•A novel Ag NPs based SPR sensor was used for colorimetric detection of Cu2+.•Flow-batch analysis was introduced to determinate Cu2+ fast and automatically.•High sensitivity and selectivity, fast detection and easy synthesis procedure.•The proposed method has been successfully applied to water research and food safety.Using flow-batch analysis, a highly sensitive and selective method for automatic colorimetric detection of copper ions (Cu2+) was produced on the basis of the surface plasma resonance (SPR) of silver nanoparticles (Ag NPs). The Ag NPs were catalytically etched by thiosulfate in the presence of Cu(NH3)42+, resulting in a color change of the solution induced by the absorbance decrease at 401 nm of the SPR peak of Ag NPs. The proposed method showed high selectivity for Cu2+ over various metallic ions, including Fe3+, Mn2+, Co2+, Ni2+, Zn2+, Pb2+, Ba2+, Cd2+, Bi3+, Sb2+, As3+, Hg2+, Cr3+ and K+. The linear range was 0.5–35 μg/L with a coefficient of 0.9954. The limit of detection was as low as 0.24 μg/L. The relative standard deviation (RSD, n=7) for the determination of Cu2+ spiked samples at concentrations of 10 μg/L was 1.21% and for 25 μg/L was 1.03%. The proposed method was successfully applied to analyze Cu2+ in lake water, tap water, rainwater and bottled water samples, as well as leaf samples for food packaging. The results were in good agreement with those obtained by graphite furnace atomic absorption spectrometry, the classical technique.

•Simultaneous determination of NO2−, NO3−, PO43−, Fe2+, Fe3+ and Mn2+ in natural waters.•The targeted elements were detected in water samples and showed good agreement with reference methods.•The advantages of low reagent consumption and high sample throughput were satisfied the criteria of Green Chemistry.•The compact automatic system is fully integrated and highly suitable for field application.An integrated system was developed for automatic and sequential determination of NO2−, NO3−, PO43−, Fe2+, Fe3+ and Mn2+ in natural waters based on reverse flow injection analysis combined with spectrophotometric detection. The system operation was controlled by a single chip microcomputer and laboratory-programmed software written in LabVIEW. The experimental parameters for each nutrient element analysis were optimized based on a univariate experimental design, and interferences from common ions were evaluated. The upper limits of the linear range (along with detection limit, µmol L−1) of the proposed method was 20 (0.03), 200 (0.7), 12 (0.3), 5 (0.03), 5 (0.03), 9 (0.2) µmol L−1, for NO2−, NO3−, PO43−, Fe2+, Fe3+ and Mn2+, respectively. The relative standard deviations were below 5% (n=9–13) and the recoveries varied from 88.0±1.0% to 104.5±1.0% for spiked water samples. The sample throughput was about 20 h−1. This system has been successfully applied for the determination of multi-nutrient elements in different kinds of water samples and showed good agreement with reference methods (slope 1.0260±0.0043, R2=0.9991, n=50).

•Atomic spectrometer was vehicle-board applied in field.•Flame atomic fluorescence spectrometer was used to analyze salty water samples.•On-line SPE was developed for preconcentration and matrix removal.•The complicate salinity effect was investigated and reported.An automatic on-line preconcentration and detection system for analysis of trace dissolved manganese (Mn) in estuarine and coastal waters was established, using preconcentration with IDA chelating resin and detection with flame atomic fluorescence spectrometer (FAFS). The rinse (pre-eluent) solution was optimized, for removing the interference ions while retaining the target element. It was found that the interference ions affected the chelating efficiency of Mn, causing variation of the detection blank and sensitivity. This effect varied when sample volume presented as preconcentration time changed. The influence at preconcentration times of 120 s, 30 s and 10 s were carefully investigated and reported. Ten folds of the foreign trace metals Zn, Cu, Ni, Fe, and Al did not show obvious interference on Mn preconcentration and detection. The method detection limit was 0.9 nmol L-1 (n = 7, preconcentration time 120 s). The linear detection range could be adjusted with designed preconcentration time. In addition to high precision and accuracy, the proposed analytical system had the advantages of high integration, and required normal site preparation, low energy supply and simple auxiliary equipment, which was appropriate for in-field operation. Compared with other common in-field applied molecular spectrometry instruments, the inherent high selectivity and multi-element applicability of FAFS highlighted the superiority and potential of the proposed analytical system. It was successfully applied to in-field vehicle-board determination of dissolved Mn in coastal waters around Xiamen, Fujian, China, and it was also used to analyze natural water samples collected from the Jiulongjiang Estuary, Fujian, China.Download high-res image (341KB)Download full-size image

A flexible, flow injection analysis method for shipboard use was developed for the on-line determination of trace dissolved aluminum (dAl) in seawater. The analytical system included a Towed Fish underway sampler and a modified flow injection analysis system with a solid phase extraction-spectrophotometric detection device. Determination was based on the spectrophotometric detection of a complex of chrome azurol S and dAl. In this study, the dAl in the samples was efficiently extracted onto an SPE cartridge, packed with iminodiacetate chelating resin. The extracted dAl was rapidly eluted with hydrochloric acid and reacted with the reagents to form a complex, which was detected at 620 nm with a 30 mm flow cell. Compared with the commonly used methods, the proposed method offered the benefits of improved sensitivity, negligible salinity effect and low cost. The experimental parameters were optimized based on a univariate experimental design, and the matrix effect of seawater was preliminarily investigated. The proposed method had high sensitivity with a detection limit of 0.80 nmol L−1. The linearity range was 1.0 to 250 nmol L−1 with a 120 s sample loading time and the upper limit was extended to 1.0 μmol L−1 when choosing longer sample loading times. The recoveries were between 96.8 and 99.8% and the relative standard deviation was 2.6% (n = 8) for an aged seawater sample spiked with 5.0 nmol L−1 dAl. The analytical results obtained with the proposed method showed good agreement with those using a reference method. The proposed method has been successfully applied to a shipboard underway analysis of dAl in the Jiulongjiang Estuary, Fujian, China.

An automated, shipboard-use system was developed for real-time speciation of iron in coastal surface waters. It comprised a towed Fish underway sampler and a modified reverse flow injection analysis system with a liquid waveguide capillary flow cell–spectrophotometric detection device. The detection was based on the reaction between ferrozine and Fe(II). The detection limits of 0.3 and 0.7 nM were achieved for Fe(II) and Fe(II+III), together with their respective dynamic linear ranges of 0.5–250 and 0.9–250 nM. The system was successfully deployed and run consecutively for about 1 week during a cruise in August 2009 to the East China Sea off the Changjiang Estuary. The distribution of operationally defined field dissolvable Fe(II) and Fe(II+III) (expressed as Fea(II) and Fea(II+III)) in these areas was obtained, which showed that both Fea(II) and Fea(II+III) concentrations decreased with salinity when there were relatively high Fea(II) concentrations (up to about 120 nM) near shore. A distinct distribution of Fea(II) to Fea(II+III) ratios was also revealed, with a ratio of 0.58 in the water off Changjiang Estuary and 0.19 in the open ocean.

This study aimed to solve the common problems in Hg isotope analysis of water samples at low concentration. The isotope composition of dissolved Hg in seawater is reported for the first time. A modified device for introducing Hg into a multi-collector inductively coupled plasma mass spectrometer and a preconcentration method for the preconcentration of dissolved Hg were developed to enhance the sensitivity of the isotopic composition analysis method. The modified cold-vapor generator was used to transfer dissolved Hg2+ from the matrix into gaseous Hg0. The purge and trap method was developed and employed to preconcentrate dissolved Hg in water samples. Keeping other parameters the same, the Hg signal generated with the modified Hg introduction device was twice as much as the commercial one (HGX 200). In the measurement of NIST SRM 3133, the external precision for δ202Hg was 0.06‰ (2SD, n = 310), and the δ202Hg value of the UM-Almadén in-house secondary standard was −0.57 ± 0.10‰ (2SD, n = 49), indicating that the modified device was stable and reliable. Factors influencing the efficiency of the purge and trap method, e.g., concentration of KMnO4 in the trapping solution, flow rate of the purge gas and purge time, were optimized. With ultrapure water (blank) and seawater (matrix) spiked with NIST SRM 3133 at Hg concentrations of 5.00–35.50 ng L−1 and 10.00–35.50 ng L−1, the δ202Hg values of the blank spike and matrix spike were 0.00 ± 0.04‰ (2SD, n = 19) and −0.02 ± 0.04‰ (2SD, n = 12), respectively. The results indicated that the purge and trap method was free from matrix interference. The results of this practical application showed good stability and reproducibility of the proposed methods.

A portable automatic flow injection (FI) based system incorporating on-line C18 solid phase extraction (SPE) cartridges and a 2 m long liquid waveguide capillary cell (LWCC) is established for simultaneous redox speciation analysis of dissolved iron in estuarine and coastal waters. Utilization of the SPE preconcentration and the LWCC enhanced the sensitivity of the ferrozine method for Fe(II) analysis. The Fe(II)–ferrozine complex was formed and extracted onto a C18 cartridge, and eluted with an HCl–ethanol solution for spectrophotometric detection with an LWCC. The determination of total Fe(II + III) was realized in a parallel flow channel after the reduction of Fe(III) to Fe(II) with ascorbic acid. The optimal combination of the pre-eluting solution and eluent was investigated to eliminate the Schlieren effect. The parameters of the FI-SPE-LWCC system were optimized based on a univariate experimental design. The effect of salinity on the method sensitivity was low enough to apply the system in both estuarine and coastal waters without adjustment. The limit of detection was 0.056 nmol L−1 for Fe(II) and 0.096 nmol L−1 for Fe(II + III). A linear determination range of 0.5–50 nmol L−1 iron was obtained with a sample loading volume of 5 mL and a sample throughput of 6 h−1. The system was successfully applied in situ in Wuyuan Bay, Xiamen, for the continuous monitoring of dissolved iron species for 20 h.

Textile wastewater shows great threats to the environment if not well pretreated before discharge. A promising technique, ozonation, was applied to remove the color in dye solutions containing C.I. Reactive Red 195 (RR195) in a semi-batch reactor. The decolorization of RR195 by the ozone process followed pseudo-first-order kinetics. Several factors which influenced the efficiency of decolorization were studied and the reaction rate constant (k) obtained with different operational parameters was compared. Our results showed that RR195 was more easily degraded in acidic than in alkaline conditions. The dyeing auxiliaries (sodium carbonate and sodium chloride) that acted as radical scavengers could enhance the decolorization process, and the ozonation time for total color removal lengthened if the initial dye concentration was higher. The analysis of the ozonation products was performed by liquid chromatography–tandem mass spectrometer and a possible degradation pathway was predicted according to the ozonation products and structure of RR195. Our results indicated that ozonation was effective in the color removal of dyes, but further treatment might be necessary since the ozonation products are high toxic.The decolorization of C.I. Reactive Red 195 in water was carried out by the ozonation process. Although the decolorization process performed well, some aromatic and triazine compounds would still remain.

A high concentration of dissolved gaseous mercury (DGM) was detected in post-desulfurized waste seawater, which was discharged from a coal-fired power plant equipped with a seawater desulfurization system and which was located in a coastal area. A large amount of DGM was converted from other forms of mercury during transformation processes, such as photo-reduction. The present study targeted the photo-reduction of mercury and the effects of various environmental parameters on DGM production in the post-desulfurized seawater discharged. The results suggested that the photo-reduction of mercury was significantly induced under UV radiation, especially with UVB. The particulate mercury on suspended solids was easily photo-reduced and considered as an important source of DGM. It was confirmed that the suspended solids in post-desulfurized seawater could enhance the reduction process of mercury under UV radiation. The pseudo-first-order rate constants of DGM production, which were determined through the concentration gradient and trial methods, were 1.39 × 10−3 min−1 and 1.45 × 10−3 min−1, respectively. The values showed no significant difference and were both much higher than the reported results, indicating that the photo-reduction of mercury in post-desulfurized seawater deserved more attention. In addition, the initial mercury level was observed when mixing the post-desulfurized seawater with fresh seawater, and this suggested that a significant amount of initial mercury would be produced when the post-desulfurized seawater was discharged into the adjacent sea area and thus becomes another considerable source of DGM.

•An on-line determination of Fe(II) and Fe(II+III) with GFAAS was achieved.•An interface of the FI system and GFAAS was established.•The sensitivities were relatively constant in the salinity range 0–10 or >10.•The MDLs were low enough for Fe speciation in most estuarine and coastal waters.•The proposed method was applied to Fe speciation in estuarine and coastal waters.An automatic on-line solid phase extraction (SPE) system employing the flow injection (FI) technique directly coupled to a graphite furnace atomic absorption spectrometer (GFAAS) was established for speciation and determination of dissolved iron in estuarine and coastal waters. Fe(II) was mixed with ferrozine solution in a sample stream to form the Fe(II)–ferrozine complex which was extracted onto a C18 SPE cartridge, eluted with eluent and detected with GFAAS. In a parallel flow channel, Fe(III) was reduced to Fe(II) with ascorbic acid and then detected in the same way as Fe(II). The home-made interface between FI–SPE and GFAAS efficiently realized the sample introduction to the furnace in a semi-automated way. Parameters of the FI–SPE system and graphite furnace program were optimized based on a univariate experimental design and an orthogonal array design. The salinity effect on the method sensitivity was investigated. The proposed method provided a detection limit of 1.38 nmol L−1 for Fe(II) and 1.87 nmol L−1 for Fe(II+III). With variation of the sample loading volume, a broadened determination range of 2.5–200 nmol L−1 iron could be obtained. The proposed method was successfully applied to analyze iron species in samples collected from the Jiulongjiang Estuary, Fujian, China. With the 2-cartridge FI–SPE system developed, on-line simultaneous determination of Fe species with GFAAS was achieved for the first time.

•Flow injection analysis of ammonium at nanomolar levels•Ammonium analysis using liquid waveguide capillary cell and CCD spectrometer•3.6 nanomolar per liter detection limit with 22 samples per hour analysis rate•In-field application at Wuyuan Bay and the South China Sea•The variation of ammonium in the surface seawater was obtained.An automated colorimetric method for the on-line determination of trace ammonium in seawater was established using a flow injection technique coupled with a 2.5-m liquid waveguide capillary cell. Using low ammonium seawater as a carrier, a sample was injected into the carrier and mixed with reagents to form indophenol blue dye, which was monitored at a wavelength of 690 nm. Different strategies of reagent injection were investigated to obtain a lower reagent blank and a higher detection sensitivity. Experimental parameters were optimized using a univariate experimental design, and the matrix effect of seawater was preliminarily investigated. The proposed method had high sensitivity with a detection limit of 3.6 nmol·L− 1. The linearity was 10 to 500 nmol·L− 1 and the upper limit could be extended to 30 μmol·L− 1 by choosing a less sensitive detection wavelength or lower reaction temperature. The recoveries were between 95.0 and 104.3% and the relative standard deviation was 4.4% (n = 7) for an aged seawater sample spiked with 50 nmol·L− 1 ammonium. The sample throughput was 22 h− 1. The analytical results obtained with the proposed method showed good agreement with those using reference fluorescence methods. Compared with the normal indophenol blue (off-line) method, the proposed method was superior due to its lower reagent consumption, greater convenience, higher sample throughput, wider linear range (10 nmol·L− 1 to 30 μmol·L− 1), as well as higher sensitivity. The method was applied in-field in Wuyuan Bay for 24 h on-line monitoring of ammonium concentrations in the surface seawater. In addition, it was also used to analyze surface seawater samples collected from the South China Sea for the study of ammonium distribution.

Extracting the iron–humic acid (Fe–HA) complex from natural waters is difficult, since there is a lack of standards and many impurities are co-extracted with the target analytes. In this study, a laboratory synthesized Fe–HA complex was used as a standard to develop a simple solid phase extraction (SPE) method for analysis of the Fe–HA complex in natural water. The Fe–HA complex in the SPE extract was separated from the matrix and analyzed by size exclusion chromatography (SEC) combined with ultraviolet spectrophotometry. The Fe–HA complex was quantified as the molar concentration of HA bound Fe, which was determined by analyzing the SEC eluent with inductively coupled plasma mass spectrometry. The results suggested that optimized extraction could be achieved with an ENVI-18 cartridge as the sorbent, methanol as the eluent, and a flow rate of sample loading of 10 ± 2 mL min−1, and without adjusting the sample pH. Under the proposed SPE conditions, the linear range of the Fe–HA complex was 0.052 to 0.301 μmol L−1. The detection limit (S/N = 3.0) was found to be 0.0012 μmol L−1. The recovery of the spiked Fe–HA complex from a natural river water sample was 76.8% and an acceptable repeatability with a relative standard deviation of 5.2% was achieved. Moreover, the developed SPE method was successfully applied to analyze the Fe–HA complex in the Jiulongjiang River. It was found that the concentrations of the complex were in the range 0.023–0.085 μmol L−1, showing a gradual decrease from upstream to downstream.

•Fluorescence detection was coupled with flow analysis and solid phase extraction for ammonium determination.•The solid phase extraction technique has been seldom used to enrich compounds for fluorescence detection.•Different storage methods for ultra-trace ammonium samples were compared during the South China Sea survey.•This method was applied on board to analyze seawater samples collected from 65 stations in the South China Sea.•A detailed ammonium profile was obtained in the SEATS station.•The proposed method offered the benefits of improved sensitivity, reduced reagent consumption, negligible salinity interference and lower cost.Combining fluorescence detection with flow analysis and solid phase extraction (SPE), a highly sensitive and automatic flow system for measurement of ultra-trace ammonium in open ocean water was established. Determination was based on fluorescence detection of a typical product of o-phthaldialdehyde and ammonium. In this study, the fluorescence reaction product could be efficiently extracted onto an SPE cartridge (HLB, hydrophilic–lipophilic balance). The extracted fluorescence compounds were rapidly eluted with ethanol and directed into a flow cell for fluorescence detection. Compared with the common used fluorescence method, the proposed one offered the benefits of improved sensitivity, reduced reagent consumption, negligible salinity effect and lower cost. Experimental parameters were optimized using a univariate experimental design. Calibration curves, ranging from 1.67 to 300 nM, were obtained with different reaction times. The recoveries were between 89.5 and 96.5%, and the detection limits in land-based and shipboard laboratories were 0.7 and 1.2 nM, respectively. The relative standard deviation was 3.5% (n = 5) for an aged seawater sample spiked with 20 nM ammonium. Compared with the analytical results obtained using the indophenol blue method coupled to a long-path liquid waveguide capillary cell, the proposed method showed good agreement. The method had been applied on board during a South China Sea cruise in August 2012. A vertical profile of ammonium in the South East Asia Time-Series (SEATS, 18° N, 116° E) station was produced. The distribution of ammonium in the surface seawater of the Qiongdong upwelling in South China Sea is also presented.

•The rFIA–LWCC method allowed both nitrite and nitrate analyzed with single sample injection.•The detection limit is 0.6 nmol L−1, low enough for surface seawater analysis.•The matrix effect from seawater was preliminarily investigated.•Analytical frequency was 5 samples per hour, with both analytes detected in triplicate.A reverse flow injection analysis (rFIA) method coupled with 1 m liquid waveguide capillary cell and spectrophotometric detection for simultaneous determination of nanomolar nitrite and nitrate in seawater was developed. The design of two analytical channels sharing the same detection system in the proposed method allowed the analysis of both nitrite and nitrate with single sample injection. Different strategies of reagent injection were investigated to obtain a higher sensitivity and a better peak shape. A dual-wavelength detection mode was chosen to eliminate the light source shifting and sample matrix interference. Experimental parameters were optimized based on a univariate experimental design and the matrix effect from seawater was preliminarily investigated. The proposed method had high sensitivity with detection limit of 0.6 nmol L−1 for both nitrite and nitrate. The linearity was 2–500 nmol L−1 for both analytes, and the upper limit could be extended by choosing a lower sensitivity detection wavelength. The analytical results of 26 surface seawater samples obtained with the proposed method showed good agreement with those using a reference method operated using an automated segmented flow analyzer. The proposed method could greatly minimize the trouble introduced by bubbles in the segmented flow analyzer. It also had the advantages of high precision and high sample throughput (nitrite and nitrate detected in triplicate; 5 h−1). Compared to normal flow injection analysis, the rFIA method is superior due to its lower reagent consumption, less dispersion of sample, as well as higher sensitivity.

A method for determining iron in seawater had been developed by coupling reverse flow injection analysis (rFIA) and catalytic spectrophotometric detection with N,N-dimethyl-p-phenylenediamine dihydrochloride (DPD). With a seawater sample or a standard solution as the carrier, the mixture of DPD and buffer was injected into the carrier stream quantitatively and discretely. After mixing with H2O2, the DPD was oxidized to form two pink semiquinone derivatives that were monitored at 514 nm wavelength with a reference at 700 nm. The method detection limit was 0.40 nmol L−1, lower than half of that of normal flow injection analysis (nFIA) method. The sample throughput was 10 h−1 with triplicate determination, compared with 4 h−1 for nFIA-DPD method. The analysis results of the certified seawaters CASS-4 (12.33 ± 0.18 nmol L−1) and NASS-5 (3.47 ± 0.23 nmol L−1) well agreed with the certified values (12.77 ± 1.04 and 3.71 ± 0.63 nmol L−1, respectively). The typical precision of the method for a 2.97 nmol L−1 iron sample was 4.49% (n = 8). Interferences from copper and salinity were investigated. An instrument was assembled based on the proposed method and applied successfully to analyze total dissolvable iron (TDFe) in surface seawater samples collected from the Pearl River Estuary, the results of which revealed non-conservative behavior of TDFe during the estuarine mixing. Results for these samples with both rFIA-DPD and nFIA-DPD methods showed good agreement with each other. The proposed method was superior to the currently used nFIA-DPD method, particularly when it is adapted for field and in situ deployment, due to its lower reagent consumption, higher sample throughput and keeping the manifold tubing clean.Highlights► Catalytic spectrophotometry was coupled to reverse flow injection analysis for iron determination. ► Double carrier streams were used in reverse flow injection analysis. ► An intercomparison experiment with normal flow injection analysis was conducted. ► The concentrations of total dissolvable iron in surface seawater of the Pearl River Estuary were obtained and non-conservative mixing processes were revealed. ► The rFIA method was more sensitive, faster and more environment-friendly.

More and more coal-fired power plants equipped with seawater flue gas desulfurization systems have been built in coastal areas. They release large amount of mercury (Hg)-containing waste seawater into the adjacent seas. However, very limited impact studies have been carried out. Our research targeted the distribution of Hg in the seawater, sediment, biota, and atmosphere, and its environmental transportation.Seawater samples were collected from five sites: 1, sea areas adjacent to the power plant; 2, near discharge outlets; 3, the aeration pool of the power plant; and 4 and 5, two reference sites. The total gaseous Hg was determined in situ with a Tekran 2537B. Analyses of total Hg (TM) followed the USEPA methods.In most part of the study area, TM concentrations were close to the reference values and Hg transfer from the seawater into the sediment and biota was not obvious. However, in the aeration pool and near the waste discharge outlets, atmospheric and surface seawater concentrations of TM were much higher, compared with those at a reference site. The concentration ranges of total gaseous Hg and TM in seawater were 3.83–8.60 ng/m3 and 79.0–198 ng/L near the discharge outlets, 7.23–13.5 ng/m3 and 186–616 ng/L in the aeration pool, and 2.98–4.06 ng/m3 and 0.47–1.87 ng/L at a reference point.This study suggested that the Hg in the flue gas desulfurization waste seawater was not only transported and diluted with sea currents, but also could possibly be transferred into the atmosphere from the aeration pool and from the discharge outlets.

An automatic gas-phase molecular absorption spectrometric (GPMAS) system was developed and applied to determine nitrite and total nitrate in water samples. The GPMAS system was coupled with a UV-light emitting diode photodiode (UV-LED-PD) based photometric detector, including a 255 nm UV-LED as the light source, a polyvinyl chloride (PVC) tube of 14 cm as the gas flow cell, and an integrated photodiode amplifier to measure the transmitted light intensity. The UV-LED-PD detector was compact, robust, simple and of low heat production, comparing with detectors used in other GPMAS works. For nitrite measurement, citric acid was used to acidify the sample, and ethanol to catalyze the quantitative formation of NO2. The produced NO2 was purged with air flow into the UV-LED-PD detector, and the gaseous absorbance value was measured. The total nitrate could be determined after being reduced to nitrite with a cadmium column. Limits of detection for nitrite and nitrate were 7 μmol/L and 12 μmol/L, respectively; and linear ranges of 0.021–5 mmol/L for nitrite and 0.036–4 mmol/L for nitrate were obtained. Related standard deviations were 1.81% and 1.08% for nitrite and nitrate, respectively, both at 2 mmol/L. The proposed method has been applied to determine nitrite and total nitrate in some environmental water samples.

A fast, sensitive, and reliable method for the determination of sulfite (SO32−) in fresh water and seawater samples was developed. The proposed method was based on the reaction of o-phthalaldehyde (OPA)–sulfite–NH3 in alkaline solution, with flow injection analysis and fluorescence detection. The experimental parameters were investigated in pure water and seawater matrixes. The detection limits (S/N = 3) were 0.006 μmol/L in pure water and 0.018 μmol/L in seawater for SO32−. The method was successfully applied to analyze SO32− in the samples of rain water and flue gas desulfurization seawater.

Profiles of the bioaccessibility of soil polycyclic aromatic hydrocarbons (PAHs) in different urban functional areas of Xiamen City, Fujian, China were investigated. A physiologically based in vitro test was used to evaluate the bioaccessibility of total and individual PAHs. There was no obvious correlation between total concentrations of PAHs and bioaccessibility during the gastrointestinal phase for the soils from different functional areas. Results showed that the bioaccessibility variation in the gastrointestinal phase (ranging from 14.6% to 63.2%) was significantly higher than that in the gastric phase (ranging from 4.9% to 21.8%). The bioaccessibility in the gastrointestinal phase was not only determined by soil organic materials but also directly related to physical and chemical properties of individual PAHs, except for two-ring PAHs. Increasing soil organic material content or decreasing ring numbers of PAHs could result in the decrease of PAH bioaccessibility. The total PAH bioaccessibility was largely contributed by individual PAHs with relatively high molecular weight.

A novel reverse flow injection analysis method coupled with a liquid waveguide capillary cell (LWCC) and spectrophotometric detection for the determination of nanomolar soluble reactive phosphorus in seawater was established. Reagent was injected into the sample stream and detected in a 2-m path length LWCC with detection wavelength set at 710 nm. Experimental parameters, including the reagent concentration, the injection volume, the flow rate and the length of the mixing coil, were optimized based on univariate experimental design. The interference of silicate and arsenate were also investigated. Under optimized conditions, the linearity and the detection limit of the proposed method were found to be 0–165.0 nM and 0.5 nM, which was estimated to be three times the standard deviation of the measurement blanks (n = 9). The relative standard deviations for the determination of 24.7 and 82.5 nM samples were 1.54% and 1.86% (n = 9), respectively. Three seawater samples were analyzed with recoveries ranging from 87.8% to 101.8%. Using the Student's t-test at the 95% confidence level, the results of the proposed method and a segmented flow analyzer reference method for determination of the two samples showed no significant difference. The proposed method had the advantages of being less reagent consuming, more sensitive and with higher throughput (15 h−1).

A method for speciation and determination of low levels of dissolved iron in rainwater was established by coupling reverse flow injection analysis with a 2-m liquid waveguide capillary cell and spectrophotometric detection. Ferrozine solution was injected into a sample stream to form an Fe(II)-ferrozine complex with Fe(II), and the absorbance of this complex was detected at both 562 nm and 625 nm with a reference wavelength at 700 nm. Fe(III) was analyzed in the same manner after being reduced to Fe(II) by ascorbic acid. The optimum conditions and the interference of Cu(I), Ni(II) and Co(II) were investigated. The limits of detection were 0.1 nM for Fe(II) and 0.2 nM for Fe(III), while the linear ranges were 0.4 – 200 nM for Fe(II) and 0.8 – 287 nM for Fe(III) at 562 nm, and can be extended to higher concentrations with the detection at a less sensitive wavelength of 625 nm. The sample throughput was 6 h−1, and the total sample volume consumed was 10 mL. This method has been successfully applied to analyze dissolved iron in rainwater of Xiamen from August to November, 2008. The lowest level of iron in rainwater was observed during typhoon events. By adopting reverse flow injection analysis coupled with liquid waveguide long path length capillary cell, the reagent consumption was low and the sensitivity was enhanced. The other advantages of this method are high sample throughput, wide linear dynamic range and high selectivity for Fe(II).

A cartridge of solid phase extraction (SPE), hydrophilic–lipophilic balance (HLB), has been used to enrich phosphomolybdenum blue (PMB) from water samples without any other additives. Based on this, the previous on-line SPE method established for the determination of nanomolar soluble reactive phosphorus (SRP) in seawater has been greatly improved. Cetyltrimethylammonium bromide (CTAB), the cationic surfactant needed for the formation of the PMB-CTAB paired compound that could be extracted on a Sep-Pak C18 cartridge using the previous method [Liang, Y., Yuan, D.X., Li, Q.L., Lin, Q.M., 2007. Flow injection analysis of nanomolar level orthophosphate in seawater with solid phase enrichment and colorimetric detection. Marine Chemistry 103, 122–130.], was not necessary. Thus the longer time and higher temperature required for the complete formation of the PMB-CTAB compound were no longer needed. In addition, with application of the sequential injection analysis technique the proposed method showed the advantages of being much faster, simpler, sample and reagent saving, as well as more convenient in operation. The PMB compound formed under room temperature was efficiently extracted on an in-line HLB cartridge, rapidly eluted by 0.15 mol/L NaOH solution, and finally determined with a laboratory-made spectrophotometer at 740 nm. Experimental parameters, including the volume of reagents added, sample loading flow rate, and eluting flow rate, were optimized. Time and temperature for the PMB reaction, and salinity effect were also studied, and these were found to have no severe effect on the detection. With variation of sample loading time at a fixed flow rate, a broadened determination range of 3.4 to 1134 nmol/L phosphate could be obtained. The recovery and the method detection limit of the proposed method were found to be 94.4% and 1.4 nmol/L, respectively. The relative standard deviation (n = 7) was 2.50% for the sample at a concentration of 31 nmol/L phosphate. Two typical seawater samples were analyzed with both the proposed method and the magnesium hydroxide-induced coprecipitation method and, using the t-test, the results of the two methods showed no significant difference.

•Spread of 1.5‰ was observed in δ202Hg values of raw coals and coal related samples.•The δ202Hg values were more negative in fly ash than those in the raw coal.•The flue gas had a significant Hg fractionation after desulfurization.•The stack emissions were enriched with lighter isotopes compared with the raw coal.In the current study, fractionation of mercury isotopes during coal combustion and seawater flue gas desulfurization (SFGD) in a coal-fired power plant using a SFGD system was investigated. Fourteen samples were collected from the power plant. The samples were pretreated with a combustion-trapping method and were analyzed with a multi-collector inductively coupled plasma mass spectrometer (MC-ICP-MS). Compared with the raw coal, the bottom ash was enriched with lighter mercury isotopes with δ202Hg values ranging from −0.45 to −0.03‰. The fly ash was enriched with lighter mercury isotopes with δ202Hg values ranging from −1.49 to −0.73‰ for Chinese coal and from −1.47 to −0.62‰ for Indonesian coal. The δ202Hg of fresh seawater and desulfurized seawater was found to be −1.32 and −0.32‰ respectively. These δ202Hg values indicated that the desulfurized seawater was enriched with heavier mercury isotopes. Based upon the calculated results obtained from the mass balance equation, it was suggested that the stack emissions were enriched with lighter mercury isotopes. Mass independent fractionation was observed in most of the samples with a Δ199Hg/Δ201Hg ratio of approximately 0.96. The results help in improving the understanding of mercury isotope fractionation during coal combustion and SFGD, and are also useful in tracing the mercury emissions from coal fired power plants.

A valve-free continuous flow method and its corresponding instrument were established, with only two multi-channel pump for delivering the sample and reagent, without any injection or solenoid valves and sample loop for selecting and adding the sample or reagent. Nitrate was reduced to nitrite with Cu-Cd reductant column, and then detected with spectrophotometric detector. The proposed method was suitable for determination of nitrate at normal level in most of estuary and coastal seawaters. With the optimum parameters, the linear range and detection limit were 5–180 μM and 0.27 μM, respectively. The nitrate samples (10 and 80 μM) were continually measured for 11 times, and the relative standard deviations were 1.4% and 1.3%, respectively. The recovery of real samples at different salinity was ranged from 99.4% to 106.1%. There were no significant differences in the analytical results between the proposed method and a reference method, i.e., flow injection analysis. In comparison with flow injection analysis, the method and instrument were less cost and easy to operate, and was suitable to be applied in general laboratories and field for continuous monitoring. The method was successfully applied to measure nitrate of seawater samples in Xiamen western harbor and underway monitor nitrate in the Jiulongjiang estuary.

The polycyclic aromatic hydrocarbons (PAHs) in flue gas emitted from a coal-fired power plant equipped with selective catalytic reduction (SCR) de-NOx device were determined. The concentrations and distributions of phase and type of PAHs in the flue gas at the outlet and inlet of the SCR device were compared. The results show that the SCR de-NOx device leads to the increase of concentrations and toxic equivalent of PAHs, especially the low rings of PAHs in the flue gas. However, the device has no significant influence on the PAHs partition between particle and the gas phase.

•Elemental mercury transfer from post-desulfurized seawater to air is studied.•Mercury isotope compositions in DGM and GEM are determined with MC-ICP-MS.•Aeration enhances the transfer and induces MIF of even mercury isotopes in GEM.Samples of dissolved gaseous mercury (DGM) in the post-desulfurized seawater discharged from a coal-fired power plant together with samples of gaseous elemental mercury (GEM) over the post-desulfurized seawater surface were collected and analyzed to study the mercury isotope fractionation during transfer from post-desulfurized seawater to air. Experimental results showed that when DGM in the seawater was converted to GEM in the air, the δ202Hg and Δ199Hg values were changed, ranging from − 2.98 to − 0.04‰ and from − 0.31 to 0.64‰, respectively. Aeration played a key role in accelerating the transformation of DGM to GEM, and resulted in light mercury isotopes being more likely to be enriched in the GEM. The ratio Δ199Hg/Δ201Hg was 1.626 in all samples, suggesting that mercury mass independent fractionation occurred owing to the nuclear volume effect during the transformation. In addition, mass independent fractionation of mercury even isotopes was found in the GEM above the post-desulfurized seawater surface in the aeration pool.

This study aimed to solve the common problems in Hg isotope analysis of water samples at low concentration. The isotope composition of dissolved Hg in seawater is reported for the first time. A modified device for introducing Hg into a multi-collector inductively coupled plasma mass spectrometer and a preconcentration method for the preconcentration of dissolved Hg were developed to enhance the sensitivity of the isotopic composition analysis method. The modified cold-vapor generator was used to transfer dissolved Hg2+ from the matrix into gaseous Hg0. The purge and trap method was developed and employed to preconcentrate dissolved Hg in water samples. Keeping other parameters the same, the Hg signal generated with the modified Hg introduction device was twice as much as the commercial one (HGX 200). In the measurement of NIST SRM 3133, the external precision for δ202Hg was 0.06‰ (2SD, n = 310), and the δ202Hg value of the UM-Almadén in-house secondary standard was −0.57 ± 0.10‰ (2SD, n = 49), indicating that the modified device was stable and reliable. Factors influencing the efficiency of the purge and trap method, e.g., concentration of KMnO4 in the trapping solution, flow rate of the purge gas and purge time, were optimized. With ultrapure water (blank) and seawater (matrix) spiked with NIST SRM 3133 at Hg concentrations of 5.00–35.50 ng L−1 and 10.00–35.50 ng L−1, the δ202Hg values of the blank spike and matrix spike were 0.00 ± 0.04‰ (2SD, n = 19) and −0.02 ± 0.04‰ (2SD, n = 12), respectively. The results indicated that the purge and trap method was free from matrix interference. The results of this practical application showed good stability and reproducibility of the proposed methods.

A portable automatic flow injection (FI) based system incorporating on-line C18 solid phase extraction (SPE) cartridges and a 2 m long liquid waveguide capillary cell (LWCC) is established for simultaneous redox speciation analysis of dissolved iron in estuarine and coastal waters. Utilization of the SPE preconcentration and the LWCC enhanced the sensitivity of the ferrozine method for Fe(II) analysis. The Fe(II)–ferrozine complex was formed and extracted onto a C18 cartridge, and eluted with an HCl–ethanol solution for spectrophotometric detection with an LWCC. The determination of total Fe(II + III) was realized in a parallel flow channel after the reduction of Fe(III) to Fe(II) with ascorbic acid. The optimal combination of the pre-eluting solution and eluent was investigated to eliminate the Schlieren effect. The parameters of the FI-SPE-LWCC system were optimized based on a univariate experimental design. The effect of salinity on the method sensitivity was low enough to apply the system in both estuarine and coastal waters without adjustment. The limit of detection was 0.056 nmol L−1 for Fe(II) and 0.096 nmol L−1 for Fe(II + III). A linear determination range of 0.5–50 nmol L−1 iron was obtained with a sample loading volume of 5 mL and a sample throughput of 6 h−1. The system was successfully applied in situ in Wuyuan Bay, Xiamen, for the continuous monitoring of dissolved iron species for 20 h.

Extracting the iron–humic acid (Fe–HA) complex from natural waters is difficult, since there is a lack of standards and many impurities are co-extracted with the target analytes. In this study, a laboratory synthesized Fe–HA complex was used as a standard to develop a simple solid phase extraction (SPE) method for analysis of the Fe–HA complex in natural water. The Fe–HA complex in the SPE extract was separated from the matrix and analyzed by size exclusion chromatography (SEC) combined with ultraviolet spectrophotometry. The Fe–HA complex was quantified as the molar concentration of HA bound Fe, which was determined by analyzing the SEC eluent with inductively coupled plasma mass spectrometry. The results suggested that optimized extraction could be achieved with an ENVI-18 cartridge as the sorbent, methanol as the eluent, and a flow rate of sample loading of 10 ± 2 mL min−1, and without adjusting the sample pH. Under the proposed SPE conditions, the linear range of the Fe–HA complex was 0.052 to 0.301 μmol L−1. The detection limit (S/N = 3.0) was found to be 0.0012 μmol L−1. The recovery of the spiked Fe–HA complex from a natural river water sample was 76.8% and an acceptable repeatability with a relative standard deviation of 5.2% was achieved. Moreover, the developed SPE method was successfully applied to analyze the Fe–HA complex in the Jiulongjiang River. It was found that the concentrations of the complex were in the range 0.023–0.085 μmol L−1, showing a gradual decrease from upstream to downstream.