ISSN:0306-1353    

DOI:10.1039/AA97303FX001

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

PART I FUNDAMENTALS AND INSTRUMENTATIONPart I: Fundamentals and Instrumentation 3 1 Light Sources There continues to be steady progress in the construction, operation and understanding of light sources used in analytical atomic spectroscopy. Most papers, however, cover ground that has been examined previously : nevertheless, this repetition serves to consolidate knowledge and, by examining the subject from a different viewpoint, in some cases promotes its application to analytical situations.An example of this is the increasing use of the pulsed- mode operation of light sources. Hollow-cathode lamps continue to occupy a pre-eminent position in practical AA analysis, whilst EDLs and lasers attract research interest for their value in AF analysis. I t is unlikely that this relative emphasis will change in the near future, except in special cases, such as the use of EDLs for AA determination of Se and As and the use of tunable lasers if their cost were reduced and the ‘tuning range increased. 1.1 HOLLOW-CATHODE LAMPS 1.1.1 Spectral Line Profiles and Discharge Mechanisms Tilch and Wollbrandt (605), using a Fabry-Perot interferometer with pressure scanning, examined the current dependence of the emission from a Ca HCL and found that the line profile could be represented by Voigt functions, within experimental error, and that the Ca atoms were at a higher temperature than the Ne filler gas.The temperature depend- ence of the collisional width of the Ne line corresponded to a Van der Waals interaction. An approximation which simplifies the calculation of the Voigt function has been described (545).Wagenaar and de Galan (1641) conclude from an investigation of 18 atomic spectral lines of analytical interest that, except for a few elements (e.g., Ca and Si), hyperfine structure is a major factor in determining line profile, The width of the resonance lines of B emitted by an HCL depend to a large extent on the filler gas (532).Using Ar, Kr and Xe, the line half-widths were found to be 0.014-0.027 nm and the line profile was Gaussian with a small blue shift indicating Doppler broadening, an effect which does not occur for Ne or He (1540), caused by the high ejection velocities of sputtered B atoms. I t was concluded that the sputtered atoms were excited by collision with rare-gas atoms and that the cross-sections for collisional excitation differ for the various gases.Agreement between Monte Carlo calculations and experimental observation suggests that collisional excitation of the sputtered B atoms in Ar, Kr or Xe occurred within a very few collisions of their ejection from the cathode surface. From an investigation of the mechanism of the hot hollow-cathode discharge at various operating currents (744), it was concluded that sputtering was the dominant process up to 0-3 A, and that evaporation accompanied by fractionation, as the cathode temperature increased above 1400°K, occurred between 0.3 and 1.0 A.The authors also inferred that interatomic collisions played a major part in the hot hollow-cathode discharge. Kagan et al.(1354) have proposed a mechanism of formation of the distribution function of fast electrons in a hollow cathode. 1.1.2 Construction Lowe (523) has described a demountable high-intensity lamp in which the cathode holder was water cooled and a continuous flow of inert gas passed through the lamp. For low-melting- point cathodes, considerable gains in intensity were achieved by cooling the cathode, while high-melting-point materials give comparable performance to sealed high-intensity lamps.A method has been described (154) for the refilling and alteration of used HCLs, although the emission intensities of these lamps were less than those of the originals. In situations where replacement HCLs are not readily available, attempts at regeneration may be justified, but in view of the difficulties in manufacturing bright stable lamps with an acceptable life, it should not be lightly undertaken.4 Part I : Fundamentals and Instrumentation 1.1.3 Operation Interest continues in the use of pulsed HCLs for AFS.Cordos and Malmstadt (176, 177) have described a power supply which produces current pulses (200 mA peak current, 10 ms prrlse width) with a reproducibility of 0.002-0.003 (RSD) giving a light output stability of 0.0008 over a 10 minute interval, with a long-term drift (several hours) of 0.002.Human (894) investigated the use of pulsed HCLs in AF using current pulses up to 900 mA and 200 ms with a Cu lamp. For currents greater than 400 mA, however, there was no further increase in signal-to-noise ratio, despite the increase in the emission intensity of the lamp, due to severe self-reversal of the emission line.Additional information on pulsed and modulated HCLs will be found in references 31, 202, 528, 918, 1215, and in ARAAS, 1971, 1, 2, and ARAAS, 1972, 2, 2. Two papers have dealt with different aspects of the application of a magnetic field to a hdlow cathode. An A1 HCL placed in a magnetic field of up to 900 G and cooled with liquid nitrogen (835) produced double the intensity for the 396-15 nm line with a slight increase in line width.A development of potential value in AA analysis, where background absorption is a problem, is the modified HCL described by Stephens (1471), which is stable in a magnetic field. Zeeman splitting of the emitted atomic lines is achieved by using a low power electromagnet so that, by modulating the current to the electromagnet, the emission from the lamp is also modulated in perturbed and non-perturbed Zeeman components, which can then serve as reference and sample beams, respectively.This technique has previously only been applied to Hg analysis (ARAAS, 1972, 2, 5). 1.2 MICROWAVE DISCHARGE TUBES 1.2.1 Temperature Control and Muld-element Lamps Elcctrodeless discharge lamps are an intense source of narrow-profile atomic spectra, but their application to AAS and AFS has been limited by the practical problems of operation (519, 811, 1303, 1449).In some cases, an increase in the light output of a lamp is accomp- anied by an increased spectral line width and hence a loss in absorption sensitivity. Attempts to improve the stability of EDLs by changing the manufacturing technique (975) and by using optically controlled feedback (1243) continue to be made, but temperature control of the lamp appears to be the most promising (26, 1244, 1304).In the case of multi-element lamps it was found (174) that the optimum temperature depended solely on the element, while the intensity of the emission could depend on the other elements present.Pate1 c't al. (26, 264) have presented the results of fluorescence measurements with a graphite rod atomizer using multi-element lamps under the following conditions : Cd-Hg-Zn at 275OC, Cd-FeHg-Tl-Zn at 285 OC and Ag-Co-Cr-Cu-Fe-Mn-Ni-Pb-Sn at 45OOC. An alterna- tive system of multi-element operation using AF employs two dual element lamps driven by a single power supply (181). 1.2.2 spectral Overlap In a few cases advantage can be taken of spectral overlap to determine one element by using the radiation from another. Norris and West (1040) have employed this approach for the determination of Cr by AA and AF using the overlap of the Ne 359.352 nm line with the Cr 359.349 nm line.* 12.3 Gas-flow Microwave Lamps Dagnall et al.(716) have described a new type of discharge lamp in which argon flows through a quartz tube at atmospheric pressure. A quartz cup containing the element of interest was suspended in the tube in a region where a micro-wave cavity excited the discharge. Although the lamp life was short, an intense emission was obtained for elements * See also: Manning, D.C., Atom. Absorp. Newsl., 1971, 10, 97.Part I : Fundamentals and Instrumentation 5 having compounds with boiling points in the range 1240-2200°C, and they could easily be recharged. Bazhov and Balshin ( 5 5 5 ) have described a lamp consisting of an opaque bulb divided into two chambers by a transparent window. One chamber was an EDL, and the other was filled with vapours of the sample.The latter was provided with a side window for the outlet of fluorescence emission. 1.2.4 Safety Stanley et al. (506) have sounded a warning note in the use of microwaves, as the health hazards are not yet predictable. They have found that radiation levels in excess of the prescribed safety standards have been detected in the vicinity of commonly used microwave cavities.Consequently, shielding precautions and on-site radiation surveys are to be recom- mended. It has been reported that an r.f. spectral lamp may be readily ignited by irradiation with light from a photographic flash-gun (1 194). 1.3 LASERS 1.3.1 Application to Atomic Fluorescence Analysis Winefordner (881) and Omenetto (1526) have demonstrated that there are advantages in using high-intensity pulsed sources to achieve near saturation of atomic energy levels in AF.These advantages are (a) the fluorescence signal is not greatly affected by source stability, ( b ) for a two-level system, the fluorescence signal is no longer dependent on the quantum efficiency, and so one can use more-reducing flames, and (c) the linearity of the analytical curve is greatly increased, particularly at high optical density. A limitation of the pulsed high-power laser is its low pulse-repetition rate.Using tunable laser excitation (200) only one source was required to excite the fluorescence spectra of 35 elements in air/H,, air/C2H1 and N20/C2Hz flames. Flame back- ground and analyte emission interferences were essentially absent, and non-resonance fluorescence was used to eliminate noise due to scatter of source radiation by the flame gases. 1.3.2 Applications to Fundamental Studies of Atomic Spectra From the point of view of the practising analyst the additional complexity and expense of laser-excited AF is not justified, as the technique has little to offer beyond what can be obtained using simpler AA techniques.The greatest value of laser-based studies is likely to be in the insight this approach provides to atomic and molecular energy levels and transitions, c.g., the fluorescence spectra of Iz excited with 528.7 nm radiation (232) and of CH in an O,/CzHz flame excited at 431.5 nm (1068); the location and pressure broadening of absorption lines (1001) and laser-saturated atomic resonance fluorescence (7 lo).Konjevic and Konjevic (997) have introduced an aidnatural gas flame into the cavity of a dye laser and used it for the determination of Na by AA down to 0.2 ng 1-'. This approach was proposed by Peterson et al. (ARAAS, 1971, 1, ref. 229) for the enhancement of absorption spectra and would appear to justify further investigation. The literature on the design and operation of lasers is voluminous, but it is not appropriate to record it here; however, typical material relating to dye lasers is given in references 1242, 1350, 1360.A comprehensive literature survey of chemical lasers has been prepared by Arnold and Rogjeska (240). 1.4 CONTINUUM SOURCES The suitability of a continuum light source for use in AA has again been investigated (604).A high-current (100 A) stabilised graphite arc which proved to be a better source than tungsten-filament, deuterium or xenon lamps has been investigated theoretically for the determination of Mg by AA (1129). The temperature of tungsten-ribbon lamps has been determined (949) by relative measurements of the spectral and integral radiant energies, in6 Part I : Fundamentals and Instrumentation the wavelength range 500-800 nm; the results in the temperature range 2000-3000°K agreed with the certificated values within 30°. I t was found that the electrical power consumed was approximately proportional to the integral radiation. With the increasing use of pulsed line sources in AF, parallel studies of the suitability of flash tubes would be of value. A source of high intensity for ultraviolet (184.9 and 194.2 nm) radiation based on a mercury vapour discharge lamp has been described (250), in which the efficiency is improved by using an electron-emitting hot-gas filament inside the source envelope and by employing high current densities (20 A cm-').

ISSN:0306-1353    

DOI:10.1039/AA9730300001

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

Part I : Fundamentals and Instrumentation 7 2 Excitation Sources and Atomizing Systems 2.1 ARCS AND SPARKS The role of an atomizing source in spectrwhemical analysis can best be compared with that of the solution process in classical wet chemical analysis. Emission sources show off their full inherent strength in direct simultaneous determination of many elements in solid samples. Progress in this field can be achieved either by extending and improving the knowledge of the actual sampling from a solid surface and the processes occurring in the excitation plasma, using well established methods, or by introducing new techniques or new combinations of known techniques.In an excellent progress report by Laqua (900), the sources for the analysis of solid materials by optical emission methods are reviewed.The hope is expressed that widespread tests of the new procedures will lead to improvements and accelerate their introduction in routine analysis. A description of the processes occurring during the spark excitation of solution samples has been given by Barnes (870). A variety of solution-introduction approaches, the rotating disc, the wick, the vacuum-cup and the porous-cup electrodes, and the spark-in-spray technique were investigated employing submicrosecond time-resolved emission spectroscopy.A study of d.c. arc plasma behaviour in respect of volatilization rate, temperature, elc'ctron density and electron field strength, when determining trace elements in silicate rocks, was reported (19, 889). The effect of various discharge gases on arc phenomena, and hence on the results of the spectral analysis, has been reported by Nickel et al.(819, 820). Spectral intensity changes occurring during the sparking period, with vacuum spectrometers, have been related to the structure of the sample (845, 846) and to the presence of oxygen from the decomposition of non-metallic inclusions (354). Some of these effects can be reduced by the use of overcritically damped medium-voltage excitation under Ar (848). An assessment of 39 metals as the counter-electrode material for the determination of gases in metals leads to the selection of W, Mo or Cu (449), with Cu being preferred because of its simple spectrum (357), except for the determination of oxygen, when a graphite electrode is recommended.Current experience of the precision and versatility of a controlled waveform source is reported (1451). (See ARAAS, 1972, 2, 8, ref. 741). The use of an a.c. arc as an atomizer for AA analyses is reported (909) and some results of an investigation of a stabilized d.c. UYC plasma device for atomic spectrometry described (1460). 2.2 PLASMAS The development of plasmas as spectrochemical excitation sources has continued in three well-defined areas, viz., the low power (<250 W, 2450 MHz) enclosed microwave discharge, the inductively coupled high-power (>1 kW, 1-40 MHz) plasma jet and a variety of d.c.discharges. The last two years have produced some very encouraging results, particularly in the determination of non-metals and refractory elements.I t is to be hoped that consolidation of this work will take place in the near future. A low-power microwave plasma, developed as an element-selective detector for gas chromatography (479, 552) is now commercially available from Applied Research Labor- atories as a complete instrument (893). Accessories for use with a microwave discharge for a "NPN isotope analyzer have been described (814).Layman and Hieftje (818, 1279, 1467) inserted a loop of tungsten wire carrying the sample into the lateral aperture of a $-wave Rroida cavity and vaporized the sample by striking an arc (100 mA modulated d.c.) with a tungsten anode positioned along the axis of the cavity. With a 100 ul sample, a detection limit of 30 ng 1-' B was claimed. A modified Evenson cavity was used by Lichte and Skogerboe (189) to obtain good detection limits for several elements, e.g., 10 1-' for B; their instrument employed continuous nebulization with Ar at 900 ml min-' and a heated spray chamber. Incorporation of the sample into a sealed EDL has been used (256) to8 Part Z: Fundamentals and Znstrumeritntion determine nanogram amounts of Cd, Tl, In, Pb and Hg-addition of Ge and ZnCL was rxommended to improve reproducibility and to act as an internal standard.A few workers (494, 957, 1348) have experimented with high-power nzicrowave-i~duced plasmas as emission sources. A Fehsenfeld cavity* was used (1348) to create a 1000 cm3 plasma with 2-5 kW microwave power at 2450 MHz. Soviet workers (494) used 200-1500 W plasmas of Ar, He, Nz or air with introduction of a wet aerosol to obtain modest detection limits for several elements.Sermin (957) used a 600 W Nz microwave discharge for several applications. High-power plasma jets continue to be investigated as sources for refractory elements including the rare earths, either in emission (349, 1151) or absorption (1126, 1217). Boumans and co-workers (370, 801) have developed a stabilised 50-70 MHz, 2 k\V 4 r plasma based on a Colpitts oscillator, with detection limits for over 20 elements in the range C.02-10 Irg 1-'. f Additions of propane or isopropanol (1217) gave improved dissociation of Zr, Ta and Nb oxides in the plasma, although this was disputed (657) for Zr in Ar with methane additions.Nebulizers used with inductively coupled plasmas varied from a modified pneumatic design from an AA spectfometer (294), and an ultrasonic nebulizer (1420) to pneumatic (349) or ultrasonic (370, 1353) systems with heated spray chambers.Recause of other differences in instrumentation it is not possible to evaluate the relative merits of these nebulizers from the detection limits reported. Measurements of Ar gas tenzperaturcy and Electroil densities (800, 802) in plasmas have again been reported.The plasma column temperature was measured as 6000OK at 10 mm above the coil, decreasing to about 4000OK in the tailflame (802), using Zn(1) 307.6/328.2 nm line-pair excitation and from Mg(I1) 2 79-6/Mg(I) 285-2 nm ionization without Abel inversion. The T i 322*2/322.4 nm line-pair with Abel inversion gave a core temperature of 5800-6400°K (800).Using a computer iimulation of inductively coupled plasmas, developed by Barnes (8, 1475), Ar flow profiles, gas and particle dynamics during sample introduction, particle decomposition, atomic radiation signal and background intensities can be studied as a function of frequency, field atensity, power levels, etc. The model also demonstrates change-over from ellipsoidal to toroidal plasma shape.A variety of d.c. plasmas has been reported, which always gives rise to the difficulty of differentiating between these and the familiar d.c. arc. I n general, the d.c. arc is extinguished between introduction of samples, whereas plasmas are usually operated continuously-this difference has been used here as a crude criterion.Kleinmann and Svoboda (61.5, 618, 887) \tudied the introduction of a de-solvated aerosol into a d.c. Ar discharge and found that at uptake rates of 6 ml min-' any increase in sample flow-rate resulted in a slight back-pressure in the heated nebulizer chamber which tended to reduce the flow. However, in the condenser, the additional volume decrease was greater than in the steady state, with the result that back- pressure decreased and a higher uptake rate was induced.Thus oscillations in sample flow could occur resulting in instability of the emission signal. This effect, which is of importance with all aerosol de-solvation systems, could be eliminated using a flow restrictor (6 mm X 0-7 mm diameter glass tube) at the condenser outlet or by incorporating a large dead volume.Yudelevich and co-workers (1396) introduced powdered samples of rock into a d.c. plasma to obtain detection limits 10-100 times better than those obtained with a d.c. arc. A spray chamber nebulizer improved the stability of the Spex plasma jet (483), and detection limits for 67 elements were found to be little better than those obtained using an N,0/C,H2 flame.A low-current air discharge was used (981) to obtain modest detection limits for several elements using photographic recording. Better results may have been obtained with photo- electric detection. An inexpensive low-voltage d.c. plasma jet was described (1556) for steel analysis, particularly for C, Mn, S and Si determination. * H. P., Rev. scient. Instrum., 1965, 36, 294. t For a review of microwave cavities see: Fehsenfeld, F.C., Evenson, K. M., and Broida, See also: Boumans, P. W. J. M., and de Boer, F. J., Spectrochim. Acta, 1972, 28B, 391Part I : Fundamentals and Znstrumentation 9 The literature on theoretical plasma studies is very extensive and every attempt has been made to exclude as much of the irrelevant material from this review as possible.A text in English (334, 919) describing the choice of excitation conditions and line-pairs also mentions an interesting pulsed-discharge plasma as well as a theoretical comparison of induction- coupled and d.c. plasmas. Atomic absorption has been used (643) to investigate the distribu- tion of excited Cs in the positive column of a low-pressure Ar discharge. Electron densities, ion temperatures, radical profiles and particle velocities have been studied in He d.c.plasmas both with (765) and without (879) an applied magnetic field. A coil placed around a d.c. discharge (1 35 1) was used to determine electron densities and collision frequencies by inductance measurements; results compared well with those given by a cylindrical Langmuir probe. A photographic technique (1476) was used to locate the distribution of elements in an arc plasma. 2.3 GLOW DISCHARGE LAMPS The glow discharge lamp (GDL) of Grimm* has been commercially developed as an alternative to arc/spark emission (reviewed in detail in ARAAS, 1972, 2, 10).Potential users are awaiting experimental verification of some of the claims made for this source, viz., linearity of calibration over a wide range, freedom from interferences and high precision.S ample-handling techniques and non-metals determination are also areas of current interest. Improvements in the analysis of powdered rocks, ores, minerals, glasses, slags and cement are described (779) using a single standard of 10 elements in CaC03 mixed with copper powder. Liquid samples were analysed (849) by evaporation on to porous copper electrodes made by pelletising a mixture of metal powder and a volatile compound, e.g., naphthalene.Evidence of good sensitivity is given by an account of the direct determination of impurities in GaAs semiconductor material (644). Depth profiles of elements in stainless steel and nickel alloys between 0.1 to 40 have been reported (222).Czakow (1049) investigated the use of hot and cold, hollow and plane cathodes for the determination of Se, F and C1 in Uz03 in addition to many other metallic elements. A photon-comting detector was compared (1 160) with a d.c. amplifier for GDL determination of A1 and S in steel. Butler (764) found minimal interferences over a wide range of copper- base alloys provided the total intensities of all major alloy element lines were added and used as an internal standard; very accurate figures were also given for unrefined (60-80%) and refined (99.5 t 0.1%) gold. An interesting application is the use of the GDL with single element and demountable resonance detectorsf (850).Hollow-cathode lamp excitation has few of the advantages claimed for GDLs; never- theless the HCL may offer some advantages, particularly in analysis of solutions that are evaporated on to the inside of the graphite cathode (891).The cathode may be uncooled (833, 891) or cooled with a water jacket (891) to permit the use of currents up to 150 mA. The Glomax (Barnes Engineering Co.) demountable HCL has been used to determine 10-80 ng Au in 0.1 ml of solution (1029).Similar detection limits were obtained (836) for [he determination of rare-earth elements in Am. An electromagnetic coil wrapped round the body of an HCL (837) permitted the regulation of excitation processes, such that as the field was increased, the electron concentration in front of the cathode wall increased, causing a rcduction in the space charge and resulting in a more arc-like spectrum.A maximum cmission intensity of atomic lines was found at a field strength of 700 oersteds. Theoretical studies on atomization and excitation in uncooled hollow cathodes (834) and sputtered discharges (645) were described. Currentholtage characteristics were used (834) to follow the process of decomposition of transition metal chlorides and exit of the vapours from the cathode cavity.*. t Grimm, W., Spectrochim. Acta, 1968, 23B, 443, and Naturwissenschaften, 1967, 22, 586. Sullivan, J. V., and Walsh, A., Spectrochim. Acta, 1965, 21, 727.10 Part I : Fundamentals and Instrumentation 2.4 LASERS A review (598) of the development of spectroscopic microanalysis for elements in steel included results on laser microprobe studies from the authors’ laboratories.The commercially available Carl Zeiss Jena LMA-1 laser microprobe has been compared (838) with a new instrument, the LMA-10, for the examination of mineral specimens. A variable-path Q-switched cell for the LMA-1 featured (329) six modes of operation where the number of spikes per pulse could be changed from about 1 to over 100 to suit the analytical problem. Comparison of laser microprobe (38 1 , 1400) vaporization with conventional sparks shows very much reduced third-element effects with the laser; nevertheless there were some indications that excitation processes in the discharge are also involved.One of the problems of laser microanalysis is the relatively small analytical signal recorded on the photographic plate. Hagenah, Laqua and Leis (839) surrounded the path of the laser plume with an Ar filled rectangular microwave cavity excited by a 2.5 kW, 2450 MHz magnetron; the resultant optically thin plasma offers sharp lines with very low bzck- ground and there is no tendency for arcing between electrodes and metallic specimens.Furthermore, the authors estimated that the residence time of the vapour in the cavity was about 1000 times greater than for a self-luminous laser-produced plume, which showed a rapid decay.Petrakiev and Georgieva (842) found that emission intensities could be increased by surrounding the disc-shaped sample with a permanent magnet in the form of a washer; ferrite and alloy magnets were tested, as was also the distance between the sample surface and the top surface of the magnet.Spectral line intensities were increased by factors varying from 1.5 to 6, but the mode of action of the magnetic field in producing this intensification was not discussed. If the t h e of excitation, after the laser is discharged, is controlled and is not determined by the electrode striking voltage, then both sensitivity and reproducibility may be increased (992).Using a 0.4 J laser with 1060 nm radiation, optimum delay times were found to be 625 us for steel and 400 us for wolframite. A laser of less than 0.4 J (1202) with a unipolar electric discharge was proposed for the examination of metallic samples, while others (1060) reported improvements in sensitivity by applying the old trick of pre-exposing the photo- graphic plate to develop the so-called ‘latent image’, (the Herchel effect).Morton, Nohe and Madsen (221) measured spectral line intensity as a function of sample weight vaporized, which was measured by optical microscopy on the assumption that an essentially conical crater was formed. If the weight of sample vaporized was taken into account, other matrix effects were low for determinations of Cu and Sn in Al, Zn, Pb, Sn and Fe-base alloys and for Ag in Pb-base alloys.The use of lasers in conjunction with AA measurements on the plume (see ARAAS, 1971, 1, 12) has received further attention from the U.S.S.R. Academy of Sciences, Moscow, (830, 1201, 1260) and Oregon State Uriversity (263). Both schools favour pulsed HCLs and carried out some of their studies on samples pelletised with graphite powder.Atomic absorp- tion with simultaneous correction for background using a non-resonance line (830) was used to plot atomization profiles for Cu: maximum absorption occurred 15 mm above the crater and was symmetrical with respect to rotation about the axis of the laser shot. Limits of detection for Cu were of the order of 1 bg g-’ or about 20 pg absolute. 2.5 FLAMES 2.5.1 Flame Types Non-metal determinations using cool-flame methods have relied on molecular absorption (1542) or emission (1543); for the determination of S by absorption (1542) the 207 nm band was preferred to that at 300 nm. Some of the new work on cool flame emission is described in Section 2.5.3. Last year (ARAAS, 1972, 2, 12), work on the determination of non-metals at vacuum- ultraviolet wavelengths was reported.The almost optically transparent Nz-separated N20/C2H2Part I : Fundamentals and Znstrumentation 11 flame was used together with a Nz-purged optical path and monochromator (614). Whilst no improvements in detection limits have been obtained for this approach over the past year, it is encouraging to find several practical applications (119, 325, 466, 633, 998, 1155, 1291).A detection limit of 10 mg 1-‘ for P was reported (1542) by PO molecular absorption in an NzO/C2H2 flame at 246 nm using the absorption at 241 nm as a background reference. The absorption spectra of CaF (530 nm), A1F (227.5 nm) and InF (234 nm) indicated that the AIF band is sharper and preferable to the well known CaF band (1542) for the determina- tion of F.Molecular band emission in N20/C2H2 or NzO/Hz flames (1248) gave better detection limits for Y, La and Gd than AAS. Band emission detection limits for Sc, Y , La, Nd, Gd, Tb, Dy and Lu were better than those obtained by atomic emission although inferior for Eu (see Table below). I n general the N20/H2 flame gave equal band intensities but considerably lower noise than NzO/CzHz.Wave1 ength/ nm La 550.1 L a 0 441.8 EU 549.4 ,, 39 9, Technique Flame Detection Limit/ mg I-’ AAS NzO/C& 3.0 FES NzO/CzHz 0.4 FES N,O/Hz 0.02 AAS N:O/C,Hz 0.0 15 FES NzO/CzHz 0.0004 FES NzO/H, 0-08 A review on high-temperature premixed flames (371) gave data, including degrees of atomization, for 25 elements in seven types of flame, Workers at Iowa State University have studied the premixed Oz/Hz flame both practically and theoretically (634) and concluded that although it would be only of limited value for AA, the low background permits the attain- ment of detection limits in emission which are comparable with those obtained by emission in N,0/CzH2 flames. Three burners were used (211, 372) in a comparative study on the determination of Ba with Ar-shielded NzO/CzHz flame.Stephens (969) described preliminary results with an O2-shielded air/CzHz flame. The surrounding O2 shield results in a higher flame temperature which in turn allows a higher CzHz flow before the onset of luminosity. The fuel-rich flame shows reducing properties and a high CN emission in the reaction zone. Although low-pressure flames are well known (ARAAS, 1971, 1, ref. 21), their potential analytical properties have not been examined hitherto. Stephens and West (714) studied the N,O/CzHz and O2/C2H2 flames at 15-300 and 10-100 torr, respectively, for several metals which form refractory oxides and found that atomization efficiency increased with decreasing pressure due to the highly reducing properties of the low-pressure flame. Laminar NzO/C2Hz and Oz/(CN)2 flames were used to study the burning characteristics of micrometre-size particles (1 342); the burner and flame characteristics were described, together with working curves for MnOn and MgO. 2.5.2 Burners and Nebulizers Few significant advances have been made in the design of burners and nebulizers over the past year. Only a few studies have been reported on ultrasonic nebulizers (1121, 1299), although these are now in common use with inductively coupled plasmas.Reports on heated spray chambers either with (375, 385) or without (2, 169, 257, 258, 442, 1147) solvent condensation continue (see Section 2.2). Published burner designs include a long path ‘multi-flame’ burner for air and N20-supported flames (728), a burner with three rows of 2.2 mm diameter holes for combustion of petroleum products (166), a burner designed to use liquid hydrocarbons as fuel (554) and a multi-port burner made by drilling a series of 3212 Part I : Fundamentals and Iizstrumeiitatioir holes on either side of an old single-slot burner which also had aluminium radiator fins attached (21).Cresser and Wilson (756) interposed an 8.5 litre tank between the cylinder valve and flow meter to overcome the problem of a fall in temperature, due to N,O expansion, at the cylinder regulator, leading to instability in flow rate.(N.B. A high cylinder-regulator pressure-about 1GO p.s,i.g.-or a low-voltage heating tape around the cylinder regulator achieves the same result). Fundamental stzidies on nebulizers include the calculation of particle diameters (1 67) for various air pressures, photography of the aerosol and microscopic measurement of the prints ,168), and more studies on the use of single-droplet injection techniques (1039, 1300, 1302).The last is one of the few devices capable of giving unambiguous data on de-solvation h e t i c s and chemical interferences. The single-droplet technique has been incorporated into a ‘null-point’ detection system, where the relative rates of introduction of sample and standard into the flame were varied, to obtain equal emission intensities, so that sample concentrations could be calculated from the relative injection rates (1039, 1302).It has been shown to t e possible (1245) to detect and size particles by forcing an aerosol of dust particles through the cavity of a He-Ne laser and focussing the scattered light on to a detector by using a hemispherical mirror. 2.5.3 Discrete Sampling Devices Ingenious devices for the introduction of measured volumes of sample into the flame continue to be produced, together with improvements to commercially available microsampling dccessorics. Delves and Reeson (478) described a time-delay circuit which permits intermittent operation of the recorder when using the Delves Cup technique; in this way the non-atomic absorption signal is not recorded and the true atomic signal is easier to read.Replacement of the nickel tube and cup with fused silica items (475) was recommended together with the use of aqua regia as an oxidant instead of HzOz, resulting in a negligible ‘smoke’ signal being obtained.Further improvements to the Perkin Elmer microsampling accessory mounts (129, 364, 366) allow smoother and more convenient operation. Pulsed electrolytic deposition (0.2 s deposition, 2 s interval, 5-100 pulses per determina- tion with 1-5 V pulse height) of Cu, Hg, Mg, l?b and Zn on to an iridium probe from 2 ml df solution allowed a concentration step and a separation from, possibly interfering, organic matter (53, 721).The iridium probe was then inserted into an air/Hz or Ar/H2 flame having d nimonic tube mounted along the optical axis of the AA spectrometer. Prudnikov and co-workers have developed several devices for sampling discrete volumes of solution. Their platinum loop inserted into an air/CzHz flame with a silica adaptor (692, lOSl) is similar to designs mentioned in earlier reviews (ARAAS, 1971, 1 and ARAAS, 1972, 2, sections 2.5.3).A micro-cup of pyrolytic graphite (3-5 mm deep, 1.5 mm inner diameter), however, has been used with an NzO/CzHz flame (379, 732, 1398, 1399) to give a notable success in the determination of all the rare-earth metals, Ba, Al, La, V, Mo and Cr together with many more easily atomized elements.Concentration detection limits, using emissior, with automatic background correction, for all elements were generally as good as or better than by conventional AES; absolute detection limits varied from 1 ng to 10 fg (lO-I4 g). An electrically heated graphite plate (1 X 0-5 mm) or micro-cup was used (1257, 1469) logether with an N20/C2Hz flame to determine many elements, including precious metals, using emission, absorption and fluorescence methods.Nebulization of small measured volumes of sample using pneumatic or ultrasonic nebulizers has been used with solutions of high solids content to overcome burner blockage. A vertical funnel-shaped adaptor leading to the capillary of a pneumatic nebulizer was used (599) for 10-500 ul of a 2% iron solution; with small volumes (e.g., 25 ul) the performance of the spray chamber is modified and there is an increase in absolute sensitivity.Others (192) have used samples of 200 ul or less with ultrasonic nebulizers and confirmed this enhancementPart I : Fundamentals and Instrumentation 13 of sensitivity. At the other end of the concentration range (1394), the injection of a sample into a flowing stream of water allowed dilution and rapid determination of K, Nn and Ca in serum.A combination of arc vaporization and AE or AA detection using an air/CzHz flame was described (664, 1059) for the analysis of solutions and solids with RSD of 0.035 and 0.054, respectively. Direct solids analysis using an iron screw impregnated with sample (533, 1547) was similar to that described previously by the same author (ARAAS, 1972, 2, ref. 581). Candoluminescence and the so-called molecular emission cavity analysis (MECA) are approaches currently under investigation (1228, 1479, 1539, 1658) for the direct analysis of solid samples utilising molecular emission spectra. The sample is packed into the hexagonal recess of an Allen screw and is inserted into a cool NJH, flame in both techniques.With candoluminescence, molecular emission is induced from the solid surface itself, while for MECA, molecular band emissions similar to those normally observed in cool flames are excited. The cavity limits the optical viewing geometry of the sample and reduces the temperature of the sample.With S? emission (1658) the signal is a peak of half-life about 5 s for organo-sulphur compounds ranging to 20 s for involatile inorganic compounds. The peak intensities and times of appearance depend on the form of sulphur present. A further development allows injection of liquids or GLC eluments through the rear of the cavity. 2.5.4 Theoretical Studies The literature contains vast numbers of reports dealing with theoretical studies on flames, and it would b2 an impossible task to include these in this book.Our correspondents have, therefore, sent us accounts of those investigations more relevant to analytical spectroscopy. ?"he techniques used for studying the physical and structural characteristics of flames have been reported (1623, 1655) : in particular methods have been described for the measurement of free atom and radial temperature distributions.Flame temperature measurements continue to interest investigators. The use of AF via resonance and thermally-assisted processes (see ARAAS, 1972, 2, ref. 616) has been extended (711), and a tunable organic dye laser (896) has been used to excite T1 fluorescence at 377.6 nm when tuned to the 377.6 nm (resonance) and 535.0 nm (antistokes direct line) wavelengths.The use of a laser has the advantage of allowing very small areas of the flame to be examined. Other methods described included thermocouple (265, 729), emissicn/ absorption of radiation by solid particles (428), radiation probe (556) and refractive index (1345) techniques. Heat transfer from a flame to its unburnt fuel (557) mixture was studied In terms of the respective emission and absorption spectra.Methyl radical concentrations in hydrocarbon flames were measured (50) using a hitherto unreported absorption peak at 212.5 nm, and the influence of water vapour on the emission spectra of flames (423) was said to have little effect on C o t emission, although black-body radiation from incandescent carbon particles was reduced considerably. Information on atomic line profiles is scarce, and it is, therefore, particularly interesting to read the accounts of de Galan and co-workers (876) and Kirkbright and co-workers (265) describing studies on analytical systems using Fabry-Perot interferometers.Measurement of the line profile of the HCL for Ca 422-7 nm (876) allowed a calculation of absorbance versus lamp current to be made which gave good agreement with experimental results, while the hyperfine structure of Mn 403.1 nm was resolved into five components.The slit profiles of an Xe arc, for the cases theoretical, as observed and after absorption by 25 mg 1-' Ca, were given, together with the experimental absorption profiles from 1000 mg 1-' Fe at 302.1 nm and 10 mg 1-' Cu at 324.7 nm.The other investigations (265) were carried out at Ca 422.7 nm; variations of AA half-widths versus concentration were used to obtain a-parameters and collisional cross-sections for Ca in Hz diffusion and air/CzHz flames. AAS working curves were calculated from a theoretical expression (713) using a flame containing the element of interest as a light source.14 Part I : Fundamentals and Instrumentation The determination of oscillator strengths and atomic transition probabilities using AAS has been discussed by Parsons (1498) from both theoretical and practical viewpoints.Delibas and Toma (440) determined the oscillator strength, f , by atomic absorption for 14 atomic lines of Cr, Mo, Sn, Bi and Pd.Results on lines not previously studied are tabulated below: L i n e / n m 360.53 520.84 386.41 319-39 223.06 227.66 276.3 1 f 0.077 0.080 0-066 0.025 0.54 1 0.036 0.036 The measurement of molecular-band to atomic-line intensity in two flames at the same temperature, but with different gas compositions, was used to identify the visible spectrum emitters (monoxide or hydroxide) of alkaline earths in air/CzH,, air/H?, NJCO, 02/N2/C0 and N,Q/CO flames (CaOH at 554, 602, 623 nm; SrOH at 606, 647, 669, 682 nm; BaO and BaQH at 487, 612 nm).Studies of the band spectra of alkali and alkaline-earth metals (1495) showed them to be qualitatively identical for 25 compounds. Tke intensity of the Na line 589.0/589-6 nm emission relative to that of Na continuum 200-500 nm was reported for several flames (796).The ionization of metals in flames, as discussed previously by Kornblum and de Galan (ARAAS, 1971, 1, ref. 617) has now been published (1640); similar studies have also been reported by Fabrikova (929). The atomization efficiency (p) of metals in flames underwent significant experimental development a few years ago (ARAAS, 1971, 1, 17).Some further advances were reported this year. Behera and Chakrabarti (500, 1437) measured p-factors in fuel-rich N20/C2H2 flames for Ti, Zr, Hf, Nb and Ta; they plotted distribution profiles, used several solvents and several metal salts. Qualitative studies (609) were reported for Mo in air/C2Hz and N20/C2H2 (504, 880, 1504), T i in N,/0,/C2H2 (832), Co in air/CzHz (926), and Sn in H2-supported flames (297, 617, 831).A new value for the dissoctazion energy of PbO was given (252) as 382 kJ mole-'. Calculated atomization efficiencies (982) using tables of thermodynamic equilibrium data gave quantitative explanations of the variation of Ca atomization with changes in water content, Ca concentration and flame composition. The problem of interferences in analytical flames has been tackled at several levels of sophistication.Chemical interferemes are probably the most studied and least accounted for by quantitative theory or experiment. Hieftje and co-workers (1501) have followed the interferences of anions on Ca in an air/CzHz flame using a single-droplet generator. Mohay (1063, 1064) has examined the same problem, again using a single-drop generator, a theoretical treatment of which has 'been written by Neukermans (1 352).Interference studies using classical techniques continue to appear, including phosphate on Ca in air/C.H, (57, 1470), anions on Ca (529, metals on Ni in air/C2Hz (625), Nb on alkaline earths in air/C2H2 (382), and various ligands on A1 (1175). The releasing action of NH,Cl on the interference of Fe on Cr in air/CzHz has been studied using two nebulizers (712), while the interferences of HC1Q4 on In was shown to be a gas-phase process (1569) as evidenced by the presence of InCl band emission.No differences between interferences obtained in FES and AAS for Mo, A1 or V could be found (196) in contrast to results obtained by other workers. A paper on the role of catalytic processes in interference described previously (ARAAS, 1972, 2, ref. 625) has now been published (1639). Reactions between solid droplets of analyte and flame radicals have been studied for Cu, Fe, Na and Ca (1499) and have been used to explain variations of Sn behaviour in different flames (617, 831).Part I: Fundamentals and Instrumentation 15 Lareral diffusion interference (629, 1473) was found for Al, Mo and V at several lines in N,O/C,H, flames on slot burners.Concomitants delay the release of these elements by forming refractory compounds so that when the atoms are finally liberated, their position in the flame will be determined by the relative rates of diffusion of the refractory solid and the atomic vapour. Experimental data shows that whenever signal enhancements occur in the centre of the flame there is a depression at the edge.Uneven distribution of aerosols in flames has been observed by Newman and Page (324) and diffusion coeficients for Li, K, Rb and Cs in O2/N2/Hz flames (1624) have been measured. Spectral interferences of Fe 213.856 nm (326), V 250.6905 nm and Si 250.6899 nm (1472) and Ne 359.253 nm with Cr 359.349 nm (1040) have been reported.In this last case Ne radiation was used both to excite A F and as a source for AAS, in the determination of Cr. Quantitative theoretical studies on physical interference continue to be made by Soviet workers (259, 351, 442, 1148 and ARAAS, 1972, 2, 18), who have investigated the effects of acids on such parameters as viscosity, surface tension, droplet diameter, heat exchange coefficient and uptake rate.Correlations between enhancement and physical properties of clrganic solvents were described (624, 1123, 1167), together with viscosity variations with elemental concentrations (884) over wide ranges of calibration. Finally, atomic fluorescence researchers continued to produce theoretical evidence that AFS is more sensitive than AAS (487, 987).A very comprehensive theory on shapes of growth curves (9) for AFS is recommended reading. Omenetto and Winefordner (13) list 14 types of AF transition together with diagrams and captions to describe them - it is hoped that their suggestions are followed so that the literature is spared a proliferation of confusing nomenclature. Considerable improvements in signal/scatter ratio were obtained (1 6) by the use of polarizers for AFS in a paper dealing with light scattering in turbulent and laminar flames.The scattering problems peculiar to non-dispersive AFS and methods for correction \.;ere reported by Larkins and Willis (530). Some other interesting applications of physics to AF included a review (1 124) which described double resonance, level crossing techniques and the Hanle effect.Magnetic depolarisation of A F in flames with fields of up to 2000 G (67) and collision-induced AF (1077) are reports for which we have little additional informa- tion. Incoherent resonance fluorescence (1130) produced by Rb atoms excited with a coherent rptical pulse was used to measure dipole moments, relaxation times and life-times. 2.6 NON-FLAME CELLS Non-flame devices can no longer be regarded as a new experimental technique since they are now in routine use for a large number of applications. The sales of commercial non-flame cells continue to increase at a rapid rate, as do the number of reports and papers from the manufacturers of these devices. The main advantages of this type of technique (214) are ,:) minimum or no sample pre-treatment (thus reducing the risk of contamination), (i9 small sample volume (biological applications) and (iii) impressive detection limits (although it should be stressed that, when studying manufacturers’ quoted limits of detection, e.g., Zn 0.5 ng 1-I, practical considerations, such as contamination, may often limit the detection limit obtainable in practice).The main disadvantage is that somewhat more operator skill is A rather heated discussion on the relative merits of graphite tube and rod devices has taken place (130), with neither side being prepared to concede the superiority of the other. Koirtyohann (1 505) has made a critical comparison of the Woodriff tube furnace, Massmann tube furnace and Varian miniature graphite furnace.A number of reviews covering various aspects of non-flame atomizers have also been reported (197, 199, 219, 612, 651, 658, 791). The use of pyrolytic graphite coatings to reduce the graphite porosity and to minimise carbide formation for certain elements has been known for some time, but a limitation of these coatings is their short life-time when determining such elements as Al, Mo and V, 11 hich require high temperatures and long atomization times.Morrow and McElhaney (1488) cquired to obtain reliable results for non-flame measurements.16 Part I: Fundamentals and Instrumentatioii have reported a method of continuously forming such coatings by adding CH, to the Ar gas supply that surrounds the heated graphite device. The pyrolysis of the CH, during the atomization period results in the formation of a type of pyrolytic coating which was found to increase substantially the life-time and precision of such devices.Another method whereby considerable gain in precision has been obtained in non-flame AA is by using optically and mechanically controlled sample introduction (1 503) (see ARAAS, 1972, 2, ref. 385).Comparatively little theoretical work on electrothermal non-flame atomization, especially with respect to inter-element effects, has been reported to date. A systematic classification of inter-element effects, vix., vaporization of analyte as molecules, molecular background absorption, scattering of source radiation, gas-phase interaction of atoms with matrix components, losses during dry ashing, carbide formation, etc., would prove extremely valuable.Computer control of non-flame AA and A F spectrometers has been reported (1295) and a comparison of lock-in amplification and photon counting has been made with respect to AF in conjunction with graphite tube and rod atomizers (627). 2.6.1 Graphite Rod Devices Resistively heated graphite rod atomizers have a relatively low power consumption (1-2 kVA) and represent the simplest form of the graphite non-flame atomizer.Simple graphite-rod systems have been described (15, 312, 316, 368, 652, 968, 1035), some of which have been left in situ during flame measurements (368, 652). Inter-element effects were found to be more severe for a low boiling point element such as Zn than for an involatile element such as Mo (1035), although this is to be expected, as most other elements will be vaporized before the Mo.The addition of Ca (as the nitrate) to solutions of Al, Ba and Si was found to enhance the signal for these elements 4-6 times (652), and this is probably due to preferential carbide formation by the Ca. The decay of the atom population at varying heights above a graphite rod (185, 1502) and computerised measurements of transient atom populations (1038) have been studied.The use of an Ar/H2 diffusion flame above a rod (185) was found to increase markedly the height above the rod where atoms could be detected, whereas the use of the Ar/H, diffusion flame using limited-field viewing (ARAAS, 1972, 2, 56, refs. 8, 9) was found to cause a decrease in the signal.The influence of heating rate on the analytical response has been studied (637, 761) with theoretical and observed results agreeing. The temperature above a graphite rod (2350'K) was found (301) to be somewhat less than that inside a miniature furnace (2770'K) at the time when the analytical signal for Ga was at a maximum. It should be stressed that the temperature above a graphite rod is very dependent on the height of measurement and that analytical measurements are usually performed under limited-field viewing conditions.The use of A F techniques in conjunction with graphite-rod atomization shows considerable potential, and has been used for the simultaneous AF determination of Ag and Cu in jet- engine oils (316). An approach (201) using continuous sample introduction into a slot in the graphite rod together with photon-counting techniques was found to be very sensitive and could be used in conjunction with a single-continuum excitation source. 2.6.2 Graphite Miniature Furnace Configurations This type of system, pioneered by Varian Techtron (6, 520, 521, 648, 792), can be considered to be intermediate between the West-type rod and the Massmann-type tube. The necessary adjustments required for fitting the Varian Techtron Mini-Massmann CRA 61 into a Unicam SP90A spectrometer have been studied and described as rather difficult (610).A modification has been made to the graphite cylinder to allow a more rapid rise in temperature and a higher final temperature in order to minimise memory effects and to increase the sensitivity for Al, Mo and V (502, 1481).Part I: Fuiidamentals and Instrumentation 17 It has been demonstrated (521, 648) that the addition of excess H31?04 or HNOJ often reduced molecular absorption and loss of volatile compounds, and that the use of a pyrolytic graphite coating considerably reduced the tendency to form carbides. Matousek (520, 792) used two Ni absorption lines to calculate the variation in the excitation temperature inside the graphite cylinder during the atomization of a Ni sample.For most of the atomization period, this calculated excitation temperature was about 3OOOC higher than that registered by a radiation thermometer, which viewed the inside of the cvlinder. Soviet workers (488, 793) have found that direct impulse vaporization from graphite crucibles using solid samples often gave better detection limits with AFS than AAS.A comprehensive theoretical treatment leading to this result has also been given (793). 2.6.3 Graphite Tube Furnaces This section covers resistively heated Massmann-type graphite-tube devices, a subject on which Massman himself has given some comprehensive reviews this year (651, 658, 791).Delves (217) has described the design of an inexpensive graphite-tube furnace, and Pickford and Rossi have improved their automated sampling system for high-purity water analysis (see ARAAS, 1972, 2, ref. 385) by using signal integration with non-resonance line background correction (476). The direct analysis of solid samples has given some encouraging results (808, 810, 1140) and the use of urea-based solid standards, instead of metallic standards, has been recom- mended as a way round the difficult calibration problem (808).A novel approach (760) to the analysis of metallic samples has used the energy from a pulsed lascr beam to vaporize a small sample from a massive piece of metal on to the inside surface of a small graphite tube.Inter-element effects and methods for their elimination have been studied for a large iiclmber of elements in a variety of matrices (795). The choice of dry ashing temperature is critical; some workers (1491) have found that pre-atomization losses can sometimes be significant at temperatures recommended by the manufacturers for the analysis of a given metal. I t has been shown that losses of Cu and Ni can occur from a chloride matrix at temperatures above 6 O O O C (300) and that losses of Cu, Mn and Ni occurred from all matrices at temperatures above 1 100°C (720).m e problem of carbide formation has been minimised by coating the inside surface of the tube with an organic Nb or T a compound (1218) which is then thermally decomposed, resulting in a very involatile carbide film being formed on the graphite tube.There is still a problem of compensating for high levels of non-specific background absorption (524, 674, 1034, 1184, 1511). Massmann et al. (651, 658, 791) have plotted the molecular absorption spectra of many species (NaBr, NaCl, MgO, SO,, SO,, etc.) within rhe graphite tube, and these consisted mainly of broad absorption bands.They also observed molecular absorption by CN and OH radicals. The complicated fine structure of the absorp- tion spectra of these species prevents the correct functioning of a normal deuterium background correction system over the wavelength regions that these species absorb. It has been shown (1507) that the optimum graphite-tube length is achieved when the median residence time of the atoms in the tube is equal to, or slightly greater than, the median time of atomization. Atomization mechanisms are a very important consideration (622, 674, 791, 794, 799, and it is generally agreed that it is important to keep the element of interest in a refractory form for as long as possible in order to effect a separation from the matrix and to ensure that adequate atomization occurs rather than volatilization of undissociated molecules.Ottaway et al. (1512) have produced evidence to support the generally held hypothesis that atomization of oxy-acid salts proceeds via the oxide, followed by carbon reduction of this oxide. Various improvements and modifications to the graphite furnace system have been reported: (a) the use of a grooved graphite tube (124) or tantalum liner (218) to prevent18 Part I: Fundamentals and Instrumelitation spreading of solutions with low surface tensions; (b) making openings in the graphite tube to allow the inert shielding gas to pass from the outside to the inside of the tube (558) to increase the tube lifetime; (c) modification of the gas-flow system (1511) to improve dry ashing and atomization; (d) reduction of the wall thickness of the graphite tube from the middle towards the ends to give more uniform heating (280, 908, 1286); ( e ) modification of the instrument optics, magnetic shielding of the photomultiplier, and improving electronic regulation in order to minimise the effect of black-body emission from the walls of the graphite tube (540, 807, 1286); (f) improving the sample-introduction crucible for the Woodriff furnace (1454, 1461) to prevent sample leakage; ( g ) use of a method to facilitate the introduction of solid samples (1140); (h) use of an improved power-unit programmer (1 139); (9 vapour-phase sample injection (1510) direct from a gas chromatographic column, so permitting the determination of volatile metallo-organic compounds; (j) use of a heating rate independent of the final pre-selected atomization temperature (15 13), which made possible the determination of Cd in a 2% w/v solution of NaCl; (A) the use of LiF windows and a vacuum monochromator (1541) to permit analytical measurements to be made at the 148.9 nm Br resonance line, and (I) the fitting of a graphite furnace to a Unicam SP1900 (945) and the conversion of a Varian Techtron Model 61 graphite-rod atomizer head to accept a graphite tube (939). 2.6.4 Metal Filament Devices These devices have low power consumption and are relatively simple, tantalum being the most commonly used material. Cantle and West (255) have advocated the use of a tungsten rod (60 mm long, 2.2 mm diameter) and have compared it with the graphite-rod atomizer, while others have used tungsten-rhenium (589, 1150) and platinum-rhodium (936) wire-loop atomizers.A graphite-rod mounting device and power unit that would accept tantalum strips, a miniature tantalum furnace and various types of graphite rods have been reported (652). A detailed study of inter-element effects, and methods for their elimination, has been made (295, 405, 795, 955, 1443, 1474) for a large number of elements in a variety of matrices using tantalum filament atomizers.Ionization (1474) of the alkali metals was found to be much less than that observed in a flame. 2.6.5 RF. Induction Heated Furnaces These devices utilise induction instead of resistive heating of a graphite tube or cup and have been advocated for some time by Morrison and Talmi (41, 1312 and ARAAS, 1972, 2, 20).The method has certain advantages over most resistively heated furnaces, e.g., a larger amount of sample can be added to the furnace, which is especially important for solid samples; the signal duration is somewhat greater; by using part of the r.f. energy to excite the atomized sample, emission as well as absorption measurements can be made.The main disadvantage is the increased complexity of the apparatus. The failure of a combination of an r.f.-heated carbon bed (see ARAAS, 1971, 1, ref. 834) and AF for the determination of Cd and Hg in air was attributed to quenching of the fluorescence radiation by CO and Nz in the atomized sample (1154). 2.6.6 Cells for Mercury Determination A brief historical survey of Hg determination by flameless atomic spectroscopy has already been given (ARAAS, 1971, 1, 20).Nearly all manufacturers of AA instruments retail accessory kits for the cold-vapour determination of Hg. The only significant development reported in the instrumentation of this type of technique was by Rains and Menis (344), who heated the absorption cell to 2OOOC in order to eliminate the need for a drying column.Surprisingly, Ar gave 25% higher absorption values than air under these conditions. Improvements to a Perkin-Elmer Flameless Mercury System (363) and an improved reduction-cell design (1 152) have been reported. A rather elaborate system utilising a copperPart I : Fundamentals and Instrumentation 19 wire loop heated by capacitator discharge (587) has been used to atomize liquid Hg samples.A rapid method for the direct analysis of solid samples for Hg has been reported (1305) whereby the sample is pyrolised in an Ar atmosphere and the Hg in the resulting gas stream is measured by AF. The quenching of excited-state Hg atoms caused by the molecular pyrolysis products is compensated for by splitting the pyrolysed sample-gas stream and directing one of the streams over PdCL to remove any Hg and then into a reference channel containing a constant bleed of Hg vapour.The magnitude of the quenching caused hv the organic breakdown products of pyrolysis can then be established and compensation made. The procedure as described will not correct for scattering interference. 2.7 HYDRIDE GENERATION TECHNIQUES The generation of hydrides of As, Bi, Ge, Sb, Se, Sn, and T e has been used to improve the limits of detection compared with conventional flame AA by approximately two orders of magnitude, Sodium borohydride is a more convenient reducing agent (536, 5SS) than the more commonly reported systems based on zinc (33, 36, 51, 207, 606, 1021, 1161).In most cases the liberated hydrides are carried in a stream of Ar to an Ar/H2/entrained-air flame; however, a better method (606), resulting in a considerably improved signal-to-noise ratio, uses an electrically heated silica tube instead of a flame to effect atomization.Christian (1443) isolated the hydrides in a liquid-nitrogen cold trap and then, by rapidly warming the cold trap, passed the liberated hydrides directly to a graphite-tube furnace. 2.8 OTHER EXCITATION AND ATOMIZING SYSTEMS Several interesting papers have been noted that do not fit into the classification of Sections 2.1-2.7. Some of these (66, 566, 673, 1120) deal with kinetic studies of excited atoms using AA methods to monitor the excited-atom population and are considered to be outsidc the terms of reference of this report. The analysis of metals and alloys by cathodic sputtering (531, 934) using either AA or AF appears to be a promising technique for both trace and macro analysis. Precisions of the order of 2 1% have been obtained at high element concentrations with detection limits in the 10-100 bg g-' region for many elements. The possibility of using the technique to determine C, P and S by AF using a solar-blind photomultiplier responding in the wave- length region 160-200 nm was also postulated (934). The radiative properties of exploding silver wires that have previously been electroplated with various other elements (from a dilute solution) have been examined (1, 1435). The resulting atomic emission could be used to detect nanogram quantities of these elements. A study of the physical processes occurring in these discharges has also been made (1188). An interesting method of atomization and excitation using a shock wave has been described (1477). A 100 bl aqueous sample was placed on a 12.5 X 47 mm strip of membrane filter-paper, dried and mounted in a shock tube near the end wall and observation window. The incident shock wave vaporized the sample and swept the resulting vapour into the observation zone, where it was excited by the reflected shock wave. Helium-driven argon shocks were said to give reflected-wave temperatures of 10,50O0K for Mach 6.8 incident shock waves. Calibration curves for Cd and Ni were linear over three decades of concentration. A corona discharge has been proposed ( 5 ) as a sampling system for AAS in place of a flame. A patent has been granted (103) for a system that uses an r.f. pulse to atomize a sample vapour, while a separate intense pulsed-light source excites AF from this vapour. The detector circuit was synchronised to produce output signals representing radiation emitted by the sample between the periods of the r.f. atomizing pulses. Various types of furnace have been used for anomalous dispersion measurements (272); determining the absorption spectrum of In1 (1239); the determination of the dissociation energy of aluminium carbide (761); spectroscopic studies over the wavelength range 120-190 nm (548) and for the study of diatomic molecules (736).

ISSN:0306-1353    

DOI:10.1039/AA9730300007

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

20 Port I: Fundamentals arid Instrumerttatioii 3 Optics Very few developments of general applicability have been reported. I n most cases authors have either described new applications, variations of established instrumentation or have yroduced a solution to their own individual problem, so in view of the limited demand for many of these modifications, their incorporation into instruments at the present time is Linlikely. 3.1 OPTICAL COMPONENTS Tn a study of the properties of materials used as windows for the vacuum ultraviolet, it has bee;: reported (547) that Suprasil W irradiated with wavelengths below 190 nm develops an ,;Ssorption band at 260 nm, which can, however, be bleached by irradiation at 250 nm. lieryllium oxide has been shown to be 90% transparent between 200-700 nm, 50% at 153 nm and 10% at about 132 nm, for a window thickness of 1 mm (230).With the increasing interest in measurements in the far ultraviolet, the possibility of fluorescence and light rcatter 'ny optical components should always be borne in mind. At the present time little use is made of fibre optics in instruments for spectrochemical analysis, as generally mirrors and lenses are found to be adequate components for manip- dating lizht beams.However, a multichannel AA instrument employing fibre optics has been described (see Section 6.2, refs. 143, 1448) and a water-cooled fibre-optic probe has been llsed to analyse the signal from a small Na flame (see also ARAAS, 1972, 2, 23). T o reduce the error, resulting from variation in the atom distribution in the flame causcd by differences between standards and samples in AE and AA, an optical system has k e n used (1109) to rotate the flame image through 90° and project it on to the entrance slit of the monochromator. 3.2 BEAM MANIPULATION 3.2.1 Modulation Most AA spectrometers modulate the emission from the light source at frequencies of a few hundred hertz either electronically or mechanically. Studies of the power spectrum of flame noise show it to be approximately inversely proportional to frequency, so that the higher the modulition frequency, the better the discrimination against this noise.In practice, however, there appears to be little gain in operating at frequencies greater than those in common usage. Wenthen and Snowman (249) have nevertheless considered the problem of the design of high-speed rotating light choppers and given details of a 10 kHz chopper.They compute that the maximum centrifugal stress on the hub was 3690 p.s.i. and that on the beryllium- copper ring, in the area of stress due to the slits, was 34,000 p.s.i. 3.2.2 Background Correction With the advent of the flameless atomizer, there has arisen an increasing demand for back- ground correction of the non-specific absorption and scatter generated by the matrix material.Most commonly this correction is effected in the spectral range 230 to 400 nm by combining the light from the spectral line source with that from a continuum such as a 150 W xenon arc or deuterium lamp (1205) with the two light paths as nearly identical as possible.This beam combination may be effected by attaching mirrors to the conventional rotating chopper (8XS, 1306) or by using a semi-silvered mirror and a vibrating quartz plate placed in the exit slit of the monochromator to provide modulation of the resonance line signal by short- range wavelength scanning (22). The vibrating quartz plate method of wavelength scanning has been used to give increased latitude in the alignment of a monochromator and to facilitate computer control of the wavelength setting (17) and for background correction in FES (203).I n all cases electronic demodulation is necessary to isolate the emission, absorption or fluorescence signal. Simul- taneous background correction (266, 812) in AA may also be obtained by monitoring thePart I : Fundamentals and Instrumentation 21 intensity of both the resonance line and a nearby non-resonance line (see also Section 6.2).Whichever method of background correction is employed, care must be taken to ensure no artefacts arise due to differences between the wavelength and bandwidth of the correctim facility and those of the analytical line (see Section 2.6.3). 3.3 WAVELENGTH SELECTION 3.3.1 Dispersive Systems In a review of diffraction gratings (946) Harrison concludes that in general the diffraction grating is more useful than either the interferometer or holographic grating. An excellent handbook on both ruled and holographic diffraction gratings is available from Jobin Yvon Irlc.Attempts to design improved diffraction gratings continue, and in one approach a rectangular corrugation profile has been shown to exhibit a surprisingly large blaze effect 19.17).This was attributed to a half-wave resonance in the corrugation region, and up to 100% power conversion can take place between the incident wave and the single-diffracted spectral order. An alternative approach (550) has been the use of sensitised polymethylmetha- crylate, which exhibits an increase in refractive index of the order of after irradiation with ultraviolet light a t 325 nm.Exposures of the order of 50-500 J cm-* were required to generate a holographic grating which had a line resolution of approximately 5000 lines mm-‘. Several papers have dealt with the practical aspects of using grating spectrometers. A yc-llow glass filter placed in front of the appropriate area of the photographic plate in a specrrograph permitted the second orders from 240-340 nm and first orders from 400-880 nm to be simultaneously recorded (20), which permitted 46 elements to be determined in a single analysis.By incorporating several collimating mirrors in a spectrograph (863) it is possible, using a single exposure, to cover a wide spectral range without loss in spectrum quality.It is well known that improving the resolution of the spectrograph will lower the detection limit by improving the line-to-background ratio. This improvement has been demonstrated by changing from the first-order diffracted spectrum to the fourth (863). The working range of grating spectrometers has been extended by setting the grating on zero order to function as a inirror (see Section 6.2, refs, 226, 522). The reciprocal dispersion at grazing incidence of a vacuum spectrograph varies appreciably over relatively short wavelength intervals.Goto, Gi:atam and Joshi (84) have derived a reduction formula which, when applied to measure- mcnts in the 10-20 nm spectral range, gave good agreement with literature values. Several papers (190, 454, 603, 737) have described applications of motor-drivzn spectrum-scanning monochromators.I n one (454), using a scanning exit slit, the signal generated by the spectral line on the photomultiplier was used to ‘gate’ the signal into the correct measuring channel. In another (190), by rotation of the diffraction grating, the scan rate was higher over unwanted regions of the spectrum but could be halted at pre-selected wavelengths with an accuracy of 0.02 nm.Scanning by rotation of the grating was also used by Norris and West (603) for the sequential measurement of four elements by AFS. In the third system (737) light entering the spectrograph was chopped mechanically and a reference waveform generated, with the signals from the spectral line and the reference waveform being fed to phase-sensitive detectors.The spectrum was scanned by a moving exit slit, and it was claimed that spectral line profiles of high quality were obtained. Cresscr and co-workers (178, 655, 1283, 1438) have extended their work on echelle monochromators in flame AE and AA analysis (see also ARAAS, 1972, 2, 24). The possible applications of echelle instruments have been reviewed by Matz (1022). 3.3.2 Interferometers Most work to be reported in this section relates to minor developments in the design and use of Fabry-Perot interferometers. To stabilise the parallelism of the flats of a Fabry-Perot interferometer, a re€erence wavelength was used to generate a correcting signal proportional22 Part I : Fundarneiztals and Instrumentation to the displacement of the fringe maximum of three auxiliary spectral sources with respect to a reference point (236).These signals were fed back to the interferometer piezo-electric drive which re-orientated the plates The stability of the finished instrument, over a 6 h period, v:.:as near to h/1000 zt 546 nm. By combining an interference filter with a fixed-spacing Fabry-Perot interferometer, an instrument was produced (238) which, operating in the region 280-395 nm, had an aperture of up to 30 mm with a transmission of 38%, a bandwidth at half-height of 0.06 nm and a free spectral range of 2 nm.The use of spherical mirrors in a Fabry-Perot interferometer has the advantage of very high resolution and good light- gathering capacity (241) while, by slight defocussing, almost linear dispersion over a few iqterference orders can be obtained.The mirrors were adjusted and permanently fixed in the required position during the shaping of the spherical surfaces, giving a resolution of 2 X 10‘ with a finesse of 35. Scanning interferometers have been used to study atomic line profiles. A simple low-cost linear scanning system using pressure variation of butane gas has been described (1195).The drive for the Fabry-Perot plate is frequently provided by a piezo-electric crystal, and in one system (239) the O-ring has been replaced by a nylon ring to give an improved performance with a 2.5 nm free spectral range scan in 1 ps. Kirkbright and Troccoli (715, 809) with a 150 \Y xenon arc lamp used a piezoelectric scanning Fabry-Perot interferometer to study atomic line widths in air/CZH2, NZQ/C2H2 and Ar-entrained air/H, flames.A scan speed of 15 Hz was used and a Techtron AA4 monochromator employed as the detection system. Other workers (68, 937) have described the application of an interferometer to AA using a continuum source, and found for absorption lines in the 320-360 nm spectral region that sensitivities were of the same order as line sources, but detection limits were worse due to lower signal-to-noise levels.I33 tes et al. (1 192) constructed a high-resolution photoelectric scanning spectrometer employing a Fabry-Perot interferometer in tandem with an echelle-grating monochromator for use in astronomical spectroscopy of the middle ultraviolet.Two papers have described instruments based on the Michelson interferometer. These instruments have replaced the mirrors of the classical design by diffraction gratings (242, 217). It is difficult to make comparisons between these instruments and the Fabry-Perot system but, generally, the interferogram is more difficult to invert to produce the spectrum and it is not likely that these interferometers will be able to contribute significantly in analytical atomic spectroscopy. 3.3.3 Non-dispersive Systems By ‘doping’ LaF, with Ce, filters have been made with transmission bands at 176 nm (237), with bandwidths and peak transmission, which were concentration and temperature dependent, of the order of 20 nm and 40%, respectively, at room temperature.The position of the peak transmission could be varied by change of host material. Coloured glass filters have been shown (245) to exhibit fluorescence when excited by a short-arc sealed-beam xenon lamp, a problem which can also arise with in‘terference filters. Qsantowski and Toft (270) described reflectance interference filters for the 200-300 nm waveband with an efficiency of up to 85% compared with the 20% of the transmission filter.The bandpass at half-height varied from 60 to 120 nm, while reflectances of visible radiation were of the order of 5%. The bandpass of dielectric narrow-lband filters can be reduced by the use of mirrors with increased phase shift dispersion (1356). These filters have a much greater transmission than conventional interference filters for the same bandpass.By oscillating a narrow-band interference filter, a sinusoidal modulation was imposcd on the transmitted intensity of a spectral line whose wavelength was slightly shorter than that of the peak transmission of the filter. This system has has been applied to flame AE (1450). A fully tunable Lyot-Ohman filter manufactured by Zeiss Oberkochen has been described by Beckers (Appl.Opt., 1971, 10, 973) (233). Incorporation of achromatic waveplates into the filterPart I: Fundamentals and Instrumentation 23 and the use of computer control has made it possible to manufacture filters that are rapidly tunable over a wide range (410-710 nm). The bandwidth varies from 0-028-0408 nm with a peak transmission ranging from 2% at 420 nm to 8% at 700 nm. Palermo and Crouch (630, 1282) have carried out a theoretical and experimental evalua- tion of resonance monochromators for AAS. They concluded that, with a line source and resonance monochromator, analytical curves were relatively independent of source broadening, while the n a r m bandpass of the resonance monochromator led to improved sensitivity of AA measurements with a continuum source. Non-dispersive wavelength selection should be of greatest benefit where signal strength is a limitation, such as in AE and AF; nevertheless very little use is made of filter systems. Walsh (597) and others (1524, 1421) have reviewed the use of non-dispersive systems for both AA and AF in the direct analysis of metals and alloys.

ISSN:0306-1353    

DOI:10.1039/AA9730300020

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

24 Part I : Fundamentals and Instrumentation 4 Detector Systems 4.1 VACUUM PHOTOTUBES 4.1.1 Photocathodes Burton and Powell (234) extended the spectral response of photodetectors into the vacuum ultraviolet by coating the envelope with fluorescent materials. The coatings, tetraphenyl bl_ieadiene and sodium salicylate, were prepared by vacuum deposition or solvent spray, and a spatial resolution of up to 20 lines per millimetre was achieved.The tetraphenyl butadiene fluorescence emission spectrum ranges from 400-500 nm with a peak intensity at 430 nm; its fluorescence efficiency between 90-380 nm was 2 or 3 times greater than that of sodium salicylate. Coating the cathode of a channel electron multiplier (243) with a depth of 300 nm of MgFz produced no significant deterioration in the electrical properties but increased the sensitivity by factors from 1.6 to 2.8 at wavelengths in the region of 50 nm.The temporal stability of alkali metal photocathodes and the effect of high light intensity on them has been discussed by Sommer (235). In a more detailed study of S10 photocathodes, Budde (948) concluded that over a period of three years the shape of the spectral response curve of a given tube could vary such that relative sensitivities from one part of the spectrum compared with another could change by up to 60%.A photomultiplier with a GaAs photocathode was recommended for the determination of K in rocks (1588). 4.1.2 Photocells Fisher et al. (248) have described a highly stable vacuum photo-emissive diode which is uniform in its response and may be used as a standard for the accurate measurement of monochromatic ultraviolet light (110-300 nm).The cell, with a window of MgFJ or LIE;, uses an opaque photocathode of Cs2Te which is blind to visible light (>360 nm) and has a quantum efficiency of 10%. 4.1.3 Photomuldpliers Kerns et al. (1198) investigated the influence of biassing circuits on the noise-power spectral density of the output of photomultiplier tubes.They found the noise power was approximately inversely proportional to frequency and attributed this to the shunting effect of inter- electrode impedance and gain fluctuation, caused by noise voltage across the biassing resistances. A method for flattening the noise spectra and improving the frequency response of the tube was described.Other workers (1241) have developed a system for testing the linearity of response of photomultipliers used in flash photolysis with a view to optimising operating conditions. Two circuits for ‘gating’ photomultipliers have been described. One of these (1347) permits differential and direct light intensity measurements to be made over 4-6 orders of magnitude with a time resolution between 0.03 and 40 us; the other (1240) has a switching time of about 5 ns with very small noise transients.Prydz (1346) has suggested the construction of a photomultiplier in which the amplitude of the single photon output pulse is a function of the part of the photocathode struck by the photon. In combination with a pulse-height analyser, the system would constitute a photon- counting one-dimensional detector.In view of the relatively wide distribution of pulse amplitudes from a conventional photomultiplier it seems unlikely that the proposed system can he of high spatial resolution. 4.1.4 Imaging Tubes Several groups of workers have continued the evaluation of television camera tubes for their suitability for multi-element analysis.The limitation of the system at the present time is the small diameter of the tube, which determines the spectral range received at one time; the resolution is - 1/500 of the tube diameter. Vidicon tubes have been successfully used inPart I: Fundamentals and Znstrumentation 25 FES by Busch, Morrison and Feldman (1329, 1338, 1496) and Zakharov et al. (1656) and Aldous, Mitchell and Jackson for up to eight elements by AA (1285, 1340, 1452).A commercial optical multichannel analyzer (1341) was claimed to have a spectral response from 185 nm to 800 nm. This system could also be linked to an image intensifier tube. A 'Spectracon' image intensifier tube (1197) was fitted with a thin mica window to permit transmission of 75-80% of the incident photo-electrons which were generated in the photo- cathode and accelerated through 40 kV before striking the window.A nuclear-track emulsion was mounted outside the window to record the image. It is claimed that this system is more linear, with an improved signal-to-noise ratio, than photographic recording of low-contrast images in astronomy. Gagne et al. (549) viewed the output of the Fabry-Perot interferometer with a television camefa; one video line was isolated from the display and presented on a cathode ray tube to show the intensity distribution across the interference pattern.The instrument can be used for the study of fast spectroscopic phenomena. 4.2 SOLID STATE DETECTORS Interest has centred on the production and performance of silicon photodiodes. Wang and Li (1199) have published details of rhe manufacture of a grating-type Au n-type silicon Schottky barrier photodiode which had a quantum yield and spectral response superior to other types of silicon photodiodes.The diode had a frequency response up to - 1 GHz and a sensitivity of 0.63 A W-* at 900. nm. I n a review of the application of silicon photodiodes to the measurement of short laser pulses (1193), a figure of merit defined as the ratio of the maximum linear current to dark current was proposed; the desired value of this parameter was lo'', but the practical limit was 2 X lo7.Another review (1238) comparing the merits of the silicon photodiode with the photomultiplier claimed that the former was smaller and had greater stability with a wider dynamic range.Two groups of workers have examined the usefulness of silicon photodiode arrays as detectors of spectral information. Horlick and Codding (628) mounted 256 diodes, each 25.4 pm wide, in the focal plane of a monochromator, and used them as integrating devices. Although non-linearity due to saturation became evident after long integration times (greater than 800 ms) the linear response range was found to extend over 3.5 orders of magnitude at 632.8 nm, with the resolution of the array approximately 0.25 nm.The application of cross-correlation techniques to the system were also investigated by these authors (635). Boumans et al. (816) connected their diodes directly to amplifiers and also obtained linear responses over approximately three orders of magnitude with a spectral resolution of the order of 3.5 ram.They found that the signal-to-noise ratio at low intensity levels was more than two orders of magnitude lower than that of photomultipliers. In comparison with vidicon-based systems, a diode array could be made large enough to view a greater spectral range but, at present, neither system nor an individual diode was as good as a slit/photo- multiplier combination for the measurement of weak spectral lines.A solar-blind photoresistor (984) has been used in a dispersionless system for the determination of Cu and Ag by flame AF. 4.3 THE PHOTOGRAPHIC PLATE As the photographic plate has been in use in spectroscopy for many years, it is inevitable that there is re-discovery of old knowledge due to the inadequate transfer of accumulated knowledge from one generation to the next.Improvements 5n technology, however, do frequently justify re-examination of old problems. For example, Torok and co-workers (1052, 1061), from a careful examination of the effect of conditions of plate development on the precision of spectral information recorded, concluded that gas agitation gave the most uniform development with higher contrast and minimum of developer.The blackening function of photographic emulsions has been evaluated (860) on the basis of the following assumptions: (1) the emulsion consists of layers, (2) the absorption cross-26 Part I : Fundamentals and Instrumentation section of the Ag halide grain satisfies a normal distribution law, (3) the blackening thresholds of grains obey a log-normal distribution law, (4) the light quanta hitting the grain satisfies Poisson statistics and ( 5 ) the cross-sections of the developed Ag halide grains are constant multiples of the absorption cross-section of the Ag halide grain.The resulting formulae approximate to the P transformation. It was shown that the average threshold was about 3 to 15 quanta per grain for most emulsions.Theoretical aspects of the photographic imaging process have also been studied by Shaw (Opt. Acta, 1973, 20, 749). An attempt to extend the spectral response of a photographic system into the infra-red employs an inter- mediate image retaining panel w?th a response up to 1.5 cLm and resolution of 8 line-pairs per mm (1196).After exposure the image is transferred to a panchromatic plate. In microdensitometry, a non-linearity arises which is a function of spatial frequency and the mode of illumination of the photographic plate (271). This effect can be overcome if imaging of the photographic plate is avoided and all transmitted light is collected. An instrument is described which uses a continuous-wave laser source and standard microscope objective, with the system’s resolution determined by the size of the scanning light spot.Ivanov and Talalaev (156) have used photographic recordings for studies in AA analysis and, in spite of lower sensitivity, have found it useful for providing fuller information in relation to multi-element analysis and inhomogeneity within the flame. 4.4 SIGNAL PROCESSING In this section no new work has been reported. Some refinement of technique and more detailed understanding of basic processes has taken place but much is repetition of previously reported work. 4.4.1 Photon Counting Murphy et al. (627) compared pulse counting with phase-sensitive amplification for atomic spectrometry employing low-background atomization systems (Ar/02/H2 flame, slotted graphite rod, graphite tube with continuous and discrete sample introduction).The phioton- counting system was found to be superior for low-intensity light sources and approximately equivalent for high-intensity sources. A theoretical and experimental comparison of photon counting with current measurement by Hawes (883) concluded that the signal-to-noise ratio, for a good counting system, shows an improvement over that for a good analog system equal to the predicted values.The improvement factor lies between 1 and 1.6 and arises principally from the more efficient use of the information generated in the system. In a study of the efficiency of the RCA C3100QF photomultiplier the quantum efficiency of the cathode at 633 nm was 5.5% in normal use but only 3.3% in the photon-counting mode (246).The difference was due to small pulses well below the single electron peak in the output pulse- height spectrum. These pulses were usually below the discriminator threshold level and not counted. 4.4.2 Modulation and Correlation Techniques In one system (25) a rapid short-wavelength repetitive scan was superimposed on a continuous wavelength scan and the resultant signal passed through a synchronous demodulating circuit.This technique, which has been in use in one form or another for several years, provides automatic background subtraction in the measurement of line spectra. In AF measurements, signal modulation can be achieved by pulsing the output of the HCL (175, 1310). By separating the output of the photodetector into two channels (175), one to measure the background alone and the other the fluorescence signal plus background, the net fluorescence signal can be determined by difference.There was an improvement in sensitivity of one to three orders of magnitude compared with that of d.c. operation. To obtain improved stability in AA measurements, a dual wavelength system has been devised (3) using a null-ratio method in which the signals from two selected wavelengths of the atomic line source are dynamically compared in a phase-locked photometric amplifier.Part I : Fundamentals and Znstrumentation 27 Hieftje and co-workers (184, 187) have studied the application of source modulation on signal-to-noise ratio of AA and AF and found that the waveform used had a significact effect on the detection efficiency, signal-to-noise ratio and practical utility of modulation in AA.Square-wave modulation in which the HCL was never completely turned off, provided greater stability than that given by an on/off system and permitted greater signal detection efficiency than a sinusoidal waveform. In the determination of Rh by AF (1871, signal processing by cross-correlation improved the signal-to-noise ratio five-fold, compared with an unoptimised system.Horlick (188, 635) has applied cross-correlation techniques to the examination of emission spectra from samples of unknown composition by comparison with spectra from samples of known composition. It has been claimed (170) that detection limits for Rb in rocks by FES can be improved by several orders of magnitude by a choice of appropriate photomultiplier, dynode-potential distribution, high-voltage supply, amplifier, monochrom'ator and atomizer burner, but the biggest contribution was by the use of a lock-in amplifier.When Fourier spectroscopy (269) was used for the study of dense spectra and the system performance was limited by photon noise, there was no advantage over conventional scanning spectroscopy arising from its multiplex facility, but there was a signal-to-noise improvement at peaks in the spectrum. I t is likely that this condition is also true of Hadamard transform spectroscopy. 4.4.3 Electronic Circuits Eppeldauer (244) described a simple operational amplifier circuit for use wi,th photo-voltaic cells, which was claimed to give high sensitivity and accurate photocurrent measurements. Further papers have described circuits for the conversion of the linear absorption signals generated in AA into logarithmic absorbance readings (204, 1284). In both absorption and emission spectroscopy, when working close to the detection limits, signal integration leads to improved precision.

ISSN:0306-1353    

DOI:10.1039/AA9730300024

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

28 Part Z: Fundamentals and Znsttumentation 5 Data Processing 5.1 EMISSION SPECTROSCOPY It is suggested (862, 872) that a fast on-line computer which is programmed to view the conventional integration of a single excitation period as a series of shorter sub-excitations would permit a statistical evaluation of the population. If there exists a correlation of the fluctuations of the corresponding partial values of different channels, a suitable compensation will then result in a new derived single value having a lower standard deviation than that of the original one.Also systematic errors with a corresponding shift in the total mean value may be indicated by an abnormal distribution. Analytical examples have been given, and the experimental modifications described (862, 872), to show that by suitable programming, all the calculations can be made simulraneously with the excitation of the sample, and logical decisions taken. The use of a computer programme in connection with a matrix correction procedure, for use when available standards are not closely similar to the unknown in composition, has been described (360).A procedure was suggested using a family of error curves, for correcting interferences due to changes of atomization or excitation conditions in the spectroscopic source (723).Witmer et d. have described further progress in the automatic reading and computer evaluation of line spectra on photographic plates (857). (See ARAAS, 1971, 1, 28, refs. 600 and 602). 5.2 ABSORPTION SPECTROSCOPY Increases in the speed of analysis using automated preparation (1453), computer-controlled sampling (14) and automated data acquisition (1326, 1327) have been reported.An inexpec- sive miniature computer that can be built from readily available components (585) for on-line use, and the application of a laboratory analog-to-digital computer for data acquisition (578) have been described. Various operating and matrix parameters have been optimised using simple testing programmes (1085, ll46), and ROOS (873) has further extended his treatment of error functions in AAS.

ISSN:0306-1353    

DOI:10.1039/AA9730300028

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

Part I : Fundamentals and Instrumentation 29 6 Complete Instruments information on current commercially available instruments has been updated from thzt supplied in ARAAS, volume 2; we are again indebted to the instrument manufacturers for their co-operation. The format used for the presentation of information on AA spectrometers in Table B has been changed to allow easier comparison of the various instruments.The new format now resembles the presentation used for emission spectrometers in Table A. The increased use of flameless AAS is reflected in the introduction of a new table describing commercially available non-flame atomizers, Table C. The information available in these tables is believed to be correct a t February, 1974. 6.1 ARC/SPARK EMISSION INSTRUMENTS Apart from the introduction of several new commercial instruments, there have been no developments reported in this field during 1973.Performance data on some Russian instruments has appeared in the literature; Aidaiw (985) reviewed some simple dispersionless instruments for AA and AF, while Makulov and Zakharova (484) evaluated the performance of the Spektr-1 and the SFPA-4 AA spectro- meters. The Beckman Klina Flame system, the Instrumentation Laboratory Model 143 and the Varian-Techtron Model AA-5 all linked to automatic sampling systems were compared for the determination of Na, K and Li; results were similar for all three instruments (583).Many adaptations to commercially available instruments have been reported. The modifi- cation of an AA spectrometer into a simple emission instrument for the determination of K, by setting the wavelength to zero, has been described (226, 522).By using the grating as a mirror and selecting the wavelength with a narrow band-pass filter in one case and a mono- chromator in the other, the problem of reduced sensitivity at longer wavelengths in the instruments was overcome. Harnly (273) described the adaptation of a spectrometer for the analysis of radioactive samples, and Malmstadt (1311) the automation of an AA spectrometer which presets optimum conditions for each element and permits more than 10 elements in a solution to be determined in less than 1 minute.A flame photometer (1084) has also been described in which errors arising from variations in the background flame composition with time are avoided by modulating the composition of the sample gas stream with a reference gas not containing the component to be determined.A repetitive wavelength scanning tech- nique has been employed (22) to produce a simple pseudo-double beam AA spectrometer which will allow compensation for non-specific absorption. Two papers by separate authors (266, 812) have described the design and construction of a dual-wavelength spectrometer.The instruments used a single fixed grating with movable exit slits, each coupled to a photomultiplier to allow 2 wavelengths to be monitored simul- taneously. This system permitted the simultaneous determination of two elements or the compensation of non-specific absorption. Instruments for multi-element analysis have been described; they employ four different approaches.(i) Flexible ultraviolet fibre optics were used to transmit the light simultaneously from 6 HCLs to the flame, and 6 flexible fibre-optic waveguides transmit the light from the flame to 6 photomultiplier tubes (143, 1448). (ii) Radiations from two multi-element HCLs were combined with a dichroic mirror and were then focussed by an achromatic lens through the atom cell on to a three-way beam splitter.Each of these beams was then split, using a medium dispersion monochromator, so that two different wavelengths can be monitored. The radiation from the exit slits was further spatially separated, by two small prisms, apex to apex, on to separate photomultipliers. Six elements can be determined simultaneously (1 509).(iiz3 Multi-element sequential flame emission has been demonstrated with a simple automatic electro-mechanical wavelength selector fitted to a single-beam commercial instrument (1493). (iv) The use of Vidicon-type detectors has been pressed by two groups of workers at several conferences (1285, 1340, 1452 and 1329, 1338, 1496) for the purpose of multi-element analysis. 6.2 FLAME-BASED INSTRUMENTSTABLE A COMMERCIALLY AVAILABLE EMISSION SPECTROMETERS Reciprocal Suppller Model Type chNa”r;n”efs cl3p;;;iTd ::::$:Ah Type of source Special features Applications Angstrom Inc., 22-101 D R. 60 0.278 210-430 P.O. BOX 248, Belleville, 0.397 21 0 - 518 M,ich. 48111. U.3.A. 0.556 21 0 - 634 V-70 D.R. 50 0.556 160-440 V-71 D.R. 22 0.556 178-320 A-70 D.R. 68 0.556 190-770 A-71 D.R. 22 0.556 210-350 Applied Quanto- D.R. 20 1.388 or 0.695 200-800 Research pact (36 lines) 0.695 or 0.35 200-400 Laboratories, Ltd., Wingate Road, ‘Luton, Beds., Quanto- D.R. 28 0.70 or 0.35 175-500 England. vac 28 (49 lines) Quanto- D.R. 28 0.70 or 0.35 175-500 vac 28C Quanto- D.R. 60 0.46 170 - 407 vac 80 (84 lines) Quanto- D.R. 60 0.695 or 0.35 190-610 meter 80 (84 lines) Quan to- meter 290008 Quanto- meter 33000 Quanto- 33000 vac D.R. 48 0-35 or 0.175 190-520 (90 lines) 0-46 or 0.23 190-630 0.56 or 0.28 190-705 0.695 Or 0.35 190-840 D.R. (64 lines) 0.695 or 0.35 190-610 (8 reference) meter 33000 D.R. As Quanto- 0.46 170-407 All ferrous and non- ferrous applications 3.0 m HV a s . spark, Optical interlock HV or LV a.c. arc, spectrum monitor.LV undirectional Various excitation d.c. arc, triggered stands and read-out capacitor discharge options wavelengths in vacuum. metals, oils, soils, and In-focus wavelength gwkogilcal specimens scanning -t- 3 nm from each receiver slit Mobile and operable in hostile environments. All wavelengths in vacuum. Various excitation stands and read-out options 1 - 5 m As model 22-101 As model V-70 As model V-70 but 0.75 mAs model 22-101 As ‘model A-70 As #model A-70 0-75 rn Low voltage i 1.5 m As model 22-101 As model 22-101. All Ferrous and nm-ferrous 0.75 mAs model 22-101 As model Y-70 including determination of C, S and P Air or argon excitation Particularly suited to High voltage stands.Typewriter and non-ferrous, e.g., Al, digital computer options Mg, Cu, Zn, and white metals, slags, powders, solutions including oils 0-6 m As Quantovac 80 As Quantovac 80 but 1A.s Quantovac 80, but 3 no air conditioner limited to 28 elements 2 !? 0.5 m As Quantovac 28 Complete computer As Quantovac 28 control.teletype or visual display output. ?I Off-line computer links 1 *O m Various, low voltage, Typewriter, teletype and All ferrous and non- 3 high voltage, multi- digital computer options.ferrous alloys, powders source (HVS, LV, Single or dual stand including slags, sinters ! d.c. arc) options. Second stand ores, rocks, ceramics, can be argon or air. Built-in instrument air- oils, etc. conditioning s soils, etc. Solutions, 2 1 . O m As Quantovac 80 As Quantovac 80 As Quantovac 80,. but 2 1.5 rn As Quantovac 80 Typewriter, teletype and As Quantometer 80 excluding determination of C, S and P 3 digital computer options.Argon and/or air stands available 4 analysis 5 9 1.0 m As Quantovac 80 Automated sequential As Quantometer 80 1.0 m As Quantovac 80 As Quantometer 33000 As Quantorac 80 -.Baird Atomic Inc.. SB-1 Phot. - 1 * 5 or 0.75 125 Middlesex Bedford. SH-1 PhOt. - 1.0 Turnpike, Mass. 01730, U .S.A. - 0.8 or 0 - 4 GW-1 Phot. GWR-1 Phot. - 0.8 or 0.4 Spectro- D.R. 30 0.6 O r 0.3 met 1000 GX-1 Phot. - 0.55 or0-28 GX-3 D.R. 16 per head 0.39 (normally 3 heads per unlit) Spectro- D.R. 60 0.294 met 11 (3.59 Spectro- D.R. 60 0-29 vac I 1 370-740 1.5 m Arc or spark 450-750 1.5 m Arc or spark 185-2400 2.0 m Arc or spark modular or RE-1 185-2400 2.0 m AS GW-1 210-590 1.0 rn Arc or spark modular 180-2250 3.0 m As GW-1 180-2250 3.0 m AS GW-1 190-432 2 - 0 m As Spectromet 1000 190-863 173-432 2.0 m As Spectromet 1000 Built-in order sorter General spectrographic analysis Built-in order sorter General Spectrographic analysis Dual gratings for General spectrographic simu I t aneou s photo- analysis graphy of two spectral regions High speed (f/15.5) gratings for rapid complex sources. examination of transient G en era1 spec t rogrep hic and/or weak sources or analysis complex spectra.Optional echelle grating for f/12.1 aperture Compact, low-oost direct Ferrous metals (except Transient, weak or determinations of S) usling C 193.1 nm, P 214.9 nm in 2nd order. Non-ferrous metals, oils reader with minimum air-conditioning requirements.Logarithmic readdout. Manual master monitor to check slit alignment Easy interchange of gratings. Easy con- version to a direct reader (GX-3) Automatic optical servo monitor continuously maintains correct slit alignment. Logarithmic read-out. Precise electponic setting of slits Automatic oplical servo monitor continuously maintains correct slit alignment. Logarithmic read-out.Manual master monitor to check slit alignment. Temperature- compensated fixed focal length. Dual stands for argon and air available As Spectromet 11. All p hsot om u I t i p I i er s in vacuum General spectrographic analysis Normal direct reading capabilities as well as photographic (see GX-1) All direct-reader applications above 190 nm All d irecl-wader applications including C, P and Sw N TABLE A COMMERCIALLY AVAILABLE EMISSION SPECTROMETERS - continued No.of ::$%::) Wavelength Focal Special features Applications Model Type channels nm per m,m range/nm length Type Of source - 1.1 or 0.54 210-780 1.5 rn Various available Wadsworth spectrograph General spectrographic Jarrell4sh 78-000 Div., Fisher 6cienTific Co., 590 Lincoln St.. 70-310 Wa I t ham, Mass. 02154 U.S.A. 75-150 V. A. Howe 8. Go. Ltd., 88 Peterborough Road, London S.W.6, England. 90-750 90-7'85 1500 Atom- counter 70-31 4 82-000 78-460 78-490 25-020 78-420 (T8-480) 82-400 (82-410) Phot. Phot. Phot. D.R. D.R. D.R. D.R. Scan. Scan. Scan. Scan. Scar;. Scan. - 1 - 0 to 0.24, depending uipon grating - 4.4 to 1.1 3 - 2 to 0.8 1.6 to 0.4 u p to 50 0.54 Up to 50 0.54 Up to 30 0-56 or 0.28 0.34 or 0-17 30+ AS 70-310 - Depends grating - si:I:ed - 3.28 in 'Varisource' unit analysis General spectrographic 180-3000 3.4 m including spark, 20 inch camera anal ys'is low- and high- 180-750 voltage a.c.arc, d.c. arcs. 200-6000 0.75 m Also versatile Ghoke of 3 gratings. Verva'lile instrument 1 -0 m 'controlled wave- Nitrogen purging exaends plarticularly suitable 2.0 m source, Optional accessories spectra 180- 1500 or form e5citation range to 175 nm.far lmeasuring itrauvs'ien't (model 66-776) permit use as direct reader or scanning spectrometer 1 electronic control- an'alyses 168-500 0.75 m As above, except Oomputer controlled Most metiallurgical 168-500 0-75 m led peak current 200-800 or 1 - 5 m As above Choice of 2 gratings. All direct-reader 190-400 applications above 200-510 or As 70310 3.4 m As above Scanning optimal.Pasy All direct-realder 190 nrn 190-250 interchange to photo- applications above graphic (70-310) and 190 nrn scanniing version b 2 2 spectrometers, some spectroscopic ?1 (70-320) Various scanning Suitable for wirh vacuum investigations rarher capab'il i t ies than for analytical 3 2 applications I i! 0.5 rn \ Depends 1 1.0 m upon 200-900 0.25 m DELTA Scan.- 1.2 195-770 0.6m Flame 3: Jobin-Yvon, VARAF Scan. - 1.8 or 0.9 200-800 0.465 mFlame Czerny-Turner mono- Liquids and solutions 3 91160 bandwidth 0.02-4 nm. 9 models with various a R Longjumeau, France. burner and read-out options Czerny-Turner mono- Liquids and solutions bandwidth 0-02-4 nm. Autorna t ic wavelength scanning device.laminar flow burner. Ultrasonic nebuliser 1 Rue du Canal, chromator, adjustable c 2 chromator, adjustable 5 i! Pneumatic nebuliser, z %. optional sM.B.L.E., PV 8300 D.R. Up to 90 0.55 or 0 46 170-430 or 1.5 in Various, optional 0ptional:zero order Steels, iron, non-ferrous 80 Rue des Vacuum 177-407 air, arc/spark stand reflected beam assembly, metals, slags, oils Deux Gares, pin sample holder, air Bruxelles 7, conditioning , rotawing Belgium.disc holder, adaptor for po int-to-fl ame exc I tat ion. Various type's of read-out 8210 D.R. U p b 60 0.55 190-700 1 . 5 m Various Air AS PV 8300 Sollids, liquids and powders. Non-ferrous metals and oils Optica 6.A.S., I35 PhoR. - 0-69-0.36 200-800 Via Gargano 21, 20139 IM'ilano, Italy. B5C D.R. 16 0.69 or 0.36 220-420 B7V D.R. 93 0.37 165-440 ESAl Scan. - ESA3 D.R. 9 ESA4 Scan. - 0.41 0.36 0.41 200 -500 160-500 (40 nm as POlY- chromator) 165-500 1.2 m 1.2 m 1.5 m 1.0 m 1.2 m 1 . 0 m All conve'ntional types available LV triggered arc and spark. HV spark, a.c. and d.c. arc LV tr'iggered arc and spark. Hill epark Controlled and n on-con4 r o I I ed H V spark, a.c. arc LV triggered arc and spark LV triggered arc, HV spark, a s .arc Stigmatic instrument with rotating Ebert grating Double spark stand both in air and inert atmosphere. Rotrode for so!utims. Air -vacuum in strum en t with all exit slits accessible from outside for adjustment. Many analytical programmes can be arrangad in parallel for easy interclhange. Computer fac i I iCi es avai la ble Scanning monochromator with one channel for analytical line and another channel for reference using reflected beam principle General purpose General purpose metal lug ical anlalysis, e.g., Al, Pb, Zn, Fe, Cu alloys.Wear metals in oils, etc. Comlplex analyses involving miany spectral lines MetaHurg4cal work. #I! material exlcita'ble with same source parameter Combined vacuum mono- Routine analysis and polychromator.All excitable elements of iron and steel. accessible With scanning Non-ferrous allloys system Scanning vacuum mono- Meta181urgfical w k . chromator with one Analysis of ferrous and channel fior analytical non-ferrous alloys line and another channel for reference. Facilities for analysing two elements simulltaneously (including C , S and P)w P TABLE A COMMERCIALLY AVAILABLE EMISSION SPECTROMETERS - continued No.of dq$z:$ Wavelength Focal Special features Applications Model Type channels nm per mm range/nm length Type Of source Rank Precision E 1000 D.R. 60 0.293-1 -155 159.6-864-3 1.5 m Various Dual spark stands. Ferrous and non-ferrous Industries Ltd., Polyvac Computer-controlled alloys. Geological Analytical Div., instrument.Dual gm't- samples. Wear metals Westwood ings give six systems in oil Industrial Estate, E 600 D.R. 25 0.25-2.0 177-310 0.6 m Condensed arc or Solid-state electvonics Low and high alloy and Ramsgate Road, polyvac high-repetition with or without tool steels, irons, slags. Margate, condensed arc computer control Kent Cf9 4JL ~ 9 0 0 D.R. 36 0.546 or 174.0-447.7 0.75 rn Various Curved entrance and Ferrous and non-ferrous England. Polyvac 0.741 236.5-607.4 exit slits.So'lid state alloys. Wear metals electronics or aomputer in oils controlled. Air or vacuum E 74213 Phot. - Not stated 191 -800 1 -57 m D.C. arc, HY spark Adjugtable slit. Spectral Analysis of high-purity Large condensed arc length 0.67 m of which specimens having Quartz/ 0.24 m can be selected complex spectra.Glass for a given exposure Determinations of trace element concentrations E 777/8 Phot. - 0.26-0.97 200-1200 1.5 rn Various Czerny-Turner mono- Routine qualitative and Flash photolysis. Examination of line profiles E 549 D.R. 12 0.5-10 200-600 0.53 mHV spark, Solid state electronti&. Non-ferrous metals. Medium condensed arc, d.c. Quartz plate for Soils.Additives and Quartz arc, thyratron- atmospheric pressure wear metals in oils chromator quantitative analysis. controlled a.c arc compensation *rr R.S.V. GmbH., SPN 3.5 Phot. 8031 Hechendorf D.R. Pilsensee, West Germany. Siemens Ltd.. Great West SPN 2.0 Phot. House, Great D.R. West SPN 1.5 Phot. D.R. Brentford, SPN 1.0 Phot. Middlesex, England. SPV 1.0 Phot. (vat) Analymat D.R. I-air Analymat D.R.Il-vac Anal yma t D . R . I I I-vac Analy- D.R. meter 30 30 15 40 40 40 Scan. 0 14 - 0 -48 o .24- o -84 0.37- 1 * 1 0.56-1 ' 7 0.4-1.7 0.31 or 0.54 0.31 or0.54 0.42 or 0.5 0.12 200 - 1000 200 - 1000 200-1000 200-1000 300- 1300 200 -650 150 - 490 110-500 200 - 800 3.5 m Glow discharge Paschen-Runge mounting General analysis 2 2 2.0 m As above As above General analysis 3 7 lamp, high, medium specially designed for or low voltage spark, a.c.or d.c. Direct reading attach- arc, continuous ment available and intermittent range below 200 nm. 1.5 m As above Direct reading General analysis i! 3 attachment available 3 1.0 m As above Direct reading General analysis 6 2 1.5 m Glow discharge Exhibits no background; General analysis s" attachment available a. 1.Om Asabove As above General analysis lamp (others no matrix effects. avai la bl e ) Linear calibration for all elements 0-1009'0 1 . 5 m Asabove As above As above 2 s 1-0 m As above As above As above e 6. 2-0 m As above As above As aboveShimadzu GE-100 Phot. - 1.66 or 0.83 200- 1 a 0 m Not stated Not stated - Seisakusho Ltd., ~ ~ - 1 7 0 phot. - 0.48 200 - 1.7 rn Not stated Not stated - 1 -c home, Chiyoda-ku, spark Tokyo 101, Japan. 14-5 Uchikanda GVM-lOO D.R. 24 0.46 170-410 1-0 m Modular unit, low Not stated - and high voltage Spectrametrics AE2 Phot. 1 Inc., D.R. 204 Andover St., D.R.10 D.R. 10 An dover , (lnter- Mass. 01810, changeable U.S.A. cassettes) ES 9 Phot. - RS 1 D.R. - 0.06 190-900 0.75 mPlasma jet Optimised AE system 0-06 190-900 0-75 rn Plasma jet high energy throughput using a high dispersion, echelle spectrometer and a high temperature plasma jet excitation source 0.06 190-900 0.75'm Plasma jet, flame 0.06 190-900 0.75 rn Plasma jet, flame or arc stand or arc stand Routine hnalysis. Rioutine quantitative multi-element analysis Qualitative and semi- qua nlt'i tative ana I y s is Spectroscopic research Routine anialysis Spex Industries 1870 Scan.- 1.6 175-1280 0-5 m - Multi-purpose unit Inc., 1702 Phot. - 1 *1 175-1500 0-75 m - - Research p*o. 7981 1704 Ph,ot. - 0.8 175-1500 1.0 m - - Research 1802 Phot. - 0.8 180-1500 1.0 rn - Direct reading Metuchen, N.J. 08840, U.S.A. Routine analysis accessory available Glen Creston, The Red House, 37 The Broadway, Stanrnore, M i dd I esex HA7 4DL, England.VEB Carl Zeiss Jena, 69 Jena, Carl-Zeiss Str. 1, German Democraric Republic. Carl Zeiss Jena Ltd., 93/97 New Cavendish St., London W1A 2AR, England. DSA-240 D.R. Up t o 5 6 0.78 210-550 0.54 mSpark Up to 11 lines analysable in single, automatic scan. Choice of reference I in e . B u i I t-in-auto m a t ic temperature and pressure compensation. Digital signal averaging, display and recording.Rapid change of analysis programme by pro- gramme store measuring range. Stigmatic depiction. Dispersion doubled by dioubled passage of light. Pre-disperser for order sorting and isolation. Gratings interchangeable. Automatic transport of plate holder. (2-24 Phot. - 0.76 210-550 0.54 mArc or spark Full range of ac c esso r i es ava i I ab'l e . PGS-2 Phot. - 0.74 or 0.37 200-2800 2.075 mAfc or spark Automatic expansion of All steels and non- ferrous alloys.Belaring metals and solders. Mineral oils and aqueous solutions General spectrographic analysis. Also examination of line profiles, hyperf itne structure, etc. General spectrographic analvsisTABLE B COMMERCIALLY AVAILABLE ATOMIC ABSORMXQN SPECTROMETERS Supplier Single/ Model double Monochromator G l ~ ~ ~ ~ g ~~~~~~) Resolution/ Wavelength "'z,$Ct; Other features beam per mm nm per mm nm range/nm expansion Beckman Instruments, 485 2500 Harbor Boulevard, Fullerton, Calif. 92634, U.S.A. 495 (Beckman Instrument GmbH, 8 Munich 45,. 1233 Frankfurter Ring 115, West Germany. Beckman R l l C Ltd., 1236 Eastfield Trading Estate, Glenrothes, Fife, KY7 4NG, Scotland. Double Double Double Double 1200 1200 1200 1200 0.2 0.2 0.2 0.2 Meter; x 5 0 Single and triple pass optics; Automatic filter selection Digital; x i 0 0 As model 485 Meter; x 5 5 Single and triple pass optics; YO T, abs.or conc. read-out Digital; x 5 5 As model 1233 Bausch and Lomb, Spec- Single Double Grating Manual stray-light and 820 Linden Avenue, tronic second-order filter Rochester, A'c2-20 selection; emission or N.Y. 14625, U.S.A. uv scrlutilon spectrometer Carl Zeiss, FMD 3 Single 0.05 193-852 Digital 4-lamp turret with 2 7082 Oberkochen, stabilised power supplies; Wurttemburg, curve correction; auto West Germany. zero; optional automatic calibration and back- ground compensation * 5 Corning-EEL, EEL 140 Single 0.25 m modified 1180 3.5 Non-linear Single lamp turret 5 St.Andrews Works, Ebert-Fastie m&er Halstead, Essex, England. EEL 240 Single 0.25 m modified 1180 3.5 Meter 4-lamp turret; integration; 5 2 GCA/ivlcPherson Instrument, EU 703 Single 1180 2.0 0.1 Modular AA; flame 3 2 spectrometer 3 II E bert-Fastie f/8 aperture 530 Main Street. emission; various detector: Acton, Mass. 01720, U.S.A. and gratings available; convertable to single or $ double beam UV a Nissei Sangyo Co.LTd., read-out; wavelength drive 2 2-Chome, Minato-Ku, Tokyo, Japan. Hiltachi Ltd., 208 Single Czerny-Turner 1440 1.8 x 20 3-lamp turret; conc. rsr 15-12 Nishi-Shimbashi, 7 2 4 g. 5Instrumentation Laboratory, 453 Inc., 113 Hartwell Avenue, Lexington, Mass. 02173, U.S.A. i ns t r u men tati o n Laboratory (U.K.) Ltd., Station House, Stamford New Road, 353 Altrincham, C hes h ire, Eng I and. Double; A . 0.33 m Ebert Channel 6. 0.16 m Ebert Dual Double: A. 0.33 m Ebert Dual Channel B. Interference Filters A . 6. 1200 A 2 5 A 0 0 3 A Digital, x250 6-lamp turret, wavelength scan, integration, ABS or conc. read-out, auto zero 1200 B. 5.0 6. 0 2 B. and calibrate, curve correction; automatic backqround correction and internal standard mode. 190 - 900 190 - 900 A. 1200 A. 2.5 A . 0.03 A. Digital; x250 As model 453 B. - 6. - B. 2.5- B. 190-900 6.0 248-766 253 Double 0.33 m Ebert 1200 2 - 5 0.03 190-900 Digital; x250 As model 453 except automatic background correction and internal standard mode. 251 Double 0.33 rn Ebert 1200 2.5 0.03 190-900 Digital; x 5 0 Integration; A'BS or cionc. read-out; auto zero; curve correction. 0.03 190- 900 Digital; x50 As model 251 -- 151 Single 0.33 m Ebert 1200 2.5 Jarrell-Ash/Fisher, Dial Atom Single 0.25 m Czerny- 1180 3.3 Meter; x 10 2-lamp turret; f/7.5 590 Lincoln Street, I I Turner aperture Waltharn, Mass. 02154, Atomsorb Single 0.25 m Ebert 1180 Meter; x20 6-lamp turret; f/3-6 aDerture U.S.A. V. A. Howe & Co. Ltd., 88 Peterborough Road, 82-500 Single 0.5 m Ebert 1180 Llondon S.W.6, England 82-810 Double Two 0.4 m Ebert 1780 2.1 Dual x 20 Various optics available Digita; Wavelength scan; integration, auto zero Channel Jobin-Yvon, VARAF Single 0.465 m Czerny- 1220 1-8 Digital 1 Rue du Canal, Turner 91 160 Longjumeau, DELTA Single 0.6 m Czerny- 1220 1.2 Digita France. 6000 Single 0.35 m Ebert Digital; x50 Automatic filter insertion; OPtlC0, pre-focussed water cooled Via Gargano 21, 20139 Milan, Italy.Automatic wavelength 5can Turner hollow cathode lamps; continuous regulation of flame temperature; auto concentration; integration W 4TABLE B COIMMERCIALLY AVAILABLE ATOMIC ABSORPTION SPECTROMETER Single/ Supplier Model double Monochromator Gr,:2:g ~ e S ~ ~ ~ ~ ~ ~ R e s o l u t i o n / W a v e i e n g t h Read-out; Scale Other features beam per mm nm per mm nm range'nm expansion w 00 Perkin-Elmer Corp., 103 Single 0.27 m Litirow 1800 1.6 0.2 190-860 Meter; x50 All mirror optics; Norwalk, integration; auto zero; Conn. 06856, U.S.A. auto flame ignition; emission PerkindElmer Ltd., 107 Single 0.27 m Littrow 1800 1.6 0 - 2 190-860 Digital; x50 As model 103 Post Office Lane, Beaconsfield, Bucks.HP9 lQA, England. Perkin-Elmer & Co. GmbH, Postfach 1120, 7770 Ueberlingen, 305B Double 0.4 m Czerny- A. U.V. 1400 A. 0-65 0.03 190-420 Meter; x l 0 0 Auto zero; curve West Germany. Turner; B. Vis. 1400 6. 1.3 420-860 correction ; integration ; 300 Single 0.4 rn Czerny- A. U.V. 2880 A. 1-0 0.2 190-420 Digital; x 4 0 Auto zero and conc.: 300s Single 0.4 m Czerny- A.U.V. 2880 A. 1.0 0-2 190-420 Meter; x 4 0 As model 300 Turner; curve correction; auto Double grating B. Vis. 1800 B. 1.6 420 - 861) flame ignition; emission Double grating Double grating auto flame ignition; 306 Double 0.4 m Czerny- A. U.V. 1400 A. 0.65 0.03 180-420 Digital; x200 Azs model 3095 p'lus 503 Double 0.4 m Czerny- A. U.V. 1400 A. 0.65 0.03 180-420 Digital; x l 0 0 As model 30Wplus Turner; 5 .Vis. 1800 8. 1-6 420 - 860 emission Turner; auto ccnc. Turner; and x 1/10 computed signal averaging Double grating B. Vis. 1400 6. 1.3 420 - 860 Double grating 5 . Vis. 1400 B. 1 - 3 420 - 860 and peak reader Pye Unicam Ltd. SP 90 Single Silice Prism - 3 at Meter; x 1 0 York Street, Series 2 200 nm Cambridge CB1 2PX, 32 at *cr England. 400 nm s Phi I i ps Electronic, Instruments Inc., 750 South Fulton Ave., Mount Vernon, N.Y. 10550, U.S.A. SP 1900 Double Ebert SP 1950 Double Ebert 1800 1800 2.2 2.2 0.1 0.1 Digital; x20 6-lamp turret; auto zero; and x 1/10 integration curve correction; conc. read-out; auto flame ignition Digital; x20 As model SP 1900 and x 1/10 except single lamp turret Rank Precision Industries ATOM- Single Silica Prism - 1.7 at Meter 6-lamp turret Westwood Industrial Estate, MK3 500 'nm Ramsgate Road, Margate, Kent CT9 4JL, ATOM- Single Czerny-Turner 1200 2.6 0.1 193-853 Digital 6-lamp turret; auto zero; 5 England.SPEK ABS, conc. and emission Gl Ltd., SPEK 200 nm e Analy!ical Div., H 1170 44.6 at E. s" 2 H 1550 read-out; curve correction; integration; auto flame iqnition 2 5 -. -Shandon Southern A3400 Single 0.25 m Czsrny- 632 6.0 0 . 2 190-860 Meter; x25 4-lamip turret; auto zero; instruments Ltd., Frimley Road, Camberley, Surrey GU16 5ET, England. emission and fluorescence Turner curve correction; integration; wavelength drive; auto flame ignition; Shandcn-i Lwbortechnik G,m bH , Frankfurt/Main 50, A3600 Single 0.25 m Czsrny- 632 Turner 6.0 0-2 190-860 Meter; x25 Single lamp turret; integration ; em i ss i,on and fluorescence West Germany.Sh imadzu-Seisa kus ho LM . , AA610 Single Czerny-Turner Meter; x 10 2-lamp turret; wavelength 14-5 Uchikana 1-chome, drive CMiyoda-Ku, Tokyo 101, Japan. Spectrametrics, I nc., A E-2 Primarily intended for 204 Andlover Street, emission. See Table A Andover, Mass. 01810, U.S.A. DR-10 ES-9 As model 'AE-2; See Table A As model A'€-2; See Table A FZS-1 As model AE-2; See Table A Varian Techtron Ry.Ltd., 1000 Single 0.25 m Czemy- 679 Springvale Road, Turner North Springvale, Victoria, Auslrali'a 3171. 1100 Single 0.25 m Czwny- Turner Varian Associ'ates Ltd., Russell House, Molesey Road, Walton-on-Th ames, 1200 Single 0.25 m Czerny- Surrey, England. Turner Varian Instrument Div., AA-6 Single 0 - 5 m Ubert 611 Hansen Way, 'Palo Alto, Calif. 94303, U.S.A. 1276 2.8 1276 2-8 1276 2.8 638 3.3 Meter; x10 Meter; x50 Digital; x50 Digital or Meter; X 0 -3-50 4-lamp turret; auto zero; emission; aperture f/8 As ~nwde'l'1000; plus linitegratbn, curve correction; peak signal retrieval As model 1100 Modular construction; 4-lamp turret; auto zero and calibrate curve correction; peak height retrieval ; integration; emission: aperture 1/10 ~- VEB Oar1 Zeiss Jena, AAS 1 Single 59 Jena, Carl Zeiss Str. 1, 3erman 'Democratic Republic. 1300 1.5 190-800 Carl Zeiss Jena L!d., 93/97 New Cavendish St., London W1A 2AR. England.TABLE C COMMERCIALLY AVAILABLE FLAMELESS ATOMIC ABSORPTION ATOMIZERS Sensitivity for 1% abs. (S.) c u Si Detection limll (d.1.) Special features Supplier Model Type M~~iusm”e”;up~ Control unit FI~S al AA spectrometers.sh ie Id ing and hydrogen Barnes H Engineering CO., Glomax anhlum Strip ully d.1. 10-11 g .S. Air cooled; inert-gas 150 PI) 30 Ommerce Road, programmable; Stdmford, Dry; Ash; Atomize; Conn. 06902, Burn off. Max. flame U.S.A. temp. 2400°C Beckman Instruments GmbH, 1268 Gralphite Furnace 100 Fully 8 Munich 45, programmable; Water cooled; inert-gas Frankfurter Ring 115, Dry; Ash; Atomize; shielding and hydrogen West G~~wwY.Burn out. Max. flame. Safety feature for failure of water o r purge gas. s. 10-10 g 8. 10-10 g Fits all AA spectrometers. temp. 3100°C ~- ~~ ~~ I ?strumentation ,Laboratory Inc., lL355 Tantalum Strip 50 Dry; Atomize. d.1. 10-11 g d.1. 10-8 g Fits most AA spectro- 11 3 Hlartwell Avenue, Max.temp. 2700°C (25 pl) (25 pll) meters. Air cooled; inert Lex’ingbn, or reducing-gas shielding. Mass. 02173, Maintains constant current during atomization. U.S.A. Permits AF determinations Fits all AA spectrometers. Optica S.A.S., CAT 6 Tantalum Strip 50 (aq.) Fully d.1. lo-’! g N.S. Water cooled; inert-gas Via Gargano 21, 40 (org.) programmable; 150 PI) shielding 20139 Milan, Dry; Ash; Atomize Italy.PerKin Elmer & Co. GmbH. HGA 72, Graphite Furnace 100 (aq.) Fully d.1. 2 x 10-12 g d.1. 5 x 10-11 g Fits only PerKin Elmer Postfach 1120, 74, 2000 50 (org.) programmable; (100 NI) (100 pi) and Zeiss AA spectro- 7770 Ueber I ingen, Dry: Ash, (two); meters. Water cooled; West Germany. Atomize. Max. inert-gas shielding. temp. 2700°C Permits ramp ashing; Perkin Elmer Corp., gas-stop; closed system 2 N orwal k , Conn. 06856, feature for failure of U.S.A. water or purge gas Pye Unicam Ltd., Graphite Furnace 100 Fully d.1. 4x10-12 g d.1. 10-11 g Water cooled; inert-gas t; York Street, programmable; (100 pl) (100 &I) shielding. Safety feature Cambridge CB1 PPX, Dry; Ash; Atomize; for failure of water or S *a s 2 2 2 (74) operation. Safety England. Burn out. Max. purge gas a temp. 3100°C b t $ (P 5 :. xRank Precision Industrim Ltd., H14575/ Graphite Furnace 100 Fully S. 5 X lo-" g N.S. Water cooled; inert-gas 2 Analytical 'Division, FA 256 programmable; shielding; power taken ;?. Westwood Industrial Estate, Dry; Ash; Wait; from Atomspek Spectro- 21 Ramsglate Road, Atomize. Max. meter directly Margate, temp. 2600°C Kent CT9 AIL, 3 England. Q- <handon Southern Instruments A3470 Graphite Rod 50 Fully d.1. 5 x 10-12 g d.1. 6 x 10-11 g Firs all AA spectrometers. Frimley Road, Dry; Ash (two); Ca m be rl ey , Surrey GU16 SET, temip. 3000°C the instrument during (5 pi) Air cooled; inert-gas shielding. Flameless accessory can be left in Ltd., programmable; (5 PI) Ens land. flame measurements 9 Atomize. Max. Varian Techlron Pty. Ltd., 63 Graph'ite Furnace 5 Fully d.1. 4 xlO-l* g d.1. 8 x 10-f' g F i t s a l l UVA spectrometers * (5 pl) wifh optical aperture f/8 $ or less. Water cooled; North Springvale, Dry; Ash; Atomize. Victoria, Max. tem p . inert-gas shielding and 2 Australia 3171. 3000°C hydrogen flame. Pyrolytic graphite coating on cups 679 Springvale Ro'ad, Graphite Cup 25 programmable; (5 1 1 ) and tubes 3 2 3 N.S. = Information not supplied. 3

ISSN:0306-1353    

DOI:10.1039/AA9730300029

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

42 Part I: Fuiidnincntals arid Iristrurtieiitatiort 7 Ancillary Information 7.1 STANDARDS Tables D, E, F and G supply updated information on commercial sources of spectrographic standards, spectrographic graphite electrodes, standard metal solutions and reagents for AAS and organometallic compounds. Of interest to those concerned with ‘blood Pb analysis is the fact that stable blood samples, analysed for Pb, are now available (United States Public Health Service, Health Services and Mental Health Administration, Bureau of Community Environmental Manage- ment, Cincinnati, Ohio, U.S.A.).In a similar way the preparation and analysis of a synthetic silico-aluminous glass doped with 28 elements, present as their oxides at a level of 1000 f i g g-I, and proposed as an analytical standard (954), is of importance; although it should be remembered that the National Bureau of Standards, YTashington, D.C., U.S.A., already supply a range of similar standards.Repas and Gegus (669) at the 7th Meeting for An,etal- lurgical Material Testing at Balatonsezplak reported the start of manufacture of analytical steel standards in Hungary. A method (777) for the production of A1 spectrographic standards has been described in which the molten metal is introduced to a sheet-iron can and then rapidly cooled by water directed at the top of the can.Uniformity of composition of repeated castings proved acceptable and spectrochemical responses of the standards and samples were similar. Reference samples for Ag, Au, Bi and Cr, at the 0.01-1 g-* level, in CdS, silica and C powder have been prepared satisfactorily by a vacuum vzporizatiod deposition technique (485).Smith (469, 472) has clearly demonstrated a fact known by some and suspected by many, viz., the instability of dilute aqueous standard solutions. Of the 28 elements studied, with concentrations at or below 10 rng 1-I, it was found that a pH of 1.5 or less was necessary in almost all cases to ensure that the metal ions remained in solution.Two reports describing adsorption characteristics of Ag, Pb, Cd, Zn and Ni on boro- silicate glass, polyethylene and polypropylene container surfaces (1 043), and Zn contamination from rubber products (128) extended the excellent work carried out by Scott and Ure last year (ARAAS, 1972, 2, 47, refs. 464, 1096). 7.2 DOCUMENTATION Ramirez-Munoz (967) has given a personal but useful description of the criteria to be followed when writing various types of papers in the field of AAS; particular attention was given to experimental papers dealing with new instrumentation, studies of the analytical behaviour of analytes and new procedures for specific analytical applications. Personal views were also expressed by Elwell (284), Price (285), Dawson (286) and Nall (287), past Chairmen of the Atomic Spectroscopy Group of the S.A.C., at a meeting to mark the tenth anniversary of the formation of the Group.These speakers reviewed the history and instrumental develop- ments of AAS and its applications to clinical biochemistry and ferrous alloy analysis. During the year many reviews of interest have appeared which are listed elsewhere in this volume.The I.U.P.A.C. general recommendations for nomenclature, symbols, units and their usage in atomic emission spectroscopy have also been published (86).Part I : Fundamentals and Iiistrumentntion 43 TABLE D SPECTROGRAPHIC STANDARDS Supplier Alcoa Hesearch Laboratories, New Kensington, Pa., U.S.A.(Alcoa of G.B. Ltd., Alcoa House. P.O. Box 15, Droitwich, Worcs., England). X Apex Smelting Co., 6700 Grant Avenue, Cleveland, Sh'io 44105, U.S.A. X X X British Non Ferrous Metals Research Association, Grove Laboratories, Denchworth Road, Wantage, Berks. OX12 9'BJ, England. B u t i d e san st a I t fur 1 Berlin 45, Unter den Materialprufung ('BIAM), Eichen 87, Germany. X x x X x x x Bureau of Analysed Samples Ltd., Newham Hall, Newby, Middlesbrough, Teesslide TS8 9EA, Englan'd x x x x X X 17 CKD Research Institute, Na Harfe 7, 190 02 Praha, 'Czechoslovakia.x x X G. L. Willan Ud., Sheffield Works, Sheffield W'orks, (Catcliffe, Rotherham, Yorks., England Henry Wiggin 8. Co. Ltd , Holmer Road, (Hereford, He ref ords h ire, Herefordshire, Englland. x x X X X Johnson Matthey Chemicals Ltd., 74 Hatton Garden, London lEClP W E , England.'Spectrornel ' powders 'Specpure' metals Moore Boundy Hamill Ltd., Station House, Potters Bar, Herts. lCN6 l k L , England. x x x x x X x x Office of Standard Reference Various Materials, National Bureau of Standards. Washingpon, DJC. 20234, U.S.A. incl. high X X X X X purity metals X Pechiney, 23 Rue Balzac, Paris 8e, France.Spex Industries Inc., P.O. Box 798, Metuchen, N.J. 08840, U.S.A. (Glen Creston, The Red x x X x x X House, 37 The Broadway, Stanmore, 'Middlesex, England. X Tyseley Metals Ltd., Kings Road, Birmingham 11, Eng Ian d . Zinc & Alliages, 34 Rue Collange, 92307 Leva IoisiPerret, 'France. X44 Part I : Fundamentals and Instrumentation 7 8 9 10 TABLE E SPECTROGRAPHIC GRAPHITE ELECTRODES Baird-Atomic, Inc., 125 Middlesex Turnpike, Bedford, Mass. 0 1730, U.S.A. General Graphites Inc., First and Monroe Street, Bay City, Mich. 48706, U.S.A. Johnson Matthey Chemicals Ltd., 74 Hatton Garden, London EClP lAE, England. Le Carbone (G.B.) Ltd., Portslade, Sussex, England. Met-Bay, Inc., 900 Harrison Street, Bay City, Mich. 48706, U.S.A. Carbon Products Division, Union Carbide Corp., 270 Park Avenue, New York, N.Y. 10017, U.S.A. (A.R.L. Ltd., Wingate Road, Luton, Beds., England). Poco Graphite, Inc., P.O. Box 2121, Decatur, Texas 76234, U.S.A. Ringsdorff-Werke GmbH., 53 Bonn-Bad Godesberg, West Germany (Mining & Chemical Products Ltd., Alperton, Wembley, Middlesex HA0 4PE, England). Spex Industries, Inc., 3880 Park Avenue, Metuchen, N.J. 08840, U.S.A. (Glen Creston, T h e Red House, 37 The Broadway, Stanmore, Middlesex, England). Ultra Carbon Corp., P.O. Box 747, Bay City, Mich. 48706, U.S.A. (Heyden & Son Ltd., Spectrum House, Aiderton Crescent, London N.W.4, England). TABLE F STANDARD METAL SOLWTIONS (MS) AND REAGENTS FOR AAS (R) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Aldrich Chemical Co., Inc., 940 W.St. Paul Ave., Milwaukee, Wis. 53233, U.S.A. (R) J. T. Baker Chemical Co., 222 Red School Lane, Phillipsburg, N.J. 08865, U.S.A. CMS, R) Barnes Engineering Co., 30 Commerce Road, Stamford, Conn. 06902, U.S.A. (MS) B.D.H. Chemicals Ltd., Poole, Dorset BH12 4NN, England. (MS, R) Bio-Rad Laboratories, 32nd and Griffin Avenues, Richmond, Calif. 94804, U.S.A. (MS) Carlo Erba, Divisione Chimica Industriale, Via C Imbonati 24, 20159 Milano, Italy.Eastman Organic Chemicals, Eastman Kodak Co., 343 State Street, Rochester, N.Y. 14650, U.S.A. (R) Fisions Scientific Apparatus Ltd., Bishop Meadow Road, Loughborough, Leics LE11 ORG, England. (MS, R) Harleco, Div. of American Hospital Supply Corp., 60th and Woodland Avenues, Philadelphia, Pa. 19143, U.S.A. (MS) Hopkin & Williams Ltd., P.O.Box 1, Romford, Essex RM1 LHA, England. (MS,R) V. A. Howe & Co. Ltd., 88 Peterborough Road, London S.W.6, England. (MS) Instrumentation Laboratory Inc., 11 3 Hartwell Avenue, Lexington, Mass. 02173, U.S .A. (MS) Johnson Matthey Chemicals LYd., 74 Hatton Garden, London EClP lAE, England. (R) Koch-Light Laboratories Ltd., Colnbrook, Bucks., England. (R) May & Baker Ltd., Dagenham, Essex R M l O 7XS, England.(R) E. Merck, D 61 Darmstadt, West Germany. (R) Spex Industries Inc., 3880 Park Avenue, Metuchen, N.J. 08840, U.S.A. (MS) (MS) 18 Ventron Corp., Alfa Products, 44 Congress Street, Beverly, Mass. 01915, U.S.A. (MS)Part I : Fundamentals and Instrumentation 45 TABLE G ORGANOMETALLICS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Angstrom, Inc., P.O. Box 248, Belleville, Mich. 48111, U.S.A. Baird-Atomic, Inc., 125 Middlesex Turnpike, Bedford, Mass. 01730, U.S .A. J. T. Baker Chemical Co., 222 Red School Lane, Phillipsburg, N.J. 08865, U.S.A. B.D.H. Chemicals Ltd., Poole, Dorset BH12 4NN, England. Messrs. Burt and Harvey Ltd., Brettenham House, Lancaster Place, Strand, London W.C.2, England. Carlo Erba, Divisione Chimica Industriale, Via C Imbonati 24, 20159 Milano, Italy. Conostan Div., Continental Oil Co., P.O. Drawer 1267, Ponca City, Okla. 74601, U.S.A. Durham Raw Materials Ltd., 1-4 Great Tower Street, London EC3R 5AB, England. Eastman Organic Chemicals, Eastman Kodak Co., 343 State Street, Rochester, N.Y. 14650, U.S.A. Hopkin and Williams Ltd., P.O. Box 1, Romford, Essex RM1 lHA, England. E. Merck, D 61 Darmstadt, West Germany. National Spectrographic Laboratories, Inc., 19500 South Miles Road, Cleveland, Ohio 44128, U.S.A. Office of Standard Reference Materials, National Bureau of Standards, Washington, Research OrganidInorganic Chemical Corp., 11686 Sheldon Street, Sun Valley, Calif. 91352, U.S.A. Ventron Corp., Alfe Products, 44 Congress Street, Beverly, Mass. 01915, U.S.A. D.C. 20234, U.S.A.

ISSN:0306-1353    

DOI:10.1039/AA9730300042

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

PART I1 METHODOLOGYIntroduction In Part 11, the term Methodology covers all aspects of the application of the techniques and instrumentation of AAS, AES and AFS to chemical analysis. The format adopted for previous volumes has been retained, with the subject matter tr-ated under the two principal headings of (1) General Techniques, covering the preparation of samples and standards and the interpretation of the experimental data, and (2) Applications, where specific methods of analysis are reviewed and tabulated. The classification of the range of applications into eight main fields of analysis also fullows the established pattern. Some duplication of entries may be found in instances where a method is relevant to more than one section. 49

ISSN:0306-1353    

DOI:10.1039/AA9730300047

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

EXPLANATION OF THE TABLES Each of the Applications sections, 2.1 to 2.8, is accompanied by a table which summarises the principal analytical features of the references from which the corresponding section is compiled. These Applications Tables form a convenient source of information for analysts interested in particular elements, matrices, sample treatments or atomization systems. In many cases, sufficient detail is given for the analytical procedure to be followed; absence of such detail usually means that the information was not directly available to the compiler of the table and the original reference should be consulted. The key to the tables is given below.EL EM ENT MATRIX CONCENTRATION TECH. AN A LY T E The elements determined are listed in alphabetical order of chemical symbol, except that, for space economy, multi-element applications (5 elements or more) are given at the end of some tables.The wavelength, in nanometres, at which the analysis was performed. An indication, necessarily brief, of the material analysed. The concentration range or level of the element in the original matrix, expressed as yo or ug/g for solids and mg/l or ug/ml for liquids.The atomic spectroscopy technique is indicated by A (absorption), E (emission) or F (fluorescence). The form of the sample, as presented to the instrument, is indicated by S (solid), L (liquid) or G (gas or vapour). ‘d.1.’ = detection limit in the analyte. SAMPLE TREATMENT A brief indication is given of the sample pre-treatment required to produce the analyte. ATOMIZATION REF. The atomization process is indicated by the abbreviations A (arc), S (spark), F (flame) or P (plasma), usually with some additional descriptive detail. The number refers to the main Reference Section, which gives the title of the paper and the name(s) of the author(s), with address. 49

ISSN:0306-1353    

DOI:10.1039/AA9730300049

出版商:RSC    

年代:1973

数据来源: RSC