Formaldehyde (FA) is a colorless and irritating toxin, which has been associated with memory loss, diabetes, Alzheimer's disease, and cancer [1, 2]. The emission of FA from food, cosmetics, packaging boxes, furniture, building materials, and other daily sources makes the exposure to FA quite common [3, 4]. According to the recommendation of the World Health Organization (WHO), the permitted content of FA in indoor air is < 80 ppb [5]. Based on China's National Standard, the maximum tolerable content of FA in drinking water is 0.9 mg/L [6]. There are increasing concerns about whether there is FA in daily contact. Therefore, it is of great importance to develop facial detection methods for FA.

There have been several methods for FA detection, such as high-performance liquid chromatography, gas chromatography, capillary electrophoresis, and electrochemical methods [7]. However, these methods need high expertise, complicated procedures, and expensive devices, which restrict their application in the fast detection of FA.

Colorimetric methods based on the color change of fluorescence probes hold great potential in addressing the above issue. Recently reported fluorescence probes include organic molecules such as hydrazine or hemocyanine [8-10], functionalized quantum dots [11], specific carbon dots [12-14], or designed composites/systems [15-18]. However, the chromogenic reactions between organic molecule probes and FA are always slow and need harmful reagents, while the fluorescent intensity variations of the other nano-probes are hard to discriminate by the naked eyes. Simple and rapid detection method for FA is still a challenge.

Recently, carbon dots with different intrinsic/emission colors provide simple colorimetric detection methods [19-23]. Herein, we present a visual detection method for formaldehyde by using a kind of acidichromic carbon dots (ACDs) and the specific reaction between formaldehyde and ammonium chloride. The ACDs are red as-prepared and turn blue upon strong acid or low pH value. FA and ammonium chloride are both weak acid that will not induce the color change of the ACDs, whilst their reaction produces H⁺ and methenamine, providing stronger acid circumstance and changing the color of ACDs from red to blue. The FA is thus detected visually in a fast and simple manner.

The ACDs are prepared according to a modified method that is detailed described in Supporting information [19]. The as-prepared ACDs are monodispersed with an average size of 3.50 ± 0.59 nm and lattice spacing of 0.165 nm, as illustrated in Fig. 1a. The size distribution is listed in Fig. S1a (Supporting information). The X-ray diffraction (XRD) result of ACDs in Fig. S1b (Supporting information) shows a wide peak at 2θ = 22.3°, which proves that the internal structure is amorphous. The FT-IR of the ACDs (Fig. S1c in Supporting information) is characterized by stretching vibration peaks of different groups, such as N−H/O−H bonds at 3469 cm^-1, C−H bonds from methyl or methylene at 2970 cm^-1 and 2879 cm^-1, C=O/C=N bonds at 1630 cm^-1, and C-O/C-N bonds at 1044 cm^-1 and 1302 cm^-1. The Raman spectroscopy of ACDs in Fig. 1b shows typical D band and G band at 1340 cm^-1 and 1580 cm^-1 with a strength ratio of 1.34, which indicates defects in the graphitized structure [24]. X-ray photoelectron spectroscopy (XPS) (Fig. S2 in Supporting information) further confirms these defects by the presence of heteroatoms, i.e., nitrogen and oxygen, and their combination with carbon atoms in the form of C-C=O (289.2 eV), C=N (285.0 eV, ), C=O (284.4 eV and 532.0 eV), C=N (397.2 eV), C-N (400.0 eV), and C−O (531.0 eV). These chemical bonds decide the photo properties of the ACDs.

As displayed in Fig. 2a, the ACDs are red in daylight and show strong red fluorescence when irradiated with UV light (365 nm). The ACDs exhibit multi-excitation (540, 570 nm) and dual emission (600, 650 nm) fluorescence features. The multi excitation originates from the absorption in the range of 500–600 nm with peaks at 540 nm and 570 nm (Fig. 2a), which are derived from the n-π* transition of the C-N/C=N bond and C-O/O=C-O bond. The multi-excitation and emission fluorescence of the ACDs are further proved by the 3D fluorescence contour map with four obvious peaks (Fig. 2b). Another fluorescence property of the ACDs is the excitation-independent fluorescence emission at different excitation wavelengths in the range of 360–580 nm, which is displayed in Fig. S3a (Supporting information). The fluorescence of ACDs is sensitive to pH values. As illustrated in Fig. 2c, the fluorescence intensity gradually increases in the range of 2–6 and decreases from 7 to 9, and the fluorescence is almost completely quenched when the pH value is higher than 10. This might be attributed to the protonation and deprotonation of surface functional groups, which leads to the change of the Fermi energy level of CDs [25-27]. Besides, the fluorescence of ACDs is varied with solvents. As shown in Fig. 2d, when dissolved in high-polarity solvents such as N,N-dimethylformamide (DMF) or dimethyl sulfoxide (DMSO), the fluorescence intensity is enhanced, and the emission peaks become wider. While in lower-polarity solvents such as acetone and ethyl acetate, the intensity decreased and the peaks shifted to shorter wavelengths. The ACDs show moderate intensity and peak width when dissolved in moderate-polarity solvents methanol and ethanol. Contrary to the polarity and pH value, ionic strength, long-time laser irradiation, and storage do not influence the fluorescence of ACDs (Figs. S3b-d in Supporting information), which demonstrates the high photo and chemical stability of ACDs.

The color of ACDs is sensitive to pH values. As demonstrated in Fig. 3a, the solution of ACDs is blue when the pH value is 2.0 and it turns purple and red within a pH value range from 3.0 to 9.0. When the pH value increases from 10.0 to 13.0, the color of ACDs gradually fades and becomes colorless.

To explain the above phenomena, the XPS of ACDs under pH values of 2.0, 4.0, and 6.0 were characterized. As shown in Fig. S4 (Supporting information), the C 1s and N 1s spectra prove that the single bond species decrease while the double bond species increase when pH value increase. For instance, the integral area of N 1s bonds quantitatively demonstrate that the relative contents of N=C (391.7 eV) increase from 0.100 to 0.480, while the relative contents of N-H (399.5 eV) and N-C (400.2 eV) decrease from 0.607 to 0.387 and from 0.294 to 0.133, respectively. Therefore, the color change of the ACDs in acidic condition are attributed to the variation of the chemical bonds. This is similar to our previous investigation [19].

As for the color fading under alkaline environment, there are two reasons. First, the ACDs show scarce absorbance in the UV–vis range under alkaline conditions, especially when pH value was higher than 10 (Fig. S5 in Supporting information). According to the color rendering principle, matter absorbs light of specific wavelength from natural light and present the color of the complementary light. The ACDs hardly absorb light under strong alkaline conditions, and thus turns colorless. Second, the lattice structure of ACDs might be changed by the added alkaline reagent. The ultra high-resolution TEM images (Fig. S6 in Supporting information) indicate that the atom arrangement of the ACDs become disorder when pH value is 10 or 12, which might be the essential reason of the color fading. In summary, we infer that the strong alkaline conditions change the intrinsic structure of the ACDs and make the ACDs absorb scarce light, which induces the color fading.

When dropped on filter papers, the ACDs can also indicate the pH variation by the color change from blue to red, as displayed in Fig. 3b. This demonstrates the capability of the ACDs as a visual indicator of pH values by its acidichromism feature.

Considering that FA reacts with ammonium chloride and produces H⁺ and methenamine, which leads to the variation of pH value. The reaction formula can be seen in Formula S2 (Supporting information). If this pH variation induces the color change of the ACDs, then FA can be detected visually. As illustrated in Fig. 4, the color of filter paper loaded with ACDs does not change when 20 µL of NH₄Cl solution (200 mmol/L) and FA solution (1.0 mol/L) are dropped singly. Interestingly, if NH₄Cl solution and FA solution (0.05–1.0 mol/L) are dropped successively, the color of the filter paper turns from red to blue in 5 min. A minimum discriminated concentration of FA is 0.04 mol/L. The stability of the visual detection method is investigated. The filter paper which loaded with ACDs were dried natural. After 3 and 7 days, the dried filter paper were used to test FA. The results in Fig. S7 (Supporting information) show color change when NH₄Cl solution (200 mmol/L) and FA solution (0.05–0.30 mol/L) were added, which demonstrates the good stability of the visual detection method. Notably, the existing visual detection methods for FA usually takes a long time and needs volatile/harmful reagent. Herein, the usage of the acidichromism feature of the as-prepared ACDs and the reaction between FA and ammonium chloride provide a fast and simple visual detection method for FA. In addition, no harmful reagents are used in the whole detection process. As well known, the exposure to FA is quite common in daily life due to the contact with furniture, chemicals, dried squids, or other FA-treated things. With the increasing concerns about the daily pollutants and health, the fast assessment of the content of FA in daily contacts is highly desired. In this work, we report a fast, simple and healthy visual semi-quantitative method for FA, which hold potentials to provide convenient household in-site monitor tools for FA. In addition, this work might open more possibility for the fast detection of other daily pollutants.

The visual detection mechanism for FA can be explained by the basic optical principle that the absorbed light determines the color of the ACDs. After mixed with FA and ammonium chloride singly, the absorption spectrum of ACDs hardly varies. When mixed with FA and ammonium chloride simultaneously, a new absorption peak at 625 nm appears, and absorption at 570 nm increases while absorption intensity at 540 nm decreases (Fig. S8a in Supporting information). With the increase of concentration of FA, the absorption at 570 nm and 625 nm increases (Fig. S8b in Supporting information). This is due to that the H⁺ produced by FA and NH₄Cl can regulate the structure of the ACDs and change the absorption spectrum, as well as the color of the ACDs.

Moreover, we also found a fluorescent determination method for FA based on the response of ACDs to FA concentrations. Experimental conditions were optimized and the adopted concentration of ACDs is 10 µg/mL, the initial pH value is 5–7, and the initial concentration of NH₄Cl is 200 mmol/L (Figs. S9a-c in Supporting information). As shown in Fig. S9d (Supporting information), formaldehyde (0–1.5 mmol/L) does not affect the fluorescence of the ACDs singly. However, after being mixed with NH₄Cl, the fluorescence of ACDs decreases with the addition of FA (Fig. S10a in Supporting information). A linear relationship with formaldehyde concentration in the range of 0.067–0.27 mol/L is found with the regression equation ΔI/I₀ = 1.5812c + 0.03322 and linear correlation coefficient R² = 0.994 (Fig. S10b in Supporting information). The detection limit is calculated to be 0.029 mol/L. The analytical method exhibits high selectivity and anti-interference. Various ions (Na⁺, K⁺, Ca²⁺, Ba²⁺, Ti⁴⁺, S^2-), phenols (4-nitrophenol), acids (benzoic acid, p-phthalic acid), aldehydes (benzaldehyde, p-hydroxybenzaldehyde), and methanol with a final concentration of 0.1 mol/L pose no effects on the fluorescence of ACDs (Fig. S10c in Supporting information) and does not interfere the determination of FA (Fig. S10d in Supporting information). Real sample, e.g., shredded squid is analyzed via the spiking method and the results in Table S1 (Supporting information) shows satisfactory recoveries of 87.6%−99.1%, indicating the success application of determination method for FA in real samples.

In summary, we demonstrated the potential of ACDs in visual detection of FA. The ACDs turns from red to blue when acidity increases. The reaction between FA and NH₄Cl produce enough acidity-enhancement to change the color of the ACDs, which indicates the contents of FA. The present method is fast and simple without using harmful reagents, which is better than traditional colorimetric methods. In addition, we also found a fluorescent method based on the response of ACDs to FA contents. The visual detection method and the fluorescence method are different in principle, equipment and application scenarios. The fluorescence detection method is suitable for accurate quantification in laboratory, while the visual method is suitable for in-site semiquantitative assessment. Since the low-toxicity and biocompatibility of carbon dots, the acidichromic carbon dots might open more applications in environmental, medical or bio-analytical fields.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Quanxing Mao: Supervision, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. Zhengliang Wang: Validation, Methodology, Investigation. Zhinan Hu: Visualization, Validation, Investigation, Formal analysis, Data curation, Conceptualization. Ziqi Yang: Validation, Methodology, Investigation, Formal analysis. Hui Li: Resources, Project administration. Yali Yao: Writing – review & editing, Funding acquisition. Zijun Yong: Validation. Tianyi Ma: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

Acknowledgments

The authors acknowledge the support from the National Natural Science Foundation of China (Nos. 21804062, 52071171, and 52202248), Liaoning BaiQianWan Talents Program (No. LNBQW2018B0048), Shenyang Science and Technology Project (No. 21–108–9–04), and Key Research Project of Department of Education of Liaoning Province (No. LJKZZ20220015). T. Ma acknowledged the Australian Research Council (ARC) through Future Fellowship (No. FT210100298), Discovery Project (No. DP220100603), Linkage Project (Nos. LP210200504, LP220100088, and LP230200897) and Industrial Transformation Research Hub (No. IH240100009) schemes, the Australian Government through the Cooperative Research Centres Projects (No. CRCPXIII000077), the Australian Renewable Energy Agency (ARENA) as part of ARENA's Transformative Research Accelerating Commercialisation Program (No. TM021), and European Commission's Australia-Spain Network for Innovation and Research Excellence (AuSpire).

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110499.