Alkali‐Metal Mediation: Diversity of Applications in Main‐Group Organometallic Chemistry

Abstract Organolithium compounds have been at the forefront of synthetic chemistry for over a century, as they mediate the synthesis of myriads of compounds that are utilised worldwide in academic and industrial settings. For that reason, lithium has always been the most important alkali metal in organometallic chemistry. Today, that importance is being seriously challenged by sodium and potassium, as the alkali‐metal mediation of organic reactions in general has started branching off in several new directions. Recent examples covering main‐group homogeneous catalysis, stoichiometric organic synthesis, low‐valent main‐group metal chemistry, polymerization, and green chemistry are showcased in this Review. Since alkali‐metal compounds are often not the end products of these applications, their roles are rarely given top billing. Thus, this Review has been written to alert the community to this rising unifying phenomenon of “alkali‐metal mediation”.


Introduction
Pioneered by Wilhelm Schlenk and Joanna Holtz in 1917, [1] organolithium compounds have made ap henomenal contribution to the development of chemistry across the whole landscape of the periodic table.T hey are masters of ad iverse range of metal-mediated reactions,f or example, operating as bases in metallation reactions,asnucleophiles in addition reactions to unsaturated molecules,a sm etal-halide exchange agents in metathesis reactions with organic halides, as ligand-transfer agents in transmetallation reactions with main-group,t ransition-metal, and lanthanide/actinide compounds,o ra sc ross-coupling partners in Pd catalysis. [2] The mediation tag stems from the action of the organolithium intermediates in these reactions,which make possible onward reactivity towards the desired organic products or organometallic compounds of other metals.B yc ontrast, the other common alkali metals sodium and potassium have contributed to amuch lesser degree to this conventional alkali-metal mediation chemistry,although, interestingly,W anklyn started investigations on organosodium and organopotassium compounds almost half ac entury before organolithium reagents revolutionized organometallic chemistry.H owever,t he general lower stabilities and lower solubilities of these compounds complicated their application. [2m-p] Interestingly,h owever, new unconventional alkali-metalmediated applications are emerging where this organolithium dominance is often challenged. At present, it is acomplicated sporadic picture,asthesenew applications cut across different themes including main-group homogeneous catalysis,s toichiometric organic synthesis,l ow-valent main-group metal chemistry,p olymerization, and green chemistry.S ystematic studies spanning the whole of Group 1( Li-Cs) are still relatively rare,but there are some studies where alkali-metal effects are clearly at play,t hat is,t here are gradations of mediation efficiency depending on the identity of the alkali metal. Remarkably,i nahigh proportion of examples, potassium appears to be outperforming its smaller sibling lithium. What follows is ac ritical selection of snapshots, mostly from the past five years,that spotlight the importance and increasing prominence of this new, often inverted, alkali-metal mediation efficiency.I ti sw ritten with the intention of promoting alkali-metal mediation as au nifying concept of tremendous scope,w ith the aim of attracting other researchers, especially early career chemists looking for exciting new research lines,w ho can take the subject to even greater heights.

Alkali Metals in Homogeneous Catalysis
Since the beginning of the 21st century,there has been an increasing interest in common alkali metal (Li, Na, K) based catalysts and the reasons are clear. These include the low toxicity and costs compared to those of transition metals and lanthanides,a nd the availability everywhere on the Earths crust and in the water of the oceans.These factors make them attractive and sustainable elements for potential use in am ultitude of transformations even though their redox chemistry and the number of oxidation states are limited. A growing number of reports are appearing where the metals of the first main group are used in catalytic applications,f or example,hydroboration, [3] intermolecular and intramolecular hydroamination, [4] hydrophosphination/hydrophosphorylation, [5] hydrosilylation, [6] hydrogenation, [7] dehydrogenation (or dehydrocoupling), [8] and Brønsted base catalysed CÀC additions. [9] Forthe last application, first reports date back to the 1950s from Pine and Wunderlich, who described the addition of alkylbenzenes to styrenes with catalytic amounts Organolithium compounds have been at the forefront of synthetic chemistry for over acentury,asthey mediate the synthesis of myriads of compounds that are utilised worldwide in academic and industrial settings.F or that reason, lithium has always been the most important alkali metal in organometallic chemistry.T oday, that importance is being seriously challenged by sodium and potassium, as the alkalimetal mediation of organic reactions in general has started branching off in several new directions.Recent examples covering main-group homogeneous catalysis,s toichiometric organic synthesis,low-valent main-group metal chemistry,polymerization, and green chemistry are showcased in this Review.Since alkali-metal compounds are often not the end products of these applications,their roles are rarely given top billing. Thus,t his Review has been written to alert the community to this rising unifying phenomenon of "alkali-metal mediation".
of Na or K. However, because of the high reactivity of the pure metals,o nly low selectivities were obtained. [9t] Today,  there are many examples where an alkali metal catalytically  converts cheap feedstock compounds,s uch as aldehydes,  amides,arenes,nitriles,imines,orallyl derivatives,into more  value added compounds.T he catalysts show good selectivity  in general, and even asymmetric additions are reported. [9n,p-r] Scheme 1p resents ag eneral mechanism for this CÀCb ond formation. Thea lkali-metal base (AM À B) deprotonates the pronucleophile (R À H) to form the reactive nucleophile (AM À R) under liberation of the conjugate acid (B À H). The high electropositivity of the alkali metal endows the AMÀR bond with strong polarity and thus installs ah igh negative charge on the carbanion. Then ucleophile (AMÀR) can then react at the partial positively charged centre of the electrophile to induce the C À Cb ond formation. Ty pically,t he resulting intermediate is ap owerful base that can form the desired product by deprotonation of either the conjugate acid (BÀH; path A) or the pronucleophile (RÀH; path B). In general, the latter species will be deprotonated when its acidic hydrogen atom has al ower pK a value than that of the conjugate base. [9j] Since alkali-metal bases are among the strongest bases, pronucleophiles with pK a values higher than 35 can be converted. Schneider and co-workers reported the allylic C(sp 3 )ÀHb ond activation of alkenes and their addition to imines.They discovered that NaHMDS (HMDS = 1,1,1,3,3,3-hexamethyldisilazide) already showed outstanding activity (Scheme 2a)a nd selectivity at room temperature,w hile surprisingly other HMDS-based complexes with various metals (Li, Mg, Ca, Sr, Sn, Cu, Ag, Zr,C e, Eu, Gd) showed no activity or less activity and selectivity (e.g.with K; internal/ external product ratio = 23:44). Thes uperior mediation efficiency of Na over both Li and Kwas explained in general terms by "a favourable mix of electronegativity,formal charge and ionic radius" which is necessary for this reaction. Schneider and co-workers were also able to detect the Scheme 2. CÀCbond formation between a) an imine and an allyl compound catalysed by NaHMDS;b)styrene and an allyl compound catalysed by LDA;c )analdehyde and toluene by activation of the benzylic group through cation-p interaction;d )conversion of benzaldehyde into N-(trimethylsilyl)benzaldimine by NaHMDS. intermediate nucleophile 1 by 23 Na NMR spectroscopy (À5.3 ppm) and assigned it to be h 3 -coordinated. [9o] In 2018, the Guan group was able to isolate an intermediate nucleophile from an LDA( LDA = lithium diisopropylamide) catalysed allylic CÀHbond alkylation with styrenes (Scheme 2b). Compound 2 was prepared by deprotonation of 1,3-diphenylpropene with in situ generated LDA, thereby giving the first structure of a p-allylic Li compound. Thecationic lithium centre is coordinated by one THF molecule and two diphenylallyl anions through h 3 -coordinations,thereby resulting in apolymeric structure.Itisnoteworthy that the reaction of compound 2 with styrene results in the formation of styrene oligomers,which indicates that the diisopropylamine is crucial to prevent polymerization (Scheme 1, path A). Moreover,the larger DA congeners of Na and Kw ere found to perform worse than their smaller sibling lithium, with yields of only 9% and 11 %, respectively. [9i] Fort he functionalization of the aromatic compound toluene and its derivatives,amethod that is based on cation-p interactions is often applied. Here,t he smaller alkali metals are supported by their biggest (non-radioactive) sibling caesium. Whereas common bases such as MHMDS (M = Li, Na, K) are not able to deprotonate toluene because of insufficient basicity (pK a % 43 of toluene in DMSO; [10] pK a % 26 of HN(SiMe 3 ) 2 in THF [11] ), the addition of caesium-containing compounds (e.g. CsF,C sTFA[ TFA = trifluoroacetate]) makes the reaction feasible. [9c,e,j] Thel arge and soft caesium cation coordinates to the electron-rich p-system and slightly but significantly polarizes the aromatic system, thereby leading to amore acidic benzyl group.This reasoning was supported by acomputational DFT study by Cundari and co-workers,w ho calculated that the Cs + -centroid (toluene) distances decrease sequentially upon benzylic C À Hc leavage along the reaction coordinate from 3.63 to 3.42 to 3.24 (neutral!transition state! product;see Figure 1a). Moreover,they showed that decreasing the size of the metal is accompanied by ahigher energetic barrier and that the cleavage is exergonic for all alkali metals except Li (G [kcal mol À1 ]: Li = 3.9;N a = À0.3;K = À4.2; Rb = À4.3;C s = À5.5).
[9s] Theg radations in coordination behaviour of the alkali metals in benzylic systems was also shown by Robertson and co-workers.T hey reported an eyecatching series of Li, Na, and Kb enzyl complexes (Figure 1b), which are broken down to monomers by the neutral, tetradentate Me 6 -TREN ligand [Me 6 -TREN = (Me 2 NCH 2 CH 2 ) 3 N].Anice trend is discernible from their crystal structures,n amely that al arger metal coordinates more towards the p-system of the aromatic ring rather than to the benzylic position. [12] Theconcept of activation by cation-p interaction has also been exploited by Walsh and co-workers for the one-pot aminobenzylation of aldehydes with toluene (Scheme 2c). This study revealed that LiHMDS was unable to catalyse the reaction at all, whereas the corresponding Na (47 %) and K (35 %) amides showed moderate conversions.T his trend is also consistent with the calculations from Cundari et al. alluded to above.The addition of 0.35 equivalents of CsTFA, the activator of the benzyl group,i mproved the reactivity remarkably for Li (94 %) and Na (95 %), in contrast to the modest mediation of K(50 %; here KHMDS is less efficient in the formation of the aldimine [13] ). In this catalytic procedure the alkali-metal base has two tasks:1 )Itc onverts the aldehyde into aldimine through nucleophilic attack at the carbonyl carbon atom followed by an aza-Petersen olefination (Scheme 2d). 2) It deprotonates the toluene at the benzylic position to generate an ucleophile that is capable of attacking the previously formed aldimine.T his one-pot synthesis elaborates av ariety of amines to av ast range of toluene derivatives through CÀCaddition of abroad range of aldehydes.B earing in mind that the generated products can be used as convenient feedstocks for important bioactive building blocks,this simple one-step procedure becomes even more attractive considering the use of cheap and non-toxic alkali metals. [9l] Alkali metals do not only cooperate with each other, they show synergistic effects with metals from essentially the entire periodic table.F or example,i n2 019 Hevia and co-workers reported acatalytic intermolecular hydroamination mediated by alkali-metal magnesiates of the general forms (AM)MgR 3 and (AM) 2 MgR 4 (AM = Li, Na, K; R= CH 2 SiMe 3 ). [4a] Note that these complexes are referred to as lower order and higher order magnesiates,r espectively,r eflecting the relative stoichiometry of the Rs ubstituent. It is noteworthy that the catalytic activity of alkaline-earth metals in intermolecular hydroamination decreases as the ionic radius decreases;thus, Mg shows poor reactivity for this transformation. [14] However, this problem can be circumvented with the help of alkali metals,a sd epicted in Figure 2a.T he formation of an alkalimetal magnesiate has two benefits:1)enhanced reactivity of the nucleophile;and 2) substrate activation by polarization of the unsaturated bond by the alkali metal, which facilitates the nucleophilic attack and additionally brings the substrate in proximity to the Mg centre.T his was evidenced by the Hevia group,w ho showed that on its own the magnesium alkyl compound Mg(CH 2 SiMe 3 ) 2 (80 8 8C, 24 h) could not catalyse the reaction between diphenylacetylene and piperidine, whereas lower order alkali-metal magnesiates allow the conversion at 80 8 8Cw ithin 18 hours.T hereby,t he reactivity increased from Li (48 %) to K( 59 %) to Na (98 %) and the catalyst loading could be decreased to 2mol %. Am ore significant boost in reactivity can be observed on utilizing higher order alkali-metal magnesiates (AM) 2 Mg-(CH 2 SiMe 3 ) 4 ,w here the reaction already takes place at room temperature and is complete within 3hours [reactivity: K(! 99 %) > Na(28 %) > Li(0 %)].S aturation of the coordination sphere of Kb yt he addition of 18-crown-6 resulted in ac omplete shutdown of the reactivity.T his finding confirms the proposal that the alkali metal functions as aL ewis base and activates the unsaturated substrate by polarization. X-ray structure determination and DOSY spectroscopy of the intermediate compound 3 (formed by the stoichiometric reaction between 4equivalents of piperidine and [(TMEDA) 2 (Na) 2 Mg(CH 2 SiMe 3 ) 4 ]) additionally supports the hypothesis.A ss hown in Figure 2c,c ompound 3 displays ac ontacted ion-pair structure,w hich is crucial for effecting communication between the two metals.D OSY NMR measurements confirmed that this structure is present for lower order alkali-metal magnesiates in solution.
Cooperativity was also revealed between alkali metals and aluminium, as evidenced by one of the best-known bimetallic compounds,n amely lithium aluminium hydride, LiAlH 4 .Recently Harder and co-workers used it as acatalyst for the hydrogenation of imines. [7c] Our own group employed various derivatives of LiAlH 4 for catalytic hydrophosphination [5a] and hydroboration [3g-i] processes.F or the latter we carried out ac omparative study on the performance of neutral (monometallic) and anionic (bimetallic) aluminium complexes (depicted in Scheme 3), and found that the

The Rise of Organosodium Mediation in Stoichiometric Organic Transformations
With more and more chemists and funding agencies understandably placing greater emphasis on the sustainable future of our planet, research into terrestrially abundant, inexpensive,and non-toxic elements is gathering momentum. Sodium trumps lithium in this category,a sit is the most abundant alkali metal in the Earthscrust and the sixth most abundant element overall, whereas lithium is ar elatively scarce element. [15] Organosodium compounds have long been inferior to organolithium compounds in terms of effectiveness and convenience across the general landscape of synthetic applications,b ut with sustainability now an added important consideration, some researchers are taking on the challenge of developing novel new lines of organosodium research, which in time could stimulate amassive growth in the use of organosodium reagents in organic transformations.
Aparticularly exciting new line of organosodium research is in the area of transition-metal-catalysed cross-coupling.The mind-set behind this research is simple:r eplace the lessreactive lithium and magnesium by the more reactive sodium so that the organometallic compounds required for the transmetallation step could be prepared from the preferred halide starting materials,organic chlorides. [16] These chlorides are less expensive than organic bromides and organic iodides, but having strong CÀCl bonds they are more challenging to use with lithium and magnesium and generally require harsh conditions.The use of sodium in the form of afine dispersion led to reactions with awide range of aryl chlorides containing different substituents (e.g. alkyl, aryl, silyl, methoxy,o r dimethylamino groups) affording the desired arylsodium compounds under mild conditions in good yields with no signs (or traces) of the homocoupled biaryl side product, as reported by Asako,N akajima, and Takai. Even sterically crowded compounds such as tri-t-butylphenylsodium could be made by this method. Since these arylsodium compounds could transfer their aryl groups to zinc (via ZnCl 2 ·TMEDA) and boron (via MeOBpin) to form arylzinc and arylboron compounds,r espectively,p alladium-catalysed Negishi and Suzuki-Miyaura cross-coupling reactions could then be performed with as econd aryl chloride by using the Pd-PEPPSI-IPr catalyst (Scheme 5).
TheH an group is also taking advantage of the clean reactivity of finely dispersed sodium for the generation of sodium phosphides R 1 R 2 PNa. These compounds can be produced in high yields and short reaction times by treatment of the sodium dispersion with either phosphine halides (R 1 R 2 P-X;S cheme 6a)o rd irectly from at ertiary phosphine (R 1 R 2 R 3 P; R 1 ,R 2 ,R 3 = alkyl, aryl;S cheme 6b). When different substituents are attached to the phosphine,t he more electron deficient CÀPb ond is generally cleaved;f or example,aphenyl group is preferentially cleaved over at olyl group,o rabenzonitrile substituent is more likely to be cleaved off than ap henyl group.T his effective and useful method can be used for the preparation of unsymmetrical tertiary phosphines (see Scheme 6d,e). Both procedures with Na show higher yields compared to those employing K, and are superior to those with Li, since the quenching of byproducts (e.g.p henyllithium or excess Li)i sa voided. The sodium phosphides are capable of undergoing reactions with various halide substrates to form tertiary phosphines and, advantageously,they exhibit better conversions when smaller halides are attached, thus opening the door to the cheaper and readily accessible chloroarenes as well as for challenging CÀF bonds.T his method is straightforward for the synthesis of valuable multidentate arylphosphines,asshown in Scheme 6. Tr ichlorobenzene can be converted into the corresponding triphosphine in an effortless one-pot synthesis in quantitative yields (Scheme 6c). Thesubstitution with different Rattachments proceeds smoothly,whereby every sodium phosphide is

Angewandte Chemie
Reviews added successively after each substitution is finished without any interim purification steps (Scheme 6d). Whereas the synthesis of unsymmetrical tertiary phosphines is quite often at edious procedure,u sing this method for tailoring the desired phosphine is ac onvenient one-pot procedure (Scheme 6e). Starting with Ph 3 P, one phenyl group can be cleaved by treatment with finely dispersed Na to access Ph 2 PNa. Subsequent addition of n-BuBr gives Ph 2 (nBu)P.The addition of another equivalent of finely dispersed Na gives (Ph)(nBu)PNa, which ultimately is converted into the desired unsymmetrical tertiary phosphine (Ph)(nBu)(1-naphthyl)P in an overall yield of 84 %u pon introducing 1-naphthyl chloride. [17] Af undamentally important reaction long synonymous with organolithium chemistry is metallation. Fore xample, using metallation to functionalise aromatic and heteroaromatic substrates is of industrial importance,especially for the elaboration of pharmaceuticals and agrochemicals. [18] In metallation, an organolithium reagent mediates the selective functionalization of an organic substrate by converting an inert CÀHbond into aC ÀEbond via alithiated intermediate containing areactive CÀLi bond that can be intercepted by an electrophile "EX" (Scheme 7).
Lithium-alkyl compounds,lithium amides,ormixed-metal complexes containing Li and another alkali metal such as potassium (so called "superbases") are usually employed for metallation applications (note that mixed-metal superbases are not covered here,s ince they have been widely reviewed elsewhere). [19] To militate against competitive nucleophilic addition processes with unsaturated organic substrates,steri-cally encumbered lithium amides are generally preferred. Equipped with two branched alkyl arms,lithium diisopropylamide (LDA) is especially suited to this role and, consequently,has been one of the most commonly utilised reagents in organic chemistry for 50 years, [20] particularly in natural product syntheses. [21] Although sodium diisopropylamide (NaDA) was first prepared as long ago as 1959, only ad ecade after its lighter sibling, it has essentially remained unused. In 2017, Collum and co-workers summed it up by remarking that "sodium diisopropylamide has been used in about ad ozen studies overall, whereas lithium diisopropylamide is probably used thousands of times daily". [22] The success of LDAisprobably afactor in this unemployment, as is the generalisation that organolithium compounds tend to be orders of magnitude more stable than organosodium compounds,afact that would also extend to the metallated intermediates formed during metallation reactions.T he greater ionic character of sodium compared to that of lithium in organic environments is also afactor,since although it may enhance reactivity it can also lead to poor solubility in organic solvents or to decomposition of the solvent through its greater basicity.Bythe judicious choice of the solvent, namely using N,N-dimethylethylamine (DMEA), NaDAh as belatedly begun to find application. [23] Prepared by mixing as odium dispersion in toluene with isoprene and diisopropylamine in DMEA, [24] NaDAwas studied by Collum and co-workers in metallation reactions of arenes,e poxides,k etones,h ydrazones,dienes,alkyl halides,and vinyl halides.Surprisingly,in most cases NaDA/DMEA shows high reactivities and chemoselectivities on apar with those of the commonly utilised LDA/THF.Examples where NaDAmediation is even advantageous include the clean elimination of 1-bromooctane to give the alkene (Scheme 8a;L DA effects am ixture of elimination and substitution);t he faster metallation of hydrazone (Scheme 8b;t he reaction with LDAe xhibits ah igher axial selectivity,a lthough is significantly slower); the rapid ortho metallation of ac arbamate in contrast to the sluggish performance of LDA/THF (Scheme 8c); and the metallation of 3-trifluoromethylphenyl chloride to give the 2sodiated intermediate at À78 8 8C, in contrast to amixture of 2and 6-lithiated intermediates with LDA/THF (Scheme 8d).
Af ollow-up kinetic study on the NaDA-mediated metallation of an assortment of alkyl halides in THF/hexane or THF/DMEA revealed that 1-halooctanes (chloro,b romo,o r iodo) undergo smooth and efficient elimination of sodium halide as opposed to amixture of elimination and substitution as often observed with LDA. [25] Va rious mechanisms have been proposed, although all involve solvated variations of the NaDAm onomer.F or example,w ith 1-bromooctane,a nE 2like elimination pathway via atrisolvated monomer is implied Scheme 6. Synthesis of sodium phosphides R 1 R 2 PNa starting from a) R 1 R 2 P-X;b )R 1 R 2 R 3 P. c) One-pot synthesis of multidentate arylphosphines. d) One-pot synthesis of unsymmetrical, multidentate arylphosphines. e) One-pot sequential synthesis of an unsymmetrical, tertiary phosphine.

Scheme 7. General equation for the classicalt wo-step metallation and electrophilic interception process.
Angewandte Chemie Reviews (Scheme 9, top);whereas this switches with 1-chlorooctane to ad isolvated monomer through ac arbenoid mechanism (Scheme 9, bottom). On the negative side,N aDAm ediation failed with n-alkyl fluorides because of competitive metallation and subsequent decomposition of the THF solvent. NaDA-mediated deprotonations of 1,3-dimethoxybenzene and related methoxylated arenes also show exclusively monomer-based mechanisms with two or three coordinated THF ligands. [26] NaDAi nD MEA has also proved efficient for the sodiation of (hetero)arenes in continuous microflow reactors (Scheme 10). [27] Establishing the optimal conditions to be af low rate of 10 mL min À1 ,t he Knochel group found that complete sodiation of the test reagent, 1,3-dichlorobenzene, could be accomplished within 0.5 seconds at À20 8 8C; whereas under conventional batch conditions this sodiation would need to be conducted at À78 8 8Ct oa void decomposition processes.T he resulting 2,6-dichlorophenylsodium can be intercepted immediately in batch reactions with ar ange of electrophiles.
This flow sodiation is successful with an impressive variety of organic substrates including haloarenes and halo-(hetero)arenes,m any of which are highly sensitive and decompose when sodiation is attempted under conventional batch conditions.4-Fluorobenzonitrile is such an example,as usually the nitrile substituent would be attacked using conventional sodiation, but under continuous-flow conditions it is sodiated ortho to the Fsubstituent. Thesubsequent batch treatment of this sodium intermediate with an aldehyde, aketone,oradisulfide leads to the desired polyfunctionalised benzonitrile in high yield (Scheme 11).
As is the case with NaDA, an ew preparative redox method for making potassium diisopropylamide,KDA,that is free from lithium impurities has enhanced its synthetic mediation qualities. [28] Accessed from a3 :1:1:0.5 mixture of potassium, diisopropylamine,T MEDA, and isoprene,t his new form of KDA, used in situ as the solvate KDA-(TMEDA), [29] has also been utilised in commercial microflow reactors for the potassiation of arenes and (hetero)arenes between À78 8 8Ca nd 25 8 8Cw ith reaction times between 0.2 s  and 24 sand acombined flow rate of 10 mL min À1 .With these substrates,p otassium-hydrogene xchange occurs selectively on the aryl ring, and the potassiated intermediates can then undergo immediate electrophilic interception with, for example,a ldehydes,k etones,a lkyl and allylic halides,a nd disulfides,t op roduce the target functionalized (hetero)arenes in high yields.T he scope of these KDA(TMEDA)-mediated reactions is extended to the lateral potassiation of methylsubstituted (hetero)arenes to generate,i nt urn, benzylic potassium intermediates and methyl-functionalised heteroarenes,asillustrated with toluene in Scheme 12.
Renewed interest in organosodium chemistry and its comparison with organolithium chemistry have not been limited to the diisopropylamide ligand. Thesodium congener of another utility amide,2 ,2,6,6-tetramethylpiperide,TMP,is also being investigated. Since its preparation in 1999, [30] NaTMP has mainly been of interest in the development of synergistic bimetallic chemistry.H owever,T akai and coworkers have recently reported examples where NaTMP, made by sodium dispersion techniques to ensure no lithium contamination is present, proves superior in terms of reactivity and selectivity compared to its lighter sibling LiTMP in Brønsted base applications for organic synthesis. [31] This includes the stereoselective Wittig reaction, where deprotonation of the phosphonium salt by LiTMP or NaTMP in THF/ hexane and the subsequent reaction with 2-naphthaldehyde gave similar high yields of the alkene product (both 88 %) but markedly different E/Z ratios (57:43 or 7:93;S cheme 13 a).
NaTMP also outperformed LiTMP in the terminal-tointernal double-bond isomerization reactions of 1-dodecene (Scheme 13 b). Whereas,LiTMP failed to catalyse the isomerization reaction in hexane at ambient temperature,N aTMP transformed 1-dodecene into 2-dodecene as amixture of E/Z stereoisomers,t hus reflecting its higher Brønsted basicity over that of LiTMP.T his same study demonstrated that NaTMP could deprotonate the heteroarenes benzofuran, benzothiophene,a nd dibenzofuran under mild conditions to produce sodium intermediates in situ that were stable enough to be intercepted with assorted electrophiles.
As mentioned earlier, the stability of alkali-metal reagents usually decreases on descending down the group.Intriguingly, however, the opposite trend was observed by the Gessner group,whose research focuses on the synthesis of alkali-metal carbenoids. [32] In 2016 they reported as eries of alkali-metal carbenoids,w ith the sodium and potassium congeners being the first structurally characterised complexes of this com-pound class (carbenoids are usually reactive intermediates in several reactions,f or example,t he Simmons-Smith reaction, and incorporate ahalogen next to the carbene unit). Thus,the decomposition of alkali-metal carbenoids is driven by the formation and precipitation of the metal halide.Gessner and co-workers found that, whereas the lithium complex undergoes decomposition even at 0 8 8C, the heavier alkali metal sodium and potassium congeners were surprisingly stable up to 30 8 8C. This superior stability is credited to the diminished Lewis acidity of the heavier alkali metals combined with the greater polarity of the M-C interactions compared to those of lithium. Thel atter effect decreases the polarization of the carbenoid CÀCl bond and concomitantly impedes the elimination of MCl. X-ray diffraction studies (the molecular structure of the potassium carboenoid is depicted in Figure 3) indeed revealed asignificant weakening of the C À Cl bond, as evidenced by its elongation compared to the bond in the precursor (by 0.05 for Na, 0.03 for K). [32b]

Angewandte Chemie
Reviews ments,w hich suggest that the dimer is retained in aromatic solvents.D espite its low Al valency, aluminyl 10 is stable for several days at 300 Kinbenzene solution, but interestingly on raising the temperature by 30 (12), where the K centre is sequestered from the anionic moiety. [34] This release of the stabilising Kc ation modifies the reactivity of the anionic moiety,leading not to CÀHbond activation but rather to C À Cbond activation and ring-opening of benzene to form as even-membered AlC 6 H 6 metallacycle in [K(2.2.2-crypt)]-[(NON)AlC 6 H 6 ]( 13; Scheme 14). In contrast to the chargeseparated monomer 12,when the dianionic moiety is atetrakis(trimethylsilyl)butylene ligand, ac ontacted ion-pair monomer 14 is formed, where the potassium engages directly with the Al centre in the shortest K-Al length reported to date [3.4549(5) ], with the remainder of the potassium coordination sphere made up of p-arene interactions with two toluene molecules. [35] Elemental potassium can also be used as an alternative source to generate the related potassium aluminyl complex [KAl( Si NON)] 2 (15; Si NON = {O(SiMe 2 N Ar ) 2 } 2À ,A r = 2,6-iPr 2 C 6 H 3 )i naredox reaction with ( Si NON)AlI (Scheme 15) as established by the Coles group. [36] Potassium···p-arene interactions are also acentral feature of the dimeric structure of 14, with the two Kc entres acting as bridges between two aluminyl anions.

Angewandte Chemie
Reviews large size and "soft" bonding character of the Kc ation also enables intermolecular interactions with aneighbouring COT ligand (h 3 -) and Ar substituent (h 5 -), which propagates the structure into ahelical polymer (Scheme 15).
Noting that all literature anionic Al complexes have to date been electronically balanced by K + cations,H arder and co-workers added the potassium reagent KHMDS to the neutral low valent compound ( DIPP BDI)Al (BDI = b-diketiminate). [37] Surprisingly,t he expected nucleophilic addition reaction did not occur, but instead deprotonation took place at abackbone Me group in the DIPP BDI ligand to generate the dianionic bisamide [H 2 C = C(NAr)-C(H) = C(Me)-NAr] 17, with the by-product presumably HMDS(H). Crystallographic characterisation of 17 revealed adimeric arrangement akin to that of 10,i nw hich formally (Al) À anions are bridged by K + cations that once again exhibit strong K + ···p-arene interactions with the DIPP groups.The mediation of the reactivity of potassium is clearly evident in the reaction of 17 with benzene (Scheme 16 a). In the absence of the potassium reagent, ( DIPP BDI)Al is inert to benzene,b ut 17 does react and in ar emarkable way,i nducing twofold CÀHb ond activation of benzene with its para-phenylene unit trapped between the two Al centres,which each carry areleased hydride ion in the product 18.O ne K + ion sits over the phenylene bridge, engaging in interactions with the DIPP p-systems,w hile the second K + ion propagates the coordination polymer by bridging through K + ···BDI and K + ···H(Al) contacts.T his 17 to 18 transformation has been likened to the ring-templatecontrolled double metallation reactions of arenes found in inverse crown chemistry.
Thea uthors strongly advocated the alkali-metal mediation theme that is the focus of this Review,a nd stressed, on the basis of DFT calculations,that it is important to take into account the role of the counter cation K + ,a si ts presence or absence can markedly alter the outcomes of anionic Al reactions,which appear to have asynergistic mixed-metal (K/ Al) origin. Coles and co-workers witnessed related K/Al synergistic reactivity in the production of the ethenetetraolate [K{Al( Si NON)(O 2 C)}] 2 (20)onexposing carbon monoxide to the potassium aluminoxane [K{Al( Si NON)(O)}] 2 (19;Scheme 16 b). [38]

Polymerization
Another important research area where alkali-metal catalysts are frequently applied is in the polymerization of lactide.P olylactide is non-toxic, biodegradable,a nd biocompatible. It is used widely in fields such as agriculture,p ackaging,f ood, and medicine.C onsidering that alkali metals are non-toxic and found in the human body (especially Na and K), catalysts based on these innocuous metals are very attractive, since traces of the catalyst used can remain in the polymer;d epending on which metal is present, this can potentially lead to harmful effects.W ithin the past two decades several suitable complexes for the ringopening polymerization of lactide have been reported and some general trends have been observed, such as the reactivity of the catalysts increases on descending the first main group from Li to Na to K. [42][43][44][45][46][47][48][49][50][51][52][53] This is attributed mainly to the larger size of the metal facilitating coordination of the substrate.H owever,t he opposite trend is observed with regard to selectivity,asitdecreases from Li to Na to K. [50,51,54] Thelower Lewis acidity and the resulting weaker bond to the substrate can prevent the growth of long and defined polymeric chains.T oo vercome this drawback, ag eneral catalyst design has been developed, which is depicted in Figure 4a.T he idea is to sandwich the catalytically active centre between the two planes,thereby embedding the cation in ad efined space to boost the interaction of the monomer and the active end of the polymer chain. One side is confined by ab ulky ligand, which forms an ionic bond to the metal. Ty pically applied ligand classes incorporate oxygen [42-48, 53, 55-60] or nitrogen donors, [49,61] or ac ombination of both types. [50,51,54,[62][63][64] Catalysts with electron richer ligands showed higher catalytic activity; [62,63] whereas more steric

Angewandte Chemie
Reviews bulk around the metal had ad etrimental influence on the performance. [46] Theo ther side of the metalsc oordination sphere is usually restricted by acrown ether.This leads to the benefit that the complexes are broken down to monomers, which then results in ab oost in the reactivity as there are more active centres available.The size of the donor has to be chosen carefully,s ince at oo-crowded cation shows less catalytic activity. [42,47,48,62] Wu et al. synthesized as eries of complexes (depicted in Figure 4b)a nd the importance of aw ell-defined space around the metal was evaluated. All compounds were synthesized by simple deprotonation of the phenol with KHMDS or NaHMDS and subsequent addition of the corresponding crown ether.Incomplexes A [46] and B, [47] ax anthyl group is attached to the phenolate in the ortho position. Theplane can be bent up (A)ordown (B)atthe sp 3hybridized carbon atom, which results in af lexible and illdefined area around the metal. Substitution of this residue by an anthryl group (C) [48] gives amore rigid and confined space.
As ar esult, C shows the highest P m value of 0.94 (A = 0.86, B = 0.79;t he P m value describes the isotacticity in the polymer). Aseries of Na and Kcomplexes of C with different substituents on the phenolate (see Figure 4c)w ere tested in the ROP of lactide with benzyl alcohol (BnOH) as ac oinitiator;aselection of results are listed in Table 1. TheK complexes show an increase in activity from 23 to 22 to 21, which indicates that the electron-donating group accelerates the reaction (entries 1-3). Therefore,ahigher electron density on the phenoxy oxygen atom enhances reactivity. Thes ame trend was observed for the Na complexes (entries 5-8), although they are slightly less active (since the NaÀOb ond is less ionic than the KÀOb ond). Adding steric bulk to the ortho position (R 1 = t-Bu) decreased the reactivity for K( entry 4) and Na (entry 8) because the active centre becomes too crowded. Changing the solvent from toluene to THF resulted in ad ecrease in reactivity,s ince there is acompeting coordination of THF and lactide towards the K + cation (entry 9). Lowering the temperature to À70 8 8Cg ave am ore controlled reaction, and the best isoselectivity (P m = 0.94) was observed for complex 21 (entry 10). Additionally, the obtained molecular-weight distributions show that the polymerizations catalysed by 21-28 are living. [48] 3. Organolithium Chemistry:G reen Shoots of aNew Future Thec ontent of this Review could leave the impression that the glory days of organolithium chemistry have long passed. However,like atrue champion, it continues to adapt and diversify to meet the challenging demands of todays world. Remarkably,i nr ecent times,s ignificant progress has been made even with regard to the Achilles heel of organolithium chemistry,n amely its inability to operate in the presence of air and/or moisture.Acentury ago,t he aforementioned Wilhelm Schlenk designed his special inertatmosphere glassware to protect organolithium compounds from these nemeses,which, on the merest contact, will rapidly cause decomposition of "RLi" to more thermodynamically stable LiÀOb onded and covalent R-H products.T aking on the seemingly "impossible" challenge of finding aw ay to perform organolithium chemistry under aerobic hydrous conditions has been overlain with ad rive towards creating am ore sustainable "Green" chemistry world. [65] In this context, this means predominately replacing hazardous organic reaction solvents by safe,g reen, and bio-renewable reaction media that are not based on crude petroleum. Three leading authorities-Capriati, Garcia-Alvarez, and Heviahave recently surveyed this emerging field in their article "the future of polar organometallic chemistry written in bio-based solvents and water". [66] Organolithium compounds unintentionally featuring water as aligand have appeared infrequently in the literature, such as 2-mercaptobenzoxazolyl(TMEDA)lithium monohydrate,reported by Snaith, Wright, and co-workers in 1990. [67] Today,examples can be found where organolithium reactions are carried out intentionally in the presence of water. Hevia, García-lvarez, and co-workers reported the chemoselective alkylation of alkyl and aryl ketones by both RLi and RMgX (Grignard) polar reagents to generate tertiary alcohols (Scheme 17 a). [68] Water was present in the bulk DES solvent (DES = deep eutectic solvent, which are eutectic mixtures made from green, nature-inspired components that form hydrogen-bond networks). Remarkably,enhanced yields and superior selectivities were obtained under wet, aerobic conditions in the DES (a 1:2choline chloride/water mixture) compared to those using as tandard inert atmosphere procedure.
Nucleophilic addition is not the only organolithium reaction type to undergo this seismic change in its practice. TheCapriati group has successfully extended it to the ortholithiation of as eries of N,N-diisopropylbenzamides by employing ac yclopentyl methyl ether/DES (1:2 choline chloride/glycerol) mixture at room temperature under air. [69] Electrophilic quenching after just two seconds by benzaldehyde,DMF,and assorted halogenating,silylating,and sulfurylating agents readily intercepted the carbanion formed by subjecting the amide to t-butyllithium. Adding to the interest of these reactions,switching the organolithium reagent from tbutyllithium to n-butyllithium led to no ortho-lithiation taking place but instead resulted in nucleophilic acyl substitutions (S N Ac) that produced mainly ketones and to asmall extent tertiary alcohols from over-addition (Scheme 17 b).
In exploiting DESs for anionic olefin polymerization, the same group have established an unprecedented air and moisture-compatible method that utilises organolithium reagents as initiators to access av ariety of polystyrenes and polyvinylpyridines (Scheme 17 c). [70] Sonication significantly accelerated the reactions,which were run in aDES comprising choline chloride and glycerol in a1 :2 stoichiometry, thereby producing polymers in high yields and with low polydispersities.S ince organolithium-mediated reactions of this type can be strongly influenced by the volume and surface of the organic droplets formed by the glycerol-or waterinsoluble reagents present, these have been labelled as "onglycerol" or "on-water" reactions,h ence the use of sonication. [71]

Summary and Outlook
In this Review we have tried to paint ap icture of the increasing utilisation of organoalkali-metal compounds in mediating new chemistry for ad iverse range of applications. Although the alkali metals often seem to be essential for the success of the applications described, their role as more of asecondary supporting nature can sometimes be glossed over. This understated appreciation contrasts with the wide acclaim given to other species used to support emerging aspects of synthesis and catalysis,s uch as popular sterically demanding tunable ligands (for example,N HCs and b-diketiminates). What may be surprising to those familiar with organoalkalimetal chemistry is that lithium is far from the centrepiece of this picture,assodium and potassium are outperforming their  lighter sibling in an increasing number of cases.T his is ar elatively new phenomenon, although in the bimetallic world of inverse crown chemistry,p otassium and especially sodium have been in the foreground of advances in template metalation, as most recently highlighted in the regioselective tetrazincation of ferrocene mediated by sodium. [72] For brevity,w eh ave limited this Review to mediations taking place in main-group chemistry,but exciting developments are also being realised in alkali-metal-mediated transition-metal chemistry.F or reasons of sustainability,E arth-abundant metals such as iron are attracting lots of activity.The potential of alkali-metal mediation in this arena is emphatically demonstrated in the reduction of dinitrogen and its hydrogenation to ammonia by aw ell-defined potassium-iron complex, as reported by the Holland group.
[73a] More recently, the same group showed that am ixed sodium-iron complex can functionalise dinitrogen through coupling reactions with hydrocarbons.
[73b] Ar eview on this related theme is likely to follow in the future. It is hoped that the snapshots presented herein will help promote alkali-metal mediation as aconcept in its own right and prompt the recruitment of many more chemists to the area. Thecanvas it offers for future research lines is vast. The origins of alkali-metal mediation are still not well understood in general, although there have been some excellent rigorous studies on af ew specific systems.T od ate,t here has been apaucity of kinetic studies and theoretical calculations.Given the often complicated nature of the aggregated structures involved in alkali-metal chemistry,s uch in-depth studies are vital to uncover transition states and map reaction profiles. More studies are also necessary in which the whole alkalimetal series is considered. Rubidium and caesium have hardly been considered at all in alkali-metal mediation, yet their propensity for engaging in p-arene interactions suggests they could make as ignificant impact in applications,s uch as lowvalent metal chemistry and homogeneous catalysis.T he synergistic effects that underpin the special chemistry of the inverse crown bimetallic complexes have mainly involved sodium mediation, and the impression at present is that the larger coordination spheres of the heavier alkali metals would make them ideal partners for constructing useful architectures with metals such as magnesium, zinc, and aluminium. Finally,t he old master lithium is far from exhausted. In fact, the recent reports of organolithium reactions taking place under aerobic and hydrous conditions represent one of the most remarkable breakthroughs in chemistry in recent times. We wait with eager anticipation to see how much this green chemistry will develop in the future.