Potassium Sulfate - an overview

Kjeldahl nitrogen

For a long time organically bound trivalent nitrogen was measured according to the method of Kjeldahl, where the water sample is evaporated to dryness and digested with sulfuric acid. Potassium sulfate is added to elevate the destruction temperature. The destruction residue is neutralized and ammonium is measured to establish the Kjeldahl nitrogen content. To calculate ‘organic nitrogen’, the result of the determination of Kjeldahl nitrogen has to be corrected for the amount of ammonium present in the sample. Kjeldahl used mercuric ions to catalyze the efficient oxidation of protein-like substances. The intoxication caused to humans in Japan in the 1953–60 periods after ingestion of mercury-containing fish from the Minamata Bay gave an impulse to highly reduce the use of mercury in the laboratories. Through this incident, first mercury was replaced by cupric ions, and later, selenium was used as catalyst, although their effectiveness remained controversial. Next the use of catalysts was abandoned in favor of potassium peroxodisulfate. In this procedure sulfuric acid is added to the sample and the sample is evaporated. After cooling potassium peroxodisulfate is added for the destruction of organic matter. The residue then had to be warmed to a temperature of 370°C for completion of the destruction. The determination is completed by the measurement of ammonium. Despite its rationalization, this method remains very laborious. An alternative for the measurement of total nitrogen is the combustion of the sample in an argon–oxygen atmosphere in the presence of ceric oxide at 1000°C. The produced nitrogen oxides are converted to nitrogen oxide, which can be determined electrochemically or by chemoluminescence. In electrochemical detection, nitrogen oxide reacts with the nitrate ion at the working electrode. Chemiluminescent detection is achieved when nitrogen oxide reacts with ozone to form nitrogen dioxide with the emission of light. The amount of light, measured with a photomultiplier, is proportional to the amount of nitrogen present in the sample. This measurement takes about 3  min per sample.

Phosphate is mostly determined photometrically through the reaction with ammonium heptamolybdate in acidic medium in the presence of potassium antimonyl tartrate as a catalyst. 12-Molybdophosphoric acid is formed that upon reduction with, for example, ascorbic acid has a blue color. The lower limit of detection is 10   μg   l−1. Chromium(VI), nitrite, and high concentrations of chloride and iron interfere. The lower recoveries caused by iron are eliminated by addition of hydrogensulfite.

Carbonate and hydrogencarbonate are determined by an acidimetric titration to endpoints of pH 8.35 for carbonate and pH 4.35 for hydrogencarbonate. The indicator electrode is usually a glass electrode that allows the pH to be followed potentiometrically. In one mode, the titrant (0.01   mol   l−1) is added at a constant rate and the pH is monitored continuously. Processing of the recorded data yields the endpoints. Automation of the determination is possible but expensive.

In modern laboratories fluoride is almost exclusively determined with a fluoride-selective electrode after addition of a total ionic strength adjustment buffer with citrate and cyclohexanediaminetetraacetic acid as decomplexing agents. The electrode signal suffers little interference from other ions. When fluoride has to be determined in concentrations as low as 1   μg   l−1, application of the electrode in a CF apparatus is advantageous. Decomposition of metal–fluoride complexes is difficult and high recoveries are required (above 80%). Quality control is necessary. It may even be necessary to carry out the decomposition of the complexes outside the analyzer in order to increase the time available for the reaction. The method suffers interference from extreme pH and multicharged cations such as Fe3+ and Al3+.

Ammonium is determined in many laboratories in a CF system in which the Berthelot reaction is implemented. In the Berthelot reaction, ammonium reacts with chlorine and phenol in the presence of sodium nitroprusside as catalyst in alkaline medium. EDTA is added to prevent interference of calcium and magnesium. Modern systems have been developed that use macroporous polytetrafluoroethylene (PTFE) membranes. In these systems a sample is introduced into a stream to which sodium hydroxide solution is added. Ammonia diffuses through the PTFE membrane into a stream of de-ionized water and the stream is fed through the flow-through cell of a conductivity meter. In this system a minimum of reagents is required and the only interference is from volatile amines.

The determination of hydrogen ions is important for rainwater. Hydrogen ion concentration is often calculated from the pH. However, this may increase the uncertainty (pH-to-hydrogen ion conversion is a logarithmic function) and decrease the accuracy because of the inaccuracy in the value of the activity coefficient (which is affected by temperature and ionic strength) that is required for the conversion from a measured activity to a calculated concentration. pH measurement therefore requires careful attention to the measuring conditions (temperature, ionic strength, avoidance of contact and diffusion potentials, etc.). Under such circumstances procedures such as Gran's plot or titrimetric methods are preferred to direct potentiometry.

Sodium and potassium are mainly determined by flame atomic emission spectrometry (FAES), flame atomic absorption spectrometry (FAAS), and inductively coupled plasma atomic emission spectrometry (ICP-AES). FAES is widely used for the determination of potassium and sodium, but interference is experienced from stray light. In FAES and FAAS, ethyne (acetylene) and air are generally used as the flame gases. After acidifying the samples with nitric acid, an electron buffer (cesium chloride or lanthanum nitrate) is added to suppress interference by phosphate. When low concentrations (below 50   μ   l−1) of potassium (in rainwater or boiler water) have to be determined, atomic absorption spectrometric methods are usually applied. For determinations at such low concentrations, sample containers and the nebulizer system must be scrupulously cleaned by soaking for ∼24   h in dilute nitric acid. FAAS is therefore preferred for low-concentration measurements. For sodium and potassium measurement in particular, simplified and cheaper AAS instruments can be used that are provided with a light source that is monochromatized with a filter instead of the more sophisticated hollow cathode lamp (HCL) or electrodeless discharge lamp (EDL). In matrix-free water samples such as boiler water, sodium can also be measured down to 1   μg   l−1 with an ion-selective electrode. The samples are not pretreated in most applications of ICP-AES. Samples with a high salt content can be reproducibly introduced with an ultrasonic nebulizer or must be diluted appropriately.

Calcium and magnesium are often determined by FAAS. Nitrous oxide and ethyne (acetylene) are the best flame gases. Prior to measurement the sample is acidified with nitric acid. Lanthanum nitrate is added to suppress interferences by oxidizible acids, aluminum and organic complexes of aluminum, phosphate, silicate, iron, etc. Strontium chloride can be used instead of lanthanum nitrate. To prevent ionization in the flame, cesium chloride, potassium chloride, or sodium chloride can be added. No pretreatment is used when the determinations are carried out with ICP-AES; because of the high temperature of the flame, all calcium phosphate compounds are decomposed. If a determination is needed only occasionally, calcium and magnesium can be determined titrimetrically with ethyleneglycol-bis(2-aminoethyl)-N,N,N′,N′-tetraacetic acid (EGTA) and Eriochrome Black T as indicator in alkaline medium. Magnesium is titrated first. Next, magnesium–EGTA complex is added and calcium displaces magnesium from this complex. The liberated magnesium is titrated as a measure of the calcium in the sample. Cyanide is added to overcome any effect of traces of iron and heavy metals.

Iron and manganese can be determined with FAAS; air and ethyne are used as flame gases. Prior to the determination of total iron and manganese, digestion with aqua regia in a microwave oven for 1   h is required. This treatment takes care of refractory compounds such as iron oxides. Iron and manganese can also be determined with ICP-AES, in which case no predigestion is required owing to the high temperature of the flame.

Silicate is determined spectrophotometrically with ammonium molybdate and ammonium vanadate. The pH of the sample must be adjusted to 7–8. Potassium cyanide is added to prevent interference of heavy metals. Oxalic acid is added to destroy molybdophosphate and vanadophosphate and to bind aluminum in a complex. As in all spectrophotometric determinations, high and variable optical absorption of the sample (due to color or turbidity) at the wavelengths of investigation causes errors; tannin, iron, and sulfide also interfere. To avoid contamination, all contact surfaces should be of polyethene. It is also possible to determine silicate using FAAS, in which case nitrous acid and ethyne must be used as flame gases. As silicates are present in colloidal form, the sample must be introduced into the AAS equipment using an ultrasonic nebulizer. Such a nebulizer is also used when silicate is measured by ICP-AES.