Particle transport in a parallel-plate semiconductor reactor: Chamber modification and design criterion for enhanced process cleanliness

Nijhawan, S.; McMurry, Peter H.; Campbell, Stephen A.

doi:10.1116/1.1288193

Cited by 14 publications

(11 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The deposition of colloidal- and nano-scale particles on surfaces is critical to numerous natural and engineered environmental, health, and industrial applications. These include provision of safe drinking water by pathogen filtration in the subsurface 1 2 or in engineered filters 3 4 , control of chronic contamination of processed food supplies 5 , improved medical screening and treatment 6 7 8 , pipeline performance assessment 9 , and improved semiconductor manufacturing 10 to name a few. Despite the wide range of applications reliant upon particle deposition on surfaces, this process remains inadequately described both conceptually and mathematically when deposition surface roughness is present, thereby precluding the development and application of predictive models for particle deposition on surfaces.…”

mentioning

confidence: 99%

Non-linear, non-monotonic effect of nano-scale roughness on particle deposition in absence of an energy barrier: Experiments and modeling

Jin

Glawdel

Ren

et al. 2015

Sci Rep

View full text Add to dashboard Cite

Deposition of colloidal- and nano-scale particles on surfaces is critical to numerous natural and engineered environmental, health, and industrial applications ranging from drinking water treatment to semi-conductor manufacturing. Nano-scale surface roughness-induced hydrodynamic impacts on particle deposition were evaluated in the absence of an energy barrier to deposition in a parallel plate system. A non-linear, non-monotonic relationship between deposition surface roughness and particle deposition flux was observed and a critical roughness size associated with minimum deposition flux or “sag effect” was identified. This effect was more significant for nanoparticles (<1 μm) than for colloids and was numerically simulated using a Convective-Diffusion model and experimentally validated. Inclusion of flow field and hydrodynamic retardation effects explained particle deposition profiles better than when only the Derjaguin-Landau-Verwey-Overbeek (DLVO) force was considered. This work provides 1) a first comprehensive framework for describing the hydrodynamic impacts of nano-scale surface roughness on particle deposition by unifying hydrodynamic forces (using the most current approaches for describing flow field profiles and hydrodynamic retardation effects) with appropriately modified expressions for DLVO interaction energies, and gravity forces in one model and 2) a foundation for further describing the impacts of more complicated scales of deposition surface roughness on particle deposition.

show abstract

mentioning

confidence: 99%

Non-linear, non-monotonic effect of nano-scale roughness on particle deposition in absence of an energy barrier: Experiments and modeling

Jin

Glawdel

Ren

et al. 2015

Sci Rep

View full text Add to dashboard Cite

show abstract

“…(2) will give the constant value of 0.55, just as the conclusions derived by Ref. [1]. So the cluster transportation behavior is believed to be irrelevant with the cluster materials in the thermal plasma¯ash evaporation process.…”

Section: Theoretical Backgroundmentioning

confidence: 81%

Modeling of clusters deposition under the effect of thermophoresis during thermal plasma flash evaporation process

Han

Yoshida²

2001

Science and Technology of Advanced Materials

View full text Add to dashboard Cite

In the thermal plasma¯ash evaporation process, due to the high quenching rate of the impinging plasma jet and the steep temperature gradient in front of the substrate, thermophoretic force is expected to have a signi®cant effect on the cluster transportation process. This paper aims at investigating the cluster transportation and deposition processes by the method of numerical simulation and presents the qualitative description of the effects of thermophoresis on the deposition ef®ciency. Eulerian approach is employed in this research and the corresponding convection±diffusion governing equation is established for the cluster concentration ®eld. The environment pressure for the plasma jet is set to be 50 Torr and the working gas is argon plasma. The velocity and temperature ®elds are simulated ®rstly and then the cluster transportation process is modeled within the established¯ow ®eld but the computational domain is only limited to a narrow region in front of the substrate. It is assumed that clusters of uniform size are generated in a certain region within the boundary layer and it is treated as the source term in the cluster transportation equation. The results of cluster concentration ®eld and the radial distribution of deposition¯ux are achieved for the clusters in the size range 1±10 nm, respectively. Results are also given for the comparative cases without considering thermophoresis effects. It is found that the concentration boundary layer is signi®cantly suppressed by the thermophoretical force, and the effect of thermophoresis plays a dominant role than that of diffusion thus almost uniform deposition ef®ciency is achieved for clusters of different sizes. q

show abstract

“…These relationships were explained in detail by theoretical and experimental studies. 8,19 NaCl particles were used as the PBMS calibration. A differential mobility analyzer was used to produce NaCl particles that range from 60 to 80 nm.…”

Section: Methodsmentioning

confidence: 99%

“…With PBMS, the particle formation in CVD processes has drawn a lot of attention over the past decade. Experimental or theoretical studies were also carried out in the system of the thermal- [5][6][7][8][9] or plasma-enhanced 10,11 chemical vapor deposition ͑PECVD͒ for the deposition of silicon films.…”

mentioning

confidence: 99%

Effects of Process Variables on TiN Particle Formation during Metallorganic Chemical Vapor Deposition

Na¹,

Kim²,

Choi³

et al. 2010

Electrochem. Solid-State Lett.

View full text Add to dashboard Cite

Titanium nitride particles formed in thermal and plasma-enhanced metallorganic chemical vapor deposition processes using tetrakis͑diethylamino͒titanium were investigated. Particle formation was studied by using a particle beam mass spectrometer calibrated by a certificate reference material ͑NaCl͒ as a function of process variables such as bubbler temperature, deposition temperature and pressure, sampling positions in the reactor, plasma conditions, and a flow rate of carrier gas. From these results, the effects of the sampling position and the thermal and plasma energies were studied. These measured results were in reasonable agreement with the transmission electron microscopy results.In recent years, the problems with particle contamination have been greatly emphasized in microelectronic processing because the particles have significant effects on device yield, performance, and reliability. 1 These particles have been the major causes of device failure because a 1% decrease in the product yield can cause more than a 2% loss in profits. 2 Generally, nanosized particles are typically produced by homogeneous nucleation and growth during fabrication processes. Therefore, to control particle contamination, it is essential to understand particle nucleation and growth in the chemically reacting environments. Chemical vapor deposition ͑CVD͒ is one of the most important processes. However, during the CVD process, the nanosized particles are generated, and these particles can have a fatal influence on the device quality. Particle measurement in the CVD process remains highly challenging because of the low pressure ambient and the particle size in nanometers. In terms of particle measurement methods, the levels of particle contamination have been conventionally monitored with either a wafer surface scanner or an in situ particle monitor ͑ISPM͒ sensor. However, the conventional wafer scanner, which utilizes an ex situ monitoring method, has difficulty detecting particles smaller than 50 nm. 3 In ISPM, the particles less than 0.1 m could not be detected because light scattering varies inversely with the sixth power of a particle diameter in this size range. 4 Hence, the particle beam mass spectrometer ͑PBMS͒ was developed in the mid-1990s to overcome these problems. Various semiconductor processes have been proven to be effective in measuring fine particles ͑0.005-0.5 m͒. With PBMS, the particle formation in CVD processes has drawn a lot of attention over the past decade. Experimental or theoretical studies were also carried out in the system of the thermal-5-9 or plasma-enhanced 10,11 chemical vapor deposition ͑PECVD͒ for the deposition of silicon films.A CVD titanium nitride ͑TiN͒ film deposited with a metallorganic precursor is widely used as a conformal barrier metal that protects shallow junctions during the formation of Al contacts on silicon as a glue layer because it is rigid and chemically stable. 12,13 For the CVD TiN process, tetrakis͑diethylamino͒titanium ͑TDMAT͒ was used by many researchers. However, because of...

show abstract

Particle transport in a parallel-plate semiconductor reactor: Chamber modification and design criterion for enhanced process cleanliness

Cited by 14 publications

References 29 publications

Non-linear, non-monotonic effect of nano-scale roughness on particle deposition in absence of an energy barrier: Experiments and modeling

Non-linear, non-monotonic effect of nano-scale roughness on particle deposition in absence of an energy barrier: Experiments and modeling

Modeling of clusters deposition under the effect of thermophoresis during thermal plasma flash evaporation process

Effects of Process Variables on TiN Particle Formation during Metallorganic Chemical Vapor Deposition

Contact Info

Product

Resources

About