Nucleation Of Cu Research Articles

Additive Manufacturing of High Strength Copper Alloy With Heterogeneous Grain Structure Through Laser Powder Bed Fusion

The low laser processability of pure copper (Cu) commonly causes high porosity, and therefore low strength of additively manufactured (AMed) Cu parts. The current work reports an effective approach to solve this problem through inoculation of pure Cu powder with cobalt (Co) submicron particles. Co addition enabled increase in laser absorptivity, and therefore dense Cu was fabricated through laser powder bed fusion even though typical AM featured coarse columnar grains were produced when Co addition was 4 wt.% or less. Significant grain refinement with heterogeneous grain structure (HGS) was achieved when the addition level was greater than the maximum solid solubility of Co in Cu (4.75 wt.%), such as 6 and 8 wt.%. HGS is characterized with ultrafine equiaxed grains close to melt pool boundaries and coarser equiaxed grains or columnar grains near centres of melt pools. Heterogenous nucleation of Cu near melt pool boundaries, where were associated with larger thermal undercooling due to faster cooling rate, was responsible for the formation of ultrafine equiaxed grains. In contrast, heterogeneous nucleation near centres of melt pools was suppressed. Detailed substructure examination indicated that heterogenous nucleation occurred on dual phase nanoparticles containing Co shell and cobalt oxide (CoO) core. The higher affinity of Co to oxygen caused formation of CoO. HGS of Cu significantly improved both strength and ductility. After direct ageing at 600°C for one hour, a tensile strength of 491.1±12.6 MPa, elongation of 27.3±2.5% and electrical conductivity of 61.5% IACS (International Annealed Copper Standard) were achieved.

The Initial Precipitation Behavior of Copper in Ferritic Stainless Steel

The precipitation of a suitable amount of Cu phase is the key to the antibacterial performance of Cu-containing antibacterial stainless steel. In order to study the precipitation behavior of Cu in ferritic stainless steel, the nucleation rate–temperature curve and phase transition kinetics curve of the Cu-rich phase were calculated theoretically. The distribution, size, and composition of Cu precipitates were characterized, and the distributions of Ni and Mn in the Cu precipitates at different annealing times were also analyzed by atom probe tomography and transmission electron microscopy. The crystal structure of the Cu precipitates was observed by high-resolution electron microscopy. With an increase in annealing temperature, the nucleation rate decreases and the time for nucleation of Cu precipitates increases monotonously. Accordingly, the size of the Cu-rich phase increases, and the number density decreases with increasing annealing time. Moreover, the contents of Ni and Mn gradually increase, and the Ni and Mn atoms aggregate on the surface of the Cu-rich phase. When the steel is annealed for 60 s, the Cu precipitate in the ferrite matrix exhibits a 9R structure with multiple twins.

Improved Copper Damascene Wires Using Direct Plate on Cobalt Process

We preset an electrochemical damascene fill process for copper which is capable of direct nucleation and fill on a cobalt substrate. This process preserves the benefits of a cobalt liner while simultaneously reducing problems inherent in the use of a PVD copper seed on small interconnect features. The process depends upon the use of an additive which enhances the action of traditional damascene suppressors. This enables improved nucleation on the Co substrate without sacrificing the ability of the bath to produce the bottom-up fill typical of damascene Cu plating. We demonstrate good fill across a wafer on a thin, resistive cobalt substrate, producing lines with better electrical properties than a control created using PVD seed followed by a traditional damascene process. As integrated circuit features get smaller, the use of electrochemical copper metallization for interconnects has become more challenging. A major reason for this is the requirement for a PVD copper seed layer on top of a cobalt liner to enable traditional damascene plating. In very small features, the seed layer takes up a substantial portion of the feature volume. This produces a serious risk of voids as either the PVD itself or subsequent electrodeposition can close off the feature top before it has been filled with metal. One way to extend the manufacturability of copper interconnects is to eliminate the PVD copper deposition step entirely and instead plate directly on the cobalt liner. Direct plating of copper onto cobalt is a challenging problem. Cobalt dissolves readily in acidic solutions used for damascene Cu plating. Further, the reduction potential of Cu(II) is such that it will galvanically corrode Co substrates exposed to it. Copper has been successfully plated on cobalt, but previous attempts use heavily complexed alkaline baths to preserve the cobalt substrate. Unfortunately, both high pH and heavily complexed copper inhibit the ability of traditional damascene additives to produce good bottom up fill. Our approach to this problem has been to use an additive which improves the nucleation of Cu on Co not by complexing the Cu in the bulk solution, but by improving the ability of damascene suppressor molecules to quickly and effectively inhibit Cu plating on newly created Cu surfaces. This fast inhibition makes it more favorable for Cu nuclei to expand across the Co surface rather than growing in a more perpendicular direction. The faster passivation of the surface with Cu in turn dramatically reduces Co corrosion loss to the bath. Because this behavior is produced using traditional damascene additives, bottom up fill can be established after nucleation occurs if plating conditions are properly chosen. Good nucleation and fill on a 36nm pitch structure can be seen in the benchtop scale fill progression tests shown in Figure 1. Nucleation is slightly rougher than is observed when plating on a Cu substrate, but fill ability is sufficient to compensate. A Sabre electrodeposition tool from Lam Research was used to scale this process up to 300mm wafer plating. In Figure 2 electrical data statistics from these wafers are compared with similar results form a pair of control wafers plated using traditional damascene process on a PVD Cu substrate. A portion of this work was performed by the Research Alliance Teams at various IBM Research and Development facilities. Figure 1

Effects of solute elements on microstructural evolution in Fe-based alloys during neutron irradiation following thermal ageing

Neutron irradiation following thermal ageing experiments were performed using Fe-Cu, Fe-Cu-Ni, Fe-Cu-Ni-Mn, and Fe-Cu-Si alloys to investigate the effects of Ni, Mn, and Si on the formation of Cu-rich clusters and their chemical composition. Atom probe tomography showed that Ni and Mn increased the number density of Cu-rich clusters. This can be caused by that Ni and Mn reduce the formation energy of Cu nuclei, thus enhancing the Cu nucleation rate. Mn may also enhance the heterogeneous nucleation of Cu because of its strong interaction with self-interstitial atoms. In addition, Mn can accelerate solute diffusion during thermal ageing. A high Ni alloying content enhanced the Mn concentration in the clusters as well as the Ni concentration, which is believed to be caused by the thermodynamic relationship between Ni and Mn. In Fe-Cu-Si alloys, the Si concentration in the clusters was very low and was comparable to the Si alloying content. This indicates that Si atoms hardly cluster with Cu.

Nucleation of Cu2Te layer by a closed space sublimation method toward the growth of Te based chalcopyrite

Nucleation of Cu-Te layers was performed by the closed space sublimation method using various source materials, source temperatures, and Si substrates with different surface orientations. The objective was to produce nuclei layers with high quality for use as nucleation centers for CuGaTe2. The grown samples were evaluated by x-ray diffraction and scanning electron microscopy. Cu2Te deposits were obtained using either a CuTe or Cu2Te source, but the latter gave a higher area coverage of Cu2Te nuclei. Highly oriented nuclei were obtained when they were grown on Si (001) at a source temperature of 640 °C and substrate temperature of 590 °C. When the source temperature was raised to 750 °C and the corresponding substrate temperature was 700 °C, nonuniform but highly oriented nuclei were obtained. Both nuclei layers exhibited a strong preference for (0001) orientation. The crystallographic features of the Cu2Te nuclei formed on Si (111) were similar to those of the Cu2Te nuclei formed on Si (001).

Laser-induced construction of multi-branched CuS nanodendrites with excellent surface-enhanced Raman scattering spectroscopy in repeated applications.

We report on the successful fabrication of multi-branched CuS nanodendrites with average branch length of about 20 nm by laser ablation of bulk Cu target in thioacetamide (TAA) solution. During the nucleation of Cu and S species, the accurate anisotropic growth should be attributed to an ultra-rapid acid etching process by laser-induced TAA hydrolyzing reaction. Interestingly, the semiconductor CuS nanodendrites provide pronounced surface enhanced Raman scattering (SERS) properties with noble-metal comparable activity and a detection limit as low as ~10-10 M, approaching the requirement (~nM) for single molecule detection. More importantly, after SERS analysis, the crystal violet (CV) probe molecules can be effectively removed from the substrate by 1064nm laser irradiation-induced moderate thermal treatment. Therefore, the unique and distinctive advantage is that the as-prepared CuS nanodendrites exhibit excellent reusability for 60 cycles of repeated SERS analyses. The low-cost CuS semiconductor nanodendrites with enhanced SERS properties should be established as a prominent SERS-based ultrasensitive probe in the repeated applications.

Effect of Ni and vacancy concentration on initial formation of Cu precipitate in Fe–Cu–Ni ternary alloys by molecular dynamics simulation**Project supported by the National Natural Science Foundation of China (Grant No. 51301102) and the Science and Technology Commission of Shanghai Municipality, China (Grant No. 15ZR1416000).

In the present work, the effects of Ni atoms and vacancy concentrations (0.1%, 0.5%, 1.0%) on the formation process of Cu solute clusters are investigated for Fe–1.24%Cu–0.62%Ni alloys by molecular dynamics (MD) simulations. The presence of Ni is beneficial to the nucleation of Cu precipitates and has little effect on coarsening rate in the later stage of aging. This result is caused by reducing the diffusion coefficient of Cu clusters and the dynamic migration of Ni atoms. Additionally, there are little effects of Ni on Cu precipitates as the vacancy concentration reaches up to 1.0%, thereby explaining the embrittlement for reactor pressure vessel (RPV) steel. As a result, the findings can hopefully provide the important information about the essential mechanism of Cu cluster formation and a better understanding of ageing phenomenon of RPV steel. Furthermore, these original results are analyzed with a simple model of Cu diffusion, which suggests that the same behavior could be observed in Cu-containing alloys.

Influence of additives on Cu thin films electrodeposited directly on Ti diffusion barrier in Cl−-free electrolytes for Cu interconnect

For Cu interconnect in Si-based microelectronic devices, seedless electrodeposition of Cu thin films directly on Ti diffusion barrier was investigated in Cl−-free, citrate (Cit)-based acidic electrolytes with polyethylene glycol (PEG) or Janus Green B (JGB) additive. Both PEG and JGB suppressed Cu electrodeposition on the Ti surface effectively, and even their inhibition against Cu-Cit complex reduction was synergetic. Accordingly they decreased the surface roughness of Cu thin film on the Ti surface considerably. Thus, seedless growth of Cu thin films on Ti was achieved from the synergistic leveling effect of both PEG and JGB in the absence of Cl− because PEG adsorption onto the Ti surface did not accompany PEG complexation with Cl− and Cu+, and CuCl contamination did not occur. 3D nucleation of Cu on the Ti surface became more progressive in more additive-concentrated electrolytes. Good interfacial adhesion between Cu thin film and Ti diffusion barrier was maintained, whereas the sheet resistance of Cu thin films was fairly increased, in spite of adding PEG and JGB.

Direct Cu Electrodeposition on the Ni Alloy Barrier Layer Prepared By Electroless Deposition on SiO2

The formation of barrier layer by electroless deposition (ELD) takes great interest for the next generation ULSI metallization, because ELD contains various advantages including low process cost, low process temperature, and the excellent conformality[1,2]. Especially, the research on the Ni or Co based alloy barrier layer is widely conducted as the barrier layer property of the film was superior[1-3].Some researches realized the all-wet metallization by using ELD barrier layer, however, the formation of the conformal Cu seed layer is still a big issue before the electrodeposition (ED) of Cu. The Cu seed layer formation on the ELD barrier layer usually conducted by ELD of Cu and/or the displacement reaction between Cu and barrier metal. In the point of the productivity of the device, however, the elimination of the Cu seed layer formation step would be preferred. Hence, in this study, we investigated the direct ED of Cu without Cu seed layer on the electroless NiWP barrier layer.NiWP barrier layer was formed on a Pd-activated SiO2 substrate by ELD in an acidic electrolyte. The SiO2substrate was activated by the Pd nanocolloids immobilized via the self-assembled monolayer on the surface. Prior to the direct ED of Cu, the native oxide of the NiWP barrier layer was removed by the coulometric reduction method.The linear sweep voltammetry (LSV) for the Cu reduction was investigated on the electroless NiWP barrier layer after the native oxide removal. As can be seen in Figure 1, the charge transfer limiting region of Cu deposition was separated into two stages: the Cu nucleation stage and the Cu film growth stage. However, because the resistivity of the electroless NiWP barrier layer was high, it was difficult to form a smooth and uniform Cu film by means of the one-step potentiostatic method. The low nucleation density of Cu or the non-uniform Cu deposition was problematic at low or high overpotential case, respectively (Figure 2a and 2b)[4]. On the other hand, the two-step potentiostatic method could successfully deposit the smooth and uniform Cu film, because the high overpotential induced higher nucleation of Cu and the low overpotential resulted in the smooth and uniform film growth (Figure 2c).The barrier layer property was studied using the ED Cu / ELD NiWP / SiO2 stacked specimen. The adhesion between each layer (ELD NiWP / SiO2and ED Cu / ELD NiWP) was excellent, which was evaluated by the tape test method. The electroless NiWP barrier layer could endure the 500°C and 30 min of thermal treatment. The minimal change in sheet resistance and the absence of silicide-related peaks in XRD analysis attested the thermal stability of the barrier layer. In addition, the patterned specimen covered with the electroless NiWP barrier layer was successfully filled by the direct ED of Cu with the conventional additive combination for the gap-filling. In this symposium, the direct Cu ED process on the electroless NiWP barrier layer will be addressed in detail. In addition, we will also discuss the trench filling by the proposed method.

Gravitational Effects on the Initial Stage of Cu Electrodeposition

The electrochemical processing has been favorably reviewed in the microelectronics industry. It is essential to control the nucleation and crystal growth in the initial stages of electrodeposition process. For this purpose many effects of additive have been examined. However, few quantitative examinations has been reported on the coupling phenomena of ionic mass transfer rate and microstructural/morphological variations through nucleation and crystal growth.This work discusses various gravitational levels as new parameters into the nonequilibrium electrochemical processing in order to create or tailor the unique physical properties in micrometer space field. It is fruitful in terms of not only physical chemistry but also industrial applications to comprehensively understand the coupling phenomena. In fact, it has been confirmed that the gravitational levels as well as the magnetic field introduce the intensive effects to the coupling behavior different from that under 1 G normal environment.Here is a report to analyze the effect of gravitational level on the initial stages of Cu electrodeposition on TaN or TiN substrate installed in a cavity-type electrolytic cell. In case of Cu on TaN substrate, cathode, anode and reference electrode are TaN substrate, copper sheet and copper wire, respectively. Electrolyte solution is 0.05 M CuSO4 – 0.05 M H2SO4. Two kinds of electrolytic cell configuration are used; one is a cathode facing downward over anode horizontally (C/A), and another is an anode over a cathode (A/C). A centrifugal field is applied to introduce a high gravitational field.Time dependence of nucleus number density shows that the nucleation of Cu on TaN is “progressive” nucleation. Under 1 G, saturated nucleus number density (N0) increases and mean diameter (dmean) becomes smaller as the electrode potential decreases. At any potential, N0 in A/C configuration is higher and dmean is smaller than C/A. The current transient shows the difference caused by two cell configurations and macroscopic natural convection due to the concentration difference should be inevitably induced with time progressed in a cavity in the A/C configuration. The image analysis indicates the nucleus distribution is considerably deviate from the Poisson distribution as long as the nearest-neighbor distance concerns. The existence of “exclusion zone” around the growing nuclei in nucleation phenomena has been found. Under high G level field, N0 becomes slightly increased and is saturated as G level increases. The effect of G level on the current transient is quite large. It is almost saturated under G level over 100 G. It is moreover found that the nucleation is saturated under high G level earlier than under low G level. The effect of ionic mass transfer rate due to the concentration difference becomes significant to enhance the nucleation rate knN0.

Formation of Cu domains on [formula omitted] revealed by stereographic maps of elastically backscattered electrons

The results of investigations of the Cu/Ru101¯0 adsorption system obtained by using directional elastic peak electron spectroscopy (DEPES), Auger electron spectroscopy and low energy electron diffraction are presented. The measurements were performed for the clean and Cu covered ruthenium surface at a coverage of 7 bilayers at room temperature. At early stages of growth the formation of two flat Cu bilayers was observed. Further adsorption leads to the nucleation of Cu(111) islands of the face centered cubic (fcc) structure. However, the experimental results reveal the formation of two fcc(111) domains which are mutually rotated by 180° arranged in the 21¯1¯111Cu0001101¯0Ru and 2¯11111Cu0001101¯0Ru orientations with respect to the substrate. Moreover, the quantitative analysis of DEPES data reveals nonequivalent (45%/33%) concentrations of these domains.

KINETICS OF GRAIN NUCLEATION IN SUPERCOOLED QUASI-TWO-DIMENSIONAL Cu

Grain nucleation kinetics of nanocrystalline Cu in a quasi-two-dimensional system was investigated using molecular dynamics. Nanoscale grains nucleate rapidly from supercooled Cu under isothermal annealing conditions. An Avrami exponent from the JMAK equation implies that the kinetics follows the continuous nucleation model. Observation of grain growth during the nucleation stage shows that the kinetics follows the classical grain growth model with a low growth exponent. The lower value of the exponent illustrates the rapid growth of grains during this stage. Using the growth exponent, the activation energy for nucleation in Cu was obtained and compared with the values obtained in earlier studies.

High Density Copper Nucleation on Ruthenium and Its Application to Direct Plating of Advanced Interconnects

Electrochemical nucleation of Cu on Ru in commercial chemistries was studied. The results show that Cu nucleation behavior is highly dependent on the suppressor species and concentrations, the acidity of the electrolyte, and the state of Ru surface. With an excellent suppressor and intermediate acid electrolyte, the Cu nucleation density as high as 1 ×1013 cm-2 can be achieved without the need of a pretreatment of the Ru surface. Direct plating of Cu on Ru with both damascene and through-silicon via structures using this highly suppressive chemistry is demonstrated.

Synthesis of Noble Metal-Doped Cu Nanoparticles by Ultrasonication

Cu nanoparticles doped with small amounts of noble metals were synthesized by ultrasonication. The noble metals selected were Pd, Pt, Ru, Rh and Ir. Mean particle size was most reduced for Cu nanoparticles doped with Pd, with Cu–Pd nanoparticles of approximately 50 nm in diameter being obtained. X-ray absorption, near-edge spectroscopy and extended X-ray absorption were used to conduct fine structure analysis, and these techniques confirmed that Cu–Pd nanoparticles possessed Cu–Pd bonds. High-angle annular dark-field scanning transmission electron microscopy and energy dispersive X-ray spectrometer mapping were used to show that Cu–Pd nanoparticles were dispersed as Pd clusters inside Cu nanoparticles. It was supposed that the surfaces of Cu–Pd nanoparticles were Cu-rich, and Pd clusters were dispersed inside these particles. It is proposed that heterogeneous nucleation of Cu onto Pd nuclei resulted when nucleation of Pd occurred.

Hybrid copper complex-derived conductive patterns printed on polyimide substrates

We synthesized new copper complexes that can be readily converted into highly conductive Cu film. Mechanochemical milling of copper (I) oxide suspended in formic acid resulted in the submicron-sized Cu formate together Cu nanoparticles. The submicrometer-sized Cu formates are reactive toward inter-particle sintering and metallic Cu seeds present in the Cu complexes assist their decomposition and the nucleation of Cu. The hybrid copper complex film printed on polyimide substrate is decomposed into dense and uniform Cu layer after annealing at 250 °C for 30 min under nitrogen atmosphere. The resulting Cu film exhibited a low resistivity of 8.2 μΩ·cm and good adhesion characteristics.

Copper Electrodeposition on Ru-N Barrier with Various Nitrogen Content for 22 nm Semiconductor Manufacturing Application

The diffusion barrier substrates of Ru sputtered in N2 and Ar atmospheres have been studied. The Ru-N film improves the properties of copper barriers and has excellent thermal stability. Copper electroplating on Ru-N barrier films was investigated with different nitrogen flow rates. The electrochemical deposition of the copper thin film onto the electrode using Ru and RuN was investigated. Scanning electron microscopic micrographs of copper electroplating on Ru and RuN barrier substrates were examined, and the copper nucleation behaviors changed with the nitrogen flow rate. It was found that a continuous copper thin film could be uniformly deposited both on Ru and RuN, with good control of thickness. The results support a predominantly progressive nucleation of Cu on the RuN surface. In addition, X-ray diffraction patterns indicated a principal (111) texture of Cu on Ru and RuN by electrodeposition, without any new phase or compound formation between the two metals. The self-annealing behavior of the Cu films on the RuN substrates that have different N contents were also invesgated in this article. Since the EIS is widely recognized as a powerful tool for the investigation of electrochemical behaviors, the tool was also used to verify the phenomena during plating.

Catalyst-Free Growth of Quasi-Aligned Nanorods of Single Crystal Cu3Mo2O9 and Their Catalytic Properties

We propose a simple and catalyst-free method to grow quasi-aligned single crystalline Cu(3)Mo(2)O(9) nanorods in terms of a mechanism differing from the conventional vapor-solid (VS) and vapor-liquid-solid (VLS) ones for chemical vapor deposition (CVD) methods by directly heating Cu foils in a mixed atmosphere of MoO(3) vapor and air. High quality Cu(3)Mo(2)O(9) nanorods can be simply grown in a temperature range from 450 to 550 degrees C whose diameter, length, and distribution density are dependent on both heating temperature and time. Interestingly, the growth rate at 550 degrees C drops significantly after 6 h. All nanorods grow along the [010] direction. On the basis of a proposed growth model, the nucleation of Cu(3)Mo(2)O(9) nanorods is believed to be governed by formation of initial polycrystalline Cu(x)O protuberances with nanoscale diameters on Cu foils which may act as growth "templates". This novel method can be applied to grow other similar tertiary transition metal oxide nanostructures on substrates with large sizes. Most importantly, these Cu(3)Mo(2)O(9) nanorods decrease the ignition temperature of Printex U model soot from 600 to 438 degrees C, being in between 200 and 450 degrees C of the exhaust of diesel-powered combustion engines, which are therefore expected to be a potential efficient and environmentally friendly catalyst for diesel exhaust combustion.

Orientation and rate dependence of dislocation nucleation stress computed using molecular dynamics

Molecular dynamics simulations are used to investigate orientation and rate dependence of partial dislocation nucleation in Cu. As the strain rate is reduced from 109 to 107 s−1, the tensile stress required for homogeneous dislocation nucleation is reduced by at most 5%. Furthermore, mild orientation sensitivity is observed in the rate dependence of the critical tensile stress. The computed resolved shear stress for partial dislocation nucleation is consistent with previous ab initio calculations of the theoretical shear strength of Cu.

Phase transition and dislocation nucleation in Cu–Nb layered composites during physical vapor deposition

Using classical molecular dynamics simulations, we have investigated the growth of {111} Cu on Nb {110} surface. Our results reveal that the deposited Cu layer initially grows as body-centered cubic (bcc) and Vernier misfits are observed in the interface of bcc Cu and bcc Nb. As it continues to grow, the bcc Cu {110} transforms into face-centered cubic (fcc) Cu {111}. The phase transition starts after the bcc Cu layer has accumulated about 3 monolayers and is finished depending on deposition parameters. Nuclei of fcc Cu {111} form in the top surface of Cu and grow in plane and toward the interface. Partial dislocations in the fcc Cu layer nucleate during the late stage of the transition, and the stacking faults grow as the Cu layer thickens.

The influence of magnetic field on phase nucleation in Cu–Be alloy

We discuss the results of small-angle polarized neutron scattering onthe diamagnetic Cu–Be alloy on annealing in the temperature range300 K<T<650 K from the quenched condition. The characteristic scale of phase nucleationdepending on temperature and time was determined. The correlation inchange of the integral cross-section of scattering neutrons on annealingwith modulation (‘on’ or ‘off’) of a constant magnetic field in the range0 Oe<H<5700 Oe was found. These experimental results may be interpreted as the magneticfield influence on the kinetics of nucleation during artificial aging of Cu–Be alloy.