Tetraphenylazadipyrromethene

This interesting chromophore scaffold forms the basis of something interesting red to near infrared emitters in the class of BODIPY derivatives. Even though in this preparation no boron is introduced, the substance itself is quite pretty and rewarding.



Step 1: Chalcone
The synthesis begins with chalcone, also known as benzalacetophenone. For this I followed the procedure given by Orgsyn pretty much exactly.

Synthesis:
A three necked round bottom flask in an ice bath equipped with an internal thermometer, a mechanical overhead stirrer (Note 1) and a stopper, placed in an icebath, is charged with 6.78 g (> 90 %, 109 mmol, 1.3 eq.) KOH and 20 g (95 %, 25 mL) undenatured ethanol and the solution left to precool until about 12 °C. Then 10.40 g (86 mmol, 1 eq.) acetophenone (Note 2) was added and stirred in, and finally 9.30 g (98 mmol, 1.1 eq.) of benzaldehyde (Note 3) were added all at once with strong stirring. As no strong exotherm was noted the internal temperature was regulated to 23 °C (target 25 °C) by replacing the ice bath with a water bath. The mixture was stirred for two hours, remaining liquid. By briefly lowering it into an ice bath it could be solidified, the resulting suspension was stirred for another hour and placed in the fridge for 18 hours. It was vacuum filtered over a fritted funnel and washed with about 100 mL of cold 20 % ethanol, then with 300 mL of cold water, then with very dilute (< 1 %) acetic acid, then with more water. The washing water had a pH of 6. After drying this crude product for 18 hours at 40-45 °C, the mass decreased from 24 g to about 17.4 g.

For recrystallization, the substance was dissolved in about 45 mL of 50 °C undenatured 95 % ethanol (Note 4). After an hour at room temperature, only little crystals had formed, but cooling to -18 °C forced good crystallization. After 10 hours it was vacuum filtered off, washed with 10 mL of cold ethanol and dried at 40-45 °C.

Yield: 15.26 g (85 %, 73.3 mmol, literature: 85 %)


Fig. 1: Purified chalcone.

Notes:
1. The mechanical stirring might not have been needed, although if the mixture crystallized sooner, maybe it would have been.
2. Orgsyn notes the acetophenone should be as pure as possible, as impurities reduce the yield more than purification of the acetophenone does. I used commercially available acetophenone that was in storage for several years without further purification.
3. The benzaldehyde was commercial and several years old, but used without further purification.
4. This is the Orgsyn recommendation (95 % ethanol saturated at 50 °C to prevent oiling out). Friends I've talked to also recommend petrol ether as a recrystallization solvent.

Analysis:
The product was used directly for the next reaction, but a TLC of the chalcone can be seen as a comparison lane in the product analysis of the next step (single spot visible).

Literature:
[1] - E. P. Kohler, H. M. Chadwell, "Benzalacetophenone", Org. Synth. 1922, 2, 1, https://doi.org/10.15227/orgsyn.002.0001

Step 2: 4-Nitro-1,3-diphenyl-1-butanone
4-Nitro-1,3-diphenyl-1-butanone, also known as γ-nitro-β-phenylbutyrophenone, is produced by Michael-addition of (deprotonated) nitromethane to chalcone. There are various choices of base, and procedures with for example potassium carbonate,^[1] potassium hydroxide,^[2] sodium hydroxide,^[3] diethyl amine^[4] and sodium methoxide^[5] are known. The procedure with sodium methoxide appealed to me the most because of the extensive description and high reported yields.

Synthesis:
1.56 g of metallic sodium are cut and cleaned from mineral oil in petrol ether. In a 50 mL flask with reflux condenser was placed 10 g of methanol predried over 3 A molecular sieves, and the flask was cooled in an ice bath. The sodium was added slowly, initially the reaction was quite vigorous, but it quickly slowed down. The second half of sodium reacted very slowly and required taking away the ice bath, heating with a heatgun and gentle heating from the hotplate. The total dissolution took about 3 hours. The resulting solution was drawn into a 10 mL syringe with Luer-lock and a wide-gauge needle.

Into a separate 100 mL flask were placed 10.4 g chalcone, 23.7 g of a 16 % nitromethane in methanol solution (equivalent to 3.8 g of nitromethane and 20.0 g of methanol) and a stir bar. Everything was dissolved by heating the solution to 40 °C.

Into this 40 °C solution was injected the methoxide solution over the course of about 30 seconds. Initially some turbidity or precipitate formed, but the exothermic reaction heated the solution almost to the boiling point, turning the solution clear and orange. The solution was left for eight minutes, then placed into a salted ice bath, and left to cool to 11 °C (which took three minutes). Then with strong stirring 4.5 g of glacial acetic acid were added. The temperature raises to 26 °C and a large amount of yellow solid forms. After stirring with a glass rod in ice for 50 minutes the mixture was filtered over a glass Büchner funnel and washed with 35 mL of 30 % ethanol, then with 25 mL of 20 % ethanol, then with water and dried on the pump. The solid becomes quite powdery but retains much solvent, as it weighs about 20.85 g at this point.


Fig. 2-4: Progression of the reaction.

The solid was recrystallized from 60 mL of 99 % ethanol at 70 °C, it was filtered hot. On cooling, agitating the filtrate caused it to set into a pasty mass. After cooling over night in the fridge it was vacuum filtered, pressed out on the filter, washed with 20 mL of 50 % ethanol and dried on the pump. Then it was dried for 24 hours at 50 °C in the drying oven.

Yield: 10.50 g (78 %) of a white powder.


Fig. 5: Nitroketone product.

Analysis:
The melting point was determined as 91-92 °C. The literature values fluctuate but are given as 68.4-69.4 °C,^[3] 99-99.5 °C,^[4] and 103 °C.^[5]
TLC on silica 60 F₂₅₄ with 7/3 n-hexane/EtOAc as the eluent and visualisation under 250 nm and 270 nm UV gave the chalcone as a single spot at R_f 0.55, the unrecrystallized product as a large spot with R_f 0.42 with some trail below and some residue on the starting line and the final product as a single spot with R_f 0.42. The literature R_f for the product is 0.55.^[3] Using vanillin-sulfuric acid as a stain, the chalcone already on light heating shows up as a yellow spot, while on stronger heating (heatgun set to 270 °C) the nitroketone shows up as a violet-purple spot.


Fig. 6: TLC plate under UV. Chalcone (left), crude product (middle), final product (right), inbetween co-spots of adjacent lanes. Slight upwards curve due to incomplete solvent evaporation from spotting.

Literature:
[1] - R. Gresser, H. Hartmann, M. Wrackmeyer, K. Leo, M. Riede, "Synthesis of thiophene-substituted aza-BODIPYs and their optical and electrochemical properties", Tetrahedron 2011, 67, 37, 7148-7155. https://doi.org/10.1016/j.tet.2011.06.100
[2] - C. You, X. Li, D. Wang, H. Chen, L. Liang, Y. Chen, Y. Zhao, H. Xiang, "Self-Assembled Aza-Boron-Dipyrromethene for Ferroptosis-Boosted Sonodynamic Therapy", Angew. Chem. Int. Ed. 2022, 61, e202210174, Angew. Chem. 2022, 134, e202210174. https://doi.org/10.1002/anie.202210174
[3] - T. Hlogyik, R. Laczkó-Rigó, É. Bakos, M. Poór, Z. Kele, C. Özvegy-Laczka, E. Mernyák, "Synthesis and in vitro photodynamic activity of aza-BODIPY-based photosensitizers", Org. Biomol. Chem. 2023, 21, 6018-6027. https://doi.org/10.1039/D3OB00699A
[4] - M. C. Kloetzel, "Reactions of Nitroparaffins. I. Synthesis and Reduction of Some γ-Nitroketones", J. Am. Chem. Soc. 1947, 69, 10, 2271-2275. https://doi.org/10.1021/ja01202a010
[5] - (a) E. P. Kohler, "The addition of aliphatic nitro compounds to unsaturated compounds", J. Am. Chem. Soc. 1916, 38, 4, 889-900. https://doi.org/10.1021/ja02261a018, (b) E. P. Kohler, "A new type of cyclic compounds", J. Am. Chem. Soc. 1924, 46, 2, 503-517. https://doi.org/10.1021/ja01667a029

Step 3: Tetraphenylazadipyrromethene
The obtained nitroketone is finally cyclized into the desired product. This step is sadly quite inefficient and low yielding. The reaction mechanism was thoroughly investigated using isotope labeling in 2012,^[1] but it is complex. From this paper I will also use the reaction conditions with ammonium acetate in ethanol, but earlier reports have used ammonium formate melts instead.^[2] Using higher boiling alcohols like butanol is also possible from what I've read.

Synthesis:
1.03 g (3.82 mmol) of 4-nitro-1,3-diphenyl-1-butanone, 10.3 g (133 mmol, 35 eq.) of commercial anhydrous ammonium acetate and 40 mL of anhydrous undenatured ethanol are placed together with a stir bar inside a 100 mL round bottom flask in an oil bath equipped with a reflux condenser and a CaCl₂ drying tube. The oil bath was heated until the ethanol started to gently reflux, and the color changed from colorless to yellowish, to pink, and then to blue. The solution was refluxed for 24 hours, then left to cool for two hours and filtered off using vacuum. The solid was washed with 20 mL of ethanol and obtained as a black paste with a metallic shine. After drying overnight they weighed 0.400 g (46 %).



Fig. 7-8: Top: Progression of the reaction during the first 30 minutes and after 10 hours. Bottom: Paste after filtration.

The solid was then recrystallized from 60 mL of boiling xylene (it is very hard to tell if everything is dissolved because the solution is such a deep dark blue), it was filtered hot, cooled to room temperature and then placed over night into the fridge. It was filtered, dried on the pump and dried at 150 °C for three hours.

Yield: 0.250 g (29 %, literature: 35 %)


Fig. 9-10: Solid product and solutions in xylene under different lighting conditions.

Discussion:
Instead of recrystallization, sublimation might also be a viable way of purification.^[2] Unfortunately, I haven't gotten around to confirm purity yet, so the yield should be viewed with some reservation. I think the color and appearance, as well as the intermediate products, are indicative of the identity of the final product, however.

Literature:
[1] - M. Grossi, A. Palma, S. O. McDonnell, M. J. Hall, D. K. Rai, J. Muldoon, D. F. O'Shea, "Mechanistic Insight into the Formation of Tetraarylazadipyrromethenes", J. Org. Chem. 2012, 77, 20, 9304-9312. https://doi.org/10.1021/jo301972w
[2] - M. A. T. Rogers, "156. 2 : 4-Diarylpyrroles. Part I. Synthesis of 2 : 4-diarylpyrroles and 2 : 2' : 4 : 4'-tetra-arylazadipyrromethines", J. Chem. Soc. 1943, 590-596. https://doi.org/10.1039/JR9430000590

Thanks Boffis! Yeah, it is certainly possible. I've done a test tube experiment with the ligand and zinc acetate in n-butanol,^[1] and there was a color change to a much deeper blue and away from the purple it had before. It reminded me a lot of a methylene blue solution. The solution wasn't fluorescent under 365 nm UV (the metal complexes usually aren't it seems) and I didn't try to isolate the result, but the color change does indicate complex formation.



The most interesting part is probably to combine it with BF₃ to produce the corresponding BODIPY-derivative because of their fluorescence. But it will involve making the boron trifluoride, which while probably doable^[2] I am not particularly looking forward to.

[1] - M. Kryjewski, B. Wicher, N. Bojanowski, E. Tykarska, J. Mielcarek, "Zinc(II) azadipyrromethene complexes substituted at the distal phenyl rings – Structure and spectroscopical properties", Polyhedron 2020, 192, 114820. https://doi.org/10.1016/j.poly.2020.114820 
[2] - W. Hellriegel, "Über Fluorotetraborat, Oxotetraborat und über eine neue Darstellung von Borfluorid.", Ber. dtsch. Chem. Ges. A/B 1937, 70, 689-690. https://doi.org/10.1002/cber.19370700418

Quote: Originally posted by Keras

Speaking of acetophenone, I was planning to synthesise some starting from a Grignard between acetaldehyde and bromobenzene followed by oxidation using whatever's handy. Does that sound sensible?

Quote: Originally posted by Keras

I was planning to synthesise some starting from a Grignard between acetaldehyde and bromobenzene followed by oxidation using whatever's handy.

Quote: Originally posted by Diachrynic

That is possible, although acetaldehyde has a tendency to oligomerize. Acetonitrile might also be a useable synthon, as it is more stable and also saves the oxidation step?

My 5 cents about the nitroketone preparation.

Unfortunately, it is not easy to prepre it in pure form, with m.p. 99-100 C.
This is confirmed value, from a number of references. All other lower m.p.'s indicate impure material. Mentioned Kohler reported higher value, namely 103 C but most probably it is some device or reading error, because he also reported m.p. of semicarbazone and he reported value by ~3 C higher than reported value elsewhere (see reference [4] given above).
I tried to prepare it using K2CO3 as a base.
First was tried "green" method: chalkone+MeNO2+K2CO3+TBACl in a mortar, without any cosolvent (see "Letters in Organic Chemistry, 2007, 4, 445-447"). Orginally they used TBABr, but I think it makes no larger difference. It was tried twice, with similar results: white powder, from which I was not able to obtain any pure sample of the nitroketone. The only crystalline material obtained (10-20 % mass of obtained powder), in form of small transparent needles, had m.p. larger than 115 C (heating was then stopped,it is not desired product). The crude product was low melting, below 60 C. All other samples obtained from EtOH "crystallization" were mixtures, with m.p. at best 70 C (not sharply and not completely).

The procedure from reference [1] was tested: chalkone+MeNO2 in MeOH + cat. amount of K2CO3. Reaction goes smoothly at r.t., but about 30 hours is needed.
Unfortunately, formed product is very "voluminous" and at the end of reaction, post-reaction mixture looks like white pudding.
Obtained white powder (after work-up) was also a mixture, with not sharp m.p. around 80 C. It had very weak odour (chalkone-like) and steam distillation was applied to remove all volatile compounds. Cooled post distillation water mixture (no odour) contained hard, rather large pieces of very pale yellow material. A sample was taken for measurement and m.p. was 97-100 C, so it is rather pure nitroketone.

As it is noticed in many references, using low molar MeNO2/chalkone ratio gives large amounts of "trimers" ( two nitrodiketones, low- ang high - melting stereoisomers). Crystallization of such mixtures is very problematic, because these nitrodiketones are less soluble in MeOH or EtOH, than nitroketone. However it is possible to do. Recommended (in literature) MeNO2/chalkone ratio is in range 5-10, to avoid formation of larger amounts of "trimers" and possibly, other compounds.

I can confirm, that all pepared white powders strongly "holds" water: suspension of ~5 g in water, after vac. suc. gives no less than 20 g of residue. Interesting, there must be some "cage" structure in it, because molar ratio water/nitroketone is very high. Also MeOH/EtOH are adsorbed strongly.
Besides, all prepared powders give solutions in alcohols with oversaturation: clear solution at r.t. deposites some sediment on cooling, which not redissolves at r.t.
Really, not very nice thing to prepare and purify. Fortunately, very cheap

[Edited on 6-11-2024 by kmno4]