摘要:

Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies By T. J. Simpson RESEARCH SCHOOL OF CHEMISTRY, AUSTRALIAN NATIONAL UNIVERSITY, P.O. BOX 4, CANBERRA, ACT 2600, AUSTRALIA 1 Introduction Recent years have seen 13C n.m.r. grow from a relatively obscure technique to one rivalling 1H n.m.r. in utility and scope. The theory of 13C n.m.r.,1s2 bio- logical3 and biochemical4 applications, and early biosynthetic work6.6 have been reviewed, two general texts718 and a compilation of spectrag have appeared, and the 13C n.m.r. spectra of a wide range of compounds have been assigned.lo Although radio-isotopes, particularly 14C, have proved extremely useful in biosynthetic studies, they have one severe disadvantage in the necessity of carrying out extensive degradations to locate the incorporated isotope in a metabolite. This is often very diffcult owing to formidable structural complexities or the presence of carbon atoms in an unreactive aromatic framework so that only a partial analysis is obtained.Furthermore, since in many cases chemical degradations are no longer necessary as structure proofs for natural products, their use for establishing 14C-labelling patterns becomes a tiresome exercise. However, since 13C n.m.r. is itself an integral component of structure elucidation, its use in biosynthetic studies is very attractive and allows the establishment of labelling patterns without recourse to extensive chemical degradations. The aim of this review is to discuss the methodology of 13C n.m.r. biosynthetic studies, both as an aid to evaluating the fast-expanding literature and to the undertaking of these studies, with emphasis on recent developments and applications.J. B. Stothers, Quart. Rev., 1965, 19, 144.* F. A. L. Anet and G. C. Levy, Science, 1973, 180, 141. * J. B. Grutzner, Lloydia, 1972, 35, 375.‘G. A. Gray, C.R.C. Critical Reviews in Biochemistry, 1973, 247. G. Lukacs, Bull. SOC. chim. France, 1972, 351. M. Tanabe, ‘Biosynthesis’, ed. T. A. Geissman, Specialist Periodical Reports, The Chemical Society, London, 1973, Volume 2, p. 241.’G. C. Levy and G. L. Nelson, ‘Carbon-13 Nuclear Magnetic Resonance for Organic Chemists’, Wiley-Interscience, New York, 1972. J. B. Stothers, ‘Carbon-13 N.M.R. Spectroscopy’, Academic Press, New York, 1972.L. F. Johnson and W. C. Jankowski, ‘Carbon-13 Nuclear Magnetic Resonance Spectro- scopy. A collection of Assigned, Coded and Indexed Spectra’, Wiley-Interscience, New York, 1972. loSee for example: E. Wenkert, J. S. Bindra, C.-J. Chang, D. W. Cochran, and F. M. Schell, Accounts Chem. Res., 1974,7,46. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies 2 Proton Satellite Method Before discussing 13C n.m.r., brief mention must be made of the proton-satellite method. Spin-spin coupling between a 13C nucleus and directly bound protons results in satellite bands appearing on either side of the main proton signal in the lH n.m.r. spectrum, so providing an indirect probe for monitoring the 13C abundance at a given position in a molecule.Any incorporation of a 13C- enriched precursor can be detected by the increase in intensity of these satellites, Sepedonin,ll griseofulvin,l2 fusaric acid,l3 piercidin A,14 molli~in,~~ and variotin16 have all been studied by this method. However, despite its advantage of only requiring readily available instrumentation, the method has severe limitations as only carbons with attached protons can be studied, and often complex lH spectra, traces of impurities, and spinning side-bands obscure the satellites. These limitations are overcome in the direct 13C n.m.r. method. 3 Carbon-13 Nuclear Magnetic Resonance The theory of 13C n.m.r. has been thoroughly reviewed',* so a brief introduction only is given here, a working knowledge of 1H n.m.r.being sufficient to under- stand 13C n.m.r. Table 1compares the relevant properties of l3C and 1H and it Table 1 Comparison of nuclear properties of 1H and 13C 1H 13c Nuclear spin * * Resonance frequency at 23.5 kG 100 MHz 25.2 MHz Natural abundance (%) 99.99 1.11 Relative sensitivity 1.oo 0.016 Normal chemical shift range for organic molecules 10 p.p.m. 200 p.p.m. may be seen that 13C, a stable isotope, natural abundance 1.1 % has a nuclear spin 1/2 and so is n.m.r. active and will show the same general splitting patterns as lH. As may be deduced from Table 1, the main difficulties in obtaining a 13C n.m.r. spectrum are its low abundance and low nuclear sensitivity, so that for a given sample l3C is 6000 times less sensitive than lH.It is the development of techniques to overcome this that has led to the huge growth in 13C studies. These include (i) the use of large samples, which became possible with the development of high-stability spectrometers taking up to 15 mm diameter n.m.r. tubes; (ii) multiscan techniques, initially multiple accumulation in the continuous- l1 J. Wright, D. G. Smith, A. G. McInnes, L. C. Vining, and D. W. S. Westlake, Canad. J. Biochem., 1969,47,945. M. Tanabe and G. Detre, J. Amer. Chem. Soc., 1966,88,4515. D. Desaty, A. G. Mclnnes, D. G. Smith, and L. C. Vining, Cunad. J. Biochem., 1968, 46, 1293. l4 M. Tanabe and H. Seto, J. Org. Chem., 1970, 35,2087. M. Tanabe and H. Seto, Biochemistry, 1970, 9,4851.l6 M. Tanabe and H. Seto, Biochim. Biophys. Acta, 1970, 208, 151. Simpson wave mode but subsequently pulsed Fourier transform (FT) n.m.r.17 which enables much faster spectral determination than conventional methods; (iii) proton-noise-decoupling,providing two gains in sensitivity. By applying a wideband proton-decoupling frequency the diffuse multiplets arising from W-1H spin-spin coupling are collapsed to a single sharp line; this also induces a nuclear overhauser effect (NOE) leading to an intensity enhance- ment up to three-fold, due to disturbance of the l3C energy-level populations for carbon atoms with attached protons. Owing to the wide range of chemical shifts observed for 13C nuclei, even com- plex molecules generally give l3C n.m.r.spectra in which every carbon has a discrete resonance. Much 13C n.m.r. assignment data has been accumulated ; nevertheless complete assignment of resonances to a new molecule requires great care, and often considerable effort, especially in biosynthetic studies where the conclusions can only be as good as the original spectral assignments. Several aids to assignment are available: (a) Known chemical shifts and substituent chemical shift efects. The chemical shifts of different functional types fall into well-defined ranges;798 carbonyl carbons resonate at low field (ca. 200 p.p.m.), aromatic and olefinic carbons at 16O-loO p.p.m., aliphatic carbons with electronegative substituents at 50-80 p.p.m., and simple aliphatics at highest field 10-30 p.p.m.Substituent effects are generally found to be additive and rules for predicting chemical shifts in hydrocarbons18 and benzenoid aromatics' are available. (b) Of-resonance and specific proton decoupling. In the proton noise-decoupled spectrum all lSC--lH coupling information is lost. In the off-resonance decoupling experiment, the lH irradiation is kept at high power levels but the centre fre- quency is moved ca. 500 Hz away from the protons being irradiated so that one-bond W-lH coupling patterns return, and the non-protonated, methine, methylene, and methyl carbons are observed as singlets, doublets, triplets, and quartets respectively. The observed or residual couplings, JR, are smaller than the actual one-bond coupling and are a function of the actual coupling, J, the decoupling power, H, and the decoupler offset Av:19 JR= JAv/H If J is known the residual coupling can be an aid to assignment.In the study of asperlin (l), a metabolite of Aspergillus nidulans, C-4, C-5, (2-6, and C-7 all have one attached proton and are oxygen-bearing and so appear in the range 55-80 p.p.m. and give doublets in the off-resonance spectrum. However, com- parison of the observed and calculated residual couplings allowed an unam- biguous assignment to be made and confirmed the incorporation pattern of sodium [2-I3C]acetate shown.20 l7 E. Breitmaier, G. Jung,and W. Voelter, Angew. Chem. Internat. Edn., 1971,10,673. l8 D. M. Grant and E. G. Paul, J. Amer. Chem.SOC.,1964, 86,2984. l9 R. R. Emst, J. Chem. Phys., 1966,45,3845. ao M. Tanabe, T. Hamasaki, D. Thomas, and L. Johnson, J. Amer. Chem. SOC.,1971, 93, 273. 499 Carbon-1 3 Nuclear Magnetic Resonance in Biosynthetic Studies If the 1H n.m.r. spectrum has been fully or partially assigned, the 1H and 13C resonance can be interrelated by single-frequency decoupling with the decoupler set exactly on a specific lH frequency. The attached carbon appears as a singlet in the 13C spectrum whereas the remaining carbons show off-resonance patterns. This process becomes tedious if several resonances require studying but can be overcome by plotting the line frequencies in the 13C spectrum as the 1H irradiating frequency is stepped through the lH n.m.r.spectrum. Where the lines cross gives the point where JR is zero, and hence the 1H and l3C frequencies can be correlated.21 Figure 1 illustrates the results obtained for NAD+ (2). (c)Lanthanide-induced shift studies. These are not as useful in 13C studies as in 1H n.m.r. because the actual shifts are of the same absolute value in both and so are relatively small in 13C n.m.r., as are solvent and anisotropy effects. However, they can be useful for separating overlapping resonances. In complicatic acid (3), the resonances due to C-3 and C-10 both occur at 46 p.p.m. However, addition of [Eu(fod)s] separated these resonances and showed that C-3 but not C-10 was enriched from [1-13C]acetate-enriched cultures of Stereum compZicatum.22 Another important use of shift reagents is to resolve the 1H n.m.r.spectrum prior to specific proton-decoupling studies on the 13C n.m.r. spectrum.23 (d)Model and derivative studies. Model compounds whose chemical shifts are known can be helpful in assigning the spectrum of a new compound, though they must be used with care. Quaternary carbons present the greatest difficulties in assignment, and studying the variation of chemical shifts in a series of closely related compounds may be the only method of reaching an unambiguous assign- ment. The 13C n.m.r. spectrum of tajixanthone (3,a prenylated xanthone meta- boiite of Aspergillus variecolor, was fully assigned by a study of eleven deriva- tives.24 Subsequent incorporation of [l-l3C]- and [2-W]-acetates indicated its biogenesis by prenylation and cleavage of an anthrone precursor via the known co-metabolite arugosin (4).(e)Synthesis of a compound enriched at a known site with l3C has been used for a* B. Birdsall, N. J. M. Birdsall, and J. Feeney, J.C.S. Chem. Comm., 1973, 316. 22 T. C. Feline, G. Mellows, R. B. Jones, and L. Phillips, J.C.S. Chem. Comm., 1974, 63. 23 B. Birdsall, J. Feeney, J. A. Glasel, R. J. P. Williams, and A. V. Xavier, Chem. Comm., 1971,1473. a4 J. S. E. Holker, R. D. Lapper, and T. J. Simpson, J.C.S. Perkin I, 1974, 2135. Simpson 80 78, 76. sdd 74, 72. 63 61-Figure 1 Plot ofpeak frequencies in the lH of-resonance selectively decoupled 13Cspectraof NAD+ as a function of the position of irradiation in the lH spectrum, expressed in p.p.m.to high frequency of internal dioxan. The position of the peaks in the lH noise-decoupled 13Cspectrum are shown by lines on the ordinate and the position of the proton peaks by lines of the abscissa. The arrows t indicate the point of collapse of the 13Cdoublet and the connection between a given lSCpeak and the assignedproton peak. Small doublet splittings are observed on some of the signals from long-range (C-H) spin couplings. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies assignment. An exact assignment of the meso-carbons of protoporphyrin IX (6) was required prior to biosynthetic studies. This was accomplished by synthesis of protoporphyrins enriched specifically at the p-, y-, and 6-meso positions respect ively.25 Z"3y" c (4) (5) (f)Partially relaxed Fourier transform (PRFT) n.m.r.The TI relaxation times for l3C atoms generally increase in the sequence methylene, methine, methyl, and quaternary carbon, so as the pulse internal time, r, is increased the negative peaks obtained in PRFTfor short values invert in the sequence above. Thiscan be of great value in assigning resonances, especially in congested spectra where off-resonance may be of limited value.2s (g) Incorporation studies. l5N is an isotope of spin 8 and so is n.m.r. active and will couple with l3C nuclei. Thus a metabolite grown in the presence of, say, Kl5N03 will exhibit 15N-13C couplings for any carbons bonded to nitrogen. Similarly, any compound with adjacent 13C nuclei will exhibit a WJ3C coupling between these nuclei.Both these points are illustrated below. 4 Biosynthetic Methodology The availability of 13C-enriched compounds has increased rapidly and a wide range, similar to that for 14C, is now available at enrichments of up to 95% and many others may be readily synthesized by standard methods from simple precursors such as WOZ, WHd, and K13CN. A. Precursor Incorporation.-Precursor efficiency may be assessed in several ways2' but for l3C studies the important criterion is dilution of added label. For 14C, dilution per labelled site is given by: A. R. Battersby, G. L. Hodgson, M. Ihara, E. McDonald, and J. Saunders, J.C.S. Chem. Comm., 1973,441; J.C.S. Perkin I, 1973,2923.*OK. Nakanishi, R. Crouch, I. Miura, X. Dominguez, A. Zamudio, and R. Villarreal, J. Amer. Chem. SOC.,1974,96, 609. *'S. A. Brown, in 'Biosynthesis', ed. T. A. Geissman, Specialist Periodical Reports, The Chemical Society, London, 1972, Vol. 1, p. 9. Simpson specific activity of precursor x no. of labelled sites specific activity of product To obtain unequivocal results in l3C studies using 90% enriched precursors, dilutions per labelled site of ca. 100 or less are required. This is due to inherent errors in l3C n.m.r. resonance intensities (see below), requiring a two-fold increase in 13C abundance to be certain that enrichment has occurred. Relatively large amounts of precursor, typically 1-20 mmol 1-l, have to be used to obtain this, especially for low precursor efficiencies.This introduces problems of expense and interference with normal metabolism; in contrast to 14C studies, non- tracer amounts are now being used. A lowering of metabolite yields is common and cases of toxicity have been reported for elevated concentrations of acetate (0.P5 and 1.6 g 1-1);z8 propionate28 (0.2 g 1-9, and mevalonate29 (0.1 g 1-1). However, in other cases, higher concentrations have been used successfully, e.g. 2 g 1-1 of acetate,24 and the problem can often be overcome by pulsed feed- ings of pre~urs0r.l~ Preliminary experiments with 14C-labelled precursors are generally carried out to ascertain the feasibility of 13C studies and to optimize conditions. Three main parameters require studying: time of precursor addition, incubation time, and amount of precursor.Maximum precursor incorporation usually occurs with addition of precursor at the start of maximum metabolite production, i.e. the start of the idiophase30 in microbial fermentations, necessitating the determination of growth and production curves. Incorporation may be very sensitive to time of addition. Figure 2 illustrates the marked variation of incorporation with day of addition of mevalonolactone into the sequiterpenoid trichothecins.29 The period of growth after addition of precursor may also be critical. The variation in dilution of [14C]acetate on incorporation into sepedonin (7)12 is shown in Figure 3, which illustrates that neither maximum yield of metabolite nor even maximum total incorporation of label is the important factor, the prime consideration being minimum dilution of label given a sufficient yield of metabolite for 1SC spectral determination. Finally, mass versus incorporation studies will determine the minimum amount of precursor that must be added to obtain a satisfactory enrichment. Table 2 shows the variation of dilution with amount of added [14C]acetate during biosynthetic studies on shanorellin (8) in ShanoreZZa spirotrichi.Incorporation of [Wlacetate and [13C]methionine indicated its origin from a tetraketide with the methyls derived from the C1-p00l.31 B. Interpretation and Presentation of Results.-Having obtained a 13C-enriched metabolite as above, the results from the 13C n.m.r.spectrum must be evaluated. ** R. J. White, E. Martinelli, G. G. Gallo, G. Lancini, and P. Benyon, Nature, 1973,243,273.’* J. R. Hanson, T. Marten, and M. Siverns, J.C.S. Perkin I, 1974, 1033. 30 J. D. Bu’Lock in ‘Essays in Biosynthesis and Microbial Development’, Wiley, London, 1967. 31 C.-K. Wat, A. G. Mclnnes, D. G. Smith, and L. C. Vining, Canad. J. Biochem., 1972, 50, 620. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies Table 2 Incorporation of radioactivity from [1-14C]acetate into shanorellin Shanorellin I 1 mmolketate pCi mmol-1 mg pCi mmol-f Dilution 0.5 210 49 8.05 26 1.o 105 63 8.85 12 2.0 52.6 44 7.90 6.7 5.0 21.1 27 8.16 2.6 Day of growth when fed MVA Figure 2 Variation of incorporation of [14C]rnevalonicacid (MVA) into trichothecin with day of addition (Reproducedfrom J.C.S.Perkin I, 1974, 1033) *a CHSCO~No D + HC02H HO EH, OH (7) Siinpson 2.5 0.54 )rCi mmol'l 2.0 .-'L Q, 1.5 /z z 0 n W n g 1.C 0.: C I 4 8 12 16 20 24 28 TIME, DAYS Figure 3 Yield and specific activity of sepedonin produced by cultures of S. chryso-spermum administered [14C]acetate (9 mmol 1-l) on the 7th day after innoculation (Reproduced by permission from Canad. J. Biochem., 1969,47,945) 0to CHS~@CH~ :%f48H3 CH3C02Na c* ~H~SCWH-CO~H 0 0 CH3 $HeOHI NHI COS-COA 0 (8) If high enrichments are obtained the labelled sites will be readily apparent by visual inspection of the spectra.In the case of radicinin (9), the enrichment was so high that only the labelled peaks were visible32 (Figure 4). 00 (9) ** M.Tanabe, H. Seto, and L. Johnson, J. Amer. Chem. SOC.,1970, 92, 2157. OMSO 200 I60 I20 80 40 0 sc/P.P.m. Figure 4 lSCN.m.r. specfraofradicininfrom (a) l8CH3CO2Na and (b) CH3l3COeNa (Reproduced by permission from J. Amer. Chem. Soc., 1970,92,2157) Incorporations are commonly given as percentage enrichments : observed l3C abundancePercentage enrichment = ( natural 13Cabundance ) -1.1 Very high enrichments, 5-60%, have been recorded but lower values, 0.5-5%, are more typical and can require more care in assessing, particularly at the lower values, because of the ‘intensity problem’.3 In 13Cn.m.r.line intensities are non-integral owing to the variable NOE and widely varying relaxation times. With FT n.m.r. the interval between scanning pulses may be shorter than the relaxation times of individual l3C nuclei, resulting in differential amounts of saturation occurring and thus variable line intensities, particularly for quaternary carbons. Addition of a free radical33 or a para-magnetic species,= e.g. chromium trisacetoacetonate, [Cr(acac)s], can partly overcome the problem. This complex quenches the NOE and shortens the relax- ation times to give more uniform line intensities and was used to advantage in 38 G. N.La Mar, J. Amer. Chem. SOC.,1971, 93, 1040. 34 R. Freeman, K. G. R. Pachler, and G. N.La Mar, J. Chem. Phys, 1971, 55,4586. Simpson studying the biosynthesis of helicobasidin (10) in HeZicobasidium m0mpa.3~ [ZW]Mevalonate was expected to label C-4 and C-12, with the remaining label distributed equally between C-8 and C-10 owing to tautomerism of the dihydroxy-quinone system. However, in the resultant spectrum, the large varia- tions in intensity made the labelling of C-8 and C-10 uncertain, but from the spectrum in the presence of 0.1 mol 1-1 [Cr(acac)s] the equal labelling of C-8 and C-10 was apparent (Figure 5). A second method uses gated decoupling. In this, the NOE is eliminated by switching off the 1H noise decoupling frequency during the interval between scanning pulses. This method was used to study the incorporation of [2-W] acetate into asperentin (1 1) by AspergiZZus The saturation problem can be eliminated by a sufficiently long delay between the scanning pulses, but as some 13C relaxation times are very long a compromise has to be made in practice to maintain reasonable spectral acquisition times.Ultimately the only reliable method is a direct comparison of the respective line intensities in the natural-abundance and enriched spectra; both sets of intensities are subject to identical NOES and relaxation considerations which should cancel out provided both spectra are standardized. This can be achieved by using identical concentrations and instrument parameters15 but a more con- venient technique is to normalize both spectra to a reliable standard.If the mater- ial is derivatized before spectral acquisition, as with asperentin (11) as the dimethyl ether or neomycin (see below) as the hexa-acetate, all the remaining line intensities can be normalized to the average value of the intensities of the introduced methyl groups in each spectrum. Correction for the difference of these averages in the respective spectra then allows direct comparison of indi-a6 M. Tanabe, K. Suzuki, and W.C. Jankowski, Tetrahedron Letters, 1973,4723. L. Cattel, J. F. Grove, and D. Shaw, J.C.S. Perkin I, 1973,2626. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies 2-'$ ACETATE NATURAL ABUNDANCE I I I I I I I iio I + Cr (acac),180 160 b 9I 618 61 NATURAL ABUNDANCE I I I I 1 I 1 I I I 1 I I I I 180 160 140 60 40 d0 &b Figure 5 13C N.rn.r.spectra of helicobasidin from (a) [2-18C]rnevalonate,(b) [2-18C]-acetate, (c) [l-W] acetate, and (d) at natural abundance, all in the presence of 0.1 M-[Cr(acac),], and (e) at natural abundance alone (Reproduced by permission from Tetrahedron Letters, 1973,4723) vidual intensities. However, derivatization is not always feasible. Incorporation of [2-13C]mevalonate was used to distinguish between the two possible foldings of the farnesyl pyrophosphate precursor of trichothecalone (12) (Scheme 1) as 14C studies provided conflicting evidence.29 Incorporation of label at G4 and C-8 rather than at C-10 indicated biosynthesis via path (6). The natural- abundance and enriched spectra were normalized to the line intensity of C-12 which was assumed to be unlabelled.This was not an ideal choice as, being one of the least intense peaks in the spectrum, it is most sensitive to errors. A more general method using all the unlabelled peaks in the spectrum for normalization has been proposed.24 Simpson * *w* Scheme 1 However, despite these operations, there remains for FT n.m.r. spectra an uncertainty in line intensities due to the digitization of data during spectral accumulation and tran~formation~~ so that a series of experiments run on the same sample under apparently identical spectrometer parameters can give intensities that vary by k20 %. This means that enrichments of less than 0.5 % must be regarded with caution in the absence of supporting data such as multiple spectral determinations, proton-satellite enrichments, or W-13C coupling.The problem is alleviated by the use of larger data-storage facilities3 but these are expensive and not always readily available. 5 Further Biosynthetic Studies A large variety of metabolite types have been studied, several having been mentioned above. Further examples are discussed below. Lasocolic acid (13) contains three unique C-ethyl groups. 14C Studies failed Me to establish the origin of these groups but addition of sodium [l-l3C]butyrate to the culture established their butyrate origin.38 Incorporation of [l-W] propionate confirmed the origin of the C-4, C-10, C-12, and C-16 methyls and [l-l3C]acetate that of the C-23 methyl.39 Considerable attention has been given to the origin of the ANSA chain in the 37 H.M. Pickett and H. L. Strass, Analyt. Chem., 1972, 44,265. 38 J. W. Westley, D. L. Preuss, and R. G. Pitcher, J.C.S. Chem. Comm., 1972, 161. 39 J. W. Westley, R. H. Evans, G. Harvey, R. G. Pitcher, D. L. Preuss, A. Stempel, and J. Berger, J. Antibiotics, 1974, 27, 288. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies rifamycins and related antibiotics. Incorporation of [l-lSC]-, [2-W]-, and [3-W]-propionates and of [1-13C]- and [2-13C]-acetates into rifamycin S(14) by Nocadia mediterranei confirmed the origin of the ANSA chain from eight propionate and two acetate units, linked in a clockwise manner2* (Scheme 2), m.CH~COLH C H3C H2COeH HO HoSOH CO2H ~~H~CO,H P3 Scheme 2 since [l-13C]acetate enriches C-17 rather than C-19. The exact assignment of the C-17 and C-19 13C resonances was crucial here and was achieved by specific proton decoup1ing.a Similar findings are reported for the related strepto- ~aricins.~'The origin of the C7N moiety, C-1 to C-4 and C-8 to C-10, is obscure, but the enrichment of C-1 and C-10 of rifamycin S on incorporation of [1-13C] glucose has been reported,42 providing the first direct evidence for the origin of this moiety, also found in the mitomycins, validamycins, and kinamycins. In this study the first use of [2J3C]malonate is reported; C-5 and C-18 are enriched but, interestingly, not the C-25 acetoxy-group.Incorporations of both l3C-labelled acetate and mevalonate into terpenes have been reported. Studies on helicobasidin, complicatic acid, and tricho- thecalone have been discussed above; incorporation of [Wlacetate into the virescenosides,e fusidic acid,U and AcrostoZagmus46 lactone have also been reported. The neomycins [e.g. (1 5)] are antibiotic metabolites of Streptomyces fradiae. The lack of crystalline derivatives and satisfactory degradations hampered 14C biosynthetic studies but incorporation of [1 -13C]glucosamine and [6-13C]glucose led to the labelling pattern shown in Scheme 3.46 The preferential incorporation 40 E. Martinelli, R. J. White, G. G. Gallo, and P. J. Benyon, Tetrahedron Letters, 1974, 1367. B. Milavetz, K.Kakinuma, K. L. Rinehart, J. P. Rolls, and W. J. Haak,J. Amer. Chem. SOC., 1973,95, 5793. 4a A. Karlsson, G. Sartori, and R.J. White, European J. Biochem., 1974, 41, 251. 49 J. Polonsky, 2.Baskevitch, N. Cognoli-Bellavita, P. Cecchivelli, B. L. Buckwalter, and E. Wenkert, J. Amer. Chem. SOC., 1972,94,4369. T. Riisom, H. J. Jakobsen, N. Rastrup-Andersen, and H. Lorck, Tetrahedron Letters, 1974,2247. 46 H. Kakisawa, M. Sato, T.4. Ruo,and T. Hayashi, J.C.S. Chem. Comm., 1973, 802. 46 K. L. Rinehart, J. M. Malik, R. S. Nystrorn, R. M. Stroshane, S. T. Truitt, M. Taniguchi, J. P. Rolls, W. J. Haak, and B. A. Roff, J. Amer. Chem. SOC., 1974, 96, 2263. Simpson 0 HOCH2 0 oHOW OH H HO (15) Scheme 3 of glucose rather than glucosamine into the deoxystreptamine residue (D) was unexpected and necessitated the proposal of a new pathway for deoxystreptamine biosynthesis.It and related amino-cyclitols are important constituents of several antibiotics.47 Incorporations of [13C ]-labelled acetates and valines into cephalosporin C (16) are as expected from 14Cstudies. The higher degree of labelling with [2-13C] acetate of C-14 compared with C-11, C-12, and C-13, which were similar, suggested the formation of the a-aminoadipyl side-chain from a-ketoglutaric acid and acetyl coenzyme A.4s The stereoselective incorporation of 13Cin (2RS,3R)-[4J3C]valine into C-249 and of (2RS,3S)- [4-13C]valine into C-17 of has been demonstrated. The (3R)-isomer specifically labels the C-2 @-methyl group of penicillin V (17) in Penicillium chrysogenum51 and the (35')- isomer labels the a-methyl group.52 H Scheme 4 17 W.T. Dobranzki, Chem. in Britain, 1974, 10, 386. Neuss, C. H. Nash, P. A. Lemke, and J. B. Grutzner, J. Amer. Chem. SOC.1971, 93, 2337; Proc. Roy. SOC.,1971, B179,335. 49 N. Neuss, C. H. Nash, J. E. Baldwin, P. A. Lemke, and J. B. Grutzner,J. Amer. Chem. Soc., 1973,95, 3797. 6o H. Kleunder, C. H. Bradley, C. J. Sih, P. Fawcett, and E. P. Abraham,J. Amer. Chem. SOC., 1973,95, 6149. 61 P. A. Lemke, C. H. Nash, and S. N. Pieper, J. Gen. Microbiol., in the press. 6* D. J. Aberhart and L. J. Lin, J.C.S. Perkin I, 1974, 2320. 511 Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies 13C N.m.r. has been used extensively in biosynthetic studies of porphyrins and vitamin BIZ (18).Early work on the incorporation of 13C-labelled por- phobilinogen, 6-amino-laevulinic acids, and Urogens I-TV has been reviewed.6 More recently, three groups have independently shown that seven of the eight methyl groups of vitamin BIZ are enriched by feeding [lWImethionine to Propiunibacterium shermanii.53 The methyl group not enriched is one of those at C-12. Battersby has suggested this to be the 12p; degradation of vitamin BIZ gives the imide (19) in which the methyl 1H n.m.r. resonances have been assigned. 42 Me, i C02H IHYCO CONH2 (19) HOCHZ Only the signal due to the 12a-methyl showed enhanced 13C satellites, Scott is in agreement with this, arguing that the C-2, C-7, C-12a, and C-17 methyls should have similar chemical shifts owing to the yeffectM of the syn-propionate side- chain whereas that at C-12p, lacking this, should be shifted downfield.Good support was obtained for this analysis from the 13C n.m.r. spectrum of the enriched vitamin BIZ after epimerization at (2-13 when one of the enriched signals, presumably the C-l2a, moved downfield by ca. 12 p.p.m. Shemin and Katz have reached the opposite conclusion using specific proton decoupling to correlate the enriched 13C resonances with the lH resonances. However, doubt has been cast on the crucial lH assignments in this case.55 69 (a)A. R. Battersby, M. Ihara, E. McDonald, J. R. Stevenson, and B. T. Golding, J.C.S. Chem. Comm., 1973, 404; ibid., 1974, 458; (b) A.I. Scott, C. A. Townsend, and R. J. Cushley, J. Amer. Chem. SOC.,1973,95, 5759; (c) C. E. Brown, D. Shemin, and J. J. Katz, J. Biol. Chem., 1973, 248, 8015. b4 D. K. Daling and D. M. Grant, J. Amer. Chem. SOC.,1972, 94,9318. 66 E. McDonald, Ann. Reports (B), 1974, 70, 597. Simpson l4C Studies clearly indicate that the mechanism of pyrrole-ring formation in prodigiosin (20) is unrelated to that operative in the porphyrins and so is of special interest. A complete assignment of the 13C spectrum of prodigiosin has been made,56 enabling 13C biosynthetic studies to be carried out. 13C-Labelled acetates, alanine, proline, glycine, and serine have been incorporated by cultures of Serratia marcescens (Scheme 5), allowinga biosynthetic path to be proposed.57 H Scheme 5 Recently metacycloprodigiosin and undecylprodigiosin have been isolated, and 13C studies indicate a similar biosynthesis.58 Specific proton decoupling and PRFT methods have been used to assign the 13C n.m.r.spectra of cytochalasin B (21) and cytochalasin D (22). Incorporation of sodium [1J3C]- and [2-W confirms previous proposals of bio- synthesis of the cytochalasins from phenylalanine, methionine, and a CIS or cl6 polyketide (Scheme 6). 6 lSC-W Spin-Spin Coupling In natural-abundance 13C n.m.r. spectra 13C-13C spin-spin coupling is not observed as the probability of 13C nuclei being adjacent is equal to the square of the natural abundance, giving satellites of 0.55 % of the intensity of the main signal.However, in enriched material the probability is much higher and the detection of a 13C-W coupling can provide conclusive evidence that two labels have been incorporated into adjacent positions in a molecule. A. Singly Labelled Precursors.-A l3C-W coupling can arise from administra- tion of singly labelled precursors in a variety of ways. 66 R. J. Cushley, D. R. Andersen, S. R. Lipsky, R. J. Sykes, and H. H. Wasserman, J. Amer. Chem. Soc., 1971,93, 6284. *' H. H. Wasserman, R. J. Sykes, P. Peverada, C. K. Shaw, R. J. Cushley, and S. R. Lipsky,J. Amer. Chem. SOC.,1973, 95, 6874. 6* H. H. Wasserman, R. J. Sykes, C. K. Shaw, and R. J. Cushley, Tetrahedron Letters, 1974, 2787. 6s W. Graf, J.-L. Robert, J. C.Vederas, C. Tam, P. H. Solomon,I. Miura, and K. Nakanishi, Helv. Chim. Actu, 1974, 57, 1801. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies 0 (21) (22) Scheme 6 Molecular rearrangement of a biosynthetic intermediate may give rise to a W-W coupling. The pyrone (23), a metabolite of Aspergz'ZZus quercinus, when enriched from [2JSC]acetate shows a coupling of 61 Hz between C-2 and C-7 (Figure 6b). This coupling probably arises from an intramolecular rearrangement of the precursor polyketide chains0 (Scheme 7). A similar coupling from head-to- head linkage of acetate units is observed on incomoration of [2-lsC]acetate into sterigmat ocystin (24).61 Folding of a pentaketide chain to give a five-membered ring results in a lsWsC coupling between C-5 and C-4a of dihydrolatumicidin (25) enriched from [2-1SC]acetate.62 Folding of a terpenoid chain enriched from [l-WIacetate gives rise to a coupling between C-1 and C5 in helicobasidin, (Figure 5), and between C-8 and C-14 in fusidic acid (26).The head-to-head linkage of farnesyl units, via squalene, gives rise to the coupling observed between C-11 and C-12.& Conversion of [2J*C]acetate into succinate in the Krebs cycle results in a (O T. J. Simpson, Tetrahedron Letters, 1975, 175. M. Tanabe,T. Hamasaki, H. Seto, and L. Johnson, Chem. Comm., 1970, 1539. H. Seto, T. Sato, H. Yonehara, and W. C. Jankowski, J. Antibiotics, 1973, 26, 609. Sitnpson L d 4 7 IL I I I I I50 100 50 0 Figure 6 lH Noise-decoupled 13Cn.m.r.spectra of pyrone (23) from (b) lSCHSCO,Na,and (c) 1SCHs13C0,Na 1sC-13C coupling between C-1 1 and C-15 in avenaciolide (27).83 Similar meta- bolic transformations give rise to 13C-l3C couplings when [2-13C]glycine is ‘3 M. Tanabe, T.Hamasaki, Y. Suzuki, and L. F. Johnson, J.C.S. Chem. Comm., 1973, 212. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies CY3A i. A ii, Reduction * HO iii, -H*O Scheme 7 OH 0 OMe 0 Y incorporated via serine into prodigiosin and [5-13C]-8-aminolaevulinicacid is incorporated into vitamin BIZvia porphobilinogen.64 B. Doubly Labelled Precursors.-This has been without doubt one of the most important developments in biosynthetic studies in recent years.Most studies have used [1,2-13C]acetate in which both the carboxyl and methyl carbons are highly enriched. This means that with this precursor all acetatederived atoms are "C. E. Brown, J. J. Katz, and D.Shemin, Proc. Nut. Acad. Sci. U.S.A., 1972, 69,2585. Simpson *e it. CH2-C02H CH3C02Na . + CHz -C02H labelled and adjacent atoms derived from incorporation of an intact acetate unit will exhibit a 13C--13C coupling. Generally, no coupling will be observed between adjacent units owing to the low probability of more than one added precursor unit being incorporated into any one metabolite molecule. This coupling can be of great use in structural and spectral assignment studies in addi- tion to providing biosynthetic information. The first application of this technique was to dihydrolatumicidin (25).On incorporation of [1 ,2-13C]acetate all 10 carbons exhibited 13C-13C couplings, confirming its origin from five acetate units.65 Feeding of a 50 :50 mixture of [l-l3C]acetate and [2-W]acetate gave rise to couplings for the carbon-carbon bonds between adjacent acetate units. Although four combinations between these singly labelled acetates are possible, only one (--CH313C0 -13CH3C0-.) gives the desired coupling. This latter technique requires high incorporations and so is of limited use. The size of the 13C--l3C coupling, generally 30-90 Hz, is related to the hybridization of the atoms involved,8 increasing with increased ‘s’ character, and so in conjunction with chemical shift data is an important source of structural information.Tenellin (28), a metabolite of Beauvaria sp., has been studied using [1,2-W]- acetate to provide both structural and biosynthetic information. An interesting feature of this study was growth of the organism using K15N03 as the sole nitro- gen source, resulting in 13C-15N couplings on C-6 and C-2. [WIMethionine indicated the origin of the C-10 and C-12 methyl groups and feeding of [1-1sC]- 65 H. Seto, T.Sato, and H. Yonehara, J. Amer. Chem. SOC.,1973,95, 8461. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies and [2-13C]-phenylalanine proved the origin of the remainder of the molecule and indicated that a carboxy-carbon migration must take place during bio- synthesis.66 0 0 PhCHZCH- CO2H HO AH2 * t CH3SCH&H&H-CO,cHI + CH3 NH2 * .I CH3-COeNa OH (28) A major advantage of doubly labelled precursors is in the elucidation of anomalous biosynthetic pathways. If the biosynthesis involves cleavage of an original acetate unit, the 13C-13C coupling is lost and the respective carbons now appear merely as enhanced singlets. When [1,2-13C]acetate is incorporated into pyrone (23), three of the nine carbons appear as singlets; the remaining six all exhibit 13C-13C coupling (Figure 6c), suggesting biosynthesis via cleavage of a pre-formed carbocyclic ring as in Scheme 7. The lack of couplings on C-1 and C-1 1of mollisin (29) enriched with [1,2-13C] acetate was interpreted in favour of a two-chain derivations' as in path (b) of Scheme 8; path (a) had been suggested on the basis of proton satellite studies 66 A.G. McInnes, D. G. Smith, C.-K. Wat, L. C. Vining, and J. L. C. Wright, J.C.S. Chern. Comm., 1974, 281 ;ibid., p. 283. 13' H. Seto, L. W. Cary, and M. Tanabe, J.C.S. Chem. Comm., 1973, 867. Simpson with singly labelled acetate. However, cleavage of a single polyketide chain, path (c), would give similar results and cannot be excluded. l3C-W Couplings from incorporation of [l,2J3C]acetate aided in the placing of the substituents on the tetronic acid ring of multicolic acid (30), a metabolite of PeniciZZium muZticoZor.68 The labelling pattern from [1-13CI- and [2-13C] acetate was not as expected from the normal pathway of fungal tetronic acid biosynthesis and suggested formation via ring cleavage of n-pentylresorcylic acid (31).The absence of couplings on C-1, C-3, and C-11 on feeding [1,2-13C] acetate confirmed this hypothesis. im CH3-COeNa --b (31) (30) Incorporation of [lY2-13C]acetate into ascochlorin (32) in Nectria coccinea resulted in only five 13C-13C couplings in the triprenyl side-chain, showing that the methyl from C-6 rather than from C-2 of mevalonate migrates during biosynt hesis.69 CI It has been suggested that the cyclopentenol (33), a metabolite of Periconia macrospinosa,and related fungal cyclopentenones are formed by ring contraction of a benzenoid precursor.70 The incorporation of singly and doubly labelled [Wlacetate shows that this is the case, but the labelling pattern obtained indi- cates a different mechanism from that pr0posed.7~ Oa J.A. Gudgeon, J. S. E. Holker, and T. J. Simpson, J.C.S. Chem. Comm., 1974, 636. 69 M. Tanabe and K. T. Suzuki, J.C.S. Chem. Comm., 1974,445. 70 W. B. Turner, ‘Fungal Metabolites’, Academic Press, London, 1971, p. 126. 7l J. S. E. Holker and K. Young, J.C.S. Chem. Comm., 1975, 525. Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies The use of a doubly labelled precursor other than acetate is illustrated by the elegant work of Battersby on the biosynthesis of uroporphyrinogen I11 (35).72 [2,1l-lsC]-PBG (34) was synthesized and showed a long-range coupling of ca. 4 Hz. This was incorporated into (35) using a cell-free enzyme system from avian blood.Analysis of the resultant 13C n.m.r. spectrum showed three doublets, A = CHzCO2H P =CHzCH2C02H PYP (34) (35) each of 5 Hz splitting, corresponding to the a-,p-, and 6-carbonsY and one doub- let of 72 Hzfor the y-carbon, indicating that PBG unit D must undergo an intramolecular rearrangement with respect to itself during biosynthesis. To date, the only other doubly labelled precursor that has been employed is [4,5-W]mevalonate. This was synthesized by Hanson and incorporated into the fungal sesquiterpenes cyclonerodiol (36a) and cyclonerotriol (36b). All three sets of 13C-J3C couplings are retained, showing that mevalonate is incorporated without rearrangement.73 Me CHgR 7 Higher Plant Metabolites All the above 13C studies have involved micro-organisms or partially purified enzyme systems.The main problems with higher plants are (a)the low incorpora- tions generally obtained and (b)dilution of the labelled metabolite by unlabelled endogenous material, which can be so large that the l3C content of the isolated substance does not differ significantly from natural abundance despite good incorporation. Despite these difilculties, two successful studies have been reported. 11 A. R. Battersby, E. Hunt, and E. McDonald, J.C.S. Chem. Comm., 1973, 442. 1s J. R. Hanson, personal communication. Simpson 1%-Labelling studies indicated that the lactone (37) was an intermediate in the biosynthesis of camptothecin (38) in Camptotheca a~uminata.'~ Owing to the absence of suitable degradations, l3C n.m.r.was used to prove specificity of incorporation. [ 1-13ClTryptamine was synthesized from K13CN and thereby the [5-13C]-lactone (37), 38 mg of which was wick-fed to intact plants. After two days growth, 20 mg of camptothecin was isolated which showed an enhance- ment of only the C-5 resonance intensity of ca. 55 %. 3 (37) (38) [1-13CJAutumnaline(39) has been synthesized and injected as the hydro- chloride (300 mg) into seed capsules (1 mg per capsule) of Colchicum autumnale (autumn crocus). After two weeks growth 1.24 g colchicine (40) was isolated. The resultant 13C n.m.r. spectrum showed ca. 2.5-fold enrichment of the C-7 signal only.75 Me 0 Me ---+ Meo%-rMe0 Me0Hoq MeO OH 0 OMe OMe However, the general extension of l3C n.m.r.biosynthetic studies to higher- plant metabolites is likely to prove difficult. This may be alleviated by the use of cell-free enzyme systems in which the problems of penetration of precursors to the active site and dilution by endogenous material are eliminated. Work with 14C-labelled material indicates that the dilution values and yields obtainable make 13C studies feasible.76 Similar considerations apply to studies using tissue cultures. A second possibility is the use of 12C-labelled compounds completely flushed of 13C which are becoming available as a by-product of 13C-enrichment and so are 74 C. R. Hutchinson, A. H. Heckendorf, P. E. Doddona, E.Hagaman, and E. Wenkert, J. Amer. Chem. SOC.,1974, 96, 5609. 76 A. R. Battersby, P. W. S. Sheldrake, and J. A. Milner, Tetrahedron Letters, 1974, 3315. D.H. Bowen, J. MacMillan, and J. Graebe, Planta, 1972, 102, 261. 521 Carbon-13 Nuclear Magnetic Resonance in Biosynthetic Studies relatively inexpensive. Their use in mechanistic77 and biosynthetic studies78 has already been advocated; elimination of a peak intensity rather than enhance- ment would be observed. Though their direct use in biosynthetic studies is impractical owing to the high enrichments that would be required to obtain significant results,79 they may be of use. Scott has suggested growth of a plant from seed or tissue culture in an atmosphere of l2CO2 which should ensure a very low (0.1 to 0.2 %) 13C content in the various pools of organic intermediates.80 Thus the l3C ‘natural abundance’ is lowered by an order of magnitude, making subsequent 13C-enrichment studies possible.It is perhaps noteworthy at this stage that it is only the fact that 13C natural abundance is so relatively low that makes any biosynthetic studies possible at all, despite making spectra difficult to determine. 8 Conclusions Besides the examples given many other metabolites have been studied. These studies include the incorporation of [Wlacetate into ochratoxin,sl palmiteolic acid,82 niob0mycin,~3 maleirny~in,~~ showdomycin,85 epoxydon,88 and thermo- zymocidin,87of [1 ,2J3C]acetate into penicillic acid,88 ovalicin,8Q sterigmatocystin,90 and bika~erin,~~ of [2-13C]mevalonate into gibberellic acid23 of DL-tryptophan- [3-13C]alanine into pyrrolnitrin,g2 and of [l-lsCJglycerate into rifamycin.93 There can be no doubt that the use of 13C methods will continue to expand rapidly, perhaps even superseding 14C in biosynthetic studies of micro-organisms. The use of doubly labelled precursors makes simultaneous determination of structure and biosynthesis possible for the first time and makes possible studies where classical 14C methods could not provide unequivocal answers. The extension of the method to higher plants and metabolic studies in man where the risk attached to the use of radio-isotopes is removed, seems imminent.77 J. Prestien and H. Gunther, Angew. Chem. Internat.Edn., 1974, 13, 276. S. B. W. Roeder, J. Magn. Resonance, 1973, 12, 343. 78 T. J. Simpson, J. Magn. Resonance, 1975, 14, 262. A. I. Scott, Science, 1974, 186, 101. Y.Maebayashi, K. Miyaki, and M. Yamazaki, Chem. and Pharm. Bull. (Japan), 1972,20, 2172. 8a A. L. Burlingame, B. Balogh, J. Welch, S. Lewis, and D. Wilson, J.C.S. Chem. Comm., 1972, 318. N. M. J. Knoll, R. J. Huxtable, and K. L. Rinehart, J. Amer. Chem. Suc., 1973, 95, 2704. s4E. F. Elstner, D. M. Carnes, R. J. Suhadolnik, G. P. Krushmen, M. P. Schweizer, and R. K. Robins, Biochemistry, 1973, 12,4992. 86 E. F. Elstner, R. J. Suhadolnik, and A. Allerhand, J. Bid. Chem., 1973, 5385. 86 K. Nabeta, A. Ichihara, and S. Sakamura, J.C.S. Chem. Comm., 1973,814. 87 F. Aragozzini, M. G. Beretta, G. S.Ricca, C. Scolastico, and F. W. Wehrli, J.C.S. Chem. Comm., 1973,788. H. Seto, L. W. Cary, and M. Tanabe, J. Antibiotics, 1974, 27, 558. M. Tanabe and K. T. Suzuki, Tetrahedron Letters, 1974,4417. H. Seto, L. W. Cary, and M. Tanabe, Tetrahedron Letters, 1974, 4491. A. G. McInnes, D. G. Smith, J. A. Walter, L. C. Vining, and J. L. C. Wright, J.C.S. Chem. Comm., 1975,66. ns L. L. Martin, C.-J. Chang, H. G. Floss, J. A. Mabe, E. W. Hageman, and E. Wenkert, J. Amer. Chem. SOC., 1972,94,8942. @3R. J. White and E. Martinelli, F.E.B.S. Letters, 1974, 49, 233.

ISSN:0306-0012    

DOI:10.1039/CS9750400497

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Photochemistry of Organic Sulphur Compounds By J. D. Coyle DEPARTMENT OF PHYSICAL SCIENCES, THE POLYTECHNIC, WOLVERHAMPTON WV1 1LY 1 Introduction This review forms part of an attempt to give a systematic description of the characteristic photochemical reactions of organic compounds on the basis of their functional groups. Introductory surveys have been presented previously for carbony11 and for olefinic2 chromophore systems. Here are described the photoreactions of several groups of compounds with sulphur-containing chromophores. 2 a-Bonded Compounds Although there are similarities between the photochemistry of thiols, sulphides, and disulphides and that of their oxygen analogues (alcohols, ethers, and per- oxides), differences arise (i) because the C-S bond is weaker than the C-0 bond, (ii) because alkoxy (RK-00) and thiyl (R3G-S-) radicals behave differently in the secondary reactions, and (iii) because the excited states are different in energy.The sulphur compounds have lower-energy n + (T*electronic transitions because of the higher energy of the sulphur non-bonded electrons (Amax for MeSH, MeSMe, and MeSSMe is at 228,212, and 255 nm respectively; cf. values of 184, 184, and 215 nm for MeOH, MeOMe, and MeOOMe). Thiols undergo homolytic S-H bond cleavage on irradiation (254 nm), despite the fact that the C-S bond is much weaker than the S-H bond. This is often encountered in excited-state reactions, that a bond to hydrogen cleaves in preference to a weaker bond, and it implies that dissociation is faster than randomization of energy.Methanethiol (MeSH) yields hydrogen (@ = 0.83), dimethyl disulphide (@ = 0.99), and methane (@ = 0.16) from the radicals obtained on cleavage.3 Anti-Markovnikov addition of thiols to alkenes4 can be of value in the preparation of sulphides (l), and ultraviolet radiation is often the J. D. Coyle and H. A. J. Carless, Chem. SOC.Rev., 1972, 1, 465. a J. D. Coyle, Chem. SOC.Rev., 1974, 3, 329. L. Bridges and J. M. White, J. Phys. Chem., 1973, 77,295. R. M. Kellogg, in ‘Methods in Free Radical Chemistry’, ed. E. S. Huyser, Marcel Dekker, New York, 1969, Vol. 2, p. 17; F. W. Staceyand J. F. Harris, Org. Reactions, 1963,13,150. Photochemistry of Organic Sulphur Compounds most convenient initiator. Heterocyclic sulphur compounds can be formed by an analogous intramolecular reaction (2).5 Dialkyl sulphides undergo primary G-S cleavage on direct or sensitized irradiation.Dimethyl sulphide (MeSMe) gives ethane (@ = 0.47) and dimethyl disulphide (@ = 0.37) by dimerization of Me. and MeS- radicals.6 With thiirans7 extrusion of sulphur often occurs (3), and in a few simple cases the 80% + cis-isomer relatively long-lived intermediate biradical (e.g. CHzCHzS0) has been trapped by added alkene. S-Aryl or S-alkyl thiocarboxylates give fragmentation products* by C-S cleavage (4) ;no photo-Fries rearrangement occurs (cf. aryl carboxylates). Ar S-COMe hv (254nm)> ArSSAr $. ArSMe (4)77% 7% (Ar=p.-MeC,H,) Both C-S and S-S homolytic cleavage occur on irradiation of dialkyl disulphides with short-wavelength U.V. radiation, but with lower-energy radiation C-S cleavage is’important only if stabilized carbon radicals are produced.Hence primary and secondary dialkyl disulphides are relatively photostable in solution towards 254 nm radiation because S-S cleavage leads to radicals which re- combine. Tertiary dialkyl disulphides give products by C-S cleavage(5), and the alkyl radicals can be trapped (6).9 J.-M. Surzur, R. Nouguier, M.-P. Crozet, and C. Dupuy, Tetrahedron Letters, 1971,2035. D. R. Tycholiz and A. R. Knight, J. Anfer. Chem. SOC.,1973,95, 1726. ? A. Padwa, Internat. J. Surfur Chem. (B), 1972, 7, 331. J. R. Grunwell, N. A. Marron, and S. I. Hanhan, J.Org. Chem., 1973,38, 1559. G. W. Byers, H. Gruen, H. G. Giles, H. N.Schott, and J. A. Kampmeier, J, Amer. Chem. Soc., 1972, 94, 1016. Coyle hv,CCI 4 BU' S-SBU' > Bu'Cl go%, 'p= 0.33 (6) 3 Thiocarbonyl Compounds Thiobenzophenone shows three U.V. absorption bands, at 599 (w), 316.5 (s), and 235 (m-s) nm, and the lowest-energy excited states are (n,w*). These species have an unpaired electron in a non-bonding orbital on sulphur, and their reactions are similar to those of thiyl radicals (hC-S.), the major types involv- ing hydrogen abstraction by sulphur or cycloaddition with unsaturated com- pounds. Unlike carbonyl excited states, thiocarbonyl excited states do not undergo a-cleavage, because there is not enough energy available to break a C-C bond; also the photoreactions of thiocarbonyl compounds often feature their ability to act as radical traps, which carbonyl compounds cannot do.Thiobenzophenone is photoreduced by propan-2-01,lO in the first instance to give diphenylmethanethiol (which is not a product of the thermal reaction) through initial hydrogen abstraction (7). hvPh2C=s 3-Me2CH-OH __3 Ph2e--SH $-Me&--OH 3/ MeaC=O + Ph2CH-SH (7) 88% 63% Intramolecular abstraction in 0-alkyl thiobenzoates leads to Norrish Type 2 elimination products (8).11 However, the preferred reaction in thioketonesl2 S Ph-C-OCH2CH2PhII hv PhCS-OH + PhCH=CH2 (8)(Pyrex) involves abstraction from the &position (9). This is a reaction of an upper excited state, and with long-wavelength light no reaction occurs.If there is no &hydrogen lo A. Ohno and N. Kito, Internat. J. Surfur Chem. (A), 1971, 1,26. l1 J. Wirz, J.C.S. Perkin II, 1973, 1307. la P. de Mayo and R. Suau, J. Amer. Chem. SOC.,1974,96,6807. 525 Photochemistry of Organic Sulphur Compounds no reaction takes place unless the C-H bond in they- or €-position is weaker than normal. The (n, T*)state then undergoes y-hydrogen abstraction, and this can be accompanied by €-abstraction with higher-energy radiation (10). S II hvPh-C -CMe2CH,CH20Et ____) (>445 nm) OEt Thiocamphor and similar rigid compounds undergo intramolecular /3-hydrogen abstraction (1l),13 and this emphasizes the wide range of internal abstractions which are commonly encountered with thioketones.Photocycloaddition of thiocarbonyls with alkenes gives thietans and 1,4-dithians.14 The (n,T*) thiocarbonyl excited state reacts with electron-rich alkenes to give both products, and the ratio depends on steric factors and on the concentration of thioketone, as expected for a reaction through a biradical inter- mediate which can be trapped by ground-state thioketone (12). The reactions with electron-deficient alkenes are more complex. A higher-energy excited state of thiobenzophenone is reactive, and a thietan is formed PhCR=CH2 major if R=H Ph and rPh2CSI R is high lS D.S. L. Blackwell and P. de Mayo, J.C.S. Chem. Comm., 1973,130. l4 A. Ohno, Internat. J. Suwur Chem. (B), 1971, 6, 183. CoyZe stereospecifically (1 3).15 Sometimes, as with thiobenzophenone and dichloro- ethylene, the (n,n*) state is unreactive; sometimes, as with thiobenzophenone 83 % and dimethyl maleate,16 both states are reactive but the (n,T*)gives thietan non- stereospecifically.With acrylonitrile,17 the thietan formed with shorter-wave- length radiation arises from thermal decomposition of a 1,3-dithian (14), of a J. warm Ph CN Q, up to 0.096 type which has been isolated at room temperature in other systems. With longer-wavelength radiation the expected thietan and 1,4-dithian are produced, together with a product arising from intramolecular attack on a phenyl ring in the biradical intermediate [similar to that obtained with alkynes; see (15)].1,3-Dienes react with thiobenzophenone to give 3-vinylthietans, which pre- dominate over the thermal Diels-Alder adducts (4-thiacyclohexenes) if the irradiation is carried out at low temperat ure.16 Alkynes lead to isothiochromenes (15) by internal attack in the biradical intermediate.18 The cyclodimerization of alkyl thiocarbonyl compounds to 1,3-dithietans is brought about by irradiation (16);19 it also occurs in an acid-catalysed thermal reaction. l6 A. Ohno, Y. Ohnishi, and G. Tsuchihashi, Tetrahedron Letters, 1969, 161. H. Gotthard, Chem. Ber., 1972,105,2008. P. de Mayo and H. Shizuka, J. Amer. Chem. SOC.,1973,95, 3942. A. Ohno, T. Koizumi, and Y. Ohnishi, Bull. Chem. SOC.Japan, 1971,44,2511. lo J. J. Woman, M. Shen,and P. C. Nichols, Canad.J. Chem., 1972,50, 3923. Photochemistry of Organic Sulphur Compounds Ph Ph Ph2C.S + PhC CH -(589nm) Ph Ph Ph Ph CH? Ph CH,Ph 4 Five-membered Aromatic Heterocycles Five-membered heteroaromatic sulphur compounds give products on irradiation in which the atoms of the ring are transposed. One main group of isomerizations, typical of 2-aryl- or 2-alkyl-thiophens (17),20 involves a transposition of positions 2 and 3 of the ring (2 e3 interchange). An intermediate cyclopropenyl thio- carbonyl compound can be envisaged, analogous to the aldehyde or ketone isolated from irradiation of some furans. The fact that pyrroles are formed if the reaction is carried out in the presence of amines supports this view.20b A second main group of isomerizations involves a shift of the sulphur atom around the ring (equivalent to 2 F1 4 or 2 F1 4/3 ;FI 5 transpositions), and these are often encountered with thiazoles (18)21 or isothiazoles, as well as with 3-arylthiophens.Electrocyclic ring-closure followed by one or two 1,3-shifts of the sulphur atom and ring-opening could account for these changes, and a bicyclic compound of this kind has been isolated after irradiation of tetrakis(trifluoromethyl)thiophen.22 *O (a) H. Wynberg, R. M. Kellogg, H. van Driel, and G. E. Beekhuis, J. Amer. Chem. Suc., 1967,89, 3498, 3501 ; (b) A. Couture, A. Delevallee, and A. Lablache-Combier, Tetra-hedron, 1975, 31,785. *I C. Riou, G. Vernim, J. J. M. DOU, and J. Metzger, Bull. Sue.chim. France, 1972,2673. s.* H. A. Wiebe, S. Braslavsky, and J. Heicklen, Canad.J. Chem., 1972, 50,2721. Coyle bP,-[ePhS = p1 80 % Other types of intermediate have been proposed, particularly those incorporat- ing a bicyclobutane unit (as in the benzvalene bond-isomer of benzene). An ylide such as (19) can account simply for a 2 3 transposition, whilst a zwitterion of type (20) can lead to migration of the sulphur atom. A second type of ylide (21) Photochemistry of Organic Sulphur Compounds allows a direct double interchange 2 4/3 @ 5, and it could also lead to a trans- formation in which only one atom (adjacent to the sulphur) remains in the same position after rearrangement. A neutral bond-isomer of the benzvalene type can explain the 3 Ft 4 trans-position which is sometimes observed with 3-arylthiophens (22).20 Evidence that carbanionic carbon is present in the intermediate(s) in thiazole transformations comes from the incorporation of deuterium from DzO,23 but overall it seems that either several mechanisms operate for thiophens and related compounds or the various types of intermediate are readily interconverted.5 Sulphur-Oxygen Compounds I Compounds containing the O=S=O unit, such as sulphonyl halides (e.g.I PhCH2S02I) or sulphones (23),24 lose sulphur dioxide on photolysis. Ana-cleavage 0 11 hY (254nm) PhSO2. + Ph* *Ph-Ph (-k PhS02H)PhSPh C6H6 ' (23)It 0 71% 7% mechanism seems reasonable, since elimination of SO2 is much more efficient when stabilized radicals are formed (24).25 R= H no reactionhv (-300nm) R = Ph efficient reaction R 02 M.Maeda and M. Kojima, Tetrahedron Letters, 1973, 3523. c4 M. Nakai, N. Furukawa, S. Oae, and T. Nakabayashi, Bull, Chem. SOC.Japan, 1972, 45, 1117. E6 P. M. Weintraub, Chem. and Ind., 1970, 1296. Coyle I The S=O unit similarly gives rise to a-cleavage on irradiation. Sulphur mon- I oxide is not eliminated, but a-cleavage can account for the photoracemization of sulphoxides (25),26 for their breakdown (26),27and for reactions of sulphinate esters (27).28 0 0 II PhCOCH,-S-Me % PhCOCH2CH,COPh + MeS0,SMe (26) 72 % 52 % 0 II hvMeC,H, -S-OCH,Ph --+ MeC6H4SO* PhCH,O* MeC6H4S02SC,H4Me 47 % Some cyclic sulphoxides undergo photochemical desulphuration.Sensitized irradiation leads to a ring-opened ketone (28), whilst direct irradiation gives a cyclic ether (29) as well.29 In both instances initial C-S(=O) cleavage probably occurs. aE R. S. Cooke and G. S. Hammond, J. Amer. Chem. SOC.,1968,90,2958. 27 H. Nozaki, T. Shirafugi, K. Kuno, and Y. Yamamoto, Bull. Chem. SOC.Japan, 1972, 45, 856. 28 M. Kobayashi, H. Minato, Y.Miyaji, T. Yoshioka, K. Tanaka, and K. Honda, Bull. Chem. SOC.Japan, 1972,45,2817. aD A. G. Schultz and R.H.Schlessinger, Tetrahedron Letters, 1973,3605. 531 -> Photochemistry of Organic Sulphur Compoundsph R,l pphPhzCO and/or Ph Ph ‘S 0II So\\0

ISSN:0306-0012    

DOI:10.1039/CS9750400523

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Photodegradation and Stabilization of Commercial Pol yolefms By N. S. Allen and J. F. McKellar DEPARTMENT OF CHEMISTRY AND APPLIED CHEMISTRY, SALFORD UNIVERSITY, SALFORD, LANCS., M5 4W’I 1 Introduction Polyolefin photochemistry has recently been discussed in a number of papers, reviews, and textbooks,lY2 and it is evident from these that there are several con- flicting views on the nature of the mechanisms involved. Broadly speaking, polyolefin photochemistry can be divided into three major areas (a)photo-degradation (or oxidation), (b) sensitized photodegradation (or oxidation), and (c) photostabilization. Although all three areas present problems of considerable technological interest, (b) and (c) are currently of particular interest because of the ever increasing problem of environmental pollution by plastics litter. The commercial polyolefins are low- and high-density polyethylene, poly- propylene, and poly (4-methylpent-l-ene).Although the end use of the polymer is the main determining factor, in general the major technical photochemical problems are concerned with the sensitized degradation of the polyethylenes and the stabilization of the other two. 2 Photodegradation A. Absorption and Emission of Light by Polyolefins.-Because of the screening effect of the upper atmosphere, no radiation of wavelengths shorter than about 290 nm reaches the earth‘s surface.2 Consequently ‘pure’ saturated polyolefins should be unaffected by exposure to normal sunlight. However, it is now gener- ally believed that it is the presence of certain impurities in the commercial poly- olefins, presumably introduced during polymerization and/or processing, that results in light absorption by the polymer and that may initiate photochemical reaction.The main light-absorbing species are believed to be carbonyl groups, present either in definite impurity compounds or as groups attached to the R. B. Fox, Pure Appl. Chem., 1972,30,87; J. F. Rabek and B. Ranby, ‘Photodegradation, Photooxidation and Photostabilization of Polymers’, Interscience, New York, 1975, and references cited therein. * For recent reviews see D. M. Pinkerton, Proc. Roy. Austral. Chem. Inst., 1972, 33; 0. Cicchetti, Adv. Polymer Sci., 1970, 7, 70; R. B. Fox,Progr. Polymer Sci., 1972, 1, 47; S.L. Fitton, R. N. Hayward, and G. R. Williamson, Brit. Polymer J., 1970, 2,217; Yu. A. Ershov, S. I. Kuzina, and M. B. Neiman, Rum. Chem. Rev., 1969,38, 12; A. King, Plastics and Polymers, 1968, 195. Photodegradation and Stabilization of Commercial Polyolefns polymer chain,+’ aromatic compounds,3*8 metallic impuritie~,~ dienes,lO trienesll, hydroperoxides,4J0J2J3 and oxygen-polymer charge-transfer com- ~1exes.l~The commercial polyolefins exhibit both fluorescence and phospho- rescence emi~sions,~~~~~~8 and these are also due to the presence of impurities that may initiate photochemical reaction in the polymer.15-17 B. Initiation and Propagation Mechanisms for Polyolefins.-(i) By Carbonyl Groups. Two main primary photochemical processes are believed to be respon- sible for the carbonyl-initiated photodegradation of polyolefins.Their definition originates from the early work of Norrish on the photochemistry of aliphatic ketones.18 Norrish Type I. This process leads to the formation of free radicals in which the excited singlet or triplet states of the carbonyl group are precursors2J8 [reaction (01. (a) A. Charlesby and R. H. Partridge, Proc. Roy. SOC., 1965, A283, 312; ibid, p. 329; (b) A. Charlesby and I. Boustead, ibid., 1970, A316, 291 ;(c) R. H. Partridge, J. Chem. Phys., 1966, 45, 5, 1679. (a)A. M. Trozzolo and F.H. Winslow, Macromolecules, 1968,1,98; (6)D. J. Carlsson and D. M. Wiles, ibid., 1969, 2, 6, 587, 597. P. J. Briggs and J.F. McKellar, Chem. and Ind., 1967,662; J. Appl. Polymer Sci., 1968,12, 1825; J. C. W. Chien and W. P. Comer, J. Amer. Chem. SOC., 1968, 90, 1001; J. E. Bonkowski, Textile Res. J., 1969, 39,243. N. S. Allen, J. F. McKellar, and G. 0. Phillips, J. Polymer Sci., Part B, Polymer Letters, 1974,12, 253; N. S. Allen, .T. F. McKellar, G. 0. Phillips, and D. G. M. Wood, J. Poly-mer Sci., Part A-I, Polymer Chem., 1974, 12,2647; N. S. Allen, J. Homer, J. F. McKellar, and G. 0. Phillips, Brit. Polymer. J., 1975, 7, 11. K. Tsuji and T. Seiki, Polymer J., 1972, 2, 606; J. Polymer Sci., Part By Polymer Letters, 1972, 10, 139. D. J. Carlsson and D. M. Wiles, J. Polymer Sci., Part B, Polymer Letters, 1973, 11, 759. (a) N. Uri, Israel J. Chew., 1970, 8, 125; (b) P.Richters, Macromolecules, 1970, 3, 262; (c) A. T. Betts and N. Uri, Chem. andznd., 1967,512; (d)C. Kujirai, S.Hashiya, H. Furuno, and N. Terada, J. Polymer Sci., Part A-I, Polymer Chem., 1968, 6, 589. lo M. U. Amin, G. Scott, and L. M. K. Tillakeratne, European Polymer J., 1975, 11, 85. l1 J. F. Heacock, F. B. Mallory, and F. P. Gay, J. Polymer Sci., Part A-I, Polymer Chem., 1968, 6,2921. la L. Balaban, J. Majer, and K. Vesely, J. Polymer Sci., Part A-1, Polymer Chem., 1969, 22, 509. lS D. C. Mellor, A. B. Moir, and G. Scott, European Polymer J., 1973, 9, 219; G. V. Hutson and G. Scott, Chem. and Znd., 1972, 725. l4 D. G. M. Wood and T. M. Kollman, Chem. and Ind., 1972, 423; (b) V. K. Milinchuk, Vysokomol. Soedineniya, 1965,7, 1293; (c) K.Tsuji and T. Seiki, J. Polymer Sci., Part A-1, Polymer Chem., 1971, 9, 3063; J. Polymer Sci., Part B, Polymer Letters, 1970, 8, 817. l5 N. S. Allen, J. F. McKellar, and G. 0. Phillips, Chem. and Ind., 1974, 300. l6 N. S.Allen, J. F. McKellar, G. 0. Phillips, and D. G. M. Wood, J. Polymer Sci.,Part B, Polymer Letters, 1974,12, 242; N. S. Allen, J. F.McKellar, and D. G. M. Wood, J. Poly-mer Sci., Part A-1, Polymer Chem., in the press. l7 (a) A. P. Pivovarov, Yu. A. Ershov, and A. F. Lukovnikov, Plast. Mussey, 1966, 10, 7; Vysok. Energii, 1968,2, 3,220; (b) D. J. Carlsson and D. M. Wiles, Macromolecules, 1974, 7,259; (c) D. J. Harper, J. F.McKellar, and P. H. Turner, J. Appl. Polymer Sci., 1974, 18, 2805. 18 J. G. Calvert and J. N. Pitts, ‘Photochemistry’, New York, John Wiley, 1966; H. J.Heller, European Polymer J., Supplement, 1969, 35. Allen and McKellar h v-CH,-C-CH,-CH,-C-+ *CH2-or -CH;+CO+*CH,-II ll 0 0 (1) Norrish Type 11.This process only occurs when the ketone possesses at least one hydrogen atom on the Ycarbon with respect to the carbonyl group. The re- action occurs via a six-membered cyclic intermediate involving intramolecular hydrogen atom abstraction and produces one olefinic and one enolic group in the polymer2J*J9 [reaction (2)]. A further possible primary process involving photoreaction of carbonyl groups in the polymer is that of intermolecular hydrogen abstraction from adjacent polymer chains. As in the Norrish type I process, this would result in the formation of macroradical centres in the chain2J8 [reaction (3)].Carlsson and Wile@ studied the photolysis of the main ketonic oxidation pro- ducts of polypropylene [i.e. (1) and (2)l. They found that whereas ketone (1) was photolysed by a Norrish type I process to give carbon monoxide and two macroradicals, ketone (2) was photolysed by a Norrish type I1 process to give acetone and an unsaturated polymer chain end [reactions (4) and (5)]. Both pro- cesses had similar quantum yields of 0.08. Carbonyl groups may also participate in the oxidative photodegradation of polyolefins by a less direct route. Trozzolo and Winslowaa postulated a mech- anism for polyethylene photo-oxidation in which the Norrish type I1 process is dominant. The photoexcited triplet carbonyl groups are quenched by triplet ground-state oxygen and the resultant excited singlet oxygen (102) then reacts with the vinyl groups produced by the Norrish type I1 process [reactions (6a-d I.J. E. Guillet, Naturwiss., 1972, 59, 503. Photodegradation and Stabilization of Commercial Potyolejins -CH,CMeCH,-C-CH,CMe 3-CH,PMeCH,* + CH,CMe +CO 790/oI I I IH 0 H H H (4) //O /IV //Me --CH=CN CH,CMeCH,-C \ -k MeCOMe (5)>85%I Me ‘H H or (2) wCH2-C \H O-OH The validity of this mechanism has recently been questioned by the work of Mill et aL2OU It is seen that the reaction scheme predicts the formation of hydro- peroxides. These workers studied the dye-sensitized photo-oxidation of atactic polypropylene in solution using dyes that are well known to generate singlet oxygen on photolysis.However, no hydroperoxide formation was detected. This crucial observation is supported by the recent work of Breck et at.20b They exposed samples of commercial polyolehs to a gas stream containing singlet oxygen and again, no hydroperoxide formation was detected. Studies of the luminescence from commercial polyolefins have made a signifi- cant contribution to our understanding of the mechanisms involved in photo- (a) T. Mill, H. Richardson, and F. R. Mayo, J. Polymer Sci., Part A-I, Polymer Chem., 1973, 11, 2899; (6) A. K. Breck, C. L. Taylor, K. E. Russell, and J. K. S. Wan, ibid., 1974, 12, 1505. Allen and McKellar initiation by carbonyl impurities.Charlesby and Partridgesa first reported the phosphorescence emission from polyolefins and this was attributed to the pre- sence of impurity carbonyl groups attached to the polymer backbone. Later Charlesby and B~ustead~~ identified other phosphorescent impurities in poly- ethylene. They found benzoic acid in the crystalline region and several poly- nuclear aromatic hydrocarbons in the amorphous region of the polymer. Allen et uL6J5J6examined the role of these phosphorescent species in the thermal and photochemical degradation of the commercial polyolehs. The phosphorescent species with excitation wavelengths greater than 290 nm were concluded to be potential photoinitiators. These species were formed only on thermal oxidation of the polymer, extending its light absorption into the U.V.region of natural sunlight. Ketonic/aldehydic type carbonyl impurities were con- cluded to be responsible for this particular emission. Further information on the nature and reactivity of the luminescent carbonyl impurities in polyolefins has come from studies of other commercial polymers. For example, the phosphorescent species in Nylon-6,6 were concluded to be carbonyl groups conjugated with ethylenic unsaturation.21 A similar conclusion was reached fi om phosphorescence studies of the thermal and photochemical oxidation of polybutadiene.22 (ii) By Hydroperoxides. Alkyl hydroperoxides have continuous absorption spectra that extend from the far-u.v. to ca. 320-330 nm in the near-u.v. They are there- fore capable of absorbing sunlight of wavelengths photochemically harmful to the polyolehs.2 Carlsson and Wiles4b studied the photolysis of polypropylene hydroperoxides generated by prior thermal oxidation of the polymer.Ketones (3) and (4) were the main carbonyl species formed [reaction (7)]. These workers found a quantum ? kv 2 N CMeCH,CMeCH,CMc --CMeCH,CMeCH,CMe -1--OH I I I I I 0,H H H 0' H (7)22 2 -CMeCH,C -CH,CMe -CMeCH, + CMeCH,CMe I II I I II I H 0 H H 0 H a1 N. S. Allen, J. F. McKellar, and G. 0.Phillips,J. Polymer Sci.,Part A-I, PoZymer Chem., 1974, 12, 1233, 2623; ibid., in the press. S. W. Beavan, P. A. Hackett, and D. Phillips, European Polymer J., 1974, 10, 925; S. W. Beavan and n.Phillips, J. Photochem., 1974, 3, 349. Photodegradation and Stabilization of Commercial Polyolejins efficiency of about four for hydroperoxide photolysis in polypropylene. Such a high quantum yield would cause appreciable chain scission in the polymer. Carlsson and Wiles postulated that photolysis of tertiary hydroperoxides present in the polymer is a key step in the photodegradation mechanism. In peroxide- free polypropylene, however, they suggested that the photolysis of ketonic impurities produces alkyl radicals that combine with oxygen to give hydro- peroxide groups. Their postulation of hydroperoxide-initiated photodegradation of polypropylene was later supported by evidence from e.s.r. studies.l'b According to Scott and co-workers,l3 the hydroperoxides formed during processing are the main photoinitiators in commercial polyolefins. They found that during processing the most rapid increase in the photodeterioration of polypropylene occurred under the conditions where there was no significant change in carbonyl concentration as monitored by i.r.spectroscopy. From his observations Scott also concludes that the hydroperoxides formed during thermal processing of the polymer are the precursors to carbonyl formation. In contrast to these conclusions on the role of hydroperoxides in polyolefin photodegradation, Mill et aL20afound that when model peroxides were added to pure atactic polypropylene they were ineffective as photosensitizers. For ex- ample, the photolysis of 10-methyl-10-nonadecyl hydroperoxide gave only the analogue of (4) and also the alcohol 10-methyl-10-nonadecanol.There was no evidence for the formation of (3) expected from the mechanism of Carlsson and Wiles.Mill et aL20aattributed the absence of ketone (3) to 'a phase effect' in which the restricted motion of the tertiary polymeric hydroperoxides in an isotactic polypropylene environment would inhibit alcohol formation. Recently, Scottlo has modified his earlier views on the role of hydroperoxides, and now suggests that unsaturated and not saturated hydroperoxides are mainly responsible for the photoinitiated degradation of polyethylene. Only in the later stages of photodegradation does he consider that the Norrish type I1 mechanism of photolysis of a carbonyl group becomes an important process [reaction (S)].*OH ll v R'CH=CH2 -kMeC-C=CH, -R'(CH,),CH,C=CHaType I1II I II I 0 R2 0 R2 (iii)By Aromatic Impurities. Both luminescence and U.V. absorption studies have indicated the presence of aromatic impurities in the polyolefins.3 For example, Allen and McKellar Carisson and Wiles* have shown that naphthalene is present in polypropylene. They have suggested that on photolysis of the polymer any polynuclear aro- matic impurity could efficiently generate singlet oxygen by the quenching of its photoexcited triplet state by ground-state oxygen. The excited singlet oxygen thus produced would then be able to attack any nearby double bonds, forming potentially photoreactive peroxides by reaction (6).However, evidence against this being an important mechanism in photo- degradation comes from e.s.r. studies.23 These have shown that there is a definite suppression of free-radical formation when the polyolefin is doped with poly- nuclear aromatic compounds. (iv) By Metallic Impurities. It is believed that certain metallic impurities, parti- cularly Ti and Fe, can efficiently sensitize the photo-oxidation of poly~lefins.~ These metallic impurities can absorb near-u.v. radiation, causing them to react with the polymer and producing free radicals via electron-transfer processes such as reaction (9). Evidence for this mechanism is given by Kujirai et al.,gd who hV Fe3+OH +Fe2+0H __+ Fe2 + *OH (9) found that the ash residue from a Ziegler-Natta catalyst (composed mainly of Ti02 and A1203) caused photochemical cross-linking in polypropylene. Initially the photo-excited ash residue abstracts a hydrogen atom from the polymer, which then allows cross-linking to occur as shown in reactions (10a-c). If /I v * (ASH) -(ASH) H 2-CH,qMeCH,W +-CH,CMeCH, N I CH, Cm CH, N conditions are such that the macroradicals formed in the polymer can react exten- sively with oxygen then these metallic impurities could act as efficient sensitizers of photo-oxidation.(v) By Oxygen-Polymer Charge-transfer Complexes. Chien24 investigated the 23 T. Takeshita, K. Tsuji, and T. Seiki, J. Polymer Sci., Part A-I, Polymer Chem., 1972, 10, 2315, 3119. a4 J. C. W. Chien, J.Phys. Chem., 1965, 69,4317. 539 Photodegradation and Stabilization of Commercial Polyolefns photo-oxidation of pure alkanes and alkenes. When saturated with oxygen, these hydrocarbons exhibit absorption bands which extend beyond 300 nm into the photoactive wavelength region for the polyolehs. Of the two hydrocarbons studied, the alkenes exhibited the more intense light absorption. To explain these interesting observations in the context of polyoleh photo-oxidation, Chien% proposed the following mechanism, in which the ionic species formed by charge transfer are responsible for the initial light-absorption process (11). Milinchuk14b carried out an e.s.r. study of polyolefins irradiated by light of wavelengths greater than 300 nm.The polymers were irradiated in the presence of oxygen and also under high-vacuum conditions. Only when they were irradiated in the presence of oxygen were free radicals detected. These observations were later confirmed by Tsuji and Seiki,l4 who tentatively attributed free-radical formation to occur via oxygen-polymer charge-transfer complexes. Wood and KoUman14" compared the rate of photo-oxidation of samples of polypropylene powder and sheet which had been stored in the presence and absence of oxygen. The sheet samples had been compression-moulded in air at 220°C. No significant difference was observed in their induction periods. Oxygen-polymer charge-transfer complexes were proposed as the photoinitia- tors. 3 Sensitized Photodegradation Ultraviolet-sensitized photodegradation shows considerable promise as a method of combating environmental pollution by plastics litter.25 Although unstabilizRd polyolefins such as polypropylene and poly(4-methylpent-1-ene)are themselves light-sensitive, if U.V.deterioration by sunlight exposure is desirable then it is importarit to be able to control and indeed monitor this effect. At present there are two promising ways of achieving this aim; one is the use of specially prepared photosensitive polymers and the other appropriate selection of photoactive additives which are incorporated in the commercial polymer dur; ng processing. A. Photosensitive Polymers.-Guillet,19~25 has synthesized cafbonyl-containing polyolefins which will degrade when exposed to outdoor sunlight but will remain intact if kept indoors.The polymers are made by polymerizing the oleh in the presence of variable amounts of carbon monoxide or a vinyl ketone monomer. The resulting copolymer is relatively stable behind window glass. This is because the ketonic groups introduced into the polymer structure do not absorb light of wavelengths greater than about 330 nm. On exposure out of doors, where the light also contains radiation in the region 290-330 nm, the Is Plastics Institute (London) Conference, 'Degradability of Polymers and Plastics', Novem- ber, 1973, and references cited therein. Allen and McKellar same ketonic groups are photoactive and can initiate degradation of the polymer by the Norrish type I1 process [reaction (2)].B. Photoactive Additives.-During the past few years a large number of additives have been claimed in the patent literature to be effective in accelerating the photo- degradation of polyolefins2~25 (prodegradants). Thus only a brief discussion of the distinctive mechanistic features of prodegradants will be given here. A wide range of derivatives of aromatic aldehydes, ketones, and quinones effectively sensitize the photodegradation of polyolefins by primary processes of hydrogen-atom abstraction26 [reaction (12)]. The radical R.initiates the oxida- I1 v * RHPh,C=O --+ Ph,C=O +Ph,&OH + R' (1 2) tive chain processes. However, Scott and Amin2s have shown that some benzo- phenone derivatives retard the degradation processes during the later stages of photo-oxidation.This retardation is ascribed by Scott to the relative stability of the ketyl radicals formed, which, in turn, leads to favourable competition between chain-initiation and -termination processes in the mechanism (13). R OH OH Retardation Apart from the carbonyl-based prodegradants, certain types of transition- metal complexes and saIts have also been shown to be effective as prodegradants for the polyolefin~.~~ Of these, the ferric complexes are particularly favoured because of their cheapness and low toxicity. According to Brackman and Birle~,~~ferric stearate sensitizes the photo-oxidation of polyolefins by first absorbing U.V. light, then undergoing electron transfer to give a carboxylic acid free radical.This species decarboxylates to give an alkyl radical, which initiates polymer degradation [reaction (14) 1. Fe3+(6,CR), -Fe2'(&CR)z 4-6,CR 4CO, $Re (14) I4 (a) A. R. Burgess, J. Nut. Bur. Srandurds, 1953, Circular 525, p, 149; (b) D. J. Harper and J. F. McKellar, J. Appl. Polymer Sci.,1973, 17, 3503; (c) M. U. Amin and G. Scott, Euro-pean Polyrne? J., 1974, 10, 1019. Photodegradation and Stabilization of Commercial Polyolefns Further evidence from the work of Scott and Amin26c has shown that certain sensitizers may also act as ‘pro-oxidants’ during processing. Indeed these workers showed that the prodegradant behaviour of the sensitizers was merely an extension of their ‘pro-oxidant activity’. On the other hand, metal complexes containing sulphur ligands showed quite a different pattern of behaviour. In particular, the dithiocarbamates behaved as antioxidants whereas some ex- hibited U.V.stabilizing effects. With regard to the factors controlling the rate of prodegradation, the work of Scott and has also been of considerable interest. For example, the behaviour of the dithiocarbamates was found to depend primarily on the nature of the complexed metal. While the nickel and cobalt complexes were found to be more stable towards U.V. light, and indeed acted as light stabilizers in the polymer, the iron complex was much less stable and acted as an effective pro- degradant at low concentrations. Interestingly, at relatively high concentrations the same compound was found to be a light stabilizer.These observations of Scott clearly demonstrate the complexity of the mechanisms involved in both the destabilizing and stabilizing action of the transition-metal chelates to be dis- cussed later. 4 Effect of Pigments on Photostability The ability to colour thermoplastic polymers such as the polyolefins is, in many cases, necessary for their commercial use. In general, pigments rather than dyes are used to colour the polyolefins. Most of the pigments used for this purpose tend to photostabilize the polyolefins, although there are certain important exceptions. Papil10~~carried out an extensive examination of the many ways in which pig- ments can affect polymer properties. It was noted that the entire additive system rather than any single stabilizer or colourant was responsible for controlling the weatherability of the polymer.This observation was later confirmed by Hill and Martinovich.28 To eliminate any pigment-stabilizer interactions, U~elmeir~~ examined the effect of a number of pigments on the U.V. stability of unstabilized polypropylene. A correlation was found between the screening power of the pigment and its ability to protect the polymer. It is well known that carbon black affords the most effective protection to U.V. degradation,2,27 and several theories have been advanced to explain its interest- ing, as well as its technically important, behaviour in the polymer. For example, apart from acting as an effective screener, which depends on pigment particle size, carbon black is believed to inhibit chain propagation because of the anti- oxidant properties of its surface phenolic gro~ps.~,~~ Additional theories suggest P.J. Papillo, Mod. Plastics, 1967, 4, 45, 31. G. R. Hill and R.J. Martinovich, Tech. Papers Reg. Tech. Conf., SOC.Plastics Engineers Phil. Sect., Oct. 2-3, 1972. O9 C. W. Uzelmeir, SOC.Plastics Engineers Trans., 1970, 26, 69. Allen and McKeIlar that the pigment has a low surface energy3O and that it may also quench certain photoactive species in the polymer.31 Apart from the screening of U.V. radiation above ca. 300 nm the high reflec- tion properties associated with the pigment constitute another important factor to be considered.White pigments generally show high U.V. reflectance within the 300-400 nm range. Notable exceptions, however, are zinc oxide and titanium dioxide (titania). Of all the white pigments used in the polyolefins, the anatase and rutile forms of titania provide an interesting contrast in their photochemical behaviour. It is well known that the two crystalline modifications of titania ewbit markedly different photoactivities when incorporated into a number of com- mercially important polymers.32 With the polyolefins, for example, while rutile is relatively inactive, anatase is markedly photosensitive in degrading the po1ymer.l6J2 Irich33 has reported that the pigment-sensitized photo-oxidation of isopropyl alcohol can be used as a model for predicting the photochemical acti- vities of different grades of the oxides of titanium and zinc in polypropylene.It is generally agreed that the photoactivity of titanium dioxide pigments is related to their semiconductor proper tie^.^^ Absorption of near-u.v. light by the titanium results in promotion of electrons into a higher energy conduction band. This leaves a positive hole in the crystal lattice. Those released electrons near the surface of the pigment particle react with oxygen to form 02 ions and possibly other species that can react with the polymer substrate. The overall result of this process is that the surface of the polymer gradually erodes away. This pheno- menon is commonly referred to as ‘chalking’. Recently, Allen et aZ.,l6 by examining titania-pigmented polyolefk by spectro- phosphorimetry, found evidence to explain this difference in photoactivity between the two crystalline forms.Only the photoactive form (anatase) quenches the long-lived phosphorescence emitted from impurities native to the polymer, indicating the mechanism shown in reactions (15a and b) for anatase-initiated Ji vI = polymer Impurity -I * I* $-TiO,(Anatase) photo-degradation. Similar phosphorescence studies were carried out on Nyl0n-6,6.3~ It was found that manganese compounds, normally coated on the 30 H. Schonhorn and P. J. Luongo, Macromolecules, 1969, 2,4, 364. 31 M. Heskins and J. E. Guillet, Macromolecules, 1968, 1, 97. 32 H. A. Taylor, W. C. Tincher, and W. F. Hamner, J. Appl.Polymer Sci., 1970, 14, 141; N. S. Allen, J. F. McKellar, G. 0. Phillips, and C. B. Chapman, J. Polymer Sci., Part B, Polymer Letters, 1974,#12, 723; R. C. Hirt, N. Z. Searle, and R. G. Schmitt, SOC. Plastics Engineers Trans., 1961, 1,21; W. L. Dills and T. B. Reeve, Plast. Technol., 1970, 16, 6, 50. 33 G. Irich, J. Appl. Polymer Sci., 1972, 16, 2387. 34 W. F. Sullivan, Progr. Org. Coatings, 1972, 1, 157; H. G. Voltz, G. Kompf, and H. G. Fitsky, ibid., 1974, 2, 223. 543 Photodegradation and Stabilization of Commercial Polyolefns surface of the anatase pigment particles to counteract their photoactivity, quench the pigmented polymer emission. S Photostabilization It is generally accepted that photostabilization of a lightsensitive polyolefin may be achieved in many ways.Indeed a judgement of satisfactory ‘photostabiliza- tion’ to light exposure will primarily depend upon the end-use to which the polyolefin is put. For example, in the case of polypropylene, it is essential that a suitable additive be incorporated in the polymer for it to be satisfactory for most commercial uses out-of-doors. A study of the patent literature will clearly vali- date this point. However, over the past few years it has become increasingly clear that these light-stabilizing additives, particularly those for polypropylene, fall into three general categories, and their nature and possible modes of action will now be discussed. It should be clearly understood that of the three major fields of poly- olefin photochemistry, photostabilization is undoubtedly the one of greatest controversy on mechanistic grounds.A. Ultraviolet Screeners.-In this case the light is prevented from being absorbed by the photoactive chromophoric species (PCS) in the polymer by, for example, carbon black or titanium dioxide in its photochemically inactive rutile form. Here the mechanism is believed to be one of simple screening of the PCS from the harmful radiation (see Section 4). B. Ultraviolet Absorbers.-The U.V. absorbers are believed to operate by a mechanism in which, like the screeners, the stabilizer prevents the light from reaching the PCS. The harmful radiation is directly absorbed by the stabilizer, then harmlessly dissipated. An effective U.V.absorber should thus have high absorbance in the wavelength range which is most harmful to the polymer. For example, a number of commercial light stabilizers based on 2-hydroxybenzo- phenone have this property. They also dissipate their energy, when photo- excited, by a mechanism that involves the reversible formation of a six-membered hydrogen-bonded ring2J8 [reaction (16)]. The result of this mechanism of light absorption and dissipation thus leaves the stabilizer chemically unchanged and able to undergo a large number of these activation-deactivation cycles. Direct evidence for such a mechanism is difficult to obtain because the role of the stabilizer is essentially that of a ‘passive’ nature. However, one feature that serves as indirect evidence for this mechanism is that the more effective of these stabi- lizers exhibit a correspondingly stronger intramolecular hydrogen bond with the carbonyl This is an essential feature of mechanism (16).A further interesting class of stabilizers that appear to operate by this mech- anism are the salicylates, particularly phenyl salicylate. Here, however, the pro- w J. H. Chaudet, G. C. Newland, H. w. Patton, and J. w. Tamblyn, SOC.Plastics En-gineers Trans., 1961, 1, 26; K. A. Leitman, T. V. Kreitzer, V. L. Maksinov, and A. F. Lukovnikov, Vysokomol. Soedineniya, 1971, A13, 1, 86. $2 Allen and McKellar 0 OHIt I keto'form 'enol'form tective mechanism is slightly different in that they are photolysed by the incident radiation into products with high absorbance in the near-u.~.~~ [reaction (17)l.OH 0 OH c\O 2,2' Dihydroxybenzophenone JI v/8 -+ (1 7) OH 0 OH 2,4' Dihydroxybenzophenone C. Excited State Quenchers.-Here the mode of action of the stabilizer is to deactivate the PCS in the polymer before it undergoes chemical reactions that result in polymer degradation. In general, these stabilizers are complex chelates of the transition metals, usually nickel. Despite their relatively lower light absorption in the region 300-400 nm, a number of these chelates are much more effective as light stabilizers in polypropylene than the 2-hydroxybenzophenones.5 Deactivation of the PCS can be achieved in a number of ways, and this adds a further complicating factor to the assessment of the stabilizing action of the quencher.For example, the photoexcited state of the PCS may be either the singlet or triplet state or it may even indirectly initiate degradation by being quenched in its triplet state by oxygen to form excited singlet-state oxygen [reaction(18)], which may then proceed to attack the polymer matri~.~*~~ The first evidence that a transition-metal chelate was both an effective light- stabilizer for polypropylene and also an efficient triplet-deactivator was reported 36 J. Jortner, J. Polymer Sci., Part A-1, Polymer Chem., 1959, 3, 7, 199; G. C. Newland and J. W. Tamblyn, J. AppI. Polymer Sci.,1964, 8, 1949. 545 Photodegradation and Stabilization of Commercial Polyolefins 3(~cs)+ 30,-~(PCS) + lo2 by Briggs and M~Kellar.~ In this case a diamagnetic chelate of structure (5) was studied, and from observing the variation in the efficiency with different sub-stituents for R it was concluded that the triplet quenching efficiency was de- pendent on the spatial configuration of the ligand around the metal atom (Ni) in the chelate. Using laser flash photolysis, Adamczyk and Wilkinsons7 found that a number of NP chelates effectively quench triplet benzophenone.The square-planar diamagnetic chelates based on structure (5) were found to be the most efficient. Their results correlate well with those of Harper and McKellar,38 who deter- mined the stabilizing efficiency of the same chelates in polystyrene. Recently, Harper et demonstrated that the diamagnetic chelate shown in structure (5) markedly quenches the lifetime of the long-lived phosphorescence from poly- propylene.The source (or sources) of this emission are believed to be carbonyl groups capable of initiating photo-oxidation of the polymer.5J6 Me Me R \/c=N / Ni Mi Several workers37~39-41 have reported that some nickel@) chelates are effective quenchers of singlet oxygen, particularly those chelates with sulphur donor ligands, e.g. 2,2’-thiobis-(4-t-octylphenolato)-n-butylamine.In this case, there- fore, a stabilizing effect by the chelate could be possible by it acting as a scaveng- ing agent for singlet oxygen, preventing the attack of 102 on the polymer matrk4 Recently, Carlsson and Wiles42 studied the effect of various stabilizers on the phosphorescence from thermally oxidized polypropylene film.In contrast to other workers they found that certain nickel(II) chelates did not operate by long- range energy transfer, but by collisional quenching and radical-scavenging A. Adamczyk and F.Wilkinson, J. Appl. Polymer Scl., 1974, 18, 1225. D. J. Harper and J. F. McKellar, J. Appl. Polymer Sci., 1974, 18, 1233. 39 D. J. Carlsson, T. Suprunchuk, and D. M. Wiles, J. Appl. Polymer Sci., 1972, 16, 615; J. P. Guillory and C. F. Cook, J. Polymer Sci., Part A-I. Polymer Chem., 1971,9, 1529. ‘O B. Felder and R. Schumacher, Angew. makromol. Chem., 1973, 31, 35; D. J. Carlsson, T. Suprunchuk, and D. M. Wiles, J. Polymer Sci., Part ByPolymer Letters, 1973, 11, 61.41 J. P. Guillory and C. F. Cook, J. Amer. Chem. SOC.,1968, 95, 15, 4885; J. Polymer Sci., Part A-I, Polymer Chem., 1973, 11, 1927. 4a D. J. Carlsson and D. M. Wiles, Macromolecules, 1974, 7, 259; J. Polymer Sci., Part B, Polymer Chem., 1974,12,2217. Allen and McKellar processes. A number of the chelates were found to be efficient scavengers of *OHand *ORradicals formed during the thermal and photo-decomposition of model and polymeric hydroperoxides. A similar effect has also been reported by Ranaweera and Scott.43 It would appear, therefore, that the series of NiII chelates which have been clearly established as effective stabilizers for poly- propylene may function either as PCS quenchers, or as radical scavengers. Indeed, some of the most effective, such as the diamagnetic chelates, may well operate by both mechanisms.l7c In addition to their function as U.V.absorbers (Section 5B), a number of investigations on the 2-hydroxybenzophenones and the corresponding hydroxy- benzotriazoles indicate that their effectiveness as stabilizers may also be due to their ability to quench PCS in the polymer. For example, early studies by Pivovarov and ~o-workers~7~ showed that the protective action of 4-t-butylphenyl salicylate and 2,2’-hydroxy(5’-methylphenyl)benzotriazole in a range of polymers is largely due to energy-transfer processes. A similar conclusion has recently been reported by George.& However, in contrast to the conclusions of Pi~ovarov~~~and George,@ other studies,41 in which the unstabilized polyolefin was irradiated by light which was first passed through solutions containing the stabilizers, have shown that the salicylates and the hydroxybenzotriazoles operate purely as U.V.absorbers while the 2-hydroxybenzophenones can also function as PCS quenchers. 6 Conclusions In this review we have discussed the evidence for the presence of several types of impurities, all of which are believed to be capable of acting as photoinitiators in a polyolefin. In our view it is evident from the literature that all these im- purities could participate in the photo-oxidation of commercial grade poly- olefins. Further, the relative importance of a particular initiator will obviously depend on the conditions of U.V.exposure, the manufacturing and processing history of the polymer, and even the nature of the polyolefin itself. During the preparation of this review an excellent book by Ranby and Rabek2 was published in which the following statement appears (p. 120). ‘Frequently the same problem is examined by various authors, using different methods, and the results obtained may refute each other. Hundreds of papers published to date have not yet given a satisfactory answer to the question of the precise mech- anism of the degradation and oxidation of polymers’. We firmly believe that this general statement can be applied to the polyolehs for the reasons given above. This should serve as a cautionary guide to future workers in attributing unique mechanisms to additives such as prodegradants and light stabilizers where the evidence emerging from the literature is that several of these additives may well be multi-functional in their mode of operation. 43 R. P. R. Ranaweera and G. Scott, Chem. and Znd., 1974, 774; J. Polymer Sci.,Part B, PoIymer Letters, 1975, 13, 71. 44 G. A. George, J. Appl. Polymer Sci., 1974, 18, 419. 547

ISSN:0306-0012    

DOI:10.1039/CS9750400533

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Alkali-metal Complexes in Aqueous Solution By D. Midgley CENTRAL ELECTRICITY RESEARCH LABORATORIES, KELVIN AVENUE, LEATHERHEAD, SURREY 1 Introduction Although sodium and potassium co-ordination compounds have been known for fifty years1 and ion-pairs have been studied extensively in non-aqueous solution,2 the aqueous chemistry of these compounds has received little systematic attention. The outstanding characteristic of alkali-metal complexes in aqueous solution is their weakness, but they are important because of the common occurrence of alkali-metal ions at high concentrations, e.g. in biological systems, in sea water, in constant ionic media, or when added in a reagent. Because the complexes are weak, they have often been ignored, and, at high, constant concentrations of alkali metal, consistent results may be obtained from a model that ignores the complex, because the fraction of ligand bound to the alkali metal is constant.In the past, the evidence for such complexes has often been indirect or difficult to interpret, but the accumulation of results and the use of ion-selective electrodes for direct study of the equilibria allow a more confident treatment of the subject. The term ‘complex’ as used here does not imply any particular kind of bonding in the associated species, which may be an ion pair, a solvent-separated ion pair, or a co-ordination compound. A. Abbreviations.-The following abbreviations are used to indicate the least protonated forms of the ligands. edta ;ethylenediamine-NNN’N’-tetra-aceticacid ; pdta :propylenediamine-NNN’N’-tetra-aceticacid ; cdta : cyclohexenediamine-NNN’N’-tetra-aceticacid ; egta: 2,2’-ethylenedioxybis [ethyl-iminodi(acetic acid)]; nta: nitrilotriacetic acid; ida : iminodiacetic acid; uda : uramil-NN-diacetic acid; ATP, ADP, AMP: adenosine tri-, di-, and mono-phosphate, respectively.Protonated and complexed forms are indicated thus: Hedta, Hzedta, Naedta, etc. 2 Inorganic Complexes Representative values of the stability constants of inorganic complexes are given N. V. Sidgwick and S. G. P. Plant, J. Chem. SOC.,1925, 209. ‘Ionsand Ion Pairs in Organic Reactions’, ed. M. Szwarc, Vol. 1, Wiley, New York and London, 1972. Alkali-metal Complexes in Aqueous Solution in Table 1.For sulphate, persulphate, thiosulphate, ferricyanide, and ferrocyanidc the stability constants decrease in the order Cs > Rb > K > Na > Li, whereas phosphates show the opposite preference, although the difference between stabilities of potassium, rubidium, and caesium complexes is often small. Complex formation is favoured by a high charge on the ligand and by the ease with which chelation can occur, e.g. cyclic condensed phosphates give stronger complexes than linear condensed phosphates. Table 1 Stability constants (log /3) of inorganic complexes at 25 "C and ,x = 0 Ligand Li Na K Rb Cs Ref. Sulphate 0.64 0.70 0.82 --a Persulphate -0.58 0.91 1.17 1.42 67 -Thiosulphate -0.58f0.02 1.oof0.04 -33 Ferrocyanide 1.78 2.08 2.3 2.65 2.85 65b --2.35f0.2 --32 Ferricyanide --1.46 k0.02 --32 -0.32f0.13 0.30f0.04 0.52f0.03 -93c Carbonate -0.55 ---d Bicarbonate -0.16 ---d Phosphate 0.72f0.04 0.59f0.04 0.48f0.06 Trimetaphosphate -0.88 no ev.--b,f Tetrametaphosphate -1.42 1.26 --b,f Triphosphate 2.87f0.06 1.64f0.06 1.37 f0.06 g Tetraphosphate 2.64 1.79 1.71 g 18 "C,E. C. Righellato and C. W. Davies, Trans. Furaday SOC.,1930,26,592; b p = 0.1 ; 3; F. S. Nakayama, J. Inorg. Nuclear Chem., 1971, 33, 1287; Gp= 0.2, R. M. Smith ..A. Alberty, J. Phys. Chem., 1956,60,180;f S. Y. Kalliney, Diss. Ah., 1970,31B, 1779; 1, J. I. Watters, S. M. Lambert, and E. D. Loughran, J. Amer. Chem. SOC.,1957, 79, 3 Organic Complexes The stability constants of alkali-metal complexes with organic ligands, almost without exception, decrease as the atomic number of the metal increases.Aliphatic and aromatic carboxylic acids do not generally form complexes with the alkali metals, although stability constants have been reported for lithium and sodium acetate^.^ The acids for which there is strong evidence of complexing are either a-hydroxy-acids or dicarboxylic acids. The evidence for association with a-amino-acids is uncertain, although sodium-leucine and -alanine complexes have been reported.* Typical stability constants are given in Table 6. For a-hydroxy-acids, the strength of the complex depends on the number of carboxylate and a-hydroxy-groups, thus the order of strengths is citrate > tartrate > malate > glycollate > lactate.Pyruvate4 has been found to be stronger than any of those ligands. A protonated complex has been D. W. Archer and C. B. Monk,J. Chem. SOC., 1964, 31 17. G. A. Rechnitz arid S. B. Zamochnick, J. Amer. Chem. SOC.,1964, 86, 2953. 550 M idg le y reported only for tartrate,5 but similar complexes would be expected with the other polycarboxylic acids. Among dicarboxylic acids, malonate,6-* succinate,6 phthalate,6 and maleate69' complexes have been reported. Chelation is probably a necessary condition for complex formation. The ease of ring formation should be important, and succinate, which gives a conformationally strained seven-membered ring, is weaker than malonate, which gives a six-membered ring. Unfortunately, no results have been reported for oxalate, which would form a five-membered ring.If chelation is necessary, protonated complexes of unsubstituted dicarboxylic acids would not be expected, and none have been reported so far. It is interesting to note that methylenedipho~phonate~ is a considerably better ligand for sodium than malonate and that methanehydroxyphosphonatelOis better than glycollate. Whereas the above complexes are weak, aminopolycarboxylates can form quite strong complexes, with stabilities: cdta > pdta > uda > edta > nta > ida. Representative stability constants are summarized in Table 2. The order is broadly as expected, the affinity increasing with the number of acetic acid groups and the charge. Table 2 Stability constants (log /3) of aminopolycarboxylate complexes Ligand P T/"C Li Na K Rb Cs Ref.edta 0.1 20 2.79 1.66 ---a 0 --2.61 ---13 b 25 2.85 1.79 0.96 0.59 0.15 14 pdta 0.5 25 4.01 2.55 f0.05 0.90k0.07 --15 cdta 0.1 25 -4.40k 0.07 1.52k0.02 --120 0.5 25 6.11 kO.l 4.66f0.1 1.83f0.05 --15 nta 0 20 3.28 2.18 -0.1 20 2.51d 1.22d 0.6e d, e uda 0 20 5.40 3.32 ---f 0.1 20 4.90k 0.02 2.72f0.01 1.23k0.03 --70 o-hydroxyphen yl -ida 0.1 20 2.20f0.02 1.OkO.1 a G. Schwarzenbach and H. Ackermann, Helv. Chim. Acta, 1947, 30, 1798; b ,LL variable; C G. Schwarzenbach, E. Kampitsch, and R. Steiner, Helv. Chim. Acta, 1945,28,828;d ref. 11 ; e G. Anderegg, Helv. Chim. Acta, 1967, 50, 2333; f G. Schwarzenbach, E. Kampitsch, and R. Steiner, Helv.Chim. Acta, 1946, 29, 364. Uramil-NN-diacetic acid is exceptional : the presence of a-hydroxy- or a-oxo- groups enables quadridentate co-ordination to occur, and the possibility of H. S. Dunsmore and D. Midgley, J. C. S. Dalton, 1972, 64. D. W. Archer, D. A. East, and C. B. Monk, J. Chem. SOC.,1965, 720. ' J. A. Barclay, M. J. Hickling, and K. White, Biochem. J., 1966, 99, 11P. S. L. Dygert, G. Muzii, and H. A. Saroff, J. Amer. Chem. SOC.,1970,74,2016. R. L. Carroll and R. R. Irani, Inorg. Chem., 1967, 6, 1994. lo H. Wada and Q. Fernando, Analyt. Chem., 1972,44, 1640. 551 Alkali-metal Complexes in Aqueous Solution resonance hybrids may increase the stability of the complex, although sub- stitution by methyl groups on one or both of the nitrogen atoms in the pyrimidine ring has little effect on the stability.ll In general, uda forms weaker complexes than edta, e.g.with thallium(1) ion and the alkaline-earth ions, and its affinity for the alkali metals is thus anomalous. Substituted iminodiacetate complexes are much weaker than those with uda.12 It is most unlikely that all the carboxylate groups in the edta series of ligands participate in co-ordination of alkali-metal ions, since, for example, trans-cdta is a stronger ligand than the more flexible edta. From the limited number of results available, it seems that alkyl-substituted edta derivatives form stronger complexes as the degree of substitution increases. Protonated complexes of edtal3J4 and cdta15 and diprotonated complexes of pdta16 and cdtal5 have been reported.Polynuclear edta complexes have been proposed.14 4 Binding by Macrocyclic Compounds Many macrocyclic compounds bind alkali-metal ions unusually strongly and with great selectivity. These electrically neutral compounds contain rings of at least twelve members, including four or more oxygen, sulphur, or nitrogen atoms, and adopt conformations such that a central cavity is formed, with the polar heteroatoms directed inwards while the hydrocarbon backbone is directed outwards. The effect of this structure is to make the inside of the molecule hydrophilic and the outside hydrophobic. Mono- bi-, and tri-cyclic structures are known. Macrocyclic compounds with ion-binding properties occur naturally, e.g.the antibiotics valinomycin [(VI) in the Figure], monactin, nonactin, nigericin, and enneatin; many others have been synthesized. The two main classes of synthetic compounds are the ‘crown’ polyethers prepared by Pedersenl’ and the macrobicyclic diamines (‘cryptates’) of Lehn, Sauvage, and Dietrich.18 The Figure shows the structures of some typical macrocyclic ligands and enables the ‘crown’ nomenclature to be understood: thus (II) is 18-crown-6, having an 18- membered ring containing six oxygen atoms, and (111)is dicyclohexyl-18-crown-6. Reviews have given many structures19 and syntheses.20 The chemistry of the macrocyclic polyethers has been reviewed by Pedersen and Frensdorff ?1 Stability constants for aqueous complexes are given in Table 3.The affnity of a macrocycle for an ion is highly dependent on the size of the cavity and the l1 H.Irving and J. J. R. F. da Silva, J. Chem. SOC.,1963, 458. la H. Irving and J. J. R. F. da Silva, J. Chem. SOC.,1963, 3308. la V. Palaty, Canad. J. Chem., 1963, 41, 18. l4 J. Botts, A. Chashin, and H. L.Young, Biochemistry, 1965, 4, 1788. l6 J. D. Carr and D. G. Swartzfager, Analyt. Chem., 1971,43, 1520. J. D. Carr and D. G. Swartzfager, Analyt. Chem., 1971, 43, 583. l7 C. J. Pedersen,J. Amer. Chem. SOC.,1967, 89, 2495, 7017. B. Dietrich, J. M. Lehn, and J. P. Sauvage, Tetrahedron Letters, 1969, 34, 2889. J. J. Christensen, R. 0.Hill, and R. M. Izatt, Science, 1971, 174, 459; A. LevCque and R. Rosset, Analusis, 1973, 2, 218. C.Kappenstein, Bull. SOC. chim. France, 1974, 89. 91 C. J. Pedersen and H. K. Frensdorff, Angew. Chem. Internat. Edn., 1972, 11, 16. Midgley a,"jb0 a:1x1 (IV) (a) m = 0, n = 1 (b) m = 1, n = 0 (c) m=n=l (d) m = 1, n = 2 (e) m=2,n = 1 (f) m= n =2 Figure Macrocyclic ligands: (I), cyclohexyZ-l5-crown-5; (II), 18-crown-6; (111), dicyclo-kexyl-18-crown-6; (IV), diazapolyoxa-macrobicycles('cryptates'); (V), tetra-azahexaoxa-macro tr icy cle;(VI) ,valinomycin Alkali-metal Complexes in Aqueous Solution radius of the ion. Far4.r. spectra, however, indicate that the forces binding sodium and potassium ions in crown complexes are nearly equa1,22 and that the solvent plays a big role in selectivity. The same order of selectivity is maintained in different solvents, but the selectivity is reduced as the polarity of the solvent decreases.In dry hexadeuteriated acetone nonactin binds sodium, potassium, and caesium ions to approximately the same degree, but in wet hexadeuteriated acetone it develops a selectivity for potas~ium.~3 The apparent stability constants of 1:l crown complexes in water24 depend on the metal but not on the ligand concentration. Although, in methanol, 2: 1complexes were found with potassium and caesium, such side-reactions could not account for the behaviour in water. The nature of the anion (chloride or hydroxide) did not affect the stability constants. Table 3 Stability constants (log /3) for macrocyclic complexes in water at 25 "C Cavity diameter Li Na K Rb Cs Ligand /A (1.36)a (1.94) (2.66) (2.94) (3.34) Ref.cyclohexyl-15-crown-5 (I) 1 .7-2.2b < 1.O <0.3 0.6 --24 18-crown-6 (II) 2.6-3.2b -<0.3 2.06 -0.8 24 cyclohexyl-18-crown-6 2.6-3.2b <0.7 0.8 1.90 -0.8 dicyclohexyl-18-crownd (ID) 2.6-3.2b isomer A 0.6 1.5-1.85 2.18 -1.25 --2.02 1.52 0.96 isomer B -1.2-1.6 1.78 -0.9 --1.63 0.87 -dicyclohexyl-2 1 -crown7 3.4-4.3b -I --1.9 dicyclohexyl-24-crown-8 ----1.9 cryptate 112 (IVa) 1.6C 4.30 2.80 <2.0 <2.0 ~2.0 cryptate 221 (IVb) 2.3c 2.50 5.40 3.95 2.55 <2.0 cryptate 222 (IVc) 2.fIc <2.0 3.90 5.40 4.35 <2.0 cryptate 223 (IVd) 3.6c <2.0 <2.0 2.2 -2.20 cryptate 332 (IVe) 4.2C <2.0 <2.0 c2.0 -<2.0 cryptate 333 (IVf) 4.P <2.0 <2.0 <2.0 -<2.0 tricyclic cryptate (V) 1.0-1.7 ____+ a ionic crystal diameter/A; cavity diameter estimated from models (C.J. Pedersen, J. Amer. Chem. SOC.,1970,92,386); cavity diameter estimated from models (ref. e); d temperature not given; e J. M. Lehn and J. P. Sauvage, Chem. Comm., 1971, 440; f J. Cheney, J. M. Lehn, J. P. Sauvage, and M. E. Stubbs, J.C.S. Chem. Comm., 1972, 1100. a2 A. T. Tsatsas, R. W. Stearns, and W. M. Risen, J. Amer. Chem. SOC., 1972,94, 5247. p3 J. H. Prestegard and S. I. Chan, Biochemistry, 1969,8,3921; J. Amer. Chem. SOC.,1970,92, 4440. 24 H. K. Frensdorff, J. Amer. Chem. SOC., 1971,93,600. 554 MidgZey The solubility of these ligands in water is fairly low, and most studies have been made in methanol or in non-polar solvents.This is especially true of the naturally occurring antibiotics, and results for these compounds in alcohols25 may be compared with those for synthetic ligands.19 The ability of the ligands to dissolve large amounts of alkali-metal salts in non-polar solvents is one of their most remarkable properties. FrensdorfP and Pedersen27 have studied distribution equilibria for crown complexes. A surprisingly large isotope effect in the distribution of 22Na and 24Na between ion-exchange resins and solutions of monactin has been reported.28 The heavier isotope is concentrated in the monactin complex by about 3.5%. Magnetic resonance studies29 of macrocyclic complexes in organic solvents show that the exchange reaction M+LgML predominates over the alternatives (*M + MLF? *ML + M and L* + MLF? ML* + L).With crown polyethers there is first a fast confor- mational transition of the ligand, then a slow complexation reacti0n.3~ 5 Spectroscopic Evidence for Alkali-metal Complexes It is usually difficult to isolate the spectra of alkali-metal complexes from the background, partly because of the limited extent of complex formation and partly because complex formation may make only a small difference to the spectrum. A. Ultraviolet Spectroscopy.-Ultraviolet spectroscopy may not reveal complex- ing that other methods show to occur, e.g. potassium ion has no effect on the U.V. spectrum of ferricyanide i0n,3lP32 although conductivity, solubility, and potassium selective electrode measurements show that complexes are formed.This discrepancy is explained in terms of the formation of outer-sphere complexes. Since the complexes are weak, it is difficult to determine their extinction coefficients, since these may differ little from those of the ligands. An extra-polation method has been used to obtain the extinction coefficients, and hence the stability constants, of thiosulphate complexes.33 Symons 34 has pointed out the dangers of explaining changes in absorbance at a single wavelength by postul- ing equilibria between species having definite extinction coefficients, with particular reference to sodium iodide solutions, W. E. Morf and W. Simon, Helv. Chim. Acta, 1971, 54,2683. m H. K.Frensdofl, J. Amer. Chem. SOC.,1971,93,4684. z7 C. J. Pedersen, J. Amer. Chem. SOC.,1970, 92, 391. as H. S. R2de and K. Wagener, Radiochim. Acta, 1972, 18, 141. 29 E. Shchori, J. Jagur-Grodzinski, Z. Luz, and M. Shporer, J. Amer. Chem. Soc., 1971, 93, 7133; D. H. Haynes, F.E.B.S. Letters, 1972, 20, 221. 30 P. B. Chock, Proc. Nat. Acad. Sci. U.S.A., 1972, 69, 1939. 31 V. E. Mironov and Yu. I. Rutkovskii, Zhur. neorg. Khim., 1966, 11, 1792. 3s W. A. Eaton, P. George, and G. I. H. Hanania, J. Phys. Chem., 1967,71,2016. 33 F. G. R. Gimblett and C. B. Monk, Trans. Furaday SOC.,1955,51,793, M. C. R. Symons,Discuss. Faraday Soc,, 1957,24, 117, Alkali-metal Complexes in Aqueous Solution B. Infrared and Raman Spectroscopy.-Infrared spectroscopy has been little used because high concentrations are necessary, and because it is difficult to com- pensate for the intense absorption of water with sufficient exactness for accurate quantitative information to be obtained.Potassium ferricyanide and ferrocyanide solutions showed no evidence of direct interaction^:^ although other methods show that complexes exist. It may be inferred that outer-sphere complexes are formed, but U.V. spectroscopy indicates an interaction for ferrocyanide but not ferricyanide, so conclusions about the nature of a complex from the absence of a change in spectrum must be drawn with caution. There is clear evidence for the formation of a potassium-monactin complex in methanol.36 Raman spectroscopy has not given direct evidence of ion-pairing in sodium sulphate, but a complex has been detected from the change in absorption of the spectrum of HS04- ion.37 Splitting in the spectra of alkali-metal sulphates at high concentrations3* was attributed to ion-pairing, and occurs at 2, 4, and 5 mol 1-1 for the lithium, sodium, and potassium salts, respectively, indicating the opposite order of strength to that found by other methods (Table 1).The stability constant of the sodium-ATP complex was calculated by an empirical method from the pH dependence of the Raman spectra of ATP solutions containing calcium or magnesium ions with and without sodium present.39 C. Nuclear Magnetic Resonance.--The need for relatively high concentrations of reactants (ca. 0.1 mol 1-l) restricts the scope of n.m.r.studies. Alkali-metal complexes are too labile for the spectrum characteristic of the complex to appear, and nuclei in the free and associated species give an average spectrum, from the shape and chemical shift of which much useful information can often be derived. 1HN.m.r. The chemical shifts and spin-spin coupling constants of the methylene and methine protons of alkali-metal malates dissolved in D20show that all the salts form complexes, whereas the tetramethylammonium salt does n0t.40 The position of the break in the step curve obtained when the chemical shifts of ethylenic and methylenic protons in edta complexes were plotted against pH indicated the relative strengths of the c0mplexes.4~ Lithium formed a sufficiently strong complex for a stability constant to be calculated.Sudmeier and Senzel.Q2 have shown that the rotational conformation of the pdta tetra-anion is sensitive to co-ordination of both ends of the ligand, and they have calculated stability constants for potassium and rubidium complexes. SIP N.m.r. Crutchfield and bani& have studied the 3lP n.m.r. spectra of tetra-86 N. Tanaka, Y. Kobayashi, and M. Kamada, Bull. Chem. SOC.Japan, 1966,39,2187. a* L. A. R. Pioda, H. A. Wachter, R. E. Dohner, and W. Simon, Helv. Chim. Acta, 1967,50, 1373. F. P. Daly, C. W. Brown, and D. R. Kester, J. Phys. Chem., 1972,76, 3664. H. Lee and J. K. Wilmshurst, Austral. J. Chem., 1964, 17, 943. a@ M. E. Heyde and L. Rimai, Blochemisfry, 1971,10, 1121.40 L. E. Erickson and J. A. Denbo, J. Phys. Chem., 1963,67,707. 41 R. J. Kula, D. T. Sawyer, S. I. Chan,and C. M. Finley,J. Arner. Chem. SOC.,1963,85,2930, 4s J. L. Sudmeier and A. J. Senzel, Anulyr. Chem., 1968,40, 1693. 4a M. M. Crutchfield and R. R. Irani, J. Arner. Chem. SOC.,1965,87,2819, Midgley methylammonium polyphosphate solutions in the presence of varying con- centrations of lithium ion. In tripolyphosphate the central phosphorus atom differed from the two outer ones, but a plot of the chemical shifts against the [Li+]/fP3010~-] ratio showed breaks at 1:l for both atoms. The phosphorus atoms in pyrophosphate were equivalent, and no break was observed in the plot, although the lithium ions still lowered the field in their vicinity.There was little effect on the chemical shift of trimetaphosphate and tetrametaphosphate. In lithium isohypophosphate complexesu both PII1and PV nuclei responded in the same manner, indicating that the ligand was bidentate. The magnitude of the chemical shift was about two-thirds that observed with lithium pyrophosphate soh t ions. SNa and 39K N.m.r. Wertz and Jardetzky45 have measured the amplitude and linewidth of the 23Na nuclear spin resonance absorption of inorganic sodium salts in water. Two classes of salts were found; (i) those with constant linewidth and linearly increasing amplitude up to saturation, e.g. chloride, fluoride, sulphate and ferricyanide ; (ii) those whose linewidth increases with concentration and whose amplitude either levels off or goes through a maximum, usually at con- centrations >6moll-l.Notable in this second group were phosphate and pyrophosphate, which had maxima at 1-2 moll-l. Organic salts showed similar divisions :48 formate, acetate, and benzoate gave no increase in line width, but a-hydroxy-and a-oxo-acids had pronounced effects. Rechnitz and Zamochnick4 found a linear correlation between the stability constants of sodium complexes with organic ligands and the linewidths reported by Jardetzky and Wertz. Eisenstadt and fried man'^^^ measurements of the relaxation rate of 23Na in a number of salts gave evidence of association which sometimes conflicted with the results of Jardetzky and Wertz; thus, pyruvate gave no stronger interaction than perchlorate, and sulphate showed ion-pairing.Measurements by other means support Jardetzky and Wertz in the former case and Eisenstadt and Friedman in the latter. James and N0ggle4~ have studied the relaxation time of the 23Na nucleus as a function of pH and metal and ligand concentration in solutions of edta, "(2-hydroxyethy1)ethylenediamine-NNW-triaceticacid, nta, and histidine. The inter- action was weak with histidine but marked with other ligands, enabling stability constants to be calculated. A series of phosphate compounds was similarly in- vestigated.49 23Na n.m.r. has been used in biological studies to indicate binding of sodium by soluble RNAS0and in tissues such as brain,5l liver,52 and mus~le.~~~~~ r4 R. L. Carroll and R.E. Mesmer, Inorg. Chem., 1967, 6, 1137. Ob J. E. Wertz and 0.Jardetzky, J. Chem. Phys., 1956,25, 357. 46 0.Jardetzky and J. E. Wertz, Arch. Biochem. Biophys., 1956,65,569; J. Amer. Chem. SOC., 1960, 82, 318. 47 M. Eisenstadt and H. L. Friedman, J. Chem. Phys., 1967,46,2182. *.9 T. L. James and J. H. Noggle, J. Amer. Chem. Soc., 1969,91, 3424. T. L. James and J. H. Noggle, Analyt. Biochem., 1972,49,208. T. L. James and J. H. Noggle, Proc. Nut. Acad. Sci. U.S.A., 1969, 62, 644. O1 F. W. Cope, J. Gen. Physiol., 1967, 50, 1353. 61 D. Martinez, A. A. Silvidi, and R. M. Stokes, Biophys. J., 1969,9, 1256. b8 F. W. Cope, Proc. Nut. Acad. Sci. U.S.A.,1965,54, 225. 557 Alkali-metul Complexes in Aqueous Solution Extension of the technique to other cations is governed by their relative Nsensitivity, in the order 7Li 23Na > B7Rb N l33Cs % 39K.Evidence of potas- sium complexes has been found using a 39Kspin-echo technique.54 D. Po1arimetry.-Alkali-metal-ion-specilk effects have been noted in the polarimetry of tartrates in various conditions,55 but without having been quanti- fied. Carr and Swartzfager15J6 have used polarimetry to study alkali-metal pdta and trans-cdta complexes. The molar rotations of the complexed species were obtained iteratively, except for the relatively strong sodium and lithium com- plexes. The values of the stability constants, p, were high compared with others in the literature, which may be due to the use of values of the fourth dissociation constants, K4, of the ligands that had been determined simultaneously with 18.The substitution of an independently obtained K4 in Carr and Swartzfager’s equations gave smaller values for 18. The difference between the polarimetrically obtained stability constants and literature values was smaller for pdta, where the acid dissociation constants also agreed more clozely. Evidence was presented for the existence of protonated species, but the apparent distribution was so strange that it must be treated with reserve, e.g. Kpdta3- and KHzpdta- complexes were found, but not KHpdta2-. A particular complex will be difEicult to detect it it is not a major component of an equilibrium mixture and its molar rotation does not differ much from that of another species.So far there has been no report of optical activity caused by the formation of a complex between an alkali metal and an optically inactive ligand. 6 Electrochemical Evidence for Alkali-metal Complexes A. Potentiometry.-The use of glass electrodes that are responsive to alkali-metal ions is well established, and calibration procedures for complexing studies have been described.56 The method has been used for all the alkali metals,ZQJ7 but commonly only for sodium and potassium. Because of the sensitivity of these electrodes to hydrogen ions, measurements can usually only be made in neutral or alkaline solutions. Liquid ion-exchange electrodes of the neutral carrier type are less affected by pH and have been used to study potassium -ATP cornplexe~.~~ Sodium amalgam electrodes have been used to study sulphate and carbonate equilibria,59 but they are far less convenient than selective ion electrodes.B. Conductivity.-Conductivity measurements are best suited to dilute solutions O4 F. W. Cope and R. Damadian, Nature, 1970, 228, 76. 66 H. T. S. Britton and P. Jackson, J. Chem. SOC.,1934, 998; M. T. Beck, B. CsiszAr, and P. Szarvas, Nature, 1960, 188, 846; L. I. Katzin and E. Gulyas, J. Phys. Chem., 1960, 64, 1739; V. Frei, Coll. Czech. Chem. Cornrn., 1962,27,2450. 66 G. A. Rechnitz and J. Brauner, Talunta, 1964, 11, 617; H. S. Dunsmore and D. Midgley,J. Chem. Soc., 1971, 3238; G. L. Gardner and G. H. Nancollas, Analyt. Chem., 1969, 41, 202. 67 G. A. Rechnitz and S. B. Zamochnick, Talanta, 1964,11, 1061.M. S. Mohan and G. A. Rechnitz, J. Amer. Chem. Soc., 1970,92, 5839. 6s R. F. Platford and T. Dafoe, J. Marine Res., 1965, 23, 63; J. N. Butler and R. Huston, J. Phys. Chem., 1970,74,2976. 558 Midgley of a single electrolyte in which only one complex is formed, and their application is therefore rather limited. Stability constants have been determined for sulphate,60 ferrocyanide,61 ferricyanide,62 and various condensed phosphate63 complexes. In all these cases the complex is charged, and no independent value of its equi- valent conductivity is known. A value may be assigned by analogy or be deter- mined simultaneously with the stability constant.64 C. E1ectromigration.-Shvedov and Nichugovsky65 have calculated stability constants for alkali-metal ferrocyanides reIative to potassium ferrocyanide from the amount of cation transferred from one compartment of the electro- migration cell to the other.The constants are not of high accuracy. D. Po1arography.-Alkali-metal ions give well-defined polarographic waves in a supporting electrolyte of tetramethylammonium hydroxide or haIide. Complexes with cdta, edta, and uda have been reported66 for lithium and sodium ions, but strong complexing can shift the half-wave potential of lithium beyond the decomposition potential of the medium, thus limiting the quantitative application of the technique. 7 Thermodynamics The standard free energy, enthalpy, and entropy changes of complex formation can be calculated from the stability constant, 16, by means of equations (1)-(3).The enthalpy change may also be measured directly by calorimetry. Values for a number of complexes are given in Table 4. AGO= -RTlnp (1) d In P/dt = AHOIRT~ (2) AG" =AH" -TAS" . (3) In principle, the calorimetric determination ofdH" is to be preferred, but very precise measurements of pH are necessary if protonic equilibria occur simultaneously with the complexing of the alkali metal. Good agreement has been reported for AH" values for potassium persulphate from osmometry67 and ion-selective-electrode68 measurements at two temperatures. The same authors' results for potassium ferricyanide and ferrocyanide68 are, however, five times larger than those of Eaton et aZ.,32who also used an ion-selective electrode, but at five temperatures, and who o-btained very similar results calorimetrically.6o I. L. Jenkins and C. B. Monk, J. Amer. Chem. SOC.,1950,72,2695. J. C. James, Trans. Faraday SOC.,1949, 45, 855. J. C. James and C. B. Monk, Trans. Faraday SOC.,1950,46, 1041. 63 C. W. Davies and C. B. Monk, J, Chem.SOC.,1949,413; C. B. Monk, ibid., p. 423; G. Kura and S. Ohashi, J. Inorg. Nuclear Chem., 1972, 34, 3899. 64 C. W. Davies, 'Ion Association', Butterworths, London, 1962. 65 V. P. Shvedov and G. F. Nichugovskii, Radiokhimiya, 1966, 8, 66. A. Bobrovsky and Yu. Zarembsky, Zhur. analit. Khim., 1972,27, 1472. 67 R. W. Chlebek and M. W. Lister, Canad. J. Chem., 1971, 49,2943. e8 R. W. Chlebek and M. W.Lister, Canad. J. Chem., 1966, 44, 437. Table 4 Standard enthalpy and entropy of complex formation at 25 "C n AN"lkcu1deg-l mol-1 AS'lcaI deg-l mol-l D' Ligand Methoda Ref. Li Na K Rb Cs Li Na K Rb Cs gsulphate C Ob 0.49 1.01 ---1.3 0.0 --C s-persulphate 2T -4.3 1.6 1.3 -1.0 -17.0 9.6 9.8 3.2 67 ;a carbonate 11T -4.46 ----12.5 ---d ferrocyanide C,5T --0.8 ----13 --32 5 ferricyanide C,5T --0.5 ----8 --32 E 3pyrophosphate C 1.o 1.4 1.7 --20.4 15.2 15.5 --e s-edta C 0.1 -1.4 ---13 3 --L 72f 2-uda 4T -7.0 -8.7 -11.8 ---1+5 -18 -35 --708 cryptate 221 (IVb) C --3.2 ---cryptate 222(IVc) C? --5.8 -11.1 -10.5 -dicyclohexyl-18-crown-6: isomer A C --0 -3.9 -3.3 -2.4 ---3.8 -4.2 -3.7 73 isomer B C --0 -5.1 4.0 ----9.6 -9.3 -73 0 C = calorimetry, nT = measurement of at n temperatures; ref.69; C R. M. Izatt, D. Eatough, J. J. Christensen, and C. H. Bartholomew, J. Chem. SOC.(A), 1969,45; ref. (d) of Table 1; V. P. Vasil'ev and S. A. Aleksandrova, Zhirr. neorg. Khim., 1973, 18, 2055; fin 10%Me,NOH; 8 in 0.1 mol 1-l Me,NNOB at 20 OC; ref. (e) of Table 3. Midgley Because the enthalpy changes are so small, it may be difficult on the basis of calorimetric evidence alone to decide whether a complex is not formed at all or ifAH z 0, e.g. lithium ~ulphate.~S For most of the ligands in Table 4, for which more than one metal ion has been studied, the order for both AH" and AS" is, algebraically, Li > Na > K > Rb > Cs. This is the order of the degree of hydration of the ions and the inverse order of their crystallographic radii, suggesting that the dominant affects are the endothermic removal of water from the hydration sheath with an accompanying gain in entropy, both of which are more pronounced for the more strongly hydrated ions.The stronger ligafids have an overall exothermic heat of reaction, but no more favourable entropy terms. Results for uda70 show enthalpy changes that are more exothermic than those for alkaline-earth and rare-earth complexes of ida and edtae71 Even transition- metal-ion complexes are rarely so exothermically formed. In contrast, the entropy changes are unfavourable, the expected gain on release of water molecules from the hydration sheaths on chelation being apparently counterbalanced by some large loss of configurational entropy, or else the complexes are of the outer- sphere kind. If uda, which is relatively rigid, suffers a large loss of configurational entropy on forming a complex, the flexible edta ion would be expected to show a more pronounced effect, but this is not the case.72 Since the interactions between a crown ligand and the metal ions small enough to fit into the cavity produced by the optimum conformation are the difference between the ions arises from the energy required for removal of the hydration sheath before entry into the complex, the smaller more strongly solvated ions giving a more endothermic heat change.This has been observed for both crown73 and cryptate74 ligands with a selectivity for potassium over sodium.Similarly, the loss of configurational entropy of the ligand should be approximately the same, but desolvation gives a larger entropy increase for the smaller ions. When the ion is too large to fit the optimum cavity, the interaction between metal and ligand will be weaker, giving less exothermic enthalpy changes and more favourable entropy changes.73 Ifthe solvation sheath is not completely removed, the changes may be further complicated. Although there are clear trends in the thermodynamic results, it is desirable that many more data are collected, especially for the weakest complexes, before much confidence can be placed in the absolute values of these quantities. 8 Kinetics There is a serious lack of direct kinetic measurements on reactions involving alkali-metal complexes in aqueous solution.The generally low stability constants 69 J. M. Austin and A. D. Mair, J. Phys. Chem., 1962, 66, 519. 70 H. Irving and J. J. R. F. da Silva, J. Chem. Soc., 1963,448. 'I1 G. H. Nancollas, 'Interactions in Electrolyte Solutions', Elsevier, Amsterdam, 1966. 78 R. G. Charles, J. Amer. Chem. SOC.,1954, 76, 5854.'* R. M. Izatt, D. P. Nelson, J. H. Rytting, B. L. Haymore, and J. J. Christensen, J. Amer. Chem. SOC.,1971, 93, 1619.'* J. P. Sauvage, Thesis, Strasbourg, 1972, (mentioned in ref. 20). 56 1 3+ Alkali-metal Complexes in Aqueous Solution and the consequent high concentration needed to study them lead to formation rates in the micro- to nano-second range.Winkler75 has given a useful account of the problems involved. Ultrasonic absorption has been used to study alkali-metal complexes of uda, nta, edta, and egta.76 The rates of formation vary relatively little (1-2 x los 1mol-1 s-I), but the rates of dissociation show a strong dependence on the metal (0.1-3 s-'). The rates of both formation and dissociation increase with the crystal radius of the ions, i.e. in the opposite sense to the increase in the stability constants. In contrast to the behaviour of transition-metal complexes, the complex form- ation does not consist in a simple substitution of a water molecule in the inner hydration sheath but involves a chelation mechanism in which several solvent molecules are successively replaced, all with comparable rates.The dissociation rate constants for complexes of the macrocyclic diamine 222 (IVc) in D2O were found to be ca. 40s-1 by means of 1H n.m.r. meas~rements.~~ Indirect or secondary effects of alkali-metal complexes on kinetics have been noted. Activated complexes, e.g. [MnO4-K-Mn04I2-, have been postulated to explain the dependence of the rate of isotopic exchange between manganate and permanganate ions on the alkali-metal cation presenL78 The rate of the reaction 2K4Fe(CN)6 + K2S208 32&Fe(CN)6 + 2KzS04 depends on the concentration of potassium ions present79 rather than on the ionic strength, as predicted by the Brernsted theory. Chlebek and Lister68 measured the stability constants of the ion-pairs formed between potassium and each anion and tried to fit the kinetic data to the following reaction schemes; (a) [Fe(CN)sP-+ Sz0s2-; (6) [KFe(CN)#-+ S20a2-; (c) [Fe(CN)epQ-+ KS~OS-;(d) [KFe(CN)6l3-+ KS~OB-.Only scheme (d) fitted all the data. Further workso supported this mechanism for the reaction in the presence of the other alkali-metal cations.Apart from the effect of the different degrees of associ-ation with the several metal ions, the rate constant itself was found to be specific to the cation, increasing in the order of the atomic number. Extrapolation of the rate data to zero alkali metal concentration suggested that the rate of reaction of the uncomplexed anions [scheme (a)] was very slow. The ligand-exchange reaction Caedta2-+ *edta4-S Ca(*edta)2-+ edta4-is retarded by the presence of sodium ion.81 The decrease in rate is not caused solely by thereduction in free edta4- concentration, but is also affected by a direct 76 R.Winkler, Structure and Bonding,1972,10, 1. M. Eigen and G. Maass, Z. phys. Chem. (Frankfurt), 1966, 49, 163; H.Diebler, M.Eigen,G. Ilgenfritz, G. Maass, and R. Winkler, Pure Appl. Chem., 1969,20, 93. 77 J. M. Lehn, J. P. Sauvage, and B. Dietrich. J. Amer. Chem. SOC.,1970, 92, 2916. 78 J. C.Sheppard and A. C. Wahl, J. Amer. Chem. SOC.,1957,79, 1020. J. Holluta and W. Herman, Z. phys. Chem., 1933, A166,453. 8o R.W. Chlebek and M. W. Lister, Cunad. J. Chem., 1967,45, 2411. R.J. Kula and G. H. Reed, Analyt. Chem., 1966,38, 697. Midgley exchange between calcium and sodium complexes.The rate constant for this reaction (34 1 mol-l s-l) is smaller than that for exchange with free edta4- (120 1 mol-l s-l), but larger than that for exchange between the calcium complex and the monoprotonated ligand (5 I mol-I s-l). Identical effects have been re- ported for the exchange between edta and a lead-pdta complex.82 9 Miscellaneous Methods DaviesM has shown how stoicheiometric activity coefficients may be interpreted in terms of ion association, with examples of sodium salts. Osmotic coefficients have been similarly treated.83 The association of sulphates has been measured through the depression of the freezing points of eutectic mixtures, e.g. water-potassium perchlorate, produced by addition of the sulphate salts.84 Ion exchange has given qualitative inf0rmation~~~8~ and solvent extraction has been useful for studying macrocyclic complexes.21#26 Permeability and conduct- ivity measurements on biological and synthetic membranes have been applied to macrocyclic complexes.87-90 10 The Study of Alkali-metal Complexes by Means of Competitive Reactions The commonest method of investigating alkali-metal complexes has been through the change in pH as the metal is added to a solution of a ligand with a suitable acidic function.With weak complexes, the method is very sensitive to errors in the acid or base concentrations, the pH, and the dissociation constants of the Iigand which should be determined in the absence of the metal but using the same apparatus and materials. Sodium tetrametaphosphate complexing was studied from the effect of sodium ions on the copper-tetrametaphosphate equilibrium, measured using a copper- amalgam ele~trode.~l Similar studies have not been made with the modern ion- selective electrodes available for several metals.An indirect polarographic study of lithium and sodium cdta complexing used thallium(i) ion as a tracer.92 Alkali-metal compounds are usually too soluble for solubility measurements to be easily interpreted, but ferricyanide complexing was inferred from the effect of alkali metals on the solubility of luteo-he~acyanoferrate(m).3~~~3 8a J. D. Carr, K. Torrance, C. J. Cnu, and C. N. Reilley, Analyt. Chem., 1967, 39, 1358. 83 W. L.Masterton and L. H. Berka, J. Phys. Chem., 1966,70, 1924. 84 J. Kenttilmaa, Suomen Kem., 1959, B32, 55; C. Sinistri, P. Franzosini, and G. Ajroldi, Ricerca Sci., 1960, 30, 1584. 8b W. Buser, Helv. Chim. Acta, 1951, 34, 1635. F. Nelson, J. Amer. Chem. SOC.,1955, 77, 813. J. M. Diamond and E. M. Wright, Ann. Rev. Physiol., 1969, 31, 581. D. C. Tosteson, Fed. Proc., 1968, 27, 1269. D. C. Tosteson, T. E. Andreoli, M. Tieffenberg, and P. Cook,J. Gen. Physiul., 1968, 51, 373s. S. G. A. McLaughlin, G. Szabo, G. Eisenman, and S. Ciani, Biophys. Sac. Abs. 14th Annual Meeting, 1970, p. 96a. O1 R. J. Gross and J. W. Gryder. J. Amer. Chem. SOC.,1955, 77, 3695. 98 R. Sundaresan, S. C. Saraiya, and A. K. Sundaram, Current Sci., 1967, 36, 255. 93 Yu. I.Rutkovskii and V. E. Mironov, Zhur. neorg. Khim., 1967,12, 3287. Alkali-metal Complexes in Aqueous Solution From the effects of lithium and sodium ions on the rate of ligand substitution between zinc(n)-4-(2-pyridylazo)resorcinol and egta, the stability constants of their egta complexes were calc~lated.~~ 11 Chemical Interference by Alkali-metal Complexing When alkali metals can compete for ligands, allowance must be made for them, especially whzn they are present at high concentrations, e.g. in the determination of stability constants by spectroph~tometry,~~ and solubility measurements,6 PH,~ in calculating the pH dependence of the potential of the ferricyanide-ferrocyanide couple,96 and in the electrode kinetics of the reduction of iodate ion on mercury.g7 Interference in the kinetics of hydrolysis,U ligand substitution,82*94 and ligand exchange81 has been found.The error produced in the dissociation constant of a weak acid by neglecting the complexing of alkali metals has been discussed by Dunsmore and Midgley5 for the case of an acid in a constant ionic medium, where concentrations as high as 3 mol 1-1 are often used. The apparent, K’, and true, K, dissociation constants are related by equation (4), where /3 is the stability constant of the alkali-metal K’= K(l + /3[MD (4) complex and [MI the concentration of free alkali metal. Since the total alkali- metal concentration is usually much greater than the total acid concentration, [MIis virtually constant, and K’will appear to be a ‘good’ constant, i.e.its standard deviation will not be significantly larger than that which would be expected from normal experimental error. Comparison of the dissociation constant with that of an acid that does not form complexes may be misleading. If the constant K’ is used in the calculation of the stability constant of the acid with a second metal, little error will entail, since K’itself compensates for the effect of the alkali metal. 12 Sea Water and Brines The composition of sea water is basic to the chemical aspects of oceanography. The concentration of salts is so high that some form of ion association must be expected, and Garrels and Thompson98 have proposed a model based on the formation of ion-pairs between sulphate, bicarbonate, and carbonate ions and the principal cat ions.Although sodium and potassium are present predominantly (99 as the free ions, the complexing of alkali metals has a considerable effect on the distribution of anions. Sodium forms much weaker complexes than calcium or magnesium, but its high concentration ensures that significant proportions of the anions are present as sodium complexes (sulphate 21 %, bicarbonate 8 %, carbonate 17%). s4 M.Tanaka, S.Funahashi, and K. Shirai, Znorg. Chem., 1968, 7, 573. 95 M. Walser, J. Phys. Chem., 1961, 65, 159. 96 J. D. Winefordner and G. A. Davison, Analyt. Chim. Acra, 1963, 28, 480. O7 P. Delahay and A. Aramata, J. Phys. Chem., 1962,66, 1194. R.M. Garrels and M. E. Thompson, Amer. J.Sci., 1962, 260, 57. Midgley Experimental studies of association in sulphate and carbonate systemsg9 have been carried out, and measurement of the free sodiumloo in sea water showed that it agreed with that predicted by the model. All the preceding work was performed at 25 "C and 1 atmosphere pressure, which are conditions that are unrepresentative of conditions in deep water. Kester and PytkowiczlOl showed that sodium sulphate has a stability constant fl = 3.42 at 2.4 "C,compared with 2.00 at 25 "C,and that fs varies with pressure at 1.5 "Caccording to equation (5), In /3 = 1.26 -0.70 x P (5) where Pis the pressure in atmospheres. Similar but unquantified trends have been observed in the association of sodium carbonate.lO2 It will be noted that the lower temperatures and greater pressures found in deep water have opposite effects on the association constant.The influence of ion-pairing on the pH1°3 and buffer capacity104 of sea water has been discussed and the ion-association model has been applied to calculations on calcium carbonate precipitation,lo5 barite saturation,lo6 and to a derivation of the history of sea-water from the solution equilibria of clay minerals.1O7 An alternative model108 uses the hypothesis of specific ionic interaction,lOg which does not assume the formation of recognizable complex species. While this approach can calculate the osmotic coefficient of sea water more simply, its application is limited for the present by the availability of the appropriate interaction coefficients.Whichever model is used, the associative interactions in sea water must be allowed for in calculations of solubility and exchange processes, and at present the Garrels-Thompson model is more versatile and conceptually simpler. 13 Alkali-metal Complexes in Biology Biological interest in the alkali metals is concentrated on sodium and potassium, whose roles in biochemistry have been reviewed by WilliamsllO and Hughes.lll The concentration of potassium inside cells is much higher than it is outside, whereas the internal sodium level is generally much lower. To maintain an activity R. M. Pytkowicz and D. R. Kester, Amer. J. Sti., 1969, 267, 217; R. M. Garrels, M. E. Thompson, and R. Siever, ibid., 1961, 259, 24.looD. R.Kester and R. M. Pytkowicz, Limn. Oceanog., 1969, 14, 686. lolD. R.Kester and R. M. Pytkowicz, Geochim. Cosmochim. Acta, 1970, 34, 1039. lo4 A. Distkhe and S. Disttche, J. Elecrrochem. Soc., 1967, 114, 330. lo3P. J. Wangersky, Limn. Oceanog., 1972, 17, 1. lo4M.Whitfield, Limn. Oceanog., 1974, 19, 235. lo6W. S. Broeker and T. Takahashi, J. Geophys. Res., 1966,71, 1575. lo6J. S. Hanor, Geochim. Cosmochim. Acta, 1969, 33, 894. lo' J. R. Kramer, Geochirn. Cosmochim. Acta, 1965, 29, 921. lo8 M. Whitfield, Marine Chem., 1973, 1, 251; J. Marine Biol. Assoc. U.K., 1973, 53, 685; Limn. Oceanog., 1974, 19, 235. loDE.A. Guggenheim and J. C. Turgeon, Trans. Faraday SOC.,1955,51, 747. 110 R.J. P. Williams, Quart. Rev., 1970, 24, 331. ll1M.N.Hughes, 'The Inorganic Chemistry of Biological Processes', Wiley-Interscience, London and New York, 1972, p.256. Alkali-metal Complexes in Aqueous Solution gradient requires energy and a selective ‘pump’ which discriminates between the two cations.87J12 The importance of macrocyclic complexes in transport across membranes has been demonstrated, although it is probable that the presence of complexes in the membrane rather than in its aqueous surroundings is the critical feature, since the distribution of these hydrophobic species would favour the former. Tostesona8 has demonstrated that the ionic permeability and selectivity for potassium over sodium of both artificial and natural membranes is increased by macrocyclic compounds, but that naturally occurring antibiotics are more effective than synthetic cyclic polyethers.The influence of the ring size of the antibiotic on cation transport in red blood cells has been discussed.84 An extensive theoretical treatment of ion-binding and transport in membranes has been given by Eisenman et al.l13 The importance of oxidative phosphorylation as a source of energy for bio- chemical reactions has led to the study of complex formation between alkali- metal ions and ATP and other phosphates (Table 5). The binding of alkali-metal Table 5 Stability constants (log p) of alkali-metal ions with phosphate compounds at 25 “C Ligand p Li Na K Rb Cs Ref. ATP a 1.7 1 .o 0.9 0.95 0.9 14 --b0.2 1.53 k 0.03 1.04f0.03 0.93f0.03 -0-2.36f0.04 2.35 f0.04 -58 ADP 0.2 1.15k0.02 0.73f0.04 0.65 k0.05 --b -bAMP 0.2 0.61 kO.04 0.34k 0.04 0.26 & 0.06 -Pyrophosphate * 3.04 2.20 2.32 2.28 2.18 14 0 3.1 f0.2 2.3 k 0.1 2.3 k 0.1 2.3fO.l -c a p variabIe; * ref.(e)of Table 1;C J. A. Wolhoff and J. T. G.Overbeck, Rec. Trav. chim., 1959, 78, 759. ions wasstudied in relation to the effect of edta and pyrophosphate on the ATPase activity of myosin.l4 Graven et al.114 have demonstrated the influence of antibiotics on oxidative phosphorylation and found that the presence of alkali-metal ions is essential for ATP hydrolysis. Jardetzky and Wertz have studied sodium complexing with many common metabolite~~6 The stability constants of complexes with some biologically important ligands are summarized in Table 6.An alternative view of cell chemistry, summarized by is that the cell can be regarded as ‘an organized non-liquid phase’, in which the water is highly 112 W. Schoner, Angew. Chem. Znternat. Edn., 1971, 10, 882. lla G. Eisenman, G. Szabo, S. Ciani, S. McLaughlin, and S. Krasne, Progr. Surface Membrane Sci., 1973, 6, 139. 11* S. N. Craven, H. A. Lardy, D. Johnson, and A. Rutter, Biochemistry, 1966,5, 1729. 116 F. W. Cope, ‘Water Structure at the Water-Polymer Interface’, ed. H. H. G. Jellinek, Plenum, New York, 1972, p. 14. 566 Midgley Table 6 Stability constants (log p) of alkali-metal-ion complexes with metabolites Ligand P Li Na K Ref. acetate 0 0.26 -0.18 - 3 glycollate 0 -0.11 - a lactate 0 -0.20 - - b C - 1.1 kO.1 - 4 pyruvate C - 2.7 2 0.1 - 4 malonate 0 - 0.83 - 8 succinate 0 - 0.3 - 6 malate 0.28 0.45 2 0.04 0.32 k0.06 0.23f0.05 40 0.17 0.38 k 0.02 0.28k0.02 0.18+0.03 57 tartrate 0.2 - 0.28 0.0 5 citrate 0.17 0.83 kO.01 0.70 k0.01 0.59k0.01 57 D-glyceric acid 2-phosphate 0.1 - 1.18 d phospho-enolpyruvate 0.1 - 1.08 d a P.B. Davies and C. B. Monk, Trans. Faraday SOC.,1954,50, 128; b P. B. Davies and C. B. Monk, Trans. Faraday SOC.,1954,50,132; C ‘Tris’ buffer; d F. Wold and C. E. Ballou, J. Biol. Chem., 1957, 227, 301. structured and the alkali-metal cations are bound by macromolecules. In this model, energy is not necessary to maintain the difference in ion concentration across the membrane.Studies of intracellular complexing by means of 23Na n.m.r. and cation-sensitive microelectrodes have indicated substantial degrees of association (30-70 %) in brain,51 kidney,S1 liver,52 muscle,53J16 and other tissues. 39K n.m.r. studies also suggest that potassium in cells is complexed.54J16 Damadian117 has shown that potassium is bound by protein in cell-free extracts. Lewis and Saroffll* have measured binding constants for sodium and potassium ions on muscle protein. 14 Analytical Applications of Alkali-metal Complexes The complexes are generally too weak to have much direct application to analysis, but ‘Thoron’, 0-(2-hydroxy-3,6-disulpho-l-naphthylazo)benzenesonicacid, forms a stable orange complex with lithium in alkaline s0lution,~1~ permitting the spectrophotometric determination of lithium in the range 1-10 pg ml-1.The colour is more intense in aqueous acetone, extending the range to 0.1 pg ml-1. A fifty-fold excess of sodium is tolerable. Carr and Swartzfager12@ have developed a method for the complexometric titration of sodium in the presence of the heavier alkali metals. The sample, after adjustment to pH 12.6 with piperidine or caesium hydroxide, is titrated with cdta 116 G. N. Ling and F. W. Cope, Science, 1969, 163, 1335. 11’ R. Damadian, Science, 1969, 165, 79. 118 M. S. Lewis and H. A. Saroff, J. Amer. Chem. SOC., 1957, 79, 21 12; H. A. Saroff, Arch. Biochem. Biophys., 1957, 71, 194. lL9P. F. Thomason, Analyt.Chem., 1956, 28, 1527. 120 J. D. Carr and D. G. Swartzfager, Analyt. Chem., 1970, 42, 1238. Alkali-metal Complexes in Aqueous Solution solution.The end-point is detected by means of a sodium-sensitive glass electrode. The method is accurate to within 2 per cent for sodium-ion concentrations as low as 10-3 moll-’. Bobrowski and Zarebskil2’ have used cdta to separate the a.c. polarograms of potassium and sodium. In 0.02 mol I-l-cdta, the separation of the half-wave potentials is increased from 25 to 80 mV, giving clearly separated peaks, the am- plitudes of which were found to be proportional to the concentrations of the ions. Busers5 has separated lithium, sodium, and potassium on a cation-exchanger in the tetramethylammonium or dimethylammonium form by eluting with uda solution.Nelsons6 showed that by adding edta to the solution, lithium, sodium, and caesium could be separated on an anion-exchange resin. Several reagents for the selective precipitation of alkali metals are available.122 Sodium is most commonly precipitated with uranyl nitrate to give a compound NaM(UO2)3(OAc)9,6H 20, where M represents zinc, magnesium, or nickel(@, but a-methoxyphenylacetic acid has also been used. Potassium, rubidium, and caesium are precipitated by hexanitritocobalt(m) ion, hexanitrodiphenylamine (dipicrylamine), and 6-chloro-5-nitrotoluene-3-sulphonicacid. The commonest reagent for these ions is tetraphenylborate, which has been used in gravimetric, voltammetric, and potentiometric determinations. Caesium can be determined in the presence of potassium by means of ~yanotriphenylborate.~~~ Macrocyclic compounds have been used to make ion-selective electrodes of the neutral carrier type.124 Although cyclic polyethers have been used, the high selectivity of valinomycin for potassium has made this compound the basis of a number of commercially produced electrodes.A neutral carrier electrode for sodium has been developed125 which has the advantage over glass electrodes of not being affected by proteins in media such as blood serum. Macrocyclic compounds have obvious potential for solvent extraction, as masking agents, and even as titrants, but applications126 have been limited, presumably by cost and availability. lZ1 A. Bobrowski and J. Zarebski, Gem. unalit., 1970, 15,457. lZa S. Kallman, ‘Treatise on Analytical Chemistry’, Part 11, Vol. 1, ed. I. M. Kolthoff and P. J. Elving, Interscience, New York and London, 1961, p. 301; D.D. Perrin, ‘Organic Complexing Reagents’, Wiley-Interscience, New York and London, 1964, p. 171. 12s A. Bauman, Tuluntu, 1968, 15, 185. lZ4 ‘Ion-selective Electrodes’, ed. R. A. Durst, National Bureau of Standards Special Publi- cation 314, U.S. Dept. of Commerce, Washington, D.C., 1969. Ips D. Ammann, E. Pretsch, and W. Simon, Anulyt. Letters, 1974, 7(I), 23. lZ6J. W. Mitchell, A.C.S. Abstracts of Papers, 167th A.C.S. National Meeting, Los Angeles,1974, Analyt. Chem. No. 173.

ISSN:0306-0012    

DOI:10.1039/CS9750400549

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

The Chemical Applications of Advances in Fourier Transform Spectrometry By the late J. Chamberlain DIVISION OF ELECTRICAL SCIENCE, NATIONAL PHYSICAL LABORATORY, TEDDINGTON, MIDDLESEX, TWI 1 OLW Completed and revised by G. W. Chantry and N. W. B. Stone NATIONAL PHYSICAL LABORATORY, TEDDINGTON, MIDDLESEX, TWl1 OLW 1 Introduction Fourier transform spectrometry (FTS) is now a well-established technique and its use will soon be routine in chemical laboratories needing coverage of the infrared region from 5 cm-1 to 10oO cm-l (or more). Recent developments in modulation techniques, in detectors and in computing systems have produced improved signal-to-noise ratios, decreased observation times, and new applications such as time-resolved spectrometry of fast chemical reactions.The basic advantages of FTS over conventional grating spectrometry are outlined below and the advances are described and illustrated by some examples of chemical interest. 2 Background After 15 years’ development, Fourier transform spectrometry is now widely used, for examination of the infrared spectral region.lS2 It owes its present standing to the work of Janine and Pierre Connes, Mertz, Strong, and Gebbie, amongst others. Although Michelson himself was able to draw some conclusions about the structure and widths of spectral lines from visibility curves recorded with his two-beam interferometer, FTS was really born when Fellgett, at the Ohio State Symposium on Molecular Spectroscopy in 1952, pointed out the multiplex advantage to be gained by using a high-speed computer to extract the spectral information from an interferogram trace.Good reviews of the subject have been given by LOW^-^ and Griffiths et aZ.,6who have emphasized the chemical aspects. A FT spectrometer consists of a source, a two-beam interferometer, a detector, and a computer (Figure 1). The interferometer part of the system is typified by a Michelson interferometer although variants are alsoused. Since an interferometer 1 R. J. Bell, ‘Introductory Fourier Transform Spectroscopy’, Academic Press, New York, 1972; G. W. Chantry, ‘Submillimetre Spectroscopy’, Academic Press, London, 1971. G. A. Vanasse and H. Sakai, Progr. Optics, 1967, 6, 259. M. J. D. Low, Analyt. Chem., 1969,41,97A.M. J. D. Low, Naturwiss., 1970, 75, 280. M. J. D. Low, Internal. Sci. and Technol., February, 1967, p. 52. P. R. Griffiths, C. T. Foskett, and R. Curbelo, Appl. Spectroscopy Rev.,1972, 6, 31. The Chemical Applications of Advances in Fourier Transform Spectrometry Broad-band source3Interferometer Ilnterference function s spectrometer II Cornputer 1 Spec t r urn 9Interference function1 -spectrum Figure 1 Block diagram showing the essentials of a FTspectrometer has circular apertures, rather than narrow rectangular slits like a conventional spectrometer it has a greater radiation throughput. This confers an advantage, usually called the Jacquinot’ advantage, over the conventional spectrometer. The basis of FTS is that the frequency distribution of the power incident on the detector (i.e.the spectrum) is obtainable by Fourier analysis of the variable part of the interfering power observed by the detector as the path-difference x between the two partial beams of the interferometer is changed.The interfering power [the interference function Z(x)] consists of a constant (x-independent) part 10and a variable (x-dependent) part F(x) called the (power) interferogram.In the ideal case this is a perfectly symmetrical function. The detected spectral power B(v”) of wavenumber v” and the interferogram F(x) are a FT pair: F(x) = 2 B(f)cos 272flx dv” = FcM{B(q) (1)K and fa B(f) = J F(x)c0~2nfxdx= Fcos{F(x)) . --oo Thus, if F(x) is measured, B(fl),given by (2), is calculable from it.The boxed region of Figure 1 is, therefore, equivalent to a spectrometer. Figure 2 shows the relationship between I(x), F(x), and B(6, In practice, (2) can be evaluated only over a finite range of x up to a maximum value D. The effect of this is to introduce finite resolution and spurious structure into the calculated spectrum.* The width and shape of the spectral window can * Under certain conditions this spurious structure does not appear (see, e.g. W. J. Burroughsand J. E. Harries, Infrured Phys., 1971, 11, 99). 7 P. Jacquinot, Reports Progr. Phys., 1960, 23, 267. Chamberhin (a1 1nter ference function (b1 In terferogram ,5(s)=jm ~(x)cos 2nLi;x dx (C1 Spectrum Figure 2 Relationships between (a) the interference function I (x), (b) the interferogram F(x), and (c) the spectrum of the detected radiation B(?) be adjusted and optimized (apodized) through the introduction into (2) of a weighting function w(x) which tapers F(x) to zero at x = D.The resolution R is given by R = l/wD , (3) where w is a number determined by the weighting function and typically is one. In the absence of weighting, w = 2. The resolving power is given by though there is a practical limit to the realizable resolving power imposed by the geometrical configuration of the optical system. * In a practical system it is not the power I(x) incident on the detector that is * The practical condition to be satisfied is that D <n/(W),where 9is the limiting solid angle of the optical system.The Chemical Applications of Advances in Fourier Transform Spectrometry recorded but Ndigitized samples of the voltage V(x)from the detector amplifier.* All detectors produce ‘noise’, and it is preferable in FTS that the limiting noise should come from the detector. This is because the Fourier spectrometer is a multiplex instrument and all (say M) elements in the observed spectral band are detected for the whole time of observation T (rather as in a spectrograph con- taining a photographic plate).t As Fellgetts showed, when detector noise is the limiting factor, this gives the multiplex spectrometer a gain in signal-to-noise ratio in the resultant spectrum of M* when comparison is made with a scanning Spectrometer that also observes the M elements in time T.When the limiting noise is not from the detector, as is frequently the case in the near-i.r., where so-called photon noise can dominate, the multiplex advantage is cancelled (for photon noise) or transformed into a disadvantage (for source or signal noise). Detector noise increases as the frequency of operation falls, so d.c. detection and amplification are not usually used to record the interference function. Instead, the radiation is modulated and a.c. detection is used. This is a common i.r. practice9 and the conventional way to modulate the radiation is by means of a periodic shutter such as a rotating sectored disc (amplitude modulation). The htderogram which results from the x-variation of the rectified output of the amplifier is once again perfectly symmetrical for the ideal case.Interferometric instruments following these principles have come into wide- spread use in the infrared, and commercial systems are available for far4.r. use.10-12 The operator has in almost all instances used his own computer. FTS has been applied to the measurement of emission, transmission, absorption, and (power) reflection spectra, thereby displacing conventional scanning spectro- meters. Chantryla has given a review of applications to inorganic structure chemistry and Chantry and Chamberlain14 have described applications to the physical chemistry of polymers. The throughput and multiplex advantages combine to give spectra which have greater radiometric accuracy at a given resolution or higher resolution without loss of signal-to-noise ratio.Alternatively, the observation time is reduced, materials with greater absorption (or weaker reflection) may be studied, or 2N samples if x is changed from -D through 0 to D, as is often the case if F(x) and V(x) are not symmetrical. However, complete two-sided records are not essential as corrections for asymmetry can be made. The number of spectral elements is given by the detected spectral bandwidth divided by the resolution: (vmax -vmin)/R. * P. B. Fellgett, J. Phys. Radium.,1958, 19, 187. R. A. Smith, f.E. Jones, and R. P. Chasmar, ‘The Detection and Measurement of Infrared Radiation,” Oxford University Press, Oxford, 1968. J. N.A. Ridyard, J. Phys. Radium., 1967, 28, C-2, 62. l1 G. W. Chantry, H. M. Evans, J. Chamberlain, and H. A. Gebbie, Infrared Phys., 1969,9, a5. Is R. C. Milward, Infrared Phys., 1969,9, 59. G. W. Chantry, in ‘Essays in Structural Chemistry’, ed. A. J. Downs, D. A. Long, and L. A. K. Staveley, MacMillan, London, 1971, p. 91. G. W. Chantry and J. Chamberlain, in ‘Polymer Science’, ed A. D. Jenkins, North-Holland, Amsterdam, 1972, p. 1330. Chamberlain weaker sources may be examined. Which of these improvements applies depends on the experimental conditions. Three further advantages are the virtually complete absence of stray radiation, the ability to place the specimen just in front of the detector rather than near the source, where it may deteriorate, and the ability to place a good-quality specimen in one arm of the interferometer and obtain a refractive index spectruml5-17 from the resultant shift and distortion of the interferogram (dispersive17 or asymmetric15J6 FTS).This is valuable throughout the infrared generally for the evaluation of integrated absorption strengths18Jg without recourse to intensity measurements, and in the far4.r. in particular where the complex permittivity can be evaluated, thereby giving con-tinuity with microwave data.14~20 In certain specialized applications, however, Fourier spectrometers are not easily used, e.g. in the measurement of the spatial variation of the spectrum of a source such as a flame. Some (chemists particularly) have considered a particular drawback to be the generation by the instrument of a record which requires transformation in a computer before the desired spectrum is available. Where interferometric technique is the only way to obtain a spectrum, this drawback is only of academic interest but where there are competing conventional methods it does represent a real psychological factor affecting the choice of spectrometer.However, recently there have been advances in modu- lation techniques, in detectors, and in computers and computing generally, all of which have enhanced the advantages just listed and have improved the quality and extended the scope of spectra obtained by FTS. Some of these developments are now becoming commercially available. With these advances, a spectrum is available from the interferometer-computer combinations in much shorter time than any conventional system could achieve, for the same spectral quality, and this particular objection should rapidly become a thing of the past.3 Recent Improvements A. Modulation.-In the majority of interferometers, the path-difference is changed slowly and monotonically from 0 to D (or -D to D). The interference record is then said to be generated aperiodically, and superimposed modulation of the radiation is required for a.c. detection. In amplitude modulation (AM) the entire magnitude of the detected signal is chopped periodically at a frequencyfo (Figure 3a). The tuned (synchronous) amplifier delivers to the analogue-to-digital converter a rectified voltage whose magnitude is directly proportional to the power in the interference function.21 This includes the steady x-independent E.E. Bell, ZnfraredPhys., 1966, 6, 57. l6 E. E. Bell, ‘Proceedings of the International Conference on Fourier Spectroscopy, Aspen, 1970’, Air Force Cambridge Special Report, No. 114, 1971, p. 71. I’ J. Chamberlain, J. E. Gibbs, and H. A. Gebbie, Infrared Phys., 1969, 9, 185. l8 J. Chamberlain, J. Quant. Spectroscopy Radiative Transfer, 1967, 7, 151. Is J. G. Chambers, A. J. Barnes, and W. J. Orville-Thomas, Chem. Physics., in press; J. G. Chambers, M. J. Phillips, A. J. Barnes, and W. J. Orville-Thomas, Adv. MoI. Relaxation Processes, in press. 2o J. Goulon, J.-L. Rivail, J. W. Fleming, J. Chamberlain, and G.W. Chantry, Chem. Phys. Letters, 1973, 18, 211. s1 J. Chamberlain, Infrured Phys., 1971, 11, 25. Vibrating mirrop Rapid-scanning ........... mirror7'-'lr PX4 c-O-D--,................. ------* ).............. AM PM I I1 I Aperiodic Periodic (01 (b) (C I Figure 3 Modulation in two-beam interferometers: (a) aperiodic generation wtth amplitude modulation, (b) aperiodicgeneration with phase modulation, (c)periodicgeneration-rapid scanning. M, = movable' mirror, M,= 'fixed' mirror, Mg'= image of M2in arm 1 of interferometer,B = beam divider Chamberlain contribution and superimposed signal-carried noise, both of which are unwanted for the Fourier transformation. Moreover, on average, the chopper halves the power available for detection. Until this year, all commercial aperiodic inter- ferometers had this type of modulation.A recent advance has been to use modulation based on interference, which gives strong discrimination in favour of the x-dependent part of the interference function. This improvement was first suggested by Mertz22 in 1958 but was not successfully applied until the period 1965-70 in the near- and mid4.r. (by J. Connes et aL23in France) and in the far4.r. (by Chamberlain, Gebbie, and their colleague^^^^^^ in the U.K.). This has been variously called internal mod~lation,~~ path-difference modulation (PDM), or phase modulation21 (PM). No chopper is used and modulation of the radiation is achieved by vibrating one of the inter- ferometer mirrors (Figure 3b).The form of modulation is selective over a broad spectral band, and so the amplitude of the vibration is usually chosen to be half a wavelength for radiation near the centre of the spectral band of interest. The result of the vibrating mirror motion is to produce a phase-modulated interference record which resembles the first derivative of the AM interference record and which is ideally antisymmetrical (Figure 4). The background con- tribution lo is not detected. Zero path-difference, manifested by the grand maxi- mum in AM, is given by a negative-going zero-crossing which occurs where the voltage variation is virtually linear with path difference. Phase modulation is most effective in improving signal-to-noise ratio where AM PM i L-I I 11--"-0)ol I I I I I I -t---t -0 t Path-difference Path-dif ference Figure 4 Aperiodic interference records: AM and PM L.Mertz, J. Phys. Radium, 1958, 19, 233. 23 J. Connes, P. Connes, and J. P. Maillard, J. Phys. Radium, 1967, 28, C-2, 120. a4 J. chamberlain and H. A. Gebbie, Znfrared Phys., 1971, 11, 56. J. Chamberlain, W. J. Burroughs, and J. E. Harries, in 'Mesospheric Models and Related Experiments', ed.G. Fiocco, Reidel, Dordrecht, Holland, 1971. 575 The Chemical Applications of Advances in Foirrier Trans form Spectrometry signal-carried noise dominates. A significant part of this noise is carried by the background component ZO and it is reduced, if not eliminated, with the suppres- sion of lo.Thus, PM has been used with enormous success in atmospheric and astrophysical spectrometry by P.Connes26 and Harries and B~rroughs.2~ Even in the laboratory, however, it offers advantage^;^^^^^^^^ the elimination of the chopper effectively doubles the detected power; the effects of source drift and fluctuation are reduced; and since the voltage interference record has a stable zero background, the digital scale may be filled with the variations of the interference signals simply by changing the amplifier gain, and no offset is required (as in AM). This latter aspect is particularly valuable in dispersive FTS, where 10may be an appreciable fraction of the maximum interfering power17 and overload due to it can occur long before the interferogram fills the digital scale.The linearity of the PM record near the principle zero-crossing enables interpolation, for the evaluation of fringe shifts, to be more accurately performed than when AM records are used and non-linear interpolation is required.24 PM is now commercially available with the Grubb-Parsons far4.r. cube interferometer,ll and Figure 5 shows an example of a spectrum from 10 to 40 cm-l obtained with such an interferometer fitted with a Rollin indium antimonide 'I30120 Il~~,,I,,,,l,,,,l,,,,1,,,,1,,,, 10 15 20 30 35 $1 NaO 250 TORR 933mm Figure 5 Transmission spectrum of N,O. The resolution (unapodized) is 0.05 cm-l. The gas cell was 933 mm in length and the gas was atpressure 250 Tonand temperature 23.5" C.The interferometer had PM and a Iiquid-helium-cooled Rollin detector. The observation time was 50 min P. Connes, Ann. Rev. Astronomy, 1970, 8, 209. 27 J. E. Harries and W. J. Burroughs, Quart.J. Roy. Meteorological SOC.,1971, 97, 519. P. Connes, in ref. 16, p. 121. 2o G. J. Davies and J. Chamberlain, J. Phys. (A), 1972, 5, 767. 576 Chamberlain detector. The ~pectrum,~OJ~ which is unapodized and has a resolution 0.05 cm-1, represents the transmission of NzO obtained from the ratio of a specimen spectrum to a background spectrum, each obtained by Fourier transformation of a single interferogram. It is possible to achieve higher resolution than this in the far-i.r. but it is not common because the weakness of available sources sets a practical limit.The ultimate resolution in a spectrum, both emission and absorp- tion, is set by the available integration time and by the relative magnitudes of the power in the resolution bandwidth compared to the overall noise level. FTS techniques permit a much higher resolution to be achieved before practical limitations set in, because of the more efficient use of the energy of the source, but it does not seem likely that resolution better than 0.01 cm-l will be attainable in practice. For very many of the observations which the chemist would want to make, high resolution is unnecessary, and then the advantages of PM FTS can be turned to providing much better radiometric precision than can be obtained with dispersive instruments.This is most desirable in many connections but one // 2 3 4 5 6 7 Y / THz Figure 6 Power absorption coeficient of a solution of p-difuorobenzene (6.16 mol I-' in carbon tetrachloride). Three independent spectra are superimposed to demonstrate the reproducibility. The resolution is 4 cm-l and the temperature of the solution was 20 f2 "C. The eflects of reflection losses at the boundary between the liquid layer and the cell windows have been eliminated. The spectrum clearly shows the broad absorption centred on 60 cm-I that is characteristic of the liquid state. The interferometer had PM and a Golay-celldetector [Adapted fromJ. Phys. (A), 1972,5,767] 30 J. W. Fleming, private communication. *l J. W. Fleming and J. Chamberlain, Infrared Phys., 1974, 14, 277; J.W. Fleming, 'I.E.E.E. Transactions on Microwave Theory and Techniques, 1974,' MTT-22, p. 1023. The Chemical Applications of Advances in Fourier Transform Spectrometry important case is the dielectric spectroscopy of liquids at submillimetric wave- numbers. Figure 6 gives an indication of the reproducibility obtainable with PM even when a Golay-cell detector is used.29 Perhaps the most striking results to be obtained by FTS have been recorded in the near-and mid-i.r. by J. and P. Connes and their colleagues, using inter- ferometers equipped with PM and other refinements. Figures 7 and 8 show examples of the spectra that have been obtained,32 these being of holmium in emission and of N20 at low pressure in absorption, respectively.The combined advantages of modern FT spectrometers have enabled J. Connes et a1.33 to record i.r. spectra with the highest resolving power yet achieved (9z lo6). This is clearly of importance to those interested in the detailed quantum pro- perties of molecules. Apart from the aperiodic type of interferometer described above, a second type, the rapid-scan interferometer, has been developed in recent years. The movable mirror is translated rapidly, but at constant speed,repetitively from- D to +D, scanning the whole interference function with each sweep.6e34p35 A com-mon mode for the drive is the sawtooth of Figure 9, which gives a periodic interference record as shown. The frequency structure of the voltage leaving the Figure 7 Holmium emission spectrum.The resolution in the upper two traces is 0.02 cm-l, in the lower trace it is 2 cm-l. The intensity scale in the upper trace is ten times that in the middle trace. For the maximum resolution 1O6 interferogram samples were required and the observation time was 10 h. The interferometer has PA4 and a PbS detector. The lines show the hyperfne structure in the holmium spectrum (Adapted from Nouv. Rev. d’Optique Appl., 1970,1, 3) G. Guelachvili and J. P. Maillard, in ref. 16, p. 151. a8 J. Connes, H. Delouis, P. Connes, G. Guelachvili, J. P. Maillard, and G. Michel, Nouv. Rev. d’Optique Appl., 1970, 1, 3. a4 L. Mertz, J. Phys. Radium, 1967, 28, C-2, 87. E. V. Loewenstein, in ‘Far Infrared Spectroscopy’, ed.K. D. Moller and W. G. Rothschild, Wiley, New York, 1971, p. 672. Chamberlain &--I /\ 5 I ig z The Chemica f App fications of Advances in Fourier Transform Spectrometry detector depends on the scan speed and the spectrum being observed, and usually lies in the audio-frequency range. Wavenumber v” is converted into frequency according tof = vP, where u is the speed of the moving mirror. The detector and amplifier must have a sufficiently wide and flat response to accomodate this range or else there will be some loss of the multiplex advantage. Samples of the periodic interference record are taken at each of the required values of x, put into store, and corresponding samples of successive scans are co-added for as long as necessary (usually until the signal-to-noise ratio is adequate).Rapid scanning (like PM) was first suggested by Mert~,~~ who applied it to astrophysical and astronomical problems where the scan rate can be made fast enough to ‘beat’ scintillation noise superimposed on the incident radiation. The work of Mertz led to the construction by Block Engineering of the near4.r. rapid-scanning cube interferometer. In the laboratory, rapid scanning enables single interferograms to be recorded very quickly6 (in a few seconds or less, depending on resolution) and co-added as necessary; as there is no chopper, all the power leaving the source is detected. However, the method becomes increasingly hard to use if D and/or N become very large, or if the rate of data acquisition becomes very high.Subject to these restrictions, the technique works very well, as Griffiths et aZ.6 have shown using the commercial Digilab instrument designed for the near- and far4.r. Rapid scanning has also been particularly rewarding in atmospheric and astrophysical spectrometry, as Hanel et aL36 have demonstrated. In addition to the usual types of spectrometry, rapid-scanning is particularly suitable for studying time-varying processes for which the variation is similar to or slower than the period of scan An important example for chemists is gas chromatography monitored by its i.r. spectrum: Figure 10 illustrates a g.c. Figure 9 Periodic interference record (upper trace) produced by periodic saw-tooth change of path-difference (lower trace).36 R. Hanel, B. J. Conrath, W. A. Hovis, V. G. Kunde, P. D. Lowman, J. C. Pearl, C. Prabhakara, B. Schlachman, and G. V. Levin, Science, 1972, 175, 305. 87 J. 0. Lephardt and P. R. Griffiths, private communication. Chamberlain result of Lephardt and GriffithP in which three distinct spectra separated by seven seconds overall are shown. The spectra were recorded ‘on the fly’. A measurement such as this would be impossible without trapping if a grating spectrometer or aperiodic interferometer were used. B.Detectors.-The second significant advance has been in detector design and construction.3*~39This has led to detectors with higher sensitivity, lower noise, and greater speed than hitherto. In addition to improving signal-to-noise ratios it is possible to use high frequencies for the modulation of the radiation, and it has proved feasible to use FTS to obtain time-resolved spectra with microsecond resolution.Cooling by liquid nitrogen (as in PbS for the near-i.r.-see Figure 8) or by liquid helium (as in InSb for the far-i.r.-see Figure 5) is common but not universal. The use of liquid refrigerants presents few problems, for modern cryostats are easily filled and have long holding times. Even the ambient-temper- ature Golay cell can give good quality spectra well into the far4.r. if used with PM (see Figure 6). However, the Golay cell has a slow response, and a parlicu- Acetone I720 Carbonyl I 800 I 1200 I 1600 Izoo0 Wavenumber Icm-1 Figure 10 Transmission spectra obtained for one microlitre of 50:50 mixture of acetone and benzene injected into gas chromatograph so that separation of acetone and benzene peaks was 7 s.The spectra were recorded as follows: (a) between leading and trailing edges of acetone; (b) between trailing edge of acetone and leading edge of benzene; (c)between leading and trailing edges of benzene. The interferometer had rapid scanning and a TGS detector 38 K. D. Moller and W. G. Rothschild, ‘Far Infrared Spectroscopy’, Wiley, New York, 1971, p. 57. ‘Infrared Detection Techniques for Space Research’, ed. V. Munro and J. Ring, Reidel, Dordrecht, Holland, 1972. 581 The Chemical Applications of Advances in Fourier Transform Spectrometry larly important ambient-temperature detector is the pyroelectric device based on triglycine sulphate.Its performance is comparable to that of the much-used Golay cell in nearly all respects except response time, which is much faster. It can, therefore, be used in rapid-scan interferometerss (see Figure 10). C, Computing.-The third major development has been in c0mputing,4~ where hardware and software with greater speed and capacity have become available. The total time Ttot taken to obtain a spectrum is the sum Ttot = T + Ttrans + Tcornp (5) of the time T taken to observe the interference record, the time Ttrans taken to transfer the data to and from the computer, and the time Tcomptaken to compute the spectrum. The value of T can range from a few seconds to a few hours depending on the spectrum under observation.It is less than the time that would be required for a comparable measurement using a grating spectrometer. The value of Tcomp depends on the size and type of computer as well as the nature of the program used with it and the quantity of data to be computed; it is rare nowadays for Tcomp to be inconveniently large. If the interferometer is operated separately from the computer (off-line) Ttrans may represent the major contri- bution to Ttot (e.g., Ttra-z 24 h for a central batch-process general-purpose computer). If there is a remote-access data-link or if the computer is local and used by the spectroscopist himself, Ttra, may be markedly reduced (although Tcomp may increase slightly owing to the need to use a smaller computer).If the interferometer is operated on-line to the computer, Ttrans can be ignored. For most general-purpose on-line computers, computation immediately follows observation (this is ‘false’ real-time computation) while for some modern special- purpose hard-wired computers, computation accompanies observation (real-time computation) and Ttot -+ T. The advantages of on-line operation are firstly that the interferogram need never be seen by the operator, who may regard the whole assembly of interferometer -detector -amplifier -A/D converter -computer simply as a spectrometer (Figure 1) and use it as such; secondly, the computer may also be used to control the interferometer. The Digilab Fourier spectro- meters6 are of this type.4 Timeresolved Fourier Spectrometry The advent of very fast detectors (response times less than 0.1 ps) has made it possible to extend FTS to the measurement of time-resolved spectra such as are emitted during reactions between transient species41 or by pulsed plasma^.^^.^^ All parts of the interferometer are fixed, with the movable mirror set to give a path-differencexi; the pulse of radiation is emitted and the corresponding detec- tor signal V(xr;t) is sampled at a series of preselected times and stored (Figure 40 J. Comes, in ref. 16, p. 83. 41 R. E. Murphy and H. Sakai, in ref. 16, p. 301. 42 J. Chamberlain, A. E. Costley, and D. D. Burgess, ‘Proceedings of the Symposium on Submillimetre Waves, 1970’, Polytechnic, New York, 1971, p.573. 43 A. E. Costley, J. Chamberlain, and D. D. Burgess, I.E.R.E. Conference Proceedings No. 22: Infrared Techniques, 1971, p. 191. Chamberlain 11). This process is repeated for each value of xi. A sequence of data is then assembled for all x for each time t;r of interest. These sequences constitute the interference records that would have been obtained if observations could have been over all x at each time tj. Fourier transformation follows as usual. It is assumed that the pulsed source is reproducible: if it is not, some form of monitor- ing and compensation is necessary. The technique has been applied in the millisecond domain by Murphy and Sakai41 (in the U.S.A.) to chemical reactions (Figure 12)and in the microsecond domain by Costley et aZ.43~~(in the U.K.) to pulsed plasmas. As an example, the time development of the spectrum emitted by a discharge in active nitrogen is shown in Figure 12.Nitrogen is exited to high vibrational levels by a microwave discharge and de-excited by a resonant transfer during collisions with C02 molecules. The C02 molecules then decay, emitting radiation in the vibrational hot bands. The spectrum is dominated by the decay over 100 ms of the C02 band at 2340 cm-1. There is also evidence for emission from N2O near 2223 cm-1. 5 Summary From what has gone before, it will be realized that FTS instruments are now superior to dispersive instruments throughout the entire i.r. region. In the far-ire, where the cost of an interferometer plus mini-computer is less than that of a grating spectrometer, the superiority is absolute, but in the middle-i.r., where the Figure 11 Construction of time-dependent interferograms with a t wo-beam interferometer.(a) Pulse V(xt,t) detected at a particular setting xt and sampled at a series of times tj (j = 1, 2, 3, . . .). (b). The interferograms that can be constructed by repeating the procedure in (a) at a series of values of xi pp A. E. Costley, R. J. Hastie, J. W. M. Paul, and J. Chamberlain, Phys. Rev. Letters, 1974, 33, 758. The Chemical Applications of Advances in Fourier Transform Spectrometry 20 I6 -I .-...-.. . . --.. .-.-. I2 h -* CY” C ., c 0 5;e 4 I 2100 2200 2300 2400 Wavcnumbcr /cm-l Figure 12 Time-dependence of spectrum emitted during reaction between active nitrogen and COz.The resolution is 40 cm-l.The times (referred to peak emission as origin) are (1) 0, (2) 35, (3) 55, and (4) 75 ms. The detector was InSb (Adapted from Proceedings of the International Conference on Fourier Spectroscopy, Aspen, 1970, Air Force Cambridge Special Report, No. 114, 1971, p. 301) cost of an FTS system is still greater than that of a grating instrument of com-parable quality, the superiority is qualified, One must ask therefore what a chemist wants from an i.r. spectrometer and then see how the two systems perform in providing what is required before one can come to form conclusions about their relative standing.It is also important to consider other instrumental approaches to the same problem: thus laser Raman spectroscopy covers the same frequency range as does i.r. spectroscopy, and there are now available tunable lasers which can provide high-resolution non-dispersive spectroscopy throughout much of the infrared. The purposes to which a chemist puts a spectroscopic instrument are multi- fariou~,~~but some of the principal ones are: I Static systems: (i) structural diagnosis of organic molecules, M See e.g. I. R. Beattie, Chem. SOC.Rev., 1975, 4, 107. Chamberlain (ii) determination of the symmetry and detailed structure of small molecules, particularly inorganic molecules, (iii) determination of equilibrium data, (iv) determination of the microstructure of complex solids, particularly polymers, (v) study of intermolecular interactions in high-pressure gases and condensed phases, (vi) detection of constituents in a mixture, especially those present at low concentrat ions, (vii) the study of adsorbed layers and of chemical-physical effects at interfaces, I1 Dynamic systems : (i) detection of transients and short-lived intermediates, (ii) detection of unstable complexes, (iii) determination of rate constants, (iv) elucidation of reaction mechanisms.For I(i) the extension of spectral range down into the farir., which FTS tech-niques have made practicable, is of little consequence, for most characteristic features lie above 200 cm-1.However, the much higher signal-to-noise ratio which is available in the ‘fingerprint’ region with FTS does permit the extension of micro-sampling techniques and does permit organic substances to be studied in normally prohibitive milieu, aqueous solution for example. In the case of I(ii), the far-i.r. capability is essential since all the vibrational fundamentals that can be observed (i.e. are not forbidden) need to be observed before an unambiguous assignment can be arrived at. Modern laser-Raman spectroscopy has made, of course, an enormous contribution in this field, but it cannot replace i.r. obser- vations entirely. Even for non-centrosymmetric molecules, it is commonly found that both the Raman and the i.r. spectrum are necessary before a confident assignment can be attempted.A good example of this can be found in the recent determination at the National Physical Laboratory (N.P.L.) of the structure of phenyl isocyanate in the liquid phase.46 Vibration-rotation spectroscopy of molecules in the vapour phase can give considerable information about the shapes of the absorbing molecules, and here the high resolution capability of modern FTS instruments is very relevant. Even if individual lines cannot be resolved, the form of the band envelope can often be interpreted unambiguously, but again it is preferable to work at the highest resolution that can be achieved. FTS systems are commercially available that can do 0.1 cm-1 in the mid-i.r., and there is talk of a commercial instrument that could achieve 0.01 cm-1.There does not seem any insuperable difficulty which would prevent the ‘home constructor’ making an instrument that could do 0.03 cm-l. However, these high- resolution instruments are, or would be, very expensive, and this factor has to be 46 G. W. Chantry, E. A. Nicol, D. J. Harrison, A. Bouchy, and G. Roussy, Spectrochim.Actu, 1974, 30A, 1717. 585 The Chemical Applications of Advances in Fourier Transform Spectrometry considered in arriving at a balance. Also, the competition from tunable lasers has to be borne in mind, for the spin-flip laser is capable of 0.01 cm-1 in two bands centred at 5 and 10 pm,47 and diode lasers capable of 0.001 cm-1 are available over much of the important mid4.r. region.48 These are not cheap, but for meaningful research in the field of structural physical chemistry there does not seem to be any choice but expensive instrumentation.It is difficult at this stage to see how the market for, and use of, very-high-resolution i.r. spectrometers will develop in the future, but it is possible that resolution up to 0.03 cm-1 in the laboratory will be the province of FTS instruments, that resolution beyond this will be the realm of the tunable diode laser, and that the spin-flip laser will be used in applications where monochromatic power is desirable, in field work and in double-resonance experiments for example. The determination of equilibrium data means in essence the determination of concentrations to high precision, and that implies the use of an instrument which gives good radiometric precision.FTS instruments are far superior to dispersive instruments in this respect, and the competition from Raman instruments is only minor, because of the narrow concentration range over which Raman instruments can be used. There is no simple equivalent in Raman spectroscopy to the variable path-length cell in i.r. spectroscopy and, of course, in the study of equilibrium systems one is not usually at liberty to change the concentration. The determination of the microstructure of solids is usually done by means of X-rays or neutron diffraction but there are some systems where these techniques give ambiguous results. This usually occurs where one is prevented from growing a suitably large single crystal, and prime examples are found in the area of macro- molecular studies.Crystalline polymers generally consist of spherulites connected by amorphous regions, and X-ray studies are therefore usually made on drawn fibres, so that there will be some degree of orientation. The interpretation of these fibre results is often not easy, and there always remains the doubt that possible changes of morphology might occur in going from the bulk material to the fibre. Also, X-ray observations at low temperature are notoriously difficult. Reser- vations must therefore be entertained about low-temperature crystal structures of polymers derived by X-ray methods. Far4.r. studies can be of enormous value here49 since the number of segments per unit cell and the symmetry of the unit cell can often be deduced from observations of the lattice spectrum in the wavenumber region 10-120 cm-l.Raman and neutron-diffraction observations have tended to be complementary rather than competitive, leaving FTS methods of unchal-lenged value. At higher wavenumbers, related effects are noticed (e.g.crystal-field splitting) but these are usually well within the compass of a good dispersive 47 C. K. N. Patel, Phys. Rev. Letters, 1972,28,649, R. A. Wood, R. B. Dennis, and J. W. Smith, Optics Comm.,1972,4, 383; R. L. Allwood, R. B. Dennis, W. J. Forth, S. D. Smith, B. S. Wherrett, and R. A. Wood, in ref. 43, p. 107, Radio and Electronic Engineer, 1972, 42,243. 4.3 See e.g. K. W.Nill, F. A. Blum, A. R. Calawa, and T. C. Harman, Appl. Phys. Letters, 1972, 21, 132, and W. H. Weber, P. D. Maker, K. F. Yeung, and C. W. Peters, Appl.Optics, 1974, 13, 1431. 48 G. W. Chantry, J. W. Fleming, E. A. Nicol, H. A. Willis, M. E. A. Cudby, and F. J. Boerio, Polymer, 1974, 15, 69. Chamberlain instrument, and FTS techniques, though desirable, are not essential. The topic of intermolecular interactions, and particularly those in liquids, has become a major research field of far4.r. spectroscopy in recent years.50 This work had its origin in the confirmation at N.P.L.of the existence of the characteristic liquid-phase absorption bands which had been predicted by Poley. Unlike all the topics so far discussed, effective investigation of liquid-phase interactions requires the experimental observation of both parts of the complex refractive index4.e.both the absorption coefficient and the refractive index. The development of asymmetric dispersive FT techniques has permitted the direct observation of both components and has therefore led to the extension of dielectric spectroscopy into the tens of gigahertz region. Much valuable chemical information is flowing from this experimental advance-one example being the elucidation of ‘chemical- relaxation processes’ by the group at Nancy.51 The detection and determination of constituents present at low concentration have always been important jobs for the i.r. spectroscopist but in recent years, with the growth of concern about the pollution of the environment, this field has become very important indeed.The successful achievement of the objectives involves a sampling problem, i.e. either a method for concentrating the trace constituents or else ensuring a sufficiently long path length, a sensitivity problem, and a radiometric accuracy problem. FTS methods solve the latter problems very well indeed, though for specific problems, i.e. monitoring a single pollutant, the superior resolving power of injection lasers can give them the edge. Nevertheless, for general surveys or for monitoring several gases, FTS methods reign supreme. At far-i.r. frequencies, FTS instruments have been used to monitor the ozone layer in the strato~phere~~ and to detect the presence of the nitrogen oxides which it is thought might affect it deleteriously.53 In the mid-i.r., FTS instruments have been used extensively to monitor atmospheric pollution, especially that due to the exhaust gases of motor cars.The study of adsorbed layers and the chemical effects at interfaces gains in motivation from the practical importance of heterogeneous catalysis, The layers involved may be only a few tens of nanometres thick and there are severe problems in getting sufficient absorption for a reasonable quality spectrum. ATR techniques can magnify the absorption, but even so the multiple-reflection forms of ATR are essential, and these require a quasi-parallel beam, with the consequent severe loss of energy throughput. The loss in energy throughput can be com- pensated for by going over to FTS techniques, and this is almost universal practice in modern research.A variant on absorption spectroscopy, namely emission spectroscopy, is also possible with FTS instruments used to study chemisorpt ion. When one considers the study of dynamic systems it might be thought that 50 G. W. Chantry, ‘Submillimetre Spectroscopy’, Academic Press, London, 1971, pp. 157-162 and pp. 177-187. 51 J. Goulon, J. L. Rivail, J. Chamberlain, and G. W. Chantry, in ‘Molecular Motions in Liquids’, ed. J. Lascombe, Reidel, Dordrecht, Holland, 1974, p. 163. 52 J. E. Harries, N. R. W. Swann, J. E. Beckman, and P. A. R. Ade, Nature, 1972, 236, 159. 53 J. E. Harries, Nature, 1973, 241, 515. The Chemical Applications of Advances in Fourier Transform Spectrometry FTS operates under a severe handicap in that it produces a spectrum which is the time average of the physical (and in general time-varying) spectrum.Thus whereas with grating instruments one can set the instrument to observe only a very narrow band of wavelengths, this is possible only in exceptional cases with FTS instruments. FTS techniques are of value as mentioned earlier if the vari- ation of spectrum occurs in a time longer than the rapid-scan time, if the pheno- mena can be reproducibly repeated many times, or if there is a way of ‘freezing-in’ the transient phenomena so that static methods may be used to study them. Standard ‘rapid-scan’ instruments can cover a length of interferogram equivalent to aresolution of 4 cm-1 in under a second, and special modifications are available that permit this time to be reduced to 1/80th of a second for an equivalent resolution of 16 cm-l.These times are sufficiently short for most gas-phase chromatography to be monitored by FTS ‘on-the-fly’; that is, in real time and without recourse to trapping. The extra dimension given to gas chromatography by the addition of an i.r. spectrometer is opening up new possibilities in diagnostic chromatography. The study of fast chemical reactions by time-resolved FTS, on the other hand, has not so far made very much progress beyond the early work of Murphy and Sakai,41 unlike the equivalent situation in physics, where the study of rapidly evolving pulsed plasmas is progressing at a rapid rate.The difference undoubtedly arises from the more severe problems which chemists experience in getting their reactants into the flash zone and the products out of it, but in the future this may well be a significant growth area in experimental kinetics. The ‘freezing-in’ technique most commonly used is matrix isolation, in which free radicals or other transients formed at a high temperature are quickly trapped in a solid argon matrix at cryogenic temperature. The spectroscopic problems arise because (i) the radicals have to be at low concentration to prevent recombination and (ii) scattering of radiation by the matrix leads to heavy energy loss. Once again, the energy-throughput advantage of FTScomes to the aid of the spectro- scopist and permits the observation of good-quality vibrational spectra of transients. In summary, therefore, the availability of FTS systems for the entire i.r. region has given chemists powerful new ways of investigating chemical systems, and the keen rivalry between FTS and other instrumental techniques will ensure that the chemist is well served with an ever-increasing range of instruments to bring to bear on his problems.

ISSN:0306-0012    

DOI:10.1039/CS9750400569

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Prostaglandins -Tomorrow’s Drugs By E. W. Horton DEPARTMENT OF PHARMACOLOGY, UNIVERSITY OF EDINBURGH, 1 GEORGE SQUARE, EDINBURGH EH8 9JZ The novel structure of the prostaglandin molecule has in recent years attracted the attention of many synthetic organic chemists. Difficulties inherent in synthe- sizing compounds which present so many stereochemical problems offer a challenge to ingenious inventiveness, whilst the possibility of synthesizing new compounds with more selective biological activity than the natural prostaglan- dins opens up wide areas of practical application in both human and veterinary medicine. The semi-systematic nomenclature1 of the natural compounds is based upon the hydrocarbon prostane (1) and the corresponding monocarboxylic acid prostanoic acid (2).The stereochemistry at positions 8 and 12 should be noted. The first two prostaglandins to be isolated from sheep vesicular glands were named prostaglandin E and prostaglandin F by Bergstrom and Sjovall in 1960.2These compounds were later designated prostaglandin El (PGE1, 3) and prostaglandin Fla (PGFla, 4) when analogues with additional double bonds were isolated, namely PGEz (5) and PGE3 (6).3The subscript numerals refer in each case to the number of double bonds in the side-chains. At an early stage it was shown that the PGE molecule could be converted chemically into a variety of related prostaglandins, most of which have now been found to occur naturally (Figure l).Thus reduction with sodium borohydride gives rise to the epimers Faand Fg.Under mildly acidic conditions the molecule is dehydrated to PGA or under more rigorous alkaline conditions to the isomer of PGA, viz. PGB. Naturally occurring prostaglandins comprise a number of families, eachdesignated by a different letter (from A to H) indicating a difference in the nature of the five-membered ring. These differences may be seen in Figures 1 and 2. The two cyclic endoperoxide prostaglandins, PGGz and PGHz,~ have hydroperoxy- and hydroxy-substituents, respectively, at the 1Sposition. Virtually all animal tissues contain one or other prostaglandin.5~6 It seems likely that in most instances prostaglandins are not stored as They are N. A. Nelson, J, Medicin. Chem., 1974, 17, 91 1. * S. Bergstrom and J.Sjovall, Acta Chem. Scand., 1959, 14, 1693, 1701. S. Bergstrom, R. Ryhage, B. Samuelsson, and J. Sjovall, Arkiv. Kemi, 1962, 19, 563. M. Hamberg and B. Samuelsson, Proc. Nut. Acad. Sci., U.S.A., 1973, 70, 899. S. Bergstrom, Science, 1967, 157, 382. E. W. Horton, ‘Prostaglandins’, Monographs on Endocrinology, Vol. 7. Springer-Verlag.Heidelberg, 1972. B. N. Davies, E. W. Horton, and P. G. Withington, Brit. J. Pharmacol., 1968, 32, 127. P. Piper and J. R. Vane, Ann. New York Acad. Sci., 1971, 180, 363. 589 Prostaglandins -Tomorrow’s Drugs 9 7 5 3 1 lo:o 12 I1 13 15 17 19 PROSTANE PROSTANOIC ACID ill (21 ----COOH . COOH\OH OH OH dH PROSTAGLANDIN E, PROSTAGLANDIN F, oi (3) (4)1”lHOH OH OH OH PG Ez PG E3 (5) (61 OH NaBHL PGE , OH OH PG Fp PG A PG B Figure 1Derivatives of prostaglandin E 590 Horton 0 OH \ , pcr, 90 -# 0 CYCLIC ENDOPEROXIDE PGC PGD Figure 2 Other prostaglandin families very readily formed from precursor unsaturated fatty acids in response to a variety of stimuli.In contrast, a Caribbean coral, Plexauru homomallu, contains vast quantities of a prostaglandin A, no less than 1.5% of its dry eight.^ This has been an extremely important source of starting material for the synthesis of prostaglandins for clinical use. Formation of the l0,ll-epoxide of PGAz (7) in turn leads to the production of the hydroxycyclopentanone PGEz (8) or its 1 1-epimer. A method for the stereospecific production of PGEz (98 %) has been devised by Corey,lO who protected the 15-hydroxy-group by forming the trixylylsilyl derivative.This allowed the preferential formation of the a-epoxide. 0 0 0--:fa-a'I bH (71 (8) The discovery of pharmacologically active agents of animal origin has in the past been a profitable starting point for the development of new drugs of clinical importance. For example, the discovery of histamine led to the development of the antihistamines for use in the treatment of allergies and, more recently, to the Hz-receptor blocking drugs that are of potential importance m the treatment of peptic ulcers. The discovery of adrenaline led in its turn to the development of more selective sympathomimetic agents, e.g. isoprenaline and salbutamol, which are of importance in the treatment of asthma, and to the synthesis of specific antagonists which block the action of adrenaline at one or other of its receptors. Thus the beta-blockers are currently of great importance in the treatment of various cardiovascular disorders.Likewise, studies on the bio- synthesis of adrenaline and noradrenaline and the recognition of an intermediary, dopamine, as a substance of importance in its own right in the central nervous system, have led to the use of I-dopa in the treatment of Parkinson's disease and a-methyl-dopa for alleviating hypertension. Moreover, inhibition of mono-amine oxidase, one of the enzymes which inactivate the catecholamines, has also found clinical usefulness.A. J. Weinheimer and R. L. Spraggins, Tetrahedron Letters, 1969, 5185. lo E. J. Corey and H. E. Ensley, J. Org. Chem., 1973,38, 3187. Prostaglandins -Tomorrow’s Drugs Various points of attack can therefore be delineated, as illustrated in the very simple diagram of Figure 3. By inhibition of biosynthesis, the formation, or excessive formation, of the natural product could be prevented. A second point of attack is inhibition of the inactivation process, leading to an accumulation of the biologically active natural product or possibly of one of its active metabolites. The third and most obvious line of attack would be to make novel compounds which act directly on the tissue receptors, these compounds being more selec- Prrcusors TISSUE Figure 3 Possible sites of chemicul attack on a biological system.(1) Inhibition of biosynthesis (or release)(2) Inhibition of metabolism (3) More selective action on tissue receptors (4) Blockade of receptors by spec$c antagonists tive than the naturally occurring compound. This approach has been particu- larly fruitful in the steroid field. Finally, by synthesizing antagonists, the actions of the endogenous compound could be blocked. Again, there could well be selective antagonists which would block some, but not all, of the actions of the natural product or products at the receptor site. How far have these approaches been applied in the prostaglandin field? What do we know of the biosynthesis of these compounds? In 1964 the enzymatic conversion of arachidonic acid into prostaglandins E2 and Fza was described.llJ2 The microsomal enzyme complex concerned was called prosta- glandin synthetase.Free arachidonic acid itself is not found in large quantities within cells. It must be released from membrane phospholipids by the action of phospholipase A2 (alternative sources are cholesterol esters). It is now known that there are at least three products of prostaglandin synthetase, i.e. PGE, PGF, and PGD (Figure4). These prostaglandins are formed from a common precursor and are not precursors one of the other. l1 S. Bergstrom, H. Danielsson, and B. Samuelsson, Eiochim. Eiophys. Am, 1964, 90, 207. la D. A. van Dorp, R. K. Beerthius, D. H. Nugteren, and H. Vonkeman, Biochim.Biophys. Ada, 1964, 90, 204. Horton COOH ARA::IDDONIC ENDOPEROXIDE OH 0 Figure 4 Products of prostaglandin synthetase Various inhibitors of prostaglandin synthetase have been discovered. Van Dorp and his colleagues found that a number of unnatural analogues of the essential fatty acids would inhibit the conversion of arachidonic acid into pro~tag1andins.l~Likewise, the acetylenic analogue of arachidonic acid (eicosa- tetraynoic acid) is a powerful inhibitor of prostaglandin synthetase.14 Both these groups of compounds are thought to occupy the arachidonic acid site on the enzyme.15 The most important advance resulted from the discovery in 1971, by J. R. Vane, that prostaglandin biosynthesis is blocked by aspirin and pharma- cologically related compounds.16 These drugs are not competitive inhibitors and it would seem unlikely from their chemical structure (Figure 5) that they act directly on the substrate receptor site. These compounds have now been used extensively for the investigation of the role of prostaglandins in both physiological and pathological conditions.Moreover, the discovery of their mode of action has provided a rational explanation for the way in which aspirin and other non-steroidal anti-inflammatory drugs produce their beneficial effects in man, reducing pain, inflammation, and fever.l5J7 The great chemical diversity of this group opens up the possibility of discovering more selective compounds, for j3 D. A. van Dorp, Ann. New York Acad.Sci., 1971, 180, 181. l4 D. T. Downing, D. G. Ahern, and M. Bachta, Biochent. Bioghys. Res. Comrn., 1970, 40, 218. j5 R. J. Flower, Pharmacol. Rev., 1974, 26, 1. l0 J. R. Vane, Nature New Biol.,1971, 231, 232. l7 S. Ferreira, and J. R. Vane, in ‘The Prostaglandins’, ed. P. W. Ramwell, Plenum Press, New York, 1974, Vol. 2, p. 1. Prostaglandins-Tomorrow’s Drugs COONa COOH0”” OoCoMe Sodium Acetylsalicylic salicylate acid (aspirin) OH C=O0 CI Paracetamol Phe nyl bu t azone Indomet hacin (butazolidin) Figure 5 Non-steroid anti-inflammatory drugs which inhibit prostaglandin synthetase example compounds which will block the formation of one kind of prostaglandin and not that of another. Some progress has been made towards this goal through the development of the bicycloheptenes, which block the formation of prosta- glandin E but not of prostaglandin F.l8 Interestingly enough, phenylbutazone, a commonly used antirheumatic drug, appears to block PGE and PGF formation, but not PGD.la Further work in this interesting area is clearly needed.Numerous metabolic pathways for the prostaglandins have been described (Figure 6).20Basically these can be divided into four groups, (a) oxidation of the 15-hydroxy-group by prostaglandin dehydrogenase, (b) reduction of the 13,14- trans-double bond by the 13,14reductase, (c) o-oxidation, and (d)p-oxidation of the a-side-chain and in some cases also of the w-side-chain, once the dicarboxylic acid has been formed.There has been only moderate success so far in synthe- sizing specific inhibitors for these various metabolic enzymes.21 The most specific enzymes for prostaglandin metabolism are the 15-hydroxy-prostaglandin dehydrogenase and the 13,14-reductase. Inhibition of these could lead to in- teresting biological results. It is, however, more difficult to predict any clinical usefulness for such compounds. Analogues of prostaglandins have been synthesized which are resistant to attack by various metabolizing enzymes. For example, 3-oxo-PGFla (9) is P. Wlodawer, B. Samuelsson, S. M. Albonico, and E. J. Corey, J. Amer. Chem. Soc., 1971, 93, 2815. R. J. Flower, in ‘Prostaglandin Synthetase Inhibitors’, ed. H. J. Robinson and J. R. Vane, Raven Press, New York, 1974, p. 9.B. Samuelsson, Ann. New York Acad. Sci., 1971, 180, 138. M. A. Marrazzi, and N. H. Andersen, in ‘The Prostaglandins’, ed. P. W. Ramwell, Plenum Press, New York, 1974, Vol. 2, p. 99. Horton 15-hydroxyprostaglandin d e hydrogenase ..*WCOOH --*-COOH H0’ ‘0H HO’ 0 PGE2 n-oxidation 0GcOoH<IN-oxidat ion =OH H0’ 0 HO’ 0 Major urinary metabolite I5-ket 0-dihydro -PGE2 Figure 6 Formation of the main human urinary metabolite of PGE, OH OH OH 15-methyl -PGE,3-0x0 -PG F,, (101 (9I 0 COOH Me Me OH OH 16.16-d1methyl-PG E, (II) resistant to /3-oxidation.22 Modifications which prevent the action of 15-hydroxy-prostaglandin dehydrogenase have proved more profitable. Synthetic 15-methyl (10) and 16,16’-dimethyl analogues (1 1) are unsusceptible to oxidation at the 15-position.2a G. Bundy, F. Lincoln, N. Nelson, J. Pike, and W. Schneider, Ann. New York Acad. Sci., 1971, 180, 76. 595 ,- Prostaglundins -Tomorrow’s Drugs The first clinical use of a prostaglandin was to induce part~rition.~3 In many hospitals in the United Kingdom, PGEz taken orally is now used routinely in childbirth. This has many advantages over the less convenient oxytocin drip. Prostaglandins in larger doses also induce abortion.24 Initially this was achievde by intravenous infusion of PGFza or PGE2, but now the preferred method is to make a single injection of the prostaglandin directly into the uterus, either intra- or extra-amniotically. 15-Methyl-PGFza has proved very effective for this purpose, though it has not yet been cleared for routine clinical use.In the middle 1960’s Ramwell and Shaw discovered that prostaglandins can inhibit gastric secretion in the rat.25 Simultaneously, Robert at the Upjohn Company carried out experiments on dogs with stomach pouches, and showed that both PGEl and PGEz administered intravenously are powerful inhibitors of the secretion of gastric juice that is secreted in response to histamine, penta- gastrin, or food.86 Unfortunately, neither of these prostaglandins is effective by mouth in man.27 On the other hand, 16,16’-dirnethyLPGEz (11) is a powerful inhibitor of gastric secretion when given by mouth to normal human volunteers or to patients with peptic ulcer.2* A single dose of 150 pg is sufficient to inhibit gastric secretion for several hours.There is now some evidence that the healing of gastric ulcers is enhanced by the administration of these compounds, and this opens the prospect of a new drug treatment for peptic ulceration. One of the earliest observations made with the newly isolated prostaglandins El and E2 was that they relax respiratory muscle. This observation was made by Dr. Main at Miles Laboratories in England in 1964.29 He showed that tracheal muscle is relaxed by prostaglandins of the E series and that a bronchodilator effect can also be demonstrated in vivo. Such results have now been extended to man, and it has been shown that aerosols of prostaglandins of the E series are about as effective as isoprenaline in causing bronchodilatation in human volunteer^.^^ The prospect of using prostaglandin or prostaglandin analogues with more selective action on respiratory smooth muscle raises the possibility of a new treatment for bronchial asthma.A further interesting observation is that asthmatic patients are unusually sensitive to prostaglandin F2a, itself a broncho- constrictor like histamine. If PGF2a is implicated in asthmatic attacks, blockade of this compound at the receptor site by a specific antagonist could offer an alternative approach to asthma therapy. In 1966, Kloeze, of the Unilever Company in Holland, made the remarkable observation that blood platelets in vitro can be prevented from aggregating by the as S.M. M. Karim, R. R. Trussell, R. C. Patch, and K. Hillier, Brit. Med. J., 1968, 4, 621. S. M. M. Karim and G. M. Filshie, Lancet, 1970, 1, 157. P. W. Ramwell and E. Shaw, in ‘Prostaglandin Symposium of the Worcester Foundation for EFperimental Biology’, ed. P. W. Ramwell and J. E. Shaw, Interscience, New York, 1968, p. 55. *t~A. Robert, in ‘Prostaglandin Symposium of the Worcester Foundation for Experimental Biology’, ed. P. W. Ramwell and J. E. Shaw, Interscience, New York, 1968, p. 47. *’ E. W. Horton, I. H. M. Main, C. J. Thompson, and P. M. Wright, Gut, 1968, 9, 655. S. M. M. Karim, D. C. Carter, D. Bhana, and P. A. Ganesan, Prostaglandins, 1973, 4, 71 I. H. M. Main, Brit. J. Pharmacol., 1964, 22, 51 1. so M. F. Cuthbert, Brit.Med. J., 1969, 4, 723. Horton presence of very small quantities of PGE1.31 It has since been observed that prostaglandin El can also inhibit platelet aggregation induced by damage to blood-vessel walls by the application of a laser beam in vivo. Moreover, PGDz is also a powerful inhibitor of platelet aggregati~n.~~ In contrast, PGEz potentiates the aggregation. Cyclic endoperoxides like PGGz and PGHz (see Figure 7) are even more powerful aggregators of platelets than PGEz, and may be essential for aggregation to take pla~e.3~ Inhibition of the formation of these endoperoxides by drugs like aspirin will prevent platelet aggregation and may be useful in the prophylaxis of thrombosis. Rapid developments in this exciting field can be anticipated, with important implications for cardiovascular medicine.In 1968, Pharriss, at the Upjohn Company, made the important observation that prostaglandin Fza is luteolytic in the rat.34 Similar luteolytic effects with this FATTY ACID CYCLO-OXYGENASE1 ENDOPEROXIDE PG H, @@-COOH OH 6OH PEROXIDASE/ PG E2 Figure 7 Mode of formation of cyclic endoperoxides PGGz and PGH, from arachidonic acid s1 J. Kloeze, in ‘Nobel Symposium 2: Prostaglandins’, ed. S. Bergstrom and B. Samuelsson, Almqvist and Wiksell, Stockholm, 1967, p. 241. 32 E. E. Nishizawa, W. L. Miller, R. B. Gorman, G. L. Bundy, J. Svensson, and M. Hamberg,Prostaglandins,1975, 9, 109. sa B. Samuelsson and M. Hamberg, in ‘Prostaglandin Synthetase Inhibitors’ ed.H.J. Robin-son and J. R. Vane, Raven Press, New York, 1974, p. 107. 33 B. B. Pharriss and L. Wyngarden, Proc. SOC. Exp. Biol. Med., 1969, 130, 92. Prostaglandins-Tomorrow’s Drugs compound have been observed in other non-primate species, and it has now been established that in the guinea-pig, sheep, pig, and cow prostaglandins almost certainly have a physiological role as luteolytic substances.35 They are synthesized by the uterus, and, acting locally, cause degeneration of the ipsilateral corpora lutea. Thus prostaglandin formation becomes maximal towards the end of the oestrous cycle in the guinea-pig, and this accounts for the loss of luteal function, the fall in progesterone secretion, and the initiation of a new cycle. Support for this finding has been obtained by the use of drugs which block prostaglandin formation and by immunizing guinea-pigs specifically against PGFza.In both cases, normal cycling is interfered with. Since luteolysis is of such fundamental importance in reproduction, it is not surprising that attempts have been made to produce prostaglandin analogues with more selective luteolytic action. There has been considerable success in this field, notably by the synthesis of various compounds by the Pharmaceutical Division of Imperial Chemical Industries. An example is I.C.I. 81008 (12), which is a highly selective luteolytic substance.36 This compound and others are now finding practical use in the veterinary field.37 The synchronization of oestrus in cattle is of considerable commercial importance I.C.I.81008 112) where artificial insemination is extensively used in large herds. This can now be achieved by prostaglandin treatment. Similarly, mares can be brought into oestrus on a selected day, a practice now common in the breeding of race- horses. A further use may be the induction of parturition in pigs to ensure that farrowing occurs not haphazardly but at a pre-selected time. All these new procedures involving the use of prostaglandins or their analogues have con- siderable economic implications. The development of a specific prostaglandin antagonist is still in the early stages. There is some evidence that poly(phloretin phosphate) (13a) [where (13b) is the phloretin molecule] antagonizes the actions of some prostaglandins in some biological systems.38 A much larger chemical and pharmacological effort should be devoted to this important area. E.W. Horton and N. L. Poyser, Physiol. Rev., 1975, in preparation. 36 M. Dukes, W. Russell, and A. L. Walpole, Nature, 1974, 250, 330. 37 Proceedings of the Upjohn Veterinary Symposium (1975) ‘The Use of Prostaglandins in Veterinary Practice’, Upjohn, Crawley. 36 K. E. Eakins and J. H. Sanner, in ‘The Prostaglandins’, ed. S. M. M. Karim, Medical and Technical Publications, Oxford, 1972, p. 262. Horton 0 0 II II I -0-R-O-P-O-R-O-P-O- I OH OH (130 1 R = phlorctin OH 0 Considering then the various points of attack outlined in Figure 3, what has been achieved and what remains to be done? Inhibitors of biosynthesis have been found and many of these, of course, were already in use before it was realised that they acted in this way, for example aspirin.Virtually no inhibitors of metabolism are of practical use at the present time, though the possibility that certain diuretics act in this way is still being investigated. Considerable pro- gress has been made in the production of more selectively active prostaglandins, notably I.C.I. 81008 and 16,16’-dimethyl-prostaglandinE2, but little progress has been made in the attempt to find antagonists that are specific for the pros- taglandin receptors. It is clear that we require more selective rather than more active compounds for practical use in Medicine.Work with the naturally occurring prostaglandins gives some hope that it may be possible, by manipulating the molecule, to produce compounds with such selectivity. Such developments in turn may lead to the discovery of antagonists which I believe would be of considerable academic interest and clinical importance. But here too, selectivity of action would be highly desirable. There is some evidence from the work of Dr. R. L. Jones in my laboratory for the existence of more than one kind of prostaglandin recept0r.3~ Antagonists which would block one but not the other, by analogy with the alpha- and beta-blocking drugs for adrenaline, would be powerful pharmacological tools. Biologically, there are still many underdeveloped areas. For example, the role of prostaglandins in relation to the central nervous system is still hardly under- stood, although it is known that prostaglandins occur there, and that they have 39 R.L.Jones, Brit.J. Pharmacol., 1975, 53,464P. Prostaglandins-Tomorrow’s Drugs some remarkable effects.40 For example, prostaglandin El inhibits leptazol convulsions in mice and prostaglandin F2a is a potentiator of spinal reflexes. At sympathetic nerve endings prostaglandin E2 inhibits noradrenaline release, and this may well prove to be a physiological mechanism;41 interference with this action or mimicking it by the action of prostaglandin analogues might well have therapeutic implications in the treatment of such diseases as hypertension. The role of prostaglandins in human semen and its importance in fertility is still not understood.There is some evidence that in some infertile males there is a lower than normal content of prostaglandin E. The recent discovery by Taylor and Kelly42 of the 19-hydroxyprostaglandin E compounds (14) in such high 0 OH OH OH (14) concentration may well lead to further developments in this field. It is perhaps speculative to suggest that such studies could eventually lead to the development of a male ‘pill’. Work on the oviduct, in which prostaglandins E have an in- hibitory acti0n,~3 is still at a very early stage. Inhibition of the various metabolic pathways, in particular, in the synthesis of arachidonic acid from dihomoy- linolenic acid and the interconversion of the primary prostaglandin44 could have interesting therapeutic possibilities.It will be apparent that the large number of natural prostaglandins that I have described in this review presents us with a field comparable in complexity to that of the steroids. Their equally diverse pharmacological activity is thought by some people to rule out the possibility of using prostaglandins in man, on the argument that a substance which does everything can have no usefulness. I do not share this pessimism. It is clear that there are considerable differences in selectivity of action amongst the natural prostaglandins. I believe, and this has already been amply demonstrated by I.C.I. 81008, that organic chemists can capitalize on this, and can design molecules not necessarily with more activity but with a selective action on, say, bronchial muscle for the asthmatic, gastric mucosa for the man with peptic ulcer, and the platelets or blood vessels for that vast legion of people who succumb daily to thrombosis and high blood pressure. ‘O E. W. Horton, Physiol. Rev.,1969, 49, 122. 41 E. W. Horton, Brit. Med. Bull., 1973, 29, 148. 48 P. L. Taylor and R. W. Kelly, Nature, 1974, 250, 665. 43 E. W. Horton and I. H. M. Main, Brit. J. Pharmacol., 1963, 21, 182. 44 C. N. Hensby, Biochim. Biophys. Acta, 1974, 348, 145.

ISSN:0306-0012    

DOI:10.1039/CS9750400589

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

INDEXES Volume 4, 1975 The indexes in this issue cover Volumes 14 (Figures in bold type refer to the volume number) Index INDEX OF AUTHORS Ahluwalia, J. C., 2, 203 Allen, N. S., 4, 533 Baker A. D., 1,355 Beattie, I. R., 4, 107 Bell. R. P.. 3. 513 Bentley, P.' H., 2, 29 Berkoff, C. E., 3, 273 Bird, C. W., 3, 309 Blandamer, M. J., 4, 55 Braterman, P. S., 2, 271 Breslow, R., 1, 553 Brown, K. S., jun., 4,263 Brundle, C. R., 1, 355 Buchanan, G. L., 3,41 Burdett, J. K., 3, 293 Burgess, J., 4, 55 Burrows, H. D., 3, 139 Cadogan, J. I. G., 3, 87 Carabine, M. D., 1, 411 Cardin, D. J., 2, 99 Carless, H. A. J., 1, 465 Cetinkaya, B., 2,99 Chamberlain, J., 4, 569 Chatt, J., 1, 121 Chivers, T., 2, 233 Collins, C.J., 4, 251 Corfield, G. C., 1, 523 Cornforth, J. W., 2, 1 Cotton, F. A., 4, 27 Coulson, E. H., 1,495 Coyle, J. D., 1, 465; 3, 329; 4, 523 Cramer, R. D., 3,273 Cross, R. J., 2, 271 Dack, M. R. J., 4, 211 Dainton, F. S., 4, 323 Doyle, M. J., 2,99 Drummond, I., 2,233Evans, D. A., 2,75 Fenby, D. V., 3,193 Ferguson, L. N., 4, 289 Fry, A., 1, 163 Green, C. L., 2, 75 Greenwood, N. N., 3, 23 1 Griffiths, J., 1, 481 Grossert, J. S., 1, 1 Groves, J. K., 1,73 Guilford, H., 2, 249 Gutteridge, N. J. A., 1,381 Haines, R. J., 4, 155 Hall, G. G., 2, 21 Hall, L. D., 4, 401 Halliwell, H. I?., 3, 373 Harmony, M.D., 1,211 Hartley, F. R., 2, 163 Hartshorn, S. R. 3, 167 Henderson, J.W., 2,397 Hepler, L. G., 3, 193 Horton, E. W., 4, 589 Hudson, M. F., 4, 363 Isbell, H. S., 3, 1 Jamieson, A. M., 2, 325 Johnson, A. W., 4, 1 Jotham, R. W., 2,457 Kemp, T. J., 3, 139 Kennedy, J. F., 2, 355 Kennewell, P. D., 4, 189 Kenny, A. W., 4,90 Kresge, A. J., 2, 475 Lappert, M. F., 2, 99 Leigh, G. J., 1, 121; 4, 155 Leznoff, C. C., 3,65 Lindoy, L. F., 4, 121 Linford, R. G., 1, 445 Lipscomb, W. N., 1, 319 Lynch, J. M., 3, 309 McKellar, J. F., 4, 533 McKervey, M. A., 3,479 Mackie, R. K., 3, 87 Maitland, G. C., 2, 181 Maret, A. R., 2, 325 Mason, R., 1,431 Mayo, B. C., 2,49 Meadowcroft, A. E., 4, 99 Menger, H. W., 2, 415 Midgley, D., 4, 549 Moore, H. W., 2, 415 Mulheirn, L. F., 1, 259 Munn, A., 4, 87 Newman, J.F., 4, 77 North, A. M., 1, 49 Page, M. I., 2, 295 Pletcher, D., 4, 471 Poliakoff, M., 3, 293 Ramm, P. J., 1, 259 Rattee, I. D., 1, 145 Redl, G., 3, 273 Rouvray, D. H., 3, 355 Sarma, T. S., 2, 203 Satchell, D. P. N., 4,231 Satchell, R. S., 4, 231 Simpson, T. J., 4, 497 Smith, E. B., 2, 181 Smith, K., 3,443 Smith, K. M., 4, 363 Stacey, M., 2, 145 Suckling, C. J., 3, 387 Suckling, K. E., 3, 387 Sutherland, R. G., 1,241 Sutton, D., 4, 443 Taylor, J. B., 4, 189 Thomas, T. W., 1,99 Thompson, M., 1, 355 Toennies, J. P., 3, 407 Tolman, C. A., 1, 337 Twitchett, H. J., 3, 209 Underhill, A. E., 1, 99 Walker, I. C., 3, 467 Waltz, W. L., 1, 241 Ward, I.M., 3, 231 White, A. J., 3, 17 Whitfield, R. C., 1, 27 Index INDEX OF TITLES Acylation by ketens and isocyanates, a mechanistic comparison 4, 231 Acylation, Friedel-Crafts, of alkenes, 1,73 Adamantane rearrangements, 3, 379 Affinity Chromatography, chemical aspects of, 2,249 Alkali-metal complexes in aqueoussolution, 4,549 Alkenes, the Friedel-Crafts acylation Of? 1,73 Aphids and scale insects, their chem- istry, 4,263 Application of electrochemical tech- niques to the study of homogenous chemical reactions, 4,471 Aqueous mixtures, kinetics of reac-tions in, 4, 55 Aryldiazonium cations, co-ordination chemistry of, 4,443 Atmosphere, interactions in, of drop- lets and gases, 1,411 Azidoquinones and related com-pounds, chemistry of, 2,415 Azobenzene and its derivatives, photo- chemistry of, 1,481 Bile pigments, 4, 363 Biomimetic chemistry, 1,553Biosynthesis of sterols, 1,259 Biosynthetic studies, carbon-1 3 nuclear magnetic resonance in, 4,497 Bredt’s rule, 3, 41 Brransted relation -recent develop-ments, 2,475 Calorimetric investigations of hydro-gen bond and charge transfer complexes, 3, 193 Cancer and chemicals, 4,289 Carbohydrate-protein complexes, gly- coproteins, and proteoglycans, of human tissues, chemical aspectsof, 2,355 Carbon-1 3 nuclear magnetic resonance in biosynthetic studies, 4,497 Carbonium ions, carbanions, and radialcs, chirality in, 2,397 Carbonyl compounds, photochemistry of, 1,465 Catalysis, homogenous, and organo- metallic chemistry, the 16 and 18 electron rule in, 1,337 -of the olefk metathesis reaction, 4, 155 CENTENARY LECTURE.Biomimetic chemistry, 1,553CENTENARY LECTURE. Quadruple bonds and other multiple metal to metal bonds, 4, 27 CENTENARYLECTURE. Rot at ionally and vibrationally inelastic scattering of molecules, 3,407CENTENARYLECTURE. Three-dimen- sional structures and chemical mech- anisms of enzymes, 1, 319 Charge transfer and hydrogen bond complexes, calorimetric investi-gations of, 3, 193 Chemical applications of advances in Fourier transform spectroscopy, 4, 569 -aspects of affinity chroma-tography, 2,249 --of glycoproteins, proteo- glycans, and carbohydrate-protein complexes of human tissues, 2, 355 Chemicals in rodent control, 1, 381 Chemistry-a to~ological subject, 2,457 -of aphids and scale insects, 4, 263 of azidoquinones and related compounds, 2,415 of dyeing, 1, 145 of homonuclear sulphur species, 2, 233 of the production of organicisocyanates, 3,209of transition-metal carbene com- plexes and their role as reaction intermediates, 2, 99 Chirality in carbonium ions, car-banions, and radicals, 2,397Chromatography, affinity, chemical aspects of, 2, 249 Cis-and trans-effects of ligands, 2, 163 Complexes, alkali-metal, in aqueous solution, 4,549Conformation of rings and neigh-bouring group effects, development of Haworth’s concepts of, 3, 1 Conformational studies on small mole- cules, 1,293Contribution of ion-pairing to ‘mem- ory effects’, 4,251 Co-ordination chemistry of aryl-diazonium cations: aryldiazenato(arylazo) complexes of transition metals, and the aryldiazenato-nitrosyl analogy, 4,443 Cyclopolymerization, 1, 523 Dielectric relaxation in polymer solu-tions, 1,49 Difluoroamino-radical, gas-phasekinetics of, 3, 17 Droplets and gases, interactions in the atmosphere of, 1,411 Drug design, quantitative, 3, 273 Dyeing, chemistry of, 1,145 Echinoderms, natural products from, 1, 1 Education, chemical, a reassessment of research in, 1, 27 Electrochemical techniques, appli-cation of to study of homogenous chemical reactions, 4,471 Electron as a chemical entity, 4, 323 Electron scattering spectroscopy,threshold, 3,467 -spectroscopy, 1,355 Electrophilic aromatic substitions, non-conventional, and related reactions, 3, 167 Elimination reactions, isotope effect studies of, 1, 163 Energetics of neighbouring groupparticipation, 2,295 Enumeration methods for isomers,3,355 Environmental protection in the dis- tribution of hazardous chemicals 4, 99 Enzymes in organic synthesis, 3, 387 -, the logic of working with, 2, 1 -, three-dimensional structures and chemical mechanisms of, 1, 319 Experimental studies on the structure of aqueous solutions of hydro-phobic solutes, 2,203 FARADAYLECTURE.The electron as a chemical entity, 4,323Fixation, of nitrogen, 1,121 Forces between simple molecules,2, 181 Formation of hydrocarbons by micro- organisms, 3, 309 Fourier transform spectroscopy, chem- ical applications of advances in, 4, 569 Index Friedel-Crafts acylation of alkenes,1, 73 Gas-phase kinetics of the difluoro- amino-radical, 3, 17 Gases and droplets, interactions in the atmosphere of, 1,411 Glycoproteins, proteoglycans, and carbohydrate-protein complexes of human tissues, chemical aspects of, 2,355Growth of computational quantum chemistry from 1950 to 1971, 2, 21 Handling toxic chemicals-environ-mental considerations, 4, 77 HAWORTH LECTURE. The MEMORIAL consequences of some projectsinitiated by Sir Norman Haworth, 2,145HAWORTH LECTURE.The MEMORIAL Haworth-Hudson controversy and the development of Haworth’s con- cepts of ring conformation and of neighbouring group effects, 3, 1 Health hazards to workers from indus- trial chemicals, 4, 82 Homogenous catalysis, and organo- metallic chemistry, the 16 and 18 electron rule in, 1,337Homogenous chemical react ions, application of electrochemical tech- niques to the study of, 4,471 Hydrocarbon formation by micro-organisms, 3,309Hydrogen bond and charge transfer complexes, calorimetric investiga- tions of, 3,193 Hydrogen isotope effects, kinetic, recent advances in the study of, 3,513Hydrophobic solutes, experimentalstudies on the structure of aqueous solutions of, 2,203 Importance of solvent internal pressure and cohesion to solution phenomena 4,211 Infrared and Raman vibrational spectroscopy in inorganic chemistry, 4,107Insect attractants of natural origin, 2, 75 Interactions in the atmosphere of droplets and gases, 1,411 Index Interactions,met al-met al, in transition- metal complexes containing infinite chains of metal atoms, 1,99 Introducing a new agriculturalchemical, 4, 77 Ion-pairing, contribution to ‘memory effects’, 4,251Isocyanates, and ketens, a mechanistic comparison of acylation by, 4, 231 -, organic, chemistry of the pro- duction of, 3,209 Isomer enumeration methods, 3, 355 Isotope effect studies of elimination reactions, 1, 163 Ketens and isocyanates, a mechanistic comparison of acylation by, 4, 231 Kinetics, gas-phase, of the difluoro- amino-radical, 3, 17 -of reactions in aqueous mixtures, 4, 55 Lanthanide shift reagents in nuclear magnetic resonance spectroscopy,2, 49 Laser light scattering, quasielastic, 2,325 Lasers, tunable, 3,293 Ligands, cis-and trans-effects of, 2, 163 LIVERSIDGE Recent advances LECTURE.in the study of kinetic hydrogen isotope effects, 3,513 Macrocyclic ligands, synthetic, trans- ition-metal complexes of, 4, 421 Mechanisms, chemical, and three-dimensional structures of enzymes, 1,319 MELDOLA LECTURE.ChemicalMEDAL aspects of glycoproteins, proteo- glycans, and carbohydrate-proteincomplexes of human tissues, 2, 355 Metalloboranes and metal-metal bonding, 3,231 Metal-metal bonding and metallo-boranes, 3,231 -bonds, multiple (especiallyquadruple), 4, 27 -interactions in transition-metal complexes containing infinite chains of metal atoms, 1, 99 Natural products from echinoderms, 1, 1 Neighbouring-group effects and ring conformation, development of Haworth’s concepts of, 3, 1 -participation, energetics of, 2, 295 Nitrogen fixation, 1,121Non-conventional electrophilic aro-matic substitutions and related reactions, 3, 167 Nuclear magnetic resonance, carbon- 13, in biosynthetic studies, 4 ,49 -spectroscopy, lanthanide shift reagents in, 2, 49 -: spin-lattice relaxation, 4, 401 NYHOLM MEMORIAL LECTURE. For- ward from Nyholm’s Marchon Lecture, 3,373 Oelfin metathesis and its catalysis, 4, 155 Olefinic compounds, photochemistry of, 3, 329 Organoboranes as reagents for organic synthesis, preparation of, 3, 443 Organometallic chemistry and homo- genous catalysis, the 16 and 18 electron rule in, 1,337Organo-transition-metal complexes: stability, reactivity, and orbital correlations, 2,271 PEDLER LECTURE.Porphyrins and related ring systems, 4, 1 Phase boundaries, reactivity of organicmolecules at, 1,229Phosphorous compounds, tervalent, in organic synthesis, 3, 87 Photochemistry of azobenzene and its derivatives, 1,481 -of carbonyl compounds, 1, 465 -of olefinic compounds, 3, 329 -of organic sulphur compounds 4, 523 -of the uranyl ion, 3, 139 -of transition-metal co-ordination compounds-a survey, 1,241Photodegradation and stabilization of commercial polyolefins, 4, 533 Polymer solutions, dielectric relax- ation in, 1,49 supports, insoluble, use in organic chemical synthesis, 3, 65 Polyolefins, commercial, photodegra- dation and stabilisation of, 4, 533 Porphyrins and related ring systems, 4, 1 605 Preparation of organoboranes: re-agents for organic synthesis, 3, 443 Prostaglandins, tomorrow's drugs,4,589Prostanoids, total synthesis of, 2, 29 Quadruple bonds and other multiple metal to metal bonds, 4, 27 Quantitative drug design, 3,273 Quantum chemistry, computational, growth of from 1950 to 1971, 2,21 -mechanical tunnelling in chem- istry, 1,211 Quasielastic laser light scattering,2, 325 Radioactive and toxic wastes: a comparison of their control and disposal, 4990 Raman and infrared vibrational spec- troscopy in inorganic chemistry,4,107Reactivity of organic molecules at phase boundaries, 1,229 Recent advances in the study of kinetic hydrogen isotope effects, 3,513Research in chemical education: a reassessment, 1,27ROBERT ROBINSON LECTURE, The logic of working with enzymes, 2, 1 Rodent control, chemicals in, 1, 381 Rotationally and vibrationallyinelastic scattering of molecules, 3,407 Scale insects and aphids, chemistry of, 4, 263 16 and 18 Electron rule in organo-metallic chemistry and homogenous catalysis, 1,337Small molecules, conformation studies on, 1,293 Solids, surface energy of, 1, 445 Solution phenomena, the importance of solvent internal pressure and cohesion, 4,211 Solvent internal pressure and cohesion, importance to solution phenomena, 4,211 Some recent developments in chem- istry teaching in schools, 1,495 Spectroscopy, electron, 1,355 -, Fourier transform, chemical applications of advances in, 4, 569 Index -, threshold electron scattering,3,467 Spin-lattice relaxation: A fourth di- mension for proton n.m.r.spectro- SCOPY, 4,401Stability, reactivity, and orbital cor- relations of organo-transition-metal complexes, 2,271 Sterols, biosynthesis of, 1,259 Structure of aqueous solutions of hydrophobic solutes, experimental studies on, 2,203 Sulphoximides, 4,189 Sulphur compounds, organic, photo- chemistry of, 4,523 -species, homonuclear, chemistry of, 2,233 Surface energy of solids, 1,445Syntheses, total, of prostanoids, 2, 29 Synthesis, organic, enzymes in, 3, 387 -, organic, preparation of organo-boranes as reagents for, 3, 443 , organic, tervalent phosphorus compounds in, 3, 87 -, organic, use of inorganic poly- mer supports in, 3,65 TATEAND LYLELECTURE.Spin-latticerelaxation: A fourth dimension for proton n.m.r.spectroscopy, 4,401Teaching of chemistry in schools,some recent developments in, 1,495 Tervalent phosphorus compounds in organic synthesis, 3, 87 Three-dimensional structures and chemical mechanisms of enzymes,1,319 Threshold electron scattering spectro-SCOPY, 3,467TILDEN LECTURE. Valence in tran-sit ion-metal complexes, 1,431 Topological subject-chemistry, 2,457 Transition-metal carbene complexes, chemistry and role as reaction intermediates, 2,99complexes, containing infinite chains of metal atoms, metal-metal interactions in, 1, 99 complexes of synthetic macro- cyclic ligands, 4,421 -complexes, valence in, 1, 431 -co-ordination compounds,photochemistry of, 1,241 Tunable lasers, 3,293 Index Uranyl ion, photochemistry of, 3, 139 Vibrational infrared and Raman Use of insoluble polymer supports in spectroscopy in inorganic chemistry, organic chemical synthesis, 3, 65 4, 107 Vibrationally and rotationallyValence in transition-metal complexes, inelastic scattering of molecules, 1,431 3, 407 6 07

ISSN:0306-0012    

DOI:10.1039/CS9750400601

出版商:RSC    

年代:1975

数据来源: RSC