α-Bromodiazoacetamides – a new class of diazo compounds for catalyst-free, ambient temperature intramolecular C–H insertion reactions

In this work, we introduce a new class of halodiazocarbonyl compounds, α-halodiazoacetamides, which through a metal-free, ambient-temperature thermolysis perform intramolecular C–H insertions to produce α-halo-β-lactams. When carried out with α-bromodiazoacetamides bearing cyclic side chains, the thermolysis reaction affords bicyclic α-halo-β-lactams, in some cases in excellent yields, depending on the ring size and substitution pattern of the cyclic amide side chains.


General introduction
All commercially available reagents and solvents were used as received except in the case of N-bromosuccinimide (NBS), which was recrystallized from water (see details below). The CH 2 Cl 2 used in the acylation of the amines and in the thermolysis reactions was obtained from an MBraun MB SPS-800 solvent purification system and is dry, but more importantly free of alcoholic stabilizers. Alcohol-stabilized CH 2 Cl 2 does not completely inhibit the reactions in question, but traces of ethyl esters (O-acylation) and ethyl ethers (carbene O-H insertion product) have been been observed spectroscopically when running the reactions in alcoholstabilized CH 2 Cl 2 . The prepared/recrystallized N-bromoimides and diazoamides were stored in a freezer at −20°C to ensure their integrity and were left in ambient temperature for about 5 minutes before exposing them to air in order to minimize weighing errors caused by condensed water.
Except for the above stated routine, no special precautions were taken to exclude air or airborne moisture at any point during the syntheses described in this work, except in the acylation reactions where it was clearly detrimental. Inert atmosphere (Ar or N 2 ) is used only where stated specifically (acylations and catalyzed reactions). Stirring refers to magnetic stirring in all the procedures. Heating was performed with fluoropolymer-coated all-metal heating mantles.
Chromatography was performed on silica gel (Merck 60, 40-63 µm) and thin layer chromatography (TLC) on aluminium sheets coated with silica gel (Fluka 6607780, 20 mm, fluorescent coating for 254 nm) or neutral alumina (Macherey Nagel Alugram Alox N, 0.20 mm, unactivated, no fluorescent coating). The plates were visualized with KMnO 4 , K 2 CO 3 and NaOH in water for the α-bromo amides (alumina) and UV-light and/or a solution of p-anisaldehyde, concentrated H 2 SO 4 and glacial CH 3 CO 2 H in 96% EtOH for the α-diazoacetamides (silica gel), followed by heating with a heat gun for both. The β-lactams (alumina) were visualized with a solution of ninhydrin and pyridine in absolute EtOH followed by heating or with a solution containing AgNO 3 , H 2 O, 2-phenoxyethanol, acetone and H 2 O 2 (30 %, aq) followed by exposing the plate to UV-light for about 10 minutes (white spots on beige background).
Nuclear magnetic resonance (NMR) spectroscopy was performed on Bruker DXP300/DXP200 spectrometers operating at 300/200 MHz for 1 H and 75/50 MHz for 13 C. The chemical shifts are reported in parts per million (δ) relative to the residual solvent signal in general accordance with the literature 1 : δ-1 H/δ-13 C (Solvent); 7.26/77.00 (CDCl 3 ), 5.32/53.80 (CD 2 Cl 2 ), 2.50/39.52 (DMSO−d 6 ) and 1.94/118.26 (CD 3 CN). Chemical shifts are given in the section for physical data for each compound. Additionally, the spectra for the reported compounds are included in the appendix where their signals are tentatively assigned to numbered structures appearing on each spectrum, based on the information gained from the performed NMR experiments. The DEPT-135 spectra were referenced using the 13 C-NMR, as noted in the spectra. The NMR data were processed with ACD/Labs NMR Processor Academic Edition.
Mass spectrometry experiments were carried out with a Fisons VG ProSpec for electron impact ionisation (EI) and a Micromass Q-TOF 2 or a Bruker Apex FTMS (FT-ICR, 4.7 T) for electrospray ionisation (ESI). The data from the three instruments are marked (EI), (ESI, Q-TOF) and (ESI, FT-ICR), respectively. The solvent used and other essential details are noted in each case. NOTE: Diazo compounds are prone to dinitrogen extrusion, a process that can be effected by means of both heat and acidic media. While the obtention of molecular ions using electron impact ionisation (EI) was straightforward, like others, 2a we did not observe molecular ions using electrospray ionisation (ESI) with added HCOOH. The ESI-data for the diazoacetamides could indicate, however, that the observed radical cations are produced upon extrusion of dinitrogen and the inclusion of a molecule of solvent (tentatively assigned). One can speculate if this owes to an interaction of the carbene produced upon nitrogen extrusion with a molecule of solvent. 3 Such a reproducible m/z, not corresponding to the molecular ion, has also been noted by others in ESI experiments with EDA. 2a Interestingly, when running the ESI without added HCOOH, simple sodium adducts of the molecular ions was observed, except in the case of 3f.
Melting points were measured using a Stuart SMP10 melting point apparatus (Barloworld Scientific Ltd., UK) and are uncorrected.
Yield measurement using an internal standard: The reactions were left to expire overnight, before evaporating the solvent under reduced pressure at or below 23 • C. 0.5 molar equivalents of 2-naphthaldehyde was then added and the exact weight noted, before dissolving the entire crude material in CDCl 3 . The solution was transferred to an NMR-tube and a 1 H-NMR spectrum recorded. The integral of the aldehyde proton of the internal standard was calculated based on its molar fraction in relation to the starting material. Thus, the yields can be read out from the integral of the signal from the β--lactam α--proton in the NMR spectra in the appendix.

N-bromophthalimide
N-bromophthalimide was prepared from phthalimide by extending a procedure published by Souza et. al 4 for the preparation of N-chloro-and N-bromosaccharin, in which bromine is generated in situ by oxidation of KBr in water. The original procedure is modified by an additional 0.25 eq of Na 2 CO 3 .
The reaction proceeds noticeably slower than in the case of sodium saccharin (as in the original publication -preparation not shown), possibly due to the fairly low solubility of phthalimide even in basic water. Another consequence of this, was a prolonged and more visible release of bromine. As the consumption of the released bromine is slower than with saccharin, it was necessary to close the reaction vessel with a glass stopper to contain the evolution of bromine gas. The yield was 65 %. The obtained spectroscopic data were in accordance with the published data. 5 The measured melting point was lower, however, possibly owing to the fact that the recrystallization from toluene left a slightly amorphous crystal mass.

Experimental procedure
Phthalimide (7.36 g, 50.0 mmol, 1.0 eq), Na 2 CO 3 (3.98 g, 37.6 mmol, 0.75 eq) and KBr (5.95 g, 50.0 mmol, 1.0 eq) were added through a solid addition funnel to a 500 mL round bottom flask followed by H 2 O (Type II, 200 mL). The flask was cooled on an ice/water bath for 10 minutes, after which potassium peroxymonosulfate (as its triple salt 2 KHSO 5 · KHSO 4 · K 2 SO 4 , oxone®) (30.8 g, 50.1 mmol, 1.0 eq) suspended in H 2 O (Type II, 75 mL) was added in small aliquots, with vigorous stirring (700 rpm), over a period of 10 minutes. The addition of the oxidant resulted in vigorous release of bromine gas, and thus the flask was capped with a glass stopper between additions. The mixture quickly turned into an orange frothing suspension with visible bromine gas above it. The suspension was left to stir for 24 hours, at which point it had become a yellow solution with a white precipitate. Stirring was continued another 24 hours, until the solution above the precipitate was almost clear. The suspension was filtered on a Büchner-funnel and the filter cake sucked dry for a period of 30 minutes. The white precipitate was dissolved in boiling toluene (150 mL), hot filtered into a beaker and left to cool slowly to ambient temperature, covered with an aluminium foil. Upon reaching ambient temperature, precipitation had begun and the beaker was placed in a refrigerator at 4°C for a period of 17 hours, before filtering the precipitate on a Büchner-funnel. The filter cake was washed with n-pentane (30 mL) to give 6.52 g (57 %) of small white crystals. The mother liquor was concentrated, filtered and the filter cake washed with toluene (5 mL) and n-pentane (10 mL) to give 0.91 g (8 %). Both crops were of equal purity by 1 H-NMR. The total yield was 65%.

N-bromosuccinimide
N-bromosuccinimide was purchased from Sigma-Aldrich and recrystallized from boiling water (10 mL g −1 ) to give thin transparent plates in accordance with the literature (purification method "Aa").

O O
The reaction was performed once following the original procedure, affording a yield comparable to that reported (∼ 15 g, 85 %). The procedure was originally performed on a 50 mmol scale and employed a recrystallization from 400 mL of boiling MeOH as the final purification step. In the modified work-up procedure, the crude precipitate (see experimental details below) is simply slurried in methanol at reflux, lowering the expenditure of MeOH considerably.
The reaction could thus be performed at various scales with a low methanol expenditure: Scale (mmol) MeOH volume (mL) Yield (g) Yield (%)  100  400  31  92  192  500  61  93  300  800  95  93  350  900  109  92 The 1 H-NMR and 13 C-NMR spectra of the obtained products were identical to those of the product obtained using the original procedure, and are in accordance with the published spectroscopic data. 7 An X-ray structure can also be found in the literature. 8

Representative experimental procedure
Tosyl chloride (100.0 g, 524.5 mmol, 1.5 eq) and tosylhydrazide (65.12 g, 349.7 mmol, 1.0 eq) were added to a 1000 mL round-bottom flask through a solid addition funnel, and dissolved/suspended in anhydrous CH 2 Cl 2 (250 mL), giving a concentration of approximately 1 M based on the limiting reactant. An oval stir bar (3.5 × 1.5 cm) was added and the suspension stirred (700 rpm) while cooling on an ice/water bath, until an internal temperature of 3°C was reached.
Addition of a solution of pyridine (41.49 g, 524.5 mmol, 1.5 eq) in anhydrous CH 2 Cl 2 (80 mL) over a period of 10 minutes, in such a way that the internal temperature did not rise above 20°C. This gave a transparent yellow solution which soon deposited a white precipitate, upon which stirring was increased to 1000 rpm, and continued for a period of 3 hours. Et 2 O (200 mL) was added and the suspension solidified. The heavy precipitation was broken up with a spatula and the suspension was transferred to a 1500 mL beaker, where it was stirred with a triangular stir bar (5 × 1 × 1 cm). H 2 O (Type II, 200 mL) and Et 2 O (200 mL) was added in the given order. Stirring was continued for 10 minutes, before filtering the precipitate on a 13 cm Büchner-funnel, where it was left to air dry with suction for a period of 5 minutes. The solid was transferred directly to a 2000 mL round bottom flask and suspended in MeOH (900 mL). The flask was equipped with a reflux-condenser and an oval stir bar (5 × 2 cm) before heating the suspension to reflux (mantle temperature 67°C) and stirring for a period of 3 hours. The suspension, which at this point would separate to slightly yellow solution above a bright white precipitate, was allowed to cool before filtering on a 13 cm Büchner-funnel. The filter cake was washed with MeOH (300 mL) and Et 2 O (300 mL). The filtrate was then concentrated on a rotary evaporator (bath temperature ∼ 45°C) to a volume of approximately 150 mL, before cooling the flask under running tap water. The precipitate was filtered on a Büchnerfunnel, washed with MeOH (200 mL) and Et 2 O (200 mL) giving a second crop. The first crop was a voluminous white powder, while the second more crystalline crop consisted of short white needles. Both crops were dried in air in a fume hood over a period of 48 hours before determining the yield. Crop I: 97.60 g (81 %), crop II: 12.34 g (10 %). The total yield was 109.94 g (92 %). Both crops were of equal purity by NMR analysis and were subsequently mixed and stored in a screw-capped PTFE jar at ambient temperature.

Synthesis route and comments
The α-bromoacetamides were synthesized by the dropwise addition of the amine in CH 2 Cl 2 to bromoacetyl bromide and tribasic potassium phosphate in dry, alcohol-free CH 2 Cl 2 . The procedure was based on a recent publication where the authors investigated the effect of different bases on the formation of amides from amines and acid chlorides. 9 The obtained α-bromoacetamides were sufficiently pure after work-up to be used in the next step without further purification.

Representative experimental procedure
A 250 mL round bottom flask was charged, in the given order, with bromoacetyl bromide (30 mmol, 1.5 eq), anhydrous CH 2 Cl 2 (20 mL) and K 3 PO 4 (50 mmol, 2.5 eq). An oval stir bar (3.5 × 1.5 cm) was added and the flask was fitted with a 100 mL addition funnel, containing the amine (20 mmol, 1 eq) in CH 2 Cl 2 (30 mL). The flask was cooled on an ice/water bath, while purging the system with argon for a period of 5 minutes. Stirring (500 rpm) was commenced and the amine solution added dropwise over a period of 30 minutes, after which the addition funnel was washed with CH 2 Cl 2 (5 mL). Stirring was then continued for a period of 3 hours, before quenching the excess acid bromide with aqueous HCl (0.5 M, 30 mL). After stirring another 5 minutes, H 2 O (10 mL) and brine (10 mL) were added and the now transparent two phases transferred to a separatory funnel and the organic phase drained off. The aqueous phase was extracted with CH 2 Cl 2 (1 × 20 mL), before washing the joint organic phases with a mixture of aqueous KHCO 3 (10 %wt, 20 mL) and brine (10 mL), and then with brine (20 mL). Drying over MgSO 4 and evaporating under reduced pressure at 30°C bath temperature, afforded the α-bromoacetamides. The crude material was stored at −20°C before submission to the next step.

2-bromo-1-(1,1-dioxidothiomorpholino)ethanone (2f)
The solubilities of both the amine thiomorpholine-1,1-dioxide and the obtained amide are relatively low in CH 2 Cl 2 . As such, the volume of CH 2 Cl 2 in which the amine was initially dissolved was increased to 7.40 mmol/30 mL. Additionally, on the same scale, the reaction mixture was diluted with CH 2 Cl 2 (100 mL) after quenching with HCl (aq), before separating the phases. A longer reaction time of 24 hours was also needed to achieve full conversion of the amine.
Yield: 1.733 g (91 %) from 7.40 mmol of thiomorpholine-1,1-dioxide. NMR spectra can be found on page 39.  7 The lower cost and similar base strength 10 of 1,1,3,3-Tetramethylguanidine (TMG), as well as its properties as a nucleophilic catalyst, 11 made it an interesting alternative to DBU in this transformation. Furthermore, in the original procedure, the crude reaction mixture was quenched with NaHCO 3 (aq) and then extracted with Et 2 O. In our experience it was problematic to perform a clean aqueous extraction in the presence of the at least 2 molar equivalents of DBU−HO 2 S-C 6 H 4 CH 3 produced during the reaction. This situation manifested itself more gravely on a > 5 mmol scale. The problem could lie in the intermediate solubility of the DBU−HO 2 S-C 6 H 4 CH 3 complex between the aqueous phase and the Et 2 O phase. What further complicated the matter of an aqueous extraction procedure, was the aqueous solubility of some of the obtained α-diazoacetamides. The use of TMG instead of DBU resulted in the partial precipitation of the TMG · HO 2 S-C 6 H 4 CH 3 salt from THF during the reaction. Thus, in the cases where the diazoamide was soluble in Et 2 O (products 3a -e), the reaction solvent (THF) could be destilled off under reduced pressure and the diazoamide extracted from the resulting solid with Et 2 O (the TMG · HO 2 S-C 6 H 4 CH 3 salt is poorly soluble in Et 2 O). A final clearing of the Et 2 O-phase by slurrying with silica gel and filtering, yielded the diazoamides with decent purities, without the need for an aqueous work-up. As noted by Toma et al., we also observed the formation of the corresponding α-tosylacetamides, resulting from the nucleophilic attack of the produced p-toluenesulfinic acid on the substrate α-bromoacetamides. This side-reaction can however be minimized by maintaining the reaction temperature below 0 • C (see procedure below).
In the case where the diazoamide was not soluble in Et 2 O (product 3f), a THF/Et 2 O (2:1) mixture could be used to extract the diazoamide with adequate results (see the experimental procedure for product 3f below).
All the obtained diazoamides were subsequently chromatographed on silica gel to ensure that they were free of contaminations (e.g. their corresponding α-tosylacetamides) that could interfere in the investigation of their halogenation and subsequent reactions. The details pertaining to the chromatography for each compound are given below.

Representative experimental procedure (products 3a -e)
To a 250 mL round bottom flask containing the α-bromoacetamide (20 mmol, 1 eq), were added through a solid addition funnel; N,N'-ditosylhydrazine (40 mmol, 2 eq), THF (35 mL) and an oval stir bar (3.5 × 1.5 cm). The suspension was cooled on an acetone/CO 2 (s) bath with stirring (400 rpm) until reaching an internal temperature of −10 • C, upon which a solution of 1,1,3,3tetramethylguanidine (100 mmol, 5 eq) in THF (10 mL) was added in aliquots using a pasteur pipette, while maintaining the internal temperature below 0 • C by the addition of CO 2 (s). The resulting yellow solution was removed from the cooling bath and allowed to warm to ambient temperature over a period of 30 minutes, during which time a white precipitate is deposited. This suspension was stirred for a period of 90 minutes, before removing the THF under reduced pressure at or below 23 • C. The resulting solid was treated with anhydrous Et 2 O (150 mL) and vigorously stirred for 10 minutes, before being filtered on a Büchner-funnel. The filter cake was washed with anhydrous Et 2 O (2 -5 × 50 mL), until leaving an almost white filter cake. Approximately 25 g (1 large tablespoon) of silica gel was added to the filtrate and the suspension stirred for 5 minutes, followed by vacuum filtration through a pad of silica gel (∼ 25 g, column 3.5 cm). The column was washed with Et 2 O (50 mL) and the Et 2 O evaporated under reduced pressure at or below 23 • C. The thus obtained crude diazoacetamides were then subjected to flash chromatography on silica gel eluting increasing amounts of Et 2 O in Pet.Ether. The products were stored at −20 • C before submission to the next step.
The compound is a solid at −20 • C (picture), and is previously reported in the literature. 12
The compound is previously reported as a reagent, 21

tert-butyl 4-(2-diazoacetyl)piperazine-1-carboxylate (3d)
Yield: 2.938 g (60 %) as a bright yellow solid starting from 19.16 mmol 2d. The compound is to the best of our knowledge previously unreported as a reagent. Our group recently reported the x-ray structure of 3e. 22 NMR spectra can be found on page 51.
The compound is previously reported in several publications. 17,20,23 NMR spectra can be found on page 53.
The compound is generally poorly soluble in common solvents, and is to the best of our knowledge previously unreported as a reagent. Our group recently published the x-ray structure of 3f. 24 NMR spectra can be found on page 56.

Experimental procedure
To a 250 mL round bottom flask containing 2-bromo-1-(1,1-dioxidothiomorpholino)ethanone (1.928 g, 7.53 mmol, 1 eq), were added through a solid addition funnel N,N'-ditosylhydrazine (5.129 g, 15.07 mmol, 2 eq), THF (50 mL) and an oval stir bar (3.5 × 1.5 cm). The suspension was cooled on an ice/water bath with stirring (400 rpm) for a period of 15 minutes, upon which a solution of 1,1,3,3-tetramethylguanidine (4.337 g, 37.65 mmol, 5 eq) in THF (10 mL) was added in bulk. The white suspension quickly became yellow, and was then stirred for a period of 70 minutes, before adding Et 2 O (100 mL) causing the disappearance of the yellow colour from the THF/Et 2 O phase, as the product was no longer soluble. Removal of some of the Et 2 O under reduced pressure, brought the yellow colour back to the liquid phase in the suspension. The mixture was filtered on a Büchner-funnel and the filter cake washed with THF/Et 2 O (2:1, 75 mL) and THF (30 mL), leaving an almost white filter cake.
A preliminary purification was performed by adding Celite (∼ 60 g, 3 tablespoons) to the filtrate, and evaporating the solvents under reduced pressure at or below 23 • C. The resulting slightly sticky powder, was suspended in CH 2 Cl 2 (15 mL) and loaded on top of 1 cm of sea-sand covering a column of silica gel (8.0 × 3.5 cm) packed in CH 2 Cl 2 . A yellow band was eluted with MeOH/CH 2 Cl 2 (5:95). To this fraction was added silica gel (∼ 50 g, 2 large tablespoons) and the solvents evaporated, leaving a pale yellow powder. The powder was loaded on top of a column (5.5 × 3.5 cm) of silica gel packed in CH 2 Cl 2 , and the product slowly eluted with CH 2 Cl 2 and from the point of elution CHCl 3 . The solvents were evaporated under reduced pressure at or below 23 • C affording 0.987 g (64 %) of a pale yellow powder which was stored at −20°C.

β-lactams
The halogenated α-bromodiazoacetic ethyl esters were reported to be stable for days at 0°C. 25 The α-halodiazoamides, however, appear less stable. In our experience the 2 -5°C temperatures, obtainable with a common ice/water bath, were not sufficiently low to work with the α-halodiazoacetamides with adequate reproducibility. Therefore, all procedures were carried out at −5°C using an acetone bath in which the temperature was controlled by a cryostat. In the summaries from the mass spectrometry experiments, the R-group in "α, α ′ -Br 2 −NR 2 " refers to the cyclic amide side chain of the respective compound.

Representative experimental procedure
The diazoacetamide (0.5 mmol, 1.0 eq) was weighed out in a 25 mL pear shaped flask, DBU (0.7 mmol, 1.4 eq) in a 10 mL beaker and the N-halo compound (0.65 mmol, 1.3 eq) in a weighing funnel. The flask was charged with a small magnetic stir bar (10 × 2 mm) and CH 2 Cl 2 (2 mL, ambient temperature), and was subsequently mounted alongside a 250 mL round bottom flask containing CH 2 Cl 2 (250 mL) in an acetone bath cooled by a cryostat to −5°C (measured with an alcohol thermometer at the surface). Stirring (300 RPM) was commenced and after 2 minutes the base in CH 2 Cl 2 (2 mL, ambient temperature) was added. After 2 minutes the N-halo compound was added in bulk through the funnel and the funnel rinsed with CH 2 Cl 2 (1 mL, ambient temperature). The addition effected a colour change from bright yellow to bright red. While the reaction was stirring, a short plug of silica gel (2 × 2 cm) was packed in CH 2 Cl 2 (ambient temperature) in a fritted flash chromatography column (15 × 2 cm) and mounted inside a holder which allows for the filling of dry ice around the column (covering ∼ 300°, leaving an observation window, see figure 5.1 on the facing page). After stirring for 5 minutes, the thermometer was placed in the 250 mL flask and CO 2 (s) was added to the bath until the internal temperature of the CH 2 Cl 2 had reached −15°C. The space between the jacket and the chromatography column was filled with dry ice, and the column equilibrated with CH 2 Cl 2 (−15°C, 20 mL). The reaction mixture was then poured directly onto the column, and the flask rinsed with two pipette volumes of CH 2 Cl 2 (−15°C). The reaction mixture was run into the column using pressurized air (ambient temperature), and the column filled up with CH 2 Cl 2 (−15°C). The column was then run with CH 2 Cl 2 (−15°C, ∼ 200 mL) under air pressure to completely elute a bright red band into a new 250 mL flask in the cooling bath. The eluted solution is bright red/pink in colour with a temperature of approximately −30°C.

Thermolysis:
The flask was placed on the bench without stirring while the temperature rose from −30°C towards ambient temperature. The solution became clear in approximately 60 minutes. In the chromatographic step, for the more polar analogues, it was necessary to add Et 2 O (5 mL) to the last 50 mL of the eluent to completely collect the red band. It should be noted that Et 2 O has a much higher elution strength than CH 2 Cl 2 under these conditions and fortunately does not seem to interfere with the C-H insertion reaction at this concentration. We have however observed that higher concentrations of Et 2 O will make intermolecular reactions (ylide formation, C-H insertion) sufficiently favoured so as to compete with the intramolecular C-H insertions.

Catalyzed reactions:
The flask was charged with a stir bar, capped with a septum and maintained in the cooling bath at −15°C (cryostat) while degassing with argon (5 minutes). The flask was then taken out of the cooling bath and a degassed solution of the catalyst in CH 2 Cl 2 (2 mL) was added via syringe under vigorous stirring. The solution was then left to heat towards room temperature.
For an illustration of the CO 2 (s)-cooled column used in this procedure, see figure 5.1 below.

6-bromo-1-azabicyclo[3.2.0]heptan-7-one (5a)
The compound was prepared following the general procedure for thermolysis given above. In CDCl 3 , the compound decomposes over time as indicated by a 1 H-NMR spectrum recorded 48 hours after the first (data not shown). This could owe to the oxidative accumulation of HCl in the CDCl 3 resulting in β--lactam hydrolysis, and would be consistent with the reported rapid hydrolysis of β-lactams fused with 5-membered rings. 14 An NMR spectrum of the crude reaction mixture, added internal standard, can be found on page 58.

7-bromo-1-azabicyclo[4.2.0]octan-8-one (5b)
The compound was prepared both following the general procedure for thermolysis and that for catalysis, given above. NMR spectra of the crude reaction mixtures and crude reactions mixture added internal standard, can be found on page 59.

Physical data
Crude reaction mixture from thermolysis, mixture of endo/exo-isomers:

8-bromo-1-azabicyclo[5.2.0]nonan-9-one (5c)
The compound was prepared following the general procedure for thermolysis given above. NMR spectra of the crude reaction mixture added internal standard, as well as the exo/endo isomers isolated by chromatography, can be found on page 61.

Chromatographic procedure
The diastereomers exo-3e and endo-3e were separated on a column of neutral alumina (10 × 2.5 cm ), using Et 2 O as the eluent.

Physical data
Isolated by chromatography:

7-bromo-4-oxa-1-azabicyclo[4.2.0]octan-8-one (5e)
The compound was prepared following the general procedure for thermolysis given above. An NMR spectrum of the crude reaction mixture, added internal standard, can be found on page 67.

Physical data
Crude reaction mixture from thermolysis:

Synthesis and thermolysis of α-chloro-and α-iodo-4d
It was desirable to investigate whether the reaction of the α-bromodiazoacetamides could be extended to its chlorine and iodine analogues. The piperazine-derivative 3d was chosen as the model compound, since the C-H insertion reaction of Br-4d gave a moderate to low yield, which provided an opportunity for its chlorine and iodine analogues to expose their differential reactivity. The halogenation reactions were carried out using DBU/NXS (X = Cl, Br, I), due to the availability of the halosuccinimides.

Dirhodium(II) catalysis results
The results were obtained following the representative experimental procedure for catalyzed reactions given above. NBS was used instead of NBP, due to the higher retention of succinimide vs phthalimide on the short column used to purify the β-diazoacetamide.  1 1,1,3

,3-tetramethylguanidine p-toluenesulfinate
The precipitation of this salt occurred during the preparations of the α-diazoacetamides as described in their experimental procedure. In one instance, the precipitate was weighed dry and it accounted for 98 %wt of the theoretical amount.