The formation of marine aerosols and cloud condensation nuclei--from which marine clouds originate--depends ultimately on the availability of new, nanometre-scale particles in the marine boundary layer. Because marine aerosols and clouds scatter incoming radiation and contribute a cooling effect to the Earth's radiation budget, new particle production is important in climate regulation. It has been suggested that sulphuric acid derived from the oxidation of dimethyl sulphide is responsible for the production of marine aerosols and cloud condensation nuclei. It was accordingly proposed that algae producing dimethyl sulphide play a role in climate regulation, but this has been difficult to prove and, consequently, the processes controlling marine particle formation remains largely undetermined. Here, using smog chamber experiments under coastal atmospheric conditions, we demonstrate that new particles can form from condensable iodine-containing vapours, which are the photolysis products of biogenic iodocarbons emitted from marine algae. Moreover, we illustrate, using aerosol formation models, that concentrations of condensable iodine-containing vapours over the open ocean are sufficient to influence marine particle formation. We suggest therefore that marine iodocarbon emissions have a potentially significant effect on global radiative forcing.

The secondary organic aerosol (SOA) yields from the laboratory chamber ozonolysis of a series of cycloalkenes and related compounds are reported. The aim of this work is to investigate the effect of the structure of the hydrocarbon parent molecule on SOA formation for a homologous set of compounds. Aspects of the compound structures that are varied include the number of carbon atoms present in the cycloalkene ring (C 5 to C 8 ), the presence and location of methyl groups, and the presence of an exocyclic or endocyclic double bond. The specific compounds considered here are cyclopentene, cyclohexene, cycloheptene, cyclooctene, 1-methyl-1-cyclopentene, 1-methyl-1-cyclohexene, 1-methyl-1-cycloheptene, 3-methyl-1cyclohexene, and methylenecyclohexane. The SOA yield is found to be a function of the number of carbons present in the cycloalkene ring, with an increasing number resulting in increased yield. The yield is enhanced by the presence of a methyl group located at a double-bonded site but reduced by the presence of a methyl group at a nondouble-bonded site. The presence of an exocyclic double bond also leads to a reduced yield relative to that of the equivalent methylated cycloalkene. On the basis of these observations, the SOA yield for terpinolene relative to the other cyclic alkenes is qualitatively predicted, and this prediction compares well to measurements of the SOA yield from the ozonolysis of terpinolene. This work shows that relative SOA yields from ozonolysis of cyclic alkenes can be qualitatively predicted from properties of the parent hydrocarbons.