GB2365384A

GB2365384A - Oil tanker with double hulled cargo tanks

Info

Publication number: GB2365384A
Application number: GB0018374A
Authority: GB
Inventors: Bjarne Thygesen; Timothy John Gunner
Original assignee: Individual
Current assignee: Individual
Priority date: 2000-07-26
Filing date: 2000-07-26
Publication date: 2002-02-20
Anticipated expiration: 2020-07-26
Also published as: GB0018374D0; GB2365384B

Abstract

An oil tanker 1 has cargo tanks 5 within a double hull 2,3 which extend to main deck level 4, wherein each tank has an upward enclosed extension 8 above deck level such that each extension encloses a volume between 10% and 20% of the total volume of each tank and where longitudinal bulkheads 6 extend trough the cargo tanks and into the upward extensions. The extensions may be dome shaped or partly hexagonal in shape and may be double walled and ceilinged. The extensions ensure a lower vapour pressure in the tanks which eliminates the need for venting polluting vapour into the atmosphere.

Description

<Desc/Clms Page number 1> "'Oil Tankers" The present invention relates to oil tankers and particularly tankers designed to carry crude oil. By "'Oil tankers" is meant tankers designed to carry volatile hydrocarbon liquids such as crude oil, naphtha, condensates, motor gasoline, aviation gasoline and spirits such as White Spirit. In crude oil tanker operations, pollution control for tankers has always been given high priority along with safety of the vessel's crew and the tanker itself, by those within the crude oil transportation industry. The effects of crude oil liquid phase pollution to the environment have long been recognised for its damaging effects but it is only in recent times that the adverse effects of hydrocarbon vapour emissions has been recognised. Work has been commenced in many directions to limit the emission and thus the impact of Volatile Organic Compounds (VOC) on the environment primarily by use of technology and process systems. In the same manner, the relevant International Marine Pollution Regulations (MARPOL) contain requirements for technical modifications to be made to the traditional vessel structures to limit or prevent the escape of hydrocarbon liquid in the event of a rupture to a vessel's hull or cargo tank. Such modifications include the requirement for oil tankers to have a double hull so that in the event of an external hull rupture the internal hull will prevent the outflow of crude oil.

In recent years the International Community has become concerned with the extent of atmospheric pollution from industrially generated gases. In this context, terms such as "Greenhouse Gases"' have been widely discussed and relate to gases which generate a "blanket" within the atmosphere thereby creating a general warming of the Earth's surface. One such Greenhouse Gas is created through the photosynthesis of Nitrogen oxides with hydrocarbon vapours and is termed ""low level" ozone.

Hydrocarbon gas evolution and escape to the atmosphere is now being regulated and controlled by many nations and communities for shore based fixed plant and ma.chinery in order to reduce the overall effects that this type of gas has on the environment. The first regulatory steps in the direction of control of atmospheric pollution for maritime transportation are to be found in Annex 6 to the International MARPOL regulations (regulation 15) However, this Annex provides limited provision for the control of escape of these gas types for the whole transportation period and currently only considers the displacement of these gases during the loading period of hydrocarbon liquids.

The present invention, however, relates to problems involving the whole transportation period and the differing handling techniques used during the transportation of crude oil and other hydrocarbon liquids referred to for example above, which could create a hydrocarbon vapour emission to the environment. The extent of current annual emission

worldwide of vapour from crude oil to the environment during transportation has been diversely estimated at between 3.1 and 7.1 million tonnes. The problem of emission reduction will be discussed below in relation to crude oil only for the sake of brevity but is of course present when carrying other volatile hydrocarbon liquids.

Due to the large number of crude oil types and their diverse quality, the ability of a specific crude oil type to evolve hydrocarbon vapours at atmospheric pressure and ambient temperature will be reflected by the quantity or concentration of volatile compounds in the specific crude oil which is, thereby, capable of generating a vapour pressure given the prevailing conditions. Generally crude oils could be divided into three groups having the nomenclature of high volatility, intermediate volatility, and low volatility.

So to illustrate the diversity of crude oil's volatility the following attempts to define the foregoing groups.

Parameter High Intermediate Low Volatility Volatility Volatility Reid Vapour B-12 5-8 0-5 Pressure psia (0.55 - 0.83 (0.34 - 0.55 (0 - 0.34 bar) bar) bar) Hydrocarbon 3.0%+ 1.5 - 2.5% 0 - 1.25% Gases C, to C5% weight

Following this generalised sub division of crude oils' volatility, a more specific illustration of the diversity of various crud & oil types volatility may be seen as Appendix 1 herewith. Appendix 1 identifies 68 differing crude oils with their associated Reid Vapour Pressures (RVP) (reported in Pound per square inch (psi) where the diversity of volatility can readily be identified. Conversion of psi to bar is accomplished by multiplying by a factor of 0.06895. Notwithstanding the foregoing, the volatility of crude oil is traditionally reflected by the analytical measurement of the crude oil's vapour pressure. This measurement/test traditionally relies upon a method or procedure and its required/defined equipment entitled "Vapour Pressure Reid Method," (I.P. 69) . The Reid Vapour Pressure test method determines the vapour pressure of a hydrocarbon liquid at a temperature of 100'F (37.8'C) with a liquid to vapour ratio of 1:4 (i.e. one part of liquid volume to 4 parts of vapour volume). Although this determination may allow comparisons to be made between the volatilities of various hydrocarbon liquids, the actual test result contains a meaningless and unusable pressure determination for the purposes of transportation of hydrocarbon liquid. Although the vapour pressure determined can be said to be Saturated Vapour Pressure, both the fixed temperature requirement and the Liquid to Vapour ratio supplies a pressure that has no relevance to the circumstances onboard an oil tanker during the transportation of a crude oil cargo onboard a tanker which are deemed full when the cargo tanks are filled to 98% of the total volume of the cargo tanks; 2% being allowed for expansion and safety.

Before considering the implications of the foregoing further, it is necessary to briefly discuss the physics of Saturated and Unsaturated Vapour Pressures. It is well known within the various Gas Laws, culminating with the Ideal Gas Law (PV = nRT) that, at a constant temperature, if the volume of a gas phase is increased the pressure in the defined phase will decrease. This law applies to circumstances where the vapour phase is Unsaturated, that is, the gas creating the pressure is not in contact with its associated liquid phase.

Had the gases in the vapour phase been in contact with its associated liquid phase (i.e. Saturated) under the foregoing circumstances then more vapour would potentially evolve from the liquid phase The result of this behaviour would be a constant pressure being maintained by the vapour phase and equilibrium being restored between the liquid and vapour phases.

It is also necessary to discuss the concept of Phase Equilibrium (Pressure) for Saturated Vapour. The state of Equilibrium (pressure) will occur when the pressure generated by the vapour phase is equal to the pressure that can be "supported" by the adjacent liquid phase. If the vapour pressure is too large for the liquid phase then vapour will condense into the liquid phase. If the vapour pressure in the vapour phase is too low then further vapour will evolve from the liquid phase to attain equilibrium. The equilibrium pressure achievable by a Saturated vapour system is depended upon two variables; namely, the

Temperature of the system and the Liquid to Vapour ratio. Clearly a Liquid/Vapour system could have two definable temperatures; namely, a temperature for the Liquid phase and a separate temperature for the Vapour phase. The temperature of the Liquid phase supplies an energy requirement for the evolution of vapour from the liquid phase whereas the differing temperature of the Vapour phase will cause the vapour to expand and contract thereby creating dis-equilibrium. between the phases. Generally the more important of these two temperatures is the liquid phase temperature supplying the "supporting" energy for vapour evolution. The practical impact of the vapour phase temperature upon transportation conditions will be discussed further below. Thus, in the event that a saturated vapour pressure exists the Ideal Gas Laws do not apply to its behaviour. Given the foregoing ten, the Ideal Gas Laws can not be used as a predictive tool/method for the calculation of pressure for alternative Liquid/vapour volume ratios for the conditions as existent onboard a tanker carrying volatile hydrocarbon liquids.

Given the foregoing discussion and its conclusions it is therefore necessary to consider an alternative vapour pressure parameter other than that of the Reid Vapour Pressure (RVP) for predictive use onboard tanker vessels. This alternative parameter is termed the Total or True Vapour Pressure (TVP).

As one aspect of the recently completed CRUCOGSA research project, extensive work was undertaken to develop a "tool" which could be readily used by a tanker's command to advise

upon the expected vapour pressure that could be attained in a vessel's tank vapour phase. This will allow a vessel to calculate the Total Vapour Pressure in a cargo tank system loaded to 98% full given the cargo's Reid Vapour Pressure and the cargo's liquid phase temperature thereby predicting the necessity for vapour release when such pressures exceed the pressure capable of being contained by the hull structure and associated safety valve systems.

The Total Vapour Pressure of a crude oil can now be calculated with the "inputs" of the cargo temperature and the Crude cargo's Reid Vapour Pressure. The assessed or required accuracy of this model is approximately +/- 1.5 p. s. i (0. 1 bar) . This scope of accuracy is developed from, amongst other criterion, the accuracy requirements of Reproducibility function associated with the Institute of Petroleum Test Method I.P. 69 --Vapour Pressure Reid Method. The model, with all pressure variables recorded in pounds per square inch (psia) , and thereby, the formula for this calculation was developed as follows: In the first instance it was necessary to develop a model that would supply an equivalent Total Vapour Pressure with a liquid to vapour ratio of 1:0.02 (98% loaded tank) at 37-80C for any recorded or measured Reid Vapour Pressure of a crude oil. This was achieved by plotting the respective vapour pressures for approximately 2020 samples to achieve an equation with a known gradient and intercept.

The logarithmic equation as developed for the samples analysed is: PT = 5.7925 *LN(PR)+5.5959 Where- PT is the Total Vapour Pressure at 37.80C for the L/V Ratio of 1:0.02 PR is the Observed Reid Vapour Pressure of the crude oil at 37.80C Having established an equation to calculate the Total Vapour Pressure at the single temperature of 37.80C (1000F) from the Reid Vapour Pressure of a crude oil, it was recognised that such an equation would be of very limited operational value to a tank vessel. Clearly the cargo liquid phase temperature varies widely and therefore a model had to be constructed to simulate the variation of Total Vapour Pressure of a crude oil in a fully loaded tank with variations in the temperature of the cargo.

This model is developed from a plot of approximately 1250 data points for the variation of the Total Vapour Pressure for 68 differing crude oils with temperature.

The discussion with regard to this model must commence with the "'RVP" plot. By reference to the mean trend of data points, it will be recognised that with the differing crude oil volatilities the trend line will retain its gradient but

will move vertically up and down but parallel to the mean trend'line. The transfer of the infinite number of pressure values at 37.8'C on the RVP graph to an equivalent position on the TVP graph is undertaken by the first equation.

Having achieved a pressure on the TVP graph, the same conditions apply as for the "RVP" graph; namely that, the mean trend line as shown on this plot will retain its gradient but move up and down parallel to the mean trend line. However in this regard a complication arises immediately this thesis is applied to the calculation of the Total Vapour Pressure at any temperature other than at either 37.8'C or remote from the mean trend line. This is due to the fact that as the trend line is moved up or down but with a parallel gradient to the mean trend line, the "y" axis intercept applicable to the mean trend line equation will alter with the movement. Thus an equation had to be generated to compensate for this function.

This Equation is: TVP Intercept = PT - 13. 41 Where: PTiS the Total Vapour Pressure at 37.80C

Applying this intercept equation to the RVP/TVP transfer equation the equation becomes: Pt = (0. 354 7 * T) + (PT- 13. 4 1) Where: Pt is the Total Vapour Pressure at a defined temperature t T is the temperature of the liquid phase in OC PT is the Total Vapour Pressure at 37.8'C Thus with the combination of the equations the final Total Vapour Pressure equation becomes: Pt = (0.3107*T) + (5.4602*LN(PR)) -5.56 The foregoing model/equat ion can now calculate the total vapour pressure of a crude oil's vapour phase (Pt) given the liquid phase temperature (T) and the Reid Vapour Pressure of the Crude Oil (PR).

Hydrocarbon vapour emissions occur during two distinct phases of transportation of a crude oil. These phases are, during the loading operations by displacement of hydrocarbon gases from the respective cargo tank by the incoming crude oil, and, during the period of transportation from the load port to the discharge port. Hydrocarbon vapour emission can also occur, but should not, given good operational practice, during a third distinct phase of operations; namely, during the discharge operations especially when undertaking crude oil washing of the cargo tanks. Generally during loading, hydrocarbon emissions of gases are from the previous cargo and enriched hydrocarbon gas mixture evolving from the cargo being loaded.

In order to consider methods to prevent the emission of hydrocarbon vapour to the atmosphere, thereby creating pollution, it is important to consider how these emissions are caused and the potential extent of the emissions.

Before a vessel tenders to load a cargo of crude oil the vessel's tanks will be filled with a mixture of hydrocarbon vapour and Inert Gas such that the cargo tank's pressure is a suitable overpressure preventing the entrance of air/oxygen to the tanks. This circumstance is required under the International SOLAS regulations for, amongst other types of specific tankers, a crude oil tanker.

The extent of hydrocarbon vapour to inert gas in a cargo tank prior to commencement of loading depend upon the volatility of the previous crude oil cargo and the extent of crude oil washing (tank cleaning) undertaken during the previous discharge programme. These variables will be considered further below when discussing the discharge programme but it can be stated that, as a result of a research project undertaken by Messrs B.P., it is understood the hydrocarbon content of the pre-loading vapour phase would be in the range of 10% to 25% with the balance of gases being from the Inert Gas injection.

Upon loading hydrocarbon vapour will evolve from the incoming cargo. Many factors will have an influence upon the evolution af these vapours such as:- (1) The crude oil type and its volatility (2) The design of the vessel particularly with respect to its pipeline and internal tank design (3) The rate of loading thereby inducing turbulence into the crude oil liquid phase (4) The temperature of the crude oil being loaded (determining the equilibrium pressure for the total vapour pressure for the cargo under the varying liquid to vapour ratios)

(5) The temperature of the air controlling the temperature of the vapour phase The modelling undertaken for the CRUCOGSA research project, however, found that during the whole loading period hydrocarbon vapours were displaced to atmosphere. At the commencement and for the majority of the loading programme the concentration of vapours being displaced represented the residue vapours from the previous cargo. As previously stated the vapour concentration in the Inert Gas volume being displaced is low. However the observations showed that when cargo tanks were about 80% full the displaced hydrocarbon vapour started to become more enriched. The enriched hydrocarbon vapour being displaced during this stage of loading was due to the increased vapour concentration as a result of release from the incoming cargo.

Such observations as those above reconcile closely with the contents of Chapter 17 of the International Safety Guide for Oil Tankers and Terminals (ISGOTT) (reference paragraph 17.6). The graph, as shown in this Chapter, records the depth of a layer above the liquid surface for a 50% concentration of hydrocarbon vapour in tank gases. It can readily be seen from this graph that released gases enriched with hydrocarbon vapour will commence at about the 80% loaded level dependent upon, amongst other criteria, the volatility of the crude oil being loaded.

The foregoing observations and conclusion were also confirmed by the CRUCOGSA research project. Enclosed herewith as Figure 2 is a set of graphs showing the increase in Total Vapour Pressure of various crude oils as the Liquid to Vapour ratio is decreased. The graph shows the status of the pressures obtained from a Liquid to Vapour ratio of 1:0.5; namely 50% full tanks, up to a Liquid to Vapour ratio of 1:0.02; namely 98% full tanks. It will be seen from the respective graphs that the pressure in the larger vapour ratios is nearly constant with change of the vapour ratio but at a Liquid to Vapour ratio of 1:0.2 (83% full tank) the vapour pressure increases in a non-linear manner. This behaviour is also consistent with temperature variation as is shown by a plot of 1sotherms for a crude oil (see Figure 3).

These findings provide evidence for an interim solution to avoid the release of enriched hydrocarbon vapours to the environment. Such a solution would be that cargo tanks should not be filled beyond a level where the total vapour pressure of the crude oil being loaded at its observed temperature for that Liquid to Vapour ratio exceeds 1.0 bar (approx. 14.5 psia) . However, as will be recognised, the consequences of filling of cargo tanks of tankers so that tanks would be well below the present 98% full capacity would increase transportation costs and create a potentially uneconomic proposition for the carriers.

With a solution in mind and from a position of maximising the volume of cargo loaded, the two main operational variables that assist in the generation of vapour evolution should be controlled; namely:- (1) Slow the rate of loading to limit turbulence in the cargo tanks. Turbulence is the manifestation of the input of kinetic energy into a liquid which will assist in the evolution of hydrocarbon vapour (consider the shaking of a container containing volatile gases under near atmospheric pressure).

(2) Reduce the temperature of the crude oil cargo being loaded. The temperature of a liquid is a result of the extent of heat energy in the liquid phase supporting gas evolution. The cooler the temperature the lower the vapour pressure.

It is to be recognised that both of the foregoing criteria will not create a total solution to the problem, namely that prevent the unnecessary escape of hydrocarbon gases to the envi ronment /atmosphere thereby creating pollution, and are outside the direct control of the vessel receiving the cargo.

The foregoing methods for hydrocarbon vapour pollution minimisation presupposes that the cargo of crude oil is stabilised with the normal hydrocarbon volatile content that

is to be associated with the relevant crude oil as prepared for s hipment. Clearly a reduction in the volatile gas content of the crude oil would, in itself, have a major impact upon the extent to which the crude oil could be loaded without displacing or enforcing release of enriched hydrocarbon gases during such an operation.

The CRUCOGSA research programme has been documenting and studying this vapour release activity and can advise that the extent of hydrocarbon vapour emissions during transportation are higher than previously documented. This is particularly the case for the large VLCC (Very Large Crude Oil Carrier) type vessels whose voyage length is usually 30 to 45 days. However, before considering the physical behaviour of crude oil and its ability to generate vapour pressures/emissions during transportation it is necessary to briefly discuss the mechanisms available onboard a tanker to control the overall pressure in its cargo tanks.

A crude oil tanker is equipped with three differing mechanisms for the control of vapour pressure within the cargo tank system; namely, a Pressure/Vacuum release valve (P/V valve), a mast riser and a Pressure/Vacuum (P/V) breaker.

The primary mechanism to protect the tank structure from an unacceptable over pressure is that of the Pressure/Vacuum release valve (P/V valve). The P/V valve is a safety valve and is not normally, onboard tankers, allowed to control the vapour pressures automatically in the cargo system.

Typically, the current designed opening pressure of a P/V valve would be set at approximately 2.2 p.s.i.g. (1.015 bar over atmospheric pressure or 1500 mmWG) but the vessel's command would control the maximum pressure in the vapour phase for the cargo tank system to 1.45 p.s.i.g. (0.1 bar above atmospheric pressure or 1000 mmWG).

These valves are designed such that they rely upon weight controlled mechanical system. As an example, the valve is equipped with a predetermined weight that is seated on the pressure orifice of the valve. The weight closes the pressure orifice until the vapour pressure is achieved to lift the weight and open the orifice to atmosphere. When the pressure reduces to the designed acceptable level the weight then closes the pressure side orifice. In principle, the valve is simple in construction and consists of few moving parts. The operation of the valve is "fail safe" for it relies upon gravity and the "mass" of the valve for the control of release pressure.

However, in practice the weight controlled mechanism in a P/V valve can be found jammed open or not properly seated on the pressure side orifice due to the fouling of the mechanism or valve seat by deposits. With the passing of time and with corrosion activity to the valve seating a gas tight fitting of the closing weight to its seating is practically impossible to achieve. Likewise regular maintenance to the numerous valves (usually one per cargo tank or space) is time consuming and can only be undertaken

during a ballast voyage when many other "more important" maintenance projects are planned for the general cargo system.

Thus, in theory P/V valves will retain the evolved hydrocarbon vapour up to the pre-set release pressure for a certain P/V valve but in practice it is more than likely that continuous leakage of hydrocarbon gas is occurring throughout the voyage period. The foregoing addresses the theory and practice of P/V valves and hydrocarbon emissions under static vessel conditions. There are circumstances when, in addition to the foregoing, "abnormal" leakages will occur through this mechanism. This circumstance is termed "breathing" and is induced by the rolling or pitching of the tanker. The rolling motion creates a wave system on the liquid surface that in turn create localised pressures and vacuums to occur with the transition of the wave past a particular point in the tank. When this point coincides with the point at which the P/V valve pipeline is connected to the specific tank vapour phase then the valve will open and close with the increased localised pressures induced by the liquid surface wave motion. Standing by in way of a P/V valve when such releases occur closely resembles the breathing action.

Unlike the P/V valve whose rate of volume release is low due to the diameter of the valve and pipeline system, the mast riser is designed to release large volumes of gas at reasonably high velocity. The main use of the mast riser is

to facilitate the release of displaced gas (inert gas and hydrocarbon mixture during the loading process) Thi s method, use and system is in the process of review and being replaced by a "Vapour Return System" to shore during loading.

The mast riser is basically a vertical pipeline with a liquid trap and dispersion nozzle design at its opening to the atmosphere. The height of the opening above the tanker's main deck should ensure that the deck remains clear of toxic gases being released from the cargo tank system. The mast riser, for normally there is only onboard a tanker, is connected to the fore and aft common Inert Gas supply main to the cargo tanks. Alternative arrangements allow the operation of the riser during a loaded voyage such that it opening can be controlled by a P/V valve type system and/or a manual valve isolating it from the common Inert Gas main pipeline.

Although the mast riser could be used for manual release of excess over pressure in the cargo tank vapour system, this is seldom done given the size and speed that the over pressure will reduce when this system is operated. Instead releases during the voyage are often undertaken by manually opening a P/V valve that supplies a more sensitive control on the gas pressure reduction in the cargo tank's vapour phase. A single P/V valve can be manually opened on the pressure side that will release pressure from the whole

vapour system given that the vapour system on a crude oil tanker is usually common to all tanks.

The P/V breaker is the main or final safety device for the vapour system to protect against over or under pressure within the vapour system. There is normally only one P/V breaker on a crude oil tanker and this is connected to the common Inert Gas main pipeline. Unlike a P/V valve which is capable of opening and closing, thereby controlling a maximum pressure regime within a tank system, when the P/V breaker operates it purely opens the vapour system to atmospheric pressure. There is no closing device or mechanism.

Thus, in the event that an over pressure occurs that the P/V valves cannot control, and the P/V breaker displaces its pre-set internal hydrostatic pressure (water/liquid column), the total gas volume creating the over pressure will be released to atmosphere until atmospheric pressure is reached within the cargo tank vapour system.

It is without doubt that hydrocarbon gases evolve from the liquid phase into the ullage or vapour phase within vessels' tanks during the transportation of crude oil. The questions remain as to how can the releases be minimised to prevent excessive pollution and, if so, how?

In order to discuss these issues to develop a solution to this operational problem, the data within the CRUCOGSA database from one voyage relating to the gas phase behaviour of a volatile crude oil has been selected as an example of the behaviour of a volatile crude oil. Figure 4, contained herewith, records the variations in the cargo system vapour phase pressure together with the cargo temperature. The pressure data is recorded every four hours throughout the 46-day voyage whereas the cargo liquid phase temperature is recorded daily (at roughly the same time every day) . The graph also records the P/V valve opening pressure together with the normal operating pressure for the vapour phase pressure of 1000 mmWG. It should be stated that the two "'y" axis of the graph are not synonymous with one another; that is to say that the recorded pressure on the one "y" axis is not that achieved by the temperature record on the other '"y" axis.

Examining the vapour pressure record it will be seen that the vapour pressure record can be broadly subdivided into three sections; namely, from day 1 to approximately day 17, from day 17 to approximately day 33 and from day 33 to day 46. Likewise the change in cargo temperature throughout the voyage also falls broadly into the three previously outlined stages.

Before examining these stages more closely a general examination of the vapour pressure record reveals that for previously defined stages one and two of the voyage the

vessel's command attempt to limit the vapour pressure within the cargo tank system to below the acceptable/usual maximum vapour pressure of 1000 mmWG. Vapour was manually released regularly throughout the voyage and in the first two stages the closing pressure selected by the vessel's command was between 400 and 800 mmWG. This observation reflects both typical tanker practice and the documented/observed practice for the general data record within the CRUCOGSA research project.

Further it will be noted that after release of the vapour pressure to the defined closing pressure, the vapour pressure increased rapidly to its original level. This was often achieved within 12 hours after the completion of release. It is also to be noted that the cargo temperature record is incomplete for approximately one week in the second defined stage referred to above. The reason for this lack of data was due to the "heavy" weather conditions experienced by the vessel during this period thereby preventing access to the vessel's main weather deck to obtain such measurements. It is to be noted that at the inception of the heavy weather period the vapour pressure also increased and this will be referred to again below when Stages 1 and 3 of the voyage are more closely examined.

Finally, it should be stated that as the vapour pressures being maintained within the tank were all gauge pressures, i.e. above atmospheric pressures, there was no need to utilise the Inert Gas system throughout the voyage in order

to maintain the required over pressure to ensure air/oxygen did not enter the vapour phase (Solas requirement) This being the case (from the records received from the vessel) the regular releases of gas from the vessel as undertaken by the vessel's command would have been solely enriched hydrocarbon vapour with limited or no Inert Gas content.

The first and second defined stages of the voyage There are four significant observations that can be made regarding the physical behaviour of the vapour phase during this defined period. These are:- (1) During the first eight days of the voyage the fluctuations of vapour pressure are only due to the vessel's command attempting to control the maximum vapour pressure within the cargo tank system. As will be discussed below the impact of the diurnal variations of air temperature had no impact on the vapour phase temperature, and therefore, pressure given that the air temperatures were so low and varied little over any 24- hour period.

(2) Over the period of the first 15/17 days the vapour pressure trend began to decline notwithstanding the increasing air, sea and cargo temperatures (see Figure

4 for data regarding the increasing cargo temperatures). This decline in the "mean" vapour pressures would be due to the extensive releases of vapour to atmosphere where the vessel's command were attempting to control the maximum vapour pressure within the tank system. It will be recalled from above that the primary vapour evolution will be from the top one third of the cargo volume due to the static head pressure of the liquid phase preventing evolution from the lower volumetric regions of the cargo (equilibrium pressure being achieved due to external static pressure).

(3) At the inception of Stage 2 of the voyage the vapour pressure started to increase although it can be assumed that the cargo temperature remained constant. As would be typical of a heavy weather period the air temperature dropped throughout the period. Thus the cause for the increase in the vapour pressure during the 2'd stage, given the general decline in the 1st stage, would be due to a combination of the increased movement of the vessel during this period (increased agitation of the cargo imparting kinetic energy into the liquid volume) and the replacement in the upper third volume of "fresh" crude oil from the lower strata in the cargo tanks.

(4) Once noticeable diurnal variations occur to the air temperature (commencing on approximately day 9/10) then

the vapour pressure also begins to fluctuate accordingly. Given that an equilibrium pressure is achieved in the vapour phase with the liquid phase and its associated temperature, the influence of the air temperature upon the vapour phase destroys the equilibrium for a short period of time until the vapour phase temperature is restored to the liquid phase temperature and equilibrium is regained. Evidence of this phenomenon will be shown and discussed below.

The final (3 d) stage of the voyage Again, the data recorded is that of the vapour pressure and air temperature every four hours throughout the period. An examination of the graph of pressures supplies two interesting observations. These are:- (1) With the higher liquid phase temperature during this period it is to be noted that the vapour pressure has generally increased accordingly. However, over the defined period it is to be noted that, as with the Stage 1 period, the general trend of the Gas pressure shows a very slight decline. The reasons for the slight decline are believed to be due to the same reasons as with Stage 1 notwithstanding the fact that at this time in the voyage the vessel's command had ceased trying to reduce the vapour pressure in the

vessel's tanks. This observation can be made by reviewing the number of manual releases (2) and the fact that the vapour pressure over an extensive period was allowed to rise above the P/V valve opening pressure. With such procedures in place the extent of hydrocarbon volume release must have been significant (the extent cannot be quantified by calculation as no time data for the various openings of the P/V valves in their "automatic" mode is available) but, due to the higher cargo temperature, the gas evolution would have increased creating a higher equilibrium gas pressure and preventing a more marked decline in vapour pressure.

(2) The effects of diurnal variation in air temperature upon the vapour pressure can be seen more clearly in this stage of the voyage.

Reference has been made here above as to the impact of sea temperature upon the liquid phase cargo temperature (distinguishing the impact of air temperature upon the liquid phase cargo temperature). Given that the vessel in question is a VLCC without any possibility to heat her cargo (this is typical for VLCCs due to the cargo/tank size), it can be seen that the cargo temperature responds closely to changes in seawater temperature. Given the standard rules of heat transfer, the larger the difference between the cargo temperature and the sea water temperature, the steeper the temperature gradient between the two volumes, the

greater the heat transfer, the more rapid the temperature rise of the cargo. The obverse also applies when the temperature difference between the respective volumes is small. Thus, it is considered fair to state that for cargoes carried on a single hulled VLCC the influence of the sea water temperature, through which the vessel is travelling, will have a significant effect upon the cargo temperature and therefore the vapour pressure capable of being generated by the cargo.

Having seen that when transporting volatile crude oils excess pressures are manually released to attain a reasonable operating pressure within the cargo tank vapour phase system, it is now thought necessary to consider how the vapour pressure in the cargo system reduces when vapour releases are instigated. As an example of a release, Figure 5, enclosed herewith, records data reported for the CRUCOGSA project and shows a graph of time versus vapour pressure during a vapour release. The planned and controlled vapour release commenced at a vapour pressure of 900 mmWG and was planned to cease at 600 mmWG. The period of time taken to achieve this reduction in vapour pressure was 40 minutes. Examining the data and diagram it can be seen that it may be divided into two clear sections; namely the primary release and the secondary release.

The primary release of vapour lasted approximately 5 minutes whereas the secondary release continued for some 35 minutes. It will be noted that the rates of pressure reduction for

the two releases are significantly different. The secondary release reduction in pressure is nearly linear with time. By adding a trend line to the secondary release data and regressing this trend line back to the ""y" pressure axis of the graph, the intercept found for the trend line is that of the Total Vapour Pressure of the crude oil at the time with its associated cargo temperature.

Given the foregoing finding it is now reasonable to state that the primary release represents the excess pressure in the cargo vapour phase system as induced by the air temperature. This excess pressure would be released rapidly until the equilibrium pressure between the vapour and liquid phases is attained. Once this pressure is attained and there remains an open pressure gradient between the internal hydrocarbon vapour pressure and the external atmospheric pressure the pressure drop will continue. However, as the equilibrium pressure is disturbed more hydrocarbon vapour will evolve from the liquid phase attempting to restore the equilibrium pressure consistent with the guiding temperature and tank V/L ratio. Due to this circumstance the rate of decline in the observed pressure is lower and, unbeknown to the vessel's command, by attempting to reduce the pressure to a level below the Total Vapour Pressure of the liquid phase of the cargo excess gases are produced and vented to atmosphere.

With the equation developed earlier in this paper this degree of excess release can now be avoided as the vessel's

command can now programme a release to a pressure equivalent to the Total Vapour Pressure of the cargo for its temperature at the time.

In the foregoing specific event, by integrating the data developing the curves together with other vapour functions for the primary and secondary phases of the release the extent of "'saved" vapour can be readily calculated. By examination of the graph alone it is thought reasonable to suggest that the ratio of the primary to secondary phase released volumes is roughly 1:6 (roughly an 85% saving of total gas/vapour emission).

It is thus concluded from the foregoing that a vessel should not load a cargo whose Total Vapour Pressure exceeded 1.0 bar (14.5 p.s.i.a.) for the observed temperature of the cargo on loading. Although the pressure of 14.5 p.s.i.a. is well below the opening pressure for the vessel's P/V valves and without the top up pressure from Inert Gas supply this would cause oxygen/air entry into a cargo tank, it can now be seen that the influence of air temperature upon the vapour phase causes significant diurnal increases to the overall vapour pressure within the cargo tank system so as to avoid air ingress. Therefore, the proposed pressure criterion should avoid the necessity to manually release vapour from the cargo tank system and takes into consideration the influences of air temperature and the impact of vessel movement in adverse weather conditions upon

the evolution of hydrocarbon gases and the generation of vapour pressure.

Although the foregoing provides some partial solutions to the problems relating to environmental pollution from tankers, all the solutions relate to changes in operational practices onboard tankers. The alternative to these solutions is the design and construction of a vessel that is capable of loading and transporting crude oil, with their ever increasing volatility characteristics, without impacting the environment.

The prevention of hydrocarbon vapour pollution to atmosphere clearly requires an alternative and novel design of the cargo tanks of oil tankers. The design must allow an adequate vapour phase volume so as to reduce the exposure of the vessel's construction to excess pressures (see figure 2) that the structure can not safely withstand, and therefore release, and simultaneously design a construction that can withstand higher vapour pressures than that of the current designs of tanker vessels. Simultaneously with the foregoing considerations the tanker must be capable of carrying a full liquid cargo thereby maintaining the economics for the transportation of volatile hydrocarbon by sea. Thus the present invention provides an oil tanker having cargo tanks within a double hull (see current MARPOL Regulations for tanker construction), the double hull extending to main deck level, wherein each tank has an upward enclosed extension

above deck level whose internal volume is between 10% and 20% of its volume within the double hull.

The upward enclosed extension of the tanks has two advantages. Firstly the cargo tanks within the requisite double hull can be filled to 100% as distinct to current levels of 98%(subject to loadline considerations). Secondly the 10 to 20% increase in total volume results in a liquid volume of 90 to 80% thus ensuring a lower resulting vapour pressure; that is, removing the necessity for venting of vapour due to excess pressure that the current hull structures can not withstand.

Preferably the form of the extension is outwardly convex by forming the tank top into a domed shape or with flat inboardly sloping sides so as to have a semi-hexagonal shape. This form results in a stronger shape than the conventional substantially flat main deck shape and can therefore withstand greater internal tank pressures of say 20-30 p.s.i.a. (currently 16.5 p.s.i.a. - 1.14 bar absolute) equivalent to 1.38-2.07 bar absolute. P/V valves could then be set to lift off at say 28 p.s.i.a. or 1.93 bar (0.93 bar gauge).

Further improvements can be made by forming the extension with a double wall or rather double ceiling to provide greater insulation to the vapour phase thereby reducing

diurnal pressure variance caused by the heating of unsaturated vapours in the vapour phase.

The tank extension can be used to enclose discharge pipelines so that deck leakages can be substantially avoided. Furthermore, fore and aft gangways can be provided under cover with in the double wall or double ceiling of the tank extension.

An example of the invention will now be described with reference to the accompanying drawings in which:- Figure 1 is a transverse cross section of an oil tanker according to the invention; Figure 2 is a graph of pressure against vapour to liquid ratio showing total vapour isotherms at different vapour to liquid ratios; Figure 3 is a graph of isotherms for Iranian Crude Oil at different temperatures; Figure 4 is crude oil voyage data showing cargo temperature against tank pressure; and

Figure 5 is a graph of time against vapour pressure during a vapour release.

Figure 1 shows an oil tanker 1 having an outer hull 2 and an inner hull 3 both extending to a main deck 4, the inner hull 3 enclosing a cargo tank 5 within which are the usual longitudinal bulkheads 6 with upper limber holes 7.

Above and extending from the main deck 4 is an upward enclosed inner extension 8 formed from sloping sides 9 and a horizontal upper deck 10. Above the enclosed inner extension is an outer extension 11 so that a double wall and ceiling is formed which encloses the top of the cargo tank 5 providing a space 13 between inner and outer extensions 8 and 11. Within space 13 are fore and aft enclosed gangways 14, pipes and valves 15 for servicing the cargo tank, the pipes extending to main deck 4 for loading and discharging. The cargo tank 5 is arranged to contain crude oil up to a liquid level so that the extension 8 encloses a volume of vapour whose volume is about 15% of the volume of the liquid. P/V valves 18 at the top of extension 8 are set to open at 26 to 28 p.s.i.a. (1.79-1.93 bar absolute).

Appendix 1 List of Crude Oils and the measured Reid Vapour Pressures (p.s.i.a.)

Crude Oil Type RVP Minimum RVP Maximum psia psia Anoa 2.2 Arab Extra Light 4.1 6.7 Arab Heavy 4.9 7.6 Arab Light 3.1 7.5 Arab Medium 4.3 7.7 Arab Super Light 5.4 8.7 Bach Ho 5.3 6.8 Basrah Light 5.8 7.9 Belida 4.9 Bintulu Neat 6.4 Bonny Light 5.8 7.5 Bonny Medium 4.7 5.3 Brent Blend 9.8 Cabinda 8 9.1 Champion Export 3.3 Dai Hung 4.7 Daquing 3.1 Djeno 6.0 6.3 Draugen 8.9 Dubai 6.7 9.2

Duri 1.4 2.4 Es Sider 8.4 Escravos 5.2 6.2 Forcados 4.6 5.6 Foroozan 4.4 5.1 Gullfaks 5.2 8.6 Hout 6.9 8.0 Iranian Heavy 6.2 10.2 Iranian Light 7.9 10.1 Isthmus 8.0 Kerapu 5.6 Khafji 7.7 10.5 Kuwait Export 7.7 11.3 Lavan 9.0 Leona 5.8 Loreto 2.8 Lower Zakum. 5.6 Marib Light 10.1 Masila Blend 3.1 3.6 Maya 5.6 6.2 Merey 2.1 Minas 2.4 3.8 Miri 4.4 Miri Light 4.8 Murban 2.6 7.0

Nanhai 5.2 5.3 Nemba 7.9 10.0 NKossa 10.9 Odudu 5.8 Olmeca 7.3 Oman 4.6 7.0 Oseberg 9.6 11.0 Palanca 8.3 8.5 Qatar Condensate 7.6 9.6 Qatar Land 6.7 15.0 Qatar Marine 7.5 Qua Ilboe 6.2 7.5 Rabi 5.3 Sahara Blend 10.4 11.7 Siberian Light 7.7 8.3 Soyo 7 Statfjord 5.8 10.6 Syrian Blend 4.9 Troll 4.5 5.6 Umm Shaif 5.5 6.9 Upper Zakum 7.7 9.4

Claims

CLAIMS 1 An oil tanker having cargo tanks within a double hull, the double hull extending to main deck level, wherein each tank has an upward enclosed extension above deck level whose internal volume is between 10% and 20% of its volume within the double hull.
2. A tanker as claimed in claim 1 wherein the extension has an outwardly convex form.
3. A tanker as claimed in claim 2 wherein the extension is dome shaped.
4. A tanker as claimed in claim 2 wherein the extension has a semi-hexagonal shape.
5. A tanker as claimed in any one of claims 1 to 4 wherein the extension is double walled and ceilinged. 6. A tanker as claimed in claim 5 wherein the extension encloses at least one gangway and pipework.

<Desc/Clms Page number 38>

Amendments to the claims have been filed as follows 1 An oil tanker having cargo tanks within a double hull, the double hull extending to the main deck level and fore and aft longitudinal bulkheads extending through the cargo tanks, characterised in that each tank has an upward enclosed extension above deck level whose internal volume is between 10% and 20% of its volume within the double hull and wherein the bulkheads extend into the upward extension. 2. A tanker as claimed in claim 1 where--#n the extension has an outwardly convex form and the bulkheads have upper limber holes. 3. A tanker as claimed in claim 2 wherein the extension is dome shaped. 4. A tanker as claimed in claim 2 wherein the extension is partly hexagonal in shape. 5. A tanker as claimed in any one of claims 1 to 4 wherein the extension is double walled and ceilinged.

<Desc/Clms Page number 39>
6. A tanker as claimed in claim 5 wherein the extension encloses at least one gangway and pipework.
7. A tanker as claimed in claim 5 wherein the walls of the extension enclose at least one gangway and pipework.
8. An oil tanker substantially as described with reference to the accompanying drawings.