4102
Cobalt ( 111)-Promoted Hydrolysis of Glycine Esters. Kinetics, Product Analysis, and Oxygen-18 Exchange Studies of the Base Hydrolysis of [ Co (en) (glyOR) ] '+Ions D. A. Buckingham, D. M. Foster, and A. M. Sargeson
Contribution f r o m the Research School of Chemistry, Australian National University, Canberra, Australia. Received December 23, 1968 Abstract: The base hydrolysis of the complexes of the type cis-[Co(en)zX(NH2CHzCOOR)IZ+(where X = C1 and Br ; R = CH3, CZHS, CH(CH3),,(CH2)CH3,C(CH3)3,and CHz(CsHs))leads to [Co(en)zglycinato]2+ primarily, along with numerous side products some of which are not yet characterized. Substantially, [Co(en)zglylZfarises rapidly after loss of X- and coincides with loss of the ester group. '*O tracer experiments with cis-[Co(en)zBr(glyOCH(CH3)z)]2+show -50% [Co(en)2glylZ+arises from coordination of the ester carbonyl oxygen while the remainder is produced by the intervention of a solvent oxygen atom in the bridging position. The latter path is interpreted as an internal nucleophilic displacement of the ester moiety by bound OH- ion. Base hydrolysis of (-)5~dCo(en)~Br(glyOCH3)]2+ leads to 50% racemization in the [C~(en)~gly]~+ product, and this factor may be related to the o'*distribution,
I
n a previous study, the Hg2+-and HOC1-induced rethe ester group by the bound OH-. Certainly, in a removal of bromide ion from [C~(en)~Br(glyOR)]*+ lated study of the base hydrolysis of [Co(NH&NH2(R = CH, and CH(CH3),) to form [Co(en)zglylZ+(gly = CH2COORl3+,the importance of the role of the internal chelated NH2CH2C02-; glyOH = monodentate NH, nucleophile was established.* CHZCOZH;gly0 = monodentate NH2CH2C02-; glyDuring the course of the work, a communication OR = monodentate (or chelated) NH2CH2C02R)was appearedg which discussed the base hydrolysis of shown to proceed r;ia the chelated ester intermediate [C~(en)~Cl(glyOC~H~)]Cl~ in terms of preliminary [Co(en), (glyOR) 13+, as had been previously proposed.2 hydrolysis of the monodentate ester moiety, followed by Incorporation of solvent water was inferred in the case hydrolysis of chloride. The variance between this of the HOC1-induced reaction, with rapid concerted disinterpretation and our own observations on [Co(en)2Brplacement of the bound water by the ester carbonyl (glyOCH(CH3)z)](C104)2, together with the over-all group. By contrast, the Hg2+-induced reaction was becomplexity of the reactions, caused us to extend this lieved to involve exclusive incorporation of the carbonyl study to cover several other complexes of the type group in the vacant coordination site of the five-coor[Co(en)sX(glyOR)IX~. dinate intermediate. These systems have the merit that the coordinated Experimental Section amino acid derivative remains bound in solution and Analar reagents were used throughout without further purificathe order of events in the reaction can be examined in tion. more detail than for labile systems such as Ni2+ and 0 ]*-Labeled glycine isopropyl ester hydrochloride was prepared by refluxing 0'8-enriched glycine hydrochloride (prepared as deCU*+complexes. Moreover, the compounds may posscribed previously') in isopropyl alcohol with thionyl chloride, sibly be considered as models for metal-ion activated Unlabeled methyl, ethyl, isopropyl, n-butyl, and benzyl esters of esterases. Whatever the propriety of this claim, it glycine were prepared similarly from the appropriate alcohols. would seem worthwhile to possess an account of the Glycine f-butyl ester hydrochloride (Puriss) was obtained from Fluka. processes involved in the Co(II1)-promoted hydrolysis Infrared spectra were recorded on a Perkin-Elmer 457 instrument. Spectrophotometric rate data were collected on both Cary of glycine esters under conditions more closely ap14 and Shimadzu RS 27 recording spectrophotometers; the Cary proaching physiological conditions than those in the in14 instrument was used in the characterization of eluate fractions duced acid hydrolyses. In general, base hydrolyses of from ion exchange of reaction mixtures. Bio-Rad analytical complexes of the type [CoNjXI2+(X = C1, Br; N = Dowex 50W X 2 (200-400 mesh) cation-exchange resin was used in the ion-exchange separation experiments. pH determinations amine) are held to occur Diu five-coordinate intermewere made on a Cambridge bench instrument. Some cobalt estid i a t e ~ ~ -which ' are different from those formed in the mations were made using a Techtron AA4 atomic absorption Hg2+-induced reactions. For the ester complexes the spectrophotometer. possibility arises of the ester carbonyl group and solvent The 0 1 8 content of COI recovered from the labeled compounds water competing for the vacated site. The latter path was determined using Atlas M-86 and GD-150 mass spectrometers. ajgSvalues for optically active complexes were measured with a leads to the hydroxo ester complexes and the possibilPerkin-Elmer 141 spectropolarimeter using a 1-dm cell. Concenity of subsequent internal nucleophilic displacement of (1) D. A . Buckingham, D. M. Foster, and A. M. Sargeson, J . Am. Chem. SOC.,90, 6032 (1968). (2) M. D. Alexander and D. H. Busch, ibid., 88, 1130 (1966). (3) R. G. Pearson and F. Basolo, J . Am. Chem. SOC., 78, 4878 (1956). (4) M. Green and H. Taube, Inorg. Chem., 2, 948 (1963). (5) D. A . Buckingham, I. I. Olsen, and A. M. Sargeson, J . Am. Chem. Soc., 89, 5129 (1967). (6) D. A. Buckingham, I. I. Olsen, and A. M. Sargeson, Aust. J . Chem., 20,597 (1967). (7) D. A . Buckingham, I. I. Olsen, and A. M. Sargeson, J . Am. Chem. Soc., 90, 6654 (1968).
Journal of the American Chemicalsociety 1 91.15 1 July 16,1969
trations of isopropyl alcohol liberated during base hydrolysis of the corresponding ester complex were determined using a Varian Aerograph 600D gas chromatograph. Rates of base uptake were determined at constant pH and temperature by pH-Stat titration with the following Radiometer apparatus: TTA3 electrode assembly, ABUl autoburet, TTTI titrator,
(8) D. A. Buckingham, D. M. Foster, and A. M. Sargeson, ibid., 91, 3451 (1969). (9) R . W . Hay, M. L. Jansen, and P. I. Cropp, Chem. Commun., 621 (1967).
4103 cator. After conversion to the chloride, it was made up to 50 ml and SBRZ titrigraph. The thermostated reaction vessel was conas before, 0.1 M in HCIOl and p = 1.0. tinuously stirred as the titrant (NaOH of appropriate concentraSupplementary Labeling Experiments. To determine the rate of tion) was added under a nitrogen atmosphere. exchange between coordinated glycine and solvent under basic Preparation of Complexes. Complexes of the type cis-[Coconditions, [C~(en)~gly]I~ (0.5 g) was dissolved in H201* (10 ml) (en)2X(glyOR)]X2 (where X = C1 and Br; R = CH3, C Z H ~CH, at p H 9.0. After 60 min the product was reprecipitated as the (CH& (CH2)3CH3,CH2C6H5,and C(CH&) were prepared from iodide salt by addition of excess NaI and cooling. This was rerrans-[Co(en)tBr~]Br.HBr or rrans-[Co(en)~Cl~]Clwith the approcrystallized from warm water a t p H 4, precipitated as the Hg142priate glycine ester hydrochloride (glycine ester p-toluenesulfonate in salt, washed with water and acetone, dried in an evacuated desicthe case of the benzyl ester), in the manner described by Alexander cator, and analyzed for 0l8(Table VID). and Busch.lO To determine whether coordinated glycine was subject to intraWhere desired, these complexes were converted to their perchlomolecular oxygen scrambling under basic conditions, carbonyl 0 1 8 rate salts by dissolution in hot 10-3 M HC104, followed by addition of excess NaC104 and cooling in an ice bath. The products were labeled [ C ~ ( e n ) ~ g l y ]was ~ + prepared from carbonyl Oi8-labeled [CO(~~)~B~(~~~OCH(CH~)Z)]B~Z by dissolving 2 g in 0.1 M HCIOl twice recrystallized from hot, dilute HClO4 by adding NaC104. (20 ml) with Hg(NO& (7 g), filtering after 2 min, and precipitating They were washed with ethanol and dried in an evacuated desic[C~(en)~gly](HgI~) by addition of excess NaI to the filtrate.' The cator. Anal. Calcd for [C0(en)~Br(glyOCH3)](C10,)2: C, 15.38; product was washed with water and methanol and dried. I t was H,4.24; N, 12.80. Found: C, 15.28; H,4.41; N , 12.93. Calcd then shaken with excess AgCl in water (100 ml), and after filtration, ) ~ )18.80; ~ ( C I OH~, )4.73; ~: N, for [ C O ( ~ ~ ) ~ B ~ ( ~ I ~ O C H ( C H ~ C, the filtrate was kept at pH 8.7 for 90 min and then neutralized 12.18. Found: C , 18.84; H , 4.99; N , 12.36. Calcd for [Co(en)2Br(glyOCH~C6Hj)]Br2: C , 26.73; H, 4.66; N, 11.99. Found: (pH 6) and [Co(en),gly](HgI,) precipitated from solution. After the C, 26.83; H , 4.76; N, 11.78. Calcd for [Co(en)~Br(glyO(CH~),- usual washing and drying procedures, a sample was retained, while the remainder was converted to the chloride by shaking with AgCl CH3)]Br2: C, 21.84; H , 5.32; N, 12.74. Found: C, 22.73; H , and made up to 10 ml, 0.1 M in HClO4. Two 5-ml aliquots were 5.93; N, 13.07. Calcd for [C0(en)~Br(glyOC(CH3)3)Br~:C, 21.84; sampled (Table VIE). H , 5.32; N, 12.74. Found: C21.84; H , 5.19; N, 12.92. Calcd C, 19.70; H , 5.33; for [Co(en)2Cl(glyOCH(CH3)~)J(C104)z~HzO: To determine whether the chelated glycine esters were subject N, 12.77. Found: C, 19.93; H , 5.33; N, 12.89. Calcd for to intramolecular oxygen scrambling during base hydrolysis, car[C0(en)~Cl(glyOC,H~)](Cl0~)~~H~0: C, 17.97; H, 5.09; N, 13.10. (3 g) was prebonyl 018-labeled [Co(en)~(glyOCH(CH3)~)](CI04)3 Found: C , 17.85; H, 5.00; N, 12.89. Calcd for [Co(en)2C1pared from carbonyl Oi8-labeled [Co(en)~Br(glyOCH(CH3)?1(glyOCH3)]C12~0.5H20:C, 21.91; H, 5.78; N, 18.26. Found: ( C 1 0 4 ) zand ~ ~ hydrolyzed in water (50 ml) at pH 8.6 using the RadiC, 22.06; H , 6.03; N, 18.45. ometer. After 5 min all base uptake had ceased. The solution Carbonyl 018-labeled ~is-[Co(en)~Br(glyOCH(CH3)~)]Br? was prewas then filtered, and reduced in volume on a rotary evaporator and pared from rrans-[C~(en)~Br~]Br .HBr and O18-labeled glycine iso[C0(en)~gly]I2precipitated by addition of NaI and cooling. The propyl ester hydrochloride. Anal. Calcd for [Co(en)?Br(glyproduct was washed with water and methanol and dried in an evacOCH(CH3)2)]Brz.H20: C, 19.51; H, 5.28; N, 12.64. Found: uated desiccator (1 g), then converted to the chloride by shaking with was prepared C, 18.78; H, 5.42; N, 12.87. cis-[Co(en)~NH3Br]Br~ AgCl and made up to 10 ml, 0.1 M in HC104, p = 1.0 (NaCIO,). as described by Werner.11 Anal. Calcd for [Co(en)2NH3Br]Br?: Two 5-ml aliquots were sampled (Table VIF). C, 11.02; H , 4.39; N, 16.07. Found: C, 10.8; H,4.3; N, 15.8. To test the enrichment of the recovered starting material [CoMeasurement of Oxygen Exchange in Ol8-Labeled [Co(en)~gly]~ (en)2Br(glyOCH(CH3)2)](C104)~ (1.5 g) was dissolved in H 2 0 (1.5 Produced cia Base Hydrolysis of [C0(en~)Br(glyOCH(CH3)r)]~ I. atom % 0l8)and hydrolyzed at pH 8.6 for 10 min, 25". The soluLabeled Complex. Carbonyl 0'8-labeled cis-[Co(en)yBr(glyOCHtion was returned to pH 5-6 with HC104, and excess NaC104 (CH3)2)]Br2(10 g) in water (1000 ml) was hydrolyzed at pH 8.6 On cooling, [C0(en)~Br(glyOCH(CHp)2)1(ClO~)~ preadded. and 25" for 90 min, by pH-Stat titration against 30% NaOH. The cipitated. This was recrystallized from unlabeled water and airsolution was then taken to pH 6 and reduced to ca. 200 ml on a dried (0.1 g). The identity of the material was confirmed by ir. rotary evaporator. On addition of excess NaI and methanol, It was dissolved in 0.01 M HC1O4 (5 ml saturated in Hg(NO&) [C~(en)~gly]I~ precipitated. This was collected, washed with cold and allowed to stand for 1 hr. Excess NaI was added and the NaI solution, methanol, and ether, and recrystallized from hot [C0(en)~gly](HgT4)obtained examined for 0 l* (Table VIG). water by addition of NaI. The final product (3.7 g) was washed as Resolution of cis-[Co(en)~Br(glyOCHa)](C104)~ and the Measurebefore and dried in an evacuated desiccator. The purity of the ment of Optical Retention in the Hydrolyzed Product. cis-[Coproduct was confirmed by comparison of its ir spectrum with that of (en)2Br(glyOCH,)](CIO& (5.5 g) was dissolved in hot dilute acetic [C~(en)~gly]I~. The dried material was shaken with excess AgCl acid (100 ml, pH 4), and ammonium d-bromocamphorsulfonate in water (20 ml) for 2 min and the precipitate of AgCl and AgI re(NH4-(+)-BCS, 3.28 g) added. On standing at room temperature moved. The filtrate was made up to 50 ml, 0.1 M in HC104 and ( -)sss-[Co(en)2Br(glyOCH3)1-(+)-(BCS)~ slowly crystallized. Suc0.60 M in NaC104. This solution was thermostated at 25" and cessive fractions were removed, washed with ice water and acetone, 5-ml aliquots were periodically withdrawn. From these, [Co(en),and dried. The first three fractions were combined (1.9 g) and once gly](Hg14) was recovered and the 0 ' 8 content of the glycine derecrystallized from hot dilute acetic acid. The diastereoisomer termined as previously described' (Table VIA). ([a]5ss 11.6') was then converted to the bromide salt by trituration 11. Labeled Solvent. The above procedure was followed using with NaBr at p H 4, and the resulting ( -)jss-[Co(en)2Br(glyOCH3)1unlabeled [C0(en)~Br(gly0CH(CH~)~)](C10~)? in enriched water Br2recrystallized to constant rotation from hot, dilute acetic acid by (1.5-2 atom % 018). A solvent sample was taken immediately adding excess NaBr and cooling. A 0.1% solution in 0.01 M after initiation of hydrolysis and its 0 1 8 enrichment determined as HCI04 gave cyJsg -0.115", whence [cy]s89-115". Anal. Calcd for described previously. (-)588-[Co(en)2Br(glyOCH3)1Br~: C, 16.55; H , 4.56; N, 13.79. In one instance (B, Table VI) cis-[Co(en)~Br(glyOCH(CH3)~)]-Found: C,16.81; H,4.78; N,13.85. (c104)2(6 g) in H2018 (30 ml) was hydrolyzed at pH 8.6 for 60 min, ( -)jg9-[C~(en)?Br(glyOCH3)]Br~ (0.1006 g) was hydrolyzed by after which time [C0(en)~gly]I2was precipitated from the solution pH-Stat titration at pH 8.6 and 25" for 90 min using 0.2 M NaOH. with NaI and twice recrystallized as in section I above, yield 1.6 g. The solution was then adjusted to pH 6 and sorbed onto an H+This was converted to the chloride salt in solution and made up to form resin. After washing, the [Co(en)~gly]~+ band was collected 25-ml volume, pH 1.0, p = 1.0 (NaC10,). (100 ml), estimated spectrophotometrically (75.479 and its rotation In another instance (C, Table VI), [C0(en)~Br(gly0CH(CH3)~)]measured (ass -0.117", whence [cy1589 - 154"). (+)S9-[Co(en),(cl04)z (7.5 g) in H2018(60 ml) was hydrolyzed at p H 8.6 for 90 300") treated in an identical manner showed gly]12(0.1033 g, [aJjgg min, neutralized, and then sorbed directly on to an H+-form resin no change in rotatory power. and the [ C ~ ( e n ) ~ g l yband ] ~ + was eluted with 1 M NaC104, pH 7-8. Kinetic Measurements. The base hydrolysis of [Co(en2)BrThe eluate band was reduced almost to dryness by rotary evaporawas fol(glyOCH(CH3)2)](C104)2and [Co(en)~Br(glyOCH~)](ClO~)~ tion and filtered to remove NaCI04; [C~(en)~gly](HgI~) precipitated lowed spectrophotometrically after rapid dissolution of a weighted from the filtrate by addition of HgIz then NaI. The product was quantity of complex in tris(hydroxymethy1)aminomethane (Tris) washed with water and methanol and dried in an evacuated desicor glycine buffer at p = 1.0 (NaC104) and 25". Some rate measurements were obtained by running the spectrum from 600 to 300 mp to spaced time intervals, while others were obtained from runs at a set wavelength. (10) M. D. Alexander and D. H. Busch, Znorg. Chem., 5, 602 (1966). Base hydrolysis of all complexes was also followed by pH-Stat (11) A. Werner and W. E. Boes, Ann., 386, 55, 178 (1912).
+.
+
Buckingham, Foster, Sargeson
/ Base Hydrolysis of [Co(en)nX(glyOR)]
2+
Ions
4104 titration of 0.2 M NaOH using the Radiometer apparatus. A weighed quantity of complex (-0.2 g) was dissolved either in water or in a solution of KNOBof known ionic strength (20 mi) and transferred to the thermostated reaction vessel. To measure the rate of alcohol production in the base hydrolysis of [C0(en)~Br(gly0CH(CH&)](C10&, the complex (0.009 g) was dissolved in 2,4,6-collidine buffer (10 ml, p H 8.55, p = 1.0) at 25", and 2-ml aliquots were periodically withdrawn, quenched with HCIOi (1 drop 70%), and then immediately frozen in methanolDry Ice. The frozen solvent was distilled under vacuum and analyzed chromatographically for isopropyl alcohol. Product Analysis by Ion-Exchange Chromatography. A. Separation of the Products of Base Hydrolysis. The complex (ca. l mmole) was hydrolyzed at known p H within the range 8-9, either by dissolution into buffer or by pH-Stat titration. A specified period of time, or a period of time sufficient to account for a selected base consumption, was allowed, and the reaction then quenched with acid. The solution was diluted and sorbed on a H+-form resin (ca. 1.5 x 25 mm when fully swollen), and eluted with NaCIOa, 1-2 M . The pH of elution was normally 5-6. The cobalt concentrations in the eluate fractions were determined spectrophotometrically and/or by atomic absorption spectroscopy. B. Azide Competition. cis-[Co(en2)Br(glyOCH(CH3)~)](ClO& (0.53 g) was dissolved in Tris buffer, pH 9.2,O.g M i n NaN3 (25 ml). After ca. 20 min, the solution was diluted to 100 ml and sorbed on a H+-form resin. Elution was with 1 M NaCIO,, pH 6-7. cis[Co(en)?Br(glyOC(CH&)]Br? (0.10 g) was dissolved in the same buffer, 0.9 M in NaNs (10 mi), and after 17 min the solution was diluted and sorbed on a Na+-form resin. Elution was with 1 M NHaCl (pH 6-7). All eluate bands were examined spectrophotometrically .
Results A. Kinetics. Two rates were observed for base hydrolysis of the [C~(en)~X(glyOR)]'+ ions. Table IA Table I. Data for Base Hydrolysis of [Co(en)~Br(glyOCH(CH&)~ClOl)s [complex],
PHfl
?M
10- Zkl,,
103
10ikl', sec-l
losky', sec-l
10- %*,f
.VI-'
M-'
sec-'
sec-'
Spectrophorometric Data in 0.1 M Tris Buffer ( p = 1.0 (NaC104). 25.0") 8.00 1.39 2.8 5.1 2.8 1.5 8.2 12.1 2.4 8.54 9.0 10.8 3.0 S . 4 8 ~ 8.5 8.52 3.35 9.6 8.0 2.9 9.14 1.66 36.2 52.5 2.6 9.13 1.75 38.5 48.2 2.8 S,59 1.49 8.9 12.3 2.5 9.W 1.60 96 137 2.9 B. Radiometer Data ( p = 0.05, 25.0") 8.5 6.6 6.6 14.8 5.3 35.0 14.8 16.7 5.9 18.5 30.8 14.8 5.8 18.3 33.0 1.5 5.1 85.5 14.5 12.1
Table 11. Base Hydrolysis of [C0(en)2X(glyOR)]X!~
5.1 3.5 3.6 2.4 3.8 3.6 3.5 3.6 8.5
--M-'
--Complex----
x
A.
8.0 8.5 8.5 8.51 9.1
and the second-order constants kl and kz are given in Table IA. The penultimate entry in Table IA refers to rates obtained for the methyl ester, [Co(en)2Br(glyOCHo)](C104)2. The agreement between these rates and those cited for the isopropyl ester suggests that hydrolysis of the ester moiety is not rate determining for either path. The reactions are not subject to general base catalysis, insofar as changing the buffer concentration does not effect the rates, and changing the buffer has no effect. Table IB gives rate constants obtained from the rate of base uptake by [Co(en)eBr(glyOCH(CH3)2)](C101)2 at constant pH in the absence of buffer. The experimental data were treated in a similar manner to that for the spectrophotometric data. The ionic strength is lower than in Table IA, and the quoted second-order rates are consequently faster, but the pattern is consistent with the spectrophotometric data in Table IA. This is supported by the inclusion in Table IB of a spectrophotometric run performed at a similarly low ionic strength. The results establish that the same series of reactions are being followed both spectrophotometrically and by pH-Stat titration, that Tris buffer is not exerting an inhibitory effect upon the system,'* and that both rates involve the consumption of hydroxide ion. Table I1 gives rate data obtained by pH-Stat titration for the series of complexes [Co(en)t~X(glyOR)]Xt (where X = C1 and Br; R = CHS, CH2CH3, CH(CH&, CH2C6Hs, (CH2)3CH3, and C(CH3)3). The secondorder rate constants kl show a sensitivity toward X, but not toward a change in R for constant X.
R
Br Br Br Br Br Br CI CI C1 fl
pH 8.6, 25.0", p
-
0.03.
10-'ki
sec-110-1ki
8.2 8.2 8.9 8.2 7.5 9.5 1. Y 2.0 2.1
1.9 13.6 11.2 15.0 14.3 9.5 6.1 5.4 7.0
Perchlorate salt.
[C~(en)~gly]~" PH 8.6 PH 9.0' 76 68 68 68 15 59 51 45 45 cp
73 69 59 54 40 40 35
.- 0.1.
11.1
9.8 10.4 6.8
pH at conclusion of hydrolysis. 0.05 M Tris buffer. [Co( C L ~ ) ~ R ~ ( ~ I ~ O C H ~ )d]0.1 ( C I,M O ~glycine ) ~ . buffer. e Spectrophotometric data. 0.1 M Tris buffer, p = 0.063. 1k = k'/[OH-].
Table I11 gives the rate dependence upon ionic strength for loss of bromide ion i n [Co(en)?BrNH3]Br2 and [C0(en)?Br(gly0CH(CH,)~)](C10& at 25 '. Observed rates were obtained from base consumption at
Table 111. Rate Dependence of Base Hqdrolysis upon presents data for [Co(en)?Br(glyOCH(CH3)2)](C104)~ Ionic StrengthY -lO%,ld, sec-----I obtained spectrophotometrically at 480 mp and in some
instances at 340 mp, Plots of log (D, - 0,) us. time were initially curved, but at longer times became linear for at least two half-lives of the second reaction. The extrapolated optical density (D,,,)attributed to the slower rate was subtracted from the observed data at shorter times to give in all cases a linear plot of log (Dext- 0,) us. time over at least two half-lives. Both rates fit the rate law R = k[complex][OH-] JournaI of the American ChemicalSociety
1 91:15 1 July 16,1969
p =
p =
p =
!J=
P =
Complex
0.04
0.1
0.2
0.5
1.0
[Co(en)?Br(NH3)]Brz [Co(en)?Br(glyOCH(CHddl(C10A
9.8 72
7.9 72
6.3
4.6 46
4.3 36
a
pH 9.0, 25.0", [complex] G
M.
(12) D. E. Allen, D. J. Baker, and R. D. Gillard, Nature, 214, 906 (1967).
4105 Table IV. Ion-Exchange Separation of Products on Base Hydrolysis of ICo(en~zCl(nlyOCH3)1Clzu
5
Band
Color
Probable charge
1 2 3 4
Blue-red Red Red Red-brown
+1 +1 +1
5
6 7
Orange Red Red
+2 +2 +3
8, 9
Brown
$3 or higher
+2
Amax, mp
Abundance
Assignment
545, 350 513, 355 520, 364 525, 465, 358 487, 347 520, 365 498, 360
Trace Trace 5% Major product 28 % 17 % Minor product
truns-[Co(en)zOH(glyO)] ~is-[Co(en)~OH(glyO)]+ [Co(en)~Cl(gl~O)l+ [Co(en)z(OHz)(gly en-H)I 2+ (?)
...
Minor products
+
[C~(en)~glyl 2+ [C~(en)~Cl(glyOCH~)] 2+
1 1
condensation products (?)
7.7 min, pH 9.0, 25", p = 0.15; eluent, 1-3 MNaC104.
elution speed in I M NaCIOl for a Dowex 50W X 2 constant pH, using K N 0 8 as the supporting electrolyte. (200-400 mesh) resin. It can be seen that in both instances an increase from Bands 1 and 2 were rapidly displaced by 1 M NaC104 ,u = 0.04 to 1.0 results in about a twofold decrease in and are tentatively assigned to trans- and cis-[Co(en)nrate, and that the [C0(en)~Br(gly0CH(CH~)~)]~+ ion (OH)(glyO)]+ species, respectively; they were also hydrolyzes about 10 times faster than the [Co(en)2BrNH3I2+ion. A similar factor is observed between the P2-[Co(trien)Cl(glyOCzH6)]2+ and /3z-[Co(trien)C1NH3]z+ ions, l 3 even though these ions base hydrolyze some 3000 times faster than the corresponding ethylenediamine complexes. Figure 1 gives the rate of isopropyl alcohol production in the base hydrolysis of [C0(en)~Br(gly0CH(CH~)~)](C10J2 at pH 8.55, p = 1.0 (NaC104), and 25". Log (C, - C,), where C represents the concentration of liberated isopropyl alcohol measured from peak areas, was plotted against time. The observed rate constant kobsd= 7 X sec-' may be compared with kl of the second entry in Table IA obtained spectrophotometrically under the same conditions of pH and ionic strength. After 200 min, 9 3 z of the theoretical amount of isopropyl alcohol in the ester complex was recovered on the gas chromatograph. B. Reactant and Product Analysis. Included in Table I1 is a result for optically pure ( -)sss-[Co(en)zBr(glyOCH3)]Brz. The consistency of the rate data obtained for this complex with those for the unresolved material and the other ester complexes in general I , I suggests that in all cases only cis complexes have been 5 IO IS 20 25 studied. Further evidence in support of this claim was minutes found by elution of [Co(en)zBr(glyOCH(CH3)z)]2+ from Figure 1. Rate of isopropyl alcohol production in the hydrolysis of a 80-cm ion-exchange column (H+-form resin) using 1 [Co(en)ZBr(glyOCH(CH,),)1(C104)~ (pH 8.55, p = 1.0, 25"). M NaC104 at pH 2. Only a single band resulted. Isopropyl alcohol concentration in arbitrary units. Under these conditions the cis- and trans-[Co(en)zNH3C1l2+ions can be separated.' These results agree observed in higher yields as hydrolysis products from with those of Alexander and Busch, lo Meisenheimer, l 4 [C~(en)~Br(glyOH)]Br~.'~ The peak at 545 mp in the and Bailar and C1app;lj the latter authors studied the trans isomer shifts to shorter wavelengths on standing at reaction of several amines with cis- and trans-[Co(en)zpH 5 , while the cis isomer reacts rapidly to give the Cl,]Cl. Only cis products arose from these experiabsorption spectrum for [ C ~ ( e n ) ~ g l y ]on ~ +acidification. ments. The former observation suggests a slow isomerization Table IV gives a product analysis for [Co(en)zC1to the cis species, and the latter resembles in many (glyOCH3)]Cl~,obtained following ion-exchange elution respects the rapid chelation of monodentate oxalate and of the hydrolyzed solution. The first six bands were bicarbonate in the well separated on elution with 1 M NaClO?. Bands 7-9 were immobile until 3 M NaC104 was used. Prob00 0 [CO(~~)Z(OHZ)(OCCO)]+ and [ C O ( ~ ~ ) ~ ( O H ~ ) ( O C O ) ] + able charge assignments were made on the basis of ions, respectively. 17,18 This aspect of the study is being
j!
I
(13) D. A . Buckingham, D. M. Foster, L. G. Marzilli, and A. M. Sargeson, Inorg. Chem., in press. (14) J. Meisenheimer and K . Kiderlen, Ann., 438, 238 (1924). (15) J. C. Bailar and L. B. Clapp, J . A m . Chem. SOC.,67, 171 (1945).
I
(16) D. A . Buckingham, D. M. Foster, A . M. Sargeson, and L. G. Warner, unpublished results. (17) P. M. Brown and G. M. Harris, Inorg. Chem., 7, 1872 (1968). (18) H Scheidegger and G . Schwarzenbach, Chimia, 19, 166 (1965).
Buckingham, Foster, Sargeson
1 Base Hydrolysis of [Co(en)aX(glyOR)] Ions =+
4106 Table V. Comparison of Observed and Calculated Base Consumption" ~~~
7 -
Equiv/mole of complex Obsd Calcd
Complex
[Co(en)~Br(glyOC(CH3)a)l 2+ [Co(en)~Br(glyOCH(CH~)~)l~ [C~(en)~Br(glyOCH~)] 2+ [C0(en)~Br(glyOCH~C~H~)12+ *
1.15 1.31 1.55 1.37 1.60 1.72 1.73
+
[Co(en)~Cl(glyOCH(CH~)~)l~ +
[Co(en)2Cl(gl~OCzHj)l 2+ [C~(en)~Cl(glyOCH~)]~+ pH 9.0,25.0", p
=
0.1.
* pH 9.1.
c
1.27 1.31 1.41 1.46 1.60 1.61 1.68
k 2 path Obsd
0.60 0.32 0.35 0.14 0.47 0.48 0.51
Mole/mole of complex------[Co(en)~gly]~+ kl path kl pathc production Calcd Calcd - obsd 0.73 0.69 0.59 0.54 0.40 0.39 0.32
0.70 0.84 0.82 0.93 0.77 0.76 0.75
-0.03 0.15 0.23 0.39 0.37 0.37 0.43
Consumption of base by species other than [ C ~ ( e n ) ~ g l ycia ] ~ +the kl path (column 5 - column 4).
investigated further. Band 3 was identified as [Co(en),higher pH's, and at higher complex concentrations, Cl(glyO)]+ by ion exchange and a spectral comparison less [ C ~ ( e n ) ~ g l yis ] ~formed. + with an authentic sample prepared by acid hydrolysis Table V shows the observed and calculated base of [C~(en)~Cl(glyOCH~)]Cl~. consumption at constant pH for a variety of ester l9 Band 4 was a major complexes. The first column gives the observed base product of the reaction of unknown constitution. It was homogeneous by ion exchange and did not contain consumption during hydrolysis, as found from base uptake on the pH-Stat. At least a portion of the product chloride ion. The absorption band at 465 mp increased is [C~(en)~gly]~+, and 1 equiv of base is required for its in intensity with a concomitant decrease in the 525-mp band on standing at pH 5. Its elution rate relative to production. The fourth column of values gives the [C~(en)~Cl(glyO)]+and [ C ~ ( e n ) ~ g l y ] ~supports + its observed figure for base consumption due to production designation as a 2+ ion. Bands 5 and 6 were identified of [ C ~ ( e n ) ~ g l y (isolated ]~+ on the ion-exchange column, by spectrophotometric comparison with the authentic Table V). If it is then assumed that all other products complexes, as [ C ~ ( e n ) ~ g l y ] and ~ + unreacted starting consume 2 equiv of base, the figure calculated for the total base consumption is that cited in column 2. In material, respectively. Small amounts of more firmly sorbed ions were retained on the column (bands 7-9) all cases, the calculated base consumption is greater and were eluted only at higher ionic strengths (3 M ) ; all than 1 equiv and agrees with the observed value. Table V, column 3, gives values for base consumed by together they represented a substantial portion of the the ks path. This value was obtained from the rate reaction products. plots by extrapolating the kz path to zero time and The complexity of the product distribution observed expressing this value as a percentage of the total for [C~(en)~Cl(glyOCH~)]Cl~ was also qualitatively observed base consumption. To determine the found with [C~(en)~Cl(glyOCzH~)](ClO& and [Co(en),maximum amount of [Co(en)?gly12+(column 5) which Cl(glyOCH(CH3),)](C104)z,but a detailed separation of the products (apart from [C~(en)~gly]~+) was not can arise by the k , path, half the extrapolated value (column 3 ) is subtracted from unity. The fact that the attempted. In general, the bromo esters, with the exception of [C0(en)~Br(gly0CHzC~HS)IBr2, gave only observed [ C ~ ( e n ) ~ g l y ] ~production + (column 4) is trace quantities of colored eluate before [C~(en)~gly]~+.either equal to or less than the calculated figure (column 5), but within experimental error never exceeds [C0(en)~Br(gly0CH~C~H~)]Br~, alone of the bromo esters, produced significant quantities of bands 1, 2, it (column 6 ) , is in agreement with the proposal that all and 4 after hydrolysis at pH 9.0 and p = 0.1, However, the [ C ~ ( e n ) ~ g l yis ] ~formed + rapidly following release of halide. Direct confirmation of this conclusion in the less of band 4 was observed at pH 8.7; this was also case of the [C~(en)~Cl(glyOCH~)]~+ ion is provided by true for [Co(en)zCl(glyOCH~)]C12treated under both the following experiment. [C0(en)~Cl(glyOCH3)]Cl~ sets of conditions. In general, [Co(en)zBr(glyOCHwas allowed to base hydrolyze at pH 9.0, p = 0.15, 25", (CH~)Z)I(C~OJZ and [Co(en)~Br(glyOC(CH3)3)1Br~ gave for 7.7 min (-2t112 for kl rate), and rapidly adjusted to predominantly [C~(en)~gly]~+, with only small amounts pH 3. It was then eluted from the ion-exchange column of other products. using 1 M NaC104; 28x [ C ~ ( e n ) ~ g l y ]was ~ + recovered C. [Co(en),glylZ+Production. Included in Table I1 compared to 35 recovery from an identical experiment is the percentage of [ C ~ ( e n ) ~ g l y ]formed ~+ from each substrate, a value obtained by ion-exchange separation left for 82 min. D. Competition with Added Azide. The products (at pH 6 ) of the products of reaction after base confrom the base hydrolysis of [Co(en),Br(glyOCH(CH3)~)]sumption had ceased. The table shows that [Co(en),(C104),at pH 9.0 in 0.9 M N a N 3 , p = 1.0, were, in order of g1yI2+ is in all cases the major product of hydrolysis elution from a H+-form resin, trans-[Co(en)z(N&]+ (except for [C~(en)~Cl(glyOCH~)]~+ at pH 9), and the extent of its formation depends upon X and to a lesser (-17 Z,E 5 6 0 335),'O [C0(en)~N3(glyO)l+(-59 %I, [Coand unreacted [Co(en)zBr(glyOCH(en)zgly]2+ (14 degree upon R, pH, and ionic strength. In all instances less [ C ~ ( e n ) ~ g l y is ] ~ formed + when X = C1 than when N o cis-[Co(en)~(N3)~1+was (CH3)z)](C104)z (-4 X = Br and the amount increases with increasing detected. 2o The [Co(en),N3(g1yO)]+ formed was stability of the ester group toward base hydrolysis. For assigned the trans configuration on the basis of its instance, for the bromo complexes 54 and 7 3 z [Coabsorption maximum at 530 mp. cis-[Co(en)zN3(en),glyI2+ are formed in the case of the benzyl and t(glyO)]+, which was observed as a product in the base hydrolysis of the halo acids [C~(en)~X(glyO)]+ in the butyl esters, respectively, at pH 9.0. This gradation is presence of N3- ion,16has an absorption maximum at less marked, but still apparent, at pH 8.6. Also,. at
x),
(19) M. D. Alexander and D. H. Busch, Inorg. Chem., 5, 1590 (1966).
Journal of the American Chemical Society
1 91:15 1 July 16,1969
x).
(20) P. J. Staples and M. L. Tobe, J . Chem. SOC.,4803 (1960).
4107 Table VI. Kinetic Data for Oxygen Exchange between Water and the Oxygen of [Co(en)~gly]~ + Recovered from Basic Solutiona Days
Atom
z
018 b
+
Days
Atom
018
+
A. H20 €3. HzO [ C O ( ~ ~ ) ~ B ~ N H ~ C H ~ C [Co(en)zBrNHzCHzC(=O)OCH(=O)OCH(CH3)zI (CH3)zI2+ 0 0.605 Solvent 1.416 0 0.597 0 0.626 5 0.496 5 0.503 10 0.359 10 0.470 15 0.309 21 0.412 25 0.288 98 0.355 40 0.282 46 0.249 57 0.280 66 0.249
'+
+
+
0 5 10
0.683 0.613 0.607
+
G.' HI. [Co(en)2BrNHzCHzC(=O)OCH(CH3)zI '+ 10 min 0.05
a
:km
2
\
\
c
" 5
\
1
. .'
0.5
+
HzO C. HI. [Co(en)2BrNHzCHzC(=O)OCH- [ C O ( ~ ~ ) ~ N H ~ C H Z C ( =2 fO ) O ] (CH3)zI '+ 60 min 0.01 Solvent 1.491 0 0.534 7 0.434 8 0.363 10 0.345 26 0.325 40 0.332 E.d HzO [Co(en)zNHzCHzC(=0 ) 0 ] 2+
91
3
0.2 C
(A,B1
10
30
20
40
60
50
DAYS
ic I
2
6
10
14
+
Figure 2. 0 1 8 exchange in (A) [ C ~ ( e n ) ~ g l y prepared ]~+ by Hgz+FSe Hz0 induced acid hydrolysis of 018-carbonyl-labeled [Co(en)zBr[CO(~~)~NH~CH~C(=O)OCH(glyOCH(CHs)2)]Brz and recovered after 90 min at pH 8.7; and (B) (CH3)zI3+ [ C ~ ( e n ) ~ g l yprepared ]~+ by base hydrolysis of 0 18-carbonyl-labeled 0 0.699 [Co(en)?Br(gIyOCH(CH3)z)]Br2for 90 min at pH 8.7. (C) represents 15 0.711 the rate obtained by subtracting from (B) the extrapolated value for its slower rate of exchange (0.1 M HCIOI, g = 1.0,25").
[C0(en)~Br(gly0CH(CH3)~)]~+ at pH 8.6. The labeling and isolation procedures corresponding to the entries in Table VIA-G are listed in the Experimental Section. a [H+] = 1.0, p = 1.0 (NaC104), 25" [ C ~ ( e n ) ~ g l y ] ~ + 0.1 M . Represents the 0 1 8 enrichment in atom z, less the atom 0I8 Table VIA presents results following the hydrolysis in COn of normal isotopic composition (0.201). Atom 0l8 of carbonyl-018-labeled [C0(en)~Br(gly0CH(CH,)~)]= 100R/(2 + R), where R = [46]/([45] + [44]). Enrichment of (C104)Zin unlabeled water. The [ C ~ ( e n ) ~ g l yformed ]~+ initially unlabeled [C0(en)~gly]2+recovered after 60 min in HzO in the reaction was isolated and recrystallized as (ca. 1.5 atom 0l8at ) pH 8.6. [C~(en)~gly]~+prepared byHg*+induced acid hydrolysis of labeled [C0(en)~Br(gly0CH(CH~)~)]Br~, [C~(en)~gly]I~. Since a direct measure of the enrichand recovered after 90 min at pH 8.7; labeled in the carboxyl poment in the parent bromo ester complex has been shown ~+ by base hydrolysis at pH 8.6 of sition. e [ C ~ ( e n ) ~ g l y ]piepared to be impracticable,' it was necessary to determine f [Co(en)z carbonyl 0 18-labeled [CO(~~)~~~~OCH(CH~)Z)](CIO~)~. indirectly the enrichment using the Hg2+-induced gly] prepared by Hg 2+-induced acid hydrolysis of [Co(en)zBrhydrolysis reaction which gives virtually full retention (glyOCH(CH3)z)](C104)2 recovered after 10 min in HzO (ca. 1.5 atom 0l8) at pH 8.6. of the carbonyl-018label.' This resulted in a figure of 0.683 atom % (Table VIE). The first entry in Table VIA then represents 88% retention of the label in the 5 10 mp. ~rans-[Co(en)~N~(glyO)]+ was estimated either base hydrolysis experiment. The discrepancy from by atomic absorption spectroscopy or spectrophoto100% was accounted for by loss of label during the metrically, € 5 3 0 230. By using the Na+ form of the resin isolation and recrystallization procedure. Subsequent and neutral eluent, it was possible to isolate the azido ester complex in the base hydrolysis of [Co(en)ZBr- enrichments are given in Table VIA, and these are plotted against time using a logarithmic ordinate in (gly0C(CH~)~)]Br~. The products of hydrolysis were in Figure 2, curve B. Two rates of oxygen exchange are order of elution, ~rans-[Co(en)~(N~)~]+ (-2 1 %), [Coapparent and extrapolation of the slower rate to zero ~ +%), and cis(en)ZN3glyO]+(-14 %), [ C ~ ( e n ) ~ g l y ](47 time gives an intercept of 0.3 1 atom %, or 45 % of the [Co(en)~N3(glyOC(CH~)3)1~+ (-9 %, , , A, 5 10 mp). original enrichment. Subtraction of this rate from the Treatment of the cis-[C~(en)~N~(glyOC(CH~)~)]~+ soobserved curve gave a linear plot, Figure 2C, for the lution with HNOz gave [ C ~ ( e n ) ~ g l y,,A,]( ~ + 485 mp, E faster exchange with a rate constant of 2 X 98). [C0(en)~Br(gly0C(CH~)~)]Br~ was stable toward sec-1. These two rates of oxygen exchange are similar to those HNO2. The azide in the azido ester complex was previously attributed' to labeling in the carboxyl and partly displaced under the conditions of elution (-6% carbonyl oxygens, respectively. [ C ~ ( e n ) ~ g l y ]was ~ + formed from the azido ester band during elution), and the failure to detect the corTable VIB presents results for exchange in [Co(en),responding azido isopropyl ester from the H+ form of g1yI2+obtained from hydrolysis of [Co(en)ZBr(glyOCHthe column was ascribed to this behavior. (CH3)2)](C181)2in 0 '*-labeled water. The first entry represents the solvent enrichment, and the second entry E. O'*-Tracer Experiments. Table VIA-C presents data obtained for the kinetics of oxygen exchange is for [Co(en)zgly]IZ at zero time. This latter figure is 44 of the solvent value, indicating the incorporation of in [C~(en)~gly]~+, isolated from base hydrolysis of
-z z
Buckingham, Foster, Sargeson
Base Hydrolysis of [Co(en)nX(glyOR)]2+ Ions
4108
almost one oxygen atom from the solvent in the product. activity, [a]58g 300". The results require base hydrolysis The discrepancy from 50% was attributed t o slight of ( -)589-[Co(en)2Br(glyOCH3)]Br2to occur with 50 exchange occurring during recovery and recrysracemization. tallization of the [C~(en)~gly]I~. Table VIC gives results of a similar experiment, except that [C~(en)~gly]*+ Discussion was isolated and purified by a more lengthy procedure The following discussion is primarly concerned with including elution from an H+-form ion-exchange column. the path leading to the formation of [C~(en)~gly]*+. In this case the enrichment is only 3 6 % of the solvent The sections are concerned with the order of events figure, indicating more extensive oxygen exchange relating to loss of halide, ester function, and the interduring isolation. In both B and C two rates for oxygen vention of solvent in the course of the reaction. The exchange in 0.1 M H+ are apparent, and extrapolation amount of [C~(en)~gly]'+product varies between 35 of the slower rate to zero time in a similar manner to that and 76 % for the different substrates, and numerous bydepicted by Figure 2 gives for B an ordinate intercept products occur, some of which are not yet identified. of 0.34 atom % 0lx(46% of half the solvent figure) and The final section is devoted to a discussion of some of for C, 0.41 atom ZOl8(58z o f half the solvent figure). these species and their origin, but in general they are Table VID-G gives the results of some supplementary considered only in regard to their role in the production 0 la-labeling experiments. D records the observation of [C~(en)~gly]'+. that unlabeled [C~(en)~gly]'+ placed in 1.5 atom Kinetics and Competition Studies. The [Co(en)2XH 2 0l8 remains unlabeled under conditions used in the (glyOR)]*+ ions show the common property of two base hydrolysis of [C0(en)~Br(gly0CH(CH~)~)]~+. E consecutive pseudo-first-order reactions on base hydrolgives the kinetic results obtained when exclusively carysis at pH 8.6, Table 11. For [Co(en)2Br(glyOCHboxyl 0 ls-labeled [ C ~ ( e n ) ~ g l y ] ~was + permitted to ex(CH3)2)I2+ these two reactions have the rate law R = change under acid conditions after being exposed to the k[complex][OH-] (Tables IA and IB), and it seems experimental conditions of base hydrolysis. The results justified by the data in Table I1 to assume that this is a of E are also shown in Figure 2 (curve A). There is general result. The faster reaction (kl)is insensitive to effectively little exchange, indicating that the oxygen variation of the ester grouping but exhibits a fourfold label has remained intact in the Co-0 position. Similar rate increase between X = C1 and X = Br. The results for the O1s-labeled chelated ester intermediate former observation suggests little or no ester involve[C0(en)~(gly0CH(CH&)]~+are given in Table VIF. ment, since if this occurred a substantial variation in Again the method of preparation ensures that the rate with changing R group might be anticipated. For labeled oxygen is initially exclusively in the Co-0 example, benzyl acetate base hydrolyzes at a faster rate positioni6 and base hydrolysis does not result in any than methyl acetate which in turn is a factor of 100 times scrambling of the oxygen label, as seen from the lack of faster than t-butyl acetate.22 A similar large variation exchange in the [C~(en)~gly]*+ product. in rate is found in both the uncoordinatedz3 and In another experiment, Table VIG, the bromo comchelated2 amino acid esters. plexes were recovered after base hydrolysis for one Also, the spectral change (545-490 mp) for the k~ half-life in labeled solvent, pH 8.6, 25'. The ir path with [C~(en)~Br(glyOR)]~+ ions indicates loss of spectrum of this material was identical with that of the Br-. A similar result obtains for the chloro ester original [C0(en)~Br(glyOCH(CH,)2>1(C10,)~.M o r e complexes. The fourfold rate difference between X = over, elution of species sorbed on the ion-exchange Br and X = C1 correlates with base hydrolysis of other resin from the filtrate failed to reveal any bromidepentaamminecobalt(II1)-halo complexes, e.g., the [COcontaining species, indicating that the above isolation (NH&XI2+ and [ C O ( ~ ~ ) ~ N H ions.24t2j ~ X ] ~ + Also, in was essentially quantitative. Entry G indicates little the presence of Nt- as a competing anion, base hydrolor no label incorporation in this material. This ysis of [C~(en)~Br(glyOC(CH~)~)lBr~ gave 9 [Co(en)Zeliminates exchange in the monodentate ester prior to N3(glyOC(CH3)3)]2+which requires that loss of Brbromide removal and also eliminates the formation of a does, at least in part, precede ester hydrolysis for this stable ester hydrate which, if formed, would have complex. A more significant experiment in this regard incorporated one-half of the solvent enrichment. is the following (Table IV). After ~ 2 t , for / ~the initial F. Base Hydrolysis of ( -)jsg-[Co(en)2Br(glyOCH3)1- hydrolysis (k,) of [C~(en)~Cl(glyOCH~>]~+, only 5 % Br2. When ( -)ja9-[Co(en)2Br(glyOCH3)]Br2 ([a]mg [C~(en)~Cl(glyO)]+was detected on ion exchange, - 115 ") was hydrolyzed for 90 min under conditions whereas 28Z of the expected 36% [ C ~ ( e n ) ~ g l y ]was ~+ identical with those used in the OIa-tracer experiments formed. This represents close to the maximum quantity the resulting [C~(en)~gly]~' isolated by ion exchange of [C~(en)~Cl(glyO)]+ formed in the reaction, since this gave -154". This may be compared with ion is known to base hydrolyze about five times optically pure ( -)js9-[Co(en)2gly]12, [ c ~ ] j a g - 305°.21 slower than [C~(en)~Cl(glyOCH~)]~+ under the same ( -)j8g-[Co(en)2Br(glyOCH3)]Br2 was shown to be conditions.16 The final product is in the latter case optically pure by treatment with Hg2+, when ( - ) j a g exclusively [ C ~ ( e n ) ~ g l y ] ~Thus, +. no more than 10% [C~(en)~gly]?+, [ o ( ] ~ -~305 ~ ', was formed. In another is possible from the hydrolysis of of [C~(en)~Cl(glyO)]+ [aljsg experiment, a solution of (+)js9-[Co(en)~gly]I~, which allows less than 30 % [C~(en)~Cl(glyOCH~)]~+, 300', was left standing for the same time and under the of the observed [C~(en),gly]~+ to come cia this route. same conditions as for base hydrolysis of (-)jagMoreover, the latter value will represent a maximum [C~(en)~Br(glyOCH~)]Br~. It was then sorbed on and (22) M. S. Newman, "Steric Effects in Organic Chemistry," John eluted from the ion-exchange resin. The [Co(en),Wiley and Sons, Inc., New York, N. Y . , 1956. g1yl2+ in the eluate showed full retention of optical (23) R. W. Hay and L. J. Porter, J . Chem. SOC.,1261 (1967).
z
(21) I. I]+ arise from the base hydrolysis of ~is-[Co(en)~(glyO)Cl]+rather than from cis- and tr~ns-[Co(en)~(OH)(glyOR)] z+.l6 The rate law is consistent with base hydrolysis of cobalt(II1) pentaammine-type complexes, and the cumulative data support an S N ~ C mechanism B for these processes. The evidence for the existence of 496,26,27
(26) F. Basolo and R. G. Pearson, “Mechanisms of Inorganic Re-
Buckingham, Foster, Sargeson / Base Hydrolysis of [C~(en)~X(glyOR)]~~ tons
41 10
deprotonated five-coordinate intermediates is suband racemic [Co(en)~gly]~+, and examining each half stantial and allows HzO or the carbonyl oxygen to be a for the position of the 01*label. competitor for the vacated coordination position. Another mechanistic possibility is that I and 11, which O18-Tracer Studies and Mechanism. After halide could be formed in the hydrolysis of the monodentate removal, there are at least two ways in which the monoOH OH dentate ester complex can react to facilitate ester I I ,.NH,CH,C-OR NH,CH,&OR hydrolysis. One possibility is that the ester carbonyl (en),Co2+ (en),Co2+’ I oxygen occupies the vacated site in the five-coordinate ‘OH OH ‘Br OH deprotonated intermediate to form the chelated ester I I1 [C~(en)~(glyOR)]~+. This species is known to hydroester, react here exclusively by expelling coordinated lyze very rapidly in base to form [ C ~ ( e n ) ~ g l y ] ~ +HOR Br- or OH-, respectively. For example, in the case of ( k O H = 1.5 X lo6 M-I sec-I for R = CH(CH3)2).16 I, the following mechanism equates the rate of alcohol The other possibility is that water competes favorably e for the intermediate to form cis-[Co(en)z(OH)(glyOR)I2+, and that the bound hydroxide then attacks II (en)2BrCo2+-NH2CHzCOR + OHthe carbonyl center in a concerted manner. These possibilities are given in Figure 3 as paths A and B. QH tr~ns-[Co(en)~(OH)glyOR]~+ was not considered as a (en~2BrCo~t--NH,CH,bOR (1) route to [ C ~ ( e n ) ~ g l y ]since ~ + studies on the related complex trans-[C~(en)~NH,(OH)]~+ indicate isomerIQH ization to the cis ion would be a slow process. 28 r As indicated in Figure 3, one method of distinguishing between paths A and B is by using an 01*label; the QH chelated ester giving rise to Co-0 18, and intramolecular I (en)2BrCo2-NHZCH2C-OR + OH- ki attack of bound OH resulting in C=0l8, in the [CoI (en)2gly]2+product. A previous study has shown that QH these two products may be distinguished; the Co-0 oxygen exchanges slowly with the solvent in 0.1 M H+ /N? while the C=O exchanges readily. (en)zCo’t CH2 Br- f H20 (2) The results of the 0lsexperiments (Table VI; Figure 2) show that close t o one oxygen atom of the solvent is incorporated. This eliminates from further consideration entry of the ether oxygen in the intermediate, followed by alkyl-oxygen bond fission. It is clear that when c~rbonyl-O~*-[Co(en)~Br(glyOCH(CH~)~)]~+ is hydrolyzed in normal water, or unlabeled complex in 0-’ HZ0l8,enrichment occurs in both the carbonyl and ?\OR QH carboxyl oxygens of [Co(en)zgly12+. The former result gives 45 retention of the 0 l8 label in the carboxyl NH position, while the latter experiments in HZOI8give 54 (en)$o’+ CH2 + HOR + OH- (3) and 42z incorporation of the monodentate ester I I carbonyl into the Co-0 position. The -50% distri0-C bution of label requires that paths A and B of Figure 3 \Q contribute about equally to the production of [Co(en),production, Brrelease, and base consumption, and glylZ+. The result cannot be explained by intramolecsupports the observed insensitivity of kl to the nature of ular interchange between the carbonyl and carboxyl the ester. The experimental results merely require that oxygens in [ C ~ ( e n ) ~ g l y ] ~ during + hydrolysis (Table the ester hydrate intermediate not be in equilibrium VIE). Nor can it be accommodated by a scrambling with the reactants since loss of oxygen label is not mechanism during hydrolysis of the chelated ester intermediate, since carbonyl-0 18-[Co(en)z(glyOCH- observed. The rate of Co-Br bond.cleavage must also exceed that for C-OR, since no [C~(en)~Br(glyO)]+ is (CH3),)I3+ shows no oxygen interchange on base formed. Reaction 1 cannot be rate determining, since hydrolysis (Table VIF). this would predict a reaction rate which was sensitive The 50% incorporation of label (R = CH(CH&) to the ester grouping and insensitive to the nature of the compares with the 50% racemization (R = CHs) in coordinated halide. However, the bromo-containing the [ C ~ ( e n ) ~ g l y ]product. ~+ Additional work using the species collected after one half-life were shown to be [Co(trien)Cl(gIyOCzHj)12+speciesI3 suggests that this is identical with the starting material and incorporated no not a coincidence and that ( +)S8s-[Co(en)zgly]2+may solvent oxygen. This result requires that solvent arise solely from path A. This aspect of the probparticipation in hydrolysis occur subsequent to Brlem is being tested by hydrolyzing ( -)js9-[Co(en)zBrrelease, and the the ester hydrate (glyOCHa)]2+ in labeled solvent, separating the active
I
+
+
/
16) competes effectively with external hydroxide ion for lysis of the ester group at pH 9. It is not surprising then that bound OH(pK, 6) in the hydroxo complex also competes efficiently with external OH- for the same site. The enormous difference in concentration between coordinated NH2- and OH- compensates for any difference in nucleophilic character of the two species. Reaction By-products. The identification of some of the other products of hydrolysis (Table IV) is uncertain, but in view of the preceding discussion we suggest that the ion separated as band 4, Table IV, could be the
II slow ( ~ ~ I ) ~ B ~ C O ~ ~ N H , C H+ ~ C O RB 7 0
II
NHzCH+2OR
+Br-
(en)zCo2+’
(4)
‘OH
+
?H
(5)
II
-
OH
I
f89t_
NH2CHzC-OR (en),&#+’
‘OH
I
QH (6)
b-C-OR
I
@H
NH2
0
Path A
(en),CoQ+
