4620
`
`J. Am. Chem. Sot. 1987, 109, 4620-4622
`Reversible Formation of Intermediates during H30+-Catalyzed
`Hydrolysis of Amides. Observation of Substantial '*O
`Exchange Accompanying the Hydrolysis of Acetanilide and
`N- C yclo hex ylacet amide
`
`H. Slebocka-Tilk,t R. S. Brown,*+ and J. Olekszyk*
`Contribution from the Department of Chemistry, University of Alberta,
`Edmonton, Alberta T6G 2G2. Canada. Received April 22, 1986
`
`Abstract: Careful mass spectrometric analysis of the I8O content of N 50% enriched acetanilide (2) and N-cyclohexylacetamide
`(3) recovered from acidic media during the course of hydrolysis reveals that both species suffer I8O loss. The percent of l8O
`exchange per tl12 of hydrolysis increases as [H30+] decreases. For 2 at 72 O C the amount of exchange increases from 0.5
`f 0.5% (per t l 1 2 ) in 1 M HCI to 9.4 i 0.5% in glycine buffer, [H30+] = 0.003 M. For 3 at 100 O C the exchange is 1.05
`f 0.3% (per t I I 2 ) at 1 M HC1 and 9.0 f 0.4% in 0.01 M HCI. When these data are used to compute k,, (the exchange rate
`constant), it shows a first-order dependence on [H,O+] followed by a plateau at high [H,O+] for both 2 and 3.
`
`Acid-catalyzed amide hydrolysis is generally considered to
`proceed by pathways involving irreversibly formed tetrahedral
`intermediates.' Numerous '*O-exchange studies with labeled
`benzamides2 and one with a ~ e t y l i m i d a z o l e ~ indicated that no
`detectable I8O loss occurs in recovered starting material during
`the course of hydrolysis in strongly acidic media. However,
`McClelland observed that 90% I80-enriched benzamide exhibited
`0.2% exchange per t I l 2 hydrolysis in 5.9% H2S04 a t 8 5 OC.
`Apparently, further increases in [H30+] do not lead to additional
`I8O exchange, a t least for b e n ~ a m i d e s . ~ ~ , ~ ~
`Recently we reported that substantial "0 loss accompanies
`H30+-catalyzed hydrolysis of labeled 1 between pH 4-5.5 The
`major question is whether t h e exchange is a consequence of the
`unorthodox geometry of l6 or is a fairly general phenomenon of
`
`2
`
`3
`
`1
`acid-catalyzed hydrolysis of amides when conducted a t low
`[H,O+].' Herein we report that two typical amides, acetanilide
`(2) and N-cyclohexylacetamide (3), indeed show increasing ex-
`change when hydrolyzed a t progressively lower [H30f]. These
`unprecedented observations are not easily reconciled in terms of
`the currently accepted mechanism of [ H30+] -catalyzed amide
`hydrolysis.
`Experimental Section
`Materials. Acetanilide was commerical (Aldrich). N-Cyclohexyl-
`acetamide was prepared by acetylation of cyclohexylamine. Both ma-
`terials were recrystallized from HzO. Labeled materials were prepared
`as follows.
`Acetyl chloride (0.785 g, 0.71 mL, 1.0 mm) was placed in a dried
`5-mL flask equipped with a microdistillation apparatus protected from
`moisture and with the thermometer replaced by a serum stopper. The
`flask was then cooled to -20 "C and 0.19 mL of H2180 (97.2 atom % I8O,
`Cambridge Isotope Laboratories) added dropwise via syringe. The flask
`was warmed to 0 'C, and a slow but exothermic reaction occurred. After
`30 min at room temperature, the flask was recooled to 0 OC and 0.45 g
`(0.29 mL) of freshly distilled PCI, added. The mixture was allowed to
`come to room temperature, and after 1 h, two layers had formed. The
`'*O-labeled acetyl chloride was distilled directly into a flask held at 0 O C
`and containing 2.2 mmol of the desired freshly distilled precursor amine.
`After extraction with methylene chloride and washing the CH2Cl2 layer
`with dilute HC1 and then saturated NaC1, the organic layer was dried
`
`+Department of Chemistry.
`*Mass Spectrometry Laboratory, Department of Chemistry.
`
`over MgSO,. Following filtration, the volatiles were removed, yielding
`essentially a quantitative yield (based on 1 equiv of amine) of amide
`(50-55% I8O enriched). The enriched amides were subsequently purified
`by recrystallization from HzO.
`Kinetics. (a) Hydrolysis. Pseudo-first-order rate constants (khyd) were
`M
`obtained by observing the rate of decrease in absorbance of 1 X
`aqueous solutions of the acetanilide (2) at the wavelength of maximum
`change (240 nm) by using a Cary 210 UV-vis spectrophotometer in-
`terfaced as previously described.8 Stock aqueous solutions of 2 (0.01 M)
`were prepared, and reactions were initiated by injecting 30 pL of the
`solution into a 1-cm quartz cuvette containing 3.0 mL of HCI or buffer
`solution (fi = 1.0 M, KCI) that had been equilibrated for 30 min at 72
`O C in the cell holder. Rate constants were obtained by nonlinear least-
`squares fitting of the Abs vs. time curves to a standard exponential model
`( A , = A,, + (Alo - A,m)e-k'). The reactions were normally followed to
`completion except for the slowest, which were followed to two half-times.
`In all cases they exhibited clean first-order kinetics.
`The rate of hydrolysis of N-cyclohexylacetamide was determined by
`an NMR technique9 using a Bruker WH-200 spectrometer. A series of
`0.05 M samples (0.02 M in the case of the lowest acid concentration)
`thermostated at 100 OC in HCI solutions (1.0, 0.5, and 0.05 N HC1; fi
`= 1.0 M KC1) were analyzed at various times to determine the solution
`composition. Immediately prior to NMR analysis, one drop of D20 was
`added to -300 pL of the solution. Careful integration of the O=C-
`CH, and >CH-N
`signals in both starting material (6 1.88 and 3.4,
`respectively) and products (6 1.84 and 2.9, respectively) provided the
`relative amounts of each material. The rate constants were calculated
`from the slope of the plot of In [ ( I amide)/([ amide + I product)] vs.
`time, where I is the integrated intensity of the appropriate peak (methyl
`or methine).
`(b) I8O-Exchange Kinetics. A typical exchange experiment was con-
`ducted as follows. A 30-mL solution of amide 2 or 3 (0.005-0.001 M,
`p = 1.0 M KCI, 50-55% I8O enriched) was divided into three portions,
`each being sealed in an ampule. The ampules were thermostated at 72
`OC (for 2) or 100 OC (for 3 ) for a time corresponding to the half-time
`(t1,2) for hydrolysis. Samples were withdrawn from the temperature bath
`and plunged into an ice bath. These were then opened and extracted with
`3 X 10 mL of purified CHzC1,, and the combined extracts were washed
`
`( I ) (a) Bender, M. L. Chem. Rev. 1960,60, 53. (b) O'Connor, C. J. Q.
`Rev. Chem. SOC. 1971, 24, 553. (c) Deslongchamps, P. In Stereoelecrronic
`Effecrs in Organic Chemislry; Pergamon: Oxford, 1983; pp 101-162.
`(2) (a) Bender, M. L.; Ginger, R. D. J. Am. Chem. SOC. 1955, 77, 348.
`(b) Smith, C. R.; Yates, K. [bid. 1972, 94, 8811. (c) Bunton, C. A.; 0'-
`Connor, C. J.; Turney, T. A. Chem. Znd. (London) 1967, 1835. (d) Bunton,
`C. A,; Farber, S. J.; Milbank, A. J. G.; O'Connor, C. J.; Turney, T. A. J .
`Chem. SOC., Perkin Trans. 2 1972, 1869.
`(3) Bunton, C. A. J. Chem. Soc. 1963, 6045.
`(4) (a) McClelland, R. A. J. Am. Chem. SOC. 1975, 97, 5281; (b) 1978,
`100, 1844, cited as footnote 47.
`(5) Somayaji, V.; Brown, R. S . J . Org. Chem. 1986, 51, 2676.
`(6) Skorey, K. I.; Somayaji, V.; Brown, R. S.; Ball, R. G. J . Org. Chem.
`1986, 51, 4866.
`(7) OH--promoted hydrolysis of amides is well-known to promote large
`amounts of exchange.l,2 The conditions herein include only the H+ domain:
`[H,O+] > lo-,.
`(8) Brown, R. S.; Ulan, J. G. J. Am. Chem. SOC. 1983, 105, 2382.
`(9) Williams, A. J. J . Am. Chem. SOC. 1976, 98, 5645.
`
`0002-7863/87/1509-4620$01.50/0
`
`0 1987 American Chemical Society
`
`SENJU EXHIBIT 2076
`LUPIN v. SENJU
`IPR2015-01099
`
`Page 1 of 3
`
`

`
`H30+- Catalyzed Hydrolysis of Amides
`
`J. Am. Chem. SOC.. Vol. 109, No. 15, 1987 4621
`
`[Buffer]" (M)
`HCI. 1.0
`
`HCI, 0.5
`
`HCI, 0.1
`
`HCI, 0.05
`
`khyd
`(s-l x 106)b
`262 f 1
`
`146f 1
`
`34.1 f 0.1
`
`16.1 f 0.1
`
`mean (%)
`54.54 f 0.10
`
`54.45 f 0.07
`
`53.32 f 0.20
`
`51.65 f 0.15
`
`0.6 f 0.4
`
`2.7 f 0.7
`
`5.7 f 0.6
`
`13.5 f 9.1
`
`13.5 f 3.5
`
`13.7 f 1.5
`
`51.43 f 0.20
`
`51.42 f 0.10
`
`49.22 f 0.12
`
`49.20 f 0.14
`
`49.65 f 0.15
`
`49.66 f 0.11
`
`6.1 f 0.7
`
`6.1 f 0.5
`
`7.7 f 0.5
`
`7.7 f 0.6
`
`9.4 f 0.6
`
`9.4 f 0.5
`
`9.5 f 1.10
`
`9.2 f 0.84
`
`4.1 f 0.28
`
`4.1 f 0.34
`
`1.64 f 0.13
`
`1.5 f 0.36
`
`Table I. Pseudo-First-Order Hydrolysis Rate Constants and IsO-Exchange Data per t 1 / 2 Hydrolysis for Acetanilide at 72 OC, p = 1.0 M KCI
`% I8O found'
`kex
`% ' 8 0 e
`(s-1 x 1 0 7 ~
`at t l / 2
`exchange/t,/2
`17.9 f 18.0
`54.57 f 0.07
`0.5 f 0.5
`54.50 f 0.07
`54.45 f 0.07
`54.44 f 0.03
`53.40 f 0.10
`53.25 f 0.10
`51.70 f 0.10
`51.60 f 0.10
`51.54 f 0.10
`51.32 f 0.08
`51.42 f 0.04
`51.41 f 0.09
`49.16 f 0.06"
`49.29 f 0.03"
`49.22 f 0.12'
`49.18 f 0.08'
`49.61 f 0.05
`49.69 f 0.1 1
`49.66 f 0.1 1
`
`10.4 f 0.1
`
`10.0 f 0.1
`
`3.51 f 0.01
`
`3.56 f 0.01
`
`1.15 f 0.02
`
`1.07 f 0.05
`
`glycine, 0.4
`[H+] = 0.032
`glycine, 0.1
`[H'] = 0.032
`glycine! 0.4
`[H+] = 0.01
`glycine," 0.2
`[H+] = 0.01
`glycine, 0.4
`[H*] = 0.003
`glycine, 0.2
`lH+l = 0.003
`[H30+] determined at 25 OC in the case of glycine buffers. bDetermined by observing rate of change of absorbance at 240 nm in duplicate. Rate
`constants derived from nonlinear least-squares fitting of Abs vs. time data to standard exponential model. Error limits from least-squares standard
`deviations. 'Initial I8O content determined by isolation of labeled acetanilide at time 0 from 1 N HCI solution (54.73 f 0.09; 54.87 f 0.08; mean
`= 54.80 f 0.15). "New sample labeled acetanilide used in this run; I8O content by isolation from glycine at time 0 (53.37 f 0.08; 53.30 f 0.09;
`mean = 53.34 f 0.14). eNormalized to 100% I8O at time 0. Error limits calculated as sum of standard deviations of the mean plus that of time 0
`sample normalized to 100%. /Calculated from percent I8O content at
`hydrolysis as in text; error limits are cumulative sums of standard devia-
`tions in "0 content and khyd.
`
`Table 11. Pseudo-First-Order Hydrolysis Rate Constants and I80-Exchange Data per tIl2 Hydrolysis for N-Cyclohexylacetamide at 100 OC, p =
`1.0 M KC1
`
`[HCII (M)
`1 .o
`
`khyd
`( S d x 106)O
`53 f 2
`
`27.7 f 1.0
`
`2.85 f 0.04
`
`0.5
`0.2
`
`0.1
`
`0.05
`
`0.02
`
`0.01
`
`%
`
`foundb
`at t1/2
`50.11 f 0.04
`50.13 f 0.02
`
`mean (%)
`50.12 f 0.05
`
`% 1 8 0
`exchangec/tli2
`1.05 f 0.3
`
`kex
`(s-1 x 108)"
`80.4 f 27.0
`
`49.37 f 0.13
`
`48.20 f 0.17
`
`47.85 f 0.20
`
`46.80 f 0.20
`
`46.10 f 0.10
`
`2.5 f 0.5
`
`4.8 f 0.5
`
`5.5 f 0.6
`
`7.6 f 0.6
`
`9.0 f 0.4
`
`42.1 f 10.5
`
`40.7 f 6.3
`
`23.4 f 3.7
`
`13.0 f 1.7
`
`7.7 f 0.7
`
`49.45 f 0.05
`49.30 f 0.04
`48.19 f 0.04
`48.21 f 0.16
`47.80 f 0.10
`47.90 f 0.15
`46.70 f 0.10
`46.90 f 0.10
`46.08 f 0.05
`46.10 f 0.10
`'khyd determined by 'H NMR analysis according to the method of Williams.' bSample separated at time 0 from 1 M HCI (50.72 f 0.04; 50.59
`f 0.03; mean = 50.65 f 0.1 I). cNormalized to 100% 180-enriched sample at time 0. Error limits calculated as sum of standard deviations of the
`mean plus that of time 0 sample normalized to 100%. "Calculated from percent l 8 0 content at il,z hydrolysis; where kh,, is not given it was
`calculated assuming a first-order dependence in [H30+] and a f5% error, which is factored into kex.
`T h e depletion is expressed as percent of '*O exchange per
`hydrolysis (normalized to 100% enrichment a t zero time) in
`column five. Given in column six a r e the k,, values, calculated
`according to kextllz = In (a/a - x), where a and a - x a r e the I 8 0
`contents a t zero time and
`respectively. T h e error limits in
`k,, a r e calculated based on the cumulative standard deviations
`in both '*O contents and khyd. T h e errors a r e largest a t low
`amounts of exchange and less so a t high amounts, but their in-
`clusion does not alter the conclusion that the amount of exchange
`increases a s [H,O+] decreases.
`Finally, shown in Figure 1 is a plot of log k,, and log khyd vs.
`-log [H30+] for both amides. T h e point of note is that the log
`k,, values for both amides tend to plateau a t high [H30+] but
`tend to a first-order dependence a t low [H30+].'o~1'
`
`with saturated NaCl until the aqueous layer was neutral. The organic
`layer was dried (MgSOJ and stripped of solvent to yield a residue which
`was subjected to direct mass spectrometric analysis with an AEI MS-12
`low-resolution mass spectrometer. The I 8 0 content of the reisolated
`material was calculated as (IM++2)/(I~++2 + IM+), where I is the peak
`intensity of the parent and enriched parent ions. Nine to sixteen separate
`determinations of the M+ and M+ + 2 intensities were recorded. Primary
`data are given in the supplementary material (Tables 1 s and 2s). Values
`given in Tables I and I1 are the averages of two independent determi-
`nations along with the cumulative standard deviations. As a check to
`exclude anomalous exchange during the extraction and analysis proce-
`dure, the I8O content of three independent samples removed at t = 0 was
`determined and compared with that of authentic material: in no case was
`the I8O content different within the experimental accuracy.
`Results
`Given in Tables I and I1 are hydrolytic rate constants and mass
`labeled 2 and 3
`spectrometric '*O-exchange data for -50%
`reisolated from solution a t the hydrolytic tljz a t various [H30+].
`Duplicate isolation experiments were performed, and t h e error
`limits quoted in column three of Tables I and I1 a r e the standard
`deviations of 9-16 scans of the M+ and M+ + 2 peaks. From the
`mean values in column four it is readily seen that a n increasing
`depletion of l 8 0 content occurs at lower [H30+] for both amides.
`
`(10) As with other reported amides,'b,2d,1' the khyd values for 2 and 3 show
`significant deviations from linearity at high [H30+]. In the regions where
`significant increases in I8O exchange occur, a first-order dependence of khyd
`on [H+] obtains.
`(11) (a) Barnett, J. W.; Hyland, C. J.; O'Connor, C. J. J . Chem. Soc.,
`Chem. Commun. 1972, 720. (b) Barnett, J. W.; O'Connor, C. J. J. Chem.
`Soc., Perkin Trans. 2 1972, 2378; 1973, 220. (c) Modro, T. A,; Yates, K.;
`Beaufays, F. Can. J . Chem. 1977, 55, 3050.
`
`Page 2 of 3
`
`

`
`-
`
`1
`
`-5.00
`
`
`
`A
`
`8
`
`8
`
`4622 J. Am. Chem. SOC., Vol. 109, No. 15, 1987
`-4.00 1
`
`w
`I
`R
`A -6.00
`W
`
`-7.00 1
`-8.00 ’ , I
`-3.00 1
`-4.00 :
`
`-0150
`
`. , I I I I , I I I
`
`0.50
`
`. .
`I I I I I , . .
`I * I I I
`1 .io
`-log[H+]
`
`I I
`2.50
`
`n
`w
`I
`m
`
`.j;;-5.00 :
`
`M
`0
`
`r(
`
`-6.00
`
`Figure 1. Plots Of log khyd (*) and log. k,, (0) VS. -log [H,O+] for
`N-cyclohexylacetamide (A) and acetanilide (B) determined at 100 and
`72 “C, respectively, p = 1 .O M KCI. The k,, values are calculated from
`percent ‘*O content at tllz hydrolysis (see text). Dashed error bars in
`exchange data of B indicate there is no satisfactory lower limit due to
`error limits exceeding the value of keX. Straight lines through khyd data
`are unit slope first-order dependence on [HsO+].
`Discussion
`Changes in C-O/C-N cleavage ratios as a function of pH have
`been noted in acid-catalyzed hydrolyses of certain imidate estersI2
`
`and amide a c e t a l ~ ~ ~ , ’ ~ and have been explained in terms of the
`involvement of tetrahedral intermediates differing in the site and
`state of protonation. In those cases, the C-O/C-N cleavage ratio
`increases at lower [H30+]. Although phenomenologically a similar
`situation is observed with 2 and 3, C-O cleavage regenerates amide
`(labeled or unlabeled), which ultimately hydrolyzes.
`With the exception of McClelland’s observations with benz-
`
`amide (0.2% I8O l o ~ s / t ~ , ~ , 5.9% H2S04, 85 oC),4a the occurrence
`of I8O exchange accompanying acid-catalyzed amide hydrolysis
`has not been demonstrated. Such exchange is well documented
`
`(12) (a) Smith, V. F.; Schmir, G. L. J. Am. Chem. SOC. 1975, 97, 3171.
`(b) Caswell, M.; Schmir, G. L. Ibid. 1979,101,7323. (c) Lee, Y . N.; Schmir,
`G. L. Ibid. 1979,101,6277. (d) Chaturvedi, R. K.; Schmir, G. L. Ibid. 1968,
`90, 44 13.
`(13) McClelland, R. A.; Patel, G. Ibid. 1981, 103, 6908.
`
`Slebocka- Tilk et al.
`
`in base hydrolysis,’J4 as well as in both acid and base hydrolysis
`of carboxylic esters,l~’~ and has been traditionally interpreted as
`implying the intermediacy of reversibly formed tetrahedral in-
`termediates.
`The generally accepted mechanism for amide hydrolysis in acid
`involves H 2 0 attack on an 0-protonated
`to produce a
`tetrahedral addition intermediate which undergoes rapid N-
`protonation and subsequent irreversible C-N cleavage (eq 1).
`
`HO
`
`/ I +
`H
`Our present results require that there be at least one inter-
`mediate (not necessarily given in eq 1) that is in equilibrium with
`starting amide and allows oxygen exchange. There are two major
`considerations in both quantitating the exchange data and relating
`it to the hydrolytic process. The first assumes that the inter-
`mediates are at equilibrium with respect to proton transfer. Thus,
`if there is a reversibly formed amide hydrate, both oxygens have
`an equivalent probability for loss (exclusive of C-’60/C-’80
`kinetic isotope effects). The fact that changes in [glycine buffer]
`at low [H30+] affect neither kbyd nor k,, suggests the various
`intermediates are at equilibrium with respect to proton transfer,
`at least in the case of acetanilide.
`A second and perhaps more serious assumption is that the
`intermediate leading to exchange is on the hydrolytic
`Inasmuch as microscopic adherance to eq 1 requires that the
`transition states leading to C-0 or C-N cleavage each have the
`same molecular composition, (H+, OHz, amide), with that scheme
`it is difficult to explain why the k,, and khyd rate constants diverge
`as a function of [H,O+]. Perhaps this indicates that there are
`two parallel processes, one leading to exchange and another to
`hydrolysis.
`More work is clearly required to determine the scope, limitation,
`and structural constraints on the exchange process prior to pro-
`posing a scheme which explains these findings. Nevertheless, the
`observation of significant exchange accompanying acid-catalyzed
`hydrolysis of these two amides challenges our current under-
`standing of this important process.
`Acknowledgment, We thank the University of Alberta and the
`Natural Sciences and Engineering Research Council of Canada
`for generous financial support. We are particularly grateful to
`the referees of this manuscript and to Professor R. L. Schowen,
`whose inciteful comments prompted a reevaluation of the phe-
`nomenon of I8O exchange accompanying amide hydrolysis.
`Supplementary Material Available: Tables 1s and 2s of original
`mass spectrometric intensity data for 2 and 3 at various [H30f]
`(10 pages). Ordering information is given on any current
`masthead page.
`
`(14) (a) Bunton, C. A,; Nyak, B.; O’Connor, C. J. J . Org. Chem. 1968,
`33,572. (b) Bender, M. L.; Thomas, R. J. J. Am. Chem. SOC. 1961,83,4183.
`(15) (a) Shain, S. A,; Kirsch, J. F. J. Am. Chem. SOC. 1968, 90, 5848. (b)
`Lane, C. A,; Cheung, M. F.; Dorsey, G. F. Ibid. 1968, 90, 6492. (c) For a
`review of early oxygen isotopic exchange reactions of organic compounds, see:
`Samuel, D.; Silver, B. L. Adv. Phys. Org. Chem. 1965, 3, 123-186.
`
`Page 3 of 3