Semiconductors: can European industry regain ground?

The glob­al mar­ket for elec­tron­ic com­po­nents will be worth $625bn in 2024. Dur­ing the Covid-19 pan­dem­ic and the result­ing short­age of chips, Europe was remind­ed of its depen­dence on Asia and the Unit­ed States. In an attempt to lim­it this depen­dence, the Euro­pean Union announced the launch of the CHIPS and Sci­ence Act in Feb­ru­ary 2022, which aimed to stim­u­late Euro­pean pro­duc­tion. Some three years lat­er, how is the sec­tor far­ing in Europe? We take stock with Sébastien Dau­vé, direc­tor of CEA-Leti, who has just launched the FAMES pilot line in Greno­ble, fund­ed by the Euro­pean Union and France.

Sébastien Dau­vé. After a decline in 2023, the mar­ket has been grow­ing since 2024, but this growth masks a two-speed evo­lu­tion. The mature semi­con­duc­tor sec­tor, such as micro­con­trollers, which were in short sup­ply for indus­try and the auto­mo­tive sec­tor in 2022–2023, is now sat­u­rat­ed. At the same time, we are see­ing an explo­sion in the mar­ket for very advanced node com­po­nents (less than 5 nm), stim­u­lat­ed by very strong demand for chips for data cen­tres and arti­fi­cial intel­li­gence appli­ca­tions: graph­ics pro­cess­ing units (GPUs) and high band­width mem­o­ry (HBM). The invest­ments in these fields are stag­ger­ing: Tai­wanese com­pa­ny TSMC, for exam­ple, has announced that it will invest $100bn in the Unit­ed States over the next four years.

The invest­ments required to deploy new indus­tri­al resources are such that we can­not imag­ine a major change in the short term. It is there­fore still char­ac­terised today by a very strong inter­de­pen­dence on a glob­al lev­el: a com­po­nent can be designed on one con­ti­nent and pro­duced on a sec­ond, while the raw mate­ri­als are sup­plied by a third. The Unit­ed States, for exam­ple, excels in the design of inte­grat­ed cir­cuits. Japan has tak­en the lead in the pro­duc­tion of wafers (the semi­con­duc­tor wafers on which elec­tron­ic com­po­nents are print­ed) and process gas­es, and Chi­na is unavoid­able for the sup­ply of rare earths. Tai­wan and South Korea, through the TSMC and Sam­sung foundries, dom­i­nate chip pro­duc­tion – TSMC is even the only one to mas­ter the most advanced nodes (2 nm), which are high­ly inno­v­a­tive and in great demand today.

Europe accounts for just under 10% of glob­al semi­con­duc­tor pro­duc­tion, thanks to man­u­fac­tur­ers such as the French com­pa­ny STMi­cro­elec­tron­ics, which ranks around 10^th in the world. Although it does not have the capac­i­ty to pro­duce advanced nodes, it is rather well posi­tioned in the design and pro­duc­tion of ‘More than Moore’ com­po­nents, con­sist­ing of sen­sors, imagers, pow­er and tele­com com­po­nents, and micro­con­trollers. This class finds appli­ca­tions in var­i­ous sec­tors of activ­i­ty, such as auto­mo­tive, indus­try, defence and health. The con­ti­nent also has near-monop­o­lies in spe­cif­ic fields: the Dutch com­pa­ny ASML, for exam­ple, is the only play­er to have mas­tered the man­u­fac­ture of EUV advanced lith­o­g­ra­phy equip­ment, which is essen­tial for foundries.

Final­ly, Europe ben­e­fits from active research and sig­nif­i­cant inno­va­tion capa­bil­i­ties, par­tic­u­lar­ly through its RTOs (Research and Tech­nol­o­gy Organ­i­sa­tions, includ­ing CEA-Leti, the Bel­gian IMEC, the Ger­man Fraun­hofer, the Finnish VTT, etc.), a unique organ­i­sa­tion­al mod­el capa­ble of dri­ving inno­va­tion from the ear­li­est stages of research through to pre-industrialisation.

We knew that this objec­tive was very ambi­tious… I would say that in the short and medi­um term, it is more a ques­tion of main­tain­ing our posi­tion in the cur­rent mar­ket and retain­ing our sov­er­eign­ty over the most strate­gic devel­op­ments when we have the means to do so: in par­tic­u­lar, those relat­ing to defence and cyber­se­cu­ri­ty, but also quan­tum com­put­ing, in which Europe is mak­ing good progress.

This is bad news for Europe, because it would be in our inter­est to have more play­ers estab­lished here. In micro­elec­tron­ics, the notion of ecosys­tem is very impor­tant. In Greno­ble, we are for­tu­nate to have an ecosys­tem that has reached crit­i­cal mass, bring­ing togeth­er the entire val­ue chain, from start-ups to large groups, and this is invaluable.

We must con­tin­ue to invest in our strengths, from tech­no­log­i­cal inno­va­tion to pro­duc­tion, but also strength­en the links between the semi­con­duc­tor and the areas of appli­ca­tion that are of vital impor­tance for Europe (indus­try, auto­mo­tive, health, etc.), which have now become ful­ly aware of the impor­tance of components.

Europe is also ahead of the game when it comes to tak­ing envi­ron­men­tal and ener­gy issues into con­sid­er­a­tion: these two con­straints rep­re­sent sig­nif­i­cant oppor­tu­ni­ties for inno­va­tion. The CEA is, for exam­ple, lead­ing the Euro­pean GENESIS project, bring­ing togeth­er 50 part­ners, with the aim of accel­er­at­ing eco-inno­va­tion in semi­con­duc­tor man­u­fac­tur­ing process­es. We also aim to reduce com­po­nent con­sump­tion by a fac­tor of 1000 by 2032.

Above all, how­ev­er, we must not for­get that the micro­elec­tron­ics mar­ket is cycli­cal by nature: what is true today will not nec­es­sar­i­ly be true tomorrow.

A major trend is emerg­ing in which Europe could play to its strengths: edge AI, arti­fi­cial intel­li­gence man­aged not in data cen­tres but on periph­er­als, smart­phones, con­nect­ed objects, indus­tri­al units, etc. These embed­ded appli­ca­tions require elec­tron­ics that are both very ener­gy effi­cient and capa­ble of per­form­ing the infer­ence phase, or even the learn­ing phase, local­ly. How­ev­er, tra­di­tion­al­ly, the units ded­i­cat­ed to cal­cu­la­tion and those ded­i­cat­ed to mem­o­ry are sep­a­rat­ed on the chips: 80 to 90% of the ener­gy is con­sumed in the trans­mis­sion of data between the two. Edge AI will there­fore require inno­va­tions in elec­tron­ic archi­tec­ture, in which Europe has a role to play. It will also be close­ly linked to sen­sors, which are a Euro­pean strength.

FAMES allows us to build 2,000 m² of addi­tion­al clean rooms and to acquire around a hun­dred new items of indus­tri­al equip­ment, rep­re­sent­ing an invest­ment of 830 mil­lion euros, sup­port­ed by both the Euro­pean Com­mis­sion and the French State. It entered into oper­a­tional ser­vice at the begin­ning of the year. We have kept per­fect­ly to the planned sched­ule, which was very tight. It is impor­tant to empha­sise this: when you know Asia, you know that oper­a­tional exe­cu­tion is for­mi­da­ble there. This proves that we are capa­ble of doing just as well.

Its pri­ma­ry pur­pose will be to pre­pare FD-SOI tech­nolo­gies for 10 or even 7 nm nodes. This tech­nol­o­gy is cur­rent­ly pro­duced by Glob­al­Foundries and STMi­cro­elec­tron­ics, in 22 and 18 nm respec­tive­ly. The tar­get mar­ket remains mod­est on a glob­al scale, but it is a par­tic­u­lar­ly inter­est­ing solu­tion for embed­ded appli­ca­tions that are look­ing for fru­gal­i­ty. But FAMES should also enable us to pre­pare the ‘next step’ for Euro­pean man­u­fac­tur­ers, by accel­er­at­ing the devel­op­ment of oth­er tech­nolo­gies deemed key for the next 5–10 years: non-volatile embed­ded mem­o­ries, which will play an essen­tial role in the afore­men­tioned nomadic AI uses, radiofre­quen­cy com­po­nents, which will sup­port the tran­si­tion to 6G appli­ca­tions, and het­ero­ge­neous 3D inte­gra­tion, which will exploit stack­ing to inte­grate new func­tion­al­i­ties on a sin­gle chip. It should be added that we will par­tic­i­pate in the oth­er pilot lines pro­vid­ed for by the Chips Act.

These pilot lines aim to pre­pare for Europe’s indus­tri­al future in the field of semi­con­duc­tors in the short, medi­um and long term. There is some­times talk of dif­fi­cul­ties in col­lab­o­rat­ing at the Euro­pean lev­el: this is far from being the case in the field of micro­elec­tron­ics. We work in close col­lab­o­ra­tion, mak­ing the most of our com­ple­men­tary skills, in order to respond effec­tive­ly to the strate­gic urgency we face.

Interview by Anne Orliac