- Email: [email protected]
Stoichiometric H2 production from H2O upon Mn2(CO)10 photolysis
Journal of Organometallic Chemistry 724 (2013) 1e6
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry journal homepage: www.elsevier.com/locate/jorganchem
Stoichiometric H2 production from H2O upon Mn2(CO)10 photolysis Jun Wei Kee, Che Chang Chong, Chun Keong Toh, Yuan Yi Chong, Wai Yip Fan* Department of Chemistry, National University of Singapore, 3 Science Drive, Singapore 117543, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 July 2012 Received in revised form 10 October 2012 Accepted 19 October 2012
Photolysis of Mn2(CO)10 in an alkane/water biphasic system has resulted in the generation of 1.80 0.16 mol of hydrogen per mol of Mn2(CO)10. Various studies including deuteration have indeed shown water to be the H2 source while kinetic studies have indicated a strong correlation between the concentration of the key intermediate MnH(CO)5 with the production rate of H2. Some of the oxygen atoms of water have been incorporated into a white solid assigned to MnCO3. A mechanism accounting for MnH(CO)5 formation from Mn2(CO)10 photolysis and subsequently H2 production from MnH(CO)5 has been proposed. Ó 2012 Elsevier B.V. All rights reserved.
Keywords: Manganese Photochemistry Water activation Hydride
1. Introduction Hydrogen production from water upon photolysis of transition metal complexes has generated considerable interest as a cleaner source of energy to replace fossil fuels. In recent years, several examples of metal complexes containing platinum [1,2], cobalt in cobaloximes [3,4], iron in hydrogenase mimics [5e9], ruthenium [10], molybdenum [11] and [12] manganese have been reported to catalytically activate water. Oxidative addition of water to a metal centre which forms an essential part of the reaction pathway has been discussed in a recent review [13]. Apart from catalysis, understanding OeH bond activation of water is also of much importance. In a previous study by Byers et al. [14], photolysis of Mn2(CO)8L2 (L ¼ CO, P(OEt)3 and PBu3) and HCl afforded MnH(CO)4L and MnCl(CO)4L along with traces of H2. Recently, our group has demonstrated stoichiometric photoactivation of water by cyclopentadienyl manganese tricarbonyl CpMn(CO)3 to produce hydrogen and hydrogen peroxide [15]. However, more examples and extensive studies are still required to understand the various photopathways, both stoichiometric and catalytic, leading to H2 generation from water especially under ambient conditions. In this paper, we report the stoichiometric hydrogen production upon photolysis of commercially-available dimanganese decacarbonyl Mn2(CO)10 in a biphasic mixture of alkane and water. Although Mn2(CO)10 is known to produce H2 upon reaction with
* Corresponding author. Fax: þ65 67791691. E-mail address: [email protected] (W.Y. Fan). 0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jorganchem.2012.10.034
HCl [14], our study demonstrates that the system still works with water. The key intermediate has been identified as MnH(CO)5 and a mechanism has been proposed to account for the general features of the experimental data. 2. Results and discussion Broadband irradiation (300e800 nm) of Mn2(CO)10 in a cyclohexane/water biphasic system resulted in the disappearance of its nCO bands (2045, 2013, 2002 and 1983 cm1) and the concomitant appearance of two new peaks at 2015 cm1 and 2007 cm1 (Fig. 1a). The corresponding 1H NMR spectrum of the sample recorded in C6D6 solvent showed a signal at 7.9 ppm. These spectroscopic assignments could be matched closely to previously reported data for MnH(CO)5 [16]. At the same time, the gas phase IR spectrum (Fig. 1b) of the headspace above the solution showed the production of CO, CO2 and MnH(CO)5 vapour [17]. The presence of MnH(CO)5 was further confirmed upon formation of MnH(CO)4PPh3 [18] in the presence of triphenylphosphine (PPh3) under mild heating. A series of control experiments has been carried out to understand how MnH(CO)5 is generated. Photolysis of Mn2(CO)10 in anhydrous cyclohexane or its thermal heating (up to 100 C) in wet cyclohexane resulted in much slower decomposition of Mn2(CO)10 and negligible signals of MnH(CO)5. In a series of deuteration experiments, Mn2(CO)10 photolysis in d12-cyclohexane/H2O yielded MnH(CO)5 while the corresponding photolysis in cyclohexane/D2O mixture afforded carbonyl signals which could be matched to the literature values for MnD(CO)5 [16] (nCO ¼ 2015 cm1 and 2006 cm1) (Fig. 2). The involvement of a cationic MnðCOÞ5 þ or MnðCOÞ6 þ intermediate
2
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
Fig. 1. FTIR spectra recorded after a 2-h broadband irradiation of Mn2(CO)10 in biphasic cyclohexane/water. (a) The production of MnH(CO)5 (2015 and 2007 cm1) taken in the cyclohexane layer (b) the production of CO (no ¼ 2143 cm1), CO2 (no ¼ 2340 cm1) and MnH(CO)5 (nCO ¼ 2033, 2028 and 2020 cm1) in the headspace above the reaction mixture. (c) XRD analysis of the white precipitate MnCO3.
has been ruled out as the photolysis of Mn2(CO)10 with tetra-nbutylammonium bromide did not give MnBr(CO)5. In the presence of tetrafluoroboric acid HBF4 in the aqueous layer, both rates of Mn2(CO)10 decomposition and MnH(CO)5 formation were not
2014.7
2005.2 Mn2(CO)10 + D2O 2006.6 Mn2(CO)10 + H2O 2030
2025
2020
2015 2010 2005 Wavenumbers (cm-1)
2000
1995
1990
Fig. 2. FTIR spectrum of MnH(CO)5 (2014.7 cm1 and 2006.6 cm1) or MnD(CO)5 (2014.7 cm1 and 2005.2 cm1) recorded at 0.5 cm1 resolution after a 2-h broadband irradiation of Mn2(CO)10 in a biphasic cyclohexane/H2O or cyclohexane/D2O respectively.
affected significantly as the pH is decreased from 6 to 2, suggesting that Hþ ions do not have a significant role in the formation of MnH(CO)5. From these data, it is evident that only light and water but no additives are required for MnH(CO)5 formation. Upon prolonged photolysis (>6 h) of the reaction mixture, a white solid has been obtained in 5e10% yield. The solid shows alkaline behaviour as it requires two equivalents of Hþ (from HCl aqueous solution) for neutralization. An XRD (X-ray powder diffraction) analysis of the solid in Fig. 1c confirms its identity to be manganese carbonate MnCO3 (JCPDS Card No.: 83-1763) [19e21]. Along with the MnH(CO)5 formation, H2 evolution has been observed throughout the photolysis until all the Mn2(CO)10 and MnH(CO)5 have decomposed. The headspace mass spectrum indicated the production of H2 at 1.80 0.16 mol per mol of Mn2(CO)10 used. D2 was also detected at 1.40 0.10 mol per mol of Mn2(CO)10 for D2O. The time profile of the H2 evolution has been shown together with those of the respective IR signals of MnH(CO)5 and Mn2(CO)10 in Fig. 3. The quantification of H2 production is carried out relative to the amount of Mn2(CO)10 used since the latter could be weighed accurately, rather than based on the IR absorbance of
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
Fig. 3. Time profile showing (a) the absorbance of D Mn2(CO)10 (at 2045 cm1) and , MnH(CO)5 (at 2007 cm1) and (b) the percentage yield of H2 per Mn2(CO)10 used at different time intervals throughout the photolysis period. The inset in (b) represents the typical mass spectrum for the m/e ¼ 1 to 10 obtained for the headspace of the photolytic mixture. (c) Time-dependent mass spectra taken of the headspace content upon a 5-h photolysis of Mn2(CO)10 in a hexane/H2O mixture, representing the signal at m/e ¼ 2. The signals obtained for a 50 Torr H2 standard and the headspace content of a similar mixture prior to photolysis are included for comparison.
MnH(CO)5 which may have a larger error due to the uncertainty in its extinction coefficient. From Fig. 3, the total yield of H2 increased with the photolytic duration of 5 h while Mn2(CO)10 decomposed within the first 3 h. A comparison of the time profiles showed that the rate of H2 evolution (given by the gradient of Fig. 3b) correlated more closely to the concentration of MnH(CO)5 rather than to Mn2(CO)10. In fact the total H2 yield obtained over 5 h (when all the MnH(CO)5 and Mn2(CO)10 have decomposed) doubled that of the 3-h yield (when all Mn2(CO)10 have decomposed) and suggested that the average rate is independent of Mn2(CO)10 but proportional to MnH(CO)5 instead.
3
To investigate further, MnH(CO)5 has been prepared in a pure form and photolysed further in the presence of water. Indeed, significant H2 signal has been detected in the mass spectrum. In contrast, the absence of photolysis or water (removed using molecule sieves 4A) did not produce H2. Thermal decomposition of MnH(CO)5 also failed to yield H2, highlighting the need for light in this step of the process. Interestingly, photolysis of MnD(CO)5 in cyclohexane/H2O produced not only the expected HD but also significant H2 and D2 in a ratio of HD:H2:D2 ¼ 3:7:1. The result is explained by a fast HeD exchange process between the acidic MnD(CO)5 (pKa w7) [22] and H2O upon mixing. Indeed, the appearance of IR signals due to MnH(CO)5 when a mixture of MnD(CO)5 and H2O was left standing in the dark for 1e3 h gave strong evidence of the exchange. The generation of H2 upon Mn2(CO)10 photolysis has been carried out in polar solvents such as thf and acetonitrile and resulted in similar yields. H2 can also be generated from acetic acid with similar efficiency upon Mn2(CO)10 photolysis. The reaction pathways mirror those in the water system as the production of MnH(CO)5 was observed. This study demonstrates the further utility of Mn2(CO)10 in activating OeH bonds in carboxylic acids. The Mn2(CO)10/H2O photolytic system has also been tested for its use in the wateregas shift process since a small quantity of CO2 has been detected. However, increasing the CO pressure from 0.1 bar to 1.5 bar not only slowed the reaction but also did not cause any increase in H2 even after long periods of irradiation. Preceding this work, numerous studies have been conducted for the initial photofragmentation of Mn2(CO)10 in various solvents [23e26]. Limiting the discussion to the photolysis of Mn2(CO)10 in non-polar solvents, it has been noted that MneMn fission [24] is much more dominant than MneCO cleavage [25] at wavelengths longer than 300 nm. The proposed mechanism will be based on the initial production of Mn(CO)5 radicals from the MneMn fission. The experimental data suggest that the production of H2 can be divided into two photoprocesses; first the generation of the key intermediate MnH(CO)5 from Mn(CO)5 radical and second, the reaction of MnH(CO)5 with H2O. In the previous study on the photoreaction of HCl with Mn2(CO)10, Byers et al. [14] has also suggested that Mn(CO)5 radicals produced upon MneMn dissociation undergo oxidative addition of HCl to give MnH(CO)5 and MnCl(CO)5 as the main products. Compare to HCl, the production of MnH(CO)5 using water has been measured to be only three times slower. Thus it is reasonable to adopt this mechanism, at least to account for MnH(CO)5 formation, by substituting HCl for H2O as illustrated in Fig. 4. Along with MnH(CO)5, a reactive manganese hydroxyl species, Mn(OH)(CO)5 is expected to have been generated. The second step of the photolysis is of vital importance as it has been shown that MnH(CO)5 concentration correlates strongly to the H2 production rate. Similar to the facile formation of Mn(CO)4PR3H complexes [18], MnH(CO)5 undergoes thermal CO dissociation followed by H2O substitution to give MnH(CO)4(H2O). A simple abstraction of the proton from the adjacent H2O ligand may liberate H2 through some as yet unknown pathway leaving a manganese hydroxyl intermediate behind. As mentioned earlier, a manganese hydroxyl species Mn(OH)(CO)5 is generated alongside MnH(CO)5 in the first step. This species is expected to be reactive since the OH ligand has been known to react with the electrophilic CO ligand to give metallocarboxylic acids. These complexes tend to decompose to give CO2 and the corresponding metal hydride [27]. In our experiments, small amounts of CO2 and MnCO3 were indeed detected; the latter is a resultant of carbonic acid formed in the presence of water. The resultant 16-electron metal hydride MnH(CO)4 coordinates to H2O and generates H2 upon photolysis again. Thus by invoking this
4
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
a CO
OC
OC
OC
OC Mn CO
OC Mn CO
CO
H
CO H2O
hv OC
OC
CO
H
Mn
H2O
CO Mn
H
c
CO
HO
H2O
CO
CO
OC hv
CO
H
CO
CO
OH2
CO +
HO
OH2
H2
CO
OC
CO
CO
HO
CO + Mn(CO)5
Mn CO
Mn
Mn
OH
OC
CO
CO
CO
CO Mn
CO
CO
CO2
CO Mn
H
hv
CO
CO OC
OC
+
Mn CO
OC
Mn
CO
CO
CO
CO
CO
CO
CO
OC
CO
CO
b
CO
CO
CO
hv
CO Mn CO
HO
+
H2
CO
Fig. 4. (a) Mechanism showing MneMn bond homolytic fission, leading to an eventual production of MnH(CO)5 and Mn(OH)(CO)5 (adopted from Ref. [14]). (b) Mechanism showing the H2 production from MnH(CO)5 and H2O. (c) Mechanism showing the H2 production from Mn(OH)(CO)5 and H2O.
pathway, it is possible to explain the generation of almost two equivalents of H2 from Mn2(CO)10 photolysis in water. 3. Conclusions Mn2(CO)10 photolysis in an alkane/water biphasic system has afforded 1.80 0.16 mol of hydrogen per mol of Mn2(CO)10. A manganese hydride MnH(CO)5 has been detected as an intermediate in the mixture. The H2 source has indeed been found to be water. The kinetic studies monitored using the IR signatures of Mn2(CO)10 and MnH(CO)5, indicate a correlation between the latter complex with the production of H2. Some of the oxygen atoms of water have been incorporated into a white solid assigned to MnCO3. A mechanism accounting for the formation of H2 via MnH(CO)5 has been proposed. We hope that this study is still potentially relevant to the development of catalytic schemes for H2 generation. 4. Experimental details 4.1. Materials and measurements All chemicals were purchased from SigmaeAldrich and used without further purification, unless otherwise noted. Chemicals used in the quantification were calibrated against their respective primary standards. 1H NMR spectra were recorded in C6D6 on a Bruker ACF300 and AMX500 NMR spectrometer with chemical shifts referenced to residual C6D6 peaks. Solution IR spectra were obtained on a Nexus 870 FT-IR spectrometer using a cell of 0.1 mm path length equipped with CaF2 windows. All photolysis experiments were conducted using a UV broadband lamp (wavelength range: 300e800 nm typical power measured at 10 cm distance: 0.80e0.90 W/cm2) at 10 cm distance away from the reaction mixture in borosilicate glass apparatus. Powder X-ray diffraction (XRD) measurement of solids was carried out using Siemens Powder XRD D5005 diffractometer.
4.2. Mass spectrometric determination of H2 and CO2 H2 gas was detected using a modified residual gas analyser (Pfeiffer Vacuum PrismaPlus QMG 200 with a high-sensitivity ion source) where the signals at m/e 1e50 were monitored over time. Prior to each analysis, a sample of pure H2 gas of known pressure (5e60 Torr) and air were injected for calibration and baseline correction respectively. For each analysis, 3.0 cm3 of the gas was injected directly using a syringe into a pre-evacuated (<1 106 Torr) 20 cm3 stainless steel chamber connected to the residual gas analyser. Thereafter, repeated 1 cm3 portions of gas samples were then introduced by means of two Swagelok ball valves to ensure that repeated sampling of the gas would not have significant depreciation of the initial pressure of gas. Subsequently, the amount of H2 present in the photolytic mixture was calculated while taking into account volume correction factors. Similarly, the amount of CO2 produced was determined by monitoring the signal at m/e 44. Pure CO2 from a cylinder was used to prepare standards of known pressure (20e60 Torr) and air were injected for calibration and baseline correction respectively.
4.3. Photolysis of Mn2(CO)10 in cyclohexane/water mixture A mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3) was photolysed in an evacuated glass apparatus (25 cm3 volume) for 2 h before the reaction mixture was sampled for IR and 1H NMR. The main metal complex intermediate was identified as MnH(CO)5 (n(CO) in cyclohexane: 2015s and 2007w; 1H NMR (C6D6) d 7.9 ppm). Upon complete photolysis, headspace mass-spectrometric studies indicated H2 production to be 1.80 0.16 mol of hydrogen per mol of Mn2(CO)10 and IR spectroscopic studies of the headspace indicated CO2 production to be 0.08 0.03 mol per mol of Mn2(CO)10 used respectively.
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
5
4.4. Photolysis of Mn2(CO)10 under a variety of conditions
4.7. Photolysis of MnH(CO)5 in cyclohexane/water
Typically, a 5-h photolysis (unless otherwise stated) was carried out for Mn2(CO)10 (0.0045 g, 1.1 105 mol) under various conditions. If anhydrous conditions were required, dried cyclohexane (1.0 cm3) which has been distilled three times from molecular sieves was used as the solvent. For the detection of MnH(CO)4(PPh3), mild heating (40e50 C) for 2 h was carried out in the presence triphenylphosphine PPh3 (104 mol) after MnH(CO)5 has been detected in an initial mixture containing Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3). The main metal complex was identified as MnH(CO)4(PPh3). (n(CO) in cyclohexane: 2062w, 1984m, 1969vs and 1958s; 1H NMR (C6D6) d 6.88 ppm doublet, 2J(PeH): 33 Hz.) For the bromide ion test, the photolysis was carried out for a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and a 1 M aqueous tetra-n-butylammonium bromide solution (0.10 cm3). For the test on the influence of acid, the photolysis was carried out for a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and aqueous tetrafluoroboric acid HBF4 (1.0 cm3) whose pH can be monitored and adjusted from 1 to 6 by varying the acid concentration. For examining the influence of different polar solvents, the photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in a solution of THF (1.0 cm3) and H2O (0.1 cm3) was conducted in an evacuated glass apparatus (25 cm3 volume) using a UV broadband lamp before the reaction mixture was sampled for headspace massspectrometric studies. The same procedures were repeated using acetonitrile. The H2 yield for THF and acetonitrile were at 1.90 0.20 and 1.87 0.20 mol per mol of Mn2(CO)10 respectively. For CO pressurization studies, a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3) was photolysed in a glass apparatus (25 cm3 volume, evacuated and filled with CO pressure of 0.1 bare1.5 bar) for 5 h before the reaction mixture was sampled for IR and 1H NMR spectroscopic studies.
MnH(CO)5 was generated from the photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and water (0.10 cm3) and then distilled into an evacuated glass apparatus (25 cm3 volume) containing cyclohexane (1.0 cm3) and water (0.10 cm3). A 2-h photolysis was carried out for the distillate and the reaction mixture was then sampled for IR spectroscopic and headspace mass-spectrometric studies.
4.5. Deuteration studies A 3-h photolysis was carried out for a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in dried cyclohexane (1.0 cm3) and D2O (0.10 cm3). The reaction mixture was sampled for IR and the product was identified as MnD(CO)5. (n(CO) in cyclohexane: 2015s and 2006w) Upon complete photolysis, D2 was detected at 1.40 0.10 mol per mol of Mn2(CO)10. Similarly, the photolysis was repeated in a mixture of 1 cm3 of d12-cyclohexane and 1 cm3 of H2O and the product was identified as MnH(CO)5. 4.6. Time profile monitoring The photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3) was conducted in an evacuated glass apparatus (25 cm3 volume) using a UV broadband lamp for a series of different time intervals before the reaction mixture was sampled for IR, 1H NMR and headspace massspectroscopic measurements. The relative amount of Mn2(CO)10 was determined by one of its infrared absorption bands at 2045 cm1. The relative amount of MnH(CO)5 was determined by following the difference between the observed absorbance at 2007 cm1 and the corresponding absorbance contributed by Mn2(CO)10. The H2 signal at m/e 2 was sampled by the residual gas analyser at different time intervals.
4.8. Photolysis of MnH(CO)5 in dried cyclohexane MnH(CO)5 was generated from a 2-h photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) and was distilled into an evacuated glass apparatus (25 cm3 volume) containing cyclohexane (1.0 cm3) and molecular sieves (1 g). A 6-h photolysis was carried out for the distillate and the reaction mixture was then sampled for IR spectroscopic and headspace mass-spectrometric studies.
4.9. HeD exchange studies of MnD(CO)5 MnD(CO)5 was distilled from a 2-h photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane/D2O into an evacuated glass apparatus (25 cm3 volume) containing cyclohexane (1.0 cm3) and H2O (0.10 cm3). The resulting mixture was left to stand in the dark for up to 3 h before an IR spectrum was obtained. Subsequently, MnD(CO)5 was distilled into a mixture of cyclohexane (1.0 cm3) and D2O (0.10 cm3). MnD(CO)5 was photolysed with cyclohexane/H2O before the head-space was sampled for headspace mass-spectrometric studies.
4.10. Attempted thermal activation of H2O using Mn2(CO)10 or MnH(CO)5 Mn2(CO)10 (0.0045 g, 1.1 105 mol) was heated at 100 C in a mixture of dodecane (1.0 cm3) and H2O (0.10 cm3) mixture for 3 days before the dodecane layer was sampled for IR spectroscopic and mass-spectrometric studies. Similarly, MnH(CO)5, synthesized from a 2-h photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol), was heated in a dodecane/water mixture at 100 C for 3 days. The reaction mixture was then sampled for IR spectroscopic and headspace mass-spectrometric studies.
4.11. Chemical analysis of solid residue A 5-h photolysis was carried out for a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3) and a white precipitate was observed. Once Mn2(CO)10 has reacted completely, the solvent and any residual MnH(CO)5 were removed under vacuum, before the residue was washed further with cyclohexane under reduced air conditions. The solid is dissolved in water and titrated with 0.1 M HCl solution with methyl orange as the indicator. Powder X-ray diffraction (XRD) measurement of the solid was carried out using Siemens Powder XRD D5005 diffractometer.
4.12. Photolysis of Mn2(CO)10 in acetic acid A mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in acetic acid (1.0 cm3) was photolysed in an evacuated glass apparatus (25 cm3 volume) for 5 h before the reaction mixture was sampled for IR.
6
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
Acknowledgements The project was supported by a research grant provided by NUS 143-000-430-112. J.W.K., C.K.T. and Y.Y.C. acknowledge the support of NUS scholarships. References [1] P. Du, K. Knowles, R. Eisenberg, J. Am. Chem. Soc. 130 (2008) 12576e12577. [2] A.J. Esswein, D.G. Nocera, Chem. Rev. 107 (2007) 4022e4047. [3] A. Fihri, V. Artero, M. Razavet, C. Baffert, W. Leibl, M. Fontecave, Angew. Chem. Int. Ed. 47 (2008) 564e567. [4] X. Hu, B.S. Brunschwig, C.J. Peters, J. Am. Chem. Soc. 129 (2007) 8988e8998. [5] C. Tard, X. Liu, S.K. Ibrahim, M. Bruschi, L.D. Gioia, S.C. Davies, X. Yang, L. Wang, G. Sawers, C.J. Pickett, Nature 433 (2005) 610e613. [6] D. Streich, M. Karnahl, Y. Astuti, C.W. Cady, L. Hammarström, R. Lomoth, S. Ott, Eur. J. Inorg. Chem. 7 (2011) 1106e1111. [7] D.J. Evans, C.J. Pickett, Chem. Soc. Rev. 32 (2003) 268e275. [8] G.A.N. Felton, A.K. Vannucci, J. Chen, L.T. Lockett, N. Okumura, B.J. Petro, U.I. Zakai, D.H. Evans, R.S. Glass, D.L. Lichtenberger, J. Am. Chem. Soc. 129 (2007) 12521e12530. [9] F. Gärtner, B. Sundararaju, A. Surkus, A. Boddien, B. Loges, H. Junge, P.H. Dixneuf, M. Beller, Angew. Chem. Int. Ed. 48 (2009) 9962e9965. [10] S.W. Kohl, L. Weiner, L. Schwartsburd, L. Konstantinovski, L.J.W. Shimon, Y. Ben-David, M.A. Iron, D. Milstein, Science 324 (2009) 74e77. [11] H.I. Karunadasa, C.J. Chang, J.R. Long, Nature 464 (2010) 1329e1333. [12] R. Brimblecombe, A. Koo, G.C. Dismukes, G.F. Swiegers, L. Spiccia, J. Am. Chem. Soc. 132 (2010) 2892e2894. [13] O.V. Ozerov, Chem. Soc. Rev. 38 (2009) 83e88. [14] B.H. Byers, T.P. Curran, M.J. Thompson, L.J. Sauer, Organometallics 2 (1983) 459e460. [15] J.W. Kee, Y.Y. Tan, B.H.G. Swennenhuis, A.A. Bengali, W.Y. Fan, Organometallics 30 (2011) 2154e2159.
[16] P.S. Braterman, R.W. Harrill, H.D. Kaesz, J. Am. Chem. Soc. 89 (1967) 2851e2855. [17] Z. Mahmood, M. Azam, A. Mushtaq, R. Kausar, S. Kausar, S.R. Gilani, Spectrochim. Acta A 65 (2006) 445e452. [18] (a) B.H. Byers, T.L. Brown, J. Organomet. Chem. 127 (1977) 181e192; (b) W. Hieber, H. Duchatsch, Chem. Ber. 98 (1965) 2933e2938; (c) W.Hieber, G.Faulhaber, F.Theubert, Z. Anorg. Allg. Chem. 314(1962) 125e143. [19] X. Wang, Y.D. Li, Mater. Chem. Phys. 82 (2003) 419e422. [20] C.H.B. Ng, W.Y. Fan, J. Phys. Chem. C 111 (2007) 9166e9171. [21] C.H.B. Ng, W.Y. Fan, J. Phys. Chem. B 110 (2006) 20801e20807. [22] C. Elschenbroich, Organometallics, third ed., Wiley-VCH, Weinheim, Germany, 2006. [23] M.S. Wrighton, Chem. Rev. 74 (1974) 401e430. [24] (a) D.A. Steinhurst, A.P. Baronavaski, J.C. Owrutsky, Chem. Phys. Lett. 361 (2002) 513e519; (b) J.L. Hughey, C.P. Anderson, T.J. Meyer, J. Organomet. Chem. 125 (1977) C49eC52; (c) R.W. Wegman, R.J. Olsen, O.R. Gard, L.R. Faulkner, T.L. Brown, J. Am. Chem. Soc. 103 (1981) 6089e6092. [25] (a) R.S. Herrick, T.L. Brown, Inorg. Chem. 23 (1984) 4550e4553; (b) S.P. Church, H. Hermann, F.W. Grevels, K. Schaffner, J. Chem. Soc. Chem. Commun. 12 (1984) 785e786; (c) L.J. Rothberg, N.J. Cooper, K.S. Peters, V. Vaida, J. Am. Chem. Soc. 104 (1982) 3536e3537; (d) H. Yesaka, T. Kobayashi, K. Yasufuku, S. Nagakuru, J. Am. Chem. Soc. 105 (1983) 6249e6252; (e) A.F. Hepp, M.S. Wrighton, J. Am. Chem. Soc. 105 (1983) 5934e5935. [26] (a) A. Hudson, M.F. Lappert, J.J. Macquitty, B.K. Nicholson, H. Zainal, G.R. Luckhurst, C. Zannoni, S.W. Bratt, M.C.R. Symons, J. Organomet. Chem. 110 (1976) C5eC8; (b) A. Hudson, M.F. Lappert, B.K. Nicholson, J. Chem. Soc. Dalton Trans. (1977) 551e554. [27] (a) D.H. Gibson, T.S. Ong, Organometallics 3 (1984) 1911e1913; (b) N. Grice, S.C. Kao, R. Pettit, J. Am. Chem. Soc. 101 (1979) 1627e1628.
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry journal homepage: www.elsevier.com/locate/jorganchem
Stoichiometric H2 production from H2O upon Mn2(CO)10 photolysis Jun Wei Kee, Che Chang Chong, Chun Keong Toh, Yuan Yi Chong, Wai Yip Fan* Department of Chemistry, National University of Singapore, 3 Science Drive, Singapore 117543, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 July 2012 Received in revised form 10 October 2012 Accepted 19 October 2012
Photolysis of Mn2(CO)10 in an alkane/water biphasic system has resulted in the generation of 1.80 0.16 mol of hydrogen per mol of Mn2(CO)10. Various studies including deuteration have indeed shown water to be the H2 source while kinetic studies have indicated a strong correlation between the concentration of the key intermediate MnH(CO)5 with the production rate of H2. Some of the oxygen atoms of water have been incorporated into a white solid assigned to MnCO3. A mechanism accounting for MnH(CO)5 formation from Mn2(CO)10 photolysis and subsequently H2 production from MnH(CO)5 has been proposed. Ó 2012 Elsevier B.V. All rights reserved.
Keywords: Manganese Photochemistry Water activation Hydride
1. Introduction Hydrogen production from water upon photolysis of transition metal complexes has generated considerable interest as a cleaner source of energy to replace fossil fuels. In recent years, several examples of metal complexes containing platinum [1,2], cobalt in cobaloximes [3,4], iron in hydrogenase mimics [5e9], ruthenium [10], molybdenum [11] and [12] manganese have been reported to catalytically activate water. Oxidative addition of water to a metal centre which forms an essential part of the reaction pathway has been discussed in a recent review [13]. Apart from catalysis, understanding OeH bond activation of water is also of much importance. In a previous study by Byers et al. [14], photolysis of Mn2(CO)8L2 (L ¼ CO, P(OEt)3 and PBu3) and HCl afforded MnH(CO)4L and MnCl(CO)4L along with traces of H2. Recently, our group has demonstrated stoichiometric photoactivation of water by cyclopentadienyl manganese tricarbonyl CpMn(CO)3 to produce hydrogen and hydrogen peroxide [15]. However, more examples and extensive studies are still required to understand the various photopathways, both stoichiometric and catalytic, leading to H2 generation from water especially under ambient conditions. In this paper, we report the stoichiometric hydrogen production upon photolysis of commercially-available dimanganese decacarbonyl Mn2(CO)10 in a biphasic mixture of alkane and water. Although Mn2(CO)10 is known to produce H2 upon reaction with
* Corresponding author. Fax: þ65 67791691. E-mail address: [email protected] (W.Y. Fan). 0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jorganchem.2012.10.034
HCl [14], our study demonstrates that the system still works with water. The key intermediate has been identified as MnH(CO)5 and a mechanism has been proposed to account for the general features of the experimental data. 2. Results and discussion Broadband irradiation (300e800 nm) of Mn2(CO)10 in a cyclohexane/water biphasic system resulted in the disappearance of its nCO bands (2045, 2013, 2002 and 1983 cm1) and the concomitant appearance of two new peaks at 2015 cm1 and 2007 cm1 (Fig. 1a). The corresponding 1H NMR spectrum of the sample recorded in C6D6 solvent showed a signal at 7.9 ppm. These spectroscopic assignments could be matched closely to previously reported data for MnH(CO)5 [16]. At the same time, the gas phase IR spectrum (Fig. 1b) of the headspace above the solution showed the production of CO, CO2 and MnH(CO)5 vapour [17]. The presence of MnH(CO)5 was further confirmed upon formation of MnH(CO)4PPh3 [18] in the presence of triphenylphosphine (PPh3) under mild heating. A series of control experiments has been carried out to understand how MnH(CO)5 is generated. Photolysis of Mn2(CO)10 in anhydrous cyclohexane or its thermal heating (up to 100 C) in wet cyclohexane resulted in much slower decomposition of Mn2(CO)10 and negligible signals of MnH(CO)5. In a series of deuteration experiments, Mn2(CO)10 photolysis in d12-cyclohexane/H2O yielded MnH(CO)5 while the corresponding photolysis in cyclohexane/D2O mixture afforded carbonyl signals which could be matched to the literature values for MnD(CO)5 [16] (nCO ¼ 2015 cm1 and 2006 cm1) (Fig. 2). The involvement of a cationic MnðCOÞ5 þ or MnðCOÞ6 þ intermediate
2
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
Fig. 1. FTIR spectra recorded after a 2-h broadband irradiation of Mn2(CO)10 in biphasic cyclohexane/water. (a) The production of MnH(CO)5 (2015 and 2007 cm1) taken in the cyclohexane layer (b) the production of CO (no ¼ 2143 cm1), CO2 (no ¼ 2340 cm1) and MnH(CO)5 (nCO ¼ 2033, 2028 and 2020 cm1) in the headspace above the reaction mixture. (c) XRD analysis of the white precipitate MnCO3.
has been ruled out as the photolysis of Mn2(CO)10 with tetra-nbutylammonium bromide did not give MnBr(CO)5. In the presence of tetrafluoroboric acid HBF4 in the aqueous layer, both rates of Mn2(CO)10 decomposition and MnH(CO)5 formation were not
2014.7
2005.2 Mn2(CO)10 + D2O 2006.6 Mn2(CO)10 + H2O 2030
2025
2020
2015 2010 2005 Wavenumbers (cm-1)
2000
1995
1990
Fig. 2. FTIR spectrum of MnH(CO)5 (2014.7 cm1 and 2006.6 cm1) or MnD(CO)5 (2014.7 cm1 and 2005.2 cm1) recorded at 0.5 cm1 resolution after a 2-h broadband irradiation of Mn2(CO)10 in a biphasic cyclohexane/H2O or cyclohexane/D2O respectively.
affected significantly as the pH is decreased from 6 to 2, suggesting that Hþ ions do not have a significant role in the formation of MnH(CO)5. From these data, it is evident that only light and water but no additives are required for MnH(CO)5 formation. Upon prolonged photolysis (>6 h) of the reaction mixture, a white solid has been obtained in 5e10% yield. The solid shows alkaline behaviour as it requires two equivalents of Hþ (from HCl aqueous solution) for neutralization. An XRD (X-ray powder diffraction) analysis of the solid in Fig. 1c confirms its identity to be manganese carbonate MnCO3 (JCPDS Card No.: 83-1763) [19e21]. Along with the MnH(CO)5 formation, H2 evolution has been observed throughout the photolysis until all the Mn2(CO)10 and MnH(CO)5 have decomposed. The headspace mass spectrum indicated the production of H2 at 1.80 0.16 mol per mol of Mn2(CO)10 used. D2 was also detected at 1.40 0.10 mol per mol of Mn2(CO)10 for D2O. The time profile of the H2 evolution has been shown together with those of the respective IR signals of MnH(CO)5 and Mn2(CO)10 in Fig. 3. The quantification of H2 production is carried out relative to the amount of Mn2(CO)10 used since the latter could be weighed accurately, rather than based on the IR absorbance of
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
Fig. 3. Time profile showing (a) the absorbance of D Mn2(CO)10 (at 2045 cm1) and , MnH(CO)5 (at 2007 cm1) and (b) the percentage yield of H2 per Mn2(CO)10 used at different time intervals throughout the photolysis period. The inset in (b) represents the typical mass spectrum for the m/e ¼ 1 to 10 obtained for the headspace of the photolytic mixture. (c) Time-dependent mass spectra taken of the headspace content upon a 5-h photolysis of Mn2(CO)10 in a hexane/H2O mixture, representing the signal at m/e ¼ 2. The signals obtained for a 50 Torr H2 standard and the headspace content of a similar mixture prior to photolysis are included for comparison.
MnH(CO)5 which may have a larger error due to the uncertainty in its extinction coefficient. From Fig. 3, the total yield of H2 increased with the photolytic duration of 5 h while Mn2(CO)10 decomposed within the first 3 h. A comparison of the time profiles showed that the rate of H2 evolution (given by the gradient of Fig. 3b) correlated more closely to the concentration of MnH(CO)5 rather than to Mn2(CO)10. In fact the total H2 yield obtained over 5 h (when all the MnH(CO)5 and Mn2(CO)10 have decomposed) doubled that of the 3-h yield (when all Mn2(CO)10 have decomposed) and suggested that the average rate is independent of Mn2(CO)10 but proportional to MnH(CO)5 instead.
3
To investigate further, MnH(CO)5 has been prepared in a pure form and photolysed further in the presence of water. Indeed, significant H2 signal has been detected in the mass spectrum. In contrast, the absence of photolysis or water (removed using molecule sieves 4A) did not produce H2. Thermal decomposition of MnH(CO)5 also failed to yield H2, highlighting the need for light in this step of the process. Interestingly, photolysis of MnD(CO)5 in cyclohexane/H2O produced not only the expected HD but also significant H2 and D2 in a ratio of HD:H2:D2 ¼ 3:7:1. The result is explained by a fast HeD exchange process between the acidic MnD(CO)5 (pKa w7) [22] and H2O upon mixing. Indeed, the appearance of IR signals due to MnH(CO)5 when a mixture of MnD(CO)5 and H2O was left standing in the dark for 1e3 h gave strong evidence of the exchange. The generation of H2 upon Mn2(CO)10 photolysis has been carried out in polar solvents such as thf and acetonitrile and resulted in similar yields. H2 can also be generated from acetic acid with similar efficiency upon Mn2(CO)10 photolysis. The reaction pathways mirror those in the water system as the production of MnH(CO)5 was observed. This study demonstrates the further utility of Mn2(CO)10 in activating OeH bonds in carboxylic acids. The Mn2(CO)10/H2O photolytic system has also been tested for its use in the wateregas shift process since a small quantity of CO2 has been detected. However, increasing the CO pressure from 0.1 bar to 1.5 bar not only slowed the reaction but also did not cause any increase in H2 even after long periods of irradiation. Preceding this work, numerous studies have been conducted for the initial photofragmentation of Mn2(CO)10 in various solvents [23e26]. Limiting the discussion to the photolysis of Mn2(CO)10 in non-polar solvents, it has been noted that MneMn fission [24] is much more dominant than MneCO cleavage [25] at wavelengths longer than 300 nm. The proposed mechanism will be based on the initial production of Mn(CO)5 radicals from the MneMn fission. The experimental data suggest that the production of H2 can be divided into two photoprocesses; first the generation of the key intermediate MnH(CO)5 from Mn(CO)5 radical and second, the reaction of MnH(CO)5 with H2O. In the previous study on the photoreaction of HCl with Mn2(CO)10, Byers et al. [14] has also suggested that Mn(CO)5 radicals produced upon MneMn dissociation undergo oxidative addition of HCl to give MnH(CO)5 and MnCl(CO)5 as the main products. Compare to HCl, the production of MnH(CO)5 using water has been measured to be only three times slower. Thus it is reasonable to adopt this mechanism, at least to account for MnH(CO)5 formation, by substituting HCl for H2O as illustrated in Fig. 4. Along with MnH(CO)5, a reactive manganese hydroxyl species, Mn(OH)(CO)5 is expected to have been generated. The second step of the photolysis is of vital importance as it has been shown that MnH(CO)5 concentration correlates strongly to the H2 production rate. Similar to the facile formation of Mn(CO)4PR3H complexes [18], MnH(CO)5 undergoes thermal CO dissociation followed by H2O substitution to give MnH(CO)4(H2O). A simple abstraction of the proton from the adjacent H2O ligand may liberate H2 through some as yet unknown pathway leaving a manganese hydroxyl intermediate behind. As mentioned earlier, a manganese hydroxyl species Mn(OH)(CO)5 is generated alongside MnH(CO)5 in the first step. This species is expected to be reactive since the OH ligand has been known to react with the electrophilic CO ligand to give metallocarboxylic acids. These complexes tend to decompose to give CO2 and the corresponding metal hydride [27]. In our experiments, small amounts of CO2 and MnCO3 were indeed detected; the latter is a resultant of carbonic acid formed in the presence of water. The resultant 16-electron metal hydride MnH(CO)4 coordinates to H2O and generates H2 upon photolysis again. Thus by invoking this
4
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
a CO
OC
OC
OC
OC Mn CO
OC Mn CO
CO
H
CO H2O
hv OC
OC
CO
H
Mn
H2O
CO Mn
H
c
CO
HO
H2O
CO
CO
OC hv
CO
H
CO
CO
OH2
CO +
HO
OH2
H2
CO
OC
CO
CO
HO
CO + Mn(CO)5
Mn CO
Mn
Mn
OH
OC
CO
CO
CO
CO Mn
CO
CO
CO2
CO Mn
H
hv
CO
CO OC
OC
+
Mn CO
OC
Mn
CO
CO
CO
CO
CO
CO
CO
OC
CO
CO
b
CO
CO
CO
hv
CO Mn CO
HO
+
H2
CO
Fig. 4. (a) Mechanism showing MneMn bond homolytic fission, leading to an eventual production of MnH(CO)5 and Mn(OH)(CO)5 (adopted from Ref. [14]). (b) Mechanism showing the H2 production from MnH(CO)5 and H2O. (c) Mechanism showing the H2 production from Mn(OH)(CO)5 and H2O.
pathway, it is possible to explain the generation of almost two equivalents of H2 from Mn2(CO)10 photolysis in water. 3. Conclusions Mn2(CO)10 photolysis in an alkane/water biphasic system has afforded 1.80 0.16 mol of hydrogen per mol of Mn2(CO)10. A manganese hydride MnH(CO)5 has been detected as an intermediate in the mixture. The H2 source has indeed been found to be water. The kinetic studies monitored using the IR signatures of Mn2(CO)10 and MnH(CO)5, indicate a correlation between the latter complex with the production of H2. Some of the oxygen atoms of water have been incorporated into a white solid assigned to MnCO3. A mechanism accounting for the formation of H2 via MnH(CO)5 has been proposed. We hope that this study is still potentially relevant to the development of catalytic schemes for H2 generation. 4. Experimental details 4.1. Materials and measurements All chemicals were purchased from SigmaeAldrich and used without further purification, unless otherwise noted. Chemicals used in the quantification were calibrated against their respective primary standards. 1H NMR spectra were recorded in C6D6 on a Bruker ACF300 and AMX500 NMR spectrometer with chemical shifts referenced to residual C6D6 peaks. Solution IR spectra were obtained on a Nexus 870 FT-IR spectrometer using a cell of 0.1 mm path length equipped with CaF2 windows. All photolysis experiments were conducted using a UV broadband lamp (wavelength range: 300e800 nm typical power measured at 10 cm distance: 0.80e0.90 W/cm2) at 10 cm distance away from the reaction mixture in borosilicate glass apparatus. Powder X-ray diffraction (XRD) measurement of solids was carried out using Siemens Powder XRD D5005 diffractometer.
4.2. Mass spectrometric determination of H2 and CO2 H2 gas was detected using a modified residual gas analyser (Pfeiffer Vacuum PrismaPlus QMG 200 with a high-sensitivity ion source) where the signals at m/e 1e50 were monitored over time. Prior to each analysis, a sample of pure H2 gas of known pressure (5e60 Torr) and air were injected for calibration and baseline correction respectively. For each analysis, 3.0 cm3 of the gas was injected directly using a syringe into a pre-evacuated (<1 106 Torr) 20 cm3 stainless steel chamber connected to the residual gas analyser. Thereafter, repeated 1 cm3 portions of gas samples were then introduced by means of two Swagelok ball valves to ensure that repeated sampling of the gas would not have significant depreciation of the initial pressure of gas. Subsequently, the amount of H2 present in the photolytic mixture was calculated while taking into account volume correction factors. Similarly, the amount of CO2 produced was determined by monitoring the signal at m/e 44. Pure CO2 from a cylinder was used to prepare standards of known pressure (20e60 Torr) and air were injected for calibration and baseline correction respectively.
4.3. Photolysis of Mn2(CO)10 in cyclohexane/water mixture A mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3) was photolysed in an evacuated glass apparatus (25 cm3 volume) for 2 h before the reaction mixture was sampled for IR and 1H NMR. The main metal complex intermediate was identified as MnH(CO)5 (n(CO) in cyclohexane: 2015s and 2007w; 1H NMR (C6D6) d 7.9 ppm). Upon complete photolysis, headspace mass-spectrometric studies indicated H2 production to be 1.80 0.16 mol of hydrogen per mol of Mn2(CO)10 and IR spectroscopic studies of the headspace indicated CO2 production to be 0.08 0.03 mol per mol of Mn2(CO)10 used respectively.
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
5
4.4. Photolysis of Mn2(CO)10 under a variety of conditions
4.7. Photolysis of MnH(CO)5 in cyclohexane/water
Typically, a 5-h photolysis (unless otherwise stated) was carried out for Mn2(CO)10 (0.0045 g, 1.1 105 mol) under various conditions. If anhydrous conditions were required, dried cyclohexane (1.0 cm3) which has been distilled three times from molecular sieves was used as the solvent. For the detection of MnH(CO)4(PPh3), mild heating (40e50 C) for 2 h was carried out in the presence triphenylphosphine PPh3 (104 mol) after MnH(CO)5 has been detected in an initial mixture containing Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3). The main metal complex was identified as MnH(CO)4(PPh3). (n(CO) in cyclohexane: 2062w, 1984m, 1969vs and 1958s; 1H NMR (C6D6) d 6.88 ppm doublet, 2J(PeH): 33 Hz.) For the bromide ion test, the photolysis was carried out for a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and a 1 M aqueous tetra-n-butylammonium bromide solution (0.10 cm3). For the test on the influence of acid, the photolysis was carried out for a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and aqueous tetrafluoroboric acid HBF4 (1.0 cm3) whose pH can be monitored and adjusted from 1 to 6 by varying the acid concentration. For examining the influence of different polar solvents, the photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in a solution of THF (1.0 cm3) and H2O (0.1 cm3) was conducted in an evacuated glass apparatus (25 cm3 volume) using a UV broadband lamp before the reaction mixture was sampled for headspace massspectrometric studies. The same procedures were repeated using acetonitrile. The H2 yield for THF and acetonitrile were at 1.90 0.20 and 1.87 0.20 mol per mol of Mn2(CO)10 respectively. For CO pressurization studies, a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3) was photolysed in a glass apparatus (25 cm3 volume, evacuated and filled with CO pressure of 0.1 bare1.5 bar) for 5 h before the reaction mixture was sampled for IR and 1H NMR spectroscopic studies.
MnH(CO)5 was generated from the photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and water (0.10 cm3) and then distilled into an evacuated glass apparatus (25 cm3 volume) containing cyclohexane (1.0 cm3) and water (0.10 cm3). A 2-h photolysis was carried out for the distillate and the reaction mixture was then sampled for IR spectroscopic and headspace mass-spectrometric studies.
4.5. Deuteration studies A 3-h photolysis was carried out for a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in dried cyclohexane (1.0 cm3) and D2O (0.10 cm3). The reaction mixture was sampled for IR and the product was identified as MnD(CO)5. (n(CO) in cyclohexane: 2015s and 2006w) Upon complete photolysis, D2 was detected at 1.40 0.10 mol per mol of Mn2(CO)10. Similarly, the photolysis was repeated in a mixture of 1 cm3 of d12-cyclohexane and 1 cm3 of H2O and the product was identified as MnH(CO)5. 4.6. Time profile monitoring The photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3) was conducted in an evacuated glass apparatus (25 cm3 volume) using a UV broadband lamp for a series of different time intervals before the reaction mixture was sampled for IR, 1H NMR and headspace massspectroscopic measurements. The relative amount of Mn2(CO)10 was determined by one of its infrared absorption bands at 2045 cm1. The relative amount of MnH(CO)5 was determined by following the difference between the observed absorbance at 2007 cm1 and the corresponding absorbance contributed by Mn2(CO)10. The H2 signal at m/e 2 was sampled by the residual gas analyser at different time intervals.
4.8. Photolysis of MnH(CO)5 in dried cyclohexane MnH(CO)5 was generated from a 2-h photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) and was distilled into an evacuated glass apparatus (25 cm3 volume) containing cyclohexane (1.0 cm3) and molecular sieves (1 g). A 6-h photolysis was carried out for the distillate and the reaction mixture was then sampled for IR spectroscopic and headspace mass-spectrometric studies.
4.9. HeD exchange studies of MnD(CO)5 MnD(CO)5 was distilled from a 2-h photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane/D2O into an evacuated glass apparatus (25 cm3 volume) containing cyclohexane (1.0 cm3) and H2O (0.10 cm3). The resulting mixture was left to stand in the dark for up to 3 h before an IR spectrum was obtained. Subsequently, MnD(CO)5 was distilled into a mixture of cyclohexane (1.0 cm3) and D2O (0.10 cm3). MnD(CO)5 was photolysed with cyclohexane/H2O before the head-space was sampled for headspace mass-spectrometric studies.
4.10. Attempted thermal activation of H2O using Mn2(CO)10 or MnH(CO)5 Mn2(CO)10 (0.0045 g, 1.1 105 mol) was heated at 100 C in a mixture of dodecane (1.0 cm3) and H2O (0.10 cm3) mixture for 3 days before the dodecane layer was sampled for IR spectroscopic and mass-spectrometric studies. Similarly, MnH(CO)5, synthesized from a 2-h photolysis of Mn2(CO)10 (0.0045 g, 1.1 105 mol), was heated in a dodecane/water mixture at 100 C for 3 days. The reaction mixture was then sampled for IR spectroscopic and headspace mass-spectrometric studies.
4.11. Chemical analysis of solid residue A 5-h photolysis was carried out for a mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in cyclohexane (1.0 cm3) and H2O (0.10 cm3) and a white precipitate was observed. Once Mn2(CO)10 has reacted completely, the solvent and any residual MnH(CO)5 were removed under vacuum, before the residue was washed further with cyclohexane under reduced air conditions. The solid is dissolved in water and titrated with 0.1 M HCl solution with methyl orange as the indicator. Powder X-ray diffraction (XRD) measurement of the solid was carried out using Siemens Powder XRD D5005 diffractometer.
4.12. Photolysis of Mn2(CO)10 in acetic acid A mixture of Mn2(CO)10 (0.0045 g, 1.1 105 mol) in acetic acid (1.0 cm3) was photolysed in an evacuated glass apparatus (25 cm3 volume) for 5 h before the reaction mixture was sampled for IR.
6
J.W. Kee et al. / Journal of Organometallic Chemistry 724 (2013) 1e6
Acknowledgements The project was supported by a research grant provided by NUS 143-000-430-112. J.W.K., C.K.T. and Y.Y.C. acknowledge the support of NUS scholarships. References [1] P. Du, K. Knowles, R. Eisenberg, J. Am. Chem. Soc. 130 (2008) 12576e12577. [2] A.J. Esswein, D.G. Nocera, Chem. Rev. 107 (2007) 4022e4047. [3] A. Fihri, V. Artero, M. Razavet, C. Baffert, W. Leibl, M. Fontecave, Angew. Chem. Int. Ed. 47 (2008) 564e567. [4] X. Hu, B.S. Brunschwig, C.J. Peters, J. Am. Chem. Soc. 129 (2007) 8988e8998. [5] C. Tard, X. Liu, S.K. Ibrahim, M. Bruschi, L.D. Gioia, S.C. Davies, X. Yang, L. Wang, G. Sawers, C.J. Pickett, Nature 433 (2005) 610e613. [6] D. Streich, M. Karnahl, Y. Astuti, C.W. Cady, L. Hammarström, R. Lomoth, S. Ott, Eur. J. Inorg. Chem. 7 (2011) 1106e1111. [7] D.J. Evans, C.J. Pickett, Chem. Soc. Rev. 32 (2003) 268e275. [8] G.A.N. Felton, A.K. Vannucci, J. Chen, L.T. Lockett, N. Okumura, B.J. Petro, U.I. Zakai, D.H. Evans, R.S. Glass, D.L. Lichtenberger, J. Am. Chem. Soc. 129 (2007) 12521e12530. [9] F. Gärtner, B. Sundararaju, A. Surkus, A. Boddien, B. Loges, H. Junge, P.H. Dixneuf, M. Beller, Angew. Chem. Int. Ed. 48 (2009) 9962e9965. [10] S.W. Kohl, L. Weiner, L. Schwartsburd, L. Konstantinovski, L.J.W. Shimon, Y. Ben-David, M.A. Iron, D. Milstein, Science 324 (2009) 74e77. [11] H.I. Karunadasa, C.J. Chang, J.R. Long, Nature 464 (2010) 1329e1333. [12] R. Brimblecombe, A. Koo, G.C. Dismukes, G.F. Swiegers, L. Spiccia, J. Am. Chem. Soc. 132 (2010) 2892e2894. [13] O.V. Ozerov, Chem. Soc. Rev. 38 (2009) 83e88. [14] B.H. Byers, T.P. Curran, M.J. Thompson, L.J. Sauer, Organometallics 2 (1983) 459e460. [15] J.W. Kee, Y.Y. Tan, B.H.G. Swennenhuis, A.A. Bengali, W.Y. Fan, Organometallics 30 (2011) 2154e2159.
[16] P.S. Braterman, R.W. Harrill, H.D. Kaesz, J. Am. Chem. Soc. 89 (1967) 2851e2855. [17] Z. Mahmood, M. Azam, A. Mushtaq, R. Kausar, S. Kausar, S.R. Gilani, Spectrochim. Acta A 65 (2006) 445e452. [18] (a) B.H. Byers, T.L. Brown, J. Organomet. Chem. 127 (1977) 181e192; (b) W. Hieber, H. Duchatsch, Chem. Ber. 98 (1965) 2933e2938; (c) W.Hieber, G.Faulhaber, F.Theubert, Z. Anorg. Allg. Chem. 314(1962) 125e143. [19] X. Wang, Y.D. Li, Mater. Chem. Phys. 82 (2003) 419e422. [20] C.H.B. Ng, W.Y. Fan, J. Phys. Chem. C 111 (2007) 9166e9171. [21] C.H.B. Ng, W.Y. Fan, J. Phys. Chem. B 110 (2006) 20801e20807. [22] C. Elschenbroich, Organometallics, third ed., Wiley-VCH, Weinheim, Germany, 2006. [23] M.S. Wrighton, Chem. Rev. 74 (1974) 401e430. [24] (a) D.A. Steinhurst, A.P. Baronavaski, J.C. Owrutsky, Chem. Phys. Lett. 361 (2002) 513e519; (b) J.L. Hughey, C.P. Anderson, T.J. Meyer, J. Organomet. Chem. 125 (1977) C49eC52; (c) R.W. Wegman, R.J. Olsen, O.R. Gard, L.R. Faulkner, T.L. Brown, J. Am. Chem. Soc. 103 (1981) 6089e6092. [25] (a) R.S. Herrick, T.L. Brown, Inorg. Chem. 23 (1984) 4550e4553; (b) S.P. Church, H. Hermann, F.W. Grevels, K. Schaffner, J. Chem. Soc. Chem. Commun. 12 (1984) 785e786; (c) L.J. Rothberg, N.J. Cooper, K.S. Peters, V. Vaida, J. Am. Chem. Soc. 104 (1982) 3536e3537; (d) H. Yesaka, T. Kobayashi, K. Yasufuku, S. Nagakuru, J. Am. Chem. Soc. 105 (1983) 6249e6252; (e) A.F. Hepp, M.S. Wrighton, J. Am. Chem. Soc. 105 (1983) 5934e5935. [26] (a) A. Hudson, M.F. Lappert, J.J. Macquitty, B.K. Nicholson, H. Zainal, G.R. Luckhurst, C. Zannoni, S.W. Bratt, M.C.R. Symons, J. Organomet. Chem. 110 (1976) C5eC8; (b) A. Hudson, M.F. Lappert, B.K. Nicholson, J. Chem. Soc. Dalton Trans. (1977) 551e554. [27] (a) D.H. Gibson, T.S. Ong, Organometallics 3 (1984) 1911e1913; (b) N. Grice, S.C. Kao, R. Pettit, J. Am. Chem. Soc. 101 (1979) 1627e1628.