This invention relates to the detection of impurities in metal agglomerates
such as scrap metals and in particular to methods of screening scrap metals for
The recycling of metals has always been an important industry, and
it is currently becoming even more important. Metals such as aluminium, copper,
zinc, brass and steel are collected as scrap and are then reprocessed for re-use.
One example is the electric arc steel making process which uses scrap almost exclusively
for its feedstock.
Because of the wide variety of sources from which scrap may be collected
its quality can vary widely and the scrap may be mixed with a number of non-metallic
impurities or contaminants. Commonly found contaminants are ceramic materials.
Such impurities contain naturally radioactive constituents to a greater extent
than metals. Alternatively, there may be radioactive contaminant materials themselves
in the scrap. To determine quickly the amount of such contamination in a lorry
load of scrap when it is presented at a scrap yard is not an easy matter. One
particular problem is the shielding effect of scrap loads themselves. There is,
as is well known, a naturally occurring background level of gamma radiation and
therefore any detection system must be set to ignore this. Unfortunately it is
found that a load of scrap lowers the measured background level of gamma radiation,
the effect being greater for greater loads. It will be realised that this reduction
in the background level measured can allow some large loads to pass as uncontaminated
when they are in fact contaminated, if the detection system is set to pass uncontaminated
small loads. Any adjustment of the detection system to compensate, runs the risk
of rejecting uncontaminated small loads. It is therefore an object of the present
invention to provide a method of analysing the quality of scrap metal with respect
to contamination quickly and easily and reliably.
Previous developments have included the system described in AU-B-0541125,
which does not concern itself with the case of radioactive contamination of a non-radioactive
mass, but with a radioactive ore, a somewhat different problem, and EP-A-0174797,
which assumes that any radioactivity of a strength to be of interest will be clearly
greater than that of the background.
According to the invention there is provided a method of detecting
the presence of radioactive impurities in a metal agglomeration as defined in claim
The agglomeration may be a load of scrap metal. The emissions from
the agglomeration may be from non-metallic contaminants, and the energy bands may
be selected from those characteristic of potassium, actinium, bismuth, lead and
thallium. The method may include measurement by scanning by a matrix of gamma ray
detectors disposed around the agglomeration. The level of the emissions measured
with the agglomeration present may be used to estimate additionally the quantity
of radioactive impurities present.
The invention will now be described by way of example only with reference
to the accompanying drawings in which:
- Figures 1 and 2 are diagrammatic front and side views of a lorry being analysed;
- Figure 3 is a possible arrangement for the analysis of scrap on a conveyor
- Figure 4 is diagram of gamma ray energy bands measured in the detection and
As has been explained the analysis of scrap for non-metallic content
is not easy and there is a requirement to be able to do this reliably, simply and
quickly in scrap yard environments. The method adopted is to measure the gamma
ray radiation from a particular load of scrap on a lorry or railway truck or from
a particular position on a conveyor belt or indeed in any other situation in which
scrap is being handled.
The method relies in part on the fact that in general the non-metallic
contaminants do have a level of radioactive activity, albeit quite low in absolute
terms. Typical levels are 0.3-1.0 Becquerels/g. The impurities commonly found
are glass, concrete, dirt or slag and it is known that these contain traces of
uranium, thorium and their decay products as well as potassium. Gamma radiation
of the isotopes and energies shown in Table 1 will be found. These are all relatively
high energy gamma rays. In general, metals contain extremely small amounts of
radioactive isotopes and therefore have a significantly lower gamma ray activity
than non-metallic materials of typically ten times lower. Any significant detection
of radioactivity therefore indicates contamination of the metal. It is also known
that metals in the dimensions and quantities of interest with respect to the invention
do not wholly absorb gamma radiation. Therefore either the radioactive contamination
of the non-metallic impurities are those generating the preponderance of such gamma
radiation in a load of contaminated scrap and the radiation being emitted, or
the radioactivity is due to a specific radioactive contamination of the scrap,
rather than the natural radioactivity of non-metals.
Turning now to Figures 1 and 2 which show respectively a side and
front view of a road vehicle containing scrap diagrammatically positioned between
a typical arrangement of radioactivity detectors. The vehicle 1 contains scrap
2 which may contain non-metallic impurities. The six detectors 10 to 15 are arranged
in a matrix around the lorry and the output from these are analysed into energy
bands and the totals in each energy band are summed. The detectors may for example
be caesium iodide scintillation counters or semiconductor counters such as high
In Figure 3 the arrangement of detectors about a conveyor belt carrying
scrap is shown. The belt 20 is carrying a supply of scrap, for example in bales
21 and the detectors 22 and 25 surround the belt. The detectors are connected
in a similar way to those of Figures 1 and 2.
The information provided by the detectors in either case can be used
in two particular ways. First, where there is a considerable amount of non-metallic
contaminant or a heavy radioactive contamination in the scrap itself the overall
reading will be relatively high, above a predetermined threshold, and although
a detailed analysis could be made by the method outlined later in all probability
the high level of contaminants shown would mean the rejection of the scrap load
or its down grading.
Where the level from non-metallic contaminants is less than the predetermined
threshold the level of ionising radiation picked up will be significantly affected
by the naturally occurring background radiation and therefore it is necessary
to resort to some simple signal processing in order to estimate the level of contamination.
As has already been mentioned, to add to analysis problems the actual measured
background level will vary with the size of the scrap load. What we do is analyse
the radiation received by energy bands. The energy bands in which radiation is
expected from typical contaminants is shown in Table 1 and by computing the amounts
of these isotopes on the basis of the radiation found at these specific energies
a level of contaminants present can be calculated.
Turning to Figure 4, this shows in (a) the normal background spectrum
that might be found for gamma ray radiation between 20KeV and 1.5MeV, the intensity
for each band being shown by the height of the respective bar shown at (1).
In (b) a scrap load has been introduced. The levels of background
radiation measured have fallen from the levels (as shown in (a)) illustrated by
the dotted lines to new levels shown solid. However, the whole spectrum falls
in the same proportion. If radioactive contamination is present it will occur at
specific energy levels and therefore will increase the intensity of some, but not
all bands. This is shown in Figure 4(b) where the energy band 5 was originally
at intensity 1 without the scrap load present, and would have fallen to the intensity
2 when the scrap load was introduced without any contamination. But this load
is contaminated with a gamma ray emitting contaminant, whose characteristics energy
falls in the band 5. The band 5 intensity rises to the level 3, which can be detected
as being a level out of ratio with the other energy bands, the ratios being of
course determined by the relative intensities with no scrap load present.
These measurements will in general give the amount of non-metallic
material or specifically radioactive contaminant present in a load of scrap on
a truck or at a particular point on a conveyor line, for example. In order to
estimate the percentage of contamination the details of the bulk density, total
weight and cross-sectional area of the scrap load itself may be needed. This might
be conveniently obtained for example from weighbridge date and type of truck,
and, for example, it would be possible to arrange for an operator to enter the
make of a truck, for the weight automatically to be supplied from the weighbridge
and for the measurements of the radioactivity to be made simultaneously. The density,
volume and cross-sectional area of the scrap could all thus be determined automatically
and an estimate of the percentage contamination of non-metallic material in the
scrap could be obtained.
1.12 and 1.76