Reagent-less and sub-minute quantification of sulfite in food samples using substrate-integrated hollow waveguide gas sensors coupled to deep-UV LED

Highlights

• A fully integrated and portable sensing platform for the sub-minute and on-site determination of sulfide in beverages.

• The analytical system was comprised of a miniaturized gas cell and a deep-UV LED for direct absorbance measurements.

• The analytical method does not employ any reagent to collect the gaseous SO₂.

• The analysis of beer, coconut water, and juice samples were performed with suitable accuracy.

Abstract

The increasing consumption of processed foods demands the usage of chemical preservatives to ensure freshness and extended shelf life. For this purpose, sodium sulfite and its derivatives have been widely used in a variety of food products to inhibit microbial spoilage and for mitigating oxidative decay. However, the excessive consumption of sulfite may cause health problems, thus requiring rapid and accurate analytical methods for the rapid identification of threshold levels. Conventionally, sulfite is volatilized from food samples by acidification followed by trapping of the gaseous SO₂ and determination using a suitable analytical technique. Herein, we propose a yet unprecedented reagent-less approach via direct absorbance measurements of gaseous SO₂ at 280 nm after sample acidification. The detection system combines a deep-UV LED and a SiC photodiode with a substrate-integrated hollow waveguide (iHWG) gas cell. Absorbance measurements were performed using a log-ratio amplifier circuitry, resulting in noise levels <0.7 mAU. This innovative concept enabled the determination of sulfite in beverages in the range of 25–1000 mg L⁻¹ with suitable linearity (r² > 0.99) and an analysis time <30 s. The limit of detection (LOD) was calculated at 14.3 mg L⁻¹ (3σ) with an iHWG providing an optical path length of 75 mm. As a proof of concept, this innovative analytical platform was employed for sulfite quantification in concentrated grape juice, coconut water and beer, with suitable accuracy in terms of recovery (83–117%) and favorable comparison with the official Monier-Williams method. Given the inherent modularity and adaptability of the device concept, we anticipate the application of the proposed analytical platform for the in-situ studies addressing sulfite and other volatilized preservatives in a wide variety of food products with tailorable detectability.

Introduction

Free sulfites usually refer to a series of sulfur-containing chemical agents (e.g., sodium sulfite, sodium bisulfite, sodium metabisulfite, potassium sulfite, etc.) widely used as preservatives by the food industry [1]. Due to their inherent oxidation properties – they are easily oxidized to sulfates – sulfites prevent the proliferation of fungi and bacteria in various types of foods, ranging from alcoholic beverages and soft drinks to fruits, seafood, jams and processed meats [2,3].

On the other hand, an immoderate consumption of sulfite may cause innumerous health issues, especially to sulfite-sensitive people, such as dermatitis, urticaria, flushing, hypotension, abdominal pain, diarrhea, anaphylactic reactions, and inflammation of the airways (as it is an activator of neutrophils) [1].For these reasons, the majority of countries follow the recommendations by the Joint FAO/WHO Expert Committee on Food Additives (JECFA), which establishes an acceptable daily intake (ADI) of sulfite (expressed as SO₂) of 0.7 mg/kg per body weight/day [1,4,5]. The control and regulation of the maximum permitted levels of sulfite in foods are established by international agencies. For instance, the European Community through the Directive 95/2/EC establishes the maximum permitted levels of sulfite for a large variety of foodstuffs, including fruit juices for bulk dispensing systems (50 ppm), beer (100 ppm), mustard (250 ppm), dried fruits (2000 ppm) and marmalades (50 ppm) [6].

Hence, analytical methods that provide accurate sulfite determination have proven to be essential. The direct determination of sulfite in food matrices is not feasible due to the sample complexity. Therefore, during sample preparation it is fundamental to separate the analyte from the matrix in order to enhance the selectivity [1]. Additionally, it may also lead to an enhancement of sensitivity caused by the increase in the final concentration of the analyte. Extraction by volatilization (i.e. distillation) is frequently employed when the analyte is likely to be volatilized (e.g. formaldehyde [7], sulfite [8], sulfide [9] and nitrite). Free sulfites are extracted from the sample by conversion to gaseous SO₂ after acidification and then collected using an appropriate reagent solution – usually aqueous-based – prior to the analysis. A variety of instrumental analytical techniques based on this procedure have been used, for example, spectrophotometry [10], flow-injection analysis (FIA) [8] electrochemistry [11], and chromatography-DAD [12]. Moreover, the officially acknowledge Monier-Williams method [13] uses titration with standard NaOH solution after the conversion of SO₂ to SO₄²⁻ by the addition of H₂O₂.

However, it is a challenge to collect gaseous SO₂ with high efficiency, as it needs to solubilize into the trapping solution, thus requiring that the mass transference at the gas/liquid interface be elevated. Usually, the SO₂ is directed using a controlled airflow and bubbled into the trapping solution. Although this procedure decreases the analysis time, the efficiency of the sample solubilization is affected by high airflows, which in turn also affects the sensitivity. Gas-diffusion microextraction (GDME) allows the gas passage from a donor solution (sample) to an acceptor (trapping) solution through a semi-permeable membrane (e.g., PTFE) [14]. This interesting approach does not require pumps, as the gaseous SO₂ is moved by diffusion. However, the time to complete the solubilization process in the trapping solution is significantly extended. Therefore, it can be concluded that the volatilization/collection approach decreases the analytical frequency, affects the accuracy and precision of the results, uses specific reagents to enhance sample collection and requires an additional step to collect the actual measurement. Additionally, most analytical methods are bulky and laborious, which limits their applicability for in situ analysis. In particular, the Monier-Williams method suffers from low repeatability and sensitivity, laborious procedures, bulkiness of instrumentation, high analysis time, and interference from other volatile compounds.

An interesting approach to eliminate the sample collection step and maintain treatment benefits from volatilization is to directly detect the gaseous SO₂. A few studies following this strategy have been reported using different approaches. In one study, a conventional UV–visible molecular absorption spectrometer was employed to detect SO₂ in the gas-phase using the molecular absorption band at 198 nm [15]. In another approach, the molecular fluorescence of gaseous SO₂ was determined using a hollow cathode lamp (213.8 nm) and an atomic fluorescence spectrometer [16]. Oliveira et al. developed an analytical method to detect SO₂ generated from total and free sulfite present in coconut water using high-resolution absorption spectroscopy [17]. To this end, a xenon arc lamp was used as a continuous source and the absorbance was measured at 215 nm. Although the obtained linear ranges using these methods were suitable for the determination of sulfite in food samples (10–1000 ppm), the complexity (automatic injectors, specific quartz tubes and extended gas cells), high cost and bulkiness of instrumentation make these methods unfeasible for routine analysis. Additionally, it has been reported that the absorption cross-sections of SO₂ below 260 nm undergo significant differences due to variations in temperature [18].

Molecular spectroscopy – including vibrational and electronic spectroscopy – has been used extensively used for sensing because it provides valuable information about structure and concentration of molecules [19,20]. In 2013, Mizaikoff and collaborators designed a new generation of miniaturized and non-flexible hollow waveguides – the so-called substrate-integrated hollow waveguides (iHWG) [21] as gas cells. The interaction between the molecules and radiation in the mid-infrared and ultraviolet range provided by the iHWG enabled the development of several arrangements for the determination of numerous gases, such as O₃, NO_x, CH₄, CO₂ and H₂S among others [[22], [23], [24], [25], [26]]. Its highly versatile (i.e. adaptable) geometry favors the fabrication of a variety of sensing systems via the combination of iHWG with a wide range of light sources (e.g. conventional broadband IR sources, tunable lasers and miniaturized lamps) and detectors. The designable optical path length and the low volume of sample required facilitates close to real-time measurements and confers portability to the device, rendering it a promising for in situ applications.

Recently, deep-UV light emitting diodes (deep-UV LEDs) have been applied in analytical platforms due to their high output stability, robustness, low power consumption and low-heat production [27]. Their relatively narrow emission bandwidth (∼30 nm) allows their direct use as a light source for spectrometric devices without the need for optical filters or monochromators, resulting in a significant decrease in complexity and cost [28]. Several recent studies have demonstrated the development of analytical platforms based on deep-UV LEDs emitting at 235, 250 and 280 nm for absorbance measurements, including HPLC [29] and CE [30] detectors as well as gas sensors [31].

Therefore, in this study we report the combination of a deep-UV LED (@280 nm) and an aluminum-coated iHWG gas cell to produce a compact, portable and low-cost analytical platform. This device was employed for monitoring gaseous SO₂ generated from acidified food samples containing free sulfites via absorbance measurements. The obtained analytical signal was associated with the concentration of sulfite in food samples, providing a rapid, simple and accurate analytical method.