Polysaccharides, including carboxylate groups (e.g., pec-
processes involved in the elaboration of food products tins, alginates), have characteristic bands, arising from
(e.g. freeze-drying, spray-drying) can be also monitored the ν(C]O) vibrations [15]. Water also absorbs in this
by FTIR [9•]. region [δ(OH) at 1635 cm−1], leading to greater or lesser
downshifts of ν(C]O), depending on the strength of the
As a whole, it can be concluded that although water is an hydrogen bonds involved [16].
undesirable compound in FTIR analysis, its presence in
certain products can be associated with food processing Fingerprint region (1500–600 cm−1): Bands arise from
and storage changes (e.g. dehydration/hydration), leading bending vibrations of groups with local symmetry [e.g.,
to chemical changes and physical transformations of sev- δ(CH2), δ(C–OH); 1500–1200 cm−1] [17]. In the
eral products. In such cases, FTIR is a very useful tool. 1200–800 cm−1 region, bands are specific for each

www.sciencedirect.com Current Opinion in Food Science 49 (2023) 100953

4 Food Physics & Materials Science

Figure 3

Current Opinion in Food Science

Solid-state infrared spectroscopy. A: amorphous and B: crystalline. An illustrative example.

carbohydrate and correspond to the C–O–C glycosidic (3600–3000 cm−1) at T = Tg [3], which can be quantified
linkage [δ(COH), ν(C–C); 1000–920 cm−1] and to the by determining the wavenumber-temperature coeffi-
anomeric region (900–800 cm−1), these latter enabling cient (measure of the downshift as a function of tem-
the elucidation of α- and β-configurations [18]. Carbo- perature) [16]. Although other vibrational modes [e.g.,
hydrate skeletal vibrations absorb below 800 cm−1, but δ(OH) (∼1030 cm−1), first overtone ν(OH)
−1
they provide limited information [17]. (∼6800 cm )], also provide information about Tg,
ν(OH) is still preferred because of the absence of com-
The following subsections report the main technologi- bination modes in that region [3].
cally relevant information that can be obtained from
carbohydrates' spectra. Computational facilities also provide powerful tools to
assess spectral features arising from physical changes.
Fan et al. [23••] shed light on the structure of freeze-
Physical changes associated with carbohydrates dried lactose stored within 25 and 95 ℃ at aw= 0.33, by
using density functional theory (DFT) calculations to
Glassy states assess the spectra of crystalline α-, β-, and α-lactose
Amorphous materials exist in the ‘rubbery’ or ‘glassy’ monohydrates (Figure 4). Like other carbohydrates,
states, and the transition among them is known as glass- amorphous lactose contains indefinite number of mole-
transition temperature (Tg), which explains several cular assemblies with varying levels of packing and or-
technological processes in the food industry, most of ders, which are determined by the organization of
them involving carbohydrates. Rubbery states (proces- hydrogen bonds. Plasticization is promoted at T ≥ Tg,
sing temperatures above Tg) are searched for the ela- and increases the spectral complexity. DFT calculations
boration of gummy candies, leathers, or in edible food- provide information about plasticization processes, che-
packaging [19,20]. On the contrary, dehydrated products mical structures, bands' assignments, and physical
(e.g., instantaneous soups, juices, and starters) require properties of the polymorphs, thus unraveling spectral
glassy ingredients (working temperatures below complexity. DFT calculations have also been employed
Tg) [21,22•]. Water has a key role as plasticizer, pro- to fully assign the experimental bands of larger carbo-
moting a Tg decrease even at low concentrations [21]. hydrates, including inulin [24], demonstrating the com-
plementarity between theoretical and experimental
Infrared spectra provide useful information about information for fulfilling the quality standards required
amorphous and rubbery states, the former leading to to elaborate food products.
broad and undefined bands, and the latter, to narrower
and well-defined ones (Figure 3). As a method for de-
termining Tg, infrared spectroscopy evaluates the for- Starch retrogradation
mation/breakage of the hydrogen bonds associated with Starch is a naturally abundant polysaccharide composed
phase transitions [16]. Below Tg (amorphous state), hy- of amylose and amylopectin, widely used in the food
drogen bonds mainly occur among carbohy- industry. Amylose is a long-chain α-D-glucose polymer
drates' hydroxyls and above Tg (rubbery state), between with almost no branches, whereas amylopectin is bran-
carbohydrates' hydroxyls and water, the latter interac- ched with short-chain glucans [25]••. Their technolo-
tions being stronger than the former. This is translated gical behavior is strongly determined by their physical
into dramatic downshifts of the ν(OH) bands properties, which are associated with their capacity to

Current Opinion in Food Science 49 (2023) 100953 www.sciencedirect.com

 IR spectroscopy for assessing food properties Borba and Gómez-Zavaglia 5

Figure 4

Current Opinion in Food Science

Schematic diagram for water plasticization occurred within the amorphous structure of lactose at a molecular level. The theoretical structures were
modeled and optimized at the DFT(B3LYP)/6-31G* level of theory.
(adapted from [21]).

interact with the surrounding water molecules. Starch increases along storage, there is a linear relationship
retrogradation (gelatinization) results from the unfolding between the 1047/1022 ratio and the storage time [25••].
of starch chains upon heating [9•] and is generally an
undesirable effect, leading to the formation of crystalline Likewise, the shifts of the ν(OH) (3500–3000 cm−1) are
double-helix structures, phases' separation, and increase related to the formation/breakage of hydrogen bonds,
of rigidity, thus affecting the food quality and shelf enabling the calculation of the energy required for ret-
life [26,27]. rogradation [29]. Considering that free hydroxyl groups
absorb above 3500 cm−1, hydroxyls' participating of in-
The physicochemical reactions behind retrogradation terstrand hydrogen bonds at 3511 cm−1, and inter-
comprise the disruption of the grain molecular organi- double-helix hydrogen bonds at 3290 cm−1, a ν(OH)
zation (swelling) as a result of the changes in the downshift indicates the formation of hydrogen bonds
starch–water interactions (rupture of most intra-/inter- (the greater the shift, the stronger the hydrogen bond).
molecular hydrogen bonds and hydrophobic interac- The length of hydrogen bonds in retrograded starch can
tions). The band at 1022 cm−1 is related to the exchange be empirically calculated as
of hydrogen bonds between starch and water, indicating 1 0
the formation of sol states before gelatinization. Based E= ×
on this, the 1000/1022 cm−1 ratio was associated with the k 0 (1)
water content, being useful for monitoring structural where ν0: standard wavenumber of free eOH groups
losses resulting from gelatinization [28]. The ν(C–OH) (3650 cm−1), ν: wavenumber of eOH in starch, and k:
mode for crystalline and amorphous states (1047 and constant (1/k = 2.625 ×102 kJ).
1022 cm−1, respectively) is also used to determine starch
retrogradation [9•]. As the degree of crystallization ν0eν=4.43×103(2.84–D) (2)

www.sciencedirect.com Current Opinion in Food Science 49 (2023) 100953

6 Food Physics & Materials Science

where ν0: wavenumber of free eOH (3650 cm−1), ν: This simple procedure enables a quick determination of
wavenumber of eOH in starch, and D: distance of the DM, which has relevant consequences in the formula-
hydrogen bond [25••]. tion of food products.

Romano et al. (2020) [9•] studied quinoa samples (also

Cellulose
containing proteins and lipids) dehydrated by spray-
Cellulose is the main component of the plant cell walls,
drying, a process promoting starch retrogradation be-
being largely used in food and pharmaceutical in-
cause of the exposure to high temperatures. In this case,
dustries [31]. It is composed of parallel unbranched D-
ν(OH) of quinoa samples (mainly arising from starch)
glucopyranose units linked by β-1,4-glycosidic bonds
downshifted after spray-drying, as a result of the
that form crystalline and highly organized microfibrils
starch–water interaction. The analysis of the 1047- and
through extensive inter-/intramolecular hydrogen bonds
1022 cm−1 bands allowed determining crystalline and
and van der Waals forces [32]. Amorphous cellulose
amorphous states also in such complex matrix.
corresponds to regions where these bonds are broken
and the ordered arrangement is lost. The complexity of
Gelatinization has a thickening effect, enabling water
cellulose increases if considered the four crystalline al-
retention in ready-to-eat meals (e.g. puddings, instant
lomorphs, namely I (Iα, Iβ), II, III (IIII, IIIII), and IV
lactic mixtures, and breakfast foods) without heating.
(IVI, IVII), I and II being the most important ones. In-
Oxidized starches have high transmittance, low viscosity,
frared spectroscopy provides useful information for
and low-temperature stability, which makes them sui-
studying amorphous/crystalline celluloses, based on si-
table for coating candles or to formulate easily melting
milar principles as those reported in subsections Glassy
candies. Acid-thinned starch (used for elaborating gums
states and Starch retrogradation [31].
and jellies) reduces hot-paste viscosity and improves
gelling, thus enhancing textural properties. Acetylation
ameliorates starch paste clarity and freeze–thaw stability. Proteins
All such chemical modifications lead to typical spectra, Infrared spectroscopy is particularly useful for de-
characterized by the absorbance of the incorporated termining proteins' structure, amide I, II, and III being
functional groups (e.g. C^H, C]H, acetyls, and alco- the most relevant bands. Amide I is almost entirely
hols) [30•]. composed of ν(C]O) (80%) arising from the peptidic
bond, being very sensitive to the secondary structure.
According to empirical frequency-structure correlations,
Degree of methoxylation in pectins β-sheets strongly absorb at 1640–1610 cm−1, giving rise
Although pectins have been discovered more than 210 to weaker bands at 1690–1680 cm−1. α-helices and
years ago, their physical–chemical and structural random coils absorb at 1650–1640 cm−1 and
properties are still a subject of investigation due to the 1660–1650 cm−1, respectively (Fig. 2). Amide II
great diversity of this polymer family and the close (1575–1480 cm−1) is associated with δ(NH) (40–60%)
relationship between structure and function. Pectins and ν(CN) (18–40%), being highly sensitive to the pro-
are linear chains of partially methyl-esterified (1→4)- tonation state of peptides but almost not sensitive to
linked α-D-galacturonic acid residues, also containing secondary structure alterations [8,10,11]. Amide III
neutral sugars (glucose, galactose). The degree of (1301–1229 cm−1) arises from ν(CN) (40%) and δ(NH)
methoxylation (DM) is an important parameter de- (60%). Although FTIR represents a quick, cost-effec-
termining their functional properties and is defined as tive, and sustainable technique for determining the
the percentage of galacturonic acid esterified with secondary structures of proteins, this is a challenging
methanol. Based on that, pectins can be classified as task. Spectral deconvolution is generally employed, but
pectins of high or low DM, the first group having > there is not a clear consensus in the literature about the
50% esterification. Pectins of high DM can form gels number of bands that should be used to deconvolve
under acidic conditions, whereas those of low DM can amide bands. When considering mathematical analysis
better interact with divalent cations through the free fundamentals, determining the spectra second deriva-
carboxyl group [15]. The bands at 1740 cm−1 [ν(C]O) tives provides a given number of minima, that in theory,
from ester groups] and 1630 cm−1 [ν(C]O) from car- would give rise to that given number of maxima, that
boxylates] can be used to determine the percentual should be used to deconvolve the spectra. However, the
DM (%DM) as identification of such peaks is not an easy task because
second derivatives increase both the sensitivity of the
A1740 bands arising from the compound and that of noise.
%DM = × 100
A1740 + A1630 (3) Therefore, distinguishing the real minima from the noise
is complicated in certain cases. This calls the attention to
where A1740: area at 1740 cm−1, and A1630: area at the need of registering experimental spectra having high
1630 cm−1. S/N ratios.

Current Opinion in Food Science 49 (2023) 100953 www.sciencedirect.com

 IR spectroscopy for assessing food properties Borba and Gómez-Zavaglia 7

Finally, it must be reminded that besides their nutri- triacylglycerides. Hydroxiperoxides are the main oxida-
tional aspects, proteins are technologically relevant be- tion products, leading to an increase of the ν(OH) bands
cause of their water-holding capacity, emulsification, (3530–3470 cm−1) [36]. The ν(CH3) and ν(CH2) bands,
gelation, thickening, and textural properties, all of them typically narrow because of the homogeneity of the
being dependent on the conformation and aggregation chains' spatial orientation, are split into four bands in
status (e.g., intermolecular β-sheet networks connected oxidized lipids (3000–2800 cm−1), representing sym-
by hydrogen bonds). Infrared spectroscopy provides metric and asymmetric CH vibrations of ν(CH3) and
critical information about structure-processing relation- ν(CH2). Unsaturated fatty acids also give rise to parti-
ships to better understand the mechanisms in- cular bands/shoulders in this region. Nonconjugated
volved [33]. double bonds (cis) and conjugated trienes arising from
lipids' oxidation, originate weak bands very close to
Fats and oils those of trans unsaturations (typical from nonoxidized
Lipids are present in different food products as phos- lipids) [37]. Lipids' oxidation also leads to ester hydro-
pholipids or triacylglycerols. Phospholipids are important lysis, producing aldehydes, whose ν(C]O) absorbs at
structures in cell membranes, representing > 87% of the lower wavenumbers, and generally occurs as
total lipids in eggs, meat, or fish. Most of them are re- shoulders [37].
moved by degumming when refining vegetable and fish
oils, the remaining oils consisting mainly of triacylgly- Emulsions
cerols. Their main spectral features include: Emulsions are part of several food products (e.g., milk,
butter, sausages, and charcuterie) and are key systems in
ν(OH), ν(NH): Hydroxyl or amine groups from certain food innovation. Indeed, reformulating traditionally
lipids (e.g. choline, ethanolamine, and glycolipids) ab- ‘unhealthy products’ rich in saturated lipids involves
sorb at wavenumbers > 3300 cm−1; their replacement with emulsions. The formulation of
emulsion gels (lipid materials in which emulsions and
ν(C]O): Triglyceride ester linkage (∼1750–1700 cm−1); hydrogels coexist) includes a lipid phase (e.g., olive,
microalgal, and chia oils), an aqueous phase, an emulsi-
ν(CH2), ν(CH3) (2950–2800 cm−1) and δ(CH) fier (e.g., proteins, carbohydrates), and a gelation pro-
(1470–1250 cm−1): Hydrocarbon chains. cess [38]. Oil-bulking agents are based on the dispersion
of oil droplets into continuous aqueous matrix-forming
The main spectral differences associated with the chain gels, a liquid oil (e.g., olive oil) being generally enclosed
lengths of saturated fatty acids and esters are better into a hydrogel network structure [39].
observed in the crystalline state than in solution, as
a result of the free rotation about the C–C bonds com- The structural characteristics of such systems can be
pared with their random motion in solution. A progres- assessed by infrared spectroscopy, providing information
sion of uniformly spaced absorption bands (∼20 cm−1) is about the structure–technological function relation-
generally observed in the 1350–1180 cm−1 region, the ship [39]. For example, the proteins' secondary structure
number of bands increasing with the chain length. leads to more ordered protein backbones (mainly com-
Similar progressions are observed in the 1070–710 cm−1 posed of α-helices) when oil-in-water emulsions are
region, and arise from the wagging and twisting–rocking produced. These structural properties are associated
vibrations of the CH2 groups. with the firmer texture of proteins in the emulsions.
Structured lipids (e.g., emulsion gels) are appropriate fat
Polymorphic forms of crystallized fats can be also dis- replacers consisting in emulsions with a gel-like network
criminated [34]. The CH2 rocking mode (∼720 cm−1) with solid-like mechanical characteristics [40]. The
splits into a doublet, enabling the differentiation of β' 3000–2500 cm−1 region provides information about the
from α- and β-forms. This supports technological appli- lipid chain order/disorder, which is related to the lipid
cations of infrared spectroscopy, considering the effects interactions and strongly determines the textural prop-
of polymorphism on melting and texture behaviors [e.g., erties of the emulsion gels [41].
small and fine crystals (β'-forms) are desirable for pro-
ducing margarines and shortenings because of the Polysaccharides are also used to formulate gel structures,
smooth/creamy textures they impart]. by immobilizing oil droplets (oil-bulking agents).
Structured lipids are stabilized by hydrogen bonds either
Lipids' oxidation is one of the most important factors between carbohydrate molecules or between carbonyls
limiting food shelf life, compromising its nutritional and water, which can be monitored by infrared spec-
value (formation of free radicals, reactive aldehydes) and troscopy [40]. In lipid membranes, carbohydrates can
sensory properties (e.g., odor, flavor, color, and tex- replace water molecules in dehydration-involving pro-
ture) [35]. Because of the greater degree of unsaturation, cesses, and this was extensively investigated by ana-
phospholipids are prompter to oxidize than lyzing structural changes in the ν(C]O) modes (e.g., as

www.sciencedirect.com Current Opinion in Food Science 49 (2023) 100953

8 Food Physics & Materials Science

carbonyl–water hydrogen bonds are stronger, the bands analytical/biophysical background, able to use it both in
of lipids' hydrated populations are downshifted) [22•]. A the academia and in the food industry. Gaining such
similar approach can be used for the study of lipid gels background would also promote the use of infrared
containing carbohydrates. As mentioned in section spectroscopy in innovative food applications and in in-
Water, the presence of water can be a challenge for terdisciplinary contexts, whose limits will be just re-
structures' elucidation because of the overlapping of stricted to the imagination and innovative spirit of
bands in the ν(OH), ν(C]O), and amide I regions. As in researchers and technologists.
other food systems, the acquisition of S/N spectra,
thorough spectral analysis, and also replacement of water CRediT authorship contribution statement
with D2O can provide solutions to overcome such chal- A.B. drafting of the paper, discussion, table, and figures.
lenges. A.G.-Z. Conceptualization, discussion, funding,
and drafting of the paper. Both authors have approved
Other applications the final version of the paper.
Katsara et al. (2021) [42] evaluated the migration of
polyethylene cheese packaging for samples stored for Conflict of interest statement
different periods at 4 ℃. The authors declare that they have no known competing
financial interests or personal relationships that could
FIR coupled with multivariate analysis is a suitable have appeared to influence the work reported in this
noninvasive alternative to X-rays for investigating toxins, paper.
microorganisms, and other food components [43]. How-
ever, its high cost, laborious setup, and the strong in- Data Availability
terference of polar molecules (e.g., water) highly limits its
utilization.
No data were used for the research described in the ar-
ticle.
Quantum chemical calculations provided useful tools for
the elucidation of certain food components, such as
sweeteners (alitame) [44] or p-coumaric, present in fruits Acknowledgements
This work was supported by the Argentinean Agency for the Scientific and
and vegetables [45]. Technological Promotion (ANPCyT) [Projects PICT(2017)/1344 and
PICT(2020)/0482]. A.G.-Z. is member of the research career CONICET.
A.B. acknowledges the financial support of Fundação para a Ciência e
Conclusions Tecnologia (FCT), Portugal, under Strategic Project (UIDB/00102/2020).
Infrared spectroscopy is a powerful analytical technique
that provides rich structural and biophysical information
Supporting information
about food components and their processing-associated
Supplementary data associated with this article can be
changes. However, it is generally underexploited be-
found in the online version at doi:10.1016/j.cofs.2022.
cause applied researchers usually lack an analytical/bio-
100953.
physical background enabling an accurate spectral
interpretation and also because food systems are struc-
turally complex. These two factors result in brief and References and recommended reading
descriptive spectral analyzes that disregard relevant Papers of particular interest, published within the period of review, have
been highlighted as:
physical and chemical information that is present in the
spectra. Considering that applied sciences are going to- •• of special interest
•• of outstanding interest.
ward interdisciplinary approaches, the involvement of
applied mathematicians/informaticians would be a great
contribution for the understanding of complex ‘mathe- 1. Larkin P: Elsevier; 2018,https://doi.org/10.1016/C2015-0-00806-1
matical functions’ (as the spectra are), providing deeper 2. Stuart BH: John Wiley and Sons Ltd; 2004.
insights for the interpretation of information that is al-
3. Santos MI, Gerbino E, Tymczyszyn E, Gomez-Zavaglia A:
ready present in the spectra. Applications of infrared and Raman spectroscopies to
probiotic investigation. Foods 2015, 4:283-305, https://doi.org/
Taking into account the simplicity and environmental- 10.3390/foods4030283

friendliness of spectral acquisition, using infrared spec- 4. Santos MI, Gerbino E, Araujo-Andrade C, Tymczyszyn EE, Gómez-
Zavaglia A: Stability of freeze-dried Lactobacillus delbrueckii
troscopy should be stimulated already in undergraduate subsp. bulgaricus in the presence of galacto-oligosaccharides
courses by establishing structure–function (nutritional and lactulose as determined by near infrared spectroscopy.
and technological) relationships, that is, connecting the Food Res Int 2014, 59:53-60, https://doi.org/10.1016/j.foodres.
2014.01.054
origin of the spectral features with the physical–chemical
5. Williams P: Grains and seeds. In Near-Infrared Spectroscopy in
changes occurring along food processing. This would Food Science and Technology. Edited by Ozaki Y, McClure WF,
contribute to form human resources with stronger Christy AA. John Wiley & Sons Inc; 2007:165-217.

Current Opinion in Food Science 49 (2023) 100953 www.sciencedirect.com

 IR spectroscopy for assessing food properties Borba and Gómez-Zavaglia 9

6. Lei T, Sun D-W: Developments of nondestructive techniques for Kunth) powder. LWT 2021, 146:111472, https://doi.org/10.1016/j.
evaluating quality attributes of cheeses: a review. Trends Food lwt.2021.111472
Sci Technol 2019, 88:527-542, https://doi.org/10.1016/j.tifs.2019.
04.013 20. Rivero R, Archaina D, Sosa N, Schebor C: Development and
characterization of two gelatin candies with alternative
7. Sadat A, Jove IJ: Peak fitting applied to Fourier Transform sweeteners and fruit bioactive compounds. LWT 2021,
• Infrared and Raman spectroscopic analysis of proteins. Appl 141:110894, https://doi.org/10.1016/j.lwt.2021.110894
Sci 2020, 10:5918, https://doi.org/10.3390/app10175918.
This paper explains how to use peak fitting to identify, separate, and 21. Romano N, Schebor C, Mobili P, Gómez-Zavaglia A: Role of
quantify the hidden peaks of amide I bands of infrared and Raman mono- and oligosaccharides from FOS as stabilizing agents
spectra for different proteins. during freeze-drying and storage of Lactobacillus delbrueckii
subsp. bulgaricus. Food Res Int 2016, 90:251-258, https://doi.org/
8. Mobili P, Londero A, Maria TMR, Eusébio MES, De Antoni GL, 10.1016/j.foodres.2016.11.003
Fausto R, Gómez-Zavaglia A: Characterization of S-layer
proteins of Lactobacillus by FTIR spectroscopy and differential 22. Romano N, Marro M, Marsal M, Loza-Álvarez P, Gomez-Zavaglia A:
scanning calorimetry. Vibrat Spectrosc 2009, 50:68-77, https:// • Fructose derived oligosaccharides prevent lipid membrane
doi.org/10.1016/j.vibspec.2008.07.016 destabilization and DNA conformational alterations during
vacuum-drying of Lactobacillus delbrueckii subsp. bulgaricus.
9. Romano N, Ureta MM, Guerrero-Sánchez M, Gómez-Zavaglia A: Food Res Int 2021, 143:110235, https://doi.org/10.1016/j.foodres.
• Nutritional and technological properties of a quinoa 2021.110235.
(Chenopodium quinoa Willd) spray-dried powdered extract. This paper gives information about the molecular mechanisms involved
Food Res Int 2020, 129:108884, https://doi.org/10.1016/j.foodres. in the interaction of oligosaccharides with lipid membranes and internal
2019.108884. bacterial structures (DNA), as determined by vibrational spectroscopy
This paper explains how to get information about the organization of techniques coupled with multivariate methods.
starch in spray-dried complex samples, also containing lipids and pro-
teins. 23. Fan F, Xiang P, Zhao L: Vibrational spectra analysis of
•• amorphous lactose in structural transformation: water/
10. Lin H, Bean SR, Tilley M, Peiris KHS, Brabec D: Qualitative and temperature plasticization, crystal formation, and molecular
quantitative analysis of sorghum grain composition including mobility. Food Chem 2021, 341:128215, https://doi.org/10.1016/j.
protein and tannins using ATR-FTIR spectroscopy. Food Anal foodchem.2020.128215.
Meth 2021, 14:268-279, https://doi.org/10.1007/s12161-020- This paper shows the complementarity of experimental infrared spec-
01874-5 troscopy and quantum chemical calculations in the study of lactose
crystallization, carried out by freeze-drying the disaccharide and equi-
11. Fevzioglu M, Ozturk OK, Hamaker BR, Campanella OH: librating it at different aw and temperatures.
Quantitative approach to study secondary structure of proteins
by FT-IR spectroscopy, using a model wheat gluten system. Int 24. Balan C, Chis MI, Rachisan AL, Baia MA: Vibrational study of
J Biol Macromol 2020, 164:2753-2760, https://doi.org/10.1016/j. inulin by means of experimental and theoretical methods. J
ijbiomac.2020.07.299 Molec Struct 2018, 1164:84-88, https://doi.org/10.1016/j.molstruc.
2018.03.047
12. Ren YQ, Lin XH, Lei T, Sun DW: Recent developments in
•• vibrational spectral analyses for dynamically assessing and 25. Lu H, Ma R, Chang R, Tian Y: Evaluation of starch retrogradation
monitoring food dehydration processes. Crit Rev Food Sci Nutr •• by infrared spectroscopy. Food Hydrocol 2021, 120:106975,
2021, 62:4267-4293, https://doi.org/10.1080/10408398.2021. https://doi.org/10.1016/j.foodhyd.2021.106975.
1947773. This work supports the use of infrared spectroscopy (changes in the
This work reports the use of vibrational spectroscopic techniques 1047/1022 cm-1 ratio) for the study of starch retrogradation, a common
(mainly NIR) and spectral data processing methods (chemometrics) for phenomenon occurring during the storage of starch-based foods.
the study of food dehydration processes and their effect on food quality
attributes. 26. Zhang H, Sun B, Zhang S, Zhu Y, Tian Y: Supramolecular
structure of A- and Btype granules of wheat starch. Food
13. Al-Mahsaneh M, Aljarrah M, Rababah T, Alu'Datt M: Using MR- Hydrocol 2015, 31:68-73, https://doi.org/10.1016/j.foodhyd.2012.
FTIR and texture profile to track the effect of storage time and 10.006
temperature on pita bread staling. J Food Qual 2018,
2018:8252570, https://doi.org/10.1155/2018/8252570 27. Zhang H, Sun B, Zhang S, Zhu Y, Tian Y: Inhibition of wheat
starch retrogradation by tea derivatives. Carbohydr Polym 2015,
14. Hong T, Yin J-Y, Nie S-P, Xie M-Y: Applications of infrared 134:413-417, https://doi.org/10.1016/j.carbpol.2015.08.018
•• spectroscopy in polysaccharide structural analysis: progress,
challenge and perspective. Food Chem X 2021, 12:100168, 28. Capron I, Robert P, Colonna P, Brogly M, Planchot V: Starch in
https://doi.org/10.1016/j.fochx.2021.100168. rubbery and glassy states by FTIR spectroscopy. Carbohydr
This work provides a comprehensive insight about principles and in- Polym 2007, 68:249-259, https://doi.org/10.1016/j.foodres.2014.
terpretation of infrared spectra for the study of polysaccharides. It fo- 07.040
cuses on the functions and properties of polysaccharides and their 29. Dankar I, Haddarah A, Omar FEL, Pujola M, Sepulcre F:
close relationship with structural features. Characterization of food additive-potato starch complexes by
15. La Cava EL, Gerbino E, Sgroppo SC, Gómez-Zavaglia A: FTIR and X-ray diffraction. Food Chem 2018, 260:7-12, https://
Characterization of pectins extracted from different varieties of doi.org/10.1016/j.foodchem.2018.03.138
pink/red and white grapefruits [Citrus paradisi (Macf.)] by 30. Awolu OO, Odoro JW, Adeloye JB, Lawal OM: Physicochemical
thermal treatment and thermosonication. J Food Sci 2018, • evaluation and Fourier transform infrared spectroscopy
83:1613-1621, https://doi.org/10.1111/1750-3841.14183 characterization of quality protein maize starch subjected to
16. Wolkers WF, Oliver AE, Tablin F, Crowe JH: A Fourier-transform different modifications. J Food Sci 2020, 85:3052-3060, https://
infrared spectroscopy study of sugar glasses. Carbohydr Res (6) doi.org/10.1111/1750-3841.15391.
2004, 339:1077-1085, https://doi.org/10.1016/j.carres.2004.01.016 This paper presents the characterization of native and modified starches
by evaluating their infrared spectra (e.g. OH, C-H, C]H, and C]H
17. Wiercigroch E, Szafraniec E, Czamara K, Pacia MZ, Majzner K, associated modes).
Kochan K, Kaczor A, Baranska M, Malek K: Raman and infrared
spectroscopy of carbohydrates: a review. Spectrochim Acta A 31. Makarem M, Lee CM, Kafle K, Huang S, Chae I, Yang H, Kubicki
2017, 185:317-335, https://doi.org/10.1016/j.saa.2017.05.045 JD, Kim SH: Probing cellulose structures with vibrational
spectroscopy. Cellulose 2019, 26:35-79, https://doi.org/10.1007/
18. Wang S, Xu H, Luan H: Multiscale structures of starch granules. s10570-018-2199-z
In Starch Structure, Functionality and Application in Foods. Edited
by Wang S. Springer; 2020:41-55. 32. Quiroz-Castañeda RE, Folch-Mallol JL: Hydrolysis of biomass
mediated by cellulases for the production of sugars. In
19. Vázquez-Sánchez AY, Corfield R, Sosa N, Salvatori D, Schebor C: Sustainable degradation of lignocellulosic biomass. Techniques,
Physicochemical, functional, and sensory characterization of applications and commercialization. Edited by Chandel AK, Silva
apple leathers enriched with acáchul (Ardisia compressa SS. IntechOpen; 2013,https://doi.org/10.5772/53719

www.sciencedirect.com Current Opinion in Food Science 49 (2023) 100953

10 Food Physics & Materials Science

33. Wei W, Hu W, Zhang X-Y, Zhang F-P, Sun S-Q, Liu Y, Xu C-H: 39. Ruiz-Capillas C, Herrero AM: Development of meat products
Analysis of protein structure changes and quality regulation of with healthier lipid content: vibrational spectroscopy. Foods
surimi during gelation based on infrared spectroscopy and 2021, 10:341, https://doi.org/10.3390/foods10020341
microscopic imaging. Sci Rep 2018, 8:5566, https://doi.org/10.
1038/s41598-018-23645-3 40. Muñoz-González I, Ruiz-Capillas C, Salvador M, Herrero AM:
Emulsion gels as delivery systems for phenolic compounds:
34. Li-Chan ECY, Ismail AA, Sedman J, van de Voort FR: Vibrational nutritional, technological and structural properties. Food Chem
spectroscopy of food and food products. applications in life, 2021, 339:128049, https://doi.org/10.1016/j.foodchem.2020.
pharmaceutical and natural sciences food science. In 128049
Handbook of Vibrational Spectroscopy. Edited by Griffiths P,
Chalmers JM. John Wiley & sons, Inc. Ltd.; 200610.1002/ 41. Pintado T, Herrero AM, Jiménez-Colmenero F, Pasqualin
0470027320. Cavalheiro C, Ruiz-Capillas C: Chia and oat emulsion gels as new
animal fat replacers and healthy bioactive sources in fresh
35. Faroux JM, Ureta MM, Tymczyszyn EE, Gómez-Zavaglia A: An sausage formulation. Meat Sci 2018, 135:6-13, https://doi.org/10.
overview of peroxidation reactions using liposomes as model 1016/j.meatsci.2017.08.004
systems and analytical methods as monitoring tools. Colloids 42. Katsara K, Kenanakis G, Viskadourakis Z, Papadakis VM:
Surf B Biointerfaces 2020, 195:111254, https://doi.org/10.1016/j. Polyethylene migration from food packaging on cheese
colsurfb.2020.111254 detected by Raman and infrared (ATR/FT-IR) Spectroscopy.
36. Rohman A, Che Man Y, Yusof F: The use of FTIR spectroscopy Materials 2021, 14:3872, https://doi.org/10.3390/ma14143872
and chemometrics for rapid authentication of extra virgin olive 43. Afsah-Hejri L, Hajeb P, Ara P, Ehsani RJ: A comprehensive review
oil. J Am Oil Chem Soc 2014, 91:207-213, https://doi.org/10.1007/ on food applications of Terahertz spectroscopy and imaging.
s11746-013-2370-5 Compr Rev Food Sci Food Saf 2019, 18:1563-1621, https://doi.org/
10.1111/1541-4337.12490
37. Rohman A, Che Man Y: Application of FTIR Spectroscopy for
monitoring the stabilities of selected vegetable oils during 44. Altunayar-Unsalan C, Unsalan O: Structural and anharmonic
thermal oxidation. Int J Food Prop 2013, 16:1594-1603, https:// vibrational spectroscopic analysis of artificial sweetener
doi.org/10.1080/10942912.2011.603874 alitame: a DFT study for molecular basis of sweet taste. J Molec
Struct 2021, 1246:131157, https://doi.org/10.1016/j.molstruc.2021.
38. Herrero AM, Ruiz-Capillas C: Novel lipid materials based on 131157
gelling procedures as fat analogues in the development of
healthier meat products. Curr Opin Food Sci 2021, 39:1-6, 45. Semidalas EC, Semidalas CE: Structure and vibrational spectra
https://doi.org/10.1016/j.cofs.2020.12.010 of p-coumaric acid dimers by DFT methods. Vibrat Spectrosc
2019, 101:100-108, https://doi.org/10.1016/j.vibspec.2019.02.007

Current Opinion in Food Science 49 (2023) 100953 www.sciencedirect.com