Simplified method of preparing column materials for gas-liquid

separated Endrin from , '- by using Chromosorb. W coated with SE-30 and QF-1. In our experience, variation between preparations can be attributed to u...
0 downloads 0 Views 373KB Size
ion

Simplified Method of Preparing Co~urnnMaterials for Gas-Liquid ~ ~ r o m a t o f Some Pesticides C. E. Mendoza, K. A. McCully, and P. J. Wales Research Laboratories, Food and Drug Directorate, Department of National Health and Welfare, Ottawa 3, Ontario, Canada

PROCEDURES ranging from simple stirring to fluidized drying have been proposed to prepare column packing materials for gas-liquid chromatography (GLC) of organic chemicals (1-6)and organic pesticides (7-14). Satisfactory separation of organochlorine pesticides was obtained with Anachrome ABS (8,9), Chromoport X X X (IO), Gas Chrom Q (9, 11, 12) and Chromosorb W (I3,14) coated with silicones, but the packed columns required lengthy conditioning with repeated injections under specified temperature and gas flow. Gas Chrom Q coated with DC-200 and QF-1 required 3-5 days conditioning and did not separate Endrin from o,p'-DDT (9). McCully and McKinley (14) separated Endrin from o,p '-DDT by using Chromosorb W coated with SE-30 and QF-1. In our experience, variation between preparations can be attributed to uneven coating of the support due to inadequate mixing or to breakage of particles during stirring. This report describes a simple method of preparing column materials to give reproducible chromatograms and to eliminate bleeding and lengthy conditioning. EXPERIMENTAL Coating the Support with Silicones. Materials and conditions used in preparation of columns are listed in Table I. SE-30 and QF-1 were dissolved in ethyl acetate (10 ml/gram of support) in a 500-ml round bottom flask (T 24/40); SE-30 was broken into smaller pieces to hasten dissolution. The solution was gently heated (Variac at 25-30 V) and occasionally stirred for 1 hour or until the gum dissolved. The support was added to the silicone solution; the slurry was refluxed on a heating mantle controlled by a Variac at 50 V and swirled occasionally for 1, 2, or 4 hours. With periodic swirling of the flask contents, ethyl acetate was evaporated (Variac at 50 V), under a gentle flow of nitrogen. The nitrogen flow was directed to the surface of the slurry by a piece of glass tubing bent in a V-shape and hung on the rim of the flask. The flask was tapped to dislodge adhering granules. The partially-dried granules were transferred to (1) R. F. Krupa, R. S. Henley, and D. L. Smead, ANAL.CHEM., 39, 851 (1967). (2) J. P. Parcher and P. Urone, J. Gas Chromatogr., 2,184(1964). (3) E. D. Smith, ANAL.CHEM., 32, 1049 (1960). (4) E. C. Homing, E. A. Moscatelli, and C. C. Sweeley, Chem. Znd. (London), 751 (1959). ( 5 ) H. H. Wotiz and S. C . Chattoraj, ANAL. CHEM.,36, 1466 (1964). (6) W. J. Zubyk and A. 2.Conner, ibid., 32,912(1960). (7) L.A. Richardson, J. R. Lane, J. T. Peeler, and J. E. Campbell, J . Dairy Sei., 50, 1073 (1967). (8) J. A. Burke, J. Ass. OfJic. Anal. Chem., 48, 1037 (1965). (9) J. A. Burke and W. Holswade, ibid., 49, 374 (1966). (10) J. L. Radomski and V. Fiserova-Bergerova. Znd. Med. Sura.. - , . d6, 281 (1967). (11) D. C. Bostwick and L. Giuffrida. J. Ass. 81%.Anal. Chem.. . 51, 34 (1968). (12) "Pesticide Analytical Manual. Food and Drug- Adrninistra. Gon,'' Vol. I, 2.32 b (1965). (13) J. R. Brown, Canadian Med. Assoc. J., 97, 367 (1967). (14) K. A. McCully and W. P. McKinley, J. Ass. Ofic. Anal. Chem., 47,652 (1964).

IA

1.8 h i

72 hr

IC I 8 1 1 1 4 8

o

2 hr

IF

4 hr

1

12

'

1 ' 16

'

20

TIME

1

1

24

IN

1

o

'

1.5 h i

1

4

'

1

a

'

1

1

12

1

1

16

1

1

20

1

,

24

MINUTES

Figure 1. Typical chromatograms obtained from c o ~ u m nof~ Chromosorb W coated by refluxing SE-30 and QF-l(10:0.4:9.6 grams) in ethyl acetate Chromatograms A, E, and C were obtained from columns packed with unheated materials; chromatograms D , E , and F, materials heate at 230 OC. Number of hours from column installation to injection QFB indicated for each chromatogram. 1. Aldrin, 2. ~ e p t a ~ ~ ~ o ~ epoxide, 3. p,p'-DDE, 4. Dieldrin, 5. o,p'-DDT, 6. p,p'-DDT, 7. Endrin I, 8. Endrin n. A , E, D , and E are chromatograms of a sohtion consisting of compounds 1 to 4 (0.2 mg each), 5 and 6 (0.5 ng each) and Endrin (2ng) a crystallizing dish and heated (35 "C) until free of clumps. Drying was completed in an oven at 110 "C for 2 hours; portions of the coated material were then preconditioned at 230 "C for 2 or 4 hours, or at 250 "Cfor 1.5 hours. Packing the Glass Column with Coated Support. The silicone-coated support was packed under vacuum into the column, 4lI2 feet X inch 0.d. borosilicate glass tubing coiled to fit an Aerograph Hi-Fi Model 550. To ensure tight and even packing, the material was poured in small portions and final packing of the column was under -25 psi pressure. RESULTS AND DISCUSSION Coated materials heated at 230 "G for 2-4 hours before being packed (Table I: columns 9, 11, 13, and 15) did not contaminate the electron capture (EC) detector (tritium source) and produced discrete resolution of pesticides as soon as the operating temperature was obtained. Typical chromatograms obtained with these columns are shown in Figure 1 (D,E, F),Figure 2 (I, J, K ) , and Figure 3 (N, Heating the materials increased the recovery or sensitivity 5 times for the organochlorine and approximately 10 times for the organophosphorus pesticides studied. Portions of the same coated materials but packed without heating (Table I: Columns 10, 12,141 bled and contaminated the EC detector. Most importantly, the unheated materials did not satisfactorily separate the pesticides and th of detection was low even after cleaning the E VOL. 40, NO. 14, DECEMBER 1968

e

1.5 hr

I

2.5 hi

2 hr

N J

48 hr

10

134 hr

iM

e 2. Typical chromatograms of methoxychlor obtained columns of Chromosorb W coated by refluxing SE-30 and (10:0.4:0.6 grams) in ethyl acetate 0

~hromatogramsG and H were obtained ffrom columns packed with unheated materials; chromatograms I to K, materials heated at 230 "6. Number of hours from column installation to injection are indicated for each Chromatogram

-

u -

0

IN

12

8

4

TIME

16

MINUTES

Figure 3. Typical chromatograms obtained from C made of Chromosorb W coated by refluxing SE-30 and (10:0.4:0.6 grams) in ethyl acetate

Q

~

Chromatograms L and M were obtained from columns packed with unheated materials; chromatograms N and 0, materials heated at 230 "C. Number of hours from column installation to injection are indicated for each chromatogram. 10. Malathion (20 ng), 11. Parathion (2.5 ng), 12.Carbophenothion(0.125 ng)

Table I. Materials and Conditions Used in Preparation of Columns Chromatogramb 1st inj.

Column Materials Column support (Chromosorb) Type Mesh We, gram W 60-80 9

Column 1

2

W

60-80

3

W W W W W W W

60-80 80-100 60-80

4 5c

6 7" 8 9d

10 10 10

80-100

20

80-100

20 20

w

1ld

W W W W

80-100 80-100 80-100 80-100

W W

80-100 80-100

W W 0

80-100 80-100

15d

IS 1P (HMDS)

9

60-80 60-80

80-100

1O d

12d 13d 14d

10 10 10

20

20 20 20

Coating, grams SE-30 QF-1 0.6 0.4 0.4 0.7 0.6 0.5 0.4

0.4 0.5 0.4 0.5 0.8 0.8

0.8 0.8 0.8

0.8 0.8

10

0.4

20

0.8

0.6 0.6 0.6

0.7 0.6 1.2 1.2 1.2 1.2 1.2 1.2 1.2 0.6 1.2 0.6

Treatment, hr Reflux Heating" 1 0 1 0 1 0 1

1 1 1

0 0 0 0

1 1

0

1 1 2 2 4 4 1

0

1

2.0(230 "C) 4.0(230 "C) 0

4.0(230 "C) 0

2.0 (230 "6) 0 0

from installation, hr 22.5

19.2 64.2 20.8 1.8 23.8 18.7 0.8 0.8

3.2 0.7 2.8 0.7 1.5 0.6 2.3 1.5 4.0 0.9

Quality Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory satisfactory Unsatisfactory Satisfactory Unsastifactor y Satisfactory Unsatisfactory Satisfactory Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory Unsatisfactory

1 0 1.2 1 0 20 0.8 1.2 1 1.5 ( 250 "C) 3.0 20 a 80-600 0.8 20 1 0 3.8 0.6 20 G 8Q-100 0.4 21 1 1.5 (250 "C) 1 .O 0.6 0 80-100 0.4 20 22 a Further heating in an oven at higher temperatures before packing, in parenthesis. 6 The chromatographic analyses were performed with an Aerograph Hi-Fi electrometer (Model 500-C) and oven (Model 550) with tritiumcoated detector under constant conditions: 190-195 "C oven, 230 OC injection block, approximately 75 ml/min Nz,190 OC detector temperature, and with electrometer attenuation 8-Xat range 1. Columns 5 and 7 are duplicates of Columns 1 and 2, respectively. d Portions of the same materials used for columns 10, 12, and 14 were heated, as indicated, before being packed into columns 11, 13, and 15, respectively. a (HMDS), Hexamethyl disilazane-treated Chromosorb. 188 (WMDS) 19

80-100

Q

~~

2226

e

ANALYTICAL CHEMISTRY

10

0.4

~

~

~

Conditioning for several days was necessary to obtain chromatograms comparable to those obtained with the heated materials only a few hours after installation. Typical chromatograms from the unheated materials are shown in Figure 1 ( A , B, C), Figure 2 (G, H), and Figure 3 (L, M ) . Column materials that were not heated before packing (Table I: Columns 1-8, 10, 12, 14, 16) did not separate Dieldrin from o,p'-DDT or Malathion from Parathion. Partial degradation of p,p'-DDT and Methoxychlor was continuously observed; satisfactory recoveries of these compounds were obtained only after lengthy conditioning of the columns. The use of different sizes of solid support and proportions of the coating material did not improve column performance; the 80 to 100-mesh support was preferred because it could be tightly packed into a glass column. The columns made of silanized Chromosorb W coated with SE-30 and QF-1 (Table I: Columns 17, 18) gave unsatisfactory resolution of some organochlorine and organophosphorus pesticides. Malathion retention time was greatly decreased as compared to those of Parathion and Carbophenothion; the multiple peaks produced indicated breakdown.

Although columns made of Chromosorb G coated with SE-30 and QF-1 (Table I: Columns 19, 20, 12, 22) gave good resolution of the test compounds, the recoveries based on peak heights were very low when compared to those from Chromosorb W columns. The p,p'-DDT breakdown was negligible. Heating of silicone-coated Chromosorb G in an oven at about 250 "C for 1.5 hours did not improve the recovery of pesticides. In separating organochlorine and organophosphorus pesticides, Chromosorb W coated with DC-200 and QF-1 (20:0.8:1.2 grams) and heated at 230 "C for 2 hours before packing was comparable to that coated with SE-30 and QF-1. ACKNOWLEDGMENT

We thank E. Coffin and S. W. Gunner of this Directorate for criticism of the manuscript.

RECEIVED for review May 28, 1968. Accepted August 7, 1968.

Calibration of the Methanol and Glycol Nuclear Magnetic Resonance Thermometers with a Static Thermistor Probe Anthony L. Van Geet Department of Chemistry, State University of New York, Buffalo, N . Y . 14214

INNUCLEAR magnetic resonance, the temperature of a sample is commonly obtained by measuring the chemical shift of methanol or of 1,Zethanediol (glycol). During our work with proton exchange ( I ) most of the experimental error was attributed to the temperature measurement, and a more accurate and direct method was desired. A thermistor probe [U. S. Patent applied for (Figure l)] fitting inside a 5-mm sample tube was built. A probe holder [Patent applied for (Figure 2)] conveniently holds the probe in place, and the probe does not touch the walls of the sample tube. Thus, the tube can spin freely around the stationary probe. The lower part of the probe is made from tubing of polyperfluoroethylene to provide thermal insulation. A resistance bridge and meter (Atkins 3FOlA-C/AB/AC/AD/AE, Atkins Technical, P. 0.Box 14405, University of Florida Station, Gainesville, Fla. 32603) provides a direct temperature reading spread out over 4 scale ranges. The meter may be placed as far from the probe as desired. The accuracy is *l OC or better between -50 and +150 "C, and the reproducibility is 0.2 "C. An alternate way to support the probe is shown in Figure 3. A sleeve bearing of polyperfluoroethylene permits the sample tube to spin while the probe rests on it. It is desirable to reinforce the rim of the thin glass sample tube with a compression ring, made by perforating the top of a pressure cap. To prevent lifting of the sample by the nitrogen flow, a cylindrical polyperfluoroethylene spinning weight of 38 grams is used ( 2 , 3 ) . The thermistor contains oxides of the transition metals, (1) A. L. Van Geet, Znorg. Chem., 7,2026 (1968). (2) A. L. Van G e t , ANAL.CHEM.,40, 1914 (1968). (3) A. L. Van Geet, Rev. Sci. Instrum., in press.

and is paramagnetic. It affects the field homogeneity whenever it is in contact with the liquid, and retuning of the Y-fine homogeneity control is necessary when the probe is raised or lowered. With the thermistor 20 mm above the receiver coil, the observed line width is 1.6 Hz. A temperature gradient occurs in the spinning sample tube. In a spinning methanol sample at -40 "C, the lowest temperature occurs at the bottom. The temperature was measured 10 and 30 mm above the bottom, and the difference was 1.0 "C. Without spinning, the difference is slightly larger. The first position coincides with the center of the receiver coil. The spectrometer is a Varian A60, with a V6031 probe and a V6040 temperature controller. The stationary temperature at the coil changes slightly when the probe is inserted into the liquid, because heat leaks through it. The effect was studied at -40 OC from the chemical shift of methanol. Insertion of the probe raises the final temperature at the coil by 1.0 "C, irrespective of the depth of insertion. For a probe which had a thinner tip by omitting Part 4 of Figure 1, the effect of insertion was only 0.7 "C, but it was more difficult to keep the tip straight, and this probe had no other advantage. For optimum accuracy of the temperature measurement, the probe is in contact with the liquid while taking the spectrum, then lowered to the position of the receiver coil to measure the temperature. On the other hand, for optimum resolution, the probe is raised above the surface of the liquid while taking the spectrum. It is necessary to keep the polyperfluoroethylene tip of the probe straight to avoid interference with the spinning. A slight curvature is easily straightened out manually by bending in the opposite direction. The tip is protected when not in use. It is inert to all solvents and dilute acids and bases. VOL. 40, NO. 14, DECEMBER 1968

2227