242
A. NELYON-SMITIL
high sensitivity, but the absorption is due only to aromatic rings and
similar structures. Thus the precise nature of the polluting oil must
be known and the error can be quite large (Harva and Somersalo,
1958). Ultra-violet fluorescence is an extremely sensitive measure for
oils rich in aromatics, although interference may be experienced from
naturally-occurring polynuclear aromatic hydrocarbons. A detection
limit of 0.001 p.p.ni. is claimed and reference has been made above to
determinations of 0.003 p.p.m. in coastal Atlantic Ocean water. Oil
adsorbed onto active charcoal can be extracted in acetone and suspended in water with the aid of a detergent, when the turbidity of the
sample is measured (Sherratt, 1956; 1962). The detection limit is
about 1.0 p.p.m. ; the method will not determine water-miscible fractions and assumes a constant particle-size in the suspension, although
it may vary from component to component.
Marine oil pollution is often heavy enough not to require confirmation in the laboratory ; the problem is then to determine the probable
nature and source of the oil. This involves either comparisons with
suspected sources, if samples can be obtained from them, or an analysis
sufficiently detailed to characterize the polluting sample. A simple
comparative method utilizes the patterns revealed under ultra-violet
light after a crude form of paper chromatography. Schuldiner (1951)
allowed spots to spread in concentric circles, whereas Herd (1953)
suspended a paper strip overnight, dipping into an ether solution of
the oil, to obtain bands of varying width and density. These papers
can be stored for several years and have been used in successful prosecutions. Johannesson ( 1955) made similar comparisons, using the
vanadium and nickel content of ashed fuel-oils to determine the source
of harbour spillages. Brunnock et al. (1968) investigated the usefulness
of vanadium, nickel, sulphur, wax and asphaltene content in identifying
beach pollution. They also give distillation curves and n-paraffin
profiles of crude oils, their residues and beach deposits, concluding that
these data make it possible to determine which crude is responsible for
the pollution, at least amongst those normally entering European
waters. It is pointed out that tank sludges accumulate over a number
of voyages, whilst fuel-oils are usually blended from several different
crudes, so that pollution from these sources poses problems of analysis
and interpretation.
Rosen and Middleton (1955) adsorbed samples of polluting oil on
silica gel, eluting aliphatics with iso-octane and aromatics with benzene.
The infra-red absorption spectrum given by each fraction between 5
and 1611 proved sufficiently distinctive to match samples from the
known source of the oil. Their later work (Rosen et al., 1959) showed
A. NELYON-SMITIL
high sensitivity, but the absorption is due only to aromatic rings and
similar structures. Thus the precise nature of the polluting oil must
be known and the error can be quite large (Harva and Somersalo,
1958). Ultra-violet fluorescence is an extremely sensitive measure for
oils rich in aromatics, although interference may be experienced from
naturally-occurring polynuclear aromatic hydrocarbons. A detection
limit of 0.001 p.p.ni. is claimed and reference has been made above to
determinations of 0.003 p.p.m. in coastal Atlantic Ocean water. Oil
adsorbed onto active charcoal can be extracted in acetone and suspended in water with the aid of a detergent, when the turbidity of the
sample is measured (Sherratt, 1956; 1962). The detection limit is
about 1.0 p.p.m. ; the method will not determine water-miscible fractions and assumes a constant particle-size in the suspension, although
it may vary from component to component.
Marine oil pollution is often heavy enough not to require confirmation in the laboratory ; the problem is then to determine the probable
nature and source of the oil. This involves either comparisons with
suspected sources, if samples can be obtained from them, or an analysis
sufficiently detailed to characterize the polluting sample. A simple
comparative method utilizes the patterns revealed under ultra-violet
light after a crude form of paper chromatography. Schuldiner (1951)
allowed spots to spread in concentric circles, whereas Herd (1953)
suspended a paper strip overnight, dipping into an ether solution of
the oil, to obtain bands of varying width and density. These papers
can be stored for several years and have been used in successful prosecutions. Johannesson ( 1955) made similar comparisons, using the
vanadium and nickel content of ashed fuel-oils to determine the source
of harbour spillages. Brunnock et al. (1968) investigated the usefulness
of vanadium, nickel, sulphur, wax and asphaltene content in identifying
beach pollution. They also give distillation curves and n-paraffin
profiles of crude oils, their residues and beach deposits, concluding that
these data make it possible to determine which crude is responsible for
the pollution, at least amongst those normally entering European
waters. It is pointed out that tank sludges accumulate over a number
of voyages, whilst fuel-oils are usually blended from several different
crudes, so that pollution from these sources poses problems of analysis
and interpretation.
Rosen and Middleton (1955) adsorbed samples of polluting oil on
silica gel, eluting aliphatics with iso-octane and aromatics with benzene.
The infra-red absorption spectrum given by each fraction between 5
and 1611 proved sufficiently distinctive to match samples from the
known source of the oil. Their later work (Rosen et al., 1959) showed