Kinetic study and real-time monitoring strategy for tempo-mediated oxidation of bleached eucalyptus fibers

Table 3 Reaction rate (k
1
), y-intercept and experimental data 
fitting (R
2
) as function of TEMPO catalyst concentration
TEMPO 
(mg/g)
k
1
(s
−1
) [10
–4
]
y-intercept
R
2
2
0.47
7.21
0.9784
4
0.98
7.26
0.9848
8
1.54
7.29
0.9858
16
3.63
7.35
0.9789
32
3.21
7.38
0.9895
Table 4 Reaction rate (k
1
), y-intercept and experimental data 
fitting (R
2
) as function of NaBr concentration
NaBr (mg/g)
k
1
(s
−1
) [10
–4
]
y-intercept
R
2
25
0.70
7.22
0.9867
50
1.32
7.24
0.9947
100
3.63
7.35
0.9798
200
3.80
7.29
0.9729

---

 Cellulose
1 3
Vol:. (1234567890)
as the accessibility of water into the fiber structure, 
leading to higher swelling at increasing refining 
degrees. Indeed, this is supported by the higher ability 
of refined pulps to retain water, represented by the 
°SR, but also for the increasing CD, indicating that 
a higher number of electron-rich sites are exposed 
per mass unit. Considering that TEMPO-mediated 
oxidation consists of an oxidation in a heterogeneous 
system, increasing the surface area of fibers may 
promote the interactions between the reagents and 
catalysts. Indeed, this becomes to the light in Fig. 
2
D, 
where it can be observed that oxidation occurs faster.
The correlation between the CC and time, properly 
evaluated by means of Eq. 
3
, lead to different k
1
for 
each refining degree (Table 
5
), but similar y-intercept, 
corresponding to the logarithm of the initial CC, and 
excellent correlation factors (R
2
), as in the previous 
cases.
The evolution of k
1
with the refining degree in 
revolutions, equivalent to applied energy, evolved 
linearly with an R
2
of 0.9877. This indicates a clear 
effect of surface area over the kinetics of TEMPO-
mediated oxidation, which can be quantified in a 13% 
for 10,000 rev of PFI. This is of particular interest, as 
4.12 × 10
–4
s
−1
is the highest k
1
obtained in the pre-
sent study but requires the application of additional 
energy. This energy was quantified in 4.44 kWh/
kg. Concretely, Carrasco et al. (
1996
) determined 
the surface area of the same pulp used in the pre-
sent study. For equivalent drainability (°SR), the 
surface area of fibers accounted for 0.98, 1.34, 2.24, 
3.50, and 4.92 m
2
/g for 0 to 10,000 PFI revolutions, 
respectively.
Out of the different parameters, temperature 
was found to have the most significant effect over 
the kinetics of the reaction. Indeed, the increase of 
the temperature from 20 to 30 °C enhanced k
1
in a 
51.52%, while increasing the TEMPO dosage from 
16 to 32 mg/g had a negative effect and increasing 
the NaBr content from 100 to 200 mg/g resulted in 
a 4.68% increase. Only in the case of mechanical 
refining the constant was increased by 13%, being 
still far from the improvement derived from a change 
on the temperature.
In addition, although the mechanical refining 
increased the reaction rate, the yield of the reaction, 
in terms of mass loss during the process, experienced 
a reduction with refining intensity. Concretely, the 
obtained yields accounted for 98.68, 94.32, 91.86, 
85.16, and 78.59% for 0 to 10,000 PFI revolutions, 
respectively, while no differences on yield were 
observed when modifying the rest of the parameters 
(temperature and/or TEMPO and NaBr dosages). The 
negative impact of refining on mass yield is mainly 
due to two phenomena. First, refining causes external 
fibrillation, and the protruding fibrils are more prone 
to degradation towards solubilized by-products than 
the fiber core. Second, the remaining xylans (which 
lack primary hydroxyl groups) can only undergo oxi-
dative cleavage during TEMPO-mediated oxidation, 
not contributing positively to CC values (Syverud 
et al. 
2011
). It is known that the surface of hardwood 
fibers is richer in xylans than that of softwood fibers 
(Syverud et al. 
2011
; Pääkkönen et al. 
2016
).
Conversion as function of time was also 
determined, aiming at glimpsing the most appropriate 
conditions in terms of reaction kinetics (Sbiai et al. 
2011
). For this, conversion was calculated according 
to Eq. 
4
and plotted according to Eq. 
5
. Figure 
4
shows the correlation between the t
final
and k
1
, 
indicating that as k
1
is increased, the required time for 
total conversion is decreased.
Starting from the most widely reported conditions 
for TEMPO-mediated oxidation, indicated in Fig. 
4
with the yellow vertical line, it is clear that few 
improvements in terms of time can be achieved 
modifying process conditions. Only temperature 
showed a significant effect on the kinetic constant, as 
well as on the required time to achieve the complete 
conversion of CH
2
OH to COO
–
groups. It becomes 
apparent that increasing the temperature from 20 °C 
to 25 or 30 °C would significantly decrease the t
final
, 
while the required extra energy is residual compared 
to other strategies such as mechanical refining. 
Furthermore, increasing the amount of TEMPO or 
NaBr would be detrimental in terms of production