Ammonia-free and one-pot synthesis of di-chloro silicon phthalocyanine and naphthalocyanine

Abstract

Silicon phthalocyanines (SiPcs) and silicon naphthalocyanines (SiNcs) are promising and versatile organic materials with applications spanning across organic electronics, photocatalysis, and photodynamic therapy. Specifically, the availability for modification of periphery sites and axial positions enables tailoring for solubility, bandgap tuning, or even biological targeting. The dichlorinated species (Cl₂-SiPc and Cl₂-SiNc) are particularly attractive due to the facile substitution of the axial chloride groups. However, conventional syntheses rely on a two-step protocol that requires the use of an ammonia gas tank at high temperatures. Here, we report a safer, one-pot synthesis of Cl₂-SiPc and Cl₂-SiNc directly from unsubstituted phthalonitrile and 2,3-dicyanonaphthalene, respectively. By adapting the alternative route proposed by Lessard et al., which uses lithium bis(trimethylsilyl) amide (LiHMDS) in place of ammonia gas, we achieved up to a 45% macrocyclic conversion to-date. Overall, the adapted ammonia-free approach provides a synthetically relevant and sustainable platform for the preparation and functionalization of silicon phthalocyanines and naphthalocyanines. We also studied the microwave process and confirmed also that the silicon phthalocyanines can be formed via microwave irradiation.

---

Introduction

Porphyrin-derived macrocycles, including phthalocyanines (Pcs) and the more conjugated family of naphthalocyanines (Ncs), have been incorporated into a diverse set of applications due to their tunable electronic and photophysical properties. These properties can be finely adjusted through peripheral and axial substitutions, as well as by varying the ring size and metal center of trivalent and tetravalent metal-based Pcs (MPcs) and Ncs (MNcs). To expand, silicon phthalocyanines (R₂-SiPc) and silicon naphthalocyanines (R₂-SiNc) represent a particularly well-studied class, demonstrating significance in organic photovoltaics (OPVs), organic thin film transistors (OTFTs), photodynamic therapy (PDT), and even photocatalysis.^1–4 The +4 oxidation state of silicon provides two axial substituents (R₂), which are most commonly chloride during the first step of forming a R₂-SiPc/R₂-SiNc. Resulting in a hexacoordinated complex, the geometry provides stability towards the Si–N bonds, and opens doors for derivatizations through bulky, polar, or even biological handle substitutions.^5,6 Moreover, axial chloride groups are synthetically attractive as these groups can be readily displaced by a variety of nucleophiles (e.g. phenols, amines, alcohols), leading to diverse libraries of symmetrical or non-symmetrical derivatives.

Conventional synthetic routes to Cl₂-SiPc and Cl₂-SiNc involve multistep protocols that rely on ammonia gas. For SiPc, phthalonitrile is first reacted with ammonia gas and sodium methoxide in methanol. After collection of the 1,3-diiminoisoindoline (DII/DI3) intermediate of this process, if formed, cyclotetramerization with silicon tetrachloride (SiCl₄) in dry quinoline occurs in a separate pot.⁵ And it is known that SiCl₄ does not form Cl₂-SiPc with phthalonitrile; however, it is known that Cl₂-SiPc can be formed via DII/DI3 and this may be due to DII/DI3 having higher Lewis basicity. Similarly, Cl₂-SiNc is typically prepared by conversion of 2,3-dicyanonaphthalene to the 1,3-diiminobenz[f]isoindoline intermediate under ammonia gas, with subsequent reaction with silicon tetrachloride in tri-n-butylamine and tetraline solvent.⁷ While the described protocols can afford conversions of up to 70% (SiPc) and 51% (SiNc),^7–9 they suffer from two key limitations: (i) a lack of commercial availability of DII/DI3 derivatives due to the complication of peripheral substitution strategies; and (ii) the use of constant bubbling ammonia gas requiring hazardous conditions.

To expand, ammonia gas presents substantial safety and practical challenges. Due to its exothermic and corrosive properties, ammonia gas may cause severe burns to the eyes and skin, and may even cause irreversible damage to moistened membranes within the respiratory tract.¹⁰ In addition, it is also capable of forming explosive mixtures and may pose flammability risks at higher concentrations in the presence of an ignition source.¹⁰ Beyond its hazards and risks, ammonia gas overall requires a specialized unit, which complicates the setup of the reaction and may limit scalability.

In this way, due to the safety and practical concerns surrounding ammonia gas as a reagent, we have developed and optimized a one-pot, ammonia-free protocol for the synthesis of Cl₂-SiPc and Cl₂-SiNc, adapted from the method of Lessard et al.¹¹ Notably, while the conventional ammonia-based protocols report modest yields, our work achieves up to a 45% conversion in macrocyclic formation using highly inactivated phthalonitrile and dicyanonaphthalene precursors, which serves as a robust and efficient starting point for peripheral functionalizations tailored to specific electronic, biomedical, and catalytic applications.

Results and discussion

Unlike adjacent protocols that benefit from electron-withdrawing substituents such as peripheral fluorines to facilitate macrocyclization, this study demonstrates that unsubstituted phthalonitrile can be efficiently converted into a phthalocyanine core. Furthermore, this approach was successfully extended to the naphthalene core, enabling the macrocyclization of a more conjugated aromatic system under identical reaction conditions. As shown in Scheme 1, phthalonitrile and 2,3-dicyanonaphthalene (2,3-DCN) are employed with LiHMDS and silicon tetrachloride to yield the axially chlorinated silicon phthalocyanine (1) (crude: 35%) and silicon naphthalocyanine (2) (crude: 45%). Despite the similarities in both procedures, compounds 1 and 2 are highly distinct in electronic structure and photophysical properties. Phthalocyanines and porphyrin cores alike, exhibit distinct colours in the solid, pigment state. The deep, dark blue solid illustrated in Fig. 1 resembles the characteristic Cl₂-SiPc colour, whereas the deeper green solid resembles the characteristic Cl₂-SiNc colour. As anticipated, the increase in conjugation in the naphthalocyanine framework results in a bathochromic shift of absorption, correlating with a reduced HOMO–LUMO energy gap, and thus, a colour shift towards the red region.

Scheme 1 Synthetic routes in the preparation of Cl₂-SiPc (1) and Cl₂-SiNc (2) without ammonia gas (i) LiHMDS, tetraline, room temperature, 24 h; and (ii) SiCl₄, 200 °C, 24 h. (b) The intermediate structures are phthalonitriles and dicyanonaphthalenes (i) dicyanonaphthalene having a p-conjugation extension (ii) is proposed based on the process established.¹¹

Fig. 1 Solid state colour of the macrocyclic products: Cl₂-SiPc (left) and Cl₂-SiNc (right).

Following the work-up and isolation of compound 1, the structure was assessed using MALDI-TOF mass spectrometry. The spectra obtained were assessed as is, without further modifications. The key, parent ion peak that was observed at 575.0 m/z represented the cationic fragment of mono-chlorinated SiPc [SiPc-Cl]⁺ (Fig. S1). The isotopic distribution of the mono-chlorinated compound can also be observed (Fig. S2). After additional sample workup, the fragment pattern decreased and a dominant peak at 609.67 m/z was detected, consistent with the molecular species (Fig. S3). Compound 2 was additionally analyzed by mass spectrometry post-work-up. The key, parent ion peak that was observed was at 775.1 m/z, representing the cationic fragment of mono-chlorinated SiNc [SiNc-Cl]⁺ (Fig. S4). Following additional work-up, the spectrum was substantially cleaned up, leading to a monoisotopic peak of 932.01 m/z (Fig. S5). In addition, the splitting pattern was chlorine driven and broadened by carbon and silicon isotopes (Fig. S6). It is known that in the MS space, due to the MS process, the Si–Cl bond is likely ‘broken’ thermodynamic wise and that is the reason for the MS outcomes. To expand from structural analysis via mass spectrometry, both compounds were subjected to UV-visible spectroscopy.

Two solutions with concentrations of 20 ng µL⁻¹ were prepared by dissolving compound 1 (Cl₂-SiPc) and compound 2 (Cl₂-SiNc) in 1,2-dichlorobenzene. Fig. 2A displays the distinct colour change into teal and light green for compound 1 and 2 dissolution. As a result, the UV absorption of compound 1 revealed a λ_max of 699 nm, while the absorption for compound 2 revealed a red-shifted λ_max of 788 nm (Fig. 2B). The absorption maxima are consistent with previously reported values of 694 nm for Cl₂-SiPc dissolved in chloroform and 770 nm for Cl₂-SiNc dissolved in dimethylformamide.^1,12 Altogether, the photophysical data support the identity and purity of the synthesized macrocycles. On the other side, due to the lack of solubility, at least a little bit can be diluted in 1,2-dichlorobenzene to enable the UV absorption data to be acquired, but not enough mass can be dissolved; then the ¹H-NMR data could not be acquired. But others in the space know that once the axial substituent is away from –Cl towards something else then the SiPc and SiNc can have more solubility and other analytics can be taken on.

Fig. 2 (A) Colour of Cl₂-SiPc (left) and Cl₂-SiNc (right) dissolved in 1,2-dichlorobenzene solution for photophysical characterization. (B) Normalized absorption spectra of Cl₂-SiPc (dotted) and Cl₂-SiNc (bolded) in 1,2-dichlorobenzene solution.

This work adapts the ammonia-free macrocyclization protocol from Lessard et al.¹¹ which studied fluorinated silicon phthalocyanines (F₂-F_x-SiPc, x = 4, 8, 16), which employed lithium bis(trimethylsilyl) amide etherate (LiHMDS·Et₂O) in place of ammonia gas.¹¹ By extending this one-pot methodology to unsubstituted dichloride SiPc and SiNc with LiHMDS, this project demonstrates that peripheral fluorination is not essential for successful conversion. To expand, our conditions afforded conversions of 35% (Cl₂-SiPc) and 45% (Cl₂-SiNc) at the 250 mg scale, which exceeded the 21–43% range reported by Lessard et al. at the 5 g scale.¹¹ Although the reported yields remain modest, the 2-fold increase highlights the synthetic relevance of the ammonia-free methodology towards dichlorinated SiPcs and SiNcs.

In our first attempt, we applied the identical conditions outlined by Lessard et al. which resulted only in ∼20% conversion for both SiPc and SiNc at the 250 mg scale, indicating room for optimization. Subsequently, alterations including reduction in solvent volume, reaction time, controlled heating during reagent additions, as well as the work-up procedure, resulted in improved conversions. Altogether, our observations suggested that higher product concentration, coupled with prolonged reaction time, facilitated intermediate formation at room temperature and subsequent macrocyclization with silicon tetrachloride, altogether effective for both phthalonitrile and 2,3-dicyanonaphthalene systems. Additionally, the following results open doors to even greater sustainable optimization via microwave chemistry.

Microwave-assisted synthesis has been recently reported as a convenient and quick alternative to longer and rigorously heated benchtop protocols.¹³ Specifically, our group has recently demonstrated that microwave irradiation reduced reaction times significantly from 3 to 24 hours to under 36 minutes, in the formation of Cl-BsubPcs and Cl-BsubNcs.¹⁴ In this paper, we have applied our previous microwave-assisted approach to the synthesis of Cl₂-SiPc, under the same conditions as the optimized benchtop procedure (see the SI, heating cycle in Fig. S6). As a preliminary result, the yield obtained (∼8 mg, 3%) used the microwave protocol of first irradiating the reaction mixture at 600 W for one cycle, followed by 5 more cycles at 500 W. This was significantly lower than the benchtop procedure since some of the reaction mixture's solvent exited out of the pressure vessel through its bushing during irradiation; therefore, we also studied Cl₂-SiPc microwave synthesis at 400 W cycles (Fig. S7). At this power, we estimate its conversion to be roughly 45–60% and no solvent exit occurred. This was done by comparing its relative phthalonitrile absorbance after microwave-assisted synthesis of Cl₂-SiPc to its initial phthalonitrile concentration (Fig. S8). Due to Cl₂-SiPc's lack of solubility, we could only estimate the reaction's conversion using HPLC-PDA by observing phthalonitrile that has been consumed. However, at our current status, we only did Cl₂-SiPc, which has no solubility; therefore for future studies, we might be able to measure the reaction's true conversion by phenoxylating the Cl₂-SiPc's axial substituent, which then improves its solubility.¹⁵ This would allow us to then monitor both the reaction's phthalonitrile and the SiPc consumption using HPLC-PDA. While further optimization parameters remain to be explored, microwave irradiation has shown its capability in achieving conversion for silicon-based Pcs and Ncs, and most importantly, at a significantly lower heating cycle of 1 hour in comparison to 24 hours.

Outside of the microwave, regarding the past studies of F/fluoro electron withdrawing substitutions and the process setup,¹¹ given this outcome of non-substitutions it was good to see that the SiPc and SiNc can be formed without electron withdrawing intermediates. Therefore, in the upcoming, we will study other substitutions such as –CR_x, –NR_x, –OR_x, –Cl, nitro, etc., as the SiPcs and SiNcs have applications. We will also try to crystallize the intermediate(s) after their reaction with LiHMDS, to acquire XRD data to confirm the proposed structure(s) (Scheme 1i).

Experimental

General

Phthalonitrile (99%) was purchased from TCI America. Lithium bis(trimethylsilyl)amide (LiHMDS (97%)), silicon tetrachloride (SiCl₄, 99%), 1,2,3,4-tetrahydronaphthalene (tetraline, anhydrous, 99%), and 1,2-dichlorobenzene (99%) were purchased from Sigma-Aldrich. Dicyanonaphthalene was prepared as outlined in the literature and recrystalized using acetone.

Synthesis of dichloride-silicon phthalocyanine (1). An oven-dried, 50 mL three-neck round bottom flask was charged with phthalonitrile (250 mg, 1.95 mmol). After purging the flask with argon for 10 minutes, LiHMDS (197 mg, 1.17 mmol) was added under an inert atmosphere and the flask was sealed with rubber septa and placed under positive argon pressure.¹¹ Next, anhydrous tetralin solvent (6 mL) was added via needle syringe transfer and the solution was set to stir rigorously (370 rpm) at room temperature for 24 hours. After that, the flask was connected to a water condenser and continued remaining under positive argon pressure. Silicon tetrachloride (0.31 mL, 2.73 mmol) was slowly added dropwise to the stirring reaction mixture (370 rpm), which was immediately heated to 200 °C. After 24 hours, the deep turquoise solution was allowed to cool to room temperature, which was then admixed with hexanes (20 mL) and methanol (50 mL) and dried under vacuum to result in a deep bluish-teal solid (crude: 95 mg (32%)). MS (MALDI) m/z: [M] calcd for C₃₂H₁₆Cl₂N₈Si, 610.06; found, 575.0. To clean up the mass spectrum, a small amount of the solid (20 mg) was stirred in hexanes (10 mL) at room temperature overnight. The solvent was removed by rotary evaporation and the product was dried under vacuum, and finally re-subjected to mass spectrometry.

Synthesis of dichloride-silicon naphthalocyanine (2). An oven-dried, 50 mL three-neck round bottom flask was charged with dicyanonaphthalene (250 mg, 1.40 mmol). After purging the flask with argon for 10 minutes, LiHMDS (141 mg, 0.84 mmol) was added under an inert atmosphere and the flask was sealed with rubber septa and placed under positive argon pressure. Next, anhydrous tetralin solvent (5 mL) was added via needle syringe transfer and the solution was set to stir rigorously (370 rpm) at room temperature for 24 hours. After that, the flask was connected to a water condenser and continued remaining under positive argon pressure. Silicon tetrachloride (0.23 mL, 1.96 mmol) was slowly added dropwise to the stirring reaction mixture (370 rpm), which was immediately heated to 200 °C. After 24 hours, the deep yellow-brown solution was allowed to cool to room temperature, and was then admixed with hexanes (20 mL). The solid was further washed with hexanes (20 mL) and methanol (30 mL) and dried under vacuum to result in a deep green solid (crude: 127 mg (45%)). MS (MALDI) m/z: [M] calcd for C₄₈H₂₄Cl₂N₈Si, 810.13; found, 775.1. To clean up the mass spectrum, the identical phthalocyanine procedure was applied to the naphthalocyanine.

These reactions were scaled up to the 500 mg scale. See the SI for the synthetic protocol and crude yield.

Conclusion

All in all, we have adapted a one-pot synthesis of dichlorinated SiPcs and SiNcs using LiHMDS in place of ammonia gas, aiming towards a safer and more sustainable alternative to the traditional protocols. Our method is not only applicable to both inactivated phthalonitrile and 2,3-dicyanonaphthalene precursors but also achieves up to a 45% macrocyclic conversion, which is up to a 2-fold increase in comparison to reported yields and via a microwave process, ∼45–60% macrocyclic conversion was also achieved. While further optimization towards higher yields and improved purification strategies remains an open direction, this method provides a safe and practical alternative for access to dichlorinated SiPcs and SiNcs, which further downstream, grants strategies towards applications in organic electronics, photocatalysis, and photodynamic therapy.

Conflicts of interest

There are no conflicts to declare.

Abbreviations

SiPc	Silicon phthalocyanine
SiNc	Silicon naphthalocyanine
OPVs	Organic photovoltaics
OTFTs	Organic thin film transistors
PDT	Photodynamic therapy
LIHMDS	Lithium bis(trimethylsilyl)amide
DII	1,3-Diiminoisoindoline
MALDI-TOF MS	Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry
UV	Ultraviolet.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5nj04250j.

Acknowledgements

The authors gratefully acknowledge NSERC of Canada and their support of this work through a Discovery Grant (DG) to T. P. B.

References

1. T. M. Grant, D. S. Josey, K. L. Sampson, T. Mudigonda, T. P. Bender and B. H. Lessard, Boron Subphthalocyanines and Silicon Phthalocyanines for Use as Active Materials in Organic Photovoltaics, Chem. Rec., 2019, 19(6), 1093–1112,  DOI:10.1002/tcr.201800178.
2. O. A. Melville, T. M. Grant and B. H. Lessard, Silicon Phthalocyanines as N-Type Semiconductors in Organic Thin Film Transistors, J. Mater. Chem. C, 2018, 6(20), 5482–5488,  10.1039/C8TC01116H.
3. T. Keleş, B. Barut, A. Özel and Z. Biyiklioglu, Synthesis of Water Soluble Silicon Phthacyanine, Naphthalocyanine Bearing Pyridine Groups and Investigation of Their DNA Interaction, Topoisomerase Inhibition, Cytotoxic Effects and Cell Cycle Arrest Properties, Dyes Pigm., 2019, 164, 372–383,  DOI:10.1016/j.dyepig.2019.01.044.
4. C. Uslan, K. T. Oppelt, L. M. Reith, B. Ş. Sesalan and G. Knör, Characterization of a Non-Aggregating Silicon(IV) Phthalocyanine in Aqueous Solution: Toward Red-Light-Driven Photocatalysis Based on Earth-Abundant Materials, Chem. Commun., 2013, 49(73), 8108–8110,  10.1039/C3CC44674C.
5. K. Mitra and M. C. T. Hartman, Silicon Phthalocyanines: Synthesis and Resurgent Applications, Org. Biomol. Chem., 2021, 19(6), 1168–1190,  10.1039/D0OB02299C.
6. R. D. Joyner and M. E. Kenney, Phthalocyaninosilicon Compounds, Inorg. Chem., 1962, 1(2), 236–238,  DOI:10.1021/ic50002a008.
7. L. Song, H. Li, U. Sunar, J. Chen, I. Corbin, A. G. Yodh and G. Zheng, Naphthalocyanine-Reconstituted LDL Nanoparticles for in Vivo Cancer Imaging and Treatment, Int. J. Nanomed., 2007, 2(4), 767–774,  DOI:10.2147/IJN.S2.4.767.
8. M. K. Lowery, A. J. Starshak, J. N. Esposito, P. C. Krueger and M. E. Kenney, Dichloro(Phthalocyanino)Silicon, Inorg. Chem., 1965, 4(1), 128,  DOI:10.1021/ic50023a036.
9. W. Zhou, N. J. Yutronkie, B. H. Lessard and J. L. Brusso, From Chemical Curiosity to Versatile Building Blocks: Unmasking the Hidden Potential of Main-Group Phthalocyanines in Organic Field-Effect Transistors, Mater. Adv., 2021, 2(1), 165–185,  10.1039/D0MA00864H.
10. National Research Council (US) Subcommittee on Acute Exposure Guideline Levels, Acute Exposure Guideline Levels for Selected Airborne Chemicals, National Academies Press (US), 2008, vol. 6.
11. M. C. Vebber, B. King, C. French, M. Tousignant, B. Ronnasi, C. Dindault, G. Wantz, L. Hirsch, J. Brusso and B. H. Lessard, From P-Type to N-Type: Peripheral Fluorination of Axially Substituted Silicon Phthalocyanines Enables Fine Tuning of Charge Transport, Can. J. Chem. Eng., 2023, 101(6), 3019–3031,  DOI:10.1002/cjce.24843.
12. L. Macor, F. Fungo, T. Tempesti, E. N. Durantini, L. Otero, E. M. Barea, F. Fabregat-Santiago and J. Bisquert, Near-IR Sensitization of Wide Band Gap Oxide Semiconductor by Axially Anchored Si-Naphthalocyanines, Energy Environ. Sci., 2009, 2(5), 529–534,  10.1039/B822954F.
13. D. A. Davis, C. Schnik, J. Silver, J. L. Sosa-Sanchez and P. G. Riby, A high-yield microwave heating method for the preparation of (phthalocyaninato)bis(chloro)silicon(IV), J. Porphyrins Phthalocyanines, 2001, 5, 376–380,  DOI:10.1002/jpp.293.
14. R. M. De Gennaro, L. Zhang, X. Ma, I. Aggarwal, J. Wadden, N. F. Farac, A. Dovijarski and T. P. Bender, Confirmation of Microwave-Assisted Synthesis of Chloro-Boron Subphthalocyanines and Chloro-Boron Subnaphthalocyanines with Rapid Screening and Data Acquisition via HPLC-PDA, ACS Omega, 2025, 10(30), 33508–33523,  DOI:10.1021/acsomega.5c04003.
15. D. M. Rasadean, T. M. Gianga, A. H. Swan, G. Kociok-Kohn and G. D. Pantos, Chiral Phthalocyanines through Axial Coordination, Org. Lett., 2018, 20, 2645–2648,  DOI:10.1021/acs.orglett.8b00851.

---