Hydroxamic Acid Chelate Ion Exchange Resin
SIR: A hydroxamic acid chelate ion exchange resin has been prepared. I t has been found to be effective with the following ions: V++5,Fe+3, Fe+2, IO+^, Ti+4, Hg+2, Cu+2, UOz+2,and Ca+2.
Figure 1 . Infrared spectrum of hydroxamic acid chelate resin.
are k n o a n to form colored chelates with a number of metal ions, and these chelates have been quite useful in analytical chemistry. Hydroxamic acids can be formed from carboxylic acids by esterification and then treatment with hydroxylamine ( 5 ) . Carboxylic acid ion exchange resins have been knonn for many years. If a hydroxamic acid could be prepared from a carboxylic acid in a resin bead, the folloaing advantages would accrue: increased selectivity in the resin, elimination of the estraction step commonly associated with hydroxamic acid chelate analysis, and further separation of the chelated ions on the column itself. EXPERIMENTAL
Reagents. All reagents used for the preparation of the resin and subsequent testing were reagent grade. Equipment. A Perkin-Elmer 237 was used to obtain t h e infrared spect r a , and a Sargent, D u m a s apparatus was used for t h e iY2 analysis. Preparation of the Chelate Resin. T h e procedure of Renfrow and Hauser ( 5 ) had to be changed slightly for the resin bead system. This required heating, stirring, and passing a stream of N2 through t h e reaction mixture before the chelate resin could be prepared. I o n Exchange Columns. Two 10cm. columns were constructed in 50ml. burets. One column was Amberlite IRC-50 and the other was t h e hydroxamic acid of hmberlite IRC-50.
since the hydroxamic acid group in the resin would be fixed in position, 3 : l and 2 : l ligand to metal chelates would probably not form as when they are free in solution. The chelate resin produced these violet colors when the vanadium or iron was added to the column. None of the compounds used anywhere in the synthesis would produce these colors when passed through either column nor would the metal ions produce any color on the carboxylic acid column. Since the colors are a well known characteristic of hydroxamic acids, it is believed that the hydroxamic acid resin was prepared. The application of infrared spectrometry to resin beads was accomplished by drying the beads, grinding them to a fine powder, and preparing KBr pellets. Differential spectra between the unreacted resin and the hydroxamic acid resin did not give satisfactory results because of the difficulty in obtaining two pellets of the same concentration and thickness. However, spectra of the individual compounds were obtained. Figure 1 is a spectrum of the chelate resin. There are three peaks that appear in the 1460-1325 cm.-l region in the chelate resin spectrum that do not appear in the original resin. These are attributed
Table I.
to the hydroxamic acid group. I t was noticed that these peaks change in nature with the change from the salt form of the resin to the acid form which is to be expected. The small peaks at 1240 em.-' and 1140 cm.-I are believed to be caused by the ester which was not converted to the hydroxamic acid. I t should be noticed that the peak a t 1140 cm.-l intensified when the resin was washed with HCl. This could be accounted for by considering that, the mineral acid might cleave some of the ester groups thereby forming the carboxylic acid group and methanol, which was not completely removed. Also traces of methanol would contribut'e to the broadening of the band at 3500-2800 em.-' which was observed. The peak of 950 cm.-I can be attributed to the N-0 stretch as done by a number of workers (2-4). Therefore, the infrared spectrum appears to show the presence of the hydroxamic acid group. Amberlite IRC-50 contains 9.5 meq. per gram of dry resin ( I ) . This would mean that the amount of nitrogen in the chelate resin would be about 701, if the resin is considered to be a 1 : l copolymer of divinylbenzene and methacrylic acid and as much as 12% if it was all polymerized methacrylic acid. I n the normal synthesis of hydroxamic acids the conversions are: acid chloride, 80-85%; the ester, 70-75%; and the hydroxamic acid step, 50-60'%. Therefore the overall reaction is 28-35% and the per cent nitrogen found would be expected to be between about 2-7y0. The actual value obtained was 3.88% which is quite reasonable. The nitrogen analysis does not prove that a hydroxamic acid group is
Comparison of Regular Resin to Hydroxamic Acid Resin
RESULTS AND DISCUSSION
The most pressing problem is to show that the hydroxamic acid is formed. Three approaches were used to establish this. The first was a visual color reaction, the second was infrared spectrometry to show the presence of the hydroxamic acid group and the third was a nitrogen analysis of the final product. The visual method is based upon the fact that the 1 : l ligand to metal chelate of vanadium(V) is an intense purple and that of the iron(II1) is reddish-purple. It was believed that
Ion 1 7 +5
Fe + 3 Fe + 2 MO+~ Ti +4 Hg + 2
c u +2
uo2 + a Ce + 4 g : 2
Solvent or pH 12N HC1 6N HC1 Hz0 H20
Hz0 H20 H20 HzO
pH = 0 H20 Hz0
Detector Benzohydroxamic acid Renzohydroxamic acid Benzohydroxamic acid Benzohydroxarnic acid Benzohydroxamic acid Benzohydroxamic acid in decanol Benzohydroxarnic acid in decanol 8-Quinolinol Salicylic acid HCl H2C201
Difference in elution volume ; ml. 2 1 5 4 3 1 5 2 0 5-1 1 5 0
0
VOL. 37, NO. 7, JUNE 1965
919
present, but it certainly does show that a nitrogen-containing group has been added to the resin. When all three of the forementioned studies are put together it seems reasonable to say that the hydroxamic acid resin has been formed. The main purpose of this work was to prepare the chelate resin. As such, no distribution studies were made nor was every available ion passed through the column. However, several ions were tested to show that the chelate resin was indeed different from the
original resin. Table I shows the results of such a study. The first 9 ions have been reported previously to react with the hydroxamic acid group while the last two are not known to react with hydroxaniic acids. LITERATURE CITED
(l!' Calmon, Calvin, Kressman, T. R. E.,
Ion Exchangers in Organic and Biochemistry," p. 122, Interscience, New York, 1957. ( 2 ) Gigiere, P. A,, Liel, I. D., Can. J . Chem. 36, 948-62 (1951). (3) Hadzi, D., J . Chem. SOC.1956, 2725.
(4)Nightingale, R . E., Wagner, E. R., J. Phys. 22, 203 (1954). ( 5 ) Renfrow, W. B., Jr., Hauser, C. R., J , 4 m , Chem, floc, 59, 2308 (1937).
GLENNPETRIE DAVIDLOCKE CLIFTON E. MELOAN Department of Chemistry Kansas State University Manhattan, Kan. RECEIVEDfor review January 15, 1965. Accepted March 29,1965. The support of the National Science Foundation summer fellowship program is gratefully acknowledged.
induction-Coupled Plasma Spectrometric Excitation Source SIR: During the past two years we have employed various experimental configurations for forming plasmas of the inductively coupled type ( 9 , 10) and have evaluated their potentialities as a source for the excitation of atomic and molecular spectra. One of these configurations was particularly useful as a pract'ical source for analytical spectrometry. I n this configuration, an induction coupled plasma is maintained by a high-frequency, axial magnetic field in a laminar flow of argon at atmospheric pressure. S o electrodes are in contact' with this discharge as opposed to the capacitively coupled plasma-i.e., highfrequency t'orch or radio-frequency discharge (6, 7)-and d.c. plasma jet. When the discharge gas in the induction-coupled plasma is pure argon, a temperature of about 16,000" K. is obtained a t or near thermal equilibrium (9, 10). To contain a discharge of this temperature without' wall contamination, a carefully controlled, laminar flow of cold argon surrounds the plasma. Atomic spectra are obtained by introducing ultrasonically generated aerosols int,o the argon flow which supports the plasma. The recent appearance of a paper by Greenfield, Jones, and Berry (3) describing a similar source has prompted this communication on our independent observations. Our experimental arrangement for maintaining the plasma and for int,roducing an aerosol into the discharge is considerably different from the Greenfield design and possesses definite advantages for analyt'ical applications. The laminar flow of gas employed to support our plasma differs distinctly from other published induction-coupled designs. Single-tube versions described by Reed (9, I O ) , Cannon ( g ) , Mironer and Kana'an and coand Hushfar (8), workers (I 4 ) have employed tangential ~
920
ANALYTICAL CHEMISTRY
gas inlets to form vortex flows of high velocity a t the walls and of low velocity in the center. The resulting recirculation of the hot gas was considered essential for operation of these designs. I n the laminar-flow plasma described in this communication, there is no recirculation. I n fact, a pure argon plasma has been operated on flow rates as high as 4.6 liters/minute in the plasma tube (which corresponds to a velocity of 38 cm./second) without any decrease in stability or intensity. Tangential inlets are also used in both tubes of the dual-tube designs of Greenfield, Jones, and Berry (3) and
\
Figure 1 .
,TRANSDUCER
Aerosol generator
Lepel (Lepel High Frequency Laboratories, Woodside, S . Y . ) , but their designs possess several inherent disadvantages in comparison to the laminarflow configuration. First, a vortex flow has more turbulence than a laminar flow and this turbulence may decrease the stability of the discharge. This loss of st,ability is drastically enhanced by minor distortions in t'he walls of the tubes, whereas our plasma has been operated successfully with the coolant tube purposely bent' and kinked. Second, the addition of aerosols of solutions or powders to a tangential flow of gas tends to cause them or their vapors to be thrown against the inner wall of the coolant tube, thus decreasing the transmittance of the tube and devitrifying the quartz. Because the coolant, flow in our design is laminar and mixes only slightly with the hot vapors in the core of the discharge, the coolant tube remains clean and useful throughout months of continuous use. The Forrest plasma torch (Forrest Electronics Corp., Las Vegas, Nev.) is the only other dual-tube, laminar-flow design to our knowledge, but no data have been published on it.s performance. As shown in Figure 1, an ultrasonic atomizing system, similar in design to that of West and Hume ( I I ) , is used to produce an aerosol of the sample solution with an average droplet diameter of about 5 microns ( 5 ) . These droplets are carried into the discharge by an independent stream of argon via a small central tube. The ultrasonic aerosol generator allows the introduction of solutions of virtually any sample concentration (so long as the viscosity is not too high) and of any degree of acidity or basicity. Thus, absolute detection limits are greatly improved. Organic solvent's may also be used if the sample container is made of an appropriate material.
