Spin-op­to­elec­tro­nics

With fur­ther mi­ni­a­tu­riza­t­i­on of micropro­ces­sors and in­te­gra­ted cir­cuits de­vice num­ber and den­si­ty in such cir­cuits in­crea­ses ex­po­nen­ti­al­ly. With ex­tre­me­ly small di­men­si­ons a phy­si­cal de­scrip­ti­on of these de­vices has to ac­count for quan­tum me­cha­ni­cal ef­fects. With on­go­ing mi­ni­a­tu­riza­t­i­on pa­ra­si­tic ef­fects in­crea­se their im­pact on de­vice func­tio­na­li­ty, oversha­dowing the in­ten­ded ope­ra­ti­on mode. With spin­tro­nics, the aim is to ex­ploit quan­tum me­cha­ni­cal ef­fects to im­pro­ve and ex­tend func­tio­na­li­ty of con­ven­tio­nal elec­tro­nic de­vices. To do this, the quan­tum me­cha­ni­cal spin or an­gu­lar mo­men­tum of the elec­tron is used for com­pu­ta­ti­on in­s­tead of its char­ge. As op­po­sed to char­ge, spin is not a con­ser­ved quan­ti­ty but ra­ther re­la­xes to equi­li­bri­um where in­for­ma­ti­on is lost.

The field of spin-op­to­elec­tro­nics ap­p­lies the con­cept of spin­tro­nics to op­ti­cal de­vices. This enables novel ef­fects, which can stron­gly en­han­ce the per­for­mance of spin-op­to­elec­tro­nic de­vices in com­pa­ri­son to their con­ven­tio­nal coun­ter­parts.

Our re­se­arch is fo­cu­sed on ef­fi­ci­ent spin in­jec­tion and de­tec­tion at room tem­pe­ra­tu­re and in ma­gne­tic re­ma­nence on the one hand, and with spin ef­fects in con­ven­tio­nal la­sers on the other hand. Our main focus is al­ways to de­ve­lop ap­p­li­ca­ti­on-ori­en­ted de­vices that can ope­ra­te wi­thout cryo­ge­nic coo­ling and strong ex­ter­nal ma­gne­tic fields.

As part of our re­se­arch on spin in­jec­tion and de­tec­tion we suc­cee­ded in me­a­su­ring the spin re­la­xa­ti­on length – i.e. the cha­rac­te­ris­tic spin decay length – as a func­tion both of tem­pe­ra­tu­re and ex­ter­nal ma­gne­tic field.

Our re­sults show that the 1/e decay length at room tem­pe­ra­tu­re and in ma­gne­tic re­ma­nence is as low as 30 nm but in­crea­ses to 50 nm at 2 T ma­gne­tic field and 80 nm at 30 K. This pro­no­un­ces the need to op­ti­mi­ze spin­tro­nic de­vices under ap­p­li­ca­ti­on set­tings as those dif­fer si­gni­fi­cant­ly from usual la­bo­ra­to­ry set­tings [1-3,13].

But most re­le­vant for ap­p­li­ca­ti­ons such as op­ti­cal data com­mu­ni­ca­ti­ons is the laser. For short-haul data trans­mis­si­on sys­tems, e.g. wi­t­hin data cen­ters, ty­pi­cal­ly ver­ti­cal-ca­vi­ty sur­face-emit­ting se­mi­con­duc­tor la­sers (VC­SELs) are used. In par­ti­cu­lar, VC­SELs are sui­ta­ble de­vices for spin-la­sers, be­cau­se the ver­ti­cal geo­me­try is ad­van­ta­ge­ous for the con­ver­si­on of the car­ri­er spin into the po­la­riza­t­i­on of the emis­si­on. To trans­mit data, in­s­tead of the in­ten­si­ty the po­la­riza­t­i­on is mo­du­la­ted in spin-VC­SELs. We could show that the re­so­nan­ce fre­quen­cy of the po­la­riza­t­i­on dy­na­mics is de­cou­p­led from the re­so­nan­ce fre­quen­cy of the in­ten­si­ty dy­na­mics [4,5,9,11,13,16]. Thus, al­ter­na­ti­ve stra­te­gies to en­han­ce the mo­du­la­ti­on band­width can be used. For ex­amp­le, this is pos­si­ble by in­tro­du­cing strain into the de­vice: Via the elas­to-op­tic ef­fect the bi­re­frin­gence in the laser ca­vi­ty can be in­flu­en­ced. The bi­re­frin­gence is the main pro­per­ty of the spin-VC­SEL to con­trol the mo­du­la­ti­on band­width [9]. In order to ob­tain high bi­re­frin­gence, se­ver­al stra­te­gies were in­ves­ti­ga­ted. The bi­re­frin­gence could be in­crea­sed from its ty­pi­cal value of some GHz up to ex­tre­me va­lues of more than 250 GHz uti­li­zing a me­cha­ni­cal ben­ding ap­proach [8]. Fur­ther­mo­re, also in­te­gra­ted me­thods were de­ve­lo­ped, which enable bi­re­frin­gence con­trol via an ad­di­tio­nal elec­tri­cal cur­rent [10] or by a cust­om de­si­gned sur­face gra­ting, which can be im­ple­men­ted du­ring the pro­cess of ma­nu­fac­tu­ring [12].

We de­mons­tra­ted ex­pe­ri­men­tal­ly, that the po­la­riza­t­i­on dy­na­mics in a spin-VC­SEL with high bi­re­frin­gence can reach dy­na­mics of at least 212 GHz [11]. This is 27 times fas­ter than the in­ten­si­ty dy­na­mics in the same laser. Ad­di­tio­nal­ly, two fur­ther be­ne­fits arise when po­la­riza­t­i­on dy­na­mics in spin-VC­SELs are used: Con­ven­tio­nal, in­ten­si­ty mo­du­la­ted la­sers are ty­pi­cal­ly dri­ven at ma­xi­mum bias cur­rent in order to enable them to reach their hig­hest pos­si­ble mo­du­la­ti­on band­width. This re­sults in high en­er­gy con­sump­ti­on and waste heat ge­ne­ra­ti­on. In con­trast, the ma­xi­mum mo­du­la­ti­on band­width of the spin-VC­SEL is de­cou­p­led from the bias cur­rent of the laser. Thus it is pos­si­ble to reach the ma­xi­mum mo­du­la­ti­on band­width al­re­a­dy clo­se­ly above thres­hold cur­rent, re­sul­ting in lower en­er­gy con­sump­ti­on [11].

Ano­ther ad­van­ta­ge of the po­la­riza­t­i­on dy­na­mics is their in­de­pen­dence from tem­pe­ra­tu­re [16]. This can save ad­di­tio­nal coo­ling ef­fort when spin-VC­SELs are used in an ap­p­li­ca­ti­on con­text such as op­ti­cal data trans­mis­si­on. The ul­tra­f­ast spin-VC­SEL was not only in­ves­ti­ga­ted ex­pe­ri­men­tal­ly, but also a very good agree­ment bet­ween ex­pe­ri­men­tal re­sults and si­mu­la­ti­ons uti­li­zing the spin-flip model pro­po­sed by San Mi­guel et al. were ob­tained, es­pe­ci­al­ly by ge­ne­ra­li­zing it towards a pa­ra­me­ter range with ex­tre­me­ly high bi­re­frin­gence va­lues [11,17]. This enables ex­ten­si­ve theo­re­ti­cal in­ves­ti­ga­ti­on by cal­cu­la­ting the be­ha­vi­or of the laser wi­thout ex­pe­ri­ments. Not only re­so­nan­ce oscil­la­ti­ons could be re­pro­du­ced by the model [4-6, 9, 11], but also con­ti­n­uous mo­du­la­ti­on [11] and even swit­ching of re­so­nan­ce oscil­la­ti­ons after an ar­bi­tra­ry num­ber of oscil­la­ti­on pe­ri­ods [7].

The in­jec­tion of the car­ri­er spin po­la­riza­t­i­on into the spin-VC­SEL was per­for­med op­ti­cal­ly in all our pre­vious ex­pe­ri­ments. This re­qui­res ad­di­tio­nal la­sers and op­ti­cal set­ups to ge­ne­ra­te and mo­du­la­te the cir­cu­lar po­la­riza­t­i­on of the in­jec­ted light, thus lo­sing the ad­van­ta­ge of a com­pact and in­te­gra­ted de­vice.

 

Fi­gu­re 3: Ul­tra­f­ast po­la­riza­t­i­on dy­na­mics in a spin-VC­SEL for mode split­tings of 112 GHz (a) and 212 GHz (b). The op­ti­cal spec­tra I show the me­a­su­red mode split­ting. S±Meas are the me­a­su­red tra­ces of the cir­cu­lar in­ten­si­ties, which are used to cal­cu­la­te the cir­cu­ar po­la­riza­t­i­on de­gree PC. PC shows an oscil­la­ti­on with the fre­quen­cy given by the mode split­ting. S±Sim are tra­ces cal­cu­la­ted using the ge­ne­ra­li­zed spin-flip model with very good agree­ment with the ex­pe­ri­men­tal data. The fre­quen­cy f~R of the oscil­la­ti­on in PC‘, which is the re­so­nan­ce fre­quen­cy oft he po­la­riza­t­i­on dy­na­mics, can be tuned using the mode spit­ting (c). From [11]: M. Lin­de­mann, G. Xu, T. Pusch, R. Mich­al­zik, M. R. Hof­mann, I. Žutić and N. C. Ger­hardt, Ul­tra­f­ast spin-la­sers, Na­tu­re 568, 212 (2019).

The­re­fo­re, for the ap­p­li­ca­ti­on in a data cen­ter the op­ti­cal spin in­jec­tion needs to be re­pla­ced by elec­tri­cal spin-in­jec­tion, which is, howe­ver, chal­len­ging due to the as­pects which have been men­tio­ned above in the con­text of the spin-LED: A spin-VC­SEL has not yet been de­mons­tra­ted at room tem­pe­ra­tu­re, as the in­jec­tion path length bet­ween in­jec­tion con­tacts and quan­tum wells in the laser ca­vi­ty is too long for the spin-po­la­ri­zed car­riers to pass wi­t­hin the spin-li­fe­time at room tem­pe­ra­tu­re. The spin-li­fe­time can be ex­ten­ded by coo­ling the de­vice, hence spin-VC­SELs could so far only be de­mons­tra­ted at tem­pe­ra­tu­res up to 230 K by Basu et al., re­qui­ring cryo­ge­nic coo­ling. As a so­lu­ti­on we have pro­po­sed a de­vice ar­chi­tec­tu­re that of­fers much shor­ter in­jec­tion path lengths, which should enable room tem­pe­ra­tu­re ope­ra­ti­on [14,15]. This con­cept com­pri­ses a high­ly re­flec­tive sur­face gra­ting re­pla­cing the upper Bragg mir­ror, which also con­tains spin-in­jec­tion con­tacts wi­t­hin its groo­ves. With the de­ve­lop­ment of room tem­pe­ra­tu­re de­vices, the spin-VC­SEL poses a pro­mi­sing al­ter­na­ti­ve to the con­ven­tio­nal data trans­mis­si­on tech­no­lo­gy, pro­mi­sing an order of ma­gni­tu­de hig­her band­width and an order of ma­gni­tu­de lower en­er­gy con­sump­ti­on.

Re­fe­rence:

  • [1] Hen­ning Sol­dat, Min­gyuan Li, Arne Lud­wig, As­trid Lud­wig, Frank Strom­berg, Heiko Wende, Wer­ner Keune, Dirk Reu­ter, An­dre­as D. Wieck, Nils C. Ger­hardt, Mar­tin R. Hof­mann, Ma­gne­tic field de­pen­dence of the spin re­la­xa­ti­on length in spin light-emit­ting di­odes, Ap­p­lied Phy­sics Let­ters 99 (5), 051102 (2011)
  • [2] Hen­ning Höpf­ner, Ca­ro­la Frit­sche, Arne Lud­wig, As­trid Lud­wig, Frank Strom­berg, Heiko Wende, Wer­ner Keune, Dirk Reu­ter, An­dre­as D. Wieck, Nils C. Ger­hardt, Mar­tin R. Hof­mann, Ma­gne­tic field de­pen­dence of the spin re­la­xa­ti­on length in spin light-emit­ting di­odes, Ap­p­lied Phy­sics Let­ters 101 (11), 112402 (2012)
  • [3] A. Lud­wig, B. Soth­mann, Hen­ning Höpf­ner, Nils C. Ger­hardt, J. Nan­nen, T. Küm­mell, J. König, Mar­tin R. Hof­mann, G. Ba­cher, A.D. Wieck, Quan­tum dot spin­tro­nics: fun­da­men­tals and ap­p­li­ca­ti­ons, in H. Zabel and M. Farle (eds.), Ma­gne­tic Na­nost­ruc­tu­res, Sprin­ger Tracts in Mo­dern Phy­sics 246, 235 (2013)
  • [4] Min­gyuan Li, Hen­drik Jähme, Hen­ning Sol­dat, Nils C. Ger­hardt, Mar­tin R. Hof­mann, Thors­ten Acke­mann, Bi­re­frin­gence con­trol­led room-tem­pe­ra­tu­re pi­co­se­cond spin dy­na­mics close to the thres­hold of ver­ti­cal-ca­vi­ty sur­face-emit­ting laser de­vices, Ap­p­lied Phy­sics Let­ters 97 (19), 191114 (2010)
  • [5] Nils C. Ger­hardt, Min­gyuan Li, Hen­drik Jähme, Hen­ning Höpf­ner, Thors­ten Acke­mann, Mar­tin R. Hof­mann, Ul­tra­f­ast spin-in­du­ced po­la­riza­t­i­on oscil­la­ti­ons with tunable li­fe­time in ver­ti­cal-ca­vi­ty sur­face-emit­ting la­sers, Ap­p­lied Phy­sics Let­ters 99 (15), 151107 (2011)
  • [6] Nils C. Ger­hardt and Mar­tin R. Hof­mann, Spin-Con­trol­led Ver­ti­cal-Ca­vi­ty Sur­face-Emit­ting La­sers, Ad­van­ces in Op­ti­cal Tech­no­lo­gies 2012, 268949 (2012)
  • [7] Hen­ning Höpf­ner, Mar­kus Lin­de­mann, Nils C. Ger­hardt, and Mar­tin R. Hof­mann, Con­trol­led swit­ching of ul­tra­f­ast cir­cu­lar po­la­riza­t­i­on oscil­la­ti­ons in spin-po­la­ri­zed ver­ti­cal-ca­vi­ty sur­face-emit­ting la­sers, Ap­p­lied Phy­sics Let­ters 104, 022409 (2014)
  • [8] To­bi­as Pusch, Mar­kus Lin­de­mann, Nils C. Ger­hardt, Mar­tin R. Hof­mann and Rai­ner Mich­al­zik, Ver­ti­cal-ca­vi­ty sur­face-emit­ting la­sers with bi­re­frin­gence split­ting above 250 GHz, Elec­tro­nics Let­ters 51, 1600 (2015)
  • [9] Mar­kus Lin­de­mann, To­bi­as Pusch, Rai­ner Mich­al­zik, Nils C. Ger­hardt, and Mar­tin R. Hof­mann, Fre­quen­cy tu­ning of po­la­riza­t­i­on oscil­la­ti­ons: Toward high-speed spin-la­sers, Ap­p­lied Phy­sics Let­ters 108, 042404 (2016)
  • [10] To­bi­as Pusch, Eros La Tona, Mar­kus Lin­de­mann, Nils C. Ger­hardt, Mar­tin R. Hof­mann, and Rai­ner Mich­al­zik, Mo­no­li­thic ver­ti­cal-ca­vi­ty sur­face-emit­ting laser with thermal­ly tunable bi­re­frin­gence, Ap­p­lied Phy­sics Let­ters 110, 151106 (2017)
  • [11] Mar­kus Lin­de­mann, Gao­feng Xu, To­bi­as Pusch, Rai­ner Mich­al­zik, Mar­tin R. Hof­mann, Igor Žutić and Nils C. Ger­hardt, Ul­tra­f­ast spin-la­sers, Na­tu­re 568, 212 (2019)
  • [12] To­bi­as Pusch, Pier­lu­i­gi De­ber­nar­di, Mar­kus Lin­de­mann, Frie­de­ri­ke Erb, Nils C. Ger­hardt, Mar­tin R. Hof­mann and Rai­ner Mich­al­zik, Ver­ti­cal-ca­vi­ty sur­face-emit­ting laser with in­te­gra­ted sur­face gra­ting for high bi­re­frin­gence split­ting, Elec­tro­nics Let­ters 55, 1055 (2019)
  • [13] Igor Zutic, Je­ong­su Lee, Chris­ti­an Go­th­gen, Paolo E. Faria Ju­ni­or, Gao­feng Xu, Guilg­he­me M. Si­pahi, Nils C. Ger­hardt, “Ch­ap­ter 16 – Se­mi­con­duc­tor Spin-La­sers”, Spin­tro­nic Hand­book: Spin Trans­port and Ma­gne­tism, Vo­lu­me 3: Na­no­s­ca­le Spin­tro­nics and Ap­p­li­ca­ti­ons, 2. Ed., edi­ted by E.Y. Tsym­bal and I. Zutic, (2019)
  • [14] Pa­tent: Mar­kus Lin­de­mann, Mar­tin R. Hof­mann, Nils C. Ger­hardt, WO002019170517A1 (2019)
  • [15] Mar­kus Lin­de­mann, Ul­tra­schnel­le Spin-La­ser (Sprin­ger), 2020
  • [16] Mar­kus Lin­de­mann, Na­ta­lie Jung, Pas­cal Stad­ler, To­bi­as Pusch, Rai­ner Mich­al­zik, Mar­tin R. Hof­mann, and Nils C. Ger­hardt, Bias cur­rent and tem­pe­ra­tu­re de­pen­dence of po­la­riza­t­i­on dy­na­mics in spin-la­sers with elec­tri­cal­ly tunable bi­re­frin­gence, AIP Ad­van­ces 10, 035211 (2020)
  • [17] Igor Žutić, Gao­feng Xu, Mar­kus Lin­de­mann, Paulo E. Faria Ju­ni­or, Je­ong­su Lee, Ve­li­mir La­bi­n­ac, Kris­ti­an Stojšić, Guil­her­me M. Si­pahi, Mar­tin R. Hof­mann, Nils C. Ger­hardt, Spin-la­sers: spin­tro­nics bey­ond ma­gne­to­re­sis­tan­ce, Solid State Com­mu­ni­ca­ti­ons 316, 113949 (2020)

Col­le­agues:

Postal Address

Ruhr-University Bochum
Faculty of Electrical Engineering
and Information Technology
Photonics and Terahertz Technology
Postbox ID 16
Universitätsstraße 150
D-44801 Bochum

Contact

Office  
Room: ID 04/327
Te­l.: (+49) (0) 234 32 - 23051
Fax: (+49) (0) 234 32 - 14167
E-Mail: ptt+office(at)rub.de
RUB campus map & travel instructions

Chair Holder

Prof. Dr.-Ing. Martin Hofmann
Building: ID 04/329
Te­l.: (+49) (0) 234 32 - 22259
Fax: (+49) (0) 234 32 - 14167
E-Mail: martin.hofmann(at)rub.de

To Top