3.2.9.3. Turbidimetry

Method and Instrumentation

There are two differences between the instrumental setups for nephelometry and turbidimetry: Illumination of the flow cell (B right), and the programming of microfluidic manipulations and data collection performed by the miniSIA 2 instrument (K). The key component of the flow cell is 100 cm long, 0.8 mm I.D. straight tubing made of green PEEK. It serves as a collimator directing a beam of light from the tungsten halogen lamp (THL) atop the tubing into the CCD detector on the bottom of the flow cell (B right). In this way, 50 μL of suspension within the flow cell are monitored in absorbance mode analogous to spectrophotometry. Therefore, the “visible light’ (400 to 800 nm) absorbed by material in the flow cell yields absorbance A related, in this work, to NTU via calibration with the formazin standard. (Construction of this Long Light Path flow cell is in Sec. 3.2.8.)

Similar to nephelometry, the flow protocol (K) comprises two parts: 1) autocalibration with a 20 NTU standard solution, 2) transport of suspension into the flow cell. However, while the autocalibration step is identical for both methods, monitoring of turbidimetry is done during the STOP flow period, on a selected section of bead suspension that has been arrested within the flow cell. Therefore, volume and flow rate of carrier solution (Pump 2) during the second step, is chosen to position the centroid of sample zone into the middle of flow cell, to maximize sensitivity of determination (Detailed explanation is in Sec. 3.2.3.). The flow protocol is executed by software program (L) that also performs data collection and processing.

Since the spectrophotometer will monitor radiation absorbed by suspended particles, it has to be set up to function in absorbance mode (M).

Results

Data collected from automated serial dilution of 20 NTU standard prepared in DI water are presented in three formats: spectra versus concentration (N), time versus concentration (O), and resulting calibration graph (P).

 

Over the entire spectrum range, the absorbance A is directly proportional to the concentration reaching maximum at 450nm (N), when signal becomes noisy due to the diminishing intensity of the THL source and low sensitivity of CCD sensors in this range. Interestingly, this corrupted peak is not specific for Formazin since it is also present on turbidity spectra of particles of Suspended Matter. We decided to monitor absorbance at 550 nm because the HT lamp light is most intense at this wavelength and because this wavelength complies with EPA norm, however wavelengths up to 850nm would be suitable.

It follows, that in variance with common practice (EPA, ISO) that conditions of turbidimetric determination should NOT be specified by wavelength of light source, but in analogy with spectrophotometry, should be aways specified by wavelength of monitored radiation.

Absorbance values versus time (O) increase sharply from baseline (BS) at the 13 seconds mark as the leading edge of suspension enters the flow cell from below. Upon reaching a maximum at 22 seconds, the flow is stopped for 15 seconds during which the suspension is arrested in the flow cell. Since during this period the absorbance (A) seems to decrease, the data collecting window (WIN) used to construct the calibration graph (P) is placed towards the end of the stop flow period. As the flow resumes, suspension is flushed from the flow cell and absorbance reaches baseline.

The calibration graph (P) in the range 0 to 20 NTU is strictly linear, of excellent reproducibility resulting in low limit of detection (LOD) and low blank values making stop flow turbidimetry well suited for the determination of suspended matter (SPM) in sea water. However, the influence of salinity of SW must now be investigated.

Sea Water calibrations for turbidimetry were obtained in the same way as was done for nephelometry by using Simulated Sea Water made of 3.5% NaCl in DI, which approximates salinity in open ocean water (35 PSU). Therefore, the 3.5% NaCl was used to prepare a 20 NTU standard and also was used as the carrier supplied by pump 2. In this way, the SSW based formazin standard was stepwise diluted with SSW. The data obtained by SSW calibration are presented in three formats: 1) spectra versus concentration (R), 2) time versus concentration (S), and 3) resulting calibration graph (T).

The spectra of SSW based calibration (R left) are identical to the spectra obtained with DI water (N), while at 550 nm the values are the same for SSW and DI spectra (R right).

There is, however, a large difference between time/concentration plots, recorded with formazin standards prepared in DI (O) and SSW (S). These differences are caused by difference in Refractive Index between DI (carrier pump 1) and SW (carrier pump 2 and sample), that distort absorbance A in response to schlieren effect. However, by placing WIN at a 30 to 33 seconds position, well away from the refractive index position, its interference is eliminated. Therefore, DI and SW based calibration data (P and T) are identical, while the slopes of DI/SW calibration lines differ only by 2%.

In conclusion, nephelometry and turbidimetry in programmable FI format yield, with formazin standards, well reproducible data, strictly linear calibration lines which differ by only by 2% between fresh water of 0 PSU and highest salinity water of 35 PSU in open ocean. Therefore, formazin in DI calibration can be used for analysis of SW samples collected in estuaries where salinity varies daily and seasonally.