Selective Recovery of a Nitrophenolic by-Product From Nitration Waste Water by Extraction

The selective recovery of certain nitrophenolic by-products, namely 2,4-DNP, picric acid or dinitrocresols, is illustrated utilizing dinitrocresol contaminated nitration waste water and non-cresol contaminated nitration waste water. The waste water utilized resulted from the production of nitrobenzene by an adiabatic nitration process. Ethylbenzene was utilized as the solvent in the extraction step.

It is noted that the isothermal or conventional nitration process does not produce picric acid, but does produce about 10% to 20% of the 2,6-dinitrophenol (2,6-DNP) isomer and some mono-nitrated phenols. The adiabatic process produces about 1% to 6% 2,6-DNP. This 2,6-DNP is generally present as the main impurity in the 2,4-DNP. Some isolation of the 2,6-DNP from the 2,4-DNP will occur when a preliminary extraction or precipitation is performed, such as used to remove cresols from the waste water as discussed above. Due to the low concentration of 2,6-DNP, i.e. less than about 500 parts per million (ppm), in the feed waste water stream as compared to the presence of about 5,000 ppm of 2,4-DNP and 5,000 ppm of picric acid, the solubility of the 2,6-DNP is not greatly exceeded when acidification is preformed. The fact that the solubility limit is not exceeded, serves to reduce the amount of 2,6-DNP which appears in the 2,4-DNP product. The normal amount of 2,6-DNP present in the 2,4-DNP is from about 1% to 4%.

The first selective recovery was of dinitrocresols from nitration waste water. Nitrobenzene was essentially removed from the waste water by solvent extraction prior to being subjected to the process of the present invention.

The cresol concentration of the cresol contaminated nitration water utilized was initially 97.4 ppm. The pH of the nitration waste water was adjusted from an initial solution pH of 9.54 to a pH of 3.7 by adding sulfuric acid to the waste water. The water was then subjected to extraction. Following the extraction and recovery of the dinitrocresols, it was determined that the cresol concentration in the waste water had been reduced to 13.2 ppm. It is noted that when a pure dinitrocresol product is to be recovered, it preferably should be done from nitrotoluene waste water which has no dinitrophenols and a high concentration of dinitrocresols contained therein. The cresol concentration continued to decrease in the water layer as the acidity of the water was lowered. As the pH, however, reached the lower acid pH levels, a substantial amount of 2,4-DNP and/or picric acid was also extracted or pulled out of the water layer and accordingly, the purity level of the recovered cresol was lower.

Following the extraction of the dinitrocresols, the water layer was recovered so that 2,4-DNP and picric acid could then be selectively recovered from the same nitration water. The recovered water was subjected to a second pH adjustment by adding sulfuric acid in an amount sufficient to bring the pH of the water to a level of 2.5. The same nitration waste water after extraction of the pH adjusted water resulted in the reduction of the concentration of 2,4-DNP in the waste water from an initial concentration level of about 5,000 ppm to a concentration of less than about 100 ppm.

Thereafter, to selectively recover picric acid from the same nitration waste water, the water layer from the second extraction was recovered and its pH adjusted to a level of 0.6. Following extraction, the nitration waste water resulted in the reduction of the concentration of picric acid in the waste water from an initial concentration level of from about 5,000 ppm to a concentration of less than about 100 ppm.

The non-cresol contaminated nitration waste water was essentially stripped of nitrobenzene by extraction before subjecting the waste water to the process of the present invention. Through the adjustment and control of the waste water pH, the concentration of 2,4-DNP and picric acid in the nitration waste water was reduced and a substantially pure form of 2,4-DNP and picric acid individually recovered. The recovery results following pH adjustment to varying levels and extraction are set forth below.

Following adjustment of a nitration waste water's pH to 4.5 and extraction of the pH adjusted water, the concentration of 2,4-DNP in the nitration waste water was substantially reduced. Approximately 65% of the 2,4-DNP present in the waste water was removed from the water layer in a substantially pure form. To recover the remaining 2,4-DNP, i.e. approximately 35% of the original concentration, so that it would not be recovered with the picric acid, the pH is again lowered to a pH of about 2.0 to 1.6 and a second extraction performed. This second extraction removes the remaining 2,4-DNP thereby allowing for the subsequent extraction of picric acid in a relatively pure state. This method yields a picric acid product which is greater than 99% pure. The water's pH is adjusted to be from about 0.5 to 1.2 for the extraction of the picric acid. Preferably, the pH is adjusted to 1.0. This final extraction removes the picric acid from a concentration level of approximately 5,000 ppm down to a level of less than about 100 ppm.

When the pH of a separate sample of nitration waste water was adjusted to 2.5 and the water subjected to extraction with the ratio of water to solvent being 5:1, the recovered solvent layer contained a composition of 99.8% 2,4-DNP and 0.2% picric acid.

Extraction utilizing a 5:1 water to solvent ratio of another sample of nitration waste water following adjustment of the pH to 4.25 resulted in the recovery of a solvent layer containing 2,4-DNP at a high purity level. More specifically, analysis of the extract at a pH level of 4.25 following the evaporation of the solvent from the organic layer revealed a material composed of 97.3% 2,4-DNP and 2.7% picric acid.

For the analysis of the recovered material, the solvent wa removed from the organic extract layer using a roto-evaporator. Thereafter, the remaining solute was dissolved in deionized water. Analysis of the extracts was performed by liquid chromatography techniques using a 40% methanol/60% 4.5 pH water solvent system. The flow rate through the column was one ml/min. Detection was performed at 254 nm. An Alltech 5-micron, Econosphere C-18 column, 4.6 mm.times.25 cm was utilized with the liquid chromatography system. Injections were generally in microliters. Data reduction was done with a Spectraphysics Model 4270 integrator.

During the analysis of the picric acid recovered from the cresol contaminated water the concentrations initially appeared in an inconsistent manner. It was found that by treating the water samples with sulfamic acid prior to injection, that the sulfamic acid assisted in removing any nitrous acid present in the sample. The retention time of the nitrous acid is nearly identical to that of picric acid and thereby caused peak overlapping problems. After the samples were treated with sulfamic acid and subjected to liquid chromatography analysis, consistent and expected levels of picric acid were observed.

As will be apparent to one skilled in the art, various modifications can be made within the scope of the aforesaid description. Such modifications being within the ability of one skilled in the art form a part of the present invention and are embraced by the appended claims.