Investigation of Black Carbon Wet Deposition to the United States from National Atmospheric Deposition Network Samples

Black carbon (BC) aerosols from burning biomass, fossil fuels, and waste are transported over large distances in the Earth's atmosphere, absorbing sunlight, altering climate, and impacting air quality. These aerosols are relatively short-lived in the troposphere and are returned to the surface by wet and dry deposition processes. Although wet deposition is considered the primary mechanism for removing BC from the atmosphere, published data is exceptionally scarce. In this study, we investigated the feasibility of determining BC wet deposition on a national/international scale using samples from the US National Atmospheric Deposition Program (NADP). The study investigated BC concentrations in precipitation by single-particle laser-induced incandescence (SP2). From October 26 th to December 1 st , 2020, we analyzed 478 NADP wet deposition samples from 209 locations, including sites in the United States, Canada and US territories, Puerto Rico, and the US Virgin Islands. Wet deposition BC concentrations varied from less than 0.3 µ g L –1 to 38.7 µ g L –1 with a median of 3.50 µ g L –1 . Associated BC wet deposition fluxes ranged from near zero to 9.1 g ha –1 wk –1 , with a median of 0.87 g ha –1 wk –1 . An analysis of the spatial variability indicated a pattern of higher BC deposition through the central United States consistent with BC transport from biomass burning during the sampling period.

Wet deposition analyses of refractory black carbon (rBC) were performed by single-particle incandescence using a single-particle soot photometer (SP2, Droplet Measurement Technologies, Boulder, Colorado) hyphenated with a desolvation nebulizer system (Fig. S3).Wet deposition rBC concentrations were determined using aqueous rBC standards (see Section S3.2) to derive linear calibration curves (Fig. S4) as a function of rBC aerosol concentration to rBC standard concentration.Water samples were pumped into the system through capillary tubing (high purity PFA, 1/16" OD 0.02" ID ) at ~ 350 µL/min by a peristaltic pump (M4, Elemental Scientific, blackblack PVC pump tubes) to a pneumatic nebulizer (Glass Expansion, Seaspray U series nebulizer -0.4 mL/min, ARG07USS04) connected to HEPA filtered compressed gas (air) set to 1000 VCCM at ~ 65 psi using a mass flow controller (MC, Alicat Scientific INC ).The nebulizer sprayed aerosol directly into a vertically mounted heated quartz glass chamber (145 ℃) with an inline cold zone chilled to 2℃ (Fig. S3).Condensed water was removed from the system by three peristaltic pumped drain tubes (red-red PVC tubing) collectively pumped at ~ 600 µL min-1.The SP2 was interfaced to the desolvation nebulizer using a borosilicate glass tee acting as an open split to allow the SP2 to maintain a sample airflow of 120 VCCM from the nebulizer system.Before running standards, the SP2 -nebulizer system was run with ultrapure water for 1 hr.The intracavity laser power output was monitored throughout the analyses and maintained at ≥ 5V.Analysis blanks consisted of ultrapure water (> 18MΩ).Blanks and rBC standards were analyzed at the beginning and end of the analysis.A blank, an quality control standard (fullerene soot), and a sample duplicate were performed after every ten samples.Fig. S3.Schematic of SP2 nebulizer/desolvation spray chamber system.

S3.2 Aqueous rBC standard preparation
Aqueous rBC standards were prepared gravimetrically from self-dispersing fullerene soot (FS).FS has been investigated as an SP2 standard to calibrate single particle incandescent peak responses to rBC mass and has been recommended as a reference material for SP2 analysis (Moteki, 2023;Baumgardner et al., 2012).However, to quantitatively disperse FS in water, FS must be partially oxidized or combined with a dispersing agent.Therefore, we prepared a self-dispersing FS by partial oxidation with nitric acid (gas).Preparation of the self-dispersing FS was as follows.
Approximately 17 mg of as-prepared fullerene soot (Sigma-Aldrich, 572497-5G) was weighed and transferred into a 50-mL polypropylene centrifuge tube (without the lid) and placed standing upright inside a 1-L closed container with approximately 200 mL of concentrated HNO3 for ~ 72 hrs.After exposure to HNO3 gas, the vial was removed from the container, reweighed, and filled with 50 g of ~ 0.002 M NH4OH (prepared with reverse osmosis water, RO, pH ~ 9.6).The partially oxidized soot was dispersed in the tube by shaking (with a lid) for ~ 1 min, followed by ultrasonication for 30 mins at room temperature.Aqueous BC standards, ranging from 0.5 x 10-9g g-1 to 50 x 10-9g g-1 (µg L -1 ), were prepared weekly from the primary standard by serial dilution in 250-mL amber-colored glass bottles with 0.002 M NH4OH.

S3.3 Aqueous rBC quantification
Broadband SP2 incandescent peak amplitudes (high-gain and low-gain) were converted to refractory BC (rBC) mass using a broadband peak amplitude to mass relationships supplied by the SP2 manufacturer (Droplet Measurement Inc.).The SP2 instrument used in the study can determine rBC masses in the range of 0.3 fg to 80 fg.The analysis is, therefore, limited to rBC masses in the 0.3 -80 fg mass window.This mass limit is based on the sensitivity of the SP2 highgain broadband detector for the low masses and the saturation of the broadband low-gain detector for the high masses.However, high concentrations of large rBC are not expected in wet-deposition samples, and single-particle rBC in the atmosphere are typically in the mass range determined by the SP2.Therefore-losses of large BC particles in the nebulizer-desolvation system may also limit the upper single-particle mass range.The nebulizer-desolvation system was custom-built to reduce losses of larger rBC particles by using a straight tube heated spray chamber in preference to a heated cyclonic spray chamber system used by some commercial nebulizer-desolvation systems.Mass distributions of rBC resulting from the self-dispersing FS standards and rBC from wetdeposition samples (Fig. S5) were similar after transport through the nebulizer system, suggesting that self-dispersing FS standards are a good match for rBC found in wet-deposition.
Individual rBC particle masses were summed for a given time interval and converted to aerosol concentrations using equation S1.

𝑟𝑟𝑟𝑟𝑟𝑟 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 (𝑔𝑔 𝑐𝑐𝑐𝑐
Aerosol concentrations were determined from the summed masses (5 sec) of incandescent rBC particles in the mass window of 0.3 fg to 80 fg.
Aqueous rBC concentrations were estimated using an external linear calibration (Eq.S2) derived from the rBC aerosol concentrations resulting from aqueous FS standards.External calibrations were conducted before sample analyses and at the end of the analysis.An analysis blank, a quality control standard, and a duplicate sample were analyzed after every ten samples.An example calibration curve is shown in Fig. S4.

S5. rBC recovery from NADP wet-dep buckets and shipping bottle
Recoveries from the NADP wet deposition buckets and shipping bottles were calculated as a percentage of the initial concentration compared to the concentration determined after 1 week following Eq.S3.    (Rolph, Stein and Stunder, 2015;Stien et al., 2015) air mass back trajectories (frequency, % trajectory endpoints) from November 22 nd 12:00 UTC, 2020, to November 17 th UTC, 2020, for (a) Missouri site M05 (Lat: 36.910800Lon: -90.318700) and (b) Arkansas site AK16 (Lat: 36.0842,Lon: -92.5868).The trajectories (from the NOAA real-time environmental applications and display system) were initiated with starting heights at 2000 m above ground level using GDAS1 meteorology, a frequency resolution of 1 x 1 degree and model vertical velocity. a. b.

Fig. S1 .
Fig. S1.Site classification of NADP NTN sites included in the study.

S2.
Wet deposition samples from Madison, Wisconsin, USADuring August and September 2021, we conducted a study at the NADP Eagle Heights test site in Madison, Wisconsin (43.11478°N, -89.43251°W) to examine the losses of refractory black carbon (rBC) to wet deposition sample containers (buckets and shipping bottles), and the variability between co-located automated wet-deposition samplers.The samplers included an ACM NTN wet-deposition sampler (Aerochem Metrics, wet/dry sampler) and two NCON NTN wet-deposition samplers (NCON Systems Inc., ADS 120 sampler).The ACM sampler (Fig.S2, KJJ1) included two side-by-side NADP HDPE buckets to collect wet and dry deposition samples, a mechanized lid, and a precipitation sensor to trigger the lid motion between buckets.The study only collected samples from the wet deposition bucket.The NCON ADS NTN samplers (Fig.2, NCON2, and NCON5) were wet-deposition-only samplers with the bucket cover resting over a splash shield during precipitation events and an infrared optical precipitation sensor.The sensor triggered the sampler's cover to open after ~ 5 precipitation drops and returned the cover within two minutes after the end of an event.

Fig. S2 .
Fig. S2.NTN automatic deposition collectors at NADP Eagle Height test site in Madison, WI, USA (Courtesy of R. Edwards).
Fig. S4.Fullerene soot external calibration.The calibration curve is comprised on two sets of calibration standards.The first set was analyzed at the beginning of the analysis and the second set at the end analysis.Responses are reported as fg rBC cm -3 unit.

classification based on population density within a 15-km radius of the sampling site
Note: 1 Country: USA -United States; CAN -Canada 2 US Territories (PR = Puerto Rico, VI = Virgin Islands) Site (based on NADP, 2022).

Table S2 2020
NADP NTN wet deposition rBC concentrations and fluxes

Table S3
Changes in BC concentrations of samples in buckets and bottles used in water leaching test.

Table S4 .
Precipitation collected, BC concentrations measured, and fluxes estimated using colocated samplers at the Eagle Heights NADP test site (43.11478°N, -89.43251°W)