Degradation of the Cellulosic Key Chromophore 5,8-Dihydroxy-[1,4]-naphthoquinone by Hydrogen Peroxide under Alkaline Conditions

Article abstract of DOI:10.1021/acs.joc.7b01827

5,8-Dihydroxy-[1,4]-naphthoquinone (DHNQ) is one of the key chromophores in cellulosic materials. Its almost ubiquitous presence in cellulosic materials makes it a target molecule of the pulp and paper industry's bleaching efforts. In the presented study, DHNQ was treated with hydrogen peroxide under alkaline conditions at pH 10, resembling the conditions of industrial hydrogen peroxide bleaching (P stage). The reaction mechanism, reaction intermediates, and final degradation products were analyzed by UV/vis, NMR, GC-MS, and EPR. The degradation reaction yielded C₁-C₄ carboxylic acids as the final products. Highly relevant for pulp bleaching are the findings on intermediates of the reaction, as two of them, 2,5-dihydroxy-[1,4]-benzoquinone (DHBQ) and 1,4,5,8-naphthalenetetrone, are potent chromophores themselves. While DHBQ is one of the three key cellulosic chromophores and its degradation by H₂O₂ is well-established, the second intermediate, 1,4,5,8-naphthalenetetrone, is reported for the first time in the context of cellulose discoloration.

Full text of DOI:10.1021/acs.joc.7b01827

Subscriber access provided by FLORIDA ATLANTIC UNIV
Article
Degradation of the cellulosic key chromophore 5,8-dihydroxy-[1,4]-
naphthoquinone by hydrogen peroxide under alkaline conditions.
Nele Sophie Zwirchmayr, Takashi Hosoya, Ute Henniges, Lars
Gille, Markus Bacher, Paul G. G. Furtmüller, and Thomas Rosenau
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.7b01827 • Publication Date (Web): 02 Oct 2017
Downloaded from http://pubs.acs.org on October 4, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Degradation of the cellulosic key chromophore 5,8ꢀdihydroxyꢀ[1,4]ꢀnaphthoquinone by
hydrogen peroxide under alkaline conditions
†
†
†
‡
†
Nele Sophie Zwirchmayr , Takashi Hosoya , Ute Henniges , Lars Gille , Markus Bacher , Paul
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
†*
Furtmüller , Thomas Rosenau
*
Corresponding author: thomas.rosenau@boku.ac.at
†
Department of Chemistry, University of Natural Resources and Applied Life Sciences,
Muthgasse 18, Aꢀ1190 Vienna, Austria
‡
Department of Biomedical Sciences, University of Veterinary Medicine Vienna,
Veterinärplatz 1, Aꢀ1210 Vienna, Austria
Keywords
Cellulose, chromophores, naphthoquinone, benzoquinone, 1,4,5,8ꢀnaphthalenetetrone, Pꢀstage,
pulp bleaching
Abstract graphic
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
,8ꢀDihydroxyꢀ[1,4]ꢀnaphthoquinone (DHNQ) is one of the key chromophores in cellulosic
5
materials. Its almost ubiquitous presence in cellulosics makes it a target molecule of the pulp and
paper industry’s bleaching efforts. In the presented study, DHNQ was treated with hydrogen
peroxide under alkaline conditions at pH 10, resembling the conditions of industrial hydrogen
peroxide bleaching (P stage). The reaction mechanism, reaction intermediates and final
degradation products were analysed by UV/Vis, NMR, GCꢀMS, and EPR. The degradation
reaction yielded C ꢀ to C ꢀcarboxylic acids as the final products. Highly relevant for pulp
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4
bleaching are the findings on intermediates of the reaction, as two of them, 2,5ꢀdihydroxyꢀ[1,4]ꢀ
benzoquinone (DHBQ) and 1,4,5,8ꢀnaphthalenetetrone, are potent chromophores themselves.
While DHBQ is one of the three cellulosic key chromophores and its degradation by H O is well
2
2
established, the second intermediate, 1,4,5,8ꢀnaphthalenetetrone, is reported for the first time in
the context of cellulose discolouration.
Introduction
Chromophores in pulp and paper are caused by oxidative damage of the polysaccharide matrix
during processing – especially oxidative bleaching – and subsequent aging processes. Residual
lignin or extractives might also be involved as precursor compounds. Often these aging processes
are summarized as “yellowing” or “brightness reversion”. In fully bleached pulps, the
1
chromophore content can be as low as 0.1 ppm, and already concentrations slightly higher result
in a yellowish tint that is well discernible for the human eye. The pulp and paper industry is
aiming for ultimately bright products and has a strong interest in reducing the chromophore
content; at the same time bleaching sequences are optimised in order to reduce the amount of
costly bleaching chemicals and to minimize oxidative damage to the cellulose. Similarly,
conservators are interested in preserving their often valuable historic cellulosic documents
endangered by darkening caused by high chromophore content. To meet all of these needs, the
chromophores first had to be identified, as done so by application of the CRI (chromophore
2
,3
release and identification) method. This approach, for the first time, provided access to the
chromophoric compounds in cellulose whereas before only likely structural units, such as
conjugated carbonyl structures or quinones had been surmised. Comparison of the structures
ACS Paragon Plus Environment

Page 3 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
allowed distinguishing between “primary” chromophores, originating from carbohydrate matter
or respective degradation products, and “secondary” chromophores, that involve process
chemicals in their formation, such as for instance sulfur species in rayon production. Within the
primary chromophores, the most predominant structures belong to the compound classes of
hydroxyꢀ[1,4]ꢀbenzoquinones, hydroxyꢀ[1,4]ꢀnaphthoquinones, and 2ꢀhydroxyꢀacetophenones.
Of these groups, the three compounds 5,8ꢀdihydroxyꢀ[1,4]ꢀnaphthoquinone (DHNQ, 1), 2,5ꢀ
dihydroxyꢀ[1,4]ꢀbenzoquinone (DHBQ, 2), and 2,5ꢀdihydroxyꢀacetophenone have been found in
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
all materials analysed and have become known as the three cellulosic key chromophores.
Once main culprits in cellulose discolouration had been determined, establishing the
chromophores’ reactivities towards common bleaching agents used in the pulp and paper
industry would be a prerequisite to optimizing bleaching sequences and developing methods for
preservation of historic paper documents. DHBQ has been subjected to inꢀdepth analyses
regarding its reaction mechanism and products when degraded by H O under nonꢀbuffered and
2
2
alkaline conditions. It was found that the main degradation products are carboxylic acids, namely
4
,5
malonic acid and acetic acid, as well as carbon dioxide, resulting from a multiꢀstep sequence
with peroxide addition and βꢀscission as the main reactions. Similarly to DHBQ, the chemistry
of DHNQ has been investigated regarding its potential in organic synthesis and has been
6
reviewed recently. However, it has not yet been examined concerning its degradation and
reactivity towards bleaching agents, such as hydrogen peroxide, ozone, and chlorine dioxide.
Both, DHNQ and DHBQ are readily deprotonated in alkaline media, and their anions (1 and 2a
a
in Scheme 1) exhibit strong resonance stabilisation and thus delocalised double bonds. This
resonance stabilisation is the key issue when investigating the fate of DHNQ in bleaching
sequences: common bleaching agents attack localised double bonds and disrupt the quinoid
electronic structure of the chromophore, finally leading to colourless products. In the case of
delocalized double bonds ꢀ as in 1 and 2 ꢀ degradation becomes much slower and bleaching
a
a
efforts less effective.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. DHNQ (1) and DHBQ (2) form highly resonanceꢀstabilized anions, 1 and 2 ,
a
a
respectively, in alkaline media.
Another reason for the peculiar stability of DHNQ is its high thermodynamic stabilisation that
results in its easy reformation from C2 to C4 fragments, namely degradation products of
6
cellulosic materials formed in bleaching sequences. Summing up, DHNQ is not only resistant to
bleaching efforts, but it is also reformed easily, making its successful and lasting removal from
pulp and other cellulosic matter particularly difficult.
In the present study, the key chromophore DHNQ, was degraded by H O under alkaline
2
2
5
conditions at pH 10 that resemble a P stage in industrial bleaching. An emphasis was put on the
intermediates and degradation products formed. Kinetic analyses of the DHNQ/H O reaction
2
2
were performed and the (radical) intermediates analysed with the help of in situ NMR and EPR
experiments. The reaction mechanism of DHNQ/H O is presented in detail. Final support of the
2
2
shown intermediates and reaction mechanism is achieved in form of quantum chemical
calculations, by calculation of activation parameters and comparison with experimental values.
Results and discussion
Kinetic analysis and quantum chemical calculations.
In previous studies on the chromophore DHBQ and its degradation by H O at different pH
2
2
4,5
values (unbuffered and mildly alkaline at pH 10), it was shown that the reaction followed a
pseudoꢀfirst order. The requirements to define reactions as pseudoꢀfirst order are (1) one of the
reagents is present in such a large excess that its concentration can be considered constant in the
course of the reaction and (2) a linear correlation for ln(A) vs. time is found. Applied to our case,
the integrated rate law for pseudoꢀfirst order reactions is as given in eq. 1, with [DHNQ] being
0
4
,7
the initial concentration of DHNQ and k being the kinetic rate constant.
ACS Paragon Plus Environment

Page 5 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
–
kt
[
DHNQ] = [DHNQ]0 e
[equation 1]
The DHNQ/H O degradation was followed in stoppedꢀflow UV/Vis measurements with
2
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
excesses of H O between 5 and 667 molar equivalents in relation to DHNQ. Based on
2
2
previously recorded UV/Vis spectra of DHNQ at pH 10, the peak at 574 nm was chosen for the
kinetic analysis. A linear correlation was found for ln(A) against time for all excesses of H O ,
2
2
meaning that a pseudoꢀfirst order was found in all DHNQ degradation reactions.
The reaction constant k can be obtained from the slope of the plot ln(A) against time. This was
done so for the first 100 s (equivalents of H O from 5 to 333) or the first 50 s (equivalents of
2
2
500 and 667). Plotting longer reaction times distorted the clearly linear, ideal correlation
obtained for 50 s to 100 s; a phenomenon that is commonly known to occur due to followꢀup
7
reactions (or “competing” reactions). For results on k at 298.15 K and the graph ln(A) vs. time,
see Figure 1.
Figure 1. Degradation of DHNQ by H O at 298.15 K followed by UV/Vis analysis at 576 nm,
2
2
measured in stoppedꢀflow mode. The obtained linear correlation for ln(A) vs. time t indicates a
pseudoꢀfirst reaction order for all molar equivalents of H O used. The reaction rate constant k
2
2
was obtained from the slope of the corresponding lines.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In addition to stoppedꢀflow UV/Vis measurements, temperatureꢀregulated UV/Vis measurements
were done to determine the activation energy (EA(exp)) of the rateꢀdetermining step of the
DHNQ/H O degradation. With the knowledge on the reaction rate obtained from the stoppedꢀ
2
2
flow UV/Vis measurements – a fast reaction, only exhibiting ideal behaviour for the first 50 to
00 seconds of the reaction – a 100ꢀfold molar excess of H O (relative to DHNQ) was chosen
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
2
for the manually operated, temperatureꢀregulated UV/Vis experiments. This way, the reaction
was already reasonably fast at lower temperatures, and measurements were still possible at
higher temperatures. The absorption was again followed at 576 nm, and the reaction
temperatures ranged between 303.15 and 343.15 K. ln(A) was plotted against time for reaction
times of up to 200 s, depending on temperature and the resulting linear range. The k values
ꢀ1
obtained from the slopes of the lines ranged between 0.00396 and 0.0351 s . An Arrhenius plot
was constructed from ln(k) vs. 1/T, as depicted in Figure 2. The experimental activation energy
EA(Exp) was calculated from the slope of the Arrhenius plot and determined to be 11.7 kcal/mol.
Figure 2. Left: ln(A) vs. time t at 303.15 to 343.15 K, determined by UV/Vis analyses at 576 nm.
k was determined from the slope of the lines and used for construction of the Arrhenius plot.
Right: Arrhenius plot obtained from ln(k) vs. 1/T [K]. The slope, being equal to –E /R, was used
A
in the calculation of EA(Exp)
.
8
⧧
⧧
⧧
The activation parameters (ꢁ H°, ꢁ S°, and ꢁ G°) were calculated based on E
and are
A(Exp)
presented in Table 1. Additionally, an Eyring plot was constructed by plotting ln(k/T) vs. 1/T.
⧧
⧧
From the intercept, ꢁ S° was calculated. The Eyring value for ꢁ S° (ꢀ33.12 cal/K·mol) was in
ACS Paragon Plus Environment

Page 7 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
⧧
good accordance with the values of ꢁ S° calculated from EA(Exp), which ranged between ꢀ32.83
⧧
⧧
⧧
and ꢀ33.26 cal/K·mol. Variation of the values of ꢁ H°, ꢁ S°, and ꢁ G° with temperature is
⧧
⧧
minimal. Noteworthy, ꢁ G° is about twice as large as ꢁ H°. With the wellꢀknown Gibbsꢀ
⧧
⧧
⧧
Helmholtz equation ꢁ G° = ꢁ H° + T*ꢁ S° in mind, this indicates a large entropic influence and
that the degree of order in the rateꢀdetermining transition state is higher than that of the reactants.
This makes it highly likely that the rateꢀdetermining step is intermolecular, involving both
DHNQ and a coreactant – evidently H O or a derived species – rather than a subsequent
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
2
fragmentation. The latter, as an intramolecular process, would have a (small) positive activation
entropy.
Table 1. Experimental pseudoꢀfirst order rate constants k at different reaction temperatures
⧧
(
313.15 to 343.15 K), obtained by UV/Vis measurements, and the activation parameters ꢁ H°,
⧧
⧧
ꢁ S°, and ꢁ G° as obtained from the Eyring plot, ln(k/T) vs. 1/T. For comparison, the Arrhenius
activation energy EA(Exp), calculated from the slope of the Arrhenius plot, was 11.7 kcal/mol.
ꢁ⧧
S° calculated from the intercept of the xꢀaxis of the Eyring plot was –33.12 cal/K·mol.
activation parameters
ꢀ1
⧧
⧧
⧧
Temperature [K] k [s ]
ꢁ H°
ꢁ S°
ꢁ G°
[kcal/mol] [cal/K·mol] [kcal/mol]
3
3
3
03.15
13.15
23.15
0.00396
0.00607
0.01152
0.02275
0.0351
11.14
11.12
11.10
11.08
11.06
ꢀ32.83
ꢀ33.28
ꢀ33.23
ꢀ33.03
ꢀ33.26
21.09
21.54
21.83
22.08
22.47
333.15
43.15
3
Degradation product analysis.
For degradation product analysis, a bulk degradation experiment was conducted under similar
conditions as the UV/Vis analyses. 1 was dissolved in borax buffer (pH 10) and H O was added
2
2
(
195 equivalents). Upon complete decolouration of the initially dark blue reaction solution, the
1
solution was freezeꢀdried and the remaining solid was subject to different analyses, H, 2D NMR
(
HSQC/HMBC), and GCꢀMS. Additionally, in situ NMR experiments were performed at a
smaller scale, using D O and 30 equivalents of H O . The products detected by NMR were
2
2
2
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
maleic acid (3), oxiraneꢀ2,3ꢀdicarboxylic acid (4), malonic acid (6), formic acid (7), and acetic
acid (8) (the latter only in traces after lyophilisation), and were confirmed by comparison with
authentic samples. Compound 7 was only found in the in situ experiment, as its high volatility
renders it impossible to detect it after the sample workꢀup that involves lyophilisation. GCꢀMS
analyses confirmed the products 3 and 6. GCꢀMS revealed another degradation product, oxalic
acid (9). Compound 9 is formed from oxidation of the intermediate tartaric acid (5) ꢀ the
hydrolysate of 4. Because product 9 lacks any nonꢀexchangeable protons, it cannot be detected in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
H NMR. The degradation products of DHNQ/H O under alkaline conditions are depicted in
2
2
Scheme 2.
O
O
O
O
O
O
O
O
O
O
O
O
O
O
1
8
H O
pH 10
O
O
O
2
2
HO
O
9
6
3
4
5
2
3
7
6
O
OH
10
4
5
O
O
O
O
O
O
2a
O
H
O
1a
O
O
O
8
9
7
O
O
O
CO2
Scheme 2. Alkaline degradation of 1 or 2 by H O , resulting in similar degradation products.
2
2
Product 3 is formed by βꢀscission of the "quinoid" side of 1 and compounds 4, 7, and 9 result
a
from 3 in the presence of H O . Compounds are depicted as their (di)anions.
2
2
Degradation mechanism and quantum chemical calculations.
Based on degradation product analysis and UV/Vis, a reasonable degradation mechanism can be
conceived as follows: the formation of 3 originates from the loss of DHNQ’s “quinoid part” – a
strong indicator being that only 3 as the cisꢀisomer of the two butenedioic acids was found, but
not the transꢀconfigured counterpart, fumaric acid. 4 can be easily formed from 3 in the presence
9,10
of an oxidizer. The similarity of the other degradation products obtained from DHNQ and the
4
,5
degradation products of DHBQ is evident ꢀ DHNQ equally results in 6 and 8 when degraded
with H O . Is it thus highly likely that DHBQ is an intermediate in the degradation of DHNQ,
2
2
even more when taking into account that 4, 7, and 9, are all linked by originating from 3.
ACS Paragon Plus Environment

Page 9 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
As concluded from the kinetic analysis, the reaction of DHNQ and H O follows pseudoꢀfirst
2
2
order kinetics, and the rateꢀdetermining first step is the attack of DHNQ by H O . The
2
2
reactions’s first intermediate is formed by addition of H O to the “quinoid” half of the molecule.
2
2
In other words, the attack of H O cancels the resonance stabilization – the superposition of
2
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
quinoid and aromatic canonic forms – and forces the molecule to “decide” which half is aromatic
5
and which quinoid, the latter reacting with the oxidant. Depending on the medium, and thus the
resulting charge of DHNQ, this results in the formation of I (from neutral DHNQ) or I (from
0
1
the dianion), and the subsequent scission of the instable peroxide linkage between C1 and C4
results in either a biꢀradical R or a bis(anion radical) R , see Scheme 3. Quantum chemical
0
1
calculations for the formation of R and of R were carried out on the DFT(M062x) level. The
0
1
values for the calculated activation energy EA(calc) were compared with the experimental one,
EA(exp). EA(calc) for R matches the experimental one very well, with values of 15.1 kcal/mol and
1
1
1.7 kcal/mol, respectively. EA(calc) for R (neutral) however, is almost twice as high as the
0 ,
experimental one, 31.6 kcal/mol. Thus, deprotonation of 1, leading to 1 , takes place before H O
2
a
2
addition, and scission of the rather labile endoperoxide between C1 and C4 in intermediate I1
results in biradical R1.
EA(calc) = 31.6 kcal/mol
EA(calc) = 15.1 kcal/mol
Scheme 3. Intermediates I , I , R , and R depicted as used in the quantum chemical
0
1
0
1
calculations. For the formation of R , the calculated E (15.1 kcal/mol) matches the experimental
1
A
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
value (11.7 kcal/mol) quite well. Compared to the anion, attack of H O to neutral DHNQ is
2
2
much disfavoured. DHNQ is thus present in its deprotonated form before H O addition, and
2
2
scission of the endoperoxide linkage in the resulting intermediate I leads to the highly transient
1
bis(anion radical) R1.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The good match of the calculated and experimental E values is a strong backꢀup for the
A
proposed structure of the shortꢀlived intermediate R . By tautomerisms ([1,5]ꢀsigmatropic proton
1
shifts) and singleꢀelectron transfer – both occurring twice due to the biradical character – a
dianionic intermediate is formed that immediately looses two hydroxide anions to form 1,4,5,8ꢀ
naphthalenetetrone (I ) as a stable intermediate (Scheme 4). All conversions between R and I
2
2
1
are exergonic (spontaneous) and do not exhibit any significant activation energy barriers
compared to the rateꢀdetermining step discussed above. 1,4,5,8ꢀNaphthalenetetrone is a strong
chromophore itself. So far, its occurrence has not been linked to chromophores in cellulosics.
Scheme 4. Formation of the stable, chromophoric intermediate I from transient bis(anion
2
radical) R1.
Experimental confirmation that I is truly an intermediate in the degradation of DHNQ by H O ,
2
2
2
1
1
was provided by synthesis of an authentic sample of I according to a literature procedure. In
2
an in situ NMR experiment I was then degraded by H O under the same conditions as DHNQ
2
2
2
(
borax buffer pH 10, 10 molar equivalents of H O ). The same degradation products and the
2 2
same product pattern were found in both cases: products 3, 4, 6, 7, and 8, while 9 was detected
1
only by GCꢀMS; 9 as it lacks protons visible in H NMR. This outcome is only possible if I was
2
indeed an intermediate in the course of the degradation of DHNQ by H O .
2
2
ACS Paragon Plus Environment

Page 11 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
The further degradation of I as shown in Scheme 5, involves addition of hydrogen peroxide to
2
,
form endoperoxide I Cleavage of the labile OꢀO bond – analogous to I ꢀR – results in R which
3
1
1
3
undergoes immediate βꢀscission of R into compound 3, a detectable product, and R . R
4
3
4
immediately forms a highly reactive aryne species that adds water to give 2ꢀhydroxyꢀ1,4ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
benzoquinone I , which is known to be readily oxidized to DHBQ (2 ) in the presence of H O .
4
a
2
2
5
From 2 the degradation continues according to wellꢀstudied and already established pathways.
a,
Interestingly, DHBQ, a cellulosic key chromophore itself, was found as relatively stable
intermediate in the degradation of DHNQ. Since both chromophores are found accompanying
each other in cellulosic materials, this poses the interesting question if at least a part of the total
DHBQ might originate from oxidative degradation of DHNQ. That the whole amount of DHBQ
can be traced back to DHNQ can safely be excluded because DHBQ is also formed by other,
1
2
already established pathways – e.g. from glucuronoxylanꢀderived hexeneuronic acids or from
1
3,14
“3ꢀketoꢀglucose moieties” in oxidatively damaged cellulose chains.
Product 3 is further oxidized to 4. In experiments with authentic samples, 4 did show signs of
hydrolysis, albeit being fairly stable. Its hydrolysate, tartaric acid (5), is the intermediate in the
1
5
formation of 9 and 7. The degradation pathway 5ꢀ9ꢀ7 was also reported in literature.
Scheme 5. Mechanism of the further degradation of 1,4,5,8ꢀnaphthalenetetrone (I ) by H O at
2
2
2
pH 10 to colorless lowꢀmolecular weight compounds.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
EPR measurements.
To gain further information about the degradation of DHNQ by H O , the reaction was analysed
2
2
by EPR. For this, the reaction conditions had to be adjusted: a more concentrated solution of 1
was prepared by using a 1:1 mixture of borax buffer (pH 10) and acetonitrile (MeCN), the final
concentration being 28 mM. To slow the reaction down, only 5 molar equivalents of H O were
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
used. Under the applied conditions two radicals of different stability were detected, one being
stable for several minutes, the other showing a remarkably high stability of several days at room
temperature (Figure 3). The less stable radical appeared as a triplet signal, the more stable radical
resulted in a singlet. The triplet signal was detected first and, upon its decline, the singlet
appeared. However, this does not automatically mean that the intermediates giving the signals
are formed in a chronological order ꢀ the singlet signal could have been present from the
beginning of the measurement, but concealed by the triplet signal.
Figure 3. EPR signals of the radical intermediates detected upon the degradation of 1 by H O .
2
2
The triplet signal (left) declined within minutes, whereas the singlet signal of the more stable
second intermediate (right) was still detectable after keeping the solution at room temperature for
6 days.
To be able to take these EPR results into account, it had to be confirmed that the altered
conditions, namely changing solvent composition, had no influence on the degradation
mechanism. The most feasible way was to perform in situ NMR experiments under EPR
conditions (borax buffer pH 10: MeCN = 1:1; 5 equivalents H O ) and to compare the obtained
2
2
spectra to spectra obtained from the bulk degradation experiments (pH 10, 195 equivalents
ACS Paragon Plus Environment

Page 13 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
H O ). By making use of a water suppression technique, we were indeed able to confirm the
2
2
formation of the products 3, 4, and 7. The MeCN resonance was not suppressed, thus the peaks
of the products 8 and 6 are superimposed by the strong solvent signal and were not visible. The
degradation products 3, 4, and 7 were found in both reaction setꢀups with comparable intensities,
thus no influence of the MeCN addition on the degradation reaction’s mechanism was observed.
Additionally, two broad peaks were found at 7.27 and 6.25 ppm, most likely representing the
radical giving the longꢀlived singlet in EPR. Comparison of the EPR signals with literature
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
6,17
indicated that the structures giving the signals in EPR are both quinoid molecules.
A first
simulation revealed good accordance between an EPR spectrum for a quinoid species with
2+2) equivalent protons and the triplet obtained (Figure 4). R forms R , a process where quick
1
8
(
1
2
recombination of the two radicals with concomitant tautomerisms causes the EPR signal to
disappear, as depicted in Scheme 3.
Figure 4. EPR signal of R , immediately after addition of H O to the chromophore solution
1
2
2
(top). The simulation for 4 (2+2) equivalent protons (bottom) is in good accordance with the
experimental results.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
At this point of the investigation we can ascertain that the stable radical species possesses 4
(2+2) equivalent protons, whether it is a species formed by oneꢀelectron oxidation of 1, oneꢀ
electron reduction of I or whether it is R cannot be conclusively decided in the moment. In all
2
3
of these species the single spins are stabilized by resonance.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Conclusion
To summarize, the main products formed in the degradation of 1 by H O at pH 10 are maleic
2
2
acid (3), oxiraneꢀ2,3ꢀdicarboxylic acid (4), and malonic acid (6), the latter originating from
DHBQ which is formed as an intermediate in the DHNQ/H O degradation. Minor products
2
2
formed are oxalic acid (9), formic acid (7), tartaric acid (5) and acetic acid (8), which is
originating from DHBQ as does 6. Activation parameters were: E (from Arrhenius plot) = 11.7
A
⧧
⧧
⧧
kcal/mol. ꢁ S° (from Eyring plot) = 33.12 cal/K·mol. ꢁ H° = 11.14 to 11.06 kcal/mol, ꢁ S° = ꢀ
⧧
3
2.38 to ꢀ33.26 cal/K·mol, and ꢁ G° = 21.09 to 22.47 kcal/mol. We were able to show that
DHNQ can be fully degraded by H O to colorless products. In the course of the degradation
2
2
reaction, two other potent chromophores are formed: DHBQ, wellꢀknown to be a cellulosic key
chromophore itself, and 1,4,5,8ꢀnaphthalenetetrone that was now reported in cellulosics’
chromophore chemistry for the first time. Applying the results of this study to industrial
bleaching conditions, it must be pointed out that successful bleaching of any cellulosic material
relies on sufficiently high excesses of H O present in the bleaching sequence to ensure the
2
2
degradation of not only DHNQ, but its followꢀup chromophores, DHBQ and 1,4,5,8ꢀ
naphthalenetetrone. This study covers only hydrogen peroxide as the bleaching agent. The
behaviour of DHNQ towards other bleaching agents, such as ClO , ozone or peracetic acid, is the
2
subject of current studies and upcoming reports.
Experimental section
General Experimental Methods. Commercial chemicals were of the highest grade available and
were used without further purification. Reagentꢀgrade solvents were used for all extractions and
workup procedures. Distilled water was used for all aqueous solutions. The stoppedꢀflow
apparatus (model SXꢀ18MV) was from Applied Photophysics (Surrey, UK). For the 20 µL
ACS Paragon Plus Environment

Page 15 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
optical observation cell with 10 mm light path length, the dead time of the instrument is 1.2 ms.
Temperatureꢀregulated UV/Vis spectra were recorded on a Hitachi Uꢀ3010 with a path length of
10 mm, slit width was 2 mm. GCꢀMS analysis was performed on an Agilent 7890A gas
chromatograph coupled with an Agilent 5975C triple axis mass selective detector (MSD; Agilent
Technologies, Santa Clara, CA, USA). A DB5ꢀMS column (30 × 0.25 mm i.d. × 0.25 ꢂm film
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
thickness; J&W Scientific, Folsom, CA, USA) was used. For H NMR analysis, a Bruker
1
13
Avance II 400 instrument ( H resonance at 400.13 MHz, C resonance at 100.61 MHz) with a 5
mm broadband probe head (BBFO) equipped with zꢀgradient with standard Bruker pulse
programs was used. Data were collected with 32k data points and apodized with a Gaussian
window function (GB = 0.3) prior to Fourier transformation. A 2.5 s acquisition time and a 1 s
relaxation delay were used. Bruker TopSpin 3.5 was used for the acquisition and processing of
the NMR data. EPR spectra of the DHNQ reaction intermediates were measured with a Bruker
EMX Xꢀband spectrometer equipped with a TE cavity.
Stopped-flow UV/Vis kinetic analysis. All measurements were performed at 25°C and pH = 10 in
borax buffer, using a 0.6 mM solution of 1 with H O concentrations between 30 to 400 mM
2
2
(
corresponding to molar equivalents of H O between 5 and 667). Measurement times were
2 2
ranging between 100 to 1000 s. The range recorded was 193ꢀ739 nm, step size was 2.2 nm.
Temperature-regulated UV/Vis kinetic analysis. A 0.2 mM solution of 1 was put in a covered
quartz cuvette and the temperature adjusted to the corresponding value (313.15 to 343.15 K).
After 10 minutes of equilibration time, H O (10%, 100 molar equivalents) was added to the
2
2
cuvette. The solution was mixed and the absorption at 576 nm was recorded for the time of 500
s. Data points were recorded every 0.5 s.
Bulk degradation of 1 by H O . To 75 mL of a borax buffer solution of 1 (30 mg, 0.16 mmol) at
2
2
pH 10 H O was added (30%, 195 molar equivalents). Upon complete decolouration, a 10 %
2
2
solution of Na S O was added dropwise until the excess peroxide was destroyed. A 15 mL
2
2
3
aliquot of the solution was used for the GCꢀMS analysis. The remaining 60 mL were frozen at ꢀ
80°C and lyophilized and used for NMR experiments in D O. For in situ NMR experiments, the
2
reaction set up was repeated in a smaller scale with 30 equivalents of H O (30%). The solution
2
2
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
was not lyophilized, but directly analyzed under application of a water suppression technique. H
NMR (H O+D O): δ 8.45 (s, 7), 6.12 (bs, unknown), 6.10 (s, 3), 3.60 (s, 4), 3.14 (s, 6), 1,99 (s,
2
2
1
3
8). C δ 184.6 (8, COOH), 174.6 (4, COOH), 176.9 (3, COOH), 173.7 (7, COOH), 132.8 (3,
CH=CH), 54.1 (4, CH (O)), 49.7 (6, CH ), 26.4 (8, CH )
2
2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
GC-MS analysis. A 15 mL aliquot of the 1/H O bulk degradation solution was used for the GCꢀ
2
2
MS analysis of degradation products. The sample pH was adjusted to neutral by addition of HCl.
200 µL of the sample solution were added to a GC vial for freezeꢀdrying. Standard addition,
derivatization and sample analysis were performed according to the literature procedure reported,
19
as were the GCꢀMS operating conditions. The NIST/Wiley 2008 database was used for
compound identification.
EPR measurements. The intermediates were generated by rapid mixing of DHNQ (28 mM) with
H O (5%, 5 molar equivalents) in a solution of buffer pH 10/MeCN (1:1) and immediately
2
2
transferred to a quartz flat cell in the EPR cavity. Then the acquisition of EPR spectra was started
with the following parameters: microwave frequency, 9.76 GHz; microwave power, 20 mW;
5
receiver gain, 4.48x10 ; modulation frequency,100 kHz; modulation amplitude, 2 G; center field,
3
479.4 G; sweep 20 G; scan rate, 28.6 G/min; time constant, 327 ms; data points, 1024;
temperature, 300 K; 15 scans were recorded. Simulation of EPR spectra was performed with the
2
3
program WINSIM. Simulation parameters: Lorentzian 0.640, Gꢀshift ꢀ0.186, line width 0.848.
Coupling constant. (2+2) equivalent protons: 2.277 Hz.
Degradation of 1 under conditions applied in EPR - in situ NMR. 1 (5.3 mg, 0.028 mM) was
dissolved in 0.6 mL of a 1:1 mixture of borax buffer (pH 10) and MeCN. A few drops of D O
2
and 4,4ꢀdimethylꢀ4ꢀsilapentaneꢀ1ꢀsulfonic acid (DSS) as an internal standard were added. H O2
2
1
(
5%, 5 molar equivalents) was added and the solution immediately used for H NMR
experiments under application of a water suppression technique. The solution was kept at r.t. for
1
6
7
days and measured again in NMR and EPR. H NMR (H O:MeCN=1:1+D O): δ 8.40 (s, 7),
2
2
.27 (bs, radical), 6.52 (bs, radical), 6.45 (s, 1), 6.06 (s, 3), 3.52 (s, 4), 3.31 (s, 6)
ACS Paragon Plus Environment

Page 17 of 18
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
15
Synthesis of 1,4,5,8-naphthalenetetrone (I ). 1 (76 mg, 0.40 mmol) was dissolved 5 ml of a 1:1
2
mixture of acetone and acetonitrile. [Bis(trifluoroacetoxy)iodo]benzene (516 mg, 1.2 mmol) was
added stepwise and upon complete addition, the solution stirred at r.t. for 3 hours. The solution
was concentrated and addition of acetone yielded I (19 mg, 25%) by precipitation. UV/Vis
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(
MeCN): λmax [nm] (log ε) 556 (0.36), 514 (0.67), 485 (0.72), 286 (2.42). H NMR (CD CN): δ
3
1
3
6
.85 (s, 4 H). C NMR: δ 186.6 (Cꢀ1/ Cꢀ3/ Cꢀ6/ Cꢀ8), 137.0 (Cꢀ2/ Cꢀ7), 127.6 (Cꢀ4/Cꢀ5/ Cꢀ9/ Cꢀ
10).
Degradation of I by H O . 5.7 mg (30 mmol) of I were dissolved in 0.8 ml of borax buffer (pH
2
2
2
2
1
0). H O (30%, 10 equivalents) was added. After 3 hours, 100 µl of D O and DSS as an internal
2
2
2
standard were added to the reaction mixture. The solution was transferred into a NMR tube and
1
the spectra recorded using a water suppression technique. H NMR (H O/D O): δ 8.44 (s, 1H, 7),
2
2
6
.85 (s, 4 H, 2), 6.24 (s, 2 H, 3), 3.60 (s, 2H, 4), 3.35 (s, 2H, 6), 2.06 (s, 2H, 8)
Computational methods. GAUSSIAN 09 was used in all calculations. The geometry optimization
of 1 and energy calculation of the intermediate R were carried out by the DFT method with the
1
M062x functional, PCM (“SCRF” key word) in water was applied. Geometry optimisations and
frequency calculations were performed by using the key word “Opt” and “Freq”. The 6ꢀ31G(d)
basis sets were used for H, C, and O, applied by the key words “6ꢀ311+G(d)” for C and “6ꢀ
3
11G(d,p)” for H.
Acknowledgement
The authors would like to thank the Austrian Research promotion Society (FFG, project 829443)
for financial support and Dr. Nora Odabas for many fruitful discussions.
Supporting Information. The supporting information file contains computational data of
compounds 1, 1 , R , R and H O , NMR spectra of the degradation experiments as well as MS
a
1
0
2
2
spectra of the degradation product.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
References
1.
Korntner, P.; Hosoya, T.; Dietz, T.; Eibinger, K.; Reiter, H.; Spitzbart, M.; Roeder, T.;
Borgards, A.; Kreiner, W.; Mahler, A. K.; Winter, H.; Groiss, Y.; French, A. D.; Henniges,
U.; Potthast, A.; Rosenau, T. Cellulose 2015, 22, 1053.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2.
Rosenau, T.; Potthast, A.; Krainz, K.; Yoneda, Y.; Dietz, T. Shields, Z. P. I.; French, A. D.
Cellulose 2011, 18(6), 1623.
3.
Rosenau, T.; Potthast, A.; Milacher, W.; Hofinger, A.; Kosma, P. Polymer 2004, 45, 6437.
Hosoya, T.; Rosenau, T. J. Org. Chem. 2013, 78, 3176.
Hosoya, T.; Rosenau, T. J. Org. Chem. 2013, 78, 11194.
Hosoya, T.; French, A. D.; Rosenau, T. Mini-Rev. Org. Chem. 2013, 10, 309.
Sandman, D. J. Mol. Cryst. Liq. Cryst. 2006, 461, 147.
4
5
6
7
.
.
.
.
8.
Atkins, P.; de Paula, J. Atkins' Physical Chemistry, 7th Edition; Oxford University Press,
2
002. ꢁH=E − (R × T), ꢁG=R × T × [23.760+ln(T/k)], ꢁS=[(E –R×T) − ꢁG]/T
a
a
9
1
1
.
Mikova, H.; Rosenberg, M.; Kristofikova, L. Chem. Listy 2001, 95, 28.
0. Ina, T.; Nippon Shokubai Co., Ltd., Japan, JP2014070028A, 2014, p. 8.
1. Yoshino, S.; Hayakawa, K.; Kanematsu, K. J. Org. Chem. 1981, 46, 3841.
12. Rosenau, T.; Potthast, A.; Zwirchmayr, N. S.; Hettegger, H.; Plasser, F.; Hosoya, T.;
Bacher, M.; Krainz, K.; Dietz, T. Cellulose 2017, 24, 3671ꢀ3687.
13. Rosenau, T.; Potthast, A.; Kosma, P.; Hosoya, T.; Henniges, U.; Mereiter, K.; French, A.
D. Cellulose 2017, 24, 1227.
14. Rosenau, T.; Potthast, A.; Kosma, P.; Saariaho, A.ꢀM.; Vuorinen, T.; Sixta, H. Cellulose
2
005, 12(1), 43.
15. Maxted, D. R. J. Chem. Soc. 1926, 2178.
16. Gille, L.; Nohl, H. Arch. Biochem. Biophys. 2000, 375, 347.
17. Lubitz, W.; Feher, G. Appl. Magn. Reson. 1999, 17, 1.
18. Duling, D. R. J. Magnet. Reson. Ser. B 1994, 104, 105.
19. Liftinger, E.; Zweckmair, T.; Schild, G.; Eilenberger, G.; Boehmdorfer, S.; Rosenau, T.;
Potthast, A. Holzforschung 2015, 69, 695.
ACS Paragon Plus Environment