اختيارية متطرفة لليثيوم والمغنيسيوم من خلال تمييز دقيق لحجم الأيونات في غشاء البولي أميد Extreme Li-Mg selectivity via precise ion size differentiation of polyamide membrane

المجلة: Nature Communications، المجلد: 15، العدد: 1
DOI: https://doi.org/10.1038/s41467-024-46887-4
PMID: https://pubmed.ncbi.nlm.nih.gov/38509082
تاريخ النشر: 2024-03-20

اختيارية متطرفة لليثيوم والمغنيسيوم من خلال تمييز دقيق لحجم الأيونات في غشاء البولي أميد

تاريخ الاستلام: 18 أكتوبر 2023
تم القبول: 13 مارس 2024
نُشر على الإنترنت: 20 مارس 2024

كوان بينغ رويو وانغ زيلين زهاو شيهونغ لين ينغ ليو دياني دوونغ تشنغ وانغ يمن يوزهانغ زو جيان جين لي جيانغ (1)

الملخص

تحقيق انتقائية عالية لـ و يعتبر أمرًا بالغ الأهمية لاستخراج الليثيوم الفعال من المحاليل الملحية، وتلعب أغشية النانوترشيح (NF) دورًا حاسمًا في هذه العملية. يكمن المفتاح لتحقيق انتقائية عالية في التصميم حسب الطلب لثقوب أغشية NF بما يتوافق مع فرق الحجم بين و الأيونات، لكن هذا يشكل تحديًا كبيرًا لأغشية NF التقليدية ومن الصعب تحقيقه. في هذا العمل، نبلغ عن تصنيع أغشية NF من البولي أميد (PA) ذات الأداء الفائق. الانتقائية من خلال تعديل عملية البوليمرة الواجهة (IP) بين البيبيرازين (PIP) وكلوريد التريميزويل (TMC) باستخدام مادة خافضة للتوتر السطحي قابلة للذوبان في الزيت تشكل طبقة أحادية عند واجهة الزيت/الماء، والتي يشار إليها باسم OSARIP. تستفيد OSARIP من تنظيم مسام الغشاء بحيث تكون جميعها أصغر من الأيونات. تحت تأثير الغربلة الحجمية فقط، استثنائي معدل الرفض لأكثر من يتم تحقيقه. وهذا يؤدي إلى مستوى عالٍ استثنائي الانتقائية، التي هي أعلى بمقدار من مرتبتين إلى ثلاث مراتب من جميع الأغشية المدفوعة بالضغط المبلغ عنها حاليًا، وأعلى حتى من المواد الإطارية الدقيقة، بما في ذلك COFs وMOFs وPOPs. قد يؤدي التحسين الكبير في أداء فصل الأيونات لأغشية NF إلى ابتكار عملية استخراج الليثيوم الحالية وتحسين كفاءة استخراج الليثيوم بشكل كبير.

تحتفظ البحيرات المالحة احتياطيات الليثيوم القابلة للاستخراج في العالم استكشاف الوصول الفعال إلى موارد الليثيوم في البحيرات المالحة أمر حاسم لتوفير إمدادات مستقرة من الليثيوم لتلبية الطلب المتزايد بسرعة على صناعة بطاريات الليثيوم أيون. أما بالنسبة لاستخراج الليثيوم من المحلول الملحي، فإن إحدى العمليات الرئيسية هي فصل و الأيونات نظرًا لوجود أيونات أخرى ذات وفرة أعلى، مثل ، و سيتم إزالته مسبقًا في شكل ترسيب من خلال التبخر المدفوع بالطاقة الشمسية فعال الفصل لا يزال يواجه تحديات كبيرة بسبب محتوى في المحلول الملحي أعلى بكثير من (على سبيل المثال، محتوى عادة ما يكون أكثر من 20 مرة من
في مياه الملح في الصين ) وأحجام و الأيونات متشابهة جدًا .
تكنولوجيا النانوفلترة المعتمدة على الضغط (NF) القائمة على فصل الأغشية هي واحدة من أكثر التقنيات اهتمامًا بين طرق المعالجة المستخدمة عادة لاستخراج الليثيوم من المياه المالحة. . ومع ذلك، فإن أغشية NF الحالية تظهر انخفاضًا نسبيًا في انتقائية الفصل، عادة أقل من 20. لذلك، في عملية استخراج الليثيوم، يجب استخدام عمليات فصل غشائية متكررة متعددة للحصول على نقاء كافٍ. في الراشح لتوليد أو ترسيب LiOH كمنتج مستهدف. الـ
الانتقائية المحدودة لأغشية NF ناتجة بشكل رئيسي عن حجم المسام غير المتساوي وتوزيعها الواسع. عادةً ما يتم تحقيق الفصل من خلال التأثير المشترك لفرز الحجم واستبعاد دونان زيادة إيجابية الشحنة في الغشاء لتعزيز استبعاد دونان قد تم استخدامها على نطاق واسع لتعزيز التنافر إلى وبذلك تحسين انتقائية الفصل . ومع ذلك، يمكن أن يزيد الاستبعاد القوي لدونان أيضًا من النفور إلى الأيونات، مما يجعل من الصعب زيادة بشكل كبير الـ الانتقائية وتقليل معدل الاسترداد في نفس الوقت لا شك أنه إذا استطعنا تنظيم مسام غشاء NF لتحقيق الفصل الذي يعتمد فقط على غربلة حجم المسام، قد يتجاوز قيود التنافر الكهربائي، وبالتالي تحقيق تحسين جذري في أداء الفصل.
تحضير أغشية NF المتطورة يعتمد على تفاعل البوليمرة البينية (IP) بين أحادي الأمين المذاب في الماء وأحادي كلوريد الأسيلي المذاب في المذيب العضوي لتشكيل طبقة بولياميد (PA) انتقائية رقيقة جداً. خلال هذه العملية، يكون انتشار المونومرات نحو الواجهة (بشكل رئيسي المونومرات الأمينية من الطور المائي إلى المذيب العضوي) بطيئًا وعشوائيًا. بينما تكون سرعة التفاعل بين الأمين وكلوريد الأسيل سريعة والتفاعل عنيف. . يؤدي ذلك إلى هيكل غير متساوٍ في طبقة الاختيار PA، مما ينتج عنه توزيع واسع لحجم المسام يصعب التحكم فيه لتحقيق تنظيم دقيق لثقوب الغشاء وفصل انتقائي عالي للأيونات المستهدفة، هناك حاجة إلى طريقة فعالة للتحكم في عملية IP لتحسين انتظام التفاعل. للأسف، لا يزال البحث في هذا المجال غير ناضج وهناك تقارير قليلة ذات صلة. لقد أبلغنا سابقًا عن استراتيجية استخدام المواد السطحية القابلة للذوبان في الماء لتشكيل طبقات أحادية مرتبة عند واجهة الماء/الزيت لتنظيم عملية IP (المشار إليها باسم SARIP) لتحسين انتظام حجم الثقوب في أغشية PA NF. . أظهرت أغشية PA NF الناتجة ذات توزيع حجم المسام الضيق بشكل كبير معدل رفض يزيد عن لكل الكاتيونات ثنائية التكافؤ. على الرغم من أن استراتيجية SARIP قد أظهرت فعاليتها في تنظيم مسام الغشاء، إلا أنه لا يزال من الصعب تلبية متطلبات الفصل عالي الانتقائية. و نطاق حجم المسام لغشاء PA NF المُعد من SARIP هو Å (بما في ذلك جميع المسام القابلة للكشف). قطر الترطيب لـ أيون هو Å، بينما قطر ستوكز لـ أيون Å (مع الأخذ في الاعتبار أن الأيون عرضة للجفاف أثناء الفصل المدفوع بالضغط بسبب طاقته المنخفضة في الترطيب. ). من الواضح أن يمكن للأيونات أن تخترق الغشاء من خلال ثقب جزئي، مما يؤدي إلى عدم كفاية الانتقائية (أقل من 50).
في عملية SARIP، بالإضافة إلى طبقة أحادية من السطحي عند واجهة الماء/الزيت، هناك أيضًا عدد كبير من جزيئات السطحي الحرة وتجمعاتها مثل الميسيلات في الطور المائي. ستؤدي وجود هذه الجزيئات الحرة وتجمعاتها إلى إضعاف جزئي للتأثير التنظيمي لطبقات السطحي الأحادية على الانتشار الواجهوي لمونومرات الأمين، مما يحد من تحسين تجانس مسام الغشاء. . في هذا العمل، نبلغ عن استخدام مادة خافضة للتوتر السطحي قابلة للذوبان في الزيت (أي فوسفات الدوديسيل (DDP)) تم إدخالها في مرحلة المذيب العضوي لتنظيم تفاعل IP، والذي يُشار إليه باسم OSARIP. الطبيعة الأمفيلية لجزيئات DDP تمكنها من تشكيل طبقات أحادية مجمعة عند واجهة الماء/الزيت. نظرًا لعدم وجود جزيئات خافضة للتوتر السطحي حرة في المحلول المائي، يمكن لطبقة DPP الأحادية عند واجهة الماء/الزيت أن تمارس تأثيرها بالكامل في تسهيل انتشار المونومرات الأمينية وتنظيم تفاعل IP. كما هو متوقع، تجعل عملية OSARIP مسام أغشية PA NF أكثر انتظامًا، وتضيق توزيع حجم المسام ليكون في النطاق Å. هذا يعني أن معظم المسام أصغر من قطر الترطيب لـ الأيون، ومعظم المسام أكبر من قطر ستوك أيون (مع الأخذ في الاعتبار أن الأيون عرضة للجفاف أثناء الفصل المدفوع بالضغط بسبب طاقته المنخفضة في الترطيب. لذلك، بالاعتماد فقط على تأثير غربلة الحجم، تظهر غشاء PA NF خصائص فائقة الارتفاع رفض يصل إلى وعالي جداً انتقائية تصل إلى 4147. الفصل الفائق العلو
انتقائية يجعل من الممكن الحصول على محلول ليثيوم عالي النقاء من خلال عملية ترشيح غشائية واحدة. قد يحول هذا تدفق التكنولوجيا الحالي لاستخراج الليثيوم من المحلول الملحي، مما يبسط العملية بشكل كبير ويحسن كفاءة الفصل.

النتائج

لتنظيم تفاعل IP بشكل فعال، تم إدخال مادة خافضة للتوتر السطحي قابلة للذوبان في الزيت، DDP، في محلول الهكسان. الطبيعة الأمفيبيلية لـ DDP تسمح بتكوين طبقة أحادية عند واجهة الزيت/الهكسان. تم ملاحظة أدنى توتر سطحي عند تركيز DDP من (انظر الشكل التكميلي 1)، مما يدل على تشكيل الكثافة الجزيئية الأكثر كثافة. وجود طبقة أحادية من DDP يمكن أن يمتص جزيئات PIP إلى واجهة الماء/الهكسان عبر التفاعل الكهروستاتيكي بين المجموعة الفوسفونية في DDP ومجموعة الأمين في PIP (الشكل 1أ). سيؤدي ذلك إلى زيادة تركيز PIP بالقرب من واجهة الزيت/الهكسان، ليصل إلى أعلى نقطة عند تركيز DDP قدره , كما يتضح من طيف الامتصاص UV-vis. (الشكل 1ب، والشكل التكميلي 2). يكشف المراقبة في الموقع لامتصاص PIP بالقرب من واجهة الزيت/الهكسان خلال تفاعل IP عن انخفاض أسرع في امتصاص PIP عندما تكون هناك طبقة أحادية من DDP عند الواجهة (الشكل 1ب). وهذا يشير إلى أن الطبقة الأحادية من DDP مفيدة في تعزيز انتشار PIP عبر الواجهة. بالإضافة إلى تأثير التراكم لطبقة DDP الأحادية، تعمل جزيئات DDP الحرة في الهكسان كحاملات نقل من خلال عملية التبادل الديناميكي مع جزيئات الطبقة الأحادية للمساعدة في انتشار جزيئات PIP عبر الواجهة . علاوة على ذلك، فإن تجميع DDP يقلل بشكل كبير من توتر واجهة الزيت/الماء، مما يفيد في تقليل عائق انتشار PIP وتعزيز التجانس المكاني لعملية الانتشار. من المهم ملاحظة أن عملية OSARIP تقضي على وجود أي جزيئات أو ميكيلات من السطح النشط الحرة في المحلول المائي، مما يتجنب أي تنافس مع تجميع السطح النشط عند الواجهة ويخلق حالة مثالية لإطلاق الإمكانات الكاملة لتجميع السطح النشط في التحكم في انتشار PIP عبر الواجهة.
في عملية IP التقليدية، يكون انتشار PIP عبر الواجهة بطيئًا وعشوائيًا، مما يجعل من الصعب على إمداد PIP مواكبة استهلاكه. لذلك، يظهر تفاعل IP عدم استمرارية مكانية وزمنية، مما يؤدي في النهاية إلى عدم تجانس كيميائي وبنيوي في غشاء PA. في عملية OSARIP، يكون انتشار PIP سريعًا ومتجانسًا مكانيًا، مما يقلل بشكل كبير من الفرق الكبير بين معدل انتشار المونومر ومعدل تفاعل IP، وبالتالي يعزز استمرارية تشكيل طبقة PA في الزمن والمكان. مقارنةً بطبقة PA الانتقائية المعدة من IP التقليدي، فإن درجة الربط المتقاطع لطبقة PA الانتقائية المعدة من OSARIP قد زادت بشكل كبير (من إلى ) بناءً على تحليل التركيب العنصري باستخدام مطياف الأشعة السينية (XPS) (الشكل 1ج، د، والأشكال التكملية 3-6)، ويظهر الغشاء شحنة سالبة أقل بسبب وجود مجموعات غير متفاعلة متبقية أقل (انظر الشكل التكميلي 7).
يعرض الشكل 1د-ز الهيكل الطوبولوجي لغشاء PA NF المعد من كل من IP التقليدي وOSARIP. بالمقارنة، يظهر غشاء PA NF المعد من OSARIP بقع محدبة صغيرة وطبقة PA الانتقائية أرق ( للغشاء المعد من IP التقليدي و للغشاء المعد من OSARIP). قمنا بمزيد من التحقيق في معلومات حجم المسام للغشاء من خلال إجراء تحليل قياسي لرفض الجزيئات المحايدة ذات الأوزان الجزيئية المختلفة. مع زيادة تركيز DDP في الهكسان، ينخفض ​​حد القطع لوزن الجزيء (MWCO) للغشاء تدريجياً ويصل إلى أدنى قيمة قدرها 134 دالتون عند تركيز حرج قدره (الشكل 2أ). حسب علمنا، لم يتم تحقيق مثل هذا MWCO الصغير بعد في غشاء PA NF المعد من تفاعل IP بين PIP وTMC المبلغ عنه حتى الآن. إن زيادة تركيز DDP إلى يزيد فعليًا من MWCO للغشاء إلى 185 دالتون.
الشكل 1 | تصنيع، التركيب الكيميائي، والهيكل الطوبولوجي لأغشية PA NF. أ مخطط لتوضيح تأثير تجميع DDP عند واجهة الزيت/الماء على انتشار PIP عبر الواجهة لتعزيز تجانس تفاعل IP. تغير امتصاص PIP بالقرب من الواجهة بالنسبة لزمن التفاعل الذي تم مراقبته بواسطة قياس UV-vis. ج التركيب العنصري الكيميائي
ودرجة الربط المتقاطع لأغشية PA NF المعدة من IP التقليدي وOSARIP. د-ز الهيكل السطحي والمقطع العرضي لأغشية PA المعدة من IP التقليدي وOSARIP ( صور SEM من الأعلى. ف، ز صور TEM مقطعية).
استنادًا إلى فرضية التوزيع اللوغاريتمي الطبيعي، تم ملاءمة توزيع حجم المسام للأغشية من منحنى الرفض للجزيئات المحايدة وتم التعبير عنه كدالة كثافة احتمالية (PDF) (انظر القسم 1.4.2 من المعلومات التكميلية). بما يتماشى مع اتجاه تغيير MWCO، فإن إدخال DDP يضيق بشكل كبير توزيع حجم المسام لأغشية PA NF وينتج عنه توزيع حجم مسام ضيق للغاية مع معامل شكل قدره 1.13 (يُعرف بأنه الانحراف المعياري الهندسي لمنحنى PDF) عند تركيز DDP قدره (الشكل 2ب). من الجدير بالذكر أن حجم المسام يقع في نطاق غشاء PA NF المعد من IP التقليدي، مما يشير إلى أن الطبقة الأحادية من DDP تحسن بشكل أساسي تجانس حجم المسام، بدلاً من تقليل حجم المسام. هذه المسام المتجانسة للغاية تمكن الغشاء من تحقيق رفض فائق لجميع الأملاح الثنائية التكافؤ بما في ذلك الملح الأنيوني لـ والملح الكاتيوني لـ (جميعها فوق 99%) (الشكل 2ج، الشكل التكميلي 8 والجدول التكميلي 1)، على الرغم من أن الغشاء مشحون سالبًا. من المثير للاهتمام أن هيكل المسام ورفض الملح لغشاء PA NF المعد من OSARIP عند تركيز DDP قدره يتأثر قليلاً بوقت التفاعل (انظر الأشكال التكملية 9 و10)،
ما يثبت بشكل أكبر تأثير تسهيل تجميع DDP على حركية IP.
عند تحليل أغشية NF الموجودة في الأدبيات (الجدول التكميلي 2)، من الواضح أن أغشية NF ذات أحجام المسام الصغيرة فعالة للغاية في رفض دون الاعتماد كثيرًا على استبعاد دونان (الشكل 2د). تؤكد هذه الاكتشافات الإمكانات الهائلة لاستخدام مثل هذه الأغشية في التطبيقات التي تتطلب إزالة فعالة للكاتيونات الثنائية التكافؤ عن طريق غربلة الحجم (مثل، الفصل لاستخراج الليثيوم من المحاليل الملحية). مقارنةً بأغشية NF الموجودة، يمتلك غشاء PA NF الخاص بنا أصغر حجم مسام فعال وتوزيع حجم مسام ضيق للغاية، مما يظهر أعلى رفض لـ . كما يظهر رفضًا مرتفعًا جدًا لكتل كاتيونات ثنائية التكافؤ أخرى، مثل لـ لـ لـ ، و لـ (الشكل التكميلي 11 والجدول التكميلي 3). في الوقت نفسه، بسبب عملية إزالة الماء، فإن العائق المكاني لمسام الغشاء لا يمكن أن يعيق اختراق الكاتيونات أحادية التكافؤ مثل ، مما يؤدي إلى رفض نسبي منخفض ( ) لها. وبالتالي، يتم تقديم منطقة انتقال حادة جدًا بواسطة الغشاء عند ربط الأيون
الشكل 2 | حجم المسام ورفض الملح الأحادي لأغشية PA NF المعدة من OSARIP. أ رفض الجزيئات المحايدة بواسطة الأغشية المعدة بتركيزات مختلفة من DDP في الهكسان. توزيع حجم المسام للأغشية المقابلة الملائمة من منحنى الرفض للجزيئات المحايدة وفقًا لمبدأ التوزيع اللوغاريتمي الطبيعي. ج رفض الملح للغشاء المعد من IP التقليدي وOSARIP. د ملخص لرفض بواسطة غشاء PA المعد من OSARIP في الظروف المثلى (خماسي أحمر) وأغشية NF
المبلغ عنها من قبل الآخرين (مربع، انظر الوصف التفصيلي في الجدول التكميلي 2). هـ مقارنة نطاق حجم المسام لأغشية PA NF المعدة من IP التقليدي، SARIP، OSARIP، والأغشية NF المتاحة تجاريًا (DK التي تنتجها سويز وNF270 التي تنتجها دوبونت). جميع الأغشية مشحونة سالبًا بناءً على نتائج قياس الجهد المتدفق. يمثل شريط الخطأ الانحراف المعياري لثلاث قياسات مكررة.
الرفض مع نصف قطر ستوك (الشكل التكميلي 11)، مما يشير إلى تفوق انتقائية الأيون-الأيون بين الكاتيونات أحادية التكافؤ والثنائية التكافؤ.
من أجل التأكيد على القدرة المذهلة لتجميع DDP في تحسين توزيع حجم المسام، قمنا بمقارنة نطاقات حجم المسام لعدة أنواع تمثيلية من أغشية PA NF بما في ذلك أغشية NF المستمدة من عمليات OSARIP وSARIP وعمليات IP التقليدية، بالإضافة إلى أغشية DK وNF270 المتاحة تجارياً (نطاقات حجم المسام المستمدة من منحنيات PDF كما أبلغ عنها آخرون). ) (الشكل 2e والشكل التكميلي 12). يتم أخذ نطاق حجم المسام لهذه الأغشية مباشرة من منحنيات PDF الخاصة بها، حيث يتم احتساب جميع المسام التي تزيد كثافتها الاحتمالية عن 0.01 للإحصائيات. يظهر بوضوح أن غشاء PA NF الذي تم إعداده من عملية IP التقليدية لديه نطاق واسع من حجم المسام Å، تليها أغشية NF270 و DK التي تتراوح أحجام مسامها بين Å و Å“، على التوالي. في المقابل، أظهر إدخال المواد السطحية عند واجهة الزيت/الماء تأثيرًا كبيرًا على تحسين نطاق حجم المسام للأغشية. عملنا السابق أظهر فعالية استخدام طبقة أحادية من السطح النشط القابل للذوبان في الماء لتنظيم تفاعل IP (مثل SARIP)، مما أدى إلى غشاء PA NF بحجم مسام يتراوح بين Å. هنا، تقوم عملية OSARIP المطورة حديثًا بتضييق نطاق حجم مسام الغشاء بشكل أكبر ليكون Åمع الأخذ في الاعتبار أن قطر الترطيب لـ أيون هو Å، فهذا يعني أن جميع مسام غشاء PA NF المُعد من عملية OSARIP أصغر تقريبًا من قطر الترطيب لـ الأيونات، مما يؤدي بالتالي إلى رفض فائق لـ أيونات.
ال تمت دراسة أداء الفصل لأغشية PA NF التي تم إعدادها من عملية OSARIP بشكل أعمق باستخدام خليط ملح ثنائي يتكون من و LiCl بتركيز إجمالي للملح يبلغ 2000 جزء في المليون و نسبة الكتلة لـ كمحلول ملحي محاكى. يظهر رفض حتى (الشكل 3أ)، بينما الرفض السلبي لـ إلى . المقابل / الانتقائية تصل إلى 828. بالمقارنة، فإن غشاء PA NF المُعد من طريقة IP التقليدية يظهر رفضاً لـ إلى و إلى ، على التوالي، مما يتوافق مع انتقائية 16.
تركيب الأيونات في مياه البحيرات المالحة الفعلية متنوع، لذا قمنا بدراسة أداء فصل غشاء PA NF تحت ظروف مختلفة. النسب. عندما تظل تركيزات الملح الكلي ثابتة عند 2000 جزء في المليون و نسبة الكتلة في خليط الملح الثنائي تزيد من 10:1 إلى 60:1، الرفض إلى يزداد قليلاً من إلى ورفض ينخفض من إلى (الشكل 3أ والجدول التكميلي 4). هذا يتوافق مع زيادة الانتقائية من 548 إلى 4147. عند الحفاظ على نسبة الكتلة 20:1 وزيادة التركيز الكلي للملح من 2000 إلى 5000 جزء في المليون، فإن رفض ينخفض قليلاً من إلى بينما رفض يخضع لتغيير كبير، ينخفض من إلى وبالمثل (الشكل 3ب والجدول التكميلي 5). وهذا يسمح لغشاء PA NF بأن يظهر مستوى عالٍ من الانتقائية تصل إلى 185 في محلول عالي الملوحة. في التطبيقات الصناعية العملية لتقنية NF لاستخراج الليثيوم من المياه المالحة، عادةً ما تكون محاليل التغذية
شكل. أداء فصل أغشية PA NF باستخدام خليط ملح ثنائي و كغذاء. نوع من الرفض لـ و و الانتقائية فيما يتعلق بـ نسبة الكتلة (تركيز الملح في محلول التغذية هو 2000 جزء في المليون). تباين الرفض و و الانتقائية بالنسبة لتركيز الملح في محلول التغذية
(الـ نسبة الكتلة هي ). ج مقارنة لـ انتقائية غشاء PA NF الذي تم إعداده في هذا العمل والأغشية المبلغ عنها من قبل الآخرين بما في ذلك غشاء NF البوليمري، وأغشية MOFs، وأغشية COFs، وأغشية POPs (انظر الوصف التفصيلي في الجداول التكميلية 7 و 8). يمثل شريط الخطأ الانحراف المعياري لثلاث قياسات مكررة.
تظهر تركيزات أعلى وتحتوي على أيونات معدنية إضافية، مثل ، و حتى في هذه الظروف، تُظهر غشاء PA NF استثنائية انتقائية تزيد عن 200 (الأشكال التكميلية 13 و 14 والجدول التكميلية 6). تؤكد هذه الأداء الملحوظ على التفوق الفريد للأغشية ذات أحجام المسام الموحدة لاستخراج الليثيوم من محلول ملحي في البحيرات المالحة.
قمنا بمقارنة انتقائية أغشية التناضح النانوي (PA NF) المعدة من OSARIP مع أكثر الأغشية تمثيلاً التي تم الإبلاغ عنها حتى الآن، بما في ذلك أغشية NF المدفوعة بالضغط، وأغشية الإطار المسامي مثل أغشية الإطارات العضوية التساهمية (COFs)، وأغشية الإطارات العضوية المعدنية (MOFs)، وأغشية البوليمرات العضوية المسامية (POPs) (الشكل 3c والجداول التكميلية 7 و8). انتقائية غشاء NF PA المُعد من OSARIP هي حالياً أعلى قيمة، والتي تتراوح بين مرتبتين إلى ثلاث مراتب أعلى من أغشية NF المدفوعة بالضغط التي تم الإبلاغ عنها حتى الآن، وأعلى بكثير من أغشية الإطار الميكرو مسامي الناشئة التي تعتبر ذات هياكل مسام مثالية موحدة.
للمزيد من التحقيق في التأثير الكبير لبنية المسام على انتقائية الأيونات في أغشية PA NF، قمنا بإعداد سلسلة من الأغشية بحجم مسام يتراوح من واسع إلى ضيق من خلال تعديل تركيز DDP في الهكسان ضمن نطاق من 0 إلى (الشكل 4أ، الشكل التكميلي 15 والجداول التكملية 9 و10). مع تضييق نطاق حجم المسام، ينخفض متوسط حجم المسام في الوقت نفسه. درسنا تغير و الرفض و
الانتقائية على متوسط حجم المسام لأغشية PA NF المحضرة باستخدام خليط الملح الثنائي من و LiCl بتركيز إجمالي للملح يبلغ 2000 جزء في المليون و نسبة الكتلة 20:1 كعلف. الـ تزداد نسبة الرفض تدريجياً مع انخفاض متوسط حجم المسام وتصل إلى أخيرًا (الشكل 4ب). ومع ذلك، فإن تظهر الرفض اتجاهًا معاكسا. يتم تقديم رفض أكثر سلبية بواسطة الغشاء الذي لديه حجم مسام متوسط أصغر. مع تغير رفض الأيونين في اتجاهين متعاكسين، / تزداد الانتقائية بشكل أسي مع انخفاض متوسط حجم المسام.
من أجل حساب الزيادة الكبيرة في الانتقائية من خلال تضييق نطاق حجم المسام، بالإضافة إلى الرفض السلبي المتزايد لـ استخدمنا نموذج انتشار الأيونات في المحلول (SDEM) لوصف نقل الأيونات عبر الغشاء في خليط ملحي ثنائي. لقد تم استخدام نموذج SDEM على نطاق واسع لشرح سلوك نقل الأيونات من خلال أغشية التناضح العكسي (RO) / التناضح النانوي (NF) / التناضح الأمامي (FO). من خلال تقديم مفهوم الحل الافتراضي وتجاهل الربط الحملاني بين نقل المذيب والمذاب عبر الغشاء، يتم دفع نقل الأيونات بشكل رئيسي بواسطة تدرجات التركيز والجهد الكهربي. تدفق الأيونات للأنواع يمكن وصفها في نموذج SDEM باستخدام معادلة نيرنست-بلانك المعدلة (1):
الشكل 4 | تأثير مسام أغشية PA على أداء الفصل لتمييز و نطاق حجم المسام لأغشية PA NF التي تم إعدادها بالظروف المذكورة بالتفصيل في الجدول التكميلي 8. تغير نفاذية الماء، رفض رفض، و الانتقائية على المتوسط
حجم المسام لأغشية PA المُعدة. ج. تغير نفاذية الأيونات كدالة لحجم المسام المتوسط. رسم توضيحي تخطيطي للحقل الكهربائي الذي ينشأ بشكل تلقائي أثناء عملية الفصل عند استخدام خليط ملح ثنائي. وكلوريد الليثيوم كمادة خام.
أين هو نفاذية الأيونات، هو تركيز الأيون في محلول افتراضي في حالة توازن حراري مع مرحلة الغشاء، هو الإحداثي عبر الغشاء الذي تم تطبيعه بواسطة سمك الغشاء، هو التكافؤ للنوع ، و هو الجهد الكهربائي غير البعدي في الحل الافتراضي. وبالتالي، فإن النقل الانتقائي لـ و عبر الغشاء يتميز باختلاف نفاذية الأيونات في نموذج SDEM. نفاذية الأيونات، يمكن أن يتم تركيبه من نتائج الرفض التجريبية، مع الأخذ في الاعتبار تركيبة التغذية ومعدل تدفق النفاذ المقاس. كما هو موضح في الشكل 4c، يمكن أن تبقى النفاذية قريبة من مع انخفاض متوسط حجم المسام بسبب تأثير الجفاف. بالمقارنة، تنخفض النفاذية على مدى ثلاثة أوامر من حيث الحجم. الفرق الكبير بين نفاذية الأيونات لـ و يؤدي إلى المستوى الفائق الارتفاع الانتقائية.
التغيير من يمكن أن يُعزى الرفض نحو اتجاه أكثر سلبية مع انخفاض متوسط حجم المسام إلى تأثير الاقتران بين نقل الأيونات عبر الغشاء. و عبر مجال كهربائي ينشأ بشكل عفوي، كما هو موضح بشكل تخطيطي في الشكل 4d. على وجه التحديد، هو الملح السائد في المحلول الملحي المحاكى وتركيز يتحدد بشكل رئيسي بواسطة تركيز. اعتراض يمكن أن تؤدي الأيونات إلى تراكم الأيونات بالقرب من سطح الغشاء. كما يزداد الرفض، تدرج التركيز لـ عبر الغشاء يزداد، مما يؤدي إلى نقل أسرع لـ مرافقة النقل لـ ، فإن نقل الأيون المضاد ذو الشحنة المعادلة ضروري للحفاظ على حيادية الشحنة في السائل المنفذ، مما يساهم في إنشاء تدرج الجهد التلقائي. بالنظر إلى الحواجز الاستيركية الأكبر لـ للتقسيم إلى مسام الغشاء، مما يسرع النقل هو الخيار الوحيد.
من المهم أن نلاحظ أن نقل الأيونات عبر المسام دون النانومتر هو عملية معقدة للغاية تتأثر بعوامل مختلفة
عوامل، بما في ذلك حجم المسام، وخصائص الشحنة، والتركيب الكيميائي، بالإضافة إلى التفاعل مع الماء وأيونات أخرى. في حالة محلول ملحي ثنائي يتكون من و LiCl ، سلوك نقل الكتلة لـ و يختلف بشكل كبير. من أجل نقل الكتلة لـ يمكن أن توفر المسام المتجانسة للغاية في غشاء PA NF رفضًا عاليًا ومستقرًا لـ من خلال فصل يعتمد فقط على الحجم. أما بالنسبة لانتقال الكتلة من يؤدي تأثير الجفاف إلى جعل مسام غشاء PA NF لها تأثير محدود في عرقلة التقسيم والنقل لـ من خلال الغشاء، بينما تلعب خاصية الشحنة والبنية الكيميائية للأغشية دورًا مهمًا، بالإضافة إلى تأثير الاقتران مع الأيون المضاد لـ تعتبر أغشية PA NF المشحونة سلبًا مفيدة للحصول على مستويات عالية النفاذية بسبب الشد عندما تكون أغشية PA NF مشحونة بشكل إيجابي عالي، فإن الجذب الكهروستاتيكي سيبطئ من هجرة . في الوقت نفسه، فإن الاستبعاد الناتج عن الشحنة الموجبة ينتج أيضًا عائقًا قويًا لتقسيم إلى مسام الغشاء. ونتيجة لذلك، فإن هناك رفضًا أعلى لـ يتم تقديمه بواسطة الغشاء، وهو غير مرغوب فيه للهدف من الاسترداد الدليل القوي على ذلك هو أن غشاء PA NF المُعد من OSARIP باستخدام بولي إيثيلين إيمين (PEI) وTMC كأحاديين، واللذان يُعتبران دائماً ذو شحنة إيجابية قوية. ، عرضت مستوى أعلى بكثير الرفض مقارنة بغشاء PA المُعد من OSARIP باستخدام PIP و TMC كأحاديين عند استخدام نفس المحلول الملحي المحاكي كغذاء (الشكل التكميلي 16 والجداول التكملية 11، 12). تشير هذه النتائج أيضًا إلى تفوق أغشية PA NF سالبة الشحنة ذات توزيع حجم المسام المتجانس للغاية لاستخراج من المحلول الملحي.

نقاش

باختصار، قمنا بتطوير طريقة بسيطة وفعالة، تُعرف باسم OSARIP، لتنظيم انتشار PIP عبر الواجهة داخل IP.
رد الفعل لإنتاج أغشية PA NF ذات مسام متجانسة للغاية. من خلال الاستفادة من التأثير الترويجي الكامل لطبقة أحادية من السطح النشط القابل للذوبان في الزيت عند واجهة الزيت/الماء على انتشار PIP، فإن تفاعل IP بين PIP وTMC ينتج عنه غشاء PA NF ذو توزيع حجم مسام ضيق للغاية يقع في النطاق بين قطر الترطيب لـ قطر الأيون وستوك أيون. هذه المسام المتجانسة للغاية تمكّن الغشاء من رفض فائق الارتفاع لـ إلى وانخفاض نسبي في الرفض إلى اختبار الأداء، باستخدام خليط ملحي ثنائي من LiCl و لتمثيل المحلول الملحي، أظهر أن غشاء PA NF المُعد من عملية OSARIP يظهر أداءً فائقًا للغاية الانتقائية تزيد عن 4000. أظهرت التحقيقات الإضافية في سلوك نقل الأيونات مع الانخفاض التدريجي في حجم المسام الفعالة أن العائق الحجمي له تأثير محدود على نفاذية ولكن تقلل بشكل كبير من نفاذية ، مما أدى إلى الزيادة الأسية في الانتقائية. بشكل عام، يقدم عملنا طريقة سهلة ودقيقة لتحسين حجم المسام لأغشية PA NF على مقياس أنغستروم، والتي يمكن تشغيلها بسهولة في العملية الصناعية الحالية لتحضير أغشية NF ذات أداء محسّن بشكل كبير لتطبيقات واسعة في معالجة المياه والفصل الكيميائي.

طرق

تصنيع أغشية PA NF

في البداية، محلول مائي يحتوي على محاليل الهكسان التي تحتوي على تم تحضير TMC وكمية معينة من فوسفات الدوديسيل (DDP) على التوالي. ثم تم تسخين النوعين من المحاليل إلى وتم الاحتفاظ بها عند هذه الدرجة الحرارة للاستخدام لاحقًا. لتصنيع أغشية PA NF عبر طريقة IP التقليدية، تم أولاً ملامسة السطح العلوي لغشاء PES للترشيح الفائق مع محلول PIP المائي لمدة 60 ثانية. ثم تم مسح أي قطرات مائية مرئية على سطح الغشاء برفق باستخدام ورقة نظيفة. بعد ذلك، تم صب محلول TMC في الهكسان الذي يحتوي على DDP على سطح PES وتم الاحتفاظ به أفقيًا لمدة 30 ثانية. ثم تم غسل الغشاء بمحلول الهكسان النقي لمدة 30 ثانية لإزالة أي TMC غير المتفاعل. تم معالجة الأغشية الناتجة في فرن عند لمدة 30 دقيقة، وتم غسل الأغشية أخيرًا بالماء النقي وتخزينها في الماء عند للاستخدام المستقبلي. عملية إنشاء أغشية PA NF عبر طريقة IP التقليدية مشابهة لتصنيع الأغشية عبر OSARIP ولكن دون إضافة DDP في الهكسان الذي يحتوي على TMC.

اختبار أداء الغشاء

في هذه الدراسة، استخدمنا نظام ترشيح عبر التدفق يحتوي على ثلاث خلايا دائرية لتقييم أداء فصل أغشية PA NF. قبل الاختبار، تم ضغط الغشاء باستخدام ماء نقي عند 6 بار لمدة ساعة واحدة. بعد ذلك، تم استخدام محاليل ملحية مختلفة بتركيز 1000 جزء في المليون كغذاء للمرور عبر الغشاء عند 4 بار بمعدل تدفق عرضي من تم الحفاظ على درجة حرارة محلول التغذية عند لحساب نفاذية أغشية PA NF، استخدمنا المعادلة (2).
أين هو حجم النفاذ الذي تم جمعه خلال فترة (ح). هو المساحة الفعالة للغشاء، والتي هي . (بار) هو الضغط المطبق. رفض الملح، (%)، من الأغشية يتم حسابه وفقًا للمعادلة (3):
أين و يستخدم للدلالة على تركيزات المحلول المتخلل والمغذي، على التوالي، والتي يتم تحديدها بواسطة الموصلية المقاسة بواسطة مقياس الموصلية (FE30K، ميتلر توليدو).
لتقييم أداء غشاء PA NF لفصل من المحلول الملحي، خلطات ملح ثنائية و LiCl بتركيزات مختلفة و متنوعة تم استخدام نسب الكتلة كحلول تغذية. حجم محلول التغذية هو 2000 مل. خلال التجربة، الضغط المطبق هو 4 بار، ومعدل التدفق المتقاطع هو تم إجراء التجربة تحت نموذج إعادة تدوير مغلق، حيث تم إعادة كل من الراشح والمخترق بشكل مستمر إلى خزان التغذية. قبل التجربة، تم استقرار عملية الترشيح لأكثر من 30 دقيقة. بعد ذلك، تم جمع 2 مل من المخترق لمزيد من التوصيف. تركيزات و تم تحديد تركيزات المحلول المنفذ ومحلول التغذية بواسطة مطيافية البلازما المقترنة بالتحريض (ICP، Agilent 5100، Bruke). تم وضع بروتوكول لقياس التركيز. و في الحل كما يلي:
إنشاء منحنيات المعايرة: منحنيات المعايرة لـ و تُنشأ باستخدام عينات قياسية بتركيزات تتراوح من 0.1 إلى 20 جزء في المليون لـ و 0.1 إلى 10 جزء في المليون لـ .
تحليل الاختراق: بالنظر إلى الفارق الكبير في و تُقاس التركيزات في المحلول بشكل منفصل. يتم عادةً تحليل محلول النفاذ مباشرةً دون تخفيف. إذا كان التركيز المكتشف يقع ضمن نطاق منحنى المعايرة، يتم قبول البيانات. خلاف ذلك، يتم تخفيف المحلول وإعادة قياسه.
تحليل العلف: يتم دائمًا تخفيف محلول العلف 10 مرات قبل التحليل. ثم يتم قياس المحلول المخفف وفقًا لنفس البروتوكول المستخدم في قياس النفاذ.
الرفض و تم تحديدها بواسطة المعادلة (3). الانتقائية يمكن حسابه من خلال رفض و معدلات النفاذية لـ و أو تركيزات و الأيونات في كل من التغذية ( ) و تتخلل ( ) تدفقات وفقًا للمعادلة التكميلية (S3):
أين هو تدفق المياه.

توفر البيانات

جميع البيانات التي تم توليدها في هذه الدراسة متاحة في المقالة، المعلومات التكميلية وملف بيانات المصدر. جميع البيانات متاحة من المؤلف المقابل عند الطلب. يتم توفير بيانات المصدر مع هذه الورقة.

References

  1. Grosjean, C., Miranda, P. H., Perrin, M. & Poggi, P. Assessment of world lithium resources and consequences of their geographic distribution on the expected development of the electric vehicle industry. Renew. Sustain. Energy Rev. 16, 1735-1744 (2012).
  2. Xu, S. et al. Extraction of lithium from Chinese salt-lake brines by membranes: Design and practice. J. Membr. Sci. 635, 119441 (2021).
  3. . et al. Materials for lithium recovery from salt lake brine. J. Mater. Sci. 56, 16-63 (2020).
  4. Razmjou, A., Asadnia, M., Hosseini, E., Habibnejad Korayem, A. & Chen, V. Design principles of ion selective nanostructured membranes for the extraction of lithium ions. Nat. Commun. 10, 5793 (2019).
  5. Song, J. F., Nghiem, L. D., Li, X.-M. & He, T. Lithium extraction from Chinese salt-lake brines: opportunities, challenges, and future outlook. Environ. Sci. Water Res. Technol. 3, 593-597 (2017).
  6. Nightingale, E. Jr Phenomenological theory of ion solvation. Effective radii of hydrated ions. J. Phys. Chem. 63, 1381-1387 (1959).
  7. Tansel, B. Significance of thermodynamic and physical characteristics on permeation of ions during membrane separation: Hydrated radius, hydration free energy and viscous effects. Sep. Purif. Technol. 86, 119-126 (2012).
  8. Li, X. et al. Membrane-based technologies for lithium recovery from water lithium resources: A review. J. Membr. Sci. 591, 117317 (2019).
  9. Wang, R., He, R., He, T., Elimelech, M. & Lin, S. Performance metrics for nanofiltration-based selective separation for resource extraction and recovery. Nat. Water 1, 291-300 (2023).
  10. Li, Y., Zhao, Y., Wang, H. & Wang, M. The application of nanofiltration membrane for recovering lithium from salt lake brine. Desalination 468, 114081 (2019).
  11. Villalobos, L. F., Zhang, J. & Elimelech, M. Nanofiltration for circularity: Fit-for-purpose design and evaluation. One Earth 6, 767-771 (2023).
  12. Zhang, H., He, Q., Luo, J., Wan, Y. & Darling, S. B. Sharpening Nanofiltration: Strategies for Enhanced Membrane Selectivity. ACS Appl. Mater. Interfaces 12, 39948-39966 (2020).
  13. Tu, Y.-M. et al. Prospective applications of nanometer-scale pore size biomimetic and bioinspired membranes. J. Membr. Sci. 620, 118968 (2021).
  14. Zhao, Y. et al. Differentiating Solutes with Precise Nanofiltration for Next Generation Environmental Separations: A Review. Environ. Sci. Technol. 55, 1359-1376 (2021).
  15. Chen, L. et al. Ion sieving in graphene oxide membranes via cationic control of interlayer spacing. Nature 550, 380-383 (2017).
  16. Shen, J. et al. Fast water transport and molecular sieving through ultrathin ordered conjugated-polymer-framework membranes. Nat. Mater. 21, 1183-1190 (2022).
  17. Sarkar, P., Modak, S. & Karan, S. Ultraselective and Highly Permeable Polyamide Nanofilms for Ionic and Molecular Nanofiltration. Adv. Funct. Mater. 31, 2007054 (2020).
  18. Peng, H. & Zhao, Q. A nano-heterogeneous membrane for efficient separation of lithium from high magnesium/lithium ratio brine. Adv. Funct. Mater. 31, 2009430 (2021).
  19. Yang, Z. et al. Dual-skin layer nanofiltration membranes for highly selective separation. J. Membr. Sci. 620, 118862 (2021).
  20. Peng, H. et al. Quaternization-spiro design of chlorine-resistant and high-permeance lithium separation membranes. Nat. Commun. 14, 5483 (2023).
  21. Lu, X. & Elimelech, M. Fabrication of desalination membranes by interfacial polymerization: history, current efforts, and future directions. Chem. Soc. Rev. 50, 6290-6307 (2021).
  22. Wang, K. et al. Tailored design of nanofiltration membranes for water treatment based on synthesis-property-performance relationships. Chem. Soc. Rev. 51, 672-719 (2022).
  23. Tan, Z., Chen, S., Peng, X., Zhang, L. & Gao, C. Polyamide membranes with nanoscale Turing structures for water purification. Science 360, 518-521 (2018).
  24. Jimenez-Solomon, M. F., Song, Q., Jelfs, K. E., Munoz-Ibanez, M. & Livingston, A. G. Polymer nanofilms with enhanced microporosity by interfacial polymerization. Nat. Mater. 15, 760-767 (2016).
  25. Liu, C. et al. Interfacial Polymerization at the Alkane/Ionic Liquid. Interface Angew. Chem. Int. Ed. 60, 14636-14643 (2021).
  26. Zhang, L. et al. Polyamide nanofiltration membrane with high mono/divalent salt selectivity via pre-diffusion interfacial polymerization. J. Membr. Sci. 636, 119478 (2021).
  27. Raaijmakers, M. J. T. & Benes, N. E. Current trends in interfacial polymerization chemistry. Prog. Polym. Sci. 63, 86-142 (2016).
  28. Muscatello, J., Müller, E. A., Mostofi, A. A. & Sutton, A. P. Multiscale molecular simulations of the formation and structure of polyamide
    membranes created by interfacial polymerization. J. Membr. Sci. 527, 180-190 (2017).
  29. Freger, V. & Ramon, G. Z. Polyamide desalination membranes: Formation, structure, and properties. Prog. Polym. Sci. 122, 101451 (2021).
  30. Freger, V. Kinetics of film formation by interfacial polycondensation. Langmuir 21, 1884-1894 (2005).
  31. Liang, Y. et al. Polyamide nanofiltration membrane with highly uniform sub-nanometre pores for sub-1 A precision separation. Nat. Commun. 11, 2015 (2020).
  32. Zhang, R. et al. Polyamide Nanofiltration Membranes from Sur-factant-Assembly Regulated Interfacial Polymerization: The Effect of Alkyl Chain. Macromol. Chem. Phys. 222, 2100222 (2021).
  33. Lu, C. et al. In situ characterization of dehydration during ion transport in polymeric nanochannels. J. Am. Chem. Soc. 143, 14242-14252 (2021).
  34. Pavluchkov, V., Shefer, I., Peer-Haim, O., Blotevogel, J. & Epsztein, R. Indications of ion dehydration in diffusion-only and pressuredriven nanofiltration. J. Membr. Sci. 648, 120358 (2022).
  35. Xu, T. et al. Highly Ion-Permselective Porous Organic Cage Membranes with Hierarchical Channels. J. Am. Chem. Soc. 144, 10220-10229 (2022).
  36. Shelley, J. C. & Shelley, M. Y. Computer simulation of surfactant solutions. Curr. Opin. Colloid Interface Sci. 5, 101-110 (2000).
  37. Liu, L. et al. A nanofiltration membrane with outstanding antifouling ability: Exploring the structure-property-performance relationship. J. Membr. Sci. 668, 121205 (2023).
  38. Yaroshchuk, A. & Bruening, M. L. An analytical solution of the solution-diffusion-electromigration equations reproduces trends in ion rejections during nanofiltration of mixed electrolytes. J. Membr. Sci. 523, 361-372 (2017).
  39. Yaroshchuk, A., Bruening, M. L. & Licón Bernal, E. E. Solution-Diffusion-Electro-Migration model and its uses for analysis of nanofiltration, pressure-retarded osmosis and forward osmosis in multi-ionic solutions. J. Membr. Sci. 447, 463-476 (2013).
  40. Epsztein, R., DuChanois, R. M., Ritt, C. L., Noy, A. & Elimelech, M. Towards single-species selectivity of membranes with subnanometre pores. Nat. Nanotechnol. 15, 426-436 (2020).
  41. . et al. Positive charged PEI-TMC composite nanofiltration membrane for separation of and from brine with high ratio. Desalination 449, 57-68 (2019).
  42. Xu, P., Hong, J., Xu, Z., Xia, H. & Ni, Q.-Q. MWCNTs-COOK-assisted high positively charged composite membrane: Accelerating enrichment and removal. Compos. Part B Eng. 212, 108686 (2021).

الشكر والتقدير

تم دعم هذا العمل من قبل المؤسسة الوطنية للعلوم الطبيعية في الصين (21988102، ج.ج.)، البرنامج الوطني الرئيسي للبحث والتطوير في الصين (2022YFB3805900، ج.ج.)، وخطة البحث والتطوير الرئيسية في مقاطعة جيانغسو (BE2O22056، ج.ج.).

مساهمات المؤلفين

صمم ي.ز. و ج.ج. التجارب؛ قام ق.ب. و ي.ز. بإجراء التجارب بما في ذلك إعداد الغشاء، اختبار الأداء وتنظيم البيانات؛ قام ر.و. بإجراء التحليل النظري؛ قام ز.ز.، ي.ل.، د.د، ز.و. و ي.هـ. بإجراء التجارب التكميلية؛ ساهم ي.ز.، ر.و.، س.ل. و ج.ج. في كتابة المخطوطة. راجع ل.ج. الورقة وقدم اقتراحات معدلة. ناقش جميع المؤلفين النتائج.

المصالح المتنافسة

يعلن المؤلفون عدم وجود مصالح متنافسة.

معلومات إضافية

المعلومات التكميلية النسخة على الإنترنت تحتوي على
المواد التكميلية المتاحة على
https://doi.org/10.1038/s41467-024-46887-4.
يجب توجيه المراسلات والطلبات للحصول على المواد إلى يوزهانغ تشو أو جيان جين.
معلومات مراجعة الأقران تشكر مجلة Nature Communications قيلوي سونغ والمراجعين الآخرين المجهولين على مساهمتهم في مراجعة الأقران لهذا العمل. يتوفر ملف مراجعة الأقران.
معلومات إعادة الطبع والتصاريح متاحة على
http://www.nature.com/reprints
ملاحظة الناشر تظل Springer Nature محايدة فيما يتعلق بالمطالبات القضائية في الخرائط المنشورة والانتماءات المؤسسية.
الوصول المفتوح هذه المقالة مرخصة بموجب رخصة المشاع الإبداعي للاستخدام والمشاركة والتكيف والتوزيع وإعادة الإنتاج في أي وسيلة أو تنسيق، طالما أنك تعطي الائتمان المناسب للمؤلفين الأصليين والمصدر، وتوفر رابطًا لرخصة المشاع الإبداعي، وتوضح ما إذا كانت هناك تغييرات قد تم إجراؤها. الصور أو المواد الأخرى من طرف ثالث في هذه المقالة مشمولة في رخصة المشاع الإبداعي للمقالة، ما لم يُشار إلى خلاف ذلك في سطر الائتمان للمادة. إذا لم تكن المادة مشمولة في رخصة المشاع الإبداعي للمقالة واستخدامك المقصود غير مسموح به بموجب اللوائح القانونية أو يتجاوز الاستخدام المسموح به، ستحتاج إلى الحصول على إذن مباشرة من صاحب حقوق الطبع والنشر. لعرض نسخة من هذه الرخصة، قم بزيارةhttp://creativecommons.org/licenses/by/4.0/.
© المؤلفون 2024

  1. (ل) تحقق من التحديثات
  2. كلية الكيمياء، الهندسة الكيميائية وعلوم المواد، مركز الابتكار لعلوم الكيمياء ومختبر جيانغسو الرئيسي لتصميم وتطبيق البوليمرات الوظيفية المتقدمة، جامعة سوتشو، سوتشو 215123، جمهورية الصين الشعبية. i-Lab، معهد سوتشو للتكنولوجيا النانوية والبيونيات النانوية، الأكاديمية الصينية للعلوم، سوتشو 215123، جمهورية الصين الشعبية. قسم الهندسة المدنية والبيئية، جامعة فاندربيلت، ناشفيل، TN 37235، الولايات المتحدة الأمريكية. مختبر CAS الرئيسي للمواد المستوحاة من الطبيعة وعلوم الواجهات، المعهد الفني للفيزياء والكيمياء، الأكاديمية الصينية للعلوم، 100190 بكين، جمهورية الصين الشعبية. ساهم هؤلاء المؤلفون بالتساوي: قوان بينغ، روييو وانغ.

Journal: Nature Communications, Volume: 15, Issue: 1
DOI: https://doi.org/10.1038/s41467-024-46887-4
PMID: https://pubmed.ncbi.nlm.nih.gov/38509082
Publication Date: 2024-03-20

Extreme Li-Mg selectivity via precise ion size differentiation of polyamide membrane

Received: 18 October 2023
Accepted: 13 March 2024
Published online: 20 March 2024

Quan Peng , Ruoyu Wang , Zilin Zhao , Shihong Lin , Ying Liu , Dianyu Dong , Zheng Wang , Yiman , Yuzhang Zhu , Jian Jin Lei Jiang (1)

Abstract

Achieving high selectivity of and is of paramount importance for effective lithium extraction from brines, and nanofiltration (NF) membrane plays a critical role in this process. The key to achieving high selectivity lies in the on-demand design of NF membrane pores in accordance with the size difference between and ions, but this poses a huge challenge for traditional NF membranes and difficult to be realized. In this work, we report the fabrication of polyamide (PA) NF membranes with ultra-high selectivity by modifying the interfacial polymerization (IP) process between piperazine (PIP) and trimesoyl chloride (TMC) with an oil-soluble surfactant that forms a monolayer at oil/water interface, referred to as OSARIP. The OSARIP benefits to regulate the membrane pores so that all of them are smaller than ions. Under the solely size sieving effect, an exceptional rejection rate of over is achieved. This results in an exceptionally high selectivity, which is one to two orders of magnitude higher than all the currently reported pressure-driven membranes, and even higher than the microporous framework materials, including COFs, MOFs, and POPs. The large enhancement of ion separation performance of NF membranes may innovate the current lithium extraction process and greatly improve the lithium extraction efficiency.

Salt lakes hold minable lithium reserves of the world . Exploring the efficient access to lithium resources in salt lakes is crucial for providing stable lithium supply to meet the rapidly growing demand of the lithium-ion battery industry. As for extracting lithium from brine, one of the key processes is to separate and ions since the other ions with higher abundance, such as , and , will be pre-removed in the form of precipitation through solar-driven evaporation . Efficient separation still faces significant challenges because the content of in brine is much higher than that of (for example, the content of is usually more than 20 times of
in salt lake brines in China ), and the sizes of and ions are very similar .
Pressure-driven nanofiltration (NF) technology based on membrane separation is one of the most concerned technologies among the commonly used treatment trains for extracting lithium from brine . However, the current NF membranes exhibit relatively low separation selectivity, usually less than 20. Therefore, in the process of lithium extraction, multiple repeated membrane separation processes have to be used to obtain sufficient purity of in the filtrate to generate or LiOH precipitation as the target product. The
limited selectivity of NF membranes is mainly caused by their uneven pore size and wide distribution separation is usually realized by co-effect of size sieving and Donnan exclusion . Increasing the charge positivity of the membrane to enhance Donnan exclusion has been widely used to enhance the repulsion to and thereby improve separation selectivity . However, strong Donnan exclusion can also increase the repulsion to ions, making it difficult to significantly increase the selectivity and simultaneously reducing the recovery rate of . Undoubtedly, if we can regulate the pores of the NF membrane to achieve separation solely relying on pore size sieving, it may break the limitation of charge repulsion, thus achieving a disruptive improvement of separation performance.
The preparation of the state-of-the-art NF membranes is based on interfacial polymerization (IP) reaction between amine monomer dissolved in water and acyl chloride monomer dissolved in organic solvent to form ultrathin polyamide (PA) selective layer . During this process, the diffusion of monomers towards the interface (mainly amine monomers from aqueous phase to organic solvent) is slow and random , while the reaction rate between amine and acyl chloride is fast and the reaction is violent . This leads to an uneven structure in the PA selective layer, resulting in a wide pore size distribution that is difficult to control . To achieve precise regulation of the membrane pores and highly selective separation of target ions, an effective method is needed to control the IP process to improve the regularity of the reaction. Unfortunately, research in this area is still immature and there are few relevant reports. We previously reported a strategy of using water-soluble surfactants to form ordered monolayers at the water/oil interface to regulate the IP process (referred to as SARIP) to improve the pore size uniformity of PA NF membranes . The resulting PA NF membranes with greatly narrowed pore size distribution exhibited a rejection rate of over for all divalent cations. Although the SARIP strategy has shown its effective in regulating membrane pores, it is still difficult to meet the requirements for high selectivity separation of and . The pore size range of the PA NF membrane prepared from SARIP is (including all detectable pores). The hydration diameter of ion is , while the Stokes diameter of ion is (considering that ion is prone to dehydration during pressure-driven separation due to its low hydration energy ). Obviously, ions still can penetrate the membrane through partial pore, thereby creating insufficient selectivity (less than 50).
In the SARIP process, in addition to the surfactant monolayer at the water/oil interface, there are also a large number of free surfactant molecules and their aggregates such as micelles in the aqueous phase. The presence of these free surfactant molecules and aggregates will partially weaken the regulatory effect of surfactant monolayers on the interfacial diffusion of amine monomers, limiting the improvement of membrane pore uniformity . In this work, we report the use of an oilsoluble surfactant (i.e. dodecyl phosphate (DDP)) introduced into the organic solvent phase to regulate IP reaction, referred to as OSARIP. The amphiphilic nature of DDP molecules enables them to form assembled monolayers at the water/oil interface. Due to the absence of free surfactant molecules in the aqueous solution, the DPP monolayer at the water/oil interface can fully exert its effect in facilitating amine monomer diffusion and regulating IP reaction. As expected, the OSARIP process makes the pores of the PA NF membranes more uniform, and the pore size distribution further narrowed to be in the range of . This means that almost all pores are smaller than the hydration diameter of ion, and most pores are larger than the Stokes diameter of ion (considering that ion is prone to dehydration during pressure-driven separation due to its low hydration energy ). Therefore, relying solely on the size sieving effect, the PA NF membrane exhibits ultra-high rejection of up to and ultra-high selectivity of up to 4147 . The ultra-high separation
selectivity of makes it possible to obtain high-purity lithium filtrate through a single membrane filtration process. This may transform the existing technology flow of extracting lithium from brine, significantly simplifying the process and improving separation efficiency.

Results

To regulate the IP reaction effectively, an oil-soluble surfactant, DDP, was introduced into the hexane solution. The amphiphilic nature of DDP allows for the formation of a monolayer at oil/hexane interface. The lowest interfacial tension was observed at a DDP concentration of (see Supplementary Fig. 1), indicating the formation of the densest molecular layer. The presence of DDP monolayer can adsorb the PIP molecules to the water/hexane interface via electrostatic interaction between the phosphonic group in DDP and the amine group in PIP (Fig. 1a). This will cause an increase in PIP concentration near the oil/hexane interface, reaching its highest point at a DDP concentration of , as evidenced by UV-vis absorption spectra. (Fig. 1b, and Supplementary Fig. 2). Further in-situ monitoring of PIP absorbance near the oil/hexane interface during the IP reaction reveals a faster decrease of PIP absorbance when there is a DDP monolayer at interface (Fig. 1b). This indicates that the DDP monolayer is helpful in promoting the trans-interface diffusion of PIP monomers. In addition to the accumulation effect of DDP monolayer, free DDP molecules in hexane act as transport carriers through the dynamic exchange process with the monolayer molecules to assist in the trans-interface diffusion of PIP molecules . Moreover, the DDP assembly significantly reduces the oil/water interface tension, which is beneficial for reducing the hindrance of PIP diffusion and enhancing the spatial uniformity of the diffusion process. It is important to note that the OSARIP process eliminates the presence of any free surfactant molecules or micelles in the aqueous solution, avoiding any competition with the surfactant assembly at interface and creating an ideal condition for unleashing the full potential of the surfactant assembly in controlling PIP transinterface diffusion.
In the traditional IP process, the trans-interface diffusion of PIP is slow and random, making it difficult for the supply of PIP to keep up with its consumption. Therefore, the IP reaction exhibits spatial and temporal discontinuity, which ultimately leads to the chemical and structural heterogeneity of the PA membrane. In the OSARIP process, the diffusion of PIP is fast and spatially uniform, which greatly reduces the significant difference between monomer diffusion rate and IP reaction rate, thereby enhancing the continuity of PA layer formation in time and space. Compared to the PA selective layer prepared from traditional IP, the cross-linking degree of the PA selective layer prepared from OSARIP is significantly increased (from to ) based on the elemental composition analysis of X-ray photoelectron spectroscopy (XPS) (Fig. 1c, d, and Supplementary Figs. 3-6), and the membrane shows less negatively charged due to less residual unreacted groups (see Supplementary Fig. 7).
Figure 1d-g displays the topological structure of the PA NF membrane prepared from both traditional IP and OSARIP. In comparison, the PA NF membrane prepared from OSARIP displays small convex spots and a thinner PA selective layer ( for the membrane prepared from traditional IP and for the membrane prepared from OSARIP). We further probed the pore size information of the membranes by conducting a standard analysis of the rejections of neutral molecules with different molecular weights. As the concentration of DDP in hexane increases, the molecular weight cut-off (MWCO) of the membrane gradually decreases and reaches a minimum value of 134 Da at a critical concentration of (Fig. 2a). To our knowledge, such a small MWCO has not yet been achieved in the PA NF membrane prepared by the IP reaction of PIP and TMC reported so far. Further increasing the DDP concentration to actually increases the MWCO of the membrane to 185 Da .
Fig. 1 | Fabrication, chemical composition, and topological structure of PA NF membranes. a Scheme for illustrating the effect of DDP assembly at the oil/water interface on PIP trans-interface diffusion to enhance the homogeneity of the IP reaction. Variation of PIP absorbance near the interface with respect to the reaction time monitored by UV-vis measurement. c Chemical elemental
composition and crosslinking degree of PA NF membranes prepared from traditional IP and OSARIP. d-g Surface and cross-sectional structure of PA membranes prepared from traditional IP and OSARIP ( Top-view SEM images. f, g Crosssectional TEM images).
Following the log-normal distribution assumption, the pore size distribution of the membranes was fitted from the rejection curve of neutral molecules and expressed as a probability density function (PDF) (see Section 1.4.2 of Supplementary Information). In keeping with the trend of MWCO change, the introduction of DDP greatly narrows the pore size distribution of PA NF membranes and results in a narrowest pore size distribution with a shape parameter of 1.13 (defined as the geometric standard deviation of the PDF curve) at a DDP concentration of (Fig. 2b). Notably, the pore size falls in the range of the PA NF membrane prepared from the traditional IP, which indicates that the DDP monolayer primarily improves the uniformity of pore size, rather than reducing the pore size. These highly uniform pores enable the membrane with an ultra-high rejection towards all divalent salts including anionic salt of and cationic salt of (all over 99%) (Fig. 2c, Supplementary Fig. 8 and Supplementary Table 1), though the membrane is negatively charged. Interestingly, the pore structure and salt rejection of the PA NF membrane prepared from OSARIP at a DDP concentration of is little affected by the reaction time (see Supplementary Figs. 9 and 10),
which further proves the facilitating effect of DDP assembly on the IP kinetics.
Upon the analysis of existing NF membranes in the literatures (Supplementary Table 2), it is clear that NF membranes with small pore sizes are highly effective at rejecting without relying much on Donnan exclusion (Fig. 2d). This discovery underscores the immense potential of employing such membranes for applications that demand efficient removal of divalent cations by size sieving (e.g., separation for Li extraction from brines). Compared with the existing NF membranes, our PA NF membrane possesses the smallest effective pore size and exceptionally narrow pore size distribution, thereby exhibiting the highest rejection to . It also exhibits very high rejection to other divalent cations, such as for for for , and for (Supplementary Fig. 11 and Supplementary Table 3). Meanwhile, due to the dehydration process, the steric hinderance of the membrane pores cannot obstacle the penetration of monovalent cations such as , leading to a relatively low rejection ( ) to it. Consequently, a very sharp transition region is presented by the membrane when correlating the ion
Fig. 2 | Pore size and single salt rejection of PA NF membranes prepared from OSARIP. a Rejection of neutral molecules by the membranes prepared with different concentration of DDP in hexane. Pore size distribution of the corresponding membranes fitting from the rejection curve of neutral molecules following log-normal distribution principle. c Salt rejections of the membrane prepared from traditional IP and OSARIP. d Summary of rejection by the PA membrane prepared from OSARIP at optimal condition (red pentagon) and the NF
membranes reported by others (square, see detail description in Supplementary Table 2). e Comparison of pore size range of PA NF membranes prepared from traditional IP, SARIP, OSARIP, and the commercially available NF membranes (DK produced by Suez and NF270 produced by Dupont). All membranes are negatively charged based on the results of streaming potential measurement. Error bar represents the standard deviation of three replicate measurements.
rejections with their stokes radius (Supplementary Fig. 11), suggesting its superior ion-ion selectivity between the monovalent and divalent cations.
In order to emphasize the impressive capability of DDP assembly in refining pore size distribution, we compared the pore size ranges of several representative types of PA NF membranes including the NF membranes obtained from OSARIP, SARIP, and traditional IP processes, as well as commercially available DK and NF270 membranes (the pore size ranges derived from the PDF curves as reported by others ) (Fig. 2e and Supplementary Fig. 12). The pore size range of these membranes is directly taken from their PDF curves, where all pores with probability density greater than 0.01 are counted for statistics. It clearly shows that the PA NF membrane prepared from traditional IP process has a wide pore size range of , followed by NF270 and DK membranes with pore size ranges of and , respectively. In contrast, the introduction of surfactants at the oil/water interface has shown a significant effect on sharpening the pore size range of membranes. Our previous work demonstrated the efficacy of utilizing a water-soluble surfactant monolayer to regulate the IP reaction (e.g. SARIP), resulting in a PA NF membrane with a pore size range of . Here, the newly developed OSARIP process further narrows the membrane pore size range to be . Considering that the hydration diameter of ion is , it means that almost all the pores of the PA NF membrane prepared from OSARIP process are smaller than the hydration diameter of ions, therefore resulting in an ultra-high rejection to ions.
The separation performance of the PA NF membranes prepared from OSARIP process was further investigated by using a binary salt mixture composed of and LiCl with the total salt concentration of 2000 ppm and the mass ratio of as simulated brine. It shows a rejection of up to (Fig.3a), while a negative rejection of to . The corresponding / selectivity is as high as 828 . In comparison, the PA NF membrane prepared from traditional IP shows a rejection of to and to , respectively, corresponding to a selectivity of 16 .
The composition of ions in actual salt lake brine is diverse, so we further studied the separation performance of PA NF membrane under different ratios. When the total salt concentration remains unchanged at 2000 ppm and the mass ratio in the binary salt mixture increases from 10:1 to 60:1, the rejection to slightly increases from to and the rejection to decreases from to (Fig. 3a and Supplementary Table 4). This corresponds to the increase of selectivity from 548 to 4147. When maintaining a mass ratio at 20:1 and increasing the total salt concentration from 2000 to 5000 ppm , the rejection of decreases slightly from to , while the rejection of undergoes a significant change, decreasing from to correspondingly (Fig. 3b and Supplementary Table 5). This allows the PA NF membrane to still exhibit a high selectivity up to 185 in high salinity solution. In practical industrial applications of NF technology for lithium extraction from brine, feed solutions typically
Fig. separation performance of PA NF membranes by using a binary salt mixture of and as feed. a Variation of the rejections of and and selectivity with respect to the mass ratio (The salt concentration of feed solution is 2000 ppm ). Variation of the rejections of and and selectivity with respect to the salt concentration of feed solution
(The mass ratio is ). c Comparison of selectivity of the PA NF membrane prepared in this work and the membranes reported by others including polymeric NF membrane, MOFs membranes, COFs membranes, and POPs membranes (see detail description in Supplementary Tables 7 and 8). Error bar represents the standard deviation of three replicate measurements.
exhibit higher concentrations and contain additional metal ions, such as , and . Even under these conditions, the PA NF membrane demonstrates exceptional selectivity of more than 200 (Supplementary Figs. 13 and 14 and Supplementary Table 6). This remarkable performance underscores the unrivaled superiority of membranes with uniform pore sizes for lithium extraction from salt lake brine.
We compared the selectivity of PA NF membranes prepared from OSARIP with the most representative membranes reported so far, including pressure-driven NF membranes, microporous framework membranes such as covalent organic frameworks (COFs) membranes, metal-organic frameworks (MOFs) membranes, and porous organic polymers (POPs) membranes (Fig. 3c and Supplementary Tables 7 and 8). The selectivity of the PA NF membrane prepared from OSARIP is currently the highest value, which is one to two orders of magnitude higher than the pressure-driven NF membranes reported so far and even much higher than the emerging microporous framework membranes that are considered to have ideal uniform pore structures.
To further investigate the significant impact of pore structure on ion-ion selectivity of PA NF membranes, we prepared a series of membranes with the pore size range from wide to narrow by tailoring the DDP concentration in hexane within the scope of 0 to (Fig. 4a, Supplementary Fig. 15 and Supplementary Tables 9 and 10). With narrowing the pore size range, the mean pore size decreases simultaneously. We studied the variation of and rejections and
selectivity over the mean pore size of as-prepared PA NF membranes using the binary salt mixture of and LiCl with a total salt concentration of 2000 ppm and mass ratio of 20:1 as feed. The rejection gradually increases along with the decrease of the mean pore size and reaches at last (Fig. 4b). However, the rejection exhibits an opposite trend. A more negative rejection is presented by the membrane with a smaller mean pore size. As the rejection of the two ions changes in opposite direction, the / selectivity increases exponentially as the mean pore size decreases.
In order to account for the significant increase of selectivity by narrowing the pore size range, as well as the increasingly negative rejection of , we utilized the solution-diffusionelectromigration (SDEM) model to describe the trans-membrane ion transport in a binary salt mixture. The SDEM model has been extensively used to explain the ion transport behaviors through reverse osmosis (RO)/NF/forward osmosis (FO) membranes . By introducing the concept of virtual solution and ignoring the convective coupling between solvent and solute transport across the membrane, the ion transport is mainly driven by the gradients of concentration and electrostatic potential. The ion flux for species , in the SDEM model can be described using the modified Nernst-Planck Eq. (1):
Fig. 4 | Effect of pores of PA membranes on the separation performance for differentiating and . a Pore size range of PA NF membranes prepared with the conditions as listed in Supplementary Table 8 in detail. Variation of water permeance, rejection, rejection, and selectivity over the mean
pore size of as-prepared PA membranes. c Variation of ions permeability as function of mean pore size. Schematic illustration of spontaneously-arising electric field during the separation process while using a binary salt mixture of and LiCl as feed.
where is the ion permeability, is the ion concentration in a virtual solution that is in thermodynamic equilibrium with the membrane phase, is the trans-membrane coordinate normalized by the membrane thickness, is the valence of species , and is the dimensionless electrical potential in the virtual solution. Thus, the selective transport of and across the membrane is characterized by the difference of ion permeability in the SDEM model. Ion permeability, , can be fitted from the experimental rejection results, given the feed composition and the measured permeate flux. As shown in Fig.4c, the permeability can maintain at near with the decrease of the mean pore size due to the contribution of dehydration effect. In comparison, the permeability decreases over three orders of magnitude. The significant difference between ion permeability of and results in the ultra-high selectivity.
The change of rejection towards a more negative direction with the decrease of mean pore size can be attributed to the coupling effect between the trans-membrane ion transport of and via a spontaneously-arising electric field, as presented schematically in Fig. 4d. Specifically, is the dominant salt in the simulated brine and the concentration of is mainly determined by the concentration. Intercepting ions can lead to the accumulation of ions near the membrane surface. As rejection increases, the concentration gradient of across the membrane increases, resulting in the faster transport of . Accompanying the transport of , the transport of the counter-ion with equivalent charge is necessary to maintain charge neutrality in the permeate, thus establishing the spontaneous potential gradient. Considering the larger steric hindrance of for partition into the pores of the membrane, accelerating the transport is the only choice.
It is important to note that the transfer of ions through subnanometer pores is a highly intricate process influenced by various
factors, including pore size, charge property, and chemical composition, as well as the coupling with water and other ions. In the case of binary salt solution comprising and LiCl , the mass transfer behavior of and differs significantly. For the mass transfer of , the highly uniform pores of the PA NF membrane can offer high and stable rejection of through a purely size-based separation. As for the mass transfer of , the dehydration effect makes the pores of the PA NF membrane have limited effect to hinder the partition and transport of through the membrane, while the charge property and chemical structure of the membranes play an important role, as well as the effect of coupling with the counter-ion of . Negatively charged PA NF membranes are beneficial for obtaining high permeance due to the traction of . When the PA NF membranes are highly positively charged, the electrostatic attraction will slow down the migration of . Meanwhile, the Donnan exclusion induced by the positive charge also produces strong hinderance for the partition of into the membrane pores. As a result, a higher rejection of is presented by the membrane, which is undesired for the target of recovering . Firm evidence to prove this is that the PA NF membrane prepared from OSARIP by using polyethylene imine (PEI) and TMC as monomers, which are always considered to have strongly positive charge , displayed a much higher rejection as compared with the PA membrane prepared from OSARIP using PIP and TMC as monomers when using the same simulated brine as feed (Supplementary Fig. 16 and Supplementary Tables 11, 12). These results also indicate the superiority of negatively charged PA NF membranes with highly uniform pore size distribution for extracting from brine.

Discussion

In summary, we developed a simple and effective method, referred to as OSARIP, to regulate the PIP trans-interface diffusion within the IP
reaction for the production of PA NF membranes with highly uniform pores. Through the utilization of the full promotion effect of oilsoluble surfactant monolayer at the oil/water interface on the PIP diffusion, the IP reaction between PIP and TMC results in a PA NF membrane with an ultra-narrow pore size distribution falling in the range between the hydration diameter of ion and Stokes diameter of ion. These highly uniform pores enable the membrane with an ultrahigh rejection of to and relatively low rejection to . The performance test, using a binary salt mixture of LiCl and to simulate the brine, revealed that the PA NF membrane prepared from OSARIP process display an ultrahigh selectivity over 4000. Further investigation into the ion transport behavior with the gradual decrease of effective pore size revealed that the steric hindrance has a limited effect on permeability but greatly reduces the permeability of , resulting in the exponential increase of selectivity. Overall, our work offers a facile and precise method for optimizing the pore size of PA NF membranes at angstrom scale, which can be easily operated in the current industrial process for preparing NF membranes with highly improved performance for wide applications in water treatment and chemical separation.

Methods

Fabrication of PA NF membranes

Initially, aqueous solution containing and hexane solutions containing TMC and a certain amount of dodecyl phosphate (DDP) were prepared, respectively. The two kinds of solutions were subsequently heated to and kept at this temperature for further use. To fabricate the PA NF membranes via traditional IP, the top surface of the PES ultrafiltration membrane was first contacted with PIP aqueous solution for 60 s . Any visible aqueous drops on the membrane surface were then gently wiped away using a clean sheet. The TMC hexane solution containing DDP was then poured onto the surface of the PES and held horizontally for 30 s . The membrane was subsequently washed with pure hexane solution for 30 s to remove any unreacted TMC. The resulting membranes were cured in an oven at for 30 min , and the membranes were finally washed with pure water and stored in water at for further use. The process of creating PA NF membranes via traditional IP is similar to the fabrication of membranes via OSARIP but without the addition of DDP in hexane containing TMC.

Membrane performance test

In this study, we utilized a cross-flow filtration system containing three circular cells to evaluate the separation performance of PA NF membranes. Before testing, the membrane was compacted using pure water at 6 bar for 1 h . Next, various salt solutions with a concentration of 1000 ppm were used as feed to pass through the membrane at 4 bar with a cross-flow rate of . The temperature of feed solution was maintained at . To calculate the permeance of the PA NF membranes, we employed Eq. (2).
where is the permeate volume collected within the period of (h). is the effective area of the membrane, which is . (bar) is the applied pressure. The salt rejection, (%), of the membranes is calculated according to Eq. (3):
where and is used to denote permeate and feed solution concentrations, respectively, which is determined by the conductivity measured by a conductivity meter (FE30K, Mettler Toledo).
For evaluating the performance of PA NF membrane for separating from brine, binary salt mixtures of and LiCl with various concentrations and various mass ratios were used as the feed solutions. The volume of the feed solution is 2000 ml . Within the experiment, the applied pressure is 4 bar, and the cross-flow rate is . The experiment is performed under a closed-loop recycling model, in which both retentate and permeate were continuously returned to the feed tank. Prior to the experiment, the filtration process was stabilized for over 30 minutes. Subsequently, 2 ml permeate were collected for the further characterization. The concentrations of and in the permeate and feed solutions were determined by inductively coupled plasma spectroscopy (ICP, Agilent 5100, Bruke). The protocol for measuring the concentration and in the solution is as follows:
Establishing calibration curves: Calibration curves for and are established using standard samples with concentrations ranging from 0.1 to 20 ppm for and 0.1 to 10 ppm for .
Permeate analysis: Considering the significant difference in and concentrations in the solution, they are quantified separately. The permeate solution is generally analyzed directly without dilution. If the detected concentration falls within the calibration curve range, the data is accepted. Otherwise, the solution is diluted and remeasured.
Feed analysis: The feed solution is always diluted 10 times before analysis. The diluted solution is then measured following the same protocol as the permeate.
The rejections of and were determined by Eq. (3). The selectivity, , could be calculated through the rejection of and , the permeation rates of and , or the concentrations of and ions in both the feed ( ) and permeate ( ) streams according to the Supplementary Eq. (S3):
where is the water flux.

Data availability

All data generated in this study are provided in the article, Supplementary Information and Source Data file. All data are available from the corresponding author upon request. Source data are provided with this paper.

References

  1. Grosjean, C., Miranda, P. H., Perrin, M. & Poggi, P. Assessment of world lithium resources and consequences of their geographic distribution on the expected development of the electric vehicle industry. Renew. Sustain. Energy Rev. 16, 1735-1744 (2012).
  2. Xu, S. et al. Extraction of lithium from Chinese salt-lake brines by membranes: Design and practice. J. Membr. Sci. 635, 119441 (2021).
  3. . et al. Materials for lithium recovery from salt lake brine. J. Mater. Sci. 56, 16-63 (2020).
  4. Razmjou, A., Asadnia, M., Hosseini, E., Habibnejad Korayem, A. & Chen, V. Design principles of ion selective nanostructured membranes for the extraction of lithium ions. Nat. Commun. 10, 5793 (2019).
  5. Song, J. F., Nghiem, L. D., Li, X.-M. & He, T. Lithium extraction from Chinese salt-lake brines: opportunities, challenges, and future outlook. Environ. Sci. Water Res. Technol. 3, 593-597 (2017).
  6. Nightingale, E. Jr Phenomenological theory of ion solvation. Effective radii of hydrated ions. J. Phys. Chem. 63, 1381-1387 (1959).
  7. Tansel, B. Significance of thermodynamic and physical characteristics on permeation of ions during membrane separation: Hydrated radius, hydration free energy and viscous effects. Sep. Purif. Technol. 86, 119-126 (2012).
  8. Li, X. et al. Membrane-based technologies for lithium recovery from water lithium resources: A review. J. Membr. Sci. 591, 117317 (2019).
  9. Wang, R., He, R., He, T., Elimelech, M. & Lin, S. Performance metrics for nanofiltration-based selective separation for resource extraction and recovery. Nat. Water 1, 291-300 (2023).
  10. Li, Y., Zhao, Y., Wang, H. & Wang, M. The application of nanofiltration membrane for recovering lithium from salt lake brine. Desalination 468, 114081 (2019).
  11. Villalobos, L. F., Zhang, J. & Elimelech, M. Nanofiltration for circularity: Fit-for-purpose design and evaluation. One Earth 6, 767-771 (2023).
  12. Zhang, H., He, Q., Luo, J., Wan, Y. & Darling, S. B. Sharpening Nanofiltration: Strategies for Enhanced Membrane Selectivity. ACS Appl. Mater. Interfaces 12, 39948-39966 (2020).
  13. Tu, Y.-M. et al. Prospective applications of nanometer-scale pore size biomimetic and bioinspired membranes. J. Membr. Sci. 620, 118968 (2021).
  14. Zhao, Y. et al. Differentiating Solutes with Precise Nanofiltration for Next Generation Environmental Separations: A Review. Environ. Sci. Technol. 55, 1359-1376 (2021).
  15. Chen, L. et al. Ion sieving in graphene oxide membranes via cationic control of interlayer spacing. Nature 550, 380-383 (2017).
  16. Shen, J. et al. Fast water transport and molecular sieving through ultrathin ordered conjugated-polymer-framework membranes. Nat. Mater. 21, 1183-1190 (2022).
  17. Sarkar, P., Modak, S. & Karan, S. Ultraselective and Highly Permeable Polyamide Nanofilms for Ionic and Molecular Nanofiltration. Adv. Funct. Mater. 31, 2007054 (2020).
  18. Peng, H. & Zhao, Q. A nano-heterogeneous membrane for efficient separation of lithium from high magnesium/lithium ratio brine. Adv. Funct. Mater. 31, 2009430 (2021).
  19. Yang, Z. et al. Dual-skin layer nanofiltration membranes for highly selective separation. J. Membr. Sci. 620, 118862 (2021).
  20. Peng, H. et al. Quaternization-spiro design of chlorine-resistant and high-permeance lithium separation membranes. Nat. Commun. 14, 5483 (2023).
  21. Lu, X. & Elimelech, M. Fabrication of desalination membranes by interfacial polymerization: history, current efforts, and future directions. Chem. Soc. Rev. 50, 6290-6307 (2021).
  22. Wang, K. et al. Tailored design of nanofiltration membranes for water treatment based on synthesis-property-performance relationships. Chem. Soc. Rev. 51, 672-719 (2022).
  23. Tan, Z., Chen, S., Peng, X., Zhang, L. & Gao, C. Polyamide membranes with nanoscale Turing structures for water purification. Science 360, 518-521 (2018).
  24. Jimenez-Solomon, M. F., Song, Q., Jelfs, K. E., Munoz-Ibanez, M. & Livingston, A. G. Polymer nanofilms with enhanced microporosity by interfacial polymerization. Nat. Mater. 15, 760-767 (2016).
  25. Liu, C. et al. Interfacial Polymerization at the Alkane/Ionic Liquid. Interface Angew. Chem. Int. Ed. 60, 14636-14643 (2021).
  26. Zhang, L. et al. Polyamide nanofiltration membrane with high mono/divalent salt selectivity via pre-diffusion interfacial polymerization. J. Membr. Sci. 636, 119478 (2021).
  27. Raaijmakers, M. J. T. & Benes, N. E. Current trends in interfacial polymerization chemistry. Prog. Polym. Sci. 63, 86-142 (2016).
  28. Muscatello, J., Müller, E. A., Mostofi, A. A. & Sutton, A. P. Multiscale molecular simulations of the formation and structure of polyamide
    membranes created by interfacial polymerization. J. Membr. Sci. 527, 180-190 (2017).
  29. Freger, V. & Ramon, G. Z. Polyamide desalination membranes: Formation, structure, and properties. Prog. Polym. Sci. 122, 101451 (2021).
  30. Freger, V. Kinetics of film formation by interfacial polycondensation. Langmuir 21, 1884-1894 (2005).
  31. Liang, Y. et al. Polyamide nanofiltration membrane with highly uniform sub-nanometre pores for sub-1 A precision separation. Nat. Commun. 11, 2015 (2020).
  32. Zhang, R. et al. Polyamide Nanofiltration Membranes from Sur-factant-Assembly Regulated Interfacial Polymerization: The Effect of Alkyl Chain. Macromol. Chem. Phys. 222, 2100222 (2021).
  33. Lu, C. et al. In situ characterization of dehydration during ion transport in polymeric nanochannels. J. Am. Chem. Soc. 143, 14242-14252 (2021).
  34. Pavluchkov, V., Shefer, I., Peer-Haim, O., Blotevogel, J. & Epsztein, R. Indications of ion dehydration in diffusion-only and pressuredriven nanofiltration. J. Membr. Sci. 648, 120358 (2022).
  35. Xu, T. et al. Highly Ion-Permselective Porous Organic Cage Membranes with Hierarchical Channels. J. Am. Chem. Soc. 144, 10220-10229 (2022).
  36. Shelley, J. C. & Shelley, M. Y. Computer simulation of surfactant solutions. Curr. Opin. Colloid Interface Sci. 5, 101-110 (2000).
  37. Liu, L. et al. A nanofiltration membrane with outstanding antifouling ability: Exploring the structure-property-performance relationship. J. Membr. Sci. 668, 121205 (2023).
  38. Yaroshchuk, A. & Bruening, M. L. An analytical solution of the solution-diffusion-electromigration equations reproduces trends in ion rejections during nanofiltration of mixed electrolytes. J. Membr. Sci. 523, 361-372 (2017).
  39. Yaroshchuk, A., Bruening, M. L. & Licón Bernal, E. E. Solution-Diffusion-Electro-Migration model and its uses for analysis of nanofiltration, pressure-retarded osmosis and forward osmosis in multi-ionic solutions. J. Membr. Sci. 447, 463-476 (2013).
  40. Epsztein, R., DuChanois, R. M., Ritt, C. L., Noy, A. & Elimelech, M. Towards single-species selectivity of membranes with subnanometre pores. Nat. Nanotechnol. 15, 426-436 (2020).
  41. . et al. Positive charged PEI-TMC composite nanofiltration membrane for separation of and from brine with high ratio. Desalination 449, 57-68 (2019).
  42. Xu, P., Hong, J., Xu, Z., Xia, H. & Ni, Q.-Q. MWCNTs-COOK-assisted high positively charged composite membrane: Accelerating enrichment and removal. Compos. Part B Eng. 212, 108686 (2021).

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21988102, J.J.), the National Key Research and Development Program of China (2022YFB3805900, J.J.), and the Key Research and Development Plan of Jiangsu Province (BE2O22056, J.J.).

Author contributions

Y.Z. and J.J. designed the experiments; Q.P. and Y.Z. performed the experiments including the membrane preparation, performance test and data organization; R.W. performed the theoretical analysis; Z.Z., Y.L., D.D, Z.W. and Y.H. performed the supplementary experiments; Y.Z., R.W., S.L. and J.J. contributed to writing the manuscript. L.J. reviewed the paper and provided revised suggestion. All authors discussed the results.

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/s41467-024-46887-4.
Correspondence and requests for materials should be addressed to Yuzhang Zhu or Jian Jin.
Peer review information Nature Communications thanks Qilei Song and the other, anonymous, reviewer(s) 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. (l) Check for updates
  2. College of Chemistry, Chemical Engineering and Materials Science, Innovation Center for Chemical Science & Jiangsu Key Laboratory of Advanced Functional Polymer Design and Application, Soochow University, Suzhou 215123, PR China. i-Lab, Suzhou Institute of Nano-Tech and Nano-Bionics, Chinese Academy of Sciences, Suzhou 215123, PR China. Department of Civil and Environmental Engineering, Vanderbilt University, Nashville, TN 37235, USA. CAS Key Laboratory of Bio-inspired Materials and Interfacial Science, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, 100190 Beijing, PR China. These authors contributed equally: Quan Peng, Ruoyu Wang.