ISSN:0003-2654    

DOI:10.1039/AN95075FX021

出版商:RSC    

年代:1950

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95075BX023

出版商:RSC    

年代:1950

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95075FP039

出版商:RSC    

年代:1950

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95075BP043

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

JUNE, 1950 THE ANALYST Voi. 75, No. 891 PROCEEDINGS OF THE SOCIETY OF PUBLIC ANALYSTS AND OTHER ANALYTICAL CHEMISTS AN Ordinary Meeting of the Society was held at 7 p.m. on Wednesday, April 5th, 1950, in the Meeting Room of the Chemical Society, Burlington House, Piccadilly, London, W.l. The chair was taken by the President, Mr. George Taylor, O.B.E., F.R.I.C. The meeting was devoted to “Bromine” and the following papers were presented and discussed: “The Determination of Bromine in Brine,” by J. Haslam, M.Sc., F.R.I.C., and G. Moses, A.M.C.T., F.R.I.C.; “The Bromine Content of the Cheshire Salt Deposit and of Some Borehole and Other Brines,” by E. C. Allberry, B.A., J. Haslam, M.Sc., F.R.I.C., and G. Moses, A.M.C.T., F.R.I.C.; “Survey of the In-shore Waters round the Coasts of Great Britain, with particular reference to the Bromine Content,” by R.0. Gibson, DSc., F.R.I.C., and J. Haslam, MSc., F.R.I.C. ; “The Analysis of Bromine and Compounds containing Bromine,” by J. Haslam, MSc., F.R.I.C. PHYSICAL METHODS GROUP THE twenty-fourth Ordinary Meeting of the Group was held at 6 p.m. on Tuesday, January 3rd, 1950, in the Chemistry Lecture Theatre of Imperial College, London, S.W.7. Mr. B. S. Cooper, the Chairman of the Group, occupied the chair and about sixty members and visitors were present. The following papers on Spectroscopy were presented and discussed : “The Determination of Strontium in Sea-Water, a Combination of Flame Photometry and Radiochemistry,” by A. A. Smales, B.Sc., A.R.I.C. ; “A Photo-electric Spectrophotometer for the Visible and Ultra-Violet Regions,” by R.A. C. Isbell, A.1nst.P.; “The Application of the Uvispek Spectrophotometer to Biochemical Problems,’’ by D. C. M. Adamson, F.R.I.C. COMMITTEES, 1950 THE Council of the Society has appointed the following Committees- FINANCE COMMITTEE G. Taylor (Chairman), F. W. F. Arnaud, Lewis Eynon, J. G. A. Griffiths, J. H. Hamence, E. B. Hughes, D. W. Kent-Jones, G. W. Monier-Williams, K. A. Williams (Honorary Secretary). PUBLIC ANALYSTS AND OFFICIAL AGRICULTURAL ANALYSTS COMMITTEE H. E. Cox (Chairman), C. A. Adams, F. W. F. Arnaud, H. H. Bagnall, W. G. Carey, H. Childs, J. F. Clark, S. Dixon, J. H. Hamence, E. S. Hawkins, E. T. Illing, A. R. Jamieson, J. King, A. Lees, J. B. McKean, C. H. Manley, S. E. Melling, G. W. Monier-Williams, H.E. Monk, C. J. Regan, J. G. Sherratt, R. W. Sutton, K. A. Williams, E. C. Wood, E. Voelcker (Honorary Secretary). ANALYTICAL METHODS COMMITTEE E. B. Hughes (Chairman), N. L. Allport, A. L. Bacharach, R. C. Chirnside, B. S. Cooper, H. E. Cox, N. Evers, L. Eynon, J. Haslam, J. H. Hamence, D. W. Kent-Jones, J. H. Lane, A. D. Mitchell, G. W. Monier-\J7illiams, J. R. Nicholls, J. E. Page, A. E. Parkes, W. H. Simmons, R. W. Sutton, G. Taylor, K. A. Williams, D. W. Wilson, D. C. Garratt (Honorary Secretary). * 285 * Ex-oficio member of all Sub-Committees.286 COMMITTEES Pol. 75 SUB-COMMITTEES- Egg Yolk Solids in Salad Cream, Mayonnaise and any other Salad Dressing, determination of- H. E. Cox (Chairman), E. B. Anderson, G. E. Forstner, C. L. Hinton, H.E. Monk, K. A. Williams, C. G. Daubney (Honorary Secretary). Essential Oils- W. H. Simmons (Chairman), A. J. M. Bailey, J. F. Charpy, T. T. Cocking, C. W. Cornwell, G. W. Ferguson, H. T. Islip, P. McGregor, W. M.. Seaber, G. E. Smith, J. H. Seager (Honorary Secretary) . Fluorimeter Development- Freezing-Point of Milk- F. J. MacDonald, R. W. Sutton, J. H. Hameince (Honorary Secretary). Meat Extracts and Similar Products, analysis oj.- H. G. Rees (Honorary Secretary). Metallic Impurities in Foodstufs- A. D. Mitchell (Chairman), D. F. H. Button, C. G. Daubney, G. R. Davies, D. Dickenson, C. L. Hinton, A. G. J. Lipscombe, R. W. Sutton, N. D. Sylvester, P. F. Wyatt, N. L. Allport (Honorary Secretary). Lead Panel-G. Taylor (Chairman), EL. E. Cox,’ C. G. Daubney, D. Dickenson, J.H. Hamence, C. L. Hinton, R. W. Sutton, N. D. Sylvester, P. F. Wyatt, N. L. Allport (Honorary Secretary). Poisons, appointed to investigate Methods of Assay for Various Substances appearing i~ the Poisons Schedules of the Poisons Regadations, 1935- G. Roche Lynch (Chairman), N. L. Allport, W. F. Elvidge, G. E. Foster, D. C. Garratt, C. H. Hampshire, E. F. Hersant, W. H. Linndl, W. A. N. Markwell, W. Mitchell, J. I(. Nicholls, R. A. Stockdale, N. Evers (Honorary Secretary). Soapless Detergents- A. MacArthur, K. A. Williams, D. C. Garratt (Honorary Secretary). Standard Methods- K. A. Williams, D. W. Kent- Jones (Honorary Secretary). Vitamin Estimations- E. B. Hughes (Chairman), A. J. Amos, A. L. Bacharach, H. E. Cox, E. R. Dawson, D. C. Garratt, H. C. H.Graves, D. W. Kent-Jones, S. K. Kon, W. H. Linnell, T. Moran, J. R. Nicholls, F. W. Norris, H. M. Sinclair, F. Wokes, E. C. Wood (Honorary Secretary). Cayotene Panel-A. L. Bacharach (Chairman), R. G. Booth, V. H. Booth, J. Green, T. Barton Mann, F. E. Moon, R. A. Morton, W. A. G. Nelson, Miss M. Olliver, S. Y. T. Thompson, N. T. Gridgeman (Honorary Secretary). Aneurine (Microbiological) Panel-A. J. Amos (Chairman), D. H. Clayson, Miss A. Jones, F. W. Norris, S. A. Price. PUBLICATION COMMITTEE J. R. Nicholls (Chairman), N. L. Allport, F. W. F. Amaud, A. L. Bacharach, R. C . Chirnside, B. S. Cooper, H. E. Cox, F. W. Edwards, B. S. Evans, Lewis Eynon, D. C. Garratt, J. H. Hamence, D. W. Kent-Jones, J. H. Lane, S. E. Melling, G. W. Monier-Williams, F. L. Okell, F. A. Robinson, W. H. Simmons, G. Taylor, L. S. Theobald, E. Voelcker, C. Whalley, K. A. Williams, E. C. Wood, G. H. Wyatt. K. A. Williams (Chairman), B. S. Cooper, J. King. G. W. Monier-Williams (Chairman), E. B. Anderson, H. E. Cox, H. D. Kay, J. King, G. Taylor (Chairman), R. Gordon Booth, Osman Jones, J. King, G. Spall, R. G. Westall, W. H. Simmons (Chairman), H. E. Cox, C. C;. Daubney, S. R. Epton, P. J. C. Haywood, G. Taylor (Chairman), N. L. Allport, R. C. Chirnside, D. C. Garratt, J. R. Nicholls,June, 19501 AIREY AND SMALES 287 LIAISON COMMITTEE E. B. Hughes (Chairman), R. C. Chimside, D. C. Garratt, J. Haslam, K. A. Williams, J. H. Hamence (Honorary Secretary). JOINT COMMITTEE OF THE SOCIETY AND THE ROYAL INSTITUTE OF CHEMISTRY The Society’s representatives are-W. G. Carey, H. E. Cox, J. H. Hamence, H. E. Monk, J. G. Sherratt, G. Taylor and E. Voelcker. DINNER COMMITTEE G. Taylor (Chairman), J. H. Hamence, D. W. Kent-Jones, K. A. Williams (Honorary Secretary) .

ISSN:0003-2654    

DOI:10.1039/AN9507500285

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

June, 19501 AIREY AND SMALES 287 Mercury Drop Control : Application to Derivative and Differential Polarography BY L. AIREY AND A. A. SMALES SYNoPsIs--Controlled disengagement of a mercury drop may be achieved either by the abrupt application of an electrical pulse to reduce the interfacial tension, or by a small lateral movement of the capillary, effectively “shearing” the drop from the mercury thread. The fundamentals of the two techniques are examined and it is concluded that the mechanical method possesses marked advantages. The modification of a standard Cambridge polarograph to permit operation either normally or as a bridge circuit for multiple electrode work, as required, is described and illustrated. The performance of the bridge circuit as used for both derivative and differential workipg is examined theoretically and practically.Illustrative examples of the utility of the two techniques are given. Possible future developments are briefly discussed and some observations on the Ilkovic equation are included. DURING the past few years several investigators1 p2 93 v4 15 have published details of bridge circuits incorporating two dropping-mercury electrodes. The general applicability of these methods, however, has been somewhat restricted by difficulties in the synchronisation of the drop rates. Heyrovskyl attempted to eliminate the objection by the use of streaming-jet mercury electrodes; these are fairly satisfactory in theory, but in practice are cumbersome. Rapidly dropping electrodes (approximately 1 sec.) have also been employed by the same author, with some success.Lingane,* in his review of polarographic development, comments that some means of synchronising the drop rates in these circuits would be a useful improvement in technique. Muller’ has observed that “knock off drops” have been made, but at present are in the category of laboratory toys. SevcW reports the use of electrical pulsing as a means of drop synchronisa- tion in his work on cathode ray polarography. This paper considers the problem of mercury drop control from three angles, vix. (a) fundamental principles governing the method, (b) the application of the technique to standard polarographic equipment, and ( c ) applications based upon nonstandard apparatus. FUNDAMENTALS OF DROP CONTROL Two general methods may be envisaged for inducing the disengagement of a pendant mercury drop from a capillary tube.Either the surface tension may be abruptly reduced to a value which renders the drop unstable, or the shape of the drop may be changed to such an extent that a condition of instability is attained. The primary process of severance from the capillary thread is the result of surface tension forces : gravitational action removes the drop from the vicinity of the capillary tube. The reduction in surface tension required288 AIREY AND SMALES : MERCURY DROP CONTROL: APPLICATION by the first method is obtained by a change of the electrical potential difference between the mercury and the solution: the distortion of the drop in the second method by a movement of the capillary tube.The two techniques are conveniently designated “electrostatic” and “electromechanical’ ’ respectively, and will be considered separately under those headings. ELECTROSTATIC CONTROL- The variation of the interfacial tension between a mercury drop and a salt solution theoretically provides a means of producing instability under any conditions. In practice, however, the extent to which the potential difference between the drop and the solution may be changed is determined largely by the presence and nature of the ions in solution. The mechanism of the process can be understood by reference to Fig. 1, which depicts the equivalent circuit of the cell assembly. If we assume an electrolyte of dilute hydrochloric acid, the passage of an electron current from B to A will charge C,; R, is practically infinite.The potential of C, will build up to a value of about 1 volt, and then reduction of hydrogen ions will commence; R, now has an appreciable finite value, which will vary somewhat as a diffusion gradient is produced. To obtain any further potential change across C,, the impressed current must be much increased because the shunting action of R, is dispropor- tionately increased with an increase of applied potential. [Vol. 75 c, C, R, R3 Fig. 1. C, Anode double layer capacitance C, R, Anode leakage resistance R, Cell electrolyte resistance R, Mercury drop leakage resistance R, Capillary thread resistance Simple equivalent circuit of polarographic cell Mercury drop double layer capacitance The presence of such a large steady current in a cell would be highly objectionable, as it causes changes in local concentration, hydrogen evolution and heating effects, particularly in R,.A practicable compromise is to discharge a condenser through the cell, giving a momentarily large current. A value of 0.1 microfarad at a potential of up to 250 volts has been found suitable when using a mercury pool. anode. If R, is increased by the use of an agar bridge, greater voltages would be needed, with attendant heating effects. Fig. 2 shows a series of instantaneous photomicrographs of a freely falling mercury drop. The salient fact is that during the last stages of its growth a drop assumes a pear shape, rupture occurs at the rapidly constricting neck and, after a few oscillations, a spherical shape is recovered within a few milliseconds. It is inferred that the elongated shape is an indispensable condition for detachment (assuming that reduction of interfacial tension nearly to zero is impossible).Hence, although a small abrupt change of potential during the last 10 per cent. or so of the life of a drop may be sufficient to detach it, the problem of detaching the drop is much more difficult during the early stages of growth. Not only must the change of interfacial tension be greater (k, a greater change of potential difference) but the reduction must be maintained for a longer time-sufficient, in fact, to allow the nearly spherical drop to extend to the required pear shape. In the presence of large con- centrations of hydrogen ion, e.g., in normal acids, it is frequently impossible to effect this without the simultaneous evolution of large amounts of hydrogen and disruptive heating of the capillary thread.In the practical utilisation of this method, it would be very desirable to be able to main- tain a constant drop time of about 3 seconds over the range 0 to -2 volts. If a time of free fall of 3 seconds at -2 volts is assumed, the drop rate at -0.6 volts would be about 6 seconds, and in acid solutions it would be quite impossible to disengage the drop after only 3 seconds of life. This applies with rather more force to the synchronisation of two approximately equal drops. Figs. 5 (a) and ( b ) respectively show the Ilkovic curves obtained at a freely falling drop and at a drop synchronised electrostatically to an external timing unit.The curves wereFig. 2. Instantaneous photomicrographs of a freely falling mercury drop Fig. 1. lnstantaneous photomicrographs of a mechanically disengaged dropFig. 3 (a). Pulsa,tor unit Fig. 7 (b) . Modified Cambridge polarographJune, 19501 TO DERIVATIVE AND DIFFERENTIAL POLAROGRAPHY 289 described on a cathode ray tube, using a modified version of the equipment designed by Randles,* and photographed in the usual way. It can be seen that some distortion of the curve is produced by the electrostatic control. The cause is rather uncertain, but is probably associated with convection stirring, which will be described later. In spite of this modification of shape, the reproducibility is good (the diagram shows a single curve only).In conclusion, it should be noted that the galvanometer in the circuit must be paralysed and protected from the current pulses. The most convenient method is to short-circuit the galvanometer with a relay operated slightly in advance of the one that injects the pulse. Some care is necessary in the disposition of the apparatus with respect to stray A.C. magnetic fields, or mechanical rectification of the induced alternating current may occur, and give rise to random fluctuations on the galvanometer. For cathode ray polarographic circuits, protection is unnecessary. ELECTROMECHANICAL CONTROL- Two possible motions of the capillary tube may be envisaged, vix., (a) vertical axial displacement, effectively “stretching” the drop, and (b) lateral movement, “shearing” the drop from the capillary thread.The primary requirement is that any such motion shall cause the minimum of stirring in the solution, and since effective disengagement is only produced by a rapid movement of the capillary tip, the shearing action is the only practicable method. Fig. 3 shows the apparatus by which this may be achieved. The upper end of the capillary tube (or tubes) is clamped rigidly to the armature of a small electromagnetic relay divested of its contacts. The relay is held open by adjustable springs, and a screw stop provides means for adjusting the travel of the armature. A small piece of thin sheet rubber, about 1/32-inch thick, is inserted into the gap to form a resilient buffer, and the adjusting screw is tightened until the rubber is lightly gripped.If the rubber buffer is omitted, it will be found that the working gap is only of the order of a few thousandths of an inch and, although initially the action may be quite satisfactory, such a small gap soon becomes wedged open by particles of almost impalpable dust. The electromagnet is “pulse” energised by the discharge of a 4-microfarad condenser. By tightening the adjusting screw of the relay armature, a condition is attained in which each discharge produces a very rapid lateral movement of the capillary tip of about 0-2 mm., followed by a highly damped oscillatory recovery. Under these conditions, which are not critical, mercury drops may be cleanly detached at any stage of their growth and under any conditions of polarisation or current flow. Fig.4 shows two typical instantaneous photomicrographs of the detachment of mercury drops at different stages in their lives. The rapidity of the process is apparent from the absence of any trace of the succeeding drop. The freedom from any marked distortion of the falling drop suggests that there is little turbulence produced by the movement. Fig. 3 (a) is a photograph of a pulsator, as it is convenient to name the unit, together with the Post Office type relay from which it was made. The real value of the latter as a starting-point lay in the facts that the yoke is made of high permeability iron showing low residual magnetism and that the “knife edge” method of supporting the armature is ideal for this purpose, permitting easy removal whenever desired.Other parts of the relay are discarded. Mainly for reasons of compactness, the 100-ohm latch-relay coil from a Siemens uniselector switch (standard Post Office equipment) was used. Included in Fig. 3 (b) is a simple control circuit. The only important requirement is that the discharge of the condenser through the pulsator coil shall be abrupt and unhindered by such features as poor contact or “bouncing” of the relay. The latter is very detrimental to the regularity of the galvanometer oscillations. The authors use as a timing control a small geared-down Klaxon split-phase (constant speed) motor to which has been fitted a pair of contacts and a cam. Closing of the contacts operates the relay which discharges a 4-microfarad condenser through the pulsator coil. Very little power is consumed in operating the contactor and the type of motive power for this is not of great importance.Electronic control of the relay by means of a multivibrator circuit has been successfully used and possesses the advantage of easy variation of drop time. Figs. 5 (d) to (f) show the Ilkovic curves delineated as described above, using electro- mechanical control with variable timing. The curves were obtained with the optimum setting of the pulsator and by comparison with Fig. 5 ( c ) , the curve for a freely falling drop, The type of coil used is not very important.290 AIREY AND SMALES : MERCURY DROP CONTROL APPLICATION [Vol. 75 it wilI be seen that the stirring effect is quite negligible, and that the four curves are all congruent within the common portions.Figs. 5 (g) and (12) illustrate the effect of progressively increasing the movement of the capillary tip to 1 to 2 mm. Figs. 5 (i) and ( j ) are curves obtained from a thallium solution in the absence of maximum suppressor, with and without tY t Fig. 3 ( b ) . Pulsator (section) and control circuit A Adjusting screw B Armature - pole face gap. Sheet rubber insertion M Motor driven contactor C R Potentiometer 50,000 ohms w/w R, 20,000 ohms W Westinghouse rectifier, H 100 Paper condenser, 4 pF., 600 v. wkg. RELAY. P.O. type 10,000-ohm. coil, tungsten contacts drop control respectively. Three curves at corresponding potentials are shown in each figure. The experiments were carried out to ascertain wht':her the initial slight stirring during drop disengagement modified in any .way the stirrhg present during maximum production.The latter stirring is always very irregular, and the photographs from which the figures were prepared suggest that there is no marked effect. Fig. 6 shows some typical curves obtained on a Cambridge polarograph using a drop controlled as described. The regularity of the drop wave is quite satisfactory, and under conditions of constant rate of mercury flow, the variation of wave height with drop time is clearly seen. Reference will be made to this and similar experiments when the IlkovicJune, 19501 TO DERIVATIVE AND DIFFERENTIAL POLAROGRAPHY 29 1 I Fig. 5 (a). Free fall Fig. 5 (b). Electrostatic control I Fig. 5 (c). Free fall, t = 4 sec. 1 Fig. 5 (d). Electromechanical control, t = 3-5 sec.Fig. 5 (e). Electromechanical control, t = 2.5 sec. Fig. 5 (f) . Electromechanical control, t = 1.5 sec. .Fig. 5 (g) . Electromechanical control. Increased amplitude of capillary movement Fig. 5 (k) . Electromechanical control. Increased amplitude of capillary movement292 AIREY AND SMALES : MERCURY DROP CONTROL: APPLICATION equation is discussed, At the optimum settin,g of the pulsator, relatively large variations in the electrical energy in the condenser (f50 per “cent.) are practically without effect on the wave; an increase in capillary movement results in a slight increase in the step height antl in extreme cases to some “raggedness” of the wave top, as might be anticipated from the Ilkovic curve shown in Fig. 5 (h). It is estimated that an increase in capillary-tip displace- ment of 10 times produces approximately a 2 per cent.increase in wave height. Control to within 0.5 per cent. should be maintained with ease. [Vol. 75 9 Fig. 6 (i) . Electromechanical control. Three curves, (a) preceding maximum, (b) at maximum and (c) following maximum maximum Fig. 5 ( j ) . Free fall. Three curves, (a) preceding maximum, (b) a t maximum and (c) following The magnetic field of the pulsator will induce an oscillatory current in the electrodes and associated wiring. It is desirable that this should decay to zero before the drop is detached, or else partial mechanical rectification may occur, leading to the possibility of erratic fluctua- tions of the galvanometer. Precise data are not available, but this factor will probably set an upper limit to the speed of action (i.e., time between injection of pulse and disengagement of drop) of a pulsator, and it should be considered in any future designs.APPLICATIONS INVOLVING STANDARD POLAROGRAPHIC EQUIPMENT I t will be appreciated from the above that of the two methods of drop control, the electromechanical is much the more practicable. A limited success has been obtained in the use of the electrostatic method on the cathode ray polarograph, but continuation of this work is not envisaged. A single controlled capillary, used in conjunction with a conventional type of polarograph, may have limited advantages in so far as the product mitt is then independent of the applied potential. Furthermore, provided that a suitable capillary is chosen, the variation of t with capillary-active substances--e.g., proteins-may be obviated.The real value of the control scheme however, lies in the ease with which two capillaries may be accurately synchronised, and following from this, it is profitable by slight modifications and additions to extend the versatility of commercial polarographs. The circuit of the Cambridge polarograph is suitable for modification to bridge working. Fig. 7 (a) shows the electrical modifications madLe; Fig. 7 (b) shows the spatial arrangement of the additional components within the instrument. Apart from a pulsator, a controller, and a suitable stand for holding two capillaries, a large 2500-ohm potentiometer (standard electronic component) and a 2-volt accumulator are required.A change from normal working to bridge conditions is effected simply by the operation of the 6-circuit switch (standard Post Office equipment). On the latter setting, change from derivative to differential working is effected simply by moving the wander lead on the delay voltage potentiometer to the zero position. To compensate for variations in the e.m.f. of the 2-volt accumulator, a small variable resistance is included in this latter circuit. A 1-volt tap is taken from this network via a press-button switch to the wiping contact of the main drum potentiometer. With this set at 1 volt, any lack of equality in the two potentials will cause, on operation of the push- button, a deflection of the instrument galvanometer. Adjustments are made to the 50-ohm variable resistor in the delay voltage circuit until balance is attained. A precision of 1 per cent.is realisable, but the ultimate reference standard is the current milliameter on the instrument. Alternatively, for work of very high accuracy, the circuit could be standardised by the use of a Weston cell and an additional galvanometer.June, 19501 TO DERIVATIVE AND DIFFERENTIAL POLAROGRAPHY 293 No information on the circuits of other types of polarographs is accessible, but it is con- sidered that modifications such as the ones described should be quite practicable. For convenience and neatness, the circuit diagram of the Cambridge instrument is not identical with the wiring layout. Fig. 7 (a). P = standardising press switch Circuit diagram of modified polarograph Additional circuits shown dotted.Switches S1, S2, S3, S4 and S5 ganged on one frame. A Polarographic cell position-normal working B Polarographic cell positions-derivative and differential working DERIVATIVE POLAROGRAPHY- Following Heyrovsky, derivative polarography is defined as polarographic work involving the automatic production and delineation, with respect to the applied voltage, of curves representing the slopes of normal waves. In contrast , differential polarography involves the elimination or minimisation of unwanted waves or residual currents by automatic subtract ion. The simple circuit described by Heyrovsky,l while adequate to illustrate the principle of the method, is unsuitable for general analytical purposes. Reference to Fig. 8 (a) will show that there is a steady current through the galvanometer even when connections are not made to the electrodes.The actual difference in potential between the two dropping electrodes is thus equal to the IR drop across the galvanometer resistance. Since it is very desirable to include the galvanometer in an Ayrton shunt circuit, and the net resistance of this varies with the setting, the true voltage difference, which we may designate as hv, becomes a function of the shunt setting. To this may be added the facts that there is a permanent deflection of the galvanometer, also variable with the shunt position, and that other than the torsion head there is no means of adjusting the zero.294 In the circuit of Fig. 8 (b) these inconveniences are all avoided, although an additional source of e.m.f.is unfortunately necessary. The tapped resistance provides values of Ahv from 0 to 50 mv. in nominal increments of 10 mv. AIREY AND SMALES : MERCURY DROP CONTROL: APPLICATION [vol. 75 (4 (b) Fig. 8. (a) Heyrovsky derivative circuit, and (b) modification Making the assumptions of identical electrodes and equality of bridge arms, a simple theoretical analysis of the circuit performance may be made. Differentiating the fundamental equation12 we have- i E = E, - QT log - nF z D - i where- E = E i = i = i, = Q = % = applied potential half-wave potential number of electrons in reduction current flowing limiting current gas constant and, re-arranging, at the half-wave potential, i.e., where i = &,, we have- Let Av be the applied delay voltage and let 6v be the voltage drop across the galvanometer network in consequence of the current flowing.Approximating the central portion of a polarogram to a straight line, the following equations hold when 6v is a maximum nF ( h v - Sv) .. * * (1) a 3 = - ' :{ I - - 2QT -T) .. .. Applying the Kirchoff laws to the network, substituting and simplifying, we have- -Fig. 6. Normal polarograms, electromechanical drop control, of 0.001 M cadmium in solution containing 0.5 M potassium chloride + 0.01 per cent. of gelatin. S = 1/50 Curve . . . . A B C D E F t, sec. . . . . 3.90 3.05 2-50 1.93 1.50 1.00 Fig. 13. Normal and differential polarograms of 8 x 10-5 lead in solution containing 0-25 M potassium nitrate + 0.01 per cent. of gelatin. Curve A, simple polarogram; curve B, residual current; curve C , applied counter current; S = + curve D, differential polarogramFig.9 (b) Fig. 9 ( L ) Derivative polarograms of 0.001 A1 cadmium in solution containing 0.5 &I potassium chloride + 0-01 per cent. of gelatin. S = 1/10. For details see opposite pageJune, 19501 TO DERIVATIVE AND DIFFERENTIAL POLAROGRAPHY 295 The significant inferences to be drawn from this equation are as follows- (a) If other factors are constant, i, cc Av. Ignoring the second term in the denominator (it is shown below that it may be neglected) and if other factors are constant, ( b ) i, cc n. (c) Similarly, i, oc 2RI 2R1 + R' Since R is variable with shunt setting, the total sensitivity will thus be compounded of a shunt sensitivity and a network sensitivity.(d) If other factors are constant, i, is not strictly proportional to i,. If (d) is considered in more detail it will be seen that the magnitude of the error depends upon the product i, x R. Approximate proportionality exists between i,, the maximum bridge current, and iD, the cell diffusion current, but a graphical plot of i, x R against iB would show discontinuities because it is necessary to reduce the sensitivity (and thus the value of R) whenever full-scale deflection of the galvanometer is attained. The magnitude Fig. 9. chloride + 0.01 per cent. of gelatin. (a) Effect of variation of Av Derivative polarograms of 0.001 M cadmium in solution containing 0.5 M potassium S = 1/10. (Facing this page) Curve . . . . A B C D E F Av, mv... 0 9 18.2 27.2 36.4 46-3 Curve . . . . A B C D (b) Effect of resistance in anode lead (Av = 18.2 mv.) Resistance, ohms 0 1000 5000 15,000 (c) Effect of bridge asymmetry (Av = 18.2 mv.) Curve . . .. A B C D E F Rl = R, R, < R2 Rl > R2 R, = R, 51 = p2 Y J -. ZD, = ZD, ZD, = ZD, i D l > ZD, $0, < ZDs R l i ~ , = R$D, Conditions of the error may be calculated in the following manner. For a two-electron reduction, the ratio of i, to i, is, with'the circuit constants shown, approximately 0.2. At any given setting of the Ayrton shunt, the bridge current giving full-scale deflection, together with the total resistance R, may be measured or calculated. The corresponding value of i, is then obtained from the ratio above and, hence, for full-scale deflection conditions the magnitude of the error fact or, nF i, 4m 2 2R1R - ? 2% + R may be computed.Proceeding on these lines, it has been calculated that for the particular gahanometer and circuit constants employed, the error factor is -0.7 per cent. at a shunt setting of 1, rising to -3.5 per cent. at a sensitivity of 1/1000. As an example, if a comparison were being made at S = 1/5 (error factor = 3.0 per cent.) between two solutions which should give deflections of 100 (full scale) and 50 respectively, the actual values obtained would be 97 and 49-25, as the error factor for the second peak is only 1-5 per cent. The observed ratio would be 1.97 instead of 2.00, an err0.r of only 1.5 per cent. A similar error occurs for other ratios. Furthermore, since i, a iB/n, the error factor, which is proportional to i, x n, will be proportional to i, and independent of n.The validity of relation (a) is shown by the results in Table I. The departure from proportionality at greater values of Av than 25 mv. is due to the non-linearity of the polaro- gram over such extended ranges. The derivative curves are presented in Fig. 9 (a) as typical examples of the performance of the apparatus. The effect of bridge setting on the over-all sensitivity is demonstrated by the results of Table 11. The constancy of the results in the last column is satisfactory. Verification of (b)296 AIREY AND SMALES : MERCURY DROP CONTROL : APPLICATION TABLE 1 VERIFICATION OF THE RELATION iBa Av Data from experiments similar to Fig. 9 (a) [Vol. 75 hv, Peak height, b/Av mv.h 0 0 9 16-2 1-69 18.2 30-'7 1.69 27.2 44.10 1.62 36.4 85.1 1-61 46.3 64.2 1.39 I has not been attempted and, for the particular instrument employed, the error in (d) is not of sufficient practical importance to justify the extreme accuracy which would be necessary to verify its existence by experiment. TABLE I1 RELATION BETWEEN GALVANOMETER DEFLECTION AND SHUNT SENSITIVITY Derivative polarograms for solution of 0-001 M cadmium in 0.5 M potassium chloride; Av = 18-2 mv., R, = 2000 ohms FOR BRIDGE CIRCUIT Shunt sensitivity, Peak height, S h R h 2 R , + R -x - h S S 2R, I 61.5 336 1-084 308 334 46.2 253 1.064 316 336 32-2 185 1.047 322 337 21.7 127 1.032 326 336 1/6 1/20 16.3 96 1.024 326 333 1/30 11.0 66 1.018 330 335 The over-all validity of equation (3) is demonstrated by the results of Table 111.It may be noted that the simple equation derived by Heyrovskyl does not adequately represent the behaviour of a derivative bridge circuit. TABLE I11 $0 1/16 VERIFICATION OF EQUATION (3) Solution of 0.001 M cadmium in 0.5 A l potassium chloride; t = 3.05 sec. iB i B Av, .-obs. - theoret. i D , iB, S R, 2% T p amps. p amps. ohms ohms mv. ZD iD 6.98 1.09 1/10 185 4000 21OC. 18.2 0.16 0.18 The postulated condition of identical electrodes is very rarely attained in practice. If the two diffusion currents are iD, and iDn it may be shown that, providing equation (3) is still valid and may be re-written as- Rxi,, = ?32iDt, &, is preferred to i,,, since R, is a fixed resistance). Note, however, that R, + R2 + R replaces 2R, + R in the denominator, which implies a changed bridge sensitivity. The effect of extraneous resistance in the circuit is dependent upon its position. If, for example, the resistance is common to the two cells, as in an agar bridge, the peak is merely widened.Fig. 9 (b) illustrates this. The instrument measures di/dV, where V is the potential difference across the mercury surface, but delineates it with respect to E , the total appliedJune, 19501 TO DERIVATIVE AND DIFFERENTIAL POLAROGRAPHY 297 potential. If, however, the position of the resistance is such as to produce bridge asymmetry, as for example, a very high resistance capillary, the unbalanced voltage drop either assists or opposes Av, with a corresponding change of peak height. Since i, is proportional to Av, the magnitude of such errors may be estimated by expressing the potential difference across the extraneous resistance as a percentage of Av. Ideally, both capillaries should have zero resistance; the usual value of about 50 ohms produces but negligible errors except at high currents.THE TECHNIQUE OF DERIVATIVE POLAROGRAPHY- Since, as is mentioned above, identical rates of mercury flow in the two capillaries would be improbable, the primary requirement is that the condition expressed by the equation R,&, = R&,* (see above) be established. Fig. 9 (c) shows a number of derivative polarograms obtained under conditions when this equation was not satisfied, together with one designed to prove its validity. In the latter case the recovery of the original shape will be observed.It will be seen that the curves obtained may, to a first approximation, be regarded as com- pounded of a derivative curve and a positive or negative differential polarogram (dotted curve). For small deviations from exact balance, the peak height measured from the fore- foot of the curve will be the algebraic sum of a derivative peak proportional to the concentra- tion of reducible ion and half of a differential wave, also proportional to the ionic concentration. Hence, except for the verification of equation (4), balance to within 5 to 10 per cent. is adequate. Necessarily, any such setting must be maintained constant for a series of experiments. The method of balancing the bridge assumes a primary polarographic curve of ideal shape.The applied cell potential is adjusted to 0.3 to 0.4 volt more negative than the half- wave potential, i e . , both electrodes are operating under conditions of limiting diffusion, and the variable resistance arm of the bridge is adjusted until there is no galvanometer deflection. In general, this must be done as a separate initial procedure, using some ion such as cadmium or thallium, as the derivative method is only of value when dealing with waves which are but poorly defined. I 2 i 3 i P 4 I 5 J Fig. 10 (a). Normal polarograms of lead + cadmium in 0.25 M potassium Cadmium, (1) 1 0 - 4 M , (2) nitrate + 0.01 per cent. gelatin solution. S = 1/10. 1-3 x lo-* M lead M , (3) 10-2M, (4) 10-1 M and (5) 1 M In other respects, the operation of the polarograph is similar to normal working. The slow recording speed must be employed, particularly when working at S = 1 owing to the sharpness of the peak.Additional damping may be secured by the use of the appropriate control, but a reduction in bridge sensitivity is simultaneously effected. Provided that the bridge resistances are all known and approximately constant, it is apparently (see Table 11) satisfactory to calculate the over-all sensitivity and make a correction table for the Ayrton shunt. Alternatively, calibration at each setting is feasible, although time-consuming. Illustrative examples of the possibilities of derivative methods are provided by Figs. 10 and 11. Fig. 10 (a) shows the ordinary polarograms of solutions of lead and cadmium in which the ratio of lead to cadmium was vaned from approximately 1/1 to l/l@.Figs. 10 (b) and (c) are the corresponding derivative polarograms, the experiment having been designed298 AIREY AND SMALES : MERCURY DROP CONTROL APPLICATION Fol. 76 to ascertain the point at which the simple proportionality between peak height and con- centration is vitiated by interference from the cadTmium peak. It will be seen that significant interference, due to a constant contribution from the fore-foot of the cadmium peak, only appears in the last example. Even under these conditions, an approximate correction for the 2 A ni Fig. 10 (b). Derivative polarograms of lead + cadmium in 0.25 M potassium nitrate + 0.01 per cent. gelatin solution. Av = 18.2 mv., S = 8 1 Cd = 104M Pb = 1.3 x M Pb = 3.9 x 104M 2 3 Cd = 104M Cd = 10-sM (A) Pb = 1.3 x 10-*M (A) Pb = 1.3 X lo-* M (B) Pb = 2.6 x 10-4 M (B) Pb = 2.6 X lO-"M ha/hb = 2.00 ha/hb = 2.00 A n I a 8 ~ 3 A 1 0 ?" Fig.10 (c). Derivative polarograms oE lead + cadmium in 0-25 M potassium nitrate + 0-01 per cent. gelatin solution. Av = 18.2 mv., S = 8 1 2 3 (A) Pb = 1.3 x lo-* M (A) Pb = 1.3 x M (A) Pb = 1.3 x lO4M (B) Pb = 2.6 X lO-4M (B) Pb = 2.13 x 10-4 M (B) Pb = 2.6 x 10-4 M Cd = 10-aM Cd = 10-IM C d = l M ha/hb = 1.99 ha/hb = 2.02 ha/hb = 1.88 interference could be obtained by making a 'blank measurement on a lead-free cadmium solution of equal concentration, Le., by subtracting the height of the cadmium peak fore-foot at the lead half-wave potential. Figs. 11 (a) and (b) respectively show the normal and derivative polarograms of lead in admixture with increasing amounts of thallium.The half-wave potentials of the two waves differ by 0.08 volt.June, 19501 TO DERIVATIVE AND DIFFERENTIAL POLAROGRAPHY 299 No consideration has been given to the use of the derivative curve for qualitative purposes. Corrections for delay voltage, net bridge resistance, cell resistance and other factors might be of importance. 2 I 4 Fig. 11 (u). Normal polarograms of lead + thallium in 0.25 M potassium nitrate + 0.01 per cent. gelatin. Thallium, (1) nil, (2) 0-5 x M , (3) 2 x M , (4) 5 x Mand ( 5 ) 20 x M S = 1/15. 2.7 x lO-4M lead I 2 3 4 Fig. 11 (b). Derivative polarograms of lead + thallium in 0.25 M potassium Mand (5) 20 x 10-4 M nitrate + 0.01 per cent. gelatin. Thallium, (1) nil, (2) 0-5 x lo-* M , (3) 20 x Av = 18.2 mv., S = i.M , (4) 5 x 2.7 x lo4 M lead DIFFERENTIAL POLAROGRAPHY- With the omission, for simplicity, of the resistances rl and r2, the circuit of Semerano and Riccoboni2 has been adopted (Fig. 12). A simple theoretical analysis is possible on the following considerations. Assuming a symmetrical bridge with identical electrodes, let a limiting current of i, flow in one cell and a current of i, + it, in the other. It is required to measure i t , and eliminate i,. If attention is focussed on the horizontal portions of the waves, the relation between cell current and voltage is expressed by di/dE = 0. With this condition, a simple application of the Kirchoff laws gives .. (5) * . .. .. R2 2R, + R i, = it, The over-all sensitivity of the bridge plus galvanometer circuit can, therefore, never exceed 50 per cent.of the sensitivity of the galvanometer alone at a comparable setting. The validity300 of the equation is shown by the results given in Table IV. requires, as before, that R1iDt = R2iDe, and the equation assumes the form AIREY AND SMALES : MERCURY DROP CONTROL : APPLICATION [Vol. 75 Inequality in the electrocies . . . . .. i, = i ' D R2 - R1+ R2 -+ R ' * TABLE 1.V VERIFICATION OF EQUATION (5) Solution of 0.001 M thallium in 0.25 M potassium nitrate; S = 1/15] t = 3.05 sec. iD. iB, R, is - R, p amps. p amps. ohms ohms i D 2R, + R 2.64 1.29 2000 127 0.487 0-484 THE TECHNIQUE OF DIFFERENTIAL POLAROGRAPHY- As for derivative work, the balancing of the bridge is the most important requirement.The purpose of the differential method is to reimove unwanted currents, whether residual or otherwise. If it is known that after such removal an ideally or nearly ideally shaped polarogram should be obtained, the procedure is merely to adjust the variable resistance bridge arm until such a shape is obtained. If, however, as in an irreversible electrode process, a distorted wave is expected, it is desirable to balance the bridge initially with e.g., a cadmium or thallium solution. As two electrodes dropping into two solutions of possibly different concentrations are now concerned, the number of variables is increased. For theoretically perfect compensation, the requirement that R,i, = R2i, must be satisfied for both the reduction and condenser currents.Ideally, therefore, the bridge should be balanced with the same cadmium or thallium solution in both cells, the unknown solution should be placed in cell 2 and the concentration of a solution containing the ion giving the unwanted wave should be adjusted until balance is again attained. Practically, this would be tedious, and no noticeable error results if the final accurate balancing is done by the variable bridge resistance. . i3 . '4 Fig. 12 Differential circuit of Semerano and Raccoboni The relatively high temperature coefficient of the diffusion current (2 per cent. per O C.) necessitates temperature equality, if not temperature constancy, in the two cells, and a thermostat is almost indispensable. As yet, insufficient evidence has been accumulated to.decide whether the two electrodes must be operated on one pulsator or whether two pulsators, each carrying one capillary and operated simultaneously, are satisfactory. The latter scheme is advantageous in that the standard Cambridge thermostat and dropping electrode stand are immediately adaptable. In the case of the derivative method, the separation of the two electrodes should be a minimum and the question does not arise.June, 19501 TO DERIVATIVE AND DIFFERENTIAL POLAROGRAPHY 301 Perfect compensation of two waves at aZZ voltages requires, apart from the conditions already discussed, that they have identical apparent half-wave potentials and shapes. Any differences in these quantities, caused by the inequality of calomel half-cells or cell resistances, will be observed by the galvanometer both in magnitude and sign, giving rise to minor derivative effects superimposed on the differential.The resistances r1 and y2 (Fig. 12) are Fig. 14 ( a ) . Normal polarograms of zinc + cadmium in 0.5 M ammonium M (1) Cadmium, nil; S = l j 5 . (3) Cadmium, chloride + 0.5 M ammonium hydroxide -+ 0-01 per cent. gelatin solution. zinc , 8 x (2) Cadmium, 8 x lo4 M ; S = 1/30. 2 x 10-3 M ; s = 1/20 designed to minimise such phenomena. However, within the ratio limit of 10/1 discussed below, the effects are small and the two compensating resistances may be discarded. As shown by the former example, nearly perfect elimination of the residual current over the whole working voltage range may be achieved. This is an attribute not possessed by the counter-current method of Ilkovic and Semerano, and more than compensates for the 50 per cent.loss in sensitivity. Fig. 14 illustrates the minimisation of a large cadmium wave preceding a zinc wave. The polarogram of Fig. 14 (6, 3) is anomalous. Fig. 14 (c) presents three consecutive repeats carried out several hours later (a somewhat smaller mercury flow rate, in consequence of a The utility of the differential technique is shown by Figs. 13 and 14. 2 3 Fig. 14 (b). Differential polarograms of zinc + cadmium in 0.5 M ammonium S = Q. chloride + 0-5 M ammonium hydroxide + 0.01 per cent. gelatin solution. 8 x 10-5M zinc Cadmium, (1) nil, (2) 8 x M and (3) 2 X 10-aM302 AIREY AND SMALES MERCURY DROP CONTROL: APPLICATION [Vol. 75 reduced head, was employed).The regularity of the galvanometer oscillations in each case is good: from the known cadmium to zinc ratio of 25/1, it will be seen that the average reduction current associated with each drop is constant to well within 0.1 per cent. It must be emphasised, however, that such regularity can only be obtained if the pulsator receives electrical excitations of constant energy and duration. As was briefly noted earlier, the phenomenon known as “relay bouncing” is particularly objectionable. On the first closure of the contacts, the condenser is partially disch.arged and the drop is probably detached. The second closure completes the discharge, causing a further shock to the capillary, which results in a slight irregular stirring and corresponding variations in the reduction current. Normally, such effects would be almost indiscernible, but the differential method magnifies Fig.14 (c). Differential polarograms of zinc + cadmium in 0-5 M ammonium S = 8. chloride + 0.5 M ammonium hydroxide + 0.01 per cent. gelatin solution. 8 :< M zinc + 2 x M cadmium them to such an extent that they may produce: a “raggedness” of the wave which almost obscures it. With reasonably careful working it is possible to carry out estimations on solutions in which the ratio of the unwanted to the desired wave is 100/1. By comparison, the simple compensation procedure of Lingane and Kerlinger9 is applicable in favourable cases up to a limit of 50/1. However, it is considered that above a ratio of 10/1 the use of automatic controlled potential electrolysis at a mercury cathode is a much more effective method of dealing with large preceding waves.The special merit of the differential method is that, unlike that of Lingane and Kerlinger, it is possible to compensate waves of any shape, such as those frequently obtained in the reduction of organic compounds. APPLICATIONS INVOLVING NON-STANDARD EQUIPMENT MULTI-TIP ELECTRODES- McGilvery, Hawkins and ThodelO have described multi-tip electrodes as a means of increasing sensitivity. Simultaneously, there occurs a proportionate increase in the condenser current, but it is claimed that the natural asynchronism of dropping rates materially reduces what would otherwise be an enhanced drop wave. Bricker and Funnanll have criticised this latter statement, claiming that an irregular wave-top results. A five-tip electrode has been made (unfortunately of a very rapid drop rate) and the drops have been shown to be capable of synchronisation by means of a pulsator. Work is in abeyance owing to the difficulty of obtaining a suitable size of capil1ar:y tubing, but there appears to be noreason why this type of electrode should not prove very profitable if the residual current iseliminated by means of a differential circuit.COMBINATIONS OF DERIVATIVE AND DIFFERENTIAL METHODS- The practicability of the separation of two waves by derivative methods depends upon the difference of half-wave potentials and the ratio of the concentrations. Instances may arise in which a separation is feasible when the two substances are present in equal amounts,June, 19501 TO DERIVATIVE AND DIFFERENTIAL POLAROGRAPHY 303 but not when present in a ratio of, for example, 1/10.Theoretically it should be possible to set up two derivative bridge circuits, one operating on the second reducible substance only. By means of a galvanometer having independent differential windings, automatic subtraction of the two derivative curves should be feasible. Numerous other combinations might be advanced; in particular it should be possible to gain the desired results with only three synchronised mercury drops, but as yet no satisfactory circuits have been designed. CATHODE RAY POLAROGRAPHY- The apparatus devised by Randless has been elaborated to permit differential operation. As an example of the performance, one part of zinc in 104 parts of cadmium may be detected- with certainty.The work is as yet in a preliminary stage and will be reported fully in a subsequent paper. MISCELLANEOUS OBSERVATIONS The Ilkovic curves of Fig. 5 show that, with proper adjustment of operation, the pulsator has no directly discernable effect on the diffusion current. It thus becomes possible to separate the variables m and t in the Ilkovic equation. The results of Table V show that the relation i a d is followed to within less than rt2 per cent. over a 2 to 1 range in flow rates. TABLE V VERIFICATION OF RELATION i, a mt Solution of cadmium in 0.5 M potassium chloride; t = 3.05 sec. h 64.0 49.6 44.0 36.7 32.9 A m, m) mg./sec. 1.41 42.9 1.278 42.0 1.066 42-4 0-831 41.5 0-706 41.5 . 42.1 + 1.9% - 1.4% Table VI shows results verifying the relation i a t i to within approximately k2.6 per cent.However, the Ilkovic curves of Fig. 5 do not exactly fit a one-sixth power law, the TABLE VI Series 1 Series 2 VERIFICATION OF RELATION i, cc t* Data from experiments similar to Fig. 6 k 43.3 41.1 39-7 37.8 35.4 33.4 43.0 42.6 40.7 37.5 37.0 32.0 t, sec. 3.9 3.06 2.60 1.93 1.60 1.00 3.9 3-66 3-05 2-15 1.68 0.75 12 t4 34-6 34.1 34.0 34.0 33-1 33.4 - mean 334 + 2.1% - 2.4% 34.3 34.3 33.8 33.0 34.3 33-6 mean 33.9 + 1.2% - 2.7% slope of the initial portions being too low. This could be qualitatively explained by the postulate that during the life of a drop there is some small accumulation (by convection and, possibly, mercury vortex action) of the depleted solution from the diffusion layer around the neck of the mercury drop.This would not be entirely removed by the falling drop,304 AIREY AND SMALES [Vol. 75 and the new drop would therefore grow through a portion of solution of concentration some- what lower than that in the bulk of the electrolyte. With increase of drop size the diffusion conditions would become those required by the Ilkovic equation. Some substantiation of this hypothesis is provided by results obtained on the cathode ray polarograph. With this instrument a polarogram is observed during the life of a drop; one such polarogram may be observed or the process may be repetitive, depending on the particular drop frequency employed. In several cases examined, single polarograms appear to be a few per cent. larger than those obtained when there has been a preceding wave.It would be of interest to examine a reduction process in which the product of the reaction remains in solution and is more dense than the bulk of the electrolyte. A small distortion of the Ilkovic equation such as is considered would not markedly affect the one-sixth power law as determined by an integration of the curve, i.e., a mean current measurement by a galvanometer. There is a suggestion of a drift in the results of Table VI, but the order of accuracy is insufficient for unequivocal deductions. DESIGN OF CAPILLARIES- Composite capillary tubes consisting of a 10-cm. length of 0.05-mm. bore tubing plus a l-cm. tip of 0-l-mm. bore, are employed with the pulsators. The free-drop time may be as great as 15 seconds, but the drop time is easily controlled to the convenient value of 3 seconds. The relatively large bore of the tip makes such capillaries much less liable to clogging, which for protein-like substances may be a decided advantage. ADDENIIUM Since the submission of this work for publication, a paper by L6v6quels has appeared. Mention is made of experiments on electromechanical drop-control, using a somewhat similar scheme to that described, and failure to devise a useful technique was reported. From an examination of the circuit diagram it would appear that the lack of success was probably due to the stirring effects arising from the cessation of the magnet-energising current after the relatively long time of energisation of the driving magnet. Acknowledgment is made to the Director, Atomic Energy Research Establishment, Ministry of Supply, for permission to publish this work. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. REFEREXES Heyrovsky, J., Anal. Chim. Acta, 1948, 2, 536; Analyst, 1947, 72, 229. Semerano, G., and Riccoboni, L., Gazz. Chim. Ital., 1942, 72, 297. Kanewskii, E. A,, J . Appl. Chem. (U.S.S.R.), 1944, 17, 514. Matheson, L. A., Nichols, N., et al., Anal. Chim. Acta, 1948, 2, 541. Sevcik, A., Cold. Czech. Chem. Commun., 1948, 13, 349. Lingane, J. J., Anal. Chem., 1949, 21, 57. Muller, 0. H., Chem, Eng. News, 1949, 27, 847. Randles, J. E. B., Trans. Farad. Soc., 1948, 44, 322. Lingane, J. J., and Kerlinger, H., I n d . Eng. Chem., Anal. Ed., 1940, 12, 750. McGilvery, J., Hawkins, R. C., and Thode, H. G., Cunad. J . Ras., 1947, 25, B 132. Bricker. C. E.. and Furman. N. H.. Anal. Ch.em.. 1948. 20. 1123. Kolthoff, I. M., and Lingane, J. J.-, “Polarography,” Interscience Publishers, Inc., New York, L6v$que, M. P., “Differential Polarography with a Single Dropping Electrode,” J . Chim. Phys., 1946, p. 144. 1949, 46, 480. ATOMIC ENERGY RESEARCH ESTABLISHMENT HARWELL, BERKS. October, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500287

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

June, 19501 DAUBNEY AND SEXTON: DETERMINATION OF EGG IN SALAD CREAM 305 The Determination of Egg in Salad Cream Part I BY C. G. DAUBNEY AND G. E. W. SEXTON SYNOPSIS-A method is proposed for determining the amount of egg yolk solids and dry egg solids in salad cream. It is based on a colorimetric deter- mination of the choline derived from egg lecithin by means of acid hydrolysis and conversion of the choline into its pink reineckate, which is soluble in acetone. The choline content of dried whole egg, sugar-dried egg, and dried egg yolk has been determined and shown to be constant in any one type of product . Mustard, a common ingredient of salad cream, causes a complication on account of the choline derived from the sinalbin of white mustard. The method described determines total choline and is applicable only when the nature and amount of any mustard present is known.Satisfactory results were attained ‘for salad creams of known composition. Work is in progress to devise a means of determining egg and mustard separately . THE determination of the egg content of certain foodstuffs is often required for the purpose of quality control and for checking their compliance with Orders issued by the Ministry of Food. The first of these Orders is the Food Standards (Salad Cream and Mayonnaise) Order, 1945 (S.R. & O., 1945, No. 1177), which lays down that in the case of salad cream, mayonnaise and salad dressing, “the product shall contain . . . not less than 1.35 per cent. by weight of egg yolk solids.” The second is the Flour Confectionery (Control and Maximum Prices) (Amendment) Order, 1950 (S.I., 1950, No.87), by which a certain maximum price may be charged for flour confectionery, which includes cakes, puddings, pastry, etc., provided “the combined fat, sugar, and dry egg solids content (or the combined content of any two of these ingredients) . . . is 45 per cent. or more.” The third is the Food Standards (Preserves) (Amendment) Order, 1949 (S.I., 1949, No. 1893), by which fruit curd shall contain not less than 1 per cent. of dried whole egg (or its equivalent as sugar-dried whole egg, liquid or frozen whole egg or shell egg). It will be noticed that in one Order it is “egg yolk solids” that is mentioned and in another “dry egg solids.” None of the Orders specifies a method for the determination of the egg content.It was in consequence of these Government Regulations that the experiments described below on egg and salad cream were carried out ; work is now in progress on flour confectionery and fruit curd, the result of which it is hoped to publish shortly. So far as the analyst is concerned, the term “egg” generally implies that of the hen. In the main, the white of egg is composed of protein and water, whilst the yolk contains, in addition to protein and water, some 30 per cent. of fatty matter and 1.5 per cent. of cholesterol. Of this fatty matter about one-third is phosphatide, chiefly lecithin. The term phosphatide (or phospholipin) refers to a group of substances in which one or more fatty acids and a nitrogenous base are combined with a third constituent, glycero- phosphoric acid.This group can be further subdivided into (a) the lecithins and (b) the cephalins. The compounds in each sub-group conform to the same type of structure but differ in the nature and position of the fatty acid radicals contained in the molecule. The lecithins are represented by the formulae- C q . OOC. R CW.0OC.R I P C H 4 - P - 0 . CH, .CH,. N i (CH,), I OH I CH.0OC.R’ p-e-rorm I CH,-O-ib-O.CHJH,.N l o i (CH& OH I OH a-form306 DAUBNEY AND SEXTON : TH:E DETERMINATION OF [Vol. 75 The cephalins are identical in structure," except that they have /%amino ethanol (cholamine) in place of choline- CH,.OOC.R CH, . OOC. R I I I 0 I II I CH-0-P-0 .CH,.C&.NH, I CH.0OC.R' I OH CH,-O-P-O.CH,.CH, 1 7 .NH, CH,. OOC.R' I OH a-form 8-form These two groups of phosphatides occur in animal tissues in varying proportions, and in egg the /3-form of lecithin is said to predominate. The phosphatides are soluble in most organic solvents except acetone, whilst cephalin is insoluble in alcohol unless water or other lipoid bodies are present. In considering the analytical chemistry of this class of substance it must be borne in mind that lecithin and cephalin appear to undergo oxidation and hydrolysis more readily than the fats. Lecithin is hydrolysed on boiling with dilute acid into glycerophosphoric acid, fatty acids and the base choline. Similar decomposition is brought about by the enzyme lecithinase which occurs in animal tissue.l $2 Cephalin decomposes similarily to give cholamine.The methods proposed from time to time for the determination of egg in various foodstuffs have usually depended on a determination of either the phosphatide or the cholesterol ~ o n t e n t . ~ In deciding on a method which was likely to be applicable to salad cream, one based on the determination of the phosphatide fraction of egg seemed to be the most promising, either by way of organically-bound phosphorus, as a measure of the total phosphatides, or by way of choline as a measure of le~ithin.~ Methods6$6 $7 98 19 based on the determination of organically-bound phosphorus require the complete extraction of the phosphatide with suitable solvents and the subsequent deter- mination of phosphorus in the extract. Such methods are, however, liable to give low results owing to the susceptibility of lecithin to decomposition by enzymes or acids, with the con- sequent splitting off of the phosphoric acid ra,dicle as mentioned above.1° Further, the physical nature of salad cream renders solvent extraction methods unsuitable.Methods based on the determination of choline after hydrolysis of the lecithin, on the other hand, present certain advantages. The products of hydrolysis will include any choline previously split off by acid or enzyme action and the experimental technique is readily applicable to salad cream. A disadvantage is that under the conditions of hydrolysis to be described, choline is also derived from white mustard (Sinapis alba) which is often an ingredient of salad cream. White mustard contains the glucoside sinalbin, which was shown by Gadamerll to be- H0.C6H4.CH,.N : C.S.C,H,,O, I O.SO$.O I (CH,), N.CH:,.CH, HO.(OCH,),.C,H,.CH : CH.CO. d This is made up of the non-volatile mustard oil, p-hydroxybenzyl isothiocyanate, linked to glucose and the acid sulphate of the choline ester of sinapinic acid.Black mustard (Sinapis nigra), on the other hand, contains the glucoside sinigrin, which is less complex, and contains no choline. An evaluation of mustard has been described by Viehoever and Nelson12 and by Terry * This formula has recently been disputed; see Hutt, H. H., Malkin, T., Poole, A. G., and Watt, P. R., Nature, 1950, 165, 314.June, 19501 EGG I N SALAD CREAM. PART I 307 and Corran,13 both of whom have based their method on a determination of the liberated sulphate after enzymatic decomposition.The non-volatile mustard oil has been studied by several workers, but it does not lend itself readily to determination. The other important ingredient of salad cream which might give rise to choline is the vegetable 0i1.l~ At the present time there are available to manufacturers various oils all of which, in the crude state, contain small but significant quantities of choline. Examination of the refined oils, as used, has shown that they yield no choline when treated by the method given below. It will be seen, therefore, that any method for the determination of egg in salad cream that depends on the choline content of the egg must also be capable of modification when white mustard is present. The method to be described determines total choline and is only applicable when the nature and amount of any mustard ingredient is known.Experiments are proceeding at the present time with the object of determining the choline from egg lecithin or from sinalbin separately. Choline may be determined in a variety of ways, which include c o l ~ r i m e t r i c , ~ ~ ~ ~ gravimetric,lg j20 volumetric21 and micro-biological methods.22 For the present purpose, a colorimetric method was found to be suitable, in which the choline is precipitated by ammonium reineckate, [Cr(NH,),(SCN),]NH,, the choline reineckate filtered off, dissolved in acetone and the (pink) colour of the solution compared with standards similarly prepared. The use of artificial standards of methyl red in a buffer solution has been suggested.EXPERIMENTAL Choline has been determined in the various forms of liquid and dried egg at present available and in salad creams containing known amounts of egg. Where mustard is present the choline content of this ingredient has been determined in the same way, using O G g . of the powder, and due correction made. METHOD- Weigh 0.5 g. of dried egg, 2.0 g. of frozen whole egg or 10 g. of salad cream into a 150-ml. flask containing a few glass beads and add 20 ml. of diluted hydrochloric acid (5 + 3). Warm the mixture on the steam-bath for 5 minutes to reduce subsequent frothing and then boil under a reflux condenser for 1 hour, in apparatus fitted with ground-glass joints. Wash down the condenser with 10 ml. of hot water and transfer the contents of the flask, when cool, to a separator.Extract the aqueous liquid by shaking with 40 ml. of methylated ether in order to remove the bulk of the fatty matter. Allow to separate and run the aqueous layer through a filter into a small conical flask. Wash the ether layer twice with 20-ml. portions of water and run the washings through the same filter. Boil the combined filtrate and washings to expel the ether and reduce the volume. After cooling, add the calculated volume of potassium hydroxide solution to neutralise the acid present, followed by 20 ml. of cold, saturated aqueous baryta solution. Add 1 drop of 1 per cent. alcoholic thymol-phthalein solution followed by glacial acetic acid added dropwise till the blue colour just disappears. Cool the liquid by standing it in a refrigerator for about 1 hour and filter by gentle suction through a filter funnel fitted with a sintered glass plate (No.3 porosity) or a hardened filter- paper. Wash the filter with water and, to the clear combined filtrate and washings, add 6.0ml. of 2 per cent. ammonium reineckate in methyl alcohol. Leave overnight in a refrigerator. Collect the pink precipitate of choline reineckate on a sintered glass filter funnel (No. 3 porosity) with gentle suction, wash twice with 10-ml. portions of ice-cold water and finally three times with 2.5-ml. portions of n-propyl alcohol. When free from solvent, dissolve the precipitate in acetone, collect the solution in a graduated flask and make up to 50ml. with acetone. Transfer to a Nessler tube and match the colour against standards prepared as follows.Dissolve sufficient choline chloride in water to give a solution con- taining 2 per cent. of choline and standardise against 0.1 N silver nitrate.23 From this prepare a dilute solution containing 1 mg. of choline per ml. Add appropriate volumes of this second solution to 150-ml. flasks, dilute to 50 ml. with water, add 6 ml. of 2 per cent. ammonium reineckate in methyl alcohol and proceed as above. In this way prepare the standards that may be necessary containing between 1 and 12 mg. of choline in 1 mg. steps. Fractional standards may be prepared therefrom by dilution with acetone. The colours should be matched within 30 minutes of dissolving the reineckate in acetone as some batches308 DAUBNEY AND SEXTON: THE DETERMINATION OF [Vol.75 of the solvent have been found to cause a marked deterioration in the pink colour on standing. The reineckate precipitates can however be left omn their separate filters after washing with the ut-propyl alcohol and then dissolved when rea'dy for matching. RESULTS Egg-Various types of egg products were examined by the above method. Half a gram was taken for each determination. TABLE :I Product Dried whole egg : Canadian .. Canadian .. Swedish (S quality) Swedish (D quality) Foreign origin . . Sample A . . 11 €3 .. 99 c .. Canadian (grade A) Canadian .. Sample D . . Sugar-dried egg : Choline, mg. per g. .. . . 16.0 ,. . . 14.0 .. . . 15.0 .. . . 14.0 .. . . 17.0 .. . . 16.0 .. . . 15.0 .. . . 16.0 .. . . 10.0 .. , .. 10.0 .. . . 12.0 Product Dried egg yolk: Chinese .. .. Chinese . . .. Foreign origin . . Foreign origin . . Foreign origin . . U.S.A. (sample F) Foreign origin . . Sample E .. Frozen whole egg : Egg albumen: Choline, mg. per g. .. . . 22.0 .. . . 22.0 .. . * 22.0 .. . . 24.0 .. . . 20.0 .. . , 21.0 .. . . 4.5 . . . . nil NOTE-Samples A-F supplied by manufacturers of salad creams as typical of egg ingredient used. Salad cram-Various commercial products. Ten grams of sample were taken for each determination. TABLE 111 Mustard Egg yo choline/g. mg./g. % % /--Jc-, Choline in (-*- Sample Egg component Present, mg. of sample, Added, Found,* 1 Dried whole egg, 16 mg. of choline/g. -- 0.18; 0.19 1.42 1.1; 1.2 2 --I 0.51; 0.55 3.25 3.2; 3.4 -- 0.90 5-62 5.6 -- 1.20 7.64 7.5t 3 4 1.5 4-4 0-55 3.1 3.0 ** 5.3 7.8 1.07 3-6 4.1 5 6 7 Dried egg yolk, 22 mg.of choline/g. 3.0 8.0 0-65 2.2 1.9 8 3.0 8.0 0.65 2.2 1-9 9 3.0 8.0 0.70 1.87 2.1 10 3.0 8.0 0.65 1-87 1.9 21 mg. of choline/g. . . 5.3 7-8 0.88 2.3 2.2 ** 5.3 7.8 1-27 3.9 4.0 11 12 13 5.3 7.8 0.90 2.3 2.3 14 Frozen whole egg, 4.5 mg. of choline/g.** 3-0 8.0 0-6 10.5 8-0 15 ** 3-0 8.0 0-66 10.5 9.1 16 5.5 9.0 0.90 9.6 9.1 17 5-5 9.0 0.75 6.2 5.7 (approx. 1 18 Sugar-dried egg, 12 mg, of cholini/g. 1.0 5.6 0.70 6.0 5.4 NOTE-Samples 1-4 specially prepared by B.F.M. I.R.A. * Corrected for mustard content. t 7.75% found by B.F.M.I.R.A. using the same method. * * Presumed choline content of egg ingredient. DISCUSSION As described, the choline is determined by matching colours in Nessler tubes. It is possible to use a Spekker absorptiometer, in which method the graph of the optical density of the solution plotted against choline is a straight line passing through the point of origin (private communication from C.L. Hinton). The method has been applied to a number of samples of dried and liquid .egg from different sources including bulk shipments and manufacturers' supplies. The results given in Table I show that the choline content of any one type of product is virtually constant,June, 19501 EGG IN SALAD CREAM. PART I For the purpose of calculation the following values may be taken- Dried whole egg . . . . 16 mg. of choline per g. Dried egg yolk . . . . 22 3) Frozen whole egg . , .. 4.5 >> Sugar-dried egg . . a . 11 > > 309 These results are in conformity with the known relationship in chemical composition between the various products.In applying the method to the determination of egg in salad cream, products prepared on a laboratory scale as well as proprietary brands have been examined. The method can be applied directly to preparations that do not contain mustard, but in the presence of mustard it is necessary to know the nature and proportion of this ingredient in order to apply a correction. A sample of white mustard yields some 9mg. of choline per gram when hydrolysed in the manner described, but most commercial flours were found to contain a lower proportion on account of blending. Work is in hand to devise a method for determining egg and mustard separately. Table I1 sets out the results obtained and satisfactory agreement is shown between the declared egg used and the amount found.It must be borne in mind that the majority of samples were commercial products and in many of them no sample of the actual egg ingredient used was available. In determining the choline, matching was carried out to the nearest 0.5 mg. standard. cream. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. The authors wish to thank the Government Chemist for permission to publish this paper and Miss B. Askew for assistance with much of the experimental work. Their thanks are also due to the British Food Manufacturing Industries Research Association, and to Messrs. Crosse and Blackwell, Heinz, Ocean Preserving Co., Rayner and Sutton, for supplies of salad REFERENCES King, E.’J., Biochem.J., 1931, 25, 799. “Methods of Analysis of the Association of Official Agricultural Chemists,” 1946 Edition, Nottbohm, F. E., and Mayer, F., 2. Unters. Lebensm., 1933, 66, 585. “Methods of Analysis of the Association of Official Agricultural Chemists,” 1945 Edition, Brooks, J., and Hawthorne, J. R., J . SOC. Chem. Ind., 1944, 63, 310. Cahn, F. J., and Epstein, A. K., I n d . Eng. Chem., Anal. Ed., 1943, 15, 281. Grossfeld, J., and Walter, G., 2. Unters. Lebensm., 1934, 68, 270. Grossfeld, J., and Peter, J., Ibid., 1935, 69, 16. Le Breton, E., Bull. SOG. Chim. Biol., 1921., 3, 539. Gadamer, J., Ber., 1897, 30, 2322. Viehoever, A., and Nelson, W. L., J.A.O.A.C., 1938, 21, 488. Terry, R. C., and Corran, J. W., Analyst, 1939, 64, 164. Hutt, H. H., and Weatherall, H., Ibid., 1944, 69, 39. Beattie, F. J. R., Biochem. J., 1936, 30, 1554. Engel, R. W., J . Biol. Chem., 1942, 144, 701. Glick, D., Ibid., 1944, 156, 643. Rooke, H. S., Lampitt, L. H., and Jackson, E. M., Biochem. J., 1949, 45, 231. Winton, “The Analysis of Foods,” 1945 Edition, New York, p. 912. Seaman, W., Hugonet, J. J., and Leibmann, W., Anal. Chem., 1949, 21, 411. Winton, “The Analysis of Foods,” 1945 Edition, New York, p. 552. Hodson, A. Z., J . Biol. Chem., 1945, 157, 383. “U.S. Pharmacopoeia,” XI11 Edition, p. 756. -, Ibid., 1934, 23, 476. Washington, D.C., p. 348. Washington, D.C., p. 347. GOVERNMENT CHEMIST’S DEPARTMENT DUDLEY HOUSE ENDELL STREET, W.C.2 December, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500305

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

310 GLISTER AND GRAINGER MODIF1E:D RAPID TECHNIQUE FOR [Vol. 75 A Modified Rapid Technique for the Separation and Determination of Penicillin Types by Partition Chromatography on Paper BY G. A. GLISTER AND A. GRAINGER SYNOPSIS-A modified rapid-development technique for the micro-chromato- graphic assay of penicillin types is presented. Problems arising out of the quantitative interpretation are discussed, and it is indicated how, by using developed mixtures of pure penicillins as standards, the results are rendered more truly representative. THE method of Goodall and Levi,l using the principle of Consden and Martin,2 for the separation and estimation of penicillin types is well known andl requires no introduction. It is recognised as the method giving the most comprehensive picture of the constituents in a mixture of penicillins.The recently published methods of Winsten and Spark3 and KlueneI-4 describe rapid development at room temperature. In view of the advantages offered by this in routine practice, our procedure was modified and, by using a simplified developing apparatus and narrow paper strips, the basic principle of filter-paper impregnated with phosphate buffer as support for the stationary aqueous phase and diethyl ether as the mobile phase being retained, development time was cut from 24 hour!; at 4” C. to 4 hours at room temperature. Quantitative interpretation of chromatographic assays of this type has been the subject of much criticism and investigation, the issue being confused by the varying shapes of the inhibition zones that are produced and by the use of undeveloped standards prepared from only one entity.Maximum zone widths, lengths and areas have all been considered as measures of activity, and various corrective factors and formulae have been evolved, but so far none has proved satisfactory. The hydroxamic acid method of Baker, Dobson and Martin6 offers a useful alternative to the microbiological finish, but has the disadvantage that it is not so readily applicable to samples of low penicillin content, such as broth culture filtrates and the more dilute extracts therefrom. The quantitative interpretation of the microbiological chromatographic technique has therefore remained only comparatively accurate for any one technique. RAPID DEVELOPMENT- Experimental work finally led to the use of micro-strips of Whatman No.1 filter-paper (30 cm. x 1 cm.) impregnated with potassium phosphate, pH 6.2, and a decreased dose of penicillin of 5 units per strip instead of the original 30 units. Development of these strips with water-saturated diethyl ether as mobile phase for 3 to 4 hours at room temperature, and using a glass jar (14 inches x 8 inches diameter) and an evaporating basin as developing chamber and trough respectively, produced chromatograms which were reproducible and comparable with those produced by other methocis, especially by that of Goodall and Levi in which strips twice the size are used and developed at 4” C. for 24 hours. QUANTITATIVE INTERPRETATION- Goodall and Levi, using standard dilutions of penicillin G, found that the ratio of the biological assay “slopes” from the inhibition zones of developed to those from undeveloped, or stationary, spots was approximately 1.2.They therefore introduced this factor as a correction to be applied to plate slopes determined by using undeveloped standards. In our early experimental work we confirmed. the factor over the range 30 to 0-3 units of penicillin G per spot (a concentration ratio of 100/1), but found that it varied considerably outside these limits. Developed standards were then introduced as indicated below. The grade of filter-paper and the pH of the phosphate buffer used to impregnate the paper, as well as the concentration of penicillin per strip and vapour conditions within the developing chamber, are all known to be factors affecting chromatographic separation and EXPERIMENTALJune, 19501 SEPARATION AND DETERMINATION OF PENICILLIN TYPES 31 1 hence the shape of the final inhibition zones of the penicillins.Observation and measurement of numerous series of typical inhibition zones obtained from the paper chromatograms indicate that in descending order of development- $-Hydroxybenzyl penicillin (X) . . Produces a circular zone. Benzyl penicillin (G) .. . . 7 All produce elliptical zones of similar shape, h2-Pentenyl penicillin (F) . . e.g., the ratios of major to minor axis are n-Amy1 penicillin (dihydro F) . J constant. ut-Heptyl penicillin (K) . . . . Produces an elliptical zone which is elon- gated out of all proportion to the others. From consideration of the zone shapes it may be seen that if penicillin G standards (developed under identical conditions as the test penicillins) are used and the maximum zone width is taken as the criterion against log units then, as there is a slight decrease in zone width for an increase in zone length, there results an over-estimation of penicillin X and an under-estimation of penicillin K.Therefore a standard should be strictly only applicable to penicillins producing zones of identical shape. Fortunately for practical purposes, penicillin X is rarely present in commercial penicillin in quantities exceeding 0-5 per cent. (of activity) and may be neglected. One or two closely associated unknown antibiotics are often present in small amounts and, as they are almost inseparable from penicillin K, they may contribute slightly to the zone length, although in practice the effect is negligible.While working with mixtures of pure penicillins G and K, a procedure was evolved by which this under-estimation of K could be overcome. An equal number of units of penicillins G and K were mixed together in the same solution, and a series of dilutions, such as those shown in Table I, was prepared, each containing the TABLE I A DEVELOPED “STANDARD CURVE,’ , , ) Penicillin K, however, has to be considered. Standard solution, containing penicillins G and K in equal amounts, subtilis units per ml. 10,000 7000 4000 2000 1000 700 400 200 100 70 40 20 Maximum zone width G zone, K zone, mm. mm. 37 35 36 34 34 33 33 31 31 30 28 21 27 26 25 24 22 20 20 20 18 19 16 18 A r \ number of units of G and K per ml.stated. Each sample was chromatographed in replicates of six strips and plated out as usual on assay plates, using B. subtiZis as test organism. Maximum widths of the inhibition zones obtained were measured and graphs plotted against log units. The results showed that in each case the graph of maximum zone width plotted against log units was linear, but that the slope for developed K was less than that for developed G, as shown in Fig. 1. The graph clearly demonstrates how penicillin K may be under-estimated by using a G slope for zone width of K. In mixtures where the proportion qf K is very small, the reverse holds, an over-estimation being obtained below the intersection. EXPERIMENTAL CONCLUSIONS- From the experimental findings it was concluded that the inclusion of penicillin K with G as developed standard would greatly improve the accuracy of the quantitative procedure, G, F and dihydro F being calculated as usual with the G standards, and K being calculated separately with a standard of similar type.Mixtures containing appreciable amounts of penicillin X should be accompanied by standards which include X.312 GLISTER AND GRAINGER : MODIFIXD RAPID TECHNIQUE FOR MODIFIED METHOD APPARATUS AND REAGENTS- [Vol. 75 Developing chamber and trough-A wide-necked cylindrical glass jar, 14 inches high and 8 inches in diam. fitted with a gas-tight lid, is suitable. The trough consists of a 3-inch porcelain evaporating basin resting on the top of an 11 to 12-inch cylinder which stands vertically in the centre of the jar.Buflered paper strips-Sheets of Whatman No. 1 filter-paper are soaked in 20 per cent. w/v potassium phosphate buffer solution, pH 6.2, blotted between blotting paper, and air- dried. The sheets are then cut by means of a razor and a steel. straight-edge into strips 30 cm. x 1 cm. in size. The cylinder is bandaged with absorbent lint. Fig. 1. Difference of slope for G and K n! :cro-pipettes-These are made from 1-mjm. bore heavy-walled glass tu,,,ig drawn out to give a long capillary tip. A calibration mark is made on the capillary at a point from which delivery of water on to No. 1 filter-paper produces a 2 to 3-mm. diameter spot. This volume is approximately 2 p1. These pipettes give relative or comparative measure- ments required by the technique.Accuracy must be maintained in the primary preparations of the standard and test solutions. Assay plates-These are prepared from sheets of plate glass 19 inches x 14 inches. Half-inch wide strips may be cemented to form walls by means of cements such as Reuter’s &chromate gelatin mixture. Standard solutions-As the diameter - log units relationship is linear over a wide range, it is only necessary to prepare two standard solutions. (a) High standard-containing 5000 B. subt& units each of sodium penicillin G and K per ml. (b) Low standard-containing 50 B. subtilis units of sodium penicillin G and K per ml. The solutions are made in 1 per cent. phosphate buffer, pH 6.5, and usually the low standard is prepared from the high one by dilution. Nutrient agar for assay plates-The medium contains: 10 g.of Peptone, 3 g. of Lab. Lemco, 2 g. of sodium chloride, 20 g. of agar and 1 litre of distilled water. Dissolve the ingredients by boiling, and adjust the pH to 7.5. While still molten, .dispense the medium into 410-ml. lots and sterillise in an autoclave. Developing solven&Anaesthetic diethyl ether saturated with distilled water at room temperature.June, 19501 SEPARATION AND DETERMINATION OF PENICILLIN TYPES 313 PROCEDURE- For ordinary work nine are used, three for each standard and three for the test. Fold each strip 2 inches from one end and make a pencil-mark a half-inch from the fold in the centre of the strip on the long end of the fold. This serves as a point for the spot. The identification mark of the sample may be marked on the strip in pencil.By means of the same micro-pipette, spot identical volumes of the standards on to their respective strips. In the case of the test sample one should aim at spotting a similar amount of penicillin. With concentrated samples this may be achieved by preliminary dilution with 1 per cent. phosphate buffer to approximately 5000 B. subtilis units per ml. When the test sample is of low activity, e.g., as with culture filtrates, it is necessary to superimpose several spots, drying between applications, in order to keep a small concentrated spot. Solutions in organic solvents may be applied direct in the same manner. When the spots are dry, hang the strips in the developing chamber which should be previously prepared by soaking the cylinder bandage with water and placing a l-inch layer of diethyl ether in the bottom of the jar.Place the short ends of the strips in the basin, with the long ends hanging down. Then fill the basin to within half an inch of the top with the wet ether and place a small glass block in the basin to hold the strips firmly. Cover the jar with the lid, and allow development to proceed for 3Q to 4 hours. In the meantime assay plates should be poured with agar seeded with B. subtilis spores and stored in a cold chamber. When development is complete, remove the strips and lay them on the surface of the B. subtiZis-seeded agar. Allow diffusion to proceed for half anqhour and then incubate the plates at 27” C. overnight. Measure the maximum zone widths with a transparent plastic rule graduated in millimetres and calculate the averages.Take the required number of buffered strips. Thus the required units per spot may be obtained. CALCULATION (PENICILLIN x PRACTICALLY ABSENT)- Mean max. zone width high G standard - Mean max. zone width low G standard = G slope log conc. ratio (e.g., 100/1) Similarly for the K slope. The mean maximum zone widths of G, F and dihydro F of the test divided by the G slope give values the antilogs of which are proportional to the activities (in B. subtilis units) of these types. Similarly, the mean maximum width of the K zone of the test divided by the K slope gives a value whose antilog is proportional ,to its activity. The individual proportionate activities expressed as percentage of the sum of these activities. E percentage activity (B.subtilis units) for each type. The percentage activity (S. aureus units) may be obtained by using this organism in place of B. subtilis in the technique. Alternatively, a series of standards could form a “developed standard curve” and a graph be plotted. Test readings could be made on the graph direct. Typical results are shown in Tables I1 to IV. TABLE I1 ANALYSIS OF ARTIFICIALLY PREPARED MIXTURES OF PENICILLINS Results expressed as activity (B. subtilis units), per cent. Known penicillin K content, 5 10 20 50 67 % Found A f 5 Goodall and Levi method, Modified method, % % 3.3 5.1 6.8 9.8 10.8 19.6 32.0 61.4 40.2 64.8314 GRAY: CRITICAL FACTORS I N THE RESORCINOL REACTION TABLE 111 COMPARISON OF RESULTS FROM CULTURE FILTRATES Wol. 75 Goodall and Levi method Modified method L A I 1 r- \ X G F,and F, K x G F,andF, K (dih ydro) (dihydro) % % Yo Yo % Yo trace 94 3 3 trace 87 5 8 trace 97 1 2 trace 95 1 4 trace 92 4 4 trace 88 6 7 trace 80 6 14 trace 75 7 18 TABLE 1:V REPLICATE ANALYSES OF THE SAME SAMPLE G F, and F, (djhydro) K 92.4 3.1 4.5 93.8 2.2 4-0 91.0 2.7 6.3 We wish to thank the Directors of the Distillers Company (Biochemicals) Limited for their kind permission to publish this paper. REFERENCES 1. 2. 3. 4. 5. Goodall, R. R., and Levi, A. A., Analyst, 1947, 72, 277. Consden, R., Gordon, A. H., and Martin, A. J. P., Biochem. J . , 1944, 38, 224. Winsten, W. A., and Spark, A. H., Science, 1947, 106, 192. Kluener, R. G., J . Bact., 1949, 57, 101. Baker, P. B., Dobson, F., and Martin, A. J . P., Private communication. THE DISTILLERS COMPANY (BIOCHEMICALS) LTD. SPEKE, LIVERPOOL November, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500310

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

314 GRAY: CRITICAL FACTORS IN THE RESORCINOL REACTION Fol. 75 Critical Factors in the Rescrcinol Reaction for the Determination of Fructose BY D. J. S. GRAY SYNOPSIS-A method is described for the colorimetric determination of fructose in fermentation media by the use of the colour reaction between fructose, resorcinol and hydrochloric acid The effect of purity of the reagents has been studied, and it has been found that the intensity of the colour produced depends not only on the purity of the reagents, but also on the time of heating and the temperature of reaction. As the fermentation media under examination contained 0.5 per cent. of ammonium nitrate, the effect of nitrate ion was also studied, and it was found that the intensity of the colour 'was increased by the presence of nitrate ion u p to 0.25 per cent., beyond which value the intensity of the colour lessened.The ammonium ion alone had no effect. Interference by glucose or kojic and gluconic acids is negligible. The fructose contents of three samples of ferm.entation media have been checked by the method of Jackson and Matthews, and gave comparable results. THE colour reaction between fructose , resorcinol and hydrochloric acid was first described by Seliwanoff.6 The reaction is not specific for fructose but is also given by other sugars that have a keto-group. Roe5 has adapted the reaction to the colorimetric estimation of fructose in blood and urine and this method has been modified by Cole. This latter modifica- tion has been described by Bacon and Bell,l but, as yet, Cole has not published his work.June, 19501 FOR THE DETERMINATION OF FRUCTOSE 316 The method depends on the conversion of fructose to 5-hydroxymethylfurfural by strong hydrochloric acid, and the subsequent reaction of this substance with resorcinol to produce a compound which has a cherry-red colour.Glucose will give the reaction only if submitted to some preliminary treatment.' The stages of the conversion of fructose to 5-hydroxymethylfurfural have been discussed by Haworth and Jones.3 This reaction has been applied to the determination of fructose in fermentation media, where it was necessary to carry out analyses quickly and on small samples. As a result of this, certain critical conditions in the application of the method have been observed. METHOD The method outlined below is a modification of Roe's procedure.REAGENTS- Alcoholic resorcinol solution (0.1 per cent.)--0-5 g. resorcinol dissolved in 500 ml. of pure 95 per cent. alcohol. Hydrochloric acid solution (s9.g.r. l*l7)-The acid must be free from iron and give no blue colour when 1 O m l . are mixed with a solution of starch and pure potassium iodide. Standard fructose solution-A stock standard solution may be prepared by dissolving 06g. of fructose in 500ml. of distilled water saturated with benzoic acid. (Benzoic acid is added as a preservative.) Working standards may be obtained by diluting this solution. PROCEDURE- Measure 2 ml. of a solution containing 0 to 60 mg. of fructose per 100 ml. into a clean, dry test tube by means of a pipette. Add 2 ml.of alcoholic resorcinol solution and 6 ml. of hydrochloric acid. Mix the contents of the tube and suspend in a water-bath maintained at a constant temperature of 80" C. (f0.2" C.) for exactly 8 minutes. Remove from the bath and cool rapidly in running water. Allow the tube to stand for 10 minutes and then transfer the contents to an absorptiometer cell and measure the absorption immediately. In the work described, a Spekker absorptiometer was used with Chance's glass filters, OB2, and 4-cm. cells. The red solution has a distinct absorption band at approximately 486 mp.2 NOTES ON THE METHOD Eficiency of the method-One hundred ml. of a solution containing 437 mg. of fructose and 401 mg. of glucose were prepared. This solution was analysed by the method described and found to contain 438 mg.fructose. Parity of alcohol-It is important that the reagents used for the reaction are of the highest purity. Errors may be introduced if the alcohol used to prepare the resorcinol solution contains aldehydes. A blank determination should give practically the same transmittance as pure water. Purity of hydrochloric acid-The commercially available "reagent" hydrochloric acid proved also to be a source of error. The method was standardised with one batch of hydro- chloric acid and reproducible results were obtained, but when another batch of acid was used the colours obtained with similar concentrations of fructose were less intense. A sample of pure hydrochloric acid was prepared and it was noted that the addition of hypochlorite to the reaction solution caused a reduction of colour intensity.It is probable that free chlorine was present in the batch of acid that gave low results. Effect of iron-The presence of as little as 2 parts of ferric iron per million in the hydro- chloric acid caused a marked increase in the intensity of the colour produced (Fig. 1). It is interesting to note that Cole,l in his modification of the method, recommends addition of ferric iron to the solution being analysed, presumably to intensify the colour produced by the reaction. In the presence of iron an orange-red colour is obtained, whereas when iron is absent this colour ranges from pink to cherry-red. Concentration of hydrochZoric acid-The concentration of hydrochloric acid in the solution must be kept constant when carrying out a series of analyses, as any variation may affect the results.The specific gravity of the acid used should be as close to 1-17 as possible. The intensity of the colour obtained increases with increasing acid concentration and the sensitivity of the method may be increased by using acid of higher specific gravity (Fig. 2). Heating time and reaction temperature-The heating time (Fig. 3) and temperature of reaction are also variable factors and affect the intensity of the colour obtained. Roe316 GRAY: CRITICAL FACTORS I N 'THE RESORCINOL REACTION [Vol. 75 recommended that the sample analysed be heated for 8 minutes at a temperature of SO" C . and used 30 per cent. hydrochloric acid. In the method described above the only alteration in procedure has been to increase the acid concentration to about 33.5 per cent.This has the advantage of increasing the sensitivity of the method and an absorption curve is obtained which is more nearly a straight line than that obtained with the lower acid concentration. Efect of nitrate ion-The fermentation media under examination contained 0.5 per cent. of ammonium nitrate. Investigations showed that the presence of 0.01 per cent. of ammonium 0.2 0.4 06 0.8 I.0 1.2 1.4 ' FRUCTOSE, mq. Fig. 1. Effect of iron. Heating time 8min. FRUCTOSE, mq. Fig. 3. Effect of heating time FRUCTOSE, m9. Fig. 2. Effect of specific gravity of hydro- chloric acid used Fig. 4. Effect of nitrate ion nitrate in the 2 ml. portion of the solution used for analysis caused a marked increase in the intensity of the colour produced by the reaction (Fig.4). This intensification of colour is caused by the presence of nitrate and no increase is observed when ammonium alone is in the solution. When the ammonium nitrate concentration was increased- to 0.25 per cent. the colour intensity was practically the same. However, a lessening of the colour intensityJune, 19501 FOR THE DETERMINATION OF FRUCTOSE 317 was noted when the ammonium nitrate concentration was increased beyond this limit. In view of the concordance of the absorption curves for fructose in the presence of concentrations of ammonium nitrate which vary between 0.01 and 0.25 per cent. in the 2-ml. portion analysed, it is possible to estimate the fructose content of fermentation media, which contain nitrate, by the preparation of a standard curve from a series of fructose solutions to which ammonium nitrate has been added.The reaction colour obtained in the presence of nitrate is similar to the colour observed when iron is present. The fructose content has been determined by the method of Jackson and Matthews* and by the resorcinol method. The results were read from a standard absorption curve prepared from a series of fructose solutions containing ammonium nitrate. The analyses given below are of three samples of fermentation media. Fructose, yo f A \ Sample Jackson and Matthews Resorcinol method A 1.6 1.5 B 0-28 0.27 C 0.44 0.44 Interftwnce by other szLbstances-Glucose does not interfere appreciably with the reaction. A solution containing 600 mg..of glucose per 100 ml. was prepared. When submitted to the analytical procedure it gave a colour corresponding to 7.5 mg. of fructose per 100 ml. The presence of kojic and gluconic acids in the solution does not appear to affect the reaction. It will be apparent that the resorcinol reaction for fructose can be used for the deter- mination of fructose under a variety of conditions. By variation of the experimental con- ditions the method can be given a greater or a lesser sensitivity. The method is simple, accurate and rapid. The author’s thanks for permission to publish this work are due to the South African Council for Scientific and Industrial Research. REFERENCES 1. 2. 3. 4. 5. 6. 7. SOUTH AFRICAN COUNCIL FOR SCIENTIFIC AND INDUSTRIAL RESEARCH Bacon, J. S. D., and Bell, D. J., Biochem. J., 1948, 42, 397. Browne, C. A., and Zerban, F. W., “Physical and Chemical Methods of Sugar Analysis,” 3rd Ed., Haworth, W. N., and Jones, W. G. M., J . Chem. Soc., 1944, 667. Jackson, R. F., and Matthews, J . A., Bur. Stand. J . Bes., 1932, 8, 403. Roe, J. H., J . Biol. Chem., 1934, 107, 15. Seliwanoff, T., Bey., 1887, 20, 181. Wolfrom, M. L., Schultz, R. D., and Cavalieri, L. F., J . Amer. Chem. SOC., 1948, 70, 614. New York, John Wiley & Sons, 1941, p. 721. NATIONAL CHEMICAL RESEARCH LABORATORY PRETORIA, SOUTH AFRICA December, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500314

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

June, 19501 FOR THE DETERMINATION OF FRUCTOSE 317 ERRATUM: May (1950) issue, p. 251. Paper by Sutton and Markland. The date of reading should be November 26th, 1949.

ISSN:0003-2654    

DOI:10.1039/AN9507500317

出版商:RSC    

年代:1950

数据来源: RSC