Materials with a low dielectric constant are required as interlayer dielectrics for the on-chip interconnection of ultra-large-scale integration devices to provide high speed, low dynamic power dissipation and low cross-talk noise. The selection of chemical compounds with low polarizability and the introduction of porosity result in a reduced dielectric constant. Integration of such materials into microelectronic circuits, however, poses a number of challenges, as the materials must meet strict requirements in terms of properties and reliability. These issues are the subject of the present paper.
Modern ultra-large-scale integration (ULSI) devices contain 108–109 transistors in an area smaller than 1 cm2 and operate at a clock frequency approaching several gigahertz. As device dimensions shrink, the switching speed of its basic element increases as a consequence of the carrier's transit time across the length of a transistor channel decreasing (figure 1). Of course, these basic elements must be interconnected to make the ULSI device functional. As the functional complexity of devices increase, the number of interconnection levels and metal length continue to increase to the extent that an advanced ULSI device may consist of 8–10 levels of metal lines. For this reason, the effective speed of the device is becoming ever more dominated by the signal propagation through the horizontal and vertical metal interconnects of components with various functions. It is here that the resistance (R) and capacitance (C) characteristics of the interconnect materials become important. In fact, the rapid increase in RC delay time is one of the main bottlenecks in deep submicron devices (figure 1). The RC delay is given by(1.1)where ρ is the metal resistivity, ϵ0 is the vacuum permittivity, k is the relative dielectric constant of interlayer dielectric (ILD), P is the metal line pitch (sum of line width and line spacing), T is the metal thickness and L is the metal line length. This equation demonstrates that RC delay can be reduced using metals with low resistivity and dielectric materials with a low dielectric constant.
The introduction of copper and low dielectric constant (low-k) materials has improved the situation as compared with the conventional Al/SiO2 technology by reducing both the resistivity of and capacitance between wires. Copper is becoming the common metallization material. Further lowering of the signal delay by introducing low-k dielectrics is one of the main challenges today (Maex et al. 2003).
A description of all the problems related to the introduction of copper and low-k dielectrics is beyond the scope of this paper and hence we will limit our discussion to approaches to decreasing the dielectric constant of ILDs.
2. Low-k dielectrics: introductory remarks
Classically, the dielectric constant of materials is described by the Clausius–Mossotti equation,(2.1)where , ϵ and ϵ0 are the dielectric constants of the material and vacuum, N is the number of molecules per unit volume (density) and α is the total polarizability, including electronic (αe), distortion (αd) and orientation (αo) polarizabilities. According to equation (2.1), the dielectric constant of materials can be reduced by decreasing the total polarizability and density. Early generations of low-k dielectrics were obtained by doping the traditional SiO2 with fluorine and carbon during the chemical vapour deposition (CVD) of the materials. Fluorine substitution lowers the k value by decreasing the polarizability and increasing the free volume. These kinds of dielectrics typically have a k-value in the range of 3–3.5.
Other types of low-k dielectrics are based on organic polymers. Saturated hydrocarbons have a lower polarizability than unsaturated, conjugated and aromatic hydrocarbons. Therefore, they potentially provide the lowest k-value without requiring the introduction of porosity. However, aliphatic C–C, C–H and C–N bonds generally become unstable at temperatures greater than or equal to 300–400 °C and, in some cases, at even lower temperatures. Only materials composed of non-aliphatic C–C, C–O, C–N and C–S bonds, aromatic structures and cross-linked or ladder structures can withstand the temperatures necessary for interconnect technology (450–500 °C). Most of the organic low-k films with sufficient thermal stability have dielectric constants close to or in the range of 2.6–3.0.
Decreasing the density produces a further reduction in the dielectric constant. Therefore, ultra low-k dielectrics with k less than 2.6 must be porous. The relative dielectric constant of porous materials, kr, depends on the porosity and dielectric constant of the film skeleton (ks):(2.2)where k1 is the dielectric constant of the material inside the pores. If the pores are empty, the first term in equation (2.2) is equal to 0. Materials with relatively small ks values provide smaller kr values at lower porosity.
3. Deposition of low-k dielectric films
(a) Matrix materials
Different types of matrix materials are used for preparing porous low-k films. The most important ones at the present time are inorganic silica-based materials (silica xerogel, aerogels), silsesquioxanes (SSQ) and organic polymers. The advantage of silica- and SSQ-based materials is that their chemical properties are similar to traditional SiO2, making it possible to use traditional technology during their integration. Organic low-k polymers have minimal frequency dispersion and their relatively low ks value provides low kr at a low porosity. However, they present more problems in integration.
Silica has a tetrahedral elementary unit (figure 2a). To reduce the kr value and make the material hydrophobic, some oxygen atoms are replaced by F or CHx groups. The addition of CHx groups introduces less polar bonds and also creates additional free volume. Such silicon oxycarbides (SiOCH) are normally deposited by CVD and they are constitutively porous. The carbon concentration in most SiOCH materials varies between 10 and 30%. The carbon concentration must be large enough to provide hydrophobic properties while still maintaining sufficient mechanical properties.
In the SSQ materials, Si and O atoms are arranged in the form of a cage or ladder (figure 2b). The cage structure creates free volume, decreasing the material's density and, therefore, its k value. The cage can include eight (T8), twelve (T12) or even more silicon atoms. The cages in polymerized SSQ are connected to each other through oxygen or CH2- groups. Other cage corners are terminated by hydrogen (HSQ), methyl (MSQ), phenyl (PSQ) or other aliphatic groups. MSQ matrix materials have a lower dielectric constant than HSQ (ca 2.8 and ca 3.0–3.2, respectively) because of the larger size of the CH3 group and the lower polarizability of the Si–CH3 bond compared with the Si–H bond. SSQ cages are metastable and tend to break down to the silica tetrahedral, especially during curing at elevated temperatures. As a result, SSQ-based low-k materials realistically represent a mixture of different SSQ cages and silica tetrahedra. Due to the low temperature stability of the cage, SSQ-based materials are prepared by spin-on-glass (SOG) technology while SiCOH materials are deposited by CVD.
The skeleton dielectric constant of both CVD SiCOH and SOG SSQ materials are defined by the polarizability of Si–O bonds and the free volume, which depends on CHx group concentration. However, the deposition of ultra low-k materials also requires the introduction of artificial porosity in both SOG and CVD technologies.
(b) Deposition of porous materials by spin-on technology
In the spin-on deposition, the film coating is performed by dispensing a liquid precursor at the centre of the substrate, which is placed on a spinner. Rotation of the substrate creates centrifugal forces that ensure distribution of material on the surface. The thickness and uniformity of the coating is a result of the balance between centrifugal forces and viscous forces, determined by the viscosity of the solution. The spinning step is followed by heating or a ‘bake’ to remove solvent at temperatures typically below 250 °C. The latter step can also initiate cross-linking of the film. Finally, a sintering at temperatures varying from 350 to 600 °C (‘cure’) is required to obtain a stable film. This ‘cure’ step induces the final cross-linking of the polymer chains and results in a mechanically stable film structure.
Numerous methods of introducing subtractive porosity into spin-on deposited materials exist, but the most important ones can be divided into two categories. The first category groups materials where porosity is introduced exclusively through sol–gel processes, while the second group includes materials where porosity is formed through the use of sacrificial particles (porogens) that are desorbed during film cure.
(i) Subtractive porosity by sol–gel based techniques
There exist two main approaches for the formation of subtractive porosity based on sol–gel techniques: the first takes advantage of ageing processes and the second relies on a hierarchical organization of the primary particles in the sol (self-assembly).
The formation of a rigid skeleton before the liquid is extracted from a wet gel is a key point in the formation of high porosity materials. Even if the gel-point is reached after material spinning, a long time is still required before the hydrolysis and condensation reactions are complete. For this purpose, an additional step (ageing) before drying the wet gel is introduced. The aim of this step is to accelerate the sol–gel reactions, typically by relying on the pH and the water content in the ambient. Once the network structure is strengthened, the solvent can be extracted without the network backbone collapsing. The level of residual porosity is generally tuned through the ratio of solvent to solid content in the sol. For subtractive porous films based on ageing techniques, there is a clear trend that increasing total porosity leads to larger pore sizes (Maex et al. 2003). The control of pore size distribution is vital for processing porous materials.
In sol–gel science, numerous studies have been performed in the synthesis of self-assembled microporous materials. Hierarchical ordering of the aggregates by preferential solvent evaporation during spin coating is reported for a solution of surfactants, a swelling agent and soluble silica. Ordered materials with dielectric constants as low as 1.3 have been synthesized with this method. In this case, the final film porosity and pore structure is related to the way in which the primary particles are assembled and ordered.
(ii) Subtractive porosity by macromolecular porogens
This technique is based on the addition of molecular or supramolecular particles to the dielectric precursor with the purpose of tailoring the thermal stability. The stability of these particles is such that they are not affected by the drying step and they are removed by pyrolysis during final film sintering or cure (typically in the range of 300–400 °C). Their volume distribution in the film at the moment of desorption represents the template for the residual pores in the layer. In the ideal case, the film's porosity is directly related to the amount of porogen as a function of the total solid part in the precursor solution and the size of the sacrificial particles is directly related to the final pore size (Baklanov et al. 2001). There are some requirements in order to maintain the relationship between sacrificial particles and pores and porosity. Firstly, the sacrificial material should be chemically compatible with the matrix material in order to avoid phase separation. Secondly, the sacrificial particles should be uniformly distributed throughout the film volume in order to avoid the coalescence of pores.
There are two ways in which the sacrificial porogens are brought into the precursor solution. One method is the dispersion of porogens in the solution. The second involves chemically linking sacrificial particles (grafting) to the network polymers. This second method allows inherent control of the volume distribution of porogens in the dielectric film.
One advantage of the nanoparticle template approach is that the film has a higher degree of cross-linking when the pores are created. The porous structure is, therefore, less affected by further densification, in comparison with sol–gel based pore formation.
(c) Chemical vapour deposition (CVD)
The semiconductor industry has long relied on insulating films of SiO2 deposited from the gas phase with silane (SiH4) oxidation. Therefore, most of the attempts at producing low-k materials with different versions of CVD have been with doped versions of SiO2. The main dopants used so far are fluorine and carbon in the form of alkyl groups. They are introduced by replacing standard silane with fluoro- and alkylsilanes like Si2H2F2 and (CH3)xSiHy with (x+y)=4. Doping a film with alkyl groups terminates some of the silicon bonds within the oxide lattice and lowers the electronic polarizability of the film. The relatively large molecular volume of the alkyl groups decreases the film density. Moreover, the removal of some Si–F and Si–alkyl groups during the deposition or post-deposition annealing of the film due to their lower thermal stability allows for the generation of additional porosity. Generally, deposited SiOCH films have a porosity of about 5–15% with a pore size of about 1–2 nm.
Various approaches have been employed to produce CVD SiCOH films with a subtractive porosity with a k value below 2.6. One method utilizes a multiphase deposition. The SiOCH precursor tetramethylcyclotetrasiloxane (TMCTS) is mixed with a thermally unstable CHx phase during deposition. This unstable phase is thermally decomposed and removed from the film during the subsequent anneal (4 h at 400 °C), leaving behind pores. The resulting porosity and dielectric constant depend on the CHx/TMCTS ratio and can be as high as 30–40% for a one-half ratio (Grill 2003; Grill et al. 2003).
4. Pore structure of low-k dielectric films
The most important properties of porous low-k dielectrics are related to porosity and pore structure. There are many established methods for determining the pore size of porous materials but traditional ‘porosimetries’ are limited in their application to thin films because of the small total pore volume. Recently, several methods have been successfully applied to the pore size determination of thin porous films. These are intrusive methods such as ellipsometric and X-ray porosimetry (Baklanov et al. 2000; Lee et al. 2002) and neutron scattering contrast matching (Hedden et al. 2004). These methods include solvent adsorption and condensation in the pores. Non-intrusive methods include small-angle neutron and X-ray scattering spectroscopy (Wu et al. 2000; Huang et al. 2002) combined with specular X-ray reflectivity (SXR) and positron annihilation lifetime spectroscopy (PALS; Gidley et al. 2000).
These techniques are based on different physico-chemical principles but systematic studies show good agreement in pore size and porosity (Kondoh et al. 2001; Grill et al. 2003). However, the information obtained by each method reflects their specific features. For instance, radiation scattering techniques give information related to the pore size but porosity cannot be determined. For this reason, the scattering techniques are normally used in combination with SXR, which allows the measurement of the film density (that is, total porosity) as(4.1)where ρ and ρs are the film and skeleton densities, respectively. This necessitates the assumption that the skeleton is identical to the dense, non-porous prototype.
The pore connectivity is one of the important characteristics of low-k films. If pores are not interconnected, this material has fewer problems related to the diffusion of technological chemistries. Some low-k films have demonstrated pores that are closed for positronium (Ps) diffusion while they are open for toluene diffusion. Mogilnikov et al. (2004) showed that limitations in Ps diffusion are related to a difference in the ground-level energy of Ps in pores of different sizes. If a material contains pores of different sizes, Ps localizes in large voids and diffusion to small necks is restricted. As a result, PALS shows that pores are closed but the necks are still large enough to facilitate the diffusion of Cu and technological chemistries.
Intrusive techniques can only be used for evaluating open pores.
(a) Pore structure of typical low-k materials
The porous low-k films can be classified as follows: mesoporous (pore size d>2 nm), microporous (d<2 nm), films with bi-modal porosity (containing both micro- and mesopores), films with ordered periodical pores and films with embedded voids (cavities) interconnected by micropores (d<2 nm).
Figure 3a shows adsorption/desorption isotherms and pore radius distributions (PRD) for a mesoporous film. Such isotherms are typical for HSQ and xerogel films and show a pronounced hysteresis loop. The saturation points for the different adsorbates are very close to one another and correspond to an open porosity of 48%. Excellent agreement of PRDs calculated from the adsorption of different adsorbates including nitrogen is demonstrated.
The MSQ-based film shown in figure 3b is characterized by a desorption isotherm with a double slope (P/Po≈0.2 and P/Po<0.05). PRD calculation gives a bi-modal porosity. The micropores appear to be an intrinsic property of the matrix material (constitutive porosity close to 8–10%) and are related to the replacement of hydrogen by the larger methyl group. Both HSQ and MSQ films have so-called H2 type isotherms that are typical for some corpuscular systems with ill-defined pore size and shape (Gregg & Sing 1982).
Figure 3c shows ellipsometric porosimetry (EP) results for a CVD SiCOH film. These films are typically microporous and their isotherms do not have a hysteresis loop. Attempts to increase the porosity of CVD films normally lead to some increase in the pore size (Grill et al. 2003). The porosity of purely microporous films normally does not exceed 10–15%. The pore structure of the above three films are almost not observable by scanning electron microscopy and transmission electron microscopy (TEM) because of their relatively small pore size and its ill-defined structure.
Figure 4a shows results obtained for low-k films with ordered cylindrical pores prepared by Asahi Kasei (Japan). Adsorption and desorption branches in isotherms are almost vertical, suggesting a well-defined structure that is confirmed by TEM inspection.
Embedded voids interconnected by micropores are evaluated using the concept of ‘pore blocking effect’ which allows the calculation of the neck and void size by comparing the adsorption and desorption isotherms (figure 4b; Baklanov et al. 2004). The validity of this approach was proven for porous SSQ materials with different void sizes. The results of the EP evaluation were compared with the results of direct observation by TEM and there was good agreement between the results.
5. Impact of porosity on integration
Porosity affects most of the film properties. Deposition of a uniform, thin and porous low-k film is only one of the challenges that have been solved by many industrial companies. The real challenge is integration of the film into integrated circuit (IC) manufacturing processes. Compared with dense SiO2, low-k materials have low stiffness and are thermally unstable, penetrable by chemicals and so on. There are several general requirements for low-k materials to be integrated: hydrophobicity, mechanical and thermal stability, chemical and physical stability under processing conditions and compatibility with other materials. There is also the very important challenge for all functional materials: reliability in the user environment.
(a) Hydrophobicity and chemical stability
A low-k material must be hydrophobic because water has a dielectric constant close to 80. Even a small amount of adsorbed water significantly increases the total k value. This is especially important for porous materials, as they have a large surface area per unit volume, where water could potentially be adsorbed. Hydrophobicity is usually achieved by the introduction of H, CHx and other organic groups.
The problem of hydrophobicity mainly arises after technological operations related to etching (patterning) hybrid low-k dielectrics and removing an organic mask used for lithography. The presence of oxygen and hydrogen radicals in the etch plasma drastically decreases the concentration of hydrophobic groups and makes low-k materials hydrophilic (Kondoh et al. 1998). The depth of such modification depends on porosity and pore size, and can reach several tens of nanometers for highly porous materials. An interesting attempt to decrease the degree of this ‘damage’ involves the use of plasmas that do not contain either oxygen or hydrogen. Carbon-rich fluorocarbon plasmas improve the situation because CFx polymers formed during the etching are deposited on the top surface and in the upper part of the pores and seal the pores. As a result, the low-k dielectrics ‘survive’ during the following O2 or H2 plasma-based resist strip.
(b) Mechanical properties
The need for mechanical stability is primarily a consequence of the introduction of Cu as the electrical conductor in the wiring of ICs. When Al was used, the substrate was coated with Al, which was then patterned using photolithography and plasma etching. The space between the Al wires was then filled with dielectric. Cu does not form volatile compounds with reactive gases and, therefore, plasma etching is difficult. As a result, the process scheme is reversed (damascene technology). First, a substrate is coated with a dielectric layer and trenches are formed by plasma etching. A Cu layer is deposited by electroplating to fill the trenches and excess Cu is polished away. In the last step of the process, the dielectric must withstand mechanical stresses during the Cu removal polish. Low-k dielectric materials must also be able to survive stresses induced by the mismatch of thermal expansion coefficients or mechanical stresses during the packaging process when fully processed circuits are connected to the outside world.
Stiffness of low-k materials quickly deteriorates as porosity increases. The Young's modulus of bulk SiO2 decreases from 70 to 80 GPa to several GPa for materials with 50% porosity (Flannery et al. 2002). As the Young's modulus of a low-k material drops below 10 GPa, integration becomes far more challenging.
To describe the effect of porosity on mechanical properties, several models have been developed. These are cellular models for foam type solids, minimum solid area models for materials with defined pore shapes, finite-element models and models based on fractal analysis (Plawsky et al. 2003). Quantitative application of these models is still difficult because of insufficient knowledge of the real structure of porous low-k films. Mechanical properties are defined not only by porosity and effective pore size; the skeleton structure and pore shape are also important contributors.
However, analysis of porous films using these models demonstrates some general tendencies in the development of porous low-k dielectrics. Jain et al. (2001) studied the effects of processing on Young's modulus of silica xerogel films. It was shown that the models for open-cell foam materials can best describe the behaviour of xerogels in the porosity range between 25 and 75%. According to this model,(5.1)where E and ρ are the modulus and density of porous material, respectively, and Es and ρs are similar values for the dense prototype. The exponent is n=2 when bending is the main mechanism of deformation but the exponent is n=1 if axial stretching is dominant. The power-law exponent can be less than 2 for the case of closed-cell foams. The mechanical properties of xerogel films strongly depend on processing conditions. The largest Young's modulus has been demonstrated for sintered xerogel films. An interesting fact is that extrapolation of the Young modulus to zero porosity gives a value close to that of dense silica (figure 5). This curve also fits the experimental data obtained for porous glasses.
Similar extrapolation to zero porosity for porous SSQ-based films gives Young's modulus values that are significantly lower than those for SiO2. This fact suggests that the measured ‘constitutive’ micropores in the SSQ-based materials are a factor of the sparse skeleton that is not able to provide mechanical properties similar to SiO2. This is probably the reason why materials based on silica skeleton like nanocrystalline silica from Catalysts & Chemicals Industrial Corporation (CCIC, Japan) give better mechanical characteristics and are becoming more popular.
(c) Thermal stability and conductivity
Low-k materials must withstand the temperatures used in interconnect technology. The temperatures can be as high as 400–500 °C. In SSQ-based materials, elevated temperatures cause the conversion of SSQ cages into silica tetrahedrals, increasing the k-value of the material. However, most hybrid low-k materials accepted by the microelectronic industry withstand temperatures up to at least 450 °C.
The thermal conductivity (Kth) of porous low-k materials is an important challenge for integration. The increased density of transistors per chip causes more joule heating and can create problems of temperature-induced reliability. The heat conduction in disordered dielectrics may be considered the propagation of anharmonic elastic waves through a continuum (Plawsky et al. 2003). This propagation occurs via interaction between the quanta of thermal phonons. The frequency (ω) of lattice waves with velocity v covers a wide range and the thermal conductivity (Kth) can be written, in general form in terms of a superposition of these waves, as(5.2)where Cp(w) is the contribution to the specific heat per frequency interval for lattice waves of that frequency and is the attenuation length or the mean free path length for the lattice waves. At temperatures above 50 K, heat transfer can be considered to be a diffusion process. The phonon free paths are very short and the temperature-dependent thermal diffusivity (α) can be written as(5.3)where Cp is the specific heat, ν is the transport velocity of the lattice waves (phonons) and is the phonon mean free path.
For some practical applications and comparative analysis, it is useful to use the relation given by Gross & Fricke (Jain et al. 2002) for the thermal conductivity of xerogels with a density greater than 100 kg m−3, as(5.4)where kd is the conductivity of the solid matrix.
One can see that this equation is similar to equation (5.1), which gives a correlation between the density and Young's modulus. Jain et al. (2002) showed that the thermal conductivity of xerogel film strongly depends on deposition conditions. As prepared xerogel film with ethanol as the solvent showed a power-law relationship with n=1.65 that is quite close to the 1.88 in equation (5.4). The sintering of the film heals the microstructure, removes the micropores and causes the skeleton to become more uniform, improving both mechanical and thermal properties. Thus, phonon scattering is reduced and the thermal conductivity increases. In this case, n is close to unity and the linear dependence of the thermal conductivity in sintered films is observed. Similarly, low-k films with constitutive porosity, such as SSQ-based films, always show lower thermal conductivity.
(d) Relationship between the properties of porous low-k films
The analysis above suggests that the mechanical and thermal properties of porous materials are changing with porosity in similar ways. The elastic modulus as determined by acoustic waves is based on the relation . The mean free path of elastic waves is much larger than any defects including pores; therefore, the measurements are not affected by scattering. Upon combining with equation (5.3), the relationship between the thermal conductivity and elastic modulus can be expressed as(5.5)Therefore, one can expect a square relation between elastic modulus and thermal conductivity in uniform porous dielectrics. This relationship was demonstrated for porous xerogel films (Plawsky et al. 2003). Equation (5.5) does not include the pore size and geometry, although one can assume that materials with large pores (at the same porosity) should have better mechanical properties and thermal conductivity because of a thicker and more continuous skeleton.
Pore size is more important for reliability during the technological operations such as pore sealing and deposition of diffusion barriers. Large pores allow easy penetration of precursors into the low-k film during the barrier deposition in physical vapour deposition, CVD and especially in atomic layer deposition. The degree and depth of plasma damage also correlates directly with pore size.
Recently, porous SSQ-based films with the same chemical composition and different pore shape were prepared using cyclodextrin (CD) derivatives as a porogen (Lee et al. submitted). The films with relatively uniform micropores with radii close to 1 nm have been prepared using a methyl group in the CD porogen. The films prepared with a benzoyl group in the CD porogen contain large cavities with radii ranging from 3 to 10 nm. It has been demonstrated that the pore size and shape do not affect the dielectric constant and mechanical properties of the films. Porosity is a major factor in determining the dielectric constant and mechanical properties such as Young's modulus and hardness.
The stress behaviour of the films seems to be strongly correlated with the pore size, shape and porosity. The films with large ‘quasi-closed’ cavities have smaller values of residual stress. In addition, the stress decreases with increasing porosity. The diffusivity of organic solvent in the microporous film (r<1 nm) increased with the film's porosity. A clear transition of the diffusion behaviour of porous film containing large voids was observed at a porogen loading of between 20 and 30%. This transition represents the percolation threshold of the cavities when they are becoming fully interconnected.
(e) Compatibility with other materials
The three major concerns are the coefficient of thermal expansion (CTE), barrier deposition and adhesion. A low-k material must be compatible with Cu in terms of CTE, as described above. This is especially an issue for organic polymers, which can have significant CTE mismatches with Cu.
A low-k film must also be compatible with the diffusion barrier, which is needed to prevent Cu penetration. Cu degrades the dielectric properties of the film, increases leakage currents and decreases the breakdown voltage. As a result, the reliability of devices significantly decreases, making their lifetime unacceptably short. Diffusivity increases with porosity. The barrier must prevent any Cu diffusion. It must be thin (nanometre scale) and entirely dense (containing no pinholes). Covering a porous material with such a barrier is non-trivial. If the material is highly porous with large pores connected to each other, the barrier may have to be unacceptably thick in order to bridge all the exposed pores. Deposition of a rigorous barrier tends to be easiest when pores are small.
Good adhesion between a low-k material and the barrier is another requirement. Otherwise, the barrier can delaminate because of the mechanical stresses induced by polishing or thermal cycling. Adhesion can also become more of an issue as the porosity of low-k material increases.
A reduction in a dielectric constant is accomplished through the selection of materials with low polarizability chemical bonds and the introduction of porosity. Different approaches have been successfully developed and they allow the preparation of a variety of thin low-k films with different porosity and pore shape.
Integration of porous materials into microelectronic circuits, however, poses a number of challenges, as the materials must meet strict requirements in terms of properties and reliability. The most important and difficult problems are related to the fast degradation of mechanical and thermal properties as a result of porosity. Although a number of advanced methods have been developed to characterize the structure and properties of porous low-k films, much work still remains to be done to understand how the porosity, the pore structure and the skeleton properties (surface area, surface chemistry and rigidity) affect the macroscopic properties important for integration and reliability.
One contribution of 18 to a Discussion Meeting Issue ‘Engineered foams and porous materials’.
- © 2005 The Royal Society