научная статья по теме OILS AND GREASE DETERMINATION BY FT-IR AND N-HEXANE AS EXTRACTION SOLVENT Химия

Текст научной статьи на тему «OILS AND GREASE DETERMINATION BY FT-IR AND N-HEXANE AS EXTRACTION SOLVENT»

ЖУРНАЛ АНАЛИТИЧЕСКОЙ ХИМИИ, 2015, том 70, № 3, с. 298-301

ОРИГИНАЛЬНЫЕ СТАТЬИ =

УДК 543

OILS AND GREASE DETERMINATION BY FT-IR AND я-HEXANE

AS EXTRACTION SOLVENT © 2015 Remo Bucci*, Anna Maria Girelli*, 1, Stefano Tafani*, Anna Maria Tarola**

*Department of Chemistry, Sapienza University of Rome P.le A.Moro5, 00185Rome, Italy 1E-mail: annamaria.girelli@uniroma1.it **Department of Management, Sapienza University of Rome Via del Castro Laurenziano 9, 00161, Rome, Italy Received 19.06.2012; in final form 21.01.2014

Recently FT-IR spectroscopy has enjoyed renewed interest for detection and quantification of contamination by oils and grease (OG), especially in water samples. This is due to the development of a new approach using less harmful solvents. This paper presents the development of a new FT-IR method that uses n-hexane as a substitute to Freon 113 in the extraction process. This solvent has a high solubility for the desired organic compounds, low miscibility with water and a low boiling point to facilitate its removal from the extracted material. Results regarding the calibration curves, recovery, precision, detection and quantitation limits are presented. Even if the recovery is not too high (45%), the preconcentration factor of 2 x 103 permits the detection of OG at levels >83 ^g/L.

Keywords: oils and grease, FT-IR, K-hexane.

DOI: 10.7868/S0044450215030184

The concentration of dispersed OG is an important parameter for water quality and safety. OG in water can be present as free floating, dissolved, emulsified or adsorbed to suspended solids. Therefore, the level of dissolved oxygen can be reduced and the quality ofwa-ter for drinking purposes may be affected. To a large extent OG in water originate from petroleum products and so their limits have been established of 0.01 mg/L for mineral water [1], 0.2 mg/L for surface waters intended for drinking water supplies [2] and 40 mg/L for discharges water in sewer [3].

In the determination of OG, absolute quantities of specific substances are not measured. Rather, groups of substances with similar physical characteristics are determined quantitatively on the basis of their common solubility in an organic extracting solvent. The 13th edition of Standard methods prescribed the use of 1,1,2-trichloro-1,2,2-trifluoroethane (Freon 113) [4] as extracting solvent for all sample types. Once extracted, these could be quantitatively determined by gravimetric or infrared (IR) methods, to provide the oil total content in water. Prior to the Montreal Protocol, IR analysis using Freon 113 was a widely used laboratory method [5, 6]. However, because of environmental problems associated to chlorofluorocarbons, it was banned by 1996; successively ,under Montreal Protocol, the use of tetrachloromethane as alternative solvent became illegal in 2010. Carbon tetrachloride is

actually not employed for its toxicological effect. Recently tetrachloroethylene (C2Cl4) has been proposed as extraction solvent alternative in OG determination at 0.1 mg/L [7, 8], when Fourier transform infrared spectroscopy (FT-IR) is employed. In the 20th edition of Standard methods, Freon 113 [9] has been replaced by K-hexane for the gravimetric procedure. Thus, it defines OG as "any material recovered as a substance extracted by K-hexane from an acidified sample". In 1999, the U.S. EPA promulgated a gravimetric method [10], with the aim to include silica gel treatment for the separate determination of total petroleum hydrocarbons after the OG extraction by K-hexane. However the gravimetric method presents such disadvantages as high detection limits (5—10 mg/L) and interference by compounds other than OG in the final weight. K-Hexane is also widely employed in Gas chromatographic procedures, where the profile of C10—C40 hydrocarbons in amounts higher than 0.01 mg/L can be obtained. The principal inconvenient of this technique is the difficulty to express the results obtained in terms of total OG [11].

In the IR technique, which takes in consideration the total number of hydrogen and carbon bonds, pure K-hexane would contribute to the oil and grease concentration measurement as it gives absorption between 3000 and 2900 cm-1. Therefore, even if the K-hexane

OILS AND GREASE DETERMINATION BY FT-IR AND k-HEXANE

299

is an appropriate extraction solvent, it results non suitable for the OG analysis by IR method. When FT-IR spectrophotometers replaced conventional dispersive spectrophotometers, a high increase in sensitivity was reached [12, 13] making FT-IR the most sensitive method but with the big limitation in Freon 113 use.

Therefore, the aim of this paper is to investigate on the possibility to validate a new method suitable especially for the laboratories that want to continue using IR methods and are forced to use an alternative solvent to Freon 113. Taking into account that currently K-hexane is the recommended extraction solvent because it gives results not statistically different from those obtained with Freon 113 [14] and that FT-IR is the most sensitive technique, a method combining this features is proposed. In this aim the evaporation of K-hexane under nitrogen flux prior to IR measurements and the dissolution of dry extracts in a not banned solvent without C—H bonds, as tetrachloroethylene, is performed.

EXPERIMENTAL

Materials. Tetrachloroethylene 99+% ACS spectrophotometry grade, K-hexadecane and isooctane, were purchased from Sigma Aldrich (Milan, Italy). Chlohridric acid was from Carlo Erba (Milan, Italy), K-hexane from Sigma.

Apparatus. All the measurements are made by spectrophotometer FT-IR Perkin Elmer FT 1600. The spectra are monitored in the range of 2700—3200 cm-1 after 16 scans employing quartz cells of1 cm path length with capacity of 1 L or 0.4 mL. Backgrounds are indicated each time from the analysis.

Calibration standard solution preparation. A calibration stock solution is obtained by mixing equal volumes of isooctane and K-hexadecane. When not in use, it is stored in a glass bottle at 4°C and remains stable for six months. This stock solution, was further diluted by weighing an adequate quantity of sample in a flask of10 mL brought to volume with tetrachlorethyl-ene (1.2 g/L). Such solution remains stable for three months. Calibration standards were, daily fresh prepared from the dilute stock solution

Aqueous samples preparation. The calibration stock solution constituted by the mixture isooctane — K-hexa-decane was diluted by weighing adequate amounts in a flask of 10 mL brought to volume with K-hexane (1.2 g/L). Spiked samples were freshly prepared by adding known volumes of this mixture to 1000 mL of samples (pure water, mineral or surface water).

Extraction from aqueous samples by n-hexane. The

aqueous samples, previously acidified with 5 mL of HCl in order to obtain a pH lower than 2, are placed in a 1 L Erlenmeyer flask. Subsequently, prefixed volumes of K-hexane are added. The solution in the flask that is closed by a lamellar plug is maintained for 30 min under mechanical agitation favouring the dispersion of micro-droplets in aqueous volume. This en-

A

Wavenumber, cm 1

Fig. 1. FT-IR spectra of calibration standards against blank solvent. In inset: hexadecane and isooctane (pure) against tetrachloroethylene solvent. Path cell is 1 cm.

hances extracting power of the organic phase, which rests on top of the liquid surface. After 15 min of resting time, a microextractor, as that reported by Zocco-lillo et al. [15], with two parallel tubes is introduced to collect the organic phase. The addition of water into the side tube causes the liquid to rise until the hexane micro layer is channelled into the upper part of the second tube. The same procedure was applied for both spiked and real samples. Successively a prefixed aliquot of organic phase is taken and submitted to N2 flux until it reaches the dryness. The residue, dissolved in tetrachloroethylene, is placed into a quartz cell (1 cm) and finally measured in FT-IR spectrometer at 2925 cm-1.

RESULTS AND DISCUSSION

Petroleum hydrocarbons and mineral oils all contain carbon—hydrogen bonds, thus giving rise to C—H stretching absorption in the range between 3200 to 2700 cm-1 of the IR spectrum such as vibrations in CH2 groups at 2930 cm-1, in CH3 groups at 2960 cm-1 and in aromatic C-H bonds at 3010-3100 cm-1. Figure 1 is reported the IR spectrum of a standard solution in tetrachloroethylene (TCE) where are shown the stretching absorbance maxima at 2925 and 2958 cm-1. Both bands are present in the spectra of pure K-hexa-decane and isooctane (see the inset in Fig. 1) but, obviously, with different sensitivities. In the spectral region are also present peaks at 2870 and 2855 cm-1 corresponding to the symmetric stretching vibrations. Consequently the quantity of OG can be estimated by measuring the peak height at the absorbance maxima or area values under the IR peaks. Both measurements depend, of course, on the baseline chosen, which is fixed, as seen in Fig. 1, between two absorbance minima at 2985 and 2820 cm-1.

Taking into account that the oils, like diesel, machine, sunflower, olive oil etc. contain CH2 groups,

300

REMO BUCCI и др.

120 100 тз 80

<D N

13 60 £

S3 40 £

20

A 0.07

2 3 4 5 6 7 8 Volume of n-hexane, mL

9 10

У = 2.25 x 10-4x - 4.54 x 10-3

Fig. 2. EfTect of «-hexane volume on normalized peak ab-sorbance value at 2925 cm-1 (A/Amax).

they will absorb at 2925—2930 cm In addition it is well known [16] that the absorbivity of CH3 stretching at almost 2960 cm-1, even if depending on the kind of compound, is generally lower than that of CH2 stretching. For this reason in this work, as in the other IR published methods [8], the operative wavenumber is fixed at 2925 cm-1.

Effect of extraction volume. Preliminary measurements were carried out at room temperature, on 500 mL of an aqueous acidified sample containing 261 ^g/L of a hexadecane and isooctane mixture. The liquid-liquid extractio

Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.

Показать целиком