Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

Bright Fibre Delivered Laser Beams for Size Reduction


The Use of High Brightness Fibre Delivered Laser Beams for Size Reduction

Paul Hilton and Ali Khan
TWI Ltd, Granta Park, Great Abington, Cambridge CB21 6AL, UK

SFEN Decommissioning Challenges - Industrial Reality and Prospects 5th International Conference & Exhibition 7-11 April 2013


This paper will describe recent advances and work to highlight the advantages (and problems) when applying high power laser cutting to aspects of decommissioning and dismantling in the nuclear sector. Work to describe single sided cutting of tubular structures using a 5kW laser, with the objective of size reduction for long term storage, was presented by TWI at ENC 2010. Since then further work to establish the cutting capability of the same laser for plate materials and structures such as rolled and fabricated beams and other welded constructions, such as waste storage containers, has been completed and the results will be presented. In addition, video of the process for selected potential applications will be shown. Further work has linked the laser cutting head to a snake arm robot, and with this system a demonstrator set up was constructed to show the cutting head entering a cell through a typical access port, cutting a hole in the side of a simulated containment wall and then proceeding to cut out a selection of tubes to gain access to a part to be removed. Again, video footage of this demonstration will be shown. The paper will also discuss aspects of establishing a safety case for use of high power laser cutting in nuclear environments.


Laser cutting is just one of a set of tools which may be useful in aspects of decommissioning, particularly that of size reduction for long term storage. Laser cutting can be compared to both mechanical cutting and also other thermal cutting processes used in decommissioning such as plasma cutting. Some of the benefits of laser cutting, when compared to mechanical cutting, include the lack of a reaction force between the cutting head and the material being cut, which means that lightweight deployment systems can be employed and as a result the mass/volume of secondary waste produced can be kept much smaller. Compared to other thermal cutting processes, laser cutting generally produces less cutting debris (due to the small kerfs possible) [1] and less fume, thus reducing the load on ventilation systems. Laser cutting also allows for single sided cutting of tubular structures and pipework [2] which is a big advantage compared to processes which must rotate around a tube in order to be able to cut it. In laser cutting where the capability to simply sever the material is more important than maintaining cut quality, the tolerance to cutting head standoff can be arranged to be high by employing high brightness lasers and long focal length cutting optics. In addition, with some thermal cutting processes, standoff constraints and the physical geometry of the cutting heads reduce flexibility of applied cutting paths. The large standoff tolerance and cut path flexibility offered by laser cutting is highly beneficial for remote cutting operations.

The reliability of today’s high power industrial lasers has been proven in the automotive industry. Very high powers (50+kW) can be transmitted down flexible fibre optics to a lightweight cutting head. Such fibre optic cables can be long and employ in-line connectors. This means that not only can the relatively expensive laser source be maintained in a clean environment but also that only relatively small amounts of fibre optic need to be used in the active areas. As the laser and the main delivery fibres can be used repeatedly, this reduces the potential cost of multiple decommissioning activities. There are also some current disadvantages to laser cutting, which are still being addressed. Amongst these includes the generic development of a safety case for use of laser cutting and within this, some of the important issues include management of the laser beam energy transmitted through the part being cut and the temperatures generated in the process. This paper will describe additional laser cutting information on plate materials and then go on to describe work performed to assist development of a safety case before concluding with descriptions of two significant demonstrations of the application of remote laser decommissioning.

Laser Cutting of Plate Material

Experimental procedures

In this series of experiments, the beam from a 5kW laser was used to cut carbon manganese steel plates, between 6 and 50mm in thickness and stainless steel plates, between 6 and 25mm in thickness. The beam from the laser was transmitted to the cutting head in use using a 150micron diameter optical fibre. In such laser cutting systems, the laser light arriving at the cutting head from the optical fibre, first expands as it leaves the fibre and is then made parallel by an optical collimation system. An additional optic then focuses the laser light to a very small spot to create the power density needed for cutting. The TWI heads developed for decommissioning use lenses of long focal lengths, (250 and 500mm), to provide a low laser beam divergence, and hence high depth of focus. The corresponding minimum focal spot diameters for these lenses were 0.3 and 0.43mm respectively.

For both lenses, the laser beam was focused through a cutting nozzle tip with an exit diameter of about 3.5mm, in which the beam was centrally aligned. Each nozzle provided a distance of 15mm between the laser beam focus and the extremity of the nozzle tip. In conventional laser cutting, these distances are only of the order 1mm. In this paper, the ‘standoff distance’ is defined as the distance between the nozzle tip and the material surface being cut. The ‘focal position’ is defined as the distance between the laser focus position and the surface of the material being cut. Compressed air was used as the cutting assist gas through the cutting nozzle. In use, each process head was mounted on the arm of a Kawasaki articulated arm robot.

During experiments two laser powers of 2 and 5kW were used with each focusing lens and the standoff distance was varied between 5 and 25mm. Cutting gas pressure was varied between 8 and 10bars. During cutting experiments the speed of the beam over the material was varied to identify the maximum cutting speed needed to sever the material in two. The basic definition of an acceptable cut in relation to decommissioning is simply the ability to separate a structure, irrespective of the cut quality. Cut surfaces that were re-fused (welded) or obstructed from free fall by the attached dross were considered as not cut. In the results below, the ‘maximum cutting speed’ is the fastest speed that achieved a cut as defined above, in a single-pass of the beam.

On the various materials investigated, to optimise a particular cut or cutting sequence, the procedure involved fixing the laser power and assist gas pressure, setting the standoff distance between the nozzle tip and the material surface (thereby adjusting the position of the point of maximum power density), and then adjusting the cutting speed until the material could not be cut in a single-pass. Using the maximum attainable cutting speed for each sequence, a slot was then cut for kerf width analysis and material removal rate calculations.

Results and discussion

Figures 1 and 2 show the results of plotting the maximum laser cutting speed against the material thickness for the CMn steel, at the two laser powers used and for each focusing lens. Two extremes of standoff distance are also included, 5mm (ie close to the nozzle tip) and 25mm (ie far from the nozzle tip). These results show, as expected, that the cutting speed drops (for either lens) as the material thickness increases. However, the differences due to choice of focusing lens are not particularly great. In addition, the results at the two extremes of standoff distance are also not large, indicating the high tolerance of the laser cutting process used in this mode of ‘severing’.

Fig 1. Cutting results on CMn steel
Fig 1. Cutting results on CMn steel b)

Fig 1. Cutting results on CMn steel 25mm stand-off (left) 5mm stand-off (right)

Fig 2. Cutting results on stainless steel
Fig 2. Cutting results on stainless steel b)

Fig 2. Cutting results on stainless steel 25mm stand-off (left) 5mm stand-off (right)

As seen in Figure 2, the trend was similar for cutting stainless steel, although higher cutting speeds were recorded for the thinner materials. The maximum cutting speeds obtained when using the 500mm focusing lens, determined for particular combinations of laser power and standoff distance, were used to cut simple kerf slots in the 6 and 12mm thickness C-Mn and stainless steel plates. Selected cross-sectional images of such kerfs for CMn steel are shown in Figure 3, and for stainless steel in Figure 4 at a laser power of 5kW.

Fig 3. Kerf cross-sections for 12mm thickness CMn steel (left) and 6mm CMn steel (right)
Fig 3. Kerf cross-sections for 12mm thickness CMn steel (left) and 6mm CMn steel (right) b)

Fig 3. Kerf cross-sections for 12mm thickness CMn steel (left) and 6mm CMn steel (right).

Fig 4. Kerf cross-sections for (left) 12mm thickness stainless steel and (right) 6mm stainless steel
Fig 4. Kerf cross-sections for (left) 12mm thickness stainless steel and (right) 6mm stainless steel b)

Fig 4. Kerf cross-sections for (left) 12mm thickness stainless steel and (right) 6mm stainless steel

The kerfs were sectioned and their areas were measured. Material removal rates (g/min) were calculated for each kerf. The material removal rate can be considered as the product of the material density, the kerf cross-sectional area and the cutting speed. These properties are influenced by the ratio between the combined energy input from laser beam absorption and gas-jet momentum to the energy released during the melting of the material. The calculated material removal rates are presented in Figure 5, for both stainless and CMn steels and at both laser powers investigated.

For constant laser focusing conditions, this relationship suggests that the material removal rate is directly proportional to the combined laser and gas-jet power, and the physical properties of the materials being cut. This can be seen in Figure 5, where higher laser power generates higher melting rate which is easily expelled by the assist gas-jet. Differences in material properties between C-Mn and stainless steels will result in different melting rates, which can affect the melt expulsion efficiency. The difference in material removal rate between C-Mn and stainless steel is most dependent of the chemical composition, and the difference appears to increase with laser power. Analysis of the kerf cross-sections indicated that the main difference in the melt ejection rates between the two materials could be attributed to side-ways burning of the C-Mn steel during cutting. Such burning probably affects the melt flow.

Fig 5. Material removal rate as a function of focal positions corresponding to standoff distances between 5 and 25mm for 6 and 12mm thickness material and two laser powers
Fig 5. Material removal rate as a function of focal positions corresponding to standoff distances between 5 and 25mm for 6 and 12mm thickness material and two laser powers b)

Fig 5. Material removal rate as a function of focal positions corresponding to standoff distances between 5 and 25mm for 6 and 12mm thickness material and two laser powers.

The side-ways burning of the surrounding material is likely due to the exothermic reaction between the higher carbon content in the C-Mn steel and 20% oxygen in the compressed air gas-jet. Side-ways burning was also present in 12mm thickness C-Mn steel, but was not as pronounced as seen in the 6mm thickness material. Nevertheless, this did result in changes to the kerf cross-sectional areas, which are reflected in the material removal rate results. As presented, the results show the removal rate to be much higher for 5kW than 2kW but it must be remembered that at 2kW, the cutting speed is significantly reduced. In remote laser decommissioning of contaminated nuclear components it is desirable to achieve minimum material removal rate.

This implies that the kerf has to be narrow and the subsequent cross-sectional area needs to be small. The result will produce a minimum level of fume and will result in longer operational life for cell filtration systems. Such optimised cutting is probably possible when size reducing flat structures but might be difficult when cutting pipework from one side or more complex shapes. Figures 6, 7, 8 and 9 show some of the geometries decommissioning cutting techniques have been established for.

Fig 6. Section from a curved vessel

Fig 6. Section from a curved vessel

Fig 7. Method to dismantle I beam

Fig 7. Method to dismantle I beam

Fig 8. T profile in CMn steel cut in a single pass

Fig 8. T profile in CMn steel cut in a single

Fig 9. Structural concrete

Fig 9. Structural concrete

Safety Case Considerations

Before deploying any new technology is an active decommissioning environment, it is an obligatory requirement to asses all possible hazards that may arise. TWI has attempted to identify possible risks associated with remote deployment of high power laser systems by way of a paper based HAZOP assessment for using fibre delivered laser beams for size reduction in a typical active cell. This centred on three key stages of the decommissioning process, which included deployment of the laser cutting tool and laser beam generator, the cutting process itself and waste and waste recovery processes. The HAZOP assessment of the first two stages above highlighted the following for future attention if a safety case for using laser cutting is to be established:

  • Effects of stray laser beams passing through the material being cut.
  • Heat and temperature generation associated with laser cutting.
  • Assessment of the variety of materials to be cut and their interaction with laser light.
  • Release of possible contents (from items such as pipes or vessels being cut).
  • Fume generation arising from laser cutting.
  • Laser safety (health hazard to operators and others).
  • Unintended laser cutting, for example of something revealed behind a part being cut.
  • Maintenance of the laser cutting head.

Evaluations to address the above are currently on-going.

Demonstrator Activities for Laser Cutting in Decommissioning

This section will address two demonstration activities TWI have recently been involved in. The first involved trials on painted mild-steel skips of the type normally used for storage of irradiated nuclear fuel elements and having a volume of about one cubic metre. Because of growing space limitations, many of these empty skips need to be size reduced. A possible solution is to place the parts from cut skips back into a skip of a similar size. As a result, a process which offers the flexibility to quickly cut skips to maximise resulting packing density is required. TWI have demonstrated cutting of such skips, using a 5kW laser beam and the type of cutting head described above, connected to an articulated arm robot. Using a pre-programmed cutting path, it was possible to reduce a single skip to resulting volume such that four size reduced skips could be comfortably stored in a single skip of the same original size.

The second is the ‘LaserSnake’ project. This project was undertaken in conjunction OC Robotics Ltd, to investigate a unique system combining the cutting power of a high power laser with the access capability and maneuverability offered by a snake-arm robot. The demonstration involved a selective and remotely-controlled approach to dismantling and decommissioning complex structures in hazardous and confined nuclear environments.

A mock-up cell (2.5m x 2.2m x 1.5m) containing a 1m long 150mm diameter access aperture, a pressure vessel wall and a subsequent arrangement of pipework, was constructed. The demonstration showed the system entering the cell, avoiding an obstacle and then cutting an access hole in the wall of the pressure vessel. The cutting head on the tip of the snake robot, then entered the pressure vessel to inspect the pipework and subsequently selected, using its on-board vision system, the targets that required cutting, before re-tracing its movements to finally withdraw itself from the cell.


  1. Pilot Guy et al., 2008: ‘Measurement of secondary emissions during laser cutting of steel equipment’, Nuclear Engineering and Design. Vol. 238, no. 8, pp2124-2134, August.
  2. Hilton P et al., 2010: ‘The laser alternative in nuclear decommissioning - tube cutting and concrete scabbling using the latest technology’, Nuclear Engineering International, Vol. 55, no. 672, July.

TWI would like to acknowledge the support of the UK’s Technology Strategy Board in completion of the above mentioned LaserSnake project.

For more information please email: