摘要:

Analyst, June, 1974, Vol. 99, $9. 367-375 367 Determination of Ammonia Levels in Water and Wastewater With an Ammonia Probe BY W. H. EVANS AND B. F. PARTRIDGE (Department of Trade and Industry, Laboratory of the Government Chemist, Cornwall House, Stamford Street, London, SE 1 9NQ) The application of an ammonia probe has been investigated for discrete laboratory measurement of ammonia levels in a variety of waters. The probe displays a Nernstian response for the range 0.2 to 40 mg 1-l of ammoniacal nitrogen in a stirred 0.1 M sodium hydroxide solution containing 0.01 M ethyl- enediaminetetraacetic acid. Recoveries of added ammonia from a wide range of water samples are satisfactory. Both these recoveries of added ammonia and repeated calibrations of the probe suggest a precision of 4 per cent.for ammoniacal nitrogen concentrations greater than 0.4 mg 1-1 and 0.015 mg 1-1 for concentrations less than 0.4 mg 1-l; the statistical limit of detection is 0.03 mg 1-l. Good agreement is obtained with existing methods based on distillation and spectrophotometric measurement for a further range of samples, but the limit of detection and the precision a t low levels suggest that accurate determination in potable waters would be difficult. The probe can also be used to determine albuminoid nitrogen by taking the difference between the ammoniacal nitrogen and the total free plus albuminoid nitrogen obtained by distillation. Values obtained in this way agree with those obtained by existing methods subject to the precision of the probe being acceptable. FREE and saline ammonia, expressed as ammoniacal nitrogen, is normally determined in the course of examination of water and wastewater; less frequently, albuminoid nitrogen is also determined.Ammoniacal nitrogen may vary from a negligible level in potable water (less than 0.01 mg 1-l) to values exceeding 40 mg 1-1 in settled sewage. The recommended methods for its determination in potable and wastewaters involve distillation of the ammonia from an alkaline medium, magnesium oxidel or sodium carbonate,2 or from a phosphate b ~ f f e r , ~ followed by spectrophotometric or titrimetric measurement of the distilled ammonia. The albuminoid nitrogen is determined subsequently on the same sample by distillation from alkaline permanganate, with the same means of measurement. A commercial probe introduced recently measures ammonia directly and has been used to determine free ammonia levels in boiler feed-waters.* An investigation into the possible laboratory application of this probe to the determination of ammoniacal and albuminoid nitrogen in swimming-pool and raw potable waters and wastewaters is described in this paper.EXPERIMENTAL The ammonia probe used in this work was an Electronic Instruments Ltd. (EIL) Labora- tory Model 8002-2, consisting of a glass pH electrode surrounded by a filling solution of ammonium chloride, in contact with a gas-penneable hydrophobic membrane. Free ammonia diffuses through this membrane until the partial pressures in the sample on the one side and the thin film of standard ammonium chloride filling solution on the other side are equal.When this steady state is reached, the ammonia - ammonium ion equilibrium in the thin film gives a value for the hydrogen-ion concentration, which is monitored by changes in potential of the glass pH electrode, referred to a silver - silver chloride electrode also dipping into the filling solution. ALKALINE pH ADJUSTMENT- Ammonium ions are converted into free ammonia exclusively at pH values exceeding 12. A dilution of 10 parts of a standard solution (see Reagents) with 1 part of 1 M sodium hydroxide solution or 1 part of 0.1 M sodium hydroxide solution produced an average difference in read- ings of 1.6 mV (Table I). This difference could not be explained by the presence of any free 0 SAC; Crown Copyright Reserved.368 EVANS AND PARTRIDGE : DETERMINATION OF AMMONIA LEVELS [An@bSt, VOl.99 ammonia in the sodium hydroxide solution as potentials obtained with standards adjusted with boiled and unboiled 1 M alkali solution were the same. Hence, the difference may be due to the buffering effect of de-ionised water giving a final pH of less than 12. Ethylenediamine- tetraacetic acid was added to the pH adjuster to prevent precipitation of hydroxides (mainly magnesium), thereby avoiding deposition on glass surfaces or on the electrode membrane ; this addition produced no significant difference in millivolt readings (Table I). TABLE I Results are meter readings in millivolts EFFECT OF DIFFERENT ALKALI CONDITIONS ON PROBE RESPONSE Ammoniacal nitrogenlmg 1-1 I Alkali conditions* 2 1 0-4 0.2 0.1 0.1 M Sodium hydroxide .. .. . . +23*6 +40.0 $63.2 +79*2 +92-2 1.0 M Sodium hydroxide . . . . . . +21*8 +39*2 +61*8 +77*2 +90*2 1.0 M Sodium hydroxide (boiled) . . . . +22.0 +39*2 $61.8 +77*2 +90*2 1.0 M Sodium hydroxide plus 0.1 M EDTA . . +22.0 + 39.0 + 62.4 + 77.2 + 90.2 1.0 M Sodium hydroxide plus 0.1 M EDTA (boiled) ,. .. .. .. . . +22.0 +39.4 +62*0 +77.2 +90.0 * 5 ml of alkaline solution used with 50 ml of standard solution. TABLE I1 EFFECT OF DIFFERENT TOTAL IONIC CONCENTRATIONS ON DETERMINATION AT Results are meter readings in millivolts DIFFERENT ALKALI STRENGTHS Concentration of added sodium chloride/M A l k a l i addition of 1 M NaOH + 0.1 M EDTA- 0 0.01 0.02 0.05 A l k a l i addition of 0.1 M NaOH + 0.01 M E D T A - 0 0.01 0.02 0.05 7 40 -57.8 - 57.8 - 58.0 - 57.8 - 55.8 - 55.2 - 54.8 - 53.8 Ammoniacal nitrogenlmg 1-1 A 7 10 4 1 0.4 0.1 -21.8 +2-0 +36.6 +59.8 +87.8 -21.8 3-2.0 +36*8 +60*0 +87*2 -21.8 +2*0 +36*6 f59.4 +87.8 -21.8 +2*0 f-36.6 +60-0 +87*8 -20.4 +3.4 +38.0 +61.8 +91*0 -20.2 +3*4 +38,0 +61.8 $90.6 -20.2 +3.6 +38.6 +61*2 +90*4 -19.8 +4.0 +38*8 +61.2 +91.4 Because the membrane is hydrophobic, ions cannot interfere directly; however, a high ionic concentration can affect the partial pressure of ammonia.It is therefore desirable to maintain the total ionic strengths within limits. Potable and surface waters seldom have total dissolved solids exceeding 1000 mg 1-1 (about 0.02 M, calculated as sodium chloride). The response to variation in the total ionic strength at the two alkali concentrations was tested by addition of sodium chloride to standard solutions containing the higher strength alkali pH adjuster, at nominal total ionic strengths of 0.13, 0.14, 0.15 and 0.18, and similarly for the lower strength alkali pH adjuster, to give nominal total ionic strengths of 0.01, 0.02, 0.03 and 0.06.The results shown in Table I1 indicate that the probe is more susceptible to changes in the total ionic strength at the lower alkali concentrations, particularly at the higher ammonia concen- trations. For this reason, and because of the possibility of the buffering effect of natural waters giving final pH values less than 12, it is recommended that the higher strength alkali pH adjuster should be used. RESPONSE TIME- For the experimental purposes described herein, equilibrium response was taken to satisfy the criterion of a shift of less than 0.2 mV min-l at constant temperature. The time taken to achieve equilibrium response for stepwise change in ammoniacal nitrogen concentra- tion, as illustrated in Tables I and 11, was 4 minutes above 1 mg 1-1 and 8 minutes in the range 0.1 to 1.0 mg 1-l.Below 0.1 mg 1-1 readings should be taken after 10 minutes; longer equilibrium times would result in loss of ammonia from the alkaline solutions.June, 19741 IN WATER AND WASTEWATER WITH AN AMMONIA PROBE 369 I t has been reported that, providing measurement is made after immersion in distilled water, the probe attains an equilibrium potential within 2 to 5 minutes.* For samples from different origins varying widely in ammoniacal nitrogen content, the time required for the necessary equilibration in distilled water between each sample would detract from the value of using the probe.The millivolt readings obtained for changes in the range 0.1 to 100 mg 1-l of ammoniacal nitrogen for 1000, 100 and 10-fold increases and decreases in concentration are shown in Table 111; readings were measured after the times specified for duplicate standard solutions at each particular level. These results indicate that hysteresis occurs when measure- ments are made on successive solutions decreasing more than 10-fold in concentration because of the diffusion of ammonia from the electrode internal filling solution, after immersion in a high concentration solution, into the low concentration sample solution.It is therefore advisable, when consecutive readings from a higher to a lower concentration differ by more than a decade, i.e., by an absolute potential greater than 60 mV, that a repeat measurement be made. TABLE I11 STABILITY OF RESPONSE TO 1000-, 100- AND 10-FOLD INCREASE OR DECREASE Results are meter readings in millivolts IN AMMONIACAL NITROGEN LEVELS Change in concentration of ammoniacal nitrogenlmgl-1 1bo to 0-1 10 to 0.1 1.0 to 0.1 100 to 10 10 to i Descending order . . + 66.0, + 77.2 + 76.0, + 80.6 + 80.8, + 80.6 - 29.4, - 29.2 + 28.0, + 28.2 Ascending order . . - 88-0, - 88.6 - 29.8, -29.4 + 28.8, + 28-8 CALIBRATION- A recent method for the determination of ammonia with this probe has involved the use of a calibration graph prepared by plotting the differences in the potential of standards from that of a reference standard concentration against the logarithm of the concentration mea~ured.~ Samples are similarly measured against the reference standard concentration and ammonia levels read from the calibration graph.With frequent calibration of the reference concentration any drift in potential is avoided. The application of this probe to the measu- rement of ammonia in the wider range of samples envisaged would require, however, consistent differences from a reference standard in the range 0.01 to 40 mg 1-1 of ammoniacal nitrogen. The level of ammonia in de-ionised water is variable and may be as high as 0.01 mg 1-l. Treatment of de-ionised water with a strong cationic exchanger produced a water containing a consistent level of 0.002 mg 1-1 of ammoniacal nitrogen, determined by distillation and spectrophotometric measurement.This treated water was used to prepare all standards at concentrations less than 1.0 mg 1-1. The potential differences of standards in the range 0.01 to 40 mg 1-1 of ammoniacal nitrogen with reference to a 1 mg 1-1 standard showed no change during a probe life of 30 days, although the absolute potentials drifted in a positive direction. The criterion of a probe life of 30 days was accepted for all values obtained, the probe membrane being readily replaceable. The calibration graph was essentially linear in the range 0.2 to 40 mg 1-1 of ammoniacal nitrogen, with a Nernstian response of 58.5 mV per decade difference, which was also obeyed up to a concentration of 100 mg 1-1.Average millivolt differences from the standard concen- tration at each level, together with the decade differences and the standard deviations from the means expressed also in terms of ammoniacal nitrogen, are shown in Table IV. This table includes the readings from 16 calibration runs taken during a membrane life of 30 days and readings from 40 calibration runs taken during a period of 6 months, during which the probe membrane was changed five times. The average calibration readings for the 1- and 6-month periods are in agreement, indicating satisfactory stability for the probe. Thus, a calibration graph, once prepared, could be used repeatedly with only an occasional check when the probe membrane is changed or to ascertain the condition of the probe.It is noticeable that at ammoniacal nitrogen concen- trations of less than 1.0 mg 1-1, averages differ depending on whether the calibration is con- ducted in an ascending or descending order of concentration. This is because of the hysteresisw 4 0 TABLE IV STATISTICAL SURVEY OF CALIBRATION OF AMMONIA PROBE Results are meter readings in millivolts Number Ammoniacal nitrogenlmg 1-1 M A .. of I \ I Average of values during 6 months Average of values during 1 month Average of ascending values during 1 month .. .. . . Average of descending values during 1 month . . .. .. Decade difference during 6 months Decade difference during 1 month Standard deviation from mean Standard deviation from mean Standard deviation from mean during 6 months, as ammoniacal during 6 months .. . . during 1 month . . . . nitrogen/mg 1-' .. .. .eadings 40 16 8 8 40 16 40 16 40 40 - 94.0 - 94.2 - 94.1 - 94.3 - - & 0.90 & 0.90 1.6 20 - 76.3 - 76.4 - 76.4 - 76.4 - - & 0.75 & 0.75 0.6 10 - 58.4 - 58.6 - 58.6 - 58.5 - - & 0*80 f0.65 0.4 4 - 35.1 - 35.0 - 35.2 - 34.8 58.9 59.2 0.75 k0.65 0.15 2 - 17.3 - 17.5 -17.6 - 17.3 59.0 58.9 & 0.60 & 0.30 0.06; 1 0 0 0 0 58.4 58.6 - - - 0-4 + 23.5 + 23.9 f24.3 +23-5 58-6 58.9 f 0.76 & 0-70 0.01 0.2 + 38.9 + 39.3 +39*8 + 38.7 56.2 56.8 f l - 5 f 1.2 0.015 0.102 + 53.4 + 53-8 + 54.7 + 52.8 53.4 53.8 f 1.9 & 1.9 0.010 0.042 + 66.0 + 67-3 + 70.5 + 64.3 42.5 43.4 j, 4.2 f 4.0 0.016 0.022 + 75.1 + 75.8 + 78-3 + 73.4 36.2 36.5 Ifi 3.3 f 4-2 0.010 0.012 + 78.8 + 78.5 + 81.4 + 75.6 25.4 24.7 f 4-9 f 5.0 0.012June, 19741 I N WATER AND WASTEWATER WITH AN AMMONIA PROBE 371 effect mentioned previously and is very marked at concentrations less than 0-1 mg 1-1.It appears that measurement at these low levels should be conducted after stabilisation in ammonia-free water, but the precision of ascending order of standards around the respective means at these levels is similar to the over-all precisions indicated in Table IV. To break the standard procedure is therefore unlikely to confer advantages for these low levels. The precision at each level of standards suggests that concentrations of ammoniacal nitrogen greater than 0-4 mg 1-1 should be measurable to an accuracy within 4 per cent. and concentrations less than 0-4 mg 1-1 to within 0.015 mg 1-l.A statistical limit of sensitivity of 0.03 mg 1-l was calculated from these readings. The temperature during these measurements varied between 22 and 26 "C. The tempera- ture effect on the calibration line will be governed by the Nernstian equation and will give millivolt decade differences of 58.2 at 20 "C, 59.2 at 25 "C and 60.2 at 30 "C. The effect of temperature variation on the slope of an average calibration graph will therefore be marginal. It has been reported, however, that absolute measured potentials change4 by 1.5 mV "C-l so that temperature changes occurring during measurement of standards or samples and the fixed concentration should be avoided and in no circumstances should exceed 1 "C. INTERFERENCES- The major ionic constituents of water, Na+, K+, Ca2+ and Mg2+, added as chlorides, and CO,z-, HC0,-, NO3- and SO,%, added as sodium salts,each at a 0.1 M concentration, had no significant effect on 10, 1.0 and 0.1 mg 1-1 of ammoniacal nitrogen.Hydrazine, often added to water central heating systems, similarly had no effect when present at 1 mg 1-1 levels. Free chlorine necessarily interferes, owing to the presence of chloramines. Dechlorination was effected by standing the sample for 10 minutes with sodium sulphite (0.5 ml of a solution containing 1.8 g 1-l) or sodium thiosulphate (0.5 ml of a solution containing 7 g 1-l) prior to the addition of pH adjuster. Total dechlorination with sodium arsenite (0.5 ml of a solution containing 2 g 1-l) was unsuccessful.DISTILLATION CONDITIONS- Evaluation of the probe required comparison of the results obtained by using this method with those obtained by existing methods of ammonia determination. It has been reported that recovery of amounts exceeding 500pg of ammoniacal nitrogen by distillation, for different pH conditions, is in~omplete.~ The recovery of ammonia, in 200ml of distillate, from 0-5 g of sodium carbonate2 and from alkaline permanganate solution1 is shown in Table V. These values confirm that satisfactory recovery of ammonia is possible for levels not greater than 400 pg of ammoniacal nitrogen without distillation into dilute acidand, for the purpose of comparative values, aliquots of samples were distilled to ensure that the ammonia did not exceed this level. Thus, 500 ml of samples were distilled for levels less than 1.0 mg 1-1 of ammoniacal nitrogen, 100 ml for levels in the range 1 to 4 mg l-l, 25 ml for levels in the range 4 to 20 mg 1-1 and 10-ml aliquots for levels exceeding 20 mg 1-l.These aliquots were diluted to 500 ml with de-ionised water and the blanks for each alkaline variant were measured daily when required. TABLE V RECOVERY OF AMMONIA BY DISTILLATION Ammoniacal nitrogen added: A, 100; B, 200; C , 400; D, 1000; E, 4000 pg Ammoniacal nitrogen recovered, per cent. Distillation conditions A B C D E From sodium carbonate , . . . 100.8 98.3 100.0 99.9 97.4 From alkaline permanganate . . 100.0 100.5 100.6 96.6 95.0 It has been noted that interference may occur in spectrophotometric measurements using Nessler's reaction on ammonia distilled from water and wa~tewaters~~~ and accordingly, all comparative values included, in addition to determination by Nessler's reaction, determina- tion by the phenol - hypochlorite reaction (indophenol procedure).Of the many variations of the latter, an adaptationa was used that was based on the method of Weatherburn.'372 APPARATUS- EVANS AND PARTRIDGE : DETERMINATION OF AMMONIA LEVELS [Analyst, Vol. 99 METHOD All glass apparatus should be stored containing de-ionised water. Ammonia probe-This is an Electronic Instruments Ltd. Laboratory Model 8002-2. The probe is stored, when not in use and overnight, in 0.1 M ammonium chloride solution as directed. Before use, the probe should be kept for at least 30 minutes in a solution of 10 parts of 0.1 mg 1-1 ammoniacal nitrogen and 1 part of alkaline EDTA pH adjuster, this solution being changed if there is a continuous drift in absolute potential.(For the experi- mental work described, when measurement was undertaken below 0-1 mg 1-1 the probe was similarly kept in de-ionised water before use.) pH or millivolt meter-Potentials are measured on a pH or millivolt meter capable of reading to 0.2 mV. Magnetic stirrer-Solutions are stirred magnetically by using a polypropylene-covered bar magnet. REAGENTS- These should be of analytical-reagent grade. Ammonia-free water-Water treated by passing it through a mixed resin bed is further purified by adding 5 g of a strong cation exchanger in the hydrogen form to 5 litres of this water and standing the mixture for 2 days3 The resulting water should not contain more than 2 pg 1-1 of ammoniacal nitrogen as determined by distillation and spectrophotometric measurement. pH adjztster-Dissolve 20 g of sodium hydroxide and 18.6 g of ethylenediaminetetraacetic acid dihydrate, disodium salt, in ammonia-free water and dilute the solution to 500 ml.Sodium thiosulphate solutions, 0.7 per cent.-Prepare daily a solution containing 0.7 g of sodium thiosulphate pentahydrate dissolved in 100 ml of de-ionised water. Standard ammonia solzttions-Dissolve 3.821 g of ammonium chloride (dried at 100 "C) in de-ionised water in a 1-litre calibrated flask and dilute to 1 litre to give a solution containing 1000 mg 1-1 of ammoniacal nitrogen. Dilute suitable aliquots with de-ionised water to give solutions containing 100, 40, 20, 10, 4 and 2 mg 1-l.Prepare daily, as required, solutions with concentrations of 1.0, 0.4, 0.2, 0.1, 0.04, 0.02 and 0.01 mg 1-1 of ammoniacal nitrogen by dilution of suitable aliquots of the above standards with ammonia-free water. PROCEDURE- Transfer 50 * 1 ml of a standard solution containing 1 mg 1-1 of ammoniacal nitrogen into a Pyrex glass beaker and add 5 ml of pH adjuster immediately before measurement. Remove the existing sample, standard or the conditioning solution and dry the probe with a tissue. Immerse the probe in the standard solution, place the beaker on a magnetic stirrer and stir by using a small bar magnet. After 4 minutes measure the potential. Transfer 50 & 1 ml of sample or distilled sample into a glass beaker, add 5 ml of pH adjuster, and similarly measure the potential after 4 minutes, or 8 minutes for levels equal to or less than 1.0 mg 1-1 of ammoniacal nitrogen. Record the difference in potential between the sample and the reference standard concentration (1 mg 1-l) and read off the ammoniacal nitrogen concentrations from the calibration graph.For a series of samples, the reference standards can be repeated as necessary to correct for any drift in absolute potential. If consecutive electrode readings involve a potential difference of greater than +SO mV, a second aliquot of the lower concentration sample must be taken and the reading repeated. If chlorine is present in a sample, add 0-5 ml of a freshly prepared sodium thiosulphate solution to 50 ml of the sample and stand the mixture for 10 minutes before addition of the pH adjuster.The calibration graph for an individual probe should be prepared by using 50ml of standard solutions in the range 0.01 to 100 mg 1-1 of ammoniacal nitrogen, the potentials in both ascending and descending order of concentration being read. A graph of the logarithm of concentration versus the average difference in potential from that of the reference standard concentration (1 mg 1-1) is linear for the range 0.2 to 100 mg l-l, and curves increasingly below the lower limit (Table IV). Allowance should be made for the ammonia level of the ammonia-free water at the lowest concentration levels in plotting this graph. When the calibration graph has been adequately defined for an individual probe, standard solutions need only be measured to ascertain the condition of the probe, or when the probe membrane has been changed.June, 19741 IN WATER AND WASTEWATER WITH AN AMMONIA PROBE TABLE VI 373 RECOVERY OF AMMONIA ADDED TO WATER AND WASTEWATERS Ammonia added as nitrogen: A, 10; B, 1; C , 0.1 mg 1-1 Ammonia recovered as nitrogenlmg 1-1 Sample level I A I as nitrogen/ A B C Sample mg 1-1 Spring water .. .. .. . . <0.01 10.5 0.99 0.10 Well water . . . . .. . . (0.01 10.0 0.99 0.10 Potable water, ex surface water . . 0.01 9-6 1.02 0.12 Borehole water . . .. .. . . <0.01 10.3 1.01 0.10 Very hard well water . . .. . . (0.01 10.1 1.01 0.11 Swimming-pool water . . .. .. 0.46 10.0 0.97 0.10 0.04 9.9 1.01 0.11 Central heating system water . . 0.08 9.4 0-99 0.13 Lake water .. .. .. .. 0.01 10.2 0.99 0.10 0.61 9.8 1.00 0.11 River water . . . . .. . . 0.37 10.0 0.99 0.11 Aged domestic sewage effluent . . 0.05 10.0 1-02 0.11 0.07 9.9 1.05 0.11 <0.01 9.9 1-01 0.07 Trade waste . . . . .. .. 0.72 9.7 1.01 0.07 1.84 9.2 0-9 - 0.62 9.6 0.96 0.11 Coefficient of variation, per cent. . . 3.2 3.3 16.0 Average recovery of nitrogen/mg 1-l . . 9.9 1.00 0.10 RESULTS The recovery of ammonia was tested by addition of standard solutions to a number of samples from a variety of origins followed by immediate measurement with the probe. Two millilitres of solutions containing 500, 50 and 5 mg 1-1 of arnmoniacal nitrogen were diluted to 100 ml with sample to give added levels of 10, 1.0 and 0.1 mg 1-1 of ammoniacal nitrogen.Allowance was made, when necessary, for the natural ammonia level to account for dilution of the water sample with standard. Swimming-pool water was pre-treated with sodium thiosulphate solution before measurement. The results are shown in Table VI. The average TABLE VII COMPARISON OF PROBE VALUES WITH VALUES FROM EXISTING METHODS FOR DETERMINATION OF AMMONIA Nitrogenlmg 1-1 r A > Sample Probe Nessler’s reaction Indophenol method A mmoniacal nityogen- Potable water . . .. .. . . 0.05 0-07 0.05 Potable water . . .. .. .. 0.24 0.22 0.22 Swimming-pool water . . . . 0.35 0.36 0-34 Ditch water . . .. .. .. 0.47 - 0.45 Raw borehole water . . . . .. 0.72 0.70 0.73 Swimming-pool water . . . . 0.17 0.20 0.18 Domestic sewage effluent . . .. 1.3 1.3 1.2 Airfield drainage run-off .. .. Domestic sewage effluent . . .. 5.2 5.7 5.3 2.8 2.9 - Slaughterhouse waste .. .. 13 14 15 Partially treated sewage . . . . 25 27 27 Potable water . . . . .. .. 0.05 0.05 0.05 0.09 0.09 0.09 Swimming-pool water . . . . 0.26 0.28 0.27 Potable water . . .. . . .. 0.45 0.43 0.47 Domestic sewage effluent . . .. 0.93 0.85 0.79 9.0 9.3 9.1 Partially treated sewage . . .. 9.5 9.5 9.6 Trade waste . . . . .. .. 12 13 12 Partially treated sewage . . . . 21 22 21 Crude sewage . . .. . . . . 44 43 44 Total ammoniacal and albuminoid nityogen-2 TABLE VIII T3 E u COMPARISON OF AMMONIACAL AND ALBUMINOID NITROGEN LEVELS IN WATER AND WASTEWATERS DETERMINED BY DIFFERENT METHODS 0 Sample Potable water, ex surface water . . Potable water .. .. .. Swimming-pool water .. .. Domestic sewage effluent . . .. Partially treated sewage . . .. Crude sewage .. .. .. c Probe (direct) 0.03 0.04 0.06 0.21 0.21 0-53 1.7 7.2 8.4 13 25 43 Ammoniacal nitrogen/mg 1-l h > Nessler’s reaction Indophenol reaction and distillation and distillation 0.03 0.03 0.06 0.05 0-06 0.06 0.22 0.2 1 0.2 1 0-19 0.53 0-63 2-0 1.7 7.1 6.6 8.0 8.9 12 12 26 26 41 42 7 Probe (by difference) 0.19 0.03 0.03 0.05 0.29 0.9 0.7 1.4 0.9 0.5 2-6 0.6 Albuminoid nitrogenlmg 1-’ Nessler’s reaction Indophenol reaction and distillation and distillation 0.19 0.18 0.01 0.0 1 0.01 0.01 0.08 0.07 0.23 0.22 0.8 0.9 1.0 1.0 1-8 1.6 0.9 1.0 0-7 0-7 2-6 2.6 2.1 2.2 A %June, 19741 IN WATER AND WASTEWATER WITH AN AMMONIA PROBE 376 recovery at each level was satisfactory, the coefficient of variation being 3.3 per cent.at the 1 and 10 mg 1-1 levels of added arnmoniacal nitrogen and 16 per cent. at the 0.1 mg 1-1 concentration. These values are in good agreement with the precision for electrode calibration. Free ammonia levels obtained by direct measurement with the ammonia probe and values obtained by distillation from 0.5 g of sodium carbonate, with subsequent spectro- photometric measurement by use of Nessler’s reaction or the indophenol method, are illus- trated for a variety of samples in Table VII. Also included is a comparison of total am- moniacal and alburninoid nitrogen obtained by distillation from alkaline permanganate, and determination both with the probe and by spectrophotometry. Because of the transient nature of ammoniacal nitrogen, all determinations were made at the same time.The satis- factory agreement obtained in these trials suggested a more stringent comparative exercise. This involved direct measurement of ammoniacal nitrogen with the probe and measurement of the total ammoniacal and albuminoid nitrogen with the probe after distillation, thus obtaining the albuminoid nitrogen level by difference. The values were compared with values obtained by consecutive distillation of the amrnoniacal and albuminoid nitrogen and deter- mination by spectrophotometric measurement; the results are shown in Table VIII. The agreement, however, may be affected by the precision of the electrode, which can introduce disproportionate errors into the albuminoid nitrogen concentrations when the ratio of free to albuminoid nitrogen is large.CONCLUSIONS The ammonia probe can be used in laboratory conditions to measure the discrete free and saline ammoniacal nitrogen present in a wide range of water samples. These include swimming-pool waters, surface waters, effluents, sewages and wastewaters. Determination should be possible within a precision of 4 per cent. for ammoniacal nitrogen levels greater than 0.4 mg l-l, and within 0.015 mg l-l for levels less than 0.4 mg 1-l. Samples can be measured in random order providing consecutive readings involve absolute potential differences of not more than +60 mV. The interference most likely to be encountered is chlorine as chloraniines, but dechlorination is readily achieved with thiosulphate. The calculated lower limit of detection of 0.03 mg 1-1 of ammoniacal nitrogen and the precision at low levels (less than 0.1 mg 1-l), which originates partly from the non-Nernstian response at these levels, suggest that its accurate determination in potable waters would be difficult. The probe can also be used to determine albuminoid nitrogen, subject to the precision of the probe being acceptable, by taking the difference between the ammoniacal nitrogen and the total free arnmoniacal plus albuminoid nitrogen obtained after distillation from alkaline permanganate. We thank Mrs. Alison Hamilton-Sharp for some technical assistance. This paper is published by permission of the Government Chemist. REFERENCES 1. 2. 3. 4. 5. 6, 7. The Department of the Environment, “Analysis of Raw, Potable and Waste Waters,” H.M. Ministry of Housing and Local Government, “Methods of Chemical Analysis as applied to Sewage “Standard Methods for Examination of Water and Wastewater,” Thirteenth Edition, American Midgley, D., and Torrance, K., Analyst, 1972, 97, 626. Sutcliffe, R. A., and Jones, G. A., Wat. Pollut. Contvol, Lond., 1968, 67, 209. Evans, W. H., David, E. J., and Patterson, S. J., Wal. Res., 1973, 7, 975. Weatherburn, M. W., AnaZyt. Chem., 1967, 39, 971. Stationery Office, London, 1972. and Sewage Effluents,” Second Edition, H.M. Stationery Office, London, 1956. Public Health Association, New York, 1970. Received November 9th, 1973 Accepted December 28th, 1973

ISSN:0003-2654    

DOI:10.1039/AN9749900367

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

376 Analyst, June, 1974, Vol. 99, p p . 376-380 The Coulometric Determination of Trace Levels of Sulphur in Gallium Phosphide BY K. GIJSBERS, L. BASTINGS AND R. VAN DE LEEST (Philifis Research Laboratories, Eindhoven, The Netherlands) Sulphur present in gallium phosphide has been determined. A t 950 "C gallium phosphide reacts with platinum and hydrogen to form Ga,Pty and platinum phosphide (PtP,) , and sulphur is catalytically converted into hydrogen sulphide. The latter is absorbed in an alkaline medium and deter- mined by controlled-potential coulometry a t a silver-gauze electrode. The determination of 0.05 pg of sulphur is shown to be possible, the blank value by this method amounting to 0.01 pg of sulphur. SULPHUR present in gallium phosphide has in the past been determined by Luke's method.' Because of the high blank value of the reagents, which amounts to 1 or 2 pg of sulphur,2 this method is not very suitable for determining small amounts of sulphur.Starting from the sensitive and reproducible sulphide determination by controlled- potential coulometry with a silver electrode (2 Ag + S2- -+ Ag,S + 2e),3 we investigated the possibility of converting the sulphur present in gallium phosphide into hydrogen sulphide. I t is known from the literature4r5 that the sulphur in both organic and inorganic sulphur- containing compounds can be converted into hydrogen sulphide by reduction with hydrogen on a platinum catalyst at elevated temperatures. EXPERIMENTAL REAGENTS- Quartz-distilled water was used throughout this work. Sodium hydroxide solution, 0.1 M-This solution is prepared from analytical-reagent grade Hydrogew-Hydrogen is purified by passing it through a liquid nitrogen cold trap so as Nitrogen-Nitrogen is purified by passing it through a copper furnace and a column sodium hydroxide.to remove any water present. filled with soda-lime in order to remove organic compounds and oxygen. CALIBRATION SOLUTIONS- The following solutions of sulphur-containing compounds were prepared. SuZ$hosaZicyZic acid solution-This solution is prepared by dissolving 80.0 mg of sulpho- Potassium suulphate solution-This solution is prepared by dissolving 54.5 mg of potassium salicylic acid in 100 ml of water. sulphate in 250 ml of water. CELL- The cell used for the determination of hydrogen sulphide is shown in Fig. 1.The working electrode is a silver-gauze electrode with an apparent surface of 30 cm2, which is kept at a constant potential of -360 mV versus a saturated calomel electrode, and the auxiliary elec- trode is a platinum foil immersed in a 1 M potassium nitrate solution. Contact with the 0.1 M sodium hydroxide solution is made via a conducting agar bridge. The potential of the working electrode is controlled by a Wenking potentiostat. The current resulting from the reaction of sulphide at the silver electrode is passed through a standard resistor (1000 Q), the voltage drop across the resistor being converted into a frequency (Anadex DF-loo), which is measured with an electronic counter (Philips PM6620). The signal is corrected for the residual current due to the reaction of contaminants, and the current integration system electrically calibrated with a constant-current source.@ SAC and the authors.GIJSBERS, BASTINGS AND VAN DE LEEST 377 PLATINUM CATALYST- Quartz-wool or hexagonal silicon carbide crystals (size 3 mm) are wetted with a chloro- platinate solution containing 2.5 g of platinum per 100 ml, dried and subsequently heated in order to decompose the chloroplatinate and to form finely divided platinum metal at the surface of the substrate. The catalyst is then heated at 1000 "C in an atmosphere of hydrogen for 1 day so as to remove contaminants (e.g., halides). Silicon carbide crystals are to be preferred as supporting material because they can be packed in a more reproducible way than quartz-wool.I c ; I I Mag L t i c stirring bar Fig. 1. Experimental set-up for the determination of sulphur in gallium phosphide. Electrodes : WE, working; AE, auxiliary; and RE, reference PROCEDURE- The gallium phosphide sample is wrapped in a platinum foil and the latter placed in a quartz crucible, which is then inserted into the quartz tube (see Fig. 1). At the beginning of the procedure the sample is not heated. Purified hydrogen is led through the tube while the catalyst is heated at 900 "C; this hydrogen is not led into the electrolysis cell. Meanwhile, purified nitrogen is led into the cell in order to remove oxygen. The presence of oxygen in the cell causes a cathodic current to flow (due to reduction of the oxygen). The flow of nitrogen is continued until the cathodic current has diminished to a value below 0.05 pA, which indicates that oxygen has been removed.The nitrogen flow is then stopped and the hydrogen, which is first passed over the heated platinum catalyst, is led into the electrolysis cell. (Care must be taken to prevent the solution from entering the inlet tube.) When the anodic current has fallen to a value below 0.05 p A (indicating the removal of contaminants), the sample in the tube is heated at 950 "C with the second resistance furnace. Gallium phosphide reacts with platinum and hydrogen while the sulphur is released partly as hydrogen sulphide. However, in order to achieve the complete conversion of the sulphur into hydrogen sulphide it is necessary to use a platinum catalyst.378 GIJSBERS et al.: THE COULOMETRIC DETERMINATION OF [Analyst, Vol.99 WORKING CONDITIONS FOR THE CATALYST- The conversion of the sulphur present in sulphosalicylic acid into hydrogen sulphide was carried out in order to check the method and to determine the optimum working conditions; 20 p1 of the sulphosalicylic acid solution (corresponding to 2.0 pg of sulphur) were evaporated to dryness in a platinum crucible, which was brought into the quartz tube and the acid vaporised at 950 "C. With the hydrogen flowing at the rate of 10 ml min-1, the vapour was led over the platinum catalyst. The experiment was repeated several times with the platinum catalyst at different temperatures. The results obtained are shown in Fig. 2, in which the integrated current due to the reaction of sulphide at the silver electrode is shown as a function of the heating time at different temperatures of the catalyst.Different hydrogen flow-rates were also tried, the temperature of the catalyst being maintained at 900 "C. From these experiments, the following optimum working conditions were determined : a hydrogen flow- rate of 10 ml min-1 and a catalyst temperature of 900 "C. 12 Heating t irne/minutes Fig. 2. Conversion of sulphur in sulphosalicylic acid into hydrogen sulphide on a platinum catalyst as a function of time and a t different temperatures of the catalyst. Hydrogen flow-rate of 10 ml min-l. Temperature/"C: a, 600; 0, 600; A, 700; 0, 800; x, 900; and +, 1000 Sulphur in inorganic compounds, c.g., potassium sulphate, can be determined in the same way, the optimum working conditions being identical with those for determining sulphur in organic compounds.The nature of the substrate of the catalyst has no influence on these conditions. RESULTS AND DISCUSSION CONVERSION YIELD - The yield from the conversion of sulphur into hydrogen sulphide is calculated by compar- ing the integrated current, due to the reaction of sulphide at the silver electrode, with a calibration graph; the latter is prepared by plotting the integrated current, from a coulometer or from a constant-current source, versus the amount of sulphide, which is calculated according to Faraday's law with n = 2. Amounts of sulphosalicylic acid and potassium sulphate corres- ponding to 2.0 pg of sulphur were analysed for sulphur according to the procedure described above.The results obtained are presented in Table I.June, 19741 TRACE LEVELS OF SULPHUR IN GALLIUM PHOSPHIDE 379 TABLE I DETERMINATION OF SULPHUR IN SULPHUR-CONTAINING COMPOUNDS (EQUIVALENT TO 2.0 pg OF SULPHUR) Experimental conditions: hydrogen flow-rate 10 ml min-l ; sample heating temperature, 950 "C; catalyst heating temperature, 900 "C; and time of analysis, 1 hour Sulphosalicylic acid found, per cent. Potassium sulphate found, per cent. 94 93 97 99 103 92 97 95 96 97 94 98 Mean 97 f 3 (la) Mean 96 f 3 (lo) The above results lead to the conclusion that the yield from the conversion of sulphur into hydrogen sulphide under the given experimental conditions is satisfactory for both types of material and that sulphur in gallium phosphide may also be expected to be converted into hydrogen sulphide, especially as the gallium phosphide is decomposed during the reaction, as described below.DETERMINATION OF SULPHUR IN GALLIUM PHOSPHIDE- Gallium phosphide reacts with platinum at 950 "C in an atmosphere of hydrogen. From X-ray diffraction analysis the reaction products were shown to be a mixture of a gallium - platinum alloy (Ga,Pt,) and platinum phosphide (PtP,). Sulphur present in gallium phosphide is released, converted into hydrogen sulphide on the platinum catalyst and determined coulometrically as sulphide in the electrolysis cell. Because of the reaction between gallium phosphide and platinum, the sulphur can be determined without interference from either gallium or phosphorus. This conclusion was checked by analysing gallium phosphide crystals* that were heavily doped with sulphur (98 p.p.m.1, which were allowed to react with different amounts of platinum. The reaction was?dlowed by monitoring the hydrogen sulphide formed.The results obtained are presented in Table 11. TABLE I1 REACTION OF GALLIUM PHOSPHIDE WITH PLATINUM Amount of gallium phosphide Amount of platinum takenlmg used/mg Sulphur found, p.p.m. Analysis tirne/hours 20 130 95 2 21 270 96 1 20 340 99 1 In order to obtain a rapid and quantitative reaction, a sufficient amount of platinum is required, irrespective of the physical nature of the gallium phosphide samples. For the determination of sulphur in gallium phosphide we used 300 mg of platinum per 20 mg of sample. The release of sulphur is not the rate-determining step because the analysis time is 1 hour for potassium sulphate and sulphosalicylic acid as well as for gallium phosphide.For comparison, gallium phosphide crystals from the same samples were analysed by Luke's method. The results for both methods are shown in Table 111. The results obtained by the two methods correspond very well, the striking difference being that with the coulometric method very small amounts of material can be analysed; down to 100 p.p.m. of sulphur in 0.55 mg of gallium phosphide (-0.05 pg of sulphur) can be determined by the coulometric method. Luke's method is not suitable for determining small amounts of sulphur because of the high blank value (1 to 2 pg of sulphur),2 which results from contamination by the chemicals *Prepared in our laboratory.380 GIJSBERS, BASTINGS AND VAN DE LEEST used, either from the atmosphere or from the analyst, or both, and attempts to eliminate these sources of contamination make the method more time consuming.TABLE I11 COMPARISON OF RESULTS OBTAINED BY LUKE’S METHOD AND THE COULOMETRIC METHOD Luke’s method’ Coulometric method A A I \ I -l Amount of gallium phosphide taken/mg Sulphur found, p.p.m. phosphide takenlmg Sulphur found, p.p.m. Amount of gallium 160 150 160 150 104 98 99 100 Mean 100 20.94 10.85 5.43 2-89 0.65 100 102 94 101 102 The advantage of the coulometric method is that it gives a low blank value (about 0.01 pg of sulphur) as the reactants are hydrogen and platinum metal, which can be purified very easily. Also, the method is carried out in a closed system, which reduces the risk of contamina- tion.The coulometric method is also sensitive, as shown in Table 111, as down to 0.05 pg of sulphur can be determined by this method, which is not possible by Luke’s method. With the method described above, sulphur can be determined in gallium arsenide (GaAs) and gallium arsenide - phosphide (GaAsxP,,) . CONCLUSION Gallium phosphide reacts with platinum at 950 “C in an atmosphere of hydrogen to form Ga,Pt, and platinum phosphide, while sulphur present in the gallium phosphide is quanti- tatively converted into hydrogen sulphide on a platinum catalyst. Measurement, by con- trolled-potential coulometry, of the hydrogen sulphide thus formed offers a sensitive method for the determination of sulphur in gallium phosphide. The reagents (platinum and hydrogen) can be purified easily and efficiently, resulting in a very low blank value (0.01 pg of sulphur) as compared with Luke’s method (1 to 2 pg of sulphur) and 0.05 pg of sulphur in gallium phosphide has been determined by this method. REFERENCES 1. 2. 3. 4. 6 . Luke, C. L., Analyl. Chem., 1949, 21, 1369. Adler, S. J., a;?d Paff, R. J,, in Willardson, R. K., and Goering, H. L., Editors, “Compound Semi- Cadersky, I., 2. analyt. Chern., 1967, 232, 103. Irimescu, I., and Chirnoaga, E., Ibid., 1947, 128, 71. Griinert, A., and Tiilg, G., TaZa&z, 1971, 18, 881. conductors, Volume 1, Reinhold Publishing Corporation, New York, 1962. Received May 21st, 1973 Amended November 8th, 1973 Accepted December 14th, 1973

ISSN:0003-2654    

DOI:10.1039/AN9749900376

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, June, 1974 BOOK REVIEWS 381 Book Reviews LIQUID SCINTILLATION COUNTING. Volume 2. PROCEEDINGS OF A SYMPOSIUM ON LIQUID SCINTILLATION COUNTING ORGANIZED BY THE SOCIETY FOR ANALYTICAL CHEMISTRY, BRIGHTON, ENGLAND, SEPTEMBER 13-16, 1971. Edited by M. A. CROOK, P. JOHNSON and B. SCALES. 1972. Price k8-75; $21.50; DM72. This book is the second volume of a series, the first one reporting a Symposium held a t Salford in 1970. It is divided into five sections corresponding to the sessions of the Symposium and it contains five plenary lectures and twenty-two shorter papers. The first chapter, by J. B. Birks and G. C. Poullis, gives an assessment of scintillator solution materials under conditions close to those used for current internal liquid scintillation counting. Solvents, primary solutes and the effect of quenching on them, and secondary solutes are assessed on the basis of relative pulse heights determined by a channels ratio method with carbon-14 as an internal source.Nearly twenty years ago a large number of compounds were synthesised and tested at Los Alamos, the pulse heights being measured by using an external radiation source. It is interesting to see how little the over-all picture has changed. This section of the book includes chapters on the absolute counting of beta-emitters, the use of an approximate mathematical model to explain the different effects of chemical and colour quenching on pulse-height distributions and a scintillation detector for labelled materials in high-pressure liquid chromatography. The second section of the book opens with a historical account of the development of the modern scintillation counter by E.Rapkin. Due to the wide use of liquid scintillation counting techniques in biochemistry and medicine the development of equipment has been mainly carried out by commercial instrument manufacturers, and a high level of sophistication has been reached. It is a competitive business. A firm that was technically successful in producing significant advances disappeared from the scene, leaving its successes to posterity in the instruments of its commercial competitors. The section continues with an account of problems experienced in chemiluminescence and how they were solved, or else remain mysteries, and concludes with two chapters on the bioluminescence assay of adenosine triphosphate and related compounds.This technique is included with scintillation counting because the equipment is suitable for the measure- ment of low light intensities, although it is not “counting” in the accepted sense, and no radio- activity is involved. However, “a fig for systems,” as J. €€. Fabre said when he included spiders in a book on insects. The third section is concerned with sample preparation for counting inorganic materials and the opening chapter by A. Dyer adequately reviews the field. The chapters that follow deal with the counting of calcium-45 in biological samples in the presence of strontium-90, the determina- tion of low levels of alpha-emitting plutonium isotopes and plutonium-241 (a low-energy beta- emitter) in urine by gel scintillation counting and the counting of carbon-14 and phosphorus-32 tracers in work on biological processes in polluted waters.This section includes a paper on radiocarbon dating with special reference to the use of liquid scintillation counting. The final chapter of the section describes a continuous automatic system for the Cerenkov counting of some fission nuclides by using substoicheiometric solvent extraction and displacement through ion- exchange resins. B. W. Fox describes sample preparation techniques in biochemistry with special emphasis on emulsion counting. The need for a full understanding of the fundamental processes involved in these emulsion systems, in order to avoid erroneous results, is emphasised in the discussion on the paper.The author of the paper deprecates the use of commercial “cocktails” based on secret recipes, which make the choosing of appropriate methods of quench correction in liquid scintillation counting impossible. The problems arising in heterogeneous counting are also evident in subse- quent papers in the section, which deal with the counting of labelled biological macromolecules and plasma samples. One speaker emphasised that counting after combustion was the best way to avoid these difficulties although there is no paper dealing specifically with this subject. It is to be hoped that it will be included in a later volume. An introductory paper by J. L. Spratt reviews the problems of data acquisition and the considerations that should decide the choice of computing facility.He suggests that excellent data handling is often performed on inappropriate data leading to “beautifully typed sheets essentially ready for publication-except that the Pp. xii + 327. London, New York and Rheine: Heyden & Son Ltd. The subject of the fourth section is sample preparation of organic materials. The final section of the book is on data processing.382 BOOK REVIEWS [Analyst, Vol. 99 figures are nonsense.” The next paper gives a description of an off-line computer programme for processing multilabel counting data with three nuclides. It includes the calculation of errors associated with external standard channels ratio quench correction and the effect on the former of irregular vial geometry. Other papers include one on drug distribution studies, in which the counting data are processed off-line to give the final pharmacokinetic constants required, and one on the computer handling of data for the radioimmunoassay of insulin labelled with iodine-125.The final paper gives a detailed account of the use of an off-line computer for processing data in connection with the applications of tritium tracers to problems in the oil industry. While this book, being one of a series, does not cover the subject completely, the papers and the discussions that follow them give much practical detail which should be particularly useful to those using this form of counting as a means to an end. The book is clearly set out and well produced. The comment one could make on price is the usual one today. D. I. COOMBER FUNDAMENTAL ASPECTS AND RECENT DEVELOPMENTS IN OPTICAL ROTATORY DISPERSION AND CIRCULAR DICHROISM.PROCEEDINGS OF NATO ADVANCED STUDY INSTITUTE HELD AT TIRRENIA (PIsA), 5-18 SEPTEMBER 1971. Edited by F. CIARDELLI and P. SALVADORI. Pp. xviii + 419. London, New York and Rheine: Heyden & Sons Ltd. 1973. Price All; $30.25; DM90. This book constitutes the published proceedings of the conference held under the auspices of NATO at Tirrenia, near Pisa, in 1971. The contributions included are only those of the invited speakers; they do, however, cover a wide range of work. The theoretical side of the study of optical activity is represented by papers from Mason, Rosenfeld and Moscovitz, Weigang and Tinoco and colleagues, while papers on the more empirical aspects of stereochemistry include a contribution from Kirk, Klyne, Scopes, Snatzke and Blout.There is an absence of studies in the carbohydrate field. Solvent effects are discussed by Legrand. Obviously, this is a useful sort of book for those who are aficionados but who were not able to spend a fortnight on the Italian littoral. Those not previously acquainted with the subject will, however, be able to pick up enough basic material from the introductory chapter(s) to attempt an understanding of the research papers. It is, in fact, refreshing to find so much historical material in the book, since it will help to counteract the impression that optical rotatory dispersion (ORD) and circular dichroism (CD) are phenomena discovered in America in the last 20 years. The book is made up of parts, so that the great names of the past, such as Arago, Biot, Fresnel, Cotton, Lowry and Kuhn, do not stand out with the clarity that would have resulted from a chapter devoted solely to the historical aspects. Because the book is a record of papers given at the conference, suitably edited, it does not include the discussions, which are often the most useful part of a conference from the point of view of the participants.On the other hand, there are contributions by many of the well known practitioners in the field, covering the subjects that have been most studied, i.e., the carbonyl group, the carboxyl group and co-ordination compounds. Out of line with the rest of the book in practical terms is the section on MCD. Although MORD and MCD measurements require modification of normal ORD or CD apparatus, the effect of these modifications is to remove the main restriction on this area of work, i.e., the need to have optically active substances to work with.Those who were prepared to resolve compounds in order to prove stereochemical points have always been a rather select band. In the study of ORD and CD they were joined by others whose interest lay in the field of optically active natural products. The techniques of MORD and MCD bring the study of optical activity within the orbit of spectro- scopists generally, as was shown by the attendance a t the Faraday Society symposium on “Magneto Optical Effects” in 1970. Thus, whilst CD studies are useful in elucidating the structure and origin of ultraviolet absorption, MCD is a technique that is likely to be developed mainly by spectro- scopists in order to obtain a better understanding of ultraviolet spectra in general, rather than by stereochemists as an extension of their field.So far as analytical applications are concerned, it is wise to be aware that the new techniques have not, in general, improved in accuracy on the time-honoured visual polarimeter, although with photoelectric instruments a reading can be obtained with much less material than is required for a visual instrument. The use of ORD and CD techniques for stereochemical correlation has analyti- cal implications and the MCD technique, by helping in the understanding of ultraviolet spectro- scopy, will influence analysis indirectly, and in particular cases it clearly could be used as a direct analytical technique if cheaper methods should fail.M. K. HARGREAVESJune, 19741 BOOK REVIEWS 383 SUGAR CONFECTIONERY AND CHOCOLATE MANUFACTURE. By R. LEES and E. B. JACKSON. The number of books in the field of sugar and chocolate confectionery is not as great as it might be, so that a contribution in this field is to be welcomed. The book, comprising some 380 pages, is set out logically with an introductory chapter dealing with basic technical considerations leading to several chapters detailing the characteristics of the more usual ingredients. The latter two thirds of the book are concerned mainly with the manu- facture of cocoa, chocolate and related products; boiled sweets, caramels, toffees and fudges; fondants, creams and crystallised confectionery ; gums, jellies and pastilles ; liquorice and cream paste; tablets, lozenges and extruded paste; marshmallow and nougat; and other types of con- fectionery.One chapter includes information on the calculation of recipes from analytical results and a collection of reference tables is included. The section covering the various types of sugar and related materials could, perhaps, with advantage have included molasses, and the possible use of sorbitol in diabetic formulations could have been mentioned as fructose is discussed in this context. Unfortunately, there appear to be one or two errors in the chapter on cocoa beans, but the section on fats and related ingredients provides a brief but useful description of several fats, antioxidants and surface-active agents commonly considered for use in confectionery production ; that on milk and milk products deals with milk, condensed milk, condensed whey, dried milk powders, sodium caseinate, lactose and butter.Gelling, whipping agents and gums are covered separately. The chapter on flavouring and colouring agents additionally includes a gr& deal of summary information on various nuts, dried fruit and other additives, which, it could be argued, merit chapters of their own. Readers may not find the chapter on cocoa, chocolate and related products entirely logical in its layout, and the order of the various sub-sections could be improved. For example, the sections on chocolate recipes and milk crumb have for some reason been interposed between those dealing with cocoa bean roasting and cocoa bean winnowing.Useful charts of possible faults and their prevention are included for many of the types of confectionery considered and there are a number of references quoted throughout, should more information be required on particular topics. While a section is included on the calculation of recipes from analytical results, no analytical methods are included in the book, and greater emphasis could, with advantage, have been devoted to aspects of quality control, including hygiene, microbiological considerations and, possibly, legislation. Not all manufacturers will find themselves in complete agreement with some of the views expressed, but nevertheless the book should prove to be a useful addition to the relatively few available and be of particular interest to newcomers in the field.Pp. xx + 379. Aylesbury: Leonard Hill Books. 1973. Price .@SO. P. H. WIGGALL INSTRUMENTAL METHODS OF FOOD ANALYSIS. By A. J. MACLEOD. Pp. vi + 802. London: The majority of books on this topic emanate from the U.S.A., so that this work by Dr. MacLeod is to be welcomed. This is an intriguing book and in many aspects original in its approach and the result of much hard work. There is no Preface, but the Introduction commences with a consideration of why foodstuffs are analysed and then discusses instrumental methods ucrsus non-instrumental methods, an argument which I believe is a non sequitur, but I must criticise the statement that “legal requirements demand the use of traditional methods’’ and that “instrumental methods are legally inadmissible as evidence in a court of law.” This statement is misleading since the number of “official” methods in the U.K.is minute and the official analyst has the discretion to use any method of value which will enable an accurate result to be obtained. All Public Analysts’ labora- tories have ultraviolet - visible spectrophotometers and gas chromatography facilities and most have a t least a selection of atomic-absorption, infrared, polarographic and electrophoresis instru- mentation available. The discussion on sampling and extraction in Chapter 2 is adequate to give the reader an insight into the difficulties involved. The theoretical aspects of paper, thin-layer, ion-exchange, gel-permeation, column and electro-chromatography are dealt with in great detail, followed by a discussion of their applications to a wide range of foodstuffs. The tables on the use of the named techniques for determining various components are useful and valuable, followed by a selected bibliography and literature references, which unfortunately do not go beyond 1970.In similar Elek Science. 1973. Price f112.384 BOOK REVIEWS [Arta2yst, VOl. 99 style, gas chromatography and ultraviolet and visible spectrophotometry are dealt with in very great detail and the above techniques comprise about two thirds of the work, I am rather surprised that atomic-absorption spectrophotometry receives such scanty treat- ment, since, as with all instrumental techniques, the instrument makers state that it will solve all our problems without too much effort-until we come across the snags in practice.The under- rated use of polarography in food analysis is given further emphasis by the author’s treatment, no mention being made of derivative or a.c. pulse techniques; short descriptions of infrared, nuclear magnetic resonance and mass spectrometry and an outline of automated analysis follow. Polarimetry and refractometry are considered to be out of date, although fairly short descriptions are given; a visit to food factories would soon alter the author’s opinion of their usefulness in a routine food laboratory. In my opinion the writing is rather loose; for example, the headings Results and Inter- pretation could be more correctly described as the Interpretation of Instrumental Data.Also, when discussing the determination of caffeine by ultraviolet spectroscopy, the interference of 5-hydroxymethylfurfuraldehyde is described (page 471) “which is not found in tea but found in coffee and thought to originate in chicory”! Again, further down the same page, consideration is given to the simultaneous determination of benzoic acid and methyl p-hydroxybenzoate ; mention should be made of the possible presence of sorbic acid, which has a maximum absorption very near to the latter and has been found together with the other two preservatives in some processed foodstuffs. An example of the uncritical approach to the literature quoted is that on page 310, in which the methyl esters of the minor fatty acids are quoted as a means of establishing the authenticity of pork products, but no mention is made that pigs may be fed on beef offals and thus, if minor fatty acids are present, the original meat can still be genuine pork meat. There are very few misprints and errors, the most serious is a “not” omitted on page 313 under analysis of pesticides and other residues in food by gas chromatography, otherwise it would appear that all possible agricultural chemical residues are being considered. A random check on literature references did not reveal any errors. On the front inside dust cover, it is stated that it is a textbook for all students taking courses in food analysis or food science. I beg to disagree-the scope and theoretical nature are, in my experience, too advanced for that group, but it can be recommended as a reference book for official and consultant laboratories, together with industrial laboratories who would like to have a book that will introduce instrumentation to newcomers and constitute a ready means of finding out how to determine a constituent of food without having to look up too many references. S. LANDSMAN Erratum JANUARY 11974) ISSUE, p. 70, line 15. For “aminohydrazone” read “amidinohydrazone”

ISSN:0003-2654    

DOI:10.1039/AN9749900381

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

384 [Analyst, VOl. 99 Erratum JANUARY (1974) ISSUE, p.70, line 15. For “atminohydrazone” read “amidinohydrazonc”

ISSN:0003-2654    

DOI:10.1039/AN9749900384

出版商:RSC    

年代:1974

数据来源: RSC