Eugenio De
Felice Zampini
Instituto
Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP), Brazil
E-mail: eugenio.zampini@ifsp.edu.br
William Akio
Oliveira
Instituto
Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP), Brazil
E-mail:
williamakio20@gmail.com
Submission: 2/17/2020
Revision:
3/3/2020
Accept: 3/6/2020
ABSTRACT
The pulp and paper segment has been seeking to increase mill efficiency
through cost reduction and better production management. One of the ways to
achieve this goal is through better control and operation of the pulp plant
digester cooking. The study aimed to verify the accuracy of the FT-NIR analyzer
in the liquor lines of a pulp and paper digester and its possible benefits for
production management. The research methodology used was the case study, where
it was analyzed and compared the results of the equipment with the values
of the company's laboratory. The analysis of the instrument
proved accurate and reliable, providing information on various properties of
liquors that enable the optimization of production management with the
stabilization of the operation. Future trends indicate that infrared technology
will continue to play an important role as a detection platform due to its
inherent flexibility and capability.
Keywords: Production management; FT-NIR analyzer; Pulp and Paper.
1.
INTRODUCTION
The pulp and paper
industry faces a constant challenge to control and manage its production in
order to keep it stable and close to the ideal, seeking new technologies and
improvements. Favaretto (2001) points out that among
these improvements, there is the search for a high efficiency of productive
resources, as a way of maximizing operational income.
Companies in this
industrial segment have been seeking to increase the efficiency of factories by
reducing production costs (RESTREPO et al., 2009), and one of the ways to
achieve this goal is through better control and operation of the cooking that
took place in the digester of the cellulose plant.
Cooking consists of
subjecting the chips, which are made up of lignin, hemicellulose and cellulose
to a chemical action of strong white liquor (a mixture of caustic soda and
sodium sulfide) and water vapor. Inside the digester, lignin is dissolved
between fibers. The digester is a pressure vessel, with an approximate 60m
height, where the chips and strong white liquor are continuously introduced
from the top (RESTREPO et al., 2009). Then the weak black liquor is formed by
the mixture of the white liquor with the dissolved lignin from the chips.
In the current Kraft
cooking process, alkaline measurements are used to determine the consumption of
chemicals during the different cooking phases. These measurements are widely
accepted as standard monitoring information, as they provide a solid basis for
managing the operation of the digester. As such, to optimize the operation of
the digester, concentrations of white liquor and effective residual alkali (ERA)
of black liquor are required in the various cooking zones.
By monitoring the
effective alkali of the white liquor, as well as the sulfidity
online and knowing the moisture content of the wood, the correct ratio between
the liquor and the wood can be controlled (LOMBARDI and LUIZ, 2017). In addition, by measuring the ERA
in various areas of the digester, faster feedback will be available regarding
the rate of alkali consumption and adjustments can be made to compensate for
changes in the properties of the wood.
Knowledge of ERA in all
cooking zones would provide better control of the kappa through cooking in the
digester, thereby reducing overcooking or undercooking. Ultimately, for Silva
et al. (2016), the income gains could be realized together with the lower kappa
variability due to the possibility of a more stable production control. The
kappa number, according to the ABNT NBR ISO 302: 2018 standard, is “an
indication of the residual lignin content or of the cellulose pulp's whitening
capacity”.
The study in question
involves the first equipment using FT-NIR (Fourier Transform - Near Infrared)
technology that was installed in South America, making it possible to obtain
accurate and real-time analyzes of liquor properties, which is why it is a
justified research in this area. The study aimed to verify the accuracy of the
FT-NIR analyzer in the liquor lines of a digester of a paper and cellulose
company and its possible benefits for the production management. The research
methodology used was the case study.
The subsequent sections
present the literature review, methodology, analysis of results and discussion,
and conclusions, respectively.
2.
LITERATURE REVIEW
In today's globalized
economy, the survival of organizations depends on their ability and speed to
innovate and make continuous improvements. As a result, organizations are
constantly looking for new management tools that will lead them to greater
competitiveness through the quality and productivity of their products,
processes and services (OTANI and MACHADO, 2008).
In order to achieve an
efficient cost management and, mainly, to follow the dynamics of changes in the
supplier and consumer markets, in addition to knowing the costs, a constant
search for reduction and continuous improvement of costs is essential for the
company (POMPERMAYER and LIMA, 2002 ).
However, without
reliable information from the process, this type of management becomes costly
and difficult. Until then, the measuring instruments installed in the cooking
process of the cellulose digester in plants in Brazil are imprecise and require
constant maintenance, or use a large amount of chemicals, as is the case with
the titration records. The FT-NIR measurement technology, in addition to
avoiding these problems, measures several parameters present in the analyzed
substances.
Infrared spectroscopy
(near-infrared) has been shown as a means for the rapid non-destructive determination
of the chemical composition (liquors) and the final yield of the pulp
(TERDWONGWORAKUL et al., 2005). Infrared, together with the Fourier transform,
is an essential analytical tool for the structural analysis of the chemistry of
paper and cellulose.
Fourier transform
infrared spectroscopy (FTIR), according to Faix
(1992), can be applied in specialized areas in microanalysis where great
sensitivity is required in the analysis of aqueous or dark solutions and even
in solid samples that require the use of a technique special reflectance for
determining quantity and in cases where recognition has time as a limiting
factor.
Many researchers have
reported that the NIR (near-infrared) technique was useful for detecting
multiple information on the chemical and physical properties of wood materials
(TSUCHIKAWA, 2007). Moreover, this can contribute to the reduction of several
factors that have an impact on the daily production of paper, as mentioned by
Duran and Fardim (2007), who point to calcification
in the measuring terminals of conditivimeters and densimeters as the main causes of errors in measurements
obtained in the process. In addition, according to Duggirala
(2005) and Grace and Tran (2007), the accumulation of scale in the kraft digester and black liquor evaporators is a major
contributor to the loss of productivity at the pulp mill.
Cruz et al. (2006)
still say that these particles are deposited in various parts of the factory's
machines and are responsible for reducing production, increasing the cost of
maintaining and operating the equipment and increasing defects in the final
product and consequently reducing its quality. “During industrial processes, a
lot of waste is generated, which can be of solid, liquid or gaseous origin”
(ALMEIDA et al., 2007).
The main advantage of
the application of the FT-NIR analyzer in the digester is its ability to
provide the information of Effective Alkali (EA) of the white liquor and sulfidity, as well as the ERA of the black liquor in the
various zones, in a reliable and accurate way. Trung
and Allison (2012) state that the control strategy based on reliable data can
significantly reduce kappa variation, overcooking and tailings, while
increasing the quality of the pulp. Still, according to the same authors, the
reduction of kappa variability can also have a significant reduction in
bleaching costs and, as a result, improvement in yield is possible. Forsström et al. (2006) highlight that the optimal cooking
kappa is a versatile tool for improving the financial development of an
eucalyptus cellulose plant.
It is also worth
remembering that this more precise control and management of production is not
done in the pulp and paper companies because there are no accurate and
real-time measurements. However, this scenario will tend to change with the
results of this first installation of the FITNIR analyzer in South America,
which is the object of study in this work.
3.
METHODOLOGY
For the development of
the research, investigation strategies were implemented, such as bibliographic
material survey and analysis, collection of process data, installation of
infrared equipment, collection of new process information and, finally, study
and comparison of elements obtained.
Initially, a
bibliographic survey was carried out with the purpose of investigating the
current applications of infrared in industrial plants, focused on the paper and
cellulose segment, the main difficulties in production management involving the
industry digester and the main benefits pointed out in studies on the most
stable operation based on parameters such as sulfidity
and kappa.
In the second step,
infrared equipment was installed in the liquor lines of the digester through
tie-ins in the local piping of a paper and cellulose company. The analyzer is
from a Canadian company called FITNIR. Then, the new information was collected
and analyzed, as well as laboratory tests after implantation of the device.
In order to assess the
accuracy of the FT-NIR equipment and to obtain comparison parameters, the
results of laboratory tests of the company's liquors were gathered. These
samples were collected at an extraction point, which is present in the
equipment by the company's professionals, at the same time that the instrument
performed the analysis of the liquors, thus allowing a correlation between the
results.
All values of EA
(effective alkali), AA (active alkali), TTA (total treatable alkali), among
others, were analyzed by the Research and Development team, following the
company's standard procedures. Inorganic solids were calculated from the
difference between total and organic solids. In total, 103 samples of white and
black liquors were collected and analyzed, with the results obtained by the
laboratory and equipment compared.
Subsequently, all the
elements achieved with this research were analyzed and compared, and from
there, possible gains that the use of infrared technology can bring to the
industrial paper and cellulose market were evaluated.
Finally, the results and conclusions of the research were systematized
in the form of writing.
4.
RESULTS, ANALYSIS AND DISCUSSION
4.1.
Overview of the process
The cooking area
(Figure 1) consists of the chip feeding system, an impregnation vessel and
digester. The equipment of the Canadian company, FITNIR, was installed close to
the process so that six sampling points were taken from the licensing lines.
One point of analysis
is for white liquor and the remaining points comprise black liquors that
involve each stage of the digestion process: upper circulation, lower
circulation, extraction, bottom circulation and transfer.
The liquors are taken
through ½” 316L stainless steel pipes to the field equipment, called FSS (Field
Sampling Station). The lamps are incidents with infrared light and the
information is carried through the optical fiber to the SRS (Spectrometer Rack
System). All parameters of the licensees' properties are analyzed and forwarded
to the company's DDCS (Digital Distributed Control System).
Figure 1: General Schematic of the Process
As a resource for the
cooking operation, the area does not have an advanced supervisory control (APC
- Advanced Process Control). The control is currently done by the operators,
but it is expected to implement an APC for the automatic control of the
operation with the approval and proof of the functioning of the FITNIR
analyzer.
4.2.
Measurement
of liquor properties
For analysis of the
composition of liquors, the table below shows the parameters considered for
each type of liquor, as well as their description.
Table 1: Measurement of liquors from the FT-NIR
analyzer
Liquor |
Measurements |
|
White Liquor |
EA, AA, TTA, Na2CO3,
Na2S, Na2SO4, Na2S2O3,
%S, %CE |
|
Black Liquor |
ERA, Organic, Inorganic & Total Solids |
|
Weak Black Liquor |
TTA, Lignin, NaOH,
Na2SO4, Na2S2O3 |
|
EA = Effective Alkali AA = Active Alkali TTA = Total Titratable Alkali %S = Sulfidity (%) %CE = Causticizing Efficiency (%) |
Na2CO3
= Sodium Carbonate Na2S =
Sodium Sulfide Na2SO4
= Sodium Sulfate Na2S2O3
= Sodium Thiosulfate NaOH = Sodium
Hydroxide |
|
The
data for the analysis of black liquors made by the company's laboratory and the
equipment are presented in the graphs with the results obtained.
Figures 2, 3, 4 and 5
show the graphic results of black liquors for the parameters of effective
residual alkali (ERA), total solids, lignin and organic solids respectively.
The graphs on the left
(Actual vs. Predicted) compare the values provided by the FITNIR equipment with
the values provided by the Canadian manufacturer company when using its own
database with information collected in several companies around the world in
its more than 20 years of operation in the paper and cellulose market. A
determination coefficient, also called R², is presented as a measure of
adjustment of the generalized linear statistical model, in relation to the
observed values.
The graphs on the right
(Lab vs. FITNIR) compare the values provided by the FITNIR equipment with the
values analyzed by the company's laboratory. In addition, in addition to
presenting the determination coefficient, they present the RMSEP (Root Mean
Square Error of Prediction), as it is the simplest and most efficient measure
of uncertainty in NIR (Near Infrared) predictions. This value is a measure of
the average uncertainty that can be expected when predicting new samples. The
RMSEP is expressed in the same units as the predicted parameter. The results of
future analyzes can then be presented as predicted values ± 2 x RMSEP. This
measure is valid as long as the new samples are similar to those used for
calibration, otherwise the forecast error can be much larger.
Figure 2: Graphical analysis for the Effective
Alkali Residue of black liquor
Figure 3: Graphical analysis for total solids of
black liquor
Figure 4: Graphical analysis for black liquor
lignin
Figure 5: Graphical analysis for black liquor
organic solids
In general, it was observed that the measurement
results of the FITNIR equipment are close to those found in the laboratory,
showing a good correlation (R²> 0.95) and acceptable RMSEP values
(values above those obtained through current
instruments).
Table 2 shows the data for laboratory and equipment
measurements for white liquor. Again, some values were omitted
for reasons of confidentiality; however, the main parameters that make up this
study are as follows.
Once again, the results are as expected, showing
great approximation between the values and enabling a more
efficient management of the process. Management that involves the reduction of
chemicals used, better performance of the equipment for the certainty of the
good execution of the process, less maintenance and rework and so on.
Table 2: Results of white liquor
Based on precise
analyzes like these, it is possible to carry out a stable control of the kappa,
with increased sulfidity and reduced demand for chips
for the same pulp production rate. This better production management implies
the growth of gross income, which represents tons of wood saved annually.
A more stable white
liquor leads to less variation in the liquor load ratio (effective alkali) per
wood, which points to a more stable kappa and higher digester yield. A more
concentrated white liquor reduces its demand and the solids formed in the
digester process, providing a better black liquor as a result.
Although not
measured by work, it is conceptually mentioned that the stability of the kappa
number at the digester outlet provides great economic gains, in addition to
causing improvements in the following production processes.
4.3.
Optimization
of operation and control
The FITNIR analyzer provides information about the process, but alone
can only provide monitoring through the analyzed data. The process control
strategies need to be integrated with the measurements to obtain the expected
and estimated returns in this study, in addition to good production management
practices. Moreover, with that, it provides incentives for further industrial
development.
5.
CONCLUSION
The pulp and paper
mills have been aiming to improve operating results with greater efficiency in
the entire production system. Applications using FT-NIR spectroscopic
techniques can contribute significantly to the achievement of this objective.
Online measurement
of key process variables in the digester promotes opportunities to optimize its
management with less variation in actual production compared to the target set
and significant reductions in operational interventions in the cooking process.
In addition to the
tangible gains pointed out in this and other studies, such as financial gains
and several management and control opportunities, the use of the equipment in
the liquor lines leads to intangible gains such as: less maintenance (robust
instrument), elimination of chemical consumption and reagents, offering
multi-component measurements quickly and accurately, ideal for monitoring the
process.
As this is the
first project with the use of FT-NIR technology in South America, more studies
are needed, but the results obtained show the tendency that infrared technology
will continue to play an important role as a detection platform due to its
inherent flexibility and capacity and, combined with advanced control, becomes
a stepping stone for industry 4.0.
REFERENCES
ALMEIDA, H. C.; SILVEIRA, C. B.; ERNANI, P. R.; CAMPOS, M. L.; ALMEIDA,
D. (2007) Composição Química de um Resíduo Alcalino da Indústria de Papel e
Celulose (DREGS). Química Nova, v.
30, n. 7, p. 1669-1672.
CRUZ, M. P.; BARBOSA, L. C. A.; MALTHA, C. R. A.; GOMIDE, J. L.;
MILANEZ, A. F. (2006) Caracterização Química do “Pitch”
em Indústria de Celulose e Papel de Eucalyptus. Química Nova, v. 29, n. 3, p. 459-466.
DUGGIRALA, P. Y. (2005) Formation of Calcium
Carbonate Scale and Control Strategies in Continuous Digesters. CD del II Coloquio
Internacional sobre Celulosa de Eucalipto.
DURAN, N; FARDIM, P. (2007) Molecular Assembly in Kraft Pulping of Eucalyptus.
Repositório da Produção Científica e
Intelectual da UNICAMP, v. 68, n. 8, p. 98-108.
FAIX, O. (1992) Fourier Transform Infrared
Spectroscopy. In: Lin S.Y., Dence C.W. Methods in
Lignin Chemistry. Springer Series in Wood Science. Springer, Berlin, Heidelberg. p. 83-109.
FAVARETTO, F. (2001) Uma contribuição
ao processo de gestão da produção pelo uso da coleta automática de dados de
chão de fábrica. Tese de doutorado. Escola de Engenharia de São Carlos,
Universidade de São Paulo.
FORSSTRÖM, A.; BASTA, J.; BLOM, C. (2006) Kappa
ótimo de cozimento–uma ferramenta versátil para a melhoria do desenvolvimento
financeiro de uma planta de celulose de eucaliptos. O Papel, v. 67, n. 6, p. 96-100.
GRACE, T. M.; TRAN, H. (2007) The effect of
dead load chemicals in the kraft pulping and recovery
system. Engineering, Pulping and
Environmental Conference, TAPPI. 3, p. 2031-2064.
LOMBARDI, L. R.; LUIZ, A. C. (2017) Avaliação da Sulfidez
do Licor Branco na Qualidade da Celulose Kraft. O PAPEL, v. 78, n. 7, p. 76-80.
OTANI, M.; MACHADO, W. V. (2008) A Proposta de Desenvolvimento de Gestão
da Manutenção Industrial na Busca da Excelência ou Classe Mundial. Revista Gestão Industrial, v. 4, n. 2,
p. 1-16.
POMPERMAYER, C. B.; LIMA, J. E. P. (2002) Gestão de custos. Finanças
empresariais. Coleção Gestão
Empresarial, Associação Franciscana de Ensino Senhor Bom Jesus. n. 4, p.
49-68.
RESTREPO, A.; YUZAK, E.; RODRIGUES, L. A.; AFFONSO, R.; GEDRAITE, R.
(2009) Considerações práticas acerca da aplicação de sistemas híbridos de controle
avançado na indústria de celulose. O
Papel, v. 70, n. 12, p. 35-51.
SILVA, M. A. A.; LIMA, R. A.; VENTORIM, L.; FERREIRA, C. R. S.;
SARCINELLI, J.; AGUILAR, R. M.; MONFARDINI, M. A. (2016) Os Benefícios da
Estabilidade da Descarga de Polpa do Digestor para o Processo de Cozimento –
Estudo de Caso. O Papel, v. 77, n. 2, p. 72-80.
TERDWONGWORAKUL, A.; PUNSUWAN, V.; THANAPASE,
W.; TSUCHIKAWA, S. (2005) Rapid assessment of wood chemical properties and pulp
yield of Eucalyptus camaldulensis in Thailand tree
plantations by near infrared spectroscopy for improving wood selection for high
quality pulp. Journal of Wood Science,
v. 51, n. 2, p. 167-171, doi: 0.1007/s10086-004-0633-3
TRUNG, T.; ALLISON, B. (2015) Advanced Online
Process Analyzer for Chemical Recovery and Pulp Mill Control. O Papel, v.
76, n. 1, p. 47-56.
TSUCHIKAWA, S. A. (2007) Review of Recent
Near Infrared Research for Wood and Paper. Applied
Spectroscopy Review Journal, v. 42, n. 1, p. 43-71.