Fibre Chemistry, Vol. 41, No. 6, 2009

METHODS OF INVESTIGATION

DETERMINATION OF ORGANIC SUBSTANCES WITH DIFFERENT BOILING POINTS IN AQUEOUS MEDIA BY CAPILLARY GAS-LIQUID CHROMATOGRAPHY

E. A. Rogova, O. A. Roshchina, T. I. Samsonova, and A. V. Genis

UDC 543.544.45.38

Studies were conducted on separation of organic substances with different boiling points and polarity when simultaneously present in water by capillary gas-liquid chromatography on a Kristall 2000 m chromatograph. The effect of the column phase and temperature conditions on the separation factor and component exit time was investigated. Column temperature programming was used to reduce the component, especially heavy component, retention time. The chromatography regime was selected and a fast method was developed for determining components with different volatility in water during one analysis with sufficiently high reproducibility and reliability of the results. It was shown that the time of the analysis decreased by more than two times with the same accuracy as with the standard methods. The method can be recommended for fast analysis of wastewaters for making a decision on the treatment method.

Capillary gas-liquid chromatography is an efficient method of determining low concentrations of organic substances. In this method, columns (capillaries) are used without any special carrier with the stationary phase applied on the inside of the capillary. This type of column ensures more efficient separation in almost all cases than packed columns and columns with porous sorbents and allows conducting a quantitative analysis in lower concentration ranges. As a result of the studies using capillary gas chromatography, methods were developed for determining contaminants in monomers and solvents and many other organic compounds in waste and natural water [1-4]. These methods include chromatographic conditions (columns, temperature) that allow determining organic substances with different boiling points separately or groups of substances with close boiling points and polarity. At the same time, the process media of the enterprises, including the wastewaters, can contain organic contaminants that strongly differ in boiling point. In this case, several methods are used, and since the conditions of determination usually differ, the equipment must be realigned and calibrated or different chromatographs must be used. This increases the duration and cost of the analysis. For quickly obtaining information on the composition of production media, fast methods that allow obtaining the necessary information after one analysis are required. The instrument base for realization of the chromatographic method now allows expanding its possibilities, including programming of the column temperature. We will examine the possibility of using capillary gas-liquid chromatography and the new generation of gas chromatographs for simultaneously determining the content of compounds with different boiling points in water to analyze the process media of chemical fibre plants. It ensures good separation of contaminants and sufficiently high reproducibility, and reliability of the results.

Scientific-Research Institute of Synthetic Fibres with Experimental Factory, Tver’. Translated from Khimicheskie Volokna, No. 6, pp. 44-49, November-December, 2009. 0015-0541/09/4106-0385 © 2009 Springer Science+Business Media, Inc.

385

We investigated substances which are the basic contaminants in monomers, monomers, solvents, and by-products used in production of fibres by the solution and melt methods. All of the compounds were toxic substances. The maximum acceptable concentration in fishery and agricultural water is from 0.001 to 10 mg/dm3. Their concentration in wastewaters is usually much higher (it sometimes attains hundreds of milligrams per cubic decimeter) and is regulated by municipal services responsible for the state of natural objects. A list of organic compounds used in the study and their boiling points and dielectric constants are reported in Table 1. Special purity or cp reagents and distilled water according to GOST 6709 were used. The chromatographic studies were conducted on a Kristall-2000M chromatograph with computer control. Detection was flame-ionization. The chromatograms were processed with the Chromatech Analytical software package for MS Windows version 3.1 or higher, which formulates a report with indication of the concentrations of the analyzed components expressed in milligrams per cubic decimeter. The sample was introduced with flow separation, which provides a narrow zone at the entry into the column and prevents overloading of the column (Fig. 1); the flow division ratio was 1:21. Part of the carrier gas flow entering the input device is directed upward and washes the membrane. Most of the stream enters the evaporation chamber, which is a quartz liner. In the chamber, the vapor sample is mixed with the carrier gas. At the input into the column, the mixed flow is separated and only a small part (1/21 in the given case) enters the column. Since the homogeneity of the vapor phase in the evaporation chamber, which is a quartz tube, is extremely important in this case, a specially shaped liner (tube) filled with silanized glass fibre, was used. Use of ordinary glass fibre is not expedient, since many of the difficultly volatile components can be partially retained in it, which negatively affects the reproducibility and reliability of the results.

Fig. 1. Diagram of the sample introduction device with flow-dividing valve: 1) membrane; 2) carrier gas; 3) discharge from membrane (fine control valve); 4) evaporation chamber; 5) carrier gas flow mixed with sample taken out of system (fine control valve); 6) capillary column. TABLE 1. Boiling Point and Dielectric Constant of the Organic Compounds Analyzed Substance

Acrylonitrile Acrolein Acetaldehyde Acetone Acetonitrile Dimethylacetamide Dimethyl sulfoxide Dimethylformamide Methanol Methyl acrylate Oxazole

Formula qm2=qmqN qm2=qmqmn qm3qnm qm3qnqm3 qm3qN qm3qnN(qm3)2 (qm3)2Sn mqnN(qm3)2 qm3nm m2q=qmqn#nqm3

HC

O

tb, °C

Dielectric constant

77.5 52.5 20.8 56.5 81 165 189 153.2 64.7 80.5 70

– – 14.820 20.7425 37.420 37.825 45.025 36.7125 32.6525 – –

78.4 197.3 77.1

24.325 34.520 6.0025

CH

HC N

Ethanol Ethylene glycol Ethyl acetate

386

q2m5nm mnqm2qm2nm qm3qnn q2m5

Man-made mixtures with fixed concentration components from working solutions prepared from certified mixtures were prepared for the study. Information on the certified mixtures used for calibrating the chromatograph is reported in Table 2 and the error of the certified values of the prepared mixtures did not exceed 1%. According to the methods in [1-3], A ZB FFAP column is recommended for determining acrylonitrile and acetaldehyde contaminants in water and a ZB WAX column is recommended for determination of dimethylformamide, dimethyl sulfoxide, dimethylacetamide, and ethylene glycol. For this reason, and also because the number of determined components has significantly expanded (Table 1), studies were conducted on separation of organic substances in water in Zebron columns of different polarity: ZB WAX (30 m × 0.25 mm × 0.25 m) with a stationary phase — PEG 20M, and ZB FFAP (50 m × 0.32 mm × 50 m) with a stationary phase — polyethylene glycol, with 2-nitroterephthalic acid. Of the other well-known strongly polar phases, the FFAP phase is advantageously distinguished by its thermal stability (the maximum working column temperature can attain 250-275°C). The ZB WAX column is distinguished by lower polarity; according to the manufacturer's information, its maximum working temperature is 250-260°C. Separation of two adjacent peaks is characterized by separation factor R, which is a measure of the efficiency of the column — it determines the sharpness of the peaks and distance between their maxima. The substances most difficult to separate by chromatography were vapors of low-boiling substances: ethyl acetate—methanol and acetonitrile—oxazole, so that their separation was most indicative in selecting the capillary column. As the data on the effect of the column phase on the efficiency of separation of these vapors (Table 3), the best results on separation of these organic compounds were obtained in the ZB FFAP capillary column. The high polarity of this phase allowed clearly separating the highly volatile components: for ethyl acetate—methanol vapors, separation exceeds 99%, and it is greater than 98% for acetonitrile—oxazole vapors. Since the determinable substances differ significantly in properties, their determination in one analysis required a special approach to selecting the chromatography conditions. Column temperature programming was used to reduce the retention time, especially for heavy components. The possibilities of the Kristall 2000M instrument and Khromatek Analitik 2.5 software allow programming a temperature change in five time intervals with a fixed rate. In our proposed method, we used two temperature isotherms. The studies showed that use of the column temperature programming method in this case does not affect the stability of the retention time of the analyzed compounds. The chromatograms characterizing separation of very volatile substances as a function of the column used are shown in Fig. 2. TABLE 2. Information on Certified Mixtures Used for Calibration of the Chromatograph Substance

Standard document for substance

Rating of substance

Acrylonitrile Acrolein Acetaldehyde Acetone Acetonitrile Dimethylacetamide Dimethyl sulfoxide Dimethylformamide Methanol (carbinol) Methyl acrylate Oxazole Ethanol Ethyl acetate Ethylene glycol

Imported CAS #107-13-01 Imported CAS #107-02-8 Imported CAS #75-07-0 TU 6-09-1707-77 TU 6-09-4326-76 TU 6-09-537 Imported CAS #67-68-5 TU 6-0914-2206-85 TU 6-09-1709-77 TU 2435-003-52470063-2003 Imported CAS #288-42-6 TU 6-09-19-122-86 TU 6-09-667-76 GOST 10164-75

For chromatography For chromatography For chromatography Cp for chromatography Cp for chromatography Cp Sp for chromatography Sp for chromatography Cp for chromatography Cp For chromatography Sp 20-5OP-2 abs. Cp for chromatography Analytically pure

Certified concentration

500 mg/dm3 0.05% 500 mg/dm3 2.5 % 0.5 % 500 mg/dm3 500 mg/dm3 500 mg/dm3 0.5 % 0.25% 0.25% 0.25% 0.25% 1000 mg/dm3

TABLE 3. Effect of Column Phase on Efficiency of Separation of Components and Analysis Time Column

Separation parameter* RAN-O REA-M

ZB WAX 0.8 ZB FFAP 1.4 ______________

1.1 1.2

Total analysis time, min

14 19

*REA-M, RAN-O — separation parameters for ethyl acetate—methanol and acetonitrile—oxazole vapors. 387

a

b

Fig. 2.

Fig. 3.

Fig. 2. Chromatograms of highly volatile components in ZB WAX (a) and ZB FFAP capillary columns (b, first part of the chromatograms): 1) acetaldehyde; 2) acetone; 3) acrolein; 4) ethyl acetate; 5) methanol; 6) ethanol; 7) methyl acrylate; 8) acrylonitrile; 9) acetonitrile; 10) oxazole. Fig. 3. Chromatograms of high-boiling components in the ZB FFAP capillary column (continuation of chromatogram b in Fig. 2): 1) dimethylformamide; 2) dimethylacetamide; 3) dimethyl sulfoxide; 4) ethylene glycol. TABLE 4. Effect of Column Temperature Conditions on Efficiency of Separation of Components and Duration of Analysis t1, C

t2, C

1 70 2 70 3 70 4 70 5 80 6 80 7 80 8 80 9 80 10 90 11 100 12 120 ______________

70 100 120 150 80 100 120 150 200 150 150 150

Conditions

Separation parameter REA/M RAN/O RDMSO/EG

2.1 2.1 2.1 2.1 1.4 1.4 1.4 1.4 1.2 1.0 0.7 –***

1.4 1.4 1.4 1.4 1.2 1.2 1.2 1.2 1.0 –*** –*** –***

–* –* 2.0 1.6 –* –* 2.0 1.7 1.7 1.7 1.7 1.7

Total duration of analysis, min

>120 >100 34.0 25.0 95.0** 50.0** 32.5 19.0 14.5 18.5 17.5 15.5

Notation: t1 — column temperature of exit of highly volatile components; t2 — column temperature of difficultly volatile components; REA/M, separation factor of ethyl acetate— methanol vapors. *The separation parameter was not calculated due to blurring of the ethylene glycol peak because of the high elution time. **Exit time of the last unblurred peak. ***The separation parameter was not calculated due to total overlapping of the peaks. Part of the overall chromatogram of component separation showing the separation of high-boiling substances: dimethylformamide, dimethyl sulfoxide, dimethyl acetamide, and ethylene glycol, is shown in Fig. 3. Special attention was focused on the fact that temperature programming did not worsen separation of the components. The effect of the column temperature regime on the efficiency of separation of the components of the following vapors: highly volatile ethyl acetate—methanol, acetonitrile—oxazole, and high-boiling dimethyl sulfoxide—ethylene glycol. 388

The results of the studies on selecting the chromatography conditions are reported in Table 4. The analysis of the experimental data showed that regimes of changing the temperature of columns 7 and 8 can be proposed from the point of view of completeness of separation of difficultly separated components (approximately 99% total separation). However, from the point of view of the duration of the analysis where the efficiency of separation decreases to 98%, regime 8 is preferred. In this regime, the initial column temperature was 80°C then beginning from 7 min 30 sec from the running time of the analysis, the temperature was increased to 150°C at the rate of 60°/min. The possibility of expanding the range of determinable concentrations to low values was investigated in the selected regime. Due to the higher ratio of the level of the useful signal to the level of fluctuation noise in comparison to packed columns, the concentrations of substances could be determined with reliable accuracy at the level of 0.5 mg/dm3 without concentration of the sample. The upper boundary of determination was 50 mg/dm3. The time from inputting the sample to exit of the most difficultly volatile component did not exceed 20 min. When the substances were concentrated above this level, the sample had to be diluted. The single dilution should not exceed 200 times. The instrument must be calibrated on man-made mixtures of close concentrations after dilution. A unified method of determining these components when simultaneously present in aqueous media was developed based on these studies. The instrument was calibrated with certified mixtures (Table 2). Calibration solutions with a mass concentration of substances in the range of 0.5 to 50 mg/dm3 were prepared for this purpose. There were at least four calibration solutions for each component. A minimum of five chromatograms for each calibration solution was made to obtain the calibration characteristics of the method (Table 5). A series of samples of wastewater provided by Tver’-Vodokanal Ltd. (Table 6, samples 1 and 2) and collected at VNIISV FGUP (sample 3) was analyzed with the method. The accuracy of measuring the mass concentrations of the substances was determined by comparing the results of the analysis of the samples of wastewater and the calibration mixture prepared from certified mixtures with the proposed method and PND F methods in the State Register. As a comparison of the results of the analysis and monitoring of the correctness of executing the measurements of the mass concentrations of the investigated substances, the unified method allows determining the components sought with the same accuracy as the PND F. The comparative data on the analysis time for a mixture of dimethylformamide (DMF), dimethyl sulfoxide (DMSO), dimethylacetamide (DMA), ethylene glycol (EG), acetaldehyde (AA), and acrylic acid nitrile using the unified method and PND F for this purpose. One chromatograph and one ZB FFAP capillary column are sufficient for conducting the analysis with the unified method. In determining dangerous substances with PND F, it is necessary to have either two chromatographs equipped with two detectors and two columns: one ZB FFAP for determining highly volatile AAN and AA and one ZB WAX for determination of DMF, DMSO, DMA, and EG. The data obtained with the Kristall 2000m chromatograph with two columns and two detectors are reported in Table 5. TABLE 5. Capillary Chromatography Calibration Characteristics for Determination of Organic Substances* in Water When Present Together No.

Component

1 Acrylonitrile 2 Acrolein 3 Acetaldehyde 4 Acetone 5 Acetonitrile 6 Dimethylacetamide 7 Dimethyl sulfoxide 8 Dimethylformamide 9 Methanol 10 Methyl acrylate 11 Oxazole 12 Ethanol 13 Ethyl acetate 14 Ethylene glycol ______________

Retention time, sec

386 294 257 283 404 830 1081 753 311 338 409 330 306 1106

Calibration Coefficient

1.6 2.1 2.7 1.6 1.2 52.1 97.9 67.7 1.2 2.6 2.1 1.0 2.8 1.8

Standard deviation

4.8 1.9 5.6 5.7 7.0 6.2 6.3 5.5 7.7 3.3 7.0 8.0 3.4 3.1

*Concentration range of 0.5-50 mg/liter.

389

TABLE 6. Comparative Data on Analysis of Contaminants in Samples of Water by PND F Methods and the Unified Method

Object analyzed

Calibration mixture Sample No.1 Tver’Vodokanal Ltd. Sample No.2 Tver’Vodokanal Ltd. Sample No. 3, VNIISV FGUP

Component determined

PND F methods

Found in sample with PND F, mg/dm3

Acetaldehyde Dimethylformamide Dimethylacetamide Dimethyl sulfoxide Ethylene glycol Acetaldehyde Dimethylacetamide Ethylene glycol Acetone Methanol Ethanol Ethylene glycol Dimethylformamide Acrylic acid nitrile Ethylene glycol

PND F 14.1:2.230-06 PND F 14.1:2.228-06 PND F 14.1:2.228-06 PND F 14.1:2.228-06 PND F 14.1:2.250-08 PND F 14.1:2.230-06 PND F 14.1:2.228-06 PND F 14.1:2.250-08 PND F 14.1:2:4.201-03 PND F 14.1:2:4.201-03 PND F 14.1:2.250-08 PND F 14.1:2.228-06 PND F 14.1:2.230-06 PND F 14.1:2.250-08

5.15±1.29 1.04±0.21 1.06±0.21 1.30±0.26 41.4±8.3 0.74±0.22 <0.5 0.48±0.14 0.84±0.25 5.70±1.14 0.58±0.17 13.7±2.7 0.55±0.16 0.67±0.08

found in sample, mg/ dm3

5.20±1.30 1.00±0.20 1.10±0.22 1.25±0.25 40.9±8.2 0.73±0.22 <0.5 0.45±0.14 0.88±0.25 5.70±1.14 31400±3100 0.55±0.16 13.5±2.7 0.58±0.17 0.66±0.08

Unified method additive found in incorposample with rated, additive, mg/ mg/dm3 dm3

5.00 1.47 1.48 2.08 20.0 0.70 0.73 2.37 0.85 5.20 30000 0.50 15.0 0.50 0.50

10.22±2.04 2.48±0.50 2.74±0.55 3.25±0.65 60.8±6.1 1.42±0.36 0.72±0.22 2.85±0.57 1.73±0.35 10.80±2.16 61300±6100 1.05±0.21 28.7±5.7 1.10±0.28 1.19±0.24

additive found, mg/dm3

5.02 1.48 1.32 2.00 19.9 0.69 0.72 2.40 0.85 5.10 29900 0.50 15.2 0.52 0.53

TABLE 7. Comparison of the Analysis Time for AA, AAN, DMF, DMSO, DMA, and EG in Water with the Unified Method and with PND F Methods (n = 2) Duration of analysis, min chromatograph with two columns and two detectors Operations

Chromatograph going into working conditions Checking calibration graph Analysis Total time

PND F 14.1:2.230-06

PND F 14.1:2.228-06

PND F 14.1:2.250-08

AA, AAN

DMA, DMF, DMSO

EG

chromatograph with one column and one detector unified method AA, AAN, DMF, DMSO, DMA, EG

30

20

20

30

20 20 70

30 30 80 230

30 30 80

40 40 110

As the reported data show, analyzing samples of water containing low-boiling and high-boiling compounds with the unified method almost halving the analysis time due to unification of the equipment and conditions for all investigated components. This allows using the method for fast evaluation of the content of harmful components in samples of wastewater to determine the method of treating them. REFERENCES 1. 2. 3. 4.

390

Method of Executing Measurements of Mass Concentrations of Dimethylacetamide, Dimethylformamide, and Dimethyl Sulfoxide in Natural and Wastewaters by Gas Chromatography, PND F 14.1:2.228-06. Method of Executing Measurements of Mass Concentrations of Acrylic Acid Nitrile and Acetaldehyde in Natural and Wastewaters by Gas Chromatography, PND F 14.1:2.230-06. Method of Executing Measurements of Mass Concentrations of Ethylene Glycol and Diethylene Glycol in Natural and Waste Waters by Gas Chromatography, PND F 14.1:2.250-08. A. F. Nersesova, T. I. Samsonova, and V. M. Pantaeva, “Quantitative determination of contaminants in acrylic acid derivatives by capillary liquid chromatography, in: Fizikokhimiya Polim., No. 11, 236 (2005).