Refined syntheses of hydrodipyrrin precursors to chlorin and bacteriochlorin building blocks

Abstract

Bromo-substituted hydrodipyrrins are valuable precursors to synthetic bromo-chlorins and bromo-bacteriochlorins, which in turn are versatile substrates for derivatization in pursuit of diverse molecular designs. 8-bromo-2,3-dihydro-1-(1,1-dimethoxymethyl)-3,3-dimethyldipyrrin (1) is a crucial precursor in the rational synthesis of the bacteriochlorin building block 3,13-dibromo-8,8,18,18-tetramethylbacteriochlorin (H₂BC-Br³Br¹³). 8-bromo-2,3,4,5-tetrahydro-1,3,3-trimethyldipyrrin (2) is a crucial precursor in the rational synthesis of the analogous 3,13-disubstituted chlorin building block (e.g. H₂C-Br³M¹⁰Br¹³). The routes to 1 and 2 share a common precursor, namely 4-bromo-2-(2-nitroethyl)-1-N-tosylpyrrole (6-Ts), which is derived from pyrrole-2-carboxaldehyde. The prior seven-step synthesis of 1 from pyrrole-2-carboxaldehyde has limited access to H₂BC-Br³Br¹³ given the large excesses of materials, extensive reliance on column chromatography, and low overall yield (1.4%). Refined procedures for synthesis of the common precursor 6-Ts as well as 1 and 2 afford the advantages of (1) diminished consumption of solvents and reagents, (2) limited or no use of chlorinated solvents, (3) limited or no chromatography, and (4) improved yields of most steps. Streamlined procedures enable the final two or three transformations to be performed without purification of intermediates. The new procedures facilitate the expedient preparation of 1 and 2 at the multigram scale.

Graphical Abstract

INTRODUCTION

The practical utilization of tetrapyrrole macrocycles relies on the availability of simple yet versatile methods of preparation of synthetic analogs of the naturally occurring molecules. Over the years we and others have been working to develop such methods, particularly for the synthesis of tetrapyrrole building blocks for use in diverse studies [for recent reviews, see references 1–6]. The presence of specific substituents at designated sites on the perimeter of the tetrapyrrole macrocycle enables subsequent elaboration to meet a broad range of molecular design objectives. Significant milestones from our group include access to porphyrins bearing up to four different meso-substituents (ABCD-porphyrins) [7, 8]; chlorins wherein a given substituent can be located at any of the six β-pyrrole or four meso-substituents [9–11]; and more recently, bacteriochlorins that bear substituents at a number of positions around the perimeter of the macrocycle [12–15]. As opportunities presented by the tetrapyrrole building blocks have emerged, the availability of scalable synthetic routes — to give at least 100 mg if not several-g quantities of the target macrocycle — has become of paramount importance. The de novo synthesis of tetrapyrrole building blocks complements semi-synthesis routes that begin with naturally occurring molecules including protoporphyrin [16], chlorophylls [17], and bacteriochlorophylls [18].

Cursory examination of the most time-consuming aspects of implementing typical tetrapyrrole syntheses has inevitably pointed to the use of chromatography to isolate small quantities of materials. Other practical limitations to larger-scale syntheses include reliance on chlorinated solvents and frequent use of reagents in sizable excess. These impediments prompted us to take up the challenge of developing refined, scalable syntheses of tetrapyrrole macrocycles. Scalability issues have since been addressed in the synthesis of diverse porphyrins, where the necessity for chromatography has largely been expunged [7, 8, 19–23]. The methodology for preparing hydroporphyrins is more recent, and issues for scalability have necessarily received less attention.

Two hydroporphyrin building blocks of great interest are a 3,13-dibromochlorin and a 3,13-dibromobacteriochlorin. The 3,13-disubstitution pattern is particularly favorable given that the long-wavelength absorption band derives from a transition that is polarized along the molecular axis that spans rings A and C, which contain the 3- and 13-positions, respectively. Introduction of auxochromic groups at sites in rings A and C enables the long-wavelength absorption band to be tuned across quite a wide spectral region [24, 25]. Given that the position of the long-wavelength absorption band is a key determinant in the energy of the excited singlet state, the ability to tune the position of the absorption band is invaluable for studies in artificial photosynthesis. Examples of both dibromo-hydroporphyrin building blocks (chlorin H₂Br³R¹⁰Br¹³ [24–26] and bacteriochlorin H₂BCBr³Br¹³ [14, 15]) have already been exploited to prepare collections of molecules for spectroscopic examination (Scheme 1). However, scalable syntheses have not been developed for the essential bromo-substituted precursors of each macrocycle.

Scheme 1.

Routes to chlorin and bacteriochlorin building blocks proceed through a common synthon (6-Ts). Differentiation occurs with the distinct α-substituent introduced upon Michael addition with enone 7a or 7b

The synthetic pathways to both dibromo-hydroporphyrin building blocks are shown in Scheme 1. The chlorin macrocycle is prepared by condensation of a Western half and an Eastern half [9]; the former is provided by a bromo-substituted tetrahydrodipyrrin (2) [26]. The two bridging meso-carbons are provided by the α-methyl group of the tetrahydrodipyrrin and the α-formyl group of the dipyrromethane (Eastern half). The bacteriochlorin macrocycle is prepared by the self-condensation of two molecules of a bromo-substituted dihydrodipyrrin (1) [14]. In this case, the two bridging meso-carbons are provided by the α-acetal group of the dihydrodipyrrin compound.

Chlorin formation is preferentially achieved with a tetrahydrodipyrrin although a dihydrodipyrrin can also be employed [27, 28]; by contrast, to date the only route to bacteriochlorins proceeds via the dihydrodipyrrin [12–15]. The formation of the tetrahydrodipyrrin (for chlorins) versus the dihydrodipyrrin (for bacteriochlorins) stems from the conditions of the cyclization of the precursor pyrrole-nitrohexanone: the reductive cyclization with Zn/HCOONH₄ affords the tetrahydrodipyrrin whereas use of NaOMe and TiCl₃ at pH 6 affords the dihydrodipyrrin. The introduction of the α-methyl versus α-acetal unit is achieved by reaction of a common pyrrole synthon (6-Ts) with an α,β-unsaturated ketone, namely mesityl oxide (7b) versus the dimethyl acetal of mesityl oxide (7a), respectively. The pyrrole synthon that is common to both syntheses bears the bromo substituent and a 2-nitroethyl unit, and ultimately is derived from pyrrole-2-carboxaldehyde (3). A further consideration for the chemistry following the branchpoint is that mesityl oxide (7b, for chlorin syntheses) is commercially available and inexpensive; by contrast, the α,β-unsaturated ketone 7a for the bacteriochlorin synthesis must be prepared by independent synthesis.

The objective in this study was first to refine the preparation of the common synthon 6-Ts, and second to extend the refinement to both of the bromo-hydrodipyrrin precursors to the chlorin and bacteriochlorin building blocks. On the latter goal, we focused chiefly on the bacteriochlorin precursor 8-bromo-dihydrodipyrrin 1. Our previous route to 1 required four column chromatography procedures, the use of large amounts of chlorinated solvents, and afforded a low overall yield (1.4% for seven steps) [14, 26]. Thus, the specific goals of refinement were to limit chromatographic purification, limit the use of chlorinated solvents such as dichloromethane, diminish consumption of solvents and reactants, and to improve the overall yield.

Compound 1 provides a number of synthetic challenges despite its simple appearance. The challenges include: (1) introduction of the acid-sensitive acetal moiety, (2) formation of the α-pyrrolic methine bridge, and (3) the presence of two different heterocycles (pyrrole and pyrroline), where pyrrole is susceptible at two sites toward electrophilic substitution, whereas the pyrroline imine unit is prone to reduction and addition reactions. The conditions developed to meet the synthesis of 1 (for bacteriochlorins) were then extended to the synthesis of 2 (for chlorins). We note that the synthesis of the desbromo analog of 2 has been refined [29]. Hydrodipyrrins 1 and 2 are representative of a broad class of hydrodipyrrins that serve as potential precursors to hydroporphyrins [30]. As such, the resulting refined syntheses should facilitate access to substantial quantities of bromo-substituted chlorins and bacteriochlorins.

RESULTS AND DISCUSSION

Synthesis of the common precursor (6-Ts) to bromo-chlorins and bromo-bacteriochlorins

The conversion of pyrrole-2-carboxaldehyde (3) to 6-Ts is shown in Scheme 2. In each of the four steps, the refined conditions identified (vide infra) are listed.

Scheme 2.

Refined stepwise synthesis of nitroethylpyrrole 6-Ts

Step 1. Bromination of pyrrole-2-carboxaldehyde.

A reported method [31] for bromination of pyrrole-2-carboxaldehyde (3) that uses Br₂ gave, in our hands, a mixture of 4-bromo, 5-bromo and 4,5-dibromo substituted products. Bromination using NBS was superior, resulting in the desired 4-bromopyrrole-2-carboxaldehyde (4) as the only detected product. Prior application of this procedure at a 0.25 M concentration of pyrrole-2-carboxaldehyde with one molar equivalent of NBS at −78 °C gave 4 in 55% yield after crystallization from hexanes/THF [26]. Here we aimed to carry out the reaction at higher concentration and simplify the workup. For the workup, a key issue was to maintain high yield while suppressing the amount of residual succinimide in the product. The reaction and workup conditions that were investigated are described in the Supporting Information. The best conditions identified entailed reaction of pyrrole-2-carboxaldehyde (1.0 M) with one molar equivalent of NBS at 0 °C for 15 min. Recrystallization from EtOH/H₂O gave no detectable succinimide and allowed the use of small volumes of solvent. The desired 4 was obtained in 81% yield.

Step 2. Protection of the pyrrole nitrogen.

Protection of the pyrrole nitrogen was found to be necessary due to the unstable nature of the unprotected analog of 6-Ts, i.e. compound 6 [26]. Considering the range of conditions for the removal of a p-toluenesulfonyl group (vide infra) as well as the known crystalline nature of 2-(2-nitrovinyl)-N-phenylsulfonylpyrroles [32], 4 was subjected to N-tosylation. Previously, the reaction was carried out at 0.15 M concentration in 68% yield [26]. We were able to increase the concentration to 1.0 M by slowly treating a NaH suspension at 0 °C with 4, followed by addition of p-toluenesulfonyl chloride. Furthermore, the reaction time was shortened from 16 h to 3 h following p-toluenesulfonyl chloride addition. These conditions gave 4-Ts in 73% yield after recrystallization from ethyl acetate/hexanes (the residual 4-Ts in the mother liquor could be isolated by concentration and a second crystallization). It is important to note that the reverse addition (adding NaH to a solution of 4) at these concentrations may result in a vigorous or even explosive reaction.

Step 3. Nitroaldol condensation.

The typical conditions for the nitroaldol condensation (Henry reaction) of pyrrole-2-carboxaldehydes require the presence of an ammonium salt and/or weak base. The condensation was previously carried out using an 18-fold excess of nitromethane in the presence of ammonium acetate under reflux [26, 29]. The large excess of nitromethane can be difficult to remove, and attempted precipitation of the product from excess nitromethane (by addition of water) resulted in “oiling out”. Furthermore, the explosive hazard of nitromethane prompted investigation of alternative reaction conditions. The reaction of 4-Ts proceeded at room temperature using potassium acetate and methylamine hydrochloride using only 2.5 equiv. of nitromethane, and the precipitated product could be removed by simple filtration. Washing the crude reaction product with water and cold ethanol resulted in pure 5-Ts in good yield (91%).

Step 4. Nitrovinylpyrrole reduction.

The previous route utilized NaBH₄ in the presence of silica in CHCl₃/2-propanol (3:1) with 5-Ts at a concentration of 0.08 M to give the product in 76% yield after column chromatography [26]. The procedure requires excess NaBH₄ (2 mol equiv.) and undesired chlorinated solvents. The seeming simplicity of the reduction of 5-Ts is confounded by the production of dimeric byproduct I (Chart 1) derived by nucleophilic attack of the nitronate anion of one product molecule (6-Ts) to the nitrovinyl moiety of one molecule of starting material (5-Ts). Here we investigated different reaction conditions for the reduction of 5-Ts (see Supporting Information). The use of LiBH₄ in THF was found to provide multiple attractive features, including few equivalents of reductant (1 mol equiv.), higher reaction concentration (0.2 M), no requirement for added silica, limited formation of dimeric byproduct I, and isolation of 6-Ts in 77% yield after recrystallization from 2-propanol. It is noteworthy that small amounts of contaminating I are not readily detected by elemental analysis given the similarity in elemental composition: I has C₂₆H₂₄Br₂N₄O₈S₂ (reduced formula C₁₃H₁₂BrN₂O₄S) whereas the desired product 6-Ts has C₁₃H₁₃BrN₂O₄S. However, I can be readily distinguished by ¹H NMR spectroscopy. Any dimeric byproduct I that is formed is removed upon recrystallization.

Chart 1.

Dimeric byproduct derived upon reduction of nitrovinylpyrrole 5-Ts

Table 1 compares the four steps of the previous route with those presented here. The refined synthesis enabled preparation of 67.5 g of the common synthon 6-Ts.

Table 1.

Transformations in previous versus refined route to common synthon 6-Ts

Step	Previous routea	Yield	Revised route	Yield	Product
1	0.25 M in THF,	55%	1.0 M in THF,	81%	4
	1 equiv. NBS,		1 equiv. NBS,
	−78 °C, 1 h,		0 °C, 15 min,
	recrystallized (hexanes/THF)		recrystallized (EtOH/water)
2	0.15 M in THF,	68%	1.0 M in THF,	73%	4-Ts
	1.2 equiv. NaH,		1.1 equiv. NaH,
	rt, 16 h,		rt, 3 h,
	recrystallized (EtOAc/hexanes)		recrystallized (EtOAc/hexanes)
3	18 equiv. CH3NO2,	crude	2.5 equiv. CH3NO2,	91%	5-Ts
	0.8 equiv. NH4OAc,		0.8 equiv. KOAc,
	reflux 3 h		0.8 equiv. MeNH2·HCl,
			rt, 2 h
			wash with EtOH/water
4	0.08 M	58%	0.2 M THF, LiBH4,	77%	6-Ts
	CHCl3/2-propanol,	steps	−10 °C, 15 min,
	silica, NaBH4, 1.5 h	3+4	recrystallized from 2-propanol
	column chromatography
1–4	overall process	22%	overall process	41%	6-Ts

^aSteps 1–4 are from reference 26.

Synthesis of the bromo-dihydrodipyrrin precursor to bromo-bacteriochlorins

Three routes were pursued for the conversion of 6-Ts to the target bromo-dihydrodipyrrin 1 (Scheme 3). In route A, 6-Ts undergoes three sequential transformations (Michael addition with the dimethyl acetal of mesityl oxide (7a) to give the N-tosylpyrrole-nitrohexanone 8a-Ts, tosyl removal to give 8a, and reductive cyclization to give 1). In route B, the two-step conversion of 8a-Ts → 1 is performed without purification of the intermediate 8a. In route C, the three-step conversion of 6-Ts + 7a → 1 is undertaken without purification of any intermediates. One issue in each route pertains to whether the tosyl group is removed prior to/concomitant with (route C) or after (routes A and B) the Michael addition with α,β-unsaturated ketone 7a. The development of refined conditions for these routes is delineated in the following section.

Scheme 3.

Three routes to 8-bromo-dihydrodipyrrin 1 for bacteriochlorin syntheses. Route A entails stepwise conversion of 6-Ts to give 1, whereas routes B and C are streamlined

Route A. Stepwise synthesis.

Step 5: Michael addition with 1,1-dimethoxy-4-methyl-3-penten-2-one (7a).

The previously reported procedures for the Michael addition of 1,1-dimethoxy-4-methyl-3-penten-2-one (7a) and 6-Ts utilized fluoride reagents (e.g. TBAF or CsF [14]) to effect reaction in acetonitrile at 0.1–0.5 M concentrations. Neat conditions would be attractive for scale-up purposes. In this regard, the analogous reaction of 6-Ts and mesityl oxide (7b) using DBU under neat conditions was reported to give high yields, whereas treatment of the analogous 2-(2-nitroethyl)pyrrole and 7b with CsF under neat conditions did not give the desired product [29]. Therefore, the Michael addition of 6-Ts and 7a was carried out under neat conditions using DBU to afford 8a-Ts in 66% yield after triturating the crude oily brown product with diethyl ether. The acetal 7a and DBU were used in 3-fold excess because the reaction does not go to completion (leftover starting material remained even after prolonged reaction times) using less of either reagent. X-ray structural characterization was obtained for compound 8a-Ts (see Supporting Information).

Step 6. Deprotection of tosylated pyrrole-nitrohexanone 8a-Ts.

Numerous conditions for cleavage of an N-arylsulfonyl group from a nitrogen heterocycle have been reported over the past decade including 5 N NaOH in aqueous dioxane [33], TBAF in THF [34], KF on basic alumina under microwave conditions [35], LiOH and α-mercaptoacetic acid in DMF [36], and K₂CO₃ in methanol [37]. Deprotection of the tosylated pyrrole-nitrohexanone 8a-Ts was carried out previously using LiOH and α-mercaptoacetic acid in DMF [14]. The procedure entailed 10 equiv. of LiOH and 4 equiv. of mercaptoacetic acid in DMF at a concentration of 0.2 M for 18 h. The high boiling point of DMF further complicates the workup. Reexamination of the reaction revealed that KF on basic alumina under microwave conditions [35] resulted in decomposition, whereas K₂CO₃ in methanol [37] resulted in an undesired, unidentified major product with the tosyl group still intact. Use of TBAF as the cleavage agent proved promising, and a study was carried out to identify suitable deprotection conditions (see Supporting Information). The best conditions entailed use of 1.0 M 8a-Ts with 1 equiv. of TBAF (1.0 M) at 66 °C for 1 h, which afforded the product 8a in 75% yield.

Application of the refined conditions using 1.2 equiv. of TBAF in a 1.0 M THF solution at reflux effected complete deprotection after 1 h. Due to the unstable nature and relatively low melting point of the product 8a (77–81 °C), purification was only achieved by column chromatography. The product so obtained is a yellow oil, which darkens in a few hours at room temperature indicating partial decomposition, but solidifies upon storage at −10 °C. The elemental analysis of this compound was unsatisfactory. However, ¹H NMR and mp analysis were identical to reported values. A short study was performed to probe the mechanism of the TBAF-mediated deprotection, which is apparently not known. The study was fruitful only in ruling out several possibilities (see Supporting Information).

Step 7. Reductive cyclization.

To obtain the desired dihydrodipyrrin 1 (and not form the unwanted tetrahydrodipyrrin), reductive cyclization is typically carried out by formation of the nitronate anion and subsequent treatment with a buffered TiCl₃ solution [27, 38]. Attempted reductive cyclization of tosyl-protected pyrrole-nitrohexanone 8a-Ts (without prior deprotection) resulted mainly in recovered starting material 8a-Ts accompanied by a small amount of deprotected 8a and the desired dihydrodipyrrin 1 (< 5% yield). Therefore, reductive cyclization was carried out by formation of the nitronate anion of the deprotected pyrrole-nitrohexanone 8a (0.4 M in THF) with 3 equiv. of NaOMe for 45 min at 0 °C, followed by treatment with a buffered TiCl₃ solution for 16 h. Previously, a freshly prepared TiCl₃ solution (8.6 wt.%, 28% HCl) was used, which required large amounts of NH₄OAc (65 equiv. versus TiCl₃) to buffer the reaction mixture to pH 6 [14]. Using TiCl₃ powder without HCl allows for significantly smaller amounts of NH₄OAc (6 equiv. versus TiCl₃). Purification was achieved by low temperature crystallization (−10 °C for 48 h) in 2-propanol to afford dihydrodipyrrin 1 in 29% yield. The mother liquor could be chromatographed to give additional product for a total yield of 35%. The synthesis of 1 via the seven sequential steps constitutes route A.

An X-ray crystal structure was obtained to confirm the final product. The crystal structure of 1 contains two symmetry independent molecules in the asymmetric unit (space group P 2₁/c). The structure of one of the molecules is shown in Fig. 1 (see Supporting Information for data). A literature survey showed that only a few crystal structures of analogous compounds have been reported, all of which are polyalkyldipyrrins lacking the pyrroline ring [39–41].

Fig. 1.

ORTEP drawing of 8-bromo-dihydrodipyrrin-acetal 1. Ellipsoids are at the 50% probability level and hydrogen atoms were drawn with arbitrary radii for clarity

Route B. Direct synthesis of dihydrodipyrrin 1 from tosyl-protected pyrrole-nitrohexanone 8a-Ts.

Route B begins with pure tosyl-protected pyrrole-nitrohexanone 8a-Ts obtained from Michael addition of 6-Ts and 7a. Here the two subsequent steps (detosylation, reductive cyclization) carried out in route A are condensed to one. To circumvent the column chromatography after detosylation of 8a-Ts, the crude reaction mixture containing deprotected pyrrole-nitrohexanone 8a was used directly in the reductive cyclization step. In this instance, reductive cyclization was carried out using an inexpensive, commercially available TiCl₃ solution (20% in 3% HCl) instead of TiCl₃ powder (which was used in route A). Use of the commercially available TiCl₃ solution simplified handling of TiCl₃ and allowed for the use of relatively small amounts of NH₄OAc (9 equiv. versus TiCl₃). Furthermore, the NH₄OAc used in this procedure was not dissolved in water, which decreases the amount of water in the reaction mixture. Purification after the two-step synthesis entailed filtration of the crude reaction mixture through a pad of silica, concentration of the filtrate, and recrystallization from 2-propanol to afford dihydrodipyrrin 1 (16% yield after two steps). This direct conversion produced a higher yield (29%) on a smaller scale (4 mmol), but the yield decreased once the reaction was scaled up (30 mmol). A possible reason for the lower yield could stem from incomplete degassing of the larger scale reaction mixture, whereupon residual oxygen oxidized the TiCl₃. A second explanation could be the quality of NaOMe, which plays an important role in anion formation of compound 8a. Larger amounts of MeOH left in the freshly prepared NaOMe can significantly decrease the overall yield.

Route C. Direct synthesis of dihydrodipyrrin 1 from 6-Ts.

Route C condenses the three individual steps (Michael addition, detosylation, reductive cyclization) of route A into one. Thus, the route begins with tosyl-protected nitroethylpyrrole 6-Ts and α,β-unsaturated ketone-acetal 7a. Direct syntheses of deprotected pyrrole-nitrohexanone 8a from 6-Ts have been reported using either CsF or TBAF [14], whereby the fluoride reagent promotes both the Michael addition and the tosyl cleavage. Direct conversion using CsF required a large excess of acetal 7a, high temperatures, and was not reproducible. Reported procedures using TBAF employed a large excess of TBAF (10 equiv.) and required chromatography.

The reaction of tosyl-protected nitroethylpyrrole 6-Ts and 1.2 equiv. of acetal 7a in the presence of three equiv. of TBAF (1.0 M) in THF solution afforded a direct conversion to the deprotected pyrrole-nitrohexanone 8a. In this manner, two steps (Michael addition, tosyl deprotection) were condensed into one. ¹H NMR spectroscopy and TLC analysis confirmed formation of 8a. The crude material was used directly in the reductive cyclization as described in route B. In this manner, dihydrodipyrrin 1 was obtained in 14% overall yield (from 6-Ts and 7a) after crystallization from 2-propanol.

Comparison of routes from 6-Ts to 1.

A comparison of the individual steps in the previous versus refined synthesis of 1 from common synthon 6-Ts is presented in Table 2. The overall yield of the refined process (5.7%) is 4 times greater than that of the prior synthesis. In addition, the concentration of each reaction is generally greater, and the chromatography steps have been decreased from four steps to one.

Table 2.

Transformations in previous versus refined stepwise route (A) from 6-Ts to 1

Step	Previous routea	Yield	Revised route	Yield	Product
5	0.1 M in CH3CN, 5 equiv. CsF,	44%	neat conditions, 3 equiv. DBU,	66%	8a-Ts
	rt, 24 h, column chromatography		rt, 15 min, trituration with ether
6	0.2 M in DMF, 10 equiv. LiOH,	45%	0.83 M in THF, 1.2 equiv. TBAF,	60%	8a
	4 equiv. HSCH2CO2H, rt, 18 h,		reflux, 1 h,
	column chromatography		column chromatography
7	5 equiv. NaOMe,	33%	3 equiv. NaOMe,	35%	1
	TiCl3 (8.6% wt, 28% HCl),		TiCl3 (20% wt, 3% HCl),
	65 equiv. NH4OAc,		9 equiv. NH4OAc,
	rt, 16 h, column chromatography		rt, 16 h, crystallized (2-propanol)
1–7	overall process	1.4%	overall process	5.7%	1

^aSteps 5–7 are from reference 14.

A comparison of the three routes (Scheme 3) is provided in Table 3. Routes A–C have comparable yields; however routes B and C eliminate an undesirable silica column purification in the synthesis of dihydrodipyrrin 1. Route C has the advantage of reducing the amount of the precious acetal 7a used in the Michael addition, employing 1.2 equivalents of 7a (versus 6-Ts) compared to 3 equivalents in routes A and B. Furthermore, routes B and C employ fewer compound isolation steps in the preparation of 1.

Table 3.

Comparison of routes A–C for converting 6-Ts to dihydrodipyrrin 1

Route	# of steps	Isolated intermediates	# of columns	Overall yield, %a	Isolated product, g
A	3	8a-Ts, 8a	1	14	1.95
B	2	8a-Ts	0	10b	1.60
C	1	—	0	14	1.18

^aBased on the amount of starting material 6-Ts.

^bBased on 16% yield obtained on the 30 mmol scale.

Chlorin precursor. Improved synthesis of the bromo-tetrahydrodipyrrin 2

The synthesis of the tetrahydrodipyrrin lacking the bromo substituent has been refined [29] whereas that bearing the bromo substituent has not [26]. The bromo-tetrahydrodipyrrin 2 was synthesized by employing the refined procedures developed for the bacteriochlorin precursor (vide supra) where possible. The Michael addition of 6-Ts with mesityl oxide (7b) was accomplished under neat conditions in the presence of DBU (Scheme 4). The product 8b-Ts was obtained in 74% yield after crystallization from 2-propanol. Note that Michael addition with tosyl-protected nitroethylpyrrole 6-Ts was accomplished in only 1 h versus 18 h using CsF in acetonitrile [26]. Treatment of 8b-Ts to conditions for reductive cyclization (Zn/HCOONH₄ [29]) afforded the tosyl-protected bromo-tetrahydrodipyrrin 2-Ts in 58% yield.

Scheme 4.

Streamlined synthesis of bromo-tetrahydrodipyrrin 2 for chlorin syntheses

Tetrahydrodipyrrins face a potential side reaction that is not present for dihydrodipyrrins, namely intramolecular cyclization between the pyrrole 3-position and the imine carbon to give a pyrrole-annulated bicyclic product [28]. Formation of the bicyclic product is promoted with acid, and can occur during the course of reactions of the precursor pyrrole-nitrohexanone aimed at forming the tetrahydrodipyrrin. The reductive cyclization conditions of Zn/HCOONH₄ were expressly developed to avoid the deleterious cyclization and smoothly afford the target tetrahydrodipyrrin [29]. Prior application of the Zn/HCOONH₄ conditions generally employed unprotected pyrrole-nitrohexanones, although tosyl-protected 8b-Ts has been examined once previously at small scale. Here, the presence of the tosyl group was carried through from the pyrrole-nitrohexanone 8b-Ts to the tetrahydrodipyrrin 2-Ts.

The motivation for carrying the tosyl group through to the tetrahydrodipyrrin stage were two-fold: (i) the cyclization of the protected precursor 8b-Ts gave a higher yield than that of the unprotected analog 8b [26]; (ii) the tosyl-protected tetrahydrodipyrrin 2-Ts is a valuable substrate for metal-mediated coupling reactions, given that many metals would be chelated by the free nitrogens otherwise.

The tosyl group has previously been cleaved in substituted analogs of tetrahydrodipyrrin 2-Ts using LiOH and α-mercaptoacetic acid in DMF [26]. The crude product 2-Ts was directly used in the subsequent tosyldeprotection using three equivalents of TBAF to afford bromo-tetrahydrodipyrrin 2 in 72% yield after column chromatography. Note that three equivalents of TBAF was necessary to afford complete deprotection versus only 1.2 equivalents for the deprotection in the dihydrodipyrrin synthesis (8a-Ts → 8a). The differences in deprotection conditions may be attributed to the fact that the deprotection occurs at different steps during the synthetic sequence: in the dihydrodipyrrin synthesis (for bacteriochlorins) the tosyl group is cleaved from the pyrrole-nitrohexanone (8a-Ts) whereas in the tetrahydrodipyrrin synthesis (for chlorins) the tosyl group is cleaved from the tetrahydrodipyrrin (2-Ts) itself. Furthermore, the cleaved product (2) did not crystallize from the crude reaction mixture (using 2-propanol or diethyl ether) and hence was purified using column chromatography. Regardless, the improvement in the Michael addition shortened the reaction time and provided the bromo-tetrahydrodipyrrin 2 in 31% overall yield.

EXPERIMENTAL SECTION

General methods

¹H NMR (300 MHz) spectra were collected at room temperature in CDCl₃ containing ~0.1% Me₄Si as reference. Melting points are uncorrected. Rotary evaporation was done at room temperature and ~100 mmHg. Drying under high vacuum was done at room temperature and 0.030 mmHg unless noted otherwise. Silica gel (40 μm average particle size) was used for column chromatography. Molecular sieves (4 Å, beads) were used as received. THF was freshly distilled from sodium/benzophenone ketyl. Anhydrous MeOH was reagent grade and used as received. 1,1-dimethoxy-4-methyl-3-penten-2-one (7a) was prepared following a literature procedure [12]. High resolution exact mass measurements were carried out using electrospray ionization (ESI) on an Agilent Technologies 6210 LC-TOF mass spectrometer. Samples were analyzed in positive-ion mode via a 1 μL flow injection at 300 μL/min in a water:methanol mixture (25:75 v/v) containing 0.1% formic acid.

Route to common synthon 6-Ts

4-bromopyrrole-2-carboxaldehyde (4).

A stirred solution of pyrrole-2-carboxaldehyde (3, 38.0 g, 400 mmol) in THF (400 mL, ACS reagent grade) was cooled to 0 °C (in an ice-water bath) under argon in a 1 L round bottom flask. NBS (71.2 g, 400 mmol; reagent grade, unrecrystallized) was added all at once. The reaction mixture was stirred for 15 min at 0 °C under argon before the solvent was removed on a rotary evaporator. (Note that after ~1 min following the NBS addition, the NBS completely dissolved and the solution turned clear. After ~5 min following the NBS addition, the clear yellow solution gave succinimide as a white precipitate.) The resulting solid (product and succinimide) was dried under high vacuum for 2 h. Water (200 mL, room temperature) was added to the flask and the suspension was filtered with a Büchner funnel. The filter cake was washed with an additional 200 mL of water (room temperature). The solid filtered material was transferred to a 1 L Erlenmeyer flask and was dissolved in 100 mL of hot ethanol (78 °C) by refluxing in a hot water bath. 900 mL of hot water (100 °C) was added all at once. Upon allowing to cool to room temperature the product crystallized from the solution. The mixture was further cooled to 4 °C for 2 h to promote more crystallization. The mixture was filtered by vacuum filtration and the filtered off-white crystals were dried under high vacuum in a vacuum desiccator for 24 h to give 56.1 g of product (81%): mp 120 °C. ¹H NMR: δ, ppm 6.91–7.03 (m, 1H), 7.05–7.17 (m, 1H), 9.48 (s, 1H), 9.68 (br s, 1H). Anal. calcd. for C₅H₄BrNO: C, 34.51; H, 2.32; N, 8.05. Found: C, 34.37; H, 2.19; N, 7.93. Yields did not significantly improve upon crystallization for durations greater than 2 h.

4-bromo-2-formyl-N-tosylpyrrole (4-Ts).

A stirred suspension of 95% NaH (10.7 g, 423 mmol) in dry THF (350 mL, distilled) in a 1 L oven-dried round bottom flask was cooled to 0 °C (in a wet ice water bath) under argon. The mixture was treated portionwise over ~15 min with 4 (61.4 g, 353 mmol). The mixture was stirred for 30 min at 0 °C before treating all at once with p-toluenesulfonyl chloride (67.3 g, 353 mmol). The reaction mixture was stirred at room temperature for 3 h, whereupon water (200 mL, room temperature) was slowly added to quench the reaction. Ethyl acetate (200 mL) was added, and the organic layer was separated. The organic layer was washed with brine (100 mL), dried (~50 g of Na₂SO₄), and concentrated to a solid on a rotary evaporator. The solid was dried under high vacuum for 2 h in a 1 L round bottom flask. The crude solid material was dissolved in 600 mL of hexanes/ethyl acetate (5:1) by refluxing in a hot water bath. Upon allowing to cool to room temperature the product crystallized from the solution. The mixture was further cooled overnight at −10 °C to promote additional crystallization. The mixture was filtered by vacuum filtration, and the filtered light brown crystals were dried under high vacuum in a vacuum desiccator to give 84.2 g of product (73%). The mother liquor could be concentrated and recrystallized to obtain an additional 10.4 g of product for a total of 94.4 g (82%): mp 83–85 °C. ¹H NMR: δ, ppm 2.44 (s, 3H), 7.10 (d, J = 1.93 Hz, 1H), 7.35 (d, J = 7.98 Hz, 2H), 7.58 (d, J = 1.93 Hz, 1H), 7.82 (d, J = 8.53 Hz, 2H), 9.95 (s, 1H). Anal. calcd. for C₁₂H₁₀BrNO₃S: C, 43.92; H, 3.07; N, 4.27; S, 9.77. Found: C, 43.69; H, 2.97; N, 4.27; S, 9.77. Alternatively, the product could be crystallized in similar yields by concentrating the organic layer to ~100 mL, adding 500 mL of hexanes, heating the resulting mixture under reflux to dissolve any solid material, and cooling to room temperature.

4-bromo-2-(2-nitrovinyl)-N-tosylpyrrole (5-Ts).

A stirred mixture of 4-Ts (84.2 g, 257 mmol) in the form of a finely ground powder, potassium acetate (20.1 g, 205 mmol), methylamine hydrochloride (13.8 g, 205 mmol), and acetic acid (1 mL) in absolute ethanol (90 mL) in a 500 mL round bottom flask was treated with nitromethane (34.5 mL, 643 mmol). The progress of the reaction was monitored by TLC (silica, hexanes/CH₂Cl₂ (3:2)) and by ¹H NMR spectroscopy in CDCl₃ (CHO: s, δ 9.95 ppm; C=NHMe: s, 8.59 ppm; CH=CHNO₂: d, J = 13.4 Hz, 8.44 ppm). The mixture was stirred for 2 h, whereupon water was added (200 mL, room temperature) and the resulting yellow precipitate was filtered by vacuum filtration. The solid filtered material was washed with water (500 mL, room temperature) followed by cold ethanol (~1 L, 0 °C) until the eluant was clear. The yellow filtered solid was dried overnight under high vacuum (86.7 g, 91%): mp 164–170 °C. ¹H NMR: δ, ppm 2.44 (s, 3H), 6.77 (s, 1H), 7.31 (d, J = 13.40 Hz, 1H), 7.36 (d, J = 8.25 Hz, 2H), 7.60 (s, 1H), 7.76 (d, J = 8.25 Hz, 2H), 8.44 (d, J = 13.40 Hz, 1H). Anal. calcd. for C₁₃H₁₁BrN₂O₄S: C, 42.06; H, 2.99; N, 7.55. Found: C, 42.23; H, 2.90; N, 7.50.

4-bromo-2-(2-nitroethyl)-N-tosylpyrrole (6-Ts).

A solution of 5-Ts (86.7 g, 234 mmol) in THF (1170 mL, HPLC grade) was cooled to −10 °C (internal temperature, using an acetone bath with a few pieces of dry ice) under argon in a 2 L round bottom flask. The solution was treated with 95% LiBH₄ (5.30 g, 234 mmol) all at once under vigorous stirring. The reaction mixture was stirred for ~15 min at −10 °C, until all of the starting material disappeared (starting material: CH=CHNO₂: d, J = 13.4 Hz, δ 8.44 ppm; product: CH₂CH₂NO₂: t, J = 7 Hz, 4.60 and 3.39 ppm), whereupon the reaction mixture was quenched by slowly adding a cold saturated aqueous NH₄Cl solution (400 mL, 0 °C). The mixture was extracted with ethyl acetate (400 mL). The organic layer was washed with brine (200 mL), dried (~50 g of Na₂SO₄), concentrated to a solid on a rotary evaporator, and dried under high vacuum for 2 h in a 2 L round bottom flask. The crude solid material was dissolved in 2-propanol (1.5 L) by refluxing in a hot water bath. (Note that any undissolved solid (e.g. dimeric byproduct I) can be removed by hot filtration.) Upon allowing to cool to room temperature the product crystallized from the solution. The mixture was further cooled overnight at −10 °C to promote more crystallization. The mixture was filtered by vacuum filtration, and the filtered light brown crystals were dried under high vacuum in a vacuum desiccator to give 67.5 g of product (77%): mp 125–127 °C. ¹H NMR: δ, ppm 2.44 (s, 3H), 3.39 (t, J = 7.01 Hz, 2H), 4.60 (t, J = 7.01 Hz, 2H), 6.10 (d, J = 1.93 Hz, 1H), 7.32 (d, J = 1.93 Hz, 1H), 7.35 (d, J = 7.98 Hz, 2H), 7.69 (d, J = 7.98 Hz, 2H). Anal. calcd. for C₁₃H₁₃BrN₂O₄S: C, 41.84; H, 3.51; N, 7.51. Found: C, 41.85; H, 3.45; N, 7.40. Data for dimeric byproduct I: mp 115–117 °C. ¹H NMR: δ, ppm 2.44 (s, 3 H), 2.45 (s, 3 H), 3.00–3.17 (m, 1 H), 3.21–3.35 (m, 1 H), 4.36–4.51 (m, 1 H), 4.69–4.86 (m, 1 H), 4.87–5.01 (m, 1 H), 5.22–5.38 (m, 1 H), 5.99 (s, 1 H), 6.00 (s, 1 H), 6.16 (s, 1 H), 6.17 (s, 1 H), 7.28–7.37 (m, 2 H), 7.37–7.44 (m, 2 H), 7.61 (d, J = 8.53 Hz, 2 H), 7.77 (d, J = 8.53 Hz, 2 H). ESI-MS: m/z 741.93978, calcd. 741.94023 (C₂₆H₂₄Br₂N₄O₈S₂).

Routes to bacteriochlorin precursor 1

6-(4-bromo-N-tosylpyrrol-2-yl)-1,1-dimethoxy-4, 4-dimethyl-5-nitrohexan-2-one (8a-Ts).

A mixture of 6-Ts (29.9 g, 80.0 mmol) and 1,1-dimethoxy-4-methyl-3-penten-2-one (7a, 38.0 g, 240 mmol, 3 equiv.) in a 250 mL round bottom flask was treated with DBU (35.9 mL, 240 mmol). The reaction mixture was stirred for 15 min under argon. A saturated solution of cold aqueous NH₄Cl (100 mL, 0 °C) was added. The mixture was extracted with ethyl acetate (200 mL). The organic layer was washed with brine (100 mL), dried (~50 g of Na₂SO₄), and concentrated on a rotary evaporator for 30 min. The crude mixture was then subjected to overnight bulb-to-bulb distillation (0.014 mmHg) at room temperature to recover unreacted 7a (9.89 g), which could be reused without further purification. Addition of diethyl ether (300 mL) to the remaining solid crude reaction mixture, trituration of the solid, and filtration afforded a pale brown solid (28.1 g, 66%): mp 122–124 °C. ¹H NMR: δ, ppm 1.14 (s, 3H), 1.23 (s, 3H), 2.44 (s, 3H), 2.59, 2.69 (AB, ²J = 18.5 Hz, 2H), 3.19 (ABX, ²J_AB = 15.7 Hz, ³J_BX = 2.0 Hz, 1H), 3.37 (ABX, ²J_AB = 15.7 Hz, ³J_AX = 11.7 Hz, 1H), 3.43 (s, 6H), 4.36 (s, 1H), 5.19 (ABX, ³J_AX = 11.7 Hz, 1H), 6.03 (s, 1H), 7.27 (s, 1H), 7.35 (d, J = 8.41 Hz, 2H), 7.66 (d, J = 8.41 Hz, 2H). Anal. calcd. for C₂₁H₂₇Br-N₂O₇S: C, 47.46; H, 5.12; N, 5.27. Found: C, 47.65; H, 5.16; N, 5.21.

6-(4-bromo-1H-pyrrol-2-yl)-1,1-dimethoxy-4, 4-dimethyl-5-nitrohexan-2-one (8a).

A sample of 8a-Ts (18.36 g, 34.47 mmol) in a 250 mL round bottom flask was treated with TBAF (41 mL, 1.0 M in THF, 41 mmol), and the reaction mixture was stirred for 45 min at reflux. A saturated solution of aqueous NaHCO₃ (100 mL) was added followed by ethyl acetate (100 mL). The mixture was extracted with ethyl acetate (2 × 100 mL). The organic layer was dried (~50 g of Na₂SO₄), concentrated to a brown oil on a rotary evaporator, dried under high vacuum for 2 h, and chromatographed (silica (100 g in a 4.5 cm diameter column), hexanes/ethyl acetate (1:0 → 3:2), 2 L) to give a light brown oil, which solidified upon storing at −10 °C (7.79 g, 60%): mp 77–81 °C. ¹H NMR: δ, ppm 1.13 (s, 3H), 1.21 (s, 3H), 2.60, 2.72 (AB, ²J = 18.8 Hz, 2H), 2.98 (d, J = 15.4 Hz, 1H), 3.30 (ABX, ²J_AB = 15.4 Hz, ³J_AX = 11.8 Hz, 1H), 3.44 (s, 6H), 4.36 (s, 1H), 5.13 (d, J = 11.8 Hz, 1H), 5.99 (s, 1H), 6.63 (s, 1H), 8.15 (br s, 1H). Anal. calcd. for C₁₄H₂₁BrN₂O₅: C, 44.57; H, 5.61; N, 7.43. Found: C, 45.53; H, 5.56; N, 6.94.

8-bromo-2,3-dihydro-1-(1,1-dimethoxymethyl)-3,3-dimethyldipyrrin (1).

In a first flask, a solution of 8a (7.79 g, 20.6 mmol) in freshly distilled THF (50 mL) and anhydrous MeOH (10 mL) at 0 °C was treated with freshly prepared NaOMe (3.33 g, 61.8 mmol). [The NaOMe was prepared by slowly adding elemental Na (~5 g) to MeOH (~50 mL). The reaction is exothermic and was refluxed using a reflux condenser until all Na has reacted. The methanol was then evaporated on a rotary evaporator and the resulting NaOMe was dried under high vacuum for 24 h.] The resulting mixture was stirred and degassed by bubbling argon through the solution for 45 min. In a second oven-dried flask purged with argon, TiCl₃ (20.0 g, 124 mmol) and freshly distilled THF (250 mL) were combined under argon. [3 equiv. of freshly prepared NaOMe were necessary to effect complete formation of the nitronate anion. Commercially available 95% NaOMe does not completely dissolve in the THF/MeOH (5:1) solvent mixture, resulting in incomplete nitronate anion formation, and increasing the amount of MeOH decreases the yield of the reaction. Incomplete formation of the nitronate anion results in leftover starting material in the crude product mixture, and the dihydrodipyrrin 1 does not crystallize.] Under vigorous stirring, a degassed solution (by bubbling argon through for 1 h) of NH₄OAc (60.0 g, 779 mmol) in water (80 mL, the minimum amount required to dissolve the NH₄OAc) was transferred via cannula to the TiCl₃ solution to buffer the solution to pH 6. Then the first flask mixture was transferred via cannula to the buffered TiCl₃ solution. The resulting mixture was stirred at room temperature for 16 h under argon. A saturated solution of aqueous NaHCO₃ (200 mL) was added and the reaction mixture was extracted with ethyl acetate (200 mL). The organic layer was washed with brine (100 mL), dried (~50 g of Na₂SO₄), concentrated on a rotary evaporator, and dried under high vacuum for 2 h. The dark brown oily residue was dissolved in a small amount of 2-propanol (3 mL) and stored at −10 °C for 48 h to give light brown crystals (1.95 g, 29%). The mother liquor was chromatographed (silica (25 g in a 2.5 cm diameter column), hexanes/ethyl acetate (3:1), 500 mL) to give additional 0.41 g of product for a total of 2.36 g (35%): mp 101–103 °C. ¹H NMR: δ, ppm 1.20 (s, 6H), 2.61 (s, 2H), 3.44 (s, 6H), 5.01 (s, 1H), 5.77 (s, 1H), 6.12 (s, 1H), 6.78 (s, 1H), 10.67 (br s, 1H). Anal. calcd. for C₁₄H₁₉BrN₂O₂: C, 51.39; H, 5.85; N, 8.56. Found: C, 51.37; H, 5.89; N, 8.37.

Route B: streamlined synthesis from 8a-Ts of 1.

A sample of 8a-Ts (16.7 g, 31.4 mmol) was treated with TBAF (37.7 mL, 1.0 M in THF, 37.7 mmol), and the reaction mixture was stirred for 1 h at reflux. The progress of the reaction was monitored by ¹H NMR spectroscopy (8a-Ts: dd, J = 8 Hz, δ = 5.20 ppm; 7: dd, J = 10 Hz, 5.13 ppm). After the starting material disappeared, a saturated solution of aqueous NaHCO₃ (100 mL) was added and the mixture was extracted with 100 mL of ethyl acetate. The organic layer was dried (~25 g of Na₂SO₄), concentrated to a brown oil on a rotary evaporator, and dried under high vacuum for 2 h. The crude material was used directly in the next step. In a first flask, a solution of the crude material in freshly distilled THF (120 mL) and anhydrous MeOH (24 mL) at 0 °C was treated with NaOMe (5.1 g, 94.2 mmol, freshly prepared as described above). The mixture was stirred and degassed by bubbling argon through the solution for 45 min. In a second 1 L round bottom flask purged with argon, TiCl₃ (160 mL, 20% wt/v in 3% HCl solution, 180 mmol), 300 mL THF, and NH₄OAc (124 g, 1620 mmol) were combined under argon and the solution was degassed by bubbling argon through the solution for 45 min. Then the first flask mixture was transferred via cannula to the buffered TiCl₃ solution. The resulting mixture was stirred at room temperature for 16 h under argon. (Note that the reaction mixture forms a thick solid material along the walls of the flask and overhead stirring may be required.) The reaction mixture was then poured over a pad of silica (100 g in a Büchner funnel). The reaction flask was rinsed with ethyl acetate (2 × 100 mL), which was poured over the silica pad. The silica pad was eluted with an additional 100 mL of ethyl acetate. The eluant was checked by TLC (silica, hexanes/ethyl acetate (1:1), top dark brown spot after staining with I₂) to confirm that all of the desired product had eluted. The filtrate was treated with 5 g of NaHCO₃ for 5 min to neutralize the mixture and absorb any remaining water. The NaHCO₃ was removed by filtration, and the filtrate was concentrated to a brown oil on a rotary evaporator. The crude product was filtered through a second silica pad (100 g in a Büchner funnel) and eluted with ethyl acetate (100 mL) with TLC monitoring as described above. The filtrate was concentrated to a brown oil on a rotary evaporator and dried under high vacuum for 1 h. The brown oil was dissolved in a small amount of 2-propanol (5 mL) and stored at −10 °C for 48 h to give light brown crystals (1.60 g, 16%). Anal. calcd. for C₁₄H₁₉BrN₂O₂: C, 51.39; H, 5.85; N, 8.56. Found: C, 51.53; H, 5.89; N, 8.33.

Route C: streamlined synthesis from 6-Ts of 1.

A mixture of 6-Ts (11.2 g, 30.0 mmol) and 7a (5.69 g, 36.0 mmol) in a 250 mL round bottom flask was treated with TBAF (90 mL, 1.0 M in THF, 90 mmol) and 4 Å molecular sieves (6 g, beads), and the reaction mixture was stirred for 24 h under argon at room temperature. A saturated solution of aqueous NaHCO₃ (100 mL) was added, and the mixture was extracted with ethyl acetate (2 × 100 mL). The organic layer was dried (~50 g of Na₂SO₄), concentrated to a brown oil on a rotary evaporator, and dried under high vacuum for 2 h. The crude material was used directly in the next step. In a first flask, a solution of the crude material in freshly distilled THF (120 mL) and anhydrous MeOH (24 mL) at 0 °C was treated with NaOMe (5.1 g, 94.2 mmol, freshly prepared as described above). The mixture was stirred and degassed by bubbling argon through the solution for 45 min. In a second 1 L round bottom flask purged with argon, TiCl₃ (160 mL, 20% in 3% HCl solution, 180 mmol), 300 mL THF, and NH₄OAc (124 g, 1620 mmol) were combined under argon and the mixture was degassed by bubbling argon through the solution for 45 min. Then the first flask mixture was transferred via cannula to the buffered TiCl₃ solution. The resulting mixture was stirred at room temperature for 16 h under argon. (Note that the reaction mixture forms a thick solid material along the walls of the flask and overhead stirring may be required.) The reaction mixture was then poured over a pad of silica (100 g in a Büchner funnel). The reaction flask was rinsed with ethyl acetate (2 × 100 mL), which was poured over the silica pad. The silica pad was eluted with an additional 100 mL of ethyl acetate. The eluant was checked by TLC (silica, hexanes/ethyl acetate (1:1), top dark brown spot after staining with I₂) to confirm that all of the desired product had eluted. The filtrate was treated with 5 g of NaHCO₃ for 5 min to neutralize the mixture and absorb any remaining water. The NaHCO₃ was removed by filtration, and the filtrate was concentrated to a brown oil on a rotary evaporator. The crude product was filtered through a second silica pad (100 g in a Büchner funnel) and eluted with ethyl acetate (100 mL) with TLC monitoring as described above. The filtrate was concentrated to a brown oil on a rotary evaporator and dried under high vacuum for 1 h. The brown oil was dissolved in a small amount of 2-propanol (5 mL) and stored at −10 °C for 48 h to give light brown crystals (1.18 g, 14%). Anal. calcd. for C₁₄H₁₉BrN₂O₂: C, 51.39; H, 5.85; N, 8.56. Found: C, 51.65; H, 5.93; N, 8.49.

Route to chlorin precursor 2

6-(4-bromo-1-N-tosylpyrrol-2-yl)-3,3-dimethyl-4-nitrohexan-2-one (8b-Ts).

A suspension of 6-Ts (13.3 g, 35.7 mmol) in mesityl oxide (7b, 10.5 g, 107 mmol, 3 equiv.) was treated with DBU (16 mL, 11.6 mmol). The resulting mixture was stirred at room temperature for 1 h, diluted with ethyl acetate, washed (3 × water, brine), dried (Na₂SO₄) and concentrated. [The thorough washing with water and use of ethyl acetate (rather than CH₂Cl₂) are essential to remove all of the DBU. The remaining DBU may cause extensive formation of byproduct during concentration.] An excess of mesityl oxide was removed under high vacuum for 16 h. The resulting crude product was dissolved in a minimal amount of 2-propanol (150 mL) and stored at −10 °C for 24 h to give brown crystals (12.54 g, 74%). The ¹H NMR data were consistent with those previously reported [26]. Anal. calcd. for C₁₉H₂₃BrN₂O₅S: C, 48.41; H, 4.92; N, 5.94. Found: C, 48.41; H, 4.92; N, 5.93.

8-bromo-3,4,5,6-tetrahydro-1,3,3-trimethyl-11-N-tosyldipyrrin (2-Ts).

A solution of 8b-Ts (10.0 g, 21.2 mmol) in THF (100 mL) was treated with HCOONH₄ (26.8 g, 424 mmol) and zinc powder (27.7 g, 424 mmol). The resulting suspension was stirred vigorously at room temperature for 2 h. The reaction mixture was filtered through a pad of silica (50 g). The filter cake was eluted with 500 mL of ethyl acetate. The filtrate was concentrated to a fluffy, light-brown solid (5.2 g, 58%). The crude product was about 95% pure as determined by ¹H NMR spectroscopy [26], and was used directly in the next reaction.

8-bromo-3,4,5,6-tetrahydro-1,3,3-trimethyldipyrrin (2).

A sample of crude 2-Ts (2.07 g, 4.89 mmol) was treated with TBAF (14 mL, 1.0 M in THF, 14 mmol, 3 equiv.), and the reaction mixture was stirred for 1 h at reflux. A saturated solution of aqueous NaHCO₃ (50 mL) was added followed by ethyl acetate (50 mL). The mixture was extracted with ethyl acetate (100 mL). The organic layer was dried (Na₂SO₄), concentrated to a brown oil on a rotary evaporator, dried under high vacuum for 2 h, and chromatographed (silica (100 g in a 3.5 cm diameter column), ethyl acetate 500 mL) to give a brown oil, which solidified upon storage at −10 °C (0.95 g, 72%). The characterization data (¹H NMR, ESI-MS) were consistent with those previously reported [26].

CONCLUSION

The prior syntheses of bromo-hydrodipyrrins required extensive column chromatography, large amounts of chlorinated solvents, and sizable excesses of solvents and reagents, which together resulted in a low overall yield and was expensive in terms of investment of time and materials. The refinements described herein include altered reaction conditions (concentration, temperature, reagents and purification techniques) for each of the individual steps. Application of the stepwise synthesis (route A) affords the valuable 8-bromo-dihydrodipyrrin-acetal 1 in significantly improved overall yield (5.7%) with limited chromatography, use of process-friendlier solvents (water, ethanol, or 2-propanol), avoidance of a problematic byproduct (I), higher concentrations, faster reaction times, and lesser excesses of solvents and reagents. Streamlined routes enable the expeditious preparation of 1. The findings upon refinement of the synthesis of 1 (for bacteriochlorins) were applied to the preparation of 2 (for chlorins). The key synthon 4-bromo-2-(2-nitroethyl)-N-tosylpyrrole (6-Ts), which is the latest intermediate shared by the chlorin and bacterichlorin syntheses, was prepared in 67.5 g quantity. Accordingly, the improvements described herein should benefit the synthesis of novel chlorins and bacteriochlorins for use in fundamental spectroscopic studies as well as a variety of applications.

Supplementary Material

X-ray structural data for 1 and 8a-Ts are deposited at the Cambridge Crystallographic Data Centre under deposition numbers CCDC 730192 and CCDC 730193, respectively. Tables of data for development of improved reaction conditions, and spectral data for synthetic compounds are available free of charge via the Internet at http://www.worldscinet.com/jpp/jpp.shtml.