ZHANG Yun, PANG Shu-feng, ZHANG Yun-hong

Institute of Chemical Physics, School of Chemistry, Beijing Institute of Technology, Beijing 100081, China

A Relative Humidity Pulse Approach to Observe Mass Transfer Processes Controlled by Bulk and Surface for Aerosols

ZHANG Yun, PANG Shu-feng, ZHANG Yun-hong*

Institute of Chemical Physics, School of Chemistry, Beijing Institute of Technology, Beijing 100081, China

With combination of a pulse relative humidity (RH) controlling system and rapid scan vacuum FTIR technique, dynamic hygroscopicity of aerosol can be studied during pulse RH process. The time-resolved FTIR spectra can provide both water content of aerosols and water vapor amount of the aerosol ambient in sub-second time resolution. Experiments were performed on sodium nitrate, magnesium sulfate and magnesium nitrate aerosols. By comparing their hygroscopicity in pulse RH process and quasi-equilibrium state, for sodium nitrate aerosols, under time resolution of 0.12 s, we didn’t see water transfer delay between aerosols and ambient environment. For magnesium sulfate aerosols, after gel formation, the water transfer speed is limited by the aerosol bulk phase. While for aged magnesium nitrate aerosols, non-soluble species generated and formed a film on the surface of aerosol particles, which slow down the water exchange rate between aerosols and ambient environment. This method turned out to be an efficient and convenient tool to elucidate the water transfer process controlled by bulk and surface for aerosols.

Relative humidity pulse； Aerosol； Mass transfer process

Introduction

Atmospheric aerosols continuously exchange water to or from gas phase during evaporation or condensation, known as aerosol hygroscopicity, which alters phase state, size, concentration and morphology of aerosols[1-4]. This greatly influences extinction ability and climate with aerosol acting as cloud condensation nuclei (CCN)[5-9]. Usually the water transport between ambient and aerosols are assumed to occur instantaneously and are estimated by solution thermodynamics. However, for high viscosity, glassy and amorphous aerosols, the time scale of water exchanging between aerosols and ambient may be significant[10]. To understand aerosol properties, time-resolved measurements need to be built on studying aerosol dynamic hygroscopicity and time scale of water transportation.

We built up a system by combing Vacuum FTIR spectrometer together with a home-made sample cell. As shown in Scheme 1, micrometer-sized aerosol particles were deposited on inner side of ZnSe windows. Sample cell was first exhausted to 0.01 kPa to exclude air. Then water vapor was guided into the cell to adjust RH. The RH change rate can be as fast as 40% per second in a RH pulse. With interferometer compartment and sample compartment of the spectrometer exhausted to 0.2 kPa, FTIR spectra can provide both aerosol condensed phase and ambient gas phase information. For example, in Scheme 1 “Spectrum A” is the raw spectrum of sucrose aerosol contains information of both aqueous droplets and water vapor. Vapor absorption bands intensity was used to calculate RH. After deducted vapor absorption band we get “Spectrum B”, it offers us information about sample’s water content, molecular structure changes, phase transition and even size changes.

Scheme 1

After aerosol sample was prepared, we first slowly change RH to get the quasi-equilibrate state water content at various RHs, drawing out the water band area to RH curves as a reference. Then RH pulses were applied, high quality spectra with time resolution of 0.12 s for one spectrum make it possible to accurately measure real-time RH and aerosol water content, which also offer us a chance to study aerosol dynamic hygroscopicity. By comparing the water band area to RH data in RH pulses and in the quasi-equilibrate state, it is easy to see whether aerosols can keep water content balance to rapidly changing ambient RH. So we designed a series experiments to study the equilibrium time scale for different aerosols, the results were shown below.

Fig.1 RH traces (a) and their corresponding water band area to RH curves (b) of sodium nitrate aerosols

Fig.2 RH traces (a) and their corresponding water band area to RH curves (b) of magnesium sulfate aerosols

Fig.3 RH traces (a) and their corresponding water band area to RH curves (b) of magnesium nitrate aerosols.

Figure 1(a) shows 6 RH pulse traces across different RH ranges and with various rate of change for sodium nitrate (SN) aerosols. Their corresponding water content was plotted versus RH in Figure 1(b). The gray line in Figure 1(b) represents the quasi-equilibrate (QE1) water content, which was obtained from the slowly dehumidification of SN aerosols. By comparing their water band area to RH curves, we can see almost all plots overlapped with the gray line. Considering the time resolution of FTIR spectra, the conclusion is even when RH change rate is as fast as 40% per second, water transport to or from SN aerosols to get equilibrium state in less than 0.12 s. Meaning that in this RH range, SN aerosols were presented as diluted aqueous droplets as shown in figure 1(b), we didn’t observed non-equilibrium between gas phase and aerosols in our experiment.

In Figure 2(a) and (b), similarly, the gray lines give the quasi-equilibrium (QE2) water content of magnesium sulfate (MS) aerosols with slowly changing RH. Compare to RH pulse B1 and B2, In RH pulse B3 and B4, the water band area to RH plots no longer overlaps with the gray line, suggesting that there is a mass transfer limited process. MS aqueous droplets were observed to form gel structure at low RHs[11-12]. In this work, when RH lower than 40%, decreasing of water content caused formation of gel structure in MS aqueous solution droplets bulk phase as shown in figure 2(b). Bulk phase gel structure slows down the water diffusion in aerosol and increases the time scale for aerosols to get equilibrium, indicating the mass transfer process is controlled by the MS aerosol bulk phase. While, water content in pulse B5 and B6 show less difference to those in quasi-equilibrate state because their RH change rates are lower. Which means with higher RH change rate is more sensitive to detect water transport limited process.

In Figure 3 (a) and (b), the gray line helps us estimate the quasi-equilibrium (QE3) water content of magnesium nitrate (MN) aerosols. RH pulses C1 and C2 were performed on the new-made MN aerosols. As their water band area to RH curves overlaps pretty good, we can say new-made MN aerosols equilibrate to gas phase in less than 0.12 s even when the concentration of solution reaches about 50% (w/w) corresponding to 35% RH[13-14]. However, after MN aerosols deposited for a period of time, the situation is much different from the new-made ones. RH pulses C3 and C4 were performed on the MN aerosols deposited for 7 hours, the water band area to RH curves in RH pulses no longer overlap with those in quasi-equilibrium state, suggesting MN aerosols need longer time to equilibrate to gas phase. This water transport delay is not caused by viscous bulk phase as we didn’t see non-equilibrium for new made MN aqueous droplets under the same concentration. There must be non-soluble species generated, which formed a film on the surface of aerosols, as shown in Figure 3 (b), the existence of the thin film slows down water transport though aerosol-gas interface, leading to the longer time needed to reach the equilibrium between aerosols and ambient.

As a conclusion, we established a new method to in-situ study aerosol dynamic hygroscopicity in pulsing RHs. We tested aerosol particles of sodium nitrate, magnesium sulfate and magnesium nitrate for comparison, our results showed that formation of gel and surface film will slow down water diffusion in aerosol bulk phase and aerosol-gas interface respectively, then increase the time scale for water transport to or from aerosols. The results suggested that the presented method is convenient and sensitive to detect aerosol equilibrium time scale under various ambient conditions.

[1] Orr C, Hurd F K, Corbett W J. Journal of Colloid Science, 1958, 13(5): 472.

[2] Kapustin V N, Covert D S. Appl. Opt., 1980, 19(8): 1349.

[3] Hemming B L, Seinfeld J H. Industrial & Engineering Chemistry Research, 2001, 40(20): 4162.

[4] Rosenoern T, Schlenker J C, Martin S T. J. Phys. Chem. A, 2008, 112(11): 2378.

[5] Fitzgerald J W. Appl. Opt., 1984, 23(3): 411.

[6] Sinclair D, Countess R J, Hoopes G S. Atmos. Environ., 1974, 8(11): 1111.

[7] Held J L, Cooper D W. Atmos. Environ., 1979, 13(10): 1419.

[8] Goodale M, Mansfield R. Neuropsychologia, 1987, 25: 97.

[9] Heintzenberg J. Tellus. B, 1989, 41(2): 149.

[10] Krieger U K, Marcolli C, Reid J P. Chemical Society Reviews, 2012, 41(19): 6631.

[11] Chan C K, Flagan R C, Seinfeld J H. Journal of the American Ceramic Society, 1998, 81(3): 646.

[12] Wang F, Zhang Y H, Li S H, et al. Analytical Chemistry, 2005, 77(22): 7148.

[13] http://www.aiomfac.caltech.edu.

[14] Zuend A, Marcolli C, Luo B P, et al. Atmospheric Chemistry and Physics, 2008, 8(16): 4559.

*通訊聯(lián)系人

O351.2

A

脈動(dòng)相對(duì)濕度方法觀測(cè)由體相及界面控制的氣溶膠傳質(zhì)過程

張 云， 龐樹峰， 張韞宏*

北京理工大學(xué)化學(xué)學(xué)院， 化學(xué)物理研究所， 北京 100081

設(shè)計(jì)搭建了可以控制樣品室內(nèi)相對(duì)濕度發(fā)生脈動(dòng)式變化的壓力控制裝置。 將這種裝置與真空紅外快速掃描技術(shù)聯(lián)用， 采集的時(shí)間分辨傅里葉變換紅外光譜可以提供在亞秒時(shí)間尺度內(nèi)， 脈動(dòng)相對(duì)濕度的準(zhǔn)確數(shù)值以及濕度變化過程中氣溶膠顆粒的含水量。 分析這些數(shù)據(jù)可以了解相對(duì)濕度脈動(dòng)變化過程中氣溶膠的動(dòng)態(tài)吸濕特性。 選擇硝酸鈉， 硫酸鎂和硝酸鎂三種無機(jī)鹽氣溶膠為研究對(duì)象， 比較了他們?cè)谙鄬?duì)濕度發(fā)生脈動(dòng)變化和準(zhǔn)穩(wěn)態(tài)變化條件下的吸濕特性。 結(jié)果發(fā)現(xiàn)， 實(shí)驗(yàn)0.12 s的時(shí)間分辨率下不足以觀察到水在硝酸鈉氣溶膠和環(huán)境之間的傳質(zhì)受阻過程。 而對(duì)于老化的硫酸鎂氣溶膠顆粒， 膠態(tài)的形成減緩了水的擴(kuò)散速率， 體相傳質(zhì)成為速控步驟。 對(duì)于老化的硝酸鎂氣溶膠顆粒， 由于在顆粒表面難溶物的生成和富集， 使得界面水的傳質(zhì)速率成為氣溶膠與環(huán)境發(fā)生傳質(zhì)的決定性因素。 這證明脈動(dòng)壓力變化裝置與快速掃描真空紅外聯(lián)用可以有效便捷地觀測(cè)區(qū)分體相和界面控制的氣溶膠傳質(zhì)過程。

脈動(dòng)相對(duì)濕度； 氣溶膠； 傳質(zhì)動(dòng)力學(xué)

2014-12-18，

2015-04-13)

2014-12-18; accepted： 2015-04-13

The NSFC (41175119, 21373026, and 21473009)

10.3964/j.issn.1000-0593(2016)03-0887-03

Biography： ZHANG Yun, (1987—), graduate student of Beijing Institute of Technology e-mail: yun5337@163.com *Corresponding author e-mail: yhz@bit.edu.cn