Application Of Near Infrared Spectroscopy To Rapid Analysis Of Pulp Wood-chips Properties

Wood-chips are raw materials for the production of chemical pulp and chemical mechanical pulp.Some of the main characteristics of pulping chips,such as type,mixture,moisture content,cellulose and lignin content,will directly affect the quality of cooking and product quality,the rapid and accurate detection of wood features is the key to the quality control of pulp production.In this paper,a portable low-resolution near-infrared spectrometer was used for studying the feasibility of analyzing several wood-chips properties rapidly.On this basis,a set of on-line analysis system of pulp material feature which can achieve on-line detection was designed.This system was developed by delphi7,C# and other programming tools relying on INSION micro-near infrared spectrometer.The main work and conclusions are as follows:(1)Feasibility study on detection of water content in wood-chips by near infrared spectroscopy(NIRS).The near infrared spectrum of four kinds of common pulp wood(Eucalyptus urophylla,Eucalyptus globulus,Acacia mangium,Acacia crassifolia)and a kind of mixed wood-chips(Eucalyptus globulus and Acacia mangium)were collected by the domestic hadamard spectrometer,and then been used for establishing the moisture content prediction model.The results show that the wood-chip spectrum collected by low-resolution near-infrared spectrometer can be used to build a reliable moisture content prediction model after chemometric pretreatment.However,the variety of wood-chips and the corresponding mixed wood-chips which used in the current production should be included as much as possible when selecting the calibration set.(2)Feasibility study on detection of lignin and total cellulose content in pulp wood-chips by NIRS.First,the spectrum of the wood-chips were collected by a hadamard low-resolution near infrared spectrometer which is low-resolution.Then,the wood-chips were ground into powder,and collected the corresponding spectrum of them were collected by a Fourier near infrared spectrometer with high resolution.Finally,the quantitative prediction models of acid-insoluble lignin and total cellulose were established by these two spectral data sets.The results show that the low-resolution spectroscopy not only can establish comparative accurate prediction model of lignin and cellulose,but also achieve high-resolution spectroscopy effect.However,because the lignin content is difficult to be measured directly and prone to making error,people should pay more attention to the determination of the accurate reference value in the establishment of the model.(3)Study on the rapid identification of wood type by NIRS.Firstly,the principal component analysis of the near infrared spectra of wood powder of four kinds of wood was carried out,then the mixed wood-chips with different mixing ratio were classified and identified.The results show that there was a significant difference between the eucalyptus and Acacia mango spectra,which could be quickly identified by near-infrared method.The multiple scattering correction of the wood-chips spectrum could eliminate the difference of batches.The near-infrared method could rapidly identify the wood's mixing ratio.(4)Designing pulp wood-chips material feature on-line analysis system.The system based on German INSION micro spectrometer as a near infrared spectroscopy acquisition device,can be used for rapid qualitative or quantitative analysis in online detection of wood-chips.Besides the spectrometer,the hardware of this system includes light source,fiber,pulverized wood-chips pretreatment platform,etc.;the software includes spectrum acquisition platform and data processing system.The results of the tests in the laboratory show that it is feasible to rapidly detect the certain characteristics of the pulp wood-chips using the current system.