Chapter 2:
Advantages of Fluorescence

Chapter 3:
Instrumentation

Chapter 4:
Variables of Fluorescence

Chapter 5:
Calibration and Standards

4.11 Linearity. Fluorescence intensity is typically directly proportional (linear) to concentration. There are, however, factors that affect this linear relationship. When concentration is too high, light cannot pass through the sample to cause excitation; thus very high concentrations can have very low fluorescence (concentration quenching; see section 4.12). At intermediate concentrations, the surface portion of sample nearest the light absorbs so much light that little is available for the rest of the sample; thus the readings will not be linear, though they will be within the range of a calibration curve.[18] See Figure 6.[19]

The linearity of a sample is related to many factors, including the chemical composition of the sample and the pathlength the light must travel (thus, the diameter of the cuvette or sample holder is a factor). An unknown sample should always be tested for linearity. Because instrument factors also affect linearity, samples must be tested on the specific instrument to be used in the study; if cuvette size is changed or optical filters are changed to different wavelengths, linearity must be tested. To test for linearity, simply take a reading for a high concentration of the sample; dilute by a factor (1:1, 1:10, etc.); and take a reading for the dilute sample. If it is linear, the reading will go down by the same factor as the dilution. If it is non-linear, the reading will most likely increase as it is diluted. If it is in the range for a calibration curve, the reading will go down, but not as much as expected by the dilution. Although a calibration curve for a specific application may prove to be non-linear, if the curve is reproducibly accurate, then the calibration is accurate.

For example, Rhodamine WT, a fluorescent tracer widely used in environmental studies, is linear to about 100 parts per billion (ppb) active ingredient using the Turner Designs 10-AU-005-CE fluorometer with the 25 mm cuvette. It is in the range for a calibration curve to about 500 ppb active ingredient.

4.12 Quenching. The term "quenching" refers to many factors that reduce, or quench fluorescence. Quenching as it relates to linearity is discussed in section 4.11. Temperature quenching is discussed in section 4.2. Factors that influence quenching can be controlled through instrument design and proper method development. For more details on quenching, click here.

Quenching factors are one reason why it is very important to treat standards, blanks, and samples in exactly the same manner. Thus, for the best accuracy, prepare all solutions in the same manner, using the same reagents and matrix solution and the same preparation techniques; measure at the same temperature after the same amount of time. Good laboratory techniques will go a long way toward producing reliable results.

4.13 Turbidity. Fluorescence measurements are significantly more immune to the effects of turbidity compared to absorption techniques like UV/VIS spectrophotometers. Turbidity is often viewed as an impurity or background in a solution (i.e., suspended solid like silt). If the interfering substance is reflective, turbidity can create light scatter and readings will increase. If the interfering substance absorbs light, fluorescence will be reduced. If the interfering substance does not absorb light, however, then the fluorescence readings will not be effected unless there is so much turbidity that the emitted light cannot penetrate the water. In extreme cases of turbidity, samples can be filtered to minimize these effects. For various applications, turbidity can easily be corrected for in the field.[21] Keep in mind that these interferences can in effect be "canceled out" by using the same matrix water for standards, blanks, and samples and treating them exactly in the same manner.

4.14 Bubbles in the Sample. Since fluorescence is such a sensitive measurement, and since accuracy of readings depend on light exciting the molecules of the compound, bubbles in the sample can result in erratic or fluctuating readings. In cuvettes, if bubbles are suspected, wait for them to settle. In continuous flow systems, take care not to introduce bubbles by avoiding turbulent waters and water too near the pump; or employ a filtering system or bubble trap. Sample intake should be through the bottom and outflow through the top. In a falling stream flow cell, bubbles are not an issue as they are part of the background measurement.

4.15 pH. Fluorescence can be affected by pH. Yet, in field studies of water, this is rarely an issue. Where it is an issue, the usual rule applies: use the same matrix water for standards, blanks, and samples and treat them in exactly the same manner. If working with pH-dependent solutions, read solutions at different pH’s to determine affects on fluorescence.

In certain studies, pH factors can be an advantage. Spectral interferences can be minimized in analytical fluorescence spectrometry by adjustment of pH prior to measurement. The pH dependence of probe molecules has been used to determine the pH of intact and cured cells.[22]

4.16 Photochemical Decay. Many fluorescent molecules can be bleached or destroyed by light (fading of dyes in the sun). Ultraviolet light, especially, can cause certain molecules to break down. Fluorescence readings decrease as the molecules are destroyed. Rate of destruction varies depending upon environmental factors, including temperature. Fluorescein, for example, is destroyed rapidly in sunlight. Rhodamine WT, however, is adequately stable for field studies.[23] For chlorophyll measurements, samples and standards need to be kept in the dark until read. All flow measurements should employ opaque delivery hose to minimize photochemical interferences. Glass provides shielding from laboratory light and ultraviolet light; the extent of shielding varies depending on the type of glass.

Fluorescence is affected by changes in temperature. As temperature increases, fluorescence decreases. Guilbault suggests that this is due to an increase of molecular motion with increasing temperature, which results in more molecular collisions and subsequent loss of energy.[24] The temperature coefficient varies depending upon the compound being measured. Using a filter fluorometer, this problem is easily resolved by measuring standards, blanks, and samples at the same temperature. For flow through studies, the researcher can set up a system to measure the temperature of samples, then compensate readings later using a spreadsheet program. Some field fluorometers like the Turner Designs 10-AU-005-CE offer a temperature compensation option, where the fluorometer measures sample temperature, then automatically compensates readings for changes in temperature.

4.31 Cuvette Size. Linearity and detection limits are affected by cuvette size (pathlength). In general, the greater the diameter of the cuvette, the lower the upper end of the linear range, and the lower the limits of detection. For example, using the Turner Designs 10-AU-005-CE with the 25 mm cuvette, Rhodamine WT is linear to 100 ppb with detection limits in the range of 0.01 ppb in DI water. When the 13 mm cuvette is used, Rhodamine WT is linear to 400 ppb with detection limits of 0.04 ppb in DI water.

The range of concentrations to be measured and the detection limits can be optimized by selecting the appropriate size cuvette. This can be especially effective for measuring substances such as aromatic hydrocarbons. Readings can be obtained in the linear range by the appropriate selection of cuvette diameter (3 mm for example) and/or by offsetting the light path slightly or employing a light attenuator (smaller window or slit).

Note that sample volume is not as important as pathlength or diameter of the cuvette. Thus, for certain applications like DNA quantitation where sample volumes are low, special low-volume cuvettes with appropriate sample windows are available.

4.32 Flow Cell Fouling. Fouling of the flow cell will result in lower, inaccurate readings over time. This comes about when the glass or quartz cuvette becomes coated as sample passes through. How long this takes depends on the quality and temperature of the sample water. There are ways to minimize this problem, including regular cleaning of the flow cell and recalibration of the instrument; or employing a filtering system where biomass is a problem. The Turner Designs TD-4100 employs a non-fouling, falling stream detection system that does not use a glass flowcell, and therefore does not require cleaning.

Continue... to Chapter 5