摘要:

DALTON FULL PAPER J. Chem. Soc. Dalton Trans. 1999 1491–1497 1491 Syntheses of highly functionalized cube-octameric polyhedral oligosilsesquioxanes (R8Si8O12) Frank J. Feher,* Kevin D. Wyndham Daravong Soulivong and Frank Nguyen Department of Chemistry University of California Irvine CA 92697-2025 USA Received 18th September 1998 Accepted 1st March 1999 Reactions of [H2N(CH2)3]8Si8O12 or its octahydrochloride salt with a variety of electrophiles including anhydrides lactones acid chlorides a,b-unsaturated esters and isocyanates aVorded functionalized R8Si8O12 frameworks in good to excellent yields. Practical methods for the synthesis of [HO(CH2)3]8Si8O12 [OCN(CH2)3]8Si8O12 and [(Ph2PCH2)2N(CH2)3]8Si8O12 are also reported. Polyhedral oligosilsesquioxanes (POSS) are an interesting class of organosilicon oligomers that can be synthesized by the hydrolytic condensation of trifunctional organosilicon monomers.1,2 This family of compounds has been known for more than 50 years,3 but until recently the majority of known POSS lacked suYcient functionality for most chemical applications.4 The pool of known POSS frameworks has expanded rapidly over the past several years. Some of this expansion is due to the discovery of new spontaneous self-assembly reactions that provide ready access to multigram quantities of several synthetically versatile POSS frameworks.5,6 Another important reason is the development of general and highly eYcient methodology for synthetically manipulating pendant groups on POSS frameworks.6–9 One of the most useful precursors to highly functionalized cube-octameric POSS frameworks is H8Si8O12 1 which is easily prepared from HSiCl3.10–12 This highly versatile framework can easily be chlorinated,13 oxidized to spherosilicates 14,15 or treated with olefins to produce a variety of hydrosilylation products.8,16–19 Unfortunately hydrosilylation of 1 with a-olefins often produces a complex mixture of inseparable products because addition of Si–H to the double bond tends to form both a and b isomers unless the olefin is 3,3-disubstituted.18,20 As part of a general eVort to develop practical routes to pure R8Si8O12 frameworks with synthetically useful functional groups we have been examining the chemistry of (H2NCH2CH2CH2)8Si8O12 2,20 which can easily be prepared in multigram quantities from an inexpensive organosilicon precursor.6,20,21 Procedures for synthesizing amino acid and peptide derivatives of 2 have appeared recently.22 In this paper we report the syntheses and characterization of many other potentially useful R8Si8O12 frameworks derived from 2.We also report a practical synthesis of (HOCH2CH2CH2)8Si8O12 3. Results and discussion The synthesis of compound 2 was first claimed in a 1991 US patent issued to Wacker-Chemie.6 This amine which is actually obtained as its octahydrochloride salt (i.e. 2?8HCl) under the conditions described in the patent is a versatile precursor to many families of functionalized R8Si8O12 frameworks. The hydrochloride salt of 2 is prepared in one step (35% yield) via the hydrolytic condensation of readily available g-aminopropyltriethoxysilane (MeOH–concentrated HCl 25 8C 5 weeks).20 It is highly soluble in water (>0.9 g mL21) slightly soluble in methanol (0.03 g mL21) and DMSO sparingly soluble in DMF and triethylamine and poorly soluble or insoluble in most other organic solvents including pyridine.It is also somewhat hygroscopic. The neutralization of 2?8HCl to the free amine is diYcult to accomplish without compromising the Si/O framework. Some of the diYculties stem from the susceptibility of silsesquioxanes to base-catalysed polymerization and redistribution reactions,23 but a far greater problem is the inherent instability of the free amine itself. Neutralization of 2?8HCl to 2 is best accomplished by eluting methanol or 14 1 ethanol–water solutions of the hydrochloride salt across a column of Amberlite IRA-400 resin. Amine 2 is marginally stable in solution and appears to decompose rapidly when the solvent is removed.Small samples of 2 can be prepared by rapidly evaporating aliquots from the stock solution but to avoid decomposition the amine should be prepared immediately before use or stored in MeOH solutions at 235 8C. Amine 2 is stable in MeOH (without molecular sieves) at 235 8C and concentrations less than 30 mg mL21 for several months. The shelf-life decreases to a few weeks at 210 8C and several hours at 25 8C. At concentrations greater than 30 mg mL21 in 1 4 MeOH–DME a white resinous precipitate forms within a couple of hours. Analysis of this solid by 13C and 29Si NMR spectroscopy (CD3CO2D) indicates >50% decomposition of the Si8O12 framework. Solutions of 2 in dry DMSO (over molecular sieves) can be prepared by neutralizing 2?8HCl in DMSO or by evaporating MeOH from a solution of 2 in MeOH–DMSO (0 8C 0.3 Torr 2 h).We suspect that there are at least two pathways for decomposition. The most rapid probably involves the formation of hydroxide via reaction of water with the free amine. Support for this pathway comes from the observation that other silsesquioxane frameworks are slowly decomposed by exposure to amine bases (e.g. Et3N) in wet solvents and the fact that decomposition is much slower in anhydrous solvents (e.g. MeOH over 3 Å sieves). However the fact that decomposition still occurs when solutions of compound 2 are prepared in DMSO and stored over molecular sieves is consistent with a second mechanism that does not require water. This mechanism might involve attack of the amine nitrogen on Si as illustrated in Scheme 1.1492 J. Chem. Soc. Dalton Trans. 1999 1491–1497 Framework degradation is also a concern whenever 2?8HCl is dissolved in water. Although 2?8HCl is very stable in neutral or acid solutions small amounts of base can produce free aminopropyl groups which appear to catalyse framework cleavage when water is the solvent. In fact a 29Si NMR sample of 2?8HCl prepared in alkaline D2O (pH 9) exhibits a prominent new set of resonances within 10 min of preparation. These resonances which grow at the expense of the resonance for 2?8HCl eventually represent approximately 30% of the total integrated intensity after 45 min at 25 8C. Attempts to achieve higher conversion led to decomposition. We have not been able to isolate this compound in pure form because it appears to decompose or revert back to 2?8HCl under many conditions.However it is believed to be disilanol 4 because of its spectroscopic similarity to 5.24 Of particular spectroscopic relevance is the 29Si NMR spectrum which exhibits 6 resonances with relative integrated intensities of 1:1:1:1:2:2 two of which have chemical shifts characteristic of RSi(OH) groups in an Si4O4 ring (d 256.0 and 256.9). Selective cleavage of a single Si–O–Si linkage in a R8Si8O12 framework is now well established,25–27 but to the best of our knowledge this transformation represents the first time selective cleavage has been observed in aqueous media. The simultaneous synthetic manipulation of many pendant groups requires remarkable eYciency if high yields of a single polyfunctional product are to be obtained.For a transformation of a pure octafunctional starting material (e.g. X8Si8O12) into a pure octafunctional product (e.g. Y8Si8O12) eight sequential chemical reactions must proceed with high conversion and without side reactions. The overall yield of Y8Si8O12 is only 90% if the yield for each reaction is 98.7% or if as little as 1.3% of all X groups fail to react. The yield of Y8Si8O12 quickly falls to 66% if conversion of X into Y is 95%. Since many Y8Si8O12 compounds cocrystallize (or coprecipitate) with derivatives containing fewer Y groups (e.g. Y7XSi8O12 or Y6X2Si8O12) the isolation of pure octafunctional products can be very diYcult if the reactions are not clean or do not proceed to completion. In spite of these obstacles a wide variety of pure R8Si8O12 frameworks can be prepared by treating 2 or its hydrochloride salt with electrophilic reagents.Provided that reactions are performed under conditions where framework degradation is avoided and competing side reactions do not occur product yields are normally quantitative by 1H and 13C NMR spectro- Scheme 1 scopy and all products can be confidently assigned as pure R8Si8O12 compounds on the basis of their NMR and IR spectra combustion analysis and/or mass spectrum. Table 1 summarizes many of our results. The reactions of compound 2 with carboxylic acid derivatives provide access to a wide range of useful compounds ranging from simple amides to R8Si8O12 frameworks possessing eight peptide or carbohydrate residues. Simple amides can be prepared by treating 2?8HCl with an acyl chloride in the presence of a tertiary amine [e.g.Et3N pyridine (i-Pr)2NEt]. For example the reaction with benzoyl chloride (Table 1 entry 1) in DMF with (i-Pr)2NEt aVords 6. Amides can also be prepared by treating 2 with anhydrides such as succinic anhydride (entry 2) and maleic anhydride (entry 3). When performed in MeOH with a generous excess of the anhydride excellent yields of pure octafunctional carboxylic acids are obtained. The reaction of 2 with maleic anhydride is particularly interesting because the product spontaneously precipitates from solution without coprecipitating partially functionalized derivatives. Octaamine 2 reacts with lactones to aVord a variety of useful compounds. These reactions are relatively slow and generally require a generous excess of lactone to achieve the high conversions necessary for octafunctionalization.In the case of caprolactone (entry 4) acylation of 2 can be accomplished by heating 2 and an excess of lactone under vacuum at 70 8C for 3 to 4 h. In the case of carbohydrate lactones (entries 5–7) acylation should be performed at 25 8C for 1 to 3 d. For all of these reactions it is desirable to use 2 rather than its hydrochloride salt in combination with Na2CO3 or a tertiary amine. This limits the amount of base present in the reaction mixture and greatly reduces the formation of ill defined silsesquioxane resins derived from base-catalysed cleavage and polymerization of the Si8O12 framework. The NMR spectra of compounds 10 11 and 12 exhibit complex concentration-dependent behavior that is consistent with two distinct environments for the pendant groups.The reasons for this behavior are not known with certainty but the two environments appear to be due to restricted rotation about the amide C–N bonds (i.e. cis/trans isomerization) and strong inter- and intra-molecular interactions between pendant carbohydrate groups (i.e. aggregation). This behavior as well as a preliminary study of lectin binding by 11 and 12 is discussed in ref. 28. Octaamine 2 reacts readily with isocyanates to aVord octafunctional ureas. For example the reactions of 2 with allyl isocyanate (entry 8) and n-butyl isocyanate (entry 9) produce 13 and 14 respectively. Freshly prepared solutions of 13 and 14 in DMSO (40 mg mL21) form gels that resist crystallization for more than a month but crystalline samples can be obtained quickly by recrystallization from hot alcohol.The reaction of compound 2 with phosgene Cl2CO is particularly noteworthy because it provides access to an octaisocyanate (i.e. 15) with many interesting possibilities as a precursor to both hybrid inorganic–organic materials 29,30 and more elaborate POSS frameworks. The synthesis of 15 can easily be accomplished on a small scale by adding saturated aqueous NaHCO3 to a biphasic reaction mixture containing phosgene in CH2Cl2 and 2?8HCl in water. Separation of the organic layer after 1 h filtration through cotton (to dry) and evaporation of the solvent aVords 15 as a colorless oil. On one occasion the yield of 15 was 80% but all other times it was considerably lower (11 to 47%). The product obtained in this fashion is spectroscopically pure and can be used to prepare other frameworks such [n-BuNHCONH(CH2)3]8Si8O12 14 and [t-BuNHCONH- (CH2)3]8Si8O12 18.Isocyanate 15 appears to be indefinitely stable at low temperatures (235 8C) and stable at elevated temperatures (25–80 8C) for short periods but it quickly produces insoluble polyureas upon exposure to traces of water. It also appears to polymerize on standing at room temperature. J. Chem. Soc. Dalton Trans. 1999 1491–1497 1493 Table 1 Reactions of compound 2 with electrophilic reagents (E) [H2N(CH2)3]8Si8O12 1 E æÆ [R1R2N(CH2)3]8Si8O12 Entry 12345 6 7 89 10 11 12 E (equivalents) a ClCOC6H5 (12) c Succinic anhydride (20) Maleic anhydride (28) e-Caprolactone (80) d-Gluconolactone (32) d-Lactonolactone (14) d-Maltonolactone (15) H2C]] CHCH2NCO (11) CH3(CH2)3NCO (11) Cl2CO (27) H2C]] CHCO2CH3 (306) CH2O/(C6H5)2PH (24) Reaction conditions (i-Pr)2NEt DMF 0–25 8C 10 h CH3OH 25 8C 10 h CH3OH 25 8C 10 h Neat 70 8C 3.4 h DMSO 25 8C 72 h DMSO 25 8C 24 h DMSO 25 8C 24 h DMSO 25 8C 10 h DMSO 25 8C 3 h NaHCO3 (aq) CH2Cl2 0 8C 1 h CH3OH 25 8C 24 h CH3OH toluene 65 8C 14 h Product 6789 10 11 12 13 14 15 16 17 R1 HHHHH H H HH CO R2 R2 R2 COC6H5 COCH2CH2CO2H cis-COCH]] CHCO2H CO(CH2)5OH CONHCH2C(H)]] CH2 CONH(CH2)3CH3 N.A.CH2CH2CO2CH3 CH2P(C6H5)2 HO O CH2OH OH OH HO O OH CH2OH OH HO HO O CH2OH O OH HO O OH CH2OH OH HO HO O CH2OH O OH HO Yield (%) b 49 58 64 23 30 53 26 90 65 11–47 73 37 a Equivalents of E per Si8O12 framework. b Isolated yield for reaction that proceeded with complete conversion of (CH2)3NH2 to (CH2)3NR1R2 as judged by 1H and 13C NMR spectroscopy.c Compound 2?8HCl was used as the starting material. Frameworks with sixteen equivalent pendant groups can be prepared by replacing all amine protons of compound 2 with an appropriate electrophile. For example the reaction of 2 with an excess of methyl acrylate produces 16 20 under conditions typically used to prepare polyamidoamino (PAMAM) dendrimers from polyamine cores (entry 11).31,32 Similarly an interesting phosphine-substituted framework with sixteen Ph2P groups can be prepared by adapting known methodology for attaching phosphine pendant groups to amine-terminated PAMAM dendrimers.33 The reaction of 2 with an excess of CH2O/Ph2PH (Table 1 entry 12) aVords 17 while the same reaction performed using 2?8HCl and excess of Et3N appears to produce a hydrochloride salt of 17. Synthesis of [HO(CH2)3]8Si8O12 3 Amine 2 and its hydrochloride salt are excellent precursors to a wide range of functionalized silsesquioxanes but there are many applications where both the basicity of an amine or amide nitrogen atom and the ability of a primary amine to react with more than one equivalent of electrophile are undesirable.In these cases an R8Si8O12 synthon with eight hydroxypropyl groups might be useful. We have therefore devised a method for preparing compound 3. The immediate precursor to compound 3 is nitrate ester 22 which is readily available in three steps from inexpensive Cl(CH2)3SiCl3. Hydrolytic condensation of Cl(CH2)3SiCl3 aVords 20 in 25–40% yield,6,8 which upon reaction with NaI aVords 21.8 Hydrolysis of 21 is very diYcult to achieve without destroying the Si8O12 framework.However first treating 21 with AgNO3 in MeCN to produce 22 then hydrogenolysis of 22 with 10% Pd/C in ethyl acetate–MeOH (800 psig H2 25 8C 3 d) aVords 3 as a waxy white solid in 85% overall yield. A related series of [RSiO3/2]n frameworks possessing hydroxyalkyl substituents can also be prepared via hydrosilylation reactions of [HSiO3/2]n with a,w-trimethylsiloxyalkenes,34 but product yields are typically lower (8–10%) and it is likely that addition of Si–H to the double bonds produces both a and b isomers. Concluding remarks We have identified practical routes to a wide variety of functionalized R8Si8O12 frameworks including [HO(CH2)3]8Si8O12 3 [OCN(CH2)3]8Si8O12 15 [(Ph2PCH2)2N(CH2)3]8Si8O12 17 and many compounds derived from reactions of [H2N(CH2)3]8Si8O12 2 with electrophilic reagents.Provided that reactions are performed under conditions where framework degradation is avoided and competing side reactions do not occur good to excellent yields of many pure R8Si8O12 compounds can be obtained. Experimental General experimental protocol and procedures have been 1494 J. Chem. Soc. Dalton Trans. 1999 1491–1497 reported earlier.24 Unless otherwise noted all manipulations were performed under an atmosphere of dry nitrogen using either standard Schlenk techniques or a Vacuum Atmospheres Corp. Dri-Lab. Unless otherwise noted reagent-grade chemicals were used without further purification for all work described here. Dimethyl sulfoxide (DMSO) was dried over activated 4 Å molecular sieves prior to use. Dimethylformamide (DMF) was distilled and stored under an atmosphere of dry nitrogen.Methanol was freshly distilled from magnesium turnings. 10% Pd/C (Aldrich) was activated under vacuum (200 8C 0.001 Torr 12 h) and used under an atmosphere of dry nitrogen. Phosgene was obtained as a 1.89 M solution in toluene (Fluka). Methyl acrylate (Aldrich) was distilled from CaH2 under nitrogen immediately before use. g-Aminopropyltriethoxysilane was obtained as a generous gift from Witco/OSi specialties. The NMR spectra were recorded on General Electric GN- 500 (1H 500.03; 13C 125.75; 31P 202.5; 29Si 99.37) Omega-500 (1H 500.22; 13C 125.79; 29Si 99.34) or Bruker DRX-500 (1H 500.03; 13C 125.74; 29Si 99.30 MHz) spectrometers infrared spectra on a Perkin-Elmer 1600 Series FTIR spectrometer mass spectra on a VG Analytical Autospec E double-focussing spectrometer or a Perceptive Biosystems matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) Voyager DE STR spectrometer.Combustion analyses (C H N) were made with a Carlo Erba Instruments Fisons Elemental Analyzer. Melting points were measured with a Laboratory Devices Mel-Temp apparatus and are uncorrected. Syntheses Compound 2?8HCl. Concentrated HCl (200 mL) was added carefully with stirring to a solution of g-aminopropyltriethoxysilane (150 mL 0.641 mol) in MeOH (3.6 L) in a 4.0 L glass bottle. The bottle was capped and allowed to stand for 6 weeks at 25 8C. The product usually begins to crystallize from the reaction mixture after 3–4 weeks but in the event that no crystals form spontaneously crystallization can be induced by agitating or adding seed crystals from a previous reaction.The product obtained in 30% yield (24.9 g) by filtering the reaction mixture washing with cold MeOH and drying (0.001 Torr 25 8C) is spectroscopically pure and suitable for most purposes. An additional 4.2 g (5%) were collected by reducing the filtrate to 1/3 of its original volume. Analytically pure 2?8HCl can be obtained as white microcrystalline powder by recrystallization from hot MeOH. 1H NMR (500.2 MHz DMSO-d6 25 8C) d 8.25 (s NH3 24 H) 2.75 (t CH2N 16 H) 1.71 (m SiCH2CH2 16 H) and 0.71 (t SiCH2 16 H). 13C-{1H} NMR (125.8 MHz DMSO-d6 25 8C) d 41.01 (s CH2N) 20.61 (s SiCH2CH2) and 8.44 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 25 8C) d 266.4 (s). Mass spectrum (MALDI-TOF dihydroxybenzoic acid (DHB) matrix) calc. for C24H72Cl8N8O12Si8 [M 1 H 2 8 HCl]1 m/z 881.29 found 881.4 (100%); [M 2 NH3 2 8 HCl]1 m/z 863.25 found 863.4 (39%) [Found (Calc.for C6H18Cl2- N2O3Si2) C 24.69 (24.57); H 6.40 (6.19); N 9.40 (9.55)%]. mp 277 8C (decomp). Compound 2 via neutralization of 2?8HCl. Amberlite IRA- 400 ion-exchange resin (37 g) was prepared by successive washing with water (4 × 200 mL) 1 M NaOH (3 × 200 mL) water (6 × 200 mL) and the elution solvent which was either MeOH 4:1:1 EtOH–DME–2-propanol or 14 1 EtOH–water (6 × 200 mL); the resin was suspended in eluent and chilled (210 8C 2 h) before use. Half of the resin beads were loaded onto a column (3.5 cm outside diameter); the other half was used to dissolve a suspension of 2?8HCl (6 g) in the minimum amount of eluent at 0 8C. Elution across the column produced a stock solution (15–20 mM 250–300 mL 0 8C) of 2 that tested negative for chloride.Small samples of 2 can be prepared by rapidly evaporating aliquots from the stock solution but to avoid decomposition the amine should be prepared immediately before use or stored in MeOH solutions at 235 8C. 1H NMR (500.2 MHz DMSO-d6 25 8C) d 4.3 (s NH2 and H2O) 2.50 (t J = 7.4 CH2N 16 H) 1.42 (m SiCH2CH2 16 H) and 0.54 (t J = 7.3 Hz SiCH2 16 H). 13C-{1H} NMR (125.8 MHz CD3OD 25 8C) d 44.87 (s CH2N) 26.88 (s SiCH2CH2) and 9.57 (s SiCH2). 29Si-{1H} NMR (99.4 MHz CD3OD 25 8C) d 266.4 (s). Mass spectrum (MALDI-TOF DHB matrix) calc. for C24H64N8O12Si8 [M 1 H]1 m/z 881.29 found 881.5 (100%); [M 2 NH3]1 863.25 found 863.5 (49%). Compound 4 in basic (pH 9) D2O. The NMR spectrum of compound 2?8HCl in basic (pH 9) D2O exhibits prominent resonances assignable to 4 within 20 min of sample preparation.Attempts to isolate pure 4 were not successful but a sample containing ª30% of 4 was obtained by agitating a solution of 2?8HCl (1.0 g 0.853 mmol) in D2O (pH 9 0.5 mL) for 20 min at 25 8C. Addition of CH3CO2H (5 mL) and evaporation (25 8C 10 Torr) produced a hard colorless resin (1.147 g) which was triturated with EtOH (40 mL × 3). The remaining solid was collected by vacuum filtration washed with EtOH and dried (25 8C 0.001 Torr) to aVord 775 mg of solid consisting of a 70 30 mixture of 2?8HCl and 4 (by 13C and 29Si NMR). Samples containing approximately equimolar 2?8HCl and 4 can be prepared by selective crystallization of 2?8HCl from MeOH– EtOH (1 4 volume ratio) or by diVusion of MeCN into a methanol solution of 2?8HCl and 4 but disilanol 4 itself has resisted crystallization.1H NMR (125.8 MHz D2O 25 8C) d 2.95 (t J = 8.0 Hz CH2N) 1.73 (m SiCH2CH2 16 H) and 0.74 (s SiCH2 16 H). 13C-{1H} NMR (125.8 MHz D2O 25 8C) d 41.68–41.48 (m CH2N) 20.74–20.38 (m SiCH2- CH2) 9.75 8.66 8.56 8.46 and 7.80 (s 1:1:2:2:2 for SiCH2 assuming 50% 2?8HCl at d 7.86). 29Si-{1H} NMR (99.4 MHz D2O 25 8C) d 256.0 256.9 265.1 265.3 266.3 266.9 (s 1:1:1:1:2:2 assuming 50% 2?8HCl at d 266.5). Compound 6. Benzoyl chloride (0.11 mL 1.02 mmol) was added dropwise to a suspension of compound 2?8HCl (100 mg 0.085 mmol) and (i-Pr)2NEt (0.30 mL 1.7 mmol) in DMF (4 mL 0 8C). After stirring overnight the crude product was precipitated by dropwise addition to 1 M HCl (aqueous 70 mL 0 8C). Filtration extraction with DMSO (4 mL) precipitation with cold saturated NaHCO3 (70 mL) washing with water and drying in vacuo (25 8C 0.01 Torr) aVorded a white solid (133 mg).Analysis by 1H and 13C NMR spectroscopy indicated quantitative formation of 6. Analytically pure material (71 mg 49%) was prepared by slowly adding DMSO (10 mL) to a solution of 6 in MeCN (100 mL) at 0 8C. 1H NMR (500.2 MHz DMSO-d6 25 8C) d 8.45 (t NH 8 H) 7.78 (d o-CH 16 H) 7.40 (t p-CH 8 H) 7.37 (m m-CH 16 H) 3.18 (m CH2N 16 H) 1.62 (m SiCH2CH2 16 H) and 0.64 (t SiCH2 16 H). 13C-{1H} NMR (125.8 MHz DMSO-d6 25 8C) d 166.27 (s CO) 134.59 130.95 128.15 127.10 (s C6H5) 41.66 (s CH2N) 22.31 (s SiCH2CH2) and 8.79 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 25 8C) d 266.5. Mass spectrum (MALDI-TOF DHB matrix) calc. for C80H96N8O20Si8 [M 1 H]1 m/z 1713.50 found 1713.68 (47%); [M 1 Na]1 1735.48 found 1735.68 (50%).Compound 7. A solution of compound 2 (88 mg 0.100 mmol) in MeOH (4.0 mL) was added to succinic anhydride (200 mg 2.0 mmol). Within 5 min of stirring at ambient temperature the anhydride completely dissolved to aVord a clear solution which was stirred overnight. Dropwise addition of the reaction mixture to water (30 mL 25 8C) and evaporation of MeOH (25 8C 0.01 Torr) aVorded a white solid which was collected by vacuum filtration washed with water dried redissolved in MeOH and evaporated to dryness. The material prepared by this method (97 mg 58%) is spectroscopically pure (by 1H and 13C NMR spectroscopy). An analytically pure sample (25 mg) J. Chem. Soc. Dalton Trans. 1999 1491–1497 1495 was prepared by slowly adding EtOH to a solution of 7 (30 mg) in MeOH (0.5 mL).1H NMR (500.2 MHz DMSO-d6 25 8C) d 7.81 (br NH 8 H) 3.01 (br CH2N 16 H) 2.40 (br CH2 16 H) 2.30 (br CH2 16 H) 1.43 (br SiCH2CH2 16 H) and 0.58 (br SiCH2 16 H). 13C-{1H} NMR (125.8 MHz DMSO-d6 25 8C) d 173.90 (s CO) 170.94 (s CO) 41.01 (s CH2N) 30.00 (s CH2) 29.17 (s CH2) 22.49 (s SiCH2CH2) and 8.75 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 25 8C) d 266.17 (s). IR (KBr) 1701 1638 (nNC]] O) 1103vs cm21 (nSiOSi). Mass spectrum (MALDI-TOF DHB matrix) calc. for C56H96- N8O36Si8; [M 2 H]2 m/z 1679.40 found 1679.2; [M 2 2H 1 Na]2 1701.38 found 1701.2; [M 2 C4H5O3]2 1579.38 found 1579.2 [Found (Calc. for C14H24N2O9Si8) C 40.40 (39.99); H 5.79 (5.75); N 6.70 (6.66)%]. Compound 8. A solution of compound 2 (440 mg 0.499 mmol) and maleic anhydride (1.40 g 0.014 mol) in MeOH (20 mL) was stirred overnight at 25 8C.The white precipitate that formed overnight was collected by filtration washed with cold methanol and dried (25 8C 0.001 Torr) to aVord 8 as a spectroscopically pure white solid (530 mg 64%). An analytically pure sample (159 mg) was prepared by recrystallizing 8 (250 mg) from MeOH (40 mL reflux to 230 8C). 1H NMR (500.2 MHz DMSO-d6 25 8C) d 9.04 (s NH 8 H) 6.37 (d J = 12.5 CH 8 H) 6.22 (d J = 12.5 Hz CH 8 H) 3.15 (t CH2N 16 H) 1.54 (m SiCH2CH2 16 H) and 0.65 (t SiCH2 16 H). 13C-{1H} NMR (128.5 MHz DMSO-d6 25 8C) d 165.47 165.40 (s CO) 132.91 131.64 (s CH) 41.41 (s CH2N) 21.82 (s SiCH2CH2) and 8.61 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 25 8C) d 266.17 (s). IR (KBr) 1720 1707 1634vs (nNC]] O) 1114vs cm21 (nSiOSi).Mass spectrum (MALDI-TOF DHB matrix) calc. for C56H80N8O36Si8; [M 1 Na]1 m/z 1687.27 found 1687.1; [M 2 C4H2O3 1 H]1 1567.29 found 1567.2; [M 2 C4H2O3 1 Na]1 1589.27 found 1589.1; [M 2 H]2 1663.28 found 1663.2; [M 2 2H 1 Na]2 1685.26 found 1685.2; [M 2 2H 1 K]2 1701.23 found 1701.1; [M 2 C4H3O3]2 1565.27 found 1565.3 [Found (Calc. for C14H20N2O9Si2) C 40.54 (40.37); H 4.90 (4. 84); N 6.53 (6.73)%]. Compound 9. e-Caprolactone (6.0 mL 50.6 mmol) was added to a solution of compound 2 (540 mg 0.630 mmol) in MeOH (20 mL). Methanol was promptly removed under vacuum (25 8C 0.02 Torr) then the reaction mixture was heated under vacuum (70 8C 0.02 Torr 3–4 h). The hard yellow non- Si-containing solid produced was removed by filtration. Dropwise addition of the filtrate to CH2Cl2 (100 ml) over 0.5 h precipitated crude 9 which was collected by filtration washed with CH2Cl2 dissolved in MeOH filtered and evaporated (25 8C 0.001 Torr) to aVord a tacky yellow solid (770 mg).Analysis of this solid by 1H and 13C NMR spectroscopy indicated the presence of only 9. Purification was performed by adding a solution of 9 (770 mg) in EtOH (6 mL) dropwise to Et2O (100 mL 0 8C). Centrifugation of the resulting turbid suspension aVorded a yellowish solid which was dissolved in EtOH decolorized with activated charcoal and evaporated (25 8C 0.001 Torr) to aVord a pale yellow product (376 mg). This was dissolved in 20 1 CH2Cl2–MeOH (5 mL) and eluted across a thin pad of silica with 20 1 CH2Cl2–MeOH (80 mL). Evaporation of the solvent (25 8C 0.001 Torr) aVorded a pale yellow solid (262 mg).Samples obtained in this fashion are suitable for most purposes. Analytically pure product was prepared by dropwise addition of MeCN (20 mL 0 8C) to a solution of 9 (83 mg) in MeOH (2 mL); vacuum filtration washing with MeCN and drying (25 8C 0.001 Torr) aVorded 9 as a fine white powder (40 mg). 1H NMR (500.2 MHz MeOH-d4 25 8C) d 8.1 (s NH) 3.54 (t J = 6.6 CH2OH 16 H) 3.15 (t J = 7.0 CH2N 16 H) 2.20 (t J = 7.6 COCH2 16 H) 1.64–1.53 (m CH2 48 H) 1.38 [m CH2(CH2)2OH 16 H] and 0.64 (t J = 8.1 Hz SiCH2 16H). 13C-{1H} NMR (125.8 MHz MeOH-d4 25 8C) d 174.02 (s CO) 60.72 (s CH2OH) 40.74 (s CH2N) 34.94 31.33 24.91 24.56 (s CH2) 21.94 (s SiCH2CH2) and 7.93 (s SiCH2). 29Si-{1H} NMR (99.4 MHz MeOH-d4 25 8C) d 266.4 [Found (Calc. for C72H144N8O28Si8?H2O) C 47.78 (47.71); H 8.21 (8.12); N 5.86 (6.18)%].Compound 10. A solution of compound 2 (500 mg 0.57 mmol) in MeOH (20.0 mL) was added to a solution of d-gluconolactone (3.20 g 18.24 mmol) in dry DMSO (20 mL); the MeOH was immediately removed by stirring under vacuum (25 8C 0.001 Torr). The solution was stirred under nitrogen (25 8C 72 h) filtered and then evaporated (30 8C 0.01 Torr) to aVord a colorless resin which was dialysed against water (25 8C 3 × 4 L) over a period of 24 h. Evaporation (30 8C 0.01 Torr) of the resulting solution aVorded 10 as a white powder. Analysis by 1H and 13C NMR spectroscopy indicated quantitative formation of 10. Final purification was performed by slowly adding an excess of MeOH to an aqueous solution of 10 and cooling to 230 8C; vacuum filtration washing with cold MeOH and drying (60 8C 0.001 Torr 3 d) aVorded 10 as a white powder in 30% yield (400 mg).1H NMR (500.2 MHz 31 mM in DMSO-d6 100 8C) d 7.38 (br NH 8 H) 4.42–3.2 (m carbohydrate 88 H) 3.13 2.80 (br CH2N 16 H) 1.56 1.53 (br SiCH2CH2 16 H) and 0.64 (br SiCH2 16 H). 13C-{1H} NMR (125.8 MHz 31 mM in DMSO-d6 25 8C) d 172.10 (s CO) 79.87 73.10 72.04 70.17 63.10 40.4 (br CH2N) 22.02 (br SiCH2CH2) and 8.34 (s SiCH2). 29Si-{1H} NMR (99.4 MHz 63 mM in DMSO-d6 25 8C) d 266.1 and 266.8. mp 160 8C (decomp.). Compound 11. A solution of compound 2 (216 mg 0.245 mmol) in MeOH (3.3 mL) was added to a solution of O-b-Dgalactopyranosyl-( 1Æ4)-D-glucono-1,5-lactone 35,36 (1.16 g 3.41 mmol) in dry DMSO (ca. 3 mL); the MeOH was immediately removed by stirring under vacuum (25 8C 0.007 Torr). The solution was stirred under nitrogen (25 8C 24 h) filtered and evaporated (30 8C 0.01 Torr) to aVord a colorless resin which was dialysed against water (25 8C 3 × 4 L) over a period of 24 h.Evaporation (30 8C 0.01 Torr) of the resulting solution aVorded 11 as a spectroscopically pure white powder (472 mg 53%). Analytically pure product was obtained by slowly adding an excess of MeOH to an aqueous solution of 11 and cooling to 230 8C; vacuum filtration washing with cold MeOH and drying (60 8C 0.001 Torr) aVorded 11 as a white powder (340 mg). 1H NMR (500.0 MHz 5 mM in DMSO-d6 25 8C) d 7.66 (br NH 8 H) 5.17–3.38 (m carbohydrate 168 H) 3.10 3.04 (br CH2N 16 H) 1.47 (br SiCH2CH2 16 H) and 0.56 (br SiCH2 16 H). 13C-{1H} NMR (125.7 MHz 5 mM in DMSOd6 25 8C) d 172.36 (s CO) 104.66 (s 19-C) 83.16 (s 4-C) 75.69 73.22 72.00 71.70 71.43 71.17 70.55 68.23 (49-C) 62.34 (s 6-C) 60.72 (69-C) 40.89 (br CH2N) 22.57 (br SiCH2- CH2) and 8.77 (br SiCH2).29Si-{1H} NMR (99.4 MHz 5 mM in D2O 25 8C) d 265.9 and 266.9 (20%). Mass spectrum (MALDI-TOF DHB-HIQ matrix) calc. for C120H224N8O100Si8 [M 1 Na]1 m/z 3624.1 found 3623.9; [M 2 C12H19O11 1 K]1 3300.95 found 3302.0; [M 2 C12H19O11 1 Na]1 3284.99 found 3284.0; [M 2 C12H19O11 1 H]1 3261.99 found 3262.1 [Found (Calc. for C120H224N8O100Si8?3H2O) C 39.63 (39.40) H 6.15 (6.34) N 3.05 (3.06)%]. mp 148 8C (decomp.). Compound 12. Amine 2 (65 mg 0.075 mmol) was treated with O-a-D-glucopyranosyl-(1Æ4)-D-glucono-1,5-lactone 35,36 (370 mg 1.088 mmol) as described above for the preparation of 11. Isolation and purification as described above aVorded 12 in 26% yield (70 mg) as an analytically pure white powder.1H NMR (500.0 MHz 5 mM in D2O 25 8C) d 7.98 (br NH 8 H) 4.95–3.30 (m carbohydrate 128 H) 3.13 (br CH2N 16 H) 1.52 (br SiCH2CH2 16 H) and 0.59 (br SiCH2 16 H). 13C-{1H} NMR (125.7 MHz 5 mM in D2O 25 8C) d 174.66 (s CO) 101.19 (s 19-C) 82.70 (s 4-C) 73.58 73.19 73.06 72.59 72.50 72.38 62.81 (s 6-C) 61.02 (69-C) 42.1 (br CH2N) 22.65 (br, 1496 J. Chem. Soc. Dalton Trans. 1999 1491–1497 SiCH2CH2) and 8.80 (s SiCH2). 29Si-{1H} NMR (99.4 MHz 5 mM in D2O 25 8C) d 265.9 266.9 (20%). Mass spectrum (MALDI-TOF DHB matrix) calc. for C30H56N2O25Si2; [M 1 K]1 m/z 3641.1 found 3642.8; [M 1 Na]1 3624.1 found 3623.9; [M 2 C12H19O11 1 K]1 3300.95 found 3299.8; [M 2 C12H19O11 1 Na]1 3284.99 found 3283.9 [Found (Calc. for C120H224N8O100Si8?4H2O) C 39.29 (39.21); H 6.28 (6.36); N 2.66 (3.05)%].Compounds 13 and 14. In a typical reaction the isocyanate (11 equivalents) was added to a solution of compound 2 (800 mg 0.908 mmol) in MeOCH2CH2OMe–EtOH–2-propanol (4:1:1 25 mL); the solution volume was reduced by a factor of two by evaporation (30 8C 10 Torr). After 3–10 h the white solid formed by the reaction was collected washed with cold CH2Cl2 then dried (25 8C 0.001 Torr). Analysis of the crude product by NMR (1H 13C and 29Si) spectroscopy indicated complete conversion of aminopropyl groups into urea functionalities and product purities greater than 95%. Pure 13 was prepared by recrystallizing the crude product from hot MeOH. Yield 1.27 g (90%). 1H NMR (500.2 MHz DMSO-d6 60 8C) d 5.86–5.79 (m CH2NHCONHCH2CH 24 H) 5.10 (dd J = 17.2 1.6 CH 8 H) 5.00 (d J = 10.3 Hz CH 8 H) 3.62 (m CH2CH 16 H) 2.98 (m CH2N 16 H) 1.45 (m SiCH2CH2 16 H) and 0.59 (t SiCH2 16 H).13C-{1H} NMR (125.8 MHz DMSO-d6 60 8C) d 157.81 (s CO) 136.56 (s CH) 113.99 (s CH2) 41.53 (s CH2N) 41.45 (s CH2N) 23.09 (s SiCH2CH2) and 8.37 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 60 8C) d 266.03 (s). IR (KBr) 1628 1583vs (nNCO) 1116vs cm21 (nSiOSi) [Found (Calc. for C56H104N16O20Si8?EtOH) C 43.58 (43.75); H 6.76 (6.96); N 14.07 (14.07)%]. mp 220 8C (decomp.). Analytically pure 14 was obtained by recrystallization from CHCl3–MeOH (1 1 230 8C) followed by recrystallization from hot EtOH. Yield 891 mg (65%). 1H NMR (500.2 MHz DMSO-d6 80 8C) d 5.70 (t NH 8 H) 5.62 [t (CH2)3NH 8 H] 2.99 (m CH2N 32 H) 1.46 (m SiCH2CH2 16 H) 1.36 (m CH2CH2CH3 16 H) 1.26 (m CH2CH2CH3 16 H) 0.87 (t CH2CH2CH3 24 H) and 0.59 (t SiCH2 16 H).13C-{1H} NMR (125.8 MHz DMSO-d6 80 8C) d 157.92 (s CO) 41.32 (s CH2N) 38.68 (s CH2N) 31.74 (s CH2) 22.97 (s SiCH2CH2) 19.04 (s CH2) 13.07 (s CH3) and 8.26 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 80 8C) d 266.01 (s). IR (KBr) 1628 1577vs (nNC]] O) 1124vs cm21 (nSiOSi) [Found (Calc. for C16H34N4O5Si2) C 45.93 (45.90); H 8.29 (8.19); N 13.05 (13.38)%]. mp 161 8C (decomp.). Compound 15. Following the general procedure described by Nowick et al.,37 saturated NaHCO3 (10 mL) was added dropwise at 0 8C with vigorous stirring to a biphasic mixture of phosgene (3 mL of 1.89 M toluene solution) in CH2Cl2 (10 mL) and compound 2?8HCl (239 mg 0.210 mmol) in water (4 mL). The pH was monitored and the reaction terminated when the aqueous layer became acidic (ca.1 h at 0 8C). The organic layer was separated dried by filtration through a wad of cotton and then evaporated (25 8C 0.001 Torr) to aVord 15 as a colorless liquid spectroscopically pure by 1H 13C and 29Si NMR spectroscopy. 1H NMR (500.2 MHz CDCl3 25 8C) d 3.33 (t CH2N 16H) 1.73 (m SiCH2CH2 16H) and 0.77 (t SiCH2 16H). 13C-{1H} NMR (125.8 MHz CDCl3 25 8C) d 122.05 (s CO) 45.05 (s CH2N) 24.88 (s SiCH2CH2) and 8.87 (s SiCH2). 29Si-{1H} NMR (99.4 MHz CDCl3 25 8C) d 267.02 (s). IR (KBr) 2274vs (nNC]] O) and 1110vs cm21 (nSiOSi) [Found (Calc. for polymerized product calculated for the formula C8H12N2O5Si2) C 35.13 (35.28); H 4.90 (4.44); N 10.17 (10.29)%]. Compound 16. Following the standard procedure for the synthesis of PAMAM dendrimers,31,32 methyl acrylate (0.8 mL 8.884 mmol) was added via syringe to a solution of freshly prepared compound 2 (65 mg 0.029 mmol) in methanol (1 mL) under N2(g).After 24 h at 25 8C analysis by 1H 13C and 29Si NMR spectroscopy indicated complete reaction of aminopropyl groups. Evaporation of the volatiles (25 8C 0.001 Torr) produced a colorless glassy solid which was triturated with dry methanol and dried (25 8C 0.001 Torr). All attempts to recrystallize or precipitate 16 as a powder were unsuccessful but samples obtained in this fashion are spectroscopically pure and can be treated with ethylenediamine to produce amineterminated PAMAM dendrimers.20 Yield 204 mg (73%). 1H NMR (500.2 MHz CD3OD 25 8C) d 3.61 (s CH3 48 H) 2.71 (t CH2CO 32 H) 2.40 (m CH2NCH2 48 H) 1.51 (m SiCH2- CH2 16 H) and 0.593 (t SiCH2 16 H).13C-{1H} NMR (125.8 MHz CD3OD 25 8C) d 174.53 (s CO) 57.23 [s CH2N(CH2)2] 52.13 (s CH2CO2) 50.34 (s CH3) 33.27 [s CH2N(CH2)2] 21.48 (s SiCH2CH2) and 10.21 (s SiCH2). 29Si-{1H} NMR (99.4 MHz CD3OD 25 8C) d 266.0 (s). Mass spectrum (MALDITOF DHB matrix) calc. for C88H160N8O44Si8 [M 1 H]1 m/z 2257.88 found 2256.5 (47%); [M 1 K]1 2295.83 found 2295.9 (62%). Compound 17. Following the general procedure described by Reetz et al.,33 H2CO (37% 2 mL 26.7 mmol) was added to a solution of HP(C6H5)2 (497 mg 2.67 mmol) in degassed MeOH (5 mL). This solution was stirred under nitrogen at 25 8C for 10 min heated to 65 8C for 10 min then evaporated to dryness (25 8C 0.001 Torr). To the resulting residue was added a solution of compound 2 (97.5 mg 0.111 mmol) in degassed 45% MeOH–toluene (7 mL).After heating at 65 8C for 14 h the reaction mixture was evaporated (25 8C 0.001 Torr) to aVord a white solid. Purification was performed by thrice adding a solution of 17 in THF (1 mL) dropwise to MeOH (10 mL). A final washing with MeOH and drying in vacuo (25 8C 0.001 Torr) aVorded 17 as a fine white powder in 37% yield (165 mg). 1H NMR (500.2 MHz CDCl3 25 8C) d 7.37–7.21 (br C6H5 160 H) 3.50 (br CH2 32 H) 2.77 (br CH2N 16 H) 1.51 (br SiCH2CH2 16 H) and 0.50 (br SiCH2 16 H). 13C-{1H} NMR (125.8 MHz CDCl3 25 8C) d 138.20 (d J = 13.3 Cquat) 133.01 (d J = 18.1 CH) 128.36 (s CH) 128.26 (d J = 6.5 CH) 58.93 (t J = 9.5 CH2N) 58.46 (d J = 5.8 Hz CH2) 19.68 (s SiCH2- CH2) and 9.47 (s SiCH2). 31P-{1H} NMR (202.5 MHz CDCl3 25 8C) d 227.6 (s). 29Si-{1H} NMR (99.4 MHz CDCl3 25 8C) d 266.3 (s) [Found (Calc.for C58H60N2O3P4Si2) C 68.70 (69.76); H 6.01 (5.97); N 2.72 (2.76)%]. When the same reaction was performed using compound 2?8HCl (78 mg 0.066 mmol) in MeOH–C6H6–NEt3 (2 5 8 mL respectively) the product obtained in high yield appeared to be the octahydrochloride salt of 17. 1H NMR (500.2 MHz pyridine-d5 25 8C) d 7.61–7.34 (br C6H5 160 H) 3.75 (br CH2 32 H) 3.07 (br CH2N 16 H) 1.86 (br SiCH2CH2 16 H) and 0.85 (br SiCH2 16 H). 13C-{1H} NMR (125.8 MHz pyridined5 25 8C) d 138.92 (d J = 13.6 Cquat) 133.62 (d J = 18.4 CH) 129.04 (s CH) 128.94 (d J = 6.5 Hz CH) 59.04 (s CH2N) 58.98 (s CH2) 20.29 (s SiCH2CH2) and 10.01 (s SiCH2). 31P-{1H} NMR (202.5 MHz pyridine-d5 25 8C) d 228.88 (s). 29Si-{1H} NMR (99.4 MHz CDCl3 25 8C) d 267.56 (s). The product tested positive for chloride (Ag1 and Beilstein test).Compounds 14 and 18 via reactions of 15 with n-BuNH2 and t-BuNH2. Isocyanate 15 was prepared as described above but the organic layer obtained after filtration through cotton was reduced to 70 mL (25 8C 10 Torr) then treated immediately with n-BuNH2 or t-BuNH2 (10 equivalents per NCO). A white suspension began to form within several minutes. After 1 h the reaction mixture was concentrated (25 8C 0.001 Torr) and then filtered to collect the crude product which was washed with cold MeOH and dried (25 8C 0.001 Torr). Analysis by 1H 13C and 29Si NMR spectroscopy indicated complete conversion of aminopropyl groups into ureido groups. Analytically pure J. Chem. Soc. Dalton Trans. 1999 1491–1497 1497 samples were obtained by recrystallization from CH2Cl2– MeOH (230 8C 7 d).A sample of 14 prepared via reaction of 15 with n-BuNH2 was indistinguishable from that prepared via reaction of 2 with n-BuNCO. For 18 yield 61 mg (18% from 2?8HCl). 1H NMR (500.2 MHz DMSO-d6 25 8C) d 5.64 (t CH2NH 8 H) 5.47 (br CNH 8H) 2.94 (m CH2N 16 H) 1.41 (m SiCH2CH2 16 H) 1.21 (s CH3 72 H) and 0.58 (t SiCH2 16H). 13C-{1H} NMR (125.8 MHz DMSO-d6 25 8C) d 157.65 (s CO) 48.83 (s Cquat) 41.15 (s CH2N) 29.23 (s CH3) 23.38 (s SiCH2CH2) and 8.67 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 25 8C) d 266.1 (s). Mass spectrum (MALDI-TOF DHB matrix) calc. for C16H34N4O5Si2; [M 1 H]1 m/z 1673.83 found 1673.84 (80%); [M 1 Na]1 1695.82 found 1695.80 (100%); [M 2 C5H9ON 1 H]1 1574.77 found 1574.77 (26%). Compound 22. A mixture of compound 21 8 (1.00 g 4.52 mmol) and AgNO3 (3.00 g 17.8 mmol) was refluxed in MeCN (50 mL) under an atmosphere of dry nitrogen for 48 h.The excess of silver salts was precipitated with saturated aqueous brine solution then solvent was evaporated to a minimum volume (25 8C 0.001 Torr). After addition of 1 1 chloroform– water and filtration through cotton the organic layer was separated dried over MgSO4 and evaporated (25 8C 0.001 Torr) to aVord 22 as a light sensitive microcrystalline pale yellow solid (710 mg 100%) which was used without further purifi- cation for the synthesis of 3. 1H NMR (500.2 MHz DMSO-d6 25 8C) d 4.47 (t J = 6.5 CH2O 16 H) 1.74 (m SiCH2CH2 16 H) and 0.72 (t J = 7.7 Hz SiCH2 16 H). 13C-{1H} NMR (125.8 MHz DMSO-d6 25 8C) d 74.92 (s CH2O) 19.71 (s SiCH2CH2) and 7.08 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 25 8C) d 266.7 (s) [Found (Calc.for C6H12N2O9Si2) C 23.29 (23.07); H 3.85 (3.87); N 8.61 (8.97)%]. Compound 3. Hydrogenolysis was accomplished by adding compound 22 (554 mg 0.443 mmol) 10% Pd/C (325 mg) ethyl acetate (10 mL) and dry MeOH (10 mL) to a small glass vessel in a Parr mini-reactor. The reactor was purged with H2 (3 × 400 psig) then charged with H2 (800 psig). After 3 d at 25 8C the reaction mixture was filtered and washed with dry MeOH (20 mL). Evaporation of the solvent (25 8C 0.001 Torr) aVorded 3 as a waxy white solid (335 mg 85%). All attempts to prepare crystalline samples or powders were unsuccessful but the product obtained in this fashion is spectroscopically pure (1H 13C NMR). 1H NMR (500.22 MHz DMSO-d6 25 8C) d 4.44 (br OH 8 H) 3.33 (t J = 6.8 CH2O 16 H) 1.46 (m SiCH2CH2 16 H) and 0.56 (t J = 7.6 Hz SiCH2 16 H).13C-{1H} NMR (125.8 MHz DMSO-d6 25 8C) d 62.90 (s CH2O) 25.82 (s SiCH2CH2) and 7.59 (s SiCH2). 29Si-{1H} NMR (99.4 MHz DMSO-d6 25 8C) d 265.9 (s). Mass spectrum (MALDI-TOF DHB matrix) calc. for C24H56O20Si8 [M 1 Na]1 m/z 911.14 found 911.0 (100%); [M 1 K]1 927.11 found 928.1 (5%). Acknowledgements These studies were supported by the National Science Foundation and Phillips Laboratory (Edwards AFB). The NMR and mass spectra were determined at University of California using instruments acquired with funding from the National Science Foundation and National Institutes of Health. References 1 M. G. Voronkov and V. Lavrent’ev Top. Curr. Chem. 1982 102 199. 2 R. H. Baney M. Itoh A. Sakakibara and T. Suzuki Chem.Rev. 1995 95 1409. 3 D. W. Scott J. Am. Chem. Soc. 1946 68 356. 4 M. G. Voronkov T. N. Martynova R. G. Mirskov and V. I. Belyi Zh. Obshch. Khim. 1979 49 1522; T. N. Martynova V. P. Korchkov and P. P. Semyannikov J. Organomet. Chem. 1983 258 277. 5 F. J. Feher and T. A. Budzichowski J. Organomet. Chem. 1989 379 33. 6 R. Weidner N. Zeller B. Deubzer and V. Frey U. S. Pat. 5 047 492 1991. 7 F. J. Feher D. Soulivong A. G. Eklund and K. D. Wyndham Chem. Commun. 1997 1185. 8 U. Dittmar B. J. Hendan U. Flörke and H. C. Marsmann J. Organomet. Chem. 1995 489 185. 9 C. Zhang and R. M. Laine J. Organomet. Chem. 1996 521 199. 10 C. L. Frye and W. T. Collins J. Am. Chem. Soc. 1970 92 5586. 11 P. A. Agaskar Inorg. Chem. 1991 30 2707. 12 P. G. Harrison C. Hall and R. Kannengiesser Main Group Met.Chem. 1997 20 515. 13 V. W. Day W. G. Klemperer V. V. Mainz and D. M. Millar J. Am. Chem. Soc. 1985 107 8262. 14 P. A. Agaskar V. W. Day and W. G. Klemperer J. Am. Chem. Soc. 1987 109 5554. 15 A. R. Bassindale and T. Gentle J. Organomet. Chem. 1996 521 391. 16 D. Herren H. Bürgy and G. Calzaferri Helv. Chim. Acta 1991 74 24. 17 A. R. Bassindale and T. E. Gentle J. Mater. Chem. 1993 3 1319. 18 T. E. Gentle and A. R. Bassindale J. Inorg. Organomet. Polym. 1995 5 281. 19 J. V. Crivello and R. Malik J. Polym. Sci. Part A Polym. Chem. 1997 35 407. 20 F. J. Feher and K. D. Wyndham Chem. Commun. 1998 323. 21 M.-C. Gravel and R. M. Laine ACS Polym. Prepr. 1997 38 155. 22 F. J. Feher K. D. Wyndham M. A. Scialdone and Y. Hamuro Chem. Commun. 1998 1469. 23 E. Rikowski and H. C. Marsmann Polyhedron 1997 16 3357.24 F. J. Feher J. J. Schwab D. M. Tellers and A. Burstein Main Group Chem. 1998 2 169. 25 F. J. Feher D. Soulivong and F. Nguyen Chem. Commun. 1998 1279. 26 F. J. Feher D. Soulivong and A. E. Eklund Chem. Commun. 1998 399. 27 F. J. Feher D. Soulivong and G. T. Lewis J. Am. Chem. Soc. 1997 119 11323. 28 F. J. Feher K. D. Wyndham and D. J. Knauer Chem. Commun. 1998 2393. 29 J. D. Lichtenhan Comments Inorg. Chem. 1995 17 115. 30 J. D. Lichtenhan in Silsesquioxane-based Polymers ed. J. C. Salamone Polymeric Material Encyclopedia CRC Press New York 1996 vol. 10 pp. 7768–7778. 31 D. A. Tomalia H. Baker J. R. Dewald M. Hall G. Kallos S. Martin J. Roeck J. Ryder and P. Smith Polym. J. 1985 17 117. 32 A. D. Meltzer D. A. Tirrell A. A. Jones P. T. Inglefield D. M. Hedstrand and D.A. Tomalia Macromolecules 1992 25 4541. 33 M. T. Reetz G. Lohmer and R. Schwickardi Angew. Chem. Int. Ed. Engl. 1997 36 1526. 34 F. Dietsche and R. Mülhaupt 215th National Meeting of the American Chemical Society Meeting Dallas TX March 1998 poster presentation. 35 K. Kobayashi H. Sumitomo and Y. Ina Polym. J. 1985 17 567. 36 T. J. Williams N. R. Plessas and I. J. Goldstein Carbohydr. Res. 1978 67 C1. 37 J. S. Nowick D. L. Holmes G. Noronha E. M. Smith T. M. Nguyen and S. L. Huang J. Org. Chem. 1996 61 3929. Paper 8/07302C

ISSN:1477-9226    

DOI:10.1039/a807302c

出版商:RSC    

年代:1999

数据来源: RSC