الفهم الجزيئي للدور الحاسم لأيونات المعادن القلوية في بدء اختزال ثاني أكسيد الكربون على سطح Cu(100) Molecular understanding of the critical role of alkali metal cations in initiating CO2 electroreduction on Cu(100) surface

المجلة: Nature Communications، المجلد: 15، العدد: 1
DOI: https://doi.org/10.1038/s41467-024-44896-x
PMID: https://pubmed.ncbi.nlm.nih.gov/38242907
تاريخ النشر: 2024-01-19

الفهم الجزيئي للدور الحاسم لأيونات المعادن القلوية في بدء اختزال ثاني أكسيد الكربون على سطح Cu(100)

تاريخ الاستلام: 30 مايو 2023
تم القبول: 9 يناير 2024
نُشر على الإنترنت: 19 يناير 2024
(أ) التحقق من التحديثات

تشاوتشاو تشانغ © هينغيو لي يانغفان شاو لين غان فييو كانغ وينهوي دوان هاين أنطون هانسن © وجيا لي (ب)

الملخص

فهم الجزيئات لواجهة الصلب-السائل هو أمر صعب ولكنه ضروري لتوضيح دور البيئة على حركية التفاعلات الكهروكيميائية. أيونات المعادن القلوية ( )، كعنصر حيوي في الواجهة، يُعتبر ضروريًا لبدء تفاعل اختزال ثاني أكسيد الكربون ( ) على المعادن المستخدمة في سك العملات، ونشاطها وانتقائيتها يمكن أن يتم تحسينه بشكل أكبر مع تغيير الكاتيون من إلى بينما الآليات الأساسية ليست مفهومة جيدًا. هنا، باستخدام محاكاة الديناميات الجزيئية من أولى المبادئ مع ترطيب صريح وطرق تحسين العينة، نقوم بالتحقيق بشكل منهجي في دور في على سطح النحاس. تناقص أحادي الاتجاه حاجز التنشيط يتم الحصول عليه من إلى الذي يُنسب إلى القدرات التنسيقية المختلفة لـ مع * . علاوة على ذلك، نوضح أن تفاعل تطور الهيدروجين المتنافس يجب أن يؤخذ في الاعتبار في الوقت نفسه لفهم الدور الحاسم لأيونات المعادن القلوية في على أسطح النحاس، حيث يتم دفعه بعيدًا عن الواجهة ومقيد بـ توفر نتائجنا رؤى مهمة في تصميم البيئات الكهروكيميائية وتبرز أهمية تضمين الحلول والتفاعلات المتنافسة بشكل صريح في المحاكاة النظرية لـ .

تفاعل اختزال ثاني أكسيد الكربون الكهروكيميائي لقد ظهرت ( ) كطريقة جذابة لإغلاق دورة الكربون في الوقت نفسه وإنتاج المواد الأولية الصناعية الهامة بسبب توافقها مع مصادر الطاقة المتجددة الخالية من الكربون. تعتبر المواد القائمة على النحاس فريدة بشكل خاص بين الأنواع المختلفة من المحفزات التي تم تطويرها اليوم، حيث يمكنها إنتاج إلكترونات متعددة ( منتجات مختزلة مثل الميثان بالإضافة إلى الكربون المتعدد منتجات تتكون أساسًا من الإيثيلين والإيثانول بغض النظر عن هياكلها المختلفة، مثل رقائق النحاس جزيئات نانوية وحتى المحفزات الجزيئية . ومع ذلك، لم تصل نشاطية واختيارية هذه المحفزات بعد إلى المستوى المطلوب للتطبيق الصناعي. لذلك،
تم بذل العديد من الجهود لتحسين أدائها الكهروكيميائي بشكل أكبر، خاصة من أجل المزيد من القيمة. المنتجات. تشمل هذه الجهود تصميم الواجهة هندسة العيوب معالجات مسبقة مثل المشتقات المؤكسدة وهندسة الإلكتروليت . ومع ذلك، فإن العوامل الحاسمة والرؤى على المستوى الذري حول الآليات الأساسية لا تزال غير واضحة.
، كرد فعل كيميائي كهربائي مائي تمثيلي، يتأثر بشكل كبير ليس فقط بالمواد الحفازة ولكن أيضًا بالبيئة عند واجهة الصلب-الإلكتروليت. منذ التسعينيات، اكتشف هوري وزملاؤه أن أيونات المعادن القلوية ( ) يمكن أن يعزز النشاط بأكثر من ترتيب من حيث الحجم وتحسينه في الوقت نفسه
الانتقائية على قطب النحاس. يصبح هذا التأثير الترويجي أكثر أهمية مع الكاتيونات الأكبر إلى تم تأكيد تأثير الترويج لأيونات المعادن القلوية. وتمتد إلى أخرى المحفزات، مثل المحفزات الانتقائية لثاني أكسيد الكربون ، ، واختيار HCOOH تم تطوير عدة تفسيرات حتى الآن. نموذج الرقم الهيدروجيني السطحي نسب الفروق بين كاتيونات المعادن القلوية إلى البروتونات البينية و تركيزات ناتجة عن التوازن بين الأنيونات لمجموعة الكربونات لكن العلاقة بين هذا النموذج وآلية التفاعل ليست مفهومة جيدًا. بعد ذلك، تم اقتراح نموذج تفاعل ثنائي القطب-المجال، الذي يبرز تأثير الاستقرار للمجال الكهربائي بين الكاتيونات المعدنية القلوية المذابة والسطح المعدني المشحون سلبًا على الوسائط القطبية. . لذلك، فإن الوسطاء الذين لديهم ثنائيات قطبية كبيرة تشمل و (النجمة * تشير إلى موقع الامتزاز) ستستقر بشكل قوي، مما يعزز انتقائية المنتجات المقابلة مثل CO على و على النحاس. ومع ذلك، يعتمد هذا النموذج على نموذج المجال الكهربائي السطحي الموحد وينسب التباين بين الكاتيونات المختلفة للمعادن القلوية إلى ميل الكاتيونات الأكبر للتجمع أكثر عند الواجهة مع تجاهل الاختلافات الجوهرية بين الكاتيونات الفردية للمعادن القلوية والتفاعلات المحلية المحتملة بين الكاتيونات والوسائط.
مؤخراً، تم إثبات المزيد أن الـ لا يمكن البدء في ، و بدون كاتيونات المعادن القلوية، حتى إلى CO و ، والتي لا يمكن تفسيرها بالكامل من خلال النماذج المذكورة أعلاه. بالإضافة إلى ذلك، أظهرت النتائج الطيفية أن إدخال ، لكن ليس تؤدي الكاتيونات إلى أعلى حقل كهربائي في طبقة ستيرن، بينما تتغير قوة حقل التفاعل الناتج عن الإذابة، وهو حقل أونساجر، مع حجم الكاتيون. تشير هذه النتائج إلى أن الكاتيونات المختلفة من المعادن القلوية لديها اختلافات جوهرية بدلاً من مجرد تفاوت في التركيز، وأن الكاتيونات عند الواجهة يمكن أن تؤثر بشكل كبير على كفاءة الـ عملية تحويل to-CO، مما قد يلعب دورًا أكثر نشاطًا بدلاً من مجرد التواجد كمشاهدين. لتقديم شرح أكثر تفصيلًا لتأثير الكاتيونات المعدنية القلوية المذابة على آليات التفاعل، تم إجراء محاكاة الديناميكا الجزيئية من أولى المبادئ (AIMD) وكشفت عن وجود تفاعلات محلية، مثل التنسيق بين الكاتيونات المذابة والوسائط السطحية. تشمل الأمثلة البارزة التفاعل بين كاتيون ويمتص جزيء على سطح الذهب ، والتفاعل بين الكاتيونات وثنائي *CO + *CO على سطح النحاس . ومع ذلك، فإن الفهم الشامل لتأثير الكاتيونات المختلفة من المعادن القلوية على
التفعيل ودورهما الحاسم في لا يزال هناك نقص في التهيئة على سطح النحاس.
في هذا العمل، درسنا بشكل منهجي تأثير كاتيونات المعادن القلوية على كل من وتفاعل تطور الهيدروجين الرئيسي المتنافس (HER)، باستخدام محاكاة AIMD مع طرق أخذ عينات محسّنة وأخذ تأثيرات الذوبان والجهد في الاعتبار بشكل صريح. من خلال حساب حواجز الطاقة الحرة لـ التفعيل، اتجاه متناقص بشكل أحادي إلى يتم العثور عليه، وهو ناتج عن القدرة على التنسيق بين كاتيون المعدن القلوي و* متوسط. تنسيق الجسر بين الكاتيونات الأكبر و يساعد في تسهيل أكثر مقارنة بالتكوين الجانبي المفضل من قبل الكاتيونات الأصغر. علاوة على ذلك، نوضح الدور الضروري لكاتيونات المعادن القلوية في على أسطح النحاس لا يمكن فهمها بالكامل دون النظر في نفس الوقت إلى تفاعل الهيدروجين التنافسي. تسلط نتائجنا الضوء على التأثير الاستقراري القصير المدى الجوهري لأيونات المعادن القلوية الفردية على وتسليط الضوء على ضرورة النظر بشكل صريح في البيئة، مثل الكاتيونات والذوبان، بالإضافة إلى التفاعلات المتنافسة، في المحاكاة الكهروكيميائية من أجل الحصول على رؤى أكثر موثوقية.

النتائج والمناقشة

استجابة الطبقة المزدوجة الكهربائية للجهد المطبق

لتوفير معيار للتحليل اللاحق، قمنا أولاً بإجراء محاكاة لنظام واجهة النحاس – الماء مع وبدون الامتزاز، يُشار إليه بـ و ، على التوالي، للتحقيق في استجابة الطبقة الكهربائية المزدوجة (EDL) لجهود مختلفة مطبقة. توضح الأشكال 1a-c توزيع كثافة الماء عمودياً على الواجهة تحت ظروف مختلفة. يمكن ملاحظة ثلاثة قمم رئيسية فوق واجهة النحاس – الماء تقريباً ، و Å، على التوالي. يُنسب الذروة الأولى (المعلمة بالخط الرمادي المتقطع في الشكل 1a-c) إلى جزيئات الماء الملتصقة كيميائيًا على سطح النحاس، بينما تشير الذروتان الثانية والثالثة إلى طبقتين من الماء مرتبتين جزئيًا. لم تُلاحظ أي ذروات واضحة بعيدًا عن الواجهة، مما يمثل طبقة الانتشار والماء الكثيف بكثافة تبلغ حوالي ، مما يدل على السماكة الكافية لطبقة الماء أيضًا. علاوة على ذلك، مع انخفاض الجهد المطبق، تنخفض شدة القمة الأولى، كما هو موضح في الشكل 1a، والذي يمكن أن يُعزى إلى زيادة التفاعل الطارد بين الركيزة المشحونة سلبًا بشكل أكبر وذرة الأكسجين المشحونة سلبًا جزئيًا. الجزيء (كما هو موضح بشكل تخطيطي في الشكل 1d). في الوقت نفسه، ستنجذب ذرات الهيدروجين بواسطة
الشكل 1 | استجابة الهيكل الواجهى للجهد المطبق.
توزيع كثافة الماء على طول الاتجاه العمودي على -واجهة الماء تحت ظروف مختلفة، مع طبقة النحاس العلوية على بعد 0. الخط الرمادي المتقطع يشير إلى الذروة الأولى للامتصاص الكيميائي
الماء. الشكل التخطيطي لتغير هيكل الواجهة تحت الظروف المقابلة لـ (أ-ج). تُصوَّر الجزيئات ككرات رمادية وحمراء، بينما تم تمثيل الجزيئات كنماذج عصا من أجل الوضوح. يتم الحصول على الجهود المطبقة من وظيفة العمل للنظام بأكمله. (ملاحظة إضافية 1).
الشكل 2 | مخطط للتمثيل * هياكل الامتزاز. التفاعلات المحلية بين * في السطح والبيئة المحيطة: مخطط لامتصاص ثنائي الأسنان لـ * على سطح النحاس في مذيب صريح مع C وواحد من ذرات الأكسجين يتنسق مع النحاس، وذرة الأكسجين الأخرى تشكل
روابط هيدروجينية (خطوط متقطعة زرقاء فاتحة) مع الماء المحيط. مخطط التكوين الجانبي لـ أو منسق مع ذرة أكسجين واحدة فقط من * (خط متقطع باللون الأرجواني). ج رسم تخطيطي لتكوين الجسر أو منسق مع كل من ذرتي الأكسجين .
سطح، مما يسبب الرابطة تدور نحو السطح، كما يتضح من التغير في ثنائيات القطب المائية السطحية داخل Åتتغير توزيع ثنائي القطب من عشوائي إلى قمة واحدة، مما يشير إلى توزيع أكثر انتظامًا لجزيئات الماء في الطبقة الأولى من الماء.
تمت ملاحظة استجابة مماثلة لترتيب الماء الواجهاتي للجهد المطبق في نظام (الشكل التوضيحي الإضافي 2)، حيث تم إدخال الجزيء على سطح النحاس النظام. من الناحية الطاقية، فإنenthalpies الامتصاص تنخفض مع انخفاض الجهد المطبق، مما يشير إلى امتصاص أكثر استقرارًا لـ (الجدول التكميلي 1). خلال الـ 15 بيكوثانية الإضافية من محاكاة AIMD، تمت ملاحظته أنه يتم امتصاصه بشكل مستقر على سطح النحاس في تكوين ثنائي الأسنان من خلال و الروابط، مع ذرة الأكسجين الأخرى موجهة لأعلى وتشكل روابط هيدروجينية مع جزيئات الماء المحيطة (تكوين الامتصاص في الشكل 1d). تحليل شحنة بادير يظهر أن الكيميائي الممتص له شحنة تبلغ حوالي ، مما يشير إلى حالة امتصاص مفعلة بالكامل. بالإضافة إلى ذلك، فإن إدخال يعزز امتصاص الماء الكيميائي (الشكل 1ب)، ولكن هذا يرجع إلى نقص التحكم في الجهد (حالة Const-Q في الشكل 1د). بعد ضبط الجهد المطبق ليكون متسقًا مع النظام من خلال تبادل الإلكترونات مع خزان الإلكترونات، فإن شدة القمة الأولى تكاد تكون متطابقة مع المرجع (الشكل 1c وحالة Const-U في الشكل 1d). تؤكد النتائج المذكورة أعلاه صحة نموذجنا وتوفر الأساس لمزيد من التحقيق في الواجهة الكهروكيميائية خلال عملية .
علاوة على ذلك، يجدر بالذكر أن التكوين ثنائي الأسنان لـ * على تختلف السطح عن ذلك الذي لوحظ في بيئة مذيبة ضمنية بالكامل. في الحالة الأخيرة، يميل إلى أن يكون م adsorbed على سطح على نطاق محتمل واسع (الشكل التكميلي 3)، بينما التكوين ثنائي الأسنان أقل استقرارًا بـ عند -0.6 فولت مقابل القطب الهيدروجيني القياسي (SHE)، وهو نطاق الجهد المعني في هذا العمل. تشير هذه الفجوة إلى المساهمة الكبيرة للروابط الهيدروجينية المحيطة في استقرار من الحل المائي الصريح، وبالتالي يثير ملاحظة تحذيرية بشأن استخدام المذيبات الكاملة غير الصريحة للتحقيق في التفسيرات الجزيئية لآليات التفاعل، خاصة عندما تلعب التفاعلات بين الوسائط والبيئة الكهروكيميائية المحيطة، مثل جزيئات الماء المذابة، دورًا كبيرًا.

تعزيز تأثيرات كاتيونات المعادن القلوية على امتصاص

أظهرت النتائج التجريبية أن المنتجات، بما في ذلك CO، لا يمكن اكتشافها على المعادن النقدية دون في الإلكتروليت . هذا يشير إلى أن يتم تثبيطه في المراحل المبكرة من -عملية تحويل CO. للتحقيق في الآلية الأساسية، هنا ركزنا أولاً على تأثير كاتيونات المعادن القلوية على الخطوة المحددة لمعدل إلى CO، ال عملية التفعيل الذي من المتوقع أن يكون له تفاعلات كهرستاتيكية
مع كاتيونات المعادن القلوية بسبب تحولها من تكوين غير قطبي خطي إلى تكوين قطبي منحنٍ. درسنا الـ النظام بمزيد من التفاصيل ثم أضاف ، أو أزواج الأيونات فيه للحفاظ على الجهد المطبق قريبًا من PZC لـ بالإضافة إلى إمكانيات البداية لـ في ضد هي التي تُعرف باسم الأنظمة. تؤكد تحليل شحنة بادير على التأين الكامل للأيونات المضافة، مع لأيونات القلوي لأيونات الفلور، على التوالي. باستخدام زاوية كمتغير جماعي (CV) خلال عملية الامتزاز، وجدنا أن المتغيرات الجماعية للحالة الأولية (IS، خطي حر ) والحالة النهائية (FS، حالة كيميائية ممتصة منحنية لا تختلف كثيرًا مع الكاتيونات المختلفة، حيث تصل إلى ذروتها عند و ، على التوالي (الشكل التكميلي 4). زاوية في IS ينحرف عن بسبب التقلبات الحرارية والتفاعلات مع الماء، والتكوين الكيميائي الممتص لـ * في FS هو مطابق لذلك في النظام (الشكل 2 والشكل التكميلي 4). هذه النتائج توضح بشكل أكبر الدور الهام للروابط الهيدروجينية التي تتكون بين والأماكن المحيطة جزيئات.
الأهم من ذلك، وجدنا أن جميع كاتيونات المعادن القلوية يمكن أن تتناسق بشكل مستقر مع * لكن مع ميزات تنسيق مختلفة. و الأيونات تتنسق مع ذرة أكسجين واحدة فقط من * (تكوين الجانب، كما هو موضح في الشكل 2ب)، بينما و سوف تنسق مع كلا ذرتي الأكسجين * (تكوين الجسر، كما هو موضح في الشكل 2c). يمكن للمرء أن يفترض بشكل طبيعي أن تنسيق الجسر ناتج عن حجم الكاتيون الأكبر، وهو ما تدعمه زيادة نصف قطر الغلاف الأول من الإذابة ( ) من إلى (الجدول التكميلي 2). إذا تم الحفاظ على اختلافات التكوين خلال عملية التنشيط، فقد يكون هذا الظاهرة سببًا جوهريًا للاختلافات في التأثيرات الترويجية على من بأحجام مختلفة. تكوين الجسر أكثر ملاءمة لـ الانحناء كنتيجة للاحتكاك الكهروستاتيكي بين وذرة الكربون المشحونة جزئيًا بشحنة إيجابية من . من وجهة نظر طاقية، اعتبرنا أولاً طاقة الامتزاز الحرة مع كاتيونات المعادن القلوية المختلفة كتقريب لوصف النشاط. النظام مع يظهر أقوى امتصاص لـ * بينما لا تختلف طاقات الامتزاز الحرة مع الكاتيونات الثلاثة الأخرى كثيرًا عن بعضها البعض (الشكل 3 والجدول التكميلي 3). وبالتالي، لم يكن من الممكن ملاحظة أي اتجاه واضح مع حجم الكاتيون، وهو ما يتماشى أيضًا مع النتائج السابقة حول كما تم إثباته مرارًا وتكرارًا ، فإن طاقة التفاعل وحدها لا يمكن أن تصف نشاط النظام بشكل كامل، بينما قد تكون طاقة التنشيط من الحالة الابتدائية إلى حالة الانتقال (TS) أكثر أهمية لتكون المعايير الطاقية الحقيقية. لذلك، من أجل الحصول على معلومات أكثر دقة وتفصيلاً حول تأثيرات الكاتيونات على عملية الامتزاز، ركزنا بعد ذلك على حاجز الطاقة الحرة وحالات الانتقال لـ عملية الامتزاز في .
الشكل 3أ يظهر ملف الطاقة الحرة لـ التفعيل في نظام في غياب ، حيث يتم الإشارة إلى حاجز حركي قدره 0.22 إلكترون فولت. للمقارنة، توضح الشكل 3c-f ملفات الطاقة الحرة لـ التفعيل مع الأيونات في المذيب. الحركية
الشكل 3 | ملفات الطاقة الحرة والهياكل التمثيلية لـ عملية التفعيل على سطح. ملف الطاقة الحرة وهياكل الحالة الابتدائية (IS) والحالة الانتقالية (TS) والحالة النهائية (FS) التمثيلية لـ عملية الامتزاز في الماء النقي ) نظام بدون تشير الخطوط الرمادية المتقطعة إلى القيمة الصفرية. تدرج الطاقة الحرة المقابل مع شريط الخطأ عند كل متغير جماعي محسوب. تشير نقاط التقاطع مع خط أساسي متقطع إلى مواقع IS وTS وFS. تمثل أشرطة الخطأ الأخطاء المعيارية لتدرجات الطاقة الحرة المحسوبة باستخدام المتوسط الكتلي عند المتغير الجماعي المقابل. ج-و ملفات الطاقة الحرة والهياكل التمثيلية عند IS وTS وFS لـ
الامتصاص مع كاتيونات مختلفة ) من ، و تمثل الظلال في جميع ملفات الطاقة الحرة الخمسة الأخطاء التراكمية ( في الهياكل المحلية، تُصوَّر الجزيئات ككرات رمادية وحمراء، بينما تُظهر الركائز النحاسية ككرات بنيّة محمرة. تُصوَّر الجزيئات على أنها عصي حمراء وبيضاء، و بما في ذلك ، و تمثلها كرات خضراء وصفراء وبنفسجية وزرقاء، على التوالي. الخطوط المتقطعة باللون الأرجواني حول توضيح تنسيقها مع ذرات الأكسجين في أغلفة الترطيب الأولى.
تظهر الحواجز اتجاهًا واضحًا ومتناقصًا بشكل أحادي من النظام إلى الأنظمة، مع تضمين ، و . هذه الاتجاه يتماشى مع الاتجاه الملحوظ للنشاط مع كاتيونات المعادن القلوية في التجارب . علاوة على ذلك، يتم تقسيم الكاتيونات عمومًا إلى مجموعتين خلال عملية التنشيط بناءً على حواجز التنشيط وخصائص التنسيق الخاصة بها مع * : كاتيونات أصغر من / وأيونات أكبر من (تم مناقشة التحقق من التكوينات المستقرة في الملاحظة التكميلية 4). فرق حاجز الطاقة الحرة بين المجموعتين حوالي 0.05 إلكترون فولت، مما يمكن أن يزيد من معدل التفاعل المقابل بمقدار 5 مرات، وهو ما يتوافق مع النتائج التجريبية. .
حاجز التنشيط في الـ عملية الانحناء ونقل الشحنة المتزامن نتيجة كسر التناظر لـ (أدنى مدار جزيئي غير مشغول، ليمو) و حالات أعلى مستوى مشغول من المدارات الجزيئية (HOMO) . في الوقت نفسه، مستوى LUMO المنخفض للزاوية يسهل التبرع العكسي للإلكترونات من الركيزة إلى ، مما يؤدي إلى الشحنة السالبة . يمكن تعزيز هذه العملية بشكل أكبر بواسطة كاتيونات المعادن القلوية (الشكل 3). في الـ النظام، السيرة الذاتية في TS تقريبًا ، لكنه يقترب من الحالة المستقرة مع كاتيونات المعادن القلوية. في الوقت نفسه، الشحنات السلبية على في عرض TS يظهر اتجاه مشابه يتناقص بشكل أحادي، بحوالي نُقل إلى في النظام، بينما حوالي يتم نقله في النظام (الجدول التكميلي 4). وهذا يعني أيضًا أن الحالة الانتقالية هي الأقرب إلى الحالة الابتدائية مع الكاتيون، الذي يستفيد من أدنى حاجز تنشيط له.
الهياكل التنسيقية لـ في الغلاف الأول من الإذابة مع وبدون تمت دراسة البيانات في TS بشكل أعمق لفهم الآلية الجزيئية الأساسية لـ التفعيل (الشكل 4أ)، مع الأنظمة المشار إليها بـ و ، على التوالي. لـ و الكاتيونات، التنسيق مع يتضمن تبادل ذرة أكسجين من في الغلاف الأول من الإذابة، مما يؤدي إلى عدد تنسيق إجمالي يبدو غير متغير. هذا التبادل يرجع أساسًا إلى أغلفة الإذابة الصلبة للأيونات الصغيرة،
التي بالكاد يمكن أن تقبل المزيد من جزيئات الإذابة مع البيئة، كما هو موضح في المنصة في دالة التوزيع الشعاعي المتكاملة (RDF) عند نصف قطر الغلاف الأول للإذابة في الشكل التكميلية 5. ومع ذلك، بالنسبة لـ و الأيونات الموجبة، زادت أرقام التنسيق بمقدار 0.9 و 2.2، على التوالي. وذلك لأن و يمكن أن يتنسيق مع أكثر من أكسجين من والتنسيق مع عدد التنسيق الأعلى لـ و الكاتيونات مع هيكل التنسيق الثنائي الفريد لها * يمكن أن يسهل عملية نقل الإلكترون ويساعد في تنشيط بشكل أكثر فعالية من الكاتيونات الأصغر، مما يوفر رؤى مهمة حول الأصل الجزيئي لتأثيرها الترويجي الأعلى على نشاط. بالإضافة إلى ذلك، هناك فرق في حاجز التنشيط بين و الأيونات الموجبة، ولكن تم العثور عليها تجريبيًا لتنشيط بالمثل ، والذي يمكن تفسيره أيضًا من خلال هيكل التنسيق الخاص بهم. مقارنةً بأعداد التنسيق الكلية لأيونات المعادن القلوية في نظام مع أولئك في الأنظمة (الشكل التوضيحي التكميلي 6)، تبقى معظمها دون تغيير باستثناء انخفاض في الكاتيون عند الواجهة بدون . في محاكاة AIMD، لاحظنا أن الكاتيون يفقد بعض المحلول الجزيئات عند واجهة (الشكل التوضيحي 7)، ولكن عندما تتنسق مع * قدرتها على الإذابة مع تعافى قليلاً في واجهة (الشكل 4أ). تم إزالة بعض المذيبات المركب يتصرف مثل الامتزاز شبه المحدد الذي قد يسد المواقع النشطة ويكون غير مواتٍ للتفاعلات السطحية، مما يؤدي في النهاية إلى تأثيره التنشيطي المماثل على مع .
الشكل 4 ب يوضح أيضًا التوزيع الإحصائي للمسافة بين كاتيونات المعادن القلوية والأكسجينين. ، و و تمت ملاحظة أن الكاتيونات تتنسق بشكل مستقر مع أكسجين واحد فقط، مما يؤدي إلى زيادة أكبر المسافة مقارنة بـ . على النقيض، بالنسبة لـ و الكاتيونات، يؤدي تكوين الجسر المفضل إلى مسافة متساوية من و . فرق آخر بين هذين المجموعتين من الكاتيونات هو كثافة التوزيعات، حيث تظهر الكاتيونات الأكبر توزيعًا أوسع، مما يتماشى مع غياب المنصات في RDF المدمج (الشكل التوضيحي التكميلي 5c،
الشكل 4 | تحليل التنسيق للغلاف الأول من الترطيب لأيونات المعادن القلوية حول TS. أ عدد الأكسجين المنسق لـ ، و الكاتيونات في -واجهة الإلكتروليت مع وبدون خريطة الكنتور لـ

توزيع المسافة بين كاتيونات المعادن القلوية والأكسجينين مستمدة إحصائيًا من لقطات حول حالة الانتقال. نتائج ، و تمثل بالأخضر والبرتقالي والبنفسجي والأزرق، على التوالي.
هذا يشير إلى تنسيق أكثر ديناميكية حول مقارنة بـ و الكاتيونات. بالإضافة إلى ذلك، لاحظنا تراكم الإلكترونات بين الـ الكاتيون والأكسجين المنسق لـ ، وهو ما يعد السمة المميزة للرابطة. ومع ذلك، لم يتم ملاحظة تراكم شحنات كبيرة لـ الكاتيون (الشكل التوضيحي التكميلي 8)، يدعم التحليل أعلاه. بهذه الطريقة، يتم تنسيق المزيد من الديناميات مع * يمكن أن يسرع أيضًا دورة التفاعل من خلال التحرك بمرونة بعيدًا للمساعدة في التنشيط في مواقع نشطة أخرى وتجنب تثبيط خطوات التفاعل الإضافية مقارنةً بالتنسيق الأكثر قوة .
استنادًا إلى محاكاة الحواجز الحركية لـ تفعيل وتحليل التنسيق لـ يمكننا أن نستنتج أن القدرة المميزة على التنسيق لأيونات المعادن القلوية هي المسؤولة عن أدوارها الترويجية المختلفة في التفعيل على سطح النحاس. التكوين الجسري بين و * يمكن أن يسهل كلاهما عملية الانحناء ونقل الإلكترون، والتنسيق المرن بينهما يمكن أن يسرع دورة التفاعل. هذه التأثيرات معًا تؤدي إلى تعزيز النشاط والانتقائية بشكل واضح. و الكاتيونات مقارنة بـ و تتوافق هذه النتائج مع اتجاه قوى مجال تفاعل أونساجر من الاستكشاف الطيفي. .
علاوة على ذلك، يمكن توقع سلوك مشابه على أسطح أو جوانب المعادن النقدية الأخرى بسبب استجابتها المماثلة لأيونات المعادن القلوية. بشكل أساسي، يمكن تقسيم تأثير الأسطح المعدنية المختلفة أو الوجوه على نظام كيميائي كهربائي إلى جانبين رئيسيين. أولاً، يمكن أن يغير المواقع النشطة، مما يؤثر على قوى الربط بين الركيزة والوسائط بسبب التغيرات في خصائص الركيزة أو عدد التنسيق للمواقع النشطة. ثانياً، يمكن أن يغير جهد الشحن الصفري، وبالتالي يؤثر على كثافة الشحن السطحي تحت جهد معين مطبق. نظرًا لأن تفعيل تشمل عمليات نقل الإلكترونات، حيث يمكن أن تؤثر التغيرات في كثافة الشحنة السطحية بشكل كبير على القوى الدافعة التي توفرها الجهود المطبقة. على سبيل المثال، في محاكياتنا، لاحظنا امتصاصًا مستقرًا للمنحني * على الأسطح دون أي قيود أو عوامل خارجية مثل الكاتيونات أو الجهود. على النقيض، التقارير السابقة عن الـ السطح يشير إلى انحناء لم يكن بالإمكان امتصاصه بشكل مستقر حتى في وجود أيونات . ومع ذلك، يتم إثبات المزيد أن المنحني يمكن أن تصبح تكوينًا مستقرًا عند كثافات شحن سطحية أعلى على سطح . على الرغم من التكوينات المختلفة المستقرة للامتزاز التي لوحظت على أسطح المعادن المختلفة، إلا أن الدور المعزز لأيونات المعادن القلوية يبقى عاملاً ثابتًا. لذلك، نتوقع أن بعض النتائج من عملنا يمكن تطبيقها على أسطح المعادن الأخرى وأوجهها، وأن التأثيرات الإضافية ستستكشف أيضًا في عملنا المستقبلي.

التأثير الضروري لـ على

التحليل أعلاه يفسر بنجاح الدور المعزز المتزايد لأيونات المعادن القلوية من إلى على التفعيل. ومع ذلك، فإن الحاجز الحركي المنخفض نسبيًا البالغ 0.22 إلكترون فولت في نظام لا يمكن أن يفسر بالكامل الدور الضروري لأيونات المعادن القلوية على الذي لوحظ في التجربة. لتحديد ذلك، نحتاج إلى النظر في رد الفعل الرئيسي التفعيل ولكن أيضًا في رد الفعل الجانبي HER. في غياب أيونات المعادن القلوية، يمكن أن تؤدي الإلكتروليتات الحمضية العالية المستخدمة في التجربة إلى تركيز عالٍ من البروتونات السطحية، مما يتسبب في هيمنة HER المتنافسة على . على الرغم من أن HER معروف بأنه رد فعل جانبي رئيسي متنافس خلال ، إلا أن الفهم الجزيئي لتأثيرات أيونات المعادن القلوية على HER نادرًا ما تم دراسته. لذلك، قمنا بمزيد من التحقيق في التأثير المحتمل لأيونات المعادن القلوية على إمكانية الوصول للبروتون ونشاط HER على سطح النحاس.
تظهر الشكل 5 توزيعات بروتون زائد مع وبدون أيونات المعادن القلوية عند الواجهة. يتم عرض المتوسطات التراكمية للنتائج في الشكل التكميلية 9. يتم حل البروتون الزائد لتشكيل أيون هيدروونيوم ، الذي يمكن أن يتحرك على طول شبكة الروابط الهيدروجينية عن طريق القفز بين جزيئات الماء المختلفة. بعد التوازن لأكثر من 20 بيكو ثانية، يصل البروتون المذاب إلى تكوين مستقر ديناميكيًا على بعد حوالي من سطح النحاس. بناءً على توزيع الماء الموضح في الشكل 1، ينتشر البروتون الزائد إلى الطبقة الثانية من الماء (الذروة الثالثة في الشكل 1a)، مما يلبي الروابط الهيدروجينية الثلاث المطلوبة من قبل أيون . ومع ذلك، عندما يتم إدخال أيونات المعادن القلوية عند الواجهة، يزداد بعد البروتون-الواجهة، باستثناء (الشكل 5a). ينتج التغيير في بعد البروتون-الواجهة عن تعطيل شبكة الروابط الهيدروجينية الواجهة والنبذ الكهروستاتيكي لأيونات المعادن القلوية المذابة . بالنسبة للأيونات الأكبر، بما في ذلك ، و ، يتم دفع البروتون الزائد بعيدًا عن الطبقة الثانية من الماء ومحبوسًا في أغلفة الترطيب الأولى لهذه الأيونات، كما يتضح من المسافات القابلة للمقارنة بين البروتون-الواجهة وأقطار أغلفة الترطيب الأولى لأيونات المعادن القلوية. المسافة المتوسطة بين البروتون والواجهة مع أقل قليلاً من ضعف نصف قطر غلاف الترطيب الأول بسبب سلوكه الجزئي في إزالة الترطيب، مشابهًا للحالة في الشكل التكميلية 7. أما بالنسبة للحالة مع ، يبدو أن بعد البروتون-الواجهة أصغر حتى من الحالة مع واحد، وهو ما يرجع إلى جانبين. من ناحية، غلاف الترطيب لـ أصغر وأصعب من ذلك الخاص بأيونات المعادن القلوية الأكبر، كما هو موضح في الأقسام السابقة، مما يعني أن شبكة الروابط الهيدروجينية الواجهة أكثر اكتمالًا نسبيًا، مما يترك مساحة لـ لتشكيل الروابط الهيدروجينية الثلاث المطلوبة عند الواجهة. بهذه الطريقة، يمكن أن يتواجد كل من و المذابان
الشكل 5 | التفاعل بين البروتون المذاب ( ) وأيونات المعادن القلوية. أ توزيع لـ في المذيب مع وبدون أيونات المعادن القلوية. تمثل الخطوط الحمراء والبنفسجية المتقطعة المسافة المتوسطة للبروتون الزائد من سطح النحاس في نطاق إنتاج 10 بيكو ثانية مع وبدون أيون ، على التوالي. يتم عرض الهياكل التمثيلية المقابلة للواجهة في
(ب). و مميزتان وممثلتان ككرات حيث يتم تمثيل O باللون الأحمر، وH بالأبيض، وK بالبنفسيجي، بينما يتم إخفاء الذرات الأخرى لزيادة الوضوح البصري. توضح الخطوط المتقطعة باللون الأرجواني حول التنسيق مع ذرات O في غلاف الترطيب الأول.
بين الطبقتين الأولى والثانية من الماء على الرغم من النبذ الكهروستاتيكي بينهما. علاوة على ذلك، بعد إدخال أيونات المعادن القلوية، تزداد كثافة الشحنة السلبية عند سطح مقارنةً بتلك الخاصة بـ واحد، مما يجذب البروتون بشكل أكبر للاقتراب من الواجهة. لذلك، يجتمع الجذب من سطح المحفز المشحون سلبًا والاحتواء من أيونات المعادن القلوية لتنتج المسافات المميزة بين البروتون والواجهة مع مختلف أيونات المعادن القلوية بالقرب من الواجهة. علاوة على ذلك، تنخفض مدة الروابط الهيدروجينية مع إدراج أيون وتتضاءل أكثر مع زيادة حجم الأيون (الملاحظة التكميلية 5)، مما يوفر دعمًا إضافيًا لتأثير أيونات المعادن القلوية المذابة على ديناميات المذيب.
بالإضافة إلى ذلك، نظرًا لأن جزيئات يمكن أن تكون أيضًا مصادر بروتون محتملة، قمنا بمزيد من التحقيق في الحواجز الحركية لخطوة فولمر من جزيئات ، مثل الحرة، و الممتصة على السطح، و -المنسقة بـ ، كما هو موضح في الشكل التكميلية 10. تظهر جزيئات الحرة والممتصة حواجز طاقة حرة عالية تبلغ 0.75 إلكترون فولت و0.95 إلكترون فولت، على التوالي، بينما يظهر -المنسق بـ أعلى حاجز يبلغ 1.05 إلكترون فولت، وهو ما يجعلها جميعًا أكثر صعوبة في الحدوث من . لذلك، من خلال دمج تحليل تأثيرات الأيونات على كل من وتفاعلات HER المتنافسة، تمكنا من شرح الدور الضروري لأيونات المعادن القلوية في ، والذي نتج عن احتواء البروتونات المشتقة من التي تقترب من السطح، بالإضافة إلى الحاجز الحركي الأعلى بكثير للبروتونات المشتقة من جزيئات مقارنةً بتفعيل .
لاحظ أن الاتجاه النوعي فقط هو ما تم تقديمه هنا. عادةً، يتطلب تحديد شبكة التفاعل معيارًا كميًا يعتمد على الفروق في حواجز التفعيل. ومع ذلك، في هذه الحالة، العامل الحاسم هو احتواء البروتونات التي تقترب من السطح بواسطة الأيونات. وهذا يعني أن تركيز المتفاعلات، المتأثر بالجهد، والأنشطة الأيونية، ونقل الكتلة، قد يلعب الدور السائد. علاوة على ذلك، في الواقع، قد ينحرف التركيز الواجهة لأيونات المعادن القلوية عن بعضها البعض تحت جهد محدد بسبب استجابات السعة المختلفة . نظرًا لأن عددًا قليلاً فقط من الأيونات يمكن إدخاله في محاكاة النانو بسبب قيود الحجم، فإنه من الصعب عكس تغيير التركيز بدقة. ومن ثم، فإن هذا يتطلب مزيدًا من الاستكشاف لكل من تطور الواجهة وطرق المحاكاة.
علاوة على ذلك، تقدم نتائجنا المستمدة من نموذج الواجهة المذاب بالكامل رؤى مهمة في أعمال المحاكاة الكهروكيميائية المستقبلية. من الجدير بالذكر أن النموذج يشير إلى تكوينات امتصاص مختلفة لجزيء في المذيبات الضمنية والصريحة، وهياكل تنسيق متميزة لأيونات المعادن القلوية المذابة والوسطاء، والتفاعل بين البروتونات وأيونات المعادن القلوية. تشير النتائج إلى أهمية النظر بوضوح في الإلكتروليت في نماذج المحاكاة، خاصة عند التعامل مع الأنواع التي يمكن أن تتفاعل مع شبكة الروابط الهيدروجينية الواجهة أو مكونات أخرى داخل طبقة الكهرباء المزدوجة، بما في ذلك الأيونات المذابة. في الوقت نفسه، قد تساعد تحقيقاتنا في سطح النحاس في الحصول على فهم أوسع لكيفية تأثير أيونات المعادن القلوية على أداء . يمكن دمج هذا البحث مع الدراسات التي أجريت على أسطح الذهب ، حيث يمكن أن تميز أيونات المعادن القلوية بين مسارات تفعيل بين آليات الكرة الداخلية والخارجية. بعد توضيح الأدوار المختلفة التي تلعبها أيونات المعادن القلوية على المعادن النقدية المختلفة وحتى المحفزات الأخرى ، من المتوقع أن يتم إنشاء علاقة أكثر نظامية بين تصميم EDL و الأداء، مما يتطلب مزيدًا من الاستكشاف.
في الختام، قدمت نتائج سلسلة من محاكاة AIMD المذابة بشكل صريح حول وHER المتنافسة رؤى منهجية حول الأدوار الترويجية والتأثيرات الضرورية لأيونات المعادن القلوية لـ . من الناحية الطاقية، فإن يقل حاجز التنشيط مع زيادة حجم الكاتيون. يمكن أن تُعزى الأدوار الترويجية المختلفة إلى قدرات التنسيق، أو قوى أونساجر، لمختلف كاتيونات المعادن القلوية. بشكل خاص، الكاتيونات الأصغر و يمكن أن تتنسق الكاتيونات مع ذرة أكسجين واحدة فقط من وفي الوقت نفسه تطلق جزيء ماء، بينما و يمكن أن تنسق الكاتيونات بشكل مرن مع كل من ذرتي الأكسجين لـ ، وهو أكثر ملاءمة لـ تنشيط ودورة التفاعل. ومع ذلك، فإن إزالة الحل الجزئية لـ الكاتيون يثبط قليلاً الميل المتزايد، مما يؤدي إلى تشابهه مع الكاتيون. من ناحية أخرى، يُنسب الدور الضروري إلى المنافسة بين HER و يمكن أن تعيق أيونات المعادن القلوية بشكل كبير الوصول إلى البروتونات على الواجهة، والحاجز الحركي الأعلى بكثير للبروتونات المستمدة من الجزيئات تساعد لتفوق عليها. نظرًا للتشابه في السلوك التحفيزي مع كاتيونات المعادن القلوية، من المتوقع حدوث هذه الظواهر مع محفزات أخرى مختارة لـ CO، مثل الفضة.
تساعد نتائجنا في فهم الأصل الجزيئي للدور المحتمل للبيئة الكهروكيميائية، وتؤكد على أهمية الترطيب الصريح، والأيونات، والاستكشاف المنهجي للتفاعلات المتنافسة في المحاكاة الكهروكيميائية، وتظهر فعالية الأدوات النظرية في استكشاف واجهة الصلب-الإلكتروليت.

طرق

نموذج القطبين مزدوجي الإلكتروليت الصلب

استنادًا إلى النتائج التجريبية، تم تحديد أن ستهيمن السطح بعد عملية اختزال كهربائي مطولة لتمثيل القطب، اخترنا سطح مصمم ك slab ثلاثي الطبقات مع سوبرسيل لمحاكاة عمليات التفعيل، وأكبر تم استخدام السوبرسيل عند النظر في التفاعلات بين كاتيونات المعادن القلوية والبروتونات. كانت المساحة فوق سطح النحاس مملوءة بجزيئات الماء بكثافة من أجل الحفاظ على كثافة الماء، تم استبدال جزيء ماء واحد بـ الجزيء. تم توسيع حجم مساحة المذيب مع إضافة الأيونات. كاتيونات المعادن القلوية ( ، أو تم وضعها في البداية Åبعيدًا عن سطح، بينما أيون سالب تم وضعه بعيدًا على الجانب الآخر من المذيب ليعمل كزوج أيوني مع الكاتيون، مع الحفاظ على الجهد المطبق قريبًا من جهد البداية. طبقة ذرية إضافية من النيون و Åتم إضافة طبقة فراغية فوق نموذج الواجهة لتعمل كقطب مضاد ومراقبة الجهد الكهروكيميائي للنظام بأكمله. .

محاكاة AIMD

تم إجراء جميع محاكاة AlMD باستخدام حزمة المحاكاة الأولية فيينا (VASP) إمكانات الموجة المعززة بواسطة البروجيكتور (PAW) و تقريب التدرج العام (GGA) مع الدالة المعدلة بيردو-بورك-إرنزرهوف (RPBE) تم استخدامهما لوصف تفاعلات الإلكترون-الأيون وطاقة التبادل-الارتباط، على التوالي. تم وصف تفاعلات فان der Waals باستخدام تصحيح DFT-D3. . تم تعيين حد الطاقة لتوسيع قاعدة الموجة المستوية إلى 500 إلكترون فولت. بالنسبة لعينات منطقة بريلوان، تم استخدام نقطة غاما فقط لمحاكاة AIMD، بينما تم استخدام تم استخدام الشبكة لحسابات الهيكل الإلكتروني تم استخدام ترموستات نوز-هوفر لأخذ عينات قانونية في الحواجز الحركية وملف الطاقة الحرة لـ تمت دراسة عملية التفعيل باستخدام مزيج من النمو البطيء وأخذ عينات في القمر الأزرق طرق مع خوارزمية SHAKE ، مع زاوية الـ الجزيء المختار كمتغير جماعي (CV). يتم مناقشة اختيار CV في الملاحظة التكميلية 2.
بالنظر إلى الجهد المطبق، فإن القطب المضاد Ne تم استخدامه لمراقبة التغيير المحتمل في نظام التفاعل بشكل فوري. تصحيح ثنائي القطب تم تطبيق ذلك في الاتجاه العمودي على الواجهة في طبقة الفراغ. يمكن اعتبار مستوى الفراغ على جانب المعدن ثابتًا بسبب الحماية الكهروستاتيكية للمعدن، وبالتالي يمكن حساب الجهد المطلق للنظام، أي الجهد المطبق، وفقًا لذلك: ضد هي ، حيث هو عزم ثنائي القطب للنظام، هو المساحة السطحية، و هو ثابت كولوم. ثم تم تطبيق تصحيحات الجهد الثابت لحساب التغير الطفيف في الجهد الكهروكيميائي لحالات التفاعل المختلفة على طول مسارات التفاعل (الملاحظة التكميلية 6) .

توفر البيانات

تم تضمين البيانات التي تدعم نتائج هذه الدراسة في النص الرئيسي والمعلومات التكميلية. تم إيداع المسارات التمثيلية لـ AIMD التي تم إنشاؤها في هذه الدراسة في مستودع بيانات Zenodo فيhttps://doi.org/10.5281/zenodo.10427196جميع البيانات الأخرى ذات الصلة التي تدعم نتائج هذه الدراسة متاحة من المؤلفين المقابلين عند الطلب.

References

  1. Nitopi, S. et al. Progress and perspectives of electrochemical reduction on copper in aqueous electrolyte. Chem. Rev. 119, 7610-7672 (2019).
  2. Huang, J. E. et al. electrolysis to multicarbon products in strong acid. Science 372, 1074-1078 (2021).
  3. He, M. et al. Selective enhancement of methane formation in electrochemical reduction enabled by a Raman-inactive oxygen-containing species on Cu. ACS Catal. 12, 6036-6046 (2022).
  4. Li, J. et al. Constraining CO coverage on copper promotes highefficiency ethylene electroproduction. Nat. Catal. 2, 1124-1131 (2019).
  5. Dinh, C. T. et al. electroreduction to ethylene via hydroxidemediated copper catalysis at an abrupt interface. Science 360, 783-787 (2018).
  6. Timoshenko, J. et al. Steering the structure and selectivity of electroreduction catalysts by potential pulses. Nat. Catal. 5, 259-267 (2022).
  7. Wang, L. et al. Electrochemical carbon monoxide reduction on polycrystalline copper: effects of potential, pressure, and pH on selectivity toward multicarbon and oxygenated products. ACS Catal. 8, 7445-7454 (2018).
  8. Chen, X. et al. Electrochemical -to-ethylene conversion on polyamine-incorporated Cu electrodes. Nat. Catal. 4, 20-27 (2021).
  9. Wang, R. et al. Partial coordination-perturbed bi-copper sites for selective electroreduction of to hydrocarbons. Angew. Chem. Int. Ed. 60, 19829-19835 (2021).
  10. Bagger, A. et al. Electrochemical reduction: classifying Cu facets. ACS Catal. 9, 7894-7899 (2019).
  11. Scholten, F. et al. Identifying structure-selectivity correlations in the electrochemical reduction of : a comparison of wellordered atomically clean and chemically etched copper singlecrystal surfaces. Angew. Chem. Int. Ed. 60, 19169-19175 (2021).
  12. Wu, Z.-Z. et al. Identification of interfaces as superior active sites for CO dimerization during electroreduction. J. Am. Chem. Soc. 144, 259-269 (2021).
  13. Akira, M. & Yoshio, H. Product selectivity affected by cationic species in electrochemical reduction of and CO at a Cu electrode. Bull. Chem. Soc. Jpn 64, 123-127 (1991).
  14. Hori, Y. et al. Formation of hydrocarbons in the electrochemical reduction of carbon dioxide at a copper electrode in aqueous solution. J. Chem. Soc. Faraday Trans. 85, 2309-2326 (1989).
  15. Resasco, J. et al. Promoter effects of alkali metal cations on the electrochemical reduction of carbon dioxide. J. Am. Chem. Soc. 139, 11277-11287 (2017).
  16. Ren, W. et al. A cation concentration gradient approach to tune the selectivity and activity of electroreduction. Angew. Chem. Int. Ed. 61, e202214173 (2022).
  17. Ayemoba, O. & Cuesta, A. Spectroscopic evidence of size-dependent buffering of interfacial pH by cation hydrolysis during electroreduction. ACS Appl. Mater. Interfaces 9, 27377-27382 (2017).
  18. Singh, M. R. et al. Hydrolysis of electrolyte cations enhances the electrochemical reduction of over Ag and Cu . J. Am. Chem. Soc. 138, 13006-13012 (2016).
  19. Malkani, A. S. et al. Cation effect on interfacial concentration in the electrochemical reduction reaction. ACS Catal. 10, 14871-14876 (2020).
  20. Gu, J. et al. Modulating electric field distribution by alkali cations for electroreduction in strongly acidic medium. Nat. Catal. 5, 268-276 (2022).
  21. Ringe, S. et al. Understanding cation effects in electrochemical reduction. Energy Environ. Sci. 12, 3001-3014 (2019).
  22. Chen, L. D. et al. Electric field effects in electrochemical reduction. ACS Catal. 6, 7133-7139 (2016).
  23. Monteiro, M. C. O. et al. Absence of electroreduction on copper, gold and silver electrodes without metal cations in solution. Nat. Catal. 4, 654-662 (2021).
  24. Qin, H.-G. et al. Quantitative understanding of cation effects on the electrochemical reduction of and in acidic solution. ACS Catal. 13, 916-926 (2022).
  25. Zhu, Q. et al. The solvation-induced Onsager reaction field rather than the double-layer field controls reduction on gold. JACS Au 2, 472-482 (2022).
  26. Groß, A. & Sakong, S. Ab initio simulations of water/metal interfaces. Chem. Rev. 122, 10746-10776 (2022).
  27. Qin, X. et al. Cation-coordinated inner-sphere electroreduction at Au-water interfaces. J. Am. Chem. Soc. 145, 1897-1905 (2023).
  28. Liu, H. et al. Promotional role of a cation intermediate complex in formation from electrochemical reduction of over Cu . ACS Catal. 11, 12336-12343 (2021).
  29. Yu, M. & Trinkle, D. R. Accurate and efficient algorithm for Bader charge integration. J. Chem. Phys. 134, 064111 (2011).
  30. Chan, K. & Nørskov, J. K. Electrochemical barriers made simple. J. Phys. Chem. Lett. 6, 2663-2668 (2015).
  31. Kronberg, R. & Laasonen, K. Reconciling the experimental and computational hydrogen evolution activities of Pt(111) through DFTbased constrained MD simulations. ACS Catal. 11, 8062-8078 (2021).
  32. Zhao, X. & Liu, Y. Origin of selective production of hydrogen peroxide by electrochemical oxygen reduction. J. Am. Chem. Soc. 143, 9423-9428 (2021).
  33. Le, J.-B. et al. Molecular origin of negative component of Helmholtz capacitance at electrified Pt(111)/water interface. Sci. Adv. 6, eabb1219 (2020).
  34. Vijay, S. et al. Unified mechanistic understanding of reduction to CO on transition metal and single atom catalysts. Nat. Catal. 4, 1024-1031 (2021).
  35. Cao, H. et al. Potential-dependent free energy relationship in interpreting the electrochemical performance of reduction on single atom catalysts. ACS Catal. 12, 6606-6617 (2022).
  36. Shin, S.-J. et al. A unifying mechanism for cation effect modulating and productions from electroreduction. Nat. Commun. 13, 5482 (2022).
  37. Le, D. & Rahman, T. S. On the role of metal cations in electrocatalytic reduction. Nat. Catal. 5, 977-978 (2022).
  38. Li, P. et al. Hydrogen bond network connectivity in the electric double layer dominates the kinetic pH effect in hydrogen electrocatalysis on Pt. Nat. Catal. 5, 900-911 (2022).
  39. Li, X.-Y. et al. Mechanism of cations suppressing proton diffusion kinetics for electrocatalysis. Angew. Chem. Int. Ed. 62, e202218669 (2023).
  40. Le, J.-B. et al. Molecular understanding of cation effects on double layer and its significance to CO-CO dimerization. Natl. Sci. Rev. 10, nwad105 (2023).
  41. Hussain, G. et al. How cations determine the interfacial potential profile: relevance for the reduction reaction. Electrochim. Acta 327, 135055 (2019).
  42. Garlyyev, B. et al. Influence of the nature of the alkali metal cations on the electrical double-layer capacitance of model Pt(111) and Au(111) electrodes. J. Phys. Chem. Lett. 9, 1927-1930 (2018).
  43. Xue, S. et al. How the nature of the alkali metal cations influences the double-layer capacitance of , and Pt single-crystal electrodes. J. Phys. Chem. C 124, 12442-12447 (2020).
  44. Qin, X. et al. Cation-induced changes in the inner- and outer-sphere mechanisms of electrocatalytic reduction. Nat. Commun. 14, 7607 (2023).
  45. Kwon, S. et al. Dramatic change in the step edges of the electrocatalyst upon exposure to CO: operando observations by electrochemical STM and explanation using quantum mechanical calculations. ACS Catal. 11, 12068-12074 (2021).
  46. Surendralal, S. et al. First-principles approach to model electrochemical reactions: understanding the fundamental mechanisms behind Mg corrosion. Phys. Rev. Lett. 120, 246801 (2018).
  47. Surendralal, S. et al. Impact of water coadsorption on the electrode potential of H-Pt(111)-liquid water interfaces. Phys. Rev. Lett. 126, 166802 (2021).
  48. Kresse, G. & Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169-11186 (1996).
  49. Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B 59, 1758-1775 (1999).
  50. Hammer, B. et al. Improved adsorption energetics within densityfunctional theory using revised Perdew-Burke-Ernzerhof functionals. Phys. Rev. B 59, 7413-7421 (1999).
  51. Grimme, S. et al. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132, 154104 (2010).
  52. Monkhorst, H. J. & Pack, J. D. Special points for Brillouin-zone integrations. Phys. Rev. B 13, 5188-5192 (1976).
  53. Hoover, W. G. Canonical dynamics: equilibrium phase-space distributions. Phys. Rev. A 31, 1695-1697 (1985).
  54. Nosé, S. A unified formulation of the constant temperature molecular dynamics methods. J. Chem. Phys. 81, 511-519 (1984).
  55. Woo, T. K. et al. A combined Car-Parrinello QM/MM implementation for ab initio molecular dynamics simulations of extended systems: application to transition metal catalysis. J. Phys. Chem. B 101, 7877-7880 (1997).
  56. Carter, E. A. et al. Constrained reaction coordinate dynamics for the simulation of rare events. Chem. Phys. Lett. 156, 472-477 (1989).
  57. Ryckaert, J.-P. et al. Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J. Comput. Phys. 23, 327-341 (1977).
  58. Neugebauer, J. & Scheffler, M. Adsorbate-substrate and adsorbateadsorbate interactions of Na and K adlayers on Al (111). Phys. Rev. B 46, 16067-16080 (1992).
  59. Chan, K. & Nørskov, J. K. Potential dependence of electrochemical barriers from ab initio calculations. J. Phys. Chem. Lett. 7, 1686-1690 (2016).

شكر وتقدير

تم دعم هذا العمل من قبل البرنامج الوطني الرئيسي للبحث والتطوير في الصين (2022YFA1203400 و 2021YFA1400100)، ومؤسسة العلوم الوطنية في الصين (12274254 و 11874036)، ومشروع الفرق البحثية والابتكارية المحلية في برنامج مواهب نهر اللؤلؤ في قوانغدونغ (2017BT01N111) ومشروع البحث الأساسي في شنتشن، الصين (JCYJ2O200109142816479 و WDZC2O200819115243002). تم توفير الموارد الحاسوبية من قبل منصة الحوسبة عالية الأداء في جامعة نانجينغ للطيران والفضاء.

مساهمات المؤلفين

قام J.L. و F.K. بتصور وتصميم المشروع. قام Z.Z. بإجراء الحسابات. قام Z.Z. و J.L. و H.L. و Y.S. و L.G. و W.D. و H.H. بتحليل النتائج. كتب Z.Z. و J.L. و F.K. الورقة، وعلق جميع المؤلفين عليها.

المصالح المتنافسة

يعلن المؤلفون عدم وجود مصالح متنافسة.

معلومات إضافية

معلومات إضافية النسخة الإلكترونية تحتوي على
المواد التكميلية متاحة على
https://doi.org/10.1038/s41467-024-44896-x.
يجب توجيه المراسلات والطلبات للحصول على المواد إلى فييو كانغ أو جيا لي.
معلومات مراجعة الأقران تشكر مجلة Nature Communications المراجعين المجهولين على مساهمتهم في مراجعة هذا العمل. يتوفر ملف مراجعة الأقران.
معلومات إعادة الطباعة والتصاريح متاحة على
http://www.nature.com/reprints
ملاحظة الناشر: تظل شركة سبرينجر ناتشر محايدة فيما يتعلق بالمطالبات القضائية في الخرائط المنشورة والانتماءات المؤسسية.
الوصول المفتوح هذه المقالة مرخصة بموجب رخصة المشاع الإبداعي النسب 4.0 الدولية، التي تسمح بالاستخدام والمشاركة والتكيف والتوزيع وإعادة الإنتاج بأي وسيلة أو صيغة، طالما أنك تعطي الائتمان المناسب للمؤلفين الأصليين والمصدر، وتوفر رابطًا لرخصة المشاع الإبداعي، وتوضح إذا ما تم إجراء تغييرات. الصور أو المواد الأخرى من طرف ثالث في هذه المقالة مشمولة في رخصة المشاع الإبداعي الخاصة بالمقالة، ما لم يُشار إلى خلاف ذلك في سطر الائتمان للمواد. إذا لم تكن المادة مشمولة في رخصة المشاع الإبداعي الخاصة بالمقالة وكان استخدامك المقصود غير مسموح به بموجب اللوائح القانونية أو يتجاوز الاستخدام المسموح به، فسيتعين عليك الحصول على إذن مباشرة من صاحب حقوق الطبع والنشر. لعرض نسخة من هذه الرخصة، قم بزيارةhttp://creativecommons.org/رخصة/بواسطة/4.0/.
© المؤلف(ون) 2024

  1. مركز جرافين شنتشن ومعهد أبحاث المواد، المدرسة الدولية للدراسات العليا بجامعة تسينغhua، شنتشن 518055، جمهورية الصين الشعبية. المختبر الوطني الرئيسي للفيزياء الكمومية ذات الأبعاد المنخفضة وقسم الفيزياء، جامعة تسينغhua، بكين 100084، جمهورية الصين الشعبية. معهد الدراسات المتقدمة، جامعة تسينغhua، بكين 100084، جمهورية الصين الشعبية. مركز العلوم الحدودية لمعلومات الكم، بكين 100084، جمهورية الصين الشعبية. قسم تحويل وتخزين الطاقة، الجامعة التقنية في الدنمارك، كنجس، لينغبي 2800، الدنمارك. البريد الإلكتروني:fykang@sz.tsinghua.edu.cn; li.jia@sz.tsinghua.edu.cn

Journal: Nature Communications, Volume: 15, Issue: 1
DOI: https://doi.org/10.1038/s41467-024-44896-x
PMID: https://pubmed.ncbi.nlm.nih.gov/38242907
Publication Date: 2024-01-19

Molecular understanding of the critical role of alkali metal cations in initiating electroreduction on surface

Received: 30 May 2023
Accepted: 9 January 2024
Published online: 19 January 2024
(A) Check for updates

Zhichao Zhang © , Hengyu Li , Yangfan Shao , Lin Gan , Feiyu Kang , Wenhui Duan , Heine Anton Hansen © & Jia Li (B)

Abstract

Molecular understanding of the solid-liquid interface is challenging but essential to elucidate the role of the environment on the kinetics of electrochemical reactions. Alkali metal cations ( ), as a vital component at the interface, are found to be necessary for the initiation of carbon dioxide reduction reaction ( ) on coinage metals, and the activity and selectivity of could be further enhanced with the cation changing from to , while the underlying mechanisms are not well understood. Herein, using ab initio molecular dynamics simulations with explicit solvation and enhanced sampling methods, we systematically investigate the role of in on Cu surface. A monotonically decreasing activation barrier is obtained from to , which is attributed to the different coordination abilities of with * . Furthermore, we show that the competing hydrogen evolution reaction must be considered simultaneously to understand the crucial role of alkali metal cations in on Cu surfaces, where is repelled from the interface and constrained by . Our results provide significant insights into the design of electrochemical environments and highlight the importance of explicitly including the solvation and competing reactions in theoretical simulations of .

The electrochemical carbon dioxide reduction reaction ( ) has emerged as an attractive method for simultaneously closing the carbon cycle and producing important industrial feedstocks due to its compatibility with renewable, carbon-free energy sources . Cu-based materials are particularly unique among the various types of catalysts developed today, as they can produce multi-electron ( ) reduced products like methane , as well as multi-carbon ( ) products consisting mainly of ethylene and ethanol , regardless of their different structures, such as Cu foil , nanoparticles , and even molecular catalysts . However, the activity and selectivity of these catalysts have not yet reached the level required for industrial application. Therefore,
many efforts have been made to improve their electrochemical performance further, especially for more valuable products. These efforts include facet design , defect engineering , pretreatments such as oxide-derived , and electrolyte engineering . However, the determining factors and atomic-level insights into the underlying mechanisms are still unclear.
, as a representative aqueous electrochemical reaction, is significantly affected not only by the catalyst but also by the environment at the solid-electrolyte interface. Since the 1990s, Hori et al. have discovered that alkali metal cations ( ) can enhance activity by more than an order of magnitude and concurrently improve
selectivity on the Cu electrode. This promoting effect becomes more significant with larger cations from to . The promoting effect of alkali metal cations has been confirmed and extended to other catalysts, such as CO-selective , , and HCOOH -selective . Several explanations have been developed so far. The surface pH model attributed the differences between alkali metal cations to the interfacial proton and concentrations resulting from the equilibration between carbonate-group anions , but the relationship between this model and the reaction mechanism is not well understood. Subsequently, a dipole-field interaction model was proposed, which emphasizes the stabilizing effect of the electric field between solvated alkali metal cations and the negatively charged metal surface on polar intermediates . Therefore, intermediates with large dipoles involving and (the asterisk* denotes the adsorption site) would be strongly stabilized, enhancing the selectivity of corresponding products such as CO on and on Cu . However, this model is based on the uniform interfacial electric field model and attributes the discrepancy between different alkali metal cations to the tendency of larger cations to accumulate more at the interface while ignoring the intrinsic difference of individual alkali metal cations and possible local interactions between cations and intermediates.
Recently, it has been further demonstrated that the cannot be initiated on , and without alkali metal cations, even to CO and , which cannot be fully explained by the above models. In addition, spectroscopic results have shown that the introduction of , but not cations, results in the highest electric field in the Stern layer, while the strength of the solvation-induced reaction field, the Onsager field, varies with the cation size . These findings suggest that different alkali metal cations have inherent differences rather than a simple concentration disparity and that cations at the interface could greatly influence the efficiency of the to-CO conversion process, potentially playing a more active role rather than merely acting as spectators. To provide a more detailed explanation of the influence of solvated alkali metal cations on reaction mechanisms, ab initio molecular dynamics (AIMD) simulations were performed and revealed the existence of local interactions, such as coordination between solvated cations and surface intermediates . Notable examples include the interaction between cation and adsorbed molecule on Au surface , and the interaction between cations and *CO + *CO dimer on Cu surface . However, a comprehensive understanding of the effect of different alkali metal cations on
activation and their critical role in initialization on the Cu surface is still lacking.
In this work, we have systematically studied the effect of alkali metal cations on both and the main competing hydrogen evolution reaction (HER), using AIMD simulations with enhanced sampling methods and explicitly taking into account the effects of solvation and potential. By calculating the free energy barriers of activation, a monotonically decreasing trend from to is found, which is due to the coordination ability between the alkali metal cation and the * intermediate. The bridge coordination between larger cations and helps to facilitate more compared to the side configuration preferred by smaller cations. Furthermore, we show that the necessary role of alkali metal cations in on Cu surfaces cannot be fully understood without simultaneously considering the competing HER. Our results shed light on the intrinsic short-range stabilizing effect of individual alkali metal cations on and highlight the necessity of explicitly considering the environment, such as cation and solvation, as well as competing reactions, in electrochemical simulations in order to obtain more reliable insights.

Results and discussion

Electrical double-layer response of applied potential

To provide a benchmark for the subsequent analysis, we first performed simulations of the Cu -water interface system with and without adsorption, denoted as and , respectively, to investigate the response of the electrical double layer (EDL) to different applied potentials. Figure 1a-c shows the density distribution of water perpendicular to the interface under different conditions. Three primary peaks could be observed above the Cu -water interface at approximately , and , respectively. The first peak (marked with the gray dashed line in Fig. 1a-c) is attributed to water molecules chemisorbed on the Cu surface, while the second and third peaks indicate two partially ordered water layers. No evident peaks were observed further away from the interface, representing the diffusion layer and the bulk water with a density of about , and indicating the sufficient thickness of the water layer as well. Furthermore, as the applied potential decreases, the intensity of the first peak decreases, as shown in Fig. 1a, which could be attributed to the increasing repulsive interaction between the more negatively charged substrate and the partially negatively charged O atom of the molecule (as schematically shown in Fig. 1d). Meanwhile, H atoms would be attracted by the
Fig. 1 | Response of interfacial structure to the applied potential.
a-c Distribution of water density along the direction perpendicular to the -water interface under different conditions, with the topmost Cu layer at the distance of 0 . The gray dashed line indicates the first peak of chemisorbed
water. d Schematic of the interfacial structure change under corresponding conditions to (a-c). molecules are depicted as gray and red spheres, while molecules are represented as stick models for clarity. The applied potentials are obtained from the work function of the whole system. (Supplementary Note 1).
Fig. 2 | Schematic of the representative * adsorption structures. Local interactions between * at surface and the surrounding environment: a Schematic of the bidentate adsorption of * at Cu surface in an explicit solvent with C and one of the O atoms coordinating with Cu , and the other O atom forming
hydrogen bonds (light blue dashed lines) with surrounding water. Schematic of the side configuration of or coordinated with only one O atom of * (magenta dashed line). c Schematic of the bridge configuration of or coordinated with both two O atoms of .
surface, causing the bond to rotate toward the surface, as reflected by the change in interfacial water dipoles within of the Cu surface (Supplementary Fig. 1). The dipole distribution changes from random to a single peak, indicating a more aligned distribution of water molecules in the first water layer.
A similar response of the interfacial water arrangement to the applied potential was also observed in the system (Supplementary Fig. 2), where a molecule was introduced onto the Cu surface of the system. Energetically, the adsorption enthalpies of decrease with decreasing applied potential, indicating the more stable adsorption of (Supplementary Table 1). During the additional 15 ps of AIMD simulation, was observed to be stably adsorbed on the Cu surface in a bidentate configuration through the and bonds, with the other O atom facing upward and forming hydrogen bonds with the surrounding water molecules (adsorption configuration in Fig. 1d). Bader charge analysis shows that the chemisorbed has a charge of about , indicating a fully activated adsorption state. In addition, the introduction of enhances the adsorption of chemisorbed water (Fig. 1b), but this is due to the lack of potential control (Const-Q case in Fig. 1d). After adjusting the applied potential to be consistent with the system by electron exchange with the electron reservoir, the intensity of the first peak is almost identical to the reference (Fig. 1c and Const-U case in Fig. 1d). The above results confirm the validity of our model and provide the basis for further investigation of the electrochemical interface during the process .
Furthermore, it is worth noting that the bidentate configuration of * on the surface differs from that observed in a fully implicit solvent environment. In the latter case, tends to be physisorbed on the surface over a wide potential range (Supplementary Fig. 3), while the bidentate configuration is less stable by at -0.6 V vs. SHE (standard hydrogen electrode), which is the concerned potential range in this work. This difference indicates the significant contribution of surrounding hydrogen bonds to the stabilization of from explicit solvation and thus raises a cautionary note regarding the use of fully implicit solvents for investigating molecular explanations of reaction mechanisms, especially when the interactions between intermediates and the surrounding electrochemical environment, such as solvated water molecules, play a substantial role.

Promoting effects of alkali metal cations on adsorption

Experimental results have shown that products, including CO , cannot be detected on coinage metals without in the electrolyte . This suggests that is inhibited in the early stages of the -to- CO conversion process. To investigate the underlying mechanism, here we first focused on the effect of alkali metal cations on the rate-limiting step of to CO , the activation process , which is expected to have electrostatic interactions
with alkali metal cations due to its transformation from a linear nonpolar configuration to a bent polar one. We studied the system in more detail and subsequently added , or ) ion pairs into it to keep the applied potential close to the PZC of as well as the onset potential of at vs. SHE) , which are referred to as systems. Bader charge analysis confirms full ionization of the added ions, with for alkali ions and for F ions, respectively. Using the angle of as a collective variable (CV) during the adsorption process, we found that the CVs of the initial state (IS, a free linear ) and the final state (FS, a chemisorbed bent ) do not vary much with different cations, peaking at and , respectively (Supplementary Fig. 4). The angle of at the IS deviates from due to thermal fluctuations and interactions with water, and the chemisorbed configuration of * at the FS is identical to that in the system (Fig. 2 and Supplementary Fig. 4). These results further demonstrate the significant role of hydrogen bonds formed between and surrounding molecules.
More importantly, we found that all alkali metal cations can stably coordinate with * but with different coordination features. and ions coordinate with only one O atom of * (side configuration, as shown in Fig. 2b), while and would coordinate with both O atoms of * (bridge configuration, as shown in Fig. 2c). One could naturally assume that the bridge coordination results from the larger cation radius, which is supported by the increasing first solvation shell radius ( ) from to (Supplementary Table 2). If the configuration differences are maintained during the activation process, this phenomenon could be an intrinsic cause of the different promoting effects on of with different sizes. The bridge configuration is more favorable for bending as a consequence of the electrostatic repulsion between and the partially positively charged C atom of . From an energetic point of view, we firstly considered the adsorption free energy with different alkali metal cations as an approximation of the activity descriptor. The system with shows the strongest adsorption of * , while the adsorption free energies with the other three cations do not differ much from each other (Fig. 3 and Supplementary Table 3). Consequently, no clear trend with cation size could be observed, which is also consistent with previous results on . As repeatedly demonstrated , the reaction energy alone cannot fully describe the activity of a system, while the activation energy from the IS to the transition state (TS) might be more important to serve as the real energetic criteria. Therefore, in order to obtain more accurate and detailed information about the cation effects on the adsorption process, we next concentrated on the free energy barrier and transition states of the adsorption process at .
Figure 3a shows the free energy profile of activation in the system in the absence of , where a kinetic barrier of 0.22 eV is indicated. For comparison, Fig. 3c-f shows the free energy profiles of activation with ions in the solvent. The kinetic
Fig. 3 | Free energy profiles and representative structures for activation process on surface. a Free energy profile and representative initial state (IS), transition state (TS), and final state (FS) structures for the adsorption process in the pure water ( ) system without . The gray dashed lines indicate the zero value. Corresponding free energy gradient with error bar at each calculated collective variable. Crossover points with dashed baseline indicate the positions of IS, TS, and FS. The error bars represent standard errors of free energy gradients calculated using block average at the corresponding collective variable. c-f Free energy profiles and representative structures at IS, TS, and FS for
adsorption with different cations ( ) of , and . The shadows in all five free energy profiles represent the cumulative errors ( at TS) estimated with the block average method (Supplementary Notes 2 and 3). In the local structures, molecules are depicted as gray and red spheres, Cu substrates are shown as reddish-brown spheres, molecules are illustrated as red and white sticks, and , including , and are represented by green, yellow, purple and blue spheres, respectively. The magenta dashed lines around illustrate their coordination with O atoms in the first solvation shells.
barriers exhibit a clear, monotonically decreasing trend from the system to the systems, with the inclusion of , and . This trend is consistent with the observed activity trend of with alkali metal cations in experiments . Moreover, cations are generally divided into two groups during the activation process based on the activation barriers and their coordination characteristics with * : smaller cations of / and larger cations of (validation of the stable configurations is discussed in Supplementary Note 4). The free energy barrier difference between the two groups is about 0.05 eV , which can increase the corresponding reaction rate by a factor of 5 , which is comparable to the experimental results .
The activation barrier in the bending process and the concerted charge transfer result from the degeneracy breaking of the (lowest unoccupied molecular orbital, LUMO) and (highest occupied molecular orbital, HOMO ) states of . Concurrently, the lower LUMO level of the bent facilitates the back donation of electrons from the substrate to , resulting in the negatively charged . This process can be further enhanced by alkali metal cations (Fig. 3). In the system, the CV at the TS is approximately , but it moves closer to the IS with alkali metal cations. Meanwhile, the negative charges on the at the TS show a similar monotonically decreasing trend, with about transferred to in the system, while only about is transferred in the system (Supplementary Table 4). This also implies that the TS is closest to the IS with the cation, which benefits its lowest activation barrier.
The coordination structures of in the first solvation shell with and without at the TS were further analyzed to understand the underlying molecular mechanism of the activation (Fig. 4a), with the systems denoted as and , respectively. For and cations, the coordination with involves the exchange of an oxygen atom from in the first solvation shell, resulting in an apparently unchanged total coordination number. This exchange is mainly due to the hard solvation shells of small cations,
which can hardly accept more solvation molecules with the environment, as shown by the platform in the integrated radial distribution function (RDF) at the radius of the first solvation shell in Supplementtary Fig. 5. However, for and cations, the coordination numbers increased by 0.9 and 2.2, respectively. This is because and could coordinate with more than one oxygen from and maintain the coordination with . The higher coordination number of and cations with their unique bidentate coordination structure with * could facilitate the electron transfer process and help to activate more effectively than smaller cations, providing important insights into the molecular origin of their higher promoting effect on activity. In addition, there is a difference in the activation barrier between and cations, but experimentally, they were found to activate similarly , which could also be explained by their coordination structure. Comparing the total coordination numbers of alkali metal cations in the system with those in the systems (Supplementary Fig. 6), most remain unchanged except for a decrease in the cation at the interface without . In the AIMD simulation, we observed that the cation loses some solvated molecules at the interface (Supplementary Fig. 7), but when it coordinates with * , its solvation ability with is slightly recovered at the interface (Fig. 4a). The partially desolvated complex behaves like quasi-specific adsorption that may block the active sites and be unfavorable for surface reactions, ultimately resulting in its similar activating effect on with .
Figure 4 b further shows the statistical distribution of the distance between alkali metal cations and two oxygens of , and and cations were observed to coordinate stably with only one oxygen, resulting in a larger distance compared to . In contrast, for and cations, the preferred bridge configuration leads to an equal distance of and . Another difference between these two cation groups is the density of the distributions, with the larger cations showing a broader distribution, consistent with the absence of platforms in the integrated RDF (Supplementary Fig. 5c,
Fig. 4 | Coordination analysis of the first solvation shell of alkali metal cations around TS. a The number of coordinated O for , and cations at the -electrolyte interface with and without the . b Contour map of the

distance distribution between alkali metal cations and two oxygens of , statistically derived from snapshots around the transition state. The results of , and are represented by green, orange, purple, and blue, respectively.
d). This suggests a more dynamic coordination around compared to and cations. In addition, we observed an accumulation of electrons between the cation and the coordinated O of , which is the characteristic feature of a bond. However, no significant charge accumulation was observed for the cation (Supplementary Fig. 8), supporting the above analysis. In this way, more dynamically coordinated with * could also accelerate the reaction cycle by flexibly moving away to help activate at other active sites and avoid inhibiting further reaction steps compared to more strongly coordinated .
Based on the kinetic barrier simulations of activation and the coordination analysis of , we could conclude that the distinct coordination ability of alkali metal cations is responsible for their different promotional roles in activation on the Cu surface. The bridge configuration between and * could both facilitate the bending and electron transfer process, and the flexible coordination between them could accelerate the reaction cycle. These effects together result in the obvious activity and selectivity enhancement of and cations compared with and . These results are consistent with the trend of Onsager reaction field strengths from the spectroscopic exploration .
Furthermore, similar behavior could also be expected on other coinage metal surfaces or facets due to their similar response to alkali metal cations . Basically, the effect of different metal surfaces or facets on an electrochemical system can be divided into two main aspects. First, it can alter the active sites, thereby affecting the binding strengths between the substrate and intermediates due to changes in substrate properties or the coordination number of the active sites. Second, it can alter the potential of the zero charge, thus influencing the surface charge density under specific applied potentials. Since the activation of involves electron transfer processes, changes in the surface charge density can significantly affect the driving forces provided by the applied potentials. For example, in our simulations, we observed stable adsorption of bent * on surfaces without any constraints or external factors such as cations or potentials. In contrast, previous reports on the surface indicated that bent could not be stably adsorbed even in the presence of ions . However, it is further demonstrated that bent can become a stable configuration at higher surface charge densities on the surface . Despite the different stable adsorption configurations observed on different metal surfaces, the promoting role of alkali metal cations remains a consistent factor. Therefore, we expect that some of the findings from our work can be applied to other metal surfaces and facets and that additional effects will also be explored in our future work.

Necessary effect of on

The above analysis successfully explains the increasingly promoting role of alkali metal cations from to on activation. However, the relatively low kinetic barrier of 0.22 eV in the system cannot fully account for the necessary role of alkali metal cations on observed in the experiment. To determine this, we need to consider not only the main reaction activation but also the side reaction HER. In the absence of alkali metal cations, the highly acidic electrolytes used in the experiment could lead to a high concentration of surface protons, causing the competing HER to dominate . Although HER is known to be a major competing side reaction during , the molecular understanding of the effects of alkali metal cations on HER has rarely been studied. Therefore, we further investigated the possible influence of alkali metal cations on proton accessibility and HER activity on the Cu surface.
Figure 5 shows the distributions of an excess proton with and without the alkali metal cations at the interface. Cumulative averages of the results are shown in Supplementary Fig. 9. The excess proton is solvated to form a hydronium ion , which can move along the hydrogen bond network by hopping between different water molecules. After equilibrating for more than 20 ps , the solvated proton reaches a dynamically stable configuration at a distance of about away from the Cu surface. Based on the water distribution shown in Fig. 1, the excess proton diffuses into the second water layer (third peak in Fig. 1a), which satisfies the three hydrogen bonds required by the ion. However, when alkali metal cations are introduced at the interface, the proton-interface distance increases, except for (Fig. 5a). The change in proton-interface distance results from the disruption of the interfacial hydrogen bond network and the electrostatic repulsion of solvated alkali metal cations . For larger cations, including , and , the excess proton is repelled away from the second water layer and confined in the first solvation shells of these cations, as indicated by the comparable proton-interface distances and the diameters of the first solvation shells of alkali metal cations. The average proton-interface distance with is slightly less than twice the radius of the first solvation shell due to its partial desolvation behavior, similar to the case in Supplementary Fig. 7. As for the case with , the proton-interface distance seems to be even smaller than in the case with a single , which is due to two aspects. On the one hand, the solvation shell of is smaller and harder than that of larger alkali metal cations, as indicated in the previous sections, which means that the interfacial hydrogen bond network is relatively more complete, leaving space for to form three required hydrogen bonds at the interface. In this way, the and solvated could coexist
Fig. 5 | Interaction between solvated proton ( ) and alkali metal cations. a Distribution of in the solvent with and without alkali metal cations. The red and purple dashed lines represent the average distance of the excess proton from the Cu surface in the production timescale of 10 ps with and without cation, respectively. The corresponding representative interface structures are shown in
(b). and are highlighted and shown as spheres where O is represented in red, H in white, and K in purple, while other atoms are hidden for visual clarity. Magenta dashed lines around illustrate the coordination with O atoms in the first solvation shell.
between the first and second water layers despite of the electrostatic repulsion between them. Furthermore, after the introduction of alkali metal cations, the negative charge density at the surface increases compared to that of single , which further attracts the proton to approach the interface. Therefore, the attraction from the negatively charged catalyst surface and the inhibition from alkali metal cations combine to result in the distinct proton-interface distances with various alkali metal cations near the interface. Furthermore, the hydrogen bond lifetime decreases with the inclusion of the cation and is further suppressed with increasing cation size (Supplementary Note 5), which provides additional support for the influence of solvated alkali metal cations on solvent dynamics.
In addition, since molecules could also be potential proton sources, we further investigated the kinetic barriers of the Volmer step from molecules, such as free , surface-adsorbed , and -coordinated , as shown in Supplementary Fig. 10. The free and adsorbed molecules exhibit high free energy barriers of 0.75 eV and 0.95 eV , respectively, while the -coordinated shows the highest barrier of 1.05 eV , which are all far more difficult to happen than . Therefore, by combining the analysis of cation effects on both and competing HER reactions, we were able to fully explain the necessary role of alkali metal cations in , which resulted from the inhibition of protons derived from approaching the surface, as well as the much higher kinetic barrier of protons derived from molecules compared to activation.
Note that only the qualitative trend is provided here. Typically, determining the reaction network requires a quantitative criterion based on differences in activation barriers. However, in this case, the critical factor is the inhibition of protons approaching the surface by cations. This implies that the concentration of reactants, influenced by potential, ionic activities, and mass transport, may play the dominant role. Furthermore, in reality, the interfacial concentration of alkali metal cations may deviate from one another under a specific potential due to different capacitance responses . Since only a few ions can be introduced at the nanoscale simulations due to scale limitations, it is challenging to reflect the concentration change accurately. Hence, this necessitates further exploration of both interfacial evolution and simulation methods.
Furthermore, our results derived from the fully explicitly solvated interfacial model offer significant insights into future electrochemical simulation work. Notably, the model indicates different adsorption configurations of the molecule in implicit and explicit solvents, distinct coordination structures of solvated alkali metal cations and intermediates, and the interaction between protons and alkali metal cations. The results suggest the importance of explicitly considering the electrolyte in simulation models, particularly when dealing with species that could potentially interact with the interfacial hydrogen bond network or other components within the electrical double layer, including solvated ions. Meanwhile, our investigations into the copper surface may aid in gaining a broader understanding of how alkali metal cations impact performance. This research can be combined with studies conducted on gold surfaces , where the alkali metal cations could differentiate the activation pathways between the inner- and outer-sphere mechanisms. After clarifying the various roles that alkali metal cations play on different coinage metals and even other catalysts, it is expected that a more systematic relationship between EDL design and performance can be established, which demands further exploration.
In conclusion, the results of a series of explicitly solvated AIMD simulations on and competing HER have provided systematic insights into the promotional roles and the necessary effects of alkali metal cations for . Energetically, the activation barrier decreases with increasing cation size. The different promotional roles can be attributed to the coordination abilities, or Onsager field strengths, of various alkali metal cations. In particular, smaller and cations can only coordinate with one oxygen atom of and simultaneously release a water molecule, whereas and cations can flexibly coordinate with both oxygen atoms of , which is more favorable for activation and reaction cycling. However, the partial desolvation of cation slightly inhibits the increasing tendency, resulting in its similarity with cation. On the other hand, the necessary role is attributed to the competition between HER and . The interfacial proton accessibility can be largely hindered by alkali metal cations, and the much higher kinetic barrier of protons derived from molecules helps to outcompete HER. Given the similarity in catalytic behavior with alkali metal cations, these phenomena are expected for other CO-selective catalysts, such as Ag
and Au. Our results help to understand the molecular origin of the possible role of the electrochemical environment emphasize the importance of explicit solvation, ions, and systematic exploration of competing reactions in electrochemical simulations, and show the effectiveness of theoretical tools in exploring the solid-electrolyte interface.

Methods

Solid-electrolyte-Ne double electrode model

Based on experimental results, it was determined that the surface would dominate after a prolonged electroreduction process . To represent the electrode, we chose the surface modeled as a three-layer slab with the supercell to simulate the activation processes, and larger supercell were utilized when considering the interactions between alkali metal cations and protons. The space above the Cu surface was filled with water molecules with a density of . In order to maintain the water density, one water molecule was replaced with a molecule. The volume of the solvent space was expanded with the addition of ions. Alkali metal cations ( , or ) were initially positioned away from the surface, while an anion was placed far away on the other side of the solvent to serve as an ionization pair with the cation, maintaining the applied potential close to the onset potential. An additional Ne atomic layer and a vacuum layer were added over the interface model to act as a counter electrode and monitor the electrochemical potential of the entire system .

AIMD simulations

All AlMD simulations were performed using the Vienna Ab initio Simulation Package (VASP) . The projector augmented wave (PAW) potential and the generalized gradient approximation (GGA) with the revised Perdew-Burke-Ernzerhof (RPBE) functional were employed to describe the electron-ion interactions and the exchange-correlation energy, respectively. The van der Waals interactions were described using the DFT-D3 correction . The energy cutoff for the plane wave basis expansion was set to 500 eV . For the Brillouin zone sampling, only the gamma point was used for AIMD simulations, while a denser grid was used for electronic structure calculations . The Nose-Hoover thermostat was used for canonical sampling at . The kinetic barriers and the free energy profile of the activation process were studied using a combination of slow-growth and blue-moon sampling methods with the SHAKE algorithm , with the angle of the molecule chosen as the collective variable (CV). The choice of CV is discussed in Supplementary Note 2.
Considering the applied potential, the Ne counter electrode was used to monitor the potential change of the reaction system on the fly. A dipole correction along the direction perpendicular to the interface was applied in the vacuum layer. The vacuum level of the metal side could be considered as constant due to the electrostatic screening of the metal, and thus the absolute potential of the system, i.e., the applied potential, could be calculated accordingly: vs SHE , where is the dipole moment of the system, is the surface area, and is Coulomb constant. Then, constant potential corrections were applied to account for the slight change in the electrochemical potential for different reaction states along the reaction paths (Supplementary Note 6) .

Data availability

The data supporting the findings of this study have been included in the main text and Supplementary Information. The representative AIMD trajectories generated in this study have been deposited in the Zenodo data repository at https://doi.org/10.5281/zenodo.10427196. All other relevant data supporting the findings of this study are available from the corresponding authors upon request.

References

  1. Nitopi, S. et al. Progress and perspectives of electrochemical reduction on copper in aqueous electrolyte. Chem. Rev. 119, 7610-7672 (2019).
  2. Huang, J. E. et al. electrolysis to multicarbon products in strong acid. Science 372, 1074-1078 (2021).
  3. He, M. et al. Selective enhancement of methane formation in electrochemical reduction enabled by a Raman-inactive oxygen-containing species on Cu. ACS Catal. 12, 6036-6046 (2022).
  4. Li, J. et al. Constraining CO coverage on copper promotes highefficiency ethylene electroproduction. Nat. Catal. 2, 1124-1131 (2019).
  5. Dinh, C. T. et al. electroreduction to ethylene via hydroxidemediated copper catalysis at an abrupt interface. Science 360, 783-787 (2018).
  6. Timoshenko, J. et al. Steering the structure and selectivity of electroreduction catalysts by potential pulses. Nat. Catal. 5, 259-267 (2022).
  7. Wang, L. et al. Electrochemical carbon monoxide reduction on polycrystalline copper: effects of potential, pressure, and pH on selectivity toward multicarbon and oxygenated products. ACS Catal. 8, 7445-7454 (2018).
  8. Chen, X. et al. Electrochemical -to-ethylene conversion on polyamine-incorporated Cu electrodes. Nat. Catal. 4, 20-27 (2021).
  9. Wang, R. et al. Partial coordination-perturbed bi-copper sites for selective electroreduction of to hydrocarbons. Angew. Chem. Int. Ed. 60, 19829-19835 (2021).
  10. Bagger, A. et al. Electrochemical reduction: classifying Cu facets. ACS Catal. 9, 7894-7899 (2019).
  11. Scholten, F. et al. Identifying structure-selectivity correlations in the electrochemical reduction of : a comparison of wellordered atomically clean and chemically etched copper singlecrystal surfaces. Angew. Chem. Int. Ed. 60, 19169-19175 (2021).
  12. Wu, Z.-Z. et al. Identification of interfaces as superior active sites for CO dimerization during electroreduction. J. Am. Chem. Soc. 144, 259-269 (2021).
  13. Akira, M. & Yoshio, H. Product selectivity affected by cationic species in electrochemical reduction of and CO at a Cu electrode. Bull. Chem. Soc. Jpn 64, 123-127 (1991).
  14. Hori, Y. et al. Formation of hydrocarbons in the electrochemical reduction of carbon dioxide at a copper electrode in aqueous solution. J. Chem. Soc. Faraday Trans. 85, 2309-2326 (1989).
  15. Resasco, J. et al. Promoter effects of alkali metal cations on the electrochemical reduction of carbon dioxide. J. Am. Chem. Soc. 139, 11277-11287 (2017).
  16. Ren, W. et al. A cation concentration gradient approach to tune the selectivity and activity of electroreduction. Angew. Chem. Int. Ed. 61, e202214173 (2022).
  17. Ayemoba, O. & Cuesta, A. Spectroscopic evidence of size-dependent buffering of interfacial pH by cation hydrolysis during electroreduction. ACS Appl. Mater. Interfaces 9, 27377-27382 (2017).
  18. Singh, M. R. et al. Hydrolysis of electrolyte cations enhances the electrochemical reduction of over Ag and Cu . J. Am. Chem. Soc. 138, 13006-13012 (2016).
  19. Malkani, A. S. et al. Cation effect on interfacial concentration in the electrochemical reduction reaction. ACS Catal. 10, 14871-14876 (2020).
  20. Gu, J. et al. Modulating electric field distribution by alkali cations for electroreduction in strongly acidic medium. Nat. Catal. 5, 268-276 (2022).
  21. Ringe, S. et al. Understanding cation effects in electrochemical reduction. Energy Environ. Sci. 12, 3001-3014 (2019).
  22. Chen, L. D. et al. Electric field effects in electrochemical reduction. ACS Catal. 6, 7133-7139 (2016).
  23. Monteiro, M. C. O. et al. Absence of electroreduction on copper, gold and silver electrodes without metal cations in solution. Nat. Catal. 4, 654-662 (2021).
  24. Qin, H.-G. et al. Quantitative understanding of cation effects on the electrochemical reduction of and in acidic solution. ACS Catal. 13, 916-926 (2022).
  25. Zhu, Q. et al. The solvation-induced Onsager reaction field rather than the double-layer field controls reduction on gold. JACS Au 2, 472-482 (2022).
  26. Groß, A. & Sakong, S. Ab initio simulations of water/metal interfaces. Chem. Rev. 122, 10746-10776 (2022).
  27. Qin, X. et al. Cation-coordinated inner-sphere electroreduction at Au-water interfaces. J. Am. Chem. Soc. 145, 1897-1905 (2023).
  28. Liu, H. et al. Promotional role of a cation intermediate complex in formation from electrochemical reduction of over Cu . ACS Catal. 11, 12336-12343 (2021).
  29. Yu, M. & Trinkle, D. R. Accurate and efficient algorithm for Bader charge integration. J. Chem. Phys. 134, 064111 (2011).
  30. Chan, K. & Nørskov, J. K. Electrochemical barriers made simple. J. Phys. Chem. Lett. 6, 2663-2668 (2015).
  31. Kronberg, R. & Laasonen, K. Reconciling the experimental and computational hydrogen evolution activities of Pt(111) through DFTbased constrained MD simulations. ACS Catal. 11, 8062-8078 (2021).
  32. Zhao, X. & Liu, Y. Origin of selective production of hydrogen peroxide by electrochemical oxygen reduction. J. Am. Chem. Soc. 143, 9423-9428 (2021).
  33. Le, J.-B. et al. Molecular origin of negative component of Helmholtz capacitance at electrified Pt(111)/water interface. Sci. Adv. 6, eabb1219 (2020).
  34. Vijay, S. et al. Unified mechanistic understanding of reduction to CO on transition metal and single atom catalysts. Nat. Catal. 4, 1024-1031 (2021).
  35. Cao, H. et al. Potential-dependent free energy relationship in interpreting the electrochemical performance of reduction on single atom catalysts. ACS Catal. 12, 6606-6617 (2022).
  36. Shin, S.-J. et al. A unifying mechanism for cation effect modulating and productions from electroreduction. Nat. Commun. 13, 5482 (2022).
  37. Le, D. & Rahman, T. S. On the role of metal cations in electrocatalytic reduction. Nat. Catal. 5, 977-978 (2022).
  38. Li, P. et al. Hydrogen bond network connectivity in the electric double layer dominates the kinetic pH effect in hydrogen electrocatalysis on Pt. Nat. Catal. 5, 900-911 (2022).
  39. Li, X.-Y. et al. Mechanism of cations suppressing proton diffusion kinetics for electrocatalysis. Angew. Chem. Int. Ed. 62, e202218669 (2023).
  40. Le, J.-B. et al. Molecular understanding of cation effects on double layer and its significance to CO-CO dimerization. Natl. Sci. Rev. 10, nwad105 (2023).
  41. Hussain, G. et al. How cations determine the interfacial potential profile: relevance for the reduction reaction. Electrochim. Acta 327, 135055 (2019).
  42. Garlyyev, B. et al. Influence of the nature of the alkali metal cations on the electrical double-layer capacitance of model Pt(111) and Au(111) electrodes. J. Phys. Chem. Lett. 9, 1927-1930 (2018).
  43. Xue, S. et al. How the nature of the alkali metal cations influences the double-layer capacitance of , and Pt single-crystal electrodes. J. Phys. Chem. C 124, 12442-12447 (2020).
  44. Qin, X. et al. Cation-induced changes in the inner- and outer-sphere mechanisms of electrocatalytic reduction. Nat. Commun. 14, 7607 (2023).
  45. Kwon, S. et al. Dramatic change in the step edges of the electrocatalyst upon exposure to CO: operando observations by electrochemical STM and explanation using quantum mechanical calculations. ACS Catal. 11, 12068-12074 (2021).
  46. Surendralal, S. et al. First-principles approach to model electrochemical reactions: understanding the fundamental mechanisms behind Mg corrosion. Phys. Rev. Lett. 120, 246801 (2018).
  47. Surendralal, S. et al. Impact of water coadsorption on the electrode potential of H-Pt(111)-liquid water interfaces. Phys. Rev. Lett. 126, 166802 (2021).
  48. Kresse, G. & Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169-11186 (1996).
  49. Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B 59, 1758-1775 (1999).
  50. Hammer, B. et al. Improved adsorption energetics within densityfunctional theory using revised Perdew-Burke-Ernzerhof functionals. Phys. Rev. B 59, 7413-7421 (1999).
  51. Grimme, S. et al. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132, 154104 (2010).
  52. Monkhorst, H. J. & Pack, J. D. Special points for Brillouin-zone integrations. Phys. Rev. B 13, 5188-5192 (1976).
  53. Hoover, W. G. Canonical dynamics: equilibrium phase-space distributions. Phys. Rev. A 31, 1695-1697 (1985).
  54. Nosé, S. A unified formulation of the constant temperature molecular dynamics methods. J. Chem. Phys. 81, 511-519 (1984).
  55. Woo, T. K. et al. A combined Car-Parrinello QM/MM implementation for ab initio molecular dynamics simulations of extended systems: application to transition metal catalysis. J. Phys. Chem. B 101, 7877-7880 (1997).
  56. Carter, E. A. et al. Constrained reaction coordinate dynamics for the simulation of rare events. Chem. Phys. Lett. 156, 472-477 (1989).
  57. Ryckaert, J.-P. et al. Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J. Comput. Phys. 23, 327-341 (1977).
  58. Neugebauer, J. & Scheffler, M. Adsorbate-substrate and adsorbateadsorbate interactions of Na and K adlayers on Al (111). Phys. Rev. B 46, 16067-16080 (1992).
  59. Chan, K. & Nørskov, J. K. Potential dependence of electrochemical barriers from ab initio calculations. J. Phys. Chem. Lett. 7, 1686-1690 (2016).

Acknowledgements

This work was supported by the National Key R&D Program of China (2022YFA1203400 and 2021YFA1400100), the National Science Foundation of China (12274254 and 11874036), the Local Innovative and Research Teams Project of Guangdong Pearl River Talents Program (2017BT01N111) and Basic Research Project of Shenzhen, China (JCYJ2O200109142816479 and WDZC2O200819115243002). Computational resources were provided by the High-Performance Computing Platform of Nanjing University of Aeronautics and Astronautics.

Author contributions

J.L. and F.K. conceived and designed the project. Z.Z. performed the calculation. Z.Z., J.L., H.L., Y.S., L.G., W.D. and H.H. analyzed the results. Z.Z., J.L. and F.K. wrote the paper, and all authors commented on it.

Competing interests

The authors declare no competing interests.

Additional information

Supplementary information The online version contains
supplementary material available at
https://doi.org/10.1038/s41467-024-44896-x.
Correspondence and requests for materials should be addressed to Feiyu Kang or Jia Li.
Peer review information Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.
Reprints and permissions information is available at
http://www.nature.com/reprints
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/ licenses/by/4.0/.
© The Author(s) 2024

  1. Shenzhen Geim Graphene Center and Institute of Materials Research, Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen 518055, People’s Republic of China. State Key Laboratory of Low Dimensional Quantum Physics and Department of Physics, Tsinghua University, Beijing 100084, People’s Republic of China. Institute for Advanced Study, Tsinghua University, Beijing 100084, People’s Republic of China. Frontier Science Center for Quantum Information, Beijing 100084, People’s Republic of China. Department of Energy Conversion and Storage, Technical University of Denmark, Kgs, Lyngby 2800, Denmark. e-mail: fykang@sz.tsinghua.edu.cn; li.jia@sz.tsinghua.edu.cn