Improvement Upon a Largely Forgotten Method for the Synthesis of N-Alkyl Urazoles

Abstract

N-Alkyl urazoles are important heterocyclic compounds that serve as important precursors to potent N-alkyl 1,2,4-triazoline-3,5-dione electrophiles. Traditional methods for urazole synthesis rely upon the use of toxic isocyanates. We have modified and optimized an overlooked and poorly described method from the literature for the synthesis of urazoles that now avoids the use of isocyanates, limits the use of solvents, and provides urazoles without the need for purification steps. A variety of urazoles are afforded in good to high yields.

1. Introduction

N-Alkyl urazoles (1, Scheme 1) are direct precursors, via oxidation, to the corresponding azo compounds N-alkyl 1,2,4-triazoline-3,5-diones (2, RTADs) [1]. RTADS are potent electrophilic reagents that readily engage in a variety of useful reactions, including Diels–Alder cycloadditions with dienes, [2 + 2] cycloadditions with suitably substituted alkenes, and ene reactions [1]. Such general organic reactivity has recently been directed towards practical applications such as “click”-type reactivity, polymer coupling, and surface modification [1,2].

The traditional method for the synthesis of N-alkyl urazoles was devised by Zinner in 1961 [3] and then refined and developed into an established organic synthesis procedure by Cookson in 1971 (Scheme 2) [4]. Unfortunately, however, this synthetic method relies upon access to appropriate alkyl isocyanates (3) as starting materials [5]. Due to the known dangers of working with isocyanates, the commercial availability of isocyanates has severely diminished, thereby hindering direct access to urazole compounds via the Cookson route. This has led to the development of alternative routes for the synthesis of urazoles that have been recently summarized by Du Prez [1].

Triazolinedione 2a is one of the most commonly used triazolinediones reported in the literature [6]. The lack of commercial availability of methylisocyanate, the Cookson precursor to N-methyl urazole (1a), is especially problematic because compound 2a is a convenient TAD derivative with which to work. Its particular usefulness derives from the unique chemical shift of the N-methyl signal in the ¹H NMR spectrum (a singlet at ~3 ppm) that makes the following reactions, and analysis of crude reaction mixtures, relatively straightforward [6]. Therefore, several years ago, our group developed a method for the synthesis of urazole 1a via the in situ generation of methyl isocyanate from N-nitroso dimethylurea 5 (see Scheme 3) as well as via the reaction of methylamine with ethyl phenyl hydrazine-1,2-dicarboxylate 6 [6]. These reactions afforded semicarbazide 4a, which could then be cyclized to the urazole 1a under standard Cookson conditions (see Scheme 2). Gratifyingly, we also found that derivatives of 4 other than just the N-methyl were also accessible from compound 6 (i.e., 1d–g) [6].

Recently, Sarlah reported a “scalable” synthesis of N-methyl urazole 1a that accessed semicarbazide intermediate 4 (R = Me) using carbonyl diimidazole (CDI) as the reactive carbonyl component (Scheme 4) [7]. While this method permits a multigram synthesis of N-methyl urazole, the process requires concentration of a reaction mixture under high vacuum followed by several recrystallization steps to afford the desired 4a. While still an excellent method for the large-scale synthesis of 1a, it is not particularly convenient for ordinary lab use, nor has it been expanded to urazole substituents other than that of N-methyl. Finally, in related work, Du Prez reported what was described as a “sustainable” synthetic route for urazoles [8]. This synthetic route eliminated the use of reaction solvents by substituting bulk heating of intermediate compounds (Scheme 5). The major limitation of this method was that the amines needed to have a relatively high boiling point (with the smallest being that of butylamine) for the method to be viable.

We were recently fortunate enough to come across a previously described method for the synthesis of urazoles that seems to have been overlooked in the literature (or at least underappreciated). In a 1967 paper, Priehradny reported that heating neat 1,6-dialkylbiurea 7 to relatively high temperatures (~250 °C) resulted in spontaneous cyclization, forming the corresponding N-alkyl urazoles (Scheme 6) [9]. Therefore, this synthesis can be considered to be a “sustainable” method akin to that described earlier by Du Prez (Scheme 5) in that solvent use is avoided in the cyclization step. The major downside to this method is that the synthesis of the required starting 1,6-dialkylbiurea 7 relied upon either the same toxic isocyanates as used in the Cookson method or the use of similarly toxic dialkyl azodicarboxylates via a two-step method (see Scheme 6). Additionally, there have also been some reports that the Priedhradny urazole synthesis method has been hard to reproduce [1,4,10]. The lack of reproducibility could be due to the lack of experimental detail provided in the original paper [9]. Therefore, for these reasons, the use of Priehradny’s synthetic method for urazoles has been largely ignored.

In this paper, we revitalize Priedhradny’s synthetic method by finding a general and highly convenient (i.e., free of purification steps) route for the synthesis of the 1,6-dialkylbiureas precursors (7) that eliminates the need for isocyanates and azodicarboxylates. We also refine the solvent-free cyclization of compound 7 to the corresponding urazoles and provide important reaction details missing from the Priehradny paper that now allow for the acquisition of practical and reproducible results. We hope that by avoiding toxic reagents, optimizing reaction steps, and providing key details on the cyclization step that ensure reproducibility, this very convenient, but previously overlooked, method might see greater use.

2. Materials and Methods

2.1. General Methods

All solvents and reagents were obtained commercially and used as received. Heating of 1,6-dialkylbisureas (7) was conducted in 5 mL conical vials left open to the atmosphere (in a suitable hood) using a metal hot plate whose temperature was closely controlled by a digitally controlled unit. Chemical shifts (¹H and ¹³C) are reported in units of parts per million downfield from tetramethylsilane (TMS). Infrared (IR) spectra were collected as solids pressed against a ZnSe attenuated total reflection (ATR) crystal. High-resolution mass spectra (HRMS) were acquired via electron spray ionization using a linear trap quadrupole Fourier transform hybrid mass spectrometer (LTQ-FTMS).

2.2. Experimental Procedures

2.2.1. N,N-Diphenyl-1,2-hydrazinedicarboxamide (8)

To a stirred mixture of 1.05 g (12.6 mmol) of hydrazine hydrate (60% by weight) and 2.76 g (2 equiv) of sodium carbonate in 125 mL of tetrahydrofuran (THF) at 0 °C, 4 g (25.5 mmol) of phenyl chloroformate was added dropwise via Pasteur pipette. The mixture was allowed to warm to room temperature and stirred overnight, after which a white precipitate appeared. A total of 50 mL of 0.5 M aq. HCl was added to the mixture, and the THF was removed via rotary evaporation to leave the white solid suspended in the aqueous layer. The mixture was filtered, and the separated solid was rinsed well with 100 mL of water. After air drying, the product was dried in the oven at 90 °C for at least two hours, after which 84.7 g (96%) of 8 was isolated as a fluffy white solid. m.p. 153–154 °C: ¹H NMR (DMSO-d₆) δ 9.44–10.10 (multiple s, 2H, NH), 7.42 (t, J = 7.8 Hz, 4H), 7.25 (t, J = 7.8 Hz, 2H), 7.13 (d, J = 7.8 Hz, 4H); ¹³C NMR (DMSO-d₆) δ 154.8, 150.5, 129.6, 125.6, 121.6. The spectra are consistent with those reported in the literature [11].

2.2.2. 1,6-Dimethylbisurea (7a)

To a stirring solution of 0.5 g (1.83 mmol) of compound 8 in 15 mL of CH₃CN, 5.50 mL (6 eq) of a 2 M solution of methyl amine in THF was added via syringe. A precipitate began to form some minutes later, and stirring was continued overnight. The resulting white precipitate was isolated via vacuum filtration and washed with 10 mL of water. Air drying afforded 0.25 g (94% yield) of 7a as a white powder. m.p. 243–244 °C: IR (ATR) cm⁻¹ 3311.6, 1662.0, 1561.1, 1413.9, 1325.6; ¹H NMR (CF₃CO₂D) δ 11.82 (br s, 4H, NH), 2.99 (s, 6H); ¹³C NMR (CF₃CO₂D) δ 164.2, 28.1; HRMS (ESI) m/z [M + H] + Calcd for C₄H₁₁N₄O₂ 147.08765; Found: 147.08695.

2.2.3. 1,6-Diethylbisurea (7b)

To a stirring solution of 0.5 g (1.83 mmol) of compound 8 in 15 mL of CH₃CN, 5.50 mL (6 eq) of a 2 M solution of ethyl amine in THF was added via syringe. A precipitate began to form some minutes later, and stirring was continued overnight. The resulting white precipitate was isolated via vacuum filtration and washed with 10 mL of water. Air drying afforded 0.28 g (88% yield) of 7b as a white powder. m.p. 238–239 °C: IR (ATR) cm⁻¹ 3297.7, 1665.0, 1554.2, 1328.6; ¹H NMR (CF₃CO₂D) δ 11.73 (br s, 4H, NH), 3.47 (q, J = 7.7 Hz, 4H), 1.28 (t, J = 7.7 Hz, 6H); ¹³C NMR (CF₃CO₂D) δ 160.3, 34.9, 11.7; HRMS (ESI) m/z [M + H] + Calcd for C₆H₁₅N₄O₂ 175.11895; Found 175.11874.

2.2.4. 1,6-Di-n-propylbisurea (7c)

To a stirring solution of 0.5 g (1.83 mmol) of compound 8 in 20 mL of CH₃CN, 0.64 g (6 eq) of n-propyl amine was added dropwise via Pasteur pipette. The solution became cloudy after approximately 10 min, and stirring was continued overnight. The resulting white precipitate was isolated via vacuum filtration and washed with 10 mL of water. Air drying afforded 0.32 g (87% yield) of 7c as a white powder. m.p. 243–244 °C: IR (ATR) cm⁻¹ 3290.0, 2961.8, 1658.5, 1560.2, 1335.3; ¹H NMR (CF₃CO₂D) δ 11.72 (br s, 4H, NH), 3.39 (t, J = 7.3 Hz, 4H), 1.68 (h, J = 7.3 Hz, 4H), 0.99 (t, J = 7.3 Hz, 6H); ¹³C NMR (CF₃CO₂D) δ 164.1, 44.8, 24.1, 11.5; HRMS (ESI) m/z [M + H] + Calcd for C₈H₁₉N₄O₂ 203.15025; Found 203.15006.

2.2.5. 1,6-Di-n-butylbisurea (7d)

To a stirring solution of 0.5 g (1.83 mmol) of compound 8 in 20 mL of CH₃CN, 0.80 g (6 eq) of n-butyl amine was added dropwise via Pasteur pipette. The solution became cloudy after approximately 10 min, and stirring was continued overnight. The resulting white precipitate was isolated via vacuum filtration and washed with 10 mL of water. Air drying afforded 0.38 g (90% yield) of 7d as a white powder. m.p. 246–247 °C: IR (ATR) cm⁻¹ 3295.0, 2956.6, 1659.6, 1555.3, 1377.0; ¹H NMR (CF₃CO₂D) δ 11.82 (br s, 4H, NH), 3.43 (t, J = 7.3 Hz, 4H), 1.63 (p, J = 7.3 Hz, 4H), 1.42 (h, J = 7.3 Hz, 4H), 0.98 (t, J = 7.3 Hz, 6H); ¹³C NMR (CF₃CO₂D) δ 163.5, 42.8, 32.9, 21.3, 14.0; HRMS (ESI) m/z [M + H] + Calcd for C₁₀H₂₃N₄O₂ 231.18155; Found 231.18157.

2.2.6. 1,6-Di-tert-butylbisurea (7e)

To a stirring solution of 0.5 g (1.83 mmol) of compound 8 in 20 mL of CH₃CN, 0.80 g (6 eq) of tert-butyl amine was added dropwise via Pasteur pipette. After being stirred overnight, the solution was concentrated via rotary evaporation to afford a pale-orange solid. The solid was taken up in 20 mL of CH₂Cl₂ and washed 2 × 20 mL with 0.5 M aq. NaOH. The organic layer was then dried over Na₂SO₄, filtered, and concentrated to afford 0.24 g (57% yield) of 7e as a white solid. m.p. 189–190 °C: IR (ATR) cm⁻¹ 3314.7, 2968.1, 1654.9, 1558.4, 1363.3, 1217.6; ¹H NMR (CDCl₃) δ 6.88 (br s, 2H, NH), 5.62 (br s, 2H, NH), 1.34 (s, 9H); ¹³C NMR (CDCl₃) δ 158.1, 50.6, 29.1; HRMS (ESI) m/z [M + H] + Calcd for C₁₀H₂₃N₄O₂ 231.18155; Found 231.18111.

2.2.7. 1,6-Dicyclohexylbisurea (7g)

To a stirring solution of 0.5 g (1.83 mmol) of compound 8 in 20 mL of CH₃CN, 1.09 g (6 eq) of cyclohexyl amine was added dropwise via Pasteur pipette. After being stirred overnight, the resulting white precipitate was isolated via vacuum filtration and washed with 10 mL of water. Air drying afforded 0.45 g (77% yield) of 7g as a white powder. m.p. 203–204 °C: IR (ATR) cm⁻¹ 3675.5, 2972.0, 1646.4, 1531.5, 1393.9, 1066.0; ¹H NMR (CF₃CO₂D) δ 11.69 (br s, 4H, NH), 3.77 (m, 2H), 1.97 (br s, 4H), 1.87 (br s, 4H), 1.72 (br d, J = 12.4 Hz, 2H), 1.35–1.50 (m, 8H), 1.26 (m, 2H); ¹³C NMR (CF₃CO₂D) δ 162.6, 53.6, 34.6, 26.6, 26.5; HRMS (ESI) m/z [M + H] + Calcd for C₁₄H₂₇N₄O₂ 283.21285; Found 283.21103.

2.2.8. 1,6-Dibenzylbisurea (7g)

To a stirring solution of 0.5 g (1.83 mmol) of compound 8 in 20 mL of CH₃CN, 1.17 g (6 eq) of benzyl amine was added dropwise via Pasteur pipette. After being stirred overnight, the resulting white precipitate was isolated via vacuum filtration and washed with 10 mL of water. Air drying afforded 0.42 g (76% yield) of 7g as a white powder. m.p. 245–246 °C: IR (ATR) cm⁻¹ 3292.2, 1660.3, 1552.1, 1298.6, 1219.0; ¹H NMR (CF₃CO₂D) δ 11.68 (br s, 4H, NH), 7.26–7.34 (m, 10H), 4.53 (s, 4H); ¹³C NMR (CF₃CO₂D) δ 163.6, 138.1, 130.89, 130.3, 129.5, 46.6; HRMS (ESI) m/z [M + H] + Calcd for C₁₆H₁₉N₄O₂ 299.15025; Found 299.15007.

2.2.9. N-Methylurazole (1a)

A total of 100 mg (0.68 mmol) of finely powdered 7a was heated in a vial to 260 °C over a period of approximately a 10 min. Visible fuming began to appear at a temperature of ~240 °C. As the sample was held at 260 °C, it melted, yielding a colorless oil. The melt was held at this temperature with occasional swirling to allow liberated methyl amine to escape. After 10 min, fuming ceased, and the vial was removed from the heat with swirling to aid in crystallization of the urazole product with cooling. The product was scraped from the vial to afford 62 mg of 1a as a white crystalline product (80% yield). m.p. 237–238 °C (lit. 232–233 °C [12]): ¹H NMR (DMSO-d₆) δ 9.59 (br s, 2H, NH), 2.84 (s, 3H); ¹³C NMR (DMSO-d₆) δ 155.3, 24.3. The spectra are consistent with those reported in the literature [13].

Repeating this process on a larger scale starting with 0.50 g of 7a afforded 0.39 g (85% yield) of 1a. Similarly, a 1.0 g sample of 7a in a 20 mL vial required heating at 260 °C for a total of 20 min to complete the melting process, and this temperature was then further held for 5 min, followed by cooling, to afford 0.63 g (80% yield) of 1a.

2.2.10. N-Ethylurazole (1b)

A total of 100 mg (0.68 mmol) of finely powdered 7b was heated in a vial to 260 °C over a period of approximately 10 min. Visible fuming began to appear at a temperature of ~240 °C. As the sample was held at 260 °C, it melted, yielding a colorless oil. The melt was held at this temperature with occasional swirling to allow liberated ethyl amine to escape. After 10 min, fuming ceased, and the vial was removed from the heat with swirling to aid in crystallization of the urazole product with cooling. The product was scraped from the vial to afford 68 mg of 1b as a white crystalline product (80% yield). m.p. 193–194 °C (lit. 195–196 °C [12]): ¹H NMR (DMSO-d₆) δ 9.90 (br s, 2H, NH), 3.37 (q, J = 7.3 ppm, 2H), 1.09 (t, J = 7.3 ppm, 3H); ¹³C NMR (DMSO-d₆) δ 154.9, 32.9, 13.3. The spectra are consistent with those reported in the literature [14].

2.2.11. N-Propylurazole (1c)

A total of 100 mg (0.68 mmol) of finely powdered 7c was heated in a vial to 260 °C over a period of approximately 10 min. Visible fuming began to appear at a temperature of ~240 °C. As the sample was held at 260 °C, it melted, yielding a colorless oil. The melt was held at this temperature with occasional swirling to allow liberated propyl amine to escape. After 10 min, fuming ceased, and the temperature was reduced to 190 °C; this temperature was held for 25 min, again with occasional swirling. After the heating period, the vial was removed from the heat with swirling to aid in crystallization of the urazole product upon cooling. The product was scraped from the vial to afford 62 mg of 1c as a white crystalline product (89% yield). m.p. 166–167 °C (lit. 168–169 °C [12]): ¹H NMR (DMSO-d₆) δ 9.90 (br s, 2H, NH), 3.30 (t, J = 7.4 ppm, 2H), 1.52 (h, J = 7.4 ppm, 2H), 0.81 (t, J = 7.5 ppm, 3H); ¹³C NMR (DMSO-d₆) δ 155.2, 39.51, 21.0, 11.0. The spectra are consistent with those reported in the literature [15].

2.2.12. N-Butylurazole (1d)

A total of 100 mg (0.68 mmol) of finely powdered 7d was heated in a vial to 260 °C over a period of approximately 10 min. Visible fuming began to appear at a temperature of ~240 °C. As the sample was held at 260 °C, it melted, yielding a colorless oil. The melt was held at this temperature with occasional swirling to allow liberated butyl amine to escape. After 15 min, fuming ceased, and the temperature was reduced to 190 °C; this temperature was held for 25 min, again with occasional swirling. After the heating period, the vial was removed from the heat with swirling to aid in crystallization of the urazole product with cooling. The product was scraped from the vial to afford 53 mg of 1d as a white crystalline product (78% yield). m.p. 170–171 °C (lit. 167–168 °C [12]): ¹H NMR (DMSO-d₆) δ 10.00 (br s, 2H, NH), 3.34 (t, J = 7.4 ppm, 2H), 1.50 (p, J = 7.4 ppm, 2H), 1.04 (h, J = 7.4 ppm, 2H), 0.87 (t, J = 7.5 ppm, 3H); ¹³C NMR (DMSO-d₆) δ 155.2, 37.6, 29.6, 19.4, 13.5. The spectra are consistent with those reported in the literature [13].

Repeating this process on a larger scale starting with 0.95 g of 7d afforded 0.49 g (75% yield) of 1d.

2.2.13. N-Cyclohexylurazole (1f)

A total of 100 mg (0.68 mmol) of finely powdered 7f was heated in a vial to 260 °C over a period of approximately 10 min. Visible fuming began to appear at a temperature of ~250 °C. As the sample was held at 260 °C, it melted, yielding a colorless oil. The melt was held at this temperature with occasional swirling to allow liberated butyl amine to escape. After 30 min, fuming ceased, and the temperature was reduced to 190 °C; this temperature was held for 25 min, allowing the product to solidify. After the heating period, the vial was removed from the heat with swirling to aid in crystallization of the urazole product with cooling. The product was scraped from the vial to afford 55 mg of 1f as a white crystalline product (85% yield). m.p. 242–243 °C (lit. 239–241 °C [16]): ¹H NMR (DMSO-d₆) δ 9.95 (br s, 2H, NH), 3.68 (dt, J = 12.4, 3.9 Hz, 1H), 2.02 (dg, J = 12.4, 3.9 Hz, 1H), 1.76 (br d, 2H), 1.55–1.65 (m, 3H), 1.25 (qt, J = 12.4, 3.3 Hz, 1H), 1.12 (dt, J = 12.4, 3.0 Hz, 1H); ¹³C NMR (DMSO-d₆) δ 155.3, 50.6, 29.5, 25.8, 25.3. The spectra were consistent with those reported in the literature [17].

2.2.14. N-Benzylurazole (1g)

100 mg (0.68 mmol) of finely powdered 7g was heated in a vial to 260 °C over a period of approximately 10 min. Visible fuming began to appear at a temperature of ~250 °C. As the sample was held at 260 °C, it melted, yielding a colorless oil. The melt was held at this temperature with occasional swirling to allow liberated butyl amine to escape. After 1 h, fuming ceased, and the temperature dropped to 190 °C; this temperature was held for 25 min, again with occasional swirling. After the heating period, the vial was removed from the heat with swirling to aid in crystallization of the urazole product with cooling. The product was scraped from the vial to afford 60 mg of 1g as a white crystalline product (94% yield). m.p. 181–182 °C (lit. 182–183 °C [12]): ¹H NMR (DMSO-d₆) δ 10.21 (br s, 2H, NH), 7.26–7.35 (m, 5H), 4.54 (s, 2H); ¹³C NMR (DMSO-d₆) δ 154.8, 136.8, 128.6, 127.6, 127.5, 41.3. The spectra were consistent with those reported in the literature [13].

3. Results and Discussion

3.1. Improved Synthesis of 1,6-Dialkylbisureas

Based upon our earlier described success in substituting phenol with amines from semicarbazide 6 (see Scheme 4) [6], we considered the possibility of forming the required 1,6-dialkylbisureas (7) via the reaction of diphenyl hydrazine-1,2-dicarboxylate 8 (Scheme 7). Compound 8 itself can easily be synthesized in high yield and purity from the reaction of hydrazine hydrate with two equivalents of phenyl chloroformate in THF [see Section 2] and isolated via vacuum filtration of the crude reaction mixture without the need for any further purification.

Addition of amines to 8 successfully and cleanly afforded the desired bisureas (7). While 2 equivalents of amine at first glance seems adequate for the reaction, we soon realized that the progress of the reaction was quickly stifled by the release of the acidic phenol, which ties up unreacted amine. Therefore, at least 4 equivalents of amine are necessary to complete the reaction, but we observed that 6 equivalents of amine (i.e., 3 equivalents per carboxylate group) were optimal to afford the corresponding bisureas in good-to-high yield in a reasonable amount of time and in a reproducible manner without the need to heat the reaction. For example, use of 4 equivalents of n-butyl amine afforded product 7d in an 80% yield, while 6 equivalents provided a 90% yield. Given that the amines used are inexpensive, the use of an excess of them did not prove to be a disadvantage. CH₃CN proved to be the optimal solvent for the reaction, although THF could be used with only slightly lower yields. Other than the di-tert-butyl derivative, the bisurea products are nearly insoluble in either solvent and can be isolated by simple filtration followed by washing them with water (to remove phenol byproduct and excess amine) to afford compound 7 as powdery, white solids. The reaction yields are summarized in

Table 1. Synthesis of 1,6-dialkylbisureas (7) ^a.

R =	Yield (%)
methyl	96
ethyl	88
n-propyl	87
n-butyl	90
tert-butyl	57
cyclohexyl	77
benzyl	76

^a Generally, 6 equivalents of the corresponding amine was added to compound 8 in THF at rt for 24 h.

. The di-tert-butyl derivative 7e was soluble in both organic solvents, however, and necessitated removal of the solvent for product isolation. The generally low solubility of the bisureas in typical organic solvents (even in DMSO) necessitated the use of deuterated trifluoroacetic acid as a solvent for ¹H and ¹³C NMR spectral analysis, in which they were all readily soluble. The NMR spectra of the isolated solids revealed essentially pure compounds free of any significant phenol contamination (see the Supplementary Materials for the ¹H and ¹³C NMR spectra).

3.2. Thermal Cyclization of 1,6-Dialkylbisureas (7) to N-Alkyl Urazoles (1)

The bisureas (7) all had high melting points (>180 °C), with most being in the range of 230–250 °C. Conversion to the urazoles (100 mg samples) occurred readily upon heating neat samples of 7 [9]. At temperatures of ~230–250 °C, amine was released and could be observed to escape via either visible fuming and/or condensation of the amine near the top of the reaction vial. A reasonable mechanism for the conversion is provided in Scheme 8. All of the powders were heated to a final pot temperature of 260 °C to induce the cyclization reaction.

The urazole products (1) all have melting points lower than 260 °C. Therefore, as the cyclization reaction took place, a melt formed. The melt was occasionally swirled to aid in removal of liberated amine. For bisureas 7a and 7b, once fuming stopped, the mixture could be cooled immediately to afford crystalline urazoles 1a and 1b. However, for the bisureas derived from medium-sized amines (i.e., 7c,d), after fuming ceased, the sample temperature was reduced to 190 °C with continued heating for 25 min to allow time for the amine to completely escape. Cooling of the melt then afforded crystalline products. Longer heating times were generally required for bisureas derived from amines with higher boiling points (e.g., benzyl and cyclohexyl). Larger-scale reactions were also conducted starting with bisureas 7a (0.5 g and 1 g) and 7d (0.95 g) to afford the corresponding urazole products in comparable yields (see Section 2).

The identities of the known urazole products were confirmed using ¹H and ¹³C NMR spectroscopy (spectra are provided in the Supporting Information) and melting-point comparison with research data. Specific heating protocols for each of the bisureas are provided in Section 2. The yields for the urazole products are provided in

Table 2. Yields from thermal cyclization of 1,6-dialkylbisureas (7) to afford N-alkyl urazoles (1) ^a.

R =	% Yield
methyl	80
ethyl	92
n-propyl	89
n-butyl	78
cyclohexyl	85
benzyl	94

^a Generally, 100 mg of sample were heated to 260° for the durations described in Section 2.

(Column 2); they generally range from good to high.

Unfortunately, the tert-butyl derivative 7e failed to provide the corresponding urazole cleanly despite many efforts to optimize the heating protocol. In addition to the formation of urazole 1e as a major product, other unknown products were also formed. The reluctance of 1e to cleanly cyclize can likely be traced to the sterically bulky tert-butyl group that inhibits the initial nucleophilic attack of the nitrogen atom on the neighboring carbonyl group in the cyclization process (see Scheme 8).

4. Conclusions

The synthesis of urazoles via thermal cyclization of 1,6-dialkylbisureas (7) is a largely forgotten synthetic method [1,9]. The reason for its neglect is presumably due to the poor experimental procedures that were provided at the time, which led many to the conclusion that the results were difficult to reproduce. Furthermore, the dependence of the method on toxic reagents such as isocyanates for the synthesis of the bisureas made it unattractive. We have now provided a high-yielding and convenient method for the synthesis of the bisurea precursors that not only eliminates the use of isocyanates but also allows for their isolation via a simple filtration process. Finally, we have provided experimental details of the thermal conversion of the bisureas into the desired urazoles that allow for good-to-high yields of the urazoles in a reproduceable manner.