Flow Injection Tutorial

Refractive index recorded by spectrophotometer, varies considerably as the interfaces between carrier ( DI water), reagents, and methanol based eluant pass through a flow cell during the measurement cycle. This sequence of interfaces form a sequence of liquid lenses that focus and defocus the light beam through the flow cell and result in a wide variation of measured absorbance (1.3.14.A.). Fortunately, the uniform diameter of Garth’s cell, identical with that of LOV channel and of holding coils ( 0.8. mm I.D), along with precise control of flow rates, makes this pattern of absorbance variations highly reproducible. This pattern is also temporally well resolved, and thus by precise positioning of data capturing window (WIN) and of baseline definition (BS) time, the influence of variation of refractive index on the absorbance of the target analyte can be completely eliminated.

Reagent blank is, for all methods of trace analysis, the ultimate obstacle in optimizing limit of detection (LOD). Therefore it is unrealistic to estimate the value of LOD based on a noise level of a spectrophotometer ( e.g. +/-1 mAU ) or on experiment, which did not involve all reagents necessary for performing a real life assay.( e.g. LOD of 0.3nM Fe - in 1.3.14.C.) Therefore the combined blank of reagents used in the automated method (DI water, hydrochloric aid, ammonium acetate buffer, ferrozine and methanol was established by recording BLANK and 0 ppB Fe values in all experiments. (DI water and buffer components were purified by deionization, or by isothermal distillation see Hatta 2018). Unfortunately, the reducing agents obtained from several vendors, were all contaminated by iron. Thus 1% solutions of these reductants contained 10ppB Fe in ascorbic acid, 5ppB Fe in sodium sulphite and 6 ppB Fe in hydroxylamine, as determined by measuring blank values of these solutions by the automated method. Since ferrozine reacts selectively with Fe(II) it was necessary make sure that calibration standards contained iron in the reduced form. Addition of 1 mL of 1% hydroxylamine to 2 ppM Fe standard increased its iron content by 2% only, thus making calibration by Fe(II) reliable. This standard was daily serially diluted by purified 0.01N HCl, to obtain low ppB level calibration standards. However, presence of Fe at ppB level in reagent grade reducing agents limits the use of ferrozine method for trace analysis of Fe(II) in real life samples, such as sea water, because trace analysis of total iron content ( Fe ( II+III) requires use of reagents with Fe content at sub ppB concentrations.

Sorbent selection was critical to successful execution of SE method. The mechanical properties of the OASIS HLB reversed phase polymer based sorbent, particle size 60 microns (Waters Corp. Milford MA) facilitated smooth automated handling of bead suspension, including reproducible packing of the microcolumn. Most importantly, the column was reliable held in place because the spherical, uniform size beads were consistently retained by the cladding of optical fiber (1.3.14.A.). In contrast, silica based sorbents used during earlier stage of this work ( C-18, C-8) failed, because their irregular shape and particle size, hindered their microfluidic manipulation and retention in microcolumn by frits or plugs

CONCLUSION: Increase in sensitivity of assay of trace nutrients by means of automated SE is of an increasing importance and of keen interest in oceanography. Hopefully, many advantages of programmable Flow Injection applied to SE , as compared to continuous flow FI or even to Sequential Injection will be recognized thus making pFI-SE an useful and widely applied tool.

Hatta. M, Measures Ch .I., Ruzicka J., Programmable Flow Injection. Principle, methodology and application for trace analysis of iron in a sea water matrix. Talanta Volume 178, 1 February 2018, Pages 698-703