Evidence for the Formation of 1,2-Dioxetane as a High-Energy Intermediate and Possible Chemiexcitation Pathways in the Chemiluminescence of Lophine Peroxides

Chemiluminescence of Lophine Peroxides

Article

Evidence for the Formation of 1,2-Dioxetane as a High-Energy

Intermediate and Possible Chemiexcitation Pathways in the

Andreia Boaro, Roberta Albino Reis, Carolina Santana Silva, Dieg

Alexander Garreta Goncalves Costa Pinto, and Fernando Heering Bartoloni*

o Ulysses Melo,

Cite This: J. Org. Chem. 2021, 86, 6633 −6647

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: A kinetic study of the chemiluminescent (CL)

reaction mechanism of lophine-derived hydroperoxides and

silylperoxides induced by a base and ﬂuoride, respectively,

provided evidence for the formation of a 1,2-dioxetane as a high-

energy intermediate (HEI) of this CL transformation. This was

postulated using a linear Hammett relationship, consistent with the

formation of negative charge on the transition state of HEI

generation (ρ > 1). The decomposition of this HEI leads to

chemiexcitation with overall low singlet excited state formation

−

−1

quantum yield (Φ from 1.1 to 14.5 × 10 E mol ); nonetheless,

−

−1

Φ = 1.20 × 10 E mol was observed with both peroxides

substituted with bromine. The use of electron-donating sub-

stituents increases chemiexcitation eﬃciency, while it also reduces

the rate for both formation and decomposition of the HEI. Diﬀerent possible pathways for HEI decomposition and chemiexcitation

are discussed in light of literature data from the perspective of the substituent eﬀect. This system could be explored in the future for

analytical and labeling purposes or for biological oxidation through chemiexcitation.

INTRODUCTION

process, generating a neutral lophine radical and a superoxide

■

ion O₂^●(Scheme 1b).

−

13−15

These radicals combine to

More than 140 years ago, Radziszewski was the pioneer in the

study of the chemiluminescence (CL) of lophine (i.e., 2,4,5-

triphenyl-1H-imidazole, Scheme 1a, L1) in basic media and in

the presence of oxygen. Such transformation is historically

important in the CL ﬁeld for being used by Wiedemann in

888 to deﬁne the term itself (“Das bei chemischen Prozessen

auftretende Leuchten wurde Chemilumineszenz genannt”, i.e.,

the light emission observed during chemical processes must be

called chemiluminescence). Despite being the ﬁrst-reported

−

produce a peroxyanion (H ) that, after cyclization, results in a

−

1,2-dioxetane intermediate (Scheme 1c, D ). The decom-

−

position of cyclic peroxide D in the so-called chemiexcitation

step generates a deprotonated benzoylamidine derivative in the

−

4,5,11,13−15

excited state (Scheme 1c, B *).

Finally, the last step

−

consists in the ﬂuorescence decay of B * to the ground state,

−

generating light and B as ﬁnal reaction products (Scheme

,9−11,13−15

c).

This CL mechanism (Scheme 1b,c) shares

CL system, the mechanism of the lophine decomposition

reaction, responsible for the generation of light, is not fully

some common aspects with bioluminescent systems compris-

ing 1,2-dioxetanones as high-energy intermediates (HEIs).

understood. Detailed mechanistic studies of other CL

transformations subsequently reported revealed that the light

observed as a product of these reactions is generated from the

exothermic decomposition of cyclic peroxide intermedi-

For instance, evidence suggests that a SET reaction between

16−18

O₂and the enolates of the ﬁreﬂy

luciferins is involved in the formation of their corresponding

and cypridine

−11

ates.

4,5,11

peroxyanions.

These peroxyanions lead to the 1,2-

To perform CL studies, a wide variety of lophine derivatives,

substituted in the 2-phenyl group (Scheme 1a, L1-6) and in 4-

phenyl and 5-phenyl (Scheme 1a, L7-13), were conveniently

obtained by condensation of their corresponding benzaldehyde

Received: February 19, 2021

Published: April 20, 2021

,12

with benzil and NH OAc in acidic conditions.

The CL

mechanism for L1 and some of its derivatives (e.g., L2, L7, and

L8, Scheme 1a) involves the reaction of lophine anion with

oxygen, probably through a single-electron transfer (SET)

2021 American Chemical Society

633

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

Article

Scheme 1. Molecular Structure of Lophine (L), Hydroperoxide (H), and Silylperoxide (S) Derivatives (a); CL Reaction

Mechanism for Lophine (L1) under Basic Condition and in the Presence of Oxygen (b,c); and for Hydroperoxides and

3,13−15

Silylperoxides in the Presence of a Base and Fluoride Respectively (d,c)

^aHEI of this sequence is the 1,2-dioxetane anion (D⁻).

dioxetanone HEIs necessary for generation of excited

states.

cence. In addition, we propose that the electronic nature of

substituents seems to play a role in determining the

multiplicity of the excited state, resulting from chemiexcitation.

,5,11

Isolated hydroperoxides (e.g., H1, H2, and H7, Scheme 1a),

prepared from their corresponding lophines by photooxyge-

nation, are also chemiluminescent in basic conditions (KOH in

RESULTS AND DISCUSSION

■

EtOH). The CL reaction of these hydroperoxides generates

benzoylamidines emitting at the same wavelength of the CL

Peroxide Preparation and Photophysical Character-

ization. Six lophine derivatives (L1−6, Scheme 1a) were

obtained from the condensation of their corresponding para-

reaction of lophines. It is postulated that in basic conditions,

−

H is deprotonated to the peroxyanion H (Scheme 1d), and

substituted benzaldehyde with benzil and NH OAc in glacial

CL arises from the decomposition of the common

,12,22

_i_n_t_e_r_m_e_d_i_a_t_e_D− (Scheme 1c).

13,20

acetic acid (Scheme 2a).

Our research group has been

However, no direct

evidence for the formation of this 1,2-dioxetane as HEI has

been reported. Kimura et al., studying the CL of H7-11

Scheme 2. Preparation of Lophines L1−6 (a), Their

Corresponding Hydroperoxides H1−6 (b), and

(

Scheme 1a) with KOH in MeOH/CH Cl 1:5, commented

2 2

−

Silylperoxides S1−6 (c), with R = H (1), OMe (2), Me (3),

that H cyclization to produce D (Scheme 1c) could be the

rate-determining step of the CL reaction. Lu et al. studied

Br (4), CF (5), and CN (6)

the isomeric silylperoxides (R)-S12 and (R)-S13 (Scheme 1a),

whose CL was triggered by addition of ﬂuoride tetrabuty-

lammonium (TBAF) in CHCl or acetone. The lower CL

quantum yield (Φ ) in acetone for (R)-S12 when compared

to (R)-S13 was attributed to the involvement of a partially

cyclic transition state (TS) at the chemiexcitation step. In

CHCl , CL eﬃciencies were equal within experimental errors,

−

suggesting that 1,2-dioxetane D plays a main role in the

formation of excited states in this solvent. To access the

occurrence of D as a HEI on this CL system, in this work, we

−

have conducted a thorough kinetic study on the decomposition

of H1-5 and S1-6 (Scheme 1a). Substitution only at the 2-

phenyl ring was intentional to observe the substituent eﬀect on

peroxides directly related to Radziszewski’s L1 chemilumines-

Methods: (a) NH OAc in glacial HOAc; (b) O in CH Cl /ACN

5:1; (c) TBDMSCl and IMI-H (cat.) in CH Cl .

634

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

Article

Table 1. Photophysical Properties for the Studied Lophines (L), Hydroperoxides (H), Silylperoxides (S), and Benzoylamidines

(

B): Maximum Absorption Wavelength (λ ), Molar Extinction Coeﬃcient (ε), Maximum Fluorescence Wavelength (λ ),

Fluorescence Quantum Yield (Φ ), Stokes Shift (Δλ), and Maximum Chemiluminescence Wavelength of the CL Reaction

(

λ ).

ε (L mol cm⁻¹

−1

Δλ (cm⁻¹)

λ_a_b_s(nm)

)

λ_F_L(nm)

Φ_F_L

λ_C_L(nm)

306

302

306

314

320

342

278

302

284

282

249

277

301

282

281

249

267

238

274

237

256

242

260

27,600 ± 300

29,000 ± 200

26,800 ± 100

30,200 ± 200

25,900 ± 600

24,520 ± 20

20,700 ± 100

24,960 ± 60

24,630 ± 80

25,600 ± 100

16,300 ± 70

21,500 ± 200

17,150 ± 60

23,200 ± 80

25,460 ± 90

17,540 ± 40

29,830 ± 30

383

394

388

385

402

431

384

394

389

384

402

380

394

386

360

401

431

384

394

387

385

437

475

0.42 ± 0.02

0.37 ± 0.01

0.41 ± 0.01

6750

7730

6910

5870

6370

6040

9930

7730

9500

9420

15,300

9800

7840

9550

(1.41 ± 0.05) × 10⁻

0.70 ± 0.03

0.74 ± 0.03

(1.9 ± 0.1) × 10⁻

(7.6 ± 0.2) × 10

(3.8 ± 0.1) × 10⁻

(6.8 ± 0.7) × 10⁻

(8.9 ± 0.4) × 10⁻

(7.0 ± 0.4) × 10

(8.2 ± 0.7) × 10⁻

(5.5 ± 0.3) × 10

(3.1 ± 0.5) × 10⁻

(7.8 ± 0.3) × 10⁻

(4.3 ± 0.4) × 10⁻

(7.3 ± 0.2) × 10

(4.9 ± 0.1) × 10

(5.8 ± 0.2) × 10

(3.4 ± 0.6) × 10

0.21 ± 0.01

522

527

528

516

530

520

519

520

502

−

7810

15,220

14,250

15,970

11,115

16,350

13,090

18,440

17,410

−2

−3

0.25 ± 0.02

Mean value ±standard deviation for at least three replicas. Obtained at 25 °C in ACN for L1−6, H1−5, and S1−6, and in ACN/THF 17:3 v/v

for B1−6. Determined as the angular coeﬃcient of λ versus concentration linear plots with the intercept set as zero (r > 0.99). Determined

abs

with 2,5-diphenyloxazole (Φ = 0.70) in EtOAc as the relative standard; for low-intensity emissions, H2 was used as the secondary standard.

−1

Calculated as the diﬀerence between λ and λ_F_Lwith both previously converted to cm . Determined for the CL reaction of H1−5 (1.0 mmol

abs

−

−1

L ) or S1−6 (0.1 mmol L ) in the presence of 1,8-diazabicyclo(5.4.0)undec-7-ene (DBU, 10 mmol L ) or TBAF ﬂuoride (1.0 mmol L ),

respectively, both in ACN. For B1−6, parameters λ , λ , and Δλ were determined directly from S1−6 solutions following the addition of TBAF.

abs FL

using this method for the preparation of substituted imidazoles

substituents increases conjugation in the structures of L5 and

L6, resulting in an approximation of their fundamental and

excited states that favors electron transitions.

23−25

for diﬀerent applications.

After puriﬁcation, these lophine

23−25,28,29

derivatives were converted to their hydroperoxides (H1−6,

Scheme 1a) by reacting with singlet oxygen generated by

The peroxides H1−5 and S1−6 exhibited blue-shifted λ

abs

photosensitization with methylene blue (Scheme 2b). This

(Figure S1c,e ), and no diﬀerence in λ compared to their

method is equivalent to the one used for the synthesis of alkyl-

corresponding lophines (Table 1 and Figure S1b,d,f). H1−5

26,27

−3

,2-dioxetanes from oleﬁns.

Finally, silylperoxides (S1−6,

showed lower Φ (in the 10 range) than L1−5 due to the

Scheme 1a) were prepared by protection of the H1−6

presence of the sp -hybridized carbon-4 bonded to the

hydroperoxyl group with tert-butyldimethylsilyl chloride

peroxidic group (Table 1). With this sp atom, the imidazole

(

TBDMSCl, Scheme 2c). Details for the preparation and

ring is no longer fully conjugated, and the phenyl rings become

characterization of the compounds are given in the

Experimental Section.

signiﬁcantly twisted and out-of-plane when compared to

lophines. Such molecules become less rigid, leading to an

increase in thermal deactivation of excited states, decreasing

The photophysical characterization of all derivatives of

lophine, hydroperoxide (except for H6 due to its unsuccessful

puriﬁcation), silylperoxide, and benzoylamidine was performed

in acetonitrile (ACN) to evaluate and compare their light

emission properties (Table 1). As expected, all lophines L1−6

the Φ_F_Lof these hydroperoxides.

As seen for L1−6, electron-withdrawing substituents also

shifted λ of H1−5 and S1−6 to longer wavelengths (Table 1

and Figure S1d,f) due to increased conjugation. The peroxides

−

presented high ﬂuorescent quantum yield (Φ > 0.3, Table

S1−6 displayed lower Φ values (in the 10 range) when

compared to their corresponding hydroperoxides (Table 1).

4,25,28

The derivative substituted with bromine L4, with

≈ 0.01, is an exception, probably a result of the heavy

Such lower Φ values can be attributed to an increased

atom eﬀect, increasing intersystem crossing and generating

number of vibrational modes associated to the organosilicon

9−31

nonemissive triplet states.

The lophine derivatives

group, which augments thermal deactivation of excited states.

containing the withdrawing substituents CF and CN (L5

CL spectra of H1−5 and S1−6 after addition of DBU (1,8-

diazabicyclo(5.4.0)undec-7-ene) and TBAF (TBAF ﬂuoride),

respectively, provided values of maximum CL emission

and L6, respectively) showed a bathochromic shift in their

maximum absorption (λ ) and ﬂuorescence (λ ) wave-

abs

lengths and higher Φ compared to other lophine derivatives

wavelengths (i.e., λ , Figure S1g,h ) that correspond to the

emission from the excited state of deprotonated benzoylami-

(

Table 1 and Figure S1a,b ). The presence of such withdrawing

635

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

after species B absorbs an excitation photon. Since there is no

Article

−

dines (B , Scheme 1c). The conjugate base B is the emitter of

report in the literature concerning ESIPT with benzoylami-

dines as substrates, this could be explored in the future,

particularly regarding the relevance of ESIPT with nitrogen as

the CL reaction of lophine derivatives, as originally noted in

previous reports.

nm range, Table 1) are expected, since all B derivatives are

fully conjugated species due to their anionic nature.

3−15

The longer λ_C_Lvalues (in the 502−530

−

proton donor. The benzoylamidine containing bromine as

,13,16

substituent (B4) presented the lowest Φ , as also noted for

Absorption and ﬂuorescence spectra were recorded after the

CL reaction between S1−6 with TBAF in order to obtain the

values of λ and λ for all ﬁnal CL reaction products (B1−6,

L4 (Table 1), in agreement with the generation of nonemissive

29−31

triplet excited states due to the heavy atom eﬀect.

Chemiluminescence Kinetics. General Remarks. We

studied the kinetics of the CL reaction of S1−6 with TBAF in

ACN/tetrahydrofuran (THF) 17:3 v/v and of H1−6 with

DBU in ACN using the peroxides under the pseudo-ﬁrst-order

condition. Kinetics were studied following light intensity over

time, and the registered proﬁles consisted of a rapid CL

increase upon addition of TBAF or DBU to a solution of S1−6

or H1−5, respectively, followed by a slower exponential decay

(Figure 1). Results obtained with S1−6 will be discussed prior

to H1−5.

abs

Table 1 and Figure S1i,j ). The benzoylamidines B1−6

exhibited lower values of λ compared to the λ of the

reaction (Table 1 and Figure S1j,h ), indicating that the emitter

B acquired a proton from the reaction medium after ground-

state formation, generating the less conjugated protonated

benzoylamidine (B, Scheme 3). In other words, B is the

−

Scheme 3. Acid−Base Equilibrium Involving the

Deprotonated Benzoylamidine (B ) and Its Conjugated

Acid (B) after Light Emission from the CL Reaction and

Possibility of Keto−Enol Tautomerization from B to an

Enol-Like Species (Enol-B)

−

Kinetics with Silylperoxides Using a Luminometer. CL

measurements with S1−6 were performed in a commercial test

tube luminometer, with TBAF stock solutions added to the

−

tube already containing S1−6 (1.27 μmol L ) through a set of

automatic injectors. CL intensity−time proﬁles (Figure 1)

were very reproducible, and the observed rate constants k

and k_o_b_s, as well as the maximum emission intensity (I_m_a_x

obs

)

were obtained ﬁtting the fall and rise in intensity using eqs 1

and 2 , respectively, for diﬀerent concentrations of TBAF

(

Tables S1 −S6).

conjugate base of a photoacid, but it is generated directly on

13−15

the excited state from the CL pathway.

The benzoylami-

I(t) = I_m_a_x(e⁽

−k_o_b_s

^t⁾)

(1)

(2)

dines B1−6 possess relatively high Stokes shift values (Δλ >

−

1,000 cm , Table 1) when compared to other studied

I(t) = I_m_a_x(1 − e ^k^o^b^s²^t⁾)

Observed rate constants k_o_b_s¹and k_o_b_s²are associated with

(

compounds, which is usually treated as evidence of excited-

24,33−35

state intramolecular proton transfer (ESIPT).

The

observed ﬂuorescence of B1−6 can occur from the

deactivation of the excited state of a keto-like tautomer and

an enol-like tautomer (enol-B, Scheme 3), the latter generated

with the intramolecular movement of the N-bonded proton

the ﬁrst and second steps of the CL mechanism (Scheme 4)

from the perspective that 1,2-dioxetane D is the intermediate

of a complex consecutive reaction. Other CL reactions have

analogous intensity −time proﬁles, where the rise (k_o_b_s) and

−

Figure 1. Examples of CL intensity−time proﬁles obtained with S4 (1.27 μmol L ) upon addition of TBAF (at 15 diﬀerent concentrations, from

−

.052 to 0.210 mmol L ) in ACN/THF 17:3 v/v at 25 °C, using a commercial tube luminometer. TBAF was added after 4 s of initiating data

acquisition.

636

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

Article

Scheme 4. Stepwise Mechanism for the CL of S1−6, with

Scheme 5. TSs Involved in the (a) Deprotection Reaction of

S1−6 by Fluoride (Scheme 4, Step 1) and (b) Generation of

the 1,2-Dioxetane HEI in the CL of S1−6 (Scheme 4, Step

−

Deprotection of the Silyl Group by Fluoride Generating H

−

(

Step 1) and Cyclization to the HEI D (Step 2)

)

These positive ρ values are consistent with negative charge

being developed in the TS of both stages (Scheme 4, steps 1

−

Decomposition of D generates B * (step 3) that decays to its B

ground state (step 4) in a process that is always faster than the

previous ones.

and 2, Scheme 5). The value of ρ < 1 for k_d_e_pshows that the

deprotection reaction (Scheme 4, step 1) is less sensitive to

substituents than the hydrolysis reaction of benzoic acid

fall (k_o_b_s¹) rate constants were attributed to the second and ﬁrst

methyl esters. In fact, the partial negative charge generated in

steps, respectively, of an intricate mechanistic se-

the peroxidic oxygen of the TS (Scheme 5a) is relatively far

from the substituent, not being so susceptible to its electronic

eﬀect. On the other hand, for the reaction of D formation

5,32,38−42

quence.

The bimolecular rate constant for the

−

deprotection of S1−6 by ﬂuoride (k , Scheme 4, step 1)

dep

was determined from a linear correlation between k and the

(Scheme 4, step 2), a negative charge is being created at the

heterocycle directly bonded to the substituted aromatic ring,

involving both nitrogen atoms in its delocalization (Scheme

5b). Substituents attached to the arylimidazolic system have a

pronounced electronic eﬀect on the stability of the TS of this

step (Scheme 4, step 2, Scheme 5b). The magnitude of ρ value

obtained for the cyclization reaction (+1.5) is a direct kinetic

evidence for the formation of a 1,2-dioxetane as a HEI in the

decomposition of this class of peroxides, related to

Radziszewski’s CL oxidation of L1 (Scheme 1b,c).

obs

concentration of TBAF (Figure S2 and Table 2). Although the

Table 2. Rate Constants for the Deprotection of S1−6 by

−

Fluoride Generating H (k ) and for the Cyclization of H

dep

−

Generating the 1,2-Dioxetane HEI D (k ), Within the

cyc

Studied CL Reaction Mechanism (Scheme 4)

cyc

dep

(× 10 L mol_s−1

−1

(× 10 L mol_s−1

−1

peroxide

R₁

σ_p

)

OMe

−0.27

−0.17

+0.23

+0.54

+0.66

1.30 ± 0.05

2.13 ± 0.04

2.38 ± 0.02

4.20 ± 0.10

5.80 ± 0.30

5.10 ± 0.06

0.76 ± 0.03

1.49 ± 0.03

1.88 ± 0.03

5.70 ± 0.09

9.70 ± 0.60

A linear dependence between k_o_b_sand TBAF concentration

can be observed but not within the entire range of studied

ﬂuoride concentration (Figure S3). For S1 and S3, the values

of k_o_b_stend to stabilize, reaching a maximum value of 2.1 and

^C^F3

−1

.7 s , respectively, at [TBAF] > 0.15 mmol L (Tables S1

and S3 and Figure S3 ). For S2, this trend is not so clear;

Values for the Hammett substituent constant (σ ) taken from ref 43.

although, for this derivative, the k

value reaches an average

obs

−

−1

of 1.8 s also above 1.5 mmol L of TBAF (Table S2 and

Figure S3). For S4 and S5, the k_o_b_svalue did not have a

signiﬁcant increase above 0.1 mmol L of TBAF, reaching

maximum values around 3.5 and 5.5 s , respectively (Tables

S4 and S5 and Figure S3 ). These maximum values for k_o_b_sare

−

−1

formation of D from H (Scheme 4, step 2) is a unimolecular

−

−1

process that does not involve ﬂuoride, the concentration of H

at a given time depends on the previous step.

29,37

Therefore,

the rate constant for the cyclization process and generation of

observed at TBAF concentrations in which the decomposition

of D (Scheme 4, step 3) becomes rate-limiting when

compared to its formation (Scheme 4, step 2). Thus, these

are the unimolecular rate constants associated with the 1,2-

−

D (k , Scheme 4, step 2) was obtained through the linear

cyc

dependence between k _o _b _s values and the concentration of

TBAF for S1−5 (Figure S3 and Table 2).

Derivatives with electron-withdrawing substituents (S4−6)

showed higher values for both rate constants, k and k

dioxetane decomposition (k ). The unimolecular decom-

udd

−

4,5,11

position of D , involving O−O and C−C cleavage,

faster for derivatives with electron-withdrawing substituents

(k_u_d_d= 5.5 s , D5 ) than the donor (k = 1.7 s , D3 ). The

lack of systematic dependence of k_o_b_swith [TBAF] for S6 may

dep

cyc

(Table 2). Such results indicate the generation of negative

−

−1

−

charge in the TSs of both reaction steps (Scheme 4, steps 1

and 2, Scheme 5). This is quantitatively assessed through linear

free-energy relationships using the Hammett substituent

udd

be related to the strong electron-withdrawing −CN group,

with rate constants being obtained close to the maximum limit

established by the unimolecular decomposition of the

constant (σ_p), obtaining the reaction constants ρ = +0.7

and +1.5 for plots with k_d_e_pand k , respectively (Figure 2).

cyc

637

The Journal of Organic Chemistry

J. Org. Chem. 2021, 86, 6633−6647

Article

Figure 2. Linear free-energy relationship with (a) k_d_e_pand (b) k_c_y_cusing the Hammett constant for para substitutions (σ ). From the angular

coeﬃcients, reaction constants ρ = +0.7 ± 0.1 and +1.5 ± 0.3 were determined for plots with k_d_e_pand k , respectively, both with fair linear

cyc

relationships (r > 0.9). Derivative S6 was not included in the correlation.

−

−1

corresponding D6 (k_u_d_dca. 4 and 9 s , Table S6 and Figure

S3 ).

diﬀerent TBAF concentrations (between 0.25 and 2.00 mmol

L ), using peroxides S1, S2, and S4 (Table S7). From the

−

Kinetics with Silylperoxides Using a Fluorescence

Spectrophotometer. We have also performed kinetic measure-

ments using a ﬂuorescence spectrophotometer, with a

ﬂuorescence quartz cuvette already containing a solution of

S1, S2, and S4 (0.025 mmol L ) in ACN, equipped with a

magnetic stirrer. Data acquisition was initiated after addition of

a TBAF stock solution in THF using a glass microsyringe. The

obtained intensity−time proﬁles (data not shown) are similar

to those registered with the luminometer (Figure 1); however,

only emission decay can be observed. Kinetics with the

ﬂuorescence spectrophotometer is not as reproducible as with

the luminometer, and the used silylperoxide concentration

must be 20 times higher due to the inherent lower sensibility of

the equipment.

linear dependence between such a decay rate constant and

TBAF concentration (Figure S4), values for k were obtained

dep

as (1.25 ± 0.04) × 10 , (2.74 ± 0.07) × 10 , and (3.4 ± 0.1) ×

−1 −1

10 L mol

for S2, S1, and S4, respectively. These

−

bimolecular k_d_e_pvalues are signiﬁcantly comparable to those

obtained with the luminometer (Table 2). Although the

deprotection step can be kinetically observed when a

ﬂuorescence spectrophotometer is utilized, the faster cycliza-

tion reaction is not accessible with this experimental setup.

The formation of the 1,2-dioxetane HEI was only observed

kinetically with the luminometer, as described in the previous

section.

The linear dependence between k_o_b_s¹and [TBAF], for S1,

S2, and S4 (Figure S4 ), reaches constant values at high

ﬂuoride concentrations, as previously observed for the

CL decay−time proﬁles obtained with the ﬂuorescence

spectrophotometer were ﬁtted to determine k

(eq 1), for

dependence between k_o_b_sand [TBAF] obtained using the

obs

638

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

Article

−

luminometer (Figure S3). At [TBAF] > 1.4 mmol L , the

water, the pK of organic hydroperoxides have been shown to

vary from 8 to 12, depending on their structure; nonetheless,

DBU seems to be basic enough to promote deprotonation of

−

decomposition reaction of D (Scheme 4, step 3) becomes

rate-limiting. This is observed for k_o_b_s¹only when using the

spectrophotometer since TBAF (i.e., the excess reagent) is

present at concentrations that are at least 1 order of magnitude

higher than those used in the luminometer. The k_u_d_dvalues

H1−5 as long as it occurs concomitant to peroxide cyclization.

−

Above 4 mmol L of DBU, the values of k

remain

obs

constant (Table S8 and Figure S6 ), as it was observed with S1,

S2, and S4 in the experiments conducted on the ﬂuorescence

spectrophotometer (Figure S4 ). The unimolecular rate

constant for the decomposition of the HEI, k_u_d_d, of the CL

reaction of H1−5 was also observed (Table 3), as well as for

−

obtained using the spectrophotometer (3.3, 2.1, and 4.0 s for

S1, S2, and S4, respectively) are in excellent agreement with

^t^h^e^kudd values determined using the luminometer (2.1, 1.8,

and 3.5 s for S1, S2, and S4, respectively). Thus, all these

results, obtained by kinetic assays performed on the

luminometer and ﬂuorescence spectrophotometer, support

the proposed overall mechanism for the CL decomposition of

S1−6 (Scheme 4).

Kinetics with Hydroperoxides Using a Fluorescence

Spectrophotometer. CL measurements with H1−5 were

performed using a commercial ﬂuorescence spectrophotometer

using a ﬂuorescence quartz cuvette already containing a

solution of H1−5 (0.20 mmol L ) in ACN, equipped with a

magnetic stirrer. Data acquisition was initiated after addition of

a stock solution of DBU in ACN using a glass microsyringe.

Other bases were used to trigger CL emission, such as pyridine

−

S1−6 in the experiments performed in the luminometer. k

udd

values for H1−5 (Table 3) are lower than those determined

−1

for S1−5 (2.1, 1.8, 1.7, 3.5, and 5.5 s , respectively), but the

systems are not exactly equivalent, particularly with respect to

solvation (S1−6 in ACN/THF 17:3 v/v and H1−5 in ACN).

Nevertheless, k is associated with the decomposition of the

udd

−

same HEI, 1,2-dioxetane D , regardless of the peroxide used to

generate it (Schemes 5b and 6). Although only the

chemiexcitation process is being represented here, the

−

conversion of D to B , involving O−O and C−C cleavage

4,5,11

(

Scheme 4, steps 3 and 4),

may or may not occur with the

−

formation of excited states (i.e., B *). Therefore, k cannot

udd

(

pK_a_H= 5.25), imidazole (pK = 6.95), 4-dimethylami-

46 46

be interpreted solely as the chemiexcitation rate constant; as a

matter of fact, it incorporates any other process associated with

the decomposition of the HEI.

nopyridine (pK_a_H= 9.20), triethylamine (pK = 10.75),

and pyrrolidine (pK = 11.27); however, only DBU (p K _a _H

3.5) aﬀorded reproducible intensity−time proﬁles (Figure

The eﬀect of substituent nature on k_u_d_dis clear, with

electron-withdrawing groups leading to higher rate constant

values (Table 3); this was observed for the S1−6 system,

although not with such a signiﬁcant trend (see previous

sections). The reaction constant ρ = +1.3 ± 0.1 was obtained

from the Hammett plot for k_u_d_d(Figure 3b), indicating that a

negative charge is associated with the TS and that this system

is more sensitive to substituents than benzoic acid deriva-

S5).

Emission decay kinetic proﬁles obtained for H1−5 at DBU

−

concentrations between 1.0 and 10.0 mmol L were ﬁtted to

determine the k_o_b_s¹values (eq 1 and Table S8). A linear

obs

dependence between k

and [DBU] was observed until 4.0

mmol L of DBU for all hydroperoxide derivatives H1−5

Figure S6). From these linear ﬁttings, a bimolecular rate

−

constant was determined (Table 3, k_D_B_U), being at least 2

tives. This information is going to be used in the

interpretation of excitation yields and in the rationalization

of the chemiexcitation mechanism discussed in the following

sections.

Table 3. Rate Constants for the Bimolecular DBU-Induced

CL of H1-5 (k_D_B_U) and for the Unimolecular 1,2-Dioxetane

Decomposition (k_u_d_d

)

Chemiluminescence and Excited State Formation

Yields. Intensity-time proﬁles obtained for S1−6 and H1−5

at diﬀerent concentrations of TBAF and DBU, respectively,

were integrated and converted to chemiluminescence quantum

yields (Φ ); see Experimental Section for details. Φ was

^kDBU (× 10 L mol s⁻¹

−1

^kudd (s⁻¹

)

peroxide

^R1

OMe

^σp

)

−0.27

−0.17

+0.23

+0.54

0.26 ± 0.02

0.40 ± 0.01

0.77 ± 0.03

1.64 ± 0.06

3.06 ± 0.08

0.25 ± 0.01

0.34 ± 0.01

0.54 ± 0.03

1.21 ± 0.01

1.68 ± 0.04

CF₃

independent of the ﬂuoride or base concentration, and mean

values of this yield were obtained for each peroxide derivative

(Table 4). Therefore, regardless of the concentration of the

used excess reagent (i.e., TBAF or DBU), the light emission

eﬃciency depended exclusively on the amount of limiting

peroxide available (i.e., S1−6 or H1−5). This is a relevant

information, considering that higher concentrations of ﬂuoride

or base could have catalyzed the decomposition of peroxides

through alternate “dark” routes. The peroxyoxalate reaction is

an example of a CL system that has its emission eﬃciency

decreased drastically with increasing concentration of base due

Values for the Hammett substituent constant (σ ) taken from ref 43.

orders of magnitude lower than k_d_e_pand k_c_y_cfor the reaction of

S1−6 with ﬂuoride (Table 2). Indeed, these results are in

accordance with the empirical observation of longer reaction

times for CL kinetics with peroxides H1−5 than with S1−6.

The reaction constant ρ = +1.2 ± 0.1 was obtained through

a Hammett plot with k_D_B_U(Figure 3a). The positive signal

and magnitude (>1) of such a ρ value indicates that a negative

charge is being generated on the TS and is susceptible to the

electronic nature of substituents, as with k_c_y_cfor the cyclization

,41,42,50

to “dark” HEI decomposition.

Considering all six peroxides S1−6, the average for Φ is

−

−6

−1

of intermediate H generated from S1−6 (Scheme 5b). This

result suggests general base catalysis by DBU, with the

abstraction of a proton coupled to cyclization and 1,2-

dioxetane formation (Scheme 6). Compared to the general

mechanism proposed for the CL of S1−6 (Scheme 4), for

H1−5, the reaction pathway would be slightly diﬀerent with

steps 1 and 2 combined on a single concerted process. In

(3.7 ± 0.7) × 10 E mol (Table 4). This value is in the

same order of magnitude than the ΦCL for the chemilumi-

nescence of (R)-S12 and (R)-S13 (Scheme 1a) in acetone

triggered by TBAF. White and Harding reported that upon

treatment with potassium superoxide in EtOH, the ΦCL for L1

−

−7

−1

is 10 to 10 E mol , while for L8 (Scheme 1a), the Φ

−

−4

−1 13

is10 to 10 E mol . The CL reaction of silylperoxides

639

The Journal of Organic Chemistry

J. Org. Chem. 2021, 86, 6633−6647

Article

Figure 3. Linear free-energy relationship for (a) k_D_B_Uand (b) k_u_d_dusing the Hammett constant for para substitutions (σ ). From the angular

coeﬃcients, reaction constants ρ = +1.2 ± 0.1 and +1.3 ± 0.1 were determined for plots with k_D_B_Uand k_u_d_d, respectively, both with fair linear

relationships (r > 0.9).

Scheme 6. TS Involved in the Formation of the 1,2-

Dioxetane HEI on the CL Reaction Mechanism of H1−5

With DBU

process eﬃciency (Scheme 4, steps 3 and 4), and this yield

−

−5

−1

goes from 1.1 × 10 to 7.8 × 10 E mol for S1−3, S5, and

S6 (Table 4). Φ values for H1−3 and H5 are close to those

determined for their corresponding silylperoxides S1−3 and S5

(

Table 4). Although there is no absolute correlation between

the chemiexcitation eﬃciency and the electronic nature of the

substituents, it was observed that derivatives containing

electron-donor substituents showed higher Φ values. This

was also observed by White and Harding, and Kimura et al.

consciously used a −NMe substituent on the 2-phenyl group

at H7−11 (Scheme 1a) to achieve higher Φ and Φ values.

The yield for singlet state generation for S1−6 and H1−5 is

low, compared to other CL systems that possess Φ close to

triggered by ﬂuoride has low eﬃciency when compared to

,5,25,32

other light-emitting systems.

,51,52

The singlet excited-state formation quantum yield (Φ ) can

unity.

be determined by correcting Φ by the ﬂuorescence quantum

Nonetheless, peroxides S4 and H4 presented Φ = 1.20 ×

−

−3 −1

yield (Φ ) of the emitter B ; see Experimental Section for

10 E mol , 2 orders of magnitude higher when compared to

peroxides containing electron-withdrawing substituents (Table

details. The Φ value is a direct measure of the chemiexcitation

640

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

Article

Table 4. Chemiluminescence Quantum Yield (Φ ) and

4). This may be understood as evidence for the formation of

Singlet Excited State Formation Quantum Yield (Φ ) for

additional singlet excited states (i.e., S ) from triplet states with

the CL Decomposition Reaction of S1−6 and H1−5

Induced by Fluoride and DBU, Respectively

the bromine substituent favoring reverse intersystem crossing

29,53,54

(RISC).

It is known that lower-energy triplet states of π/

π* nature can be produced on the decomposition of aryl-

Φ_C_L(× 10⁻E mol⁻¹

)

Φ (× 10 E mol⁻¹

−5

)

peroxide

R₁

substituted 1,2-dioxetanes generating the corresponding

OMe

3.8 ± 0.2

3.9 ± 0.2

4.7 ± 0.3

4.2 ± 0.7

2.9 ± 0.6

2.8 ± 0.3

7.1 ± 0.2

5.7 ± 0.3

8.5 ± 0.7

3.9 ± 0.6

3.6 ± 0.7

7.8 ± 0.4

6.7 ± 0.4

6.4 ± 0.4

120 ± 20

1.4 ± 0.3

1.1 ± 0.5

14.5 ± 0.4

9.8 ± 0.5

11.6 ± 0.9

120 ± 20

1.7 ± 0.3

carbonyl residues, in addition to the generally observed triplet

11,55−58

n/π* states.

In our case, a possible lower π/π* triplet

−

state of the aryl-substituted B * emitter containing a bromine

group could enable RISC to the S and explain the increased

CF₃

OMe

Φ for S4 and H4. This hypothesis suggests that the yield of

generation of overall triplet states (Φ ) is higher than Φ for

both peroxidic systems (S1−6 and H1−5); however, in the

absence of a bromine substituent, RISC cannot occur and an

increase in singlet excites states is not evident. Recently, it has

been shown that triplet excited states are likely the emitters in

^C^F3

the solid-state thermochemiluminescence of H1, supporting

−1

For reactions with [S1 −6 ] = 1.27 μmol L and diﬀerent TBAF

our proposition that in solution, (nonemissive) triplet states

could also be generated. However, our results are not

conclusive, and additional experimental and theoretical data

are required to verify the occurrence of triplet excited states on

the decomposition of S1−6 and H1−5. Moreover, the possible

concentrations (see Tables S1 −S6 for details). For reactions with

−1

[

H1−5] = 0.2 mmol L and DBU concentrations from 1.0 to 10.0

−1

mmol L .

Scheme 7. Possible Mechanisms for the Chemiexcitation Step During the Decomposition of a 1,2-Dioxetane HEI in the

Lophine Peroxide CL Reaction from Kimura et al., Lu et al., and This Work

Asterisk at the top of the benzoylamidine species indicates an electronic excited state, whether it is singlet or triplet.

641

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

Article

occurrence of RISC with S4 and H4 should also be further

investigated.

biradical anions that have the formal negative charge and the

unpaired electron in diﬀerent carbonyl groups. The EBT

generates two diﬀerent resonance structures for the same

species in the excited state, regardless of the path that is taking

place (Scheme 7). Regarding the hydroperoxides H7−11

Possible Chemiexcitation Mechanisms for the De-

composition of Lophine Peroxides. Regardless of the

originating peroxide, S1−6 or H1−5, the corresponding 1,2-

−

dioxetane D is generated as the HEI. Considering our results

studied by Kimura et al., the presence of the strong electron-

−

discussed above, we can assume that the D decomposition

donating group −NMe on the 2-phenyl ring may promote

follows the chemiexcitation mechanism proposed for the

thermal decomposition of 1,2-dioxetanes and 1,2-dioxeta-

chemiexcitation mainly through the intramolecular CIEEL

−

sequence, justifying the Φ between 0.11 and 0.79 E mol

,5,60,61

−2

nones.

Such mechanism have shown Φ < 10

reported by these authors.

−

mol , where the peroxidic bond cleavage enters a biradical

region associated with triplet states.

With respect to our results, the Φ values for the

,5,59,60

This hypothesis

silylperoxides and hydroperoxides with electron-donating

substituents are higher than the ones with electron-with-

drawing groups (Table 4). For example, with S2 and its

could be better supported by future computational simulation

studies. For a while, our strongest evidence is the increased Φ

value observed for the bromine substituted S4 and H4, when

compared to other derivatives. For the uncatalyzed decom-

position of 1,2-dioxetanes, it has been determined exper-

methoxy group, Φ is up to 5.5 and 7-times higher than the

values for S5 and S6, with their corresponding CF and CN

substituents. With H2 and H5, comprising the OMe and CN

imentally that excited carbonyl residues are formed with Φ ca.

substituents, Φ is 8.5-times higher for the former. This

−1

4,11,62,63

.5 E mol .

The determination of the actual Φ values

suggests that the CIEEL contribution to the overall

−

for the CL of S1−6 and H1−5 could be performed, but an

adequate experimental approach has not yet been proposed. If

the role of triplet states is conﬁrmed, this should be considered

in the future use of lophine silylperoxides as sensors for

decomposition of D increases with increased electron-

donating ability of the substituent at the 2-phenyl ring,

which in turn facilitates intramolecular electron transfer and

,11,69−71

O−O bond cleavage (Scheme 7).

Thus, our reasoning

biological systems. The conjugation of a lophine silylperoxide

suggests that with S1−6 and H1−5, the transformation would

still occur mainly through a mechanism that resembles the

thermal decomposition of alkyl-1,2-dioxetanes and 1,2-

to a peptide or protein would allow the use of this class of

peroxides as CL probes for labeling biomolecules. While

singlet excited states are important for analytical and labeling

,5,11,59−63

dioxetanones

but with the CIEEL sequence becom-

,64

purposes,

triplet states generated in situ could be used for

ing slightly more relevant with electron-donating substituents

at the 2-phenyl ring.

In this intramolecular CIEEL landscape, the diﬀerence

between the chemiexcitation eﬃciencies for hydro- and

silylperoxides bearing the same substituent (e.g., H2 and S2)

assessing biological oxidation through chemiexcitation instead

of light absorption, also known as “dark” photochemistry.

Kimura et al. postulated that the HEI produced from H7−

1 decomposes through a [2 + 2] cycloelimination reaction,

generating a biradical anion intermediate (Scheme 7). They

proposed that excited states are formed by an intramolecular

electron reorganization (or electron back-transfer, EBT)

between donor and acceptor moieties of the molecule, leading

must be accounted for. Consistently, the Φ value for the

silylperoxide is smaller than the one for the corresponding

hydroperoxide (Table 4), even if the diﬀerence is not that

,11

signiﬁcant for yields of this nature.

For a given substituent,

−

to excited states. With H7−11, the formation of a biradical

the same HEI cyclic peroxide D is going to be generated,

regardless of the starting peroxide (Schemes 5b and 6). One

must remember that the solvating media for the CL

transformation with S1−6 was ACN/THF 17:3 v/v, while

pure ACN was used with H1−5. Adam, Matsumoto, and

Troﬁmov proposed that an increase in viscosity prevents C−C

bond rotation, frictionally reducing ground-state formation of

anion intermediate prior to excited states was explicitly

addressed. Such a view is a bit diﬀerent from the fully

concerted [2 + 2] cycloelimination reaction that directly

produces triplet states in the thermal decomposition of 1,2-

,11,59−63

dioxetanes.

Lu et al., when discussing that a partially cyclic TS may be

involved on the chemiexcitation process for CL with (R)-S12

and (R)-S13, considered that bond cleavage and formation

should occur simultaneously for the generation of excited

states (Scheme 7) in a view that is more equivalent to the fully

molecules and increasing chemiexcitation yields. This

dependence of Φ in viscosity has been debated more

,73,74

recently,

and it seems that chemiexcitation in the

intramolecular system is compromised when conformational

,11,59−63

concerted thermal decomposition of 1,2-dioxetanes.

changes compete with the concerted EBT and C−C bond

Their conclusion is supported by our observation that triplet

states are generated in higher yields than singlet states, as seen

cleavage. Thus, it is hypothesized that the observed

diﬀerences in Φ values for hydro- and silylperoxides bearing

by the increased Φ value observed for the bromine induced

the same substituent are actually related to the inﬂuence of a

RISC (Table 4).

solvent-cage eﬀect on the overall EBT eﬃciency and not to

One possibility that has not been formally considered so far

is that chemiexcitation on the decomposition of D may

the starting peroxide structure. Our interpretation could be

better supported by additional experimental data, employing

−

happen via the intramolecular chemically initiated electron

diﬀerent mixtures of benzene/diphenylmethane as reaction

66−68

8,72

exchange luminescence (CIEEL) mechanism.

The intra-

media.

This binary solvent system allows for diﬀerent

molecular CIEEL mechanism is used to rationalize the induced

viscosities without signiﬁcantly aﬀecting the solvent reorgan-

decomposition of aryl-1,2-dioxetanes that have high Φ

ization energy and has been successfully applied for both

−169,70

8,72

73,74

values,

−1

reported as 0.6 E mol

and even 1.0 E

intramolecular

and intermolecular CIEEL systems.

mol . In our proposition (Scheme 7), an intramolecular

electron transfer induces O−O bond cleavage and generation

of a biradical anion. Then, C−C bond cleavage can occur

through two diﬀerent paths (Scheme 7A,B), leading to new

So far, for the rationalization of the CL reaction with

peroxides derived from lophine derivatives, it is suggested that

electron-withdrawing groups on the 2-phenyl ring increase the

rate of formation of the 1,2-dioxetane HEI (Schemes 5b and

642

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

). They also increase the rate of decomposition of the cyclic

peroxide (conversion of D to B , Scheme 4, step 3), which

may occur through a concerted [2 + 2] cycloelimination step

Article

EXPERIMENTAL SECTION

■

−

General Section. All chemicals used in this work were purchased

from commercial sources as analytical-grade reagents and used

without further puriﬁcation. Deionized water was obtained through a

Milli-Q Millipore purifying system (conductivity 18.2 MΩ cm).

Quartz cuvettes (Hellma, QS Suprasil) had a volume of 3.0 mL and

an optical path of 10.00 mm, with two polished sides for absorption

and four sides for ﬂuorescence or chemiluminescence assays. Small

sample volumes (μL) were transferred using Hamilton glass

microsyringes. All the reported CL parameters were obtained as

mean value ± standard deviation for at least three replicas at the same

experimental condition. The purity of the peroxides used in kinetic

(

Scheme 7). However, with increasing electron-donating

ability of the substituent, the intramolecular CIEEL pathway

Scheme 7) becomes more evident. As discussed above, this is

(

not observed kinetically but can be inferred based on the

increased singlet excited-state quantum yields of derivatives

with electron-donating groups (Table 4). The CIEEL

mechanism was originally devised to explain the eﬃcient

6−68

formation of excited states in intermolecular systems;

however, further investigation

52,75

showed that this process is

measurements (over 97% in mass) was determined by iodometry

−

actually orders of magnitudes less eﬃcient than previously

using the method of spectrophotometric quantiﬁcation of I reported

0−52,81

reported. An alternative charge transfer induced luminescence

elsewhere.

,75−77

Equipment. H and ¹³C NMR spectra were recorded at 25 °C

using a 500 MHz Varian or 300 MHz Bruker DPX spectrometer.

Elemental analyses were carried out on a Thermo Scientiﬁc Flash EA

mechanism

was proposed to rationalize the low

eﬃciency of singlet excited state formation in intermolecular

systems, although the CIEEL mechanism is still successful to

explain the high chemiexcitation eﬃciency for many intra-

112 analyzer. For UV−vis absorption and ﬂuorescence emission

−7,52,69−71

assays, Varian Cary 60 and Varian Cary Eclipse spectrophotometers

were used, respectively; both equipment cell holders were maintained

at 25 °C using a Varian Cary PCB 1500 system. Chemiluminescence

measurements were performed using the spectrophotometer men-

tioned above with the emission monochromator set at the mirror

position and with the excitation lamp shut oﬀ. For fast CL kinetic

measurements, a Berthold Sirius single-tube luminometer with a set of

automatic injectors was used.

molecular systems.

Nevertheless, research studies

have reported that electron and/or charge transfer and back-

transfer are not necessarily involved on the highly eﬃcient

78−80

imidazopyrazinone-based chemi-/bioluminescence.

For

this system, eﬃcient chemiexcitation occurs when the

decomposition of the 1,2-dioxetanone HEI is delayed, allowing

the molecule to spend more time on a region of the potential

energy surface where ground and excited states are

degenerated, thus augmenting the possibility of transition

Lophine L1−6 Preparation. Based on the method of Benisvy et

al., as previously reported by our group,

23−25

for the preparation of

79,80

lophine derivatives, a 50 mL round-bottom ﬂask was charged with 2.7

g (12.9 mmol) of benzil, 7.4 g (96 mmol) of ammonium acetate, 50

mL of glacial acetic acid, and 13.2 mmol of the corresponding

substituted benzaldehyde. The mixture was kept under reﬂux and

magnetic stirring for 4 h, using an oil bath for heating, and afterward,

100 mL of ice-cold deionized water was added to induce precipitation.

The precipitate was ﬁltered out under vacuum, washed with deionized

water (5 × 15 mL), and dried under suction. The crude product was

between electronic states.

Concerning our results with

S1−6 and H1−5, we have observed that electron-donating

−

substituents decrease the rate of decomposition of B to D

involved in the chemiexcitation step (Scheme 4, step 3). Thus,

alternatively to the CIEEL process, this rationale could explain

the increased Φ_Svalues observed for electron-donating

substituted hydro- and silylperoxides (Table 4). As it can be

seen, there are plenty of opportunities to continue the

investigation on the CL decomposition of peroxides derived

from lophine derivatives, both experimentally and theoretically.

dissolved in 300 mL of EtOAc, dried under MgSO , ﬁltered,

concentrated through rotary evaporation, and dried under vacuum

−

(

10 mbar, 30 min). Puriﬁcation was attained using recrystallization

from hot EtOAc or CH Cl /hexane. The obtained cotton wool-like

crystals were slightly yellow, being ﬁltered under vacuum, washed with

cold pentane (2 × 50 mL), and dried under vacuum (10 mbar, 30

−

CONCLUSIONS

■

min).

In summary, we have provided evidence that a 1,2-dioxetane

,4,5-Triphenylimidazole (L1). 1.98 g (52%), slightly yellow cotton

−

is generated as a HEI on the CL of lophines hydro-

wool-like crystals, R = 0.53 (SiO , Hex/EtOAc 3:1). Anal. Calcd for

peroxides (Scheme 6) and silylperoxides (Scheme 5b). At high

ﬂuoride or DBU concentrations, the unimolecular decom-

position of this HEI becomes rate-limiting. Electron-with-

C H N : C, 85.1; H, 5.4; N, 9.4. Found: C, 84.5; H, 5.4; N, 9.3. H

NMR (DMSO-d , 300 MHz): δ 12.71 (s, 1H), 8.13−8.09 (m, 2H),

7.55−7.23 (m, 13H). C{ H} NMR (DMSO-d , 75 MHz): δ 145.6,

−

drawing groups increase both the rate of 1,2-dioxetane D

137.1, 135.2, 131.1, 130.4, 128.7, 128.5, 128.3, 127.8, 127.1, 126.5,

25.2.

-(4-Methoxyphenyl)-4,5-diphenylimidazole (L2). 2.77 g (66%),

slightly yellow cotton wool-like crystals, R = 0.67 (SiO , Hex/EtOAc

formation and decomposition, while the latter may occur

through a concerted [2 + 2] cycloelimination. With stronger

electron-donating groups at the 2-phenyl ring, a higher

:1). Anal. Calcd for C H N O: C, 80.9; H, 5.6; N, 8.6. Found: C,

22 18 2

contribution of the intramolecular CIEEL mechanism at the

0.9; H, 5.5; N, 8.5. H NMR (DMSO-d , 300 MHz): δ 12.52 (s,

−

decomposition of D may be responsible for the increased Φ

1H), 8.03 (d, 2H, J = 8.8 Hz), 7.56−7.22 (m, 10H), 7.05 (d, 2H, J =

values that were observed, although there are other possible

mechanistic alternatives. Nevertheless, additional experiments

and theoretical simulations are required to rationalize excited

state formation with peroxides derived from lophine and its

13 1

.9 Hz), 3.82 (s, 3H). C{ H} NMR (DMSO-d , 75 MHz): δ 159.4,

145.6, 136.8, 135.3, 131.2, 128.6, 128.4, 128.2, 127.6, 127.1, 126.7,

126.4, 123.1, 114.1, 55.2.

4,5-Diphenyl-2-(p-tolyl)imidazole (L3). 1.35 g (34%), slightly

yellow cotton wool-like crystals, R = 0.58 (SiO , Hex/EtOAc 3:1).

−

−1

derivatives. The Φ = 1.20 × 10 E mol observed with

Anal. Calcd for C H N : C, 85.1; H, 5.8; N, 9.0. Found: C, 83.6; H,

bromine-substituted derivatives S4 and H4 may be possibly

related to the presence of triplet states and RISC, although this

is still under investigation. If indeed conﬁrmed, the possibility

of modulation in excited-state multiplicity could be explored in

the future, whether there is interest in light emission or “dark”

photochemistry induced by the decomposition of hydro-

peroxides or silylperoxides.

.6; N, 9.5. H NMR (CDCl , 500 MHz): δ 7.79 (d, 2H, J = 8.2 Hz),

.53 (d, 4H, J = 7.1 Hz), 7.34−7.26 (m, 6H), 7.23 (d, 2H, J = 7.9

Hz), 2.38 (s, 3H). C{ H} NMR (CDCl , 125 MHz): δ 146.4, 138.9,

29.7, 128.7, 128.0, 127.5, 127.2, 125.4, 21.5.

-(4-Bromophenyl)-4,5-diphenylimidazole (L4). 4.5 g (93%),

slightly yellow cotton wool-like crystals, R = 0.53 (SiO , Hex/

EtOAc 3:1). Anal. Calcd for C H N Br: C, 67.2; H, 4.0; Br, 21.3; N,

643

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

Article

.5. Found: C, 67.9; H, 4.1; Br, 21.4; N, 7.4. H NMR (DMSO-d ,

(m, 2H), 7.49−7.47 (m, 2H), 7.38−7.32 (m, 5H), 1.62 (s, 1H).

13 1

00 MHz): δ 12.80 (s, 1H), 8.07−8.02 (m, 2H), 7.71−7.67 (m, 2H),

.53 (d, 4H, J = 6.9 Hz), 7.37 (s, 6H). C{ H} NMR (DMSO-d , 75

MHz): δ 144.5, 131.7, 129.5, 128.5, 127.1, 121.4.

,5-Diphenyl-2-(4-(triﬂuoromethyl)phenyl)imidazole (L5). 2.1 g

45%), slightly yellow cotton wool-like crystals, R = 0.65 (SiO , Hex/

EtOAc 3:1). Anal. Calcd for C H F N : C, 72.5; H, 4.1; N, 7.7.

Found: C, 71.8; H, 3.8; N, 8.1. H NMR (CDCl , 500 MHz): δ 8.03

C{ H} NMR (CDCl , 125 MHz): δ 191.7, 171.7, 134.4, 133.6,

131.8, 131.3, 130.6, 130.4, 129.8, 129.0, 129.0, 128.8, 127.8, 126.8,

15.7.

-Hydroperoxy-4,5-diphenyl-2-(4-(triﬂuoromethyl)phenyl)-4-imi-

dazole (H5). 0.184 g (85%), colorless crystals, R = 0.78 (SiO , Hex/

(

EtOAc 3:1). Anal. Calcd for C H F N O : C, 66.7; H, 3.8; N, 7.1.

22 15 3 2 2

Found: C, 69.5; H, 4.8; N, 6.9. H NMR (CDCl , 500 MHz): δ 8.41−

(

d, 2H, J = 8.1 Hz), 7.70 (ddd, 2H, J = 8.1, 4.4, 2.0 Hz), 7.57−7.52

8.39 (m, 2H), 7.97 (d, 2H, J = 8.1 Hz), 7.66−7.63 (m, 1H), 7.57−

7.53 (m, 2H), 7.50−7.48 (m, 2H), 7.44 (d, 2H, J = 8.2 Hz), 7.37−

m, 4H), 7.34 (dt, 6H, J = 13.6, 7.7 Hz). ¹³C{ H} NMR (CDCl , 125

MHz): δ 168.0, 144.5, 133.1, 131.0, 128.8, 128.0, 126.0, 126.0, 125.4.

-(4,5-Diphenylimidazol-2-yl)benzonitrile (L6). 2.60 g (62%),

slightly yellow cotton wool-like crystals, R = 0.23 (SiO , Hex/

EtOAc 4:1). Anal. Calcd for C H N : C, 82.2; H, 4.7; N, 13.1.

Found: C, 81.2; H, 5.5; N, 11.2. H NMR (DMSO-d , 500 MHz): δ

.33 (m, 3H), 1.59 (s, 1H). C{ H} NMR (CDCl , 125 MHz): δ

192.2, 171.3, 134.1, 133.9, 133.2, 130.5, 130.1, 129.9, 129.4, 129.1,

128.5, 126.8, 125.5, 125.5, 115.9.

Silylperoxide S1−6 Preparation. Based on the method of Lu et

al., a 25 mL round-bottom ﬂask was charged with 10 mL of CH Cl ,

3.00 (1 H, s), 8.25 (d, 2H, J = 8.4 Hz), 7.93 (d, 2H, J = 8.4 Hz),

H1-6 (0.92 mmol), imidazole (4.6 mmol), and TBDMSCl (3.7

mmol). The mixture was kept under vigorous magnetic stirring at

room temperature for 1 h. The reaction was followed using thin-layer

chromatography (20:1 hexane/EtOAc), and the presence of S1−6

.58−7.20 (m, 10H). C{ H} NMR (DMSO-d , 125 MHz): δ 143.7,

38.1, 134.7, 134.3, 132.8, 130.6, 129.6, 128.7, 128.5, 128.3, 128.2,

27.1, 126.8, 125.5, 118.9, 110.1.

Hydroperoxide H1−6 Preparation. Based on the method of Lu

et al., a 100 mL Schlenk reaction tube was charged with 80 mL of a

:1 CH Cl /ACN mixture and 0.61 mmol of L1-6. A catalytic amount

−1

was tested using a TBAF solution in THF (1.0 mol L ), which

caused a rapid light ﬂash on the thin-layer chromatography spot

containing the peroxide. After that, the solution was ﬁltered and the

CH Cl removed by rotary evaporation. The obtained solid was

2 2

of methylene blue was added, the mixture was kept under magnetic

stirring, and molecular oxygen was bubbled using a glass tube with a

fritted glass tip. The solution was irradiated under a 665 nm light-

emitting diode array, while the temperature was maintained at 4 °C

using a water−ice bath. The singlet oxygen produced through

photosensitization reacted with the imidazolyl moiety of L1-6 by an

ene reaction, producing hydroperoxides H1−6. The reaction was

followed using thin-layer chromatography, and the organic peroxide

was revealed with a potassium iodide solution in acetic acid. After

completion, the reaction mixture was ﬁltered through a 2.5 × 10 cm

silica gel layer and eluted with CH Cl . Hydroperoxides were puriﬁed

solubilized with hexane and washed with an equal volume of

deionized water (5×) to remove the residual salt. To the organic

layer, MgSO was added, the mixture was ﬁltered, and the solvent

−3

removed by rotary evaporation and dried under vacuum (10 mbar,

0 min). Silylperoxides S1−6 were puriﬁed by ﬂash column

chromatography (SiO , hexane/EtOAc 9:1), and after solvent

−

removal, the clear colorless crystals were dried under vacuum (10

mbar, 90 min).

-((tert-Butyldimethylsilyl)peroxy)-2,4,5-triphenyl-4-imidazole

(

S1). 0.190 g (26%), colorless crystals, R = 0.61 (SiO , Hex/EtOAc

through recrystallization from CH Cl /pentane, ﬁltered under

:1). Anal. Calcd for C H N O Si: C, 73.3; H, 6.8; N, 6.3. Found:

27 30 2 2

vacuum using a fritted glass Buchner funnel, and washed with cold

C, 74.3; H, 6.9; N, 6.6. H NMR (CDCl , 500 MHz): δ 8.50−8.48

pentane (2 × 50 mL). The obtained colorless cotton-like crystals were

_d_r_i_e_d_u_n_d_e_r_v_a_c_u_u_m_f_o_r₃_h₍₁₀−3 mbar). We were unable to purify

(m, 2H), 8.24 (ddd, 2H, J = 7.3, 3.0, 1.6 Hz), 7.59−7.50 (m, 4H),

.46−7.43 (m, 2H), 7.35−7.31 (m, 2H), 7.30−7.27 (m, 3H), 0.85 (s,

H6 adequately for characterization, even after many recrystallization

cycles; nonetheless, H6 was still used for the preparation of S6.

-Hydroperoxy-2,4,5-Triphenyl-4-Imidazole (H1). 0.140 g (63%),

colorless crystals, R = 0.41 (SiO , Hex/EtOAc 3:1). Anal. Calcd for

13 1

H), 0.21 (s, 3H), 0.16 (s, 3H). C{ H} NMR (CDCl , 125 MHz):

δ 191.3, 171.6, 134.8, 132.5, 132.0, 131.6, 131.1, 130.0, 129.8, 129.2,

28.8, 128.7, 128.6, 126.2, 115.2, 26.2, 18.5, −5.3, −5.4.

-((tert-Butyldimethylsilyl)peroxy)-2-(4-methoxyphenyl)-4,5-di-

phenyl-4-imidazole (S2). 0.113 g (26%), colorless crystals, R = 0.36

SiO , Hex/EtOAc 9:1). Anal. Calcd for C H N O Si: C, 71.1; H,

C H N O : C, 76.8; H, 4.9; N, 8.5. Found: C, 73.1; H, 4.8; N, 7.6.

H NMR (CDCl , 500 MHz): δ 8,41−8.39 (m, 2H), 7.92−7.91 (m,

(

28 32

(

H), 7.60 (ddd, 1H, J = 6.7, 3.8, 1.1 Hz), 7.55−7.50 (m, 4H), 7.35−

13 1

6.8; N, 5.9. Found: C, 71.9; H, 6.8; N, 6.0. H NMR (CDCl

MHz): δ 8.46 (d, 2H, J = 4.4 Hz), 8.23−8.21 (m, 2H), 7.53−7.50 (m,

H), 7.45−7.42 (m, 2H), 7.33−7.27 (m, 5H), 7.03 (d, 2H, J = 8.5

, 500

.32 (m, 4H), 7.2 (t, 2H, J = 7.8 Hz), 1.66 (s, 1H). C{ H} NMR

CDCl , 125 MHz): δ 191.4, 172.5, 134.7, 133.3, 132.3, 130.8, 130.3,

30.0, 129.9, 129.6, 128.9, 128.9, 128.5, 126.8, 115.5.

-Hydroperoxy-2-(4-methoxyphenyl)-4,5-diphenyl-4-imidazole

H2). 0.157 g (71%), colorless crystals, R = 0.23 (SiO , Hex/EtOAc

Hz), 3.90 (s, 3H), 0.85 (s, 9H), 0.21 (s, 3H), 0.16 (s, 3H). C{ H}

NMR (CDCl , 125 MHz): δ 192.3, 190.8, 171.2, 135.1, 132.5, 132.0,

(

−

31.1, 129.8, 129.1, 128.8, 128.6, 126.1, 114.1, 55.6, 26.2, 18.4, −5.3,

5.3.

-((tert-Butyldimethylsilyl)peroxy)-4,5-diphenyl-2-(p-tolyl)-4-imi-

dazole (S3). 0.128 g (30%), colorless crystals, R = 0.57 (SiO , Hex/

:1). Anal. Calcd for C H N O : C, 73.7; H, 5.1; N, 7.8. Found: C,

22 18 2 3

3.5; H, 5.1; N, 7.6. H NMR (CDCl , 500 MHz): δ 8.40−8.38 (m,

H), 7.92−7.89 (m, 2H), 7.60−7.57 (m, 1H), 7.53−7.50 (m, 4H),

.34−7.31 (m, 3H), 6.70−6.67 (m, 2H), 3.82 (s, 3H), 1.69 (m, 1H).

EtOAc 9:1). Anal. Calcd for C H N O Si: C, 73.6; H, 7.1; N, 6.1.

C{ H} NMR (CDCl , 125 MHz): δ 190.8, 171.9, 162.9, 135.1,

28 32 2 2

Found: C, 73.9; H, 7.1; N, 6.1. H NMR (CDCl , 500 MHz): δ 8.39

33.1, 132.1, 130.9, 130.3, 129.4, 128.9, 128.8, 126.7, 122.8, 115.3,

13.8, 55.5.

-Hydroperoxy-4,5-Diphenyl-2-(p-Tolyl)-4-Imidazole (H3). 0.168

g (76%), colorless crystals, R = 0.38 (SiO , Hex/EtOAc 3:1). Anal.

Calcd for C H N O : C, 77.2; H, 5.3; N, 8.2. Found: C, 72.6; H,

(d, 2H, J = 7.7 Hz), 8.23 (d, 2H, J = 7.4 Hz), 7.51 (t, 1H, J = 7.1 Hz),

.44 (t, 2H, J = 7.4 Hz), 7.31 (dd, 7H, J = 26.8, 4.9 Hz), 2.46 (s, 3H),

.86 (s, 9H), 0.21 (s, 3H), 0.16 (s, 3H). C{ H} NMR (CDCl , 125

MHz): δ 191.1, 171.6, 134.9, 132.5, 131.1, 130.1, 129.8, 129.4, 129.1,

128.8, 128.5, 126.1, 115.1, 26.2, 21.9, 18.4, −5.3, −5.3.

.9; N, 8.4. H NMR (CDCl , 500 MHz): δ 8.40 (d, 2H, J = 7.3 Hz),

-(4-Bromophenyl)-4-((tert-butyldimethylsilyl)peroxy)-4,5-di-

.81 (d, 2H, J = 8.0 Hz), 7.59 (t, 1H, J = 7.3 Hz), 7.53−7.51 (m, 4H),

phenyl-4-imidazole (S4). 0.090 g (22%), colorless crystals, R = 0.59

.33−7.32 (m, 3H), 6.98 (d, 2H, J = 8.0 Hz), 2.32 (s, 3H), 1.66 (s,

_H₎_.₁3C{1H} NMR (CDCl , 125 MHz): δ 191.1, 172.4, 142.9, 134.9,

(SiO , Hex/EtOAc 9:1). Anal. Calcd for C H BrN O Si: C, 62.2; H,

2 27 29 2 2

.6; N, 5.4. Found: C, 63.0; H, 5.7; N, 5.5. H NMR (CDCl , 500

33.2, 130.9, 130.3, 130.0, 129.5, 129.0, 128.9, 128.9, 127.3, 126.8,

15.3, 21.8.

-(4-Bromophenyl)-4-hydroperoxy-4,5-diphenyl-4-imidazole

H4). 0.190 g (77%), colorless crystals, R = 0.48 (SiO , Hex/EtOAc

:1). Anal. Calcd for C H BrN O : C, 61.9; H, 3.7; N, 6.9. Found:

C, 61.8; H, 3.9; N, 6.6. H NMR (CDCl , 500 MHz): δ 8.38 (dt, 2H,

MHz): δ 8.40−8.37 (m, 2H), 8.27−8.25 (m, 2H), 7.69−7.67 (m,

H), 7.55−7.51 (m, 1H), 7.48−7.44 (m, 2H), 7.38−7.34 (m, 2H),

(

7.32−7.29 (m, 3H), 0.88 (s, 9H), 0.23 (s, 3H), 0.19 (s, 3H). C{ H}

NMR (CDCl , 125 MHz): δ 191.6, 170.7, 134.6, 132.7, 131.9, 131.4,

130.9, 130.5, 129.9, 129.3, 128.8, 128.6, 126.9, 126.2, 115.3, 26.2,

J = 8.5, 1.6 Hz), 7.76−7.73 (m, 2H), 7.64−7.60 (m, 1H), 7.55−7.52

18.4, −5.3, −5.4.

644

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

Paulo, Brazil; orcid.org/0000-0001-

Article

-((tert-Butyldimethylsilyl)peroxy)-4,5-diphenyl-2-(4-

absorbance, ﬂuorescence, and NMR spectral data, and

examples of CL time proﬁ les and linear correlations with

observed rate constants (PDF)

(

triﬂuoromethyl)phenyl)-4-imidazole (S5). 0.046 g (40%), colorless

crystals, R_f= 0.61 (SiO , Hex/EtOAc 9:1). Anal. Calcd for

C H F N O Si: C, 65.9; H, 5.7; N, 5.5. Found: C, 66.2; H, 5.9;

N, 5.5. H NMR (CDCl , 500 MHz): δ 8.60 (d, 2H, J = 8.0 Hz),

.25−8.23 (m, 2H), 7.78 (d, 2H, J = 8.1 Hz), 7.57−7.53 (m, 1H),

AUTHOR INFORMATION

■

.48−7.45 (m, 2H), 7.35−7.29 (m, 5H), 0.85 (s, 9H), 0.20 (s, 3H),

.16 (s, 3H). ¹³C{ H} NMR (CDCl , 125 MHz): δ 192.0, 170.4,

Corresponding Author

Fernando Heering Bartoloni − Centro de Cien

Humanas, Universidade Federal do ABC, Santo André

34.8, 134.3, 133.6, 133.4, 133.0, 130.9, 130.3, 130.0, 129.5, 128.9,

cias Naturais e

28.7, 126.3, 125.6, 125.6, 125.1, 115.3, 26.2, 18.4, −5.4, −5.4.

-(4-((tert-Butyldimethylsilyl)peroxy)-4,5-diphenyl-4-imidazole-

9210-580, Sao

-benzonitrile (S6). 0.110 g (61%), colorless crystals, R = 0.72 (SiO ,

7304-0992 ; Email: fernando.bartoloni@ufabc.edu.br

Hex/EtOAc 9:1). Anal. Calcd for C H N O Si: C, 71.9; H, 6.2; N,

.0. Found: C, 71.6; H, 6.2; N, 8.9. H NMR (CDCl , 500 MHz): δ

Authors

Andreia Boaro − Centro de Cien

Universidade Federal do ABC, Santo André 09210-580, Sao

Paulo, Brazil

Roberta Albino Reis − Centro de Cien

Humanas, Universidade Federal do ABC, Santo André

09210-580, Sao Paulo, Brazil

Carolina Santana Silva − Centro de Cien

Humanas, Universidade Federal do ABC, Santo André

9210-580, Sao Paulo, Brazil

Diego Ulysses Melo − Centro de Cien

Humanas, Universidade Federal do ABC, Santo André

.58−8.56 (m, 2H), 8.24−8.22 (m, 2H), 7.82−7.80 (m, 2H), 7.57−

cias Naturais e Humanas,

.54 (m, 1H), 7.48−7.45 (m, 2H), 7.33−7.29 (m, 5H), 0.84 (s, 9H),

.19 (s, 3H), 0.16 (s, 3H). C{ H} NMR (CDCl , 125 MHz): δ

92.3, 170.0, 135.7, 134.3, 133.0, 132.4, 130.9, 130.3, 130.0, 129.6,

28.9, 128.7, 126.4, 118.6, 115.5, 115.2, 26.1, 18.4, −5.4, −5.4.

Quantum Yield Determination. Fluorescence quantum yields

cias Naturais e

(

Φ ) were determined using the relative method with 2,5-

diphenyloxazole in EtOAc as the standard. Chemiluminescence

quantum yields (Φ ) were obtained from the integrated CL

cias Naturais e

4,25

intensity−time proﬁles, using the method reported elsewhere.

Singlet excited-state formation quantum yields (Φ ) were determined

4,25

cias Naturais e

by numerically dividing Φ by Φ , as described elsewhere.

Chemiluminescence Kinetic Measurements for S1−6 Using

−

9210-580, Sao Paulo, Brazil

Alexander Garreta Goncalves Costa Pinto − Centro de

Ciencias Naturais e Humanas, Universidade Federal do

ABC, Santo André 09210-580, Sao Paulo, Brazil

the Luminometer. The 1.5 μmol L stock solutions of S1−6 were

prepared in ACN, and the 1.5 mmol L⁻TBAF stock solution in THF

−

was obtained by diluting 15 μL of a commercial 1.0 mol L TBAF

solution in THF up to 10 mL of the solvent. The luminometer test

tube was charged with 1.7 mL of the silylperoxide stock solution and

placed in the equipment measurement chamber. After 4 s of initiating

data acquisition, the two automatic injectors of the luminometer were

used to add, simultaneously, the TBAF stock solution and pure THF

solvent. By changing the setup volumes of the two injectors

accordingly, it was possible to have diﬀerent ﬁnal concentrations of

TBAF, diluting the stock solution directly into the test tube

containing S1−6. Combined, the added volume from the two

injectors was always 300 μL so as to keep the ﬁnal volume in the test

tube at 2.0 mL and the ﬁnal concentration of silylperoxide at 1.27

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.joc.1c00230

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

This study was ﬁnanced in part by the Coordenaca

−

μ mol L . The used TBAF concentrations are reported in Tables S1 −

Aperfeic

oamento de Pessoal de Nível SuperiorBrasil

S6 .

(CAPES)Finance Code 001; D.U.M. and R.A.R. are

Chemiluminescence Kinetic Measurements for S1, S2, and

S4 Using the Fluorescence Spectrophotometer. Using stock

thankful to CAPES for the fellowships. The authors are

thankful to the Fundac a̧ o de Amparo a Pesquisa do Estado de

Sao Paulo (FAPESP) for the provided grants (A.B. 2016/

10585-4 and 2019/15871-3, F.H.B. 2012/13807-7). The

authors are grateful to the Multiuser Central Facilities at

UFABC for the experimental support.

solutions of 0.025 mmol L⁻for S1, S2, and S4 and 25.0 mmol L for

−1

TBAF, CL kinetic assays were performed by varying the TBAF

−

concentration from 0.25 to 2.0 mmol L , keeping peroxide

concentrations constant. Data acquisition was initiated after quick

addition of the TBAF stock solution to induce the peroxide

decomposition. The voltage of the photomultiplier was 1000 V, and

the opening of the emission slit was 20 nm for the kinetic experiments

with all studied silylperoxides.

REFERENCES

■

1) Radziszewski, B. R. Untersuchungen Uber Hydrobenzamid,

Amarin Und Lophin. Ber. Dtsch. Chem. Ges. 1877 , 10 , 70 −75.

(2) Wiedemann, E.; Messerschmitt, J. B. Ueber Fluorescenz Und

Phosphorescenz II. Abhandlung. Gu

Gesetzes. Ann. Phys. 1888, 270, 446−463.

3) Lu, G.; Wada, J.; Kimoto, T.; Iga, H.; Nishigawa, H.; Kimura,

Chemiluminescence Kinetic Measurements for H1−5 Using

the Fluorescence Spectrophotometer. Using stock solutions 0.2

_m_m_o_l_L₋1 for H1−5 and 0.2 mol L−1 for DBU, CL kinetic assays were

ltigkeit Des Talbot ’ Schen

performed by varying the DBU concentration from 1.0 to 10.0 mmol

−

L , keeping peroxide concentrations constant. Data acquisition was

initiated after quick addition of the DBU stock solution to induce the

peroxide decomposition. The voltage of the photomultiplier was 800

V, and the opening of the emission slit was 10 nm for the kinetic

experiments with all studied hydroperoxides.

(

M.; Hu, Z. Chemiexcitation of the Chemiluminescence of Lophine

Peroxide Anions via a Partially Cyclic Transition State. Eur. J. Org

Chem. 2014, 2014, 1212−1219.

(

4) Vacher, M.; Galván, I. F.; Ding, B.-W.; Schramm, S.; Berraud-

Pache, R.; Naumov, P.; Ferré, N.; Liu, Y.-J.; Navizet, I.; Roca-Sanjuán,

D.; Baader, W. J.; Lindh, R. Chemi- and Bioluminescence of Cyclic

Peroxides. Chem. Rev. 2018, 118, 6927−6974.

ASSOCIATED CONTENT

sı Supporting Information

The Supporting Information is available free of charge at

https://pubs.acs.org/doi/10.1021/acs.joc.1c00230.

■

(5) Bastos, E. L.; Farahani, P.; Bechara, E. J. H.; Baader, W. J. Four-

membered cyclic peroxides: Carriers of chemical energy. J. Phys. Org.

Chem. 2017, 30, No. e3725.

Kinetic CL data for the induced decomposition of

silylperoxides (S1−6) and hydroperoxides (H1−5),

(6) Augusto, F. A.; Francés-Monerris, A.; Galván, I. F.; Roca-

Sanjuán, D.; Bastos, E. L.; Baader, W. J.; Lindh, R. Mechanism of

645

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

activated chemiluminescence of cyclic peroxides: 1,2-dioxetanes and

Article

Intramolecular Electron Transfer in Catalyzed 1,2-Dioxetane

Decomposition. J. Org. Chem. 2010, 75, 6574−6580.

(27) Nery, A. L. P.; Weiß, D.; Catalani, L. H.; Baader, W. J. Studies

,2-dioxetanones. Phys. Chem. Chem. Phys. 2017, 19, 3955−3962.

7) Augusto, F. A.; de Souza, G. A.; Khalid, M.; Baader, W. J.

on the Intramolecular Electron Transfer Catalyzed Thermolysis of

1,2-Dioxetanes. Tetrahedron 2000 , 56 , 5317 −5327.

Efficiency of electron transfer initiated chemiluminescence. Photo-

chem. Photobiol. 2013, 89, 1299−1317.

(8) Bastos, E. L.; da Silva, S. M.; Baader, W. J. Solvent cage effects:

(28) Nakashima, K. Lophine Derivatives as Versatile Analytical

Tools. Biomed. Chromatogr. 2003, 17, 83−95.

(29) Anslyn, E.V.; Dougherty, D. A. Modern Physical Organic

Chemistry; University Science Books, 2006.

Basis of a general mechanism for efficient chemiluminescence. J. Org.

Chem. 2013, 78, 4432−4439.

(9) Ciscato, L. F. M. L.; Augusto, F. A.; Weiss, D.; Bartoloni, F. H.;

Albrecht, S.; Brandl, H.; Zimmermann, T.; Baader, W. J. The

chemiluminescent peroxyoxalate system: state of the art almost 50

years from its discovery. Arkivoc 2012, 2012, 391−430.

(30) Adam, W. Determination of Chemiexcitation Yields in the

Thermal Generation of Electronic Excitation from 1,2-Dioxetanes. In

Chemical and Biological Generation of Excited States; Adam, W., Ed.;

Elsevier, Inc., 2012.

(

10) Roda, A. Chemiluminescence and Bioluminescence: Past, Present

and Future; Roda, A., Ed.; Royal Society of Chemistry, 2010.

11) Baader, W. J.; Stevani, C. V.; Bastos, E. L. The Chemistry of

Peroxides, Parts 1 & 2; Waldvogel, S., Ed.; John Wiley & Sons, Inc.,

006.

12) Davidson, D.; Weiss, M.; Jelling, M. The Action of Ammonia

on Benzil. J. Org. Chem. 1937, 02, 319 −327.

13) White, E. H.; Harding, M. J. C. Chemiluminescence in Liquid

(

31) Wilson, T. Chemiluminescence in the Liquid Phase: Thermal

Cleavage of Dioxetanes. Int. Rev. Sci.: Phys. Chem., Ser. Two 1976, 9,

65−311.

32) Stevani, C. V.; Silva, S. M.; Baader, W. J. Studies on the

(

Mechanism of the Excitation Step in Peroxyoxalate Chemilumines-

cence. Eur. J. Org Chem. 2000, 2000, 4037−4046.

Solutions: The Chemiluminescence of Lophine and Its Derivatives.

(33) De Carvalho, F.; Neto, M. D. C.; Bartoloni, F. H.; Homem-de-

Mello, P. Density Functional Theory Applied to Excited State

Intramolecular Proton Transfer in Imidazole-, Oxazole-, and Thiazole-

Based Systems. Molecules 2018, 23, 1231.

Photochem. Photobiol. 1965, 4, 1129−1155.

(14) White, E. H.; Harding, M. J. C. The Chemiluminescence of

Lophine and Its Derivatives. J. Am. Chem. Soc. 1964 , 86 , 5686 −5687.

(34) Sedgwick, A. C.; Wu, L.; Han, H.-H.; Bull, S. D.; He, X.-P.;

15) Sonnenberg, J.; White, D. M. Chemiluminescent and

James, T. D.; Sessler, J. L.; Tang, B. Z.; Tian, H.; Yoon, J. Excited-

State Intramolecular Proton-Transfer (ESIPT) Based Fluorescence

Sensors and Imaging Agents. Chem. Soc. Rev. 2018, 47, 8842−8880.

Thermochemiluminescent Lophine Hydroperoxide. J. Am. Chem.

Soc. 1964, 86, 5685−5686.

(16) White, E. H.; Rapaport, E.; Seliger, H. H.; Hopkins, T. A. The

(35) Duarte, L. G. T. A.; Germino, J. C.; Berbigier, J. F.; Barboza, C.

chemi- and bioluminescence of firefly luciferin: An efficient chemical

A.; Faleiros, M. M.; De Alencar Simoni, D.; Galante, M. T.; De

Holanda, M. S.; Rodembusch, F. S.; Atvars, T. D. Z. White-Light

Generation from All-Solution-Processed OLEDs Using a Benzothia-

zole-Salophen Derivative Reactive to the ESIPT Process. Phys. Chem.

Chem. Phys. 2019, 21, 1172−1182.

production of electronically excited states. Bioorg. Chem. 1971, 1, 92−

17) da Silva, L. P.; da Silva, J. C. G. E. Theoretical study of the

22.

correlation between superoxide anion consumption and firefly

luciferin chemiluminescence. Chem. Phys. Lett. 2013, 577, 127−130.

36) Chen, C.-L.; Chen, Y.-T.; Demchenko, A. P.; Chou, P.-T.

(18) Branchini, B. R.; Behney, C. E.; Southworth, T. L.; Fontaine, D.

Amino Proton Donors in Excited-State Intramolecular Proton-

M.; Gulick, A. M.; Vinyard, D. J.; Brudvig, G. W. Experimental

support for a single electron-transfer oxidation mechanism in firefly

bio- luminescence. J. Am. Chem. Soc. 2015, 137, 7592−7595.

Transfer Reactions. Nat. Rev. Chem. 2018, 2, 131−143.

(37) El Seoud, O. A.; Baader, W. J.; Bastos, E. L. Practical Chemical

Kinetics in Solution. In Encyclopedia of Physical Organic Chemistry;

John Wiley & Sons, Inc., 2016.

(19) Kondo, H.; Igarashi, T.; Maki, S.; Niwa, H.; Ikeda, H.; Hirano,

T. Substituent Effects on the Kinetics for the Chemiluminescence

Reaction of 6-Arylimidazo[1,2-a]Pyrazin-3(7H)-Ones (Cypridina

Luciferin Analogues): Support for the Single Electron Transfer

(38) Augusto, F. A.; Bartoloni, F. H.; Cabello, M. C.; dos Santos, A.

P. F.; Baader, W. J. Kinetic Studies on 2,6-Lutidine Catalyzed

Peroxyoxalate Chemiluminescence in Organic and Aqueous Medium:

Evidence for General Base Catalysis. J. Photochem. Photobiol. Chem.

SET)-Oxygenation Mechanism with Triplet Molecular Oxygen.

Tetrahedron Lett. 2005, 46, 7701−7704.

20) Philbrook, G. E.; Maxwell, M. A. Chemiluminescent Intensities

from Substituted Lophines. Tetrahedron Lett. 1964, 5, 1111−1116.

21) Kimura, M.; Nishikawa, H.; Kura, H.; Lim, H.; White, E. H.

39) Cabello, M. C.; El Seoud, O. A. A.; Baader, W. J. Effect of Ionic

019, 382, 111967.

Liquids on the Kinetics and Quantum Efficiency of Peroxyoxalate

Chemiluminescence in Aqueous Media. J. Photochem. Photobiol. Chem.

Maximization of the Chemiluminescence Efficiency of 1,4,5-

40) Souza, G. A.; Lang, A. P.; Baader, W. J. Mechanistic Studies on

018, 367, 471−478.

Triarylhydroperoxy-4H-Isoimidazoles. Chem. Lett. 1993, 22, 505−

(

08.

the Peroxyoxalate Chemiluminescence Using Sodium Salicylate as

22) Benisvy, L.; Blake, A. J.; Collison, D.; Davies, E. S.; David

Base Catalyst. Photochem. Photobiol. 2017, 93, 362−369.

Garner, C.; McInnes, E. J. L.; McMaster, J.; Whittaker, G.; Wilson, C.

A Phenol-Imidazole pro-Ligand That Can Exist as a Phenoxyl Radical,

Alone and When Complexed to Copper(II) and Zinc(II). Dalton

Trans. 2003, 10, 1975−1985.

(41) Silva, S. M.; Casallanovo, F. l.; Oyamaguchi, K. H.; Ciscato, L.

F. L. M.; Stevani, C. V.; Baader, W. J. Kinetic Studies on the

Peroxyoxalate Chemiluminescence Reaction: Determination of the

Cyclization Rate Constant. Luminescence 2002, 17, 313−320.

(23) Reis, R. A.; Boaro, A.; Orfao, R. B.; Melo, D. U.; Bartoloni, F.

42) Stevani, C. V.; Lima, D. c. F.; Toscano, V. G.; Baader, W. J.

H. The Decomposition of Triphenylimidazole-Para-Acetate Follows

Specific Base Catalysis and Can Be Conveniently Followed by

Fluorescence. Luminescence 2019, 34, 234−242.

Kinetic Studies on the Peroxyoxalate Chemiluminescent Reaction:

Imidazole as a Nucleophilic Catalyst. J. Chem. Soc., Perkin Trans. 2

1996, 2, 989−995.

(24) Orfao, R. B.; De Carvalho, F.; Homem-De-Mello, P.; Bartoloni,

(

43) Quina, F. H.; Bastos, E. L.. Fundamental Aspects of

F. H. On the Fluorescent, Steric and Electronic Factors Affecting the

Detection of Metallic Ions Using an Imidazolyl-Phenolic Derived

Fluorescent Probe. J. Braz. Chem. Soc. 2017, 28, 1896−1904.

Quantitative Structure-Reactivity Relationships. In Encyclopedia of

Physical Organic Chemistry; Wang, Z., Ed.; John Wiley & Sons, Inc.,

2017.

(25) Alves, J.; Boaro, A.; da Silva, J. S.; Keslarek, V. B.; Cabral, C. A.;

Orfao

, R. B.; Ciscato, L. F. M. L., Jr.; Bartoloni, F. H. Lophine

(44) The use of Hammett plots for the determination of organic

reactions mechanisms is, beyond doubt, extremely valuable (see refs

29 and 43, for example). It should be noted that this linear free-energy

relationship also made signiﬁcant contributions to the study of

chemiluminescence and its reaction pathways. Examples of Hammett

Derivatives as Activators in Peroxyoxalate Chemiluminescence.

Photochem. Photobiol. Sci. 2015, 14, 320−328.

26) Ciscato, L. F. M. L.; Bartoloni, F. H.; Weiss, D.; Beckert, R.;

Baader, W. J. Experimental Evidence of the Occurrence of

(

646

J. Org. Chem. 2021, 86, 6633−6647

The Journal of Organic Chemistry

(65) Premi, S.; Brash, D. E. Chemical Excitation of Electrons: A

Article

plots being used in chemiluminescence, under very diﬀerent contexts,

can be found in refs 25 and 26 .

Dark Path to Melanoma. DNA Repair 2016 , 44 , 169 −177.

(

45) Linnell, R. Dissociation Constants of 2-Substituted Pyridines. J.

Org. Chem. 1960, 25, 290.

46) Walba, H.; Isensee, R. W. Acidity Constants of Some

Arylimidazoles and Their Cations. J. Org. Chem. 1961, 26, 2789−

791.

47) Hall, H. K. Correlation of the Base Strengths of Amines. J. Am.

Chem. Soc. 1957, 79, 5441−5444.

48) Kaupmees, K.; Trummal, A.; Leito, I. Basicities of Strong Bases

in Water: A Computational Study. Croat. Chem. Acta 2014, 87, 385−

95.

49) Jonsson, M. Thermochemical Properties of Peroxides and

Peroxyl Radicals. J. Phys. Chem. 1996, 100 , 6814 −6818.

50) Boaro, A.; Bartoloni, F. H. Peroxyoxalate High-Energy

(66) Koo, J.-Y.; Schuster, G. B. Chemically Initiated Electron

Exchange Luminescence. A New Chemiluminescent Reaction Path for

Organic Peroxides. J. Am. Chem. Soc. 1977, 99, 6107 −6109.

(67) Koo, J. A.; Schmidt, S. P.; Schuster, G. B. Bioluminescence of

the Firefly: Key Steps in the Formation of the Electronically Excited

State for Model Systems. Proc. Natl. Acad. Sci. U.S.A. 1978, 75, 30−

68) Koo, J.-Y.; Schuster, G. B. Chemiluminescence of Diphenoyl

Peroxide. Chemically Initiated Electron Exchange Luminescence. A

New General Mechanism for Chemical Production of Electronically

Excited States. J. Am. Chem. Soc. 1978, 100, 4496−4503.

(69) Schaap, A. P.; Handley, R. S.; Giri, B. P. Chemical and

Enzymatic Triggering of 1,2-Dioxetanes. 1: Aryl Esterase-Catalyzed

Chemiluminescence from a Naphthyl Acetate-Substituted Dioxetane.

Tetrahedron Lett. 1987, 28, 935−938.

Intermediate Is Efficiently Decomposed by the Catalyst Imidazole.

Photochem. Photobiol. 2016, 92, 546−551.

(70) Schaap, A. P.; Gagnon, S. D. Chemiluminescence from a

(51) Bartoloni, F. H.; de Oliveira, M. A.; Augusto, F. A.; Bastos, E.

Phenoxide-Substituted 1,2-Dioxetane: A Model for Firefly Bio-

L.; Baader, W. J. Chemiluminescence Efficiency of Catalyzed 1,2-

Dioxetanone Decomposition Determined by Steric Effects. J. Org.

Chem. 2015, 80, 3745−3751.

luminescence. J. Am. Chem. Soc. 1982, 104, 3504−3506.

(71) Nery, A. L. P.; Röpke, S.; Catalani, L. H.; Baader, W. J.

Fluoride-Triggered Decomposition of m-Sililoxyphenyl-Substituted

Dioxetanes by an Intramolecular Electron Transfer (CIEEL)

Mechanism. Tetrahedron Lett. 1999, 40, 2443−2446.

(52) de Oliveira, M. A.; Augusto, F. A.; Ciscato, L. F.; Bastos, E. L.;

Baader, W. J. Revision of Singlet Quantum Yields in the Catalyzed

Decomposition of Cyclic Peroxides. J. Org. Chem. 2012, 77, 10537−

72) Adam, W.; Matsumoto, M.; Trofimov, A. V. Viscosity

10544.

Dependence of the Chemically Induced Electron-Exchange Chem-

iluminescence Triggered from a Bicyclic Dioxetane. J. Am. Chem. Soc.

(

53) Turro, N. J. Modern Molecular Photochemistry; The Benjamin/

Cummings Publishing Co. Inc.: Menlo Park, 1978.

54) Aizawa, N.; Harabuchi, Y.; Maeda, S.; Pu, Y.-J. Kinetic

(

000, 122, 8631−8634.

prediction of reverse intersystem crossing in organic donor −acceptor

73) Khalid, M.; Souza, S. P., Jr.; Ciscato, L. F. M. L.; Bartoloni, F.

molecules. Nat. Commun. 2020, 11, 3909.

H.; Baader, W. J. Solvent viscosity influence on the chemiexcitation

efficiency of inter and intramolecular chemiluminescence systems.

Photochem. Photobiol. Sci. 2015, 14, 1296−1305.

55) Eckersley, T. J.; Rogers, N. A. J. Photochemical studies on

bicyclic βγ -unsaturated ketones: II . Okadi-π -methane rearrangements

originating from some state other than T (π -π *). Tetrahedron 1984,

(74) Khalid, M.; de Souza, S. P., Jr.; Bartoloni, F. H.; Augusto, F. A.;

Baader, W. J. Chemiexcitation Efficiency of Intermolecular Electron-

transfer Catalyzed Peroxide Decomposition Shows Low Sensitivity to

Solvent-cavity. Photochem. Photobiol. 2016 , 92 , 537 −545.

56) Mirbach, M. J.; Henne, A.; Schaffner, K. Rearrangement of an

0, 3759−3768.

excited β ,γ -unsaturated ketone generated by dioxetane thermolysis.

(75) Catalani, L. H.; Wilson, T. Electron transfer and chemilumi-

On the T (π ,π *) and T (n,π *) reactivities. J. Am. Chem. Soc. 1978,

nescence. Two inefficient systems: 1,4-dimethoxy-9,10-diphenylan-

thracene peroxide and diphenoyl peroxide. J. Am. Chem. Soc. 1989,

57) Zimmerman, H. E.; Keck, G. E.; Pflederer, J. L. Molecular

00, 7127−7128.

111, 2633−2639.

rearrangements of excited states generated by intramolecular

dioxetane scission. Photochemistry without light. Mechanistic and

exploratory organic photochemistry. J. Am. Chem. Soc. 1976, 98,

(76) Isobe, H.; Takano, Y.; Okumura, M.; Kuramitsu, S.; Yamaguchi,

K. Mechanistic insights in charge-transfer-induced luminescence of

1,2-dioxetanones with a substituent of low oxidation potential. J. Am.

Chem. Soc. 2005, 127, 8667−8679.

58) Zimmerman, H. E.; Keck, G. E. Photochemical rearrangement

574−5581.

(77) Takano, Y.; Tsunesada, T.; Isobe, H.; Yoshioka, Y.; Yamaguchi,

K.; Saito, I. Theoretical studies of decomposition reactions of

dioxetane, dioxetanone, and related species. CT induced lumines-

cence mechanism revisited. Bull. Chem. Soc. Jpn. 1999, 72, 213−225.

without light. 95. Dioxetane effected photochemistry by direct

intramolecular excitation. J. Am. Chem. Soc. 1975, 97, 3527−3528.

(59) Schramm, S.; Karothu, D. P.; Lui, N. M.; Commins, P.; Ahmed,

E.; Catalano, L.; Li, L.; Weston, J.; Moriwaki, T.; Solntsev, K. M.;

Naumov, P. Thermochemiluminescent Peroxide Crystals. Nat.

Commun. 2019, 10, 997.

(78) da Silva, L. P.; Pereira, R. F. J.; Magalhaes, C. M.; Silva, J. C. G.

E. Mechanistic Insight into Cypridina Bioluminescence with a

Combined Experimental and Theoretical Chemiluminescent Ap-

proach. J. Phys. Chem. B 2017, 121, 7862−7871.

(60) Farahani, P.; Baader, W. J. Unimolecular Decomposition

79) Magalhaes, C. M.; da Silva, J. C. G. E.; da Silva, L. P. Study of

Mechanism of 1,2-Dioxetanedione: Concerted or Biradical? That Is

coelenterazine luminescence: Electrostatic interactions as the

the Question! J. Phys. Chem. A 2017, 121, 1189−1194.

controlling factor for efficient chemiexcitation. J. Lumin. 2018, 199,

(61) Farahani, P.; Oliveira, M. A.; Galván, I. F.; Baader, W. J. A

Combined Theoretical and Experimental Study on the Mechanism of

39−347.

Spiro-Adamantyl-1,2-Dioxetanone Decomposition. RSC Adv. 2017, 7,

80) Silva, L. P.; Magalhaes, C. M. Mechanistic insights into the

efficient intramolecular chemiexcitation of dioxetanones from TD-

62) Bechara, E. J. H.; Wilson, T. Alkyl Substituent Effects on

7462−17472.

DFT and multireference calculations. Int. J. Quantum Chem. 2018,

Dioxetane Properties. Tetraethyl-, Dicyclohexylidene-, and 3,4-

19, No. e25881.

Dimethyl-3,4-Di-n-Butyldioxetanes. A Discussion of Decomposition

Mechanisms. J. Org. Chem. 1980, 45, 5261−5268.

81) Cotton, M. L.; Dunford, H. B. Studies on Horseradish

Peroxidase. XI. On the Nature of Compounds I and II as Determined

from the Kinetics of the Oxidation of Ferrocyanide. Can. J. Chem.

(63) Adam, W. Determination of Chemiexcitation Yields in the

1973, 51, 582−587.

Thermal Generation of Electronic Excitation from 1,2-Dioxetanes.

Chemical and Biological Generation of Excited States; Adam, W.,

Cilento, G., Eds.; Academic Press, 1982; pp 115−152.

(64) Boaro, A.; Ageitos, L.; Torres, M.; Bartoloni, F. H.; de la

Fuente-Nunez, C. Light-Emitting Probes for Labeling Peptides. Cell

Rep. Phys. Sci. 2020, 1, 100257.

647