Final Report Summary - IAQSENSE (Nanotechnology based gas multispectral sensing system for environmental control and protection)
Executive Summary:
Indoor air quality is a crucial factor for the human’s apperception. It is proven that air quality affects the health and the well-being of people. Up to now, only expert lab systems could measure air quality.
IAQSense Project has managed the R&D of new nanotechnology based sensor systems, completely unique solutions for miniaturized, low cost and mass produced sensors.
The main technology innovation is based on surface ion mobility dynamics, it separates each gas component and its concentration like a mass spectrometer and allows high sensitivity (ppb) fast (1s) multi-gas detection.
The IAQSense Project demonstrated that such a device can characterize, monitor and improve indoor air quality, in a way for which no affordable sensor exists. To enlarge the field of detection, the sensor system includes other patented biomolecule detection and security related nanotechnologies .
• The IAQsense objectives were to develop sensor systems, including all electronics, pattern detection firmware and validate these systems and their impact in real environments, firstly in buildings, but also show the potential in vehicles and for health and hazard scenarios.
• High-potential markets have been discovered and validated with potential users, product range has been defined for Building automation and IoT markets.
• Validation results with VOC generators and real building environments have confirmed the huge impact of real-time VOC monitoring for comfort and energy saving while guiding launching priorities.
• Ability for mass production and low cost were mandatory requirements for the development and silicon foundries have been contacted for steps beyond the project.
• The development includes a highly integrated IC for real-time gas pattern processing with special architecture for very low power and energy management; test and validation; predictive numerical models developed in the project; and implementation into buildings to identify and model the impact of this disruptive technology, before industrialization.
• The deliverables of IAQSense are nanotechnology embedded sensor systems, with on-board knowledge-based smart sensing. It creates a fast, low cost, high resolution system, with high added value, ready for mass production.
By developing 3 technologies, a range of products, 4 components and 5 different IoT compatible sensing boards, IAQSense brings breakthrough solutions close to the market for volatile molecule sensing in buildings, chemical threats detection, medical applications. Recognized as such by the Consortium Partners and market operators it has favoured the creation of a Spin-off Company managing the interest of the Partners. It provides a one stop shop team for further development, integration, customization/functionalization and managing commercial exploitation.
IAQSense results will have the following impact:
• provide solutions for real-time analysis and control of air quality on every place, home, office, vehicles,
• participate in the optimum reduction in energy consumption,
• participate in IoT and pervasive sensing trends and
• paves the way for integration into wearable devices for applications for health and security.
The consortium is composed of 4 SMEs, 3 industrial, 3 research institutes gathering all patent owners, specialists in marketing, project management and European industrial leaders in nanotechnologies & microelectronics. The Technical Committee are users of this innovative technology in the construction area. Dissemination actions has provided more contacts in automotive, medical and security markets while generating further opportunities for breakthrough projects.
Project Context and Objectives:
S/T quality, relevant to the topics addressed by the call
CONCEPT AND OBJECTIVES
Context and current challenges
For the last 20 years, there has been a growing concern regarding pollutants in closed environments and the difficulty in identifying these pollutants and their critical levels, without heavy, expensive equipment. Currently, a critical question to be addressed is the new regulation for ‘air tightness’ of energy efficient buildings, new or renovated, and the growing number of new materials in construction.
“Indoor Air Quality in Office Buildings: A Technical Guide” - Health Canada 1995 at this time indicated that “Several thousand chemicals, synthetic and natural, are Volatile Organic Compounds (VOC). Over 900 have been identified in indoor air, with over 250 recorded at concentrations higher than 1 ppb.”
Further this report also indicates that “indoor levels may be 10 times higher than those outdoors, in range of few mg/m3. Identification and measurement of individual VOCs are expensive and time-consuming, and the total cost is underestimated as the VOCs, at very low concentrations, are very difficult to identify.”
Research shows that “VOCs at concentrations below threshold limit values can cause discomfort, including fatigue, headaches, drowsiness, dizziness, weakness, joint pains, peripheral numbness or tingling, euphoria, tightness in the chest, unsteadiness, blurred vision, and skin and/or eye irritation. In the exposure range of 0.3-3 mg/m3, odours, irritation, and discomfort may appear, above 3 mg/m3 complaints, above 25 mg/m3, temporary discomfort and respiratory. Office levels cover all ranges. And hypersensitive individuals can have severe reactions at very low concentrations of VOCs emitted by their environment.”
In the EU, European Collaborative Action has published 26 reports (1996 – 2007) connected to IAQ.
Report n°19 indicates 63 VOCs that may commonly be found indoor and makes a recommendation to “Develop and evaluate improved strategies and systems to control ventilation rates, including demand-controlled systems using pollutant sensors for Indoor and Outdoor air quality and more accurate measurement of air flow within the system”. Report n°20 shows the relationship between levels, odours and comfort.
“Report on Carcinogens 12th Edition 2011” – USA lists 250 carcinogens and their threshold limits.
“Valeurs guides de qualité d’air intérieur (AFSSET (now ANSES)/CSTB 2007)” - France lists 222 VOC, close to the AgBB German standard. ANSES has also published a list of guidelines with health threshold values.
Furthermore, the French Centre for the IAQ (OQAI) has made a classification of 58 pollutants (classified as very high priority & high priority) in the dwellings according to several criteria such as the apparition frequency, the potential of acute risks and chronic risks etc... (www.oqai.fr).
For security reasons, for example in airports and in public or private buildings, the detection of potential threats like chemical/bio, hazards and narcotics is increasingly required. Extremely sensitive and selective sensors are needed for these specific purposes.
In addition to VOC, bio molecules are also present in closed environments. The presence of bio molecules can be an indication of the presence of mould, viruses, etc. indicating the spread of a disease or the presence of allergen factors. Detection of bio molecules requires specific sensing methods.
Given the absence of affordable but sensitive and selective gas sensors on the market to tackle the current challenges, the project aimed to develop new advanced sensor systems to precisely monitor the composition of the air in terms of both chemical and bio contaminants.
IAQSense identified all the potential applications for the novel technologies to be further developed in the project. A common set of specifications has be defined for design and production. Furthermore, a survey has been conducted through the partners of the project, the enable the attached Technical Committee members to refine the technical requirements and the market potential. The development has started on this basis.
A broader review on a worldwide basis has be conducted, to update the potential applications knowing development and application results, to segment the market, to identify the key potential users and to adjust the final product technical specifications to the market needs.
This was also be an opportunity to review the emerging competition, new maturing technologies, new trends in the market and to adjust the product offer.
The IAQSense offer is very unique, and latent needs has also be revealed by this new offer, especially the real-time monitoring opportunities, triggering new applications, never thought of before. It is an opportunity for the consortium as a whole, a chance that should not be underestimated and has be checked carefully through new ideas, new product innovation and a market attraction survey.
More recently opportunities to enter the IoT trends and markets has been identified and developments targeted.
Origin of the project
Regarding the requirements for multi-gas detection, some of the IAQSense project partners have developed a silicon based technology, integrating the equivalent of a spectrometer but at a nano-scale.
Based on surface ion mobility dynamics this is particularly suitable to identify each individual gas in the ambient air and its density. This technology has been patented by EFFICIENCE Marketing (UK patent application number: 1200626.8 “Ionensensors”, 16/01/12 – France and Germany in 2013). This nanotechnology has a high potential to be both affordable and low-cost. One additional patent targeting nano-mechanics (US Patent 7,335,942 B2 – 2008 “Field Effect Transistor Sensor cantilevers”. This patent is now NANO property in Germany while the third Nanorods technology developed by ISL is free (already the subject of publication).
All the patents’ owners and technology stakeholders are involved in the project. Therefore all patents are available as the background of this project.
Based on this solid background, the IAQSense project emanates from the will to demonstrate the relevancy of these devices applied to indoor air quality characterization and monitoring, for which no efficient sensors currently exist.
Furthermore security market and pervasive sensing of chemical threats for the industry, for the protection of public buildings, crowded areas has been targeted.
And last but not least wearables for allergy prevention and occupational health has been also investigated.
The project should go to industrialization for mass markets for the above 3 markets.
Later on, the large Automotive market should benefit from these technologies (Cabin monitoring and Air Quality Control)
S/T Objectives
The S&T objectives of the project are the development of the sensor, its integrated electronics, and several wireless micro modules, validated with a set of the most common molecules present Indoor.
Research in new algorithms for building automation has been explored as contribution of this smart sensor technology for the optimisation of health, comfort and energy savings.
Software Tools for extension of the set of detectable molecules (retrofit download) to be provided.
The technologies has achieved a TRL6 level at least by the end of the project for industrialization and mass production preparation.
The sensing microsystem developed during the project should be able to detect:
• separately any VOC below threshold limits (below 100ppb),
• biomolecules down to ppb levels,
• ultra-traces of hazard and narcotics down to ppt levels.
During the project, the entire system has been developed and derivatives defined for different applications i.e.:
• primarily for buildings (monitoring IAQ),
• health,
• vehicles, and
• security (detection of hazards and/or narcotics).
To work as smart sensor and to detect specific threats and toxic gas, gas patterns are validated and implemented as firmware in an Integrated Circuit. The firmware is upgradable.
Project Results:
Functional specifications and methodology – EFFICIENCE Marketing
Objectives
This WP aims at confirming the primary (serving the building industry) and secondary (health, vehicles, safety) objectives of this project, specifying the spectrum of molecules to detect and then to deduce the specification guidelines for the technological development, electronics, embedded software and the target micro-module configuration at project completion.
Functional specifications (Define from existing market needs, and potential innovative applications) EM; NANO, ISL, MICRO, CEA
For the Functional Specifications priority was given to building applications. This corresponds to the composition of the Technical Committee.
Expected experts in indoor air pollution and health impacts, industrial managers in construction, building automation, ventilation, sensing, have been interviewed. Their profile range from innovative SME’s to Large Companies with international profile. Big companies like Bouygues and Schneider were participating.
Unfortunately contacts with company like Siemens were not successful but further contacts beyond the project are expected on technical specifications of future products.
After issuing the Functional Specifications and a White Paper which takes into account the key points during the first months of the Project, a lot of contacts in different exhibitions, conferences, direct call of people interested from India, USA, Canada, China have been seen. They all confirm the interest for the concept which is developed by IAQSense.
Except these confirmations, the subject was closed during the first period of the Project.
Methodology and Quality Plan EM; ALL PARTNERS
This task aims at defining the entire methodology to be used during the project and the set of indicators, such as availability of gas head samples, characterization of data log available and matching with theoretical models, availability of electronics and micro-modules, in-situ test reporting, package for industrial porting.
The Methodology for validation and approval of key documents and deliverables (review meetings or review procedures) has been defined and applied.
The Quality Plan was mainly defined for implementation of Quality systems within Start-up like Nano Analytik for process quality management. The other industrial side, Id-Mos has been checked for availability of ISO9000 certification which covers all procedures along the semiconductor engineering cycle. For Nano Analytik, after meetings in Ilmenau between EFFICIENCE Marketing (in charge of implementation of Quality plans) and Nano Analytik in February 2014, the whole process of implementation of the Quality System in production has been set-up with a dedicated person in charge. It applies to internal and external (subcontracting) procedures and purchasing actions (validated suppliers).
The quality system of Nano Analytik is not documented for public release.
The risk management is under direct responsibility of the Project Coordinator. It is especially critical in this project as some tasks are dependant and creating critical path in the project management.
Two such tasks have been disconnected from the critical path in order the insure parallel engineering work (Demo Boards and Tests in Buildings). These decisions have proven very effective as several boards and several remarkable reports on building applications have been delivered.
Deliverables (delivered)
Functional specification
Electrical, mechanical, environmental specifications for each technology – Target molecules (priority list and modelling list)
Methodology and Quality Plan
Project management – Milestones and review meetings – Risk management – Responsibilities
Technologies
Objectives
This Work Package 2 process development is focused on the manufacturing technology of the sensors and test structures for technology verification. It will require development of manufacturing sub-processes of: spectrometer-FET-transistor, transistors on the tip and piezo-resistive cantilever beams functionalised with nanotubes.
Working samples of the sensing head are delivered concurrently with a characterisation platform.
Description of work
The key specifications of the sensor should be reached during the work:
• sensitivity <100ppb for the spectrometer and selectivity between the molecules of the reference set,
• sensitivity to 1ppb with a set of 4 functionalization layers for the transistor on tip,
• sensitivity to ppt with a set of 2 functionalization layers for the cantilever with nanorods,
• 70eV ioniser with controlled plasma sample reaching the sensing layer.
Design of building blocks NANO, Partners involved: EM, ISL, MICRO, FACET
This task aims at designing building blocks for sensing head according to functional specifications (
The testing and characterization equipment, realized by FACET will be in line with the sensor design and geometry. The sensors are developed and produced by NANO up to the end of the project
The “Post Kick-Off meeting a “Design & Specification meeting”” (Sofia beginning Sept 2013) has been held in fact ahead of the kick-off meeting (17th Sept 2013).
It has defined the key orientations for the Spectrometer on Chip:
• Departure from circular structure as created by ISSP-BAS to have more freedom for form factor of a linear structure
• Ntype only development, also a departure from both N and P of the ISSP-BAS structure
• Integrated ionizer on chip using suspended filament
• On the known reliability problem, due to the electrolysis of electrodes in harsh environment nothing was decided, leaving this question to the designers
• A suitable CMOS process was expected to be available in Ilmenau thanks to cooperation contract between TU Ilmenau and its spin-off, Nano Analytik
For the Cantilever, the base process was already in place and the main task was the further functionalization and technology transfer from ISL (single die) to one shot functionalization of a full wafer.
During the course of the project several questions were raised and solved concerning the Spectrometer on Chip but increasing the work load on both process and batch processing and characterization work as well.
• The need for both P and Ntype of Spectrometer as signature of dynamics of ions is different for positive and negative ions which both coexist in the Tin Oxide layer. This question has been solved with additional batches (2 different N and P substrate to simplify the process).
• The conditions for ion dynamics on the surface of the Tin Oxide has been examined more deeply leading to higher layer resistivity to respect the so-called “Debye Length”. New batches are involved. The investigations were done theoretically, by modelling and by referring to pre-project studies indicating different categories of resistivity
• Explaining the difference between the modeling and the experimental results, especially a parasitic effect of ten folds increase of static current of the Spectrometer when biased for relatively long time has led to understanding the role of the porous Tin Oxide layer, with creation of H+ ions on the surface which through the porous layer penetrate inside the Silicon Dioxide layer to change deeply the “Work Function” of the SiO2. In addition, this phenomenon is slowing down the transistor by a factor 1000 blocking the interaction Tin Oxide / Channel with ion dynamics. This study has led to the introduction of a “Chemical Barrier” which will probably be introduced in a full, mass production process.
In addition, to solve the reliability problem this chemical barrier is mandatory
• The solution for the ionizer went through 3 steps:
o Design of the initial “filament” ionizer was not feasible, not “soft”, and not effective
o A second design has been decided using a “field ionization” using 200nm holes (technology derived from NASA mars lander). Designed and tested it has been proved to be unreliable at atmospheric, humid pressure
o On the end a proven technique of UVC LED surface ionization using the work function of the Tin Oxide has been chosen, taking the opportunity to use low cost UV LED sources developed for Curing by several companies (including LG). Compared to initial specification it is a 5.2eV (240nm) ionizer with soft ionization (molecules are not broken up)
All this needed work has led to much more work than expected with collaboration between Nano Analytik, EFFICIENCE Marketing specialists, Stefan Andreev team to identify and solve the questions raised through process trimming and characterization work much more extensive than expected.
Technology 1: Integrated Mobility Spectrometer
A novel and unique sensor system based on mobility, dynamics of ions and operates as spectrometer on chip.
Better knowledge of the technology’s intrinsic capability was the objective of the first run of fabrication starting as follows:
• define the structure able to reach the performance (all technical partners),
• proceed with device simulation (MICRO),
• prepare variants for DOE plan including intrinsic parameters and environmental ones.
(Key influencing parameters are the dimensions of (a) collection layer, (b) FET dimensions, (c) ionizer dimensions and accelerating electrode place, (d) the overall device cavity).
The relationship between target specifications and key parameter variation have be analysed thanks to the characterization platform prepared by FACET, an electrical test made by Stefan Andreev on Owlstone gas generator and the support of ISL for analysis in the gas generator.
The results were planned to provide:
• an understanding of the influence of key parameters through DOE analysis,
• a choice of optimal parameters,
• the influence of environmental parameters.
Comments:
The DOE plan can be done a fully stabilized version of the Spectrometer that can be reached within a full standard process, beyond the Project.
Technology-Type 2 Field effect transistor using Tip
The sensing principle has similarities with the sensor type 1, based on change of an electric field due to the adsorption of certain species on a porous layer but work is static only.
The basic principle of this sensor is a FET placed on a micro-scale tip to enhance the overall sensing sensitivity, operated in constant voltage mode and with the source-drain current changing with the amount of adsorbed species. The species are adsorbed by special functionalization coating which ensures selectivity.
Comments:
Technology -Type 3 (TT3) Nanostructured collectors on cantilevers for trace and ultra-trace sensing
Technology
The sensor works as nano-balance. The cantilever is an oscillating and capable of detecting atto mass-changes. The detection is based on reference, non-adsorbing device to be compared to the adsorbing one in terms of oscillating frequency. This way one chip should be composed of multiple different active layers/cantilevers and one reference using 4 places on the chip.
Development
The processing technology for this sensor-type is known by NANO. The coating process will be executed either by ISL on site, or the technology will be transferred to NANO in order to produce the entire sensor unit with one micro-machining process plan.
Production
The production of the sensor is based on standard SOI-technology.
The realization of the process of wafer coating needed two variants to reach a perfect homogeneous coating (etching after deposition).
It pays in return as the performances of one cantilever with different individual gases are very sensitive.
The question of selectivity when a mix of gas is reaching the cantilever and identification of gas types is still under investigation and will be performed with boards developed during the project. Two hypothesis:
• The speed of adsorption of each gas is different and an acquisition of data with a cycle inject/purge will create dynamic events which are representative of the molecules
• A network of 8 cantilevers with different functionalization (on top of the nanotube coatings) will be able to create a static pattern
• The boards developed during the project are able to investigate the two modes or a combination of the two thanks to 8 cantilever on board and 2 electrostatic pumps.
Production of sensor head samples NANO
This task aims at designing building blocks, Design for Manufacturing oriented, definition of packaging/filtering for different applications, setup of mask sets, design of the process steps for each block.
The principle initially described for the task has been followed:
1 - Batch design and production in Nano Analytik
2 – Electrical and first gas test of the samples in Sofia
3 – Test at low concentration in ISL
Several different different batches have followed this path and be tested thanks to the platform of task 2.3 below and gas generators (Owlstone in Sofia, Proprietary in ISL).
Preparation of the test environment and the platform for system integration FACET, NANO, MICRO
The test platform has 3 objectives:
• ensure the electrical test of the different components (technologies) and sub-components of the sensing head to release working heads for further characterisation,
• provide the characterisation platform to the different actors of the characterisation step. It will provide datalogging of the responses to different gases or molecules,
• the platform should also provide the necessary flexibility to install signal processing of the datalogs (batch processing) to prepare the future integrated real time algorithms.
The platform has been built around the National Instruments boards and Labview software platform.
The characterization itself and the interpretation of the results have been done under control of Nano Analytik (with link to the design) and Microsystems Ltd - Varna (with link to the model). FACET providing the necessary upgrade of the platform as needed for datalogging or signal processing.
The team of Prof Stefan Andreev has done the first tests (electrical and preliminary gas tests) on the different types of Spectrometers. He was supported by EFFICIENCE Marketing and FACET for debugging of the samples behaviour and adaptation of the platform hardware and software. See description of characterization and results.
Several technical meetings were held in Sofia for that purpose with technical meetings in Ilmenau to transfer the acquired knowledge to Nano Analytik design and production.
Process porting package reporting for mass production NANO
This task aims at preparing the portage of the process for the sensing heads mass production
Shifting to outside foundry.
While Nano Analytik has succeeded to make several batches of 2 different Spectrometer on Chip which were adding a lot to the initial knowledge on Spectrometer on Chip, some points have been considered important enough to insure porting the process to external foundry:
• the clean room used by NANO has a limited capability in terms of process layers, lead time and volume of production.
• In terms of economics a standard foundry provides hundred folds better cost,
This has lead to consultation of several foundries:
- Xfab (not ready at this time - 2015)
- Semefab (quote obtained)
- AMS via CMP plus post-processing as needed (quote obtained)
Which offer different degree of integration of CMOS+MEMS.
It has been considered favourably for the integration step beyond the project.
Description of boards
All boards are hardware and software compatible with RaspBerry PI-2 / PI-3.
The basic software (drivers, initialization, data acquisition) have been developed for Windows 10 - IoT installed on RaspBerry PI.
5 boards have been developed :
- 2 reference boards using components for sources external to IAQSense
o IAQCore based board for detection of TVOC and CO2
o MSS board using a network of 8 surface stress membranes
- 3 boards using IAQSense technologies:
o Spectrometer on Chip + processing ASIC
o Cantilever based network of 8 resonating cantilevers plus processing/actuating ASIC
o Cantilever based network of 8 resonating cantilevers plus FPGA based lock-in amplifier
The technologies developed by IAQSense are used on the boards:
- Spectrometer on Chip developed and produced by Nano Analytik (P and Ntype)
- ASIC IM452A for Spectrometer on Chip developed by Id-Mos (Xfab foundry)
- ASIC IM309D for Cantilevers network developed by Id-Mos (Xfab foundry)
- Optionally TIP transistor can be packaged for interfacing IM452A on the same board and realizing an enhanced Spectrometer or a field sensing probe (Kelvin probe)
The attachment of board to a Raspberry PI provides embedded solutions for:
- Drivers, Initialization and data acquisition
- Neural network or AI event identification
- Pattern recognition based on event libraries
- Firmware development, compilation, and downloading in case of IM452A ASIC with embedded µC.
The following solutions are recommended for knowledge processing and pattern recognition:
- MatLab and neural network toolbox
- Weka and machine learning algorithms (Random Forest and derivates)
Solutions for software/firmware development:
- MSP430 development and debug: GNU Compiler Collection (GCC) for Spectrometer ASIC firmware development
- Windows 10: Visual Studio 2015 for all RaspBerry PI drivers
IAQCore based board
Application Features
The Application is in charge of driving the HW to:
• Initialise display + IaqCore chip + leds + Buttons
• Launch a periodic timer to read the IaqCore data (5mn power-on then 1s rate of output data
• Red and Green LEDs display the state of the TVOC and CO2 trigger levels and blink at different speeds showing the change trend
• Buttons: Button 1 for new storage session – Button 2 clear all the stored files
• Display (on external screen)
o State of the sensor (ready or not) – Rvalue – TVOC and CO2 level
• Store the measures in a file in CSV format
o Data are stored on CSV in the FTP server of the board
MSS network board (8 stress membranes)
Features
8 channels MSS with input connectors for Silicon Membranes.
Each membrane can be changed, functionalized (DIY polymer deposition), accessible inside injection chamber.
• Each channel has a separated ADC and DAC for managing sensor bridge amplification, offset compensation and conversion (16 bits).
2 electrostatic pumps (on board supplied and driven) provide static or dynamic flow of analytes and purge for static pattern analysis or dynamic slopes analysis.
• One pump is driven to feed the pollutant, the other feed the fresh air alternatively
Application Features
The Application is in charge of driving the HW to:
• Initialise the different devices
o Chip selects for the different ADCs of the Membrane sensors
o Interrupts of the different ADCs (data ready)
o Driving of the pumps
• Define the default DAC values for each sensor
o Before launching a measure campaign, the MSS voltage offset is compensated.
• Launch a measurement campaign for each of the 8 sensors
• The global cycle is 60s split in 30s of pollutant feeding and 30s of fresh air feeding. (100k samples per MSS membrane)
• Data are stored in CSV format for further analysis in the ftp server
Spectrometer on Chip Demo board
Features
The DemoBoard supports 4 sensors
• 2 type P sensors + 1 type P sensor reference
• 2 type N sensors + 1 type N sensor reference
A processing ASIC (IM452A) drives the sensors, acquires the signal from each sensor and provides processing thanks to the embedded µC (MSP430)
An external Flash and RAM are added on the board to store program and one dataset.
Software
The software is prepared on RaspBerry PI and provides initialization, parameters setting and firmware downloading:
Initialization
• Initialize the acquisition parameters (sensor parameters, sequence and acquisition speed)
• Provides transparent mode (learning mode)
• Or Full acquisition and compression mode with pattern recognition
Dowloading
• Download firmware developed offline
• and database prepared on base on learning under gas generator
Principle of learning and pattern recognition
• Using Matlab Neural Toolbox or Weka machine learning, under gas generator with different set of gases. Extraction of key events
• Downloading reference events in IM452A for real-time pattern recognition
8 Cantilever network with ASIC detection
The demo board support 8 resonating cantilevers (100kHz).
An ASIC (IM309D) is acquiring, filtering and detecting the 8 signals form each cantilever in parallel. It provides the driving of each cantilever with separate driving sine waves on the individual resonating frequency (thermal actuator).
A set of 8 ADC and DAC insure dataset acquisition and analogue settings while a set of 8 function generators provide the sine wave for the actuators.
Each cantilever is delivered with surface coating with CuO nanotubes for extended surface and sensitivity and are individually accessible for further individual functionalisation (additional surface coating). It provides very sensitive detection and selectivity at ppb levels of concentrations.
2 electrostatic pumps are providing injection/purge for dynamic detection of events due to molecules on top of the static detection providing means for further selectivity and uncertainty lifting.
Software
Provides from RaspBerry PI:
• The search for resonating frequency on each cantilever and further setting of function generators.
• Initialization and data acquisition
Principle of learning and pattern recognition
• Using Matlab Neural Toolbox or Weka machine learning, under gas generator with different set of gases. Extraction of key events in static mode or synchronised dynamic mode
8 Cantilever network with Lock-in amplifier
separate board has been developed using digital lock-in amplifier to provide further improvement of S/N ratio for below ppb detection.
It uses the same set of 100kHz coated cantilevers that can be further functionalized.
It serves, with the other cantilever as a base for further study of CMOS integration and very low concentration detections.
Description of components
Spectrometer on Chip:
- An Ntype spectrometer
o in DIL14
o comprises a 100x100µm Ntype sensing device with Tin Oxide as sensing layer and a reference transistor
o Compatible to interface ASIC IM452A
- An Ptype spectrometer
o in DIL14
o comprises a 100x300µm Ptype sensing device with Tin Oxide as sensing layer and a reference transistor
o Compatible to interface ASIC IM452A
- These spectrometer can be further functionalized for gas spectrum extension with different types of polymers (available with deposition equipment in Chemical lab Sofia)
ASIC 452A : Dedicated to Spectrometer on Chip
o 4 simultaneous drive and acquisition channels for Spectromers
o 2 multiplexed channels for P and N reference transistors
o UV LED booster and drive
o 200ksps acquisition rate (adjustable) with 12 fast ADC plus offset DAC setting.
o MSP430 embedded processor (16 bits) for real-time processing
o Uploadable software for real-time processing:
- Spectrometer sequence and activation of drivers
- Denoising and compression
- Event and pattern recognition according to uploaded knowledge base
- Transparent mode for external machine learning
Description of software tools
- Drivers for all boards
- Development and debug tools for MSP430 boards
o TI tool for MSP430 board for cantilever network
o Open sources tool trimmed to ASIC IM452A embedded MSP430 and peripherals
- Matlab and toolbox
o Specific programme for machine learning of Spectrometer on PC-Matlab
o Specific programme for machine learning of Spectrometer version RaspBerry PI-Matlab/Simulink/Toolbox
o Event knowledge base for Spectrometer
- Demo software for MSP430 based 8-Cantilever board
Documentation provided
- White Paper
- Datasheets of components
- Detailed description of ASIC IM452A (on request)
- Datasheets of boards and connection to RaspBerry PI
- Report on optimum position of VOC sensors in buildings (on request)
Work performed on Building integration and Control algorithms
Objectives:
Installation of miniaturized sensors developed in IAQSense project in given environment able to recreate a real indoor environment such as a dwelling both in the CEA and ACCIONA facilities.
Test of these sensors close to a real indoor environment with a possible indoor control such as temperature, air change rate, level of pollutants, etc....
Focus on the best positioning of the miniaturized sensors at different scale (building, room) and location (centre of the room, on the walls, etc....).
The objective of this work was to find some appropriate position for IAQ Sensors. For that, a bibliographical review was first done with a lot of recommendation found at a room scale. Then some tests of different position of sensors were run in an experimental house, at a room scale first and at the building size in a second time.
The tests done at the room scale confirmed and completed the recommendation found in the literature: The position of the sensors can be on the walls not particularly in the centre of the room. They can be positioned at a height corresponding to the breathing zone (0.5m-1.7m). Avoid positioning the sensors close to the windows, the corners, the doors, the heating or cooling systems, the solar task, the forced convection/draught, pollutant sources & sinks.
At the building scale, the strategy will depend on the number of sensors available. Starting from the “ideal solution” with no limitation in terms of costs, architectural constraints ... it is recommended to position one sensor per room, or at least in each area were we spent the most of the time (living zone). At the opposite, when only one sensor can be set, the transit zone might be the more representative of the indoor pollution and so the sensor can be positioned in this area. This area being between the air inlet and the exhaust air. In this configuration, the relevance of the measurements coupled with the reactivity of the pollution detection will strongly depend on the building configuration (geometry, doors open or closed, ventilation ...). In the case of multi-storey building, an intermediate solution could be to have one sensor per storey which could be positioned either in the living zone if this is a sort of open space, or in a transit zone if there are plenty of rooms.
A summary was proposed to provide some recommendations for the positioning of sensors at different scales as follows:
“An adequate location of the sensors is very important because they measure the parameters that will manage the equipment needed to improve the indoor air quality.
Theoretically, the optimum location is in the centre of the room but this is not a practical place to introduce sensors. After the installation sensors in 4 points of the room, the conclusions regarding the most realistic locations are:
- The thermo-hygrometric conditions change a lot in a room mainly with the height due to temperature gradient made by the fluctuations in the air density. To measure the temperature and humidity at the middle of a wall (with the appropriate insulation and R-value) is enough to obtain similar values than the inhabitants can perceive.
- The air pollutant composition (VOC and CO2) should be measured at the breathable height to know the concentration of those pollutants that can be breathed. Using the same location than before, it is able to have representative values but it is important to be sure that the sensor is not joined to an emission source.”
A detailed report on sensor positioning and findings is available on request.
Optimization of the IAQ via a regulation system
Focus on the VOC sensor performance through management strategies of HVAC systems.
The management strategy will consist in increasing the indoor air quality (IAQ) while minimizing the energy consumption.
This management is a control algorithm that has be developed previously, consequently there is a necessity to define the inputs of this algorithm (VOC, CO2, temperature, and so on...) and the outputs (I.e.: HVAC system) to reach a valuable performance both in IAQ and energy efficiency.
The major goal of the IAQsense project was to develop a new IAQ sensor based on nanotechnologies and able to monitor several major pollutants in the indoor air. The first task consisted in defining an acceptable positioning of IAQ sensors in an indoor environment. Then, after applying these recommendations of positioning, the goal of the next task was to develop a new control command solution (CCS), using the data delivered by the sensor in order to improve the indoor air quality in regards with a low energy consumption.
A control command solution which can value the data coming from the IAQSense sensors has been developed and tested in this task by CEA. This control command solution has the objective to tackle both IAQ improvement and energy consumption reduction. To develop and test the efficiency of our control command solution (CCS), two parallel steps were proposed in this study with on one hand a numerical approach and on the other an experimental step. Both have been tested at a room scale.
In the numerical approach, the building simulation tool Energy+ was used to simulate both energy consumption and IAQ in one room. It has been coupled to Matlab through the platform ‘BCVTB’ in order to develop, test and optimize the Control Command Strategy. This numerical part has shown the great ability of a good ventilation strategy to reach both energy efficiency and indoor air quality objectives.
The second step allowed to test at real scale the algorithm developed through simulation. The tests were carried out in our experimental platform located in Le Bourget du lac, at INES. The algorithm has been implemented in one room of an experimental house. It controls the mechanical ventilation system depending on two parameters values (the carbon dioxide, and the total volatile organic compounds).
The results are very interesting because the algorithm has the expected behaviour both in the simulations and in the measurement campaigns. It is able to regulate the IAQ thanks to essentials parameters: the carbon dioxide and the TVOC. The first pollutant is an indicator of the air exchange rate and of the occupancy rate. It permits to contribute to a good ventilation. The second one is a global estimation/’picture’ of the chemical pollutants in a room. Thanks to specific concentration thresholds based on health sanitary criterion, the algorithm turned ‘on’ or ‘off’ the mechanical ventilation system (MVS).
This mechanical ventilation was the actuator chosen to improve both IAQ and EE. The ventilation was modulated preventing the pollutants from going over these sanitary values in regards with a low energy consumption. The numerical approach has been implemented and has shown the great ability of a good ventilation strategy to reach both an energy efficiency and an indoor air quality objective. The ventilation strategy tested in the numerical model with the chosen configuration shows its ability to reach the IAQ objective and at the same time to maintain an energy consumption either smaller or at least in the same order of magnitude as the usual ventilation strategies from 0,3 to 0,5 h-1. In the experimental part, the first tests done allowed to ensure that the CCS reacts correctly regarding the algorithm implemented at real scale. Thus, these results showed a great opportunity to promote more the IAQ by arguing its compatibility with the energy efficiency.
The work done in this task will help our partners to provide an additional function to the sensor developed. It could contribute to improve the IAQ without degrading the energy consumption in buildings thanks this control command solution
Test of sensing heads in gas or molecule environments - analysis of patterned data log with algorithms
Summary
Several batches of spectrometer MOSFET sensors are produced in an iterative manner to achieve the expected response to the gas molecules.
The experimental work for spectrometer MOSFET sensors was organized by:
- Establishing identical measurement equipment at producer (nano), sensor behavior and descriptors identification lab (Sofia lab) and final experimental lab for sensitivity measurements (ISL).
- Setting the SVN server for data collection was established at uS.
- Organizing the training workshop for the NI platform users.
- Creating a plan and procedure for distribution of the sensor batches.
- Systematic performing the experiments and collecting data for the task 3.2
The sensor response to the gases is relevant to the expected behavior derived from the simulation model. Based on the first experiments, the simulation model was improved.
Based on the first experiments, a feedback to the producer was provided and the final experiment conditions and regimes was specified for systematic testing.
The experimental work with the nano-functionalized cantilevers was continued during the 2nd period. Improved nano-functionalized cantilevers and detection electronics were used.
The experiments confirmed extremely high sensitivity of nano-functionalized cantilevers to some explosive materials as well as to selected number of analytes of interest.
Deliverables
“Report on characterization environment / parameters and results” was delivered in Period 1 based on the experiments with first sensor batches 1 and 2. The results obtained at ISL were reported in June 2016 as an amendment of that report.
Outcomes
Spectrometer on chip based on MOSFET sensor
NANO developed and produced working MOS structures sensitive to chemical substances.
SOFIA LAB performed more detailed experiments, proving possibility for discrimination by type and concentration. Proof of concept of using descriptors and pattern recognition was done. Important outcome of this task is the establishment of SOFIA LAB as a permanently working entity which will provide the sustainability of the project results in the after-project period.
ISL LAB performed experiments with concentrations below exposure limits, proof of higher sensitivity was done.
MICRO developed all needed MATLAB tools for analysis of the experiment results. Based on the experimental data, considerable set of descriptors were derived for the following pattern recognition model.
Nano-functionalized cantilevers
A very high-sensitive detection system based on nano-functionalized cantilevers was developed and tested in gas environment with extremely low-level of analytes of interest.
The system proved very high sensitivity for explosive materials. The results allow direct practical exploitation in a commercial product.
Modelling
Summary
The algorithms for descriptors extraction and pattern recognition were developed on MATLAB first, and validated with the experiment data.
The software implementation was done by creating a transportable C/C++ code which was implemented on a common contemporary microcontroller MSP432xx. The software code was validated using a dedicated platform.
The used Artificial Neural Network (ANN) allow software engine to be updated for recognition of new data (if systematically gathered).
The Software part of the Spectrometer on Chip was proven for functionality and implementation in the further industrial product.
Deliverables
“Pre-defined patterns and test comparison report” – delivered in March 2016
“Software for customization of the gas spectrum detection” – delivered in March 2016
Outcomes
Fully functional recognition engine based on ANN is developed and validated.
MATLAB scripts for ANN generation and training as well as exporting the C++ source files of the net.
Transportable C++ code for the computation of the descriptors and pattern recognition engine is available.
The principle of the developed tools and software architecture allows easily upgrading of the software on the client side just by importing the updated ANN coefficients after each next training. The latest makes the software very flexible and convenient with respect to the usage of the end-product.
The software part of the spectrometer on chip product is ready for future commercialization.
Development of innovative technologies for functionalization of cantilever surface at wafer scale
Summary
The nanostructuration protocol was developed at ISL before the project IAQSense, and transferred to piezoresistive cantilevers during the project. A close cooperation between ISL and Nanoanalytik have permitted to define which way the nanostructuration could be performed on piezoresistive cantilevers without damaging the electronic transducing part of these cantilevers. The protocol for nanostructuration on large wafer scale was defined and optimized at ISL with the aim to allow the fabrication of hundreds of nanostructured cantilevers in one same batch, and showing similar characteristics in terms of resonance frequency and physical properties. Once optimized the protocol was transferred to Nanoanalytik fort including the synthesis in the process of cantilever fabrication
Deliverables
“Technological report on the nanostructuration of collecting layers on cantilever surfaces: Technology and process description of nanostructuration of specific layers on cantilever based sensor” as been delivered mid-June 2016.
Dissemination
Internal and public website (FSRM, EM)
Publication on www.iaqsense.eu of
- 2 IAQSense Flyers
- 3 IAQSense Presentations
- 3 IAQSense Newsletters
- 2 IAQSense Events
- 5 IAQSense Highlight documents
- 6 IAQSense Posters
Dissemination
- Dissemination plan updated in March 2016 (FSRM). Final version delivered M36 (August 2016, FSRM).
- Organization of an internal workshop for the project parnter in Ilmenau (DE), July 2015 (nano, EM).
Content:
System training with the new measurement setup for characterization of chip transistor spectrometer.
Participants :
S. Andreev (microsistemi) I. Atanasov (nanoanalytik), C. Iroulart (EM), T. Ivanov (Facet), M. Holz (nanoanalytik) , P. Popov (Facet), N. Nikolov (microsistemi), I. Rangelow (TU Ilmenau).
- Organization of 3 public workshops to disseminate project results (FSRM)
23-6-2015 – Workshop with Partners and Technical Committee – Paris
12-05-2016 – Sensor&Test, Nurembers – Special Session “Novel sensor system for Indoor Air Quality control”
07-07-2016 – Indoor Air, Gent – presentation “Changing the game in the management of Indoor Air Quality” and “Real-time monitoring for improved health, comfort and energy efficiency”.
- Publication of 3 newsletters (FSRM, with inputs of all partners)
Aim: to disseminate public project results
Target group: colleagues, business contacts, customers, press, clusters, projects, the wider public.
Distribution: sent by e-mail to 250 addresses and published on www.iaqsense.eu
Issue I: April 2016
Content :
• The IAQ project
• The Integrated Spectrometer on Chip and the Processing ASIC
• CANT 2005 –8 resonating cantilever network
• IM309xx – Signal processing seriesfor resonatingcantilever (8 channels)
• IAQSense at Sensor&Test Nürnberg
• Common Booth IAQSense-PAMIAQ projects
Issue II: June 2016
Content :
• Software elaborated for the customization of the gas spectrum detection
• Functional Cantilevers
• IAQSense at nanoFIS, Graz (AT), June 27-29, 2016
• IAQSense at Indoor Air 2016, Ghent (NL), July 3-8, 2016
Issue III: August 2016
Content:
• Project End
• Five Demo Boards
• Spin-off company IAQSense Ltd created for further project exploitation
- New IAQSense Project Flyer (FSRM)
Aim: to give general information on the project, objectives, expected results, partners
Target group: scientific community, EC, industry, students, general public
Distribution: by all partners to their contacts and at exhibitions/congresses where their institution was present, at the IAQSense booth and IAQSense workshops.
Download: www.iaqsense.eu/publications
- Flyer « Spectrometer on Chip » (FSRM, EM, nano analytic, ID-mos)
Aim: Information on the technical details of the spectrometer on chip
Target group: Industry
Distribution:Technical Committee, customers
Download : www.iaqsense.eu/publications
- 23 presentations, papers, posters in 10 countries. 1 booth (Organisation: FSRM. Content: ALL)
01/2014: CMEES Cluster, Sofia
05/2014: Visit IMEC and meeting with Bulgarian authorities
06/2014: ECTP, Special session “Indoor Environment Quality
09/2014: CMEES Cluster, Sofia
05/2015: COST Action EuNetAir meeting
05/2015: Meeting Technical Committee
09/2015: Common Meeting Sensindoor/IAQSense
11/2015: NIDays, sensor measurement system
12/2015: COST Action EuNetAir meeting
03/2016: Nanoscale Conference
04/2016: Smart Cities, Stand Eureka Euripides
05/2016: Sensor + Test
07/2016: Indoor Air 2016
07/2016: NanoFIS
Deliverables: Dissemination Plan - FSRM - Dissemination Activities – FSRM
White paper on the project (EM)
Task finished in M3. Published on website
Cooperation with related European RTD Initiatives (EM)
Objectives for the period:
- Cooperation with related projects and clusters
- To realize a common workshop and a booth with project SensIndoor
Results:
• EuNet Air (Cost Action)
• Poster of M. Holz, nanoanalytik, in Cluster meeting May 2015, Lynköping
• Posters of N. Nikolov, microsistemi, and M. Holz in Cluster meeting Dec 2015, Sofia
• Cluster Modelling and Characterization
• Participation of M. Holz, nanoanalytik, in cluster meeting, Nov 2015, Brussels
• Eureka
• Poster of I. Atanasov, nanoanalytik, in Smart Cities Innovation Event, April 2016, Stockholm
• Collaboration with EC funded project SensIndoor:
• Meeting to discuss possible common actions (June 2015, Freiburg DE)
• Several telcos to organize the events
• Common booth at Sensor&Test, Nuremberg (May 2016)
• Joint session at Sensor&Test, Nuremberg (May 2016)
• Joint session at IndoorAir 2016, Ghent (July 2016)
Links to all related clusters and projects: http://www.iaqsense.eu
Deliverables for the period :
Cooperation Results with other European RTD Initiatives (EM: M36)
IPR and exploitation of results (nano)
Objectives:
- To organize an exploitation workshop with a moderator from the EC
- To collect the claims of the partners on the project foreground
- To inform the Technical Committee about the project results and to collect feedback
- To elaborate the final IPR and exploitation plan
Results:
- Exploitation workshop with a moderator from the EC organized M19 (March 2015) by FSRM and EM in Brussels.
- Tables with background, foreground, exploitable results and claims filled in by all partners (EM). More details: see D8.36 IPR and Exploitation Plan.
- Workshop with the Technical Committee organized June 2015 in Paris (FSRM, EM)
Content:
• 6 presentations on IAQSense project and results by partners EM, nano analytik, Id-mos and Eeleo.
• Feedback by the Technical Committee and discussion.
Participants :
• 7 partners from the project: EELEO, nano analytik, EM, ID-MOS, FSRM and
• 8 members of the Technical Committee from international companies : AERECO, SCHNEIDER ELECTRIC
• CIAT, EXPLORAIR, BOUYGUES.
By agreement of all partners, a spin-off company named IAQSense Ltd has been created in the end of the project to assure continued use of the project results. Its strong international team is in charge of CMOS and MEMS integration, mass production, test and calibration. IAQSense Ltd will follow the trends towards IoT and pervasive sensing in the development of a complete range of environmental sensors for indoor, outdoor and wearables.
- Details: see IPR and Exploitation Plan.
Deliverables:
IPR and exploitation plan proposal (EM, M36)
The exploitation of the results are managed through a spin-off company regrouping main disciplines for development of products and services and introduction on the market.
Components are available
Boards are available and should be further characterized, optimized for commercial delivery
Medium term CMOS integration of Sensors (Spectrometer, MSS, Cantilevers) should be organized beyond the project to reach small factor application markets (wearables) and mass produced sensors for Buildings, vehicles, pervasive security networks.
Creation of a Spin-off – End of August 2016
Activities
- Chemical Lab
- Development
- Integration
- Functionalization
- Commercial
In charge of development of current and future product
Components
- Production Spectrometer
- Production ASIC IM452A
- Production Cantilevers
- CMOS Integrated Spectrometer + Processing
- CMOS + MEMS Cantilever + Processing
- CMOS + MEMS MSS
Boards
IoT / RaspBerry PI compatible
- 8-MSS Boards (VOC and threat detection)
- 8-Cantilever boards (VOC and Threat detection)
- 2-Cantilever board (PM detection)
- TIP based Field mapping
- 4-Spectrometer on Chip board (VOC)
Chemical Lab service – Accredited ISO 17025
Functionalization service (on INERA Equipment and Polymer service)
Electronics and AI software support
Development of Wearable
CAD for CMOS + MEMS
Commercial:
- Exploitation of contacts gained during IAQSense Project
- Going on with Technical Committee members
Chemical lab description
The Chemical Lab is built using the Assets acquired by the IAQSense Project.
The key equipment is the gas generator (Owlstone) which provides defined concentration of gases using permeation tubes.
A set of permeation tubes have been acquired during the project covering the key gases identified for VOC detection in building. Some short life time tubes should be acquired at the time of a measurement session.
The objective of a post project installation of the Chemical Lab is to organise a Lab according to the ISO 17025 quality standard for measuring lab by having all procedures, quality system and safety conditions accordingly.
The accredited lab should be able to deliver all services for internal development step and external services to gas sensors and gas detectors manufacturer.
The type of planned services:
- Characterization service for new sensors
- Test, measurement and calibration for production of sensors
- Recalibration of gas detectors
One of the capacity of the lab is application of ISO 16000-29 for TVOC and specific gas set sensors.
The lab is installed in Sofia. A lab manager, chemist, gas sensing specialist and with knowledge on ISO 17025 has been hired for this operation.
Design of analogue interface FACET, ID-MOS
Objectives : The objectives is to Design of sensor head interface with processing electronics, focus on low power consumption, sleep mode for the first sensor based on a spectrometer principle. By heating a wire filament, the electrons are collected by the trap electrode. Depending on each molecule type, the voltage level of this electrode can reach 70V. The spectrometer itself is based on a floating gate over a FET. After stabilization, the ions collected by the gate are moving from one side to the other at different speeds by alternating the drain/source positively and negatively. Depending on their physical properties the gate potential will modulate the current in the channel. During the operating time window, in the range of 10μs, the analogue electronic has to provide the high voltage (10V – 70V), enough current for the sensitive device (5mA) and to detect the output current. Then the signal is amplified, filtered and converted dynamically. (ADC 16bits with 1Mhz sampling).
Main functions :
• I to V converter and programmable Amplifier
• STEP-UP DC-DC Converter
• SENSOR Bridge Driver
• LED Driver
• Temperature sensor amplifier
• ADC / DAC
• BLOCK Diagram
Results : Block Diagram / Layout
Deliverables
• Analogue Specification
Digital Architecture and design of pattern processing function on-chip (ID MOS, FACET, NANO)
Objectives : The digital part of the future one chip solution will be designed by ID-MOS keeping in mind the interface with the analogue block and future choice of compatible technology for the one chip solution.
The output signals coming from the analogue block will be converted with an ADC to:
- Identify the status of the sensing head and to feedback conditioning control signals,
- Process the incoming signal from the different parts of the sensing head.
The precision of the ADC will ensure precision and selectivity of the gas spectrum, it is a critical part of this design.
For this task of digital signal processing the algorithms used during the characterisation phase of the sensing head will be examined for their portability on-chip.
Results : Methodology/Block Diagram/ Layout
The industry standard TOP-DOWN methodology was employed to design the circuit.
The design is based on RTL VHDL modelisation.
RTL models and final the netlist are validated using a single or several test bench(es) featuring automatic checks. The layout was obtained by automatic Place and Route, and checked by DRC /LVS tools.
Integration results of the digital part: 23 000 equivalent Nand Gates – 9.86 mm2 total area in 0.35um technology
Deliverables
• Pre-defined pattern processing arch. Digital part Specifications
Global Integrated Circuit (IC) - Global architecture / foundry - Test architecture - Characterisation of the IC - ID MOS, FACET, NANO
Objectives :
The analogue and digital blocks developed separately is merged in one chip defining:
- the global architecture of the chip
- the test architecture
- the global layout for tape-out (data for production in subcontracted foundry).
ID-MOS is managing the subcontracting for semiconductor foundry and all related engineering tasks.
The external foundry will provide wafers ready for probing (test of individual chips directly on wafer).
After packaging the final test and the characterisation at chip level is done by FACET.
In this way ID-MOS manage the complete design and production of the IC product cycle and provides ready tested chips.
Results : Block Diagram / Layout/ Package / Silicium
Deliverables
• ASIC : Design Summary Report (DSR) - Layout Summary Report (LSR)
• First silicon delivery
• TOP Circuit specification - DSP specification
The ASIC coming from foundry as wafer was packaged in August 2106 and made available for the Spectrometer demo board. Currently the board is under characterization.
Preparing micro module (Combining Sensor, IC and pattern recognition software, Harvesting, RF) NANO, EELEO
Objectives : Deliver demonstration boards and development kits for the key technologies Spectrometer on Chip and Cantilevers
Results : 5 boards have been delivered for evaluation and future optimisation for introduction on the market as IoT compatible boards
Monitoring system - Integration of the sensor system in in-situ monitoring system. NANO, ACCIONA
Objectives : Use of VOC sensors for monitoring in buildings
Results : suitable VOC sensors have been delivered for investigations in buildings during 2015-2016 period.
Potential Impact:
Impacts
Socio-economic impacts
Right from the beginning of the project several potential applications of “real-time VOC sensing” were identified, starting from the priority target of providing solutions for buildings.
The potential market for buildings (home, offices, public buildings) and vehicles have been estimated as TAM, SAM and SOM for the different segments while targeting a final product at 30€ pricing.
These hypothesis have been confirmed and the target price as well through further investigations on the market trends.
In addition impact of new trends like IoT market development has been estimated with the possibility to enter the IoT market through compatible boards.
The building related market has been checked and confirmed through the Technical Committee attached to the project and involving professionals in building construction, ventilation and building automation.
Furthermore the security market and the medical ones have been checked, with a particular interest in providing wearable solutions for the occupational health and allergy/asthma prevention.
In buildings:
- Health and Comfort, Sick building syndrome
The prevention of health problems related to carcinogens is one impact which can be linked economically to the cost of medical treatments.
The effect on Comfort, work efficiency, absenteeism and feelings at work is largely connected to the Sick building syndrome due to the air tight new constructions (at the origin of the Project). Several studies in Europe and USA have estimated the huge economic impact of such situation.
- Energy consumption
The same “air tight constructions” being done to reduce the energy consumption in buildings, the question of optimisation of energy versus pollution (ventilation) has been raised.
The rules for design of new low energy buildings is to insure a ventilation level (air renewal) which is enough to insure average pollution removal and low enough to keep the energy. This is a compromise that is too low when there is normal human presence and too high when they are absent. This valid, at different time of the day, for residential buildings or offices.
The studies made by the two project Partners, CEA and ACCIONA have demonstrated that with good sensing distribution and suitable algorithms an optimisation of the pollution level at any time and lowering of the energy consumption is possible with significant improvement.
Medical
- Allergy and asthma
Having in mind the very large number of people (Millions in each country) affected by allergy and Asthma, the possibility to detect the environmental quality of the premises where you go is major improvement. Two cases of growing interest have been discussed with specialists in Europe and USA:
• The occupational health with impacts over decades (see ERS organisation – European Respiratory Society)
• The growing usage of flagrance in the public places creating high risks of allergy (USA) and general topic of asthma. We are currently talking with Doctors specialized in allergy in the USA to create a suitable wearable.
Our plan for product introduction
The high interest for sensors at the right level of sensitivity and selectivity and the growing trend towards IoT and pervasive sensing has created an huge opportunity for the technologies developed in the IAQSense project.
- A reasonable Turn-over and Business Plan has been setup for components and boards (see separate Business Plan)
- A structure which is product development and business oriented is the counter part of these opportunities. This is the reason for the creation of a “Spin-off” Company.
- A road map for product introduction is considering the constraints:
o Mature products to be introduced first (polymer customised networks)
o Products with limited development in the second place
o Products with high level of integration and complexity will need more time (wearables)
Societal implications of the project so far
- Protection against chemical accidents or chemical attacks
Growth of chemical industry and frequent accidents (leakage, explosions) is rising the question of preventive actions to protect the nearby population.
The growth of terrorist actions is connected to preventive detection of chemicals and explosives.
In general protection of key assets in modern society is critical for economic survival.
The answer lies in pervasive sensing in industry, public areas with IoT connections. Network of sensors with suitable costs (low) and performances (high) are required. This is one of the topics in SEC H2020 calls IAQSense partners missed in 2016.
Market studies show two trends:
• More sophisticated stationary sensors at the right cost to create pervasive, real-time communicating networks of sensors
• “Body” Integration with wearables and smartphone integration
- Comfort in general and intellectual efficiency
The three factors we mentioned, growing pollution and growing allergy in the population, introduction of new chemicals in new materials or directly in the air (flagrance), trends to confinement in buildings are subjects to fight with introduction of smart sensors
- Management of growing “smart cities”
The trend for higher and higher growth of urban population is creating new challenges and new needs for which the following topics apply
o Smart sensing network
o Citizen data
o Pervasive sensing
o Providing relevant information on the IoT channel
In addition the citizen is on the centre of a distributed network of information for himself and for all the population of the City.
- Self-awareness of pollution is required to take decision
o Information network on pollution
o Wearables collecting, displaying, delivering pollution information
Dissemination and exploitation plan
The dissemination of the project results has been done through the following actions:
- Creation of a website (www.iaqsense.eu) to inform the public about IAQSense actions. This site has published :
o A white paper about VOC detection objectives, real-time operation and technologies
o News about presence at different exhibitions
o Highlights on project results and products datasheets
- Press release and interviews:
o Has led to several publications worldwide
- Newsletters informing a large set of professionals about the development steps and the final results of IAQSense project
- Participation in Clusters
- Participations in several dedicated conferences and exhibitions
- Communicating with the Technical committee
- Exchanging with other projects (existing or created on purpose)
- Face to face exchange with current and new contacts for development and commercial purpose
The dissemination continues through the exploitation plan:
- Creation of a Spin-off called IAQSense Ltd (done in beginning Sept 2016)
- Protection of Brand name and creation of commercial websites
- Development of products and services around components and boards
- Initializing commercial and distribution actions
- Extension of product range features through CMOS integration and Functionalization techniques
The results exploited through the above structure are:
- Chemical lab (ISO 17025 accredited) as support for the development and external services for characterization, test and production support of gas sensors
- CMOS integration of technologies developed along the project and in collaboration (6 type of concerned components)
- Coating and Functionalization techniques
- Machine learning and pattern recognition techniques
List of Websites:
www.iaqsense.eu
Contacts:
• Stefan Andreev - Manager of IAQSense Ltd - andreev.stefan@abv.bg
• Mathias Holz - m.holz@nanoanalytik.de
• Paul-Emile Latimier - paul-emile.latimier@id-mos.com
• Claude Iroulart - claude.iroulart@efficience-marketing.com
• Jacques Montes - jacques.montes@eeleo.com
Under construction:
Commercial website: www.iaqsense.com and national versions www.iaqsense.fr www.iaqsense.de
Indoor air quality is a crucial factor for the human’s apperception. It is proven that air quality affects the health and the well-being of people. Up to now, only expert lab systems could measure air quality.
IAQSense Project has managed the R&D of new nanotechnology based sensor systems, completely unique solutions for miniaturized, low cost and mass produced sensors.
The main technology innovation is based on surface ion mobility dynamics, it separates each gas component and its concentration like a mass spectrometer and allows high sensitivity (ppb) fast (1s) multi-gas detection.
The IAQSense Project demonstrated that such a device can characterize, monitor and improve indoor air quality, in a way for which no affordable sensor exists. To enlarge the field of detection, the sensor system includes other patented biomolecule detection and security related nanotechnologies .
• The IAQsense objectives were to develop sensor systems, including all electronics, pattern detection firmware and validate these systems and their impact in real environments, firstly in buildings, but also show the potential in vehicles and for health and hazard scenarios.
• High-potential markets have been discovered and validated with potential users, product range has been defined for Building automation and IoT markets.
• Validation results with VOC generators and real building environments have confirmed the huge impact of real-time VOC monitoring for comfort and energy saving while guiding launching priorities.
• Ability for mass production and low cost were mandatory requirements for the development and silicon foundries have been contacted for steps beyond the project.
• The development includes a highly integrated IC for real-time gas pattern processing with special architecture for very low power and energy management; test and validation; predictive numerical models developed in the project; and implementation into buildings to identify and model the impact of this disruptive technology, before industrialization.
• The deliverables of IAQSense are nanotechnology embedded sensor systems, with on-board knowledge-based smart sensing. It creates a fast, low cost, high resolution system, with high added value, ready for mass production.
By developing 3 technologies, a range of products, 4 components and 5 different IoT compatible sensing boards, IAQSense brings breakthrough solutions close to the market for volatile molecule sensing in buildings, chemical threats detection, medical applications. Recognized as such by the Consortium Partners and market operators it has favoured the creation of a Spin-off Company managing the interest of the Partners. It provides a one stop shop team for further development, integration, customization/functionalization and managing commercial exploitation.
IAQSense results will have the following impact:
• provide solutions for real-time analysis and control of air quality on every place, home, office, vehicles,
• participate in the optimum reduction in energy consumption,
• participate in IoT and pervasive sensing trends and
• paves the way for integration into wearable devices for applications for health and security.
The consortium is composed of 4 SMEs, 3 industrial, 3 research institutes gathering all patent owners, specialists in marketing, project management and European industrial leaders in nanotechnologies & microelectronics. The Technical Committee are users of this innovative technology in the construction area. Dissemination actions has provided more contacts in automotive, medical and security markets while generating further opportunities for breakthrough projects.
Project Context and Objectives:
S/T quality, relevant to the topics addressed by the call
CONCEPT AND OBJECTIVES
Context and current challenges
For the last 20 years, there has been a growing concern regarding pollutants in closed environments and the difficulty in identifying these pollutants and their critical levels, without heavy, expensive equipment. Currently, a critical question to be addressed is the new regulation for ‘air tightness’ of energy efficient buildings, new or renovated, and the growing number of new materials in construction.
“Indoor Air Quality in Office Buildings: A Technical Guide” - Health Canada 1995 at this time indicated that “Several thousand chemicals, synthetic and natural, are Volatile Organic Compounds (VOC). Over 900 have been identified in indoor air, with over 250 recorded at concentrations higher than 1 ppb.”
Further this report also indicates that “indoor levels may be 10 times higher than those outdoors, in range of few mg/m3. Identification and measurement of individual VOCs are expensive and time-consuming, and the total cost is underestimated as the VOCs, at very low concentrations, are very difficult to identify.”
Research shows that “VOCs at concentrations below threshold limit values can cause discomfort, including fatigue, headaches, drowsiness, dizziness, weakness, joint pains, peripheral numbness or tingling, euphoria, tightness in the chest, unsteadiness, blurred vision, and skin and/or eye irritation. In the exposure range of 0.3-3 mg/m3, odours, irritation, and discomfort may appear, above 3 mg/m3 complaints, above 25 mg/m3, temporary discomfort and respiratory. Office levels cover all ranges. And hypersensitive individuals can have severe reactions at very low concentrations of VOCs emitted by their environment.”
In the EU, European Collaborative Action has published 26 reports (1996 – 2007) connected to IAQ.
Report n°19 indicates 63 VOCs that may commonly be found indoor and makes a recommendation to “Develop and evaluate improved strategies and systems to control ventilation rates, including demand-controlled systems using pollutant sensors for Indoor and Outdoor air quality and more accurate measurement of air flow within the system”. Report n°20 shows the relationship between levels, odours and comfort.
“Report on Carcinogens 12th Edition 2011” – USA lists 250 carcinogens and their threshold limits.
“Valeurs guides de qualité d’air intérieur (AFSSET (now ANSES)/CSTB 2007)” - France lists 222 VOC, close to the AgBB German standard. ANSES has also published a list of guidelines with health threshold values.
Furthermore, the French Centre for the IAQ (OQAI) has made a classification of 58 pollutants (classified as very high priority & high priority) in the dwellings according to several criteria such as the apparition frequency, the potential of acute risks and chronic risks etc... (www.oqai.fr).
For security reasons, for example in airports and in public or private buildings, the detection of potential threats like chemical/bio, hazards and narcotics is increasingly required. Extremely sensitive and selective sensors are needed for these specific purposes.
In addition to VOC, bio molecules are also present in closed environments. The presence of bio molecules can be an indication of the presence of mould, viruses, etc. indicating the spread of a disease or the presence of allergen factors. Detection of bio molecules requires specific sensing methods.
Given the absence of affordable but sensitive and selective gas sensors on the market to tackle the current challenges, the project aimed to develop new advanced sensor systems to precisely monitor the composition of the air in terms of both chemical and bio contaminants.
IAQSense identified all the potential applications for the novel technologies to be further developed in the project. A common set of specifications has be defined for design and production. Furthermore, a survey has been conducted through the partners of the project, the enable the attached Technical Committee members to refine the technical requirements and the market potential. The development has started on this basis.
A broader review on a worldwide basis has be conducted, to update the potential applications knowing development and application results, to segment the market, to identify the key potential users and to adjust the final product technical specifications to the market needs.
This was also be an opportunity to review the emerging competition, new maturing technologies, new trends in the market and to adjust the product offer.
The IAQSense offer is very unique, and latent needs has also be revealed by this new offer, especially the real-time monitoring opportunities, triggering new applications, never thought of before. It is an opportunity for the consortium as a whole, a chance that should not be underestimated and has be checked carefully through new ideas, new product innovation and a market attraction survey.
More recently opportunities to enter the IoT trends and markets has been identified and developments targeted.
Origin of the project
Regarding the requirements for multi-gas detection, some of the IAQSense project partners have developed a silicon based technology, integrating the equivalent of a spectrometer but at a nano-scale.
Based on surface ion mobility dynamics this is particularly suitable to identify each individual gas in the ambient air and its density. This technology has been patented by EFFICIENCE Marketing (UK patent application number: 1200626.8 “Ionensensors”, 16/01/12 – France and Germany in 2013). This nanotechnology has a high potential to be both affordable and low-cost. One additional patent targeting nano-mechanics (US Patent 7,335,942 B2 – 2008 “Field Effect Transistor Sensor cantilevers”. This patent is now NANO property in Germany while the third Nanorods technology developed by ISL is free (already the subject of publication).
All the patents’ owners and technology stakeholders are involved in the project. Therefore all patents are available as the background of this project.
Based on this solid background, the IAQSense project emanates from the will to demonstrate the relevancy of these devices applied to indoor air quality characterization and monitoring, for which no efficient sensors currently exist.
Furthermore security market and pervasive sensing of chemical threats for the industry, for the protection of public buildings, crowded areas has been targeted.
And last but not least wearables for allergy prevention and occupational health has been also investigated.
The project should go to industrialization for mass markets for the above 3 markets.
Later on, the large Automotive market should benefit from these technologies (Cabin monitoring and Air Quality Control)
S/T Objectives
The S&T objectives of the project are the development of the sensor, its integrated electronics, and several wireless micro modules, validated with a set of the most common molecules present Indoor.
Research in new algorithms for building automation has been explored as contribution of this smart sensor technology for the optimisation of health, comfort and energy savings.
Software Tools for extension of the set of detectable molecules (retrofit download) to be provided.
The technologies has achieved a TRL6 level at least by the end of the project for industrialization and mass production preparation.
The sensing microsystem developed during the project should be able to detect:
• separately any VOC below threshold limits (below 100ppb),
• biomolecules down to ppb levels,
• ultra-traces of hazard and narcotics down to ppt levels.
During the project, the entire system has been developed and derivatives defined for different applications i.e.:
• primarily for buildings (monitoring IAQ),
• health,
• vehicles, and
• security (detection of hazards and/or narcotics).
To work as smart sensor and to detect specific threats and toxic gas, gas patterns are validated and implemented as firmware in an Integrated Circuit. The firmware is upgradable.
Project Results:
Functional specifications and methodology – EFFICIENCE Marketing
Objectives
This WP aims at confirming the primary (serving the building industry) and secondary (health, vehicles, safety) objectives of this project, specifying the spectrum of molecules to detect and then to deduce the specification guidelines for the technological development, electronics, embedded software and the target micro-module configuration at project completion.
Functional specifications (Define from existing market needs, and potential innovative applications) EM; NANO, ISL, MICRO, CEA
For the Functional Specifications priority was given to building applications. This corresponds to the composition of the Technical Committee.
Expected experts in indoor air pollution and health impacts, industrial managers in construction, building automation, ventilation, sensing, have been interviewed. Their profile range from innovative SME’s to Large Companies with international profile. Big companies like Bouygues and Schneider were participating.
Unfortunately contacts with company like Siemens were not successful but further contacts beyond the project are expected on technical specifications of future products.
After issuing the Functional Specifications and a White Paper which takes into account the key points during the first months of the Project, a lot of contacts in different exhibitions, conferences, direct call of people interested from India, USA, Canada, China have been seen. They all confirm the interest for the concept which is developed by IAQSense.
Except these confirmations, the subject was closed during the first period of the Project.
Methodology and Quality Plan EM; ALL PARTNERS
This task aims at defining the entire methodology to be used during the project and the set of indicators, such as availability of gas head samples, characterization of data log available and matching with theoretical models, availability of electronics and micro-modules, in-situ test reporting, package for industrial porting.
The Methodology for validation and approval of key documents and deliverables (review meetings or review procedures) has been defined and applied.
The Quality Plan was mainly defined for implementation of Quality systems within Start-up like Nano Analytik for process quality management. The other industrial side, Id-Mos has been checked for availability of ISO9000 certification which covers all procedures along the semiconductor engineering cycle. For Nano Analytik, after meetings in Ilmenau between EFFICIENCE Marketing (in charge of implementation of Quality plans) and Nano Analytik in February 2014, the whole process of implementation of the Quality System in production has been set-up with a dedicated person in charge. It applies to internal and external (subcontracting) procedures and purchasing actions (validated suppliers).
The quality system of Nano Analytik is not documented for public release.
The risk management is under direct responsibility of the Project Coordinator. It is especially critical in this project as some tasks are dependant and creating critical path in the project management.
Two such tasks have been disconnected from the critical path in order the insure parallel engineering work (Demo Boards and Tests in Buildings). These decisions have proven very effective as several boards and several remarkable reports on building applications have been delivered.
Deliverables (delivered)
Functional specification
Electrical, mechanical, environmental specifications for each technology – Target molecules (priority list and modelling list)
Methodology and Quality Plan
Project management – Milestones and review meetings – Risk management – Responsibilities
Technologies
Objectives
This Work Package 2 process development is focused on the manufacturing technology of the sensors and test structures for technology verification. It will require development of manufacturing sub-processes of: spectrometer-FET-transistor, transistors on the tip and piezo-resistive cantilever beams functionalised with nanotubes.
Working samples of the sensing head are delivered concurrently with a characterisation platform.
Description of work
The key specifications of the sensor should be reached during the work:
• sensitivity <100ppb for the spectrometer and selectivity between the molecules of the reference set,
• sensitivity to 1ppb with a set of 4 functionalization layers for the transistor on tip,
• sensitivity to ppt with a set of 2 functionalization layers for the cantilever with nanorods,
• 70eV ioniser with controlled plasma sample reaching the sensing layer.
Design of building blocks NANO, Partners involved: EM, ISL, MICRO, FACET
This task aims at designing building blocks for sensing head according to functional specifications (
The testing and characterization equipment, realized by FACET will be in line with the sensor design and geometry. The sensors are developed and produced by NANO up to the end of the project
The “Post Kick-Off meeting a “Design & Specification meeting”” (Sofia beginning Sept 2013) has been held in fact ahead of the kick-off meeting (17th Sept 2013).
It has defined the key orientations for the Spectrometer on Chip:
• Departure from circular structure as created by ISSP-BAS to have more freedom for form factor of a linear structure
• Ntype only development, also a departure from both N and P of the ISSP-BAS structure
• Integrated ionizer on chip using suspended filament
• On the known reliability problem, due to the electrolysis of electrodes in harsh environment nothing was decided, leaving this question to the designers
• A suitable CMOS process was expected to be available in Ilmenau thanks to cooperation contract between TU Ilmenau and its spin-off, Nano Analytik
For the Cantilever, the base process was already in place and the main task was the further functionalization and technology transfer from ISL (single die) to one shot functionalization of a full wafer.
During the course of the project several questions were raised and solved concerning the Spectrometer on Chip but increasing the work load on both process and batch processing and characterization work as well.
• The need for both P and Ntype of Spectrometer as signature of dynamics of ions is different for positive and negative ions which both coexist in the Tin Oxide layer. This question has been solved with additional batches (2 different N and P substrate to simplify the process).
• The conditions for ion dynamics on the surface of the Tin Oxide has been examined more deeply leading to higher layer resistivity to respect the so-called “Debye Length”. New batches are involved. The investigations were done theoretically, by modelling and by referring to pre-project studies indicating different categories of resistivity
• Explaining the difference between the modeling and the experimental results, especially a parasitic effect of ten folds increase of static current of the Spectrometer when biased for relatively long time has led to understanding the role of the porous Tin Oxide layer, with creation of H+ ions on the surface which through the porous layer penetrate inside the Silicon Dioxide layer to change deeply the “Work Function” of the SiO2. In addition, this phenomenon is slowing down the transistor by a factor 1000 blocking the interaction Tin Oxide / Channel with ion dynamics. This study has led to the introduction of a “Chemical Barrier” which will probably be introduced in a full, mass production process.
In addition, to solve the reliability problem this chemical barrier is mandatory
• The solution for the ionizer went through 3 steps:
o Design of the initial “filament” ionizer was not feasible, not “soft”, and not effective
o A second design has been decided using a “field ionization” using 200nm holes (technology derived from NASA mars lander). Designed and tested it has been proved to be unreliable at atmospheric, humid pressure
o On the end a proven technique of UVC LED surface ionization using the work function of the Tin Oxide has been chosen, taking the opportunity to use low cost UV LED sources developed for Curing by several companies (including LG). Compared to initial specification it is a 5.2eV (240nm) ionizer with soft ionization (molecules are not broken up)
All this needed work has led to much more work than expected with collaboration between Nano Analytik, EFFICIENCE Marketing specialists, Stefan Andreev team to identify and solve the questions raised through process trimming and characterization work much more extensive than expected.
Technology 1: Integrated Mobility Spectrometer
A novel and unique sensor system based on mobility, dynamics of ions and operates as spectrometer on chip.
Better knowledge of the technology’s intrinsic capability was the objective of the first run of fabrication starting as follows:
• define the structure able to reach the performance (all technical partners),
• proceed with device simulation (MICRO),
• prepare variants for DOE plan including intrinsic parameters and environmental ones.
(Key influencing parameters are the dimensions of (a) collection layer, (b) FET dimensions, (c) ionizer dimensions and accelerating electrode place, (d) the overall device cavity).
The relationship between target specifications and key parameter variation have be analysed thanks to the characterization platform prepared by FACET, an electrical test made by Stefan Andreev on Owlstone gas generator and the support of ISL for analysis in the gas generator.
The results were planned to provide:
• an understanding of the influence of key parameters through DOE analysis,
• a choice of optimal parameters,
• the influence of environmental parameters.
Comments:
The DOE plan can be done a fully stabilized version of the Spectrometer that can be reached within a full standard process, beyond the Project.
Technology-Type 2 Field effect transistor using Tip
The sensing principle has similarities with the sensor type 1, based on change of an electric field due to the adsorption of certain species on a porous layer but work is static only.
The basic principle of this sensor is a FET placed on a micro-scale tip to enhance the overall sensing sensitivity, operated in constant voltage mode and with the source-drain current changing with the amount of adsorbed species. The species are adsorbed by special functionalization coating which ensures selectivity.
Comments:
Technology -Type 3 (TT3) Nanostructured collectors on cantilevers for trace and ultra-trace sensing
Technology
The sensor works as nano-balance. The cantilever is an oscillating and capable of detecting atto mass-changes. The detection is based on reference, non-adsorbing device to be compared to the adsorbing one in terms of oscillating frequency. This way one chip should be composed of multiple different active layers/cantilevers and one reference using 4 places on the chip.
Development
The processing technology for this sensor-type is known by NANO. The coating process will be executed either by ISL on site, or the technology will be transferred to NANO in order to produce the entire sensor unit with one micro-machining process plan.
Production
The production of the sensor is based on standard SOI-technology.
The realization of the process of wafer coating needed two variants to reach a perfect homogeneous coating (etching after deposition).
It pays in return as the performances of one cantilever with different individual gases are very sensitive.
The question of selectivity when a mix of gas is reaching the cantilever and identification of gas types is still under investigation and will be performed with boards developed during the project. Two hypothesis:
• The speed of adsorption of each gas is different and an acquisition of data with a cycle inject/purge will create dynamic events which are representative of the molecules
• A network of 8 cantilevers with different functionalization (on top of the nanotube coatings) will be able to create a static pattern
• The boards developed during the project are able to investigate the two modes or a combination of the two thanks to 8 cantilever on board and 2 electrostatic pumps.
Production of sensor head samples NANO
This task aims at designing building blocks, Design for Manufacturing oriented, definition of packaging/filtering for different applications, setup of mask sets, design of the process steps for each block.
The principle initially described for the task has been followed:
1 - Batch design and production in Nano Analytik
2 – Electrical and first gas test of the samples in Sofia
3 – Test at low concentration in ISL
Several different different batches have followed this path and be tested thanks to the platform of task 2.3 below and gas generators (Owlstone in Sofia, Proprietary in ISL).
Preparation of the test environment and the platform for system integration FACET, NANO, MICRO
The test platform has 3 objectives:
• ensure the electrical test of the different components (technologies) and sub-components of the sensing head to release working heads for further characterisation,
• provide the characterisation platform to the different actors of the characterisation step. It will provide datalogging of the responses to different gases or molecules,
• the platform should also provide the necessary flexibility to install signal processing of the datalogs (batch processing) to prepare the future integrated real time algorithms.
The platform has been built around the National Instruments boards and Labview software platform.
The characterization itself and the interpretation of the results have been done under control of Nano Analytik (with link to the design) and Microsystems Ltd - Varna (with link to the model). FACET providing the necessary upgrade of the platform as needed for datalogging or signal processing.
The team of Prof Stefan Andreev has done the first tests (electrical and preliminary gas tests) on the different types of Spectrometers. He was supported by EFFICIENCE Marketing and FACET for debugging of the samples behaviour and adaptation of the platform hardware and software. See description of characterization and results.
Several technical meetings were held in Sofia for that purpose with technical meetings in Ilmenau to transfer the acquired knowledge to Nano Analytik design and production.
Process porting package reporting for mass production NANO
This task aims at preparing the portage of the process for the sensing heads mass production
Shifting to outside foundry.
While Nano Analytik has succeeded to make several batches of 2 different Spectrometer on Chip which were adding a lot to the initial knowledge on Spectrometer on Chip, some points have been considered important enough to insure porting the process to external foundry:
• the clean room used by NANO has a limited capability in terms of process layers, lead time and volume of production.
• In terms of economics a standard foundry provides hundred folds better cost,
This has lead to consultation of several foundries:
- Xfab (not ready at this time - 2015)
- Semefab (quote obtained)
- AMS via CMP plus post-processing as needed (quote obtained)
Which offer different degree of integration of CMOS+MEMS.
It has been considered favourably for the integration step beyond the project.
Description of boards
All boards are hardware and software compatible with RaspBerry PI-2 / PI-3.
The basic software (drivers, initialization, data acquisition) have been developed for Windows 10 - IoT installed on RaspBerry PI.
5 boards have been developed :
- 2 reference boards using components for sources external to IAQSense
o IAQCore based board for detection of TVOC and CO2
o MSS board using a network of 8 surface stress membranes
- 3 boards using IAQSense technologies:
o Spectrometer on Chip + processing ASIC
o Cantilever based network of 8 resonating cantilevers plus processing/actuating ASIC
o Cantilever based network of 8 resonating cantilevers plus FPGA based lock-in amplifier
The technologies developed by IAQSense are used on the boards:
- Spectrometer on Chip developed and produced by Nano Analytik (P and Ntype)
- ASIC IM452A for Spectrometer on Chip developed by Id-Mos (Xfab foundry)
- ASIC IM309D for Cantilevers network developed by Id-Mos (Xfab foundry)
- Optionally TIP transistor can be packaged for interfacing IM452A on the same board and realizing an enhanced Spectrometer or a field sensing probe (Kelvin probe)
The attachment of board to a Raspberry PI provides embedded solutions for:
- Drivers, Initialization and data acquisition
- Neural network or AI event identification
- Pattern recognition based on event libraries
- Firmware development, compilation, and downloading in case of IM452A ASIC with embedded µC.
The following solutions are recommended for knowledge processing and pattern recognition:
- MatLab and neural network toolbox
- Weka and machine learning algorithms (Random Forest and derivates)
Solutions for software/firmware development:
- MSP430 development and debug: GNU Compiler Collection (GCC) for Spectrometer ASIC firmware development
- Windows 10: Visual Studio 2015 for all RaspBerry PI drivers
IAQCore based board
Application Features
The Application is in charge of driving the HW to:
• Initialise display + IaqCore chip + leds + Buttons
• Launch a periodic timer to read the IaqCore data (5mn power-on then 1s rate of output data
• Red and Green LEDs display the state of the TVOC and CO2 trigger levels and blink at different speeds showing the change trend
• Buttons: Button 1 for new storage session – Button 2 clear all the stored files
• Display (on external screen)
o State of the sensor (ready or not) – Rvalue – TVOC and CO2 level
• Store the measures in a file in CSV format
o Data are stored on CSV in the FTP server of the board
MSS network board (8 stress membranes)
Features
8 channels MSS with input connectors for Silicon Membranes.
Each membrane can be changed, functionalized (DIY polymer deposition), accessible inside injection chamber.
• Each channel has a separated ADC and DAC for managing sensor bridge amplification, offset compensation and conversion (16 bits).
2 electrostatic pumps (on board supplied and driven) provide static or dynamic flow of analytes and purge for static pattern analysis or dynamic slopes analysis.
• One pump is driven to feed the pollutant, the other feed the fresh air alternatively
Application Features
The Application is in charge of driving the HW to:
• Initialise the different devices
o Chip selects for the different ADCs of the Membrane sensors
o Interrupts of the different ADCs (data ready)
o Driving of the pumps
• Define the default DAC values for each sensor
o Before launching a measure campaign, the MSS voltage offset is compensated.
• Launch a measurement campaign for each of the 8 sensors
• The global cycle is 60s split in 30s of pollutant feeding and 30s of fresh air feeding. (100k samples per MSS membrane)
• Data are stored in CSV format for further analysis in the ftp server
Spectrometer on Chip Demo board
Features
The DemoBoard supports 4 sensors
• 2 type P sensors + 1 type P sensor reference
• 2 type N sensors + 1 type N sensor reference
A processing ASIC (IM452A) drives the sensors, acquires the signal from each sensor and provides processing thanks to the embedded µC (MSP430)
An external Flash and RAM are added on the board to store program and one dataset.
Software
The software is prepared on RaspBerry PI and provides initialization, parameters setting and firmware downloading:
Initialization
• Initialize the acquisition parameters (sensor parameters, sequence and acquisition speed)
• Provides transparent mode (learning mode)
• Or Full acquisition and compression mode with pattern recognition
Dowloading
• Download firmware developed offline
• and database prepared on base on learning under gas generator
Principle of learning and pattern recognition
• Using Matlab Neural Toolbox or Weka machine learning, under gas generator with different set of gases. Extraction of key events
• Downloading reference events in IM452A for real-time pattern recognition
8 Cantilever network with ASIC detection
The demo board support 8 resonating cantilevers (100kHz).
An ASIC (IM309D) is acquiring, filtering and detecting the 8 signals form each cantilever in parallel. It provides the driving of each cantilever with separate driving sine waves on the individual resonating frequency (thermal actuator).
A set of 8 ADC and DAC insure dataset acquisition and analogue settings while a set of 8 function generators provide the sine wave for the actuators.
Each cantilever is delivered with surface coating with CuO nanotubes for extended surface and sensitivity and are individually accessible for further individual functionalisation (additional surface coating). It provides very sensitive detection and selectivity at ppb levels of concentrations.
2 electrostatic pumps are providing injection/purge for dynamic detection of events due to molecules on top of the static detection providing means for further selectivity and uncertainty lifting.
Software
Provides from RaspBerry PI:
• The search for resonating frequency on each cantilever and further setting of function generators.
• Initialization and data acquisition
Principle of learning and pattern recognition
• Using Matlab Neural Toolbox or Weka machine learning, under gas generator with different set of gases. Extraction of key events in static mode or synchronised dynamic mode
8 Cantilever network with Lock-in amplifier
separate board has been developed using digital lock-in amplifier to provide further improvement of S/N ratio for below ppb detection.
It uses the same set of 100kHz coated cantilevers that can be further functionalized.
It serves, with the other cantilever as a base for further study of CMOS integration and very low concentration detections.
Description of components
Spectrometer on Chip:
- An Ntype spectrometer
o in DIL14
o comprises a 100x100µm Ntype sensing device with Tin Oxide as sensing layer and a reference transistor
o Compatible to interface ASIC IM452A
- An Ptype spectrometer
o in DIL14
o comprises a 100x300µm Ptype sensing device with Tin Oxide as sensing layer and a reference transistor
o Compatible to interface ASIC IM452A
- These spectrometer can be further functionalized for gas spectrum extension with different types of polymers (available with deposition equipment in Chemical lab Sofia)
ASIC 452A : Dedicated to Spectrometer on Chip
o 4 simultaneous drive and acquisition channels for Spectromers
o 2 multiplexed channels for P and N reference transistors
o UV LED booster and drive
o 200ksps acquisition rate (adjustable) with 12 fast ADC plus offset DAC setting.
o MSP430 embedded processor (16 bits) for real-time processing
o Uploadable software for real-time processing:
- Spectrometer sequence and activation of drivers
- Denoising and compression
- Event and pattern recognition according to uploaded knowledge base
- Transparent mode for external machine learning
Description of software tools
- Drivers for all boards
- Development and debug tools for MSP430 boards
o TI tool for MSP430 board for cantilever network
o Open sources tool trimmed to ASIC IM452A embedded MSP430 and peripherals
- Matlab and toolbox
o Specific programme for machine learning of Spectrometer on PC-Matlab
o Specific programme for machine learning of Spectrometer version RaspBerry PI-Matlab/Simulink/Toolbox
o Event knowledge base for Spectrometer
- Demo software for MSP430 based 8-Cantilever board
Documentation provided
- White Paper
- Datasheets of components
- Detailed description of ASIC IM452A (on request)
- Datasheets of boards and connection to RaspBerry PI
- Report on optimum position of VOC sensors in buildings (on request)
Work performed on Building integration and Control algorithms
Objectives:
Installation of miniaturized sensors developed in IAQSense project in given environment able to recreate a real indoor environment such as a dwelling both in the CEA and ACCIONA facilities.
Test of these sensors close to a real indoor environment with a possible indoor control such as temperature, air change rate, level of pollutants, etc....
Focus on the best positioning of the miniaturized sensors at different scale (building, room) and location (centre of the room, on the walls, etc....).
The objective of this work was to find some appropriate position for IAQ Sensors. For that, a bibliographical review was first done with a lot of recommendation found at a room scale. Then some tests of different position of sensors were run in an experimental house, at a room scale first and at the building size in a second time.
The tests done at the room scale confirmed and completed the recommendation found in the literature: The position of the sensors can be on the walls not particularly in the centre of the room. They can be positioned at a height corresponding to the breathing zone (0.5m-1.7m). Avoid positioning the sensors close to the windows, the corners, the doors, the heating or cooling systems, the solar task, the forced convection/draught, pollutant sources & sinks.
At the building scale, the strategy will depend on the number of sensors available. Starting from the “ideal solution” with no limitation in terms of costs, architectural constraints ... it is recommended to position one sensor per room, or at least in each area were we spent the most of the time (living zone). At the opposite, when only one sensor can be set, the transit zone might be the more representative of the indoor pollution and so the sensor can be positioned in this area. This area being between the air inlet and the exhaust air. In this configuration, the relevance of the measurements coupled with the reactivity of the pollution detection will strongly depend on the building configuration (geometry, doors open or closed, ventilation ...). In the case of multi-storey building, an intermediate solution could be to have one sensor per storey which could be positioned either in the living zone if this is a sort of open space, or in a transit zone if there are plenty of rooms.
A summary was proposed to provide some recommendations for the positioning of sensors at different scales as follows:
“An adequate location of the sensors is very important because they measure the parameters that will manage the equipment needed to improve the indoor air quality.
Theoretically, the optimum location is in the centre of the room but this is not a practical place to introduce sensors. After the installation sensors in 4 points of the room, the conclusions regarding the most realistic locations are:
- The thermo-hygrometric conditions change a lot in a room mainly with the height due to temperature gradient made by the fluctuations in the air density. To measure the temperature and humidity at the middle of a wall (with the appropriate insulation and R-value) is enough to obtain similar values than the inhabitants can perceive.
- The air pollutant composition (VOC and CO2) should be measured at the breathable height to know the concentration of those pollutants that can be breathed. Using the same location than before, it is able to have representative values but it is important to be sure that the sensor is not joined to an emission source.”
A detailed report on sensor positioning and findings is available on request.
Optimization of the IAQ via a regulation system
Focus on the VOC sensor performance through management strategies of HVAC systems.
The management strategy will consist in increasing the indoor air quality (IAQ) while minimizing the energy consumption.
This management is a control algorithm that has be developed previously, consequently there is a necessity to define the inputs of this algorithm (VOC, CO2, temperature, and so on...) and the outputs (I.e.: HVAC system) to reach a valuable performance both in IAQ and energy efficiency.
The major goal of the IAQsense project was to develop a new IAQ sensor based on nanotechnologies and able to monitor several major pollutants in the indoor air. The first task consisted in defining an acceptable positioning of IAQ sensors in an indoor environment. Then, after applying these recommendations of positioning, the goal of the next task was to develop a new control command solution (CCS), using the data delivered by the sensor in order to improve the indoor air quality in regards with a low energy consumption.
A control command solution which can value the data coming from the IAQSense sensors has been developed and tested in this task by CEA. This control command solution has the objective to tackle both IAQ improvement and energy consumption reduction. To develop and test the efficiency of our control command solution (CCS), two parallel steps were proposed in this study with on one hand a numerical approach and on the other an experimental step. Both have been tested at a room scale.
In the numerical approach, the building simulation tool Energy+ was used to simulate both energy consumption and IAQ in one room. It has been coupled to Matlab through the platform ‘BCVTB’ in order to develop, test and optimize the Control Command Strategy. This numerical part has shown the great ability of a good ventilation strategy to reach both energy efficiency and indoor air quality objectives.
The second step allowed to test at real scale the algorithm developed through simulation. The tests were carried out in our experimental platform located in Le Bourget du lac, at INES. The algorithm has been implemented in one room of an experimental house. It controls the mechanical ventilation system depending on two parameters values (the carbon dioxide, and the total volatile organic compounds).
The results are very interesting because the algorithm has the expected behaviour both in the simulations and in the measurement campaigns. It is able to regulate the IAQ thanks to essentials parameters: the carbon dioxide and the TVOC. The first pollutant is an indicator of the air exchange rate and of the occupancy rate. It permits to contribute to a good ventilation. The second one is a global estimation/’picture’ of the chemical pollutants in a room. Thanks to specific concentration thresholds based on health sanitary criterion, the algorithm turned ‘on’ or ‘off’ the mechanical ventilation system (MVS).
This mechanical ventilation was the actuator chosen to improve both IAQ and EE. The ventilation was modulated preventing the pollutants from going over these sanitary values in regards with a low energy consumption. The numerical approach has been implemented and has shown the great ability of a good ventilation strategy to reach both an energy efficiency and an indoor air quality objective. The ventilation strategy tested in the numerical model with the chosen configuration shows its ability to reach the IAQ objective and at the same time to maintain an energy consumption either smaller or at least in the same order of magnitude as the usual ventilation strategies from 0,3 to 0,5 h-1. In the experimental part, the first tests done allowed to ensure that the CCS reacts correctly regarding the algorithm implemented at real scale. Thus, these results showed a great opportunity to promote more the IAQ by arguing its compatibility with the energy efficiency.
The work done in this task will help our partners to provide an additional function to the sensor developed. It could contribute to improve the IAQ without degrading the energy consumption in buildings thanks this control command solution
Test of sensing heads in gas or molecule environments - analysis of patterned data log with algorithms
Summary
Several batches of spectrometer MOSFET sensors are produced in an iterative manner to achieve the expected response to the gas molecules.
The experimental work for spectrometer MOSFET sensors was organized by:
- Establishing identical measurement equipment at producer (nano), sensor behavior and descriptors identification lab (Sofia lab) and final experimental lab for sensitivity measurements (ISL).
- Setting the SVN server for data collection was established at uS.
- Organizing the training workshop for the NI platform users.
- Creating a plan and procedure for distribution of the sensor batches.
- Systematic performing the experiments and collecting data for the task 3.2
The sensor response to the gases is relevant to the expected behavior derived from the simulation model. Based on the first experiments, the simulation model was improved.
Based on the first experiments, a feedback to the producer was provided and the final experiment conditions and regimes was specified for systematic testing.
The experimental work with the nano-functionalized cantilevers was continued during the 2nd period. Improved nano-functionalized cantilevers and detection electronics were used.
The experiments confirmed extremely high sensitivity of nano-functionalized cantilevers to some explosive materials as well as to selected number of analytes of interest.
Deliverables
“Report on characterization environment / parameters and results” was delivered in Period 1 based on the experiments with first sensor batches 1 and 2. The results obtained at ISL were reported in June 2016 as an amendment of that report.
Outcomes
Spectrometer on chip based on MOSFET sensor
NANO developed and produced working MOS structures sensitive to chemical substances.
SOFIA LAB performed more detailed experiments, proving possibility for discrimination by type and concentration. Proof of concept of using descriptors and pattern recognition was done. Important outcome of this task is the establishment of SOFIA LAB as a permanently working entity which will provide the sustainability of the project results in the after-project period.
ISL LAB performed experiments with concentrations below exposure limits, proof of higher sensitivity was done.
MICRO developed all needed MATLAB tools for analysis of the experiment results. Based on the experimental data, considerable set of descriptors were derived for the following pattern recognition model.
Nano-functionalized cantilevers
A very high-sensitive detection system based on nano-functionalized cantilevers was developed and tested in gas environment with extremely low-level of analytes of interest.
The system proved very high sensitivity for explosive materials. The results allow direct practical exploitation in a commercial product.
Modelling
Summary
The algorithms for descriptors extraction and pattern recognition were developed on MATLAB first, and validated with the experiment data.
The software implementation was done by creating a transportable C/C++ code which was implemented on a common contemporary microcontroller MSP432xx. The software code was validated using a dedicated platform.
The used Artificial Neural Network (ANN) allow software engine to be updated for recognition of new data (if systematically gathered).
The Software part of the Spectrometer on Chip was proven for functionality and implementation in the further industrial product.
Deliverables
“Pre-defined patterns and test comparison report” – delivered in March 2016
“Software for customization of the gas spectrum detection” – delivered in March 2016
Outcomes
Fully functional recognition engine based on ANN is developed and validated.
MATLAB scripts for ANN generation and training as well as exporting the C++ source files of the net.
Transportable C++ code for the computation of the descriptors and pattern recognition engine is available.
The principle of the developed tools and software architecture allows easily upgrading of the software on the client side just by importing the updated ANN coefficients after each next training. The latest makes the software very flexible and convenient with respect to the usage of the end-product.
The software part of the spectrometer on chip product is ready for future commercialization.
Development of innovative technologies for functionalization of cantilever surface at wafer scale
Summary
The nanostructuration protocol was developed at ISL before the project IAQSense, and transferred to piezoresistive cantilevers during the project. A close cooperation between ISL and Nanoanalytik have permitted to define which way the nanostructuration could be performed on piezoresistive cantilevers without damaging the electronic transducing part of these cantilevers. The protocol for nanostructuration on large wafer scale was defined and optimized at ISL with the aim to allow the fabrication of hundreds of nanostructured cantilevers in one same batch, and showing similar characteristics in terms of resonance frequency and physical properties. Once optimized the protocol was transferred to Nanoanalytik fort including the synthesis in the process of cantilever fabrication
Deliverables
“Technological report on the nanostructuration of collecting layers on cantilever surfaces: Technology and process description of nanostructuration of specific layers on cantilever based sensor” as been delivered mid-June 2016.
Dissemination
Internal and public website (FSRM, EM)
Publication on www.iaqsense.eu of
- 2 IAQSense Flyers
- 3 IAQSense Presentations
- 3 IAQSense Newsletters
- 2 IAQSense Events
- 5 IAQSense Highlight documents
- 6 IAQSense Posters
Dissemination
- Dissemination plan updated in March 2016 (FSRM). Final version delivered M36 (August 2016, FSRM).
- Organization of an internal workshop for the project parnter in Ilmenau (DE), July 2015 (nano, EM).
Content:
System training with the new measurement setup for characterization of chip transistor spectrometer.
Participants :
S. Andreev (microsistemi) I. Atanasov (nanoanalytik), C. Iroulart (EM), T. Ivanov (Facet), M. Holz (nanoanalytik) , P. Popov (Facet), N. Nikolov (microsistemi), I. Rangelow (TU Ilmenau).
- Organization of 3 public workshops to disseminate project results (FSRM)
23-6-2015 – Workshop with Partners and Technical Committee – Paris
12-05-2016 – Sensor&Test, Nurembers – Special Session “Novel sensor system for Indoor Air Quality control”
07-07-2016 – Indoor Air, Gent – presentation “Changing the game in the management of Indoor Air Quality” and “Real-time monitoring for improved health, comfort and energy efficiency”.
- Publication of 3 newsletters (FSRM, with inputs of all partners)
Aim: to disseminate public project results
Target group: colleagues, business contacts, customers, press, clusters, projects, the wider public.
Distribution: sent by e-mail to 250 addresses and published on www.iaqsense.eu
Issue I: April 2016
Content :
• The IAQ project
• The Integrated Spectrometer on Chip and the Processing ASIC
• CANT 2005 –8 resonating cantilever network
• IM309xx – Signal processing seriesfor resonatingcantilever (8 channels)
• IAQSense at Sensor&Test Nürnberg
• Common Booth IAQSense-PAMIAQ projects
Issue II: June 2016
Content :
• Software elaborated for the customization of the gas spectrum detection
• Functional Cantilevers
• IAQSense at nanoFIS, Graz (AT), June 27-29, 2016
• IAQSense at Indoor Air 2016, Ghent (NL), July 3-8, 2016
Issue III: August 2016
Content:
• Project End
• Five Demo Boards
• Spin-off company IAQSense Ltd created for further project exploitation
- New IAQSense Project Flyer (FSRM)
Aim: to give general information on the project, objectives, expected results, partners
Target group: scientific community, EC, industry, students, general public
Distribution: by all partners to their contacts and at exhibitions/congresses where their institution was present, at the IAQSense booth and IAQSense workshops.
Download: www.iaqsense.eu/publications
- Flyer « Spectrometer on Chip » (FSRM, EM, nano analytic, ID-mos)
Aim: Information on the technical details of the spectrometer on chip
Target group: Industry
Distribution:Technical Committee, customers
Download : www.iaqsense.eu/publications
- 23 presentations, papers, posters in 10 countries. 1 booth (Organisation: FSRM. Content: ALL)
01/2014: CMEES Cluster, Sofia
05/2014: Visit IMEC and meeting with Bulgarian authorities
06/2014: ECTP, Special session “Indoor Environment Quality
09/2014: CMEES Cluster, Sofia
05/2015: COST Action EuNetAir meeting
05/2015: Meeting Technical Committee
09/2015: Common Meeting Sensindoor/IAQSense
11/2015: NIDays, sensor measurement system
12/2015: COST Action EuNetAir meeting
03/2016: Nanoscale Conference
04/2016: Smart Cities, Stand Eureka Euripides
05/2016: Sensor + Test
07/2016: Indoor Air 2016
07/2016: NanoFIS
Deliverables: Dissemination Plan - FSRM - Dissemination Activities – FSRM
White paper on the project (EM)
Task finished in M3. Published on website
Cooperation with related European RTD Initiatives (EM)
Objectives for the period:
- Cooperation with related projects and clusters
- To realize a common workshop and a booth with project SensIndoor
Results:
• EuNet Air (Cost Action)
• Poster of M. Holz, nanoanalytik, in Cluster meeting May 2015, Lynköping
• Posters of N. Nikolov, microsistemi, and M. Holz in Cluster meeting Dec 2015, Sofia
• Cluster Modelling and Characterization
• Participation of M. Holz, nanoanalytik, in cluster meeting, Nov 2015, Brussels
• Eureka
• Poster of I. Atanasov, nanoanalytik, in Smart Cities Innovation Event, April 2016, Stockholm
• Collaboration with EC funded project SensIndoor:
• Meeting to discuss possible common actions (June 2015, Freiburg DE)
• Several telcos to organize the events
• Common booth at Sensor&Test, Nuremberg (May 2016)
• Joint session at Sensor&Test, Nuremberg (May 2016)
• Joint session at IndoorAir 2016, Ghent (July 2016)
Links to all related clusters and projects: http://www.iaqsense.eu
Deliverables for the period :
Cooperation Results with other European RTD Initiatives (EM: M36)
IPR and exploitation of results (nano)
Objectives:
- To organize an exploitation workshop with a moderator from the EC
- To collect the claims of the partners on the project foreground
- To inform the Technical Committee about the project results and to collect feedback
- To elaborate the final IPR and exploitation plan
Results:
- Exploitation workshop with a moderator from the EC organized M19 (March 2015) by FSRM and EM in Brussels.
- Tables with background, foreground, exploitable results and claims filled in by all partners (EM). More details: see D8.36 IPR and Exploitation Plan.
- Workshop with the Technical Committee organized June 2015 in Paris (FSRM, EM)
Content:
• 6 presentations on IAQSense project and results by partners EM, nano analytik, Id-mos and Eeleo.
• Feedback by the Technical Committee and discussion.
Participants :
• 7 partners from the project: EELEO, nano analytik, EM, ID-MOS, FSRM and
• 8 members of the Technical Committee from international companies : AERECO, SCHNEIDER ELECTRIC
• CIAT, EXPLORAIR, BOUYGUES.
By agreement of all partners, a spin-off company named IAQSense Ltd has been created in the end of the project to assure continued use of the project results. Its strong international team is in charge of CMOS and MEMS integration, mass production, test and calibration. IAQSense Ltd will follow the trends towards IoT and pervasive sensing in the development of a complete range of environmental sensors for indoor, outdoor and wearables.
- Details: see IPR and Exploitation Plan.
Deliverables:
IPR and exploitation plan proposal (EM, M36)
The exploitation of the results are managed through a spin-off company regrouping main disciplines for development of products and services and introduction on the market.
Components are available
Boards are available and should be further characterized, optimized for commercial delivery
Medium term CMOS integration of Sensors (Spectrometer, MSS, Cantilevers) should be organized beyond the project to reach small factor application markets (wearables) and mass produced sensors for Buildings, vehicles, pervasive security networks.
Creation of a Spin-off – End of August 2016
Activities
- Chemical Lab
- Development
- Integration
- Functionalization
- Commercial
In charge of development of current and future product
Components
- Production Spectrometer
- Production ASIC IM452A
- Production Cantilevers
- CMOS Integrated Spectrometer + Processing
- CMOS + MEMS Cantilever + Processing
- CMOS + MEMS MSS
Boards
IoT / RaspBerry PI compatible
- 8-MSS Boards (VOC and threat detection)
- 8-Cantilever boards (VOC and Threat detection)
- 2-Cantilever board (PM detection)
- TIP based Field mapping
- 4-Spectrometer on Chip board (VOC)
Chemical Lab service – Accredited ISO 17025
Functionalization service (on INERA Equipment and Polymer service)
Electronics and AI software support
Development of Wearable
CAD for CMOS + MEMS
Commercial:
- Exploitation of contacts gained during IAQSense Project
- Going on with Technical Committee members
Chemical lab description
The Chemical Lab is built using the Assets acquired by the IAQSense Project.
The key equipment is the gas generator (Owlstone) which provides defined concentration of gases using permeation tubes.
A set of permeation tubes have been acquired during the project covering the key gases identified for VOC detection in building. Some short life time tubes should be acquired at the time of a measurement session.
The objective of a post project installation of the Chemical Lab is to organise a Lab according to the ISO 17025 quality standard for measuring lab by having all procedures, quality system and safety conditions accordingly.
The accredited lab should be able to deliver all services for internal development step and external services to gas sensors and gas detectors manufacturer.
The type of planned services:
- Characterization service for new sensors
- Test, measurement and calibration for production of sensors
- Recalibration of gas detectors
One of the capacity of the lab is application of ISO 16000-29 for TVOC and specific gas set sensors.
The lab is installed in Sofia. A lab manager, chemist, gas sensing specialist and with knowledge on ISO 17025 has been hired for this operation.
Design of analogue interface FACET, ID-MOS
Objectives : The objectives is to Design of sensor head interface with processing electronics, focus on low power consumption, sleep mode for the first sensor based on a spectrometer principle. By heating a wire filament, the electrons are collected by the trap electrode. Depending on each molecule type, the voltage level of this electrode can reach 70V. The spectrometer itself is based on a floating gate over a FET. After stabilization, the ions collected by the gate are moving from one side to the other at different speeds by alternating the drain/source positively and negatively. Depending on their physical properties the gate potential will modulate the current in the channel. During the operating time window, in the range of 10μs, the analogue electronic has to provide the high voltage (10V – 70V), enough current for the sensitive device (5mA) and to detect the output current. Then the signal is amplified, filtered and converted dynamically. (ADC 16bits with 1Mhz sampling).
Main functions :
• I to V converter and programmable Amplifier
• STEP-UP DC-DC Converter
• SENSOR Bridge Driver
• LED Driver
• Temperature sensor amplifier
• ADC / DAC
• BLOCK Diagram
Results : Block Diagram / Layout
Deliverables
• Analogue Specification
Digital Architecture and design of pattern processing function on-chip (ID MOS, FACET, NANO)
Objectives : The digital part of the future one chip solution will be designed by ID-MOS keeping in mind the interface with the analogue block and future choice of compatible technology for the one chip solution.
The output signals coming from the analogue block will be converted with an ADC to:
- Identify the status of the sensing head and to feedback conditioning control signals,
- Process the incoming signal from the different parts of the sensing head.
The precision of the ADC will ensure precision and selectivity of the gas spectrum, it is a critical part of this design.
For this task of digital signal processing the algorithms used during the characterisation phase of the sensing head will be examined for their portability on-chip.
Results : Methodology/Block Diagram/ Layout
The industry standard TOP-DOWN methodology was employed to design the circuit.
The design is based on RTL VHDL modelisation.
RTL models and final the netlist are validated using a single or several test bench(es) featuring automatic checks. The layout was obtained by automatic Place and Route, and checked by DRC /LVS tools.
Integration results of the digital part: 23 000 equivalent Nand Gates – 9.86 mm2 total area in 0.35um technology
Deliverables
• Pre-defined pattern processing arch. Digital part Specifications
Global Integrated Circuit (IC) - Global architecture / foundry - Test architecture - Characterisation of the IC - ID MOS, FACET, NANO
Objectives :
The analogue and digital blocks developed separately is merged in one chip defining:
- the global architecture of the chip
- the test architecture
- the global layout for tape-out (data for production in subcontracted foundry).
ID-MOS is managing the subcontracting for semiconductor foundry and all related engineering tasks.
The external foundry will provide wafers ready for probing (test of individual chips directly on wafer).
After packaging the final test and the characterisation at chip level is done by FACET.
In this way ID-MOS manage the complete design and production of the IC product cycle and provides ready tested chips.
Results : Block Diagram / Layout/ Package / Silicium
Deliverables
• ASIC : Design Summary Report (DSR) - Layout Summary Report (LSR)
• First silicon delivery
• TOP Circuit specification - DSP specification
The ASIC coming from foundry as wafer was packaged in August 2106 and made available for the Spectrometer demo board. Currently the board is under characterization.
Preparing micro module (Combining Sensor, IC and pattern recognition software, Harvesting, RF) NANO, EELEO
Objectives : Deliver demonstration boards and development kits for the key technologies Spectrometer on Chip and Cantilevers
Results : 5 boards have been delivered for evaluation and future optimisation for introduction on the market as IoT compatible boards
Monitoring system - Integration of the sensor system in in-situ monitoring system. NANO, ACCIONA
Objectives : Use of VOC sensors for monitoring in buildings
Results : suitable VOC sensors have been delivered for investigations in buildings during 2015-2016 period.
Potential Impact:
Impacts
Socio-economic impacts
Right from the beginning of the project several potential applications of “real-time VOC sensing” were identified, starting from the priority target of providing solutions for buildings.
The potential market for buildings (home, offices, public buildings) and vehicles have been estimated as TAM, SAM and SOM for the different segments while targeting a final product at 30€ pricing.
These hypothesis have been confirmed and the target price as well through further investigations on the market trends.
In addition impact of new trends like IoT market development has been estimated with the possibility to enter the IoT market through compatible boards.
The building related market has been checked and confirmed through the Technical Committee attached to the project and involving professionals in building construction, ventilation and building automation.
Furthermore the security market and the medical ones have been checked, with a particular interest in providing wearable solutions for the occupational health and allergy/asthma prevention.
In buildings:
- Health and Comfort, Sick building syndrome
The prevention of health problems related to carcinogens is one impact which can be linked economically to the cost of medical treatments.
The effect on Comfort, work efficiency, absenteeism and feelings at work is largely connected to the Sick building syndrome due to the air tight new constructions (at the origin of the Project). Several studies in Europe and USA have estimated the huge economic impact of such situation.
- Energy consumption
The same “air tight constructions” being done to reduce the energy consumption in buildings, the question of optimisation of energy versus pollution (ventilation) has been raised.
The rules for design of new low energy buildings is to insure a ventilation level (air renewal) which is enough to insure average pollution removal and low enough to keep the energy. This is a compromise that is too low when there is normal human presence and too high when they are absent. This valid, at different time of the day, for residential buildings or offices.
The studies made by the two project Partners, CEA and ACCIONA have demonstrated that with good sensing distribution and suitable algorithms an optimisation of the pollution level at any time and lowering of the energy consumption is possible with significant improvement.
Medical
- Allergy and asthma
Having in mind the very large number of people (Millions in each country) affected by allergy and Asthma, the possibility to detect the environmental quality of the premises where you go is major improvement. Two cases of growing interest have been discussed with specialists in Europe and USA:
• The occupational health with impacts over decades (see ERS organisation – European Respiratory Society)
• The growing usage of flagrance in the public places creating high risks of allergy (USA) and general topic of asthma. We are currently talking with Doctors specialized in allergy in the USA to create a suitable wearable.
Our plan for product introduction
The high interest for sensors at the right level of sensitivity and selectivity and the growing trend towards IoT and pervasive sensing has created an huge opportunity for the technologies developed in the IAQSense project.
- A reasonable Turn-over and Business Plan has been setup for components and boards (see separate Business Plan)
- A structure which is product development and business oriented is the counter part of these opportunities. This is the reason for the creation of a “Spin-off” Company.
- A road map for product introduction is considering the constraints:
o Mature products to be introduced first (polymer customised networks)
o Products with limited development in the second place
o Products with high level of integration and complexity will need more time (wearables)
Societal implications of the project so far
- Protection against chemical accidents or chemical attacks
Growth of chemical industry and frequent accidents (leakage, explosions) is rising the question of preventive actions to protect the nearby population.
The growth of terrorist actions is connected to preventive detection of chemicals and explosives.
In general protection of key assets in modern society is critical for economic survival.
The answer lies in pervasive sensing in industry, public areas with IoT connections. Network of sensors with suitable costs (low) and performances (high) are required. This is one of the topics in SEC H2020 calls IAQSense partners missed in 2016.
Market studies show two trends:
• More sophisticated stationary sensors at the right cost to create pervasive, real-time communicating networks of sensors
• “Body” Integration with wearables and smartphone integration
- Comfort in general and intellectual efficiency
The three factors we mentioned, growing pollution and growing allergy in the population, introduction of new chemicals in new materials or directly in the air (flagrance), trends to confinement in buildings are subjects to fight with introduction of smart sensors
- Management of growing “smart cities”
The trend for higher and higher growth of urban population is creating new challenges and new needs for which the following topics apply
o Smart sensing network
o Citizen data
o Pervasive sensing
o Providing relevant information on the IoT channel
In addition the citizen is on the centre of a distributed network of information for himself and for all the population of the City.
- Self-awareness of pollution is required to take decision
o Information network on pollution
o Wearables collecting, displaying, delivering pollution information
Dissemination and exploitation plan
The dissemination of the project results has been done through the following actions:
- Creation of a website (www.iaqsense.eu) to inform the public about IAQSense actions. This site has published :
o A white paper about VOC detection objectives, real-time operation and technologies
o News about presence at different exhibitions
o Highlights on project results and products datasheets
- Press release and interviews:
o Has led to several publications worldwide
- Newsletters informing a large set of professionals about the development steps and the final results of IAQSense project
- Participation in Clusters
- Participations in several dedicated conferences and exhibitions
- Communicating with the Technical committee
- Exchanging with other projects (existing or created on purpose)
- Face to face exchange with current and new contacts for development and commercial purpose
The dissemination continues through the exploitation plan:
- Creation of a Spin-off called IAQSense Ltd (done in beginning Sept 2016)
- Protection of Brand name and creation of commercial websites
- Development of products and services around components and boards
- Initializing commercial and distribution actions
- Extension of product range features through CMOS integration and Functionalization techniques
The results exploited through the above structure are:
- Chemical lab (ISO 17025 accredited) as support for the development and external services for characterization, test and production support of gas sensors
- CMOS integration of technologies developed along the project and in collaboration (6 type of concerned components)
- Coating and Functionalization techniques
- Machine learning and pattern recognition techniques
List of Websites:
www.iaqsense.eu
Contacts:
• Stefan Andreev - Manager of IAQSense Ltd - andreev.stefan@abv.bg
• Mathias Holz - m.holz@nanoanalytik.de
• Paul-Emile Latimier - paul-emile.latimier@id-mos.com
• Claude Iroulart - claude.iroulart@efficience-marketing.com
• Jacques Montes - jacques.montes@eeleo.com
Under construction:
Commercial website: www.iaqsense.com and national versions www.iaqsense.fr www.iaqsense.de
