Part of the  

Solid State Technology


The Confab


About  |  Contact

Posts Tagged ‘node’

Next Page »

Lithographic Stochastic Limits on Resolution

Monday, April 3rd, 2017


By Ed Korczynski, Sr. Technical Editor

The physical and economic limits of Moore’s Law are being approached as the commercial IC fab industry continues reducing device features to the atomic-scale. Early signs of such limits are seen when attempting to pattern the smallest possible features using lithography. Stochastic variation in the composition of the photoresist as well as in the number of incident photons combine to destroy determinism for the smallest devices in R&D. The most advanced Extreme Ultra-Violet (EUV) exposure tools from ASML cannot avoid this problem without reducing throughputs, and thereby increasing the cost of manufacturing.

Since the beginning of IC manufacturing over 50 years ago, chip production has been based on deterministic control of fabrication (fab) processes. Variations within process parameters could be controlled with statistics to ensure that all transistors on a chip performed nearly identically. Design rules could be set based on assumed in-fab distributions of CD and misalignment between layers to determine the final performance of transistors.

As the IC fab industry has evolved from micron-scale to nanometer-scale device production, the control of lithographic patterning has evolved to be able to bend-light at 193nm wavelength using Off-Axis Illumination (OAI) of Optical-Proximity Correction (OPC) mask features as part of Reticle Enhancement Technology (RET) to be able to print <40nm half-pitch (HP) line arrays with good definition. The most advanced masks and 193nm-immersion (193i) steppers today are able to focus more photons into each cubic-nanometer of photoresist to improve the contrast between exposed and non-exposed regions in the areal image. To avoid escalating cost and complexity of multi-patterning with 193i, the industry needs Extreme Ultra-Violet Lithography (EUVL) technology.

Figure 1 shows Dr. Britt Turkot, who has been leading Intel’s integration of EUVL since 1996, reassuring a standing-room-only crowd during a 2017 SPIE Advanced Lithography ( keynote address that the availability for manufacturing of EUVL steppers has been steadily improving. The new tools are close to 80% available for manufacturing, but they may need to process fewer wafers per hour to ensure high yielding final chips.

Figure 1. Britt Turkot (Intel Corp.) gave a keynote presentation on "EUVL Readiness for High-Volume Manufacturing” during the 2017 SPIE Advanced Lithography conference. (Source: SPIE)

The KLA-Tencor Lithography Users Forum was held in San Jose on February 26 before the start of SPIE-AL; there, Turcot also provided a keynote address that mentioned the inherent stochastic issues associated with patterning 7nm-node device features. We must ensure zero defects within the 10 billion contacts needed in the most advanced ICs. Given 10 billion contacts it is statistically certain that some will be subject to 7-sigma fluctuations, and this leads to problems in controlling the limited number of EUV photons reaching the target area of a resist feature. The volume of resist material available to absorb EUV in a given area is reduced by the need to avoid pattern-collapse when aspect-ratios increase over 2:1; so 15nm half-pitch lines will generally be limited to just 30nm thick resist. “The current state of materials will not gate EUV,” said Turkot, “but we need better stochastics and control of shot-noise so that photoresist will not be a long-term limiter.”

TABLE:  EUVL stochastics due to scaled contact hole size. (Source: Intel Corp.)


From the LithoGuru blog of gentleman scientist Chris Mack (

One reason why smaller pixels are harder to control is the stochastic effects of exposure:  as you decrease the number of electrons (or photons) per pixel, the statistical uncertainty in the number of electrons or photons actually used goes up. The uncertainty produces line-width errors, most readily observed as line-width roughness (LWR). To combat the growing uncertainty in smaller pixels, a higher dose is required.

We define a “stochastic” or random process as a collection of random variables (, and a Wiener process ( as a continuous-time stochastic process in honor of Norbert Wiener. Brownian motion and the thermally-driven diffusion of molecules exhibit such “random-walk” behavior. Stochastic phenomena in lithography include the following:

  • Photon count,
  • Photo-acid generator positions,
  • Photon absorption,
  • Photo-acid generation,
  • Polymer position and chain length,
  • Diffusion during post-exposure bake,
  • Dissolution/neutralization, and
  • Etching hard-mask.

Figure 2 shows the stochastics within EUVL start with direct photolysis and include ionization and scattering within a given discrete photoresist volume, as reported by Solid State Technology in 2010.

Figure 2. Discrete acid generation in an EUV resist is based on photolysis as well as ionization and electron scattering; stochastic variations of each must be considered in minimally scaled areal images. (Source: Solid State Technology)

Resist R&D

During SPIE-AL this year, ASML provided an overview of the state of the craft in EUV resist R&D. There has been steady resolution improvement over 10 years with Photo-sensitive Chemically-Amplified Resists (PCAR) from 45nm to 13nm HP; however, 13nm HP needed 58 mJ/cm2, and provided DoF of 99nm with 4.4nm LWR. The recent non-PCAR Metal-Oxide Resist (MOR) from Inpria has been shown to resolve 12nm HP with  4.7 LWR using 38 mJ/cm2, and increasing exposure to 70 mJ/cm2 has produced 10nm HP L/S patterns.

In the EUVL tool with variable pupil control, reducing the pupil fill increases the contrast such that 20nm diameter contact holes with 3nm Local Critical-Dimension Uniformity (LCDU) can be done. The challenge is to get LCDU to <2nm to meet the specification for future chips. ASML’s announced next-generation N.A. >0.5 EUVL stepper will use anamorphic mirrors and masks which will double the illumination intensity per cm2 compared to today’s 0.33 N.A. tools. This will inherently improve the stochastics, when eventually ready after 2020.

The newest generation EUVL steppers use a membrane between the wafer and the optics so that any resist out-gassing cannot contaminate the mirrors, and this allow a much wider range of materials to be used as resists. Regarding MOR, there are 3.5 times more absorbed photons and 8 times more electrons generated per photon compared to PCAR. Metal hard-masks (HM) and other under-layers create reflections that have a significant effect on the LWR, requiring tuning of the materials in resist stacks.

Default R&D hub of the world imec has been testing EUV resists from five different suppliers, targeting 20 mJ/cm2 sensitivity with 30nm thickness for PCAR and 18nm thickness for MOR. All suppliers were able to deliver the requested resolution of 16nm HP line/space (L/S) patterns, yet all resists showed LWR >5nm. In another experiment, the dose to size for imec’s “7nm-node” metal-2 (M2) vias with nominal pitch of 53nm was ~60mJ/cm2. All else equal, three times slower lithography costs three times as much per wafer pass.


Edge Placement Error Control in Multi-Patterning

Thursday, March 2nd, 2017


By Ed Korczynski, Sr. Technical Editor

SPIE Advanced Lithography remains the technical conference where the leading edge of minimum resolution patterning is explored, even though photolithography is now only part of the story. Leading OEMs continue to impress the industry with more productive ArFi steppers, but the photoresist suppliers and the purveyors of vacuum deposition and etch tools now provide most of the new value-add. Tri-layer-resist (TLR) stacks, specialty hard-masks and anti-reflective coatings (ARC), and complex thin-film depositions and etches all combine to create application-specific lithography solutions tuned to each critical mask.

Multi-patterning using complementary lithography—using argon-fluoride immersion (ArFi) steppers to pattern 1D line arrays plus extreme ultra-violet (EUV) tools to do line cuts—is under development at all leading edge fabs today. Figure 1 shows that edge placement error (EPE) in lines, cut layers, and vias/contacts between two orthogonal patterned layers can result in shorts and opens. Consequently, EPE control is critical for yield within any multi-patterning process flow, including litho-etch-litho-etch (LELE), self-aligned double-patterning (SADP) and self-aligned quadruple-patterning (SAQP).

Fig.1: Plan view schematic of 10nm half-pitch vertical lines overlaid with lower horizontal lines, showing the potential for edge-placement error (EPE). (Source: Y. Borodovsky, SPIE)

Happening the day before the official start of SPIE-AL, Nikon’s LithoVision event featured a talk by Intel Fellow and director of lithography hardware solutions Mark Phillips on the big picture of how the industry may continue to pattern smaller IC device features. Regarding the timing of Intel’s planned use of EUV litho technology, Phillips re-iterated that, “It’s highly desirable for the 7nm node, but we’ll only use it when it’s ready. However, EUVL will remain expensive even at full productivity, so 193i and multi-patterning will continue to be used. In particular we’ll need continued improvement in the 193i tools to meet overlay.”

Yuichi Shibazaki— Nikon Fellow and the main architect of the current generation of Nikon steppers—explained that the current generation of 193i steppers, featuring throughputs of >200 wafers per hour, have already been optimized to the point of diminishing returns. “In order to improve a small amount of performance it requires a lot of expense. So just improving tool performance may not decrease chip costs.” Nikon’s latest productivity offering is a converted alignment station as a stand-alone tool, intended to measure every product wafer before lithography to allow for feed-forward tuning of any stepper; cost and cost-of-ownership may be disclosed after the first beta-site tool reaches a customer by the end of this year.

“The 193 immersion technology continues to make steady progress, but there are not as many new game-changing developments,” confided Michael Lercel, Director of Strategic Marketing for ASML in an exclusive interview with SemiMD. “A major theme of several SPIE papers is on EPE, which traditionally we looked at as dependent upon CD and overlay. Now we’re looking at EPE in patterning more holistically, with need to control the complexity with different error-variables. The more information we can get the more we can control.”

At LithoVision this year, John Sturtevant—SPIE Fellow, and director of RET product development in the Design to Silicon Division at Mentor Graphics—discussed the challenges of controlling variability in multi-layer patterning. “A key challenge is predicting and then mitigating total EPE control,” reminded Sturtevant. “We’ve always paid attention to it, but the budgets that are available today are smaller than ever. Edge-placement is very important ” At the leading edge, there are multiple steps within the basic litho flow that induce proximity/local-neighbor effects which must be accounted for in EDA:  mask making, photoresist exposure, post-exposure bake (PEB), pattern development, and CD-SEM inspection (wherein there is non-zero resist shrinkage).

Due to the inherent physics of EUV lithography, as well as the atomic-scale non-uniformities in the reflective mirrors focusing onto the wafer, EUV exposure tools show significant variation in exposure uniformities. “For any given slit position there can be significant differences between tools. In practice we have used a single model of OPC for all slit locations in all scanners in the fab, and that paradigm may have to change,” said Sturtevant. “It’s possible that because the variation across the scanner is as much as the variation across the slit, it could mean we’ll need scanner-specific cross-slit computational lithography.” More than 3nm variation has been seen across 4 EUVL steppers, and the possible need for tool-specific optical proximity correction (OPC) and source-mask optimization (SMO) would be horrible for managing masks in HVM.

Thin Films Extend Patterning Resolution

Applied Materials has led the industry in thin-film depositions and etches for decades, and the company’s production proven processing platforms are being used more and more to extend the resolution of lithography. For SADP and SAQP MP, there are tunable unit-processes established for sidewall-spacer depositions, and chemical downstream etching chambers for mandrel pull with extreme material selectivity. CVD of dielectric and metallic hard-masks when combined with highly anisotropic plasma etching allows for device-specific and mask-specific pattern transfers that can reduce the line width/edge roughness (LWR/LER) originally present in the photoresist. Figure 2 from the SPIE-AL presentation “Impact of Materials Engineering on Edge Placement Error” by Regina Freed, Ying Zhang, and Uday Mitra of Applied Materials, shows LER reduction from 3.4 to 1.3 nm is possible after etch. The company’s Sym3 chamber features very high gas conductance to prevent etch byproducts from dissociation and re-deposition on resist sidewalls.

Fig.2: 3D schematics (top) and plan view SEM images (bottom) showing that control of plasma parameters can tune the byproducts of etch processes to significantly reduce the line-width roughness (LWR) of minimally scaled lines. (Source: Applied Materials)

TEL’s new SAQP spacer-on-spacer process builds on the work shown last year, using oxide as first spacer and TiO2 as second spacer. Now TEL is exploring silicon as the mandrel, then silicon-nitride as the first spacer, and titanium-oxide as second spacer. This new flow can be tuned so that all-dry etch in a single plasma etch chamber can be used for the final mandrel pull and pattern transfer steps.

Coventor’s 3D modeling software allows companies to do process integration experiments in virtual space, allowing for estimation of yield-losses in pattern transfer due to variations in side-wall profiles and LER. A simulation of 9 SRAM cells with 54 transistors shows that photoresist sidewall taper angle determines both the size and the variability of the final fins. The final capacitance of low-k dielectric in dual-damascene copper metal interconnects can be simulated as a function of the initial photoresist profile in a SAQP flow.


High-NA EUV Lithography Investment

Monday, November 28th, 2016


By Ed Korczynski, Sr. Technical Editor

As covered in a recent press release, leading lithography OEM ASML invested EUR 1 billion in cash to buy 24.9% of ZEISS subsidiary Carl Zeiss SMT, and committed to spend EUR ~760 million over the next 6 years on capital expenditures and R&D of an entirely new high numerical aperture (NA) extreme ultra-violet (EUV) lithography tool. Targeting NA >0.5 to be able to print 8 nm half-pitch features, the planned tool will use anamorphic mirrors to reduce shadowing effects from nanometer-scale mask patterns. Clever design and engineering of the mirrors could allow this new NA >0.5 tool to be able to achieve wafer throughputs similar to ASML’s current generation of 0.33 NA tools for the same source power and resist speed.

The Numerical Aperture (NA) of an optical system is a dimensionless number that characterizes the range of angles over which the system can accept or emit light. Higher NA systems can resolve finer features by condensing light from a wider range of angles. Mirror surfaces to reflect EUV “light” are made from over 50 atomic-scale bi-layers of molybdenum (Mo) and silicon (Si), and increasing the width of mirrors to reach higher NA increases the angular spread of the light which results in shadows within patterns.

In the proceedings of last year’s European Mask and Lithography Conference, Zeiss researchers reported on  “Anamorphic high NA optics enabling EUV lithography with sub 8 nm resolution” (doi:10.1117/12.2196393). The abstract summarizes the inherent challenges of establishing high NA EUVL technology:

For such a high-NA optics a configuration of 4x magnification, full field size of 26 x 33 mm² and 6’’ mask is not feasible anymore. The increased chief ray angle and higher NA at reticle lead to non-acceptable mask shadowing effects. These shadowing effects can only be controlled by increasing the magnification, hence reducing the system productivity or demanding larger mask sizes. We demonstrate that the best compromise in imaging, productivity and field split is a so-called anamorphic magnification and a half field of 26 x 16.5 mm² but utilizing existing 6’’ mask infrastructure.

Figure 1 shows that ASML plans to introduce such a system after the year 2020, with a throughput of 185 wafers-per-hour (wph) and with overlay of <2 nm. Hans Meiling, ASML vice president of product management EUV, in an exclusive interview with Solid State Technology explained why >0.5 NA capability will not be upgradable on 0.33 NA tools, “the >0.5NA optical path is larger and will require a new platform. The anamorphic imaging will also require stage architectural changes.”

Fig.1: EUVL stepper product plans for wafers per hour (WPH) and overlay accuracy include change from 0.33 NA to a new >0.5 NA platform. (Source: ASML)

Overlay of <2 nm will be critical when patterning 8nm half-pitch features, particularly when stitching lines together between half-fields patterned by single-exposures of EUV. Minimal overlay is also needed for EUV to be used to cut grid lines that are initially formed by pitch-splitting ArFi. In addition to the high NA set of mirrors, engineers will have to improve many parts of the stepper to be able to improve on the 3 nm overlay capability promised for the NXE:3400B 0.33 NA tool ASML plans to ship next year.

“Achieving better overlay requires improvements in wafer and reticle stages regardless of NA,” explained Meiling. “The optics are one of the many components that contribute to overlay. Compare to ArF immersion lithography, where the optics NA has been at 1.35 for several generations but platform improvements have provided significant overlay improvements.”

Manufacturing Capability Plans

Figure 2 shows that anamorphic systems require anamorphic masks, so moving from 0.33 to >0.5 NA requires re-designed masks. For relatively large chips, two adjacent exposures with two different anamorphic masks will be needed to pattern the same field area which could be imaged with lower resolution by a single 0.33 NA exposure. Obviously, such adjacent exposures of one layer must be properly “stitched” together by design, which is another constraint on electronic design automation (EDA) software.

Fig.2: Anamorphic >0.5 NA EUVL system planned by ASML and Zeiss will magnify mask images by 4x in the x-direction and 8x in the y-direction. (Source: Carl Zeiss SMT)

Though large chips will require twice as many half-field masks, use of anamorphic imaging somewhat reduces the challenges of mask-making. Meiling reminds us that, “With the anamorphic imaging, the 8X direction conditions will actually relax, while the 4X direction will require incremental improvements such as have always been required node-on-node.”

ASML and Zeiss report that ideal holes which “obscure” the centers of mirrors can surprisingly allow for increased transmission of EUV by each mirror, up to twice that of the “unobscured” mirrors in the 0.33 NA tool. The holes allow the mirrors to reflect through each-other, so they all line up and reflect better. Theoretically then each >0.5 NA half-field can be exposed twice as fast as a 0.33 NA full-field, though it seems that some system throughput loss will be inevitable. Twice the number of steps across the wafer will have to slow down throughput by some percent.

White two stitched side-by-side >0.5 NA EUVL exposures will be challenging, the generally known alternatives seem likely to provide only lower throughputs and lower yields:

*   Double-exposure of full-field using 0.33 NA EUVL,

*   Octuple-exposure of full-field using ArFi, or

*   Quadruple-exposure of full-field using ArFi complemented by e-beam direct-writing (EbDW) or by directed self-assembly (DSA).

One ASML EUVL system for HVM is expected to cost ~US$100 million. As presented at the company’s October 31st Investor Day this year, ASML’s modeling indicates that a leading-edge logic fab running ~45k wafer starts per month (WSPM) would need to purchase 7-12 EUV systems to handle an anticipated 6-10 EUV layers within “7nm-node” designs. Assuming that each tool will cost >US$100 million, a leading logic fab would have to invest ~US$1 billion to be able to use EUV for critical lithography layers.

With near US$1 billion in capital investments needed to begin using EUVL, HVM fabs want to be able to get productive value out of the tools over more than a single IC product generation. If a logic fab invests US$1 billion to use 0.33 NA EUVL for the “7nm-node” there is risk that those tools will be unproductive for “5nm-node” designs expected a few years later. Some fabs may choose to push ArFi multi-patterning complemented by another lithography technology for a few years, and delay investment in EUVL until >0.5 NA tools become available.


Air-Gaps for FinFETs Shown at IEDM

Friday, October 28th, 2016


By Ed Korczynski, Sr. Technical Editor

Researchers from IBM and Globalfoundries will report on the first use of “air-gaps” as part of the dielectric insulation around active gates of “10nm-node” finFETs at the upcoming International Electron Devices Meeting (IEDM) of the IEEE ( Happening in San Francisco in early December, IEDM 2016 will again provide a forum for the world’s leading R&D teams to show off their latest-greatest devices, including 7nm-node finFETs by IBM/Globalfoundries/Samsung and by TSMC. Air-gaps reduce the dielectric capacitance that slows down ICs, so their integration into transistor structures leads to faster logic chips.

History of Airgaps – ILD and IPD

As this editor recently covered at SemiMD, in 1998, Ben Shieh—then a researcher at Stanford University and now a foundry interface for Apple Corp.—first published (Shieh, Saraswat & McVittie. IEEE Electron Dev. Lett., January 1998) on the use of controlled pitch design combined with CVD dielectrics to form “pinched-off keyholes” in cross-sections of inter-layer dielectrics (ILD).

In 2007, IBM researchers showed a way to use sacrificial dielectric layers as part of a subtractive process that allows air-gaps to be integrated into any existing dielectric structure. In an interview with this editor at that time, IBM Fellow Dan Edelstein explained, “we use lithography to etch a narrow channel down so it will cap off, then deliberated damage the dielectric and etch so it looks like a balloon. We get a big gap with a drop in capacitance and then a small slot that gets pinched off.

Intel presented on their integration of air-gaps into on-chip interconnects at IITC in 2010 but delayed use until the company’s 14nm-node reached production in 2014. 2D-NAND fabs have been using air-gaps as part of the inter-poly dielectric (IPD) for many years, so there is precedent for integration near the gate-stack.

Airgaps for finFETs

Now researchers from IBM and Globalfoundries will report in (IEDM Paper #17.1, “Air Spacer for 10nm FinFET CMOS and Beyond,” K. Cheng et al) on the first air-gaps used at the transistor level in logic. Figure 1 shows that for these “10nm-node” finFETs the dielectric spacing—including the air-gap and both sides of the dielectric liner—is about 10 nm. The liner needs to be ~2nm thin so that ~1nm of ultra-low-k sacrificial dielectric remains on either side of the ~5nm air-gap.

Fig.1: Schematic of partial air-gaps only above fin tops using dielectric liners to protect gate stacks during air-gap formation for 10nm finFET CMOS and beyond. (source: IEDM 2016, Paper#17.1, Fig.12)

These air-gaps reduced capacitance at the transistor level by as much as 25%, and in a ring oscillator test circuit by as much as 15%. The researchers say a partial integration scheme—where the air-gaps are formed only above the tops of fin— minimizes damage to the FinFET, as does the high-selectivity etching process used to fabricate them.

Figure 2 shows a cross-section transmission electron micrograph (TEM) of what can go wrong with etch-back air-gaps when all of the processes are not properly controlled. Because there are inherent process:design interactions needed to form repeatable air-gaps of desired shapes, this integration scheme should be extendable “beyond” the “10-nm node” to finFETs formed at tighter pitches. However, it seems likely that “5nm-node” logic FETs will use arrays of horizontal silicon nano-wires (NW), for which more complex air-gap integration schemes would seem to be needed.

Fig.2: TEM image of FinFET transistor damage—specifically, erosion of the fin and source-drain epitaxy—by improper etch-back of the air-gaps at 10nm dimensions. (source: IEDM 2016, Paper#17.1, Fig.10)


D2S Releases 4th-Gen IC Computational Design Platform

Friday, September 30th, 2016


By Ed Korczynski, Sr. Technical Editor

D2S ( recently released the fourth generation of its computational design platform (CDP), which enables extremely fast (400 Teraflops) and precise simulations for semiconductor design and manufacturing. The new CDP is based on NVIDIA Tesla K80 GPUs and Intel Haswell CPUs, and is architected for 24×7 cleanroom production environments. To date, 14 CDPs across four platform generations are in use by customers around the globe, including six of the latest fourth generation. In an exclusive interview with SemiMD, D2S CEO Aki Fujimura stated, “Now that GPUs and CPUs are fast-enough, they can replace other hardware and thereby free up engineering resources to focus on adding value elsewhere.”

Mask data preparation (MDP) and other aspects of IC design and manufacturing require ever-increasing levels of speed and reliability as the data sets upon which they must operate grow larger and more complex with each device generation. The Figure shows a mask needed to print arrays of sub-wavelength features includes complex curvilinear shapes which must be precisely formed even though they do not print on the wafer. Such sub-resolution assist features (SRAF) increase in complexity and density as the half-pitch decreases, so the complexity of mask data increases far more than the density of printed features.

Sub-wavelength lithography using 193nm wavelength requires ever-more complex masks to repeatably print ever smaller half-pitch (HP) features, as shown by (LEFT) a typical mask composed of complex nested curves and dots which do not print (RIGHT) in the array of 32nm HP contacts/vias represented by the small red circles. (Source: D2S)

GPUs, which were first developed as processing engines for the complex graphical content of computer games, have since emerged as an attractive option for compute-intensive scientific applications due in part to their ability to run many more computing threads (up to 500x) compared to similar-generation CPUs. “Being able to process arbitrary shapes is something that mask shops will have to do,” explained Fujimura. “The world could go 193nm or EUV at any particular node, but either way there will be more features and higher complexity within the features, and all of that points to GPU acceleration.”

The D2S CDP is engineered for high reliability inside a cleanroom manufacturing environment. A few of the fab applications where CDPs are currently being used include:

  • model-based MDP for leading-edge designs that require increasingly complex mask shapes,
  • wafer plane analysis of SEM mask images to identify mask errors that print, and
  • inline thermal-effect correction of eBeam mask writers to lower write times.

“The amount of design data required to produce photomasks for leading-edge chip designs is increasing at an exponential rate, which puts more pressure on mask writing systems to maintain reasonable write times for these advanced masks. At the same time, writing these masks requires higher exposure doses and shot counts, which can cause resist proximity heating effects that lead to mask CD errors,” stated Noriaki Nakayamada, group manager at NuFlare Technology. “D2S GPU acceleration technology significantly reduces the calculation time required to correct these resist heating effects. By employing a resist heating correction that includes the use of the D2S CDP as an OEM option on our mask writers, NuFlare estimates that it can reduce CD errors by more than 60 percent, and reduce write times by more than 20 percent.”

In the E-beam Initiative 2015 survey, the most advanced reported mask-set contained >100 masks of which ~20% could be considered ‘critical’. The just released 2016 survey disclosed that the most complex single-layer mask design written last year required 16 TB of data, however platforms like D2S’ CDP have been used to accelerate writing such that the average reported write times have decreased to a weighted average of 4 hours. Meanwhile, the longest reported mask write time decreased from 72 to 48 hours.

IoT Demands Part 2: Test and Packaging

Friday, April 15th, 2016

By Ed Korczynski, Senior Technical Editor, Solid State Technology, SemiMD

The Internet-of-Things (IoT) adds new sensing and communications to improve the functionality of all manner of things in the world. Solid-state and semiconducting materials for new integrated circuits (IC) intended for ubiquitous IoT applications will have to be extremely small and low-cost. To understand the state of technology preparedness to meet the anticipated needs of the different application spaces, experts from GLOBALFOUNDRIES, Cadence, Mentor Graphics and Presto Engineering gave detailed answers to questions about IoT chip needs in EDA and fab nodes, as published in “IoT Demands:  EDA and Fab Nodes.” We continue with the conversation below.

Korczynski: For test of IoT devices which may use ultra-low threshold voltage transistors, what changes are needed compared to logic test of a typical “low-power” chip?

Steve Carlson, product management group director, Cadence

Susceptibility to process corners and operating conditions becomes heightened at near-threshold voltage levels. This translates into either more conservative design sign-off criteria, or the need for higher levels of manufacturing screening/tests. Either way, it has an impact on cost, be it hidden by over-design, or overtly through more costly qualification and test processes.

Jon Lanson, vice president worldwide sales & marketing, Presto Engineering

We need to make sure that the testability has also been designed to be functional structurally in this mode. In addition, sub-threshold voltage operation must account for non-linear transistor characteristics and the strong impact of local process variation, for which the conventional testability arsenal is still very poor. Automotive screening used low voltage operation (VLV) to detect latent defects, but at very low voltage close to the transistor threshold, digital becomes analog, and therefore if the usual concept still works for defect detection, functional test and @speed tests require additional expertise to be both meaningful and efficient from a test coverage perspective.

Korczynski:  Do we have sufficient specifications within “5G” to handle IoT device interoperability for all market segments?

Rajeev Rajan, Vice President of Internet of Things (IoT) at GLOBALFOUNDRIES

The estimated timeline for standardization availability of 5G is around 2020. 5G is being designed keeping three classes of applications in mind:  Enhanced Mobile Broadband, Massive IoT, and Mission-Critical Control. Specifically for IoT, the focus is on efficient, low-cost communication with deep coverage. We will start to see early 5G technologies start to appear around 2018, and device connectivity,

interoperability and marshaling the data they generate that can apply to multiple IoT sub-segments and markets is still very much in development.

Korczynski:  Will the 1st-generation of IoT devices likely include wide varieties of solution for different market-segments such as industrial vs. retail vs. consumer, or will most device use similar form-factors and underlying technologies?

Rajeev Rajan, Vice President of Internet of Things (IoT) at GLOBALFOUNDRIES

If we use CES 2016 as a showcase, we are seeing IoT “Things” that are becoming use-case or application-centric as they apply to specific sub-segments such as Connected Home, Automotive, Medical, Security, etc. There is definitely more variety on the consumer front vs. industrial. Vendors / OEMs / System houses are differentiating at the user-interface design and form-factor levels while the “under-the-hood” IC capabilities and component technologies that provide the atomic intelligence are fairly common. ​

Steve Carlson, product management group director, Cadence

Right now it seems like everyone is swinging for the fence. Everyone wants the home-run product that will reach a billion devices sold. Generality generally leads to sub-optimality, so a single device usually fails to meet the needs and expectations of many. Devices that are optimized for more specific use cases and elements of purchasing criteria will win out. The question of interface is an interesting one.

Korczynski:  Will there be different product life-cycles for different IoT market-segments, such as 1-3 years for consumer but 5-10 years for industrial?

Rajeev Rajan, Vice President of Internet of Things (IoT) at GLOBALFOUNDRIES

That certainly seems to be the case. According to Gartner’s market analysis for IoT, Consumer is expected to grow at a faster pace in terms of units compared to Enterprise, while Enterprise is expected to lead in revenue. Also the churn-cycle in Consumer is higher / faster compared to Enterprise. Today’s wearables or smart-phones are good reference examples. This will however vary by the type of “Thing” and sub-segment. For example, you expect to have your smart refrigerator for a longer time period compared to smart clothing or eyewear. As ASPs of the “Things”come down over time and new classes of products such as disposables hit the market, we can expect even larger volumes.​

Jon Lanson, vice president worldwide sales & marketing, Presto Engineering

The market segments continue to be driven by the same use cases. In consumer wearables, short cycles are linked to fashion trends and rapid obsolescence, where consumer home use has longer cycles closer to industrial market requirements. We believe that the lifecycle norms will hold true for IoT devices.

Korczynski:  For the IoT application of infrastructure monitoring (e.g. bridges, pipelines, etc.) long-term (10-20 year) reliability will be essential, while consumer applications may be best served by 3-5 year reliability devices which cost less; how well can we quantify the trade-off between cost and chip reliability?

Steve Carlson, product management group director, Cadence

Conceptually we know very well how to make devices more reliable. We can lower current densities with bigger wires, we can run at cooler temperatures, and so on.  The difficulty is always in finding optimality for a given criterion across the, for practical purposes, infinite tradeoffs to be made.

Korczynski:  Why is the talk of IoT not just another “Dot Com” hype cycle?

Rajeev Rajan, Vice President of Internet of Things (IoT) at GLOBALFOUNDRIES

​​I participated in a panel at SEMICON China in Shanghai last month that discussed a similar question. If we think of IoT as a “brand new thing” (no pun intended), then we can think of it as hype. However if we look at the IoT as as set of use-cases that can take advantage of an evolution of Machine-to-Machine (M2M) going towards broader connectivity, huge amounts of data generated and exchanged, and a generational increase in internet and communication network bandwidths (i.e. 5G), then it seems a more down-to-earth technological progression.

Nicolas Williams, product marketing manager, Mentor Graphics

Unlike the Dot Com hype, which was built upon hope and dreams of future solutions that may or may not have been based in reality, IoT is real business. For example, in a 2016 IC Insights report, we see that last year $63.4 billion in revenue was generated for IoT systems and the market is growing at about 20% CAGR. This same report also shows IoT semiconductor sales of over $15 billion in 2015 with a CAGR of 21.1%.

Jon Lanson, vice president worldwide sales & marketing, Presto Engineering

It is the investment needed up front to create sensing agents and an infrastructure for the hardware foundation of the IoT that will lead to big data and ultimately value creation.

Steve Carlson, product management group director, Cadence

There will be plenty of hype cycles for products and product categories along the way. However, the foundational shift of the connection of things is a diode through which civilization will only pass through in one direction.

IoT Demands Part 1: EDA and Fab Nodes

Thursday, April 14th, 2016

The Internet-of-Things (IoT) is expected to add new sensing and communications to improve the functionality of all manner of things in the world:  bridges sensing and reporting when repairs are needed, parts automatically informing where they are in storage and transport, human health monitoring, etc. Solid-state and semiconducting materials for new integrated circuits (IC) intended for ubiquitous IoT applications will have to be assembled at low-cost and small-size in High Volume Manufacturing (HVM). Micro-Electro-Mechanical Systems (MEMS) and other sensors are being combined with Radio-Frequency (RF) ICs in miniaturized packages for the first wave of growth in major sub-markets.

To meet the anticipated needs of the different IoT application spaces, SemiMD asked leading companies within critical industry segments about the state of technology preparedness:


*  Electronic Design Automation (EDA) – Cadence and Mentor Graphics,

*  IC and complex system test – Presto Engineering.

Korczynski:  Today, ICs for IoT applications typically use 45nm/65nm-node which are “Node -3″ (N-3) compared to sub-20nm-node chips in HVM. Five years from now, when the bleeding-edge will use 10nm node technology, will IoT chips still use N-3 of 28nm-node (considered a “long-lived node”) or will 45nm-node remain the likely sweet-spot of price:performance?

Timothy Dry, product marketing manager, GLOBALFOUNDRIES

In 5 years time, there will be a spread of technology solutions addressing low, middle, and high ends of IoT applications. At the low end, IoT end nodes for applications like connected smoke

detectors, security sensors will be at 55, 40nm ULP and ULL for lowest system power, and low cost. These applications will be typically served by MCUs <50DMIPs. Integrated radios (BLE, 802.15.4), security, Power Management Unit (PMU), and eFlash or MRAM will be common features. Connected LED lighting is forecasted to be a high volume IoT application. The LED drivers will use BCD extensions of 130nm—40nm—that can also support the radio and protocol-MCU with Flash.

In the mid-range, applications like smart-meters and fitness/medical monitoring will need systems that have more processing power <300DMIPS. These products will be implemented in 40nm, 28nm and GLOBALFOUNDRIES’ new 22nm FDSOI technology that uses software-controlled body-biasing to tune SoC operation for lowest dynamic power. Multiple wireless (BLE/802.15.4, WiFi, LPWAN) and wired connectivity (Ethernet, PLC) protocols with security will be integrated for gateway products.

High-end products like smart-watches, learning thermostats, home security/monitoring cameras, and drones will require MPU-class IC products (~2000DMIPs) and run high-order operating systems (e.g. Linux, Android). These products will be made in leading-edge nodes starting at 22FDX, 14FF and migrating to 7FF and beyond. Design for lowest dynamic power for longest battery life will be the key driver, and these products typically require human machine Interface (HMI) with animated graphics on a high resolution displays. Connectivity will include BLE, WiFi and cellular with strong security.

Steve Carlson, product management group director, Cadence

We have seen recent announcements of IoT targeted devices at 14nm. The value created by Moore’s Law integration should hold, and with that, there will be inherent advantages to those who leverage next generation process nodes. Still, other product categories may reach functionality saturation points where there is simply no more value obtained by adding more capability. We anticipate that there will be more “live” process nodes than ever in history.

Jon Lanson, vice president worldwide sales & marketing, Presto Engineering

It is fair to say that most IoT devices will be a heterogeneous aggregation of analog functions rather than high power digital processors. Therefore, and by similarity with Bluetooth and RFID devices, 90nm and 65nm will remain the mainstream nodes for many sub-vertical markets, enabling the integration of RF and analog front-end functions with digital gate density. By default, sensors will stay out of the monolithic path for both design and cost reasons. The best answer would be that the IoT ASIC will follow eventually the same scaling as the MCU products, with embedded non-volatile memories, which today is 55-40nm centric and will move to 28nm with industry maturity and volumes.

Korczynski:  If most IoT devices will include some manner of sensor which must be integrated with CMOS logic and memory, then do we need new capabilities in EDA-flows and burn-in/test protocols to ensure meeting time-to-market goals?

Nicolas Williams, product marketing manager, Mentor Graphics

If we define a typical IoT device as a product that contains a MEMS sensor, A/D, digital processing, and a RF-connection to the internet, we can see that the fundamental challenge of IoT design is that teams working on this product need to master the analog, digital, MEMS, and RF domains. Often, these four domains require different experience and knowledge and sometimes design in these domains is accomplished by separate teams. IoT design requires that all four domains are designed and work together, especially if they are going on the same die. Even if the components are targeting separate dice that will be bonded together, they still need to work together during the layout and verification process. Therefore, a unified design flow is required.

Stephen Pateras, product marketing director, Mentor Graphics

Being able to quickly debug and create test patterns for various embedded sensor IP can be addressed with the adoption of the new IEEE 1687 IP plug-and-play standard. If a sensor IP block’s digital interface adheres to the standard, then any vendor-provided data required to initialize or operate the embedded sensor can be easily and quickly mapped to chip pins. Data sequences for multiple sensor IP blocks can also be merged to create optimized sequences that will minimize debug and test times.

Jon Lanson, vice president worldwide sales & marketing, Presto Engineering

From a testing standpoint, widely used ATEs are generally focused on a few purposes, but don’t necessarily cover all elements in a system. We think that IoT devices are likely to require complex testing flows using multiple ATEs to assure adequate coverage. This is likely to prevail for some time as short run volumes characteristic of IoT demands are unlikely to drive ATE suppliers to invest R&D dollars in creating new purpose-built machines.

Korczynski:  For the EDA of IoT devices, can all sensors be modeled as analog inputs within established flows or do we need new modeling capability at the circuit level?

Steve Carlson, product management group director, Cadence

Typically, the interface to the physical world has been partitioned at the electrical boundary. But as more mechanical and electro-mechanical sensors are more deeply integrated, there has been growing value in co-design, co-analysis, and co-optimization. We should see more multi-domain analysis over time.

Nicolas Williams, product marketing manager, Mentor Graphics

Designers of IoT devices that contain MEMS sensors need quality models in order to simulate their behavior under physical conditions such as motion and temperature. Unlike CMOS IC design, there are few standardized MEMS models for system-level simulation. State of the art MEMS modeling requires automatic generation of behavioral models based on the results of Finite Element Analysis (FEA) using reduced-order modeling (ROM). ROM is a numerical methodology that reduces the analysis results to create Verilog-A models for use in AMS simulations for co-simulation of the MEMS device in the context of the IoT system.

EUV Resists and Stochastic Processes

Friday, March 4th, 2016


By Ed Korczynski, Sr. Technical Editor

In an exclusive interview with Solid State Technology during SPIE-AL this year, imec Advanced Patterning Department Director Greg McIntyre said, “The big encouraging thing at the conference is the progress on EUV.” The event included a plenary presentation by TSMC Nanopatterning Technology Infrastructure Division Director and SPIE Fellow Anthony Yen on “EUV Lithography: From the Very Beginning to the Eve of Manufacturing.” TSMC is currently learning about EUVL using 10nm- and 7nm-node device test structures, with plans to deploy it for high volume manufacturing (HVM) of contact holes at the 5nm node. Intel researchers confirm that they plan to use EUVL in HVM for the 7nm node.

Recent improvements in EUV source technology— 80W source power had been shown by the end of 2014, 185W by the end of 2015, and 200W has now been shown by ASML—have been enabled by multiple laser pulses tuned to the best produce plasma from tin droplets. TSMC reports that 518 wafers per day were processed by their ASML EUV stepper, and the tool was available ~70% of the time. TSMC shows that a single EUVL process can create 46nm pitch lines/spaces using a complex 2D mask, as is needed for patterning the metal2 layer within multilevel on-chip interconnects.

To improve throughput in HVM, the resist sensitivity to the 13.54nm wavelength radiation of EUV needs to be improved, while the line-width roughness (LWR) specification must be held to low single-digit nm. With a 250W source and 25 mJ/cm2 resist sensitivity an EUV stepper should be able to process ~100 wafer-per-hour (wph), which should allow for affordable use when matched with other lithography technologies.

Researchers from Inpria—the company working on metal-oxide-based EUVL resists—looked at the absorption efficiencies of different resists, and found that the absorption of the metal oxide based resists was ≈ 4 to 5 times higher than that of the Chemically-Amplified Resist (CAR). The Figure shows that higher absorption allows for the use of proportionally thinner resist, which mitigates the issue of line collapse. Resist as thin as 18nm has been patterned over a 70nm thin Spin-On Carbon (SOC) layer without the need for another Bottom Anti-Reflective Coating (BARC). Inpria today can supply 26 mJ/cm2 resist that creates 4.6nm LWR over 140nm Depth of Focus (DoF).

To prevent pattern collapse, the thickness of resist is reduced proportionally to the minimum half-pitch (HP) of lines/spaces. (Source: JSR Micro)

JEIDEC researchers presented their summary of the trade-off between sensitivity and LWR for metal-oxide-based EUV resists:  ultra high sensitivity of 7 mJ/cm2 to pattern 17nm lines with 5.6nm LWR, or low sensitivity of 33 mJ/cm2 to pattern 23nm lines with 3.8nm LWR.

In a keynote presentation, Seong-Sue Kim of Samsung Electronics stated that, “Resist pattern defectivity remains the biggest issue. Metal-oxide resist development needs to be expedited.” The challenge is that defectivity at the nanometer-scale derives from “stochastics,” which means random processes that are not fully predictable.

Stochastics of Nanopatterning

Anna Lio, from Intel’s Portland Technology Development group, stated that the challenges of controlling resist stochastics, “could be the deal breaker.” Intel ran a 7-month test of vias made using EUVL, and found that via critical dimensions (CD), edge-placement-error (EPE), and chain resistances all showed good results compared to 193i. However, there are inherent control issues due to the random nature of phenomena involved in resist patterning:  incident “photons”, absorption, freed electrons, acid generation, acid quenching, protection groups, development processes, etc.

Stochastics for novel chemistries can only be controlled by understanding in detail the sources of variability. From first-principles, EUV resist reactions are not photon-chemistry, but are really radiation-chemistry with many different radiation paths and electrons which can be generated. If every via in an advanced logic IC must work then the failure rate must be on the order of 1 part-per-trillion (ppt), and stochastic variability from non-homogeneous chemistries must be eliminated.

Consider that for a CAR designed for 15mJ/cm2 sensitivity, there will be just:

145 photons/nm2 for 193, and

10 photons/nm2 for EUV.

To improve sensitivity and suppress failures from photon shot-noise, we need to increase resist absorption, and also re-consider chemical amplification mechanisms. “The requirements will be the same for any resist and any chemistry,” reminded Lio. “We need to evaluate all resists at the same exposure levels and at the same rules, and look at different features to show stochastics like in the tails of distributions. Resolution is important but stochastics will rule our world at the dimensions we’re dealing with.”


Measuring 5nm Particles In-Line

Monday, November 30th, 2015

By Ed Korczynski, Sr. Technical Editor

Industrial Technology Research Institute (ITRI) ( worked with TSMC ( in Taiwan on a clever in-line monitor technology that transforms liquids and automatically-diluted-slurries into aerosols for subsequent airborn measurements. They call this “SuperSizer” technology, and claim that tests have shown resolution over the astounding range of 5nm to 1 micron, and with ability to accurately represent size distributions over that range. Any dissolved gas bubbles in the liquid are lost in the aerosol process, which allows the tool to unambiguously count solid impurities. The Figure shows the compact components within the tool that produce the aerosol.

Aerosol sub-system inside “SuperSizer” in-line particle sizing tool co-developed by ITRI/TSMC. (Source: ITRI)

Semiconductor fabrication (fab) lines require in-line measurement and control of particles in critical liquids and slurries. With the exception of those carefully added to chemical-mechanical planarization (CMP) slurries, most particles in fabs are accidental yield-killers that must be kept to an absolute minimum to ensure proper yield in IC fabs, and ever decreasing IC device feature sizes result in ever smaller particles that can kill a chip. Standard in-line tools to monitor particles rely on laser scattering through the liquid, and such technology allows for resolution of particle sizes as small as 40nm. Since we cannot control what we cannot measure, the IC fab industry needs this new ability to measure particles as small as 5nm for next-generation manufacturing.

There are two actual measurement technologies used downstream of the SuperSizer aerosol module:  a differential mobility analyzer (DMA), and a condensation particle counter (CPC). The aerosol first moves through the DMA column, where particle sizes are measured based on the force balance between air flow speed in the axial direction and an electric field in the radial direction. The subsequent CPC then provides particle concentration data.

Combining both data streams properly allows for automated output of information on particle sizes down to 5nm, size distributions, and impurity concentrations in liquids. Since the tool is intended for monitoring semiconductor high-volume manufacturing (HVM), the measurement data is automatically categorized, analyzed, and reported according to the needs of the fab’s automated yield management system. Users can edit the measurement sequences or recipes to monitor different chemicals or slurries under different conditions and schedules.

When used to control a CMP process, the SuperSizer can be configured to measure not just impurities but also the essential slurry particles themselves. During dilution and homogeneous mixing of the slurry prior to aerosolization, mechanical agitation needs to be avoided so as to prevent particle agglomeration which causes scratch defects. This new tool uses pressured gas as the driving force for solution transporting and mixing, so that any measured agglomeration in the slurry can be assigned to a source somewhere else in the fab.

TSMC has been using this tool since 2014 to measure particles in solutions including slurries, chemicals, and ultra-pure water. ITRI, which owns the technology and related patents, can now take orders to manufacture the product, but the research organization plans to license the technology to a company in Taiwan for volume manufacturing. EETimes reports ( that the current list price for a tool capable of monitoring ultra-pure water is ~US$450k, while a fully-configured tool for CMP monitoring would cost over US$700k.


Monolithic 3D processing using non-equilibrium RTP

Friday, April 17th, 2015


By Ed Korczynski, Senior Technical Editor, Solid State Technology

Slightly more than one year after Qualcomm Technologies announced that it was assessing CEA-Leti’s monolithic 3D (M3D) transistor stacking technology, Qualcomm has now announced that M3D will be used instead of through-silicon vias (TSV) in the company’s next generation of cellphone handset chips. Since Qualcomm had also been a leading industrial proponent of TSV over the last few years while participating in the imec R&D consortium, this endorsement of M3D is particularly relevant.

Leti’s approach to 3D stacking of transistors starts with a conventionally built and locally-interconnected bottom layer of transistors, which are then covered with a top layer of transistors built using relatively low-temperature processes branded as “CoolCube.” Figure 1 shows a simplified cross-sectional schematic of a CoolCube stack of transistors and interconnects. CoolCube M3D does not transfer a layer of built devices as in the approach using TSV, but instead transfers just a nm-thin layer of homogenous semiconducting material for subsequent device processing.

Fig. 1: Simplified cross-sectional rendering of Monolithic 3D (M3D) transistor stacks, with critical process integration challenges indicated. (Source: CEA-Leti)

The reason that completed transistors are not transferred in the first place is because of intrinsic alignment issues, which are eliminated when transistors are instead fabricated on the same wafer. “We have lots of data to prove that alignment precision is as good as can be seen in 2D lithography, typically 3nm,” explained Maud Vinet, Leti’s advanced CMOS laboratory manager in an exclusive interview with SST.

As discussed in a blog post online at Semiconductor Manufacturing and Design ( last year by Leti researchers, the M3D approach consists of sequentially processing:

  • processing a bottom MOS transistor layer with local interconnects,
  • bonding a wafer substrate to the bottom transistor layer,
  • chemical-mechanical planarization (CMP) and SPE of the top layer,
  • processing the top device layer,
  • forming metal vias between the two device layers as interconnects, and
  • standard copper/low-k multi-level interconnect formation.

To transfer a layer of silicon for the top layer of transistors, a cleave-layer is needed within the bulk silicon or else time and money would be wasted in grinding away >95% of the silicon bulk from the backside. For CMOS:CMOS M3D thin silicon-on-insulator (SOI) is the transferred top layer, a logical extension of work done by Leti for decades. The heavy dose ion-implantation that creates the cleave-layer leaves defects in crystalline silicon which require excessively high temperatures to anneal away. Leti’s trick to overcome this thermal-budget issue is to use pre-amorphizing implants (PAI) to completely dis-order the silicon before transfer and then solid-phase epitaxy (SPE) post-transfer to grow device-grade single-crystal silicon at ~500°C.

Since neither aluminum nor copper interconnects can withstand this temperature range, the interconnects for the bottom layer of transistors need to be tungsten wires with the highest melting point of any metal but somewhat worse electrical resistance (R). Protection for the lower wires cannot use low-k dielectrics, but must use relatively higher capacitance (C) oxides. However, the increased RC delay in the lower interconnects is more than offset by the orders-of-magnitude reduction in interconnect lengths due to vertical stacking.

M3D Roadmaps

Leti shows data that M3D transistor stacking can provide immediate benefit to industry by combining two 28nm-node CMOS layers instead of trying to design and manufacture a single 14nm-node CMOS layer:  area gain 55%, performance gain 23%, and power gain 12%. With cost/transistor now expected to increase with sequential nodes, M3D thus provides a way to reduce cost and risk when developing new ICs.

For the industry to use M3D, there are some unique new unit-processes that will need to ramp into high-volume manufacturing (HVM) to ensure profitable line yield. As presented by C. Fenouillet-Beranger et al. from Leti and ST (paper 27.5) at IEDM2014 in San Francisco, “New Insights on Bottom Layer Thermal Stability and Laser Annealing Promises for High Performance 3D Monolithic Integration,” due to stability improvement in bottom transistors found through the use of doping nickel-silicide with a noble metal such as platinum, the top MOSFET processing temperature could be relaxed up to 500°C. Laser RTP annealing then allows for the activation of top MOSFETs junctions, which have been characterized morphologically and electrically as promising for high performance ICs.

Figure 2 shows the new unit-processes at <=500°C that need to be developed for top transistor formation:

*   Gate-oxide formation,

*   Dopant activation,

*   Epitaxy, and

*   Spacer deposition.

Fig. 2: Thermal processing ranges for process modules need to be below ~500°C for the top devices in M3D stacks to prevent degradation of the bottom layer. (Source: CEA-Leti)

After the above unit-processes have been integrated into high-yielding process modules for CMOS:CMOS stacking, heterogeneous integration of different types of devices are on the roadmap for M3D. Leti has already shown proof-of-concept for processes that integrate new IC functionalities into future M3D stacks:

1)       CMOS:CMOS,

2)       PMOS:NMOS,

3)       III-V:Ge, and

4)       MEMS/NEMS:CMOS.

Thomas Ernst, senior scientist, Electron Nanodevice Architectures, Leti, commented to SST, “Any application that will need a ‘pixelated’ device architecture would likely use M3D. In addition, this approach will work well for integrating new channel materials such as III-V’s and germanium, and any materials that can be deposited at relatively low temperatures such as the active layers in gas-sensors or resistive-memory cells.”

Non-Equilibrium Thermal Processing

Though the use of an oxide barrier between the active device layers provides significant thermal protection to the bottom layer of devices during top-layer fabrication, the thermal processes of the latter  cannot be run at equilibrium. “One way of controlling the thermal budget is to use what we sometimes call the crème brûlée approach to only heat the very top surface while keeping the inside cool,” explained Vinet. “Everyone knows that you want a nice crispy top surface with cool custard beneath.” Using a laser with a short wavelength prevents penetration into lower layers such that essentially all of the energy is absorbed in the surface layer in a manner that can be considered as adiabatic.

Applied Materials has been a supplier-partner with Leti in developing M3D, and the company provided responses from executive technologists to queries from SST about the general industry trend to controlling short pulses of light for thermal processing. “Laser non-equilibrium heating is enabling technology for 3D devices,” affirmed Steve Moffatt, chief technology officer, Front End Products, Applied Materials. “The idea is to heat the top layer and not the layers below. To achieve very shallow adiabatic heating the toolset needs to ramp up in less than 100 nsec. In order to get strong absorption in the top surface, shorter wavelengths are useful, less than 800 nm. Laser non-equilibrium heating in this regime can be a critical process for building monolithic 3D structures for SOC and logic devices.”

Of course, with ultra-shallow junctions (USJ) and atomic-scale gate-stacks already in use for CMOS transistors at the 22nm-node, non-equilibrium thermal processing has already been used in leading fabs. “Gate dielectric, gate metal, and contact treatments are areas where we have seen non-equilibrium anneals slowly taking the place of conventional RTP,” clarified Abhilash Mayur, senior director, Front End Products, Applied Materials. “For approximate percentages, I would say about 25 percent of thermal processing for logic at the 22nm-node is non-equilibrium, and seen to be heading toward 50 percent at the 10nm-node or lower.”

Mayur further explained some of the trade-offs in working on the leading-edge of thermal processing for demanding HVM customers. Pulse-times are in the tens of nsec, with longer pulses tending to allow the heat to diffuse deeper and adversely alter the lower layers, and with shorter pulses tending to induce surface damage or ablation. “Our roadmap is to ensure flexibility in the pulse shape to tailor the heat flow to the specific application,” said Mayur.

Now that Qualcomm has endorsed CoolCube M3D as a preferred approach to CMOS:CMOS transistor stacking in the near-term, we may assume that R&D in novel unit-processes has mostly concluded. Presumably there are pilot lots of wafers now being run through commercial foundries to fine-tune M3D integration. With a roadmap for long-term heterogeneous integration that seems both low-cost and low-risk, M3D using non-equilibrium RTP will likely be an important way to integrate new functionalities into future ICs.

Next Page »