Modification of graphite furnace power supply to ... - ACS Publications

Modification of graphite furnace power supply to allow interruption of analytical cycle. E. W. Cooper, and J. V. Dunckley. Anal. Chem. , 1976, 48 (12)...
0 downloads 0 Views 264KB Size
Table I. Comparison between the Detection of Odors Using the Sniffer and by Directly Sniffing the GC Outputa Amount

injected, ng

Method

n-Hexylacetate

2-Heptanol

100

Sniffer Direct

++ ++

++ ++

10

Sniffer

0

0 0 1 Sniffer 0 0 Direct 0 0 0 ++ indicates the compound was clearly detected; +, only faintly detected; and 0, no detectable odor.

Direct

0

I

I

I

5

IO

15

TIME IN MINUTES

Figure 2. Gas chromatogram of n-hexylacetate (a)and 2-heptanol (6). The event marks ( c )represent the first detection of odor while the marks (a) represent the disappearance of detectable odor

gallon jug (GI. A piece of YIe-in. stainless steel tubing from a nominal 50 to 1splitter in the flame ionization detector oven extends 1 cm into the room and 2 mm into the vacuum port (B) of the filter pump. A %-inch male Swagelok fitting was threaded into the filter pump at D to facilitate plumbing, and the ball check valve in the vacuum port was removed. Nylon tubing ( V 4 - h . 0.d.) (F) was used to connect the components and a Whitey B-1RS4 (Whitey Co., Oakland, Calif.) valve

was installed at E to control the air flow. RESULTS AND DISCUSSION Good results were achieved with a 12 l./min air flow through the filter pump and this caused 250 ml/min room air to flow into the vacuum port a t C. Although this represented a 500fold dilution of the gas chromatographic effluent, it did not appear to result in a severe reduction in odor intensity as compared to directly sniffing the effluent. Without the sniffer, the low velocity of the effluent carrier gas was subjected to room drafts which complicated the detection of peak odors. Figure 2 shows a gas chromatogram of n-hexylacetate ( a ) and 2-heptanol ( b ) which were simultaneously sniffed blindly. Each peak represented approximately 5 wg of material and the event marks (c) represent the first detection of odor and ( d ) the last detectable odor. This sniffer results in a marked improvement in the sensory resolution of gas chromatographic peaks compared with sniffing the effluent directly. Table I shows a comparison between the sensitivity of the sniffer and

+

sniffing the GC output directly. Both methods yielded an apparent threshold of approximately 10 ng for both n-hexylacetate and 2-heptanol with less than a factor of 10 difference between them. When certain odorous compounds were present in high concentrations in the gas chromatographic effluent (e.g., phenyl ethyl alcohol in grape essence) or had an extremely low odor threshold (e.g., geosmin in beets), the sniffer became contaminated and it was necessary to clean the filter pump. Washing with water followed by isopropyl alcohol and then 1,1,2-dichloro-1,2,2,-difluoroethane (Le., Freon 113) worked well. Contamination of the sniffer was also minimized by removing the sniffer from over the gas chromatographic effluent when strongly absorbant peaks emerged or by replacing it with a clean duplicate. The sniffer worked equally well when attached to the effluent from a Llewellyn ( 3 ) type helium separator on a gas chromatograph-mass spectrometer interface. LITERATURE C I T E D (1) R. Teranishi, P. Issenberg, I. Hornstein, and E. L. Wick, “Flavor Research: Principles and Techniques,” Marcel Dekker, New York, N.Y., 1971. (2) H. S. Knight, Anal. Chem., 30, 2030 (1958). (3) R. A. Flath, D. R. Black, D. G Guadagni, W.H. McFadden, and T. H. Shultz, J. Agric. Food Chem., 15, 29 (1967).

RECEIVEDfor review February 27,1976.Accepted June 23, 1976.

Modification of Graphite Furnace Power Supply to Allow Interruption of Analytical Cycle E. W. Cooper* and J. V. Dunckley Electronic Engineering Department, Dunedin Hospital, Dunedin, New Zealand

The graphite furnace is now well established as a useful atom generator for the analysis of cations by atomic absorption spectrometry (1-3). Some of the variety of experimental furnaces described in the literature (4-1 7) have been developed and marketed by manufacturers of atomic absorption spectrometers either as an integral part of the instrument or, more commonly, as an “add on” accessory. These usually operate by automatic sequential switching of the furnace power in accordance with a preset program, generally in three or more steps with increasing current loadings for specific time intervals. For adequate control of the analytical process, it is desirable that time, temperature, and rise rates be reproducible. A common operating sequence is dry-ash-atomize wherein the sample (e.g., 5 pl of solution) is applied to the furnace, 1822

dried at low temperature, ashed at a higher temperature, and finally atomized. It seemed that the graphite furnace had considerable unexploited potential as a chemical reaction vessel. This was realized by placing switches to interrupt the analytical cycle at the end of the drying and ashing cycles. One of these switches (Sx) a t the end of the drying cycle would allow for multiple sample application to the furnace as well as the application of reagents to the dried specimen, while a further switch (Sy) to interrupt a t the end of the ash cycle would allow for the addition of a variety of reagents to render ash constituents more or less volatile, or to modify the influence of matrix cations. Thus, in operation the sample would be placed in the furnace, the selected cycle interrupt switch closed, and the cycle “start” button pressed. The analytical cycle would then proceed automatically to the interrupt stage

ANALYTICAL CHEMISTRY, VOL. 48, NO. 12, OCTOBER 1976

..

---

1

I

1

Figure 1. Circuit modifications which permit interruption of the analytical cycle are enclosed within dotted lines

and then shut down. On again pressing the start button, this sequence would be repeated. After opening the interrupt switch and pressing the start button, the full analytical sequence will take place. The power supply-Varian Techtron Model 63-was modified as follows: Timer Board Assembly (Instruction Manual Figure 5 - 3 . 4 ~was ) altered as shown in Figure 1.The operating sequence is commenced by pressing the START button which completes the circuit for a pulse from capacitor C28 (0.047 pf) to trigger the two SCR's D11 and D12, thus energizing relays A and B whose contacts B1 of relay B close, the timing sequence is commenced for the period as set by the DRY time control. At this same instant, relay A is energized, switching on the transistor QlO (Figure 5-3.2 block diagram) to complete the circuit for the rectified pulses from T2 to operate the triac control for the carbon rod voltage from transformer, T3. On completion of the DRY time cycle, the positive pulse from the idifferential output of MA2 is applied to the base of Q2, which switches off the SCR D12, thus de-energizing relay

B, while at the same time Q3 is turned on resetting the ramp generator MA1 whose output will return to zero. Without interruption by the switch Sx (Figure l),the pulse a t the output of MA2 would switch on the SCR D14 for the analysis to proceed. Because of the similar circuitry used for the ASH timing circuit, a corresponding modification has been applied in the circuit immediately following the I.C. MA4. Two switches were interposed in the circuit, the first (Sx) at the junction of R14, R28, and the gate of SCR D14, leaving R28 and the gate connected a t all times, but allowing the circuit to be broken a t this point to prevent the pulse from switching on D14. The second break is made by switch Sy at the junction between (212, R27, and the RAMP/STEP switch, but again allowing R27 and the RAMP/STEP switch to remain in normal circuit. These two switches allow each step in the analysis to be interrupted to allow chemical manipulation of the analytical environment. It is likely that the power supply-timer units of other manufacturers' furnaces may be similarly modified to extend their analytical capability in like manner.

LITERATURE CITED (1) F. J. M. J. Maessen, F. D. Posma, and Johannes Balke, Anal. Chem., 46, 1445 (1974). (2) F. J. M. J. Maessen and F. D. Posma, Anal. Chem., 46, 1439 (1974). (3) B. C. Begnoche and T. H. Risby, Anal. Chem., 47, 1041 (1975). (4) H. Massmann, Spectrochim. Acta, Part B, 23, 215 (1968). (5) R . Woodriff and G. Ramelow, Spectrochim. Acta, Part B, 23, 665 (1968). (6) T. S.West and X. K. Williams, Anal. Chlm. Acta, 45, 27 (1969). (7) M. P. Bratzel, R . M. Dagnall, and J. D. Winefordner, Anal. Chim. Acta, 48, 197 (1969). (8)B. V. L'vov "Atomic Absorption Spectrochemical Analysis", Adam Hilger, London, 1970. (9) M. Donega and T. E. Burgess, Anal. Chem., 42, 1527 (1970). (IO) H. T. Delves, Analyst (London),95, 431 (1970). (11) M. D. Amos, P. A. Bennet, K. G. Brcdie, P. W. Y. Lung, and J. P. Matousek, Anal. Chem., 43, 211 (1971) (12) J. B. Headridge and D. R. Smith, Talanta, 18, 247 (1971). (13) J. P. Matousek, Am. Lab., June, 45 (1971). (14) J. P. Pemsler and E.J. Rapperport, Anal. Chim. Acta, 58, 15 (1972). (15) Y. Talmi and G. H. Morrison, Anal. Chem., 44, 1455 (1972). (16) C. J. Molnar, R. D. Reeves, J. D. Winefordner, M. T. Glenn, J. R. Ahlstrom, and J. Savory, Appl. Spectrosc., 26, 606 (1972). (17) Tibor Kantor, S. A. Clyburn, and Claude Veillon, Anal. Chem., 46, 2205 (1974).

RECEIVED for review April 2, 1976. Accepted June 16, 1976.

External Reference Signal in X-ray Energy Spectrometry P. J. Van Espen and F. C. Adams* Department of Chemistry, University of Antwerp (U.I.A.), 8 2 6 10 Wilrijk, Belgium

The prlecise and accurate analysis by x-ray fluorescence spectrometry requires a well stabilized high voltage supply to ensure i3 constant and easily reproducible fluorescing beam intensity. This condition is generally fully recognized in wavelength dispersive fluorescence spectrometry. Most generators used in energy-dispersive systems are much less stabilized. The rationale for this is that a very high precision cannot be reached anyway with this type of instrument. The reasons cited are lower counting rates and thus higher statistical errors and the occurrence of increased background levels and interferences. In many situations, the generator instability is the dominant source of error and, hence, represents the major limitation on precision and the accuracy. The remedy to this situation is quite simple, if one is willing to exchange ithe generator for a better stabilized one and, consequently, to considerably increase the total cost of the system. As an alternative, methods have been proposed which rely on

the measurement of a spectral component which bears a relation to the primary beam intensity. The coherently and incoherently scattered primary beam have been proposed for this. It can only serve the purpose, however, when the sample is either infinitely thick or of strictly the same weight and composition because the scattered intensity is proportional to the primary beam intensity times the number of scattering atoms in the beam. Alternatively, an internal standard incorporated into the sample can be used, as is often done in ion induced x-ray emission (1). This is not possible when the sample has to be measured nondestructively without any preseparations or pretreatment. Moreover, the internal standard should be distributed very homogeneously throughout the sample. We have successfully solved the stability problem by using a thin wire as an external standard which is reproducibly positioned in the radiation path just below the sample. Similar

ANALYTICAL CHEMISTRY, VOL. 48, NO. 12, OCTOBER 1976

1823