Substantial organic impurities at the surface of synthetic ammonium sulfate particles

. Ammonium sulfate (AS) particles are widely used for studying the physical–chemistry processes of aerosols and for instrument calibrations. Small quantities of organic matter can greatly inﬂuence the studied properties, as observed by many laboratory studies. In this work, monodis-perse particles (200–500 nm aerodynamic diameter) were generated by nebulizing various AS solutions and organic impurities were quantiﬁed relative to sulfate using a high-resolution time-of-ﬂight aerosol mass spectrometer (HR-ToF-AMS). The organic content found in AS solutions was also tentatively identiﬁed using a liquid chromatography– tandem mass spectrometer (LC–MS). The results from both analytical techniques were consistent and demonstrated that the organic impurities contained oxygen, nitrogen, and/or sulfur, their molecular masses ranged from m/z 69 to 420, and they likely originate from the commercial AS crystals. For AS particle sizes ranging from 200 to 500 nm, the total mass fraction of organic compounds (relative to sulfate) ranged from 3.8 % to 1.5 %, respectively. An inorganic– organic mixture model suggested that the organic impurities were coated on the AS particle with a surface density of 1.1 × 10 − 3 g m − 2 . A series of tests were performed to remove the organic content (using pure N 2 in the ﬂow, ultra-pure water in the solutions, and very high AS quality), showing that at least 40 % of the organic impurities could be removed. In conclusion, it is recommended to use AS seeds with caution, especially when small particles are used, in terms of AS purity and water purity when aqueous solutions are used for atomization.

Most of the ammonium sulfate particles used in the laboratory are commercial.About 90% of ammonium sulfate is produced by 3 different processes: (1) as a byproduct caprolactam production, (2) from synthetic manufacture by combining anhydrous ammonia and sulfuric acid, and (3) as a coke oven byproduct by reacting the ammonia recovered from coke oven offgas with sulfuric acid.No detailed information is given about the potential organic compounds during these manufacture processes.

S3 Comparison between EXP P1 and EXP P8 :
The effect of the gas supplier (used for nebulizing the AS solution), was investigated by replacing compressed air (EXP P1) by pure N2 (EXP P8) Figure S3: [Org]/[Sulfate] mass ratios obtained from the AMS measurements in EXP P1 using compressed air and in EXP P8 using pure N2 (Linde Gas, 99.999 %).The AS concentration in the solution was the same for both experiments presented in this figure (0.5M).

S6 Influence of the AAC rotational speed on the chemical content
To understand the influence of the rotational speed on the detection of organic traces, AS aerosols at da = 300 nm were selected in EXP P7 (Table 1) under three different rotational conditions: 190, 285 and 369 rad/s.In Figure S6: , the ratios of mass concentrations Org/Sulfate and CxHyNx /Sulfate are shown as a function of the rotational speed of the concentric cylinder in green and blue dots, respectively.Their averages are 2.9±0.1% and 0.48±0.04%.No significant difference has been observed with varying the rotational speed of AAC.

S7 Multi-charging corrections for the [Org]/[Sulfate] using the DMA for particle size selection
Under the hypothesis that organic compounds were coated homogenously on the surface of AS aerosol particles, the density was defined as ρorg,S.The correction was done by removing the effects of multi-charged modes, so the corrected [Org]/[Sulfate] were represented only by the first mode.In a first step, [Org]/[Sulfate] were calculated using equation S5.1: Where   is the number concentration at diameter d, ∑ ( 2   ) is the total surface of all particles in the first mode measured by SMPS.∑ ( is the total volume of all particles in the first mode.  is the volume density of sulfate in AS aerosols, 96.06 g.cm -3 . Experimentally, [Org]/[Sulfate] mass concentration ratios were measured by the HR-ToF-AMS which considered all the particle sizes (including multi-charged modes) as described by equation SI7.2: In the last step, combining Equation S7.1 and Equation S7.2, the corrected [Org]/[Sulfate] were obtained (equation S7.3), they gather the total amounts of organics and sulfate on the first DMA mode: Where all diameter information was recorded by the SMPS.

S9 Comparison between experiments C9 and C10: LC/ESI+-MS base peak chromatogram of acetonitrile liquidliquid extracts of aqueous AS solutions from ACROS Organics™ and EMSURE®,. S10 LC/ESI+-MS-MS of the most intense ions found in the acetonitrile liquid-liquid extracts of aqueous AS solution and their fragments (experiment C9).
Precursor Where SS is the supersaturation, D and d are the droplet diameter and the particle diameter, respectively,  is the hygroscopicity,  / is the surface tension at the interface between droplet and air,   is the molar mass of water, R is the ideal gas constant,   is the density of water, and T is the temperature, fixed at 298.15 K.
For the estimation of SS in the presence of organic impurities, one needs to evaluate the values of effective kappa eff and  / on AS particles.To do so, two extreme hypotheses are explored for the organic fraction: -Hypothesis 1: the organic fraction is considered non-surface-active with  = 0.1.In this case, the surface tension is considered the same as pure water, i.e., 72 mN.m -1 ; and the effective  (  ) is calculated following the ZSR (Zdanovskii, Stokes, and Robinson) assumption (Stokes and Robinson, 1966).With a mass fraction [Org]/[Sulfate] = 3.8%, the organic fraction in the particle,  is 2.8%.And thus:   =  × 0.1 + (1 − ) × 0.61 = 0.59 -Hypothesis 2: the organic fraction is considered as extremely surface-active using surfactin as the proxy for surfactant.In this case,  is fixed at 0.61 and the surface tension varies according to the equilibrium surface tension isotherms of aqueous solutions of surfactin at various concentrations from Ekström et al., 2010 (Table S11).
Table S11: Simulation of the um surface tension isotherms of aqueous solutions of surfactin at various concentrations (Ekström et al., 2010).
Surfactin concentrations (mol.L -1 ) Following hypothesis 1 and hypothesis 2, the supersaturation was calculated along the droplet activation, and the corresponding Köhler curves are shown in Figure S11 where the CCN activation curve of pure AS particles were added for comparison.The results show that whereas the critical supersaturation is not significantly impacted by the presence of organic impurities under hypothesis 1, it is highly impacted under hypothesis 2, with a potential error of more than 70%.

S12 Investigation of the concentrations of organic surface-active species in an aqueous solution of AS
Organic surface-active species are amphiphile molecules, also called surfactants.Their potential presence in a concentrated 500 g.L -1 solution of AS (Across 99.5 %) were quantified, following a method adapted from Noziè re et al., 2017.This quantification method enables to differentiate surfactants by class, cationic, anionic, and non-ionic.Very low concentrations of surfactants were detected, as shown in Table S12.No anionic surfactants were detected.
As the concentrations are below the quantification limit for cationic surfactants (65 nM) and for non-ionic surfactants (75 nM), these results should be interpreted with caution.Nevertheless, an upper limit of 140 nM for surfactants concentration was considered to investigate the implications of these results.
A concentration of 140 nM of surfactants in a 500 g.L -1 AS solution is equivalent to a [Orgsurfactant]/[Sulfate] mass ratio of 1×10 -5 %, taking cetyltrimethylammonium chloride (CTAC) as a standard for cationic surfactants and Triton X-100 as a standard for non-ionic surfactants.Because this ratio is 5 orders of magnitude lower than the total organic fraction detected in AS particles, it is likely that surfactants play a negligible role in organic impurities.
However, small concentrations of highly surface-active species can induce non-negligible effects on surface tension, and thus on CCN activity (Ovadnevaite et al., 2017;Noziè re et al., 2017).We have thus investigated a rough estimation of the surface tension induced by the amount of surfactants observed.Considering AS particles with dm = 130 nm (under our experimental conditions), the mass ratio of 1×10 -5 % [Orgsurfactant]/[Sulfate] is equivalent to a concentration of 6×10 -7 mol.L -1 of surfactants.Note that this concentration is an upper value for the concentration of surfactants upon activation of the particles.For any surface-active molecule, the surface tension of a 6×10 -7 mol.L -1 solution is the same as that of pure water as shown by surface tension isotherms (see for example Fig. 3 in Ekström et al., 2010;Fig.2 in Frossard et al., 2019;or Fig. 2 in Arabadzhieva et al., 2020).It is thus concluded that the CCN activity of AS particles with dm = 130 nm should not be affected by the presence of surfactants in AS crystals.
Table S14 Concentrations of surfactants by class in 10 mL of a 500 g.L -1 solution of AS (ACROS 99.5 %).The quantification limits of cationic and non-ionic surfactants are shown in the table.

Figure S4- 1 :
Figure S4-1: Linear plot of the AMS signals of the sum of NO2 + and NO + fragments (i.e.nitrate signal) versus the total organic signal during EXP P1.

Figure S6 :
Figure S6: Effect of the AAC rotational speed on the organic content in AS particles (at da = 300 nm) measured by the HR-ToF-AMS (in EXP P7).

and their associated molecular formula and retention times.
In this table, the bold m/z were studied in more details using LC/ESI + -MS-MS (shown in S9).

of the error made on the critical supersaturation during the CCN activation of AS aerosol particles if one assumes 100% AS, omitting the presence of organic impuritiesThis
Considering ideal solutions, a simple calculation using the -Köhler equation (equation S11.1) fromPetters and  Kreidenweis, 2007, was performed.