GB2257377A - Oil-spill recovery equipment - Google Patents

Abstract

Spilt oil is recovered from the sea by towing an inflated boom 1 - 5 across the surface so that oil is collected in a bag 7 which forms an extension rearwards of the boom along its whole length. Oil enters the bag through a gap between water-filled tube 5 and air-filled tube 4, past a valve 14 in the form of an inflatable tube divided into sections spaced along its length. Each section is individually inflated with air from tube 3 or allowed to deflate by a two-way control valve, in response to water or oil being detected in the region immediately upstream thereof by a corresponding pair of capacitance sensing wires 17, 18. Thus the bag 7 collects mainly oil. The ends of the boom are connected to manned tugs, small enough to be dropped, with the boom in a folded deflated state, from an aircraft. Streamers 22 dangling from fabric 6, which encloses the various tubes of the boom, assist coalescing of the oil globules collected. The two-way control valve comprises two valve discs (25, 26) mounted at opposite ends of a centrally pivotted lever to close either of two flow passages (Fig. 8), depending whether a solenoid coil around one passage in energised or not. The discs are mild steel washers encapsulated, with the lever, in rubber. A flat enclosing housing defines flow paths between the two passages (29, 30) controlled by the valve discs and a common passage (31) leading to a section of the inflatable tube 14. <IMAGE>

Description

IMPROVEMENTS IN AND RELATING TO OIL-SPILL RECOVERY The history of recent oil spills shows that the task of recovery starts too late and takes too long, that a depressingly small fraction of the oil is collected and that dispersants can leave residues which may be more ecologically damaging than oil itself.

It is useful to ask why this problem is so difficult.

Liquids should be easier to handle than solids and gases.

Oil should not be an exception. But at sea we have to deal with very large volumes floating as slicks which are much wider than the dimensions of convenient engineering objects.

Half a metre is a big pipe but 5 kilometres is a small oil slick. The problem gets worse as the oil layer gets thin compared to the inlet diameter of suction devices because they take in too much air and water and too little oil. The thickness of an oil layer in a slick is often only a millimetre or so. The problem gets worse still if the surface of the sea is moving relative to the collection equipment and of course it nearly always is. Finally the differences between oil and water are not very large and tend to reduce with time. Oil spill recovery is an urgent difficult job likely to be required almost anywhere at sea and at short notice. To summarise: oil slicks have large volumes, are wide, thin, mobile, remote and unpredictable.

What is needed is something which is also very wide but easily and rapidly transported over long distances, which is clever at telling thin layers of oil from water and which moves exactly in sympathy with the waves.

This invention relates to equipment which, in its preferred form, can be packed into volumes small enough to be dropped from aircraft, which can be inflated and linked so as to cover frontages wider than the biggest oil slicks, which has intelligent electronic discrimination between oil and water, which can store substantial quantities of oil for subsequent transfer to other vessels.

This invention relates in one aspect, to novel equipment for oil-spill recovery and to an improved method of recovering oil from the surface of a body of water.

In its broadest equipment aspect, it comprises an inflatable boom supporting a storage bag for recovered oil, the bag having a valve-controlled entrance means. Desirably the boom includes at least one air-filled tube supporting an upper surface of the bag and at least one water-containing tube supporting a lower surface of the bag. Conveniently there are a plurality of air-filled tubes increasing in cross-sectional size in the direction towards the entrance to the bag.

Suitably the valve is located between an air-filled tube and a water-containing tube. The valve is desirably electronically controlled to open on a sensing of oil at the entrance and to close on a sensing of water at the entrance.

Valve control on the basis of sensing dielectric constant is particularly preferred.

The valve-controlled entrance can utilise pressurised fluid power for its control and an air-filled tube can be the source of such power.

In its broadest method aspect, the invention resides in towing a floating boom on the surface of oil-contaminated water to scoop the oil towards a collection area defined by the boom and is characterised in that the collection area is a valve-controlled bag located beyond the boom in the direction opposite to the direction of tow.

Suitably the control of the valve to the bag is effected electronically based on a sensing of a property of water that is measurably different from the same property of oil. Conveniently, the said property is the dielectric constant.

Preferably the valve used in the method is located between air-filled and water-containing tubes and suitably there are a plurality of air-filled tubes disposed one after the other in the direction opposite to the direction of tow.

Conveniently the air-filled tube closest to the source of the tow is smaller in diameter than the other or others.

Usefully, streamers can be supported in the water just outside the collection area to trail in oil as it approaches the valve of the bag.

Most conveniently, a precursor of the method involves lifting the boom in a collapsed pack to the location of the oil-contaminated water by aircraft and inflating the boom on the water following dropping of the pack from the aircraft.

Suitably a Z-fold pack is used sized to be received in the hold of a Hercules C-130 H aircraft.

The invention will now be further described, by way of example, with reference to the accompanying drawings, in which: Figures 1 and 2 are plan views of equipment according to the invention in use to recover oil from a spill, Figure 3 is an enlarged sectional view of a bag defining the boom of the equipment shown in Figures 1 and 2, Figure 4 is a partial view from below of one end of the equipment shown in Figure 1, Figures 5 and 5A are front views of a fabric grill defining separated entrances to the bag of Figure 3, Figure 5 showing the grill partially collapsed and Figure 5A the grill fully erected, Figure 6 shows equipment packed for air lifting to a site of use, and Figures 7, 8 and 9 show one form of electricallycontrolled valve for the equipment.

The best way to combine air dropping and very wide frontage is to use a inflatable combination of fabric tubes filled with air or water.

Figure 1 shows a plan view of a boom which can be towed by tugs 2, 2A at each end. It will take the shape of a curve which will be something between an arc and a catenary, distorted by local wave action. A length of 110 metres round an arc of 90 degrees would sweep a 1000 metre width of sea. Although this would be wider than any present mobile oil-cleaning equipment it is still small compared to the sea space needed for large tankers and the size of slicks that they can make.

Figure 2 shows how the swept frontage can be further increased without limit by tying adjacent tugs 2, 2A to one another so that any width of oil slick can be cleared. Each tug can have side-thrust capability so that the total frontage of the boom can be adjusted to the width of the slick (shown schematically at 3 in Figures 1 and 2) or to follow the banks of an estuary.

Figure 3 shows a sectional view of the deployed structure. The arrangement illustrated is built round five long tubes 1 to 5. Many fabrics are made in rolls with limited width, perhaps 1.5 metres, but with lengths up to 2000 metres or even more. The largest two tubes can, therefore, conveniently be made from 1.5 m width fabric with a lap joint to produce tubes 5 and 4 of a diameter of 0.4 metres. The upper four tubes 1 to 4 will be filled with air to a pressure of about 5000 N/m2 above ambient. This will prevent collapse under a head of 0.5 metre of water. The inflation time for a kilometre length of 0.4 m diameter tube will be about 60 seconds with 10 kilowatts of compressor power. The lower tube 5 will be filled with water.The reduced diameters of the forward air-filled tubes 1 to 3 form a concave upper surface which will produce a downward force if the wind is blowing from the left. The large plan area and rather low mass and inertia will give the structure a high frequency response in pitch and heave. It will seem to be 'glued on' to the wave slope so that wave action will cause very little relative velocity.

The air-filled tubes 1 to 4 will be linked to the lower tube 5 by a loop of fabric 6 in the shape of a flattened letter "S" running from the top of tube 4, over the top of tubes 3 and 2, back underneath tubes 1 to 4, further for a distance of several metres to form a storage volume 7 and then turning back under tube 5 finishing at its top.

Two strong braids 8 and 9 run the full length of the equipment. Their working load will be about 7 tonnes and they should be made of a light material (such as Parafil or Dyneema) which will not stretch under load more than the safe extension of the fabric. A flat braid is more suitable than a round section rope for reasons of packing and folding discussed later. One braid will run inside tube 1 slightly forward of the six o'clock position and will be retained there by a bonded strip. The second braid will run inside tube 5. The forward braid will normally take about 0.75 of the towing force but either must be able to take the full load alone.

In Figure 4 (viewed from below) light ties 10 made from a fine nylon rope in a W form can be seen running from gussets close to the tow braid of tube 1 to the tow braid of tube 5. The pitch of the W will be about 1 metre. At a tow speed of 1 metre second the working tension in these ties will be less than 1000 Newtons, even if the entire towing force is provided from the braid 8 in tube 1. Similar ties could be used between tubes 4 and 5 but in normal operation they would remain slack.

In normal operation the boom will be moving to the left in Figure 3. As the drag will be dominated by the water-filled tube 5, the ties 10 will be in tension and the geometry of the inflated structure will be well defined.

The gap above tube 5 will be filled with a structural member which must hold the tubes apart at a distance of about 50 mm and which must also allow the passage of oil.

It can be made from z-shaped lengths of fabric 11 bonded alternately to two other flat strips 12 as shown from the front in Figures 5 and 5A. It can pack flat as shown in part in Figure 5.

When the structure is under tow, there will be a downwards force on tube 5 of about 50 Newtons/metre so that the fabric 11 will be kept in tension and the tubes 3 and 5 pulled apart. At the entrance to the valve section there will be a positive pressure tending to push fluid into the oil storage volume. It will have a steady component due to the stagnation pressure of the forward velocity and fluctuating component due to wave action. There will also usually be a small negative pressure inside the oil bag because of its drag, which gives some useful suction. Aft of the zig-zig structure of fabric 11 at the entrance to the storage bag 7 is a pneumatically operated flap valve 14 which can admit oil or exclude water. A section of the valve is shown in Figure 3. It consists of a length of fabric folded over on to itself. When inflated it will fill the gap between tube 4 and tube 5.Even when uninflated, it will serve as a check valve to prevent the flow of fluid out of the storage volume 7 if the pressure gradient should ever reverse.

At intervals of a few metres along the length of the flap valve 14 are short lengths of cylindrical closed-cell foam to divide the valve into several hundred separate compartments. Each valve compartment can be independently inflated according to whether or not there is oil or water at the entrance to the gap between tubes 3 and 5. The dielectric constant of most oils is about 2 while that for water is 80. Local measurements of the capacitance between insulated wires can be used to operate an electromagnet valve which sends air to each flap valve compartment from tube 3 which serves as a common air line for all the flap valves. Inflating a flap valve will block one of the entrances to the bag when water is present. Venting the flap valve will admit oil.It can be helpful to delay the opening of the valve until the entrance region is almost all oil whereby the retained water fraction is kept reasonably low.

Figures 7, 8 and 9 show in plan and in sections on lines VIII-VIII and IX-IX are valve pack 50.

The valve pack 50 exhibits three ports. An entrance port 29 is connected to air tube number 3. An exit port 30 leads to the interstitial space between air tubes 3 and 4 which is connected to atmosphere. A switching port 31 is connected to the fabric flap-valve 14 which controls the entrance to the oil storage bag 7.

The electromagnetic valve elements of the pack 50 are two large mild steel washers 23 and 24 with conical forms 25 and 26 pressed in the area of their holes. The washers are mounted on a see-saw (not shown) which can pivot about an axis 27.

If capacitance-sensing wires 17 and 18 (see Figure 3) read a sufficiently low value (indicating an accumulation of oil in the gullet) an operating current is led to a coil 28 and will attract washer 23 so as to close entrance 24 while pulling washer 29 away from its seating so as to open the t exit port, 30. There will now be a passage between the switching port 31 and the exit port 30. This connects the fabric flap-valve 14 to atmosphere via the interstitial space between the air tubes 3 and 4. The air in the flapvalve 14 is released, the flap-valve collapses and oil is pushed into the storage bag 7 by the stagnation-pressure of the towing velocity.

When most of the oil has been removed from the gullet, the capacitance between wires 17 and 18 will rise. The current in coil 28 will be cut off. Air pressure on the washers 23, 24 will swing the see-saw so as to move washer 23 away from its seating so as to open the entrance port 29 and, at the same time, to move washer 24 towards its seating to close exit port 30. Air can now flow from the air tube 3 through entrance port 29 into the fabric flap-valve 14.

The fabric flap-valve will inflate and so close the entrance to the oil storage bag. This state will be maintained until a fresh accumulation of oil in the gullet is sensed.

The frame of the casing of the valve pack 50 can be made from two light-alloy channel pressings 32 and 33 which are moulded into a rubber or plastics housing 34. This can also protect the electronic circuitry and the drive coil.

The washers and their see-saw are also encapsulated in rubber which desirably is thin in the region of the magnetic path so as to reduce the need for high ampere-turns in the drive coil 28.

Some flexibility in the see-saw will improve valve seating.

Power to drive the capacitance-sensing circuitry and the electrically operated valves is supplied along wires 21 and 22. Conveniently generators located in the tugs power these valves. The voltage can start at, say, 48 volts with local sub-regulation to 24 volts at the valves. It is desirable that the electronic pack 50 is kept very thin and a design no more than 20 mm thick is convenient. Careful attention to waterproofing is required. It is fortunate that total reliability is not needed. If a few flap valves fail in their closed condition oil can flow along the boom to another entrance slit where the valve is open. If some valves stay open, the result is a proportional increase in the retained water fraction when clear water is being swept.

The problem of separating oil from water becomes more difficult as the difference between the density of the oil and water gets small. A downward flow velocity will drag oil droplets against their small buoyancy force. The limit for spherical droplets using the Stokes formula for viscous drag can be calculated. The force on a sphere of diameter D in a fluid of viscosity P moving with velocity Vz is F = 3 IT p D Vz which must be less than the buoyancy force, uD3 Ap g 6 If the droplet is not to be dragged down it is necessary to have: VZ < D2 Ap g 18p [p for water at 100C is about 0.0013 falling to 0.0009 at 30 C.] A first rough estimate of the vertical component of water velocity can be made by assuming that the stream lines are formed from 2 equal arcs tangential to the forward and aft flow lines. If this were true the maximum descending velocity Vz would be about 0.75 of the towing velocity VT.

This allows us to express the droplet escape diameter De in terms of density difference:

This means that at a temperature of 300C and a density difference Ap of 100 kg/m3 all spherical droplets greater than 3.3 mm in diameter should be retained. Similar results should apply for the carrot-shaped drops sometimes observed.

Increasing the diameter of the water-filled tube 5 without also increasing the size of the air tubes will paradoxically increase the amount of oil escaping because of the greater descending velocity. This is why some boom designs lose oil.

This simple analysis will be wrong if the flow is turbulent and the problem will become more complicated for other structures which do not move in such close sympathy with the waves. Nevertheless it illustrates the desirability of low forward velocity and small dimensions for the water tube.

The separation process can be aided by exploiting viscosity and stickiness. Any oil in contact with the underside of the S-shaped fabric 6 will flow more slowly and so form a layer which is thicker than the oil layer in the sea in front of the boom. The effect can be increased by the attachment of streamers 22 to the underside of fabric 6 in the region between tubes 1 and 2. They should not be quite long enough to reach the gap between tubes 3 and 5.

One strand (eg of knitting wool) every 100 mm would be an acceptable arrangement. Big sticky globules should in turn catch smaller ones.

Some water will inevitably be collected along with the oil but with a reasonable density difference partial separation will occur inside the bag. The option of releasing water through intelligently operated valves in the bottom of the storage volume 7 will allow periodic release of trapped water.

Packed volume A Z-fold pack is convenient for air dropping because it has a rectangular rather than cylindrical shape and because it can be packed from both ends. The hold dimensions of the Hercules C-130 H are: Length 12.62, Width 3.07, Height 2.74 metres. [In the later stretched L-130-30 Super Hercules the length is increased to 16.89 metres but this version is less common.] Cargo size must allow a clearance of 75 mm all round together with either a gap of 350 mm along one side, or a gap of 760 mm square over the top. The latter seems more suitable so that the working dimensions of the base of the equipment should fit.

Length 12.47 m, Width 2.92 m, Height 1.98 m.

Either side of the 760 mm square is room for a bridge to hold the package together. If 2 metres is allowed at each end of the 12.47 metre length for the tugs there is 8.47 metres of folded fabric between the tugs. For a boom with a length of 1100 metres to be accommodated in the space available means that the thickness for each fold must not exceed 2.92 x 8.47 1 1100 = 22.5 mm.

The crew of the aircraft would be unhappy if this rather long pack were to tilt and jam half way out of the hold so it will be necessary to plan the dropping operation carefully. It is desirable to aim to keep the pack horizontal and the acceleration forces low.

The Chinook helicopter can carry an underslung load and so makes less demands on package volume. It can just carry an 11 tonne load but with only 1 hour flight time.

Fabric A fabric material which is thin, strong, moderately impermeable and easily bonded is desired. The present preference is a nylon cloth with a layer of polyurethane sealing. In one example, type N4084/4 from Carrington Performance Fabrics, a thickness of 0.3 mm has a tensile strength of about 24,000 Newton/m along the roll length and 18,000 across it. This fabric is much lighter than the material normally used for oil-booms. However it gives a large factor of safety (18 times) for the expected pressure requirement and thin fabrics are stressed less than thick ones by folding. There seems to be no obvious justification for anything thicker. The cost in 1991 is 4/m2. This material has been used for a water-retaining gusset suffering severe flexing at the wave-frequency for a critical part of a wave tank which has been used for more than a decade.

If the thickness of the electronics package and any other hardware can be kept to 20 mm or less there would be room for 8 layers of fabric within the 22.5 mm limit. With careful packing round the ropes and cables it should therefore be possible to deploy a 1100 metre boom sweeping a 1000 metre width of sea from a package that can be dropped from a Hercules.

Sweep Rate to Power Ratio Sweep-rate is measured in square metres per second and is calculated from the product of the width of the sweep and the forward tow velocity. While power consumption does not seem to be of great importance in the context of an oil spill, it has major implications once the philosophy of air transport is accepted. The power P consumed by a structure of width W and draught D with a drag coefficient of Cd towed at a speed V in fluid of density p is given by P = 1p WDV3 Cd 2 The sweep rate is WV. The sweep-rate to power ratio is then 2 p D V2 Cd The choice of D will depend on the flow behaviour of the oil/water layer and perhaps the expected wave height. The value of 0.4 metres is a reasonable first estimate (influenced by the available fabric widths) but different draughts and in particular reduced draughts may be used. To maximise the sweep-rate to power ratio low towing speeds are required with very wide collection widths. Furthermore, low forward speeds induce less agitation of the oil-water interface and so should release less oil below the boom. If an initial value for Cd of 0.5 is taken, then 100 kilowatts of towing power at one metre per second could be expected to sweep a frontage of one kilometre.

Tugs Naval Architects have long been conditioned to shapes of vessels designed to have minimum drag. Using booms with frontal widths of 1 kilometre, the tug drag becomes irrelevant. The important design factors for the tugs are as follows: (1) The dimensions of the transportation space available (e.g. a Hercules hold).

(2) Power to produce the required towing force and side trust and to drive air compressors and water pumps.

(3) Sea keeping and manoeuvrability with unpredictable tow loads.

(4) Protection for the crew.

(5) Storage of fuel, food and water.

(6) Communication with support aircraft and other tugs.

(7) Connections to the fabric tubes or to other tugs.

The plan area dictated by the dimensions of the Hercules hold is only 2.9 x 2 metres. A draught of 1 metre would give a tug displacement of 5.8 tonnes, much more than is likely to be required. The plan area of the tug could, desirably, by increased by the inflation of external buoyancy bags to give it an effective length of perhaps 5 metres and a beam of 4 metres. This will provide much more resistance to pitch and roll.

At speeds of 1 metre/second conventional rudders are ineffective. However excellent handling properties can be provided by the use of two propellers one placed as far forward and the other as far aft as possible. If the propeller mountings are provided to rotate about a vertical axis then each tug can be driven forwards, backwards or sideways and can yaw about any centre.

It is desirable that the line of action of the tow ropes should pass close to or perhaps a little below the centres of roll and pitch so that tension variations do not induce excessive tug rotation. Unfortunately it will be necessary to use quite large propellers, perhaps 0.7 metres in diameter so that their thrust line will have to be applied - perhaps 0.5 metre below the pitch centre.

The--total drag on the boom is likely to be about 105 Newtons. For single boom applications the tugs will be pulling outwards at 450 so that each will be applying about 7 x 104 Newtons. The pitch stiffness of a rectangular plan vessel of length L and width W is p g WL2 4 For an initial choice of dimensions this gives a pitch righting moment of 2.5 x 105 Nm/radian. If the thrust moment is 3.5 x 104 Nm the change in pitch angle will be 8 degrees. It should not be necessary to apply full thrust sideways.

Deployment Packed boom-tug combinations can be pre-positioned at any airfield from which a suitable transport aircraft (e.g.

a Hercules) can operate. At the alarm of an oil spill, each boom-tug combination would be fuelled and provisioned with food and water for 2 people in eachtug.

The underside of each packed combination could have bogie wheels so that it can easily be moved from its store to the aircraft in a shorter time than the preflight checks.

On arrival at the spill, the pilot and surface crew would choose the best dropping point in the light of wind, current and geography. The surface crew could strap themselves into well-sprung astronaut-style seats and be dropped with the equipment. They could, of course, descend by conventional parachute.

Tug/boom combinations can also be deployed directly from bases at oil terminals or from tanker decks. The packed shape is not elegant but with the power of two tugs, the surface speed should be at least 10 knots. Depending on the distribution of suitable aircraft bases, surface deployment may be quicker for spills within say 40 miles of an oil terminal.

It is quite difficult to judge position from sea level so for large spills some form of air support is highly desirable even for surface deployed booms. Once the correct position has been reached, about 200 metres down wind of the slick, the tugs would separate from the Z-fold fabric pack and the crews would inflate the side pontoons so as to increase righting moments. They would then start the air and water pumping so as to inflate the various tubes and at the same time would start to move away from the Z-fold fabric package.

The Z-folds would be held together by fairly weak inter-fold links (eg patches of double-sided adhesive tape).

Inflation will provide all the necessary separation force so that unpeeling will occur only at the outermost fold and the central pack will remain in one lump.

The tugs now move apart from one another, heading slightly towards the slick so as to form the proper arc.

Once the Z-fold was completely unpacked the tugs would get final heading instructions from the air and begin the sweep.

Very wide oil slicks will be cleared by dropping more than one pack and linking pairs of tugs. Bringing them close will reduce the drag so as to allow more rapid movement to a separated patch of oil. The amount of equipment and personnel that can be brought to the scene will rise roughly with the square of time. Additional help can be obtained from surface vessels which can draw oil out of the storage bags. These might be other air-dropped tugs which could temporarily store volumes until conventional tankers can arrive.

If arrival of the equipment at a spill is early enough to catch the oil layer while it is still fairly thick, the booms can be used to 'corral' a pool of oil much larger than their own bag volume by forming a semicircular fence at the leeward perimeter and drifting downwind with the slick. It could reasonably be hoped to contain a depth of 200 mm of oil in this way. The prevention of overtopping would be very important and would be assisted by the tight heave response, by the low relative velocity and by wave suppression of the oil layer. One pack could contain a pool of 38,000 tonnes but the volume rises with the square of the number of booms in use so that 2 units could contain 150,000 tonnes and 4 units could contain 600,000 tonnes.

Conventional suction equipment would find it much easier to recover oil from this sort of layer thickness than from the usual few millimetres thickness of freely spreading, opensea slicks.

With 2 crew per tug the operation could, in desperate emergency, continue for several days. Crews can be changed by helicopter. Tug fuel consumption will be about 20-25 litres/hour.

Sea-Keeping Characteristics Rigid structures like destroyers, oil-platforms and even large container ships can be and sometimes are destroyed by wave action. But waves cannot break anything which can yield through a distance greater than the wave amplitude. The leading tubes have a large plan area and very low inertia. One can expect them to pitch and heave in exact response to waves of considerable steepness. This reduces the chance of overtopping except by extremely steep plunging breakers which are uncommon in water deeper than 1.6 times the wave height. It would be possible to contain oil from an overtopping wave by the addition of a sixth small air-filled tube placed on the after portion of the storage volume. This must not be too large if wave-induced tension is to be avoided.

An extremely steep plunging breaker could conceivably capsize a part of the structure bringing the water-filled tube 5 above the air-filled ones. There would, however, be a large righting moment both from the weight of water in tube 5, the buoyancy of the air in tube 4 and the tension of the tow rope in tube 1.

Flotsam Tolerance The surface of the sea can contain an extraordinary variety of objects many of which have sharp edges. Floating crates with protruding splinters and nails are common. The probability of meeting such an object will be several hundred times higher than in the operation of an inflatable boat because of the extra width. To ensure flotsam does not prevent oil sweeping it may be necessary to use helicopters to remove some flotsam from the sweep path.

The proposed fabric is reasonably tough, however it would not be reasonable to expect any moderately-priced fabric to survive all attacks from sharp members encountered at sea.

The most vulnerable regions are the forward surfaces of tubes 1 and 5, both of which have a double fabric cover and which could if needed have a third layer. Tube 5 will get some protection from the W ties.

The effects of holes which lead to air or water losses greater than the pumps can provide need to be considered.

It is possible that operation could continue with the complete loss of the forward air tube provided that the empty fabric does not flutter in the flow. However damage to the water tube is more serious. The ideal protection would be to sub-divide the tubes into separate compartments after the initial inflation or after damage. This could be done by including a number of what the pipe-line industry calls pigs. These would be inflatable piston-like bags which could move along the tubes when partially inflated and be fixed in place by further inflation when they had reached the site of damage.

The underside of the oil store 7 is also exposed to oblique attack. However the fabric can yield easily and the pressure gradient is low so that the initial effect of small holes or slits will be the rather slow release of the lower water layer and so an inadvertent improvement in the collected water fraction.

A brave, agile and well-paid person wearing special boots, perhaps trained at wind-surfing, could walk along the top of the air tubes in moderate sea-states and crawl along them in quite rough ones. This could allow the clearance of flotsam. Very long slits in the oil storage bag could be repaired by lines passed under the boom used to lash the loose fabric to the tubes either side of the slit.

It would be awkward to go astern to rescue anyone falling overboard because one would not want to reverse the pressure gradient at the entrance to the storage volume. It therefore might be prudent to trail one or more rescue lines from tug to tug behind the storage bags. Buoyant, oilabsorbent cladding on this line could perform a final cleanup of small escaped droplets.

Further aspects Although manned operation of the tugs would normally be preferred it will be appreciated that remote control (eg by radio) is possible.

Figure 3 shows, by the lines "WL", the water level to be expected under typical operating conditions in still water. The tube 40 shown in Figure 3 is a desirable addition which aids in stabilising the boom and in particular provides a dam to hold back any oil which may spill over the leading array of air-filled tubes 1 to 4 onto the upper surface of the bag 7. The tube 40 as illustrated has an upper compartment 41 which is air-filled and a lower compartment 42 which is water-filled and this is a suitable arrangement although other means for providing a "dam" are clearly possible.

Removing collected oil from the bag 7 can be assisted by the provision of a false bottom in the bag 7 (shown with the chain line 6a in Figure 3). By dividing the boom in sub-lengths over each of which the space between the true and false bottoms of the bag can be independently inflated, the sources of pressurised air available on the tugs can be used to propell the oil along the boom (eg towards either or both tugs) where removal from the bag 7 can be more easily achieved. The false bottoms 6a are thus used somewhat as a peristaltic pump to move the oil sideways of the direction of tow.

Conclusions Despite the very slow forward speed needed for high oil-recovery fractions, each module of equipment according to a preferred embodiment of this invention could have very high effective sweep rate - about 1000 square metres per second -- because of the extreme width. The inflated structure and thin fabric give this width from a package which is light enough and small enough to be dropped from the air giving rapid response anywhere in the world. After deployment the flexibility of the structure will give it good survival characteristics. It will follow wave motion very closely so that the oil/water interface is not disturbed and overtopping minimised. The intelligent electronic discrimination exploits the largest difference between the properties of oil and water and should minimise the retained water fraction.

Claims (19)

1. Oil-spill recovery equipment comprising an inflatable boom supporting a storage bag for recovered oil, the bag having a valve-controlled entrance means.

2. Oil-spill recovery equipment according to claim 1, in which the boom includes at least one air-filled tube supporting an upper surface of the bag and at least one water-containing tube supporting a lower surface of the bag.

3. Oil-spill recovery equipment according to claim 1 or 2, in which there are a plurality of air-filled tubes increasing in cross-sectional size in the direction towards the entrance to the bag.

4. Oil-spill recovery equipment according to claim 2, in which the valve is located between an air-filled tube and a water-containing tube.

5. Oil-spill recovery equipment according to claim 4, in which the valve is electronically controlled to open on a sensing of oil at the entrance and to close on a sensing of water at the entrance.

6. Oil-spill recovery equipment according to claim 5, in which the valve control means includes means sensing dielectric constant.

7. Oil-spill recovery equipment according to claim 2, or any claim dependent thereon, in which the valvecontrolled entrance means utilises pressurised fluid power for its control.

8. Oil-spill recovery equipment according to claim 7, in which an air-filled tube is used as the source of such fluid power.

9. Oil-spill recovery equipment substantially as hereinbefore described with reference to, and as illustrated in the accompanying drawings.

10. A method of recovering oil from the surface of a body of water which comprises towing a floating boom on the surface of oil-contaminated water to scoop the oil towards a collection area defined by the boom, characterised in that the collection area is a valve-controlled bag located beyond the boom in the direction opposite to the direction of tow.

11. A method according to claim 10, in which control of the valve to the bag is effected electronically based on a sensing of a property of water that is measurably different from the same property of oil.

12. A method according to claim 11, in which the said property is dielectric constant.

13. A method according to claim 10 or 11, in which the valve is located between air-filled and water-containing tubes.

14. A method according to claim 13, in which there are a plurality of air-filled tubes disposed one after the other in the direction opposite to the direction of tow.

15. A method according to claim 14, in which the airfilled tube closest to the source of the tow is smaller in diameter than the other or others.

16. A method according to any one of the preceding claims 10 to 15, in which streamers are supported in the water just outside the collection area to trail in oil as it approaches the valve of the bag.

17. A method of recovering oil from the surface of a body of water substantially as herein described with reference to the accompanying drawings.

18. A method as claimed in any of claims 10 to 17, in which a precursor of the method involves lifting the boom in a collapsed pack to the location of the oil-contaminated water by aircraft and inflating the boom on the water following dropping of the pack from the aircraft.

19. Any novel oil-spill recovery equipment or method for oil-spill recovery which is disclosed herein.