Getting a grip on the deinking process

PPI, Mar 1998 by McKinney, Roland

Control of the deinking process has become a far more sophisticated affair in recent years, but the industry still has a very long way to go

Athough recycling has become one of the popular causes of the 1990s, paper recycling has been practiced for hundreds of years and is one of the main reasons for the low survival rate of ancient documents. Centuries ago, manuscripts and works of art considered "inferior" were recycled, but many of these recycled artifacts would be considered priceless today.

Over the last 30 years, paper recovery and re-use has grown rapidly. In 1965, about 18 million tons of paper and board were recovered for recycling, mainly from industrial sources such as printers and converters. In 1980, this had increased to an estimated 44 million tons, but by 1996 it climbed to almost 120 million tons, much of which was post-consumer grades. Only a proportion of the paper that is re-used is deinked though. The fraction varies from country to country, depending on the structure of the industry, but in 1996 approximately 26 million tons of scrap paper were deinked worldwide.

Even though the "average" quality of recovered paper used in deinking mills has fallen as recovery rates have increased, the quality of recycled paper has steadily improved to the point where some products are almost indistinguishable from those made from primary fibers. One of the important contributors to better quality has been the use of process control systems.

Priming the system

Feedback control systems have been used for thousands of years. For example, the water clock of Ktesibios (circa 300 BC) used a float regulator. More recently, James Watt used a flyball governor to control the speed of a steam engine in 1769. The first online sensors in the paper industry were installed in the 1950s and these were followed by computer-based control systems in the early 1960s. Although computer process control is not new, it is changing as fast as new applications for microchips can be developed and it is one of the most rapidly developing fields in the industry.

From a control perspective, process variables can be classified as shown in Figure 1. Input variables act on the process and the levels of output variables are determined by the level of the input variables. Manipulated variables are inputs which are adjusted by a control system in order to affect output controlled variables. Other input variables are known, and can be measured, whereas others are not known. Uncontrolled output variables are not controlled either because they are not important, or because they cannot be controlled.

Controlled variables may be controlled to a target value, or set point, and a classic feedback loop incorporating a set point is illustrated in Figure 2. Feedback regulation is widely used. The controlled output variable signal is fed back and compared with the set point. The resulting difference (error) is used to adjust the manipulated variable, to bring the output variable closer to the set point.

Deinking control

Initially, control systems for deinking plants were developed from those used in pulp mills and on paper machines. More recently, complete systems and sub-systems have been designed specifically for recycling mills. For example, a model-based predictive control system for deinking mills was introduced in 1994 by Aspen Technology, while Honeywell introduced their Recycle Deinking Advanced Control Package at about the same time. New and more reliable sensors are being developed to help extend the capabilities of control systems, such as the Residual Peroxide Analyser from BTG. "Intelligent" sensors and control valves are also extending control capabilities, making this an area of constant change and improvement.

Overall, the function of a recycle line is to remove non-fibrous materials that are either likely to cause manufacturing problems or result in paper quality defects. In addition, the recycle line should:

minimize variations in quality created by using scrap paper from many different sources to give a consistent recycled pulp

maximize fiber recovery so that yield losses are minimized

minimize chemical use.

Consequently, there are variations in the systems used to produce recycled pulp for use in different grades of paper. A typical newsprint recycling line (including deinking) is illustrated in Figure 3. (For simplicity, only major process unit operations are illustrated.)

As is illustrated, deinking systems are made up of a sequence of unit operations, although only some of these are controlled. Some of the control loops are complex, while others are simple. Some of the important loops are described below. There can be as many as 35-40 control input variables in a complex deinking system, with numerous unknown input variables. As knowledge of factors affecting deinking develops, the number of controlled input variables will increase, provided suitable sensors are developed.

Pulp power

Pulping is a critical step in recycling and deinking. It determines how well fibers are separated from each other, which has a subsequent impact on yield. If contaminants are not well separated from the fiber, removing them will increase fiber losses. But if pulping conditions are severe, contaminants can be reduced in size, making their subsequent removal less efficient. In deinking, ink separation from fiber is primarily accomplished in the pulper, and its ease of removal in the first deinking stages is determined mainly by pulping conditions. In grades such as newsprint, the cost of chemicals used during pulping is high, giving considerable potential for cost savings by the use of computer-based control systems. So a control loop for pulping must not relate solely to the output variables such as consistency, temperature and pH after pulping, it should also be optimized with respect to chemical consumption and the efficiency of downstream unit operations.