Hydrogen Evolution Reaction, Electrochemical CO₂ Reduction, and Oxidative Photodegradation of Organic Dyes Catalyzed by Co(II) Trimethoxy-Meso-Arylporphyrin

Abstract

In search of robust catalysts for redox transformations such as the hydrogen evolution reaction (HER) or CO₂ to CO reduction, we stepped on the previously reported meso-tetrakis(3,4,5-trimethoxyphenyl)porphyrinato cobalt(II) complex [Co(TTMPP)]. We prepared [Co(TTMPP)] in good yields and characterized it by IR, UV-vis absorption, photoluminescence spectroscopy, and cyclic voltammetry (CV). The [Co(TTMPP)] was used as a homogeneous catalyst for the electrochemical formation of H₂ (HER) in DMF (N,N’-dimethylformamide)/TFA (trifluoroacetic acid) and DMF/EtN₃BF₄ solutions, with high faradic efficiencies (FE). Additionally, the reduction of CO₂ to CO in DMF under a CO₂ atmosphere was catalyzed in DMF/TFE (TFE = 2,2,2-trifluoroethanol) and DMF/PhOH with high FE and only traces of H₂ as a by-product. Turnover frequencies of 15.80 or 9.33 s⁻¹, respectively were determined from CV experiments or controlled potential electrolysis in the presence of 1eq. TFE. They were lower with PhOH as proton source with 13.85 or 8.31 s⁻¹, respectively. Further, [Co(TTMPP)] as a solid catalyst (suspension) allowed the photodecomposition of the organic dyes methylene blue (MB) and rhodamine B (RhB) using H₂O₂ under visible light irradiation. The photocatalyst was photostable over five cycles. A photocatalytic mechanism was proposed based on trapping experiments of reactive oxygen species.

1. Introduction

Cobalt porphyrin complexes combine the three oxidation states Co(I)/Co(II)/Co(III) with at least three of the intrinsic oxidation states of the porphyrin ligand Por⁻/Por²⁻/Por³⁻ and thus show rich electrochemistry [1,2,3,4,5,6,7,8,9]. Consequently, they have been used for redox catalysis, [2,4,9,10,11,12,13,14,15,16,17,18,19,20,21] photocatalysis [9,11,12,13,17,18,19,20,21], and for related applications such as molecular sensing [8,21,22,23]. This last application owe Co(por) systems to the binding of additional ligands in the axial positions of the coordination plane defined by the tetradentate porphyrin ligand. This is less pronounced for the oxidation state Co(II) but very important to stabilize the oxidation state Co(III) [24], comparable to the biologically important B₁₂ system (a Co corrin) [25]. Amongst the Co porphyrins, the meso-tetraphenyl porphyrin complexes with meso-tetrakis(phenyl)-porphyrin cobalt(II) [Co(TPP)] (Scheme 1a) as the parent compound, have turned out to be the most interesting group, since the phenyl groups allow vast substitution to vary the redox potentials [1,2,3,4,5,6,7,8,9,10,26,27,28], to confine the metal center through bulky substituents, to introduce charged moieties such as the SO₃⁻ group, and to modify their solubility in water or organic solvents [2,3,4,9,10,11,12,13,14,27,29,30]. Amongst important electrocatalytic processes catalyzed by [Co(TPP)] derivatives, the hydrogen evolution reaction (HER) which is the catalytic reduction of protons [2,9,10,31,32,33,34,35,36,37,38,39,40,41], the reduction or evolution of O₂ [2,32,39,40,42,43,44,45], and the CO₂ to CO reduction [2,9,15,16,46,47,48,49,50,51,52,53,54,55,56,57,58,59] have gained enormous importance in view of the growing need to produce the environmental benign fuels H₂ and O₂ for fuel cell applications and energy conversion and to use waste CO₂ for the production of CO as a versatile C1 building block for base chemicals. Another important application of [Co(TPP)] and derivatives as catalysts is the oxidative degradation of organic pollutants [9,10,11,12,13,17,18,19,60].

We and others have contributed to this field by studying [Co(TPP)] catalysts containing various substitution patterns at the meso-phenyl groups (Scheme 1a) in catalytic [2,14,26,61,62,63,64] and electrocatalytic reactions [2,4,9,10,21,49,52,56] and in oxidative degradation reactions of organic dyes [9,10,11,12,13,18,19].

Herein, we report the preparation of the previously reported [64,65,66,67] meso-tetrakis(3,4,5-trimethoxyphenyl)porphyrinato cobalt(II) complex [Co(TTMPP)] (Scheme 1b), its characterization through elemental analyses, ESI-MS(+) and FT-IR, UV-vis absorption, and fluorescence spectroscopy alongside with its electrochemical behavior. We also report the use of [Co(TTMPP)] as a catalyst in the electrocatalytic evolution of H₂ (HER), for the electrocatalytic reduction of CO₂ to CO, and for the photo-assisted oxidative degradation using H₂O₂ of the dyes methylene blue (MB) and rhodamine B (RhB) (Scheme 1c,d).

2. Results and Discussion

2.1. Synthesis and Analysis

The free base porphyrin H₂TTMPP was prepared and purified, adopting a reported procedure [68] with an isolated yield of 70%. Elemental analysis (see Experimental Section) and FT-IR spectroscopy (Figure S1, Supplementary Materials) showed its purity. The Co(II) complex [Co(TTMPP)] was synthesized in 93% yield using the dimethylformamide procedure [9] and analyzed by elemental analysis,¹H NMR (Figure S2), ESI-MS(+) (Figure S3), and FT-IR spectroscopy (Figure S1), for details, see Experimental Section. The ¹H NMR spectrum showed broad signals due to the paramagnetic character of the Co(II) complex [64].

2.2. Photophysical Properties

The UV-vis absorption spectrum of H₂TTMPP is characterized by an intense absorption band at 424 nm known as the Soret band and four less intense absorption bands at 520, 556, 598, and 652 nm known as the Q bands(Figure 1a) [9,12]. For [Co(TTMPP)], the Soret band is found at 414 nm, thus blue-shifted compared to H₂TTMPP alongside with two Q bands at 532 and 567 nm, compared to the four observed for the free base. These changes are due to the increase in symmetry from C_2v to D_4h upon coordination [9,12,13,18,69]. The optical band gap (Eg-op) was calculated using the 1240/λ_gap method to 1.84 eV (λ_gap = 674 nm) for H₂TTMPP and 2.08 eV (λ_gap = 596 nm) for [Co(TTMPP)]. The value found for [Co(TTMPP)] is typical for Co(II) porphyrins [8,9,11,12,13].

Upon excitation at 430 nm, both H₂TTMPP (λ_max = 658 and 721 nm) and [Co(TTMPP)] (λ_max = 651 and 714 nm) show photoluminescence (PL) in CH₂Cl₂ at room temperature (Figure 1b). They can be assigned to the S₁[Q(0,0)] → S₀ and S₁[Q(0,1)] → S₀ transitions in agreement with previous studies on [Co(TPP)] and derivatives [9,12,13,18]. The PL quantum yields (Φ_PL) are 0.082 for H₂TTMPP and 0.027 for [Co(TTMPP)] with lifetimes of 7.1 ns for H₂TTMPP and 1.3 ns for [Co(TTMPP)] which are in the typical range for meso-arylporphyrins and their Co(II) complexes [3,9,12,13,18].

2.3. Electrochemical Characterization of [Co(TTMPP)]

Cyclic voltammetry on [Co(TTMPP)] was performed in N,N-dimethylformamide (DMF) at room temperature, showing two reversible reduction waves and two reversible oxidation waves (Figure 2). The first reversible one-electron oxidation at 0.26 V vs. SCE is attributed to the Co(II)/Co(III) redox couple in line with previous reports [5,7,9]. From other [Co(TPP)] derivatives and further M(II) porphyrins it is also known that after oxidation solvents coordinate to Co(III), here DMF, thus influencing the potential [9,28,62,70].

A second, slightly larger oxidation wave was observed at 0.97 V and is assigned to a porphyrin-centered process (Por^2−-/Por¹⁻), in agreement with previous reports [5,7,9]. The first reduction E_1/2 = −0.86 V is attributed to the Co(II)/Co(I) redox couple [5,9,28], the second at E_1/2 = −2.04 V to a Por²⁻/Por³⁻ redox couple in line with previous reports [9,28]. The values for [Co(TTMPP)] are very similar to those of the recently studied 4-CF₃ derivative meso-tetrakis(4-(trifluoromethyl)phenyl)porphyrinato Co(II) [Co(TMFPP)] (−2.05, −0.91, 0.30, and 0.98 V) [9] in line with a small to the marginal influence of the substituents on the meso-phenyl rings found also in other studies [3,4,5,9,27,28,70].

Applying scan rates from 250 to 1000 mV/s (Figure 3a) allowed us to calculate the diffusion coefficient (D) of the complex from the Randles–Sevcik equation (Equation (1)) which applies for fully homogeneous diffusion-controlled electrochemical process saying that the peak current (I_p) for a faradaic electron transfer varies linearly with the square root of the scan rate (ν^1/2). D can be calculated from the slope of I_p vs. ν^1/2 (Figure 3b).

I_p = 0.4463 F A (F/RT)^1/2 D^1/2 n_p^3/2 [C₀] v^1/2

(1)

where I_p is the peak current, F is the Faraday constant (96,485 C mol⁻¹), R is the universal gas constant (R = 8.314 J K⁻¹ mol⁻¹), T = 298 K, n_p is the number of electrons transferred (here, n_p = 1), and A is the active surface area of the electrode (0.00785 cm²). Note that our plots are reported as a function of the current density, bypassing the need of the area value in equation 1. Here, [C₀] is the concentration of the analyte (here [C₀] = 1 mM), and ν is the scan rate in V s⁻¹.

D for the Co(II)/Co(I) reduction was determined to 1.8 10⁻⁷ cm S⁻¹, which is larger than the 2.3 10⁻⁸ cm S⁻¹ determined for the porphyrin reduction (Por^2−/Por³⁻). The difference is attributable to the growth of the negative charge. These results are very similar to those obtained previously with the [Co(TMFPP)] complex (1.98 10⁻⁷ cm S⁻¹ and 1.1 10⁻⁸ cm S⁻¹, respectively) [9]. D for the first oxidation was determined to 7.2 × 10⁻⁷ cm S⁻¹. The pretty large value is in keeping with the assumed additional DMF ligand for the oxidized complex [Co(TTMPP)(DMF)]⁺.

2.4. Electrocatalytic H₂ Production

In a recent study, the electrocatalyzed H₂ evolution using the Co(II) complex [Co(bapbpy)Cl]⁺ (bapbpy = 6,6′-bis(2-aminopyridyl)-2,2′-bipyridine) was carried out in DMF and different mechanisms were described depending on the strength of the acid used as proton source [71]. We thus studied the electrocatalytic activity of [Co(TTMPP)] as homogeneous catalyst for H₂ production in DMF using trifluoroacetic acid (TFA) and triethylammonium tetrafluoroborate (HNEt₃BF₄) as proton sources. Upon addition of TFA or HNEt₃BF₄ in DMF, the CVs of [Co(TTMPP)] show catalytic reduction waves at around −2 V which coincide with the second reduction wave of [Co(TTMPP)], whereas the first reduction wave at around −0.9 V remains unchanged (Figure 4 and Figure S4, Supplementary Material). As for the previously reported TPP-CF₃ complex [Co(TMFPP)] [9], the first reduced species [Co(I)(Por²⁻)]⁻ is not catalytically active for [Co(TTMPP)] and efficient proton reduction required the double reduced species which can be described as either [Co(0)(Por²⁻)]²⁻ or [Co(I)(Por³⁻)]²⁻. The catalyst [Co(TCPP)] (H₂TCPP = meso-tetra-para-X-phenylporphine) showed a behavior similar to [Co(TTMPP)] for X = Cl, whereas the X = OMe or H derivatives reduced protons catalytically already at around −1 V [10]. Maybe the character of the second reduction is very sensitive to the substitution pattern. Alternatively, the reactive two-electron reduced species might be generated through a disproportionation reaction: 2 [Co(Por)]⁻ = [Co(Por)] + [Co(Por)]²⁻ for the complexes with X = OMe or H and this reaction is again depending on the substitution pattern. This remains to be studied in more detail and will be further discussed in Section 2.5.

Controlled potential electrolysis (CPE) at −2 V for 2 h in aqueous DMF under an Ar atmosphere gave a faradaic efficiency (FE) of 76% in the presence of three equivalents TFA (

Table 1. Important parameters for the electrocatalytic H₂ evolution ^a.

H+ Source	Time CPE	ECPE	Icat/Ip	FEH2	TON	TOF (h−1)
3 eq. TFA	2 h	−2.00	10.86	76	11.04	5.52
3 eq. HNEt3+	2 h	−1.89	13.65	88	14.60	7.30

^a Under CPE conditions using [Co(TTMPP)] as catalyst in 0.1 M n-Bu₄NBF₄/DMF under an Ar atmosphere. E_CPE = applied potential, FE = faradaic efficiency, TON = turnover number, TOF = turnover frequency.

). Gas chromatography (GC) confirmed the production of H₂ (Figure S5) during CPE. The catalytic current enhancement (I_cat/I_p) where I_cat is the catalytic current after the addition of a proton source and I_p is the peak current in the absence of acid is 10.86, the turnover number (TON) is 11.04, and the turnover frequency (TOF) 5.52 h⁻¹. In the presence of three equivalents HNEt₃BF₄, CPE at −1.89 V for 2 h in DMF under an argon atmosphere we recorded a higher FE of 88%. In addition, I_cat/I_p (13.65), the TON (14.6), and TOF (7.3 h⁻¹) values are superior for HNEt₃BF₄ over TFA.

The FE for [Co(TTMPP)] in the presence of HNEt₃BF₄ is also slightly higher than the one found for the [Co(TMFPP)] complex (85% using DMF/acetic acid) [9] and overall these values are comparable to those of other Co(II) porphyrin derivatives with various meso-substituents [10,31,32,36,40,41,71]. For example, the complexes [Co(TMAP)](ClO₄)₂ (H₂TMAP = meso-tetrakis(N,N,N-trimethylanilinium-4-yl)porphine), [Co(TMPyP)](ClO₄)₂ (meso-tetrakis(N-methylpyridinium-4-yl)porphine), and [Co(TpyP)] (meso-tetrakis-4-ylporphine) showed FEs around 90% when using TFA as proton source [41]. For [Co(TMAP)] (H₂TMAP = meso-tetrakis(N,N,N-trimethylanilinium-4-yl)porphine) an FE of almost 68% was reported when using acetic acid as proton source [10]. Our experiments revealed a high sensitivity to the efficiency of the proton source and it is clear that direct comparison with results of studies using different proton sources is difficult.

Markedly higher TONs of 28 and 35 have been recently reported for the above-mentioned polypyridyl Co(II) complex [Co(bapbpy)Cl]⁺ [71] at overpotentials of >0.8 V, whereas at an overpotential of <0.6 the TON drops to 4 (with HBF₄ as proton source).

2.5. Electroreduction of CO₂ to CO

2.5.1. Catalytic Behavior under CO₂

The electrocatalytic behavior of [Co(TTMPP)] as a homogeneous catalyst in CO₂ reduction was studied in CO₂-saturated DMF solutions with trifluoroethanol (TFE) or phenol (PhOH) as H⁺ sources. Cyclic voltammograms in the presence of CO₂ show catalytic currents at around −2 V (Figure 5 and Figure S6). Upon the addition of protons without CO₂, similar catalytic currents were observed. Slightly higher currents were found for the combination of CO₂ and protons with I_cat/I_p values of 5.6 for TFE and 4.9 for PhOH. As observed for the proton reduction, the first reduced species [Co(I)(Por²⁻)]⁻ seems not to be catalytically active for the CO₂ reduction, whereas the double reduced species ([Co(0)(Por²⁻)]²⁻ or [Co(I)(Por³⁻)]²⁻) appears to be active. The same was previously found for the TPP-CF₃ derivative [Co(TMFPP)] [9].

The catalytic performance was studied under CPE conditions at −1.94 V and −1.93 V in DMF under a CO₂ atmosphere in the presence of PhOH (Figure 5) or TFE (Figure S6) and faradaic efficiencies (FEs) of 95% and 88% were found (

Table 2. Faradaic efficiencies of the CO₂ reduction ^a.

H+ Source	Time CPE	ECPE	FECO2	FEH2
1 eq. TFE	2 h	−1.94	95	Not detected
1 eq. PhOH	2 h	−1.93	88

^a Under CPE conditions using [Co(TTMPP)] as catalyst in 0.1 M n-Bu₄NBF₄/DMF under a CO₂ atmosphere. E_CPE = applied potential, FE = faradaic efficiency.

). GC confirmed the production of CO and no detectable amounts of H₂ (Figure S7). From this, we conclude that in the presence of CO₂, the protons only cause a lowering of the overpotential of the CO₂ to CO reduction but are not reduced [9,15,47,50,51,52,53,54,57,58]. For the CF₃-substituted derivative [Co(TMFPP)], we recently obtained an FE of 90% in aqueous DMF [9], whereas for the standard [Co(TPP)] only 50% were found under similar conditions [15]. Remarkably, when immobilized on carbon nanotubes, [Co(TPP)] gave efficiencies of 83% or 93% at −1.15 and −1.35 V, respectively [15]. We can thus conclude that both CF₃ and OMe substitution enhanced the efficiency for the CO₂ reduction using [Co(TPP)] derivatives, whereas the applied potentials are in the same range as for [Co(TPP)]. Support with an electron-conducting material could pave the way for the use of [Co(TTMPP)] as an electrocatalyst at less negative potentials [15,16,30,42,46,51,54,72].

Mechanistic studies suggest CO₂ activation by the singly reduced species [Co(I)(Por²⁻)]⁻ with subsequent protonation to [Co(II)(COOH)(Por²⁻)]. Further protonation and H₂O loss lead to [Co(III)(CO)(Por²⁻)] [2,58] and one-electron reduction back to the parent Co(II) complex [2]. As a side reaction, [Co(I)(Por²⁻)]⁻ reacts with H⁺ forming a hydrido complex [Co(III)(H)(Por²⁻)] which reacts with H⁺ to yield H₂ [2]. As pointed out above, we did not observe H₂ production. The proton source just facilitated the CO₂ to CO conversion and from our experiments, we conclude that the potentials of the two one-electron reduction steps for the CO₂ reduction might be markedly different depending on the porphyrin substitution pattern and the proton source [58]. This remains to be studied in more detail.

2.5.2. Benchmarking of the Catalyst

The efficiency of an electrocatalyst is a function of its overpotential, the inherent turnover frequency (TOF, cycles completed per second), the number of turnovers (TON, the maximum number of cycles for one mol of catalyst), and the speed of catalysis expressed as the maximum TOF (TOF_max) [73,74]. The ratio I_cat/I_p measured at different scan rates gives a good estimation of the TOF_max. The catalytic plateau current (I_cat) can be expressed as in Equation (2) assuming the electron transfer to the catalyst is fast and the typical S-shaped feature of the current is observed [73,74].

I_cat= n_catF A [C₀] (D k_cat [CO₂])^1/2

(2)

The catalysis follows the first-order rate in both the catalyst and substrate. Combining Equations (1) and (2), the maximum turnover frequency TOF_max = k_cat [CO₂] can be determined using Equation (3) from the cyclic voltammograms recorded in CO₂-saturated DMF solution in presence of TFE or PhOH.

TOF_max = k_cat [CO₂] = (F v np³/R T)(0.4463/n_cat)²(I_cat/I_p)²

(3)

For both Equations (2) and (3), n_cat is the number of electrons required for the catalytic reaction (n_cat = 2), n_p = number of transferred electrons (here = 1), ν is the scan rate, F is the Faraday constant (96,485 C mol⁻¹), A is the surface area of the electrode (A = 0.00785 cm²), C_cat is the catalyst concentration ([C₀] = 10⁻³ M), D_cat is the diffusion constant of the catalytically active species, k_cat is the rate constant of the catalytic reaction, and [CO₂] is the concentration of CO₂ in DMF (Figures S8 and S9).

The thus determined TOF_max values for [Co(TTMPP)] are 15.80 s⁻¹ and 13.85 s⁻¹ for 1 equivalent TFE and 1 equivalent PhOH, respectively. TOFs for Co-based catalysts were reported to range from 0.2 to >1000 s⁻¹ [47] and for supported [Co(TPP)] values of 2.5 to 8.7 s⁻¹ were reported depending on the C-support [48,50]. Thus, [Co(TTMPP)] showed good performance in solution even without support. At the same time, it is difficult to compare homogeneous and heterogeneous catalysts and most [Co(TPP)] electrocatalysts were used as supported [47,48,49,50,51] or encapsulated [45,46] heterogeneous catalysts.

Catalytic Tafel plots allow examining the catalytic performances against both kinetic (TOF_max) and thermodynamic (overpotential, η) descriptors [2,73,74]. In DMF, the standard potential of the CO₂/CO couple can be described through Equation (4) [73,74]:

E⁰_{CO2/CO,DMF,HA} = −0.259 − 0.0529 pka(HA,DMF)

(4)

with pka^TFE = 24.0 and pka^PhOH = 18.8 in DMF [7] and overpotential can be determined as shown in Equation (5):

η = |E⁰_CO2/CO − E_app |

(5)

The TOF was plotted against the overpotential (Figure 6) applying Equation (6):

TOF = TOF_max/(1 + exp(F/(R T)(E⁰_CO2 − E_cat))exp(−(F η)/(R T))

(6)

In this electroanalytical method, competing factors such as substrate depletion and catalyst inhibition are minimized by analyzing the foot of the catalytic wave to determine the observed catalytic rate constant (k_cat) [74]. From plots of I/Ip versus 1/{1 + exp[(F/RT)(E − E_cat/2)]}, k_cat can be calculated from the slope of the linear portion of the curve, which gives access to the maximum TOF value, where TOF = k_cat[CO₂] under saturation conditions [2,73,74,75]. Under CPE conditions, scan-rate-independent TOFs of 9.33 and 8.31 s⁻¹ were determined, whereas the same method gave slightly larger values (15.80 and 13.85 s⁻¹) from CVs (

Table 3. Catalytic parameter for the CO₂/CO reduction ^a.

H+ Source	ECPE	Icat/Ip	TOF(s−1)		TON b
			CPE	CVs
1eq. TFE	−2	5.6	9.33	15.80	113,760
1eq. PhOH	−2	4.9	8.31	13.85	99,720

^a From the Tafel plots (Figure 6). ^b From the CVs Figures S8 and S9.

). These values are comparable to those found for other Co porphyrin derivatives with various meso-substituents [2,15,16,46,47,48,49,50,51,54,56,72]. UV-vis absorption spectra of the solution before and after the CPE experiments show that the [Co(TTMPP)] complex is almost quantitatively retained after 2 h of CPE, indicating good stability of the catalyst (Figure S10).

2.6. Photocatalytic Degradation of Methylene Blue and Rhodamine B Using H₂O₂

UV-vis absorption spectroscopy allowed following the oxidative photodegradation of the two dyes methylene blue (MB) and the rhodamine B (RhB) in H₂O using H₂O₂ as oxidant and [Co(TTMPP)] as solid heterogeneous catalyst Figure 7 and Figure S11).

The degradation efficiencies after 90 min under visible light irradiation were determined to be 92.6% for RhB and 84.1% for MB. In the absence of light, degradation was lower than 1%. These values are similar to those recently reported for the [Co(TMFPP)] derivative [9], and markedly higher compared to those of [Co(TMPP)(4-CNpy)] and [Co(TClPP)(4-CNpy)] which were up to 80% for MB after 300 min reaction time [18]. Comparison of R² showed that the reactions follow pseudo-first-order kinetics with ln(C₀/C_t) = k t, where C₀ is the initial dye concentration, C_t is the dye concentration at time t, and k is the rate constant. The rate constants k were calculated to 0.023 for RhB and 0.017 min⁻¹ for MB (Figure 7b).

Recycling experiments with five cycles showed only a slight reduction of activity (Figure S12) with efficiencies decreasing from 91.2% to 88.2% for PhB and from 84.1% to 80.3% for MB after five cycles (90 min each). Importantly, parts of the loss of catalytic activity are caused by the unavoidable loss of photocatalyst during recovery.

To get more insight into the photodegradation mechanism of RhB and MB using our catalyst, the influence of potentially active species such as superoxide radicals (^•O₂⁻), hydroxyl radicals (^•OH), or holes (h⁺) in the reaction [17,60] was investigated using several types of scavengers: L-ascorbic acid for (^•O₂⁻), isopropyl alcohol (IPA) for (^•OH), and sodium ethylene diamine tetraacetate Na₂(H₂EDTA) for hole trapping [76]. The addition of IPA led to a marked decrease in the degradation efficiency to 50.1% for RhB and 44.3% for MB. The addition of L-ascorbic acid reduces the rates even to 28.3% for RhB and 21.3% for MB. In contrast to this, the addition of Na₂(H₂EDTA) reduced the efficiency only to 90.2% for RhB and 81.5% for MB. We therefore conclude that holes play only a minor role in the photodegradation of RhB and MB, whereas (^•OH) and (^•O₂⁻) are the predominant species.

3. Experimental Section

3.1. Materials

N,N-Dimethylformamide (DMF, >99.8%, extra dry over molecular sieves), N,N-diisopropylethylamine (DIPEA), Co(OAc)₂·4H₂O, p-chloranil, NEt₃, CH₂Cl₂, CHCl₃, n-hexane, EtOH, trifluoroacetic acid (TFA), and BF₃ OEt₂ (all Acros Organics); HNEt₃BF₄, rhodamine B (RhB), methylene blue (MB), H₂O₂, 3,4,5-trimethoxybenzaldehyde, pyrrole, H₂O₂ (30%), ethyl acetate (AcOEt), isopropyl alcohol (IPA), L-ascorbic acid, sodium ethylene diamine tetraacetate Na₂(H₂EDTA) (all Sigma-Aldrich, Merck, Darmstadt, Germany); CF₃CH₂OH (TFE) and phenol (PhOH) (both Alfa-Aesar, Thermo Fisher, Kandel, Germany); n-Bu₄NBF₄ and n-Bu₄NPF₆ (purriss. Fluka, Merck, Darmstadt, Germany), were used as received.

3.2. Syntheses

3.2.1. Meso-Tetrakis(3,4,5-Trimethoxyphenyl)Porphyrin (H₂TTMPP)

Here, 3,4,5-trimethoxybenzaldehyde (363 mg, 1.85 mmol) and pyrrole (127 μL, 1.85 mmol) were added to a 250 mL of distilled CHCl₃ in a double-necked round bottom flask under argon and shielded from light. BF₃·OEt₂ (192 μL, 0.0015 mol) was added, and the reaction was maintained at room temperature for 2 h. A few drops of NEt₃ and 179 mg. of p-chloranil (1.66 mmol, 0.75 equivalents) were added, and the solution was heated to reflux (light protection was removed). After 1 h, the obtained solution was cooled to room temperature. The reaction mixture was evaporated to dryness, the residue dissolved in CHCl₃, filtered over silica, and recrystallized from CHCl₃/n-hexane (8:2). Yield: 126 mg (0.12 mmol, 70%) of a purple solid. Anal. calcd. for C₅₆H₅₄O₁₂N₄ (974.37): C, 68.98; H, 5.58; N, 5.75; found: C, 68.92; H, 5.55; N, 5.71%; MS (ESI(+), CH₂Cl₂): m/z = 974.38 for [M]⁺; UV-vis (CH₂Cl₂): λ_max (ε 10⁻³ M⁻¹cm⁻¹): 424(365), 520(78),556(41),598(24),652(32); ¹H NMR (500 MHz, CDCl₃) δ = 9.01 (s, 8H, β-pyrrole), 7.48 (s, 8H, arylH), 4.23 (s, 24H, OCH₃), 4.01 (s, 12H, OCH₃) ppm, −2.73 (s, 2H, NH) ppm. FT-IR (solid, $\stackrel{=}{\mathsf{\nu }}$, cm⁻¹); 3328 (w), 2973 (s), 2942 (s), 2885 (m), 1745 (m), 1604 (m), 1505 (s), 1462 (m), 1288 (vs), 1235 (vs), 1164 (s), 1107 m), 1032 (s), 956 (s), 802 (vs), 734 (vs), 592 (s), 530 (s), 421 (m).

3.2.2. Meso-Tetrakis(3,4,5-Trimethoxyphenyl)Porphyrinato Cobalt(II) [Co(TTMPP)]

H₂TTMPP (0.200 g, 0.2 mmol) was dissolved in a mixture of CHCl₃ (8 mL) and EtOH (2 mL), followed by the addition of Co(OAc)₂·4H₂O (0.300 g, 1.2 mmol) and N,N-diisopropylethylamine (DIPEA) (0.073 mL, 0.42 mmol). The reaction mixture was stirred under an inert atmosphere of argon and under reflux at 55 °C for 1 h followed by extraction with CHCl₃ (15 mL) and H₂O (15 mL) six times (interleaved). The product was purified by silica gel column chromatography using CH₂Cl₂/ethyl acetate (7:3) as eluent. The resulting solid was filtered, washed with n-hexane, and finally dried under vacuum to yield 191 mg (0.18 mmol, 92.6%) of product. Anal. calcd. for C₅₆H₅₂N₄O₁₂Co (1031.29): C, 65.18; H, 5.08; N, 5.43; found: C, 65.15; H, 5.06; N, 5.41%; MS (ESI(+), CH₂Cl₂): m/z = 1031.29 for [M]⁺; UV-vis (CH₂Cl₂): λ_max (ε 10⁻³ M⁻¹cm⁻¹): 414(356), 536(47), 573 sh(18); ¹H NMR (500 MHz, CDCl₃): δ = 16.32 (s, 8H, β-pyrrole); 12.67 (s, 8H, arylH); 5.48 (s, 12H, OCH₃), 5.01 (s, 24H, OCH₃); FT-IR (solid, $\stackrel{-}{\mathsf{\nu }}$, cm⁻¹); 3063 (w), 2962 (s), 2928 (vs), 2859 (s), 1725 (vs), 1606 (m), 1456 (s), 1381 (m), 1273 (vs), 1123 (vs), 1072 (vs), 1043 (m), 957 (m), 740 (s), 699 (m), 648 (w).

3.3. Methods and Instrumentation

UV-vis absorption spectra were recorded on a WinASPECT PLUS (validation for SPECORD PLUS version 4.2) scanning spectrophotometer (Analytic, Jena, Germany) using 10 mm path length cuvettes. FT-IR spectra were measured on a Perkin Elmer Spectrum Two FT-IR spectrometer (Perkin Elmer, Darmstadt, Germany). The ¹H NMR spectra were measured on Bruker DPX 500 spectrometers (Bruker, Rheinhausen, Germany) in solution in deuterated solvents based on the solvent peak as an internal standard. Elemental analysis and mass spectrometry were practiced in the nanobio chemistry platform of the ICMG, Grenoble, France. A Fluoromax-4 spectrofluorometer (Horiba Scientific, Loos, France) to record photoluminescence (PL) spectra at room temperature in CH₂Cl₂. PL quantum yield (Φ_PL) was determined using the optical method [77] with [Zn(TPP)] as standard (Φ_PL = 0.031). The lifetimes were measured upon irradiation at λ = 405 nm using the single photon counting technique and the fluorescence decay was fitted to single exponentials with the PicoQuant FLUOFIT software (PicoQuant, Berlin, Germany).

3.4. Electrochemistry

Cyclic voltammetry experiments were performed using a CH-660B potentiostat (CH Instruments, Austin, TX, USA) or a Metrohm µstat400 (Metrohm, Filderstadt, Germany) at room temperature. All measurements were performed in DMF (freshly distilled) with a solute concentration of approximately 10⁻³ M and n-Bu₄NBF₄ (0.1 M) as supporting electrolyte. A three-electrode cell was set up with a glassy carbon working electrode, a Pt wire as counter electrode, and an Ag/AgNO₃ reference electrode. Potentials were converted into values for the saturated calomel electrode (SCE) by applying equation 7 [6,9,13,78]:

E(SCE) = E(Ag/AgNO₃) + 360 mV

(7)

NHE potentials are converted into the current SCE scale by subtracting about 240 mV, whereas SCE differs from the ferrocene/ferrocenium couple by +160 mV [78].

3.5. Electrocatalytic CO₂ Reduction

The experiments were performed at room temperature under a CO₂ atmosphere in a conventional three-electrode cell sealed with Apiezon M vacuum grease (Sigma Aldrich/Merck, Darmstadt, Germany). A glassy carbon electrode plate (2 cm², 0.25 mm thickness) was used as working electrode in the cathodic compartment. A 0.5 mm diameter platinum wire (10 cm length) was used as counter electrode in the anodic compartment. The cell was charged with the catalyst and then purged with argon or CO₂ for a minimum of 15 min before controlled potential electrolysis was carried out. Constant magnetic stirring was applied during electrolysis.

3.6. Gas Chromatography (GC)

Gas detection was performed using GC/MS gas chromatography (Perkin Elmer Clarus 560; Perkin Elmer, Darmstadt, Germany) instrument with a thermal conductivity detector fitted with RT-QPlot pre-column + molecular sieve 5 Å column. Temperature was held at 150 ºC for the detector and 80 ºC for the oven. The carrier gas was helium. Manual injections of 100 μL were performed during the experiment via a gas-tight Hamilton microsyringe. The total volume of the cell was 173 mL.

3.7. Faradaic Efficiency, Turnover Number, Turnover Frequency Calculation

The faradaic efficiency (FE) of CO₂ and the hydrogen evolution reaction (HER) were calculated using Equation (8):

FE = Z n F/Q

(8)

where Z is the amount of product in mol, n is the number of electrons (2 for both CO and H₂), F is the Faraday constant, and Q is the number of electrons (or charge) passed through the solution during electrolysis.

3.8. Gas Phase Analyses

Gas phases were analyzed by GC and the Turnover Number (TON) was calculated based on the total amount of the products in mmol in the gas phase (CO and H₂) by different porphyrin catalysts, divided by the total amount of each individual catalyst in the electrolysis solution (Equation (9)).

TON = n(product)/n(catalyst)

(9)

The turnover frequency (TOF) was calculated using TON divided by the time of the electrolysis (Equation (10)):

TOF = n(product)/n(catalyst)/t

(10)

The n(catalyst) is calculated based on the following equation (Equation (11)):

n_(ca𝑡) = C_{(𝐶𝑎𝑡)} × 𝑉_𝑠𝑜𝑙

(11)

3.9. Photo-Decomposition of RhB and MB with H₂O₂

The photocatalytic reaction was performed in a quartz tube reactor (Sigma Aldrich, Merck, Paris, France). A total of 25 mg (0.024) mmol of [Co(TTMPP)] was dispersed in aqueous solutions of MB (M_W = 319.85 g/mol) or RhB (479.03 g/mol) (both 0.05 mmol). Before irradiation, the suspensions were stirred for 90 min in the dark in order to reach an adsorption–desorption equilibrium of the dye molecules on the surface of the catalysts. Then H₂O₂ (0.006 mmol) was added and the solutions were irradiated. During the photoreaction, about 3 mL of suspension was collected at different time intervals and centrifuged to remove solid materials. The concentrations of the dyes were determined by recording the UV-vis absorption of the supernatant at 555 nm (RhB) and 654 nm (MB). The efficiency was calculated using (Equation (12)):

Yield (%) = (C₀−C_t/C₀) ∗ 100

(12)

where C₀ is the initial concentration of dyes and C_t is the concentration at different time intervals.

For the catalyst recycling experiments, the solid catalyst was filtered off after each cycle and washed with water and EtOH (five times each). Then the catalyst was dried at 60 °C for 12 h and re-dispersed in fresh MB or RB solutions.

3.10. Oxidative Photodegradation Mechanism–Trapping Experiments

A small number of scavengers (5 mmol/L) were added in the dark to the aqueous RhB or MB solution before adding the solid catalysts and starting the irradiation.

4. Conclusions

The previously reported meso-tetrakis(3,4,5-trimethoxyphenyl)porphyrin (H₂TTMPP) and its cobalt(II) complex [Co(TTMPP)] were synthesized and for the first time characterized by IR spectroscopy, UV-vis absorption and photoluminescence spectroscopy, as well as by cyclic voltammetry (CV). CVs of [Co(TTMPP)] showed two fully reversible reduction waves at E_1/2 = −0.88 V and E_1/2 = −2.03 V vs. SCE assignable to Co-centered reduction. The first oxidation process at 0.3 V is reversible and assigned to the Co(II)/Co(III) couple. A second two-electron oxidation follows at 0.96 V and is very probably porphyrin-based. Using [Co(TTMPP)] as a homogeneous catalyst for the electrochemical formation of H₂ from DMF/TFA and DMF/EtN₃BF₄, we found faradaic efficiencies (FE) of 76% and 88%, respectively, upon electrolysis at −2 V. At similar potentials, the reduction in CO₂ to CO in DMF under a CO₂ atmosphere was catalyzed in the presence of TFE and PhOH as proton sources with high FEs of 95% and 88%, respectively, good turnover frequencies of 15.80 s⁻¹ (TFE) and 13.85 s⁻1 (PhOH), and only traces of H₂ as a by-product. Remarkably, the reaction rates of both H⁺ and CO₂ reduction reactions were higher than for the parent [Co(TPP)] complex, although the applied potentials were quite similar. We found that the performance of both H⁺ and CO₂ reduction is strongly dependent on the proton source and in future experiments we will study this in more detail. Additionally, the application of supported, thus heterogenized, Co(II) porphyrins seems to be an interesting venue to achieve lower potentials, higher turnover numbers and frequencies, and higher stability.

Further, the dyes methylene blue and rhodamine B were photodecomposed using H₂O₂ and [Co(TTMPP)] as solid heterogeneous catalysts with an efficiency of 84.1% and 92.6%, respectively, in 90 min under visible light irradiation. Trapping experiments of reactive oxygen species (ROS) show that holes play only a minor role in the photodegradation of RhB and MB, whereas (^•OH) and (^•O₂⁻) are the predominant species. The [Co(TTMPP)] as a solid photocatalyst was found to be photostable over five reaction cycles. Also for the degradation, C-based supports might pave the future way to even more efficient catalysts.