Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2024 Jan 4;9(2):2519–2527. doi: 10.1021/acsomega.3c07025

Colorimetric Detection of Furfural with Enhanced Visible Absorption of Furfural-DNPH in Basic Conditions

Hyunjoo Park †,‡,*, Eunyoung Kim †, Taehyun Jun †, Sang-Hyun Pyo §, Shin-Hyun Kim ‡,*
PMCID: PMC10795146  PMID: 38250383

Abstract

graphic file with name ao3c07025_0005.jpg

Furfural is an intermediary toxic aldehyde compound produced during heat-induced food processing and storage. Furfural is also formed by the degradation of cellulosic insulation in oil-immersed electric potential transformers, whose level is an important indicator of aging for replacement. In this study, we report a new means to detect the trace level of furfural in a colorimetric manner. Furfural is reacted with dinitrophenylhydrazine (DNPH) in acid solutions. The colorless furfural-DNPH compound turns orange-colored as the solution changes to basic. The delocalization of the π-electron in the DNPH-aldehyde derivatives at the basic condition causes the shift of the absorption peak from 318 to 470 nm, which renders the solution orange-colored. The color and absorbance are saturated in 20 min of incubation. There is high linearity between the absorbance and the concentration of furfural in the range of 0–0.2 mM, which enables the quantitative detection of furfural. The limit of detection is estimated to be as low as 1.76 μM for the absorbance analysis and 10 μM for the naked eyes. The colorimetric assay protocol is applicable to the detection of various aromatic aldehydes, which show strong π-electron delocalization and is not applicable to aliphatic aldehydes due to lack of delocalization. This simple assay can be conducted in typical 96-well microplates using a microplate reader, which provides a low-cost and high-throughput screening. Therefore, we believe that our method is potentially applicable for the quantitative detection of aromatic aldehydes in various samples from foods, electronic devices, and so on.

Introduction

Furfurals such as hydroxymethylfurfural (HMF) and furfural (FF) are intermediary aldehyde compounds that occur during the formation of pigments (melanoidins), which are closely linked to mechanisms such as caramelization and Maillard reactions.1−4 Furfurals are generated mainly by the 1,2-enolization pathway via 3-deoxyosone,4,5 where the degradation of hexoses, through direct enolization in the Maillard reaction, is the initial step.4,6 These compounds are commonly produced by heat-induced food processing and storage and are present in trace levels in foods depending on the processing condition and storage time. Subsequently, they can serve as indicators of the extent of the Maillard reaction and markers of heating processes in many products that contain sugars.2−4,7 Above a certain limit, when adsorbed in the body, these compounds have negative impacts on human health, affecting the central nervous system, kidneys, liver, and other organs.8 This makes it important to develop a simple, fast, and sensitive method that can determine their concentration in various foods, such as baked goods, milk, and beers, and to investigate the relationship between the degree of heat treatment and the content of furfural compounds.1−4 Furfural analysis is also a useful tool for assessing the aging of solid insulation in oil-immersed transformers since furanic compounds found in transformers are solely formed by the degradation of cellulosic insulation.9

Furfurals are usually detected using traditional chromatographic methods such as gas chromatography (GC), high-performance liquid chromatography (HPLC), and spectrophotometric methods.1−4 However, these methods are often hindered by matrix effects and insufficient selectivity and often require several analytical steps including extraction and derivatization before any analysis can be conducted.3 It is known that the detection limits can be improved using precolumn derivatization reactions with various compounds, such as dinitrophenylhydrazine (DNPH), which is the most popular derivatization reagent for the HPLC-UV analysis of aldehydes.1,3,10 However, even though these methods can provide accurate and reproducible results, their applications are still limited by several shortcomings, such as a time-consuming sample preparation protocol and the use of expensive instrumentation.10

Meanwhile, colorimetric methods for the in situ detection of aldehydes have been developed with many important applications, including for chemical toxin detection, security screening, food inspection, and disease monitoring, serving as alternatives to traditional analytical methods for rapid and inexpensive determination.11−15,22−24 Various methods have been reported to enhance the sensitivity and selectivity of aliphatic and aromatic aldehyde detection, such as fluorescent supramolecular polymers, chemiluminescence (CL) detection of aldehydes derivatized with 2,4-dinitrophenylhydrazine (DNPH), MnO2 nanosheets, and supramolecular complex.16−20 Generally, the colorimetric detection of aldehydes is based on a nucleophilic addition to a carbonyl group by, for example, an amine such as aniline, DNPH, or 4-amino-3-hydrazino-5-mercapto-1,2,4-triazole (Purpald), which results in the formation of an imine, which gives a different UV–vis absorption band. However, these methods for the detection of aldehydes are not applicable to furfurals due to the absence of imine in the product. For colorimetric detection of furfurals, different strategies have been developed. For example, polymeric films have been prepared by the radical polymerization of 4-vinylaniline, 2-hydroxymethyl methacrylate, and ethylene dimethyl methacrylate, of which aniline groups react with an aldehyde of furfural in acidic media through the Stenhouse reaction, generating a deep-red cyanine derivative.1 Based on this mechanism, this may be limited only to the detection of furfural, which can form conjugation through the resulting product from furfural and aniline (1:2 equiv), and is therefore not applicable to HMF and other aldehydes. Based on the Winkler method, a solution containing p-aminobenzoic acid, barbituric acid, and hydrochloric acid is used to detect furfural and HMF in sugarcane liquor using a diffuse reflectance technique coupled with limited-area spot-testing on a paper platform.7 Recently, a Seliwanoff reagent (resorcinol-HCl) has been employed for reaction with HMF in honey, which yields a red-color xanthenoid complex through the condensation reaction of two equivalents of resorcinol and HMF.21 However, these methods are only applicable to specific aldehydes of furfural or HMF, and the response time for the color change is as long as approximately 1 h. Therefore, it is important to develop a colorimetric chemosensor for the simple, fast, and quantitative detection of various aromatic aldehydes.

In this work, we propose a furfural-DNPH colorimetric chemosensory system based on the coupling reaction of furfural and DNPH. Although the furfural-DNPH derivative itself forms a fully conjugated structure with no color at acidic conditions, it turns orange in basic conditions due to π-electron delocalization. This extended conjugation often results in the absorption of visible light, leading to a color change. In the case of the furfural-DNPH derivative, the extended conjugation likely imparts an orange color to the solution, and in a basic medium, DNPH may undergo deprotonation and the resulting species may have a different color compared to the protonated form. This response amplification in a basic aqueous medium is inspired by the colorimetric detection of α-ketoglutarate with DNPH by forming DNP-hydrazone.22,23 Our method provides a relatively short response time for coloration. Moreover, high linearity between the absorbance and the concentration of furfural enables quantitative detection. With the naked eyes, the color change is recognizable for the concentration of 10 μM and the limit of detection is approximately 1.76 μM for the spectral analysis. Our method provides colorimetric response for not only furfural but also various aromatic aldehydes, which can provide strong π-electron delocalization in the final products. Importantly, our simple and fast analysis protocol is facile for a low-cost and high-throughput screening through the use of conventional microwell plates and a microplate reader. It has been proven that our method is directly applicable to the detection of furfural in beers. Therefore, we believe that our method is potentially applicable to the quantitative detection of furfural, HMF, and aromatic aldehydes in various samples from foods, electrical equipment, and so on within a short time.

Experimental Section

Materials

Ultrapure water was obtained from a Milli-Q water purification system (Millipore). Furfural (99%), 5-hydroxymethylfurfural (99%), benzaldehyde (99%), 4-hydroxy-benzaldehyde (99%), vanillin (99%), acetonitrile (≥99%), butyraldehyde (98%), pentaldehyde (98%), methanol, hydrochloric acid (HCl, 35%), 2,4-dinitrophenylhydrazine (DNPH, 97%), sodium hydroxide (97%), 4-amino-3-hydrazino-5-mercapto-1,2,4-triazole (Purpald, 99%), sodium periodate (98%), formaldehyde (37 wt % in H2O purity), deuterium oxide (D2O, 99.9 atom %), dimethyl sulfoxide-d6 (99.8 atom %), and chloroform-d (99.8 atom %) were provided by Sigma-Aldrich (St. Louis, MO). All chemicals were used without further treatment.

Colorimetric Reaction of Formaldehyde and Furfural with Purpald

The Purpald colorimetry assay was performed following a modified method from previous reports.24,25 10 mM Purpald and 2 mM NaIO4 solutions in 2 N NaOH and 0.2 N NaOH, respectively, were freshly prepared. Furfural solutions of 0, 1 (0.01 mM), 5 (0.05 mM), 10 (0.1 mM), 20 (0.21 mM), and 100 ppm (1.0 mM) and formaldehyde at 0, 1 (0.03 mM), 5 (0.17 mM), 10 (0.33 mM), 20 (0.67 mM), and 100 ppm (3.3 mM) were prepared, respectively. For colorimetric analysis, 100 μL of aldehydes and 50 μL of Purpald solution were placed and mixed in the wells of 96-well microplates at room temperature. The reactions and color development were monitored and compared at 0, 1, 5, and 10 min. After 10 min, 50 μL of 2 mM NaIO4 was added to each well.

Colorimetric Reaction of Formaldehyde and Furfural with DNPH

The DNPH colorimetry assay was performed following a modified method from previous reports.22,23 0.5 mM DNPH solution in a mixture of acetonitrile and 2 N HCl (1/1) was freshly prepared. Furfural solutions of 0, 1 (0.01 mM), 5 (0.05 mM), 10 (0.1 mM), 20 (0.2 mM), and 100 ppm (1.0 mM) and formaldehyde at 0, 1 (0.03 mM), 5 (0.17 mM), 10 (0.33 mM), 20 (0.67 mM), and 100 ppm (3.3 mM) were prepared, respectively. For colorimetric analysis, 100 μL of aldehydes and 50 μL of DNPH solution were placed and mixed in the wells of 96-well microplates at room temperature. The reactions and color development were monitored and compared at 0, 1, and 5 min. After 5 min reaction, 50 μL of 5 N NaOH was added to each well, and color development was monitored and analyzed in the absorbance range of 250–600 nm using a UV–vis microplate spectrophotometer (Multiscan GO, Thermo Scientific).

Amplified Colorimetric and Spectroscopic Analyses by UV–Vis Spectroscopy

In the DNPH colorimetry assay, the color development and absorbance shift for the reaction of furfural and DNPH in an acid and base were monitored at an absorbance range of 250–600 nm using a UV–vis spectrometer (UV–vis–NIR, PerkinElmer, Lambda 1050). A 0.5 mM DNPH solution in a mixture of acetonitrile and 1 N HCl (1/1) was freshly prepared. Furfural solutions of 0, 0.05 (0.0005 mM), 0.1 (0.001 mM), 0.5 (0.005 mM), 1 ppm (0.01 mM), 5 (0.05 mM), 10 (0.1 mM), and 20 (0.21 mM) were prepared in 0.5 N hydrochloric acid. 300 μL of furfural solution and 150 μL of 0.5 mM DNPH solution were placed and mixed in a UV quartz cuvette (path length 5 × 5 mm2). After reaction for a given time, the samples were mixed by adding 100 μL of 5 N NaOH. Absorbance spectra of the reactions of furfural and DNPH at different reaction times were obtained in the absorbance range of 250–600 nm using a UV–vis spectrometer. The methods were employed for aliphatic aldehydes (formaldehyde, butanal, and pentanal) and aromatic aldehydes (4-hydroxy-benzylaldehyde, vanillin, benzylaldehyde, 5-hydroxymethylfurfural, and furfural).

The effects of these aldehydes’ concentrations were simultaneously examined in the 96-well microplate using a microplate reader (Multiskan Go, Thermo Scientific). The 150 μL aldehyde solutions at different concentrations were placed and mixed with a 75 μL 5 mM DNPH solution in the 96-well microplate, followed in 10 min by the addition of 50 μL of 5 N NaOH. All of the data were obtained from two independent experiments, and the average of the replicates is provided.

Spectral characterization of colorimetric response amplification, caused by π-electron delocalization in a basic solution, was carried out. The formation of furfural-DNP hydrazone in an acid and a new conjugated form after the addition of 5 N NaOH was elucidated by 1H NMR using 600 MHz NMR (a Bruker AVANCE NEO 600, Bruker Biospin, Germany) at 25 °C. All of the chemical shifts are reported in ppm (δ) relative to the compound tetramethylsilane and referenced to the chemical shifts of the residual solvent resonances (1H). The furfural-DNP hydrazine solid was precipitated from the reaction solution by the addition of deionized water, followed by simple filtration and washing using deionized water. The resulting precipitate was dissolved in DMSO-d6 and measured by 1H NMR. Then, the previous acidic solution was added, with 5 N NaOH prepared in D2O, and measured by 1H NMR.

Results and Discussion

Colorimetric Responses for the Reaction of Furfural with DNPH

Purpald has been used as one of the most popular reagents for the colorimetric determination of aldehyde, which is stable at room temperature and more sensitive than other methods, such as those using acetyl-acetone or chromotropic acid.25,26 Purpald combines with aldehydes such as formaldehyde in an alkaline solution to form a colorless intermediate that is oxidized by ambient oxygen to form an intensely purple tetrazine, which serves as the colorimetric product. Commercially available formaldehyde test kits based on Purpald chemistry are extremely sensitive. We studied the effectiveness of Purpald for the detection of furfural in comparison with the detection of formaldehyde (Figure S1). No color development was observed for furfural in the Purpald-based colorimetric assay, while strong purple colors were developed for formaldehyde, even though potassium periodate was added as an oxidant to the reaction solution to oxidize the Purpald-aldehyde adduct to form chromogens. This may be because the Purpald-furfural adduct lacks resonant structures for the delocalization of electrons, while Purpald-formaldehyde possesses them (Figure S1A). That is, Purpald is inappropriate and other reagents are required for the colorimetric detection of furfural.

DNPH-derivatization has been used to increase the response of UV absorption in aldehyde detection and coupled with HPLC and GC. The addition of 2,4-DNPH results in a condensation reaction between the aldehyde carbonyl group (furfural and formaldehyde) and 2,4-DNPH in acidic conditions to form aldehyde-DNP-hydrazone through a well-known mechanism (Scheme S1). To study the colorimetric response of the DNPH-derivatization, we mixed the solution of either furfural or formaldehyde in 0.5 N HCl (300 μL) and 0.5 mM DNPH in a mixture (150 μL) of acetonitrile and 2 N HCl (final 1 N, 1/1), where the concentration of furfural or formaldehyde was varied in the range of 0–1 or 0–3.3 mM, respectively. There was no significant change in color for both furfural and formaldehyde for the entire range of concentration, which remained similar to the pale-yellow color of the pure DNPH solution used (top rows of Figure 1A,B). Also, the condensation reaction causes no meaningful change in absorbance (Figure S2); there was a slight increase at 270 nm and a slight decrease at 325 nm from DNPH, while the spectrum band was gradually broadened in the range of 375–450 nm during the reaction, which is not dramatic at all and insufficient for colorimetric detection. Meanwhile, we were motivated by the derivatization of α-keto acid with DNPH in an acidic condition, followed by a pH change to a basic condition to allow color development.22,23 When the pH was increased by adding 5 N NaOH (100 μL) in the acidic solution of the furfural-DNPH adduct, an orange color developed (bottom row of Figure 1A). The color change in the basic condition is attributed to the increased redistribution of the pi-electron over two nitrogen groups of four resonant structures (Figure 1C). This new observation enables the colorimetric and spectral analyses of aldehydes, especially furfural, through the typical DNPH-derivatization of aldehydes. A weak color change is observed for formaldehyde (bottom row of Figure 1B). The basic condition may induce deprotonation of acidic protons on the hydrazone nitrogen atoms, leading to the formation of an anion. The deprotonation can affect the distribution of electrons within the formaldehyde-DNPH derivative, influencing the conjugation and UV absorption spectrum (Figure S10). The color changes from colorless under acidic conditions to a pale-yellow color under basic conditions for the formaldehyde-DNPH adduct without significant color amplification.

Figure 1.

Figure 1

Open in a new tab

(A, B) Photographs showing colorimetric response to the reaction of furfural (A) and formaldehyde (B) at the denoted concentrations with DNPH in an acidic condition (top row) and basic condition (bottom row). (C) Reaction schemes for furfural-DNPH and formaldehyde-DNPH in acidic conditions and their possible electron delocalization in basic conditions. (D) Absorbance spectra of furfural-DNPH in an acidic solution for 10 min reaction and in a basic solution for four different incubation times as denoted, where the reaction occurs by mixing 0.05 mM furfural solution and 0.5 mM DNPH solution in the volume ratio of 2:1. There is no spectral change during the incubation in a basic condition for 10 min. (E) Absorbance spectra of the furfural-DNPH in an acidic solution (10 min) and a basic solution after reaction during a 0–20 min time course in acidic conditions. (F) Reaction time-dependent change of the absorbance at 465 nm, where the solid line is a fit with an exponential function.

To investigate the origin of the color change and kinetics of the reaction, the absorbance spectra were measured. For the concentration of 0.05 mM furfural, the furfural-DNPH derivative was produced in acidic conditions, which revealed no absorption peak in the visible range for wavelengths larger than 400 nm (Figure 1D). The absorption spectrum of the furfural-DNPH adduct exhibited a profile similar to that of pure DNPH (Figure S3). Upon addition of NaOH, the absorption peak is observed at 465 nm, which develops an orange color. The spectral change instantly occurs upon the increase of pH, and no more change is observed during incubation for 10 min. It is confirmed that the strong absorption near 465 nm originates from the furfural-DNPH derivative, neither furfural nor DNPH (Figure S3).

As the addition of the NaOH solution quenches the reaction between furfural and DNPH, the time-dependent extent of the reaction can be characterized by the absorption spectra. As the reaction time before the quenching increases from 0 to 20 min, so does the absorbance at 465 nm (Figure 1E); for the 0 min reaction, only DNPH was used without furfural to exclude any reaction. The absorbance rapidly increased in the initial 5 min and then gradually saturated. The temporal change of absorbance, A(t), follows the exponential function (Figure 1F)

graphic file with name ao3c07025_m001.jpg 1

where A0 and A∞ are the initial and equilibrium absorbances, respectively, and τ is the characteristic time scale. The value of τ is estimated as 8.23 min so that the reaction reaches equilibrium at 3τ = 24.7 min. For the on-site colorimetric analysis, we set the reaction time as 10 min for the further analysis of furfural, which provides consistent quantitative data. It seems that the reaction is first order on the concentration of furfural from the exponential change as the concentration of DNPH was set to be 1 order higher.

Sensitivity of Detection

To evaluate the sensitivity of detection, the correlationship between the furfural concentration, cfurfural, and absorbance, A, was investigated. The concentration of the furfural was varied in the range of 0.0005–0.2 mM, whereas that of DNPH was fixed at 0.5 mM. For the maximum concentration of furfural, 0.2 mM, the molar ratio of DNPH to furfural is 1.25 for the mixing ratio of two solutions in 2:1, which guarantees a sufficient amount of DNPH to produce the furfural-DNPH adduct. Also, the constant concentration of DNPH allowed the comparison of absorbance at 465 nm without hindrances due to DNPH intensity or baseline changes. For the entire range of concentrations, there was no phase separation or precipitation.

Under these optimal conditions, absorbance spectra were obtained for the full range of furfural concentrations (Figure 2A). The spectra show the main peak at 465 nm, which becomes more pronounced along with the concentration of furfural. Also, the spectra reveal a low shoulder peak at 550 nm. The absorbance shows a very high linearity with the concentration of furfural at both 465 and 550 nm in two different concentration ranges: (1) 0 mM ≤ cfurfural ≤ 0.01 mM and (2) 0.01 mM < cfurfural ≤ 0.2 mM (Figure 2B). Therefore, the concentration of furfural can be quantitively analyzed from the measurement of absorbance. The linear correlationship for the shoulder peak at 550 nm is complementary to that for the main peak at 465 nm for the quantification.

Figure 2.

Figure 2

Open in a new tab

(A) Absorbance spectra of the reaction solutions of furfural and DNPH at 10 min at the acidic condition, followed by a change to the basic condition, where the concentration of furfural was adjusted in the ranges of 0 mM ≤ cfurfural ≤ 0.01 mM (the left panel) and 0.01 mM < cfurfural ≤ 0.2 mM (the right panel). (B) Linear correlationships between absorbances at 465 and 550 nm and furfural concentration in the two ranges.

The limit of detection (LOD) was calculated as 3.3σ/S = 0.00176 mM (=0.169 ppm) from the standard deviation σ and slope S of the linear fits for absorbance at 465 nm in the low range of concentration of 0 mM ≤ cfurfural ≤ 0.01 mM.27 The limit of quantification (LOQ) was estimated as 10σ/S = 0.00535 mM (=0.514 ppm). These values of LOD and LOQ are low enough for the detection of furfural in foods, including fruits and vegetables; the typical concentration of furfural is 0.8–14 ppm in wheat bread, 0.6–33 ppm in cognac, 22 ppm in rum, 10–37 ppm in malt whisky, 2–34 ppm in port wine, 55–255 ppm in coffee, and 0.01–4.93 ppm in juices.28 With naked eyes, the color change, distinguished from 0 mM furfural, is identifiable for 0.01 mM (=0.961 ppm) (Figure S4), which is also valuable for simple equipment-free on-site detection.

Origin of Amplification of Colorimetric Response in Basic Conditions

The condensation of furfural aldehyde with DNPH in acidic conditions results in the formation of the furfural-DNP hydrazine with a pale-yellow color (Figure 1C). The hydrazone structure (N–N bond) in the molecular structure does not allow the flow of pi-electrons. Since the N–N bond is a σ bond, pi-electron movement through the furan-DNPH adduct is not free; thus, the electrons are localized. As a consequence, the HOMO and LUMO energy levels of furfural and DNPH are maintained, and the energy in the visible region is not absorbed. Therefore, there is no change in color and absorption wavelength shift.

Following the pH change with the addition of the NaOH solution, a rapid color change and amplification were observed. When hydrogen (H+) in the hydrazone is removed by the base (−OH), an electron pair is formed, which can move through the pi bond and the unshared electron pair, thus eventually leading to the formation of four resonance structures (Figure 3A). Subsequently, the electron delocalization occurs in a large area over the nitrogen groups and aromatic furan rings of the furfural-DNP hydrazone structure in the basic condition. With an increasing degree of conjugation, the energy gap decreases by elevating the energy level of HOMO and lowering the energy level of LUMO. Therefore, electron transfer is caused by low energy.

Figure 3.

Figure 3

Open in a new tab

(A) Chemical structures of furfural-DNP hydrazine expected in acidic and basic conditions. (B) 1H NMR spectrum of the furfural-DNPH derivatives in DMSO-d6 (black line, FF-DNPH), and after the addition of 5 N NaOH prepared in D2O (red line, FF-DNPH in the base).

The change in the chemical structure of the furfural-DNP hydrazine accompanying the change from acidic to basic conditions was elucidated by 1H NMR (Figure 3B). The furfural-DNP hydrazine solid was recovered from the reaction solution, which was then dissolved in DMSO-d6 and measured by 1H NMR to study the chemical structure at acidic conditions. Meanwhile, the acidic solution was added with 5 N NaOH prepared in D2O to study the chemical structure at the basic condition with 1H NMR. As expected, the pale-yellow color immediately changed and became an orange color. Figures S5–S8 show the 1H NMR spectra of furfural, DNPH, and furfural-DNPH derivatives individually, allowing for comparison. Furthermore, these observations provide the positive indication that the method is specifically detecting furfural, dismissing the possibility of humin formation during the reaction (Figures S7, S8).

The observed alterations in chemical shifts across varying pH levels confirmed the formation of conjugated bonds between furfural and DNPH, a process driven by pi-electron delocalization (Figure 3). All protons experience diamagnetic shielding from nearby electrons when subjected to an external magnetic field. These electrons produce a localized magnetic field opposite to the applied field, and this is directly related to the electron density surrounding the proton. In the presence of a strong base, the electron density around the protons in the furfural-DNPH adducts increases. This intensifies the shielding effect of the atom to which a proton is attached due to the pi-electron delocalization. As a result, a stronger external field is needed to achieve resonance in NMR. This indicates that in a basic medium, the chemical shifts of all protons were detected at higher fields.

Colorimetric Detection of Aromatic Aldehydes

The protocol developed for the detection of furfural involves the amplification of colorimetric chemosensory response and the shift in absorption wavelength by pi-electron delocalization in the presence of a strong base. That is, the pi-electron delocalization should be stable over the two nitrogen groups in the aldehyde-DNPH adduct molecule. As seen in the possible electron delocalization (Figure 1C), the furfural-DNPH adduct could be stabilized by the furan of furfural in basic conditions, while it may be limited for formaldehyde. It is therefore anticipated that our protocol is applicable to aromatic aldehydes for the amplification of colorimetric chemosensory response while being inappropriate for aliphatic aldehydes. To compare the absorption responses and spectral profiles, aliphatic and aromatic aldehydes were employed at the same concentration of 0.05 mM, and the same assay protocol was followed. For the aliphatic aldehydes, formaldehyde, butanal, and pentanal were used, whose aldehyde-DNPH adducts in the basic condition showed absorbance spectra similar to that for DNPH without any adducts (Figure 4A). There is no absorption band in the range of 450–600 nm, which is completely different from the furfural-DNPH adduct. Meanwhile, furfural, 5-hydroxymethylfurfural (5-HMF), benzaldehyde, 4-hydroxy-benzaldehyde, and vanillin were used for aromatic aldehydes (Figure 4B). All of the adducts of the aromatic aldehyde-DNPH in the basic condition showed spectral bands in the range of 450–600 nm. The absorbance was in the order of 4-hydroxy-benzaldehyde > vanillin > benzaldehyde >5-HMF > furfural, which indicates that benzene-based aldehydes have a higher absorption intensity than the furan-based ones. The higher absorption intensity of 4-hydroxy-benzylaldehyde and vanillin than that of benzaldehyde is attributed to the pi-electron delocalization through the hydroxyl group in the aromatic ring. There was a small variation in the wavelength of the absorption peak in the range of 458–480 nm (Figure 4B and Table 1).

Figure 4.

Figure 4

Open in a new tab

(A) Molecular structures of aliphatic aldehydes (formaldehyde, butanal, and pentanal) and absorbance spectra of the adducts of the aliphatic aldehyde-DNPH in the basic condition. The absorbance spectrum for DNPH in the basic condition is included for comparison. (B) Molecular structures of aromatic aldehydes (furfural, 5-hydroxymethylfurfural (5-HMF), benzaldehyde, 4-hydroxy-benzaldehyde (4-H-benzaldehyde), and vanillin) and absorbance spectra of the adducts of the aromatic aldehyde-DNPH in the basic condition. (C) Photograph of a 96-well microplate showing color development for eight aromatic and aliphatic aldehydes employed in the assay, where the concentration of the aldehyde was varied in the range of 0.0005–0.2 mM. (D) Linear correlationship between aldehyde concentration and absorbance at 470 nm for five aromatic aldehydes, where absorbance was measured using a microreader after incubation for 5 min in the basic condition.

Table 1. Summary of Assay Profile and Results of Aliphatic and Aromatic Aldehydes.

    Absb for 0.01 mM
 Absb for 0.1 mM
 regressionb
aldehyde λmaxa 5 min 10 h 5 min 10 h 5 min 10 h
furfural 465 0.2155 0.1758 0.5960 0.4707 y = 3.8737x + 0.1825 y = 3.044x + 0.1457
R2 = 0.9967 R2 = 0.9964
5-HMF 469 0.2320 0.1963 0.7482 0.6254 y = 4.9482x + 0.1908 y = 4.1161x + 0.1549
R2 = 0.9914 R2 = 0.9872
benzaldehyde 458 0.2192 0.1741 0.6080 0.2264 y = 3.4933x + 0.1984 y = 0.5883x + 0.1606
R2 = 0.9759 R2 = 0.7747
4-hydroxy benzaldehyde 477 0.2426 0.2116 1.1756 1.0839 y = 9.2215x + 0.1773 y = 8.5045x + 0.1488
R2 = 0.9964 R2 = 0.9952
vanillin 480 0.2464 0.2089 1.1117 1.0473 y = 8.8441x + 0.1759 y = 8.3574x + 0.1437
R2 = 0.9985 R2 = 0.9972
formaldehyde N.D.c 0.1898 0.1461 0.2392 0.1955 y = 0.5752x + 0.1798 y = 0.479x + 0.1401
R2 = 0.9941 R2 = 0.9485
butanal N.D.c 0.1891 0.1407 0.2165 0.1584 y = 0.4575x + 0.1778 y = 0.2178x + 0.136
R2 = 0.9663 R2 = 0.9689
pentanal N.D.c 0.1852 0.1363 0.2585 0.1664 y = 0.6911x + 0.1774 y = 0.3027x + 0.1339
R2 = 0.9794 R2 = 0.9733
Open in a new tab
a

Measured in a 5 × 5 mm2 UV quartz cuvette.

b

Measured at 470 nm in a 96-well microplate at 5 min or 10 h after addition of 5 N NaOH.

c

Not detected.

The assay protocol can be successfully performed in a 96-well microplate instead of a quartz UV cuvette (Figure 4C). The color change is discernible with naked eyes at the concentration of 0.01 mM or higher for all five aromatic aldehydes. As the average wavelength for the absorption peaks is 470 nm for aromatic aldehydes, the absorption intensity at 470 nm was used for the analysis of the microplate with a Microplate Reader. Linearity between the concentration and absorbance at 470 nm was conserved for all aromatic aldehydes in the range of 0.0005–0.2 mM (Figure 4D and Table 1). This verifies that the assay method could be employed for the high-throughput screening and analysis of aromatic aldehydes, while it would not be suitable for aliphatic ones.

The furfural-DNPH adduct is stable in the basic condition for 10 min as confirmed (Figure 1D). However, it turns out that the adducts are unstable for long-term incubation in the basic condition. The absorbance at 470 nm decreased after the incubation of 10 h at the concentrations of both 0.01 mM and 0.1 mM for all of the aromatic aldehydes (Table 1). The rate of the absorbance decrease depended on the aldehyde. 4-Hydroxy-benzaldehyde and vanillin showed a relatively slow decrease, whereas benzaldehyde exhibited a fast decrease from 0.6080 to 0.2264 for the concentration of 0.1 mM. This might be caused by the reverse reaction to make a more stable form in the basic condition. Nevertheless, the assay protocol guarantees a stable signal at least for 10 min so that it may be sufficiently valid for the high-throughput screening and analysis of aromatic aldehydes within a short analysis time course. While our design allows for the quantitative analysis of furfural alone, it is important to note that applying this method to complex matrices requires careful consideration and validation. Furfural is rarely formed with other types of aldehydes such as aromatics, one at a time. Additionally, our method demonstrates selectivity, making it applicable only to aromatic aldehydes and excluding aliphatic aldehydes. In experiments with mixed solutions of furfural and HMF, we observed a proportional increase in their concentrations (Figure S9). Conversely, when furfural and formaldehyde were mixed (Figure S10), the DNPH peak interrupted the maximum wavelength of formaldehyde (400 nm). Therefore, we anticipate that furfural can still be detectable in a mixed matrix of furfural and aliphatic aldehydes.

This colorimetric analysis protocol was applied to detect furfural in beer. The fresh beer was kept at 60 °C for 1 week for artificial aging, and the furfural concentration was evaluated for the fresh and aged beer using HPLC29 (Table S1); the furfural concentration is increased from 0.82 to 10.32 μM by aging. Both the fresh and aged beer were subjected to a colorimetric analysis protocol (Figure S11). After adding 0.5 mM DNPH at the acidic condition, the fresh and aged beers were incubated for 5 min and the pHs of the media were increased to the basic condition. Both the fresh and aged beers showed color change due to the presence of furfural. However, the aged beer showed a much stronger color than the fresh one as the furfural concentration was 12.6 times larger. In order to verify the practicability of this methodology, it was applied to the detection of furfural in power transformer insulating fluids (Figure S12). As a result, a high correlationship of absorbance was observed at 465 nm from the model solution in power transformer insulating fluids. The potential for quantitative detection extends to various practical applications where furfural is formed.

Conclusions

We have developed a new colorimetric chemosensory system for the quantification of furfural and other aromatic aldehydes based on a coupling reaction with DNPH, followed by signal and color amplification. The colorimetric response was significantly amplified by the pi-electron delocalization of the DNPH-aldehyde adduct, induced by changing the pH to a basic condition. This is the first DNPH-based colorimetric furfural detection with high sensitivity amplified by pi-electron delocalization. There was a very good linear relationship between absorbance and furfural concentration, which enables the quantitative analysis of furfural. The LOD for furfural is as low as 1.7 μM for the absorbance analysis, and the color change is discernible for the concentration of 10 μM with the naked eyes. The assay is applicable to the colorimetric and spectroscopic analyses of various aromatic aldehydes, which provide pi-electron delocalization in a basic condition with the help of an aromatic ring. The simple and facile protocol was successfully conducted using a 96-well plastic microplate and microplate reader, securing the reliability of the analysis for a small volume of samples. Therefore, this work stands out due to its significant advantages and superiority in the field of furfural detection. The incorporation of pi-electron delocalization adds a distinctive feature to the methodology, providing enhanced sensitivity and selectivity for furfural over other aldehydes, particularly aliphatic aldehydes. The protocol’s ability to achieve high-throughput screening and facile detection, especially in transformer oil, showcases its practical applicability. This makes the protocol well-suited for the quantitative detection of aromatic aldehydes like furfural in various samples, including those from foods and electronic devices.

Acknowledgments

This work was financed by the Research Fund for the Korea Electric Power Research Institute.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.3c07025.

  • Reaction mechanism of DNPH and furfural; reaction schemes and colorimetric responses of furfural and formaldehyde with Purpald; time-dependent absorbance spectra and colorimetric response of the mixture of furfural solution and DNPH solution; absorbance spectra of furfural-DNPH in an acid and base; colorimetric responses to the reaction of furfural with DNPH; 1H NMR spectrum of furfural, DNPH, furfural-DNPH derivatives, and furfural-DNPH derivatives in the basic condition; absorbance spectra of furfural and HMF mixed solution and HMF and linear correlationships between maximum absorbance; absorbance spectra of furfural and formaldehyde mixed solution and formaldehyde and linear correlationships between maximum absorbance; series of photographs of fresh and aged beer samples and furfural content of fresh beer and artificially aged beer; and UV–vis spectrum and color development of furfural-DNPH reaction solutions through in situ extraction from model furfural in power transformer insulating fluid (PDF)

Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.

Supplementary Material

ao3c07025_si_001.pdf (737.1KB, pdf)

References

  1. Rico-Yuste A.; González-Vallejo V.; Benito-Peña E.; De las Casas Engel T.; Orellana G.; Moreno-Bondi M. C. Furfural Determination With Disposable Polymer Films and Smartphone-based Colorimetry for Beer Freshness Assessment. Anal. Chem. 2016, 88 (7), 3959–3966. 10.1021/acs.analchem.6b00167. [DOI] [PubMed] [Google Scholar]
  2. Cepeda-Vázquez M.; Blumenthal D.; Camel V.; Rega B. Multivariate Optimization of Headspace Trap for Furan and Furfural Simultaneous Determination in Sponge Cake. Talanta 2017, 164, 708–715. 10.1016/j.talanta.2016.10.073. [DOI] [PubMed] [Google Scholar]
  3. Xing Q.; Ma Y.; Fu X.; Cao Q.; Zhang Y.; You C. J. Effects of Heat Treatment, Homogenization Pressure, and Overprocessing on the Content of Furfural Compounds in Liquid Milk. Sci. Food. Agric. 2020, 100 (14), 5276–5282. 10.1002/jsfa.10578. [DOI] [PubMed] [Google Scholar]
  4. Srivastava R.; Bousquières J.; Cepeda-Vázquez M.; Roux S.; Bonazzi C.; Rega B. Kinetic Study of Furan and Furfural Generation during Baking of Cake Models. Food. Chem. 2018, 267, 329–336. 10.1016/j.foodchem.2017.06.126. [DOI] [PubMed] [Google Scholar]
  5. Rannou C.; Laroque D.; Renault E.; Prost C.; Sérot T. Mitigation Strategies of Acrylamide, Furans, Heterocyclic Amines and Browning during the Maillard Reaction in Foods. Food. Res. Int. 2016, 90, 154–176. 10.1016/j.foodres.2016.10.037. [DOI] [PubMed] [Google Scholar]
  6. Perez Locas C.; Yaylayan V. A. Origin and Mechanistic Pathways of Formation of the Patent Furan a Food Toxicant. J. Agric. Food. Chem. 2004, 52 (22), 6830–6836. 10.1021/jf0490403. [DOI] [PubMed] [Google Scholar]
  7. Milani M. I.; Rossini E. L.; Castoldi K.; Pezza L.; Pezza H. R. Paper Platform for Reflectometric Determination of Furfural and Hydroxymethylfurfural in Sugarcane Liquor. Microchem. J. 2017, 133, 286–292. 10.1016/j.microc.2017.03.046. [DOI] [Google Scholar]
  8. Janzowski C.; Glaab V.; Samimi E.; Schlatter J.; Eisenbrand G. 5-Hydroxymethylfurfural: Assessment of Mutagenicity, DNA-damaging Potential and Reactivity towards Cellular Glutathione. Food Chem. Toxicol. 2000, 38 (9), 801–809. 10.1016/S0278-6915(00)00070-3. [DOI] [PubMed] [Google Scholar]
  9. Miyagi K.; oe E.; Yamagata N. Evaluation of Aging for Thermally Upgraded Paper in Mineral Oil. J. Int. Counc. Electr. Eng. 2011, 1 (2), 181–187. 10.5370/JICEE.2011.1.2.181. [DOI] [Google Scholar]
  10. Kishikawa N.; El-Maghrabey M. H.; Kuroda N. Chromatographic Methods and Sample Pretreatment Techniques for Aldehydes Determination in Biological, Food, and Environmental Samples. J. Pharm. Biomed. Anal. 2019, 175, 112782 10.1016/j.jpba.2019.112782. [DOI] [PubMed] [Google Scholar]
  11. Li Z.; Fang M.; LaGasse M. K.; Askim J. R.; Suslick K. S. Colorimetric Recognition of Aldehydes and Ketones. Angew. Chem. 2017, 129 (33), 9992–9995. 10.1002/ange.201705264. [DOI] [PubMed] [Google Scholar]
  12. Liang J.; Li H.; Wang J.; Yu H.; He Y. Cascade Chromogenic System with Exponential Signal Amplification for Visual Colorimetric Detection of Acetone. Anal. Chem. 2020, 92 (9), 6548–6554. 10.1021/acs.analchem.0c00149. [DOI] [PubMed] [Google Scholar]
  13. Cennamo N.; De maria L.; et al. Monitoring of Low Levels of Furfural in Power Transformer Oil with a Sensor System Based on a POF-MIP Platform. Sensors 2015, 15 (4), 8499–8511. 10.3390/s150408499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Li J.; Hou C.; Huo D.; Yang M.; Fa H.; Yang P. Development of a Colorimetric Sensor Array for the Discrimination of Aldehydes. Sens. Actuators, B 2014, 196, 10–17. 10.1016/j.snb.2014.01.054. [DOI] [Google Scholar]
  15. Wang B.; Anslyn E. V.. Chemosensors: Principles, Strategies, and Applications, 1st ed.; Wiley: Hoboken, NJ, 2011. [Google Scholar]
  16. Shi B.; Chai Y.; Qin P.; Zhao X.; Li W.; Zhang Y.; Wei Q.; Lin Q.; Yao H.; Yao H.; Qu W. Detection of Aliphatic Aldehydes by a Pillar[5]arene-based Fluorescent supramolecular Polymer with Vaporchromic Behavior. Chem. Asian. J. 2022, 17, e202101421 10.1002/asia.202101421. [DOI] [PubMed] [Google Scholar]
  17. EL-Maghrabey M.; Suzuki H.; kishikawa N.; kuroda N. Sensitive chemiluminescence detection approach for determination of 2,4-dinitrophenylhydrazine derivatized aldehydes using online UV irradiation – luminol CL reaction. Application to the HPLC analysis of aldehyde in oil samples. Talanta 2021, 233, 122522. [DOI] [PubMed] [Google Scholar]
  18. Liu M.; Deng J.; Lac C.; Chen Q.; Zhao Q.; Zhang Y.; et al. Synthesis, characterization of conjugated oilgo-phenylene-ethylnylenes and their supramolecular interaction with β-cyclodextrin for salicylaldehyde detection. Talanta 2012, 100, 229–238. 10.1016/j.talanta.2012.08.008. [DOI] [PubMed] [Google Scholar]
  19. Zhao Q.; Shen T.; Liu Y.; Hu X.; Zhao W.; Ma Z.; Li p.; Zhu x.; Zhang Y.; Liu M.; Yao S. Universal Nanoplatform for Formaldehyde Detection based on the Oxidase-Mimicking Activity of MnO2 Nano sheets and the In Situ Catalysis-Produced Fluorescence Species. J. Agric. Food chem. 2021, 69, 7303–7312. 10.1021/acs.jafc.1c01174. [DOI] [PubMed] [Google Scholar]
  20. Feng L.; Musto C. J.; Suslick K. S. A Simple and Highly Sensitive Colorimetric Detection Method for Gaseous Formaldehyde. J. Am. Chem. Soc. 2010, 132 (12), 4046–4047. 10.1021/ja910366p. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Besir A.; Yazici F.; Mortas M.; Gul O. A Novel Spectrophotometric Method based on Seliwanoff Test to Determine 5-(Hydroxymethyl) Furfural(HMF) in Honey: Development, in House Validation and Application. LWT 2021, 139, 110602 10.1016/j.lwt.2020.110602. [DOI] [Google Scholar]
  22. Sassine M.; Picquet-Varrault B.; Perraudin E.; Chiappini L.; Doussin J. F.; George C. A New Device for Formaldehyde and Total Aldehydes Real-Time Monitoring. Environ. Sci. Pollut. Res. 2014, 21, 1258–1269. 10.1007/s11356-013-2010-5. [DOI] [PubMed] [Google Scholar]
  23. Wong S. J.; Ringel A. E.; Yuan W.; Paulo J. A.; Yoon H.; Currie M. A.; Haigis M. C. Development of a Colorimetric α-ketoglutarate Detection Assay for Prolyl Hydroxylase Domain(PHD) Proteins. Biol. Chem. 2021, 296, 100397 10.1016/j.jbc.2021.100397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Rosini E.; Caldinelli L.; Piubelli L. Assays of D-Amino Acid Oxidase Activity. Front. Mol. Biosci. 2018, 4, 102. 10.3389/fmolb.2017.00102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Quesenberry M. S.; Lee Y. C. A Rapid Formaldehyde Assay Using Purpald Reagent: Application Under Periodation Conditions. Anal. Biochem. 1996, 234 (1), 50–55. 10.1006/abio.1996.0048. [DOI] [PubMed] [Google Scholar]
  26. Dickinson R. G.; Jacobsen N. W. A New Sensitive and Specific Test for the Detection of Aldehydes: Formation of 6-Mercapto-3-Substituted-s-Triazolo[4.3-b]-s-Tetrazines. J. Chem. Soc. D 1970, 24, 1719–1720. 10.1039/c29700001719. [DOI] [Google Scholar]
  27. Shrivastava A.; Gupta V. B. Methods for the Determination of Limit of Quantitation of the Analytical Methods. Chron. Young sci. 2011, 2 (1), 21–25. 10.4103/2229-5186.79345. [DOI] [Google Scholar]
  28. Angerer J.; Bernauer U. Opinion on Furfural. SCCS 2012, 1461, 12. 10.2772/77636. [DOI] [Google Scholar]
  29. ASTM international . Standard Test Method for Furanic Compounds in Electrical Insulation Liquids; ASTM D5837-99(reapproved2005); West conshohocken, PA; 2005. 10.1520/D5837-99. [DOI]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao3c07025_si_001.pdf (737.1KB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES