January 28, 2003

Mechatronics Techniques Applied to Paper Recycling Research at NC State University

 

An experimental machine made with an inverted belt sander is being used by Dr. M. K. Ramasubramanian of NC State University to develop a waste paper sorting system.  The green laser spot can detect the difference between white and colored paper. (Photo: M. K. Ramasubramanian)

Paper or plastic grocery bags?  Aluminum, plastic or glass soda containers?  Paper towels or rags for that spill?  Decisions we make every day have an impact on the waste stream.  With landfill space at a premium, these decisions take on added importance.  Paper recycling is one area where such decisions affect the quality of our environment and our lives.

According to the American Forest and Paper Association website, Americans use more than 100 million tons of paper each year, 45 percent of which is recovered by recycling.  More paper is recovered in the United States for recycling than all other materials combined.

With this high level of paper recycling activity, the process should be well developed and efficient.  But decisions relating to paper use and reuse are legion and complex, and they could be improved.  Separation of different grades of paper is one challenge for municipal solid waste managers.  Should the consumer be required to separate white paper, colored paper, newspaper and glue-contaminated paper?  Can the paper be sorted more efficiently and accurately at a paper processing facility?  What mechanisms can be used to improve this process?

Enter Dr. M. K. Ramasubramanian, associate professor of mechanical and aerospace engineering and director of the mechatronics program in mechanical and aerospace engineering at NC State University. Ramasubramanian has been working on a project that he hopes will result in a machine that can sort paper for recycling much more efficiently than the methods currently in use.

According to Ramasubramanian, “Paper is a significant component of the solid waste system; 40 percent of municipal solid waste consists of paper and paper products.”  The U.S. Department of Energy (DOE) has included sorting waste paper for recycling in its Agenda 2020 program, which is a vision for our energy utilization in the year 2020 in the U.S.

As a participant in Agenda 2020, Ramasubramanian has been studying paper sorting for several years.  “Each kind of paper has an entirely different chemical composition,” he explained.  “Different grades of paper require different processing techniques during recycling.”  Lignin is the key ingredient in paper.  Some papers, such as newsprint, have a very high lignin content, while others, such as fine writing paper, have low lignin content.  Recyclers cannot mix papers with high and low lignin content.

Dyed or colored paper presents another set of problems, as does glossy, plastic-coated or glued paper.  Color, for example, must be removed from a piece of paper before it can be recycled.  The quality of a bale of paper meant for recycling depends on the ‘purity’ of the content,” said Ramasubramanian.  “To achieve this content purity we are developing a new sorting process.”

Currently, paper for recycling is sorted manually by workers standing along a conveyor belt, but the process is very slow and expensive.  The conveyor belt can only run at 75 feet per minute to allow the sorters to do their work. According to Ramasubramanian, “A lot of paper that could be recovered goes into the landfill just because of the processing cost.”  Ramasubramanian and Dr. Richard A. Venditti, assistant professor of wood and paper science at NC State, aim to develop an automated system that will move the conveyor at 1,200 feet per minute and will sort the paper using a machine based on mechatronics.

Mechatronics combines sensors, actuators, microprocessors and control algorithms in one interactive unit to perform tasks.  For the paper recycling system, sorting must be based on lignin content, dye in the paper and glossiness.  “Plastic-coated paper must go to the landfill because the plastic is a contaminant in the recycling process,” said Ramasubramanian.  Although sensors for gloss detection in moving objects as well as high speed vision hardware are available, the challenge is in adapting them to recognize various types of paper in real-time, based on color and gloss through software design.  Much of his recent research has focused on developing a high-speed vision processing system that can detect colors.  A high-speed camera creates images of the different sheets of paper in different orientations flying by the sensors.  “We do real-time image processing to identify the shape geometry of color boundaries,” said Ramasubramanian.  “We must track not only the color but also where it is on the conveyor for the actuator to grab and put that sheet in a separate bin, making a complex processing challenge.”

In the laboratory Ramasubramanian and his research team have created a test rig comprising a small inverted belt sander that can run 750 feet per minute with a vision system mounted above the sander.  With this unit they have had success with tracking white paper as it goes under the vision system and with identifying colored papers. 

The biggest challenge in developing this system was identification of the lignin content by non-contact methods in a dry sheet of paper passing under a sensor.  Two sheets of paper may look white, but the lignin content can be at the two extremes, and they cannot be mixed in recycling.  “We have developed a unique sensor that is laser-fluorescence-based,” said Ramasubramanian.  “If you excite lignin molecules with laser light in the visible region — we use 532 nanometers — they will fluoresce at a higher wavelength, with a peak at 630 nanometers in this case.  We have developed a system that amplifies the fluorescence yield and produces a meaningful analog output quantifying lignin content in real-time.  For the first time, we are able to correlate the intensity with absolute lignin content in real-time,” Ramasubramanian said.

The next phase of the research, scheduled to take place over the next four years, involves building a miniconveyor that will contain the completed identification system as well as the actuation system that will sort the papers.

Advantages of Ramasubramanian’s system include energy savings realized by recovery of fibers that might have gone to the landfill, saving landfill space and saving virgin pulp and forests.  After his preliminary study is complete, paper recycling companies in Washington state and Maryland have agreed to let Ramasubramanian use their facilities for large-scale testing.  The companies realize that high-speed sorting makes both business and environmental sense as well as yields energy savings.  “The economic benefits are very clear with high-speed automation,” said Ramasubramanian.

Because photocopy, or xerographic, ink is bonded onto the surface of paper using a plastic coating (thermoplastic fusing), Ramasubramanian calls photocopied paper a “recycler’s nightmare.”  Ramasubramanian is involved in another project, funded by the National Science Foundation, in which he is experimenting with removing thermoplastically fused inks from paper surfaces, which is a major problem in recycling xerographically and laser printer papers. Ramasubramanian hopes to use a process called acoustic microcavitation to release xerographic ink from the paper prior to the pulping process.  “This novel machine could be like an ultrasonic eraser,” he said.  “I could envision one beside every photocopier — sort of a “decopier” that would remove the ink and stack the cleaned paper, ready for recycling.”  This idea of sorting and deinking at the source is revolutionary, but possible.

Many challenges exist with recycling and product recovery processes, but as resources grow scarce and the consequences of overflowing landfills increasingly affect the quality of our lives, they are challenges we must face. 

— rudd —

Media Contacts:

Dr. M. K. Ramasubramanian, (919) 515-5262, rammk@eos.ncsu.edu

Linda E. Rudd, (919) 515-3848, linda_rudd@ncsu.edu