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