Food Additives
Food Additives
Long Wu, Chenghui Zhang, Yingxi Long, Qi Chen, Weimin Zhang & Guozhen
Liu
To cite this article: Long Wu, Chenghui Zhang, Yingxi Long, Qi Chen, Weimin Zhang & Guozhen
Liu (2021): Food additives: From functions to analytical methods, Critical Reviews in Food Science
and Nutrition, DOI: 10.1080/10408398.2021.1929823
REVIEW
ABSTRACT KEYWORDS
Food additives refer to all kinds of trace substances used in food or food processing to preserve analytical methods; food
flavor or enhance food taste, appearance, or other qualities. At present, artificial synthetic food additives; food additives
additives have gradually replaced the natural food additives and many problems related to food regulation; molecular
structure and technique
additives, involving the abuse of food additives, excessive additives or even toxic additives.
Obviously, food additives can bring people great sensory enjoyment and commercial convenience,
but they may also cause potential risks to human health. So, it is of high significance to conduct
quantitative analysis on the content of food additives. According to their functions and the regula-
tory requirements of food additives, this review starts from the classification and structures of vari-
ous food additives involving colorants, preservatives, antioxidants, sweeteners, emulsifiers,
stabilizers, thickeners, gelling agents. It then summarizes and discusses analytical methods for
quantification of food additives including modern immunoassays and other biotechnological
methods. The proposed review aspires to fill in the knowledge gap of food additives between aca-
demia and industry by covering all kinds of analytical methods for quantifying food additives.
CONTACT Long Wu longquan.good@163.com College of Food Science and Engineering, Hainan University, 58 Renmin Ave, Meilan District, Haikou City,
Hainan Prov 570228, P.R. China; Guozhen Liu liuguozhen@cuhk.edu.cn School of Life and Health Sciences, The Chinese University of Hong Kong,
Shenzhen 518172, P.R. China
ß 2021 Taylor & Francis Group, LLC
2 L. WU ET AL.
modern quantification techniques. This review will start where the E-number, main usage and maximum dosage can
be referred. Below section details different food additives
from the classification of food additives, then cover analysis
and their functions.
of food additives based on all aspects of analytical methods,
including both traditional and modern biotechnological
methods (Figure 1). Colorants
Colorants are a kind of additives that added to food to rem-
Food additives and their functions edy colors lost during preparation, enhance perception of
flavor or to make food look more attractive (Coultate and
Till now, there are more than 25,000 compounds of food
Blackburn 2018). Colorants present in both natural and syn-
additives being used all around the world (Anderson 1986).
thetic forms, where natural food colors refer to naturally
According to the compositions, food additives are generally
available dyes obtained from vegetables, animals or minerals.
divided into two major categories of natural additives and For instance, curcumin is mainly used for coloring food like
synthetic additives. Wherein, natural food additives are beverages, sauces and confectionery, which can be added
mainly produced by purifying the ingredients from plant or only lower than 0.01 g/kg. Additionally, canthaxanthin used
animal sources. While chemically synthesized additives are to dye tanning pills, candies, and carbonated drinks has the
based on chemical raw materials, from which organic or maximum dosage from 0.001 to 0.03 g/kg. On the other
inorganic matter can be extracted and purified. Additionally, hand, synthetic food colors are popularly also known as arti-
based on their functions, food additives can be divided into ficial food colors, which are manufactured chemically.
several groups such as antioxidants, bleach, sweeteners, pres- Wherein, quinoline yellow used for coloring cold fruits, ice
ervatives, colorants, thickeners and so on. Food additives creams, sweets, and beverage products should be controlled
from different group may have overlap due to mul- under 0.1 g/kg in food processing. Tartrazine mainly added
tiple effects. in soups, ice creams, sweets, gum and mustard yogurt has
To regulate food additives and inform consumers, addi- maximum dosage from 0.015 to 0.018 g/kg.
tives were classified and assigned unique numbers in differ-
ent countries. The numbering method in European Union
gives each additive a unique number called an "E number", Preservatives
which is used in Europe for all approved additives. This Preservatives are used to prevent or inhibit spoilage of food
numbering scheme has now been adopted and extended by due to contamination of fungi, bacteria and other microor-
the Codex Alimentarius Commission to internationally iden- ganisms and keep food safe for longer time (Carocho,
tify all additives, regardless of whether they are approved for Morales, and Ferreira 2018). Similar to food colors, preser-
use (Codex Alimentarius Commission 1989). Using this vatives can also be divided into two main groups of artificial
method, the additives can be easily classified into groups and natural ones. Commonly used artificial preservatives
that have different functions (Figure 2). Food additives men- include sorbic acid, which is used in dairy products, soybean
tioned in following contents are all contained in Table 1, products, processed vegetables, cooked meat products and
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 3
Table 1. Typical representatives of four food additives with E-number (not given: —).
Maximum dosage
Category Common substance (E-number) Mainly used range (g/kg)
Colorants Natural food colors Natural carotene (E160a) Dairy products, frozen drinks, processed fruits, dried —
vegetables and soybean products
Anthocyanins (E163) Jams, sugar confectioneries, jellies, soft drinks and —
frozen products
Curcumin (E100) Beverages, sauces and confectionery < 0.01
Canthaxanthin (E161g) Tanning pills, fruit-spreads, candies, sirups, sauces, 0.001 0.03
carbonated drinks
Artificial Tartrazine (E102) Soups, sauces, ice creams, ice lollies, sweets, 0.015 0.018
food colors chewing gum, marzipan, jam, jelly, marmalade and
mustard yogurt etc.
Quinoline yellow (E104) Cold fruits, ice creams, cake, chocolate, bread, < 0.1
cheese sauces and beverage products etc.
Sunset yellow FCF (E110) Jam, dairy products, cocoa products, starch 0.005 0.0887
desserts, compound condiments, beverage products
Carmoisine (E122) Beverage, wine, candy, green plum, bayberry, 0.005 0.015
sandwich, ice cream etc.
Ponceau 4R (E124) Beverage, wine, soda, candy, pastry, soya drink, ice 0.001 0.01
cream, yogurt
Allura red (E129) Candy coating, fried chicken, meat enema, western 0.004 0.0085
ham, jelly, biscuit sandwich
Preservatives Natural Nisin (E234) Meat, dairy products, vegetable protein products, 0.15 0.5
preservative canned goods, coffee beverage, tea, soy sauce
Natamycin (E235) Yogurt, cheese, raw ham, dried sausage, cakes < 0.1
Lysozyme (E1105) Cheese, drink, baby food, meat and fish products —
Artificial Sorbic acid (E200) and potassium Dairy products, soybean products, processed 0.075 2.0
preservative sorbate (E202) vegetables, cooked meat products, aquatic
products, etc.
Benzoic acid (E210) and sodium Condiments, pickled products, beverage products, 0.2 2.0
benzoate (E211) fruit wine
P-hydroxybenzoates (E214-219) Jam and sauce products, carbonated drinks, etc 0.012 0.5
Sodium nitrite (E250) Soybean products, meat products, aquatic products, 1.0 10
pastry and puffed foods, etc.
Propionic acid (E280) and sodium Soybean products, wet flour products, bread, 0.25 2.5
propionate (E281) pastry, vinegar and soy sauce, etc.
Antioxidants Ascorbic acid and its salts Peeled fresh fruits and vegetables, wheat flour, 0.2 5.0
(E300-302) fruits and vegetables products
Tocopherol (E306) and its geometric Meat, fish, nuts, vegetables, fruits, beverages and —
isomers (E307-309) canned food
Propyl gallate (E310) Nuts and canned seeds, gum-based candy, grilled 0.1 0.4
meat, fried noodles, etc.
Tert-butyl hydroquinone (E319) Moon cakes, instant rice noodles products, biscuits, 0.2
baked food fillings
Butyl hydroxyanisole (E320) Fat, oil and emulsified fat products, coarse grain, 0.2
instant rice noodles products, etc.
Butylated hydroxytoluene (E321) Fried noodles, gum-based candy, air-dried aquatic 0.2 0.4
products, etc.
Sweeteners Natural sweetener Sorbitol (E420) Dairy products, jams, wet flour products, baked 0.5 30
products, beverages and soybean products, etc.
Mannitol (E421) Candy, chewing gum 0.2
Thaumatin (E957) Candied fruits, candy, biscuits, canned meat, etc. —
Stevioside (E960) Flavor fermented milk, candy, condiments, canned 0.17 10.0
fruits, flavored sirup and tea products
Maltitol, maltitol sirup (E965) Processing fruits, frozen surimi products, soybean —
products, etc.
Lactitol (E966) Dairy products, Spices —
Xylitol (E967) Dairy products, tea products, alcoholic drinks, —
seasonings, starch products, processed fruits and
vegetables, etc.
Artificial sweetener Aspartame (E951) Dairy products, frozen fruits and vegetables, cereals 0.3 4.0
and starch desserts
Cyclamate (E952) Canned fruits, jams, mixed wine, instant noodle 0.65 8.0
food and condiments
Saccharin and its salts (E954) Frozen drinks, dehydrated mango, dried figs, cold 0.15 5.0
fruits, cooked beans and dried fruits
Sucralose (E955) Prepared dairy products, jams, sufu, coarse cereals 0.25 5.0
products, Baked products, etc.
Neotame (E961) Dairy products, frozen fruits and vegetables, cereals 0.01 0.35
and starch desserts
Emulsifiers, stabilizers, Polyglycerol esters (E475) Soy milk drinks, ice-creams, spreads, breads, cakes, 5 10
thickeners toffee, chocolate
Sorbitan esters (E491-495) Milk, ice-creams, breads, cakes, biscuits, soya 0.05 10
products, dried yeast, drinks
Sodium alginic acid (E400) Milk products, cheese, margarine, freezing drinks 10
(continued)
4 L. WU ET AL.
Table 1. Continued.
Maximum dosage
Category Common substance (E-number) Mainly used range (g/kg)
Agar (E406) Ice cream, low-fat spreads, dairy products, salad —
dressings, mayonnaise
Pectin (E440) Jam, jelly, cheese, candy, sauce, yoghurt 3.0
Gellan gum (E418) Pudding, jelly, sugar, drinks, dairy products, jam —
products, bread
Flavor enhancers Monosodium glutamate (E621) Soup, sausage, fish cake, chili sauce, canned food, 0.2 0.5
puffed food, jelly
Guanylic acid (E626) Meat, ham, bacon, soy sauce, food condiments 0.5
Glazing agents Carnauba wax (E903) Candy, dried food, baked goods, marshmallows, —
sugar products, fruit juice, broth sauce
Shellac (E904) Apples, oranges, gum sugar base, roasted coffee, 0.2
chocolate, wafer biscuits
aquatic products. The maximum dosage of sorbic acid is mainly used in fruits, jams, beverages, desserts and dairy
0.075 to 2.0 g/kg in food processing. Benzoic acid and its products. The maximum dosage of aspartame is from 0.3 to
salts, are commonly used in preserving condiments, pickled 4.0 g/kg; cyclamate has maximum dosage from 0.65 to 8.0;
products, beverage products and fruit wine, which have the and saccharin sodium is from 0.15 to 5.0 g/kg.
maximum additive amount from 0.2 to 2.0 g/kg. The natural
preservatives involving nisin, which is added to meat, dairy
products, vegetable protein products, canned goods, coffee Emulsifiers, stabilizers, thickeners and gelling agents
beverage, tea and soy sauce. The maximum additive amount Since many functions of food additives are overlapped, and
of nisin is 0.15 to 0.5 g/kg. Moreover, the maximum dosage thickeners and gelling agents have nearly the same functions
of natamycin should be controlled lower than 0.1 g/kg in (Abid et al. 2018). Therefore, emulsifiers, stabilizers, thick-
food processing. eners and gelling agents are categorized as one group from
E400 to E499 in the E-number classification (Figure 2).
Antioxidants Emulsifiers, also called emulgents, act as an interface
between the conflicting components in water and oil
There is some overlap between preservatives and antioxi- (McClements and Jafari 2018), and are used in creams and
dants because they perform the similar effect on the food sauces, bakery, and dairy products. Examples of emulsifiers
preservation. Antioxidants mainly prevent or inhibit the oxi- are polyglycerol esters, which has the maximum dosage
dation process to keep the food from spoilage (Lorenzo et
range from 5 to 10 g/kg. Stabilizers are substances or chemi-
al. 2018). Natural antioxidant additives are commonly added
cals that allow food ingredients, which do not mix well, to
to meat, fish, nuts, vegetables, fruits, beverages and canned
remain in a homogenous state after blending (Tekin Pulats€ u,
food. For example, the additive amount of ascorbic acid as
Sahin, and Sumnu 2018). The common food stabilizer
well as its salt-like ascorbate and its geometric isomers
includes agar, alginic acid and its sodium, potassium,
should be controlled from 0.2 to 5.0 g/kg. The artificial anti-
ammonium and calcium salts. Thickeners or thickening
oxidants are often added to oils, cheese, and chips to sup-
agents are substances added to food preparations for
press the formation of hydroperoxides (Carocho, Morales,
increasing their viscosity without changing other properties
and Ferreira 2018). For instance, the phenol derivatives of
like taste, for example, pectin has the maximum dosage of
propyl gallate should be controlled in dosage from 0.1 to
0.4 g/kg. Moreover, tert-butyl hydroquinone has the max- lower than 3.0 g/kg. Gelling agents like konjac, karaya gum,
imum dosage of 0.2 g/kg; and butyl hydroxyanisole should and gellan gum are incorporated into foods to produce spe-
be controlled lower than 0.2 g/kg too. cific structure, flow, stability and eating qualities desirable
for consumers.
Sweetener
Glazing agents, flavor enhancer
Sweetener, a sugar substitute, is a food additive that provides
a sweet taste like that of sugar while containing significantly Other additives like glazing agents which fall into this cat-
less food energy, making it a zero-calorie or low-calorie egory, have other functions for improving whipping, leaven-
sweetener (Martyn et al. 2018). Natural sweeteners like sorb- ing and color permanence (Teixeira 2018). One glazing
itol, xylitol, and mannitol are derived from sugars, which agent called shellac has the maximum dosage range of lower
are widely used in dairy products, tea products, alcoholic than 0.2 g/kg in food processing. Additionally, the food fla-
drinks, seasonings, candy, starch products, processed fruits vor enhancers like monosodium glutamate and guanylic acid
and vegetables. Wherein, the maximum dosage of sorbitol is function in enhancing the existing flavor (Abdel-Moemin,
from 0.5 to 3.0 g/kg; the additive amount of mannitol should Regenstein, and Abdel-Rahman 2018), where monosodium
be controlled under 0.2 g/kg. Artificial sweetener such as glutamate has the maximum dosage from 0.2 to 0.5 g/kg and
aspartame, cyclamate, saccharin sodium and sucralose are guanylic acid should be lower than 0.5 g/kg.
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 5
Figure 4. Chemical structures of the representative synthetic food colorants (from 1 to 6: tartrazine (E102), quinoline yellow (E104), sunset yellow FCF (E110), amar-
anth (E123), ponceau 4 R (E124) and indigo carmine (E132)).
chromatography, gas chromatography and mass/liquid chro- developed methods can also realize quantitative and qualita-
matography. These methods behave advantages of high sen- tive detection of food additives with simple operations, rapid
sitivity, good stability and reproducibility, but still suffer detection, high sensitivity and good specificity. The follow-
shortcomings such as professional operations, complex sam- ing section will discuss different analytical methods for five
ple pretreatment, relative long test period and high cost classes of food additives respectively, including colorants,
(Taghdisi et al. 2019). Besides, spectrophotometry like Uv- preservatives, sweeteners, antioxidants, emulsifiers as well as
vis absorbance and fluorescence spectroscopy possess simple their application and properties.
operation and good sensitivity, however, the stability is a Then, different analytical methods including biosensors
great concern, and the interference sources are complicated are comprehensively summarized and their detection per-
(Nguyen and Waterhouse 2019). High performance capillary formance for food additives, especially for the additives with
electrophoresis exhibits simple operation, high accuracy and low concentration, are compared and evaluated.
sensitivity but requires high quality of electrophoresis paint
(Papetti and Colombo 2019). Recently, ion chromatography
Analysis of food colorants
has been intensively used in analysis of food additives,
which can determine low level of components even at high Food colorants, also known as dyes or pigments that impart
substrate concentration, thus simplify or eliminate sample color when it is added to food or drink, are usually divided
pretreatment process and achieve simultaneous determin- into natural ones and artificial ones. Compared with natural
ation of multiple components (Meher, Labhsetwar, and pigments, artificial pigments have the advantages of high
Bansiwal 2018). stability, good color uniformity, low microbial pollution and
With the development of nanotechnology, many newly cost of production. Because of the advantages, artificial pig-
developed analytical methods such as surface enhanced ments such as tartrazine (E102), quinoline yellow (E104),
Raman spectroscopy (SERS), fluorescence assay, electro- sunset yellow FCF (E110), amaranth (E123), ponceau 4 R
chemical or electroluminescence assay, enzyme-linked (E124) and indigo carmine (E132) are widely applied in
immunosorbent assay (ELISA), lateral flow immunoassay food industry. However, continuous intake of synthetic colo-
(LFIA) and their related biosensors have been constructed rants may do certain toxicity to human (Amchova,
to analyze food additives, which have greatly improved the Kotolova, and Ruda-Kucerova 2015). Experiments demon-
performance of the analytical methods. These newly strated that a high dosage of the dye such as amaranth
8 L. WU ET AL.
might increase the incidence of malignant tumors in rats. genotoxicity, or developmental toxicity in the amounts at
Therefore, the synthetic pigments allowed in some countries which it is used (Joshi and Pancharatna 2019). However, it
are regulated with strictly limited dosage. was banned or restricted as a food additive in countries like
Moreover, all these synthetic pigments include functional Norway, Finland and Sweden due to its lack of evidence for
group like –OH, –NH2 or benzene ring (–C6H5), which are food safety.
optically or electrochemically active from the chemical struc- In recent years, several analytical methods have been
ture (Figure 4). It means that the substances are easily acti- reported for the determination of sunset yellow, especially
vated to an excited state under light or electricity, resulting the electrochemical sensor and assays based on SERS. For
in specific optical or electrical signals. From this perspective, example, Tran et al. developed an electrochemical sensor for
the detection of colorants is more favored with optical or sunset yellow measurement by modifying graphite flakes on
electrochemical methods. the glassy carbon electrode (Tran et al. 2019). By using dif-
ferential pulse voltammetry, sunset yellow can be detected
with the linear range of 50 1000 nM with the calculated
Tartrazine LOD of 19.2 nM. The recoveries of the method are
Tartrazine (E102), known as lemon yellow, is a kind of 93.35 107.7% in commercial soft drinks, which suggests its
water-soluble azo dye and often used in the coloring of food accuracy and stability in the analysis of sunset yellow.
and beverage. Studies have shown that intake of tartarzine Furthermore, Li et al. constructed a method by using bimet-
can cause a series of changes in biochemical markers and it allic nanoparticles functionalized graphene for the electrode
does great harm to asthma patients and children at higher modification (Li et al. 2019a). Following that, the differential
doses (Elekima et al. 2019). The acceptable daily intake pulse voltammetry was adopted for the determination of
(ADI) for tartrazine is allocated as 7.5 mg/kg/day by JECFA sunset yellow. Further study revealed that the oxidation
in 1964, but many countries have banned or restricted tar- mechanism of sunset yellow is pH dependent and the elec-
trazine (Walton et al. 1999). To regulate food safety, many trochemical process is adsorption-controlled, which sug-
electrochemical and optical methods have been developed to gested that the morphology of modified material is
analyze tartrazine in food. extremely important to achieve high sensitivity of detection.
For instance, He et al. prepared TiO2-reduced graphene The Co-Ni modified electrode showed the linear range of
oxide composite-modified electrodes for the sensitive detec- 0.008 10.0 lM and the LOD of 2.0 nM for sunset yellow.
tion of tartrazine (He et al. 2018a). The new method showed The nanomaterials based electrochemical methods increased
that the composites increased the electrochemical active area active area and enhanced electron transfer efficiency, behav-
and enhanced the sensitivity with a LOD of 8.0 nmol L1 ing accurate and sensitive determination of the used dyes in
(S/N ¼ 3). The method was successfully applied for the soft drinks without laborious pretreatment.
detection of trace tartrazine in carbonated beverage samples. Besides, Ai et al. synthesized flower-shaped silver nano-
Similarly, de Lima et al. reported the differential pulse vol- particles as SERS substrates for the simultaneous detection
tammetry method for tartrazine detection using gold nano- of sunset yellow, tartrazine, food blue and acid red (Ai et al.
particles and Nickel(II)-phthalocyanine-tetrasulfonic (de 2018). The SERS substrate behaved high SERS activity and
Lima et al. 2019). In this work, a large and well-defined the four different food colorants can be distinguished by
redox peak of tartrazine was obtained with the LOD of principle component analysis. The LODs of the four colo-
0.055 mmol L1. The results demonstrated the excellent rants were measured to be 107 mol L1. Based on gold
accuracy and reliability for the detection of tartrazine in nanorods, Ou et al. reported a SERS method for the detec-
juice samples. Besides, Tiwari et al. described the diffuse tion of sunset yellow and allura red in beverages (Ou et al.
reflectance Fourier transform infrared (DRS-FTIR) spectros- 2018). Based on the substrate, it showed strong peak at
copy for the detection of tartrazine (Tiwari and Deb 2019). 1592 cm1 and 1219 cm1 for sunset yellow and allura red.
Based on the prior extraction of tartrazine with modified sil- With little sample preparation, the method can realize the
ver nanoparticles, the method achieved linear calibrations in convenient, cost efficient and quick detection of sunset yel-
the range of 1 to 160 ng mL1 with a LOD of 2.44 ng mL1. low and allura red as low to 0.10 mg L1. Generally, the
The above developed methods proved to be good alternative above SERS methods have demonstrated great advantages in
techniques, providing low cost and reliable results for the terms of simplicity and sensitivity in the analysis of colo-
control of harmful food additives. As those electrochemical rants, however, the stability and the background interference
and optical methods are sensitive to environment, great may be the main restrictions for their wide applications.
attention should be paid on the influence of nanomaterials
and the substrates on the signal outputs.
Amaranth
Amaranth (E123) usually appears as a trisodium salt in the
Sunset yellow drinks, candy and pastries. Excessive intake of amaranth can
Sunset yellow (E110), a water-soluble synthetic azo dye, is have adverse effects on health, such as high genotoxicity and
often used in combination with amaranth to give a brown cytotoxicity. Since 1976, it has been banned in the United
coloring in chocolates. Under EU and WHO/FAO guide- States by the Food and Drug Administration as a suspected
lines, the daily acceptable intake is 0 4 mg/kg. It was carcinogen. However, its use is legal in some other countries
reported that sunset yellow behaved no carcinogenicity, like United Kingdom and China but the maximum amount
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 9
is strictly limited due to the potential risks to health poses great challenges to the food safety supervision.
(Oplatowska-Stachowiak and Elliott 2017). According to the Therefore, it is an urgent to develop effective methods for
hygienic standard of food additives by FAO/WHO in 1994, the sensitive detection of indigo carmine.
the daily acceptable intake of amaranth is less than 0.5 mg/ For example, Manjunatha et al. discussed the electro-
kg. Therefore, it is critical to measure the amount of amar- chemical behavior of indigo carmine at poly (glycine) modi-
anth in food to control food quality and ensure consum- fied carbon paste electrode by cyclic voltammetry
er’s health. (Manjunatha 2018). The method showed the oxidation
So far, many optical or electrochemical techniques are potential of indigo carmine at 407 mV with the LOD of
used for the determination of amaranth in food. For 110 nM under differential pulse voltammetry, which can be
example, Li et al. reported a molecularly imprinted polymer applied to the detection of indigo carmine in real samples.
based electrochemical sensor for the selective and sensitive Furthermore, by coupling flow injection analysis and boron-
detection of amaranth (Li et al. 2019b). The polymer doped diamond electrode, Deroco et al. developed a multiple
ensured the selectivity and the electrochemical method pulse amperometry method for indigo carmine detection
attributed to the sensitivity. The method showed a linear with a well-defined oxidation peak at around 0.67 V
relationship for amaranth in the range of 0.006–10 lM, and (Deroco et al. 2018). The oxidation process involves the
the limit of detection (LOD) of 2 nM. The proposed sensor transfer of two electrons and two protons, yielding the oxi-
can distinguish amaranth from other structurally similar dized species dehydroindigo. The method showed the indigo
substances like sunset yellow and tartrazine. He et al. devel- carmine concentration ranges of 70.0 1000 nM with LOD
oped a kind of manganese dioxide-graphene nanocomposites of 40.0 nM, which was applied in commercial candies at a
modified glassy carbon electrode for amaranth analysis 95% confidence level. These electrochemical methods can be
based on ionic liquids (He et al. 2018b), revealing that the considered of great analytical interest with shorter analysis
oxidation of amaranth is an adsorption-controlled and irre- time and lower cost of instrumentation.
versible electrode process involving two-electron and one- Chen et al. created an Au AuOx substrate by in situ
proton. The method showed a range of 0.02 lM10 lM and oxidation reduction cycle for the SERS detection of indigo
a LOD value of 1.0 nM (S/N ¼ 3) for amaranth in soda sam- carmine (Chen et al. 2018). The novel substrate greatly
ples. The above electrochemical methods provided a rapid, improved the SERS performance through both electromag-
simple, low-cost and effective approach to analyze amaranth netic and chemical enhancements. For indigo carmine, the
in food samples. substrate could exhibit a characteristic peak at 1582 cm1,
For the optical methods, Liu et al. presented a novel which further showed a range of 107 to 103 M in soft
fluorescence nanosensor for selective and sensitive determin- drink sand ice cream. Besides, Li et al. proposed a resonance
ation of amaranth using carbon dots. The method showed a Rayleigh scattering technique for indigo carmine analysis
linear range of 0.2 30 lM for amaranth with the LOD of using acridine orange (Li et al. 2016). In this work, the weak
0.021 lM (Liu et al. 2019). It reported that amaranth can scattering intensity of acridine orange can be enhanced with
efficiently and sensitively quench the C-dots fluorescence at the presence of indigo carmine, with a linear range of
523 nm by the inner filter effect and non-radiative energy 2 32 lM and the LOD of 2.4 108 M. Without any sig-
transfer mechanisms. The proposed sensor has several nificant interference, the method can achieve sensitive and
advantages such as simplicity in design, fabrication and selective determination of trace level of indigo carmine.
operation, and rapid response. Moreover, by simulating the From the electrochemical methods to the scattering tech-
theoretical spectrum and their characteristic peaks with niques, it can be seen that the characteristic signal and good
density functional theory (DFT), Zou et al. presented a sur- robustness are the indispensible part in analytical methods.
face-enhanced Raman spectroscopy method for the rapid Fortunately, the above methods show specific signal toward
and quantitative detection of amaranth based on silver the analyte and possess the advantages like convenient oper-
nanorods array substrates (Zou et al. 2018). The substrates ations, rapid detection, high sensitivity and reproducibility,
showed characteristic peaks of amaranth located at showing their potential applications in indigo carmine detec-
1345 cm1 and 1361 cm1 with the linear ranges from 1.0 to tion in foodstuffs.
100.0 mg L1 and the LOD of 1.0 mg L1. Those optical
methods are simple, sensitive, rapid and effective, but the
stability and accuracy still remain to be developed due to Other colorants
the potential background interference. Other commonly used colorants include ponceau 4 R
(E124), azorubine (E122), quinoline yellow (E104), and
allura red (E129). They contain azo groups as their chromo-
Indigo carmine phore in the molecular structure, which can produce charac-
As a kind of synthetic colorant, indigo carmine (E132) is an teristic reduction peak by cathodic voltammetric
organic salt derived from indigo by sulfonation. Recent stud- determination. Meanwhile, the –OH on aromatic ring can
ies show that indigo carmine might cause irritations vomit- also make them electrochemically oxidizable by using anodic
ing and diarrhea to human beings and the ADI value for voltammetry (Lipskikh et al. 2018). For example, Silva et al.
indigo carmine is 5.0 mg kg1 d1 (Europe Food Safety described an electrochemical sensor based on ionic liquid
Authority (EFSA)), 2014). Moreover, it is difficult to extract and carbon black combined with chitosan film (Silva, Wong,
them completely from the food samples for analysis, thus and Fatibello-Filho 2019). In this work, an oxidation peak at
10 L. WU ET AL.
þ0.9 V could be observed corresponding to Ponceau 4 R, health than the natural preservatives, which would behave
which was recorded as calibration signal for measurements. adverse effect on children’s behavior (Anand and Sati 2013).
Particularly, the oxidation signal of Ponceau 4 R was greatly Thus, many methods have been developed to analyze preser-
improved by the constructed film, yielding the LODs of vatives for food safety supervision and risk assessment.
0.9 nM. By coupling with molecularly imprinted technique Amongst, due to the lack of active groups, unique polarity
(MIT), Piri et al. developed a simple electrochemical sensor and molecular charge, chromatography and capillary electro-
for azorubine determination in water sample (Piri et al. phoresis are the most popular methods for the analysis of
2018). The method behaved a detection linear range of preservatives.
112 mg L 1 and a LOD of 0.57 mg L 1. The molecularly
imprinted polymer can greatly induce the interference from
food samples, making it more accurate and selective to azor- Benzoic acid and sodium benzoate
Benzoic acid (E200) and sodium benzoate (E211) are used
ubine. The MIP coupled electrochemical method can greatly
as preservatives mainly in pickled products and beverages.
enhance the selectivity of electrochemical sensor, however,
The maximum amount used in foods ranges from 0.2 to
the poor conductivity of the polymer can in turn restrict its
2.0 g/kg under the China National Standards GB 2760-2014.
sensitivity.
Excessive benzoic acid will destroy the VB1 in food and
Besides, Sivasankaran et al. investigated the fluorescence
make the calcium insoluble, which can destroy the absorp-
quenching behavior of quinoline yellow toward copper
nanoclusters (Sivasankaran et al. 2019). The blue fluores- tion of calcium by the human body. Furthermore, long-term
cence of copper nanoclusters can be selectively quenched by intake of such preservatives will increase the risk of cancer
(Javanmardi et al. 2019). Therefore, it is necessary to ensure
quinoline yellow, and the fluorescence intensity decreased
a relative low level of these preservatives in food to meet
with the increase of quinoline yellow concentration. The
regulatory standards.
results showed a linear range from 5.50 to 0.20 lM with the
Timofeeva et al. proposed a simple and highly available
LOD of 0.11 lM. Yao et al. reported a fast and quantitative
microextraction method coupled with high performance
SERS method for allura red detection in candy samples (Yao
liquid chromatography for benzoic and sorbic acids detec-
et al. 2018). The SERS substrate of silver nanorod array was
tion in beverages and sauce samples (Timofeeva et al. 2019).
firstly developed with highly sensitive and reproducible,
Air-assisted dispersive liquid-liquid microextraction proced-
which serves characteristic peaks of allura red at 488, 589,
ure with organic phase solidification was developed as a new
751, 1221, 1270, 1326, 1495, and 1578 cm1, respectively.
method for pretreatment, which made it eco-friendly, simple
The intensity at 1221 cm1 was used to determine allura red
and easy to operate with relatively low LODs. The microex-
with the linear range of 0.8 100 mg L 1 as well as a LOD
traction recoveries of benzoic and sorbic acids ranged from
of 0.05 mg L 1. The robust fluorescence and SERS detection
93% to 105% and 96% to 101%, respectively. The method
can make them perfect methods in fast preliminary testing
showed linear detection ranges of 0.1 150 mg L1 and
of foods, providing potential application in food safety con-
0.05 100 mg L1 for benzoic acid and sorbic acid with
trol field. Those optical methods may suffer from the
LODs of 0.03 mg L1 and 0.02 mg L1, respectively.
instability caused by either food matrices or substrate mate-
Moreover, microfluidic device is widely used in food ana-
rials. Take SERS-based assays for an example, the food
lysis, and the paper-based device has the advantages such as
matrices can lead to strong fluorescence, and the uniform €
lightweight and low in cost. For example, Oztekin exploited
SERS substrates pose crucial influence on the stability of
a silica capillary instead of a column to achieve rapid sample
SERS signal. Hence, it is a favorable strategy to develop vari- €
separation (3 min) and analysis of benzoic acid (Oztekin
ous substrates for different food additives according to their
2018). The method exhibited a detection linear range of
functions and structures.
0.005 0.4 mM and a LOD of 0.405 mg L 1. Besides, Liu et
al. developed a microfluidic paper-based analytical device via
Analysis of preservatives an integrated microfluidic platform for benzoic acid detec-
tion (Liu et al. 2018). In the work, twenty-one food samples
Microbial, enzyme or chemical changes can easily cause were measured to analyze benzoic acid with the deviations
food spoilage due to the high nutrient content in foods. of concentration measurements below 6.6%. The reported
Fortunately, introducing food preservatives can prevent or capillary electrophoresis methods provided a compact and
delay food spoilage. The commonly used preservatives reliable tool for the analysis of benzoic acid, suggesting the
include benzoic acid, sorbic acid, propionic acid and their advantages such as simple procedures, low cost and good
salts, P-hydroxy phenyl acid polymer and sodium nitrite reproducibility. Obviously, the sensitivity of this method
(Figure 5). The work mechanism of these preservatives lies remains to be improved.
in inhibiting growth of the bacteria or their spe-
cific enzymes.
Antioxidants can also act as preservatives, and antioxi- Parabens
dants like BHA, BHT, TBHQ and PG will be discussed in a Parabens, known as p-hydroxybenzoates (E214-219), are
separate section because they have similar chemical struc- intensively used in food, cosmetic and pharmaceutical prod-
tures and are usually used in food together. Besides, the arti- ucts. Some animal studies have reported adverse reproduct-
ficial preservatives may have more obvious harm to human’s ive effects of parabens (Jagne, White, and Jefferson 2016).
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 11
Figure 5. Chemical structures of the representative synthetic preservatives (from 1 to 9: benzoic acid (E210), sodium benzoate (E211), ethylparaben (E214), propyl-
paraben (E216), methylparaben (E218), sodium nitrite (E250), propionic acid (E280) and sodium propionate (E281) and sorbic acid (E200)).
As such, possible detrimental health effects to human could transport capability of blood in human body with adverse
be caused with the use of foods, cosmetics, and pharmaceut- effects. Moreover, nitrite can be readily converted to potent
ical products. To address the concern, many regulatory carcinogenic N-nitrosamines by reacting with the amino
agencies have established exposure limits for the use of para- acids of proteins. The lethal level of nitrite is about 22 mg/
bens in those products. Therefore, it is necessary to monitor kg (body weight), and the maximum allowed nitrite concen-
parabens in such products to control and verify the safety. tration in meat products is 200 ppm (Parthasarathy and
Most of the analytical methods for parabens are chroma- Bryan 2012). Hence, a variety of analytical methods, espe-
tography technique coupled with certain separation system. cially chromatography and capillary electrophoresis methods
For example, Becerra-Herrera et al. reported the gas chro- have been developed for nitrite analysis.
matography mass spectrometry for the simultaneous detec- For example, Lin et al. constructed a new capillary liquid
tion of parabens like n-butylparaben, methylparaben (E218), chromatography based on polymer-based monolithic column
ethylparaben (E214), propylparaben (E216) coupling with (Lin, Hsu, and Fuh 2019). Without derivatization process,
rotating disk adsorptive extraction (Becerra-Herrera et al. the proposed method obtained comparable LOD/LOQ val-
2018). The method showed recoveries of 79%91% with ues of 1.50/4.25 and 1.25/4.00 lg g1 for nitrite and nitrate.
relative standard deviations less than 10% and LODs lower Moreover, for more than 200 analyses, the separation per-
than 0.05 lg L 1 for the analytes. The proposed method formance showed no noticeable decline, which suggested it
was very suited to the detection of parabens with fast extrac- is an acceptable method to monitor the occurrence of nitrite
tion and low volume of derivatizing agent. Van Overmeire and nitrate in vegetables. Antczak-Chrobot et al. reported an
et al. reported a liquid chromatography-tandem mass spec- ion chromatographic method to determine nitrite and
trometry for the simultaneous determination of four para- nitrate in sugar by-products (Antczak-Chrobot, Ba˛k, and
bens (methyl-, ethyl-, propyl-, and butyl-paraben) (Van Wojtczak 2018). The method needn’t to enzymatically con-
Overmeire et al. 2019). Using negative electrospray ioniza- verted nitrate to nitrite, behaving recoveries of nitrate and
tion, the four parabens were analyzed with LODs of nitrite better than 90% and 96%. The detection linear ranges
0.1 0.2 ng g1 and LOQs of 0.2 0.5 ng g1, which could for nitrite and nitrate are 0.05 15.0 mg L1 and
be an alternative to conduct simultaneous detection of para- 0.1 50 mg L1 as well as LODs of 0.015 mg L1 and
bens in food matrices. 0.046 mg L1, respectively. Owing to the simplicity, the
small volume of samples and short time of analysis, the
technique is suitable for routine control of nitrite and nitrate
Nitrite concentration. Tezcan proposed a sample stacking-capillary
Nitrite, mostly sodium nitrite (E250), is used for curing electrophoresis method for nitrate contents determination
meat because it prevents bacterial growth and gives the (Tezcan 2018). The measurement could be completed in less
product a desirable pink-red fresh color. Unfortunately, than 3 min without requiring any preliminary preparation
nitrite could bind with hemoglobin to reduce the oxygen method. The nitrate concentration in commercial milk
12 L. WU ET AL.
samples ranged from 0.20 to 1.50 mg L1 with the LOD of sensitive and more suitable for on-site analysis of
0.03 mg L1 (S/N ¼ 3), showing the desirable separation per- antioxidants.
formance and detection sensitivity of electrophor-
esis technique.
Tocopherols
Tocopherols are a kind of natural compounds in seed oils
Other preservatives found to have vitamin E activity, which provide rich infor-
Other preservatives such as sorbic acid (E200) and potas- mation about the identity and the quality of vegetable oils.
sium sorbate (E202), propionic acid (E280) and propionates Amongst, a-, c- and d-tocopherol usually serve as antioxi-
(E281-283) are also reported to be detected by many analyt- dants and are widely used in foods. Particularly, it is of great
ical methods. For instance, by coupling capillary electro- value to analyze tocopherols for the quality control of edible
phoresis with UV spectrophotometric method, de Jesus et al. oils and for oil admixtures verification (Lechner, Reiter, and
realized the simultaneous detection of propionate and sor- Lorbeer 1999). Among various detection methods, the low
bate with the separation time less than 5 min (de Jesus et al. cost, high sensitivity and simple operations of the electro-
2018). In this work, propionate and sorbate were detected at chemical detection platform make it an attractive alternative
235 nm and 250 nm with the LOQs of 4.3 and 1.5 mg kg1 to determine tocopherols.
in bread. The method was demonstrated to be quick, simple, For instance, Riman et al. proposed an electrochemical
and easy for the simultaneous determination of propionate detector for HPLC by using the pencil graphite as disposable
and sorbate. Further, Guadalupe et al. proposed a simple working electrode (Riman et al. 2019). The electrochemical
method for propionate and sorbate analysis by capillary elec- detector in combination of HPLC exhibited nanomolar con-
trophoresis with capacitively coupled contactless conductiv- centrations of tocopherols. The combined method was
ity detection (Guadalupe et al. 2018). The separation of highly applicable to separate and detect the tocopherols,
analytes can be achieved within 7 min with good resolution. showing linear detection ranges of 0 500 nM with LODs of
The method behaved LOQs of 0.11 and 0.21 g kg1, as well 2.2, 1.2, 0.8 nM for a-, c- and d-tocopherol. Moreover, the
as the recoveries varied from 88% to 94% and 86% to 102% proposed detector behaved excellent electrochemical per-
for propionate and sorbate. The capillary electrophoresis formance identical to that of commercial UV-VIS HPLC
method is appropriate for the analysis of inorganic and detector. Similarly, Sys et al. developed a new electrochem-
organic compounds that lack a strong chromophore group, ical method for the determination of the sum of tocopherols
such as propionate and sorbate. (Sys et al. 2017). Based on extraction of tocopherols into
glassy carbon paste electrode, the detection was conducted
by square wave stripping voltammetry. Within 5 min and
Analysis of antioxidants 15 min accumulation, the method showed different linear
Food antioxidants are substances that can prevent or slow ranges of 0.5 40 lM and 0.05 10 lM as well as the LODs
damage to food caused by free radicals and unstable mole- of 0.1 lM and 0.033 lM, respectively. The electrochemical
cules in food produces. Commonly, the antioxidants used in method can be recommended as an alternative for analysis
food produces include tocopherols (E306-309), propyl gallate of the total quantity of tocopherols in food samples.
(PG, E310), tert-butyl hydroquinone (TBHQ, E319), buty-
lated hydroxyanisole (BHA, E320) and butylated hydroxyto- Propyl gallate
luene (BHT, E321), which can prevent food from turning Propyl gallate (PG, E310) is widely employed in foodstuffs
rancid and assure excellent quality and pleasant appearance, and plays an important role in preserving food products
odor, and taste for the consumer (Figure 6). It has been such as fried foods, edible fats and oils as a common anti-
reported that high doses of antioxidants like BHA and BHT oxidant. However, it was reported that high doses of PG
do harmful effects on animals and may induce the promo- may cause potential health hazards such as apoptosis and
tion of cancer (Williams, Iatropoulos, and Whysner 1999). DNA cleavage (Vikraman et al. 2013). The ADI recom-
Thus, it is important to determine and quantify the mended by JECFA is 0 1.4 mg/kg body weight and the
unavoidable additives in food samples. maximum level in food is 200 mg kg1. Therefore, it is of
Most of the antioxidants possess phenolic hydroxyl great importance to develop analytical methods for quantita-
group, which could act as an active electron donor to scav- tive determination of PG.
enge free radicals. Besides the GC-MS or LC-MS (Kibi et al. For example, Wu et al. proposed an electrochemical bio-
2019), the food antioxidants can be electrochemical active sensor for the simultaneous determination of BHA and PG
and are easy to be determined by the electrochemical meth- in foodstuffs with well-defined oxidation peak at 624 and
ods due to their specific structure. Moreover, from the con- 655 mV (Wu et al. 2016). By modifying spiny Au-Pt nano-
stitute their excellent electrochemical properties in terms of tubes and horseradish peroxidase onto electrode, it behaved
high electron transfer rate for diverse redox systems, low synergetic catalase-like activity toward BHA and PG. The
noise and high sensitivity, these features have enabled method showed a wide linear range of 0.3 50 mg L1 and
expanding the application scope of electrochemical methods 0.1 100 mg L1 for BHA and PG with LODs of 0.046 mg
in antioxidants analysis. In this section, electrochemistry L1 and 0.024 mg L1. Moreover, BHA and PG were dem-
related methods will be introduced because they are more onstrated by a simple visual detection method, which
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 13
Figure 6. Chemical structures of food antioxidants (from 1 to 5: a-, c- and d- tocopherol (E307-309), propyl gallate (PG, E310), tert-butyl hydroquinone (TBHQ,
E319), butylated hydroxyanisole (BHA, E320) and butylated hydroxytoluene (BHT, E321)).
involved Au-Pt nanotubes as catalyst and TMB as indicator. fats and oils is no more than 0.02%. Here, new developed
Due to the strong oxidation resistance of BHA and PG, electrochemical methods with simple operations and rapid
TMB could be less oxidized and result in the decrement of detection will be introduced.
optical density. Both the electrochemical and visual detec- For example, Yue et al. developed an electrochemical sensor
tion methods revealed the rich electron in the structure of for selective detection of TBHQ by integrating molecularly
BHA and PG. imprinted polymer, PdAu nanoparticles and reduced graphene
Using nickel nanoparticles modified electrode, oxide together to form the sensing interface (Yue et al. 2019).
Sivasankaran and Kumar developed an electrochemical sen- The sensor showed favorable accuracy and satisfactory precision
sor for the analysis of PG (Sivasankaran and Kumar 2019). for the analysis of TBHQ, behaving the linear range of
Simply, nickel nanoparticles modified surface improved the 0.5 60 mg L1 and a LOD of 0.046 mg L1. The detection
oxidation peak current of PG molecules. The method could results were further verified by HPLC with good recoveries rang-
achieve two linear detection ranges for PG from 4.0 104 ing from 99.4 to 108.5% in spiked edible oil samples. Besides,
M to 1.0 103 M and from 2.0 106 to 3.0 104 M Cao et al. designed a ratiometric electrochemical sensor for
with the calculated LOD of 5.85 108 M. Finally, the TBHQ detection based on MnO2 and reduced graphene oxide
method was verified by standard spectrophotometry for (Cao et al. 2019). Wherein, MnO2 acted as the inner reference
quantification of PG in spiked food samples. The electro- electrochemical oxidation peak at 0.572 V, and the oxidation
chemical sensor behaved good selectivity, sensitivity and sta- peak for TBHQ is about 0.06 V. The sensor showed two linear
bility toward PG detection, which is helpful for regulating ranges of 1.0 50.0 lM and 100.0 300.0 lM with a LOD of
and reducing the risk of overuse of PG in foods. 0.8 lM. The ratiometric method can avoid the interference of
other antioxidants in matrix samples and achieve rapid and
accurate detection of TBHQ in edible oil.
Tert-butyl hydroquinone
Tert-butyl hydroquinone (TBHQ, E319) is known as an arti-
ficial antioxidant with two phenolic hydroxyl groups and a Butylated hydroxyanisole
tert-butyl group. The chemical structure indicates the stabil- Butylated hydroxyanisole (BHA, E320) consists of a mixture
ity and strong oxidation resistance of TBHQ. Thus, TBHQ of two isomers, 2-tert-/3-tert-butyl-4-hydroxyanisole, which
is widely adopted as an antioxidant in unsaturated vegetable is widely used in fat, oil and emulsified fat products as an
oils and many edible animal fats. Study has shown that antioxidant. It was reported that BHA increased the inciden-
excessive use of TBHQ may do genotoxicity and carcinogen- ces of hyperplasia of the forestomach, and the higher dose
icity for man (Van Esch 1986). FDA sets a maximum limit induced a significant increase in the incidence of papilloma
for TBHQ of 200 mg/kg in foods and the limit of TBHQ in and squamous cell carcinoma for rats (Ito et al. 1983). FDA
14 L. WU ET AL.
suggests that the levels of BHA are limited to 2 1000 mg are also used in food stuffs. Due to the safety concerns, peo-
kg1. So, it is essential to develop simple and sensitive ana- ple are more favorable to natural antioxidants like ascorbic
lytical methods for BHA. acid and tea polyphenols. The artificial ones are more likely
Bavol et al. reported an electrochemical method for the simul- to cause long-term toxicity for human health (Van der
taneous of BHA, PG and TBHQ using flow injection analysis Heijden, Janssen, and Strik 1986). Therefore, many research-
coupled with multiple-pulse amperometry (Bavol et al. 2018). ers showed intense attention on antioxidants analysis, espe-
Without preliminary separation, the method could realize con- cially for the synthetic ones.
tinuous application of three potential pulses at a single-injection Pedroso et al. proposed an electrochemical sensor for
step. As a result, it provided low LOQs of 0.85 lM, 1.45 lM, and dodecyl gallate detection based on molecularly imprinted
2.51 lM for BHA, PG and TBHQ, respectively. Aghdam et al. polymer film (Pedroso et al. 2017). Electropolymerized
developed a new microfluidic-based method based on a minia- molecularly imprinted polymers (MIPs) were synthesized
turized chemical analysis system for antioxidants detection in after modifying carbon nanotubes on electrode. Using
edible oils (Aghdam et al. 2019). The method showed recoveries square wave voltammetry, the sensor showed linear range
of 93.2 102% and 95.4–97% for BHA in sunflower oil and olive from 0.50 to 8.0 nM with a LOD of 0.22 nM and a LOQ of
oil. Without any pretreatment, the detection linear range for 0.67 nM. In this work, carbon nanotubes increased the active
BHA was 10 100 lM and the LOD was 1.4 lM. Squissato et al. surface area and enhance the electron transfer and the MIPs
presented an effective voltammetry method for the direct detec- ensured the selectivity for the recognition of dodecyl gallate
tion of BHA, PG, TBHQ and BHT in biodiesel (Squissato, in complex matrix. Sivasankaran et al. fabricated a glassy
Richter, and Munoz 2019). In the work, partial least square carbon electrode chemically modified with gold nanopar-
regression was applied to resolve overlapped signals. Owing to ticles for octyl gallate detection (Sivasankaran et al. 2016).
the PLS approach, there is no need to modify electrode to achieve By combining the advantages of gold nanoparticles and self-
the separation of the peaks. assembled monolayer of dodecane thiol, the sensitivity of
The above electrochemical methods, including can be eas- the proposed method is largely improved. The method
ily applied in the detection of food samples without high behaved a linear detection range of 1.2 0.2 lM and a LOD
demands on instrumentation, which showed great potential of 8.3 nM, which was demonstrated to be applicable in food
in food analysis when applying a simple extraction proced- samples such as margarine, butter and sunflower oil.
ure or an analysis model.
Analysis of sweeteners
Butylated hydroxytoluene
Butylated hydroxytoluene (BHT, E321) is another kind of Obesity epidemic is a global health problem which has
synthetic phenolic antioxidant commonly applied in food caused wide attention recently, and low-calorie foods with
and pharmaceutical industries. Usually, BHT is used in com- sweet taste are favored by the individuals of all ages. Thus,
bination with BHA to enhance the antioxidant activity. low and no-calorie sweeteners with similar sweet taste as
Neverthless, long-term exposure to BHT will cause chronic sugar has played a critical part in food industry. Wherein,
poisoning, causing adverse effects like metabolic disorders artificial sweeteners like acesulfame (E950), aspartame
(Branen 1975). The new developed detection of BHT mainly (E951), cyclamate (E952), saccharin (E954), sucralose (E955)
focused on all kinds of electrochemical methods. and neotame (E961) are non-caloric replacements used to
For example, Emam et al. developed a molecularly enhance sweet tastes for food. However, the safety of artifi-
imprinted electrochemical gas sensor to analyze BHT (Emam cial sweeteners has been controversial. Therefore, it is of
et al. 2018). The electrochemical sensor was constructed via great significance to develop new methods for artificial
three steps: (1) modify a thin layer of graphene onto the sub- sweeteners analysis to control food quality and ensure con-
strate, (2) facilitate polymer template and target molecule by sumer’s health.
electrochemical polymerization and (3) remove the target The sweeteners can be divided into three groups based
molecules to produce molecularly imprinted cavities. The on their different charging states in solution: positively
method can achieve good sensitivity and selectivity for BHT charged (aspartame and neotame), negatively charged (ace-
with the LOD of 0.02 ppb. Karimi-Maleh et al. suggested a vol- sulfame K and cyclamate Na), and uncharged (saccharin
tammetric method for the detection of ferulic acid and BHT and sucralose) (Figure 7). Also, as the sweeteners show
(Karimi-Maleh et al. 2019). The strategy was based on the diversiform structures, various methods ranging from chro-
twofold modification of electrode, NiO-embedded single-wall matography to electrochemical methods and then to optical
carbon nanotube and n-methyl-3-butylimidazolium bromide. methods have been proposed.
The method can resolve the oxidation signal of BHT with a
LOD of 0.1 mM, which was favorable for the direct analysis of
Acesulfame, cyclamate and saccharin
BHT in corn milk, wheat flour, and corn cider samples.
Acesulfame (E950), cyclamate (E952) and saccharin (E954)
are often used in the form of sodium or calcium salt, which
Other antioxidants are non-glucose, low-calorie products with a highly sweet
Other antioxidants like natural ones (tea polyphenols, ascor- taste. Since they are chemically synthesized and they may
bic acid) and artificial ones (dodecyl gallate, octyl gallate) create potential health risks, their presence in food will
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 15
Figure 7. Chemical structures of sweeteners (from 1 to 6: acesulfame K (E950), aspartame (E951), cyclamate (E952), saccharin (E954), sucralose (E955) and neo-
tame (E961)).
cause consumers’ mistrust. Consequently, many govern- However, it is used not as widely as aspartame in food prod-
ments set maximum ADI value for acesulfame and sac- ucts. Both of the two sweeteners remain safety controversy,
charin. Therefore, it is encouraging to investigate the level and their safety has been studied extensively since their dis-
of these sweeteners using effective methods. covery with research that includes animal studies, clinical
Liu et al. proposed a sweetness sensor to detect the nega- and epidemiological research (O’Donnell 2012). In this
tively charged sweeteners like acesulfame K (Liu et al. 2020). sense, it is relevant to develop new analytical methods to
In the sensor, lipid polymer membranes act as the taste- verify quality control and monitoring of these compounds
sensing part, and the potential change in membrane caused in food matrixes.
by target adsorption was defined as outputs. The method For instance, Le et al. proposed the electrochemical oxi-
can well predict acesulfame K in a concentration of dation of a binary mixture of caffeic acid (CAF) and aspar-
0.2 0.7 mM with R2 ¼ 0.99 when coexist with salty sub- tame (ASP) on preanodized screen-printed carbon electrodes
stances, which suggested it was useful for the evaluation of to form copolymer of poly(CAF-ASP) (Le, Su, and Cheng
acesulfame K food and beverages. Based on nanodiamond 2019). Using the DPV method, the copolymer modified elec-
paste modified with 0.1 pM of a, b, and c-cyclodextrin, trode was applied to ASP determination resulting with a lin-
Stefan-van Staden et al. designed three sensors for pattern ear range from 0.05 to 10 lM as well as a LOD of
recognition of aspartame, acesulfame K and cyclamate Na 0.0076 lM and a LOQ of 0.0256 lM. The proposed electro-
(Stefan-van Staden, Moscalu-Lungu, and van Staden 2019). chemical platform showed easy operations in analyzing ASP
Nanodiamond makes the sensor channel more stable than in drink samples, offering potential applications in the food
nano-graphite and the modified cyclodextrins give the industry. Bathinapatla et al. presented a highly sensitive elec-
selectivity of analytes. All the three targets can be detected trochemical sensor for the detection of neotame using DPV
at the pM levels with no sampling and fast screening, with a modified glassy carbon electrode (Bathinapatla et al.
revealing the high sensitivity of the sensors in food samples. 2015). By fabricating the electrode with copper nanoparticles
Besides, Seyinde et al. adopted a reverse phase HPLC decorated multiwalled carbon nanotubes, it behaved an obvi-
method for quantitative determination of saccharin ous oxidation peak of neotame at a potential of 1.3 V. The
(Seyinde, Ejidike, and Ayejuyo 2019). This work simply peak currents of DPV plots showed a linear ranged from
reported the use of ahypersil C-18 column and an eluent 0.05 to 2.0 mM for the detection of neotame. The detection
consisting of 10% acetic acid in water. Based on the method. results in food samples were verified by capillary electro-
the saccharin content in beverages was found to be phoresis method with a 96% confidence level, revealing that
l.23 ± 0.03 mg mL1, which showed acceptable analytical lin- it is a promising alternative in analysis of sweeteners.
earity, precision and accuracy.
Sucralose
Aspartame and neotame Sucralose (E955) is an artificial sweetener and a kind of
Aspartame (E951) and neotame (E961) are two artificial commonly used sugar substitute, which is used as a replace-
sweeteners with similar chemical structures. Chemically, ment for, or in combination with other artificial like aspar-
neotame owns a 3,3-dimethylbutyl group attached to the tame and acesulfame potassium. Usually, sucralose is applied
amino group of the aspartic acid portion of the molecule. in products such as candy, breakfast bars and soft drinks.
16 L. WU ET AL.
Despite being generally recognized as safe by regulatory analysis (PCA) models. The results reveal that FTIR can be
agencies including the FDA and the Joint FAO/WHO applied for the rapid detection of sodium alginate in fish
Expert Committee Report on Food Additives, based on and other meat products. Devi et al. reported an ampero-
information from clinical studies, high doses of sucralose metric immunosensor for the detection of monosodium glu-
could do effects on various organs but on illnesses such as tamate based on Au@MoS2/chitosan nanocomposites
obesity, diabetes, and cancer remain yet-unknown (Rodero, modified electrode (Devi et al. 2019). In the method, the
de Souza Rodero, and Azoubel 2009). Therefore, it is neces- nanocomposites acted as a conductive matrix and the anti-
sary to analyze the content of sucralose to control the glutamate antibody was immobilized on to its surface to dis-
food quality. cern the glutamate. It showed a detection range of
Voss et al. described a silylation derivatization method 0.05 200 mM with the LOD and LOQ of 0.03 and 0.1 mM.
and gas chromatography mass spectrometry (GC-MS) for By combining amperometric method with immunoassay, the
detection of sucralose in groundwater samples (Voss, sensor behaved satisfactory selectivity, reproducibility
Newman, and Miller-Schulze 2019). In the work, a deuter- and stability.
ium-labeled internal standard was employed to account for Many methods have been explored to determine food
reaction and sample processing imprecision. As such, this additives with the prospect of good selectivity, sensitivity
proposed method is robust for sucralose analysis with the and stability (Jiang et al. 2019; Nair et al. 2020; Xie et al.
LOD of 21.8 ng L1. The quality control measures are com- 2019). However, most of them may suffer from the weak-
petitive with previously published methods such as GC-MS, ness like time-consuming, complex pretreatment, compli-
HPLC-MS/MS for the quantification of sucralose by deriva- cated operations and high cost. It seems that high analytical
tization. Encouragingly, Heider et al. reported a service- performance (e.g. sensitivity and stability) will make other
learning approach related with analytical techniques to allow performance such as convenience and low cost compro-
students to learn analytical skills toward sucralose in water mised. To compare the performance of analytical methods
(Heider et al. 2018). As sucralose neither absorb light in the in food additive analysis, Table 3 summarized the detection
visible or UV region, nor electrochemically active or photo- parameters such as linear range, limit of detection (LOD)
luminesce, and the water matrix precludes the use of infra- and limit of quantification (LOQ), as well as their advan-
red spectroscopy due to water interferes in the OH region. tages and disadvantages.
Students finally chose reversed-phase liquid chromatography
with mass spectrometry to measure sucralose. From this
point, the analytical skills from the molecular structure to Prospects and challenges
the properties and to the techniques are thirstily required to With the development of novel materials, especially the
design a method for the analysis of food additives. functional nanomaterials, the performance of analytical tech-
niques has been greatly improved. As we discussed above,
all the analytical methods showed remarkable advancement
Analysis of other food additives
in the analysis of food additives. However, there are still
Together with the food additives discussed above, other many remaining challenges to be solved such as complex
additives are commonly detected include emulsifiers, stabil- sample pretreatment, high-cost, time-consuming and tedious
izers, flavor enhancers and so forth. Amongst, the favored operations and relatively low detection efficiency. For the
detection methods are chromatography, spectrophotometry, food additives, the complexity may come from three aspects:
electrophoresis, optical and electrochemical methods. Based (a) a large variety of food additives make them complex; (b)
on the differences in the structure and property of food various foods using food additives make the matrices com-
additives, they are often analyzed by the fitting technique or plex; (c) the similar structures of food additive make them
their coupled ones to achieve the best results. complex to be identified specifically. The limitations of ana-
For example, Oellig et al. presented a high-performance lytical techniques are dependent on detection principles
thin-layer chromatography method coupled with fluores- themselves and could be greatly influenced by the perform-
cence detection for the detection of acetic acid esters ance of materials. Thus, it is an effective means to develop
(E472a), lactic acid esters (E472b), citric acid esters (E472c), methodology and new functional materials to improve the
and mono- and diacetyl tartaric acid esters (E472e) (Oellig, technique performance.
Link, and Schwack 2020). Under UV 366 nm, direct visual For example, to meet the challenges of detection accur-
comparison of the emulsifier pattern was obtained, which acy, most commonly used techniques like mechanical,
can be developed for sensitive and selective screening of microwave and thermal approach are applied to extract the
E472. The visual fingerprint was demonstrated to be very analytes and minimize interferences from food samples.
valuable and suitable for a fast and simple quality control of Those careful processing procedures will inevitably include
emulsifier in food samples. Using Fourier transform infrared laborious and time-consuming operations, which seems to
spectroscopy (FTIR), Clarke developed a new method to be contradictory to the convenience of analytical methods.
measure sodium alginate in fish meat products (Clarke However, if we analyze the food additives starting from their
2017). After drying and extracting the sample, FTIR was structure, they may fall into an optimal detection method.
adopted to directly quantify the amount of sodium alginate Like this, a certain food additive such as sucralose is neither
with partial least squares (PLS) and principal component optically nor electrochemically active, and its structure
Table 3. Comparison of analytical methods for food additives detection (not given: —).
Performance
Methods Food additives Advantage Disadvantage References Examples Linear range: LOD LOQ
HPLC preservative Good selectivity, high Expensive equipment, Timofeeva et al. (2019) benzoic acids linear range: 0.1–150 mgL1 0.03 mgL1 0.1 mgL1
sensitivity, recovery and high cost of inspection, (correlation coefficient: 0.998)
reproducibility long cycle and inability to sorbic acids linear range: 0.05–100mgL1 0.02 mgL1 0.05 mgL1
meet batch inspection (correlation coefficient: 0.998)
sweeteners —————————
antioxidant Riman et al. (2019) Gentisic acids linear range: 1 1000nmolL1 0.4 nmolL1 1.3nmolL1
(correlation coefficient: 0.999)
Caffeic acids linear range: 3 1000nmolL1 1.1 nmolL1 3.6 nmolL1
(correlation coefficient: 0.999)
Dihydrocaffeic acids linear range: 1 1000nmolL1 0.4 nmolL1 1.4 nmolL1
(correlation coefficient: 0.999)
GC-MS or LC-MS sweeteners Rapid, sensitive and accurate Cumbersome, expensive, Voss, Newman, and Sucralose linear range: 99 2004ngL1 0.8ngL1 0.7 1.9ngL1
in quantification time-consuming, and Miller-Schulze (2019)
antioxidant need to be completed Kibi et al. phenolic linear range: Good 10 80 lgL1 ——————
in lab (2019)Kivilompolo, antioxidants
Obůrka, and (GC-MS)
Hy€otyl€ainen (2007)
Spectrometry colorants High sensitivity, wide linear Expensive, lack of Ntrallou, Gika, and Amaranth, ponceau ————————— 0.2–6.0 mgL1 ——————
range, fast and convenient, stability, require Tsochatzis (2020) 4R, sunset yellow,
suitable for on-site detection, sample database. tartrazine,
less sample brilliant blue
preservative consumption —————————
antioxidant —————————
Capillary preservative Simple, rapid, low cost and Relative low Lin, Hsu, and NO2-(3 days) Accuracy: 98.17 101.26% 0.06mgL1 0.17 mgL1
electrophoresis less sample consumption sensitivity and Fuh (2019) NO3-(3 days) Accuracy: 96.00 101.22% 0.05mgL1 0.16 mgL1
sweeteners reproducibility —————————
Electrochemical colorants Simple, require low volume Relative low Sierra-Rosales, Toledo- Tartrazine Sensitivity: 3.5lAlmolL1 0.22lmolL1 ——————
method samples, high sensitivity, low reproducibility and Neira, and Sunset yellow Sensitivity: 3.9 lAlmolL1 0.12lmolL1 ——————
detection limit, strong anti- selectivity, short working Squella (2017) Carmoisine Sensitivity: 4.1 lAlmolL1 0.11lmolL1 ——————
preservative interference ability and high life, —————————
sweeteners recovery rate the electrode and —————————
antioxidant electrolyte are difficult —————————
to reuse.
ELISA colorants Simple operations, low cost, Tedious procedures, Ntrallou, Gika, and Sudan I ————————— 0.07 lgL1 ——————
high sensitivity, fast relative low Tsochatzis (2020)
and convenient. reproducibility, require
antibody, and easy to
be interfered
by matrix.
Surface-enhanced colorants Rapid, sensitive and simple, Lack of stability, require Ntrallou, Gika, and Tartrazine, sunset ————————— 0.1 lgL1 ——————
Raman scattering rapid response, low volume either complex data Tsochatzis (2020) yellow, brilliant
samples and on- processing or high-quality Blue, acid red
sweeteners site detection SERS substrate —————————
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION
17
18 L. WU ET AL.
contains rich hydroxyl groups with strong polarity. So, it University Start-up Scientific Research Projects of China (KYQD(ZR)-
may be not appropriate for optical and electrochemical 21044). Guozhen Liu acknowledges the financial support by the
University Development Fund (UDF01002012) from the Chinese
methods, but the chromatography method with derivative
University of Hong Kong (Shenzhen).
materials is favorable for the substance. From this aspect,
how to choose right method with proper functional materi-
als for additives detection is equally important as developing
new methods. References
Besides, researchers have been paying increasing attention
Abdel-Moemin, A. R., J. M. Regenstein, and M. K. Abdel-Rahman.
on addressing the limitations of analytical methods. By 2018. New Food Products for Sensory-Compromised Situations.
designing different strategies, analytical performances have Comprehensive Reviews in Food Science and Food Safety 17 (6):
been improved in terms of sensitivity, stability, accuracy and 1625–39. doi: 10.1111/1541-4337.12399.
selectivity. For example, traditional Raman scattering is a Abid, M., H. Yaich, H. Hidouri, H. Attia, and M. A. Ayadi. 2018.
Effect of substituted gelling agents from pomegranate peel on col-
typical optical detection method with low signal response.
our, textural and sensory properties of pomegranate jam. Food
However, the signal intensity of Raman scattering can be Chemistry 239:1047–54. doi: 10.1016/j.foodchem.2017.07.006.
greatly enhanced (106 fold) with the aid of noble metal Aghdam, A. A., M. R. Majidi, H. Veladi, and Y. Omidi. 2019.
nanomaterials like nanosilver or nanogold. The improved Microfluidic-based separation and detection of synthetic antioxi-
Raman scattering is called surface enhanced Raman scatter- dants by integrated gold electrodes followed by HPLC-DAD.
ing (SERS), which is a very popular method in food analysis Microchemical Journal 149:104059. doi: 10.1016/j.microc.2019.
104059.
due to its high robustness. Further, by combining mathem- Ai, Y.-J., P. Liang, Y.-X. Wu, Q.-M. Dong, J.-B. Li, Y. Bai, B.-J. Xu, Z.
atical algorithms, SERS can achieve the multicomponent Yu, and D. Ni. 2018. Rapid qualitative and quantitative determin-
detection for additives, which make the analysis simpler and ation of food colorants by both Raman spectra and Surface-
more convenient. Moreover, the SERS technique can also enhanced Raman Scattering (SERS). Food Chemistry 241:427–33.
couple with other detection methods like electrochemistry, doi: 10.1016/j.foodchem.2017.09.019.
Amchova, P., H. Kotolova, and J. Ruda-Kucerova. 2015. Health safety
immunoassay and microscopy to meet the specific require- issues of synthetic food colorants. Regulatory Toxicology and
ments for the analysis of food additives. Hence, it is another Pharmacology : RTP 73 (3):914–22. doi: 10.1016/j.yrtph.2015.09.026.
way to strengthen the analytical performance by combining Anand, S. P., and N. Sati. 2013. Artificial preservatives and their harm-
with other methods. ful effects: Looking toward nature for safer alternatives.
World Health Organization outlined an “ASSURED” cri- International Journal of Pharmaceutical Sciences and Research 4 (7):
2496.
teria of “(i) Affordable, (ii) Sensitive, (iii) Specific, (iv) User- Anderson, D. M. W. 1986. Evidence for the safety of gum arabic
friendly, (v) Rapid and robust, (vi) Equipment-free, and (vii) (Acacia senegal (L.) Willd.) as a food additive-a brief review. Food
Deliverable to users who need them” for point of need devi- Additives and Contaminants 3 (3):225–30. doi: 10.1080/
ces (Urdea et al. 2006). Wherein, a variety of analytical 02652038609373584.
methods trend to miniaturization, integration, automation Antczak-Chrobot, A., P. Ba˛k, and M. Wojtczak. 2018. The use of ionic
chromatography in determining the contamination of sugar by-
and intelligence. Theoretically, microsystems behave high
products by nitrite and nitrate. Food Chemistry 240:648–54. doi: 10.
efficiency, fast analysis time and low reagent consumption 1016/j.foodchem.2017.07.158.
on separation. Thus, miniaturization and integration of Bathinapatla, A., S. Kanchi, P. Singh, M. I. Sabela, and K. Bisetty. 2015.
already existing techniques is an alternative to the traditional Fabrication of copper nanoparticles decorated multiwalled carbon
techniques. Also, it is believed that the combination with nanotubes as a high performance electrochemical sensor for the
detection of neotame. Biosensors & Bioelectronics 67:200–7. doi: 10.
effective pretreatment techniques with new analytical techni-
1016/j.bios.2014.08.017.
ques can develop automatic and intelligent sensing techni- Bavol, D., A. Economou, J. Zima, J. Barek, and H. Dejmkova. 2018.
ques, which predicts a development direction of intelligent Simultaneous determination of tert-butylhydroquinone, propyl gall-
sensor to integrate artificial intelligence with sensors. While ate, and butylated hydroxyanisole by flow-injection analysis with
such ideas promise the application of those techniques in multiple-pulse amperometric detection. Talanta 178:231–6. doi: 10.
resource-limited field like medical treatment for extreme 1016/j.talanta.2017.09.032.
Becerra-Herrera, M., V. Miranda, D. Arismendi, and P. Richter. 2018.
POC, these requirements and tactics are very important for Chemometric optimization of the extraction and derivatization of
developing analytical platforms in food additives. Only all parabens for their determination in water samples by rotating-disk
the relevant techniques work together with the structures of sorptive extraction and gas chromatography mass spectrometry.
analytes, can it be possible to develop more efficient meth- Talanta 176:551–7. doi: 10.1016/j.talanta.2017.08.071.
ods to analyze food additives. Branen, A. L. 1975. Toxicology and biochemistry of butylated hydrox-
yanisole and butylated hydroxytoluene. Journal of the American Oil
Chemists’ Society 52 (2):59. doi: 10.1007/BF02901825.
Cao, W., Y. Wang, Q. Zhuang, L. Wang, and Y. Ni. 2019. Developing
Disclosure statement an electrochemical sensor for the detection of tert-butylhydroqui-
No potential conflict of interest was reported by the authors. none. Sensors and Actuators B: Chemical 293:321–8. doi: 10.1016/j.
snb.2019.05.012.
Carocho, M., M. F. Barreiro, P. Morales, and I. C. Ferreira. 2014.
Funding Adding molecules to food, pros and cons: A review on synthetic
and natural food additives. Comprehensive Reviews in Food Science
This work was supported by the National Natural Science Foundation and Food Safety 13 (4):377–99. doi: 10.1111/1541-4337.12065.
of China (31801638), the fund of Key Laboratory of Fermentation Carocho, M., and I. C. Ferreira. 2013. A review on antioxidants, proox-
Engineering (Ministry of Education) (202105FE09), and the Hainan idants and related controversy: Natural and synthetic compounds,
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 19
screening and analysis methodologies and future perspectives. Food detection of Amaranth. Colloids and Surfaces. B, Biointerfaces 172:
and Chemical Toxicology: An International Journal Published for the 565–72. doi: 10.1016/j.colsurfb.2018.09.005.
British Industrial Biological Research Association 51:15–25. doi: 10. He, Q., J. Liu, X. Liu, G. Li, P. Deng, J. Liang, and D. Chen. 2018a.
1016/j.fct.2012.09.021. Sensitive and selective detection of tartrazine based on TiO2-electro-
Carocho, M., P. Morales, and I. C. Ferreira. 2015. Natural food addi- chemically reduced graphene oxide composite-modified electrodes.
tives: Quo vadis? Trends in Food Science & Technology 45 (2): Sensors 18 (6):1911. doi: 10.3390/s18061911.
284–95. doi: 10.1016/j.tifs.2015.06.007. Ishidate, M., Jr, T. Sofuni, K. Yoshikawa, M. Hayashi, T. Nohmi, M.
Carocho, M., P. Morales, and I. C. Ferreira. 2018. Antioxidants: Sawada, and A. Matsuoka. 1984. Primary mutagenicity screening of
Reviewing the chemistry, food applications, legislation and role as food additives currently used in Japan. Food and Chemical
preservatives. Trends in Food Science & Technology 71:107–20. doi: Toxicology: An International Journal Published for the British
10.1016/j.tifs.2017.11.008. Industrial Biological Research Association 22 (8):623–36. doi: 10.
Chen, H.-C., C.-H. Chen, C.-S. Hsu, T.-L. Chen, M.-Y. Liao, C.-C. 1016/0278-6915(84)90271-0.
Wang, C.-F. Tsai, and H. M. Chen. 2018. In situ creation of surface- Ito, N., S. Fukushima, A. Hagiwara, M. Shibata, and T. Ogiso. 1983.
enhanced raman scattering active Au-AuO x nanostructures through Carcinogenicity of butylated hydroxyanisole in F344 rats. Journal of
electrochemical process for pigment detection. ACS Omega 3 (12):
the National Cancer Institute 70 (2):343–52.
16576–84. doi: 10.1021/acsomega.8b02677. Jagne, J., D. White, and F. Jefferson. 2016. Endocrine-disrupting chemi-
Clarke, A. D. 2017. Measurement of total sodium alginate in restruc-
cals: Adverse effects of bisphenol A and parabens to women’s health.
tured fish products using fourier transform infrared spectroscopy.
Water, Air, & Soil Pollution 227 (6):182. doi: 10.1007/s11270-016-
EC Nutrition 11:33–45.
2785-3.
Codex Alimentarius Commission. 1989. Class names and the inter-
Javanmardi, F., J. Rahmani, F. Ghiasi, H. Hashemi Gahruie, and A.
national numbering system for food additives. CAC/GL :36–1989.
Coultate, T., and R. S. Blackburn. 2018. Food colorants: Their past, Mousavi Khaneghah. 2019. The Association between the
present and future. Coloration Technology 134 (3):165–86. doi: 10. Preservative Agents in Foods and the Risk of Breast Cancer.
1111/cote.12334. Nutrition and Cancer 71 (8):1229–40. doi:10.1080/01635581.2019.
de Jesus, F. F. S., A. G. Coelho, L. de Assis Pallos, J. A. F. da Silva, D. 1608266. PMC:31044613
Daniel, and D. P. de Jesus. 2018. Simple and fast method for simul- Jiang, K., D. Nie, Q. Huang, K. Fan, Z. Tang, Y. Wu, and Z. Han.
taneous determination of propionate and sorbate in bread by capil- 2019. Thin-layer MoS2 and thionin composite-based electrochemical
lary electrophoresis with UV spectrophotometric detection. Journal sensing platform for rapid and sensitive detection of zearalenone in
of Food Composition and Analysis 72:93–6. doi: 10.1016/j.jfca.2018. human biofluids. Biosensors & Bioelectronics 130:322–9. doi: 10.
06.010. 1016/j.bios.2019.02.003.
de Lima, L. F., C. C. Maciel, A. L. Ferreira, J. C. de Almeida, and M. Joshi, V., and K. Pancharatna. 2019. Food colorant Sunset Yellow
Ferreira. 2019. Nickel (II) phthalocyanine-tetrasulfonic-Au nanopar- (E110) intervenes developmental profile of zebrafish (Danio rerio).
ticles nanocomposite film for tartrazine electrochemical sensing. Journal of Applied Toxicology: JAT 39 (4):571–81. doi: 10.1002/jat.
Materials Letters 262:127186. 3747.
Deroco, P. B., R. A. Medeiros, R. C. Rocha-Filho, and O. Fatibello- Karimi-Maleh, H., R. Farahmandfar, R. Hosseinpour, J. Alizadeh, and
Filho. 2018. Selective and simultaneous determination of indigo car- A. Abbaspourrad. 2019. Determination of ferulic acid in the pres-
mine and allura red in candy samples at the nano-concentration ence of butylated hydroxytoluene as two phenolic antioxidants using
range by flow injection analysis with multiple pulse amperometric a highly conductive food nanostructure electrochemical sensor.
detection. Food Chemistry 247:66–72. doi: 10.1016/j.foodchem.2017. Chemical Papers 73 (10):2441–7. doi:10.1007/s11696-019-00793-y.
12.006. Kibi, M., S. Nishiumi, T. Kobayashi, Y. Kodama, and M. Yoshida.
Devi, R., S. Gogoi, S. Barua, H. S. Dutta, M. Bordoloi, and R. Khan. 2019. GC/MS and LC/MS-based tissue metabolomic analysis
2019. Electrochemical detection of monosodium glutamate in food- detected increased levels of antioxidant metabolites in colorectal
stuffs based on Au@MoS2/chitosan modified glassy carbon elec- cancer. Kobe Journal of Medical Sciences 65 (1):E19.
trode. Food Chemistry 276:350–7. doi: 10.1016/j.foodchem.2018.10. Kivilompolo, M., V. Obůrka, and T. Hy€ otyl€ainen. 2007. Comparison of
024. GC–MS and LC–MS methods for the analysis of antioxidant phen-
Elekima, I., O. E. Nwachuku, D. Ukwukwu, H. U. Nwanjo, and N. olic acids in herbs. Analytical and Bioanalytical Chemistry 388 (4):
Nduka. 2019. Biochemical and histological changes associated with 881–7. doi: 10.1007/s00216-007-1298-8.
azo food dye (Tartrazine) in male albino rats. Asian Journal of Le, A. V., Y. L. Su, and S. H. Cheng. 2019. A novel electrochemical
Research in Biochemistry 5:1–14. doi: 10.9734/ajrb/2019/v5i130083. assay for aspartame determination via nucleophilic reactions with
Emam, S., A. Adedoyin, X. Geng, M. Zaeimbashi, J. Adams, A.
caffeic acid ortho-quinone. Electrochimica Acta 300:67–76. doi: 10.
Ekenseair, E. Podlaha-Murphy, and N. X. Sun. 2018. A molecularly
1016/j.electacta.2019.01.020.
imprinted electrochemical gas sensor to sense butylated hydroxyto-
Lechner, M., B. Reiter, and E. Lorbeer. 1999. Determination of toco-
luene in air. Journal of Sensors 2018:1–9. doi: 10.1155/2018/3437149.
pherols and sterols in vegetable oils by solid-phase extraction and
Europe Food Safety Authority (EFSA). 2014. Scientific opinion on the
subsequent capillary gas chromatographic analysis. Journal of
re-evaluation of indigo carmine (E 132) as a food additive. EFSA.
Chromatography. A 857 (1–2):231–8. doi: 10.1016/s0021-
Journal 12 (7:3768):1–57.
Guadalupe, C., A. F. Fernanda Souza de Jesus, L. de Assis Pallos, J. 9673(99)00751-7.
Alberto Fracassi da Silva, and D. Pereira de Jesus. 2018. Capillary Lin, S. L., J. W. Hsu, and M. R. Fuh. 2019. Simultaneous determination
electrophoresis with capacitively coupled contactless conductivity of nitrate and nitrite in vegetables by poly (vinylimidazole-co-ethyl-
detection for the determination of propionate and sorbate in bread. ene dimethacrylate) monolithic capillary liquid chromatography
Journal of Separation Science 41 (20):3932–7. doi: 10.1002/jssc. with UV detection. Talanta 205:120082. doi: 10.1016/j.talanta.2019.
201800705. 06.082.
Heider, E. C., D. Valenti, R. L. Long, A. Garbou, M. Rex, and J. K. Lipskikh, O. I., E. I. Korotkova, Y. P. Khristunova, J. Barek, and B.
Harper. 2018. Quantifying sucralose in a water-treatment wetlands: Kratochvil. 2018. Sensors for voltammetric determination of food
Service-learning in the analytical chemistry laboratory. Journal of azo dyes-A critical review. Electrochimica Acta 260:974–85. doi: 10.
Chemical Education 95 (4):535–42. doi: 10.1021/acs.jchemed. 1016/j.electacta.2017.12.027.
7b00490. Liu, L., Z. Mi, H. Li, C. Li, Q. Hu, and F. Feng. 2019. Highly selective
He, Q., J. Liu, X. Liu, G. Li, P. Deng, and J. Liang. 2018b. Manganese and sensitive detection of amaranth by using carbon dots-based
dioxide Nanorods/electrochemically reduced graphene oxide nano- nanosensor. RSC Advances 9 (45):26315–20. doi: 10.1039/
composites modified electrodes for cost-effective and ultrasensitive C9RA04494A.
20 L. WU ET AL.
Liu, C. C., Y. N. Wang, L. M. Fu, and K. L. Chen. 2018. Microfluidic Papetti, A., and R. Colombo. 2019. High-performance capillary electro-
paper-based chip platform for benzoic acid detection in food. Food phoresis for food quality evaluation. In Evaluation technologies for
Chemistry 249:162–7. doi: 10.1016/j.foodchem.2018.01.004. food quality, 301–77. Cambridge, UK: Elsevier.
Liu, Y., X. Wu, Y. Tahara, H. Ikezaki, and K. Toko. 2020. A quantita- Parthasarathy, D. K., and N. S. Bryan. 2012. Sodium nitrite: The “cure”
tive method for Acesulfame K using the taste sensor. Sensors 20 (2): for nitric oxide insufficiency. Meat Science 92 (3):274–9. doi: 10.
400. doi: 10.3390/s20020400. 1016/j.meatsci.2012.03.001.
Li, Q., J. Yang, X. Tan, Z. Zhang, X. Hu, and M. Yang. 2016. A simple Pedroso, M. M., M. V. Foguel, D. H. S. Silva, M. D. P. T. Sotomayor,
and rapid resonance Rayleigh scattering method for detection of and H. Yamanaka. 2017. Electrochemical sensor for dodecyl gallate
indigo carmine in soft drink. Luminescence: The Journal of Biological determination based on electropolymerized molecularly imprinted
and Chemical Luminescence 31 (5):1152–7. doi: 10.1002/bio.3085. polymer. Sensors and Actuators B: Chemical 253:180–6. doi: 10.1016/
Li, L., H. Zheng, L. Guo, L. Qu, and L. Yu. 2019a. Construction of j.snb.2017.06.127.
novel electrochemical sensors based on bimetallic nanoparticle func- Piri, S., F. Piri, M. R. Yaftian, and A. Zamani. 2018. Imprinted
tionalized graphene for determination of sunset yellow in soft drink. Azorubine electrochemical sensor based upon composition of MnO2
Journal of Electroanalytical Chemistry 833:393–400. doi: 10.1016/j. and 1-naphthylamine on graphite nanopowder. Journal of the
jelechem.2018.11.059. Iranian Chemical Society 15 (12):2713–20. doi: 10.1007/s13738-018-
Li, L., H. Zheng, L. Guo, L. Qu, and L. Yu. 2019b. A sensitive and 1459-z.
selective molecularly imprinted electrochemical sensor based on Pd- Poms, R. E., C. L. Klein, and E. Anklam. 2004. Methods for allergen
Cu bimetallic alloy functionalized graphene for detection of amar- analysis in food: A review. Food Additives and Contaminants 21 (1):
anth in soft drink. Talanta 197:68–76. doi: 10.1016/j.talanta.2019.01. 1–31. doi: 10.1080/02652030310001620423.
009. Riman, D., M. I. Prodromidis, D. Jirovsky, and J. Hrbac. 2019. Low-
Lorenzo, J. M., M. Pateiro, R. Domınguez, F. J. Barba, P. Putnik, D. B. cost pencil graphite-based electrochemical detector for HPLC with
Kovacevic, A. Shpigelman, D. Granato, and D. Franco. 2018. Berries near-coulometric efficiency. Sensors and Actuators B: Chemical 296:
extracts as natural antioxidants in meat products: A review. Food 126618. doi: 10.1016/j.snb.2019.05.095.
Research International (Ottawa, ON) 106:1095–104. doi: 10.1016/j. Roca-Saavedra, P., V. Mendez-Vilabrille, J. M. Miranda, C. Nebot, A.
foodres.2017.12.005. Cardelle-Cobas, C. M. Franco, and A. Cepeda. 2018. Food additives,
Manjunatha, J. G. G. 2018. A novel poly (glycine) biosensor towards contaminants and other minor components: Effects on human gut
the detection of indigo carmine: A voltammetric study. Journal of microbiota-a review. Journal of Physiology and Biochemistry 74 (1):
Food and Drug Analysis 26 (1):292–9. doi: 10.1016/j.jfda.2017.05.002. 69–83. doi: 10.1007/s13105-017-0564-2.
Martyn, D., M. Darch, A. Roberts, H. Y. Lee, T. Yaqiong Tian, N. Rodero, A. B., L. de Souza Rodero, and R. Azoubel. 2009. Toxicity of
Kaburagi, and P. Belmar. 2018. Low-/no-calorie sweeteners: A Sucralose in Humans: A Review. International Journal of
review of global intakes. Nutrients 10 (3):357. doi: 10.3390/ Morphology 27 (1): 239–244. doi: 10.4067/S0717-
nu10030357. 95022009000100040.
McClements, D. J., and S. M. Jafari. 2018. Improving emulsion forma- Seyinde, D. O., I. P. Ejidike, and S. Ayejuyo. 2019. HPLC determin-
tion, stability and performance using mixed emulsifiers: A review. ation of benzoic acid, saccharin, and caffeine in carbonated soft
Advances in Colloid and Interface Science 251:55–79. doi: 10.1016/j. drinks. International Journal of ChemTech Research 12 (4):15–23.
cis.2017.12.001. doi: 10.20902/IJCTR.2019.120403.
Meher, A. K., N. Labhsetwar, and A. Bansiwal. 2018. An improved Sierra-Rosales, P., C. Toledo-Neira, and J. A. Squella. 2017.
method for direct estimation of free cyanide in drinking water by Electrochemical determination of food colorants in soft drinks using
ion chromatography-pulsed amperometry detection (IC-PAD) on MWCNT-modified GCEs. Sensors and Actuators B: Chemical 240:
gold working electrode. Food Chemistry 240:131–8. doi: 10.1016/j. 1257–64. doi: 10.1016/j.snb.2016.08.135.
foodchem.2017.07.041. Silva, T. A., A. Wong, and O. Fatibello-Filho. 2019. Electrochemical
Nair, M. S., D. V. Nair, A. K. Johny, and K. Venkitanarayanan. 2020. sensor based on ionic liquid and carbon black for voltammetric
Use of food preservatives and additives in meat and their detection determination of Allura red colorant at nanomolar levels in soft
techniques. In Meat quality analysis, 187–213. UK: Elsevier Science. drink powders. Talanta 209: 120588.
Nguyen, T. H., and A. L. Waterhouse. 2019. A production-accessible Sivasankaran, U., and K. G. Kumar. 2019. Communication—
method: Spectrophotometric iron speciation in wine using ferrozine Electrochemical sensing of synthetic antioxidant propyl gallate: A
and ethylenediaminetetraacetic acid. Journal of Agricultural and cost effective strategy using nanoparticles. Journal of the
Food Chemistry 67 (2):680–7. doi: 10.1021/acs.jafc.8b04497. Electrochemical Society 166 (2):B92–B94. doi: 10.1149/2.0791902jes.
Ntrallou, K., H. Gika, and E. Tsochatzis. 2020. Analytical and sample Sivasankaran, U., J. Radecki, H. Radecka, and K. Girish Kumar. 2019.
preparation techniques for the determination of food colorants in Copper nanoclusters: An efficient fluorescence sensing platform for
food matrices. Foods 9 (1):58. doi: 10.3390/foods9010058. quinoline yellow. Luminescence: The Journal of Biological and
O’Donnell, K. 2012. Aspartame, neotame and advantame. In Sweeteners Chemical Luminescence 34 (2):243–8. doi: 10.1002/bio.3601.
and Sugar Alternatives in Food Technology, 117–36. Sivasankaran, U., A. E. Vikraman, D. Thomas, and K. G. Kumar. 2016.
Oellig, C., K. Link, and W. Schwack. 2020. Characterization of E 472 Nanomolar level determination of octyl gallate in fats and oils. Food
food emulsifiers by high-performance thin-layer chromatography Analytical Methods 9 (7):2115–23. doi: 10.1007/s12161-015-0356-7.
with fluorescence detection and mass spectrometry. Journal of Squissato, A. L., E. M. Richter, and R. A. Munoz. 2019. Voltammetric
Chromatography A 1618:460874. doi: 10.1016/j.chroma.2020.460874. determination of copper and tert-butylhydroquinone in biodiesel: A
Oplatowska-Stachowiak, M, and C. T. Elliott. 2017. Food colors: rapid quality control protocol. Talanta 201:433–40. doi: 10.1016/j.
Existing and emerging food safety concerns. Critical Reviews in Food talanta.2019.04.030.
Science and Nutrition 57 (3):524–48. doi:10.1080/10408398.2014. Stefan-van Staden, R. I., A. Moscalu-Lungu, and J. F. van Staden. 2019.
889652. Pattern recognition of sweeteners in biological fluids, beverages, and
Ou, Y., X. Wang, K. Lai, Y. Huang, B. A. Rasco, and Y. Fan. 2018. ketchup using stochastic sensors. Electroanalysis 31:1–8.
Gold nanorods as surface-enhanced Raman spectroscopy substrates Sys, M., B. Svecova, I. Svancara, and R. Metelka. 2017. Determination
for rapid and sensitive analysis of allura red and sunset yellow in of vitamin E in margarines and edible oils using square wave anodic
beverages. Journal of Agricultural and Food Chemistry 66 (11): stripping voltammetry with a glassy carbon paste electrode. Food
2954–61. doi: 10.1021/acs.jafc.8b00007. Chemistry 229:621–7. doi: 10.1016/j.foodchem.2017.02.068.
€
Oztekin, N. 2018. Simultaneous determination of benzoic acid and sor- Taghdisi, S. M., N. M. Danesh, M. A. Nameghi, M. Ramezani, M.
bic acid in food products by capillary electrophoresis. Food and Alibolandi, and K. Abnous. 2019. An electrochemical sensing plat-
Health 4 (3):176–82. form based on ladder-shaped DNA structure and label-free aptamer
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION 21
for ultrasensitive detection of ampicillin. Biosensors & Bioelectronics determination of propyl gallate in vegetable oils. Food Analytical
133:230–5. doi: 10.1016/j.bios.2019.03.044. Methods 6 (3):775–80. doi: 10.1007/s12161-012-9485-4.
Teixeira, A. Z. A. 2018. Sodium content and food additives in major Voss, S., E. Newman, and J. P. Miller-Schulze. 2019. Quantification of
brands of Brazilian children’s foods. Ci^encia & Sa ude Coletiva 23 sucralose in groundwater well drinking water by silylation derivati-
(12):4065–75. doi: 10.1590/1413-812320182312.21812016. zation and gas chromatography-mass spectrometry. Analytical
Tekin Pulats€u, E., S. Sahin, and G. Sumnu. 2018. Characterization of Methods 11 (21):2790–9. doi: 10.1039/C9AY00442D.
different double-emulsion formulations based on food-grade emulsi- Walton, K., R. Walker, J. J. van de Sandt, J. V. Castell, A. G. Knapp, G.
fiers and stabilizers. Journal of Dispersion Science and Technology 39 Kozianowski, M. Roberfroid, and B. Schilter. 1999. The application
(7):996–1002. doi: 10.1080/01932691.2017.1379021. of in vitro data in the derivation of the acceptable daily intake of
Tezcan, F. 2018. A sample stacking-capillary electrophoresis method food additives. Food and Chemical Toxicology: An International
for simultaneous determination of nitrate and thiocyanate ions of Journal Published for the British Industrial Biological Research
ultra-heat-treated milk samples. Turkish Journal of Chemistry 42 (4): Association 37 (12):1175–97. doi: 10.1016/s0278-6915(99)00107-6.
1184–90. doi: 10.3906/kim-1801-15. Wiley, D., and C. N. Y. Nee. 2020. Food ingredients. In Food and
Timofeeva, I., D. Kanashina, K. Stepanova, and A. Bulatov. 2019. A Society, 377–91.Elsevier: UK.
simple and highly-available microextraction of benzoic and sorbic Williams, G. M., M. J. Iatropoulos, and J. Whysner. 1999. Safety assess-
acids in beverages and soy sauce samples for high performance ment of butylated hydroxyanisole and butylated hydroxytoluene as
liquid chromatography with ultraviolet detection. Journal of antioxidant food additives. Food and Chemical Toxicology: An
Chromatography. A 1588:1–7. doi: 10.1016/j.chroma.2018.12.030. International Journal Published for the British Industrial Biological
Tiwari, S., and M. K. Deb. 2019. Modified silver nanoparticles- Research Association 37 (9–10):1027–38. doi: 10.1016/S0278-
enhanced single drop microextraction of tartrazine in food samples 6915(99)00085-X.
coupled with diffuse reflectance Fourier transform infrared spectro- Wissgott, U., and K. Bortlik. 1996. Prospects for new natural food colo-
scopic analysis. Analytical Methods 11 (28):3552–62. doi: 10.1039/
rants. Trends in Food Science & Technology 7 (9):298–302. doi: 10.
C9AY00713J.
1016/0924-2244(96)20007-X.
Tran, Q. T., T. T. Phung, Q. T. Nguyen, T. G. Le, and C. Lagrost.
Wu, L., W. Yin, K. Tang, D. Li, K. Shao, Y. Zuo, J. Ma, J. Liu, and H.
2019. Highly sensitive and rapid determination of sunset yellow in
Han. 2016. Enzymatic biosensor of horseradish peroxidase immobi-
drinks using a low-cost carbon material-based electrochemical sen-
lized on Au-Pt nanotube/Au-graphene for the simultaneous deter-
sor. Analytical and Bioanalytical Chemistry 411 (28):7539–49. doi:
mination of antioxidants. Analytica Chimica Acta 933:89–96. doi:
10.1007/s00216-019-02155-9.
10.1016/j.aca.2016.06.020.
Urdea, M., L. A. Penny, S. S. Olmsted, M. Y. Giovanni, P. Kaspar, A.
Xie, Y., T. Chen, Y. Guo, Y. Cheng, H. Qian, and W. Yao. 2019. Rapid
Shepherd, P. Wilson, C. A. Dahl, S. Buchsbaum, G. Moeller, et al.
2006. Requirements for high impact diagnostics in the developing SERS detection of acid orange II and brilliant blue in food by using
world. Nature 444 (S1):73–9. doi: 10.1038/nature05448. Fe3O4@Au core–shell substrate. Food Chemistry 270:173–80. doi: 10.
Van der Heijden, C. A., P. J. C. M. Janssen, and J. J. T. W. A. Strik. 1016/j.foodchem.2018.07.065.
1986. Toxicology of gallates: A review and evaluation. Food and Yamjala, K., M. S. Nainar, and N. R. Ramisetti. 2016. Methods for the
Chemical Toxicology: An International Journal Published for the analysis of azo dyes employed in food industry-A review. Food
British Industrial Biological Research Association 24 (10–11):1067–70. Chemistry 192:813–24. doi: 10.1016/j.foodchem.2015.07.085.
doi: 10.1016/0278-6915(86)90290-5. Yao, Y., W. Wang, K. Tian, W. M. Ingram, J. Cheng, L. Qu, H. Li, and
Van Esch, G. J. 1986. Toxicology of tert-butylhydroquinone (TBHQ). C. Han. 2018. Highly reproducible and sensitive silver nanorod
Food and Chemical Toxicology: An International Journal Published array for the rapid detection of Allura Red in candy. Spectrochimica
for the British Industrial Biological Research Association 24 (10–11): Acta Part A: Molecular and Biomolecular Spectroscopy 195:165–71.
1063–5. doi: 10.1016/0278-6915(86)90289-9. doi: 10.1016/j.saa.2018.01.072.
Van Overmeire, I., K. Vrijens, T. Nawrot, A. Van Nieuwenhuyse, J. Yue, X., X. Luo, Z. Zhou, and Y. Bai. 2019. Selective electrochemical
Van Loco, and T. Reyns. 2019. Simultaneous determination of para- determination of tertiary butylhydroquinone in edible oils based on
bens, bisphenols and alkylphenols in human placenta by ultra-high an in-situ assembly molecularly imprinted polymer sensor. Food
performance liquid chromatography-tandem mass spectrometry. Chemistry 289:84–94. doi: 10.1016/j.foodchem.2019.03.044.
Journal of Chromatography. B, Analytical Technologies in the Zou, Q., Y. Yao, W. Wang, J. Li, C. Yan, and C. Han. 2018. Study on
Biomedical and Life Sciences 1121:96–102. doi: 10.1016/j.jchromb. detection of pigment Amaranth based on surface-enhanced Raman
2019.05.012. scattering. In 2018 Cross Strait Quad-Regional Radio Science and
Vikraman, A. E., Z. Rasheed, L. Rajith, L. A. Lonappan, and G. K. Wireless Technology Conference (CSQRWC), pp. 1–3. IEEE. doi:
Krishnapillai. 2013. MWCNT-modified gold electrode sensor for the 10.1109/CSQRWC.2018.8455384.