Flow-batch analysis

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy Trends in Analytical Chemistry, Vol. 35, 2012 Trends Flow-batch analysis Paulo Henrique Gonçalves Dias Diniz, Luciano Farias de Almeida, David Peter Harding, Mário César Ugulino de Araújo Since its inception, flow-based analysis has dramatically transformed the way that sample analysis is handled. What used to be manual, slow and prone to error has become automated, rapid and precise. Though many improvements and modifications have already been explored, a most promising development has been the flowbatch analyzer. Through use of an instantaneous stop-mixing chamber and computer-controlled, flow-based sampling, it achieves extraordinarily high sampling rates. We review its historical evolution, components, features and operation, and analytical applications. ª 2012 Elsevier Ltd. All rights reserved. Keywords: Concentration gradient; Flow analysis; Flow-batch analyzer; Instantaneous stop-mixing chamber; Multi-component simultaneous analysis; Sample analysis; Sample pre-treatment; Screening analysis; Standard-addition method (SAM); Titration Paulo Henrique Gonçalves Dias Diniz, Luciano Farias de Almeida, David Peter Harding, Mário César Ugulino de Araújo* Universidade Federal da Paraı́ba, CCEN, Departamento de Quı́mica, P.O. Box 5093, 58051-970 João Pessoa, PB, Brazil * Corresponding author. Tel./Fax: +55 83 3216 7438; E-mail: [email protected] 1. Introduction Since the publication of the first flow-based techniques, several approaches to automated flow analysis have been developed and reviewed in the literature [1–6]. In parallel with these developments, hybrid systems or analyzers exploiting characteristics of both flow and batch have also been developing, with most of them exploring the flexibility and the versatility of mixing chambers [7–12]. The first hybrid system was proposed in 1969 by Javier and co-workers [7] using an approach that involved features of both batch and flow injection; it included the use of an automatic stopped-flow system for phosphate determinations. Other hybrid systems applied to flow titrations have also been developed, exploring the use of chambers [8–12]. Different approaches to calibration of flow systems have been described in the literature and were recently reviewed [13–16]. In 1999, Honorato et al. [11] developed the flow-batch analyzer (FBA), created as an alternative to the then existing methods for flow titrations. The basic technique joins flow, discrete (or batch), and programmable multi-commutation principles to the sampling and analytical process. In the past 12 years, new, surprisingly interesting applications of this basic design have been successfully explored. The main advantages derive from the complete programmability of the process – from the control of flows and samplings to instantaneous stop-chamber mixing, detection and evacuation, the process never leaves the control screen. Programmable commutation and instantaneous stop-chamber sample preparation, followed by detection give to FBA systems an immense operational advantage. The main component of this manifold is the mixing chamber [14] (MC), into which different solutions can optionally be defined, added, or removed with complete control, because of a loop system in which the samples and the reagents are always returning to their containers, allowing intermittent sampling. FBA systems are already being mentioned in flow-analysis textbooks [5,6]. Trojanowicz [5] has included FBA in the classification of flow-based analytical systems in which six kinds are presented: continuous flow (CF), flow injection (FI), sequential injection (SI), multi-commutation in flow injection (MCFI), stopped flow (SF) and batch-flow injection (BFI). Trojanowicz [5] describes the hybrid system proposed by Honorato et al. [11] without conceptual loss, but, in FBA systems, the fluids are sequentially or simultaneously (drawn) into the chamber instead of being injected. In 1994, van der Linden [18] proposed definitions and classification of analytical methods based on flowing media. The main criterion for classification of these 0165-9936/$ - see front matter ª 2012 Elsevier Ltd. All rights reserved. doi:http://dx.doi.org/10.1016/j.trac.2012.02.009 39 Author's personal copy Trends Trends in Analytical Chemistry, Vol. 35, 2012 methods is the type of sampling involved, which is divided into two major groups: (1) analysis with continuous sampling; and, (2) analysis with intermittent sampling, where sample portions are injected into the system from a sampling loop. Different combinations of features [e.g., confluence type, propulsion fluid(s), pumping or aspiration, and segmentation scheme] define the peculiarities of each method. Amplifying these recommendations, Zagatto [19] in 2002 classified flow-analysis methods and procedures as shown in Fig. 1. In Fig. 1, we have placed the typical FBA system appropriately. In this article, we present a comprehensive, critical discussion of the fundamentals, the applications, and the outlook for FBA. 2. Basic components A basic FBA system usually comprises five essential parts: (1) a propulsion system, which propels the fluids towards the analysis. Most FBA systems described in the literature employ peristaltic pumps. Nevertheless, FBA systems employing an SIA valve [20–24] or a piston [25] have been used for this purpose; (2) a transport system to all units, which constitutes the FBA system, usually pumping or conducting tubes with diameter 0.2–2.0 mm; (3) a multi-commutation system (valves), three-way solenoid valves for directing the fluids towards the mixing chamber or continuing the loop flow; (4) a mixing/reaction chamber, where aliquots (e.g., sample, reagents, calibration solutions, buffer, and indicators) are added and subsequently processed; and, (5) a detection system, monitoring a given property of a sample, its product, or the analytical matrix. The solutions or mixtures prepared in the mixing chamber are usually sent to the detector, but several approaches also have been developed with the detection system as part of the mixing chamber [17,20–22,24,26–30]. Fig. 2 shows the main components of a typical FBA system. A personal computer is used to control the peristaltic pump, valve actuator and detector. In Fig. 3, we see that the fluids are transported (in continuous flow) through flexible tubes, from their respective containers through the peristaltic pump and back again. Solenoid valves serve to commutate each respective flow to the mixing chamber. One of the solenoid valves directs the final mixed/prepared solution to detection and subsequent discard. As mentioned before, the mixing chamber is the main FBA component, and precisely carries out calibrationsolution preparations, sample conditioning, and analyte/ reagent additions, and is where even detection is possible. Fig. 4 shows the details of a typical mixing chamber. The next section describes the fundamentals of FBA with comprehensive descriptions of the analyzer and its operating modes. 3. Fundamentals In Fig. 3, the manifold (flexible tubes) of the typical FBA is at start-up, where reagents and samples are always recycling (returning to their respective flasks), and being propelled by a peristaltic pump. The dotted lines inside the solenoid valves indicate closed channels. A single valve positioned at the outlet of the mixing chamber for aspiration is also connected to a channel of the peristaltic pump to flush the MC, sending mixtures towards detection (valve ON), or cleaning solution to clean and to Figure 1. Classification of the flow-based analytical methods with an appropriate placement for flow-batch systems (adapted from [18]). 40 http://www.elsevier.com/locate/trac Author's personal copy Trends in Analytical Chemistry, Vol. 35, 2012 Trends Figure 2. The main components of a flow-batch analysis (FBA) system: (a) peristaltic pump; (b) valve actuator; (c) personal computer; (d) threeway solenoid valve; (e) mixing chamber; (f) magnetic actuator; and, (g) detector. Figure 3. Basic manifold of a typical flow-batch analyzer (FBA). C1–C4, Containers of samples, reagents or standard solutions; PP, Peristaltic pump; V1–V5, Solenoid valves; MS, Magnetic stirrer; MC, Mixing/reaction chamber; MA, Magnetic actuator; D, Detector; W, Waste. discharge the detection unit (valve OFF). FAAS, FES, and ICP-OES/MS systems have their own cleaning systems. Switched ON, the inlet solenoid valves can send accurate amounts of the fluids towards the mixing chamber, simultaneously or sequentially, (while the detector is washed). Afterwards, the MC mixture is drawn towards the detector and sent out as waste. In FBA systems, the flow channels transport the fluids while sample processing is accomplished in the mixing/ reaction chamber, before detection (or monitoring), just http://www.elsevier.com/locate/trac 41 Author's personal copy Trends Trends in Analytical Chemistry, Vol. 35, 2012 Figure 4. Typical mixing chamber: (a) solid, (b) internal, and (c) transversal view. The bottommost channel is always for mixture exit/detection/ discard. as in batch analysis. Detection/monitoring can be accomplished at the MC or afterwards. The addition of fluids to the chamber is by three-way solenoid valves acting as fluid conductors to direct the fluids towards the MC. The ON/OFF timing of the valves is programmed on a microcomputer (or dedicated microcontroller PIC) to deliver precise fluid volumes to the chamber, making possible simultaneous or sequential multi-commutation of the fluids. Sample processing is performed in discrete batch mode inside the MC. Highly sensitive measurements can be made because the physical and chemical equilibriums inherent in the analytical process are reached without dispersion or dilution of the sample. Analytical signal measurements can be performed in continuous [11] or stopped flow [22] modes, directly in the MC [30]. In general, FBA allows implementation of wellestablished classical methods while minimizing operator errors. Also, by automating the method, it offers greater flexibility to configure the particular sample. The ability to work with a very wide range of analyte concentrations is an intrinsic advantage of FBA. Both sample and reagent concentrations are preset and controlled by changing the timing and flow-rate parameters on the control software. 42 http://www.elsevier.com/locate/trac The FBA system can perform separate analytical processes without changing the physical configuration of the system, and is programmable to the point of permitting linear and non-linear concentration-gradient techniques for samples and reagents (which go beyond multi-tasking). FBA samples may be kept inside the analyzer for extended periods of time without significant sampling-rate reduction, making FBA suitable for slow kinetic reactions. As such, FBA systems combine intrinsic features of flow and batch analysis. The result is a robust, automated system with high analytical frequency and low reagent/sample consumption that allows varied applications with good figures of merit. However, as with other automatic analyzers, FBA systems present some inherent inconveniences. Most FBA systems use peristaltic pumps to propel the fluids. A time-consuming step, an average 5 s for emptying the sample channel, is required when the sample is changed (the whole analytical sample path must be emptied before the next sampling). The physical properties of the pumping tubes also change with time. When the peristaltic pump tubing is changed, a completely new calibration of the flow system must be performed for Author's personal copy Trends in Analytical Chemistry, Vol. 35, 2012 Trends Figure 5. Piston pump coupled to the mixing chamber for propulsion of fluids in a flow-batch analysis (FBA) system. (a) Stepper motor (b) mixing chamber (c) internal view of a mixing chamber with a piston inserted inside. The red arrows indicate the directions of piston movement (adapted from [25]). flow-rate parameters, and the system instructions set, as described in one paper [31]. Roller pulsation inherent in the tube system has serious implications for valve-activation times of less than 1 s. Pulsation negatively affects volume accuracy for each and every aliquot delivered to the MC. This is always the case when fluid volumes are below 10 lL; any dead volumes in the conducting tubes (proportional to the tube length between each valve and the mixing chamber) must be minimized to avoid serious measurement problems. In order to overcome these drawbacks, Almeida and co-workers [25] proposed a new FBA system using piston propulsion – a piston-propelled flowbatch analyzer (PFBA) (Fig. 5). Using a piston to drive the system has proved to be convenient because it costs much less, is portable and self-flushing, and avoids altogether the possibility of pulsation problems occasioned with peristaltic pumps. Due to the hydrophobic characteristic of PTFE materials used in the mixing chambers, valves and transmission lines of FBA manifolds, a cleaning step may be unnecessary. However, carry-over may occur in some applications if organic species in the sample matrix adhere to the surface of the PTFE materials. In this case, a cleaning step can be performed by switching ON a dedicated cleaning solution line valve for an appropriate time, homogenizing the content in the FB chamber (FBC) for a similar time span, and then switching ON a discard valve in order to empty the FBC and the line between the FBC and the detection unit (or spectrophotometer). This procedure should be repeated (one, two or three times) until the baseline signal is restored. 4. Analytical applications Several analytical procedures and methods based on FBA have been developed providing a wide range of applications. Table 1 summarizes the analytical applications performed with FBA and their respective figures of merit. 4.1. Titration Classical titrations are generally laborious and time consuming, and require a good amount of reagent, sample, and patience. They are also prone to error. The precise physical characteristics of timed stop-chamber filling, mixing, and measurement (with flushing after each discrete sample) answer these complaints while FBA automated titrations [11,12,20,24,29–34] conform remarkably to IUPAC titration guidelines [35]. To illustrate the versatility, Honorato and co-workers [11] applied a single-dimension optimization algorithm, based on the Fibonacci-sequence method for end-point search when determining acidity in white wines by titration with NaOH using an m-cresol purple indicator. Reducing the number of data points in Fe2+ ion determination for steel, iron ores, and similar samples using permanganimetric titration, Honorato and co-workers [32] designed and utilized a prior (preliminary) assay scheme to predict the analyte concentration, by processing known amounts of sample and titrant, and evaluating the titrant excess. Classical metronidazole determinations in drugs by spectrophotometric titration use perchloric acid to titrate, and malachite green as indicator. In order to circumvent the inherent drawbacks in this determination, Medeiros http://www.elsevier.com/locate/trac 43 http://www.elsevier.com/locate/trac Ref. Analyte Sample Detection technique Detection limit or range Sampling rate (/h) White wines UV-Vis 5.2–7.3 g/L 20 [20] N-total Plant materials UV-Vis 1.00–6.00 % N 14 [21] Fe(III) Estuarine waters UV-Vis 10.0–50.0 lg/L 30 [22] Aluminum Plant tissues UV-Vis 0.22 mg/L 15 [23] Iron Hydrated ethanol fuel FAAS 0.04 mg/L 10 [24] Cocaine Apprehended drug samples UV-vis 29.4 mg/L 12 [25] Manganese Mineral waters GFAAS 0.0048 ng 80 [26] Amoxicillin Pharmaceuticals LED-based photometry 5.1 mg/L 50 [27] Catecholamines Pharmaceuticals Chemiluminescence N/A 28 [28] Fe(II) Pharmaceuticals LED-based photometry 1.0–10.0 mg/L 120 [29] Bromine Petrochemicals Coulom./amp. N/A N/A [30] Acid content Fruit juice LED-based photometry N/A N/A [31] Ca2+ and Mg2+ River and dam waters Flame photometry N/A 120 [32] Iron Alloys and ores UV-Vis 1.0–10 mmol/L 20 Acidity determination in wines; algorithm based on Fibonacci method for end-point search In-line individual sample-matrix matching; analysis of samples with highly variable acidity Determination of Fe(III) in estuarine waters with high saline variability In-line specific pH conditioning of digested vegetal samples Internal standard calibration; iron determination in hydrated ethanol fuel Spectrophotometric determination of cocaine using cobalt thiocyanate as a complexing reagent Piston-propelled flow-batch analyzer; automatic preparation of calibration solutions Spectrophotometric determination of amoxicillin by using a reaction of diazotised o-nitroaniline in an alkaline medium Chemiluminescent determination of dopamine, norepinephrine and epinephrine in pharmaceutical preparations Miniaturized analyzer using an urethane–acrylate photo-resist ultraviolet-lithographic technique Determination of bromine index and bromine number for petrochemicals Photometry based on a twin-LED assembly as radiation source and photo-detector; determination of acid content of fruit juices by using titration with NAOH Screening analysis; water hardness quality control; LED-based photometry at 650 nm Titration of ferrous ions by permanganate; analyte concentration preview by prior assay Author's personal copy Tartaric acid Trends in Analytical Chemistry, Vol. 35, 2012 [11] Remarks Trends 44 Table 1. Analytical applications of flow-batch analysis Pharmaceuticals UV-Vis N/A 60 [34] Ascorbic acid Natural orange juices Coulometry 0–9.0 · 10 [36] Monosodium glutamate Foods Nephelometry 9.7 · 10 [38] Copper Alcoholic beverages FAAS 1.0–7.0 mg/L 70 [39] Sodium Alcohol fuel Flame photometry 0.5–2.5 mg/L 80–140 [40] Chlorine Chlorinated waters Biamperometry 8.261 · 10 [41] N/a Wine UV-Vis N/A N/A [42] N/a Distilled spirits UV-vis N/A 120 [43] Cu2+, Mn2+ and Zn2+ Pharmaceuticals UV-vis N/A 60 [44] Levodopa and carbidopa Pharmaceuticals UV-Vis N/A 18 [45] Albumin and total protein Blood serum UV-vis N/A 50 and 60 [46] Glycerol Biodiesel Fluorescence 0.036 mg/L 14 [47] Hydroxyproline Sausages UV-Vis 0.12 lg/mL 1 5 5 mol triiodide/L g/L 7 mol/L N/A 9 10 Exploitation of concentration gradients; spectrophotometric determination of metronidazole Determination of ascorbic acid in natural orange juice using a coulometric flow cell for generation of triiodide ions as a titrant Determination of monosodium glutamate in dehydrated broths using nephelometric detection Copper determination in alcoholic beverages standard addition automatic method Alcoholic grade determination, standard addition method In-line electrochemical reagent generation; standard-addition technique; determination of total available chlorine in commercial formulations and chlorinated water samples Screening analysis for quality control of wines; multivariate chemometric analysis Adulteration detection in distilled spirits using SIMCA classification models Preparation of calibration-standard mixtures for simultaneous multicomponent spectrometric analysis by using PLS Simultaneous enzymatic determination of levodopa and carbidopa, using chemometric modeling Kinetics-independent spectrometric analysis; non-linear calibration modeling; exploitation of concentration gradients Liquid-liquid extraction of glycerol and simultaneous oxidation with periodate, generating formaldehyde that reacts with acetylacetone Hydrolysis of hydroxyproline and spectrophotometric detection in separate chambers Trends 45 N-total, Total nitrogen; UV–Vis, UV–Vis spectrophotometry; ICP-AES, Inductively coupled plasma atomic emission spectrometry; FAAS, Flame atomic absorption spectrometry; GFAAS, Graphite furnace atomic absorption spectrometry; Turbid./nephel., Turbidimetry/nephelometry; Coulomb./amp., Coulometry/amperometry; PLS, Partial Least Squares; SPA, Successive Projections Algorithm; SIMCA, Soft Independent Modeling of Class Analogy; PLS, Partial Least Squares. N/A, Not Applicable. Author's personal copy Metronidazole Trends in Analytical Chemistry, Vol. 35, 2012 http://www.elsevier.com/locate/trac [33] Author's personal copy Trends Trends in Analytical Chemistry, Vol. 35, 2012 and co-workers [33] proposed a flow-batch-multi-commuted titrator successfully exploiting concentration gradients generated inside the MC. The proposed methods avoid analytical curves. Another titration application of the FBA system for determination of the bromine index and bromine number relative to the total content of reactive olefin in petrochemical samples was explored by Pasquini and co-workers [29]. In this procedure, two pairs of platinum electrodes were used for coulometric generation of reagent (titrant), and biamperometric end-point detection. The titration was performed in batch, with excess titrate being detected by biamperometry following analysis of the titration curve. Oliveira and co-workers [34] employed triiodide ions freshly prepared by electro-generation in a coulometric flow cell in instantaneously controlled amounts and subsequently used them as the titrant in an FBA system for determination of ascorbic acid in orange-juice samples. The procedure was carried out by introducing fixed amounts of sample together with iodide solution into the mixing chamber of the FBA system. Successive amounts of triiodide were then generated until a significant excess was obtained. The titration equivalence point was precisely identified using statistical line-regression analysis. 4.2. Sample pre-treatment The following procedures for liquid-liquid extraction [24], in line matching of pH [22], acidity [20] and salinity [21] demonstrate flow-batch techniques for sample pre-treatment. A pre-treatment strategy was developed by Honorato and co-workers [22] using an FBA system to perform in-line pH conditioning of vegetable-sample digests for spectrometric determination of aluminum with eriochrome cyanine R (ECR) as the chromogenic reagent. In-line individual pH sample conditioning was performed, because Al-ECR complex absorbance depends on pH. The individual pH sample conditioning was based on a feedback procedure involving absorbance monitoring of m-cresol purple indicator in the pH range 1.2–2.8. In plant materials, the determination of total nitrogen using Kjeldahl digests normally presents high acid variability. To overcome this drawback, an FBA system was implemented for in-line acidity matching [20]. To adjust pH, thymolphthalein was used as the pH indicator and a 2.0 mol/L NaOH / 0.5 mol/L Na2B4O7 solution was incrementally added until the processed sample achieved a pH value of about 12. To minimize systematic errors caused by salinity variations in estuarine waters during spectrophotometric catalytic determination of Fe3+, Carneiro and co-workers [21] proposed an FBA system employing a matrix-matching technique. The method was based on iron-catalyzed oxidation of N,N-dimethyl-p-phenylenediammonium dichloride by H2O2. The rate of the indi46 http://www.elsevier.com/locate/trac cator reaction was sensitive to variations in salinity, so different amounts of NaCl were added to every sample to permit an individual conductance adjustment, which provided both better and more easily reproducible conditions for the indicator reaction. Silva and co-workers [24] proposed a successful FBA procedure for spectrophotometric cocaine determination in a biphasic medium using cobalt thiocyanate as the complexing reagent. In this reaction, two phases were formed: the superior aqueous phase (pink) contained an excess of cobalt thiocyanate solution; and, the lower organic phase layer (blue) contained the cocaine–cobalt thiocyanate complex from which the signal was measured. During the crystallization of organic compounds, the presence of some substances even at trace levels can produce a delay in the precipitation process. Flow batch was used to implement and control both nucleation and crystallization [36], which permitted nephelometric detection in the determination of monosodium glutamate (MSG) for food samples. The method was based on the inhibitory effect of MSG on the crystallization of l-lysine in an isopropanol/acetone mixture. The calibration curve was prepared on-line. 4.3. Analyte or standard addition Analytical approaches employing samples with different matrix compositions generally give inaccurate results, with statistically different tendencies, or matrix effects. Procedures implementing the standard-addition method (SAM) are in the IUPAC recommendation [37]. Notwithstanding the advantages, this method is slow and relatively laborious. These drawbacks were circumvented by automating the SAM using FBA, as described for the first time by Almeida and co-workers [38]. A flow–batch SAM was employed for copper determination by flame atomic absorption spectrometry (FAAS) in a group of sugar-cane-based alcoholic beverages, known as ‘‘Cachaça’’, which present significant compositional differences and matrix effects for each brand. Copper determinations by analytical curve using matrixmatching standards can yield inaccurate results, thus requiring the use of the SAM [38]. Alcohol fuels also present significant variability in matrix composition and consequent matrix effects in flame-emission spectrometry (FES) analysis, justifying the use of the SAM. A flow-batch SAM analyzer developed for alcohol fuels in sodium determination by flame photometry has also been developed [39]. Nascimento and co-workers [40] developed a biamperometric FBA system for determination of total available chlorine in commercial bleaches and chlorinated tap-water samples without the need for unstable chlorine standards. The procedure was based on standard additions of an in line electrochemical-generated reagent. The system used an electrochemical flow cell for Author's personal copy Trends in Analytical Chemistry, Vol. 35, 2012 in-line generation of triiodide ions, which reacted with chlorine in the samples. 4.4. Screening analysis Flow-batch screening analysis and analyte quantification can use the same manifold configuration, which minimizes set-up costs while providing a more complete picture [31,41,42]. An FBA water-hardness screening/quantification system was developed and employed for quality control of river and dam waters in Paraı́ba, Brazil [31]. Detection was based on spectrophotometric monitoring at 650 nm for the well-known quantitative reaction of calcium/ magnesium ions with EDTA, using eriochrome black T as indicator. Nascimento and co-workers [41] accomplished screening analysis of wines employing an FBA system with spectrophotometric detection in the UV-vis range. Due to the strong absorption of the pure samples in the spectral region studied, the samples of wines were automatically diluted (with distilled water) and mixed in the chamber before recording their spectra. The study was evaluated using chemometric models for exploratory analysis and classification. Nascimento and co-workers [42] also proposed an automatic FBA system for adulterated distilled spirits using UV-vis spectrophotometry and soft independent modeling of class analogies (SIMCA) classification models. Automatic dilutions of the non-adulterated samples, and simulated adulterations using water, ethanol, and methanol were performed in an FBA system. 4.5. Multi-component simultaneous analysis FBA systems have also facilitated an increasing variety of simultaneous determinations, which effectively eliminate the needs for comprehensive examination and interference separation [43,44]. Visani and co-workers [43] developed an FBA system with UV-vis spectrophotometric detection and automatic preparation of both standard calibration and validation mixtures. The mixtures were evaluated applying 4-(2piridilazo)resorcinol as the complexing reagent for simultaneous determination of Cu2+, Mn2+ and Zn2+ ions in polyvitaminic and polymineral pharmaceutical formulations. Grünhut and co-workers [44] developed simultaneous determination of levodopa and carbidopa in pharmaceutical preparations using an enzymatic FBA system with spectrophotometric detection. This study was based on oxidization (to their respective dopaquinones) of both analytes using the enzyme polyphenol oxidase. 4.6. Concentration gradients Analytical procedures using FBA-gradient techniques can be easily implemented where only one standard solution is available for slow kinetics reactions (involving Trends non-linear calibration), and for titrations where the sample is incrementally added to the mixing chamber by programmed multi-commutation [33,45]. Medeiros and co-workers [33] proposed metronidazole spectrophotometric determination in drugs using concentration gradients generated in a flow-batch-multicommutated titrator. The titration used perchloric acid as titrant and malachite green as indicator. Souza and co-workers [45] developed an FBA system that exploits concentration gradients for total-protein and albumin determination in blood serum involving slow kinetics reactions. The study was evaluated successfully using kinetics-independent spectrometric analysis with non-linear calibration modeling. 4.7. Chemiluminescence Flow-batch controlled chemiluminescent detection was developed for determination of dopamine, norepinephrine, and epinephrine in pharmaceutical preparations. The method was based on the inhibitory effect of these catecholamines on a luminol-potassium hexacyanoferrate (III) chemiluminescent system in an alkaline medium. The optimization was carried out using an experimental Box-Behnken design at 28 samples/h [27]. 4.8. Internal-standard addition A flow-batch manifold coupled to an atomic absorption flame spectrometer was evaluated for assessing the iron content of hydrated automotive ethanol using the internal-standard method. Nickel was selected as the internal standard, since it is usually absent in samples, and because it requires similar atomization conditions. Samples were collected randomly from service stations in Pernambuco, Brazil, and iron concentrations were determined using the proposed procedure at about 10 samples/h [23]. 4.9. Fluorescent determination of free glycerol in biodiesel Commercial biodiesel has residual amounts of glycerol that can cause engine, fuel-tank and filter damage. A flow-batch method to determine free glycerol in biodiesel used in-line analyte extraction, and a fluorescent product for automatic signal analysis. The preparation of standards and samples, and derivations and analysis were fully automated. A sampling rate of 14 samples/h and a detection limit of 0.036 mg/L glycerol agreed with the reference method (ASTM D6584-07) at a 95% confidence level [46]. 4.10. Miniaturization of the flow-batch analysis system Technology always tends toward miniaturization and portability, particularly when considering medical applications, or field/outpost applications far from local access. A complete FBA system has been miniaturized and tested [28], using deep UV-lithography on urethanehttp://www.elsevier.com/locate/trac 47 Author's personal copy Trends Trends in Analytical Chemistry, Vol. 35, 2012 acrylate photoresin, in determining Fe(II) in iron-based supplements (oral solutions) using the 1,10-phenanthroline method. The micro FBA (lFBA), using a lowcost microprobe and discrete (100 lL) high frequency sampling rates, was compared to a manual UV-Vis photometric reference procedure with no statistically significant differences. 4.11. Automated two-system arrangement The presence of hydroxyproline is often used as an analytical approach to assess the amount of collagen present in meat products. The versatility of FBA systems was underlined in this study which used two stop chambers in series [43]. The FBA system was simple, and used both chambers (hydrolysis and detection) simultaneously. The reference method used 16 h of hydrolysis, while, in the proposed method, the hydrolysis time was 15 min. For this purpose, a pressure hydrolysis chamber was used with a low-cost halogen lamp [43]. 5. Conclusions When FBA systems are compared to other types of flow analyzer, their versatility stands out. Given the immense amount of control and flexibility available for whichever type of system is used in applying FBA or PFBA, one could conceivably perform markedly different analytical procedures without significantly changing the physical configuration. Using FBA, for example, one could carry out an automatic spectrometric titration for determination of iron in a given sample, and immediately afterwards perform simultaneous determinations of carbidopa and levodopa in pharmaceutical samples. To do this, it would be necessary only to actuate more solenoid valves, and to adjust the control software. Operational parameters in FBAs are changed and specified for different analytical systems, purely and simply as a function of valve-switch-timing control. Analytical curves can be adjusted to the working calibration solution used. This is possible because, in FBA, only one standard solution is necessary to construct a calibration curve. All dilutions are carried out automatically (controlled by the timing of the valves for each application). These programmed automatic dilutions can be applied to the samples as well, so that samples with analyte concentrations out of the linear range of detection can be analyzed after dilution. Virtually any analytical procedure or technique coupled to any detection system can be automated using FBA, and new applications are being explored with continued success. Acknowledgements The authors gratefully acknowledge Capes and CNPq Brazil scholarships and research fellowships. The 48 http://www.elsevier.com/locate/trac authors also thank Herbertty Vieira and Renato Navarro for collaboration in the design of figures. References [1] J. Ruzicka, E.H. Hansen, Trends Anal. Chem. 27 (2008) 390. [2] V. Cerdà, C. Pons, Trends Anal. Chem. 25 (2006) 236. [3] Y.L. Yu, Y. Jiang, M.L. Chen, J.H. Wang, Trends Anal. Chem. 30 (2011) 1649. [4] P.C.A.G. Pinto, M. Lucia, M.F.S. Saraiva, J.L.F.C. Lima, Anal. Lett. 44 (2011) 374. [5] F.A. Iñón, M.B. Tubino, in: M. Trojanowicz (Editor), Advances in Flow Analysis, Wiley-VCH, Weinheim, Germany, 2008, p. 6. [6] C. Pasquini, M.V. Rebouças, in: S.D. Kolev, I.A. McKelvie (Editors), Comprehensive Analytical Chemistry. Advances in Flow Injection Analysis and Related Techniques, Elsevier, Amsterdam, The Netherlands, 2008, p. 619. [7] A.C. Javier, S.R. Crouch, H.V. Malmstadt, Anal. Chem. 41 (1969) 239. [8] J.A. Sweileh, P.K. Dasgupta, Anal. Chim. Acta 214 (1988) 107. [9] J. Bartroli, L. Alerm, Anal. Lett. 28 (1995) 1483. [10] L. Alerm, J. Bartroli, Anal. Chem. 68 (1996) 1394. [11] R.S. Honorato, M.C.U. Araújo, R.A.C. Lima, E.A.G. Zagatto, R.A.S. Lapa, J.L.F.C. Lima, Anal. Chim. Acta 396 (1999) 91. [12] A.J C. Garcia, B.F. Reis, J. Autom. Methods Manage. Chem. 2006 (2006) 1. [13] P. Koscielniak, J. Kozak, Crit. Rev. Anal. Chem. 36 (2006) 27. [14] P. Koscielniak, M. Wieczorek, J. Kozak, J. Koziol, Anal. Lett. 44 (2011) 398. [15] P. Koscielniak, M. Wieczorek, J. Kozak, M. Herman, Anal. Lett. 44 (2011) 411. [16] J. Kozak, M. Wojtowicz, A. Wrobel, P. Koscielniak, Talanta 77 (2008) 587. [17] E.A.G. Zagatto, J.M.T. Carneiro, S. Vicente, R. Fortes, J.L.M. Santos, J.L.F.C. Lima, J. Anal. Chem. 64 (2009) 524. [18] W.E. van der Linden, Pure Appl. Chem. 66 (1994) 2493. [19] E.A.G. Zagatto, J.F. van Staden, N. Maniasso, R.I. Stefan, G.D. Marshall, Pure Appl. Chem. 74 (2002) 585. [20] J.M.T. Carneiro, R.S. Honorato, E.A.G. Zagatto, FreseniusÕ J. Anal. Chem. 368 (2000) 496. [21] J.M.T. Carneiro, A.C.B. Dias, E.A.G. Zagatto, R.S. Honorato, Anal. Chim. Acta 455 (2002) 327. [22] R.S. Honorato, J.M.T. Carneiro, E.A.G. Zagatto, Anal. Chim. Acta 441 (2001) 309. [23] J.E. Silva, F.A. Silva, M.F. Pimentel, R.S. Honorato, V.L. Silva, M.C.B.S.M. Montenegro, A.N. Araújo, Talanta 70 (2006) 522. [24] M.J. Silva, E.V. Anjos, R.S. Honorato, M.F. Pimentel, A.P.S. Paim, Anal. Chim. Acta 629 (2008) 98. [25] L.F. Almeida, M.G.R. Vale, M.B. Dessuy, M.M. Silva, R.S. Lima, V.B. Santos, H.G.D. Diniz, M.C.U. Araújo, Talanta 73 (2007) 906. [26] S.K.B. Freitas, V.L. Silva, A.N. Araújo, M.C.B.S.M. Montenegro, B.F. Reis, A.P.S. Paim, J. Braz. Chem. Soc. 22 (2011) 279. [27] M. Grünhut, V.L. Martins, M.E. Centurión, M.C.U. Araújo, B.S.F. Band, Anal. Lett. 44 (2011) 67. [28] S.S. Monte-Filho, M.B. Lima, S.I.E. Andrade, D.P. Harding, Y.N.M. Fagundes, S.R.B. Santos, S.G. Lemos, M.C.U. Araújo, Talanta 86 (2011) 208. [29] C. Pasquini, E.V. Aquino, M.V. Rebouças, F.B. Gonzaga, Anal. Chim. Acta 600 (2007) 84. [30] M.B. Silva, C.C. Crispino, B.F. Reis, J. Braz. Chem. Soc. 21 (2010) 1854. [31] R.A.C. Lima, S.R.B. Santos, R.S. Costa, G.P.S. Marcone, R.S. Honorato, V.B. Nascimento, M.C.U. Araújo, Anal. Chim. Acta 518 (2004) 25. [32] R.S. Honorato, E.A.G. Zagatto, R.A.C. Lima, M.C.U. Araújo, Anal. Chim. Acta 416 (2000) 231. Author's personal copy Trends in Analytical Chemistry, Vol. 35, 2012 [33] E.P. Medeiros, E.C.L. Nascimento, A.C.D. Medeiros, J.G. Veras Neto, E.C. Silva, M.C.U. Araújo, Anal. Chim. Acta 511 (2004) 113. [34] S.C.B. Oliveira, E.C.S. Coelho, T.M.G. Selva, F.P. Santos, M.C.U. Araújo, F.C. Abreu, V.B. Nascimento, Microchem. J. 82 (2006) 220. [35] H. Fraiser, G.H. Nancollas, Compendium on Analytical Nomenclature, Blackwell, Oxford, UK, 1987. [36] C.C. Acebal, M. Insausti, M.F. Pistonesi, A.G. Lista, B.S.F. Band, Talanta 81 (2010) 116. [37] K. Danzer, L.A. Currie, Pure Appl. Chem. 70 (1998) 993. [38] L.F. Almeida, V.L. Martins, E.C. Silva, N.T. Moreira, M.C.U. Araújo, Anal. Chim. Acta 486 (2003) 143. [39] L.F. Almeida, V.L. Martins, E.C. Silva, N.T. Moreira, M.C.U. Araújo, J. Braz. Chem. Soc. 14 (2003) 249. [40] V.B. Nascimento, T.M.G. Selva, E.C.S. Coelho, F.P. Santos, J.L.S. Antônio, J.R. Silva, E.N. Gaião, M.C.U. Araújo, Talanta 81 (2010) 609. Trends [41] J.A. Nascimento, A.G.G. Dionı́zio, E.C.L. Nascimento, S.K.B. Freitas, M.C.U. Araújo, Quı́m. Nova 33 (2010) 351. [42] E.C.L. Nascimento, M.C.U. Araújo, R.K.H. Galvão, J. Braz. Chem. Soc. 22 (2011) 1061. [43] V. Visani, S.R.R.C. Barros, H.A. Dantas Filho, L.F. Almeida, R.A.C. Lima, W.D. Fragoso, T.C.B. Saldanha, M.C.U. Araújo, Eclét. Quı́m. 34 (2009) 37. [44] M. Grünhut, M.E. Centurión, W.D. Fragoso, L.F. Almeida, M.C.U. Araújo, B.S.F. Band, Talanta 75 (2008) 950. [45] M.C. Souza, V.L. Martins, L.F. Almeida, O.D. Pessoa Neto, E.N. Gaião, M.C.U. Araújo, Talanta 82 (2010) 1027. [46] M.B. Lima, M. Insausti, C.E. Domini, M.F. Pistonesi, M.C.U. Araújo, B.S.F. Band, Talanta 89 (2012) 21. [47] J.A. Belloni, M.S. Di Nezio, M.F. Pistonesi, M.E. Centurión, Talanta 89 (2012) 526. http://www.elsevier.com/locate/trac 49