Nucleation and Growth of Nanoparticles in the Atmosphere.pdf

of 55
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
rXXXX American Chemical Society A | Chem. Rev. XXXX, XXX, 000–000 REVIEW Nucleation and Growth of Nanoparticles in the Atmosphere Renyi Zhang,* ,†,‡,§ Alexei Khalizov, † Lin Wang, ‡ Min Hu, § and Wen Xu † † Department of Atmospheric Sciences and Department of Chemistry, Center for Atmospheric Chemistry and Environment, Texas A&M University, College Station, Texas 77843, United States ‡ Department of Environmental Science & Engineering and Institu
  r XXXX American Chemical Society  A | Chem. Rev.  XXXX, XXX, 000 – 000 Nucleation and Growth of Nanoparticles in the Atmosphere Renyi Zhang,*  , †  , ‡  , §  Alexei Khalizov, † Lin Wang, ‡ Min Hu, § and Wen Xu † † DepartmentofAtmosphericSciencesandDepartmentofChemistry,CenterforAtmosphericChemistryandEnvironment,TexasA&MUniversity, College Station, Texas 77843, United States ‡ Department of Environmental Science & Engineering and Institute of Global Environment Change Research, Fudan University,Shanghai 200433, China § State Key Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences and Engineering,Peking University, Beijing, 100871, China CONTENTS 1. Introduction B2. Overview of Vapor Nucleation D2.1. Nucleation Theories and ComputationalApproaches D2.1.1. Classical Nucleation Theory D2.1.2. Kinetic Theories F2.1.3. Molecular Dynamics and Monte CarloMethods F2.1.4. Density Functional Theory G2.1.5. Nucleation Theorem G2.2. Nucleation Experiments H2.2.1. Adiabatic Expansion Approaches H2.2.2. Di ff  usion Chamber H2.2.3. Laminar Flow Chamber I2.2.4. Turbulent Mixing Chamber I2.2.5. Continuous Generation of NucleatingVapors from Chemical Reaction sources I2.2.6. Comparison between ExperimentalResults and Nucleation Theories I3. Nucleation of Nanoparticles in the Atmosphere K 3.1. Atmospheric Measurements K 3.1.1. Concentrations and Size Distributions of Atmospheric Nanoparticles L3.1.2. Chemical Composition of AtmosphericNanoparticles M3.1.3. Measurements of Charged and NeutralAtmospheric Clusters P3.2. Laboratory Studies R3.2.1. Binary Nucleation of H 2 SO 4  H 2 O R3.2.2. TernaryNucleationofH 2 SO 4  H 2 OInvolv-ing Ammonia and Amines T 3.2.3. Nucleation of H 2 SO 4  H 2 O Assisted byOrganic Acids U3.2.4. Nucleation of Iodine Oxides W3.2.5. Ion-Induced Nucleation X3.2.6. Chemical Composition, Reactivity, and Thermodynamics of Nucleating Clusters Y 3.2.7. Other Species AA3.3. Theoretical and Computational Studies AA3.3.1. Quantum Chemical Calculations AA3.3.2. Molecular Dynamics and Monte CarloSimulations AD3.4. Parameterizations of AtmosphericNucleation AF4. Growth of Nanoparticles in the Atmosphere AG4.1. Role of the Kelvin (Curvature) E ff  ect in Growthof Nanoparticles AH4.2. Condensation AI4.2.1. Condensation of Sulfuric Acid AI4.2.2. Condensation of Low-VolatilityOrganics AI4.3. Heterogeneous Reactions AJ4.3.1. Ammonia AJ4.3.2. Amines AJ4.3.3. Aldehydes AL4.3.4.  α -Dicarbonyls AM4.3.5. Alcohols AN4.3.6. Other Species AO5. Numerical Treatment of Ambient NanoparticleNucleation and Growth Rates AP5.1. Measured Nucleation and Growth Rates AP5.2. Condensation Sink of Low-Volatility Vapor AP5.3. Combined Growth Including Condensationand Intramodal/Extramodal Coagulation AP5.4. Derivation of Nucleation Rates fromAtmospheric Measurements AQ6. Summary and Future Research Needs ARAuthor Information ASBiographies ASAcknowledgment AT Glossary of Acronyms AT References AT  Received:  May 17, 2011  B | Chem. Rev.  XXXX, XXX,  000–000 Chemical Reviews REVIEW 1. INTRODUCTION This review intends to critically assess recent  󿬁 ndings relatedto nucleation and growth of atmospheric nanoparticles, with anemphasis on the understanding of these processes at a funda-mental molecular level. Aerosols (small particles suspended inair) can be directly emitted into the atmosphere from primary sources or be formed in the atmosphere through nucleation of gas-phase species. Aerosol nucleation events produce a largefraction of atmospheric aerosols. New particle formation occursin two distinct stages, 1 i.e., nucleation to form a critical nucleusand subsequent growth of the critical nucleus to a larger size(>2  3 nm) that competes with capture and removal of thefreshly nucleated nanoparticles by coagulation with pre-existingaerosols. Nucleation is generally de 󿬁 ned as creation of molecularembryos or clusters prior to formation of a new phase during thetransformation of vapor f liquid f solid. This process is char-acterized by a decrease in both enthalpy and entropy of thenucleating system (i.e.,  Δ H   < 0 and  Δ S  < 0). Hence, althoughthermodynamicallyfavorableaccordingtothe 󿬁 rstlawofthermo-dynamics, (i.e., exothermic) nucleation is hindered in entropy according to the second law of thermodynamics. A free energy  barrier, Δ G ( Δ G = Δ H    T  Δ S >0),isofteninvolvedandneedsto be surmounted before transformation to the new phase becomesspontaneous. Another major limitation in the nucleation andgrowth of atmospheric nanoparticles lies in signi 󿬁 cantly elevatedequilibrium vapor pressures above small clusters and nanoparti-cles,alsoknownastheKelvin(curvature)e ff  ect,whichconsiderably restricts growth of freshly nucleated nanoparticles.Formation of molecular clusters occurs through randomcollisions and rearrangements of atoms or molecules of theexisting phase (Figure 1a). Growth of a cluster can be repre-sented as a reversible, stepwise kinetic process. After reaching acritical size (the critical cluster or nucleus), further growth of thecluster becomes spontaneous. At each step, formation anddecomposition of a cluster can be described by fundamentalkineticrate theories. A cluster can form  homogeneously  within thesrcinal phase or  heterogeneously  on various irregularities, such aspre-existing small particles or ions, which assist in surmountingthe free energy barrier associated with formation of an interface betweenthesmallclusterofthenewphaseandthesrcinalphase(Figure 1b). The lifetime of clusters is extremely short, but sincea very large number of clusters form and dissociate at any time, afew can reach the critical size and continue to grow sponta-neously to form larger particles. Atmospheric nucleation of aerosols from vapors 1,2 is, in principle, analogous to that of freezing of liquids, 3 crystallization of supersaturated solutions, 4 and formation of vapor bubbles inside the bulk liquid; 5 allproceed by the same basic mechanism. The common feature of the nucleation process is that there exists a dividing surface 6,7 atthe critical nucleus that separates the properties of the srcinalandnewphases.Fromanenergeticperspective,thefreeenergyof cluster formation,  Δ G  , increases with cluster size prior to butdecreases after the critical nucleus, reaching a maximal value atthe critical size, i = i *.Hence,thecriticalnucleuscanbeidenti 󿬁 edif the free energy surface leading to cluster growth is available 6 ð ∂ Δ G = ∂ i Þ i ¼ i   ¼  0 :  ð 1 : 1 Þ The properties of the critical nucleus are central to nucleationtheory. The rate at which nucleation occurs is related to thechemical makeup of the critical nucleus and the gaseous con-centrations of the nucleating species and is an important variablein simulations of aerosol formation in atmospheric models. 1 Nucleation from the vapor phase is  homomolecular   when asingle type of a gas is involved in formation of a critical nucleusand  heteromolecular   when several types of gases are involved information of a critical nucleus. In the absence of existingheterogeneities, homomolecular nucleation requires an extre-mely high supersaturation. For instance, homogeneous nuclea-tion of pure water vapor requires a supersaturation of a few hundred percent. Since such a condition is hardly realized in theatmosphere, homomolecular nucleation of water vapor, leadingtoformationofclouddroplets,isalwaysheterogeneousinnature,taking place on pre-existing water-soluble seeds, i.e., cloudcondensation nuclei (CCN). In fact, clouds would have neverformed in the Earth ’ s atmosphere in the absence of CCN.Homogeneous nucleation of atmospheric nanoparticles, thefocusareaofthisreview,isalwaysheteromolecular,involvingtwo(binary), three (ternary), or possibly more mutually interacting Figure 1.  Schematic representation of the transformation from the molecular complex through the critical nucleus to 2  3 nm nanoparticle (top) andassociated free energy variation (bottom). (Reprinted with permission from ref 1. Copyright 2010 American Association for the Advancement of Science.)  C | Chem. Rev.  XXXX, XXX,  000–000 Chemical Reviews REVIEW  vapors (multicomponent). The abundance, volatility, and reac-tivity likely determine the potential of a chemical speciesas a nucleation precursor. Atmospheric aerosol formation isclosely linked with the gas-phase chemistry because the abun-dances required for nucleation to occur are achieved through agradual increase in the concentration of the nucleating vaporsproduced from photo-oxidation of atmospheric gases, such assulfur dioxide and volatile organic compounds (VOCs), includ-ing many saturated, unsaturated, or aromatic hydrocarbons,SO 2  þ  OH  f O 2  ,H 2 O H 2 SO 4  ð 1 : 2 Þ  VOCs  þ  OH f O 2 oxidized organics  ð 1 : 3 Þ The most common nucleatingspecies is sulfuricacid because of its low vapor pressure at typical atmospheric temperatures, which is further reduced in the presence of water due to thelarge mixing enthalpy of these two substances. 8  10 The pre-sence of gaseous H 2 SO 4  in concentrations exceeding 10 5 molecules cm  3 has been shown as a necessary condition toobserve new particle formation in the atmosphere. 11,12 Inaddition to sulfuric acid, a number of other nucleating pre-cursors, including atmospheric ions, ammonia, amines, organicacids, and iodine oxides, have been proposed to be involved information of the critical nucleus under di ff  erent ambientenvironments. The size and chemical make up of atmosphericcritical nuclei are not well-known presently, because of the lack of existing analytical methods to directly probe the criticalnucleus. Indirect measurements and theoretical calculationssuggest that the critical nucleus has a diameter on the order of 1nmandconsistsofarelativelysmallnumberofmoleculesheldtogether by noncovalent van der Waals (vdW) interactions.Since the molecules of known nucleating vapors possess asigni 󿬁 cant dipole moment and/or contain a hydrogen atomconnected with an electronegative atom (nitrogen or oxygen),electrostatic, polarization, and hydrogen-bonding interactionshave been recognized to play a signi 󿬁 cant role in formation of the smallest clusters. As clusters grow, proton transfer from anacid moiety (e.g., H 2 SO 4 ) to a base moiety (e.g., H 2 O or NH 3 ) becomes possible because the resulting ion pair is stabilized by interactions withsurrounding polar molecules (e.g., H 2 O) within the cluster. Formation of an ion pair can signi 󿬁 cantly increase the nucleation rate by reducing the free energy of the critical nucleus. However, current understanding of the role of proton transfer and other possible chemicalprocesses in the nucleation of atmospheric clusters is stillinadequate. Aerosolnucleationevents,whicharere 󿬂 ectedasepisodeswith very high concentrations (up to 10 4 particle cm  3 or higher) of nanoparticles generated in a short period of time, are frequently observed in the free troposphere and under remote, urban,forested, and marine environments of the lower troposphere.Thermodynamically stable larger clusters and small nanoparti-cles formed during a nucleation event need to grow quickly sothat they are not scavenged by coagulation through collisions with existing larger particles. The surface of pre-existing particlesalso acts as a condensation sink for nucleating vapors, reducingtheir concentration and inhibiting nucleation. Whereas conden-sation of low-volatility vapors and reversible partitioning of semivolatile vapors are commonly recognized as the majorcontributors to growth of aerosols, the role of heterogeneouschemical reactions between gas-phase chemical compounds andparticles is not well understood and is a subject of intensiveresearch. 13  When reaching a size of about 50  100 nm, aerosols become e ffi cient light scatterers and CCN. 14 Overall, during theatmospheric lifetime, the size of particles may vary over 5 ordersof magnitude, from a lower limit of about 1 nm corresponding tostable molecular clusters to an upper limit of about 1 mm forcloud droplets. Growth of nanoparticles driven by condensation,partitioning, heterogeneous chemical reactions, and coagulationis another focus area of this review. Atmospheric aerosols have profound impacts on the Earth  atmosphere system, in 󿬂 uencing the weather, climate, atmo-spheric chemistry and air quality, ecosystem, and publichealth. 15 Those particles cool the atmosphere by directly scattering a fraction of the incoming solar radiation back tospace, an e ff  ect commonly referred to as direct climate forcing.By acting as CCN and ice nuclei (IN), aerosols play animportant role in controlling cloud formation, development,and precipitation, impacting the albedo, frequency of occur-rence, and lifetime of clouds on local, regional, and globalscales, 16  21  which is often referred to as indirect climateforcing. Presently, the aerosol direct and indirect e ff  ectsrepresent the largest uncertainty in climate predictions. 22  Also,chemical reactions occurring on the surface or in the bulk of aerosols 23,24 may alter the properties of aerosols and thegaseous composition of the atmosphere. For example, hetero-geneous reactions on particle surfaces convert inactive chlorinespecies into photochemically active forms in the middle atmo-sphere (between 20 and 50 km altitudes), leading to depletionof stratospheric ozone, 25  35  which acts as a UV shield. In thelower atmosphere (below 20 km), particle-phase reactions canmodulate formation of tropospheric ozone, 36  41  which is a key criteria air pollutant. On the regional and local scales,  󿬁 neparticulate matter (i.e., aerosols smaller than 2.5  μ m or PM 2.5 )represents a major contributor to air pollution. 42 Elevatedconcentrations of PM 2.5  cause degradation in visibility, exacer- bate accumulation of pollutants in the planetary boundary layer(PBL), and adversely a ff  ect human health. 43 Increasing evi-dence has implicated aerosols not only in aggravation of existing health symptoms but also in the development of serious chronic diseases. 44  When inhaled, aerosols can amplify the adverse e ff  ect of gaseous pollutants, such as ozone, 45 andthesmallestparticlescausethemostseverehealthimpacts 46  becausethey have higher probability than larger particles to deposit in thepulmonary region and penetrate into the bloodstream. 47,48 Several previous review articles have provided a detailedaccount of di ff  erent aspects of new particle formation in theatmosphere, including  󿬁 eld measurements of atmospheric aero-solsandnucleationevents, 49  51 coastalnewparticleformation, 52 the relation between laboratory,  󿬁 eld, and modeling nucleationstudies, 53,54 andtheroleofdi ff  erenttypesofnucleationprocessesin the atmosphere. 55 Over the past few years, there has beensubstantial research progress in the area of atmospheric aerosolnucleation, including development of novel detection methodsfor atmospheric nanoparticles and clusters, as summarized in arecent review by Bzdek and Johnston. 56  Advances in analyticalinstruments have led to a number of laboratory and  󿬁 eld studiesthat produced exciting yet often contradictory results regardingthe compositions of the critical nucleus and the role of sulfuricacid and other species in the nucleation and growth of nanoparticles. 57  61 In the present review, we  󿬁 rst provide the background information on theoretical and experimental ap-proaches towardinvestigation ofhomogeneous vapornucleation  D | Chem. Rev.  XXXX, XXX,  000–000 Chemical Reviews REVIEW and then introduce recent advances in nucleation and growth of atmospheric nanoparticles. Throughout this review, we strive topresent the various nucleation aspects from a fundamentalchemical prospective. Since there is a vast body of literature inthe area of atmospheric aerosol nucleation, we do not attempt to be inclusive to cover all available publications on this subject.Instead, we choose in this review to focus on the studies thatmake the most important advances in this  󿬁 eld.In section 2, we introduce the nucleation theories andillustrate how the predicted nucleation rates are related withresults of laboratory experiments for a number of simple nucle-ating systems. Nucleation of atmospheric aerosols, includingambient measurements, laboratory experiments, and theoreticalstudies, is described in section 3. Results of laboratory experi-ments and ambient measurements of nanoparticle growth arepresented in section 4, and section 5 provides the numericalapproaches developed to connect measured aerosol nucleationand growth rates. Section 6 contains the concluding remarks anddescribes future research needs. A glossary of acronyms isprovided at the end of the review. 2. OVERVIEW OF VAPOR NUCLEATION 2.1. Nucleation Theories and Computational Approaches In the absence of heterogeneities, formation of a new phaseoccursthroughrandom 󿬂 uctuationsinthevapordensity,generatingclusters that can grow or decay by gaining or losing a monomermolecule. Growth of the cluster can be represented by a reversible,stepwise kinetic process in a single or multicomponent system ::: C s f þ  A  i  1 , k  þ i    1 i    1 r s k   i C  s f þ  A  i  , k  þ i i r s k   i  þ  1 C i þ 1 :::  ð 2 : 1 Þ  where A  i  1  denotes a monomer species to be added to the clusterC i  1  at the ( i    1)th step and  k  i  and  k  i + represent the clusterdecomposition and association rate constants, respectively. A complete nucleation theory can be established to describe theevolutionofthepopulationofclusters,i.e.,theratesandmechanism bywhichtheseclustersgrowanddecay.Asre 󿬂 ectedbyequation1.1,thefreeenergyofthenucleatingsystemreachesamaximum(i.e.,thenucleation barrier) when the critical nucleus forms. In addition, amulticomponent system may exhibit multiple nucleation barriers,leading to further complication in the identi 󿬁 cation of the criticalnucleus onthebasisofthefreeenergysurfaceoftheclustergrowth.Kinetically,atthecriticalnucleus,theratetoformthe( i +1)thcluster is equal to that of decomposition of the critical nucleus toform the ( i    1)th cluster, i.e. k   i  ½ C  i   ¼  k  þ i  ½  A  i ½ C  i   ð 2 : 2 Þ  where [A  i ] and [C i ] are the number concentrations of theassociating monomer and the cluster of size  i  , respectively.Furthermore, since the molecular  󿬂 ux between adjacent clustersachievestheminimumatthecriticalnucleus(commonlyreferredto as a bottleneck  7 ), another practical approach to locate thecriticalnucleus istovariationallyminimizethemolecular 󿬂 uxasafunction of the cluster sited F  i = d i  ¼  0  ð 2 : 3 Þ  where  F  i  is the number of clusters growing from a size  i  to a size i  +1 per second. Therate ofnucleation,  J   ,isde 󿬁 ned asthe rate of growth of the critical nucleus  J   ¼  k  i þ ½ C  i  ð 2 : 4 Þ The association and decomposition rate constants can becalculated employing the kinetic rate theories, such as transitionstate theory (TST). 62 For each cluster, the association rate isrelated to the dissociation rate by detailed balance 63  70 k  þ i    1 k   i ¼  Q  C  i Q  C i  1 Q   A i  1 exp  D C  i kT    ð 2 : 5 Þ  where  Q  C i * is the partition function of the critical nucleus,  Q  C i  1 and Q   A  i  1 arethepartitionfunctionsoftherespective( i  1)thclusterandmonomer,  k   is the Boltzmann constant,  T   is the temperature, and  D C i *isthebindingenergyofthecriticalnucleusrelativetomonomerand ( i    1)th cluster. The decomposition rate constant of eachcluster can be calculated according to the following expression 63  70 k   i  ¼  kT hQ  q C  i Q  C  i exp   Δ  EkT    ð 2.6 Þ  where  Q  C  i * q is the partition function of the transition state,  h  is thePlanck constant, and Δ  E  is the transition state energy relative to thecritical nucleus. In the case that the association reaction proceeds withoutanactivationbarrier(aloosetransitionstate),thelocationof thetransitionstatecanbedeterminedvariationallybyminimizingthedecomposition reaction rate constant using canonical variationaltransition state theory (CVTST). 71 The partition functions requiredfor eqs 2.5 and 2.6 can be evaluated by treating the rotational andtranslational motion classically and treating vibrational modes quan-tum mechanically. Vibrational frequencies, moments of inertia, andreactionenergiescanbetakenfromquantumchemicalcalculations. 72  An example of such an approach is the dynamical nucleation theory (DNT) of Kathmann et al., 73  75  which uses CVTST to locatetransition states and calculate evaporation rate constants  k  i  for eachstage of the nucleation process.Depending on the assumptions and approximations made,three major types of theoretical approaches have been estab-lished to characterize the nucleation process. Phenomenologicaltheories, e.g., classical nucleation theory, attempt to obtain thefree energy of formation of the critical nucleus from macroscopicparameters, such as the surface tension and the bulk liquid density.Some kinetic theories derive the cluster distribution and hence thenucleation rate by calculating rate constants for association anddecomposition of clusters, avoiding explicit evaluation of clusterformation energies from macroscopic parameters. Molecular-scaleapproaches, including molecular dynamics, Monte Carlo simula-tions,anddensityfunctionaltheory,apply  󿬁 rstprinciplestocalculatethe cluster structure and free energy of cluster formation. 2.1.1. Classical Nucleation Theory.  The classical nuclea-tion theory (CNT) was formulated by Becker and D € oring 76 andFrenkel 77 on the basis of the kinetic theory of nucleationestablished by the work of Volmer and Weber 78 and Farkas. 79 CNT includes the thermodynamic and kinetic components by evaluatingthefreeenergychangeofformationofanascentphasecluster and calculating the nucleation rate. The phenomenologi-calapproachtoCNTdescribesthenucleationprocessintermsof the change in Gibbs free energy of the system upon transfer of   i molecules from the vapor phase to an  i -mer cluster of radius  r  Δ G  ¼   ikT   ln  S  þ  4 π  r  2 σ   ð 2.7 Þ  where  S  =  p  A   /p  A S is the saturation ratio,  p  A   is the vapor pressureof substance A in the gas phase,  p  A S is the vapor pressure of 

Trial Civpro

Jul 23, 2017
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks