Physical organic chemistry of transition metal carbene complexes. 3

Physical organic chemistry of transition metal carbene complexes. 3. Thermodynamic and kinetic acidities of chromium carbenes (CO)5Cr:C(OCH3)CH3 and ...
0 downloads 0 Views 767KB Size
12526

J . Am. Chem. SOC.1993,115, 12526-12532

Physical Organic Chemistry of Transition Metal Carbene Complexes. Thermodynamic and Kinetic Acidities of (C0)5Cr=C(OCH3)CH3 and (C0)5Cr=C(OCH3)CH2Ph in Aqueous Acetonitrile Claude F. Bernasconi* and Weitao Sun Contribution from the Department of Chemistry and Biochemistry of the University of California, Santa Cruz, California 95064 Received August 23, 1993'

Abstract: Rate constants for proton transfer from (methoxymethylcarbene)pentacarbonylchromium(O), (CO)5Cr=C(OCH3)CH3, to OH- and amine bases and from (benzylmethoxycarbene)pentacarbonylchromium(O), (CO)sCr=C(OCH3)CH2Ph, to OH-, amines, carboxylate ions, and H2O were determined in 50% acetonitrile-50% water (v/v) at 25 'C. Intrinsic rate constants (k,) were deduced from extrapolations of Bransted plots. For (CO)5Cr=C(OCH3)CH3 they are log k, = 3.70 (secondary amines) and 3.04 (primary amines) and for (CO$3=C(OCH3)CH2Ph log k, = 1.86 (secondary amines), 1.51 (primary amines), and -0.8 (RCOO-). Theselog k,values are consistent with significant resonance stabilization of the respective anions. Kinetic isotope effects on the order of 2.5-3.0 for the deprotonation of either metal carbene complex by OH- and of 5.6 for the deprotonation of (CO)SCr==C(OCH,)CHZPh by piperidine were measured. These relatively low values may indicate substantial coupling of proton transfer to heavy atom motion, including bond changes in the C O ligands. The effect of changing solvent from water to 50% acetonitrile-50% water is to induce 2-fold and 4.8-fold increases in the rate of deprotonation of ( C O ) S C F C ( O C H ~ ) C H ~by OH- and piperidine, respectively. Possible reasons for these solvent effects, including electrostatic stabilization of the transition state of the piperidine reaction, are discussed.

It has been known for over 25 years that (methoxymethylcarbene)pentacarbonylchromium(O), l-CH3, a prototype Fischer carbene complex,2is a rather strong carbon acid. The first relevant

on a stopped-flow kinetic determination of the rate of deprotonation of 1-CH3 by OH-.* Because of the instability of 1-CH3 in basic solution-at 0.1 M K O H the decomposition half-life is less than 1 s-rates of proton transfer could only be determined with highly basic proton acceptors (OH-, piperidine) that allow measurements in the direction l-CH3 l-CHZ-. Measurements in the direction of 1-CH2- l-CH3 would be highly desirable in order to expand the range of proton acceptors amenable to study. Such measurements require a capability of reacting 1-CHz- with acid within a second after it has been formed by deprotonation of 1-CH3with OH-, Le., before decomposition sets in. Such a capability has recently been acquired in our laboratory. This paper presents the results of such a study and also of the proton transfer kinetics of (benzy1methoxycarbene)pentacarbonylchromium(O), 1-CHzPh. Due to the low solubility of 1-CHzPh in water, our investigation was carried out in 50% acetonitrile 50% water (v/v).

- -

report was Kreiter's observation of a rapid conversion of I-CH3 to 1-CD3 in a dilute NaOCHs/CH3OD solution.3 Casey and Anderson4 subsequently showed that in T H F the acidity of l-CH3 is approximately the same as that ofp-cyanophenol, an acid whose pKa in water is 7.97.5 In quoting this work, numerous authors have, rather misleadingly, referred to 1-CH3 as a carbon acid with a pKa of 8 which of course cannot be correct since p-cyanophenoxide ion and l-CHz- should respond quite differently to a solvent change from T H F to water.6 The pKa of l-CH3 in water was reported in 1989 to be 12.3.8 A possible reason why the pKa in water has not been determined much earlier is the instability of basic aqueous solutions of 1-CHs9 which precludes equilibrium measurements by classical spectrophotometric procedures. The reported pKa in water was based Abstract published in Advance ACS Abstracts, December 1, 1993. (1) This is part 3 in this series. Part 2: Bernasconi, C. F.; Stronach, M. W. J . Am. Chem.Soc. 1993, 115, 1341. (2) DBtz, K. H.; Fisher, H.; Hofmann, P.; Kreissl, F. R.; Schubert, U.; Weiss, K. Transition Metal Carbene Complexes; Verlag Chemie: Deerfield Beach, FL, 1983. (3) Kreiter, C. G. Angew. Chem., Int. Ed. Engl. 1968, 7, 390. (4) Casey, C. P.; Anderson, R. L. J . Am. Chem. SOC.1974, 96, 1230. (5) Fickling, M. M.; Fischer, A,; Mann, B. R.;Packer, J.; Vaughan, J. J . Am. Chem. SOC.1959, 81, 4226. (6) A pK, = 8 for 1-CH3 has even been quoted in a highly acclaimed text.' (7) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G . Principles and Applications of Organotransition Metal Chemistry; University Science Books: Mill Valley, CA, 1987; p 806. (8) Gandler, J. R.; Bernasconi, C. F. Organometallics 1989, 8, 2282. (9) This contrastswith thesatisfactory stability ofthe l-CHs/l-CHz-system in THF.4

0002-7863/93/1515-12526$04.00/0

,WH3

(CO),Cr=C, CH2Ph

lCH2Ph

Results General Features. When I-CH3 is placed into a KOH solution two reactions are observed. The faster of the two is in the millisecond time range. Figure 1 shows the spectral changes associated with this process as measured in a stopped-flow spectrophotometer. Addition of HCl to the reaction solution 150 ms after the reaction is initiated regenerates 1-CH3 as determined by its UV/vis spectrum and confirmed by an HPLC analysis. Similar observations were made with 1-CHzPh(Figure 2). Thespectral changes are attributed toreversibledeprotonation of the carbene complexes, a conclusion for which more evidence will be derived from the kinetics results. With both carbene 0 1993 American Chemical Society

J . Am. Chem. SOC.,Vol. 115, No. 26, 1993

Transition Metal Carbene Complexes

40

1

12521

A

30

m '

20

10

mmm3am3804D42D 0

Wavelength, nm

0

Figure 1. Time-resolvedspectra of the proton transfer reaction of l-CH3 (1.20 X lo-' M) with 0.1 M KOH in 50% acetonitrile50%water (v/v) at 25 "C.The first spectrum was obtained within 1 ms after mixing and

subsequent spectra at 10"

0.02

0.04 0.06 [KOHI, M

0.08

0.1

Figure 3. Plot of k o w vs [KOH] for the proton transfer reaction of with KOH in 50% acetonitrile-50% water (v/v) at 25 "C.

intervals.

12

I

33

350

370

390

410

430

Wavelength, nm

Figure 2. Time-resolvedspectra of the proton transfer reaction of 1-CH2Ph (1.20 X lo-' M) with 0.04 M KOH in 50% acetonitrile-50% water (v/v) at 25 O C . The first spectrum was taken within 2.5 ms after mixing

0

200

and subsequent spectra at 100-ms intervals.

[KOHI-I, M-'

complexes the deprotonation is followed by a slower process with a half-life of several seconds. A detailed kinetic investigation of this slower process, which leads to the formation of acetaldehyde in the case of l-CH3 and to 0-methoxystyrene in the case of l-CHZPh, will be the subject of a future report. Kinetics. Most experiments were performed in 50% acetonitrile-50% water (v/v) at 25 O C and an ionic strength of 0.1 M maintained with KCl. Rates were measured in K O H solutions and various amine buffers; with 1-CHZPh kinetic determinations were also done in carboxylate buffers. Pseudo-first-order conditions, with the carbene complex as the minor component, were used throughout. Under these conditions the reaction may be represented by eq 1 with the observed pseudo-first-order rate

qo + kPHIOH-I + kg[Bl 1-CHZR

+

kyaH+ +'*!k

1-CHR

7

(1)

k!/'[BH]

constant for equilibrium approach being given by eq 2.

= kyO

+ kyHIOH-] + k?O + k2aH++ k;[B] +

-

k!r[BHl (2) l-CH3. Runs with K O H were conducted in the I-CH3 l-CH2- direction. The raw data are summarized in Table S 1 of the supplementary materialalo Figure 3 shows a plot of k,W vs [OH-] which yields kyH = slope = 456 f 12 M-l s-1, kFo = intercept = 0.91 f 0.09 s-l, and = kpH/kFo = 501 f 71 (10) See paragraph concerning supplementary material at the end of this

eH

paper.

Figure 4. Plot of AOD,/AOD vs l/[KOH] according to eq 3 for the

proton transfer reaction of l-CH3 with KOH in 50% acetonitrile501 water (v/v) at 25 "C.

eH

M-I. An independent check of is provided by measuring the absorbance change (AOD) induced by the reaction as a function of [OH-]. These AOD measurements are included in Table S 1.IO AOD is related to by eq 3 with AOD, being the absorbance

eH

-AOD, AOD

-

1 [OH-]

eH

-t

(3)

change at high [OH-] where the equilibrium strongly favors l-CHZ-. Figure 4 shows a plot according to eq 3 from which = 289 f 20 M-1 was determined. In view of the fact that stock solutions of 1-CH3are not entirely stable (see Experimental Section) and the absorbances at the end of the reaction are based on an extrapolation because of the decomposition process referred to earlier, the agreement between the two values is remarkably good. In our subsequent discussion the kinetically determined value will be used. The reaction with piperidine was measured by two methods. In the first, 1-CH3 was mixed with piperidine solutions that contained 0.1 M KOH, Le., the reaction was conducted in the 1-CH3 l-CH2- direction. In the second, the reaction was conducted in the l-CHS1-CH3 direction; Le., 1-CHz- was generated in a 0.02 M K O H solution and ca. 1 s later was mixed with 1:l piperidinebuffersat p H 11.01 ("pH-jumpexperiments"). In contrast to the experiments with OH- as the base, the absorbance changes were independent of amine concentration, an observation which is consistent with proton transfer as elaborated upon in the Discussion. Plots of k o u vs piperidine

eH

eH

(R = H or Ph)

,k

600

400

-

-

Bernasconi and S u n

12528 J. Am. Chem. SOC., Vol. 115, No. 26, 1993

Table 11. Effect of Changing the Solvent from Water to 50% Acetonitrile-50% Water and the Temperature from 20 to 25 "C on the Reaction of 1-CH3 with OH- and Piperidinea

[piperidine], M

0

0.0075

0.0025

0.01

base OHOHOHPip Pip Pip

/$

300

ap

100

0 [piperidine], M

Figure 5. Plots of k , vs~[piperidine] for the proton transfer reaction of l-CH3 with piperidine: (0)in 0.1 M KOH, bottom x axis; (0)at pH 11.01, top x axis.

solvent H20 H20 50%MeCN H20 H20 50% MeCN

Base n-BuNH2 MeOCH2CH2NH2 H2NCOCH2NH2 piperidine piperazine HEPAC OH-

peH

10.40 9.39 8.14 11.01 9.97 9.33 16.64d

k:, M-1 s-L 122 22.6 5.01 906 f 21 246 85.5 456 f 1ZC

1

1

M-l

25

25 20 25 25

lo4 lo4 los lo4 lo4 lo5

0.1 M (KCI), p e H = 12.50. Error limits are standard deviations in the slopes of koM vs [BH] for amines and koM vs [OH-] in OH- reaction. HEPA = N(2-h droxyethy1)piperazine. pK, = 15.19, see text. e ky = kyH;k!," = kFg. /In units of s-I; error limits are standard deviation of the intercept of the plot of k o vs~[OH-].

slope = ky

eH

eH

eH/eH.

-

(1 1) Allen, A. D.; Tidwell, T. T. J . Am. Chem. Soc. 1987, 109, 2774. (12) Barbosa, J.; Sanz-Nebot, V. Anal. Chim. Acra 1991, 244, 183.

11.12

11.01

peH 12.3 12.8 12.5

ky, M-I s-l

12.3

12.8 12.5

135b 229 456 69b 188 906

( );;

+ k1 BHuH+ -= k r 1 + K;H

(4)

plot of slope vs a H + is shown in Figure 6; it yields ky = 75.5 f 1.6 and p e H = 10.40 f 0.05. The p e H value was confirmed by measurements of absorbance changes as a function of pH. The relevant relationship is given by eq 5. Average AOD values

-AOD, AOD

"p =

concentration for the two kinds of experiments are shown in Figure 5; the raw data are given in Table S2.10 The slope of the first plot (906 M-I s-l) is approximated by ky (k!?[BH]