ANALYTICAL CHEMISTRY, VOL. 51, NO. 1, JANUARY 1979
method described in ref. 16 (or in ref. 14 if any major differences between corn and chow are discounted) would appear to indicate that fluorescence detection possesses a higher specificity than UV detection for the problem a t hand. A comparative study of the two detection methods using the same extraction and clean-up procedures would provide a definitive answer to this question. However, from the point of view of carrying out routine laboratory analyses of grain samples, selection of one method over another is often largely a matter of personal convenience-even in those cases where fluorescence detection is significantly more sensitive for detection of standards. Since the size of the grain sample from which the extract is made can be arbitrarily large (on the scale of these analyses), detector sensitivity can be traded off against the effectiveness of the clean-up procedure: an increase in the sample size combined with a correspondingly more thorough cleanup will offset a loss in detector sensitivity without a sacrifice in the overall ability of the method to detect contamination in grain (in terms of ppb). An examination of the equivalent amounts of grain injected onto the HPLC column reported in this work (0.2 g) and the two papers describing UV detection of zearalenone in grain (1 g in ref. 14, 5 g in ref. 16) clearly demonstrates this tradeoff and points u p the options open to the analyst in detecting contamination of grain.
ACKNOWLEDGMENT We thank Larry M. Seitz, U S . Grain Marketing Research Center, Manhattan, Kansas, whose help has been vital to the completion of these studies.
69
LITERATURE CITED M. R. Berman and R. N. Zare, Anal. Chem., 47, 1200 (1975). G. J. Diebold and R. N. Zare, Science, 196, 1439 (1977). G. J. Diebold, N. Karny, R. N. &re, and L. M. Seitz, J. Assoc. Off. Anal. Chem., in press. G. J. Diebold and R. N. Zare, "Detection of Aflatoxin 8 , in Contaminated Corn", March 1978 ACS Symposium Series (to be published). C. J. Mirocha and C. M. Christensen, in "Mycotoxins", I. F. H. Purchase, Ed., Elsevier, Amsterdam, 1974, Chapter VI. S. V. Pathre and C. J. Mirocha, in "Mycotoxins and Other Fungal Related Food Problems". J. V. Rodricks, Ed. Adv. Chern. Ser. 149, American Chemical Society, Washington, D.C., 1976, Chapter 10. L. Stoloff, S. Henry, and 0. J. Francis, Jr , J . Assoc. Off. Anal. Chem., 59, 118 (1976). R. M. Eppley, J . Assoc. Off. Anal. Chern., 51, 74 (1968). L. Stdoff. S. Nesheim, L. Yen, J. V. Rodricks, M. Stack, and A. D. Campbell, J , Assoc. Off. Anal. Chem., 54, 91 (1971). F. Thomas, R. Eppiey, and M. W. Trucksess, J . Assoc. Off. Anal. Chem., 58, 114 (1975). L. M. Seitz and H. E. Mohr, J . Assoc. Off. Anal. Chern.. 59, 106 (1976). C. J. Mirocha. B.Schauerhamer, and S. V. Pathre, J . ASSOC.Off. Anal. Chem., 57, 1104 (1974). M. Mulaiyandi, J. P. Barrette, P. L. Wavrock, J . Assoc. Off. Anal. Chem., 59, 959-(1976). C. L. Holder, C. R. Nony, and M. C. Bowman, J. Assoc. Off. Anal. Chem., 60. 272 (1977). G. Engstiorn, J: L. Richard, and S. J. Cysewski, J . Agric. Food Chem., 25, 833 (1977). T. E. Mller and E. Josefsson, J. Assoc. Off. Anal. Chem., 61. 789 (1978). L. M. Seitz and H. E. Mohr, Cereal Chem., 54, 179 (1977). Compare the absorbance reported for aflatoxin by J. A. Robertson, W. A. Pons, Jr., and L. A. Goldbiatt, J . Agric. Food Chem., 15, 798 (1967) with that for zearalenone reported in reference 12. While the maximum absorbances are approximately the same, the detection limit on TLC plates for zearaienone (see reference 14) is about two orders of magnitude less than that for aflatoxin (see reference 1)
RECEIVED for review July 27,1978. Accepted October 18,1978. The support of the National Cancer Institute under Grant No. 2 R01 CA23156-02 is gratefully acknowledged.
Silicon Emitter for Field Desorption Mass Spectrometry T. Matsuo" and H. Matsuda Institute
of Physics,
College of General Education, Osaka University, Toyonaka, Osaka, 560 Japan
I. Katakuse Department of Physics, Faculty of Science, Osaka University, Toyonaka, Osaka, 560 Japan
A new ion emitter for field ionization and desorption mass spectrometry has been developed. Very fine silicon whiskers grown on a 60-pm (diameter) tungsten (or tantalum) wire can be used as a strong and efficient emitter. The process of growth is as follows. A small amount of gold is deposited by evaporation on 60-1m tungsten wire. After being preheated in a vacuum chamber, the gold coated tungsten wire is heated by direct electric current in silane gas (SiH, 5 YO -t Ar 95 Y O ) at a pressure of 180 Torr. Silicon whiskers grow in about 1 min. Pre-treatment, careful control, and high voltage supply during the whisker growth are unnecessary. The total ionizing efficiencies of silicon emitters in F I operation and in FD operation have been measured. The ionization efficiency for acetone in F I operation was 5 X 10" A/lorr and that for AMP in FD operation was 1.4 X lo-'' C/pg. The results for other test samples are also described.
obtain good F D emitters. Essential requirements of FD emitters are pointed out in Ref. 2 as: (1) high ionization efficiency, (2) large surface area, and (3) sufficient strength. The high-temperature activated carbon emitter is one solution that is now widely adopted as a standard emitter (3-5). Nickel and cobalt emitters (6, 7 ) also have recently been used in some cases. It seems to be desirable t o add one more requirement which is, (4) easy production. A silicon emitter shortly reported in Ref. 8 would be a possible solution which satisfies the above four requirements. In our early work, silicon whiskers were grown on the 64-pm tantalum wire in silane gas (at 50 Torr) within about I 5 min. After this initial work, systematic studies were done to obtain better silicon emitters. The detailed technique of producing silicon emitters and the ionization efficiency measurement of silicon emitters for some test samples (oligopeptides, nucleotides, organic salts) are described in this paper.
Significant progress has been made in field desorption mass spectrometry (FD-MS) since Beckey first introduced it in 1969 ( I ) . One of the most important problems in FD-MS is to
Silicon Whisker Growing System. The apparatus is shown schematically in Figure 1. The growing processes are as follows. A 60-pm tungsten wire is spot welded on the wire support (1-mm covar wire). Gold is deposited by evaporation on the part of
EXPERIMENTAL
0003-2700/79/0351-0069$01 OO/O
1978 American Chemical Society
70
ANALYTICAL CHEMISTRY, VOL. 51, NO. 1, JANUARY 1979
3 kV. Mass spectra were obtained by scanning the magnetic field. An UV recorder (San-ei Instrument Co. Ltd.) was used with an amplifier bandpass of 100 kHz. The electron multiplier gain was determined by measurement and was 9 X lo4 for 1.5 kV and 8 X lo5 for 2.0 kV. The high resistance of the pre-amplifier was 10: n. The residual gas pressure a t the ion source was nearly 2 X Torr and that of the main analyzer was less than 1 X Torr. The calibration of the mass marker was done by polystyrene. Sample Preparation. Acetone, cholesterole, uridine, AMP, and benzenesulfonic acid sodium salt were purchased from Wako Chemical Co. (Tokyo). Bradykinin and oligopeptides (H-ProLeu-Gly-NH2, H-Lys-Lys-Gly-Lys-Lys-Gly-OH) were supplied by the Institute for Protein Research, Osaka University. The solvent and the concentration for each sample are given in Table I. A known amount of sample was loaded on the silicon emitter using the syringe technique (10).
Vacuum Chamber
i
3;
1-7% ; *r
ti,
5’1.1
95%
fu 11
Pressure Gauge
11 g , u > {j
5,
?
Power Supply
rJ5 ~0
Flgure 1. Growing system of silicon emitters
RESULTS AND DISCUSSION Temperature and Pressure Dependency of Growing Conditions. The influences of temperature and pressure on t h e shape of silicon whiskers were investigated by taking scanning electron microphotographs. The change of pressure of silane gas is not critical and gives only a different growing time. T o obtain good silicon whiskers, it takes about 10 min at 50 Torr, 1 min at 200 Torr, and 20 s a t 400 Torr. T h e pressure of 180 Torr was adopted for ease of handling. O n the other hand, the temperature dependency is very high. The scanning electron microphotographs of three different cases are shown in Figure 2. Best ionization efficiency is attained under t h e condition given in Figure 2a. This experimental condition has been adopted in our routine production. Ionization Efficiency. T o increase t h e ionization efficiency, it is clearly indispensable t o produce a very strong electric field in t h e vicinity of the top of the whiskers. T h e electric field strength on t h e needle surface is expressed in a n approximation similar t o that given by Gomer (11) as:
tungsten wire on which silicon whiskers are to be grown. A suitable thickness of the gold layer is several hundred angstroms. This gold coated tungsten wire is set in a small vacuum chamber and evacuated. By a direct electric current (0.45 V, 0.90 A), the wire is pre-heated for about 10 s. The temperature of the wire rises to about 800 “C. Then the chamber is filled with silane gas (SiH, 5% + Ar 95%) to a pressure of 180 Torr. Immediately thereafter the tungsten wire is again heated by a dc current, the silicon whiskers begin to grow. One minute is enough to obtain good silicon whiskers. The dc current is supplied by a regulated constant voltage power supply. The out-put voltage is adjusted so that very fine and uniform whiskers grow. A suitable condition was (0.45 V, 0.90 A) and in this case very faint light from the tungsten wire could scarcely be observed in a dark room. The same voltage of the power supply can be applied for pre-heating though higher temperature is obtained because the pressure is lower. The length, diameter, and growing time of the whiskers are dependent upon the pressure of silane gas and the temperature of the wire. These dependencies are discussed later. In our system, 24 silicon emitters are produced successively a t a time. In order to keep a good yield ratio, each emitter holder should be connected to the constant voltage power supply by thick (1.5 mm) copper wires of exactly equal length. Mass Spectrometer. The Matsuda-type second-order double focusing mass spectrometer was used for the analysis of mass spectra (9). The resolution was about 1000 with a 0.2-mm main slit and a 0.2-mm detector slit. Accelerating voltage was normally
a
b
Figure 2. Scanning electron microphotographs of silicon emitters. Supply voltage for heating was of silane gas was 180 Torr in all cases. Magnifications are given in the lower part (U = lrn)
C
(a) 0.45 V, (b) 0.52 V, (c) 0.70 V. Pressure
ANALYTICAL CHEMISTRY, VOL. 51, NO. 1, JANUARY 1979
71
Table I. Results of Sensitivity Measurements of Silicon Emitters compound
solvent
284 644 1059
0.36
1.1
386
C,H,OH C,H,OH H,O + C,H,OH CHCl,
e. uridine
244
H2O
f.
347 180
H,O H, 0
a. H-Pro-Leu-Gly-NH, b. H-Lys-Lys-Gly-Lys-LyS-Gly-OH c. Bradykinin d. cholesterole
adenosine-5'-monophosphate
g. benzenesulfonic acid sodium salt
I
+
ionization efficiency, C/pg 2.5 X lo-'' 1.0 x 8.0 x 10-14 9.0 x l o - " 8.0 x 1.4 X 10.''
