Identification and Determination of Plasticizers in Lacquers by

plasticizers in nitrocellulose, vinyl, and acrylic- type lacquers by programmed tem- perature gas liquid chromatography. The analysis is conducted on ...
0 downloads 0 Views 352KB Size
Identification and Determination of Plasticizers in Lacquers by Programmed Temperature Gas Chrcbmatog ra phy G. G. ESPOSITO Coating and Chemical Laboratory, Aberdeen Proving Ground, Md.

b The physical charsxteristics of lac-

imum and maximum requirements which should be verified for quality control. A method (1) for the assay of isolated butyl benzyl phthalate using gas chromatography and a procedure (8) for identifying plasticizers in plastics have been described. The latter utilized two Apiezon K columns isothermally, but some of the chromatogram peaks were broad and asymmetric. The quantitative data in this work were brief, not substantiated by analysis of known compositions and not related to the whole material. The literature does not contain an analytical method of general applicability to the determination of plasticizers in lacqiier formulations. This paper describes the use of programmed temperature gas chromatography for identifying and determining the plasticizers frequently encountered in lacquers. The chromatogram peaks are narrow, symmetrical. and congruous for quantitative use. Lacquers of known composition based on nitrocellulose, vinyl, and acrylic resins were prepared with various phthalate plasticizers and tricresyl phosphate. Good agreement between known and analytical values was obtained. A high temperature gas chromatograph unit capable of analyzing high-boiling materials

quers are greatly influenced b y their plasticizer content. A procedure is described for the identification and determination of seven plasticizers in nitrocellulose, vinyl, and acrylictype lacquers by programmed temperature gas liquid chromatography. The analysis is condlxted on lacquer samples after treatmisnt to remove the resins; the plasticizers are identified from their relative retention times and determined by the internal standard technique. The method is simple, rapid, and accurate.

P

are important modifiers of synthetic resins and many of the physical characteristics of the coating end products will depend as much on the plasticimr as on the resin. Because of their vital role in adapting plastic and resinous materials to their varied end uses, analytical detection and measurement are important. There are a large number of plasticizer materials available, most of them highboiling esters. Only % limited number are encountered in lacquers, where they usually make up from 5 to 15y0of the nonvolatile portion. Most lacquer specifications set minLASTICIZERS

C

D E

1

I ;

I li I

IO

F

I

I

I

15

20

25

M I N U T E S

Figure 1. Separation of plasticizers on a silicone grease column A. B. C.

D.

Dimethyl phthalate Diethyl phthalate Dibutyl phthalate Dit utyl sebacate

E. F. G. H.

was employed in this study, but polymeric-type plasticizers are not sufficiently volatile and cannot be determined by the technique described. Since phthalate esters are widely used for plasticizing coating resins, along with polymeric phthalate alkyd resins, it has been difficult to differentiate by chemical methods between the phthalate content from each source, and in most quality control work, only total phthalate content has been reported, which has little significance. The chromatographic method permits the determination of the phthalate plasticizers indepentlcntlg- of phthalate resins. The unit was ralihrated with known individual plasticizers with an internal standard and the relative retentions and correction factors were established. In the analysis of lacquers, a weighed amount of internal standard is added t o a weighed sample, which is diluted slightly with acetone, and a preliminary physical separation of plasticizer from resin is made by adding carbon tetrachloride and petroleum ether dropwise to precipitate the resinous portion. After evaporation of a portion of the

Butyl benzyl phthalate Di-(2-ethylhexyl) phthalate Tricresyl phosphate Di-(2-ethylhexyl) sebacate

I 10

15

20

M I N U TE S

Figure 2.

Analysis of nitrocellulose

Ia c q uer A. B.

Dibutyl sebacate (internal standard) Di-(2-ethylhexyl) phthalate

VOL. 35, NO. 10, SEPTEMBER 1963

1439

C

I

0

1

I

I

5

10

15

M I N U T E S

Figure 3. Analysis of nitrocellulose lacquer containing two plasticizers A. 6. C.

Diethyl phthalato Dibutyl phthalate Dibutyl sebacate (internal standard)

solvent, the plasticizers are separated, identiiied, and measured by programmed temperature gas-liquid chromatography using a &foot column of silicone grease. The average lapsed time for each analysis is less than one hour. APPARATUS AND MATERIALS

The equipment used to obtain the chromatograms was a Model 500 linear programmed temperature gas chromatograph ( F & hf Scientific Co.) equipped with a Brown Electronik recorder (Minneapolis-Honeywell) and a disk integrator (Disc Instruments, Inc.).

210" C., the attenuation is set a t a point of low sensitivity, and 10 to 20 pl. of sample are injected onto the column. The temperature programming niechanism is engaged immediately for a column heating rate of 4" C. per minute. After the solvents have emerged, the sensitivity is reset according to the type and amount of plasticizer present. When the final column temperature of 290" C . is reached, it is maintained until the chromatogram has completely developed. I The plasticizers are identified by calculating their retention times, relative to butyl sebacate, and by comparing them to the calibration chart shown in Table I. The detector response is a function of molecular weight and structure, and response correction factors are obtained by chromatographing known weights of each plasticizer with the internal standard. The area of each peak is measured and corrected for detector response, and the plasticizer is determined from the known concentration of internal standard that has been added. RESULTS

The separation of plasticizers is illustrated in Figure 1. All plasticizers produced essentially one peak, except tricresyl phosphate. Figures 2 and 3

Table 1.

Chromatographic Unit.

EXPERIMENTAL

Approximately 100 mg. of internal standard (butyl adipate or butyl sebacate) are accurately weighed into a 25-ml. Bask followed by 2 to 3 grams of lacquer, which is also accurately weighed. One milliliter of acetone is added and a magnetic stirring bar is inserted. With rapid stirring of the sample, 3 ml. of carbon tetrachloride are added dropwise (80 to 100 drops per minute) followed by 3 ml. of petroleum ether added in the same manner. The sample is then filtered through rapid filter paper into tl. 25-ml. beaker. An antibumping stone is added and the beaker placed in a 70" C. water bath. It is removed as soon as boiling subsides. The chromatographic unit is provided with a column consisting of a 6-foot length of 1/4-inch copper tubing packed with 20% by weight of silicone grease on acid and alkali-washed Chromosorb W. The det,ector cell temperature is set a t 300" C., the detector cell current a t 160 ma., injection port temperature a t 330" C., and helium flow a t exit is adjusted to 120 cc. per minute. The chromatographic column is heated to 1440

ANALYTICAL CHEMISTRY

Relative Retention Time Data for Plasticizers

Relative retention time (dibutyl sebacate = 1) Dimethyl phthalate 0.16 Diethyl phthalate 0.26 Dibutyl phthalate 0.67 Butyl benzyl phthalate 1.32 Di-( Zethylhexyl) phthalate 1.65 (DOW 1.78-1.97" Tricresyl phosphate Dioctyl sebacate 2.20 Several peaks produced in this range. 5

I

I

1

I

0

5

10

15

M I NU TE S

Figure 4. A. 6. C.

Analysis of acrylic lacquer

Dibutyl adipate (internal standard) Dibutyl phthalate Butyl benzyl phthalate

show the analysis of nitrocellulose lacquers containing phthalate plasticizers. The lacquer analysis illustrated in Figure 3 is from a sample containing two plasticizers, diethyl phthalate and dibutyl phthalate, showing that both can be determined in one analysis. All nitrocellulose lacquers examined produced a tailing peak in the first few minutes of operation and the pen had to be returned to zero manually. Figure 4 shows the analysis of an acrylic-type lacquer plasticized with butyl benzyl phthalate. A vinyl chloride lacquer containing tricresyl phosphate produced the Chromatogram shown in Figure 5. Quantitative data were obtained on known lacquers prepared from nitrocellulose, acrylic, and vinyl resins plasticized with diethyl phthalate, dibutyl phthalate, di-(2-ethylhexyl) phthalate, butyl benzyl phthalate, and tricresyl phosphate. The tabulated results of the analyses shown in Table I1 agree within =t0.25% of the actual plasticizer content. Changes in attenuation

B

L 0

I

I

5

10

I 15

1 20

M I N U T E S

Figure 5. A. 8.

-

Analysis of vinyl lacquer

Dibutyl rebacate (internal standard) Tricresyl phosphate

I

25

Table II.

Type of lacquer Nitrocellulose

Acrylic Vinyl

Analysis of Known Lacquers

Plasticizer used Tricresyl phosphate Di-(2-ethylhexyl) phthalate Di-( Zethylhexyl) phthalate Diethyl phthalate Dibutyl phthalate Dibutyl phthalate Butyl benzyl phthalate Tricresyl phosphate

Present, % 12.5 12.0 10.0 6.4 6.7 15.0 19.0 9.3

Found, % 12.6, 11.7, 10.0, 6.4, 6.5, 14.7, 18.8, 9.3,

12.1 11.8 10.3 6.2 6.9 14.9 18.9 9.3

the chromatograms. Dibutyl adipate is available in higher purity and was used for the butyl benzyl phthalate analysis because a small amount of impurity in the butyl sebacate caused some interference. The butyl benzyl phthalate used to prepare the known contained 5y0dibutyl phthalate and no detectable amount of dibenzyl phthalate. If dibenzyl phthalate is to be determined, it is best to repeat the analysis isothermally a t 180’ C. ACKNOWLEDGEMENT

are not noted on the chromatograms but are taken into consideration in the calculation. DISCUSSION

It is not likely i;hat all plasticizers can be completely separated as there are certain to be combinations which cannot be resolved. So-called “methyl abietate” is a mixture of the methyl esters of rosin acids and appears as several peaks immediately following that of butyl sebacake and overlapping butyl benzyl phtha.ate. This investigation illustrates the suitability of a

silicone grease column as described for the separation of the plasticizers that are most frequently encountered in lacquer coatings. Additional information can be obtained by the use of a polar, 6-inch 20% polyester column which will affect the order of elution of some of the plasticizers and may be needed for some other combinations. Both dibutyl sebacate and dibutyl adipate were used as internal standards in this study. Dibutyl sebacate is preferred for establishing relative retention times and for quantitative work because of the position it occupies on

The advisory assistance of C. F. Pickett., director of the laboratory, and M. H. Swam of the analytical section is acknowledged and appreciated. LITERATURE CITED

(1) Cook, C . D., Elgood, E. J., Shaw, G. C., Solomon, D. H., ANAL. CHEM. 34,1177 (1962). (2) Lewis, J. S., P:tton, H. W., in “Gas

Chromatography, V. J. Coates, H. J. Nobels, I. S. Fagerson, eds., p. 145, Acadenlic Press, New York, 1958. RECEIVEDfor review April 17, 1963. Accepted June 10, 1963.

Cornparkon of Pitch Resins from Different Sources by Combined Pyrolysis and Gas Liquid Chromatography CLARENCE KARR, Jr,, JOSEPH R. COMBERIATI, and WILLIAM C. WARNER Morgantown Coal Research Center, Bureau of Mines,

b Pitch resins from various sources can be effectively compared by the technique of combined pyrolysis and gas liquid chromatography. This method has the advantage of indicating differences in the fundamental structures of the resin!;. Examples are given for the analysis of the benzenesoluble, petroleum ether-insoluble resins isolated from, low-temperature lignite, subbiturninoils and bituminous coal tars, and a commercial electrode binder pitch.

I

has been a growing interest in the utilization of materials obtained from coal tar pitches and petroleum pitches. Among these pitch fractions are the petroleum ether-insoluble portions variously known as resins to the coal chemist and asphaltenes to the Fetroleum chemist. A method of comparing pitch resins from various sources would be highly desirable if it reflected differences in the fundamental structures of the resins. Such a method apperirs to be available in the technique of combined pyrolysis and gas liquid chroma,tography that was N RECENT Y E A F ~ Sthere

U. S.

Department o f the interior, Morgantown, W. Va.

developed by the Bureau of Mines for an investigation of the structures of resins from the pitch of various lowtemperature coal tars (4). In the work reported here, all of the resin samples were obtained from coal tar pitches. However, there does not appear to be any reason why this technique would not be equally applicable to resins or asphaltenes obtained from petroleum pitches. The four samples studied were all the benzene-soluble, petroleum ether-insoluble resins from a low-temperature Texas lignite tar, a low-temperature Nugget, Wyo., subbituminous coal tar, a low-temperature West Virginia bituminous coal tar, and an electrode binder pitch from a hightemperature (coke oven) bituminous tar. The low-temperature tars were prepared by fluidized-bed carbonization a t about 500’ C.; the pitch binder was a commercial sample. EXPERIMENTAL

Apparatus and Procedure. The resin sample was powdered and placed in a thin layer on a watch glass, which was then put in a vacuum drying oven for 4 hours a t 100’ C.,

the oven atmosphere being nitrogen a t 20 mm. of Hg pressure. This treatment was necessary to remove any trace of solvents. About 75 mg. of dried resin was placed in a 4 m m . 0.d. tube, the air was evacuated, and the tube was sealed. The sealed tube was inserted within the heating coil of the pyrolysis chamber (4). The chamber, the helium carrier gas inlet, and the connection between the chamber and the gas liquid chromatographic column were all preheated to 175’ C. The coil voltage was set high enough so that upon firing the tube would shatter within a few seconds from the pressure of the gaseous pyrolysis products. From calibration runs with a thermocouple, the temperature a t the moment of shattering was established. In the work reported here, this temperature was within a few degrees of 530’ C. This sealed-tube technique was required because the lower-molecularweight resins, M = 400 to 500, evaporate before they decompose. The volatile pyrolysis products were swept immediately into the gas liquid chromatographic column by the helium stream. The column was made from a 20-foot length of 1/4-inch copper tubing that was filled with 75 grams of packing made from %yoApiezon L grease on VOL. 35, NO. 10, SEPTEMBER 1963

0

1441