Proceeding Articles

The paper is intended as a tutorial introduction to some of the principles used in model building techniques, which place emphasis on  modelling of the process behaviour as shown to the outside world by records of inputs to and outputs from the process, rather than by attempting to model details of the physics or chemistry internal to the process.

The best way to achieve a well-controlled process is to consider the control during design of the equipment. The steps in designing for control are stated, then expanded in a discussion of the design of flow boxes. An analysis shows how the control of a flow box is affected by the geometry of the box and the characteristics of the air pad supply system.

Automatic controllers have been subjected to fairly intensive theoretical study and some attempts have been made to establish human transfer functions for certain manual control actions. The paper industry is fairly specialised, however, in that the response of the system is very slow. This paper is an attempt to study the human operators’ characteristics in the control of basis weight. The paper is in two parts: the first is an ergonomic study of the process of papermaking and the second is a detailed study of basis weight control. A simulator for basis weight can described that matches the actual operation of the papermachine and has proved extremely useful for training purposes.

The Industrial Democracy Project in Norway is a long-term research sponsored jointly by the Confederation of Employers and the Trades Union Council. The field experiment reported took place in the chemical pulp department of an integrated papermill as one of a series of four experiments carried out in different industrial settings. Extensive task fragmentation and bureaucratisation in modern industry have produced widespread feelings of alienation in the work force, owing to an increasing mismatch between technologically based task requirements and human needs. Emerging theories of socio-technical systems, including a list of psychological job requirements, offers a frame of reference for understanding these problems. Previous experience suggests that full commitment to productive aims can be achieved only under conditions that allow for a high level of self-regulation and learning. In process technology (including pulp and paper), the dependence relationships among the state characteristics of the materials form a complex network. In the present case, this resulted in uncontrolled variations being transmitted along the process. Having identified the optimum unit for experimentation, individual jobs were redesigned in order to facilitate group learning, which would permit the work groups to increase their control of the process. Results of the socio-technical analyses before and after the experiment are reported and reference is made to the variance matrix technique.

Despite considerable advance during the last twenty years in the technology of paper and papermaking, there has been scarcely any explicit consideration of the subject that underlies all of our knowledge and all of our problems in these fields the structure of paper. A brief survey of papermaking literature brings home forcibly our state of ignorance . None of the classic reference or textbooks includes even a reference to structure in its contents list or index. Even the few papers explicitly referring in their titles to structural considerations are indexed under alternative headings . An enquirer from another industry or technology might be forgiven for inferring that paper is a structureless body defying the normal laws of physical analysis and description!

This is an exaggeration, of course, for ideas about structure are implicit in a wide range of theories, discussions and practical operations on paper and papermaking, but the absence of explicit reference to structure does clearly indicate the superficial and empirical nature of our current knowledge and of the technology of paper and papermaking. At the same time, there are now many indications of a revolutionary change in this situation.

The statistical geometry of fibrous networks is described in terms of the fibre and sheet dimensions and geometric probability. The method has been developed for random- two-dimensional structures and extended to cover deviations from randomness (orientation and flocculation). It is also applied to a multiplanar structure as a first approximation to three-dimensional structures . Further approximations to three-dimensional networks are discussed. Experimental results for two-dimensional structures are presented.

This contribution illustrates the structure of paper by presenting a selection of light micrographs of the surfaces of papers. In addition to illustrating a wide range of papers, the effects are portrayed of progressive beating; coating; creping; machine glazing and the glazing of handsheets; parchmentising; supercalendering ; and watermarking. The two-sidedness of the surfaces of several papers is clearly brought out and the contribution of the surface of mechanical printing papers to their printability is illustrated.

Of particular interest are the effects of progressive beating on the density of the sheet, the closeness of its surface structure and its response to glazing. It was observed that cellulosic material has the ability to replicate fine scratches on a surface with which it is dried in contact.

An examination has been made of eight paper samples by means of the scanning electron microscope at the Pulp and Paper Research Institute of Canada. This is the first step in a programme of examination of paper surfaces and fractures that is planned for this instrument. The scanning microscope(^1,2) bridges a gap between the capabilities of the light microscope and the conventional transmission electron microscope and the results obtained in this study may therefore be of interest to those attending a symposium on paper structure . The capabilities and limitations of the two more conventional techniques are described in detail in other papers of the symposium and will not be discussed here.

A study of the morphology and energetics of fracture can shed some light on adhesion . Two approaches are made and both lead to the same general conclusions . In the first, the fracture of liquids of increasing viscosities or their cavitation leads to the ideas that fracture of polymers is governed by flaws or imperfections, not by molecular parameters. The second approach is that of Griffith, who postulated his now famous ‘crack theory’. The fracture of materials is again seen as a statistically controlled process dependent on the presence of flaws. Qualitatively, it is known that fracture of paper is explicable by stochastic considerations, too; however, quantitatively, the crack or flow theory fails.

In order to explain the mechanical failure of paper, it becomes necessary to apply the statistical approach to the bonds between the molecules . These are the hydrogen bonds joining cellulose molecules together. It is found that, if a variance is assumed to exist around the mean value of the bond energy, the rupture energy of paper can be related to its content of hydrogen bonds. Thus, the adhesion of the elements of paper is quantitatively explicable in terms of characteristics of this bond.

Using a technique developed by the authors, an examination by electron microscopy of coniferous wood tracheids has shown that the secondary alterations in the texture of the cell wall layers during drying play an important role in fibre-to-fibre bonding. Differences between late wood and early wood are indicated, together with the typical behaviour of each of the cell wall layers in such bonding. The effects of beating, of drying a sheet of paper and reforming it from the disintegrated pulp are dealt with, particularly regarding their influence on sheet strength.