Synthesis of N-unsubstituted 1,2,3-triazoles via a cascade including propargyl azides, allenyl azides, and triazafulvenes

About thirty NH-1,2,3-triazoles with at least one additional functional group in a side chain at C-4 were prepared from propargyl substrates. These reactions included propargyl azides and their [3,3]-sigmatropic rearrangement to generate short-lived allenyl azides, which cyclized to form triazafulvenes that could be trapped by addition of N-or O-nucleophiles. In most cases, simple substrates and cheap sodium azide were utilized as starting compounds, and the syntheses were performed by using a one-pot procedure without isolation of any dangerous azides. This method to prepare NH-1,2,3-triazoles turned out to be compatible with quite different substitution patterns of the propargyl substrate.

We earlier discovered a one-pot cascade to transform propargyl halides or sulfonates 1 into Nunsubstituted 1,2,3-triazoles 5 that bear at C-4 a side chain with the functional group Nu (Scheme 1). 28This method includes nucleophilic substitution of 1 with the help of stoichiometric amounts of sodium azide.The resulting azide 2 is not isolated but slightly heated to undergo [3,3]sigmatropic migration of the azido group with formation of allenyl azide 3.Such intermediates can be detected by NMR spectroscopy or even isolated in special cases; 29 however, 3 tends to rapid ring closure to generate triazafulvene 4, which is trapped by addition of the nucleophile NuH to yield the final product 5.The intermediacy of very short-lived species 4 becomes plausible after treating optically active 3 with R 1 = H and R 2 = Me with the nucleophile MeOH to give completely racemic 5 via achiral 4. 28 In the presence of a Brønsted base, the one-pot cascade to prepare NH-1,2,3-triazoles can also transform the substrates 1 into products 8. 30,31 Allenyl azide 6 is formed by prototropic rearrangement of 2 in this case, and subsequent ring closure followed by nucleophilic addition of NuH at the resulting intermediate 7 leads to the heterocyclic product 8, which is an isomer of 5 (for R 1 ≠ R 2 ).There is no need that the base for the step 2 → 6 is identical with the nucleophile NuH, which is necessary for the formation of 8. Scheme 1. One-pot cascades to transform the starting compounds 1 into the N-unsubstituted 1,2,3triazoles 5 and 8.The reaction can also be performed with propargyl tosylate instead of 1a.b Heated at 70 °C for 60 h in the second step.c Propargyl chloride instead of 1a was utilized.d Recondensation of the reaction mixture after the first step, and heating at 50-60 °C for 3 days in the second step.e Heated at 50-55 °C for 24 h in the second step.f Heated at 60 °C for 10 days in the second step.
After the transformation 1a → 2a, the one-pot reaction sequence to prepare triazoles 5a-n was finished by adding an excess of the O-or N-nucleophiles NuH and heating the homogeneous mixture for 2-3 hours at 65-70 °C.Addition of sodium hydroxide was necessary in the cases of weak nucleophiles, such as water or phenol, to form stronger nucleophilic species (entries 1 and 9).Otherwise, low yields of the desired products, for example 5a and 5i, were obtained.We assume that in the absence of efficiently competing reagents, the product of type 5 will function as nucleophile to trap the short-lived intermediate 4.This would lead to unwanted oligomeric or polymeric products.Even when aliphatic alcohols with pKa > 14 are used as nucleophiles NuH, addition of sodium hydroxide could be advantageous, although only small proportions of the more nucleophilic alcoholates are generated.In the presence of strong bases, the reaction cascade can switch from 2 → 3 → 4 → 5 to the base-catalyzed sequence 2 → 6 → 7 → 8.In the case of the parent compounds 1a and 2a (R 1 = R 2 = H), however, the products 5 and 8 are identical.Basic aqueous solutions, in which the NH-1,2,3-triazoles 5 form triazolide salts, can be utilized to remove any non-salt impurities by washing with an organic solvent like diethyl ether.Such a procedure will liberate the reaction mixture from any not consumed dangerous azide of type 2. On the other hand, separation of the desired product 5 and inorganic salts is very difficult if 5 is present as triazolide or triazolium salt.Thus, pH adjustment of the reaction mixture is necessary to form the uncharged compound 5 before extraction with an organic solvent such as diethyl ether is started.An estimation of the pKa is useful to reach the isoelectric point of the NH-1,2,3-triazole 5.For example, pKa = 8.6 was calculated as an average value of the three tautomers of 4alkoxymethyl-1,2,3-triazoles depicted in Table 1. 48In most one-pot sequences, the mixtures were neutralized to pH = 6-7, and even when pH adjustment is optimized, extractions of 5 with the help of Soxhlet apparatus or continuous extraction with perforator equipment were often necessary, since the desired NH-1,2,3-triazoles are highly polar compounds.Especially in the case of 4aminomethyl-1,2,3-triazole (5k), water solubility is excellent and solubility in diethyl ether or chloroform proved to be very low (entry 11).Probably, the formation of internal salts, 4-(ammoniummethyl)-1,2,3-triazolides, is responsible for these properties of NH-1,2,3-triazoles bearing an aminomethyl group in the 4-position.Thus, extractive separation of inorganic salts and 5k was impossible.Consequently, the reaction mixture was carefully recondensed under reduced pressure after treatment of 1a with sodium azide.Thereafter, the condensate including a solution of 2a was heated with aqueous ammonia, which enabled simple workup by evaporation and isolation of the pure product 5k by recrystallization or sublimation.
Best yields of the desired products 5a-o were achieved when a large excess of the nucleophile NuH is utilized to trap the highly reactive intermediate of type 4.This is no problem in the case of cheap alcohols (see entries 2-5 for example), which can be easily separated from 5 in the workup.However, a compromise is necessary for more expensive or less volatile nucleophiles.Fortunately, amines such as dioctylamine are stronger nucleophiles, and an excess of only 1.33 equivalents led to an acceptable yield of the desired product 5n (entry 14).In the case of very volatile nucleophiles like ammonia, methylamine, or dimethylamine, the second step has to be performed in an autoclave to avoid early loss of NuH.Alternatively, lower temperatures (50-60 °C) and longer reaction times can be used (entries 11-13).When heated in the presence of organic azides, unsaturated compounds, such as allyl and propargyl alcohols, can undergo unwanted 1,3-dipolar cycloaddition to generate disubstituted 1H-1,2,3-triazoles.Obviously, this side reaction could not efficiently compete with the desired formation of 5g and 5h (entries 7 and 8), although the yields were somewhat lower than those of heterocycles 5b-e, which resulted from saturated alcohols (entries 2-5).
4][5][49][50][51] This leads to a single set of (broad) NMR signals for NH-1,2,3-triazoles and excludes the possibility to separate and isolate the tautomers at ambient temperature. 52,53We also obtained a single set of broad 13 C NMR signals for the carbon atoms C-4 and C-5 in each case of 5a-o and did not receive any hint that separation of the tautomers can be achieved at room temperature.Nevertheless, isolation of such tautomers by liquid column chromatography at ambient temperature was recently claimed for several NH-1,2,3-triazoles, including 5a and 5k, without any comment on the usually (very) rapid equilibration and any citation of the corresponding previous investigations. 54ne-pot procedures to prepare NH-1,2,3-triazoles from substituted propargyl halides or sulfonates.By using similar conditions as described in Table 1, we were able to synthesize NH-1,2,3-triazoles 5p-u (Scheme 2).Thus, propargyl bromides with substituents in 3-or 1-position as well as substituted propargyl chlorides and tosylates could be utilized as substrates.In the case of the tertiary substrate 1t, the yield of the desired product 5t was significantly lower.We assume that the transformation of 1t into the corresponding tertiary propargyl azide of type 2 was hampered by a competing elimination reaction, which should produce an enyne.
NH-1,2,3-triazoles from functionalized propargyl azides and methanol.In order to clarify whether the sequence 2 → 3 → 4 → 5 is affected or prevented by additional functional groups, for example, by an unwanted further [3,3]-sigmatropic rearrangement of intermediate 3, we tested the substrates 2v-z,aa-dd (Scheme 3).The starting compounds 2w,bb, 55 and 2dd 56 were described in literature and prepared by simple treatment of the corresponding bromides or mesylates with sodium azide.Similarly, the azide 2v was available from the appropriate and known 57 mesylate, whereas 2x was synthesized from hexa-4,5-dien-2-yn-1-ol 58 and phosphorus tribromide followed by the reaction of the resulting propargyl bromide with sodium azide. 59The amine 2y was accessible by treating 4-chlorobut-2-yn-1-ylamine hydrochloride 60 with an excess of sodium azide. 59In the cases of 2z and 2aa, the propargyl chlorides, which were necessary to synthesize the azido compounds, were prepared by the reaction of 1,4-dichlorobut-2-yne with ethylene glycol or catechol in the presence of potassium hydroxide. 59When 2-bromo-2-phenylacetaldehyde 61,62 was reacted with phenylethynylmagnesium bromide, the resulting bromohydrin was transformed into the azidohydrin on exposure to sodium azide followed by mesylation and a second nucleophilic substitution to produce diazide 2cc as a mixture of anti/syn diastereomers. 59heme 3. Treatment of functionalized propargyl azides with methanol to produce 4methoxymethyl-1,2,3-triazoles.
On gentle heating with an excess of methanol the propargyl azides 2v, 2w, and 2x afforded in moderate to good isolated yields the NH-1,2,3-triazoles 5v, 5w, and 5x, respectively.Obviously, unwanted side reactions of the intermediate allenyl azides 3v-x, such as electrocyclization of the carbon scaffold or another [3,3]-sigmatropic migration of the azido group, did not hamper the desired ring closure to form 4v-x and the interception products 5v-x.On the other hand, it is known from previous experiments that short-lived azidoallenes 3w and 3x can be trapped by [4+2]cycloaddition when solutions of the precursors 2w or 2x in tetrahydrofuran were heated in the presence of tetracyanoethene. 63The allenyl azide 3bb did not undergo an unwanted Cope rearrangement because of the mild conditions of its formation and subsequent ring closure.Thus, substrate 2bb led to NH-1,2,3-triazole 5bb via intermediates 3bb and 4bb.This is in contrast to the reported 64 isomerization of 5-thiocyanatohept-1-en-6-yne by flash vacuum pyrolysis at 260-400 °C.In this case, the [3,3]-sigmatropic migration of the thiocyanato group to generate an allenyl isothiocyanate was always accompanied by a succeeding Cope rearrangement.
On heating solutions of 2y, 2z, or 2aa in methanol, the NH-1,2,3-triazole products did not include those of intramolecular trapping of the intermediates 4y, 4z, or 4aa, namely pyrrole or 1,4dioxocine compounds.Instead, the methanol interception products 5y, 5z, and 5aa were isolated in moderate to good yields.Obviously, intramolecular trapping of the corresponding triazafulvene species 4 is highly disfavored, and even heating of 2y, 2z, or 2aa in non-nucleophilic organic solvents like chloroform or toluene yielded at the most trace amounts (<1%) of such products.The NH-1,2,3-triazole 5y was previously prepared by treating 4-methoxybut-2-yn-1-yl benzenesulfonate first with sodium azide and then with aqueous ammonia.Further reactions of NH-1,2,3-triazoles.When the one-pot synthesis of 1,2,3-triazoles was performed with the substrates 1a, sodium azide, and 2-chloroethanol, a second ring closure occurred under the reaction conditions, and the final product 11 was isolated (Scheme 5).Obviously, this heterocyclic compound was generated via intermediate 5ee.Attempts to prepare 11 by treatment of 5a with 1,2-dibromoethane in the presence of sodium hydroxide only led to very low yields.The Staudinger reaction of azide 5o with the help of triphenylphosphine and subsequent hydrolysis afforded the amino compound 5k.In contrast to the phosphorus reagent and the byproduct, the desired product shows excellent water solubility, and thus separation and purification of 5k was convenient.Oxidation of the methoxymethyl compound 5b with aqueous potassium permanganate yielded the carboxylic acid 12.When the one-pot synthesis of 5b was combined with the oxidation of the crude 4-methoxymethyl-1,2,3-triazole and the known 69 decarboxylation of 12, the parent 1,2,3-triazole (13) could be prepared in 33% yield based on the completely consumed sodium azide (first step).Treatment of 5b with boiling aqueous hydrogen iodide induced ether cleavage and quantitatively furnished the salt 14, which could be isolated as an air-sensitive solid.Hydrolysis of 14 led to 4-hydroxymethyl-1,2,3-triazole (5a) that confirmed the structure of 14. Scheme 5. Further reactions of NH-1,2,3-triazoles.

Conclusions
The one-pot cascade including propargyl azides, allenyl azides, and triazafulvenes turned out to be a convenient method to prepare NH-1,2,3-triazoles.Although this method includes several steps, the formation of the desired products is not hampered by additional functionalities in the propargyl substrates and is compatible with the installation of quite different N-or O-functional groups in the side chain at C-4 of the triazoles.Some of these functionalized NH-1,2,3-triazoles show excellent water solubility, which is a challenge for appropriate workup.36][38][39][40][41][42]45

Experimental Section
General.Alternative methods to prepare compounds 5b, 70 5i, [70][71][72] 5o, 37 and 5y 30 were reported in the literature.Since we already described the synthesis of 5a 30 and 5k 28 , these procedures are not repeated here.Melting points were determined with a Pentakon Dresden Boetius apparatus, and are uncorrected.Refractive indices were measured with a refractometer from Carl Zeiss.FTIR spectra were recorded on a Bruker IFS 28 FTIR spectrophotometer.IR measurements were made on solutions in KBr cuvettes or as potassium bromide pellets. 1 H NMR spectra were recorded on Varian Gemini 2000 or Unity Inova 400 spectrometers operating at 300 and 400 MHz, respectively.Using the same spectrometers, 13 C NMR data were achieved at 75.4 and 100 MHz.NMR signals were referenced to TMS (δ = 0) or solvent signals and recalculated relative to TMS.The multiplicities of 13 C NMR signals were determined with the aid of DEPT135 experiments.MS spectra were measured on a MS9 spectrometer from AEI/Manchester.GC-MS spectra were acquired with Shimadzu quadrupole mass spectrometer.Ionization was performed by EI (70 eV).For the previous separation, a Shimadzu GC-17A gas chromatograph with thermal conductivity detector and DB-1 column (30 m) was used.HR-MS (ESI) spectra were recorded on Applied Biosystems Mariner 5229 mass spectrometer or Bruker micrOTOF-QII spectrometer.For elemental analyses, a Vario EL elemental analyzer from Elementar Analysensysteme GmbH Hanau or a Vario Micro Cube from Elementar were used.Elemental analyses of explosive azides could not be conducted.Flash column chromatography was performed on silica gel 60 (0.04-0.063 mm) from Machery-Nagel.Separations by HPLC were carried out with Knauer HPLC-pump 64 and UV-detector ( 254 nm).
General procedure for the synthesis of 5b−i.In the first step, sodium azide (1.0 eq) was dissolved in water (15.0 eq); 3-halo-1-propyne (1.1 eq, X = Br, Cl) and defined organic solvent (10 eq, shown in Table 1) were added and stirred at about 30 °C for 18−24 h.In the second step, sodium hydroxide (10 eq) and corresponding alcohol/phenol (10 eq), if needed, were added and heated to 65−70 °C for 2−3 h.Workup A. The solution was cooled to ambient temperature, neutralized with hydrochloric acid to pH 6−7, and solvents evaporated under vacuum.The residue was dissolved in water, extracted with diethyl ether, and dried over magnesium sulfate.Diethyl ether was evaporated and the crude products were purified by bulb tube distillation (5b−f) or by recrystallization (5i) to afford pure products.Workup B. The solution was diluted with water and washed with diethyl ether at pH 12 to remove organic side products.It was neutralized with hydrochloric acid to pH 6−7 and extracted with diethyl ether.The organic layer was washed with water and dried over magnesium sulfate.The volatiles were removed under reduced pressure to afford pure products (5g,h).Hz, 3H, CH3), 3.60 (q, 3 J 7.0 Hz, 2H, OCH2Me), 4.71 (s, 2H, OCH2), 7.79 (s, 1H, CH), 13.98 (s, 1H, NH). 13   Hz, 3H, CH3), 3.57 (q, 3 J 7.0 Hz, 2H, CH2Me), 3.69 (m, 4H, O(CH2)2O), 4.72 (s, 2H, OCH2), 7.67 (s, 1H, CH), NH signal could not be detected. 13 General procedure for the synthesis of 5l−n.In the first step, sodium azide (1.0 eq) was dissolved in water (15.0 eq); 3-bromo-1-propyne (1a, 1.1 eq) and 1,4-dioxane (10 eq) were added and stirred at about 30 °C for 18−24 h.In the second step, an aqueous solution of amine (80 eq) or rather pure dioctyl amine (1.3 eq) were added and heated to 50−60 °C for 1−10 days.Workup A. The solution was cooled to ambient temperature, and solvents were evaporated under vacuum.The residue was dissolved in water and extracted with diethyl ether.The volatiles were evaporated, and the crude products were purified by bulb tube distillation (5l,m).Workup B. The solution was diluted with water and extracted with diethyl ether at pH 7. The organic layer was washed with water and dried over magnesium sulfate.The volatiles were removed under reduced pressure.Dioctyl amine was evaporated by short way distillation at 150 °C/0.01 mbar to afford pure product (5n).In the first step, sodium azide (0.24 g, 3.7 mmol) was dissolved in methanol (10 mL) and water (3 mL); 1-bromo-2-tridecyne 73 (1q, 1.0 g, 3.9 mmol) and methanol (10 mL) were added and stirred at 30 °C for 24 h.In the second step, methanol (150 mL) was added, and the mixture was heated to 60 °C for 4 days.The solution was cooled to ambient temperature and methanol evaporated under vacuum.The residue was dissolved in water and extracted with methyl tert-butyl ether.The organic layer was washed with water and dried over magnesium sulfate.The volatiles were removed under reduced pressure to afford the pure product.

4-(N-Methylamino)methyl-1,2,3-triazole (5l
Yellow oil; yield (91%).IR (film): 3443 (NH) cm −1 .General procedure for the synthesis of 5r−t.The 3-halo-1-propynes 1r, 74 1s, 75 or 1t 76 (1.0 eq) were dissolved in a solution of dioxane/water (3:1); sodium azide (6.0 eq) and ammonium chloride (3.0 eq) were added and stirred at 85 °C for 16−20 h.The workup was carried out as described at workup B of 5b−i.General procedure for the synthesis of 5v−dd.Propargyl azide 2v−z,aa−dd 59 was dissolved in methanol and stirred at 60 °C for 1−17 days.The volatiles were removed under reduced pressure to afford the pure products 5v−z,bb.In case of 5aa, the crude product was suspended in chloroform to crystallize the pure product.The workup of 5cc was carried out as described at workup B of 5b−i.The residue of 5dd, after evaporation of the solvents, was dissolved in a solution of methanol/water (3:1).This mixture was triturated with diethyl ether, thereby, resulting in the precipitation of a solid, which was washed with methanol to afford the pure product 5dd.(11).Sodium azide (6.5 g, 0.1 mol) was dissolved in water (80 mL); 3-bromo-1-propyne (1a, 8.3 mL, 0.11 mol) in ethylene glycol dimethyl ether (400 mL) was added, and the mixture was stirred at 22 °C for 18 h.The reaction mixture was charged with 2-chloroethanol (1.0 L, 14.93 mol) and heated to 65−70 °C for 3 h.The volatiles were removed under reduced pressure.The residue was dissolved in water (1.5 L), potassium hydroxide (6.17 g, 0.11 mol) was added and heated under reflux for 18 h.Water was evaporated, the residue was suspended in chloroform (150 mL), and undissolved salts were removed by filtration.The solvents were evaporated under reduced pressure and the crude product was purified by bulb tube distillation at 140−150 °C/0.01 mbar to afford the pure product, which is sensitive to oxidation; thus, the ether functionality is easily transformed into a lactone (6,7-dihydro- Staudinger reaction of 5o.4-Azidomethyl-1,2,3-triazole (5o, 0.68 g, 5.5 mmol) was dissolved in dry diethyl ether (20 mL), charged with a solution of triphenylphosphine (1.57g, 6.0 mmol) in dry diethyl ether (20 mL), and stirred at room temperature for 24 h.The diethyl ether was evaporated, the residue diluted with tetrahydrofuran (40 mL) and water (10 mL), and refluxed for 1.5 h.The volatiles were removed under reduced pressure, the residue was stirred in water (50 mL), and washed with organic solvent.The water was evaporated to afford the pure product 5k (0.48 g, 4.9 mmol, 89%), which was identical with the previously synthesized 28 compound.
Oxidation of 5b followed by decarboxylation to form 13. 3-Bromo-1-propyne (1a, 23.8 g, 0.200 mol) was dissolved in methanol (100 mL), charged with sodium azide (11.8 g, 0.1815 mol) in water (50 mL), and stirred at room temperature for 14 h.The mixture was added to a solution of sodium hydroxide (36.3 g, 0.908 mol) in methanol (700 mL) and heated under reflux for 2 h.The methanol was evaporated and the residue diluted with water (400 mL).Potassium hydroxide (17.7 g, 0.315 mol) was added to the solution, which was charged in portions with potassium permanganate (41.1 g, 0.259 mol).The mixture was stirred for 12 h at room temperature and,   thereafter, heated at 70 °C for 3 h.The suspension was filtered and the clear filtrate dissolved in hydrochloric acid.During the evaporation of the volatiles, carboxylic acid 12 69 crystallized in the cold solution.The solid was separated by filtration and heated at 300 °C in an open apparatus of recondensation to remove carbon dioxide.The triazole 13 (4.18g, 60.5 mmol, 33%, based on sodium azide) was isolated by recondensation (5•10 −3 mbar) at room temperature.Synthesis of 14. 4-Methoxymethyl-1,2,3-triazole (5b, 12.0 g, 0.11 mol) and hydroiodic acid (24.0 g, 57% aqueous solution) were heated under reflux and methyl iodide and water removed by distillation for 3 h until yellow hydroiodic acid was noticed.The solution was cooled to ambient temperature and the volatiles were removed under reduced pressure (5•10 −3 mbar) to afford the crude product (34.4g, 0.10 mmol, 96%).The pure product was obtained by recrystallization (MeCN).(20 mL) was added dropwise to a solution of potassium bicarbonate (3.03 g, 30 mmol) in water (20 mL), and the mixture was stirred at room temperature for 24 h followed by heating at 50 °C for 6 h.After evaporation, the residue was dissolved in water, and sulfuric acid was added to generate a pH of 5−6.Water was removed under reduced pressure, and the crude product was extracted by Soxhlet extraction with diethyl ether for 2 days.The diethyl ether was evaporated and a distillation (5•10 −3 mbar) afford the pure product 5a (0.61 g, 6,2 mmol, 61%), which was identical with the previously prepared 30 compound.