PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Inorg Chem. Author manuscript; available in PMC 2010 July 20.
Published in final edited form as:
PMCID: PMC2879077
NIHMSID: NIHMS127128

Synthesis, Structure and Reactivity of Two–Coordinate Mercury Alkyl Compounds with Sulfur Ligands: Relevance to Mercury Detoxification

Abstract

The susceptibility of two-coordinate mercury alkyl compounds of the type X–Hg–R (where X is a monodentate sulfur donor) towards protolytic cleavage has been investigated as part of ongoing efforts to obtain information relevant to understanding the mechanism of action of the organomercurial lyase, MerB. Specifically, the reactivity of the two-coordinate mercury alkyl compounds PhSHgR, [mimBut]HgR and {[HmimBut]HgR}+ (HmimBut = 2-mercapto-1-t-butylimidazole; R = Me, Et) towards PhSH was investigated, thereby demonstrating that the ability to cleave the Hg–C bond is very dependent on the nature of the system. For example, whereas the reaction of PhSHgMe with PhSH requires heating at 145 °C for several weeks to liberate CH4, the analogous reaction of PhSHgEt with PhSH leads to evolution of C2H6 over the course of 2 days at 100 °C. Furthermore, protolytic cleavage of the Hg–C bond by PhSH is promoted by HmimBut. For example, whereas the reaction of {[HmimBut]HgEt}+ with PhSH eliminates C2H6 at elevated temperatures, the protolytic cleavage occurs over a period of 2 days at room temperature in the presence of HmimBut. The ability of HmimBut to promote the protolytic cleavage is interpreted in terms of the formation of a higher coordinate species {[HmimBut]nHgR}+ that is more susceptible to Hg–C bond cleavage than is two-coordinate {[HmimBut]HgR}+. These observations support the notion that access to a species with a coordination number greater than two is essential for efficient activity of MerB.

Introduction

In view of the potent toxicity of organomercury compounds,1 Nature has developed a detoxification procedure that is achieved by the combined action of two enzymes, namely (i) organomercurial lyase (MerB), which causes protolytic cleavage of the otherwise inert Hg–C bond, and (ii) mercuric ion reductase (MerA), which reduces Hg(II) to less toxic elemental mercury, Hg(0).1,2 The active site of MerB features cysteine ligation3 and, to emulate this aspect, we have employed the [S3]–donor tris(2-mercapto-1-t-butylimidazolyl)hydroborato ligand, [TmBut], to provide insight into the mechanism of action of the enzyme.4 For example, we demonstrated that the Hg–C bonds of the tris(2-mercapto-1-t-butylimidazolyl)hydroborato mercury alkyl complexes [κ1–TmBut]HgR (R = Me, Et) are readily cleaved by a thiol (Scheme 1). The facility with which the Hg–C bonds are cleaved under mild conditions was proposed to be a consequence of the mercury center of two-coordinate [κ1–TmBut]HgR being able to access higher coordination numbers due to the multidentate nature of the [TmBut] ligand.4,5 Herein, we provide further evidence that supports this suggestion by describing the susceptibility of a series of linear two-coordinate mercury alkyl compounds with a common S–Hg–C coordination environment towards protolytic Hg–C bond cleavage by PhSH.6

Results and Discussion

X–ray diffraction studies on [κ1–TmBut]HgR indicate that only one of the sulfur donors of the [TmBut] ligand coordinates to the mercury in the solid state, such that the metal adopts a linear two–coordinate S–Hg–C coordination geometry.4 In view of the observation that two-coordinate mercury alkyl compounds of the type X–Hg–R are generally inert towards protolytic cleavage of the Hg–C bond,7,8 the high reactivity of [κ1–TmBut]HgR towards PhSH (as a simple mimic for a cysteine S–H group) was attributed to the ability of mercury to access non-linear κ2– or κ3–isomers in which the Hg–C bond is more susceptible to cleavage.4 In this regard, 1H NMR spectroscopic studies demonstrate that [κ1–TmBut]HgR is fluxional and that higher coordination numbers are accessible.4 To provide further evidence for the proposal that increased coordination number facilitates protolytic cleavage of Hg–C bonds, it was deemed appropriate to probe the reactivity of well-defined two–coordinate X–Hg–R complexes in which X is a strictly monodentate sulfur donor. Therefore, we report here the reactivity of a series of two–coordinate mercury alkyl compounds, namely PhSHgR, [mimBut]HgR and {[HmimBut]HgR}+ (HmimBut = 2-mercapto-1-t-butylimidazole), towards PhSH.

Reactivity of PhSHgR towards PhSH

As part of an investigation of thimerosal (sodium ethylmercury thiosalicylate), we have recently demonstrated that the phenylthiolate mercury alkyl complexes, PhSHgMe and PhSHgEt, possess two-coordinate linear geometries at mercury.9,10 As such, these complexes provide suitable reference points to evaluate the reactivity of the Hg–C bond in organomercury compounds with a two–coordinate linear S–Hg–C coordination geometry. Significantly, despite the similar coordination geometries (Table 1), [κ1–TmBut]HgR and PhSHgR (R = Me, Et) react very differently towards PhSH. Thus, whereas the Hg–C bonds of [κ1–TmBut]HgR are cleaved rapidly by PhSH at room temperature,4 the phenylthiolate complexes PhSHgR are inert under these conditions. At elevated temperatures, however, PhSH cleaves the Hg–C bond of PhSHgR (R = Me, Et) to give (PhS)2Hg11 and RH (Scheme 2).12 The greater reactivity of [κ1–TmBut]HgR towards PhSH is, therefore, consistent with the notion that cleavage of the Hg–C bond is promoted by access to species with coordination numbers greater than two.

Table 1
Mercury coordination environments in two-coordinate mercury alkyl complexes with sulfur donors.

While the Hg–C bonds of both PhSHgMe and PhSHgEt are cleaved by PhSH, the facility of these reactions differ considerably, with the Hg–Et bond being considerably more susceptible to cleavage than that of the Hg–Me bond. For example, whereas PhSH protolytically cleaves the Hg–Et bond of PhSHgEt over a period of 2 days at 100°C, the corresponding reaction of PhSHgMe proceeds only slowly at temperatures of ca. 145°C. The greater reactivity of the Hg–Et bond relative to the Hg–Me bond in this system is noteworthy because protonolysis of RHgI by HClO4 and H2SO4 exhibits the opposite trend, with the reactivity decreasing in the sequence Me > Et > Pri > But.13 On the other hand, the ease of cleaving the Hg–R bond of a series of asymmetric dialkyl mercury compounds, RHgR', by AcOH decreases in the irregular sequence Et > Pri > Me > But (for a common spectator R' group).14 The observation that the preference for cleaving Hg–Me and Hg–Et bonds in RHgX molecules may be switched by varying X and/or the acid is particularly noteworthy and it is evident that a detailed understanding of these effects is of considerable relevance to mercury detoxification.

Reactivity of [mimBut]HgR towards PhSH

While comparison of the reactivity of [κ1–TmBut]HgR and PhSHgR provides evidence that the additional sulfur donors of the [TmBut] ligand are responsible for facilitating cleavage of the Hg–C bond, a better comparison is to evaluate the reactivity of [κ1–TmBut]HgR relative to a two-coordinate mercury alkyl complex that features a monodentate sulfur ligand that more closely resembles the [κ1–TmBut] ligand than does phenylthiolate. Therefore, we sought mercury alkyl compounds that incorporate the 2-mercapto-1-t-butylimidazolyl ligand, namely [mimBut]HgR (R = Me, Et), which may be hypothetically regarded as being derived from [κ1–TmBut]HgR by dissociation of the neutral borane, HB(mimBut)2 (Scheme 3). Such complexes are conveniently obtained via the reaction of RHgCl with HmimBut in aqueous NaOH solution (Scheme 4).15 The molecular structure of [mimBut]HgEt has been determined by X-ray diffraction (Figure 1) and the coordination geometry at mercury is comparable to that of [κ1–TmBut]HgEt (Table 1).

Figure 1
Molecular structure of [mimBut]HgEt.

In contrast to the phenylthiolate derivatives, the 2-mercapto-1-t-butylimidazolyl complexes [mimBut]HgR react rapidly with PhSH at room temperature. However, instead of cleaving the Hg–C bond, PhSH cleaves the Hg–S bond to give PhSHgR (Scheme 4). The different reaction pathway is most likely a consequence of the presence of the sp2 nitrogen lone pair on the [mimBut] ligand, protonation of which provides an alternative mechanism than one involving direct reaction with the Hg–C bond (Scheme 4). Supporting this suggestion, the proposed {[HmimBut]HgR}+ (R = Me, Et) intermediates may be synthesized independently via addition of HmimBut to [RHg][BF4],16 as illustrated in Scheme 5. The molecular structure of the ethyl derivative {[HmimBut]HgEt}[BF4] has been determined by X-ray diffraction (Figure 2) and the Hg–S and Hg–C bond lengths are similar to those of the neutral counterpart, [mimBut]HgEt (Table 1).

Figure 2
Molecular structure of {[HmimBut]HgEt}[BF4] (only the cation is shown).

With respect to the reactivity of {[HmimBut]HgR}[BF4], it is significant that treatment of {[HmimBut]HgR}[BF4] with NaSPh generates PhSHgR (Scheme 5), thereby providing evidence for the second step of the proposed mechanism for the reaction of [mimBut]HgR with PhSH (Scheme 4).

Reactivity of {[HmimBut]HgEt}[BF4] towards PhSH

Since the HmimBut ligand of {[HmimBut]HgR}+ is not susceptible to protonation, {[HmimBut]HgR}+ cannot react with PhSH in a manner analogous to that of [mimBut]HgR and the site of reactivity switches from the Hg–S bond to the Hg–C bond. Thus, treatment of {[HmimBut]HgEt}+ with PhSH results in elimination of C2H6 (Scheme 6). The initial mercury product is postulated to be {[HmimBut]HgSPh}+, although this species has not been isolated due to the existence of a subsequent exchange equilibrium that results in the formation of, inter alia, {[HmimBut]2Hg}[BF4]2, the molecular structure of which is shown in Figure 3. In addition, (PhS)2Hg, the accompanying redistribution product, was identified by mass spectrometry (m/z = 421.1 {M + 1}+).11

Figure 3
Molecular structure of {[HmimBut]2Hg}[BF4]2 (only the cation is shown).

Protolytic Cleavage of PhSHgR and {[HmimBut]HgR}+ by PhSH is Promoted by HmimBut

The observation that both neutral and cationic two-coordinate compounds, i.e. PhSHgR and {[HmimBut]HgR}+, are less susceptible to protolytic cleavage of the Hg–C bond than is [TmBut]HgR provides a strong indication that the more facile cleavage reaction of [TmBut]HgR is a consequence of the ability of the mercury center to access a higher coordination number, an effect which has been attributed to an increase in the negative charge on the carbon atom.17 Further evidence to support the proposal that an increase in coordination number enhances protolytic cleavage of the Hg–C bond is provided by the observation that cleavage of the Hg–C bond of both PhSHgEt and {[HmimBut]HgEt}[BF4] by PhSH is promoted by addition of HmimBut. For example, a mixture of PhSHgEt and PhSH readily eliminates ethane at room temperature in the presence of HmimBut.18,19 Likewise, HmimBut promotes elimination of ethane from a mixture of {[HmimBut]HgEt}[BF4] and PhSH at room temperature (Scheme 7). Since HmimBut alone does not cleave the Hg–C bond under these conditions,20 the ability of HmimBut to promote protolytic cleavage may be rationalized by the generation of higher coordinate species that are more susceptible to Hg–C protolytic cleavage than their two-coordinate counterparts.21 Other studies also suggest that coordination of substrates to mercury enhances the susceptibility of protolytic cleavage of Hg–C bonds. For example, I catalyzes the protolytic cleavage of allyl mercury iodide,22 while the nature of the buffer (i.e. formate, acetate, phosphate) has been shown to have an effect on the rate of protolytic cleavage of an aryl–mercury bond.23

Support for the proposal that HmimBut is capable of coordinating to the mercury centers of PhSHgMe, PhSHgEt and {[HmimBut]HgMe}+ is provided by the observation that the 1H NMR spectroscopic signals for the mercury alkyl groups of these complexes shift in the presence of HmimBut. For example, 1H NMR spectra of PhSHgMe and PhSHgEt in the presence of variable concentrations of HmimBut are illustrated in Figures 4 and and55,24 thereby demonstrating that HmimBut binds rapidly, and reversibly, to the mercury centers of these two-coordinate compounds, such that the observed chemical shifts are a weighted average of two- and three-coordinate species.

Figure 4
1H NMR spectrum of PhSHgMe in the presence of increasing amounts of HmimBut (* = mesitylene internal reference).
Figure 5
1H NMR spectrum of PhSHgEt in the presence of increasing amounts of HmimBut (* = mesitylene internal reference).

In addition to 1H NMR spectroscopy, 199Hg NMR spectroscopy also provides evidence for coordination of HmimBut to PhSHgR. For example, the 199Hg NMR spectroscopic signal for PhSHgMe progressively shifts from −557 ppm25 to −540 ppm upon increasing the concentration of HmimBut. Likewise, the 199Hg NMR spectroscopic signal for PhSHgEt progressively shifts from −730 ppm25 to −680 ppm in the presence of HmimBut. O'Halloran has noted that 199Hg NMR chemical shifts of mercury thiolate complexes typically become deshielded as coordination number increases,21e and thus the observed variation of 199Hg chemical shifts are in accord with coordination of HmimBut to PhSHgR generating three-coordinate species that are in rapid equilibrium with the two-coordinate species in solution.

Further evidence for the existence of three-coordinate mercury species in solution is provided by mass spectrometric studies suggest that {[HmimBut]2HgSPh}+ (m/z = 623.3)} is a component of the product mixture resulting from the reaction of {[HmimBut]HgEt}[BF4] with PhSH in the presence of HmimBut, although ligand redistribution giving {[HmimBut]4Hg}[BF4]2 and (PhS)2Hg11 is facile (Scheme 7). Likewise, the corresponding reaction of {[HmimBut]HgEt}[BF4] with p-ButC6H4SH in the presence of HmimBut liberates ethane and yields {[HmimBut]4Hg}[BF4]2 and [p-ButC6H4S]2Hg.26 The molecular structures of {[HmimBut]4Hg}[BF4]2 (Figure 6) and [p-ButC6H4S]2Hg (Figure 7) have been detemined by X–ray diffraction.27 As expected, the Hg–S bond lengths in tetrahedral {[HmimBut]4Hg}2+ (2.54 Å average) are substantially longer than the corresponding values in two-coordinate {[HmimBut]2Hg}2+ (2.35 Å average). For comparison, these bond lengths are virtually identical to the mean values for two-coordinate (2.34 Å) and four-coordinate (2.55 Å) mercury thiolate compounds listed in the Cambridge Structural Database.28

Figure 6
Molecular structure of {[HmimBut]4Hg}[BF4]2 (only the cation is shown).
Figure 7
Molecular structure of [p-ButC6H4S]2Hg.

Conclusions

In summary, comparison of the reactivity of the two-coordinate mercury alkyl compounds [κ1–TmBut]HgR, PhSHgR, [mimBut]HgR and {[HmimBut]HgR}+ towards PhSH indicates that the susceptibility towards cleavage of the Hg–C bond is very dependent on the nature of the system. Thus, whereas the Hg–C bond of [κ1–TmBut]HgR is readily cleaved by PhSH at room temperature, the Hg–C bonds of PhSHgR and {[HmimBut]HgR}+ are inert under comparable conditions. On the other hand, [mimBut]HgR is reactive towards PhSH at room temperature, but it is the Hg–S bond that is preferentially cleaved to give PhSHgR. Although {[HmimBut]HgEt}+ does not react with PhSH at room temperature, addition of HmimBut promotes cleavage of the Hg–C bond, thereby supporting the notion that access to geometries with a coordination number greater than two is required for the efficient activity of MerB.

Experimental Section

General Considerations

All manipulations were performed using a combination of glovebox and Schlenk techniques under a nitrogen or argon atmosphere. Solvents were purified and degassed by standard procedures. Reactions monitored by NMR spectroscopy were prepared in an NMR tube equipped with a J. Young valve. NMR spectra were measured on Bruker 300 DRX, Bruker 400 DRX and Bruker Avance 500 DMX spectrometers. For solutions in organic solvents, 1H NMR spectra are reported in ppm relative to SiMe4 (δ = 0) and were referenced internally with respect to the protio solvent impurity (δ 7.16 for C6D5H, and 2.50 for d6–Me2SO).29 13C NMR spectra are reported in ppm relative to SiMe4 (δ = 0) and were referenced internally with respect to the solvent (δ 128.06 for C6D6).29 199Hg NMR chemical shifts are reported relative to HgMe2 (δ = 0) but in view of the toxicity of the latter compound, the spectra were referenced externally with respect to HgI2 (1 M in d6-DMSO, δ = −3106).30 Coupling constants are given in hertz. IR spectra were recorded as KBr pellets on a Nicolet Avatar DTGS spectrometer, and the data are reported in reciprocal centimeters. Mass spectra were obtained on a JMS-HX110/110 Double Focusing mass spectrometer using fast atom bombardment (FAB). HmimBut,31 PhSHgMe9 and PhSHgEt9 were obtained by the literature methods. HgI2 (Aldrich), MeHgCl (Aldrich), EtHgCl (Strem), PhSH (Aldrich), PhSNa (Fluka) and AgBF4 (Strem) were obtained commercially. CAUTION: All mercury compounds are toxic and appropriate safety precautions must be taken in handling these compounds.

Reactivity of PhSHgEt towards PhSH in the presence and absence of HmimBut

(a) A solution of PhSHgEt (25 mg, 0.074 mmol) in C6D6 (0.7 mL) was treated with PhSH (20 µL) and heated at 100°C for a period of 2 days. The solution was allowed to cool to room temperature, thereby depositing a white precipitate over a period of 3 days. The mother liquor was decanted and the white solid was washed with pentane (2 × 1 mL) and dried in vacuo to give (PhS)2Hg as a white solid (9 mg, 29%). (PhS)2Hg was identified by comparison of the 1H NMR spectrum of a solution in d6–DMSO with that of an authentic sample:11a 1H NMR (d6–DMSO) 7.06 [t, 2H, 3JH-H= 7 Hz (C6H5S)2Hg], 7.15 [t, 4H, 3JH-H= 7 Hz (C6H5S)2Hg], 7.36 [d, 3JH-H= 4 Hz, 4H of (C6H5S)2Hg]; 1H NMR (C6D6) 6.85 [m, 6H of (C6H5S)2Hg], 7.25 [m, 4H of (C6H5S)2Hg].

(b) A solution of PhSHgEt (40 mg, 0.12 mmol) in C6D6 (3 mL) was treated with PhSH (40 µL) and mesitylene (10 µL) as an internal standard. The resulting solution was divided equally into four NMR tubes, to which three were treated with HmimBut (2 mg, 0.013 mmol; 10 mg, 0.064 mmol; 20 mg, 0.13 mmol) and the reactions were monitored by 1H NMR spectroscopy. The formation of small quantities of C2H6 was observed immediately for all three samples which contained HmimBut, but complete elimination required a period of several days: the mixture containing 2 mg of HmimBut went to completion over a period of 12 days at room temperature, while the mixtures containing 10 and 20 mg of HmimBut were complete after 1 week. In contrast, the solution to which no HmimBut was added proceeded to only ca. 50% conversion over a period of 2 weeks at room temperature and complete elimination of C2H6 required heating for 1 day at 145°C.

Reactivity of PhSHgMe towards PhSH in the presence and absence of HmimBut

A solution of PhSHgMe (20 mg, 0.062 mmol) in C6D6 (1.5 mL) was treated with PhSH (20 µL) and mesitylene (20 µL) as an internal standard. The resulting solution was divided equally into two NMR tubes, to which one was treated with HmimBut (10 mg, 0.064 mmol). The reactions were heated at 145°C and monitored by 1H NMR spectroscopy. In the presence of HmimBut, elimination of methane was complete after two days, whereas in the absence of HmimBut, elimination of methane proceeded only to ca. 90% completion after three weeks.

Comparison of the reactivity of PhSHgMe and PhSHgEt towards PhSH

A solution of PhSHgMe (10 mg, 0.031 mmol) and PhSHgEt (10 mg, 0.030 mmol) in C6D6 (0.7 mL) was treated with PhSH (20 µL) and mesitylene (20 µL) as an internal standard. The reactions were heated at 145°C and were monitored by 1H NMR spectroscopy, thereby demonstrating the complete formation of C2H6 after heating the solution for 1 day. In contrast, liberation of CH4 required a period of three weeks.

Synthesis of [mimBut]HgMe

A solution of HmimBut (200 mg, 1.28 mmol) in aqueous NaOH (30 mL of 75 mM) was added to a suspension of MeHgCl (321 mg, 1.28 mmol) in water (20 mL) over 15 minutes resulting in the immediate formation of a white precipitate. The suspension was stirred for 3 hours, allowed to settle for 30 minutes, and filtered. The precipitate was dried in vacuo to give [mimBut]HgMe as a white powder (290 mg, 61 %). 1H NMR (C6D6) 0.42 [s, 3H, 2JHg-H = 176 Hz, {C3N2H2[C(CH3)3]S}HgMe], 1.45 [s, 9H, {C3N2H2[C(CH3)3]S}HgMe], 6.66 [br d, 1H, 3JH-H= 2 Hz, {C3N2H2[C(CH3)3]S}HgMe], 6.92 [br d, 1H, 3JH-H= 2Hz, {C3N2H2[C(CH3)3]S}HgMe]. 13C{1H} NMR (C6D6) 8.6 [1C, {C2N2H2[C(CH3)3]CS}HgCH3], 29.6 [3C, {C2N2H2[C(CH3)3]CS}HgMe], 55.8 [1C, {C2N2H2[C(CH3)3]CS}HgMe], 117.8 [1C, {C2N2H2[C(CH3)3]CS}HgMe], 125.9 [1C, {C2N2H2[C(CH3)3]CS}HgMe], 144.4 (tentative) [1C, {C2N2H2[C(CH3)3]CS}HgMe]. IR Data (KBr pellet, cm-1): 3172 (w), 3111 (w), 2970 (m), 2908 (m), 1679 (w), 1561 (w), 1511 (m), 1475 (w), 1468 (w), 1447 (w), 1438 (w), 1417 (s), 1404 (m), 1393 (m), 1367 (s), 1343 (vs), 1297 (m), 1251 (vs), 1230 (m), 1221 (m), 1180 (w), 1141 (w), 1123 (vs), 1043 (s), 1021 (m), 914 (w), 843 (w), 817 (w), 771 (m), 720 (s), 690 (vs), 632 (w). Mass spectrum: m/z = 373.1 {M+1}+.

Synthesis of [mimBut]HgEt

A solution of [HmimBut] (200 mg, 1.28 mmol) in aqueous NaOH (30 mL of 40 mM) was added to a suspension of EtHgCl (339 mg, 1.28 mmol) in water (20 mL) over 15 minutes resulting in the immediate formation of a white precipitate. The suspension was stirred for 16 hours, allowed to settle for 30 minutes, and filtered. The precipitate was dried in vacuo to give [mimBut]HgEt as a white powder (263 mg, 53%). Crystals of composition [mimBut]HgEt suitable for X-ray diffraction were obtained from CH3CN. 1H NMR (C6D6) 1.07 [t, 3H, 3JH-H = 8Hz, {C3N2H2[C(CH3)3]S}HgCH2CH3], 1.27 [q, 2H, 3JH-H = 8Hz, {C3N2H2[C(CH3)3]S}HgCH2CH3], 1.46 [s, 9H, {C3N2H2[C(CH3)3]S}HgEt], 6.66 [br d, 1H, 3JH-H= 2Hz, {C3N2H2[C(CH3)3]S}HgEt], 6.93 [br d, 1H, 3JH-H= 2Hz, {C3N2H2[C(CH3)3]S}HgEt]. 13C{1H} NMR (C6D6) 13.8 [1C, {C2N2H2[C(CH3)3]CS}HgCH2CH3], 25.7 [1C, {C2N2H2[C(CH3)3]CS}HgCH2CH3], 29.6 [3C, {C2N2H2[C(CH3)3]CS}HgEt], 55.8 [1C, {C2N2H2[C(CH3)3]CS}HgEt], 117.7 [1C, {C2N2H2[C(CH3)3]CS}HgEt], 126.0 [1C, {C2N2H2[C(CH3)3]CS}HgEt], 144.7 [1C, {C2N2H2[C(CH3)3]CS}HgEt]. IR Data (KBr pellet, cm-1): 3165 (w), 3103 (w), 2988 (w), 2970 (m), 2926 (w), 2864 (w), 1683 (w), 1566 (w), 1476 (w), 1445 (w), 1418 (m), 1406 (m), 1394 (m), 1368 (s), 1338 (vs), 1295 (m), 1248 (vs), 1232 (m), 1178 (m), 1141 (m), 1123 (vs), 1044 (s), 1022 (s), 966 (w), 952 (w), 913 (w), 845 (w), 800 (m), 722 (s), 692 (vs), 682 (s), 633 (w). Mass spectrum: m/z = 387.1 {M+1}+.

Reactivity of [mimBut]HgMe towards PhSH

A solution of [mimBut]HgMe (10 mg, 0.027 mmol) in C6D6 (0.7 mL) was treated with PhSH (10 µL) and mesitylene (10 µL) as an internal standard. The reaction was monitored by 1H NMR spectroscopy, thereby demonstrating the formation of PhSHgMe and HmimBut in quantitative yield over a period of 1.5 hours.

Reactivity of [mimBut]HgEt towards PhSH

A solution of [mimBut]HgEt (10 mg, 0.027 mmol) in C6D6 (0.7 mL) was treated with PhSH (10 µL) and mesitylene (10 µL) as an internal standard. The reaction was monitored by 1H NMR spectroscopy, thereby demonstrating the formation of PhSHgEt and HmimBut in quantitative yield over a period of 1.5 hours. Over the period of a day, PhSHgEt reacts further with excess PhSH to yield (PhS)2Hg (see above).

Synthesis of {[HmimBut]HgMe}[BF4]

A mixture of MeHgCl (750 mg, 2.99 mmol) and AgBF4 (582 mg, 2.99 mmol) was treated with CH2Cl2 (15 mL) resulting in the immediate deposition of a white precipitate. The suspension was stirred 3 hours, allowed to settle for 30 minutes and filtered into a solution of [HmimBut] (466 mg, 2.98 mmol) in CH2Cl2 (15 mL). The resulting solution was stirred for 1 hour and solvent removed in vacuo to give {[HmimBut]HgMe}[BF4] as a white powder (680 mg, 50 %). 1H NMR (C6D6) 0.48 [s, 3H, 2JHg-H = 194Hz, H{C3N2H2[C(CH3)3]S}HgMe], 1.24 [s, 9H, H{C3N2H2[C(CH3)3]S}HgMe], 6.15 [d, 1H, 3JH-H= 2Hz, H{C3N2H2[C(CH3)3]S}HgMe], 6.75 [d, 1H, 3JH-H= 2Hz, H{C3N2H2[C(CH3)3]S}HgMe], 12.07 [br, 1H, H{C3N2H2[C(CH3)3]S}HgMe]. 13C{1H} NMR (C6D6) 8.9 [1C, H{C2N2H2[C(CH3)3]CS}HgCH3], 28.6 [3C, H{C2N2H2[C(CH3)3]CS}HgMe], 58.8 [1C, H{C2N2H2[C(CH3)3]CS}HgMe], 118.4 [1C, H{C2N2H2[C(CH3)3]CS}HgMe], 119.8 [1C, H{C2N2H2[C(CH3)3]CS}HgMe], 147.6 (tentative) [1C, H{C2N2H2[C(CH3)3]CS}HgMe]. IR Data (KBr pellet, cm-1): 3191 (m), 2981 (m), 2919 (m), 2736 (w), 1574 (s), 1469 (s), 1420 (m), 1374 (s), 1325 (m), 1246 (s), 1220 (s), 1139 (s), 1055 (s), 914 (m), 734 (m), 686 (m). Mass spectrum: m/z = 373.1 {M}+ (M = {[HmimBut]HgMe}).

Synthesis of {[HmimBut]HgEt}[BF4]

A mixture of EtHgCl (500 mg, 1.89 mmol) and AgBF4 (367 mg, 1.89 mmol) was treated with CH2Cl2 (25 mL) resulting in the immediate deposition of a white precipitate. The suspension was stirred 3 hours and filtered into a flask containing HmimBut (221 mg, 1.42 mmol). The resulting solution was stirred 1 hour at room temperature and the volatile components removed in vacuo. The residue was extracted into C6H6 (20 mL) and filtered. The volatile components were removed by lyophilization to give {[HmimBut]HgEt}[BF4] as a white powder (480 mg, 72 %). Crystals suitable for X-ray diffraction were obtained by vapor diffusion of pentane into a THF solution of the compound. 1H NMR (C6D6) 1.15 [s, 9H, H{C3N2H2[C(CH3)3]S}Hg], 1.17[m, 3H, HgCH2CH3], 1.61[m, 2H, H{C3N2H2[C(CH3)3]S}HgCH2CH3], 6.33 [d, 1H, 3JH-H= 2Hz, H{C3N2H2[C(CH3)3]S}HgEt], 6.90 [d, 1H, 3JH-H= 2Hz, H{C3N2H2[C(CH3)3]S}HgEt], 12.27 [br, 1H, H{C3N2H2[C(CH3)3]S}HgEt]. 13C{1H} NMR (C6D6) 13.8 [1C, H{C2N2H2[C(CH3)3]CS}HgCH2CH3], 28.5 [1C, H{C2N2H2[C(CH3)3]CS}HgCH2CH3], 59.9 [1C, H{C2N2H2[C(CH3)3]CS}HgEt], 119.7 [1C, H{C2N2H2[C(CH3)3]CS}HgEt], obscured by solvent [1C, H{C2N2H2[C(CH3)3]CS}HgEt], 145.1 [1C, H{C2N2H2[C(CH3)3]CS}HgEt]. IR Data (KBr pellet, cm-1): 3287 (m), 3183 (m), 3158 (m), 2984 (m), 2928 (m), 2868 (m), 2742 (w), 1730 (w), 1618 (w), 1577 (s), 1480 (m), 1460 (m), 1431 (w), 1408 (w), 1373 (m), 1338 (m), 1284 (w), 1250 (m), 1222 (s), 1182 (s), 1143 (vs), 1129 (s), 1106 (vs), 1068 (vs), 1044 (vs), 958 (s), 913 (m), 818 (w), 785 (w), 755 (s), 696 (m). Mass spectrum: m/z = 387.1 {M}+ (M = {[HmimBut]HgEt})

Reactivity of {[HmimBut]HgMe}+ towards NaSPh

A mixture of {[HmimBut]HgMe}[BF4] (25 mg, 0.055 mmol) and NaSPh (10 mg, 0.076 mmol) was treated with C6D6 (0.7 mL). The reaction was monitored by using 1H NMR spectroscopy which demonstrated the formation of PhSHgMe and HmimBut within 20 minutes at room temperature.

Comparison of the reactivity of {[HmimBut]HgEt}[BF4] towards PhSH in the presence and absence of HmimBut

A solution of {[HmimBut]HgEt}[BF4] (15 mg, 0.032 mmol) in C6D6 (1.5 mL) was treated with PhSH (15 µL) mesitylene (2 µL) as an internal standard. The solution was divided into two NMR tubes, to which one was treated with HmimBut (5 mg), and the two samples were monitored by 1H NMR spectroscopy. For the sample that was treated with HmimBut, 1H NMR spectroscopy demonstrated the complete loss of the mercury ethyl signal and the formation of ethane over a period of 2 days. For the sample without added HmimBut, 1H NMR spectroscopy demonstrated that {[HmimBut]HgEt}+ was unperturbed and there was no formation of ethane over a period of 2 days and only small amounts (< 5 %) could be detected after a period of 10 days at room temperature. However, quantitative elimination of ethane was achieved over a period of 10 days at 60 °C.

Synthesis of {[HmimBut]2Hg}[BF4]2

{[HmimBut]HgEt}[BF4] (15 mg, 0.032 mmol) was treated with a solution of PhSH (20 µL) in C6D6 (0.7 mL) and heated at 60 °C for a period of 3 days. A white precipitate was deposited upon cooling to room temperature. The mother liquor was decanted, and the solid was washed with pentane (2 × 0.5 mL) and dried in vacuo to give {[HmimBut]2Hg}[BF4]2 as a white powder (4 mg, 37% yield). 1H NMR (C6D6) 1.25 [s, 18 H, {HC3N2H2[C(CH3)3]S}2Hg], 5.97 [t, 2H, JH-H = 2 Hz, {HC3N2H3[C(CH3)3]S}2Hg], 6.32 [t, 2H, JH-H = 2Hz, {HC3N2H2[C(CH3)3]S}2Hg], 12.11 [br, 2H, {NHC3N2H2[C(CH3)3]S}2Hg] (note: the chemical shifts are influenced by HmimBut due to exchange). The accompanying redistribution product, (PhS)2Hg, was observed by 1H NMR spectroscopic analysis of the sample prior to isolating {[HmimBut]2Hg}[BF4]2.

Synthesis of {[HmimBut]4Hg}[BF4]2

(a) A solution of {[HmimBut]HgEt}[BF4] (15 mg, 0.032 mmol) in C6D6 (0.7 mL) was treated with HmimBut (5 mg, 0.032 mmol) and PhSH (5 µL). Over a period of several days crystals of composition {[HmimBut]4Hg}[BF4]2 suitable for X–ray diffraction were deposited. The formation of (PhS)2Hg and ethane was demonstrated by 1H NMR spectroscopy. 1H NMR for {[HmimBut]4Hg}[BF4]2 (C6D6): 1.47 [s, 36H, {HC3N2H2[C(CH3)3]S}4Hg], 5.95 [d, JH-H= 2 Hz, 4H of {HC3N2H3[C(CH3)3]S}4Hg], 6.06 [d, JH-H= 2Hz, 4H, {HC3N2H2[C(CH3)3]S}4Hg], 12.1 [br, 4H of {NHC3N2H2[C(CH3)3]S}4Hg] (note: the chemical shifts are influenced by HmimBut due to exchange).

(b) A solution of {[HmimBut]HgMe}[BF4] (5 mg, 0.011 mmol) in C6D6 (0.7 mL) was treated HmimBut (5 mg, 0.032 mmol) and PhSH (5 µL). The solution was heated at 110 °C for a period of 2 weeks resulting in the formation of {[HmimBut]4Hg}[BF4]2.

Reactivity of {[HmimBut]HgEt}[BF4] towards p-ButC6H4SH in the presence and absence of HmimBut]

A solution of {[HmimBut]HgEt}[BF4] (12 mg, 0.03 mmol) and mesitylene (10 µL) in C6D6 (1.5 mL) was treated with p-ButC6H4SH (20 µL). The solution was split into two NMR tubes, one of which contained HmimBut (3 mg, 0.02 mmol). The solution to which HmimBut was added reacted over a period of 1 day at room temperature to eliminate ethane and generate [p-ButC6H4S]2Hg and {[HmimBut]4Hg}[BF4]2, as demonstrated by 1H NMR spectroscopy. In contrast, in the absence of additional HmimBut, elimination of ethane was very slow, with < 10% after 3 days at room temperature. The sample was heated at 60°C, resulting in ca. 60% conversion over 12 days. Complete elimination of ethane was achieved by heating at 110°C for 3 hours, and [p-ButC6H4S]2Hg and {[HmimBut]2Hg}2[BF4] were identified by 1H NMR spectroscopy.

1H NMR Spectroscopic evidence for reversible binding of HmimBut to PhSHgMe

(a) A solution PhSHgMe (7 mg, 0.02 mmol) in C6D6 (0.7 mL) was treated with mesitylene (10 µL) and successive portions of HmimBut (4 × 2 mg, 0.01 mmol). The resulting solution was monitored by 1H NMR spectroscopy, thereby demonstrating that the chemical shift of the mercury methyl signal is a function of the concentration of HmimBut (0 mg HmimBut, δCH3 = 0.081; 2 mg HmimBut, δCH3 = 0.099; 4 mg HmimBut, δCH3 = 0.126; 6 mg HmimBut, δCH3 = 0.145; 8 mg HmimBut, δCH3 = 0.172) due to rapid and reversible coordination of HmimBut.

(b) A solution PhSHgMe (ca. 50 mg, 0.15 mmol) in C6D6 (0.7 mL) was treated with successive portions of HmimBut (2 × 10 mg, 0.06 mmol). The resulting solution was monitored by 199Hg NMR spectroscopy, thereby demonstrating that the chemical shift of the mercury signal is a function of the concentration of HmimBut (0 mg HmimBut, −557 ppm; 10 mg HmimBut, −542 ppm; 20 mg HmimBut, −540 ppm).

1H NMR Spectroscopic evidence for reversible binding of HmimBut to PhSHgEt

(a) PhSHgEt (7 mg, 0.02 mmol) in C6D6 (0.7 mL) was treated with mesitylene (10 µL) and successive portions of HmimBut (4 × 2 mg, 0.01 mmol). The resulting solution was monitored by 1H NMR spectroscopy, thereby demonstrating that the chemical shifts of the mercury ethyl signals are a function of the concentration of HmimBut (0 mg HmimBut, δCH2 = 0.944, δCH3 = 0.825; 2 mg HmimBut, δCH2 = 0.958, δCH3 = 0.837; 4 mg HmimBut, δCH2 = 0.986, δCH3 = 0.863; 6 mg HmimBut, δCH2 = 1.019, δCH3 = 0.8923; 8 mg HmimBut, δCH2 = 1.066, δCH3 = 0.930).

(b) A solution PhSHgEt (200 mg, 0.59 mmol) in C6D6 (0.7 mL) was treated with successive portions of HmimBut (2 × 25 mg, 0.16 mmol). The resulting solution was monitored by 199Hg NMR spectroscopy, thereby demonstrating that the chemical shift of the mercury signal is a function of the concentration of HmimBut (0 mg HmimBut, −730 ppm; 25 mg HmimBut, −684 ppm; 50 mg HmimBut, −680 ppm).

1H NMR Spectroscopic evidence for reversible binding of HmimBut to {[HmimBut]HgMe}+

A solution of {[HmimBut]HgMe}[BF4] in C6D6 was titrated with a solution of HmimBut in C6D6 and was monitored by 1H NMR spectroscopy. Evidence for rapid reversible binding is provided by the observation that the signals due to the [mimBut] and alkyl groups shift, while no signals due to uncoordinated HmimBut are observed.

Synthesis of [p-ButC6H4S]2Hg

HgO (1.00 g, 4.62 mmol) was treated with a solution of p-ButC6H4SH (1.59 mL, 9.3 mmol) in EtOH (30 mL). The resulting orange suspension turned white over a period of 1 hour and was stirred for an additional 3 hours at room temperature. After this period, the mixture was filtered and the precipitate was washed with EtOH (30 mL) and dried in vacuo to give [p-ButC6H4S]2Hg as a white powder (2.30 g, 94% yield). Crystals of composition [p-ButC6H4S]2Hg suitable for X-ray diffraction were obtained from CH3CN. 1H NMR (C6D6) 1.15 [s, 18H, [p-ButC6H4S]2Hg], 6.99 [d, 3JH-H = 8 Hz, 4 H [p-ButC6H4S]2Hg], 7.33 [d, 3JH-H = 8 Hz, 4 H [p-ButC6H4S]2Hg]. 13C NMR (C6D6) 31.4 [6 C, [p-{(CH3)3C}C6H4S]2Hg], 34.4 [2 C [p-{(CH3)3C}C6H4S]2Hg], 126.4 [4 C SCC4H4CBut], 131.0 [2 C SCC4H4CBut], 133.2 [4 C SCC4H4CBut], 149.2 [2 C SCC4H4CBut]. IR Data (KBr pellet, cm-1): 3071 (w), 2961 (vs), 2901 (m), 2866 (m), 1491 (s), 1461 (m), 1396 (m), 1361 (m), 1268 (m), 1200 (w), 1119 (s), 1080 (w), 1009 (m), 829 (m), 820 (m), 807 (m), 737 (m), 722 (w). Mass spectrum: m/z = 530.6 {M }+.

X-ray structure determinations

Single crystal X-ray diffraction data were collected on either a Bruker Apex II diffractometer or a Bruker P4 diffractometer equipped with a SMART CCD detector. Crystal data, data collection and refinement parameters are summarized in Table 2. The structures were solved using direct methods and standard difference map techniques, and were refined by full-matrix least-squares procedures on F2 with SHELXTL (Version 6.10).32

Table 2
Crystal, intensity collection and refinement data.

Supplementary Material

1_si_001

Acknowledgments

We thank the National Institutes of Health (GM046502) for support of this research. The National Science Foundation (CHE-0619638) is thanked for acquisition of an X–ray diffractometer.

Footnotes

Supporting Information Available: Experimental details, computational data and crystallographic data in CIF format. This material is available free of charge via the Internet at http://pubs.acs.org.

References

1. (a) Clarkson TW, Magos L. Crit Rev Toxicol. 2006;36:609–662. [PubMed](b) Mutter J, Naumann J, Guethlin C. Crit Rev Toxicol. 2007;37:537–549. [PubMed](c) Clarkson TW. Env Health Persp Suppl. 2002;110:11–23. [PMC free article] [PubMed](d) Clarkson TW. Crit Rev Clin Lab Sci. 1997;34:369–403. [PubMed](e) Langford NJ, Ferner RE. J Human Hypertension. 1999;13:651–656. [PubMed](f) Boening DW. Chemosphere. 2000;40:1335–1351. [PubMed](g) Magos L. Metal Ions in Biological Systems. 1997;34:321–370. [PubMed](h) Hutchison AR, Atwood DA. J Chem Crystallogr. 2003;33:631–645.(i) Alessio L, Campagna M, Lucchini R. Am J Ind Med. 2007;50:779–787. [PubMed](j) Clarkson TW, Vyas JB, Ballatorl N. Am J Ind Med. 2007;50:757–764. [PubMed](k) Risher JF, De Rosa CT. J Env Health. 2007;70:9–16. [PubMed](l) Onyido I, Norris AR, Buncel E. Chem Rev. 2004;104:5911–5929. [PubMed](m) Ozuah PO. Curr Probl Pediatr. 2000;30:91–99. [PubMed]
2. (a) Moore MJ, Distefano MD, Zydowsky LD, Cummings RT, Walsh CT. Acc Chem Res. 1990;23:301–308.(b) Walsh CT, Distefano MD, Moore MJ, Shewchuk LM, Verdine GL. FASEB J. 1988;2:124–130. [PubMed]
3. Lafrance-Vanasse J, Lefebvre M, Di Lello P, Sygusch J, Omichinski JG. J Biol Chem. 2009;284:938–944. [PubMed]
4. Melnick JG, Parkin G. Science. 2007;317:225–227. [PubMed]
5. (a) Stradeit H. Angew Chem Int Ed Engl. 2007;46:2–5.(b) Miller SM. Nature Chemical Biology. 2007;3:537–536. [PubMed](c) Omichinski JG. Science. 2007;317:205–206. [PubMed]
6. Some of this research has been previously communicated. See reference 4.
7. (a) Kreevoy MM. J Am Chem Soc. 1957;79:5927–5930.(b) Kreevoy MM, Hansen RL. J Am Chem Soc. 1961;83:626–630.
8. Barone V, Bencini A, Totti F, Uytterhoeven MG. Organometallics. 1996;15:1465–1469.
9. Melnick JG, Yurkerwich K, Buccella D, Sattler W, Parkin G. Inorg Chem. 2008;47:6421–6426. [PubMed]
10. For simplicity, this description does not take into account the longer distance secondary Hg•••S interactions that are often observed in mercury compounds.
11. (PhS)2Hg has been previously reported. See, for example: Carlton L, White D. Polyhedron. 1990;9:2717–2720.Canty AJ, Kishimoto R. Inorg Chim Acta. 1977;24:109–122.
12. PhSHgMe has been synthesized by reaction of MeHgCl with PhSH, but subsequent elimination of methane was not observed; however, the conditions employed were not reported. In contrast, PhSHgPh has been reported to react with PhSH to give (PhS)2Hg and PhH. See: Cross RJ, Jenkins CM. Environ Pollut. 1975;8:179–184.
13. Kreevoy MM, Hansen RL. J Am Chem Soc. 1961;83:626–630.
14. Nugent WA, Kochi JK. J Am Chem Soc. 1976;98:5979–5988.
15. The related methimazole complex (mimMe)HgMe has been reported. See: Norris AR, Taylor SE, Buncel E, Bélanger-Gariépy F, Beauchamp AL. Can J Chem. 1983;61:1536–1541.
16. Brower KR, Gay B, Konkol TL. J Am Chem Soc. 1966;88:1681–1685.
17. Ni B, Kramer JR, Bell RA, Werstiuk NH. J Phys Chem A. 2006;110:9451–9458. [PubMed]
18. In the presence of excess PhSH and HmimBut, the mercury product could possibly be a dynamic mixture of Hg(SPh)2, [Hg(SPh)3], [Hg2(SPh)6]2– and [Hg(SPh)4]2–. See: Christou G, Folting K, Huffman JC. Polyhedron. 1984;3:1247–1253.Bowmaker GA, Dance IG, Harris RK, Henderson W, Laban T, Scudder MC, Oh SW. J Chem Soc Dalton Trans. 1996:2381–2388.
19. In view of the observation that coordination of HmimBut has a significant effect on the facility of protolytically cleaving the Hg–C bond, it is possible that protolytic cleavage in the absence of this reagent does not occur with two-coordinate PhSHgR but occurs preferentially via low concentration species with higher coordination numbers, such as PhSHgR(PhSH) or [PhSHgR]2.
20. At elevated temperatures (100°C), a solution of PhSHgEt and HmimBut eliminates ethane.
21. The three-coordinate intermediates are represented with a “T-shaped”, rather than a “Y-shaped”, geometry on the basis that the former type of structure is often observed in organomercury compounds.a,b,c Indeed, this geometry is observed in [TmBut]HgR upon consideration of the secondary bonding interaction involving one of the mercaptoimidazole groups.d Furthermore, DFT geometry optimization calculations on {[HmimBut]2HgCH3}+ indicate a structure in which the S–Hg–S and S–Hg–C angles are 89° and 157°, respectively, which are more in accord with a “T-shaped”, rather than a “Y-shaped”, geometry. Three-coordinate mercury compounds with “Y-shaped” geometries and 120° bond angles are, nevertheless, known in situations where the three ligands contribute equally to the bonding (e.g. [Hg(SBut)3]).e (a) Casa JS, García-Tasende MS, Sordo J. Coord Chem Rev. 1999;193-195:283–359.(b) Holloway CE, Melník M. J Organomet Chem. 1995;495:1–31.(c) Holloway CE, Melnik M. Main Group Metal Chem. 1994;17:799–885.(d) reference 4.(e) Wright JG, Natan MJ, MacDonnell FM, Ralston DM, O'Halloran TV. Prog Inorg Chem. 1990;38:323–412.
22. Kreevoy MM, Goon DJW, Kayser RA. J Am Chem Soc. 1966;88:5529–5534.
23. Gopinath E, Bruice TC. J Am Chem Soc. 1987;109:7903–7905.
24. Note that 2JHg–H (235 Hz) for PhSHgMe does not change significantly in the presence of HmimBut.
25. The 199Hg NMR chemical shifts for PhSHgR are within the range observed for two-coordinate mercury thiolate compounds. See references 9, 21d, and Almagro X, Clegg W, Cucurull-Sánchez L, González-Duarte P, Traveria M. J Organomet Chem. 2001;623:137–148.Carlton L, White D. Polyhedron. 1990;9:2717–2720.
26. The dithiolate complex [p-ButC6H4S]2Hg can also be independently prepared via reaction of p-ButC6H4SH with HgO, a method analogous to that for other Hg(SR)2 complexes. See: Carlton L, White D. Polyhedron. 1990;9:2717–2720.
27. For other structurally characterized (ArS)2Hg compounds, see: Alsaadi BM, Sandström M. Acta Chem Scand. 1982;A36:509–512.Gruff ES, Koch SA. J Am Chem Soc. 1990;112:1245–1247.Block E, Brito M, Gernon M, McGowty D, Kang H, Zubieta J. Inorg Chem. 1990;29:3172–3181.Kato M, Kojima K, Okamura T, Yamamoto H, Yamamura T, Ueyama N. Inorg Chem. 2005;44:4037–4044. [PubMed]Chen JX, Zhang WH, Tang XY, Ren ZG, Zhang Y, Lang JP. Inorg Chem. 2006;45:2568–2580. [PubMed]Ueyama N, Taniuchi K, Okamura T, Nakamura A, Maeda H, Emura E. Inorg Chem. 1996;35:1945–1951.Chen JX, Zhang WH, Tang XY, Ren ZG, Li HX, Zhang Y, Lang JP. Inorg Chem. 2006;45:7671–7680. [PubMed]Almagro X, Clegg W, Cucurull-Sánchez L, González-Duarte P, Traveria M. J Organomet Chem. 2001;623:137–148.
28. Manceau A, Nagy KL. Dalton Trans. 2008:1421–1425. [PubMed]
29. Gottlieb HE, Kotlyar V, Nudelman A. J Org Chem. 1997;62:7512–7515. [PubMed]
30. Kidd RG, Goodfellow RJ. In: NMR and the Periodic Table. Harris RK, Mann BE, editors. Academic Press; New York: 1978. p. 268.
31. Cassidy CS, Reinhardt LA, Cleland WW, Frey PA. J Chem Soc Perkin Trans. 1999;2:635–641.
32. (a) Sheldrick GM. SHELXTL, An Integrated System for Solving, Refining and Displaying Crystal Structures from Diffraction Data; University of Göttingen, Göttingen, Federal Republic of Germany. 1981(b) Sheldrick GM. Acta Cryst. 2008;A64:112–122. [PubMed]