Iodine chemistry
is an important driver of new particle formation
in the marine and polar boundary layers. There are, however, conflicting
views about how iodine gas-to-particle conversion proceeds. Laboratory
studies indicate that the photooxidation of iodine produces iodine
oxides (I
x
O
y
), which are well-known particle precursors. By contrast, nitrate
anion chemical ionization mass spectrometry (CIMS) observations in
field and environmental chamber studies have been interpreted as evidence
of a dominant role of iodic acid (HIO
3
) in iodine-driven
particle formation. Here, we report flow tube laboratory experiments
that solve these discrepancies by showing that both I
x
O
y
and HIO
3
are involved in atmospheric new particle formation. I
2
O
y
molecules (
y
= 2,
3, and 4) react with nitrate core ions to generate mass spectra similar
to those obtained by CIMS, including the iodate anion. Iodine pentoxide
(I
2
O
5
) produced by photolysis of higher-order
I
x
O
y
is hydrolyzed,
likely by the water dimer, to yield HIO
3
, which also contributes
to the iodate anion signal. We estimate that ∼50% of the iodate
anion signals observed by nitrate CIMS under atmospheric water vapor
concentrations originate from I
2
O
y
. Under such conditions, iodine-containing clusters and particles
are formed by aggregation of I
2
O
y
and HIO
3
, while under dry laboratory conditions,
particle formation is driven exclusively by I
2
O
y
. An updated mechanism for iodine gas-to-particle
conversion is provided. Furthermore, we propose that a key iodine
reservoir species such as iodine nitrate, which we observe as a product
of the reaction between iodine oxides and the nitrate anion, can also
be detected by CIMS in the atmosphere.